Volume 137 1994 Tijdschrift voor Entomologie A journal of systematic and evolutionary entomology since 1858 OLOG SE, = N AA EVER 1 (ed } * > (es dy TE > A Tag N or Published by the Nederlandse Entomologische Vereniging Tijdschrift voor Entomologie A journal of systematic and evolutionary entomology since 1858 Scope The ‘Tijdschrift voor Entomologie' (Netherlands Journal of Entomology) has a long tradition in the publication of original papers on insect taxonomy and systematics. The editors particularly invite papers on the insect fauna of the Palaearctic and Indo-Australian regions, especially those including evolutionary aspects e.g. phylogeny and biogeography, or ethology and ecology as far as meaningful for insect taxonomy. Authors wishing to submit papers on disciplines related to taxonomy, e.g. descriptive aspects of morphology, ethology, ecology and applied entomology, are requested to contact the editorial board before submitting. Usually, such papers will only be published when space allows. Editors E. J. van Nieukerken (elected 1986) and J. van Tol (1985) Co-editors A. W. M. Mol (1990) and R. T. A. Schouten (1990) Advisory board M. Brancucci (Basel), N. E. Stork (London) and M. R. Wilson (Cardiff). The ‘Tijdschrift voor Entomologie’ is published in two issues annually by the ‘Nederlandse Entomologische Vereniging’ (Netherlands Entomological Society), Amsterdam. Editorial address c/o National Museum of Natural History, Postbus 9517, 2300 RA Leiden, The Netherlands. Correspondence regarding membership of the society, subscriptions and possibilities for exchange of this journal should be addressed to: Nederlandse Entomologische Vereniging c/o Instituut voor Taxonomische Zoölogie Plantage Middenlaan 64 1018 DH Amsterdam The Netherlands Subscription price per volume Hfl. 300,— (postage included). Special rate for members of the society. Please enquire. Instructions to authors Published with index of volume 137 (1994). Graphic design Ontwerpers B.V, Aad Derwort, 's-Gravenhage Tijdschrift voor Entomologie Contents of volume 137 Articles 143 155 161 173 271 325 27 57 331 Asche, M. & M. D. Webb Review of the southern Palaearctic and palaeotropical leafhopper genus Hengchunia Vilbaste (Homoptera, Cicadellidae). Baker, D. B. A new genus of nomadine bees from North Africa (Hymenoptera: Apoidea, Anthophoridae). Boer, A. J. de The taxonomy and biogeography of the lorida group of the genus Baeturia Stäl, 1866 (Homoptera, Tibicinidae). Boer, A. J. de Four species added to the Baeturia nasuta group, with notes on taxonomy and biogeography (Homoptera, Tibicinidae). Garrison, R. W. A revision of the New World genus Erpetogomphus Hagen in Selys (Odonata: Gomphidae). Jong, H. de The phylogeny of the subgenus Tipula (Savtshenkia) (Diptera: Tipulidae), with special reference to the western Mediterranean fauna. Keffer, S. L.: see Polhemus, J. T. Kjeerandsen, J. Three new species of PseudexechiaTuomikoski from Tanzania and Thailand (Diptera: Mycetophilidae). Kovac, D.: see R. Rozkosny Malicky, H.: see Weaver, |. S. Nieser, N.: see Polhemus, J. T. Patocka, J. Die Puppen der Spanner Mitteleuropas (Lepidoptera, Geometridae): Charakteristik und Bestimmungstabelle der Gattungen. Polhemus, D. A.: see Polhemus, J. T. Polhemus, J. T. & D. A. Polhemus Four new genera of Microveliinae (Heteroptera) from New Guinea. Polhemus, J. T., N. Nieser & S. L. Keffer Synonymical notes on the Nepa cinerea Linnaeus complex (Nepidae: Heteroptera). 87 R.Rozkosny & D. Kovac Adults and larvae of two Ptecticus Loew from Peninsular Malaysia (Diptera, Stratiomydiae). 87 Tol,)J. van The Odonata of Sulawesi and adjacent islands. Part 3. The genus Macromia Rambur (Corduliidae). 95 Weaver, J. S. & H. Malicky The genus Dipseudopsis Walker from Asia (Trichoptera: Dipseudopsidae). Webb, M. D.: see Asche, M. Referees for volume 137 B. Aukema, J. Belle, P. L. Beuk, P. Chandler, J. P. Duffels, C. Dufour, R. de Jong, C. Hauser, |. M. Kerzhner, C. Michener, A. Neboiss, P. Skou, K. D. P. Wilson, M. R. Wilson. Dates of publication Volume 137 (1), pages I-142, 15 July 1994 Volume 137 (2), pages 143-336, 15 December 1994 © Nederlandse Entomologische Vereniging, Amsterdam Issn 0040-7496 NEW TAXA DESCRIBED IN TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137: + DIPTERA Pseudexechia inthanonensis Kjærandsen ................329 Pseudexechia lanceostylus Kjerandsen ...................327 Pseudexechia longistylus Kjerandsen .....................325 Ptecticus malayensis Rozkosny & Kovac ................. 79 Tipula (Savtshenkia) alpha de Jong … … 311 Tipula (Savtshenkia) omega de Jong … … 315 Tipula (Savtshenkia) trinacia de Jong … … … … … … 317. HETEROPTERA Aegilipsicola Polhemus & Polhemus … … … 60 Aegilipsicola rapida Polhemus & Polhemus............ 61 Neusterinsifer Polhemus & Polhemus .................... 67 Neusterinsifer compactus Polhemus & Polhemus ....69 Neusterinsifer cyclops Polhemus & Polhemus.......... 69 Neusterinsifer gladius Polhemus & Polhemus......... 70 Neusterinsifer nabire Polhemus & Polhemus.......... 71 Neusterinsifer sepik Polhemus & Polhemus ............ 72 Tanyvelia Polhemus & Polhemus......................... 58 Tanyvelia missim Polhemus & Polhemus … … … … … 58 Tarsovelia Polhemus & Polhemus … … … … … … … … 63 Tarsovelia alta Polhemus & Polhemus................... 64 Tarsovelia arfak Polhemus & Polhemus................. 65 Tarsovelia dani Polhemus & Polhemus.................. 65 HOMOPTERA Baeturia bemmeleni de Boer... … 10 Baeturia daviesi de Boer... … 157 Baeturiawortwindde BOT... 20 [Biterumalgibberosa de BOER 1")... 169 Baeturiaihamiltoni de Boer … tuna 8 Baeturia hartonoi de Boer … … … … 19 [Dzeruniaypioramide BOT... 15 Baeturia retracta de Boer … … …… 170 Baeturia silveri de Boer … … …… 13 Baetuniarsplendida de Boer….............l 00)... 167 Daeruntanwepeneri.de BOF... serres 13 Hengchunia helleri Asche & Webb ...................... 147 Hengchunia indica Asche & Webb ...................... 150 Hengchunia javana Asche & Webb...................... 152 Hengchunia pakistanica Asche & Webb … … … … 150 HYMENOPTERA Herhammobates Baker... 155 Aethammobates prionogaster Baker........................ 156 ODONATA Erpetogomphus agkistrodon Garrison..................... 199 Erpetogomphus bothrops Garrison ......................... 212 Erpetogomphus elaphe Garrison ........... 205 Erpetogomphus heterodon Garrison ....................... 230 Erpetogomphus leptophis Garrison......................... 204 Erpetogomphus liopeltis Garrison ........................ 210 Macromia celebicavan Toll.::....2::cc.0cssesscessssessecesesss 88 TRICHOPTERA Dipseudopsis adiaturix Weaver & Malicky … … … 102 Dipseudopsis flinti Weaver & Malicky.................. 116 Dipseudopsis lucasi Weaver & Malicky … … … … … 123 Dipseudopsis malaisei Weaver & Malicky … … … 124 Dipseudopsis martinovi Weaver & Malicky … … … 124 Dipseudopsis robustior andamanensis Weaver 8clMalicley: an. an EE near con 134 Dipseudopsis schmidi Weaver & Malicky … … … … 134 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 INSTRUCTIONS TO AUTHORS The Tijdschrift voor Entomologie publishes original papers dealing with systematic and evolutionary ento- mology. The editors particularly invite papers on the insect fauna of the Palaearctic and Indo-Australian re- gions, especially those including evolutionary aspects e.g. phylogeny and biogeography, or ethology and ecology as far as meaningful for insect taxonomy. Authors wishing to submit papers on disciplines rela- ted to taxonomy, e.g. descriptive aspects of morpho- logy, ethology, ecology and applied entomology, are requested to contact the editorial board before sub- mitting. Usually such papers will only be accepted when space allows. Papers in English are preferred, but papers written in French or German will also be considered. It is our policy that papers are reviewed by at least one external referee. Authors will generally be notified of accep- tance within two months. For the first submission two printed copies are re- quired, including photocopies of figures reduced to A4 format. Diskette and original artwork should not be sent until the paper is accepted. Manuscripts should preferably be printed on A4 size paper, on one side only, double spaced, with a left margin of at least 4 cm. After acceptance of a paper, authors are re- quested to prepare a manuscript according to stan- dards given in a separate style-sheet. These instruc- tions include e.g. the wordprocessing programs we can handle and the codes that should be included or are allowed in your files. The editors can handle the most common wordprocessing programs for MS-DOS, Windows and Macintosh systems, but prefer Wordperfect (any version). Lengthy papers (more than 40 pages in print) are only accepted when space allows. Publication can be speeded up by paying a page-charge of NLG 50. Otherwise no page-charge is asked. Membership of the "Netherlands Entomological Society’ is not oblig- atory. Text preparation A cover page should provide the names of the aut- hors and a proposal for a running title. The second page starts with author names (in all capitals), use & for ‘and’, on a new line the name of institute (as short as possible), with multiple authors using superscript ' etc. The title is brief and informative, typed in all ca- pitals, with order and family of the taxon treated in parentheses. The abstract starts with a bibliographical reference (authors, year, title), see recent issues for layout. Then the abstract follows, without the word ‘abstract’ and without indentation. The written in English, should be concise, yet cover the main results abstract, vi of the paper, including new taxa and nomenclatorial changes. The name and address of one and only one of the authors follows, if needed preceded by the word ‘Correspondence:’. This should also be the ad- dress for the galley proofs. The abstract ends with a list of key-words. The text proper starts on a new page, the introduc- tion (if any) starts without heading. Use a maximum of three categories of headings, all to be typed left, and using capitals and lower-case letters. The first type may also be typed in small capitals, the second and third type may be typed in bold. After the head- ing the paragraph follows without blank. A fourth category may be used in descriptions of species, etc. It is separated from the following paragraph by a long dash (+) to be typed as double dash. New paragraphs should »ot be indented in word- processors. Scientific names of genera and species should be typed in italics or underlined. No underli- ning or italics are allowed for any other text. Abbreviations of museums (‘codens’) etc. should be typed in SMALL CAPS, if available. References In the text they are given as Lopes (1982a), (Lopes 1982) or (Brown & White 1975: 24). All cited papers should be listed alphabetically at the end of the paper under the heading ‘References’, papers not cited in the text should be omitted from the list of references. Examples for format: Boer, P. J. den, 1970. On the significance of dispersal pow- er for populations of carabid-beetles (Coleoptera, Carabidae). — Oecologia 4: 1-28. Karsholt, O. & E. S. Nielsen, 1976. Systematisk fortegnelse over Danmarks sommerfugle. — Scandinavian Science press, Klampenborg, 128 pp. Johansson, R. & E. S. Nielsen, 1990. Tribus Nepticulini. — In: Johansson, R. et al. The Nepticulidae and Opostegidae (Lepidoptera) of NW Europe. — Fauna en- tomologica scandinavica 23: 111-238, pls. Titles of journals should not be abbreviated. Type long dashes as double dash ‘- -’, or in WordPerfect use ‘en dash (4,33). Do not try to type indentation, just end each reference with a hard return. Nomenclature The latest edition of the ICZN Code should be fol- lowed. The composition of new names should prefer- ably be explained in a paragraph ‘Etymology’, includ- ing indication of gender of generic names and kind of specific name (adjective, noun in apposition, etc.). Use standard abbreviations: Sp. n., gen. n., comb. n., syn. n., sp. rev., nom. n., etc. For all genus and spe- cies-group names the authority (preferably with year of description) should be mentioned once. Author’s names are not abbreviated. In new taxa the type material should be listed im- mediately after the name. Only holotype, lectotype, neotype, paratype and paralectotype are allowed. Label data should not be quoted literally (except for primary types), but arranged in a standardized se- quence. Material should be listed alphabetically or chronologically under the present day countries or other geographical units. Long lists of non-type mate- rial should be summarized. Geographical names should be written according to present day spelling, original spelling or label names may be given in brackets. Use standard transcription for non-latin scripts (e.g. Pinyin for chinese, BSI for cyrillic, etc.) or refer to the “Times Atlas of the World’. Abbreviations (‘codens’) for depositories preferably follow R. H. Arnett, Jr., G. A. Samuelson & G. M. Nishida (1993: The insect and spider collections of the world. Second edition. Sandhill Crane Press, Gainesville). Otherwise, they should be listed under “Material and methods’ or in the introduction. Data for primary types of previously described spe- cies follow directly the reference to the original de- scription as: Elachista subnitidella Duponchel, [1843]: 326, pl. 77: 8. Lectotype & [designated by van Nieukerken & Johansson 1987: 471]: [Austria, Vienna region], Duponchel coll., Genitalia slide EVN 2522 (MNHN) [ex- amined]. Illustrations All illustrations, including photographs, graphs, maps, etc. should be serially numbered as figures. No subdivision with letters is recommended. Illustrations are to be reduced to column width (65 mm), 1.5 X column width (102.5 mm) or text width (135 mm). Line figures should be mounted in blocks, or are printed singly. When all figures are mounted in full- page blocks (after reduction: 135 X 195 mm includ- ing caption), they may be printed after the text, oth- erwise the approximate place in the text should be in- dicated with pencil in the margin of the manuscript. Line-drawings are numbered with pre-stencilled or pre-printed figures, which should not be too large af- ter reduction, preferably using a font like ‘Garamond’ or ‘Times’. Photographs should be unmounted glossy prints. Numbering of photos should be left to the dis- cretion of the editors. Captions should be typed on a separate sheet (or in a separate file), consult a recent issue for style. Colour plates will only be printed at the author’s expense. Tables Tables should be typed on separate sheets (or files), starting with the captions. When using a wordproces- sor: start with a practical TAB setting, and use only one [TAB] code for each next column. No formatting with spaces is allowed. No lines should be added. Extensive and long tables should be avoided. Proofs, reprints Authors receive one proof only, which should be corrected and returned immediately. When correc- tions are few, sending per telefax is recommended. Authors receive 50 reprints free of charge. Additional reprints can be ordered when proofs are returned. Members of the Netherlands Entomological Society receive a considerable dis- count. Covers can be ordered at extra cost. All correspondence should be addressed to: Tijdschrift voor Entomologie National Museum of Natural History Postbus 9517 NL-2300 RA Leiden The Netherlands (Phone +31-71-143844, telefax +31-71-133344). vii i pompino mim anne au ve sd a last ent LL ua de Sater aa wits sails ef | | poser. pal hemd Ya qui nm shew nun se ann ue | LE bl» TTI K Pl di a Wik R M) ERN atleta Dan BR ai opa herbe wie. att PES (le ga n o” AT ROET dirkk STARE haag e 4 Pa TR n | Li ie ? ge M by win ij ; 6 phar eS: ila pil) stede TRE VENTE NM or DUT AM Lui je A Py. VPE taupe: pn ip “ it N y lab; ri. wre LR M dele +4 au Hel Eu a by a put TEL ET N Aa | at Ne yale, Gv dé iù “y ai TE OL LI Ù | DRE | “para ehr PI aday i aw UM UT "1 ny ibe fa. oo ed, REN iS N ralf a We ted Lp gui; NPT UA A bh ea Lew we à berk, 4 TRA anced + tro u a | ‘a a ni | i Dell lu write ‘(in N Cagli, ly Tijdschrift voor _ Entomologie A journal of systematic and evolutionary entomology since 1858 Tijdschrift voor Entomologie A journal of systematic and evolutionary entomology since 1858 Scope The ‘Tijdschrift voor Entomologie' (Netherlands Journal of Entomology) has a long tradition in the publication of original papers on insect taxonomy and systematics. The editors particularly invite papers on the insect fauna of the Palaearctic and Indo-Australian regions, especially those including evolutionary aspects e.g. phylogeny and biogeography, or ethology and ecology as far as meaningful for insect taxonomy. Authors wishing to submit papers on disciplines related to taxonomy, e.g. descriptive aspects of morphology, ethology, ecology and applied entomology, are requested to contact the editorial board before submitting. Usually, such papers will only be published when space allows. Editors E. J. van Nieukerken (elected 1986) and J. van Tol (1985) Co-editors A. W. M. Mol (1990) and R. T. A. Schouten (1990) Advisory board M. Brancucci (Basel), N. E. Stork (London) and M. R. Wilson (Cardiff). The ‘Tijdschrift voor Entomologie’ is published in two issues annually by the ‘Nederlandse Entomologische Vereniging’ (Netherlands Entomological Society), Amsterdam. Editorial address c/o National Museum of Natural History, Postbus 9517, 2300 RA Leiden, The Netherlands. Correspondence regarding membership of the society, subscriptions and possibilities for exchange of this journal should be addressed to: Nederlandse Entomologische Vereniging c/o Zoölogisch Museum Amsterdam Plantage Middenlaan 64 1018 DH Amsterdam The Netherlands Subscription price per volume Hfl. 300,— (postage included). Special rate for members of the society. Please enquire. Instructions to authors Published with index of volume 136 (1993). Graphic design Ontwerpers B.V., Aad Derwort, ’s-Gravenhage A. J. DE BOER Institute for Systematics and Population Biology (Zoölogisch Museum), University of Amsterdam THE TAXONOMY AND BIOGEOGRAPHY OF THE LORIAE GROUP OF THE GENUS BAETURIA STÄL, 1866 (HOMOPTERA, TIBICINIDAE) Boer, A. J. de, 1993. The taxonomy and biogeography of the doriae group of the genus Baeturia Stal, 1866 (Homoptera, Tibicinidae). — Tijdschrift voor Entomologie 137: 1-26, figs. 1-102, table 1. [ISSN 0040-7496]. Published 15 July 1994 The /oriae group is proposed for a presumably monophyletic group of 9 species. Species of the group combine characters seemingly synapomorphic for the species of Baeturia Stal, 1866, with characters seemingly synapomorphic for the species of Gymnotympana Stal, 1861. Though the generic relationships of the loriae group are not clear, it is included here in the genus Baeturia, on account of shared male genital characters. One species (B. loriae Distant, 1897) is rede- scribed and eight species (B. bemmeleni, B. daviesi, B. fortuini, B. hamiltoni, B. hartonoi, B. pi- grami, B. silveri, and B. wegenert) are described as new. B. loriae is transferred back from Gymnotympana to Baeturia. A key to the males is presented. A redescription of B. tenuispina Blöte, 1960 is included in this publication, since a character of the clasper indicates a relation- ship with some species of the loriae group. Nevertheless, the attribution of that species to the loriae group is far from certain. The loriae group is endemic to New Guinea, most species are restricted to the central and eastern mountain ranges of that island. A. J. de Boer, Institute for Systematics and Population Biology (Zoölogisch Museum), University of Amsterdam, PO Box 4766, 1009 AT Amsterdam, The Netherlands. Key words. — Baeturia, loriae group, taxonomy, biogeography, New Guinea. Current studies of the tibicinid cicadas of eastern Malesia show, that a number of New Guinean and Australian genera form a monophyletic group: ‘the Baeturia and related genera complex’ (de Boer 1990, 1991). The largest genus of this group, Baeturia Stal, 1866, appears to be non-monophyletic in its present concept, but, within that genus, several monophylet- ic species groups have been recognized. Taxonomic and biogeographic revisions have been published so far for the B. nasuta group (de Boer 1982), the B. con- viva group (de Boer 1986), the B. bloetei group (de Boer 1989) and the B. viridis group (de Boer 1992). These pages present the revision of a presumably monophyletic group of nine species, for which the name Baeturia loriae group is proposed. This group is of particular interest from a taxonomic point of view, since its species combine several characters regarded apomorphic for Baeturia, with characters apomor- phic for Gymnotympana Stal, 1861. The distribution of these characters within the group is such, that any reconstruction of the in-group relationships will un- avoidably show several conflicting characters. The B. loriae group is endemic to northern, eastern and cen- tral New Guinea. B. tenuispina Blöte has a character of the clasper in common with some species of the B. loriae group, but differs in most other respects. This species is not at- tributed to the B. loriae group, but described here, since it seems more related to this group than to any of the other species groups of Baeturia. MATERIAL AND METHODS The material used for this study comes from the following institutions: AMS — Australian Museum, Sydney; BMNH — Natural History Museum [formerly: British Museum (Natural History)], London; BPBM — Bernice P. Bishop Museum, Honolulu; CSIRO — Commonwealth Scientific and Industrial Research Organisation, Australian National Insects collection, Canberra; KBIN — Koninklijk Belgisch Instituut voor Natuurwetenschappen, Brussel; MSNG — Museo Civico di Storia Naturale ‘G. Doria’, Genova; MVM — Museum of Victoria, Melbourne; RMNH — Nationaal Natuurhistorisch Museum (formerly: Rijksmuseum van Natuurlijke Historie), Leiden; SMN — Staatliches Museum für Naturkunde, Stuttgart; ZMA — Institute for Systematics and Population Biology (Zoölogisch Museum), Amsterdam. The following sources have been used for tracing TypscH RIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 the localities: ‘Atlas van tropisch Nederland’ (1938), ‘The Times Atlas of the World (Comprehensive Edition)’ (1968), ‘Papua New Guinea a travel survi- val kit by Tony Wheeler, 1988, Lonely Planet Publications, Victoria, Australia and a ‘List of New Guinea localities’ published by the Bernice P. Bishop Museum (1966). To examine the male genitalia, the pygofer was pulled out, after overnight softening, with a sharp needle inserted between the pygofer and the 8th ab- dominal segment. The aedeagus was pulled out at the same time, by inserting the needle between the clasp- ers. Body and tegmen lengths of all specimens were measured, other measurements are based on a maxi- mum of ten specimens, when available. PHYLOGENY Classification of the B. loriae group The generic classification of the loriae group pre- sents a major problem. As stated above, the species combine characters of Baeturia with characters of Gymnotympana. Some of these characters appear to be unique for either Baeturia or Gymnotympana (or part of these genera). The distribution of these char- acters within the loriae group is not homogeneous; some species have more in common with Baeturia, others more closely resemble Gymnotympana. The following characters are found in the loriae group and indicate a relationship with Baeturia, the first four are presumed apomorphies for that genus. (1) a strongly curved aedeagus, tapering to its apex, with bluntly rounded lateral lobes and oval aede- agus pore. This aedeagus shape is unique for Baeturia. (2) a narrow vertex, with large ocelli close together (fig. 3). Two species of Gymnotympana have an even narrower vertex. In other species of the complex, the vertex is broader than in Baeturia. (3) a broad and triangular middle spine on fore fe- mur (fig. 19) is shared by all species of Baeturia. (4) a fairly narrow and frontally bluntly rounded postclypeus, distinctly protruding beyond vertex lobes and, in most species, distinctly and angu- larly swollen ventrally (figs. 3-4). A similar swel- ling only occurs in some species of Baeturia, the narrow, protruding, and rounded shape of the postclypeus is possibly apomorphic for Baeturia. (5) a rather stout caudodorsal beak, often angularly bent and curved over anal valves. Other species of the complex have a much shorter and erect beak. The curved caudodorsal beak is probably synapomorphic for several of the Baeturia spe- cies groups together. (6) almost straight parallel claspers, separated at the ho base, and with a broad and rounded ventral hol- low in the apical part. Such separated claspers are unique for Baeturia and Scottotympana de Boer, 1991: (7) a distinct and angular clasper heel (in B. bemme- leni, B. hamiltoni, and B. wegeneri). A clasper heel is found in many species of Baeturia, but not in other genera. (8) rows of setae on veins of tegmina and wings (in B. hamiltoni and B. wegeneri). Similar rows of setae are found in many Baeturia species, but not in any of the related genera. However, such setae also occur, though more distinctly, in several ge- nera of the Prasiini, a possible sister group of the ‘Baeturia and related genera complex’. (9) dense brown speckling on body (in B. daviesi and B. wegeneri). Such speckling, though strong- ly variable in density, only occurs within Baeturia. (10) irregularly shaped brown spots on tegmina (in B. pigrami and B. silveri, fig. 59). Such spots are on- ly found in some species of Baeturia (e.g. B. gut- tulinervis, B. guttulipennis and B. nasuta). (11) a distinct crest between the aedeagal lobes. A si- milar crest occurs in some species of the B. con- viva group and some species related to that group. Other characters found in the B. loriae group do not occur in other species groups of Baeturia, but seem to indicate a relationship with Gymnotympana. The first two must possibly be regarded as apomorp- hies for Gymnotympana, others are either restricted to part of that genus, or occur in several other genera as well. (1) sexual dimorphism in wing venation (males with very slender anal fields, compare figs. 59 and 84) is shared by all species of Gymnotympana, but does not occur in other genera of the complex. (Similarly slender anal fields are recorded for Prasia Stäl and Plautilla Stàl (Boulard 1975), however. In Prasia this character is restricted to the males, but the literature does not record such dimorphism for Plautilla). (2) a sharp ventrolateral fold in tergites 4-7 (more or less distinct in all species) is shared by nearly all species of Gymnotympana, but not found in rela- ted genera. (3) an enlarged distal part of male operculum. Both, the broadly rounded shape (as found in B. bem- meleni, B. hamiltoni, and B. wegeneri), as the triangular shape (as found in B. daviesi, B. for- tuini, and B. pigrami), are found in Gymnotympana, but do not occur in related ge- nera. (4) a very short,. often rudimentary meracanthus (found in all species with enlarged opercula) is shared by Venustria Goding & Froggatt, 1904 and most species of Gymnotympana. This char- acter might be coupled to the character descri- bed immediately above. (5) alaterally elongate basal part of male operculum (in six species) is shared by nearly all Gymno- tympana species and not found in other genera. (6) abroad hyaline border along hind margin of teg- men (in B. fortuini and B. loriae, fig. 84) is sha- red by all species of Gymnotympana, Scottotym- pana, and Venustria. (7) an inwards curving ventral corner of tymbal, with strongly inwards directed connecting bar between abdomen and tymbal (only found in 2. fortuini) is found in most species of Gymno- tympana, but not in other related genera. (8) furthermore, all species except B. daviesi and B. wegeneri have a more or less distinct colour pattern of dark streaks and spots on head, pronotum, and abdomen (figs. 8-9). This colour pattern is much more distinct than in other species of Baeturia. An even more distinct colour pattern is found in ma- ny Gymnotympana species, in Scottotympana and in two species of Chlorocysta Westwood, 1851. Since these patterns are very variable between the various species, it is not clear whether the patterns found in the loriae group can be considered ho- mologue with those in Gymnotympana. The loriae group thus, seems to combine some ap- parently convincing synapomorphies of Baeturia and Gymnotympana. At this point of investigation, with several groups of the ‘Baeturia and related genera complex’ still undescribed, and with only vague indi- cations as to the possible sister groups of the complex, it is not possible to fully evaluate the meaning of the character distribution in the loriae group. Current phylogenetic analysis, involving all species of the ‘Baeturia and related genera complex’, however, shows that a monophyletic loriae group as part of the genus Baeturia, is the most parsimonious solution. Furthermore, this solution is in accordance with the distribution of all male genital characters. Nevertheless, the remarkable number of conflic- ting characters concentrated in the loriae group is upsetting to any phylogenetic reconstruction. Whatever generic allocation of the loriae group is fa- voured, the parallel evolution and / or the successive gain and loss of several characters must be presumed. Biochemical study of the loriae group could be inte- resting as a test for a hybrid origin of that group. Monophyly of the B. loriae group The above discussed combination of characters found in the loriae group involves all those characters DE Boer: The loriae group of Baeturia that would, by non-existence of that group, be used as apomorphies for either Baeturia or Gymnotympana. As a result neither Baeturia nor Gymnotympana can be properly defined by shared apomorphies in exclu- sion of the loriae group. But also for the loriae group itself, no unambiguous apomorphies could be found. The group can be best defined by the unique combi- nation of two characters: sexual dimorphism in wing, and an S-curved aedeagus with lateral lobes at the ba- se of the curvation. Furthermore, there are three char- acters shared by all, or nearly all, species of the group, and not widely distributed outside the group, that could indicate a close relationship between the spe- cies. These characters are: 1) a distinct crest between the lateral lobes of aedeagus, only shared by some spe- cies of the B. conviva group and two species related to that group; 2) an enlarged and angularly rounded ventral part of the lateral lobe of pygofer, shared by only a few species of Gymnotympana (e.g. G. stridens); and 3) a rather short and broad eighth apical area of tegmen, tending to be shorter than in Gymnotympana or other species of Baeturia. The latter character is so- mewhat variable within the species and cannot be used as a diagnostic character. Ingroup phylogeny As long as the problems concerning the phylogene- tic position of the loriae group, relative to other spe- cies groups of Baeturia and to the genus Gymnotympana, are not satisfactorily solved, charac- ters that occur either in Baeturia or Gymnotympana and also in some species of the loriae group cannot be properly evaluated as plesiomorphous or apomorp- hous, and can therefore not be used in a phylogenetic reconstruction of the group. Only the distribution of characters that appear to be unique for some species of the loriae group allows some remarks on ingroup phylogeny. B. bemmeleni, B. hamiltoni, and B. wegeneri almost certainly form a monophyletic group. These species share an almost identical clasper, characterized by a nearly rectangular dorsal corner (figs. 11, 26, and 30), which is regarded synapomorphic. Furthermore, these species share an almost identical male opercu- lum; very broad, slightly domed and broadly rounded at its apex (figs. 14, 22, and 38). A very similar oper- culum was found in G. olivacea Distant, 1905, but in that species the lateral margin of operculum is almost continuous with the lateral crest of operculum base. Some undescribed Gymnotympana species, related to G. olivacea, have a similarly broad and curved, but more angular, square-shaped, operculum. B. daviesi, B. fortuini, B. hartonoi, B. loriae, B. pi- grami, and B. silveri, most probably form a monophy- letic group on account of a shared dorsal protrusion on the clasper. This protrusion is partly, or complete- TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 136° wbemmeleni A hamiltoni @hartonoi AQ» @tenuispina Bwegeneri Fig. 1. Localities of Baeturia bemmeleni, B. hamiltoni, B. hartonoi, B. tenuispina, and B. wegeneri. ly fused with the upwards curving proximal part of the dorsal margin of clasper, and merges with the clasper base, were in many other Baeturia species the clasper heel is formed. B. tenuispina Blöte has a simi- larly shaped protrusion at the corner of its clasper heel (fig. 97) and should possibly be included in this group. In B. daviesi, B. pigrami, and B. silveri the dorsal protrusion is very distinct and finger-shaped (figs. 40, 50, 64); in B. hartonoi, this protrusion is more angu- lar and strongly curved mesiad (fig. 70); and in B. for- tuini it is rudimentary and only recognizable in the specimens from Mt. Missim (fig. 79). The dorsal pro- trusion is believed to be lost in the lectotype of B. lor- sae and the remaining specimens of B. fortuin. The clasper of these six species is very broad in lat- eral view; the dorsal margin of the clasper curves up- wards from clasper base, so that no clasper heel is formed. This clasper-shape is probably also synapo- morphous. B. fortuini, B. loriae, and B. hartonoi possibly form a monophyletic group on account of the shared elon- gate lateral lobes of aedeagus (figs. 73, 82, 92), and the very broad, angular and laminiform clasper. B. fortuini and B. loriae are very closely related and probably sister species. They share triangular bronzed spots in the tegmina (fig. 84) and a very long and slender aedeagus (figs. 82, 92). Possibly these two species are synonyms, and the lectotype of B. loriae is just an aberration. The relationships between B. daviesi, B. pigrami, and B. silveri are less clear. These species share a trian- gular clasper, with convex dorsal margin (lateral view), which must probably be regarded more primi- tive than the squarely laminiform clasper of the three species mentioned above. Especially the claspers of 2. daviesi and B. pigrami are very similar, sharing a broadly rounded and outcurving dorsal lobe, but the clasper of B. hartonoi has a very similar outcurving lobe. B. pigrami and B. silveri are conspicuous by the brown patches along veins of tegmina. But B. guttuli- nervis Blöte and B. guttulipennis Blöte, two otherwise very different and not closely related species, have similar patches. Table 1. Altitudinal ranges of the species of the B. loriae group in meters above sea level. B. bemmeleni 35-1350 B. fortuini 1100-2100 B. hamiltoni 50-1200 B. hartonoi 100 B. loriae 1300 B. pigrami 120-1260 B. silveri 1800-2590 (8500 ft) B. wegeneri 250 eee 0° + daviesi e fortuini = /oriae v pigrami a silveri 4° 8° 12% 148° Fig. 2. Localities of Baeturia daviesi, B. fortuini, B. loriae, B. pigrami, and B. silveri. BIOGEOGRAPHY The B. loriae group is endemic to New Guinea, and distributed in the northeastern part of Irian Jaya, and in central and eastern Papua New Guinea. Many specimens come from, or close by, the central and eastern mountain ranges. The group seems to consist mainly of montane species, probably with very res- tricted areas of distribution, often reaching altitudes of over 1000 m (table 1). It is remarkable, that the subdivision made above, between B. bemmeleni, B. hamiltoni, and B. wegeneri versus the remaining species, is a subdivision between a mainly western and a mainly eastern species group (figs. 1, 2). TAXONOMY Description of the B. loriae group The species of the B. loriae group are predominant- ly brown coloured, sometimes greenish tinged. B. we- generi and B. daviesi are densely speckled with small brown spots, the other species are conspicuous by a colour pattern of regularly distributed dark markings. These species generally have two dark spots or longi- tudinal streaks between eyes and lateral ocelli and of- ten darkened lateral corners of postclypeus; a light immaculate middorsal band on pronotum, someti- DE BoER: The loriae group of Baeturia mes bordered by almost black lines, and dark brown streaks in and along the oblique fissures. The dark markings on the mesonotum form two semi-circular paramedian spots, at pronotal collar, two blackened dents in front of cruciform elevation and lateral bands, converging from pronotal collar to edges of elevation. The cruciform elevation and a triangular area in front of the elevation is generally light och- raceous and unmarked. On the male abdomen, dark markings tend to form dorsal, and sometimes lateral bands, while a latero-ventral row of dark spots is often very clear. The dorsal markings tend to be interrupted middorsally by a narrow lighter coloured line. Ventral side of male abdomen generally light ochraceous, seg- mental hind margins often red. Female abdomen mo- re irregularly dark spotted all over, though the light middorsal band is often more conspicuous. Females are shorter than males (on average), but have a larger, more robust head and thorax and often longer tegmina. Head in dorsal view (fig. 3): Postclypeus distinctly protruding beyond vertex lobes and broadly rounded at frontal margin. Postclypeus 1.2-2.2X as broad as long and 0.5-0.7 X as broad as distance between eyes, strongly varying within the species. Head 0.7-0.8 X as wide as pronotal collar and only slightly narrower than anterior width of pronotum. Vertex narrow, ocelli large and close together. Distance between la- teral ocelli less than 1.5 X the width of frontal ocellus and 0.7-1.4 X as long as distance between eye and la- teral ocellus. Eye 0.6-0.8 X as wide as distance be- tween eyes. Head 0.9-1.2 X as long and 2.2-2.6 X as wide as distance between eyes. Legs: Ochraceous, sometimes with longitudinal brown stains on femora and tibia. Fore femur (fig. 19) with row of three sharply pointed spines, dimi- nishing in length towards tibia. Proximal spine about as long as distance to middle spine. Middle spine broad, triangular. Tegmina and wings: Hyaline, though tegmina in some species with brown markings. Venation och- raceous, costa often reddish. Tegmina with 8, wings with 6 apical areas. Costal area of tegmen very nar- row. Anal fields of wing in male distinctly narrower than in female (compare figs. 59 and 84). Tymbal organ (fig. 6): Six or seven parallel trans- verse sclerotized ridges spanning the tymbal from dorsal to ventral margin. Short intercalary ridges for- ming a midlateral band across tymbal. Opercula: Basal part of operculum vaulted, with distinct crest around rectangular distolateral corner. Distal part of male operculum either very long and domed, and completely covering tymbal cavity in ventral view; or short, oblong-shaped, laminiform and flat against body, only partly covering tymbal ca- vity. Distal part of operculum medially extending TITDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 mn is À NA Fies. 3-7. Baeturia hamiltoni sp. n. — 3, Head in dorsal view; 4, postclypeus in lateral view; 5, female genital segment in later- al view; 6, tymbal; 7, female operculum. 6 Fig. 8. Baeturia hamiltoni sp. n., habitus male. beyond meracanthus. Distal part of female opercu- lum short, sickle-shaped and erect, medially exten- ding just to base of meracanthus. Basal part of female operculum forming narrow rim around base of mer- acanthus. Abdomen: With tergites 4-7 of the male abdomen sharply folded at the ventrolateral edge of abdomen, though these folds are not always distinct an all speci- mens. Male genitalia: Pygofer with stout and often angu- larly bent caudodorsal beak. Lateral lobes of pygofer with small and bluntly rounded protuberances, that do not extend beyond pygofer margin. Lateral lobes DE Boer: The loriae group of Baeturia Fig. 9. Baeturia hamiltoni sp. n., habitus female. strongly folded inwards. Claspers parallel and direc- ted posteriad, slightly bending down towards their apices. Apical part of clasper with small, rounded and sharply edged clasper hollow. Aedeagus strongly S- curved, with two narrowly elongate lateral lobes at ba- se of S-curvation, and a distinct crest berween lateral lobes. Aedeagus in lateral view broad at base, gradual- ly narrowing to pointed apex. Female genitalia: Caudodorsal beak sharply poin- ted at apex. Ovipositor sheats reaching just beyond apex of beak (fig. 5). Etymology: This revision is to form part of an area cladistic study of the New Guinean region, in which TiJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 the general patterns of distribution of organisms are compared to the geological history of that island. The new species described in this publication are named in honour of geologists, who contributed to a pale- ogeographic reconstruction of the area. The species are named after R. W. van Bemmelen, H. L. Davies, A. R. Fortuin, W. B. Hamilton, H. M. S. Hartono, C. J. Pigram, E. A. Silver, and A. Wegener. Key to the males 1. Clasper with distinct angular clasper heel; apical part of clasper slender in lateral view, with rectan- gular dorsal corner (fig. 11). Distal part of oper- culum broad and long, weakly domed, and bro- adly rounded at apex (fig.14) … 2 — Clasper without clasper heel; apical part of clasp- er in lateral view broader than clasper base, often with dorsal protrusion, partly fused to proximal part of dorsal margin (fig. 40). Distal part of op- erculum either oblong and flat (fig. 41), or trian- gular, domed and pointed (fig. 51) at apex … … 3 Body densely speckled all over, not forming an unspeckled or lighter coloured middorsal band … En en B. wegeneri — Body not speckled, but with regularly distributed dark markings and an immaculate middorsal band over part or whole length of body … … …. 4 3. Body length 25.4-29.7 mm. Caudodorsal beak broad and truncate at apex ............. B. hamiltoni — Body length 17.1-21.3 mm. Caudodorsal beak narrowly truncate or pointed at apex .................. ERE De B. bemmeleni 4. Aedeagus very long, with elongate and slender lateral lobes (fig. 82). Clasper quadrangular in lateral view; dorsal part of clasper angular, lamin- tormui(fips:70, 79480389) rennen 5 — Aedeagus short and broad, lateral lobes not elon- gate (fig. 45). Clasper triangular in lateral view; dorsal part of clasper swollen, with weakly convex dorsal margin (figs 40, 50, 64) ......................... 7 Postclypeus not swollen in lateral view. Tegmen immaculate. Aedeagus angularly bent to apex - (fig. 73). Clasper with dorso-lateral lobe (fig. 70) B. hartonoi — Postelypeus distinctly swollen in lateral view. Tegmen with triangular brown spots in apical ar- eas (fig. 84). Aedeagus not angularly bent to apex (fig. 81). Clasper without dorso-lateral lobe (fig. LE) OCR PRET PRE SRE ET 6 6. Distal part of operculum triangular, domed and narrowly rounded at apex (fig. 77). Margin of ae- deagus pore concave at apex (fig. 83) .B. fortuini — Distal part of operculum square-shaped, flat against body (fig. 91). Margin of aedeagus pore not concave at apex (fig. 93) .................. B. loriae to A 7. Tegmina with brown markings along veins (fig. DON RE, RE 9750050090900090000000000RR0 8 — Tegmina immaculate ......................... B. daviesi 8. Postclypeus angularly swollen in lateral view. Distal part of operculum triangular and slightly domed. Hind margin of tegmen with distinct hyaline border2................ Sees B. pigrami — Postclypeus not swollen in lateral view. Distal part of operculum square-shaped and flat against body. Hind margin of tegmen with narrow hya- line border'.........2......... eee eee B. silveri Description of the species Baeturia hamiltoni sp. n. (figs. 1, 3-20) Type material. — Holotype d: ‘Neth. Ind.-American New Guinea Exped. Rattan camp, 1200 m, 6.iii.1939, L. J. Toxopeus’, RMNH. - Paratypes: IRIAN JAYA: NEW GUINEA (W): Araucaria camp, 800 m, 8.iii.1939, L. J. Toxopeus, Id, 32, RMNH; same data but 9.11.1939, 16, 19; 10.11.1939, 3257 11.919397 IC MP 19395125; 16.10.1939, 1d, 29; 20.11.1939, 19; 21.11.1939, 26; 22.11.1939, 26, 29; 24.11.1939, 29; 25.11.1939, 16; 29.11.1939, 13, all RMNH; same data but 19.iii.1939, 29; 28.1.1939, 2d, all ZMA; Bernhard camp, 100 m, 11.iv.1939, L. J. Toxopeus, 2 9, RMNH. Other material. — PAPUA: NEW GUINEA (NE): Ambunti, Sepik R., 50 m, 10.v.1963, R. Straatman, 1d, BPBM. B. hamiltoni is a large brown coloured, species with a light middorsal band on head and thorax, sometimes continuing over abdomen. The species is easily recognized by a small and almost black medial spot on pronotum, at pronotal col- lar. Females have a glistering middorsal band of silvery setae on the abdomen, making the light and immaculate middor- sal band more conspicuous than in males. Description Body of males brown, abdomen sometimes red- dish, with distinct colour pattern of darkened mar- kings and an immaculate middorsal band on head and thorax, sometimes continuing over abdomen. Females on average slightly smaller than males, with more robust head and thorax, but shorter abdomen and considerably longer tegmina. Male abdomen 1.3- 1.4x as long as head and thorax, of female 0.9-1.0x. Tegmina of males 1.1-1.2X as long as total body length, of females 1.4-1.5x. Head (fig. 3): Light brown, with traces of red around ocelli, longitudinal dark brown to black streaks on vertex lobes, between eyes and lateral ocel- li, and often brown spots at margins of postclypeus. Postclypeus very stout, 1.3-1.9X as long as broad, distinctly protruding beyond vertex lobes and broad- ly rounded at anterior margin. Postclypeus in lateral view (fig. 4) strongly inflated, and angularly protrud- ing; dorsal part of anterior margin (lateral view) al- DE Boer: The loriae group of Baeturia Figs. 10-20. Baeturia hamiltoni sp. n. — 10, pygofer in lateral view; 11, clasper in lateral view; 12, clasper in dorsal view; 13, male caudodorsal beak in dorsal view; 14, male operculum; 15, aedeagus from aslant; 16, aedeagus in lateral view; 17, aedea- gus from behind; 18, detail crest between lateral lobes of aedeagus; 19, fore femur; 20, female caudodorsal beak in dorsal view. Lettering: bp = basal part of operculum; c = crest between lateral lobes of aedeagus; cb = caudodorsal beak; dm = distomedi- al margin of operculum; di = distal margin of pygofer; dc = crest around distolateral corner of basal part of operculum; do = dorsal margin of pygofer; dp = distal part of operculum; he = clasper heel; ho = clasper hollow; la = lateral margin of opercu- lum; m = meracanthus; me = medial margin of operculum; p = protuberance on lateral lobe of pygofer; ve = ventral margin of pygofer. most straight, ventral part strongly concave. Head very narrow between eyes, distance between lateral ocelli only slightly larger than width of frontal ocel- lus. Distance between lateral ocelli 0.8-1.2 x distance between eye and lateral ocellus. Head narrower than anterior part of pronotum. Thorax: Pronotum brown, with broad dark brown streaks, in and along oblique fissures, and broad, im- maculate medial band. Markings variable in intensity and shape, and sometimes completely absent. All specimens with a small dark brown or black middor- sal spot at margin of pronotal collar. Mesonotum light brown, with two paramedian dark spots at pro- notum margin, darkened lateral streaks converging from pronotal collar to corners of cruciform elevation and two black spots in front of elevation. Cruciform elevation, and a narrow triangular area in front of el- evation, often reaching to pronotal collar, light och- raceous and immaculate. Legs: Fore femur (fig.19) with row of three erect spines, diminishing in length towards tibiae. Tegmina and wings: Hyaline, though slightly bronzed, venation ochraceous or reddish tinged, but costa often bright red. Veins densely set with short se- tae. Tegmina with very short apical areas, 8th area al- most square-shaped, and with a very narrow hyaline border along hind margin. This border slightly broad- er in wings. Tymbal organs (fig. 6): Tympanum not curving in- wards; connecting bar between tymbal and abdomen quite short and almost parallel to body axis. Six trans- verse sclerotized ridges spanning the tymbal, from dorsal to ventral margin. À 7th ridge, close to proxi- mal tymbal margin, almost reaching ventral margin. TijpSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Six short intercalary ridges seem to form a band across tymbal. Opercula: Male operculum (fig. 14) ochraceous brown with darkened streaks. Basal part of opercu- lum almost oblong-shaped, but at half-width slightly longer than at lateral or medial margin. Distal part of operculum very broad, shell-shaped, and almost flat against body. Operculum broadly rounded at apex and reaching to, or beyond, margin of abdominal seg- ment 3. Distomedial margin and distal % of lateral margin convex, proximal % of lateral margin straight. Medial margin short and straight, recurving to base of meracanthus. Opercula well separated medially; 1st sternite slightly elevated between opercula, with dis- tinct distal lobe. Meracanthus very short, reaching to about % the operculum length. Female operculum (fig. 7) sickle-shaped and erect, with weak crest along distal margin; area between operculum and base of meracanthus broad and flat. Abdomen: Male abdomen light brown, with some darker brown spots on lateral sides and latero-ventral rows of dark spots on segments 3-7. Segments 2-7 darkened dorsally as well, though these markings of- ten interrupted by a narrow lighter middorsal band. Segment 8 almost entirely dark brown, but pygofer ochraceous. Ventral side of abdomen light brown, immaculate. Segmental hind margins often reddish. Anterior margin of 2nd tergite convex medially; ante- rolateral sides of 2nd tergite distinctly swollen adja- cent to tymbal. Female abdomen more irregularly spotted all over. Light middorsal band more conspic- uous than in males, often glistering with short, silvery setae. Ovipositor sheaths reaching just beyond apex of caudodorsal beak (fig. 5). Female caudodorsal beak (fig. 20) stout, slightly erect and rounded at apex. Male genitalia: Pygofer slender in lateral view as in fig. 10. Dorsal margin straight, angularly bending into stout, short and slightly erect caudodorsal beak. Distal margin straight angularly bending into margin of beak. Lateral lobes of pygofer strongly curved in- wards, with bluntly rounded lateral protuberances. Ventral margin angularly convex, but concave near base of pygofer. Caudodorsal beak in dorsal view (fig. 13) broad and truncate at apex. Clasper in lateral view (fig. 11) broad at base, with very distinct and angular clasper heel, strongly bent down at half-length, form- ing almost right angled dorsal corner. Dorsal margin of clasper concave between clasper heel and this dor- sal corner. Distinct crest running along % of dorsal margin, and bending outwards around aedeagus to- wards clasper base. Clasper hollow short, broadly rounded at apex. Clasper in dorsal view (fig. 12), wich dorsal margins of clasper bases very close together, leaving only a narrow gap for anal valves. Aedeagus in lateral view (fig. 16) strongly concave around aedea- gus pore, lateral lobes long and slender. Distinct crest, 10 at base of S-curvation, between lateral lobes (figs. 15, 18). Aedeagus pore (fig. 16) large and oval-shaped. Measurements: Body length d: 25.4-29.7 mm (x 27.6 mm + 1.3), 2: 23.5-28.7 mm (x 26.6 mm + 1.4); tegmen length d: 28.0-35.0 mm (x 32.5 mm + 1.8), 2: 32.1-40.5 mm (x 37.7 mm + 2.3); head length d : 2.3-2.6 mm (x 2.4 mm), 9: 2.4-3.1 mm (x 2.6 mm); pronotum length d: 3.1-3.8 mm (x 3.5 mm), 9: 3.5-4.3 mm (x 4.0 mm); mesonotum length d: 5.8-6.6 mm (x 6.3 mm), 2: 6.5-7.8 mm (x 7.1 mm); head width d: 5.5-6.2 mm (x 5.8 mm), 9: 5.9-6.9 mm (x 6.6 mm); width of pronotal collar d: 7.1-8.5 mm (x 8.0 mm), 9: 8.3-9.9 mm (x 9.2 mm). Distribution (fig. 1). — B. hamiltoni is known from a very restricted area in Irian Jaya and from one local- ity in NW Papua. Etymology. — The species is named in honour of the geologist W. B. Hamilton. Remark. — Three females from Araucaria camp that have been described as B. guttulinervis Blöte (allo- and paratypes) are very similar in size and colour pattern to the females of B. hamiltoni, but have darkened patches along the venation of tegmen, are slightly brown speckled on the pronotum and miss the dark middorsal spot at pronotal collar. Baeturia bemmeleni sp. n. (figs. 1, 21-29) Type material. — Holotype d: ‘NEW GUINEA NETH. Bodem, 100 m, 11 km SE of Oerberfaren, July 7- 17. 1959’; ‘T. C. Maa Collector’, BPBM. — Paratypes: IRIAN: NEW GUIN- EA (W): Boven Digul gebied [Upper Digul area], 400 km N. Merauke, Dr A. Kalthofen, legit 1926, 16, ZMA; Boven Sermowai riv., + 400 m, 8.iv.1911, K. G. [K. Gjellerup], 26, RMNH. Other material. — PAPUA: NEW GUINEA (NE): Eliptamin valley, 1200-1500 m, 16-31.vii.1959, W.W. Brandt, 1 9, BPBM; NEW GUINEA (SE): Kiunga, Fly riv., 35 m, viii.1969, J. and M. Sedlacek, 12, BPBM. B. bemmeleni closely resembles B. hamiltoni in shape of male operculum and clasper, but is consider- ably smaller and less intensely coloured. The female from Eliptamin Valley deviates somewhat in shape and colour pattern, but the shape of its 8th apical ar- ea of tegmen suggests that this specimen belongs to the B. loriae group and it presumably belongs to this species. Description Body of males ochraceous or reddish brown, with light brown patches, smaller than in foregoing spe- cies. Distribution of patches resemble markings in 2. hamiltoni; leaving a light middorsal band on head and 24 “25 26 DE Boer: The loriae group of Baeturia Figs. 21-30. 21-29: Baeturia bemmeleni sp. n. — 21, pygofer in lateral view; 22, male operculum; 23, female operculum; 24, male caudodorsal beak in dorsal view; 25, female caudodorsal beak in dorsal view; 26, clasper; 27, aedeagus from behind; 28, aedeagus in lateral view; 29, detail crest between lateral lobes of aedeagus. — 30: Baeturia wegeneri sp. n., clasper. thorax immaculate. This band sometimes interrupted by a vaguely darkened middorsal spot, at margin of pronotal collar. Females greenish brown, specimen from Eliptamin Valley densely brown speckled. Both females show the light brown middorsal marking on pronotum. Male abdomen 1.3-1.6X as long as head and thorax, of females 1.1 x. Male tegmen 1.1-1.2 x as long as total body length, of females 1.3 x. Head: Ochraceous, with brown markings on vertex lobes and along sides of postclypeus, but densely speckled in female of Eliptamin Valley. Postclypeus strongly protruding, almost triangular in dorsal view, 1.8-2.3 X as wide as long. Postclypeus in lateral view, angularly inflated as in B. hamiltoni, though in some specimens more rounded at anterior margin. Head about as wide as anterior part of pronotum. Distance between lateral ocelli 0.7-0.9 as long as distance between eye and lateral ocellus. Thorax: Pronotum light brown, with dark brown spots along oblique fissures and a light, but narrow, middorsal band. Female from Eliptamin Valley brown speckled between medial fissures and showing this band only on proximal half of pronotum. Slightly darkened middorsal spot, at margin of pronotal col- lar, marking end of immaculate band; this spot less clear than in B. hamiltoni. Spot U-shaped in one male from Sermowai riv. and the female from Kiunga riv.; connecting two dark lines, running partly along the immaculate middorsal band. Mesonotum variable in colour pattern between individual specimens, though always with a light, immaculate cruciform elevation and, with exception of the densely speckled female from Eliptamin Valley, with a light middorsal band or narrow triangular area in front of elevation. Two dark spots in front of cruciform elevation. Tegmina and wings: Hyaline, veins slightly red- dened. Tegmina with short 8th apical area, and with narrow hyaline border along hind margin. Wings with slightly broader hyaline border. Tymbal organs: As in B. hamiltoni, not curving in- wards, but with connecting bar between tymbal and abdomen distinctly longer and more curved inwards. Six transverse sclerotized ridges spanning the tymbal, from dorsal to ventral margin. A 7th ridge, close to proximal tymbal margin, almost reaching ventral margin. Six short intercalary ridges seem to form a band across tymbal. Opercula: Male operculum (fig. 22) closely resem- bling that of B. hamiltoni, broadly rounded, shell- shaped, and ochraceous brown with some darkened streaks. Basal part of operculum almost oblong, but weakly tapering mesiad at half its width. Distal part weakly domed, almost flat against body, its rounded apex reaching well beyond margin of abdominal seg- ment 3. Distomedial margin and distal % of lateral margin weakly convex, proximal % of lateral margin more strongly convex. Meracanthus very short, reach- ing to about % the operculum length. Female opercu- 11 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 31-38. Baeturia wegeneri sp. n. — 31, pygofer in lateral view; 32, pygofer from aslant; 33, first sternite; 34, caudodorsal beak in dorsal view; 35, aedeagus from behind; 36, aedeagus in lateral view; 37, detail crest between lateral lobes of aedeagus; 38, operculum. lum (fig. 23) sickle-shaped and erect, with broad and flat area between base of meracanthus and medial edge of operculum. Abdomen: Male abdomen ochraceous brown, stained with dark brown spots, concentrated in near- ly continuous middorsal band. Sternites immaculate. À latero-ventral row of clear dark spots on segments 3-7. Segmental hind margins bright red. Anterior margin of 2nd tergite convex medially; sides of 2nd tergite distinctly swollen and adjacent to tymbal. Female abdomen darker brown with irregular brown markings. Latero-ventral row of dark spots less con- spicuous than in males. Segmental hind margins och- raceous. Female caudodorsal beak (fig. 25) long, slen- der, and pointed at apex. Male genitalia: Pygofer (fig. 21) closely resembling that of B. hamiltoni, slender in lateral view. Dorsal margin straight, angularly bending into straight, stout and slightly erect caudodorsal beak. Distal margin straight, angularly bending into margin of beak. Lateral lobes of pygofer strongly incurved, with well- developed, bluntly rounded lateral protuberances. Ventral margin angularly convex, but concave near base of pygofer. Caudodorsal beak (fig. 24) short, sharply pointed or narrowly truncate at apex. Clasper 12 a in lateral view (fig. 26) almost identical to that of B. hamiltoni, with very distinct angular clasper heel, and rectangular dorsal corner. Dorsal margin of clasper concave between clasper heel and this dorsal corner. Distinct crest running along % of dorsal margin, and bending outwards around aedeagus towards clasper base. Clasper hollow short, broadly rounded at apex. In dorsal view, dorsal margins of clasper bases wider apart than in B. hamiltoni. Aedeagus in lateral view (fig. 28) slender, strongly concave along margin of pore. Lateral lobes long and slender. Distinct crest at base of S-curvation, between lateral lobes (fig. 29). Aedeagus seen from behind (fig. 27) very slender. Aedeagus pore narrow and oval. Measurements: Body length d: 17.1-21.3 mm (x 18.4mm + 1.7), 2: 19.3 and 21.7 mm; tegmen length 3: 19.8-20.4 mm, 9: 27.8 and 28.6 mm; head length 3: 1.5-1.8 mm (x 1.7 mm), 2: 1.9 and 2.2 mm; pro- notum length d: 1.9-2.2 mm (x 2.1 mm), 9: 2.8 and 3.2 mm; mesonotum length d: 3.4-3.8 mm (x 3.7 mm), 9: 5.0 and 5.3 mm; head width 6: 4.2-4.5 mm (x 4.3 mm), 9: 5.0 and 6.0 mm; width of pronotal collar d: 5.2-5.4 mm (x 5.3 mm), 2: 6.9 and 7.2 mm. Distribution (fig. 1). — B. bemmeleni is known from several localities in the northern and central parts of Irian Jaya. Etymology. — The species is named in honour of the geologist R. W. van Bemmelen. Baeturia wegeneri sp. n. (figs. 1, 30-38) Type material. — Holotype d : ‘NEW GUINEA: NE, May R. Petrol sta., 250 m, 3.vi.1963’ [print]; ‘Dry Forest [print]; ‘R. Straatman Collector BISHOP’ [print], BPBM. — Paratype: same locality and collector as holotype but 100 m, 31.v.1963, 1d, BPBM. B. wegeneri is an ochraceous brown coloured spe- cies, with densely brown speckling all over its body, as characteristic for many Baeturia species, and in this respect quite different from most other species of the B. loriae group. However, B. wegeneri is very similar to the two foregoing species in shape of male opercu- lum and genitalia. Description Body ochraceous brown, densely brown speckled all over. Abdomen strongly inflated, 1.7 X as long as head and thorax. Tegmina about as long as body length. Head: Ochraceous, brown speckled. Postclypeus broad and oblong in dorsal view, 1.4-1.6 as broad as long. Postclypeus angularly inflated ventrally (as in fig. 4). Head narrower than anterior part of prono- tum. Distance between lateral ocelli 0.8-1.0X dis- tance between eye and lateral ocellus. Thorax: Pronotum ochraceous, brown speckled in middorsal band and on pronotal collar. Mesonotum grey-brown, densely brown speckled. Two dark spots in front of speckled cruciform elevation, almost lost in speckling. Tegmina and wings: Hyaline, veins ochraceous or slightly reddish and set with short setae. Tegmen with short and squarish 8th apical area, and narrow hyaline border along hind margin. Wing with slightly broad- er hyaline border. Tymbal organs: Not curving inwards, connecting bar between tymbal and abdomen quite short and al- most parallel to body axis. Six transverse sclerotized ridges spanning the tymbal, from dorsal to ventral margin. A 7th ridge, close to proximal tymbal mar- gin, almost reaching ventral margin. Six short interca- lary ridges seem to form a band across tymbal. Operculum (fig. 38): Broad and rounded as in foregoing species, ochraceous with some darkened stains near lateral margin. Basal part of operculum al- most oblong, but weakly tapering towards mesiad at half its width. Distal part slightly domed, almost flat against body, broadly rounded at apex and reaching DE Boer: The loriae group of Baeturia just beyond margin of abdominal segment 3. Distome- dial margin and distal # of lateral margin convex, prox- imal % of lateral margin more strongly convex. Opercula well separated medially, by distinctly smaller and anteriorly lobate first sternite (fig. 33). Meracan- thus short, reaching to about 4 of operculum length. Abdomen: Ochraceous brown, tergites densely brown speckled all over, sternites unspeckled. Latero- ventral row of dark spots almost inconspicuous, by speckling. Segmental hind margins bright red. Anterior margin of 2nd tergite convex middorsally; sides of 2nd tergite weakly swollen and adjacent to tymbal. Genitalia: Pygofer in lateral view (fig. 31) resem- bling that of foregoing species. Dorsal margin straight, angularly bending into stout and slightly erect caudo- dorsal beak. Distal margin straight and more gradual- ly bending into margin of beak. Lateral lobes of pygof- er strongly curved inwards, with well-developed, bluntly rounded, lateral protuberances. Ventral mar- gin straight, not concave to base; ventral margins con- verge to sharp angle at base of pygofer opening (fig. 32). Caudodorsal beak slightly longer than in B. bem- meleni, very slender in dorsal view (fig. 34) and sharp- ly pointed apically. Clasper in lateral view (fig. 30) strongly resembling that of two foregoing species, with a very distinct angular clasper heel and rectangular dorsal corner. Clasper strongly bent down to apex at half-length. Distinct crest running along % of dorsal margin. Clasper hollow short and broadly rounded at apex. Dorsal margins of clasper bases wider apart than in B. hamiltoni. Aedeagus in lateral view (fig. 36) con- cave along margin of pore, with broader basal lobes than in B. bemmeleni. Distinct crest at base of S-curva- tion, between basal lobes (fig. 37). Aedeagus from be- hind (fig. 35) broader than in B. bemmeleni, with a short and broad oval pore. Measurements: Body length: 26.0 and 27.3 mm; tegmen length: 26.4 and 26.6 mm; head length: 2.0 and 2.3 mm; pronotum length: 2.6 and 2.7 mm; mesonotum length: 5.3 mm; head width: 4.7 and 4.8 mm; width of pronotal collar: 6.4 and 6.5 mm. Distribution (fig. 1). — B. wegeneri is only known from May river, in central west Papua New Guinea. Etymology.— The species is named in honour of the geologist A. Wegener. Baeturia silveri sp. n. (figs. 2, 39-48) Type material. — Holotype d : ‘NEW GUINEA: NE Purosa, 20-26 km SE Okapa, 1800-2020 m, 28.viii.1964’ [print]; ‘J. and M. Sedlacek Collectors BISHOP’ [print], BPBM. — Paratypes: same data as holotype 16, 19, BPBM. Other material. — PAPUA: NEW GUINEA (NE): Kandep, 13 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 39 SS 42 2 Figs. 39-48. Baeturia silveri sp. n. — 39, pygofer in lateral view; 40, clasper; 41, male operculum, Kandep; 42, male caudodor- sal beak in dorsal view; 43, female caudodorsal beak in dorsal view; 44, aedeagus from behind; 45, aedeagus in lateral view; 46, detail crest berween lateral lobes of aedeagus; 47, male operculum, holotype; 48, female operculum. Lettering: d = distal margin of operculum; | = lateral margin of operculum; m = medial margin of operculum. Western Highlands, 8000 ft, 23.xii.1961-14.ii.1962, W. W. Brandt, 1d, CSIRO. B. silveri can be recognized by the brown speckled tegmina and short, oblong male operculum. Description Body of males light brown or ochraceous, with a colour pattern of dark brown spots and a narrow, im- maculate, middorsal band as characteristic for most species of the B. loriae group. Female slightly reddish brown with irregularly distributed dark spots. Male abdomen 1.4-1.6 as long as head and thorax, of fe- male 1.1 x. Tegmina of males 1.2-1.4 x as long as to- tal body length, of female 1.5 x. Head: Ochraceous brown, with dark brown streaks on vertex lobes, almost black between eyes and lateral ocelli. Female with black markings between prono- tum and ocelli. Postclypeus sometimes darkened to- wards lateral margins, 1.6-1.9 x as broad as long and broadly rounded at anterior margin. Postclypeus not, or only slightly, swollen ventrally, anterior margin (lateral view) straight or weakly convex. Head nar- rower than anterior part of pronotum. Distance between lateral ocelli 0.9-1.3 x distance between eye and lateral ocellus. Thorax: Pronotum light brown, darkened along fissures; with a lighter, ochraceous, middorsal band, sharply bordered by irregular black streaks. Middorsal band abruptly widening near pronotal collar, and ending at a vaguely darkened middorsal spot. Middorsal band reddish brown and brown speckled in the female, but clearly recognizable by its black lin- ing. Pronotal collar red-brown, brown speckled. Mesonotum with two distinct semi-circular dark spots at pronotum margin and darkened, greenish tinged, broad lateral streaks, converging from prono- tum margin to corners of cruciform elevation. Elevation and narrowly pointed triangular area in front of elevation, light ochraceous. Mesonotum of female, including cruciform elevation, reddish tinged and densely brown speckled. Tegmina and wings: Hyaline, venation reddish ochraceous. Tegmen with irregular brown markings, predominantly along venation. Hyaline border along hind margin of tegmen broader than in foregoing species and almost as broad as in wing. The 8th apical area of tegmen slightly longer, more slender, than in other species of this group. Tymbal organ: Not curving inwards, connecting bar between tymbal and abdomen quite short and al- most parallel to body axis. Five transverse sclerotized ridges spanning the tymbal, from dorsal to ventral margin. A 6th ridge almost reaching ventral margin, and a 7th, most proximal, ridge running from dorsal margin to about half the tymbal width. Six intercalary ridges seem to form a band across tymbal. Traces of red between tymbal ridges. Opercula: Male operculum (fig. 47) quite different from that of the three foregoing species, more resem- bling the opercula found in other groups of Baeturia. Basal part of operculum oblong, equally long at later- al and medial margins. Distal part angularly oblong, wider than long and flat against the body, almost completely covering tymbal cavity. Lateral and medi- al margin almost straight, distal margin weakly con- vex. Distolateral corner broadly rounded, distomedial corner angular. Opercula wide apart medially, separ- ated by broad and rounded Ist sternite. Meracantus almost as long as operculum. Operculum of Kandep specimen (fig.41) much longer and more directed mesiad, with longer lateral margin. Female opercu- lum (fig. 48) sickle-shaped and erect. Distal margin convex. Abdomen: Male abdomen light brown, ochraceous in Kandep specimen. Two parallel dorsal rows of dark spots along a narrow light middorsal band and a very clear latero-ventral row of dark spots on segments 3- 7. Lateral and ventral sides of abdomen slightly brown stained. Abdomen of Kandep specimen with midlateral band of dark spots, as in B. fortuini and B. loriae. Anterior margin of 2nd tergite convex medial- ly; sides of 2nd tergite distinctly swollen and adjacent to tymbal. Female abdomen irregularly brown spot- ted all over, though with unstained light middorsal band. Female caudodorsal beak (fig. 43) pointed at apex. Male genitalia: Pygofer in lateral view rounded, with stout caudodorsal beak (fig. 39). Dorsal margin straight almost continuous with erect caudodorsal beak. Distal margin straight or slightly concave, broadly rounded into margin of beak. Lateral lobes of pygofer curved inwards, with well-developed, bluntly rounded protuberances, that curve slightly upwards towards pygofer margin. Ventral margin angularly convex, but concave near base of pygofer. Caudodorsal beak (fig. 42) rounded, almost truncate at apex. Clasper in lateral view (fig. 40) strongly rounded and quite different from the three foregoing species. Clasper gradually widening distad to clasper base, so that no clasper heel is formed. Dorsal margin of clasper rounded, forming an almost globular, slightly inwards curved, protrusion at about % its length and, distad to this protrusion, convex to apex. Clasper slightly curved outwards, towards dorsal mar- gin and around aedeagus. Clasper hollow narrow tri- angular, narrowly rounded at apex. Aedeagus longer DE Boer: The loriae group of Baeturia than in foregoing species, in lateral view (fig. 45) strongly concave along margin of pore, with long and slender basal lobes. Distinct crest at base of S-curva- tion between basal lobes (fig. 46). Aedeagus pore (fig. 44) large and oval-shaped, but truncate at apex. Measurements: Body length d : 20.0-22.0 mm (x 20.9 mm), 9: 18.8 mm; tegmen length d: 25.9-28.8 mm (x 27.2 mm), 9: 28.6 mm; head length d: 1.5- 1.7 mm, 9: 1.8 mm; pronotum length d: 2.2-2.5 mm, 9: 2.5 mm; mesonotum length d 4.2-4.6 mm, 2: 5.0 mm; head width d: 3.8-4.3 mm, 9 : 4.4 mm; width of pronotal collar d : 5.4-6.2 mm, 9: 6.2 mm. Distribution (fig. 2). — B. silveri is known from on- ly two localities, both in the central mountain ranges of Papua New Guinea. Etymology. — The species is named in honour of the geologist E.A. Silver. Baeturia pigrami sp. n. (figs. 2, 49-59) Type material. — Holotype d: ‘Museum Leiden Neth New Guinea Exp. Star Range, 1260 m, Sibil, 21.viii.1959, op licht’ [print], RMNH. Other material. — PAPUA: NEW GUINEA (NE): Feramin, 150-120 m, 1-6.vi.1959, W.W. Brandt, 29, BPBM. Of this species only one male is available, it resem- bles B. silveri in colour patterns of body and tegmina, and in clasper shape. However, B. pigrami is much larger, has very large triangular opercula, and an an- gularly swollen postclypeus. Two females, with simi- lar colour patterns and from the same area, probably belong to this species. Description Body of male dark reddish brown, with very dis- tinct light middorsal band on head and thorax. This band much broader in females, and continuing over abdomen. Male abdomen strongly inflated and 1.5 x as long as head and thorax of females 1.0. Tegmina of male 1.2 x as long as total body length, of females 1.4-1.5x. Head: Light brown with dark stains on vertex lobes, mainly between lateral ocelli and eyes. Vertex and post- clypeus with long setae, more densely set than in fore- going speces. Postclypeus unstained, light ochraceous in females, slightly brown suffused in male. Male post- clypeus 1.2X as broad as long, in females 1.5 and 2.2 X. Postclypeus angularly swollen, almost conical in lateral view (as in fig. 4). Vertex in male very narrow, with ocelli close together. Head narrower than anterior part of pronotum. Male head 1.2 X as long as distance between the eyes, in female 0.9X. Distance between lateral ocelli in male 0.7 X distance between eye and lateral ocellus, in females 1.1-1.2x. 15 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 56 Figs. 49-58. Baeturia pigrami sp. n. — 49, pygofer in lateral view; 50, clasper; 51, male operculum; 52, aedeagus from behind; 53, aedeagus in lateral view; 54, detail crest between lateral lobes of aedeagus; 55, male caudodorsal beak in dorsal view; 56, female operculum; 57, female caudodorsal beak in dorsal view; 58, tymbal. Thorax: Pronotum dark brown with light ochra- ceous middorsal band, narrow and sharply defined in male, much broader and more diffuse in females. Pronotal collar slightly reddish and, especially in male, stained with brown. Male mesonotum dark brown, but light ochraceous in triangle-shaped area in front of reddish cruciform elevation. Female mesonotum with continuous broad ochra- ceous middorsal band. Two black spots in front of cruciform elevation in male and female. Tegmina and wings (fig. 59): Hyaline, though teg- mina with irregular brown markings, predominantly along veins, venation reddish brown. Eight apical ar- brown ea short and broad, especially in females. Tegmen and wing with very narrow hyaline border along hind margin. lymbal (fig. 58): Five transverse sclerotized ridges spanning the tymbal from dorsal to ventral tymbal margin. À Gth ridge almost reaching ventral tymbal margin, and a 7th, most proximal, ridge running from dorsal margin to only about half the tymbal width. Six intercalary ridges seem to form a band 16 across tymbal. Traces of red between tymbal ridges. Opercula: Male operculum (fig. 51) very large, red- dish ochraceous with dark stains. Basal part of oper- culum with greatest length at % its width, slightly ta- pering towards lateal margin, more strongly tapering to base of meracanthus. Distal part triangle-shaped and slightly domed, narrowing towards narrowly rounded, almost pointed apex and reaching to half- length 3rd abdominal segment. Lateral margin of male operculum convex near base, concave at half- length, and weakly convex towards apex. Distomedial margin concave near apex and convex to short and straight medial margin. Opercula close together me- dially, separated by narrow and weakly elevated 1st sternite. Meracanthus very short, reaching to about % the operculum length. Female operculum (fig. 56) sickle-shaped, slightly angular at distal margin, and erect. Base of operculum forming narrow rim around base of meracanthus. Abdomen: Male abdomen brown, dark brown stained dorsally and on sides of 7th and 8th segments, though lighter and reddish tinged, in very narrow DE BOER: The loriae group of Baeturia Fig. 59. Baeturia pigrami n. sp., female tegmen and wing. middorsal band. A latero-ventral row of slightly dark- ened patches, hardly conspicuous. Segmental hind margins slightly red. Anterior margin of 2nd tergite convex middorsally. Lateral parts of 2nd tergite, with auditory capsules, and anterolateral part of 3rd ter- gite, weakly incurving towards ventral corner of tym- bal. Tergites 4-7 with sharp latero-ventral fold. Female abdomen ochraceous, densely stained with ir- regular brown spots laterally and ventrally, but with broad and immaculate middorsal band. Female cau- dodorsal beak very characteristic in dorsal view (fig. 57), with greatest width at about % its length and sharply pointed at apex. Male genitalia: Pygofer in lateral view (fig. 49). Dorsal margin of pygofer bending gradually into erect caudodorsal beak. Distal margin straight, angularly bending into margin of beak. Lateral lobes of pygofer strongly curving inwards, with well-developed, blunt- ly rounded protuberances. Ventral margin angularly convex, but strongly concave to base of pygofer. Caudodorsal beak in dorsal view (fig. 55) very slender, narrowly rounded, almost pointed at apex. Clasper (fig. 50) closely resembling that of B. silveri, not forming a clasper heel and slightly bent down to apex. Dorsal margin of clasper strongly bent upwards from clasper base, forming a long finger-shaped, slightly in- curved protuberance. This protuberance distinctly longer than in B. silveri. Clasper strongly bent out- wards distally of this protuberance, and forming a broadly rounded dorsal crest. Clasper hollow longer than in B. silveri. Aedeagus in lateral view (fig. 53) slightly concave along margin of pore, with narrow and elongate lateral lobes. Aedeagus with distinct crest at base of S-curvation, between lateral lobes (fig. 54). Aedeagus pore broad and short, oval-shaped (fig. 52). Measurements: Body length d: 27.0 mm, 9: 20.9 and 22.4 mm; tegmen length d : 32.0 mm, 9:31.3 and 31.8 mm; head length d: 2.6 mm, 9: 2.1 and 2.3 mm; pronotum length d: 2.9 mm, 9: 3.5 and 3.6 mm; me- sonotum length d: 6.1 mm, 2: 6.1 and 6.6 mm; head width d': 5.4 mm, 9: 5.4 and 5.7 mm; width of prono- tal collar d : 7.7 mm, 9: 7.8 and 8.4 mm. Distribution (fig. 2). — The only male known of this species comes from Sibil, Star Range, in the cen- tral mountain ranges of New Guinea. Two females, that probably belong to this species, come from near- by Feramin. Etymology. — The species is named in honour of the geologist C. J. Pigram. Baeturia daviesi sp. n. (figs. 2, 60-66) Type material. — Holotype d : ‘Papua, Mt Lamington, G. H. Muray, 1924’ (written); ‘W. W. Froggatt Collection’ (print); ‘CANB' (print), CSIRO. Of this species only one male is available. B. davie- si is very similar to B. pigrami in the shapes of the male operculum and genitalia, but B. davies: is distin- guished by a distinctly smaller size, immaculate teg- mina and the absence of distinct colour markings, characteristic to B. pigrami and most other species of 17 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 > ie 60 / 61 \ 64 Figs. 60-67. — 60-66, Baeturia daviesi n. sp., 60, pygofer in lateral view; 61, pygofer from aslant; 62, operculum; 63, caudo- dorsal beak in dorsal view; 64, clasper; 65, aedeagus from behind; 66, aedeagus in lateral view. — 67, Baeturia hartonoi sp. n., fore femur. the B. loriae group, though its abdomen is densely brown speckled, as in B. wegeneri. Description Body dull grey-brown with vaguely lighter colou- red middorsal band, slightly more distinct on prono- tum. Abdomen 1.4X as long as head and thorax. Tegmina 1.1 X as long as body length. Head: Greyish brown, with vaguely darkened spots between eyes and lateral ocelli. Postclypeus triangu- larly protruding, 1.5 X as wide as long. Postclypeus angularly swollen ventrally, almost conically protrud- ing in lateral view (as in fig. 4) Head narrower than anterior part of pronotum. Vertex very narrow, with ocelli close together; distance between lateral ocelli about as wide as frontal ocellus and 0.8 X the distance between eye and lateral ocellus. Thorax: Pronotum greyish brown with vaguely lighter coloured, middorsal band, most conspicuous over proximal half. Mesonotum brown with a green- ish tinge, without special colour markings, apart from two dark spots in front of slightly reddish cruciform elevation. Legs: Proximal spine of fore femur rather short, about as long as distance to middle spine. 18 Tegmina and wings: Hyaline and immaculate, veins ochraceous. Tegmina with fairly short, square, 8th apical area, and very narrow hyaline border along hind margins. This border is slightly broader in wing. Tymbal organ: Tymbal not curved inwards to- wards distal margin; connecting bar between tymbal and abdomen almost parallel to body axis. Six trans- verse sclerotized ridges spanning the tymbal from dorsal to ventral tymbal margin. Most proximal ridge narrowing considerably towards the ventral tymbal margin and only just reaching it. Six short intercalary ridges seem to form a band across the tymbal. Traces of bright red between the ridges, especially on dorsal half of tymbal. Operculum (fig. 62): Very similar to that of B. pi- grami, but smaller, with dark stains near apex. Basal part of operculum somewhat quintangular, with its greatest length at base of lateral margin of distal part, weakly tapering to distolateral corner and more strongly tapering to base of meracanthus. Distal part large, triangle-shaped and slightly domed, curved to abdomen near its almost pointed apex, and reaching to about half-lenght abdominal segment 3. Proximal % of lateral margin slightly convex, distal 4 almost straight to apex. Distomedial margin concave near DE Boer: The loriae group of Baeturia Figs. 68-75. Baeturia hartonoi sp. n. — 68, pygofer in lateral view; 69, caudodorsal beak in dorsal view; 70, clasper; 71, post- clypeus in lateral view; 72, operculum; 73, aedeagus in lateral view; 74, aedeagus from behind; 75, detail crest between later- al lobes of aedeagus. apex and convex to almost rectangular medial corner. Medial margin short and straight. Opercula close to- gether medially, separated by narrow and distally lo- bate Ist sternite. Meracanthus very short, reaching to about % the operculum length. Abdomen: Light brown with brown speckling, darker brown dorsally and lateroventrally, with a slightly lighter coloured narrow middorsal line, though without a distinct row of latero-ventral spots. Segmental hind margins bright red. First tergite only slightly shorter than 2nd. Anterior margin of 2nd ter- gite weakly convex middorsally. Ventrolateral part of 2nd tergite, auditory capsules, and anterolateral cor- ners of 3rd tergite, curved inwards towards ventral corner of tymbal. Anterolateral part of 2nd tergite ad- jacent to tymbal. Genitalia: Pygofer in lateral view (fig. 60) slender. Dorsal margin straight, continuous with straight, slender and erect caudodorsal beak. Distal margin of pygofer straight, angularly bending into margin of beak. Lateral lobes of pygofer bent inwards, with small, bluntly rounded protuberances. Caudodorsal beak very short, pointed at apex (fig. 63). Ventral margin convex, but weakly concave to base of pygof- er; ventral margins converging to sharp angle at base of pygofer opening (fig. 61). Clasper (fig. 64) much resembling that of B. pigrami, without clasper heel, but with dorsal margin strongly bent upwards from clasper base, forming a long finger-shaped, slightly in- curved protuberance. Dorsal margin strongly bent outwards distad to protuberance, forming a broad and distinctly laterally protruding crest. Clasper slightly bending down to rounded apex. Apical part of clasper with small clasper hollow. Aedeagus in lat- eral view (fig. 66) slightly concave along margin of pore, with narrow and elongate lateral lobes. Aedeagus with distinct crest at base of S-curvation, between lateral lobes. Aedeagus pore broad and short, oval-shaped (fig. 65). Measurements: Body length: 23.9 mm; tegmen length: 25.6 mm; head length: 2.0 mm; pronotum length: 2.8 mm; mesonotum length: 5.1 mm; head width: 4.8 mm; width of pronotal collar: 6.5 mm. Distribution (fig. 2). — The only specimen available comes from Mt. Lamington on the Papuan Peninsula. Etymology. — The species is named in honour of the geologist H.L. Davies. Baeturia hartonoi sp. n. (figs. 1, 67-75) Type material. — Holotype d : ‘PNG: NEW GUINEA: SE: S Highlands Distr.: Kutubu: Tugiri, 1000 m, 7-9.ii.1978° (print); J. L. Gressitt Collector BISHOP Museum’ (print), BPBM. Of this species only one male specimen is available. B. hartonoi is a small species, that can be recognized by an elongate apical part of aedeagus and a lateral protuberance on the clasper. 19 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Description Body ochraceous brown, head and thorax slightly reddish. Abdomen with conspicuous dark dorsal and midlateral bands, and slightly inflated. Abdomen 1.3x as long as head and thorax. Tegmina 1.3X as long as body length. Head: Brown, slightly darkened around ocelli, but not as distinctly marked as in some other species of the 8. loriae group. Postclypeus distinctly protruding beyond vertex lobes, broadly rounded at anterior margin, and 1.8 x as broad as long. Postclypeus only very slightly swollen ventrally, anterior margin (later- al view) weakly convex (fig. 71). Head narrower than anterior part of pronotum. Distance between lateral ocelli about 1.5X the width of frontal ocellus and 1.2 X che distance between eye and lateral ocellus. Thorax: Pronotum, plain brown with no special markings. Mesonotum almost plain grey-brown, cru- ciform elevation slightly lighter ochraceous. Two vaguely darkened spots in front of elevation hardly visible. Legs: Proximal spine of fore femur strongly bent, almost adjacent to femur, and shorter than distance to middle spine (fig. 67). Tegmina and wings: Hyaline and immaculate, ve- nation ochraceous. Tegmen with narrow hyaline bor- der along hind margin, wing with fairly broad border. Tymbal organ: Tymbal not curved inwards to- wards distal margin; connecting bar between tymbal and abdomen almost parallel to body axis. Six sclero- tized transverse ridges spanning the tymbal from dor- sal to ventral tymbal margin. À 7th, most proximal, ridge almost reaching ventral tymbal margin. Seven short intercalary ridges seem to form a midlateral band across tymbal. Operculum (fig. 72): Short and oblong shaped as in B. silveri. Basal part of operculum oblong; equally long at lateral and medial margins. Distal part broad- er than long, flat against the body. Operculum direct- ed mesiad, so that connecting bar between tymbal and abdomen becomes partly visible in ventral view. Lateral margin very short, bending gradually into slightly convex distal margin. Distomedial corner an- gular. Medial margin almost straight. Medial part of operculum reaching beyond margin of abdominal segment 2. Opercula widely separated medially, by broad and rounded Ist sternite. Meracanthus reach- ing to about 3/4 the operculum length. Abdomen: Ochraceous, greenish tinged, darkened in dorsal and lateral bands, though less intensely in narrow middorsal line. Latero-ventral row of dark spots most clear on segments 3-5 on right side, and on segments 3-4 on left side of body, but faded in successive segments. First tergite very short medially. Anterior margin of 2nd tergite weakly convex medial- ly. Lateral parts of 2nd tergite not incurved. 20 Genitalia: Pygofer in lateral view (fig. 68). Dorsal margin straight and angularly bending into slender and erect caudodorsal beak. Distal margin broadly rounded into margin of beak. Lateral lobes of pygofer strongly bent inwards, with small and bluntly round- ed protuberances. Ventral margin weakly convex, but distinctly concave towards base of pygofer. Caudo- dorsal beak in dorsal view (fig. 69) slender and point- ed at apex. Clasper very broad, squarely rounded in lateral view (fig. 70), abruptly broadening dorsally, distally of clasper base, so that no clasper heel is formed. Dorsal margin angularly bending down at about half-lenght, forming a small and slightly in- curving dorsal protrusion on clasper. This protrusion seems to function as a support or lock for the aedea- gus, since it curves inwards, around aedeagus. This in contrast with the dorsal protrusions on the claspers of the three foregoing species, which lie proximally of the aedeagus. Dorsal margin of clasper bending into almost straight distal margin, at rounded, slightly in- wards curved, distodorsal corner. Claspers slightly di- verging towards pointed apices. Apical part of clasper directed downwards, with narrow clasper hollow. Lateral side of clasper forming a small and angular, la- miniform lateral protrusion, unique for this species. Aedeagus elongate in apical part of its S-curvation (fig. 71), with narrowly elongate lateral lobes and truncate apex. Aedeagus with distinct crest between lateral lobes (fig. 75), crest very low mid-between ba- sal lobes, and strongly concave (seen from behind (fig. 74). Aedeagus pore narrow and sharply incised, truncate at aedeagus apex. Measurements: Body length: 19.0 mm; tegmen length: 24.5 mm; head length: 1.8 mm; pronotum length: 2.5 mm; mesonotum length: 4.3 mm; head width: 4.3 mm; width of pronotal collar: 5.8 mm. Distribution (fig. 1). — The only specimen known comes from Lake Kutubu, in central Papua New Guinea, just south of the central mountain ranges. Etymology. — The species is named in honour of the geologist H.M.S. Hartono. Baeturia fortuini sp. n. (figs. 2, 76-87) Type material. — Holotype d: ‘Coll RI. Sc. N.B. Papua New Guinea Morobe pr.: Anguaia, 1800 m, 21.v.1988 (st. 050) J. van Stalle, LG. no.: 27363’ (print, orange label), KBIN. — Paratypes: PAPUA: NEW GUINEA (NE): Bulldog rd., 60 km S Wau, 2070 m, 22-31.v.1964, J. Sedlacek, 16, BPBM; Dowalo, W Zenag, 2000 m, 4.11.1971, J.L. Gressitt, 29, BPBM; Vagau, Herzog Mts., 4000 ft, 4-17.1.1965, M.E. Bacchus, 1, BMNH; U. Watut SW, 1500 m, 3.v.1968, J.L. Gressitt, 1d, BPBM; same data but 1100-1600 m, 30.iv.1968, 1 4, BPBM. Other material. — Mt Missim, 2100 m, 15.iii.1968, P. 79 À 80 DE BOER: The loriae group of Baeturia Figs. 76-83. Baeturia fortuini sp. n. — 76, pygofer in lateral view; 77, male operculum; 78, first sternite; 79, clasper, Mt. Missim; 80, idem, holotype; 81, aedeagus from aslant; 82, aedeagus in lateral view; 83, aedeagus from behind. Colman, 1d, 12, BPBM; Wau, ix,1965, J. Sedlacek, 19, BPBM; Wau, Morobe Dist, 1200-1300 m, 14-17.1.1963, J. Sedlacek, 1 2, BPBM; Wau, Morobe Dist., Mt. Missim, 2000 m, 1.v.1966, O. R. Wilkes, 13, BPBM. The most striking character of this species is the row of bronzed triangular spots along the hind mar- gins of the tegmina. This characer is also found in B. loriae, described next. B. fortuini is very similar to B. loriae, but can be separated from that species by its larger and triangle-shaped male operculum. Description Body of males light brown or greenish tinged, with dark markings, forming dorsal and lateral bands on abdomen, and with narrow and light middorsal band over whole length of body. Females provided with ir- regular dark patches, and a narrow and light middor- sal band. Male abdomen 1.4-1.7X as long as head and thorax, of females 0.9-1.0x. Tegmina of males 1.2-1.4x as long as total body length, of females 1.4- Wel he Head: Greenish or brown. Vertex and postclypeus with very long setae, as in B. pigrami. Dark spots on vertex lobes, between eyes and ocelli, and on lateral parts of postclypeus. Postclypeus triangularly pro- truding and 1.4-2.2X as wide as long. Postclypeus angularly swollen ventrally (as in fig. 4). Head nar- rower than anterior part of pronotum. Distance between lateral ocelli 1.1-1.4X as long as distance between eye and lateral ocellus. Thorax: Greenish, with light brown streaks, partly along oblique fissures, and with two pairs of parame- dian black spots, at both ends of a broad, greenish tinged, and immaculate middorsal band. Long setae, predominantly on these black areas and on pronotal collar. Colour pattern on mesonotum very variable. Cruciform elevation and small triangle-shaped area in front of elevation light brown or bright green, lateral parts of mesonotum either completely, or only the distal half, dark brown to castaneous. Legs: Ochraceous, fore femora with dark brown streaks. Proximal spine of fore femur shorter than dis- tance to middle spine. TIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Fig. 84. Baeturia fortuini sp. n., male tegmen and wing. Tegmina and wings (fig. 84): Hyaline, though teg- men with a regular row of seven triangular dark spots, close to hind margin, with one spot in each apical ar- ea from apical area 2 to 8. The 8th apical area of teg- men generally broad and almost square-shaped. Venation red. Tegmina and wings with fairly broad hyaline border along hind margins. Tymbal organ: Tymbal somewhat incurved ven- trally, connecting bar between abdomen and tymbal directed inwards. Six transverse sclerotized ridges spanning the tymbal, from dorsal to ventral tymbal margin. Most proximal ridge narrowing considerably towards ventral margin and only just reaching it. Some specimens with a 7th ridge, hardly separated from proximal tymbal margin, running from dorsal margin to about half-width of tymbal. Six short inter- calary ridges seem to form a midlateral band across tymbal. Opercula: Male operculum (fig. 77) ochraceous with longitudinal dark streaks on distal part. Basal part of operculum almost oblong, slightly elongate at half-width and slightly tapering towards medial mar- gin. Distal part very large, triangular, and slightly domed, narrowing towards narrowly rounded and al- most pointed apex, and reaching to about half-length of abdominal segment 3. Lateral margin of opercu- lum convex near operculum base, concave at half- length and convex again near apex. Operculum strongly curved to body along distomedial margin. Distomedial margin weakly convex. Medial margin short and straight. Opercula close together medially, separated by globularly protruding Ist sternite (fig. 78). Meracanthus very short, reaching to about % the operculum length. Female operculum (fig. 85) sickle- shaped and erect, with weak crest along its distal mar- gin. Basal part of female operculum forming a narrow rim around base of meracanthus. Abdomen: Male abdomen light brown or greenish. Dark brown markings forming two dorsal bands, sep- arated by a narrow and lighter coloured middorsal line, and lateral bands over abdomen. These bands are interrupted by lighter, often reddish coloured, segmental hind margins. Latero-ventral row of dark spots on segments 3-7 very clear. Ventral side of male abdomen immaculate, with bright red segmental hind margins. First tergite quite long and weakly curved. Anterior margin of 2nd tergite medially straight. Lateral parts of 2nd tergite hardly inflated, forming a weak crest along anterolateral margin, au- ditory capsules somewhat removed from ventral cor- ner of tymbal. Female abdomen more intensely and irregularly dark stained. Dark lateral bands less con- spicuous than in males. Light middorsal line often very clear, broader than in males. Segmental hind margins ochraceous or reddened. Ventral side of fe- male abdomen dark brown stained. Female caudo- dorsal beak (fig. 86) light ochraceous, long and slen- der, pointed at apex. Male genitalia: Pygofer in lateral view (fig. 76) rather stout, with very long and slender caudodorsal beak. Dorsal margin concave, gradually bending into convexly bent beak. Distal margin straight, almost forming right angle with margin of beak. Lateral lobes of pygofer strongly curving inwards towards dis- tal margin, with well-developed, angularly rounded, protuberances. Ventral margin weakly convex, dis- tinctly concave near base of pygofer. Caudodorsal beak in dorsal view (fig. 87) long and slender, point- ed at apex. Clasper in lateral view (fig. 80) square- shaped as in B. hartonoi, dorsally abruptly broaden- ing, distally of clasper base, so that no clasper heel is formed. Dorsal part of clasper forming a broad and la- miniform protrusion, broadly rounded at its proximal corner, and rectangular at its distal corner. Claspers very narrow and directed downwards towards weakly diverging apices, and with a narrow and rounded clasp- er hollow. Both specimens from Mt. Missim with quite different clasper (fig. 79), more resembling those of B. silveri and B. loriae, with broad and laminiform, but less angular, dorsal protrusion and forming a very small, slightly inwards curved, protrusion at its rectan- gular distal corner. Aedeagus in lateral view (fig. 82) very long and slender, with extremely long and slender lateral lobes, and a very distinct ridge between these lobes (fig. 81). Aedeagus pore slightly larger than in foregoing species. Margin of aedeagus pore distinctly concave at aedeagus apex (fig. 83). Measurements: Body length d: 22.3-24.5 mm (x 23.4 mm +0.9), 9: 21.6-22.8 mm (x 22.2 mm + 0.4); tegmen length d: 27.0-31.3 mm (x 29.8 mm + 1.5), 2: 31.4-36.6 mm (x 34.6 mm + 1.7); head length d: 1.7-2.0 mm (x 1.9 mm), 2: 2.1-2.3 mm (x 2.2 mm); pronotum length d: 2.5-3.0 mm (x 2.8 mm), ?: 3.2- 3.5 mm (x 3.4 mm); mesonotum length d: 4.5-5.4 mm (x 4.9 mm), 2: 5.7-6.4 mm (x 6.1 mm); head width d: 4.6-5.2 mm (x 5.0 mm), 9: 5.5-5.9 mm (x 5.7 mm); width of pronotal collar d: 6.2-7.4 mm (x 6.8 mm), 2: 7.6-8.2 mm (x 7.9 mm). Distribution (fig. 2). — B. fortuini seems restricted to a small area in Papua New Guinea, just south and west of the Huon Gulf. Etymology. — The species is named in honour of the geologist A. R. Fortuin. Baeturia loriae Distant, 1897 stat. n. (figs. 2, 88-93) Baeturia loriae Distant, 1897: 382. Gymnotympana loriae, Distant 1906: 158; Metcalf 1963: 193: Identification of types: In his description Distant (1897) indicates that the type series of this species contains at least one male and one female, originating from Moroka and Paumomu river, and collected by Loria. Two males from Moroka could be traced: one with a type label was found in the BMNH, the other, with identification label: ‘ Baeturia loriae Dist’, was located in the MSNG. Females, or material from Paumomu river, that could possibly belong to the type series, were not found. The specimen from the BMNH is hereby designated lectotype, the other speci- men belongs to a different species, possibly Baeturia DE Boer: The loriae group of Baeturia bicolorata Distant. The lectotype bears the following labels: ‘Type’ (round label, red margin); ‘loriae Dist (written); ‘New Guinea SE Moroka, 1300 m, Loria, vii-xi.93’ (print); ‘Distant coll. 1911. 383’ (print). Classification. — The reasons for the re-allocation of this species in the genus Baeturia are given in the paragraph on phylogeny of the B. loriae group Only one male of this species is known. B. loriae close- ly resembles B. fortuini in body markings, and shares the triangular dark spots in apical areas of tegmina with that species. B. loriae can be separated from B. fortuini by its flat and broader, almost rectangular, operculum. Description Head and thorax brown, abdomen olive green. Dark, almost black, markings forming regular dorsal and lateral bands on abdomen. A narrow and imma- culate middorsal band over whole length of body. Abdomen 1.5 as long as head and thorax. Tegmen 1.2 x as long as total body length. Head: Grey-brown, with dark brown spots on vertex lobes and lateral parts of postclypeus. Postclypeus broadly protruding, 1.8 as wide as long, anterior margin broadly rounded. Postclypeus distinctly swol- len and ventrally conically protruding, anterior margin (lateral view) broadly rounded. Head narrower than anterior part of pronotum. Distance between lateral ocelli 1. 4 distance between eye and lateral ocellus. Thorax: Pronotum ochraceous, with brown streaks in and along oblique fissures and two pairs of dark brown spots at both ends of immaculate and lighter coloured middorsal band. Mesonotum with two large and triangle-shaped dark spots, in front of cruciform elevation, and two brown straeks laterad of elevation. Tegmina and wings: Hyaline, veins ochraceous. Tegmina, as in B. fortuini. with a regular row of sev- en triangular dark spots, close to hind margin, one spot in each apical area from area 2 to 8. The 8th ap- ical area of tegmen short, almost square-shaped. Tegmina and wings with a fairly broad hyaline border along hind margin. Tymbal organ: Six transverse sclerotized ridges spanning the tymbal, from dorsal to ventral margin. Most proximal ridge narrowing considerably towards ventral tymbal margin and only just reaching it. A 7th ridge, hardly separated from proximal tymbal margin, running from dorsal margin to about half width of tymbal. Six short intercalary ridges seem to form a midlateral band across tymbal. Operculum (fig. 91): Ochraceous. Basal part of op- erculum oblong. Distal part very large and angular, flat against body. Lateral margin straight, bending into convex medial margin, at almost rectangular dis- tal corner. Medial margin bending angularly back to base of meracanthus, at medial corner. Meracanthus 23 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 85-93. 85-87, Baeturia fortuini sp. n., 85, female operculum; 86, female caudodorsal beak in dorsal view; 87, male cau- dodorsal beak in dorsal view. — 88-93, Baeturia loriae, 88, pygofer in lateral view; 89, clasper; 90, caudodorsal beak in dorsal view; 91, operculum; 92, aedeagus in lateral view; 93, aedeagus from behind. very short, reaching to about % the operculum length. Abdomen: Greenish brown. Dark brown markings forming two dorsal bands, separated by a lighter nar- row middorsal line, and lateral bands over abdomen. These bands are interrupted by lighter, often reddish coloured, segmental hind margins. Latero-ventral row of dark spots on segments 3-7 very clear. Ventral side of male abdomen immaculate, with bright red segmental hind margins. Anterior margin of 2nd ter- gite straight medially. Lateral parts of 2nd tergite hardly inflated, forming a weak crest along anterolat- eral margin, auditory capsules somewhat removed from ventral corner of tymbal. Genitalia: Pygofer in lateral view (fig. 88) slender, with long and slender, slightly bent, caudodorsal beak. Dorsal margin of pygofer straight, angularly bending into beak. Distal margin straight and broad- ly rounded into margin of beak. Lateral lobes of py- gofer strongly bent inwards, with bluntly rounded protuberances. Ventral margin weakly convex, dis- tinctly concave near base of pygofer. Margins of cau- dodorsal beak in dorsal view (fig. 90) converging from base, but parallel in apical part of beak. Caudodorsal beak rounded at apex. Clasper in lateral view (fig. 89) angular, square-shaped, as in the Mt. Missim specimens of B. fortuini. Dorsal margin of clasper very slighdy bending upwards, distally of bh EN clasper base, so that no clasper heel is formed. Straight dorsal margin bending almost rectangularly into straight distal margin. Distal margin concavely incurved, just before reaching downwards directed apical part of clasper. Clasper hollow small and nar- rowly rounded at clasper apex. Aedeagus in lateral view (fig. 92) very long and slender, with extremely long and slender lateral lobes, and a very distinct ridge between lobes. Aedeagus pore oval (fig. 93). Measurements: Body length: 24.1 mm; tegmen length: 30.0 mm; head length: 1.9 mm; pronotum length: 2.9 mm; mesonotum length: 5.0 mm; head width: 5.2 mm; width of pronotal collar: 6.9 mm. Distribution (fig. 2). — The only specimen available comes from Moroka on the Papuan Peninsula. Baeturia tenuispina Blöte, 1960 (figs. 1, 94-102) Baeturia tenuispina Blöte, 1960: 72, figs. 22-23. — Duffels & Van der Laan 1985: 254. Material examined. — PAPUA: NEW GUINEA (NE): Gewak, Salawaket Range, 1530 m, 6.ix.1956, E.J. Ford Jr., 16, BPBM; Gurakor, 7.vii.1965, H. Pyka, 1d, 72, SMN; Kokoda, 1200 ft, viii. 1933, L.E. Cheesman, 14 holotype Baeturia tenuispi- na Blöte, 34, BMNH; same data, 16 paratype, RMNH; same DE Boer: The loriae group of Baeturia Figs. 94-102. Baeturia tenuispina Blöte. — 94, pygofer in lateral view; 95, pygofer from aslant; 96, male operculum; 97, clasp- er; 98, male caudodorsal beak in dorsal view; 99, aedeagus from behind; 100, aedeagus in lateral view; 101, female opercu- lum; 102, female caudodorsal beak in dorsal view. data but vi-x.1933, 96, 652, BMNH; Kokoda, xi.1957, 29, BMNH; Kokoda, 400 m, 15-20.x1.1965, J. and M. Sedlacek, 16, BPBM; Kokoda-Pitoki, 400 m, 25.11.1956, J.L. Gressitt, 14, BPBM; Sangeman Village nr Busu R., NE of Lae, 25 m, 30.viii.1957, D. Elmo Hardy, 34, BPBM; Wau, Morobe Dist., 1200 m, 29-30.ix.1963, J. Sedlacek, 13, 1 2, BPBM; PA- PUA: NEW GUINEA (SE): Bori nr Sasambota, Popondetta Subdist., 31.x.1963, D.K. McAlpine, 1d, AMS; Mt Lamington Dist, Northern Division, 1925, C.T. McNamara, 12, AMS; same data but vii.1927, 2 9; i-11.1929, 23, 19, all AMS; Popondetta, 25 m, v.1966, Shanahan- Lippert, 16, BPBM; same data but vi.1966, 2d, BPBM; D’ENTRECASTAUX ISLANDS: GOODENOUGH: Goodenough Id., x.1943, F/D, C. Ralph, 16, MvM. Males of B. tenuispina are easily recognized by the long spine-shaped caudodorsal beak and a small fin- ger-shaped protrusion on the clasper heel. This pro- trusion forms the only indication that the species might be related to the B. loriae group. Description Body ochraceous to castaneous brown, covered with brown speckling. Females on average shorter than males, but with equally large head and thorax. Abdomen of males 1.3-1.5 X as long as head and tho- rax, Of females 1.0-1.2 x. Tegmen of males 1.1-1.2 x as long as body length, of females 1.3-1.4 x. Head: Ochraceous, dark brown or blackish speck- led. Postclypeus angularly protruding, 1.7-2.3 as wide as long, anterior margin convex. Postclypeus slightly swollen ventrally, anterior margin (lateral view) weakly convex. Thorax: Pronotum ochraceous brown speckled medially, between medial pair of oblique fissures. Mesonotum greyish brown, densely brown speckled but without any distinct black spots in front of cruci- form elevation. Tegmina and wings: Hyaline, venation ochraceous or reddish, often quite densely set with short setae. Tegmen with 8 apical areas, the 8th apical area long- er than in most species of the B. loriae group. No sex- ual dimorphism in wings. Tegmen with narrow hya- line border along hind margin, this border distinctly broader in wing. Tymbal organs: Tymbal not curved inwards to- wards distal margin; connecting bar between tymbal and abdomen almost parallel to body axis. Seven scle- rotized transverse parallel ridges spanning the tymbal from dorsal to ventral margin and an 8th, most prox- imal ridge nearly reaching ventral margin. Seven short intercalary ridges seem to form a midlateral band across tymbal. Tymbal reddish coloured at dor- sal margin and dorsally between ridges. Opercula: Male operculum (fig. 96) quite large and almost completely covering tymbal cavity in ventral view. Distal part of male operculum broad, angularly rounded, oblong and curved towards body. Lateral 25 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 margin short, convexly bending into long weakly convex distal margin. Distomedial and medial mar- gins weakly convex. Medial corners rounded. Meracan- thus reaching to about % operculum length. Female op- erculum (fig. 101) with very short, sickle-shaped and erect distal part. Medial margin short and straight Abdomen: Male abdomen light brown or ochra- ceous, densely brown speckled dorsally; ventral part, and sometimes lateral parts of segments 4-5, unspeck- led. Latero-ventral row of darkened spots generally very distinct on segments 3-4, but often only vaguely visible on segments 5-7. Segmental hind margins bright red. First tergite quite short. Anterior margin of 2nd tergite medially convex. Lateral parts of 2nd tergite distinctly swollen adjacent to tymbal. Female abdomen greyish brown and densely speckled, latero- ventral row of spots less distinct. Segmental hind margins often ochraceous. Ovipositor sheaths just reaching apex of caudodorsal beak. Female caudodor- sal beak in dorsal view (fig. 102) very long and slen- der, triangle-shaped and sharply pointed at apex. Male genitalia: Pygofer (figs. 94-95) very character- istic, with broadly rounded lateral lobes and long, slender and erect caudodorsal beak. Dorsal margin concave to base, but convexly bent into caudodorsal beak. Distal margin angularly convex, forming a nar- rowly rounded angle with straight margin of beak. Ventral margin convex. Lateral lobe of pygofer weak- ly curving inwards towards end of distal margin, forming a weakly swollen elongate and bluntly rounded protuberance. Caudodorsal beak in dorsal view (fig. 98) extremely long and slender, its parallel margins converge at about half-length of beak, to- wards pointed apex. Claspers conspicuous by slender finger-shaped protuberance on corner of clasper heel (fig. 97). Dorsal crest very stout and strongly outcurv- ing, abruptly ending at distolateral corner of clasper. Apical part of clasper slightly curved down and with small clasper hollow. Aedeagus (fig. 100) very small with narrow lateral lobes, in lateral view weakly con- cave along its pore. Aedeagus pore oval (fig. 99). Measurements: Body length d: 18.3-23.3 mm (x 20.3 mm + 1.4), 2: 16.7-20.0 mm (x 18.5 mm + 1.2): tegmen length d:: 20.0-24.9 mm (x 22.7 mm + 1.2), 2: 23.6-26.5 mm (x 24.9 mm + 1.2); head length d: 1.6-2.0 mm (x 1.8 mm), 9: 1.6-1.9 mm (x 1.8 mm); pronotum length d: 2.3-3.2 mm (x 2.7 mm), 9: 2.7- 3.0 mm (x 2.9 mm); mesonotum length d: 4.2-5.4 mm (x 4.6 mm), 2: 4.2-4.9 mm (x 4.6 mm); head width d: 4.4-5.2 mm (x 4.8 mm), 2: 4.5-5.0 mm (x 4.8 mm); width of pronotal collar d: 5.5-7.1 mm (x 6.3 mm), 2: 6.2-6.9 mm (x 6.6 mm). Distribution (fig. 1). — B. tenuispina is distributed in the eastern parts of Papua New Guinea and is also recorded from Goodenough Island of the D’Entre- casteaux islands. 26 ACKNOWLEDGEMENTS For the loan of material I am indebted to: Mr B. J. Day and Mr M. S. Moulds (AMS); Dr W. J. Knight and Mr M. D. Webb (BMNH); Mr G. M. Nishida and Mr K. Arakaki (BPBM); Dr B.P. Moore (CsIRO); Mr J. van Stalle (KBIN); Dr V. Raineri (MSNG); Dr A. Neboiss (MVM); Mr J. van Tol (RMNH); and to Mr F. Heller (SMN). I would like to thank Mr G. Verlaan for technical assistance and and Mr D.A. Langerak for the drawing of figures 8, 9, 59, and 84 and the maps (figs. 1 and 2). I am indebted to Prof. Dr F.R. Schram and Dr J. P. Duffels (Institute for Systematics and Population Biology, Amsterdam) for their critical reading and comments on the manuscript. REFERENCES Blöte, H. C., 1960. The genus Baeturia Stäl as represented in New Guinea (Homoptera, Cicadidae). — Zoologische Mededelingen 37: 61-80. Boer, A. J. de, 1982. The taxonomy and biogeography of the nasuta group of the genus Baeturia Stäl, 1866 (Homoptera, Tibicinidae). — Beaufortia 32: 57-78. Boer, A. J. de, 1986. The taxonomy and biogeography of the conviva group of the genus Baeturia Stâl, 1866 (Homoptera, Tibicinidae). — Beaufortia 36: 167-182. Boer, A. J. de, 1989. The taxonomy and biogeography of the bloetei group of the genus Baeturia Stäl, 1866 (Homoptera, Tibicinidae). — Beaufortia 39: 1-43. Boer, A. J. de, 1990. Aedeastria, a new cicada genus from New Guinea, its phylogeny and biogeography (Homoptera, Tibicinidae), preceded by a discussion on the taxonomy of New Guinean Tibicinidae. — Beaufortia 40: 63-72. Boer, A. J. de, 1991. Scottotympana, a new cicad genus from New Guinea, with the description of three new species, their taxonomy and biogeography (Homoptera, Tibicinidae). — Beaufortia 42: 1-11. Boer, A. J. de, 1992. The taxonomy and biogeography of the viridis group of the genus Baeturia Stal, 1866 (Homoptera, Tibicinidae). — Bijdragen tot de Dierkunde 61 (3): 163-183. Distant, W. L., 1897. Viaggio di Lamberto Loria nella Papuasia orientale, xvii. Additions to our knowledge of the Cicadidae of New Guinea. — Annali del Museo civico di Storia naturale Giacomo Doria Genova (2) 17: 378-383. Distant, W. L., 1906. A synonymic catalogue of Homoptera. Part 1. Cicadidae 1906: 1-207 (Trustees Indian Museum, London). Duffels, J. P. & P. A. van der Laan, 1985. Catalogue of the Cicadoidea (Homoptera, Auchenorhyncha) 1956-1980. — Series Entomologica 34: i-xvi, 1-414. Metcalf, Z. P., 1963. General catalogue of the Homoptera, VIII. Part 2. Tibicinidae: i-iv, 1-492. North Carolina State College, Raleigh, N.C. Receiveid: 2 July 1993 Accepted: 28 July 1993 JAN PATOCKA Institut für Waldökologie der SAW, Zvolen, Slovakia DIE PUPPEN DER SPANNER MITTELEUROPAS (LEPIDOPTERA, GEOMETRIDAE): CHARAKTERISTIK UND BESTIMMUNGSTABELLE DER GATTUNGEN Patoëka, J., 1994. Die Puppen der Spanner Mitteleuropas (Lepidoptera, Geometridae): Charakteristik und Bestimmungstabelle der Gattungen. — Tijdschrift voor Entomologie, 137: 27-56, figs. 1-227. [ISSN 0040-7496]. Published 15 July 1994. Characteristics of the pupae of the family Geometridae (Lepidoptera) and an identification key for its 161 Central European genera are given. The present system of this family is discussed from the view of pupal characters. The species examined are listed in an appendix. Institut für Waldökologie der SAW, Stürova 2, 960 53 Zvolen, Slovakia. Key words. — Geometridae; Pupae; Taxonomy; Key for genera. Die Familie der Spanner, Geometridae (Lepido- ptera) ist eine der größten und wichtigsten Schmet- terlingsfamilien. Viele ihrer Vertreter sind ernste Schädlinge in Land- und Forstwirtschaft, einige sind sogar Vorratsschädlinge. Die Mehrzahl der Arten ist wegen ihrer Gebundenheit an spezifische Standorte auch vom ökologischen Gesichtspunkt wichtig und kommt als Indikatoren der bedrohten Umwelt in Betracht. Die Kenntnis ihrer jüngeren Entwicklugs- stadien ist für viele Fragen unbedingt notwendig; z. B. für das Studium der Schädlinge, der Prädatoren und Parasitoide, der Bioindikatoren, Bodenfauna, verschiedener Synusien usw. Die bisherige Kenntnis der Puppen der mitteleuropäischen Geometridae ist lückenhaft. Speziell die Puppen dieser Familie behan- delt Khotko (1977) und erwähnt 25 Gattungen. In der vorliegenden Arbeit werden zum Vergleich 161 mitteleuropäische Gattungen, also mehr als 90 % ih- res Bestandes berücksichtigt. An Khotko’s Arbeit knüpfen die Veröffentlichungen des Verfassers an, die zahlreiche weitere Arten der Unterfamilie Laren- tiinae (Patocka 1980-1994) und mehrere Gruppen der Ennominae (Patocka 1978, 1985, 1986 a, b, 1992, 1993) behandeln. Auf diese Weise ist zwar die Mehrzahl der Arten dieser beiden Unterfamilien ein- gehend bearbeitet, es fehlt jedoch eine zusammenfas- sende Bestimmungstabelle der Gattungen. Diese Lücke soll die vorliegende Arbeit ausfüllen. Einge- hendere Angaben über viele Geometridenpuppen fin- den sich bei Ljungdahl (1919) und Nordström et al. (1941), über die Arten an Obstbäumen Speyer (1958), an Tannen Patoëka et al. (1960), an Eichen Patotka (1980) und über die Puppen in der Bodenstreu in Kiefernwäldern Koehler (1937). Einzelbeschreibun- gen von Geometridenpuppen sind in weiteren Litera- turquellen verstreut. Die Handbücher über die Schmetterlinge Mitteleuropas, wie Forster & Wohl- fahrt (1981) geben nur kurze und für eine Bestim- mung nicht ausreichende Puppenbeschreibungen. Die Puppen der kanadischen Geometridae wurden im Rahmen einer Monographie dieser Familie von McGuffin (1967-1981) bearbeitet. Das hier benützte System und die Nomenklatur folgen grundsätzlich Leraut (1980) und Gustafsson (1987), die morphologische Terminologie richtet sich nach Mosher (1916) und McGuffin (1967- 1981). Das untersuchte Material (die hier berück- sichtigten Gattungen und Arten sind im Anhang auf- gelistet) stammt großenteils aus der Sammlung des Verfassers, zu welcher viele Fachkollegen durch Überlassung von Puppen und Puppenexuvien beige- tragen haben, wofür ihnen auch an dieser Stelle herz- lich gedankt sei. Viele Arten konnte sich der Verfasser von Museen ausleihen, insbesondere von dem Museum für Naturkunde der Humboldt-Universität in Berlin (Dr. W. Mey), von dem Naturhistorischen Museum Wien (Dr. M. Lödl), von dem Zoologi- schen Museum Kopenhagen (O. Karsholt), von der Zoologischen Staatssammlung München (Dr. W. Dierl) und aus der Entomologischen Abteilung des Nationalmuseums in Prag (Dr. F. Krampl). Den ge- nannten Institutionen und Herren dankt der Ver- fasser verbindlichst, ebenso wie auch der Alexander von Humboldt-Stiftung Bonn für die Erteilung eines Stipendiums, welches ihm das Studium des Materials in München, sowie der Hanskarl-Göttling-Stiftung der Bayerischen forstlichen Versuchs- und For- schungsanstalt, welche einen Aufenthalt in Berlin er- möglichten. T1JDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Die Puppen und ihre Exuvien wurden mit Hilfe ei- nes Stereomikroskopes und z. T. eines Abbeschen Zeichenapparates untersucht und gezeichnet. CHARAKTERISTIK DER FAMILIE GEOMETRIDAE Eine Übersicht der Puppenmorphologie dieser Familie geben die Abb. 1-6. Die Puppen der mittel- europäischen Geometridae sind kleine bis mittel- große (etwa 5-25 mm lang) Pupae obtectae. Sie sind schlank (Abb. 86) bis gedrungen (Abb. 94), meistens in der Mitte am breitesten, vorne abgerundet und hinten mehr zugespitzt (Abb. 1, 2, 88, 94). Manche Formen (z.B. Cyclophora, Abb. 130) sind vorne durch einen Höcker an der Basis der Vorderflügel breiter und sehen in Ventralansicht eher keilartig aus. Die Mehrzahl der Puppen ist einfarbig gelb-, rot-, oder schwarzbraun bis fast schwarz gefärbt und + glän- zend. Es handelt sich hierbei um Arten die sich im Boden oder in der Bodenstreu, in einem Erdkokon bzw. einer Erdhöhle verpuppen. Die übrigen Puppen sind grün, gelblich oder sandfarben, + glanzlos und oft mit dunklen Zeichnungen versehen. Das sind Arten, die sich oberirdisch bzw. am Boden in einem leichten Gespinst, oder nur mit Einzelfäden befestigt verpuppen. Eine Gattung (Abraxas) ist glänzend schwarz mit satt gelben Querstreifen am Abdomen, die sich ebenfalls oberirdisch verpuppt. Die Oberfläche der Puppen ist fein bis grob skulp- turiert. Meist ist der Kopf und Thorax + gerunzelt oder gefurcht und das Abdomen (manchmal auch der Thorax, insbesondere das Metanotum) mit kleinen punktförmigen Grübchen, den sog. Punktgrübchen (Abb. 210) besetzt. Selten entspringen von diesen Microbörstchen (Apeira, Abb. 74). Bei wenigen Puppen (z. B. bei der Gattung Geometra, Abb. 78) kommt eine Stachelskulptur vor. Die Körperborsten sind kaum sichtbar bis relativ groß und stark (Abb. 31). Der Vertex wächst entweder mit der Frons ohne eine Epicranialnaht zusammen (Vertreter der Unterfamilie Ennominae und einige andere, Abb. 4), oder er ist durch eine deutliche Epicranialnaht abge- trennt (Mehrzahl der übrigen Unterfamilien, Abb. 3). Beim Schlüpfen löst sich der Vertex bei dieser Gruppe von der Frons ab. Die Frons ist in wenigen Fällen mit einem Paar Höcker (Abb. 8, 9, 86, 125) versehen, oder kegelförmig vorgezogen (Abb. 43), an- sonst meist abgerundet. Zwei Paare von Frontal- (Abb. 4) und ein Paar von Clypealborsten (Abb. 116) sind meist deutlich und groß ausgebildet. Von der Frontoclypealnaht sind nur Teile nahe der Antennen deutlich erkennbar. Die Komplexaugen sind groß und bestehen aus einem + glatten, ventralen Streifen und einem oft mehr rauhen Dorsalteil (Abb. 95, 96). Die Mandibeln sind meist kaum differenziert (Abb. 112, 116), bei Theria jedoch stark hervortretend (Abb. 22). Das Labrum erscheint meist halbkreis- (Abb. 46), schaufel- (Abb. 117) oder trapezförmig (Abb. 47), gemeinsam mit dem schwach abgegrenz- ten Postclypeus oft fünfeckig (Abb. 116). Das Labium ist entweder ganz verdeckt (Abb. 87) oder klein (Abb. 209) bis groß (Abb. 101), als eine drei- oder fünfeckige Platte sichtbar. Die Maxillar- und Labialpalpen sind stets verdeckt. Die Genae grenzen an die Vorderbeine; diese Grenze immer kürzer als die Grenze zwischen Auge und Vorderbein. Die Grenzen zwischen Vorderbeine und Antennen kurz oder lang (Abb. 19, 21). Bei den sonst ähnlichen Puppen der Herminiinae, Noctuidae, grenzen die Vorderbeine nicht an die Antennen, weil die Mittelbeine die Augen berühren. Die Vorder- und Mittelbeine sind meist relativ lang, die Vorderbeine zuweilen (Abb. 16), die Mittelbeine meistens unge- fähr so lang wie die Antennen und Vorderflügel. Die Enden der Hinterbeine sind oft sichtbar. Die Vorderschenkel sind sichtbar, groß (Abb. 1, 125) oder klein, schmal (Abb. 94, 77, 165), oft aber ganz verdeckt (Abb. 16, 94a). Die Proboscis ist meist so lang oder wenig kürzer als die Vorderflügel, selten stark verkürzt und deutlich kürzer als die Mittelbeine, bzw. auch als die Vorderbeine, (Abb. 16, 98, 99). In wenigen Fällen überragt die Proboscis (Abb. 9, 12, 94a, 123) und noch seltener überragen auch die Mittelbeine und Antennen (Abb. 13) die Spitzen der Vorderflügel und den Kaudalrand des 4. Abdominal- segmentes. Bei Rhodometra (Abb. 123) verschiebt die vergrößerte Proboscis das Labrum bis nahe am Frontalende des Kopfes, was oft auch bei Sphingidae mit langer Proboscis vorkommt. Die Antennen sind meist ungefähr so lang wie die Vorderflügel (Abb. 1). Die Vorderflügel überragen den Kaudalrand des 4. Abdominalsegmentes nicht, sind voneinander ge- trennt (Ausnahme: Apeira, Abb. 76) und zeichnen sich zuweilen durch hervorstehende (Abb. 94) oder dunkel gestreifte bzw. helle Adern aus. Selten liegt an ihren Wurzeln je ein Höcker (Abb. 123, 125, 130). Die Hinterflügel säumen die Vorderflügel meist zu % bis 4 des 4. Abdominalsegmentes, nicht aber bis zu deren Spitzen, wie es bei der Familie Drepanidae, die den Geometridae sonst ähnlich ist, vorkommt. Eine Ausnahme ist die Gattung Archiearis. Die thorakalen Spiracula sind klein und spaltför- mig. Oft bildet sich kaudal von ihnen am Mesono- tum, ein auffälliger, tomentöser (Abb. 31, 91), rauh skulpturierter (Abb. 54) oder ‘ohrartig hervortreten- der, kantiger Höcker (Abb. 33, 35). Das Metanotum besitzt frontal in der Mitte meist einen mitteltiefen und abgerundeten Ausschnitt. Die Frontolaterale Lappen des Metanotums sind meist abgerundet (Abb. 102, 103), zuweilen spitz oder winkelig (Abb. 153), insbesondere bei den Vertretern der Unterfami- lie Sterrhinae. Die Vorderecken des 1. Abdominalseg- mentes sind meist abgerundet, Ihre Seiten + parallel. Die Spiracula sind am 2.-3. Abdominalsegment sicht- bar oder + verdeckt. Die abdominalen Spiracula sind klein bis mittelgroß, meist schmal elliptisch, oft mit breiten elliptischen Höfen und nicht selten + erha- ben. Die 5.-7. Abdominalsegmente sind beweglich. An der Basis des 5. Segmentes frontal des Spiracu- lums finden sich oft Leisten, Gruben und verdichtete Punktgrübchen (Abb. 34, 38, 40, 45, 49-55, 138, 140-142), die gute Unterscheidungsmerkmale bie- ten. An der intakten Puppe sind sie jedoch durch den Kaudalrand des 4. Abdominalsegmentes oft verdeckt; dieser sollte deshalb, um eine Untersuchung zu er- möglichen, mit einer Nadel beseitigt werden. An der Puppenexuvie sind diese Gebilde auch ohne diesen Eingriff meist gut sichtbar. Bei der Gattung Rheuma- ptera gibt es Querleisten und Rinnen an der Basis der Dorsalseite des 5. Abdominalsegmentes (Abb. 139). Zwischen dem 9. und 10. Segment besitzt die Mehr- zahl der Spannerpuppen einen + tiefen, rinnenartigen Quereinschnitt, die sog. Dorsalrinne, welche an ihrer Kaudalseite meist einen oder mehrere Ausläufer auf- weist, deren Ränder oft verdunkelt und + tomentös sind. Seltener bildet die Dorsalrinne auch einen oder einige Ausläufer frontalwärts (Frontalausläufer, Abb. 6). An den Seiten entspringt von dieser Rinne oft noch ein dorso-kaudalwärts gerichteter Lateralein- schnitt (Abb. 5). Dorsalseite des 10. Abdominalsegmentes ist ver- schieden lang (Abb. 70, 71), seine Ventralseite oft + breit mit gewölbten Seiten, abgeflacht, und der manchmal erhabene frontolaterale Teil durch flache Furchen abgegrenzt. Die Analnaht ist meist lang, ihre oft erhabenen Seiten wallartig, manchmal auch mit Längs-, seltener Querrunzelung. Das sog. Analfeld groß und auffällig. Die Genitalnaht und ihr Genital- feld, besonders bei den Männchen sind oft groß. Der Absturz zur Basis des Kremasters verläuft steil bis schräg (Abb. 137, 182). Der Kaudalausläufer des 10. Segmentes, der Kremaster, welcher bei der Puppe ei- ne Haft- und Stützfunktion besitzt (Patocka 1978), ist bei den Puppen der Geometridae immer vorhan- den und + groß, seltener klein (Abb. 17) und mehr oder weniger scharf von dem 10. Abdominalsegment abgegrenzt bzw. abgesetzt. In der Dorsalsicht ist er oft haufen- (Abb. 119) oder zungenförmig (Abb. 80), dreieckig (Abb. 178), seltener trapez- (Abb. 18, 164) oder ankerförmig (Abb. 7). Ventral an seiner Basis gibt es oft ein Paar Basalhöcker, dazwischen eine Vertiefung bzw. Mulde. Bei der Unterfamilie Geo- metrinae ist der Kremaster ventral charakteristisch ra- dial gefurcht (Abb. 75, 79), ansonst ist er + glatt oder verschiedenartig skulpturiert (gerunzelt, gefurcht, ge- rippt; Abb. 120, 122). Der Kremaster trägt meist 4, seltener 1-3 oder 5-7 (bei Apeira eine Anzahl) Paare von Borsten, die oft am Ende hakenartig gekrümmt Patocka: Puppen der Spanner Mitteleuropas sind. Die Bezeichnung von 4 Paaren von Primärbor- sten am Kremaster (nach McGuffin 1967-1981) zei- gen die Abb. 5-6. Bei vielen Geometriden-Puppen wachsen die starken Borsten D, zu einem Stiel zusam- men und nur ihre Spitzen bleiben als eine Gabelspitze frei (Abb. 32), die manchmal auch vermehrt sind (Abb. 28). Während bei vielen Vertretern der Unter- familie Larentiinae die übrigen Borsten noch erhalten bleiben (Abb. 177, 178) verschwinden sie bei zahlrei- chen Ennominae vollständig (Abb. 71). Bei Bupalus bleibt nur eine einfache, stumpfe Spitze übrig (Abb. 42). Die Puppen der Spanner findet man z. T. an der Nahrungspflanze, durch die Häkchen am Kremaster und einen Gürtelfaden befestigt (z.B. bei Cyclophora) oder in einem lockeren Gespinst (Abraxas, Ennomos, bei den Comibaenini in dem Sack der Raupe) direkt an der Nahrungspflanze oder auch am Boden. Die Mehrzahl der Puppen ruht jedoch in einem + festen Erdkokon oder einer Erdhöhle in der Bodenstreu oder im Boden selbst. Die Puppen variieren in der Größe insbesondere nach Geschlecht, Generation oder relativer Abundanz. Die Raupen leben an ver- schiedenen Gruppen von Pflanzen, sogar an Sporen- pflanzen bzw. an Detritus. Es gibt ebenso polyphage, wie relativ streng oligophage, an spezielle Pflanzen- gattungen gebundene Arten (z. B. Clematis, Actaea, Berberis, Rhamnus, Hypericum, Veratrum). Manche Gruppen (Eupitheciini, Perizomini) bevorzugen Blüten und Früchte. Viele Arten leben an Laubhöl- zern, an denen sie oft auch schädlich werden, aber auch an Nadelhölzern, und eine zahlreiche Gruppe von Larentiinae lebt an Galium. Die Spanner bewohnen verschiedenartige Biotope, sehr viele Wälder und steppenartige, mehrere auch ruderale Standorte, Sümpfe bzw. Torfmoore, Felsen und Alpenwiesen im Hochgebirge. Viele Arten sind für bestimmte Biotope recht charakteristisch. BESTIMMUNGSTABELLE DER GATTUNGEN Einige Gattungen, wie Orthostixis Hübner, 1822, Costaconvexa Agenjo, 1949, Coenocalpe Hübner, 1825, Bichroma Gumppenberg, 1887, Eilicrinia Hübner, 1825, Mannia Prout, 1915, konnten infolge Materialmangels nicht berücksichtigt werden. 1 Kaudalhalfte des Kremasters in Dorsalsicht ein- fach zugespitzt (selten mit zwei Börstchen am Ende):(AbbS42) ee ee Bupalus Leach — Kaudalhälfte des Kremasters in Dorsalsicht mit 2-4 Dornen oder Zähnen, jedoch ohne Börstchen oder Häkchen (Abb. 7, 14, 20). Die Dornen oft gestielt (Abb. 32, 52), zuweilen auch Verzweipbi(Abb 28) EN nno d 29 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 30 = Kaudalhälfte des Kremasters mit 2-8 (seltener mehr) Häkchen oder Borsten (Abb. 6, 72, 80, 89, 136). Die Endborsten (D,) zuweilen + dornartig umgebildet (Abb. 89, 143, 218). SEON Le We eee PAT RUE 3 = Frons ohne ein Höckerpaar, selten kegelför- mig vorgezogen (Abb. 43), sonst abgerundet (ATD WEA ENE 4 Proboscis überragt deutlich den Kaudalrand des 4. Abdominalsegmentes (Abb. 9) … … …. Re eine ns Siona Duponchel = Proboscis überragt den Kaudalrand des 4. Abdominalsegmentes nicht (vgl. Abb. 1) ...... NEE ehe. Aethalura McDunnough Einzelne Abdominalsegmente dorsal mit gro- ßen, elliptischen, dunkel umrandeten Gru- benttAb DEREN TM Dt anne 5 = Abdominalsegmente ohne solche Gruben (elba ee e i ere 6 Am 5.-7. Abdominalsegment dorsal fehlen die Gruben ..................... Lithostege Hübner — Auch am 5-7. Abdominalsegment gibt es dunkel umrandete Gruben (Abb. 11) … … … EERE enen Chesias Treitschke Proboscis überragt den Kaudalrand des 4. Abdominalsegmentes (Abb. 12, 13) … … … 7 = Proboscis überragt nicht den Kaudalrand des 4. Abdominalsegmentes (Abb. 1)............... 9 Sowohl die Antennen wie auch Mittelbeine überragen den Kaudalrand des 4. Abdomi- meals TN CREME PE SET 8 - Nur die Antennen, nicht jedoch die Mittel- beine überragen den Kaudalrand des 4. Ab- dominalsegmentes............. Odezia Boisduval = Weder Antennen noch Mittelbeine überra- gen den Kaudalrand des 4. Abdominalseg- mentes (Abb. 12) … … … Schistostege Hübner Kremaster länger als basal breit (Abb. 15)... boon Carsia Hübner — Kremaster kürzer als basal breit (Abb. 13, 14) ee A neet Aplocera Stephens Patocka: Puppen der Spanner Mitteleuropas 9 (6) Kremaster ankerartig, stumpf abgerundet, an seinen Seiten je ein größerer und oft auch kleinerer Zahn (Abb. 7). Puppe schlank........ OSE EEE MAURA NG Archiearis Hübner = Kremaster am Ende mit gabelartigen, schräg kaudalwärts gerichteten Dornen (Abb. 25, 32, 52), diese zuweilen weiter verzweigt (Abb. 28). Wenn die Dornen + gegeneinan- der stehen (Abb. 17, 20), nur in einem Paar und die Puppe ist gedrungen ................... 10 Weder Labium, noch Vorderschenkel sicht- babbel GS) PIRRO ee ee 11 = Labium, Vorderschenkel, oder beides sicht- bat (Vel Abbiate 23 11 (10) 8. Abdominalsegment dorsal mit einem Paar spitzeriHlöcker er ner, Nychiodes Lederer = 8. Abdominalsegment dorsal ohne Höcker … 10 (9) 12 (11) Kremaster mit zwei Paaren schlanker, wenig divergierender Dornen, außerdem in seinem Basalteil noch ein Paar Höcker (Abb. 23) ..... CONE Na Artiora Meyrick = Kremaster mit 1 Paar Dornen, die zuweilen noch verzweigt sein können, außerdem höch- stens mit kurzen Zähnen an den Seiten (Abb. 172202237262) EEE ONE 13 13 (12) Kremaster klein, kurz, seine Dornen stark di- vergicrend (Abbi 17, 18,20) em. 14 = Kremaster groß, am Ende mit langen Dornen oder mit einer gestielten Gabel (Abb. 25, 27, ZONE EEL ENNE N 16 14 (13) Proboscis kürzer als die Vorderbeine (Abb. JO) EN Operophtera Hübner = Proboscis langer als die Vorderbeine (Abb. 1) 15 (14) Grenzen zwischen Vorderbeine und Anten- nen etwa 2X so lang wie die Grenzen zwi- schen Vorderbeine und Augen (Abb. 21). Mandibeln erhaben .............. Theria Hübner = Grenzen zwischen Vorderbeine und Anten- nen höchstens wenig länger (oft kürzer) als die Grenze mit den Augen (Abb. 19). Mandi- beln nicht erhaben. ............ Alsophila Hübner Abb. 1-21. — 1,2,3,5,6, Horisme vitalbata, 4, Synopsia sociaria; 7, Archiearis notha, 8, Aethalura punctulata; 9, 10, Siona line- ata; 11, Chesias legatella, 12, Schistostege treitschkei, 13, 14, Aplocera praeformata; 15, Carsia sororiata, 16, 17, Operophthera brumata; 18, 19, Alsophila quadripunctaria, 20, 21, Theria rupicapraria. 1, Habitusbild der Puppe, Ventral; 2, Dorsalsicht; 3, 4, Frons und Umgebung; 5, Abdominalende in Lateralsicht; 6, idem, in Dorsalsicht; 7, Kremaster in Dorsalsicht; 8, Kopf in Ventralsicht; 9, Frontalhälfte der Puppe in Ventralsicht; 10, 14, 15, 17, 18, 20, Abdominalende in Dorsalsicht; 11, Abdominalhälfte der Puppe in Dorsalsicht; 12, 13, idem in Ventralsicht; 16, Kopf und Beine in Ventralsicht; 19, 21, Augen und Umgebung. A,-A,, = 1.-10. Abdominalsegment, Af = Analfeld, An = Analnaht, Ant = Antennen, As = abdominale Spiracula, Cl = Clypeus, D,, D, = Dorsalborsten 1,2, Dr = Dorsalrinne, Es = Epicranialnaht, Fa = Frontalausläufer der Dorsalrinne, Fr = Frons, G = Genae, Gn = Genitalnaht, Hb = Hinterbeine, Hfl = Hinterflügel, Ka = Kaudalausläufer der Dorsalrinne, Kr = Kremaster, L, = Lateralborste 1, Lb = Labium, Lbr = Labrum, Le = Lateraleinschnitt, Mb = Mittelbeine, Msn = Mesonotum, Mtn = Metanotum, O = Auge, Pcl = Postclypeus, Pn = Pronotum, Pr = Proboscis, Sd, = Subdorsalborste 1, Ts = thorakales Spiraculum, V = Vertex, Vb = Vorderbeine, Vfl = Vorderflügel, Vs = Vorderschenkel ail TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 32 16 (13) 9. Abdominalsegment mit einer Querreihe* von Dornen (Abb. 24).. Eurrhanthis Hübner — 9. Abdominalsegment ohne Dornen (Abb. DO) enso nts te th ee AURA monte o 1194 17 (16) 10. Abdominalsegment mit je einem spitzen lateralen Fortsatz (manchmal ein weiterer an der Mitte des Kremasters, Abb. 25) .............. EIER ER RN N. no ARNO I TRI Lycia Hübner — 10. Abdominalsegment ohne, Kremaster oft mit einem Paar von laterale Fortsätze (Abb. 18 (17) Dorsalrinne vorhanden, Kaudalausläufer + tomentös, Lateraleinschnitt deutlich (Abb. DO; Ds AB dte 19 = Dorsalrinne und Lateraleinschnitt nicht vor- handemt@Xbb32) ER 21 19 (18) Erhebung kaudal des thorakalen Spiraculums flach, länglich elliptisch, stark tomentös (Abb. 31). Puppe 12-15 mm lang.................. IMRE N TT Mit Erannis Hübner = Erhebung kaudal des thorakalen Spiraculums ‘ohrartig’, kantig (Abb. 33, 35). Puppe über 17 RI ARRE e 20 20 (19) Leiste frontal des Spiraculums am 5. Abdo- minalsegment doppelt (Abb. 34). Erhebung kaudal des thorakalen Spiraculums steiler frontal- als kaudalwärts (Abb. 33). Puppe 17 TT) Er EE Apocheima Hübner — Leiste frontal des Spiraculums am 5. Abdo- minalsegment einfach (Abb. 36). Erhebung kaudal des thorakalen Spiraculums weniger steil frontal- als kaudalwärts. (Abb. 35) Pup- pe 20526fmm lang... Biston Leach 21 (18) Puppe 17-21 mm lang.. Phigalia Duponchel — Ruppeal0-15immlang ine 29) 22 (21) Kremaster mit deutlichen lateralen Fort- sätzen (Abb. 41). Vertiefung frontal des Spiraculums am 5. Abdominalsegment mit zahlreichen Punktgrübchen frontal der Querleiste (Abb. 40)....... Selidosema Hübner = Kremaster ohne, bzw. nur mit angedeuteten lateralen Fortsätzen (Abb. 32). Vertiefung am 5. Abdominalsegment frontal des Spira- culums mit Querleisten, ohne Punktgrüb- chen (Abb. 37, 38) ............ Agriopis Hübner 23 (10) Dorsalrinne mit 1-2 großen Kaudalausläu- Patocka: Puppen der Spanner Mitteleuropas fern ADD 2607829) RER ee 24 = Dorsalrinne mit mehreren Kaudalausläufern, oder nicht vorhanden (Abb. 56, 62, 70) .. 28 24 (23) Vorderschenkel nicht sichtbar (vgl. Abb. 82) = Vorderschenkel sichtbar (vgl. Abb. 1) … 26 25 (24) Dorsalrinne mit nur einem großen Kaudal- ausläufer. Kremaster ohne laterale Fortsätze (Abba2.6)/ ie eee Chondrosoma Anker = Dorsalrinne mit zwei großen Kaudalausläu- fermi (Abb 2A Lignyoptera Lederer 26 (24) Frons breit kegelförmig vorgezogen (Abb. 43). Dorsalrinne auch an der Frontalseite ausgebuchtet (Abb. 44) … … Pelurga Hübner = Frons nicht vorgezogen. Dorsalrinne nur an der Kaudalseite mit einem großen, spitzen Vorsprung (Abb#29);727 702720 27 27 (26) Puppe mehr gedrungen, schwärzlich, Exuvie schwarzbraun. Puppenlänge 6-8 mm ............ MEER TEE REN Phibalapteryx Stephens - Puppe schlanker, bräunlich, Exuvie gelb- braun. Puppenlänge 9-10 mm … MAC RC RON MIE ape Cataclysme Hübner 28 (23) Dornen am Kremaster voneinander entfernt entspringend (Abb 30) tear. 29 = Dornen am Kremaster gabelartig, gestielt (Abb#285 672024206) e eee 30 29 (28) Kremaster an der Dorsalseite fein skulptu- Hert. Püuppenur 6-8 ma lang... Lr Pygmaena Boisduval - Kremaster an der Dorsalseite grob skulptu- riert. Puppe über 10 mm lang... jose SUR HT BR IN Gnophos Treitschke s. lat. 30 (28) Proboscis wesentlich kürzer als die Mittel- beine. Labium sehr klein. Puppe 7-9 mm lang. A. ER RER Tephronia Hübner = Proboscis nicht oder wenig kürzer als die Mittelbeine (im Zweifelsfall Labium Mittel- oro) sis SR IE EN CR RER. 31 31 (30) Dornen am Kremaster + länger als ihr Stiel (Abb. 102, 206). Puppen nur 6-8 lang... 152 = Dornen am Kremaster + kiirzer als ihr Stiel (Abb. 52, 62, 67). Meist größere Puppen ..... 32 EEE 33 Abb. 22-57. — 22, 39, Theria rupicapraria, 23, Artiora evonymaria, 24, Eurranthis plumistaria, 25, Lycia zonaria, 26, Chondrosoma fiduciaria; 27, Lignyoptera thaumastaria, 28, 31, Erannis defoliara, 29, Phibalapteryx virgata; 30, Gnophos ob- fuscatus, 32, 38, Agriopis aurantiaria, 33, 34, Apocheima hispidaria, 35, 36, Biston strataria, 37, Agriopis bajaria, 40, 41, Selidosema plumaria; 42, Bupalus piniarius, 43, 44, Pelurga comitata; 45, 46, Ectropis crepuscularia, 47-49, Calospilos sylvata; 50, Cleora cinctaria, 51, Fagivorina arenaria; 52-54, Serraca punctinalis 55, Ematurga atomaria, 56, 57, Chiasmia clathrata. 22, Labrum, Mandibulae; 23-30, 32, 41, 42, 44, 52, 56, Abdominalende in Dorsalsicht; 31, 33, 35, Erhebung kaudal des thorakalen Spiraculums; 34, 36, 38, Basis des 5. Abdominalsegmentes, Lateralsicht; 29, Kopf und Beine in Ventralsicht; 40, 45, 49-51, 53, 57, Basis des 5. Abdominalsegmentes, Lateralsicht; 43, Kopf in Ventralsicht; 46, 47, Labrum, Labium; 48, Abdominalende in Lateralsicht; 54, Erhebung kaudal des thorakalen Spiraculums. 33 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 34 = Dorsalrinne nicht entwickelt (Abb. 70, 71) » EUR RU CR aq 47 33 (32) Kaudal des thorakalen Spiraculums je eine re- lativ große, breite, schr rauhe, erhöhte Fläche (Abb. 54). An der Basis des 5. Abdominalseg- mentes lateral zwischen zwei dunklen Quer- leisten wenigstens 12 Zellen in einer Reihe (ADD DNS) tel 34 = Kaudal des thorakalen Spiraculums eine klei- ne, längliche oder eine unauffallige Erhebung (vgl. Abb. 106, 111). Wenn diese jedoch grö- fer und rauher ist (Abb. 227) und auch sonst, gibt es an der Basis des 5. Abdominal- segmentes lateral anders aussehende Bildun- gen (Abb. 51, 57, 65, 68, 69) oder nur etwa oaZellenil(Abb. SO)... Ur Lee 36 34 (33) Mesonotum ohne Punktgrübchen ................ RE eneen Hypomecis Hübner = Mesonotum mit Punktgrübchen.............. 35 35 (34) Spiraculum am 5. Abdominalsegment liegt in der Nähe der 10.-15. Zelle (von der Ven- tralseite gezählt, Abb. 53). Puppe meist über 14 mm lang, eher glänzend ... Serraca Moore = Spiraculum am 5. Abdominalsegment liegt in der Nähe der 2.-6. Zelle (Abb. 55). Puppe meist matter, unter 13 mm lang O a ARE REN i EEE eee Ematurga Lederer 36 (33) Abdomen ohne Punktgrübchen (vgl. Abb. DIR Et een Lomaspilis Hübner = Abdomen mit Punktgrübchen (vgl. Abb. 51) on ERBE. nn EN ee her 37 37 (36) Am 5. Abdominalsegment frontal des Spira- culum eine länglich ovale, von einer dunklen Leiste umgebene Fläche (Abb. 45) … … … 38 — Am 5. Abdominalsegment vor dem Spiracu- lum andere Bildungen: Zellen (Abb. 50), Leisten (Abb. 65, 68, 69) oder nur Punkt- grübchen (Abb. 57) bzw. eine glatte Fläche Allos, SE EA A 39 38 (37) Labium ganz klein (Abb. 46) bzw. fehlend. Vorderschenkel gut sichtbar (vgl. Abb. 1). Lateraleinschnitt undeutlich ......................... RR da Ectropis Hübner = Labium (Abb. 47) und Lateraleinschnitt (Abb. 48) deutlich. Vorderschenkel verdeckt (eIPADDNS OD) een Calospilos Hübner 39 (37) Frontal des Spiraculums am 5. Abdominal- Patoëka: Puppen der Spanner Mitteleuropas segment nur eine glatte Fläche, weder mit Leisten, noch Punktgrübchen (Abb. 51) … … Nh SERE rio dA MOE Fagivorina Wehrli = Frontal des Spiraculums am 5. Abdominal- segment eine dunkle Leiste und vor ihr etwa fünf Zellen (Abb. 50)............ Ascotis Hübner = Frontal des Spiraculums am 5. Abdominal- segment Leisten und Punktgrübchen (Abb. 65, 69), oder nur Leisten (Abb. 68) oder Punkterübchen (Abb.5y).... 2... 40 40 (39) 9. Abdominalsegment mit Punktgrübchen (ADDED ORG Eeen A 4] = 9. Abdominalsegment ohne Punktgrübchen (AbbS62A69) RI NA RENTE 43 41 (40) Frontal des Spiraculums am 5. Abdominal- segment keine glatte Fläche, nur Punktgrüb- chen (Abb. 57) Chiasmia Hübner — Frontal bzw. frontodorsal des Spiraculums am 5. Abdominalsegment eine glatte Fläche und davor liegen Punktgrübchen bzw. eine Querleiste(Abb 59 OO) Me 42 42 (41) An der Basis des 5. Abdominalsegmentes la- teral nur Punktgrübchen ohne eine dunkle Querleiste (Abb. 59). Stiel am Kremaster re- lativ kurz Abb. 58)... Parectropis Sato = An der Basis des 5. Abdominalsegmentes la- teral eine dunkle Querleiste mit einer Reihe von dicht gedrängten Punktgrübchen (Abb. 60). Stiel des Kremasters lang (Abb. 61) … … Be OR aI US i PE, Tephrina Gueneé 43 (40) Dorsalrinne ohne Kaudalausläufer … … … … … A DEN ee Euconista Lederer — Dorsalrinne mit Kaudalausläufern … … … 44 44 (43) Frontal des Spiraculums am 5. Abdominal- segment liegen nur Punktgrübchen (Abb. DIE TNA SIEM IRENE OENE AR 45 - Frontal des Spiraculum am 5. Abdominalseg- ment liegt eine glatte Fläche, dann Punkt- grübehen (Abbsa9)e 2.2... me nn. 46 = Frontal des Spiraculums am 5. Abdominal- segment liegt zuerst eine glatte Fläche, dann eine dicke Leiste und eine Vertiefung mit Leisten und Punktgrübchen (Abb. 226). N eae ae Retr Odontognophos Wehrli 45 (44) Kaudal des thorakalen Spiraculums eine längliche tomentöse Erhebung (vgl. Abb. 31) en Isturgia Hübner Abb. 58-87. — 58, 59, Parectropis similaria, 60, 61 Tephrina arenacearia; 62, Deileptenia ribeata; 63, Arichanna melanaria; 64, Stegania dilectaria; 65, Peribatodes rhomboidaria; 66, Semiothisa signaria, 67, 68, Itame brunneata; 69, 71, Diastictis artesiaria; 70, Narraga fasciolaria, 72, Abraxas grossulariata, 73, Ennomos autumnaria; 74, 76, Apeira syringaria; 75, 77, 78, Hemistola biliosata; 79, Jodis putata; 80, 81, Pseudoterpna pruinata, 82-84, Thalera fimbrialis 85, Chlorissa cloraria; 86, Hemithea aesti- varia; 87, Epirrita christyi. 58, 61-63, 66, 67, 70-73, 80, Abdominalende in Dorsalsicht; 59, 60, 64, 65, 68, 69, 75, 78, 89, Basis des 5. Abdominalsegmentes in Lateralsicht; 75, 79, Abdominalende in Ventralsicht; 76, 82, 86, Habitusbild in Ventralsicht; 77, Kopf und Beine in Ventralsicht; 81, Abdominalende in Lateralsicht; 84, 85, Augen und Umgebung. 35 a 36 = Kaudal des thorakalen Spiraculums keine t6- mentöse Erhebung, nur grobe Runzeln......... CSA I CLIC APR DA TEN, Lobophora Curtis 46 (44) Dorsalrinne mit 8-9 mittelgroßen kaudalen Ausbuchtungen (Abb. 62). Lateraleinschnitt ohne Runzeln um die Spitze. Ventralseite des Kremasters mit großen Basalhöckern ............ sn Deileptenia Hübner — Dorsalrinne mit etwa 20 kleinen Ausbuch- tungen (Abb. 63). Lateraleinschnitt mit star- ken Runzeln um die Spitze. Basalhöcker an der Ventralseite des Kremasters schwach oder fehlen Amnerne. EN Arichanna Moore 47 (32) 5. Abdominalsegment frontal des Spiracu- lums mit einer schlank ovalen Leiste. Auch der Kaudalrand des 4. Segmentes dort ver- dick ABDAGA) N. ra AE md 48 — 5. Abdominalsegment frontal des Spiracu- lums mit etwa fünf Zellen, die kaudal durch eine dicke Querleiste begrenzt sind (Abb. 50) sonde Cleora Curtis — 5. Abdominalsegment frontal des Spiracu- lums mit Punktgrübchen und/oder Leisten (AbbI97X60N69, 68569) NN... 49 48 (47) Puppe 6-8 X 2,5 mm........ Stegania Gueneé - Buppe 9210) 73 mm..." Lygdia Gueneé 49 (47) Frontal des Spiraculums am 5. Abdominal- segment eine fast skulpturlose, glatte Fläche und vor ihr Punktgrübchen bzw. Leisten VADDRDONGON ian 50 = Frontal des Spiraculums am 5. Abdominal- segment Punktgrübchen und/oder Leisten, ohne eine glatte Fläche dazwischen (Abb. 57, GSMO) Nee AN IT RR 52 50 (49) 9. Abdominalsegment mit Punktgrübchen Ge EABDRO IDEE. Je nani Alcis Curtis = 9. Abdominalsegment ohne Punktgrübchen (ab Aion 2 DR RE N ENG 51 51 (50) Lateral an der Basis des 5. Abdominalseg- mentes eine dicke Querleiste und vor ihr dichte Punktgrübchen (Abb. 65) .................. PERREN TOTI n iDan) Peribatodes Wehrli = Lateral an der Basis des 5. Abdominalseg- mentes nur Punktgriibchen, keine Querleiste (NEN Paradarsia Warren 52 (49) An der Basis des 5. Abdominalsegmentes la- teral nur Punktgrübchen (Abb. 57).......... 53 Patocka: Puppen der Spanner Mitteleuropas = An der Basis des 5. Abdominalsegmentes la- teral Leisten und dazwischen manchmal Punktgrübchen (Abb. 68, 69).................. 54 53 (52) Stiel des Kremasters lang (Abb. 66). Ventral- seite des Kremasters ohne deutliche Basal- höcker. Metanotum ohne Punktgrübchen (vel2Abb 102) ER. Semiothisa Hübner = Stiel des Kremasters kurz (Abb. 70). Ventral- seite des Kremasters mit deutlichen Basal- héckern. Metanotum mit Punktgriibchen Ggl'Abb 210) Narraga Walker 54 (52) Stiel des Kremasters lang und schlank (Abb. VA) SITA EN CA Diastictis Hübner n Stiel des Kremasters kurz und dick (Abb. 67) MILA AIEE MUUR, BAER EN: Itame Hübner Puppe glänzend braunschwarz, Kaudalstrei- fen des 5.-8. Abdominalsegmentes leuchtend gelb. Häkchen am Kremaster gleich groß, dick, zerstreut, Sd, entspringt nahe der Basis des Kremasters (Abb. 72) … … Abraxas Leach 55 (1) = Puppe nicht glänzend schwärzlich mit gelben Seitenstreifen. Wenn die Häkchen am Kremaster gleich groß und dick sind, ent- springt Sd, nicht nahe der Basis des KTEMASTÉ TS ln ee 56 56 (55) Mittelteil der Puppe stark verbreitert (Abb. 76) Thorax und Abdomen dorsal mit dichten winzigen Börstchen besetzt (Abb. 74) … … … PER AN ER ES TRE Apeira Gistl - Puppe nicht verbreitert, ohne Bérstchen....... aa EN MORE N rc 57 (56) Am 9. Abdominalsegment an Stelle der Borsten D, (und manchmal auch D,) zahnar- tige Spitzen. 9. Abdominalsegment dorsal stark längsgefurcht (Abb. 73). Puppen etwa 15-22 mm lang, matt.... Ennomos Treitschke = Am 9. Abdominalsegment D' borstenartig oder riickgebildet. Puppe oft kleiner und/- oderlelänzend ne Re 58 58 (57) Kremaster ventral stark radial gefurcht, und/- oder seine Borsten, auch D,, entspringen fä- cherartig von den Seiten der Spitze, sind + gleich groß (D, manchmal etwas stärker) und am Ende stark spiralig eingerollt (Abb. 75, 79, 80). Vorderschenkel nicht sichtbar oder klein (Abb. 82, 86, 88). Puppen bunt, oft grün gefärbt und/oder mit Zeichnungen....... Abb. 88-114. — 88, Chlorissa cloraria, 89, Epirrita christyi, 90, Larentia clavaria, 91, Selenia dentaria, 92, Hypoxystis pluviaria; 93, Plagodis pulveraria, 94, Opisthograptis luteolata, 94a, Therapis flavicaria, 95, 100, Angerona prunaria; 96, 101, Crocallis elinguaria, 97, Epione parallelaria, 98, Synopsia sociaria, 99, Aspilates gilvaria, 102, Epione repandaria; 103, 105-107, Dyscia conspersaria, 104, 108, Pseudopanthera macularia; 109, Petrophora chlorosata; 110-112, Epirrhanthis diversata, 113, Hylaea fa- sciaria, 114, Odontopera bidentata. 88, 94, 94a, Habitusbild in Ventralsicht; 89, 90, 105, 108, 109, 110, 114, Abdominalende in Dorsalsicht; 91, 106, 111, 113, Erhebung kaudal von dem thorakalen Spiraculum; 92, Basis des 5. Abdominalsegmentes in Lateralsicht; 93, 107, Abdominalende in Lateralsicht; 95, 96, Augen und Umgebung; 97, Lateraleinschnitt und Dorsalrinne in Lateralsicht; 98- 101, Kopf und Beine in Ventralsicht; 102-104, Metanotum, 1. Abdominalsegment; 112, Labrum, Labium. 37 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 DE NR EL CHOR 59 _ Kremaster anders gestaltet, ventral meist nicht regelmäßig radial gefurcht. D, wenn vorhanden, entspringt von seiner Dorsalseite. Borsten oft ungleich ‘groß oder lang. Im Zweifelsfall Vorderschenkel groß (Abb. 1), Puppen oft zeichnungslos und + glänzend .... 59 (58) Proboscis so lang wie die Mittelbeine (Abb. 82, 86). Abdomen ohne stachelige Skulptur . = Proboscis kiirzer als die Mittelbeine, doch länger als die Hälfte dieser (Abb. 77). Wenigstens die mittleren Abdominalsegmen- te mit stacheliger Skulptur (Abb. 78) … … 60 = Proboscis kürzer als die Hälfte der Mittel- beine. Abdominalsegmente ohne stachelige Skulptur-----Garantati Aplasta Hübner 60 (59) Puppe einfarbig grün, Exuvie weißlich grün- gelbE ste M ei Hemistola Warren = Puppe mit Zeichnungen und dunklen Pinna- culafderBorsten ee 61 61 (60) Puppe grün, am Rücken z. T. rotbraun, 15- 20 mm lang. Hinterbeine meist nicht sicht- bart ur SEO Geometra Linnaeus - Puppe nicht grün gefärbt, 11-14 mm lang. Spitze der Hinterbeine sichtbar, groß ...... 62 62 (61) Flügeladerung deutlich schwarz abgesetzt, Grundfärbung weißgräulich........................- On Thetidia Boisduval - Flügeladern kaum dunkler, Grundfärbung hell rostbraun ................ Comibaena Hübner 63 (59) Kremaster mittellang mit 4 Paaren von Häk- chen ADDSSO) ee N 64 — Kremaster schr lang mit 5-7 Paaren von Hläkcheni(Abb79) Free: Jodis Hübner 64 (63) Dorsalrinne und Lateraleinschnitt vorhan- den, deutlich (Abb. 80, 81) .......................... PERREN NN Pseudoterpna Hübner = Dorsalrinne und Lateraleinschnitt nicht vor- kauden yelyAbbyg3) nee. 65 65 (64) Puppe recht schlank, mehr als 4 mal so lang wie breit. Frontalhöcker relativ groß (Abb. 86). Puppe eintönig sandfarben oder grün- lich, nur der Rückenstreifen am 1.-8. Abdominalsegment dunkel. Spiracula hell .... EE Hemithea Duponchel - Puppe + gedrungen, Frontalhöcker klein (Abb. 82, 88). Puppe meist bunter gezeich- net, auch am Kopf und Thorax. Spiracula Gunkelivetleckr (ADD #05) Fe er 66 66 (65) Grenze zwischen Vorderbein und Antenne etwa 2X so lang wie die Grenze zwischen Vorderbein und Gena (Abb. 84). Kopf und Thorax ventral, auch die Adern der Vorder- flügel sehr kontrastreich dunkel gezeichnet. 38 Puppe relativ schlanker, 13-15 mm lang........ ee inte de Thalera Hübner _ Grenze zwischen Vorderbein und Antenne etwa 3X so lang wie die Grenze zwischen Vorderbein und Gena (Abb. 85). Kopf und Thorax ventral oft weniger kontrastreich ge- zeichnet. Puppe mehr gedrungen, 9-11 mm langer LU. Re Chlorissa Stephens 67 (58) Vorderschenkel nicht sichtbar (Abb. 99, 101) sica la uh SE er. 68 = Vorderschenkel sichtbar (Abb. 1, 165, 166) sel insel PRE oa OENE 98 68 (67) D, am Kremaster + parallel, Spitze hakenför- mig, wenn divergierend, dann erst im Kaudalteil (Abb. 109) oder Kremaster kurz und stumpf abgerundet (Abb. 119) ......... 72 = D, am Kremaster stark divergierend, Spitze oft mehr dorn- als hakenförmig, manchmal + gestielt. Kremaster länger und/oder spitz (Abb; 89.1167; 1681/4) ra 69 69 (68) Mittlerer Kaudalausläufer der Dorsalrinne vergrößert, tiet (Abb. 89,1 WA) 70 = Mittlerer Kaudalausläufer der Dorsalrinne klein oder flach (Abb. 167, 168, 171) bzw. Dorsalrinne nicht entwickelt (Abb. 93) ... 73 70 (69) Labium nicht sichtbar (Abb. 87). Kremaster lang vorgezogen, D, einen Winkel von fast 180° bildend (Abb. 89) ...... Epirrita Hübner = Labium groß (vgl. Abb. 135), Stiel des Kremasters kürzer, D, einen spitzen Winkel bildend....:::1:::2 e 71 71 (70) Sd, vorhanden, entspringt nahe der Basis des Kremasters, Kaudalausläufer der Dorsalrinne spitz (ADD S10) ee Larentia Treitschke = Sd, am Kremaster fehlt. Kaudalausläufer der Dorsalrinne stumpf abgerundet (Abb. 174) EEE NE 131 72 (68) Proboscis überragt deutlich den Kaudalrand des 4. Abdominalsegmentes (Abb. 94 a) ....... tenui PERS Therapis Hübner — Proboscis überragt den Kaudalrand des 4. Abdominalsegmentes nicht (Abb. 94)....... 73 73 (72) Dorsalrinne nicht entwickelt (Abb. 93) … 74 = Dorsalrinne entwickelt, mindestens stärker und tiefer als der Dorsaleinschnitt zwischen dem 8. und 9. Abdominalsegment (Abb. 105: 109: LIO); nate RE 75 74 (73) Punktgrübchen am Abdomen fehlen (Abb. 92). Proboscis kürzer als die Vorderbeine (vel RAD DROS) ee Hypoxystis Prout = Punktgrübchen am Abdomen vorhanden (vgl. Abb. 102). Proboscis länger als die Vor- derbeine (vgl. Abb. 94)....... Plagodis Hübner 75 (73) Kremaster mit 4 Paaren von Häkchen (Abb. 105, 114). D, + größer als die übrigen Bor- SUE ss rooster Ne Al 76 - Kremaster mit 2 Paaren von Häkchen. D, lang, schlank. Puppe 10-11 mm lang............ RA ABATE PETE Pachycnemia Stephens _ Kremaster mit 2 Paaren von Häkchen. D, kürzer, relativ dick (Abb. 164). Puppe 16-18 TRAME venten RN TAR Colotois Hübner 76 (75) Adern der Vorderflügel stark hervortretend. Vorderschenkel meist etwas sichtbar (Abb. JA) ABuppelsandtarben zitti oberon ANN Opisthograptis Hübner — Adern der Vorderflügel nicht oder kaum her- VORLESEN eend een een Tif. 77 (76) Vorderbeine grenzen aneinander (Abb. 98, OO) RR TE LH tar 78 — Vorderbeine grenzen nicht aneinander (Abb. GVA. TOO eer ee Se oO aR 79 78 (77) Mittelbeine grenzen aneinander (Abb. 98). Kremaster wenig länger als basal breit........... RO A I IVATO Synopsia Hübner — Mittelbeine grenzen nicht aneinander (Abb. 99). Kremaster länger als basal breit (Abb. 105) eerd NER ee Aspilates Treitschke 79 (77) Mittelbeine grenzen aneinander (Abb. 100, LOI) cas ee ADT 80 = Mittelbeine grenzen nicht aneinander (vgl. FAD DMO RE er en ee en nr 82 80 (79) Kremaster in Lateralsicht stumpf. Dorsalrin- ne mit flachen, breiten Ausläufern. Puppe unter to monlanc nn... Cleorodes Warren — Kremaster in Lateralsicht eher spitz. Dorsal- rinne mit tieferen Kaudalausläufern. Puppe Hoer ion TREE 81 81 (80) Puppe fein skulpturiert, am Rücken glän- zend, Augen glatt (Abb. 95). Antennen und Mittelbeine kaudal stumpf, abgerundet (ADD 100)... Angerona Duponchel = Puppe grob skulpturiert, am Riicken kaum glänzend. Augen dorsal grob skulpturiert (Abb. 96). Antennen und Mittelbeine kaudal spitzakeiltormig (Abb. 101). GERNE A EE Crocallis Treitschke 82 (79) Kremaster (in Dorsalsicht, ohne Häkchen) fast zweimal so lang als basal breit (Abb. 105, 107). Kaudal der thorakalen Spiracula je eine relativ große tomentöse Erhebung (Abb. 91, INS) onse A PES AU 83 = Kremaster kaum länger oder kürzer als basal breit. Im Zweifelsfall der Höcker kaudal des thorakalen Spiraculum klein, unauffällig, kattmttomentöstt...... een 85 83 (82) Dorsalrinne mit starken, abgerundeten Kau- dalausläufern (vgl. Abb. 114)...................... OE sia, Charaspilates Wehrli = Dorsalrinne mit undeutlichen, kleinen bzw. mehr spitzen Kaudalausläufern (Abb. 105) … RS e Stel DRE ce 84 Patocka: Puppen der Spanner Mitteleuropas 84 (83) 1. Abdominalsegment ohne Punktgrübchen (Abb. 103). Lateraleinschnitt groß, tief (Abb. NOTA Aro ee Dyscia Hübner = 1. Abdominalsegment mit starken Punkt- grübchen (Abb. 102). Lateraleinschnitt schwach entwickelt oder klein (Abb. 97) ...... A Nn Epione Duponchel 85 (82) Metanotum mit Punktgrübchen (Abb. 104) — Metanotum ohne Punktgrübchen (Abb. 102, TORNE 87 86 (85) Pro- und Mesonotum mit Punktgrübchen (Abb. 104). D, am Kremaster hakenförmig (Abba1O8)E essen Pseudopanthera Hübner = Pro- und Mesonotum ohne Punktgriibchen (vgl. Abb. 210). D, am Kremaster schlank dornförmig, in der kaudalen Hälfte vonein- ander divergierend (Abb. 109) #77 denn EN an Petrophora Hübner 87 (85) Thorakale Spiracula kaudal mit je einem gro- ßen, elliptischen, tomentösen Höcker (Abb. RON RAR ISTE DE ne Selenia Hübner = Thorakale Spiracula kaudal ohne oder nur mit kleinen und schmalen, bzw. nicht to- mentösen Höckern (Abb. 111, 113) … … 88 88 (87) 1-2. Abdominalsegment ohne Punktgrüb- chen (vgl. Abb. 103). D, am Kremaster meist + weit voneinander entfernt (Abb. 120-122) BER Eos Ee 96 — 1-2. Abdominalsegment mit Punktgrübchen (vgl. Abb. 102). D, am Kremaster entsprin- gen nahe beieinander (Abb. 110, 114, 119) Nn 89 89 (88) D, am Kremaster in Dorsalsicht deutlich län- ger als der Kremaster selbst, parallel (Abb. 110). Labrum auffallend lang , abgerundet (ADD AND) PRE Epirrhanthis Hübner — D, am Kremaster nicht länger als der Kre- master selbst oder distal stark divergierend (Abb. 114, 115, 119). Labrum meist kurz undbreit Abb. IC IP) EE. 90 90 (89) 10. Abdominalsegment (ohne Kremaster) in Dorsalsicht fast 2 mal länger als das 9. Abdo- minalsegment, mit einer gebogenen, feinen Querfurche (Abb. 114). Puppe 17-20 mm lang, a SENS Odontopera Stephens = 10. Abdominalsegment (ohne Kremaster) nicht oder wenig länger als das 9. Abdomi- nalsegment, ohne durchgehende Querfurche. Puppentofe ktirzer als ls mm... 91 91 (90) Lateraleinschnitt groß, wenigstens bis zur Mitte des 10. Abdominalsegmentes reichend (Abba) enon Re een 92 = Lateraleinschnitt klein und kurz (Abb. 118) Dors ase RL SNR AT ARE SA NU BE OR 95 92 (91) Kaudal des thorakalen Spiraculums je Sh 39 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 längliche, stark tomentöse Erhebung (Abb. 113). Lateraleinschnitt distal schmal. Kre- master auch im distalen Bereich dorsal skulp- turiert, Sd, entspringen etwa in seiner Mitte ne di Neue Hylaea Hübner = Kaudal des thorakalen Spiraculums ebenfalls je eine längliche tomentöse Erhebung. Late- raleinschnitt distal + breit. Kremaster dorsal auch im distalen Bereich skulpturiert, Sd, entspringeinahe seiner Basis 2... oo Menophra Moore = Kaudal des thorakalen Spiraculum liegt keine oder eine ganz unauffällige und kaum tomen- töse Erhebung (Abb. 111). Kremaster im di- stalen Bereich + glatt. Sd, entspringt meist nicht so nahe der Kremasterbasis ............. 93 93 (92) Labrum mit wenig konvergierenden Seiten (Abb. 117). Puppen fein skulpturiert, relativ stark glänzend. Kremaster im Basalteil schwach gerunzelt ... Puengeleria Rougemont = Labrum trapezformig mit stark konvergieren- den Seiten (Abb. 116). Puppe nur mäßig glänzend, + gröber skulpturiert. Kremaster dorsal an der Basis stark gerunzelt............ 94 94 (93) Puppe 12-20 mm lang. 8. Abdominalseg- mentohne Punktgrübchen.....................2.. POR ER RAI aa ER RIE) Campaea Lamarck = Puppe 8-11 mm lang. 8. Abdominalsegment miestarken Punktgrübchen \..................... EN Cepphis Hübner 95 (91) Sd, entspringen nahe der Basis des Kremas- ters, dieser kurz, stumpf, abgerundet, D, groß (Abb. 19)... u... Lomographa Hübner — Sd, entspringen etwa von der Mitte des Kre- masters, dieser relativ lang, spitz (Abb. 118) a ni ii Cabera Treitschke 96 (88) Sd, entspringen am Kremaster auf der Höhe von D, in der Mitte des Kremasters oder di- stal davon (Abb. 121, 122). Lateraleinschnitt Schmale an N EE 97 - Sd, entspringen am Kremaster mehr frontal als D,, nahe der Basis (Abb. 120). Lateralein- schnitt kurz, breit, stumpf abgerundet .......... RER a, Sciadia Hübner D, mehr als halb so lang wie der Kremaster, dieser kürzer als basal breit. 97 (96) Patotka: Puppen der Spanner Mitteleuropas Dorsalrinne mit 2-3 Kaudalausläufern (Abb. 224). Labium sehr groß (Abb. 225) Perconia Hübner = D, mehr als halb so lang wie der Kremaster, dieser kürzer als basal breit. Dorsalrinne mit mehreren Kaudalausläu- fern (Abb. 122). Labium groß................... ic strie Gnophos Treitschke s. lat. = D, weniger als halb so lang wie der Kre- master, dieser länger als basal breit. Dorsal- rinne mit mehreren Kaudalausläufern (Abb. 121). Labium groß Crocota Hübner An der Basis der Vorderflügel je ein starker Hocker (Abb: 123,125, 130) 7.2... 99 — An der Basis der Vorderflügel kein Höcker (OAN oo EA PE oe eee 101 Proboscis überragt den Kaudalrand des 4. Abdominalsegmentes (Abb. 123). Kremaster mit relativ langen Borsten (Abb. IRA i are ere Rhodometra Meyrick — Proboscis überragt den Kaudalrand des 4. Abdominalsegmentes nicht (Abb. 125, 130). Kremaster mit kurzen Borsten (Abb. 127 ANN I 100 Frons mit langem, zweispitzigem Fortsatz (Abb. 125). Die Borsten am Ende des Kremasters einzeln (Abb. 127) … N RR NO a Timandra Duponchel = Frons ohne Fortsatz (Abb. 130). Die Borsten am Ende des Kremasters in einer Gruppe (Abb. 128) … Cyclophora Hübner Puppe + matt, grün oder leder- bzw. sand- farben, dann mit dunkler Sprenkelung oder anderer Zeichnung bzw. schwarzen Spiracula. Exuvie grünweiß, gelbweiß oder trüb)bräunlichgelb 227% 102 = Puppe + glinzend, gelb-, rot- bis schwarz- braun, zeichnungslos. en 113 102 (101) Frontolaterale Lappen des Metanotums spitax(Abb4 53). retina 128 = Frontolaterale Lappen des Metanotums abgerundet (Abbr 199)... 165 103 (165) Adern der Vorderfliigel stark hervortre- TEN dla ek nnen ENNE 76 = Adern der Vorderflügel nicht stark hervor- PÉTER RC ree 104 100 (99) 101 (98) Abb. 115-145. — 115, 117, Puengeleria capreolaria, 116, Campaea honoraria; 118, Cabera pusaria; 119, Lomographa temera- ta, 120, Sciadia tenebraria, 121, Crocota nivearia, 122, Gnophos dilucidaria, 123, 124, Rhodometra sacraria, 125-127, Timandra griseata, 128, 130, Cyclophora linearia, 129, Ourapteryx sambucaria; 131, Eulithis populata, 132, 133, Ecliptopera silaceata; 134, Electrophaes corylata; 135, Dysstroma truncata; 136, Cidaria fulvata, 137, Thera variata; 138, Rheumaptera cer- vinalis 139, 140, R undulata; 141, Philereme transversata, 142, Triphosa dubitata, 143, Pterapherapteryx sexalata, 144, Xanthorhoe designata, 145, Epirrhoe alternata. 115-118, 137, Abdominalende in Lateralsicht; 116, 117, 133, 135, Labrum, Labium; 119-122, 124, 127-129, 131, 132, 134, 136, 143, Abdominalende in Dorsalsicht; 123, 125, 130, Habitusbild in Ventralsicht; 126, Kopffortsatz in Lateralsicht (Umriss); 138, 140, 141, 142, Basis des 5. Abdominalsegmentes in Lateralsicht; 139, Basis des 5. Abdominalsegmentes, Dorsalsicht; 144, 145, Auge und Umgebung. 41 T1JDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 104 (103) D, am Kremaster an einem kurzen Stiel, eine Winkel von fast 90° bildend (Abb. 129) APuppe über ld3mmilang................. N it, Ourapteryx Leach — D, am Kremaster nicht oder kaum gestielt. Puppe unter 16 mm lang … 105 105 (104) Kremaster mit 4 Paaren von Borsten (Abb. NE) BE OEREN EEE 106 = Kremaster mit 6 Paaren von Borsten (Abb. LE CNET) RE N A ce 111 106 (105) 10. Abdominalsegment dorsal stark ver- kürzt und + längsgefurcht oder längsge- rippt (Abb. 131). Puppen über 12, oft über lSamimnblang in Eulithis Hübner = 10. Abdominalsegment stark verkürzt, je- doch nicht längsgefurcht. Puppenlänge unter 12 mm ....... Xanthorhoe Hübner p.p (X. munitata (Hübner)) = 10. Abdominalsegment nicht stark ver- kürzt, kaum längsgefurcht (Abb. 132)....... dass nRT 107 107 (106) Labium nicht sichtbar oder ganz klein (ALDE 32209) Men 108 — Labium relativ groß (Abb. 135) ......... 110 108 (107) Dorsalrinne ohne Kaudalausläufer (Abb. SV) ae ER Ecliptopera Warren = Dorsalrinne mit Kaudalausläufern, wenig- stens der mittlere gut sichtbar (Abb. 134, DI) OD oi ap Ines ch 109 109 (108) Dorsalrinne mit deutlichen spitzen tomen- tösen Ausläufern (Abb. 134). Puppe deut- lich dunkel gefleckt … Electrophaes Prout = Dorsalrinne mit undeutlichen Ausläufern, nur der mittlere etwas mehr ausgeprigt (Abb. 208). Puppe undeutlich oder nicht eRe Glew. wel. mk el 156 110 (107) Puppe dünnschalig, nicht gesprenkelt, nicht bereift, Exuvie matt gelblich ............ ie LT J 8 rique Dysstroma Hübner = Puppe dickschalig, dunkel gesprenkelt, deutlich bereift, Exuvie hellbraun ............. ne RE EEE FAN Chloroclysta Hübner = Puppe dickschalig, braun mit Zeichnun- gen, insbesondere an den Adern der Vor- derfliigel, nicht bereift, nur 7-8 mm lang REA xi ie ARCS re EE EES EME ee 150 Patocka: Puppen der Spanner Mitteleuropas 111 (105) Dorsalrinne und Lateraleinschnitt nicht sichtbar (Abb. 136) ..... Cidaria Treitschke = Dorsalrinne und Lateraleinschnitt + sicht- bari(Abbssl3 EE Me EE 112 112 (111) Borsten am Kremaster satt braunrot, Exu- vie weißlichgelb. Dorsalrinne schwach, meist nur mit einzelnen, kleinen Kaudal- auslaufen mn Plemyria Hübner = Borsten am Kremaster bräunlich. Exuvie bräunlich weiß oder gelb. Dorsalrinne mit mehreren meist deutlichen Kaudalausläu- fermi VAD Dw Se Thera Stephens 113 (101) An der Basis des 5. Abdominalsegmentes, dorsal oder lateral, dunkle Leisten, die Querrinnen oder -vertiefungen begrenzen (Abb: IS SM) PR en 114 = An der Basis des 5. Abdominalsegmentes keine Leisten, Querrinnen oder -vertiefun- gen, höchstens dichte Punktgrübchen (Abbs22l0) Hera 116 114 (113)An der Dorsalseite der Basis des 5. Abdominalsegmentes Querleisten und da- zwischen eine Querrinne. Sie laufen lateral + in die Nähe der Spiracula herab (Abb. 133-240) Rheumaptera Hübner = Nur lateral an der Basis des 5. Abdominal- segmentes Querleisten und Vertiefungen (ABBIA) PR nn ne 115 115 (114) 1-2 laterale Vertiefungen am 5. Abdomi- nalsegment ungeteilt (Abb 141)................ NE STERN Henn Philereme Hübner = Laterale Vertiefung an der Basis des 5. Ab- dominalsegmentes in mehrere Zellen un- terteilt (Abb. 142)........ Triphosa Stephens 116 (113) Am Kremaster - außer den Borsten D,, die starke Endhäkchen, Enddornen bzw. eine gestielte Endgabel bilden - gibt es nur ein Paar kleiner Borsten (L, oder D,) (Abb. 143 MIL ONS D) er By, ~ Am Kremaster, außer D, noch 2-3 Paaren von Borsten (Abb. 167, 183, 201, 218) … te nee une: 123 117 (116) Kremaster mit einem Paar von lateralen Höckern. D, zusammen gestielt, stark di- vergierendi(Abb: 9) ent nn STE ee erahnen Pterapherapteryx Curtis Abb. 146-178. — 146, Catarhoe cucullata, 147, C. rubidata; 148, Epirrhoe alternata; 149, 150, Nycterosea obstipata; 151, 152, Camptogramma bilineata; 153, 154, Scopula marginepunctata; 155, Idaea aureolaria, 156, I. rufaria; 157, I. deversaria; 158, 159, Rhodostrophia vibicaria, 160, 161, Baptria tibiale, 162, Glacies alticolaria, 163, Gnophos unicoloraria, 164, Colotois pen- naria; 165, 167, Nothacasis sertata, 166, 171-173, Anticollix sparsata, 168, Acasis appensata, 169, Trichopteryx polycommata; 170, 7. carpinata; 174-176, Anticles badiata, 177, Discoloxia blomeri, 178, Venusia cambrica. 146, 148, 150, 152, 154, 157, 159, 161-164, 167, 168, 171, 172, 174, 177, 178, Abdominalende in Dorsalsicht; 147, Dorsalrinne; 149, 151, 153, Pronotum, 1. Abdominalsegment; 158, Erhebung kaudal von dem thorakalen Spiraculum; 160, Metanotum, 1. Abdominalsegment; 165, Kopf und Beine in Ventralsicht; 166, Vorderschenkel; 169, 170, 175, Abdominalende in Lateralsicht; 176, Punktgrübchen an der Basis des 2. Abdominalsegmentes. 43 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 = Kremaster ohne Lateralhöcker. Enddornen schlanker, nicht so lang gestielt, weniger divergierend (Abb. 146, 150)............. 118 118 (117) Grenzen zwischen Vorderbeine und An- tennen viel länger als die Grenzen zwi- schen Vorderbeine und Augen. Grenzen zwischen Antennen und Augen kaum län- ger als die zwischen Antennen und Vorder- beine (Abbi ee. 119 = Grenzen zwischen Vorderbeine und Antennen nicht oder wenig länger als die Grenzen zwischen Vorderbeine und Au- gen. Grenzen zwischen Antennen und Augen deutlich länger als die zwischen Antennen und Vorderbeine (Abb. 145) .... kde 120 119 (118) Am Kremaster außer D, noch L, vorhan- den, nahe bei D, (vgl. Abb. 148) … … … … … tab. aes Xanthorhoe Hübner - Am Kremaster außer D, noch D,, von D, ENTIELN GS AT OLIO 120 120 (118) Dorsalrinne nur mit einem großen Kau- dal- und Frontalausläufer in der Mitte (ADD RAGA) Catarhoe Herbulot = Dorsalrinne mit mehreren Kaudalausläu- ferm\(Abb®l4S sli OMS 2) 121 121 (120) D, so lang wie oder länger als der Kremas- ten (Ab 48) nen Epirrhoe Hübner - D, kürzer als der Kremaster, dieser + länger als basal breit (Abb. 150, 152) ........... 122 122 (121) Frontolaterale Lappen des Metanotums spitz (Abb. 149). Dorsalrinne im Mittelteil wenig verbreitert (Abb. 150)..................... EN ROTTE da PAT Nycterosea Hulst _ Frontolaterale Lappen des Metanotums abgerundet. Dorsalrinne im Mittelteil stark verbreitert (Abb. 152)....................... EEE ER Camptogramma Stephens 123 (116) Frontolaterale Lappen des Metanotums spitzwinkelig (Abb. 153, 160). (Wenn je- doch das 1. Abdominalsegment mit deutli- chen Punktgrübchen versehen und die Skulptur relativ grob ist, suche weiter un- ters 12 boues EB zun 124 SS Frontolaterale Lappen des Metanotums stumpf, abgerundet (Abb. 151, 181).. 129 124 (123) D, gleichlang und selten stàrker als die übrigen Borsten des Kremasters (Abb. SID ZI EEE EEE 125 D, deutlich länger als die übrigen Borsten des Kremasters (Abb. 154, 159, 161) … … EE A e ie AE a D 126 125 (124) Dorsalrinne nur mit einem mittelgroßen Vorsprung in der Mitte. Labium klein (Abb. 193). Lateraleinschnitt kurz, abge- ründet....:....... NIUE i 149 = Dorsalrinne mit kleinen Vorsprüngen, der mittlere aber sehr lang, bis nahe zur Basis des Kremasters reichend. (Labium an der untersuchten Puppe nicht vorhanden) ...... sana MER Glossotrophia Prout = Dorsalrinne mit mehreren Vorsprüngen, keiner von ihnen nähert sich der Basis des Kremasters (Abb. 155, 157). Wenn nur ei- ner dorsal in der Mitte oder gar keiner vor- handen ist, dann fehlt das Labium und/- oder der Lateraleinschnitt sehr groß .......... hes, bate en Idaea Treitschke 126 (124) Kremaster rauh skulpturiert, D, entsprin- gen weit voneinander entfernt. Vorder- schenkeliklein® So 97 - Kremaster + fein skulpturiert. D, entsprin- gen nahe beieinander, Vorderschenkel gross late inne RR 127 127 (126) Sd, entspringen nahe der Basis des Kre- masters (Abb. 161). Labium klein (vgl. Abbi 192) 232 Baptria Hübner = Sd, (wenn vorhanden) entspringen weit von der Basis des Kremasters entfernt (Abb. 154, 159). Labium meist groß (vgl. Abb:135).:..:22..22 12190) 128 128 (127) Kaudal des thorakalen Spiraculums je eine große, elliptische, stark tomentöse Erhe- bung (Abb. 158). Puppe sandfarben oder PIEDE: son menen Rhodostrophia Hübner - Kaudal des thorakalen Spiraculums eine längliche, weniger auffällige Erhebung (vgl. Abb. 113). Puppe rotbraun, glänzend reed Ee Scopula Schrank 129 (123) Kremaster mit 4 Paaren steifer, gleich gro- Ser Borsten, am Ende nicht hakenartig. D, entspringen weit voneinander entfernt (Abb. 162). Punktgrübchen fehlen… … … … RT vanen, BEE Glacies Milliere, 1874 = Wenigstens ein Teil der Borsten am Kremaster hakenartig. Punktgrübchen am Abdomen + vorhanden … … 130 130 (129) Vorderschenkel klein, oft recht schmal. Labium klein oder fehlend (Abb. 165). D, am Kremaster länger und stärker als die übrigen Borsten, + divergierend (Abb. 1672168: A EA) ne 131 = Vorderschenkel, Labium, oder beide mit- telgroß bis groß (Abb. 1), sonst D, am Kremaster nicht viel länger oder stärker als die übrigen Borsten. (Abb. 193)......... 135 131 (130) Dorsalrinne in der Mitte beiderseits, insbe- sondere kaudalwärts, mit abgerundeten Ausschnitt. Borsten am Enddrittel des Kremasters konzentriert (Abb. 174). Punktgrübchen an der Basis des 1.-3. Ab- dominalsegmentes stark vergrößert (Abb. Moers. Anticlea Stephens pp. (A. badiata Denis et Schiffermiiller) = Dorsalrinne mit einem oder mehreren fla- chen Kaudalen Ausschnitten. Borsten auch an der Basalhälfte des Kremasters (Abb. 167, 168, 171). Punktgrübchen an der Basis des 1. -3. Abdominalsegmentes nicht VO Er 132 132 (131) Metanotum mit Punktgrübchen (Abb, EI AAU ON AAE 133 — Metanotum ohne Punktgrübchen (Abb. US tre RR NEE ER SEE 134 133 (132)Kremaster in Dorsalsicht länger als basal breit. Borsten Sd, entspringen nahe der Basis des Kremasters (Abb. 167) … … … …. bode NN, Nothacasis Prout — Kremaster in Dorsalsicht kürzer als basal breit. Sd, entspringen etwa in der Mitte des Kremasters (Abb. 168) … … … REN eN EE bte Acasis Duponchel 134 (132) Puppe ohne oder mit sehr kleinem La- bium. Lateraleinschnitt undeutlich oder in der Form eines Grübchens (Abb. 169, AOC a Trichopteryx Hübner — Puppe mit deutlicherem mittelkleinem Labium. Lateraleinschnitt rinnenförmig, mit einem kurzen Kaudalausläufer (Abb. ND) ER Anticollix Prout 135 (130) Kremaster kürzer als basal breit, sehr grob skulpturiert (gefurcht, gerippt). D, haken- förmig, kaum divergierend, stärker und länger als die übrigen Borsten, basal + von- einander entfernt, der Kremaster dazwi- schen + konkav (Abb. 163). Epicranial- naht zwischen Vertex und Frons nicht sichtbar (Abb. 4). Puppen über 11 mm — Kremaster kürzer als breit, stumpf, grob skulpturiert. D, stärker als die übrigen Borsten, basal nahe beieinander, ohne Ein- schnitt dazwischen, ziemlich parallel lau- fend. Epicranialnaht sichtbar. Puppenlän- pesi Dimmi. Epilobophora Inoue = Kremaster oft langer als breit, nur schwach oder mäßig skulpturiert. D, entspringen entweder nahe beieinander und divergie- ren, oder sind nicht größer als die übrigen Borsten des Kremasters. Puppen manch- mal auch kürzer als 10 mm. Epicranialnaht relativ deutlich (Abb. 3) … … … … … … … 136 136 (135) Kremaster mit 3 Paar Borsten (Abb. 178- ISO) ee A a ca 137 _ Kremaster mit 4 Paar Borsten (Abb. 190, DOME DANS) ARE UM ME EU 144 137 (136) Lateraleinschnitt nicht entwickelt, oder als einfache Rinne ohne einen größeren Patocka: Puppen der Spanner Mitteleuropas Kaudalausläufer (Abb. 175, 182) ....... 138 = Lateraleinschnitt deutlich entwickelt, mit einem starken Kaudalausläufer (Abb. 184) ER Ae stelo RA 142 138 (137) Borsten am Kremaster gleich groß (Abb. 7D) OS oe Des Gymnoscelis Mabille - D, viel größer als die übrigen Borsten am Kremaster (Abb. 180) ........................ 139 139 (138) Punktgrübchen an der Basis des 1.-4. Ab- dominalsegment stark vergrößert und ver- tiere (Ab DC) En Ri 140 = Punktgrübchen an der Basis des 1-4. Abdominalsegmentes nicht vergrößert (ABB AS DES OUEN MII) NE 141 140 (139) Kremaster am Ende abgerundet, schaufel- förmig (Abb. 177). Puppe 7-8 mm lang... RENE EUER De GIA Discoloxia Warren = Kremaster am Ende spitz (Abb. 178). Puppe 8-9 mm lang........... Venusia Curtis 141 (139) Metanotum ohne Punktgrübchen (vgl. Abb 194) na Euchoeca Hübner = Metanotum mit Punktgriibchen (Abb. LEDER Hydrelia Hübner 142 (137) Kremaster in Dorsalsicht nicht länger als basal breit (Abb. 183)..... Euphyia Hübner = Kremaster in Dorsalsicht länger als basal breit Abb 185) eee AR ER 143 143 (142) Dorsalrinne nur mit drei Kaudalausläufern (Abb) Mesoleuca Hübner = Dorsalrinne mit mehreren Kaudalausläu- fern (ABBS TROIE een 146 144 (136) Puppenlänge über 11 mm. Punktgrübchen fehlen am Metanotum und meist auch am 1. Abdominalsegment (Abb. 187, 219); falls vorhanden, dann Kremaster recht lang und die Borsten am 9. Abdominalsegment deutlich (Abb. 186) ........................... 145 - Metanotum oder wenigstens das 1. Abdo- minalsegment mit deutlichen Punktgrüb- chen, oder Puppen unter 10 mm lang....... 145 (144) D, am Kremaster entspringt weit ventral von Sd (Abb#186,220)e em 146 = D, entspringt etwa auf derselben Höhe wie SAAB ASS MI SI) Pre a 147 146 (145) Kremaster lang zungenförmig (Abb. 186). Borsten am 9. Abdominalsegment relativ kräftig (Abb. 186) ...... Scotopteryx Hübner = Kremaster in Dorsalsicht dreieckig, nicht länger als basal breit (Abb. 220) ................ NERA I AI Pareulype Herbulot 147 (145) Grenzen zwischen Vorderbeinen und An- tennen viel länger als zwischen Vorderbei- nen und Augen; Grenzen zwischen Anten- nen und Augen kaum länger als zwischen Antennen und Vorderbeinen (Abb. 144) .. 45 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 46 nai Xanthorhoe Hübner pp: (X montanata (Denis et Schiffermüller), X. incursata (Hübner)) = Grenzen zwischen Vorderbeinen und Antennen nicht oder kaum länger als zwi- schen Vorderbeinen und Augen; Grenzen zwischen Antennen und Augen viel länger als zwischen Antennen und Vorderbeinen (Abbate Entephria Hübner 148 (144) Grenzen zwischen Vorderbeinen und Antennen deutlich länger als zwischen Vorderbeinen und Augen (etwa gleich lang wie die Grenzen mit Augen und Genae zu- sammen). Grenzen zwischen Antennen und Augen etwa gleich lang wie zwischen Antennen und Vorderbeine (Abb. 144, 191). Kleine Arten, unter 9, meist unter 8 mm Länge, Skulptur oft grob, D, am Kremaster hakenförmig, nicht oder nur wenig divergierend. (Abb. 196-201) Lateraleinschnitt und Labium meist gut entwickelte N... 149 = Grenzen zwischen Vorderbeinen und Antennen nicht oder kaum länger als zwi- schen Vorderbeinen und Augen; Grenzen zwischen Antennen und Augen deutlich länger als zwischen Antennen und Vorder- beinen (Abb. 145). Wenn die Puppen un- ter 8 mm lang, fehlt oft das Labium und der Lateraleinschnitt. D, oft gabelartig di- vergierend (Abb. 203-205, 220)......... 151 149 (148) Labium sehr klein (Abb. 192). Dorsalrinne mit einem einzigen, kleinen Kaudalausläu- fer (Abb. 193). Borsten am Kremaster + gleich groß, nur im distalen Drittel (Abb. 193). Frontolaterale Lappen des Metano- tums ziemlich spitz (Abb. 194) ................. TEEN Calliclystis Dietze = Labium größer (Abb. 1), Dorsalrinne mit mehreren Ausläufern (Abb. 196, 199, 200), oder D, deutlich größer als die übri- gen Borsten des Kremasters (Abb. 198). Frontolaterale Lappen des Metanotums + abeerundet, (Abb:210) tronen 150 150 (149) Metanotum ohne, 1. Abdominalsegment ohne oder mit schwach angedeuteten Punktgrübchen (Abb.195). Dorsalrinne Patocka: Puppen der Spanner Mitteleuropas mit kleinen Kaudalausläufern, Borsten am Kremaster fast gleich groß, nur in der di- stalen Hälfte (Abb. 196) … … ………………. en ee Chloroclystis Hübner Skulptur + grob, 1. Abdominalsegment und oft auch das Metanotum (bzw. ganze Thorax dorsal) mit deutlichen Punktgrüb- chen. Sonst meist entweder die Kaudal- ausläufer der Dorsalrinne groß (Abb. 201) oder D, länger als die übrigen Borsten des Kremasters (Abb. 197, 198, 200) bzw. Borsten auch in der Basalhälfte des Kre- master Eupithecia Curtis 151 (148) D, am Kremaster deutlich, weit voneinan- der entfernt, einem stumpfen Winkel bil- dend (Abb. 190). Metanotum mit Punkt- grübchen (wel -Abb32 10)... N A NE Anticlea Stephens p. p. (Anticlea erivata (Denis et Schiffermüller)) D, entspringen nahe beieinander (Abb. DANG D8) kk OS a, 152 152 (151) D, in Dorsalsicht länger als der Kremaster, dornartig, divergierend und + lang gestielt. Oft kleinere Arten, unter 9 mm Länge (Abb. 203-205) ............ Perizoma Hübner D, in Dorsalsicht nicht länger als der Kremaster, oft hakenartig und + parallel laufend (Abb. 213, 216, 223), wenn diver- gierend (Abb. 218, 220), dann oft größere Arten, uber: 9 mmilang str te Me... 153 153 (152) Labium fehlt, oder sehr klein (vgl. Abb. 87, 209). Dorsalrinne und Lateralein- schnitt + schwach ausgebildet. D, hakenar- tig, viel stärker als die übrigen Borsten (Abb. 207, 208, 211). Oft kleinere Arten, 6-8:mmlange ee Be 154 Labium mittelgroß, bis groß (Abb. 1, 135). Dorsalrinne und Lateraleinschnitt meist deutlich und tief. D, stärker oder auch gleichstark wie die übrigen Kremas- térborstent rn ee 157 154 (153) Metanotum mit Punktgrübchen (vgl. Abb. 210). Puppe nur etwa 6-7 lang........... 155 Metanotum ohne Punktgrübchen (vgl. Abb. 195). Puppe oft länger als 7 mm ...... en 156 155 (154) D, am Kremaster viel näher an L, als an Sd, Abb. 179-211. — 179, Gymnoscelis rufifasciata, 180, 181, Hydrelia flameolaria, 182, Euchoeca nebulata, 183, Eyphyia, 185, Mesaleuca albicillata, 186, 187, Scotopteryx moeniata, 188, Xanthorhoe montanata; 189, Entephria caesiata; 190 Anticlea deri- vata; 191-194, Calliclystis v-ata, 195, 196, Chloroclystis rectangulata; 197, Eupithecia absinthiata; 198, E. pulchellata, 199, E. tenuiata; 200, E. abietaria; 201, E. pusillata, 202, Perizoma alchemillata; 203, P. bifasciata; 204, P. didymata; 205, P. peralle- lolineata; 206, P. blandiata; 207, P. incultaria; 208, 209, Eustroma reticulata; 210, 211, Asthena albulata. 179, 180, 183, 185, 186, 188-190, 193, 196-208, 211, Abdominalende in Dorsalsicht; 181, 187, 194, 195, 210, Metanotum, 1. bzw. 2. Abdominalsegment; 182, 184, Abdominalende in Lateralsicht; 191, Augen und Bereich; 192, 209, Labrum, Labium. 47 TiJDSCHRIET VOOR ENTOMOLOGIE, VOLUME 137, 1994 Abb. 212-225. — 212, Minoa murinata; 213, Cosmorhoe ocellata; 214, Nebula salicata; 215, N. nebulata; 216, Colostygia apta- ta, 217, Hydriomena ruberata, 218, Lampropteryx otregiata; 219-221, Pareulype berberatæ 222, Spargania luctuata, 223, Melanthia procellata; 224, 225, Perconia strigillaria, 226, 227, Odontognophos dumetata. 212-218, 220, 222-224, Abdominalende in Dorsalsicht; 219, Metanotum, 1-2. Abdominalsegment; 221, 226, Basis der 5. Abdominalsegmentes in Lateralsicht; 225, Labrium, Labium; 227, Hécker kaudal von dem thorakalen Spiraculum. 48 entspringend (Abb. 212).........................- A Minoa Treitschke - D, am Kremaster etwa in der Mitte zwi- schen Sd, und L, entspringend (Abb. 211) REATO MOTTA] LIA Asthena Hübner 156 (154) D, am Kremaster entspringen frontal von Sd, (Abb. 207)....... Perizoma Hübner p.p. BE ne. (P. incultaria Herrich-Schäffer) = D, am Kremaster entspringen kaudal von Sdm(Abb. 208). 22082... Eustroma Hübner 157 (153) Borsten am Kremaster fast gleich lang und gleich dick, schlank. Sd, entspringt nahe der Basis des Kremasters (Abb. 213) ......... EN ARE WE RR Cosmorhoe Hübner — Borsten am Kremaster nicht gleich lang und gleich groß (D, deutlich größer, Abb. 216, 218). Wenn doch gleich groß, dann entspringt Sd, + auf der Höhe von D,, weit von der Basis des Kremasters entfernt (ASS DIT TO anna 158 158 (157) D, entspringt am Kremaster frontal von SCH Abb2 NR... Hydriomena Hübner - D, entspringt + auf der Höhe von Sd, (AN) ova TA TO) na 159 — D, entspringt kaudal von Sd,. (Abb. 220, DOD DDS) ARE EET 161 159 (158) D, nicht, oder nur wenig länger als die übrigen Borsten (Abb. 214, 216) … … 160 = D, deutlich länger und stärker als die übri- gemsBorsten des Kremästers.........le.ecesen MARE OM CE eu Orthonama Hübner 160 (159) D, kaum länger, manchmal etwas stärker als die übrigen Borsten am Kremaster, die- ser kaudal + stumpf (Abb. 214, 215) ........ E ENE ES Nebula Bryard - D, deutlich länger und stärker als die übri- gen Borsten des Kremasters, dieser kaudal zugespitzt (Abb. 216) .. Calostygia Hübner 161 (158) Metanotum und 1. Abdominalsegment ohne Punktgrübchen (Abb. 219) … … 146 = Metanotum und 1. Abdominalsegment mit Punktgrübchen (Abb. 210).......... 162 162 (161) Kremaster in Dorsalsicht länger als basal breit, D, oft dornartig, divergierend (Abb. DNS) RCE Lampropteryx Stephens = Kremaster in Dorsalsicht nicht länger als basal breit, D, hakenartig (Abb. 220, 223) REEL 163 163 (162) Kremaster in Dorsalsicht kürzer als basal bret) hakenartıe (Abb 223) ii RT Melathia Duponchel = Kremaster in Dorsalsicht nicht oder kaum kiirzer als basal breit, D, mitunter dornar- WE? (ON oes a 970 2 EEE EN 164 164 (163) Dorsalrinne mit deutlichen Kaudalaus- läufern, auch oft ein Frontalausläufer in Patocka: Puppen der Spanner Mitteleuropas der Mitte vorhanden (Abb. 6). D, + ha- kentormig en. Horisme Hübner = Dorsalrinne mit relativ kleinen Kaudal- und keinem Frontalausläufer. D, am Kremaster eher dornartig, gebogen (Abb. DADA SERRE Spargania Guéneé, 1857 165 (102) Proboscis kürzer als die Vorderbeine. Vorderflügel grenzen kurz aneinander… … RE ANSE RER RAT Lythria Hübner = Proboscis länger als die Vorderbeine. Vorderflügel grenzen nicht aneinander...... DISKUSSION In der vorliegenden Arbeit wurde auf eine Bestim- mungstabelle der in den derzeitigen Systemen üblichen Unterfamilien der Geometridae, sowie der Tribus nach Herbulot (1962-1963) verzichtet. Die bisherigen Tabellen zur Bestimmung der Unterfamilien von Khotko (1977), sowie von McGuffin (1967-1981) kann man für größere oder kleinere Zahl der Gattun- gen, jedoch nicht für alle erfolgreich benutzen. Die Tabelle von Khotko ist allzu kompliziert - zur Bestim- mung von fünf Unterfamilien sind dort mehr als zwei Textseiten nötig. Die Tabelle von McGuffin ist einfa- cher, es gibt jedoch bei der Bestimmung mancher Puppen der mitteleuropäischen Fauna Schwierig- keiten, besonders was die Unterscheidung der Un- terfamilien Sterrhinae und Larentiinae betrifft. Noch schwieriger ist dies bei den Tribus im Sinne von Herbulot. Nur für einen Teil von ihnen gelang es mir anhand der Puppenmerkmale eine benutzbare Diagnose aufzustellen und auch dort ist diese nicht für alle Gattungen nach dem Herbulot'schen System gül- tig. Unter die von puppenmorphologischer Hinsicht uneinheitlichen Tribus gehören z. B. die Pseudoter- pnini (Unterfamilie Geometrinae) und die Lythriini (Unterfamilie Larentiinae). Mehrere Gattungen soll- ten anhand der Puppenmorphologie in andere Tribus gehören als in welche sie von Herbulot gestellt wur- den, wie z. B. : Calospilos, Lomaspilis, Lygdia und Stegania (siehe Patoc’ka 1986a), oder Odontognophos, Cleorodes, Crocota, Synopsia und Menophra (siehe Patoc’ka 1993). Die Gattungen des Herbulot'schen Systems ent- sprechen dagegen in der Mehrzahl auch auf Grund der Puppenmerkmale sehr gut. Auf wenige Aus- nahmen soll hier aufmerksam gemacht werden: Die Art obstipata weicht von vittata stark ab, es wird für sie deshalb der Gattungsname Nycterosea (im Sinne von Forster-Wohlfart 1981) benützt. Ebenso wurde es mit den Gattungen Dysstroma, Chiasmia, Diastictis und Phigalia gemacht. Außerdem ist auch die Gattung Xanthorhoe puppenmorphologisch unein- heitlich. Insbesondere die Art X. munitata entspricht 49 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 viel besser der gattung Zulithis. Die Arten Xanthorhoe montanata und incursata weichen von den übrigen stärker ab. Ähnlich ist es auch mit der Gattung Perizoma, wo besonders: die Art incultaria den übri- gen puppenmorphologisch unähnlich ist. Stark von- einander abweichende Arten enthält auch die Gattung Anticlea. Dagegen sind die größeren oder großen Gattungen Cyclophora, Sterrha, Idaea und Eupithecia hinsichtlich der Puppenmorphologie rela- tiv einheitlich gebaut. Die Verhältnisse in der Gattung Gnophos s. lat. (vgl. Sauter 1990) werden in einer besonderen Arbeit (Patocka im Druck) bespro- chen. LITERATUR Forster, W. & T. A. Wohlfahrt, 1981. Die Schmetterlinge Mitteleuropas 5, Spanner (Geometridae). — Franckh’sche Verlagshandlung, Stuttgart: 1-311. Gustafsson, B. (ed), 1987. Catalogus Lepidopterorum Sueciae. — Riksmuseet, Stockholm: 1-140. Herbulot, C., 1962-1963. Mise à jour de la liste des Geometridae de France. — Alexanor, 2: 117-124, 147- 154, 3: 17-24, 85-93. Khotko, E. I., 1977. A key to the spanpupae (Lepidoptera, Geometridae). — Academia Scientiarum Beloruss. SSR, Minsk: 1-80. Koehler, W., 1937. Beitrag zur Kenntnis der, unter Streudecke der Kiefernbestande überwinternden Schmetterlingspuppen. — Institut des recherches des forêts dominiales Pologne, Traxaux et comptes rendus Warszawa, Ser. A. Nr. 29: 1-81 Leraut, P. 1980. Liste systématique et synonymique des Lépidoptères de France, Belgique et Corse. — Alexanor, Supplément: 1-334. Ljungdahl, D., 1919. Nägra puppenskrivningar. — Entomo- logisk Tidskrift 40: 97-119. McGuffin, W. C., 1967-1981. Guide to the Geometridae of Canada. — Memoirs of the Entomological Society of Canada 50: 1-103, 86: 1-159, 101: 1-191, 117: 1-153. Mosher, E., 1916. A classification of the Lepidoptera based on characters of the pupa. — Bulletin of the Illinois State Laboratory of Natural History 12: 1-159. Nordström, F., E. Wahlgren & A. Tulgren, 1941. Svenska Fjärilar. — Nordisk Familjeboks Förlags Aktiebolag Stockholm: 1-354. Patocka, J., 1978. Zur Puppenmorphologie und -Taxono- mie der Unterfamilie Ennominae insbesondere der Tri- bus Bistonini (Lepidoptera, Geometridae). — Vestnik C- eskoslovenské Spoleénosti Zoologické 42: 143-151. Patocka, J., 1980. Die Raupen und Puppen der Eichen- schmetterlinge Mitteleuropas. — Monografiën zur ange- wandten Entomologie, Verl. Paul Parey, Hamburg & Berlin: 1-188. Patocka, J. 1980-1983. Beitrag zur Puppentaxonomie der mitteleuropäischen Larentiinae (Lepidoptera, Geometri- dae). — Biolögia, Bratislava, 35: 97-109, 36: 583-593, 37: 559-570, 38: 117-131. Patocka, J., 1985. Beitrag zur Kenntnis der Puppen der Unterfamilie Ennominae (Lepidoptera, Geometridae). — Biolögia, Bratislava, 40: 997-1012. Patocka, J., 1986a. Zur Kenntnis der Puppen der mitteleu- 50 ropäischen Spanner aus der Tribus Abraxini und Semio- thisini (Lepidoptera, Geometridae). — Biolögia, Bratis- lava, 41: 579-595. Patoëka, J., 1986b. Zur Kenntnis der Puppen der Tribus Boarmiini (Lepidoptera, Geometridae) in Mitteleuropa. — Acta Entomologica Bohemoslovaca 83: 301-315. Patoëka, J., 1992. Über einige Puppen der Spanner aus der Tribus Ennomini (Lepidoptera, Geometridae, Ennomi- nae). — Entomologische Berichten, Amsterdam 52: 171- 176. Patoëka, J., 1993. Über einige Puppen der Spanner aus der Tribus Boarmini (Lepidoptera, Geometridae, Enno- minae). — Entomologische Berichten, Amsterdam 53: 114-120. Patoëka, J., 1994. Die Puppen der mittel- und westeuropäi- schen Spanner aus der Tribus Asthenini (Lepidoptera, Geometridae, Larentinae). — Entomologische Berichten, Amsterdam 54: 13-19. Patoëka, J., im Druck. Die Puppen der Gnophos-Gruppe (Lepidoptera, Geometridae) Mitteleuropas. — Zeitschrift der Arbeitsgemeinschaft österreichischen Entomologen: im Druck. 3 Patoëka, J., J. Burgan, M. Capek & M. Stolina, 1960. Die Tannenschmetterlinge der Slowakei. — Ed. SAV Bratis- lava: 1-214. Sauter, W., 1990. Zur Systematik der Gnophos-Gruppe (Lepidoptera, Geometridae). — Nota Lepidopterologica 12: 328-343. Speyer, W., 1958. Lepidopteren-Puppen an Obstgewächsen und in ihrer näheren Umgebung. — Mitteilungen aus der Biologischen Bundesanstalt für Land- und Forstwirt- schaft 93: 1-40. Received: 23 April 1993 Revised manuscript accepted: 10 April 1994 ANHANG Übersicht der Gattungen und Arten der Familie Geome- tridae, deren Puppen als Unterlage für die Aufstellung dieser Gattungstabelle untersucht wurden. Archiearis Hübner, 1823 parthenias (Linnaeus, 1761) (Slowakei, Holland) notha (Hübner, 1803) (Slowakei) puella (Esper, 1787) (Slowakei) Alsophila Hübner, 1825 aescularia (Denis & Schiffermüller], 1775) (Slowakei) quadripunctaria (Esper, 1787) (Slowakei) Aplasta Hübner, 1823 ononaria (Fuessly, 1783) (Deutschland) Pseudoterpna Hübner, 1823 pruinata (Hufnagel, 1767) (Slowakei) Geometra Linnaeus, 1758 papilionaria (Linnaeus, 1758) (Böhmen) Comibaena Hübner, 1823 bajularia ({Denis & Schiffermüller], 1775) (Slowakei) Thetidia Boisduval, 1840 smaragdaria (Fabricius, 1787) (Slowakei) Hemithea Duponchel, 1829 aestivaria (Hübner, 1799) (Slowakei, Böhmen) Chlorissa Stephens, 1831 viridata (Linnaeus, 1758) (Polen) cloraria (Hübner, 1813) (Slowakei) etruscaria (Zeller, 1848) (Slowakei) Thalera Hübner, 1823 5 fimbrialis (Scopoli, 1763) (Osterreich) Hemistola Warren, 1893 biliosata (Villers, 1789) (Slowakei) Jodis Hübner, 1823 lactearia (Linnaeus, 1758) (Slowakei) putata (Linnaeus, 1758) (Böhmen, Osterreich) Cyclophora Hübner, 1822 pendularia (Clerck, 1759) (Slowakei, Böhmen) albiocelaria (Hübner, 1789) (Slowakei) annulata (Schulze, 1775) (Slowakei) albipunctata (Hufnagel, 1767) (Slowakei, Böhmen) puppillaria (Hübner, 1799) (Frankreich) ruficiliaria (Herrich-Schäffer, 1855) (Slowakei) porata (Linnaeus, 1767) (Österreich) quercimontaria (Bastelberger, 1897) (Slowakei) punctaria (Linnaeus, 1758) (Slowakei) suppunctaria (Zeller, 1847) (Slowakei) linearia (Hübner, 1799) (Slowakei) Timandra Duponchel, 1829 griseata W. Petersen, 1902 (Slowakei) Scopula Schrank, 1802 immorata (Linnaeus, 1758) (Slowakei) umbelaria (Hübner, 1813) (Österreich) virgulata (Denis & Schiffermüller, 175) (Slowakei) ornata (Scopoli, 1763) (Slowakei) rubiginata (Hufnagel, 1767) (Slowakei) marginepunctata (Goeze, 1781) (Slowakei) corrivalaria (Kretschmar, 1862) (Deutschland) immutata (Linnaeus, 1758) (Slowakei) ternata Schrank, 1802 (Slowakei) Glossotrophia Prout, 1913 asellaria (Milliere, 1868) (Italien) Idaea Treitschke, 1825 rufaria (Hübner, 1799) (Österreich) serpentata (Hufnagel, 1767) (Slowakei) aureolaria (Denis & Schiffermüller, 1775) (Österreich) extersaria (Herrich-Schäffer, 1848) (Italien) typicata (Gueneé, 1857) (Italien) filicata (Hübner, 1799) (Österreich) laevigata (Hübner, 1799) (Deutschland) obsoletaria (Rambur, 1833) (Italien) biselata (Hufnagel, 1767) (Slowakei) calunetaria (Staudinger, 1859) (Schweiz) inquinata (Scopoli, 1763) (Deutschland) dilutaria (Hübner, 1799) (Slowakei) fuscovenosa (Goeze, 1781) (Slowakei) Patocka: Puppen der Spanner Mitteleuropas humiliata (Hufnagel, 1767) (Slowakei) seriata (Schrank, 1802) (Österreich) subsericeata (Haworth, 1809) (Italien) emarginata (Linnaeus, 1758) (Deutschland) aversata (Linnaeus, 1758) (Slowakei) degeneraria (Hübner, 1799) (Slowakei) deversaria (Herrich-Schäffer, 1847) (Slowakei) Rhodostrophia Hübner, 1823 vibicaria (Clerck, 1759) (Slowakei) Rhodometra Meyrick, 1892 sacraria (Linnaeus, 1767) (Bosnien) Lythria Hübner, 1823 purpuraria (Linnaeus, 1758) (Slowakei) rotaria (Fabricius, 1798) (Russland) Cataclysme Hübner, 1825 riguata (Hübner, 1813) (Slowakei) Phibalapteryx Stephens, 1829 virgata (Hufnagel, 1767) (Slowakei) Scotopteryx Hübner, 1825 _ moeniata (Scopoli, 1763) (Österreich) coarctaria (Denis & Schiffermiiller], 1775) (Slowakei) chenopodiata (Linnaeus, 1758) (Slowakei, Dänemark) luridata (Hufnagel, 1767) (Slowakei) Larentia Treitschke, 1825 clavarıa (Haworth, 1809) (Polen) Orthonama Hübner, 1825 vittata (Borkhausen, 1794) (Deutschland) Nycterosea Hulst, 1896 È obstipata (Fabricius, 1794) (Osterreich) Xanthorhoe Hübner, 1825 biriviata (Borkhausen, 1794) (Slowakei) designata (Hufnagel, 1767) (Slowakei) munitata (Hübner, 1809) (Schweiz) spadicearia ([Denis & Schiffermiiller], 1775) (Slowakei) ferrugata (Clerck, 1759) (Slowakei) quadrifasciata (Clerck, 1759) (Slowakei) montanata (Denis & Schiffermüller], 1775) (Slowakei) fluctuata (Linnaeus, 1758) (Slowakei) incursata (Hübner, 1813) (Schweiz) Catarhoe Herbulot, 1951 rubidata (Denis & Schiffermüller), 1775) (Slowakei) cuculata (Hufnagel, 1767) (Slowakei) putridata (Herrich-Schäffer, 1852) (Frankreich) Epirrhoe Hübner, 1825 hastulata (Hübner, 1813) (Slowakei) pupillata (Thunberg, 1788) (Deutschland) tristata (Linnaeus, 1758) (Slowakei) alternata (Müller, 1764) (Slowakei) rivata (Hübner, 1813) (Dänemark) molluginata (Hübner, 1813) (Slowakei) galiata (Denis & Schiffermüller], 1775) (Slowakei) Camptogramma Stephens, 1831 51 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 bilineata (Linnaeus, 1758) (Dänemark, Slowakei) Entephria Hübner, 1825 cyanata (Hübner, 1809) (Schweiz) flavicinctata (Hübner, 1813) (Deutschland) infidaria (La Harpe, 1852) (Schweiz) caesiata (Denis & Schiffermüller, 1775) (Slowakei) Anticlea Stephens, 1831 badiata (Denis & Schiffermüller), 1775) (Slowakei) derivata ([Denis & Schiffermüller], 1775) (Slowakei) Mesoleuca Hübner, 1825 albicillata (Linnaeus, 1758) (Slowakei) Pelurga Hübner, 1825 comitata (Linnaeus, 1758) (Slowakei) Lampropteryx Stephens, 1831 suffumata ([Denis & Schiffermüller], 1775) (Slowakei) otregiata (Metcalfe, 1917) (Mähren) Cosmorhoe Hübner, 1825 ocellata (Linnaeus, 1758) (Slowakei) Nebula Bruand, 1846 salicata (Hübner, 1799) (Slowakei) tophaceata (Denis & Schiffermüller, 1775) (Mähren) nebulata (Treitschke, 1828) (Österreich) achromaria (La Harpe, 1852) (Slowakei) Eulithis Hübner, 1821 prunata (Linnaeus, 1758) (Slowakei) testata (Linnaeus, 1761) (Slowakei, Böhmen) populata (Linnaeus, 1758) (Böhmen) pyropata (Hübner, 1822) (Polen) pyraliata ({Denis & Schiffermüller), 1775) (Slowakei) Ecliptopera Warren, 1894 silaceata ([Denis & Schiffermüller], 1775) (Slowakei) capitata (Herrich-Schäffer, 1839) (Slowakei) Chloroclysta Hübner, 1825 siterata (Hufnagel, 1767) (Slowakei) miata (Linnaeus, 1758) (Slowakei, Deutschland) Dysstroma Hübner, 1825 citrata (Linnaeus, 1761) (Deutschland, Slowakei) truncata (Hufnagel, 1767) (Slowakei) infuscata (Tengström, 1869) (Schweden) Cidaria Treitschke, 1825 fulvata (Forster, 1771) (Slowakei) Plemyria Hübner, 1825 rubiginata (Denis & Schiffermüller], 1775) (Slowakei) Thera Stephens, 1831 obeliscata (Hübner, 1787) (Slowakei) variata ([Denis & Schiffermüller], 1775) (Slowakei) britannica (Turner, 1925) (Slowakei) stragulata (Hübner, 1809) (Slowakei) cognata (Thunberg, 1792) (Österreich) juniperata (Linnaeus, 1758) (Slowakei) cupressata (Geyer, 1831) (Frankreich) 52 Eustroma Hübner, 1825 reticulata (Denis & Schiffermüller, 1775) (Slowakei) Electrophaes Prout, 1923 corylata (Thunberg, 1792) (Slowakei) Colostygia Hübner, 1825 _ aptata (Hübner, 1813) (Österreich) _ austriacaria (Herrich-Schäffer, 1856) (Österreich) tempestaria (Herrich-Schäffer, 1856) (Österreich) multistrigaria (Haworth, 1809) (Schweiz) Hydriomena Hübner, 1825 furcata (Thunberg, 1784) (Slowakei) impluviata ([Denis & Schiffermüller], 1775) (Slowakei) ruberata (Freyer, 1831) (Österreich, Deutschland) Horisme Hübner, 1825 vitalbata ([Denis & Schiffermüller), 1775) (Slowakei, Österreich) tersata ([Denis & Schiffermüller), 1775) (Slowakei, Deutschland) aemulata (Hübner, 1813) (Slowakei) aquata (Hübner, 1813) (Deutschland) corticata (Treitschke, 1835) (Österreich) radicaria (La Harpe, 1855) (Deutschland) Melanthia Duponchel, 1829 procellata (Denis & Schiffermüller], 1775) (Slowakei) alaudaria (Freyer, 1846) (Slowakei) Pareulype Herbulot, 1951 berberata ([Denis & Schiffermüller), 1775) (Dänemark, Slowakei) Spargania Guenee, 1857 luctuata ([Denis & Schiffermüller), 1775) (Slowakei) Rheumaptera Hübner, 1822 hastata (Linnaeus, 1758) (Slowakei) cervinalis (Scopoli, 1763) (Slowakei) undulata (Linnaeus, 1758) (Slowakei, Österreich) Triphosa Stephens, 1829 dubitata (Linnaeus, 1758) (Slowakei) sabaudiata (Duponchel, 1830) (Österreich) Philereme Hübner, 1825 vetulata ([Denis & Schiffermüller], 1775) (Slowakei) transversata (Hufnagel, 1767) (Slowakei) Euphyia Hübner, 1825 biangulata (Haworth, 1809) (Slowakei) unangulata (Haworth, 1809) (Böhmen) frustata (Treitschke, 1828) (Österreich) Epirrita Hübner, 1822 dilutata (Denis & Schiffermüller], 1775) (Slowakei) christyi (Allen, 1906) (Slowakei) autumnata (Borkhausen, 1794) (Slowakei) Operophtera Hübner, 1825 brumata (Linnaeus, 1758) (Slowakei) fagata (Scharfenberg, 1805) (Slowakei) Perizoma Hübner, 1825 affinitata (Stephens, 1831) (Polen) alchemillata (Linnaeus, 1758) (Slowakei) hydrata (Treitschke, 1829) (Schweiz) blandiata ({Denis & Schiffermüller), 1775) (Slowakei) flavofasciata (Thunberg, 1792) (Slowakei) didymata (Linnaeus, 1758) (Slowakei, Böhmen) obsoletaria (Herrich-Schäffer, 1838) (Österreich, Schweiz) sagittata (Fabricius, 1787) (Dänemark, Böhmen) incultaria (Herrich-Schäffer, 1848) (Österreich) parallelolineata (Retzius, 1783) (Österreich) Baptria Hübner, 1825 _ tibiale (Esper, 1791) (Österreich) Eupithecia Curtis, 1825 tenuiata (Hübner, 1813) (Slowakei) inturbata (Hübner, 1817) (Schweiz, Deutschland) haworthiata Doubleday, 1856 (Dänemark, Deutschland) immundata (Lienig & Zeller, 1846) (Slowakei, Deutsch- land) plumbeolata (Haworth, 1809) (Deutschland) abietaria (Goeze, 1781) (Deutschland) analoga Djakonov, 1926 (Deutschland, Polen, Slowakei) linariata (Denis & Schiffermüller), 1775) (Deutschland, Slowakei) pulchellata Stephens, 1831 (Deutschland) pyreneata Mabille, 1871 (Slowakei, Deutschland) laquaearia Herrich-Schäffer, 1848 (Slowakei, Deutschland) irriguata (Hübner, 1813) (Slowakei, Deutschland) liguriata Milliere, 1884 (Frankreich) exiguata (Hübner, 1813) (Deutschland) insigniata (Hübner, 1790) (Deutschland) valerianata (Hübner, 1813) (Slowakei, Deutschland) pygmaeata (Hübner, 1799) (Deutschland) undata (Freyer, 1840) (Schweiz) variostrigata Alpheraky, 1878 (Schweiz) silenata Assman, 1849 (Böhmen, Deutschland) carpophagata Staudinger, 1871 (Italien) venosata (Fabricius, 1787) (Slowakei, Deutschland) schiefereri Bohatsch, 1893 (Italien) silenicolata Mabille, 1866 (Italien) alliaria Staudinger, 1870 (Slowakei) egenaria Herrich-Schäffer, 1848 (Deutschland) extraversaria Herrich-Schäffer, 1852 (Deutschland) centaureata ([Denis & Schiffermüller}, 1775) (Slowakei, Deutschland) queneeata Miller, 1862 (Österreich) gratiosata Herrich-Schäffer, 1861 (Frankreich) thalictrata (Püngeler, 1902) (Schweiz, Italien) acteata Walderdorf, 1869 (Deutschland) selinata Herrich-Schäffer, 1861 (Deutschland, Slowakei) trisignaria Herrich-Schäffer, 1848 (Slowakei, Deutschland) intricata (Zetterstedt, 1839) (Slowakei, Deutschland) veratraria Herrich-Schäffer, 1848 (Slowakei, Deutschland) cauchiata (Duponchel, 1830) (Deutschland) satyrata (Hübner, 1813) (Slowakei, Deutschland) cretaceata Packard, 1874 (Italien) pernotata Guenée, 1857 (Schweiz) absinthiata (Clerck, 1759) (Slowakei, Deutschland) goossensiata Mabille, 1869 (Dänemark, Deutschland) assimilata Doubleday, 1856 (Slowakei, Deutschland) expallidata Doubleday, 1856 (Deutschland) vulgata (Haworth, 1809) (Slowakei, Deutschland) Patocka: Puppen der Spanner Mitteleuropas tripunctaria Herrich-Schäffer, 1852 (Slowakei, Deutsch- land) denotata (Hübner, 1813) (Slowakei, Deutschland) subfuscata (Haworth, 1809) (Slowakei, Deutschland) icterata (Villers, 1789) (Deutschland) succenturiata (Linnaeus, 1758) (Deutschland) denticulata (Treitschke, 1828) (Deutschland) impurata (Hübner, 1813) (Deutschland) orphnata W. Petersen, 1909 (Slowakei, Deutschland) subumbrata (|Denis & Schiffermüller), 1775) (Deutsch- land) semigraphata Bruand, 1851 (Deutschland) millefoliata Rössler, 1866 (Deutschland) santolinata Mabille, 1871 (Frankreich) simpliciata (Haworth, 1809) (Slowakei, Deutschland) sinuosaria Eversmann, 1848 (Slowakei, Polen, Deutschland) druentiata Dietze, 1902 (Frankreich) distinctaria (Herrich-Schäffer, 1848) (Slowakei, Deutsch- land) graphata (Treitschke, 1828) (Schweiz, Italien, Österreich) pimpinellata (Hübner, 1813) (Deutschland) pauxillaria Boisduval, 1840 (Deutschland) gelidata (Möschler, 1860) (Böhmen, Deutschland) nanata (Hübner, 1813) (Dänemark, Deutschland) innotata (Hufnagel, 1767) (Deutschland) fraxinata Grewe, 1863 (Deutschland) unedonata Mabille, 1868 (Spanien) virgaureata Doubleday, 1861 (Slowakei, Deutschland) abbreviata Stephens, 1831 (Deutschland) dodoneata Guenee, 1857 (Deutschland) pusillata ([Denis & Schiffermüller), 1775) (Slowakei, Deutschland) ericeata (Rambur, 1833) (Slowakei) phoeniceata (Rambur, 1834) (Spanien) scopariata (Rambur, 1833) (Frankreich) oxycedrata (Rambur, 1833) (Italien) lanceata (Hübner, 1825) (Slowakei, Deutschland) lariciata (Freyer, 1842) (Slowakei, Deutschland) tantillaria Boisduval, 1840 (Slowakei, Deutschland) Gymnoscelis Mabille, 1868 rufifasciata (Haworth, 1809) (Slowakei, Deutschland) Chloroclystis Hübner, 1825 v-ata (Haworth, 1809) (Slowakei, Deutschland) chloerata (Mabille, 1870) (Slowakei, Deutschland, Däne- mark) rectangulata (Linnaeus, 1758) (Slowakei, Deutschland) debiliata (Hübner, 1817) (Böhmen, Dänemark, Deutsch- land) Anticollix Prout, 1938 sparsata (Treitschke, 1828) (Dänemark, Deutschland) Chesias Treitschke, 1825 legatella (Denis & Schiffermüller], 1775) (Mähren, Schwe- den) Carsia sororiata (Linnaeus, 1758) (Schweden) Aplocera Stephens, 1827 plagiata (Linnaeus, 1758) (Slowakei) efformata (Guenee, 1857) (Dänemark) praeformata (Hübner, 1826) (Slowakei) 53 Tij DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Odezia Boisduval, 1840 atrata (Linnaeus, 1758) (Deutschland) Schistostege Hübner, 1825 treitschkei Kovács, 1957 (Ungarn) decussata (Denis & Schiffermüller, 1775) (Österreich) Lithostege Hübner, 1825 griseata ([Denis & Schiffermüller), 1775) (nach Khotko, 1977) Disciloxia Warren, 1895 blomeri (Curtis, 1832) (Slowakei) Venusia Curtis, 1839 cambrica (Curtis, 1839) (Slowakei) Euchoeca Hübner, 1823 nebulata (Scopoli, 1763) (Slowakei) Asthena Hübner, 1825 albulata (Hufnagel, 1767) (Slowakei) anseraria (Herrich-Schäffer, 1855) (Slowakei) Hydrelia Hübner, 1825 flammeolaria (Hufnagel, 1767) (Slowakei) Minoa Treitschke, 1825 murinata (Scopoli, 1763) (Slowakei) Lobophora Curtis, 1825 halterata (Hufnagel, 1767) (Slowakei) Trichopteryx Hübner, 1825 polycommata (Denis & Schiffermüller], 1775) (Slowakei) carpinata (Borkhausen, 1794) (Slowakei) Epilobophora Inoue, 1943 sabinata (Geyer, 1831) (Schweiz) Nothocasis Prout, 1937 sertata (Hübner, 1817) (Slowakei) Pterapherapteryx Curtis, 1825 sexalata (Retzius, 1783) (Slowakei) Acasis Duponchel, 1845 appensata (Eversmann, 1842) (Slowakei) Abraxas Leach, 1815 grossulariata (Linnaeus, 1758) (Slowakei) Calospilos Hübner, 1825 sylvata (Scopoli, 1763) (Slowakei) Lomaspilis Hübner, 1825 marginata (Linnaeus, 1758) (Slowakei) Ligdia Guenée, 1857 adustata ({Denis & Schiffermüller], 1775) (Slowakei) Stegania Guenée, 1857 dilectaria (Hübner, 1790) (Slowakei, Frankreich) Semiothisa Hübner, 1818 notata (Linnaeus, 1758) (Slowakei) alternaria (Hübner, 1809) (Slowakei) signaria (Hübner, 1809) (Slowakei) Chiasmia Hübner, 1823 clathrata (Linnaeus, 1758) (Slowakei) glarearia (Brahm, 1791) (Slowakei) Diastictis artesiaria (Denis & Schiffermüller, 1775) (Slowakei) Isturgia Hübner, 1823 limbaria (Fabricius, 1775) (Slowakei) Narraga Walker, 1861 fasciolaria (Hufnagel, 1767) (Slowakei) Itame Hübner, 1823 wauaria (Linnaeus, 1758) (Slowakei) brunneata (Thunberg, 1784) (Böhmen, Schweden) Tephrina Guenée, 1845 arenacearia (Denis & Schiffermüller, 1775) (Slowakei) murinaria (Denis & Schiffermüller, 1775) (Slowakei) Enconista Lederer, 1853 miniosata (Duponchel, 1829) (Frankreich) Pygmaena Boisduval, 1840 fusca (Thunberg, 1792) (Schweiz) Cepphis Hübner, 1823 advenaria (Hübner, 1799) (Polen) Petrophora Hübner, 1811 chlorosata (Scopoli, 1763) (Slowakei, Böhmen) Plagodis Hübner, 1823 pulveraria (Linnaeus, 1758) (Slowakei) dolabraria (Linnaeus, 1767) (Slowakei) Pachycnemia Stephens, 1829 hippocastanaria (Hübner, 1799) (Holland) Opisthograptis Hübner, 1823 luteolata (Linnaeus, 1758) (Slowakei) Epione Duponchel, 1829 repandaria (Hufnagel, 1767) (Slowakei) paralellaria ({Denis & Schiffermüller], 1775) (Slowakei) Pseudopanthera Hübner, 1823 macularia (Linnaeus, 1758) (Slowakei) Hypoxystis Prout, 1915 pluviaria (Fabricius, 1787) (Slowakei) Therapis Hübner, 1823 flavicaria (Denis & Schiffermüller, 1775) (Österreich) Epirrhanthis Hübner, 1823 diversata (Denis & Schiffermüller, 1775) (Böhmen) Ennomos Treitschke, 1825 autumnaria (Werneburg, 1859) (Slowakei) quercinaria (Hufnagel, 1767) (Slowakei) alniaria (Linnaeus, 1758) (Slowakei, Deutschland) fuscantaria (Haworth, 1809) (Osterreich, Deutschland) erosaria ([Denis & Schiffermüller], 1775) (Slowakei) quercaria (Hübner, 1813) (Italien) Selenia Hübner, 1823 dentaria (Fabricius, 1775) (Slowakei) lunularia (Hübner, 1788) (Slowakei, Böhmen) tetralunaria (Hufnagel, 1767) (Slowakei) Apeira Gistl, 1848 syringaria (Linnaeus, 1758) (Böhmen, Slowakei) Artiora Meyrick, 1892 evonymaria (Denis & Schiffermüller, 1775) (Slowakei) Odontopera Stephens, 1831 bidentata (Clerck, 1759) (Slowakei) Crocallis Treitschke, 1825 tusciaria (Borkhausen, 1793) (Deutschland) elinguaria (Linnaeus, 1758) (Slowakei) Ourapteryx Leach, 1814 sambucaria (Linnaeus, 1758) (Slowakei, Böhmen) Colotois Hübner, 1823 pennaria (Linnaeus, 1761) (Slowakei) Angerona Duponchel, 1829 prunaria (Linnaeus, 1758) (Slowakei) Chondrosoma } fiduciaria Anker, 1854 (Osterreich) Apocheima Hübner, 1825 hispidaria ([Denis & Schiffermüller), 1775) (Slowakei) Phigalia Duponchel, 1829 pilosaria ((Denis & Schiffermüller), 1775) (Slowakei) Lycia Hübner, 1825 hirtaria (Clerck, 1759) (Slowakei) graecaria (Staudinger, 1870) (Bosnien) isabellae (Harrison, 1914) (Böhmen) pomonaria (Hübner, 1790) (Slowakei) zonaria (Denis & Schiffermiiller], 1775) (Slowakei) Biston Leach, 1815 strataria (Hufnagel, 1867) (Slowakei) betularia (Linnaeus, 1758) (Slowakei) Agriopis Hübner, 1825 leucophaearia ([Denis & Schiffermüller], 1775) (Slowakei) bajaria ([Denis & Schiffermüller], 1775) (Slowakei) aurantiaria (Hübner, 1799) (Slowakei) marginaria (Fabricius, 1776) (Slowakei) Erannis Hübner, 1825 defoliaria (Clerck, 1759) (Slowakei) Nychiodes Lederer, 1853 obscuraria (De Viller, 1789) (Frankreich) Patocka: Puppen der Spanner Mitteleuropas Eurranthis Hübner, 1823 plumistaria (De Viller, 1789) (Frankreich) Peribatodes Wehrli, 1943 rhomboidaria ((Denis & Schiffermiiller], 1775) (Slowakei) umbraria (Hübner, 1809) (Italien) secundaria (|Denis & Schiffermüller}, 1775) (Slowakei) Cleora Curtis, 1825 cinctaria ({Denis & Schiffermiiller], 1775) (Slowakei) Deileptenia Hübner, 1825 ribeata (Clerck, 1759) (Slowakei) Alcis Curtis, 1826 repandata (Linnaeus, 1758) (Slowakei) maculata (Staudinger, 1890) (Slowakei) Arichanna Moore, 1868 melanaria (Linnaeus, 1758) (Slowakei, Böhmen) Hypomecis Hübner, 1821 roboraria ([Denis & Schiffermüller), 1775) (Slowakei) viertlii (Bohatsch, 1883) (Slowakei) Serraca Moore, 1887 punctinalis (Scopoli, 1763) (Slowakei) Fagivorina Wehrli, 1943 arenaria (Hufnagel, 1767) (Slowakei) Ascotis Hübner, 1825 selenaria (Denis & Schiffermüller, 1775) (Slowakei) Ectropis Hübner, 1825 crepuscularia ([Denis & Schiffermüller], 1775) (Slowakei) Paradarsia Warren, 1897 consonaria (Hübner, 1799) (Slowakei) Parectropis Sato, 1980 similaria (Hufnagel, 1767) (Slowakei) Aethalura Mc Dunnough, 1920 punctulata ((Denis & Schiffermüller], 1775) (Slowakei) Ematurga Lederer, 1853 atomaria (Linnaeus, 1758) (Slowakei) Tephronia Hübner, 1825 sepiaria (Hufnagel, 1767) (Deutschland) Odontognophos Wehrli, 1951 dumetata (Treitschke, 1827) (Slowakei) Selidosema Hübner, 1823 brunnearia (Villers, 1789) (Böhmen) plumaria ([Denis & Schiffermüller], 1775) (Slowakei) Bupalus Leach, 1815 piniaria (Linnaeus, 1758) (Slowakei) Crocota Hübner, 1823 lutearia (Fabricius, 1794) (Frankreich, Schweiz) niveata (Scopoli, 1763) (Österreich) 55 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Menophra Moore, 1887 nycthemeraria (Geyer, 1831) (Deutschland) Synopsia Hübner, 1825 sociaria (Hübner, 1799) (Slowakei) Cleorodes Warren, 1894 lichenaria (Hufnagel, 1767) (Deutschland) Gnophos Treitschke, 1825 furvatus ([Denis & Schiffermüller), 1775) (Slowakei) obfuscatus ([Denis & Schiffermüller], 1775) (Österreich) ambiguatus (Duponchel, 1838) (Österreich) pullatus ([Denis & Schiffermüller], 1775) (Österreich) glaucinarius (Hübner, 1799) (Österreich) variegatus (Duponchel, 1830) (Österreich, Schweiz) intermedia Wehrli, 1917 (Slowakei) dilucidaria (Denis & Schiffermüller}, 1775) (Österreich, Schweiz) serotinaria (Denis & Schiffermüller), 1775) (Österreich, Schweiz) zelleraria (Freyer, 1836) (Österreich) caelibaria (Herrich-Schäfer, 1852) (Österreich) operaria (Hübner, 1813) (Österreich) Sciadia Hübner, 1822 A tenebraria (Esper, 1806) (Österreich, Schweiz) Glacies Milliere, 1874 i alticolaria (Mann, 1853) (Osterreich) Siona Duponchel, 1829 lineata (Scopoli, 1763) (Slowakei) Chariaspilates Wehrli, 1953 formosaria (Eversmann, 1837) (nach Khotko, 1977) Aspitates Treitschke, 1825 gilvaria ((Denis & Schiffermüller], 1775) (Slowakei) Dyscia Hübner, 1825 conspersaria (Fabricius, 1775) (Slowakei) Perconia Hiibner, 1823 strigillaria (Hübner, 1787) (Mähren) Cabera Treitschke, 1825 pusaria (Linnaeus, 1758) (Slowakei) exanthemata (Scopoli, 1763) (Slowakei) Lomographa Hübner, 1825 bimaculata (Fabricius, 1775) (Slowakei) temerata ([Denis & Schiffermüller], 1775) (Slowakei) Campaea Lamarck, 1816 margaritata (Linnaeus, 1767) (Slowakei) honoraria ([Denis & Schiffermüller], 1775) (Böhmen) Hylaea Hübner, 1822 fasciaria (Linnaeus, 1758) (Slowakei) Puengeleria Rougemont, 1903 capreolaria ({Denis & Schiffermüller), 1775) (Slowakei) Lignyoptera thaumastaria Rebel, 1901 (Bosnien) Theria Hübner, 1825 rupicapraria ([Denis & Schiffermüller], 1775) (Slowakei) primaria (Haworth, 1809) (Deutschland) JOHN T. POLHEMUS & Dan A. POLHEMUS University of Colorado Museen, ‘Englewood, U.S.A. & Bishop Museum, Honolulu, U.S.A. FOUR NEW GENERA OF MICROVELINAE (HETEROPTERA) FROM NEW GUINEA Polhemus, J. T. & D. A. Polhemus, 1994. Four new genera of Microveliinae (Heteroptera) from New Guinea. — Tijdschrift voor Entomologie 137: 57-74, figs. 1-30. [ISSN 0040-7496]. Published 15 July 1994. Four new genera of Microveliinae are described from New Guinea. These genera and their con- stituent species are as follows: Tanyvelia gen. n. containing type species 7. missim sp. n. from Papua New Guinea; Aegilipsicola gen. n. containing type species A. rapida sp. n. from Papua New Guinea; Tarsovelia gen. n. containing type species 7. alta sp. n. from Papua New Guinea, plus 7. arfak sp. n. from Irian Jaya, and 7. dani sp. n. from Irian Jaya; and Neusterinsifer gen. n. containing type species N. compacta sp. n. from Papua New Guinea, plus N. sepik sp. n. from Papua New Guinea, N. cyclops sp. n. from Papua New Guinea, N. nabire sp. n. from Irian Jaya, and N. gladius sp. n. from Irian Jaya. Habitus figures of the above new genera are provided, ac- companied by illustrations of the male genital structures and distribution maps for all new spe- cies. Correspondence: Dr Dan A. Polhemus, Dept. of Natural Sciences, Bishop Museum, P.O. Box 19000-A, Honolulu, HI. 96817, U.S.A. Key words. — Veliidae; Microveliinae; New Guinea; taxonomy; new genera; new species; keys; distribution. The assemblages of Microveliinae occurring on the islands of the Malay Archipelago exhibit a profound shift in taxonomic composition as one progresses from west to east through the region. On the Greater Sunda Islands, Celebes and the Philippines such fau- nas are dominated by members of the genera Pseudovelia Hoberlandt and Microvelia Westwood, accompanied by several other more ecologically spe- cialized and less speciose genera. On New Guinea and surrounding islands, by contrast, Pseudovelia is absent and there occurs instead a group of previously unde- scribed endemic genera which fill the typical Pseudovelia niches. In addition, New Guinea also supports numerous endemic species of in the genus Microvelia sensu lato, but as noted by Andersen (1982) this generic grouping is almost certainly poly- phyletic. A detailed analysis on a world basis of the subgroups contained within it will be necessary be- fore it can be certain that the Papuan and Australian species presently held in Microvelia are in fact conge- neric with the Asian forms. New Guinea thus represents a major center of di- versification for the Microveliinae, with a fauna de- rived independently from that occurring in the Asian tropics. In the present report we describe four new genera of endemic Papuan Microveliinae, all ofwhich are easily separable from Microvelia sensu lato on the basis of distinct apomorphies. The relationships among these genera are still unresolved, however, | pending the completion of ongoing revisions of the diverse Australian and New Guinea Microvelia fau- nas. Key to genera of New Guinea Microveliinae 1. Middle tarsi subequal in length to middle tibia (fig. 9); light markings on hemelytra restricted to basal angles; fore femur of male at least slightly modified for phoresy .............. Tarsovelia gen. n. — Middle tarsi distinctly shorter than middle tibia (figs. 1, 5, 17); light markings on hemelytra may be restricted to basal angles, or may occur also on distal portions; fore femur may or may not be modified for phoresy (figs. 3, 6, 24-26) ........... 2 2. Claws extremely long; light hemelytral markings entirely bright greyish to light bluish pruinose; gula long, rostral cavity demarcated by strongly laisedicarina Re ee Aegilipsicola gen. n. — Claws relatively short; light hemelytral markings not bright pruinose; gula short, rostral cavity not demarcated by strongly raised carina ............... 3 3. Antennae extremely long (fig. 1); comb on fore tibia 3/4 the length of the tibia (fig. 3); only mi- cropterous morph known ....... Tanyvelia gen. n. — Antennae not extremely long (fig. 17); comb on fore tibia less than 3/4 length of tibia (except in Neuterinsifer compacta), if comb on fore tibia 3/4 length of tibia then male proctiger bearing elon- gate process and female abdominal tergites VII 57 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 and VIII forming an anal plate (see below); only apterous and macropterous morphs known, latter with large hemelytral light markings basally, and often occurring in all closed cells ..................... 4 4. Female abdominal tergites VII-VIII deflected ventrad forming an anal plate (fig. 23); male gen- italia highly modified, bearing a long sinuate an- teriorly directed sword-like process (figs. 18-22) ot ak EEE Es, Neusterinsifer gen. n. — Female abdominal tergites VII-VIII not deflected ventrad, not forming an anal plate; male genitalia may be modified, but without a long sword-like DIOCESS ee cesser Microvelia Westwood Tanyvelia gen. n. (figs. 1-4, 27) Diagnosis Size. — Micropterous form, length of males 2.73- 2.91 mm, females 2.81-3.13 mm; general body char- acteristics and size sexually dimorphic, males smaller, female abdomen more robust. Colour. — Ground colour blackish brown, tinged with orange brown, without silvery pubescence; ante- rior pronotal lobe yellowish, pruinose, sharply demar- cated from dark collar and posterior lobe. Micropterous wing pads brown, set with anterolater- al light streaks. Structural characters. — Only micropterous form known, shape elongate (fig. 1). Eyes globose, exserted bur usually not beyond anterolateral pronotal angles, separated by about twice the width of an eye, ap- pressed to anterior pronotal margin, with short ocular setae. Head declivant anteriorly, recessed into prono- tum, posterior margin sloping caudo-dorsally, with usual three pairs of facial trichobothria; gular region moderately long, plainly visible, rostral cavity closed posteriorly. Rostrum reaching to middle of mesoster- num, segment I short, enclosed in rostral cavity, seg- ments I and IV subequal in length and about three times longer than II, segment III about 8 times as long as II. Antennae slender, very long, about 3/4 of body length. Pronotum slightly raised medially, with weak me- dian longitudinal carina; collar weakly formed, set off by a ragged row of small dark foveae, terminating under eyes laterally; anterior and posterior lobes set off by a transverse row of small foveae, evanescent medially; anterior lobe with scattered stiff erect dark setae; posterior lobe with numerous small foveae, hu- meri not prominent, broadly rounded posteriorly, not modified, covering metanotum. Thoracic venter not diagnostic, with weakly formed tubercles on ei- ther side of mesosternal midline on posterior margin opposing an unmodified metasternum. Metasternal scent gland opening (omphalium) small but visible, 58 marked by a small tubercle; scent channels promi- nent, curving slightly anterad to base of metacetabu- lae. Abdomen without silvery setae or other adorn- ment, except weak longitudinal pruinose line on mid- line of tergites IV-VII; with short paired longitudinal carinae on basal half of tergite II. Abdominal sternites set off from laterosternites by hair-free glabrous oval lacunae. Micropterous wing pads elongate. Legs slender, long, hind legs very long; anterior fe- mur set beneath with short light setae, unmodified in males; anterior tibia of male with a comb of minute black setae occupying 3/4 of tibial length; middle fe- mora set ventrally with 4-5 very long slender setae; all tarsi long (fig. 3), claws moderately long; both up- and down curving arolia large, evident. Male genital segments moderately large, protrud- ing, modified; proctiger unmodified (fig. 4); para- meres small, slender symmetrical, acuminate (fig. 2). Female tergite VIII on same plane as VII, truncate posteriorly; first gonocoxae small, barely exposed, plate-like; tergite IX of both sexes triangular, protrud- ing posteriorly. Type-species: Tanyvelia missim sp. n. Remarks Comparative notes. — Separable from other genera of Papuan Microveliinae by the characters given in the key. Similar in general appearance and ecological habits to Baptista Distant, but distinguished by the presence of a long grasping comb that runs nearly the entire length of the male fore tibia (fig. 3), and the ab- sence of modifications to abdominal sterna VI and VII. Among the Papuan genera, Tanyvelia appears most closely allied to Aegzlipsicola, but the pronotal collar is not as pronounced, the legs, although long and slender, are slightly stouter, and the fore tibial grasping comb is shorter. The proctiger of Tanyvelia is broadly rounded distally rather than coming to a rounded point as in Aegzlipsicola (compare figs. 4 and 7), and the paramere is slender and elongate rather than short and stout (compare figs. 2 and 8). Etymology. — The generic name Tanyvelia is de- rived from tany- (Gr.), long, referring to the slender body shape, and Velia, the nominate genus of the family. Gender feminine. Distribution. — New Guinea (fig. 27). Tanyvelia missim sp. n. (figs. 1-4, 27) Type material: Holotype, micropterous male: Papua New Guinea, Morobe Prov., upper Poverty Creek on Mt. Missim, nr. Wau, 1600 m., 18 Sept. POLHEMUS & POLHEMUS: Microveliinae from New Guinea au co: € ee v u u te roe Figs. 1-4. Tanyvelia missim, gen. n., sp. n. . — 1. Male, dorsal habitus; 2. Male foreleg; 3. Male paramere; 4. Male proctiger. Figs. 5-8. Aegilipsicola rapida, gen. n., sp. n. — 5. Male, dorsal habitus; 6. Male foreleg; 7. Male proctiger; 8. Male paramere 59 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 1983, CL 1830, D. A. and J. T. Polhemus (BPBM). - Paratypes: 23 micropterous males, 12 micropterous females, same data as holotype (JTPC). Diagnosis Size. — Micropterous male, length 2.73-2.91 mm (X = 2.81, n = 10); width 0.79-0.90 mm (x = 0.86, n = 10). Micropterous female, length 2.81-3.13 mm (x = 2.96, n = 10); width 0.83-0.94 mm, (x = 0.88, n = 10). Colour. — Micropterous male: ground colour black, venter and connexiva tinged with brown. Head black, brown ventrally; rostrum luteous on basal three segments, piceous distally. Pronotum with anterior lobe entirely yellowish including propleura, except narrowly embrowned medially; disc and collar black. Abdomen black, tergites mostly pruinose, lighter ven- trally. Legs and antennae luteous to yellow brown, distally darker. Structural characters. — Micropterous male: head of moderate length, declivant anteriorly; length 0.47; width of eye/interocular space, 0.16/0.30. Pronotum length:width, 0.74 : 0.81. Abdominal tergites not shining; tergites, II-VI subequal in length (0.16- 0.19), VII longer (0.30). Abdominal venter set with short appressed setae; ventrite VII with a short V- shaped depression on caudal half. Legs, antennae thickly clothed with short to moderate length setae, with scattered longer setae. Legs unarmed. Antennal formula, segments I : II : III : IV; 0.51: 0.40 : 0.70 : 0.70. Proportions of legs as follows: Femur, tibia, tarsal 1, tarsal 2 of fore leg, 0.79 : 0.77 : 0.30 : 0.0; of mid- dle leg, 1.07 : 1.05 : 0.14 : 0.28; of hind leg, 1.16 : 1.47 : 0.23 : 0.30. First genital segment with an antero-ventrally di- rected tuft of stiff dark setae. Paramere small, long, triangular (fig. 2). Micropterous female: Similar to male in most re- spects, but somewhat larger; connexiva vertical or re- flexed over dorsum except in gravid specimens. Remarks Comparative notes. — Recognized among the Pa- puan Microveliinae by its elongate form with long legs and antennae (fig. 1), relatively unmodified male foreleg with a long tibial grasping comb (fig. 3), acu- minate paramere (fig. 2), and elongate male proctiger (fig. 4). Biological notes. — Upper Poverty Creek at the Mt. Missim type locality was a small, rocky headwater stream dropping steeply through a boulder-strewn bed heavily shaded by primary rain forest. Etymology. — The name ‘7255577 is a noun in appo- 60 sition and refers to the Mt. Missim type locality. Distribution. — Eastern New Guinea (fig. 27). Aegilipsicola gen. n. (figs. 5-8, 28) Diagnosis Size. — Macropterous form, length of males 2.70- 2.84 mm, females 3.16-3.31 mm, general body char- acteristics not sexually dimorphic, but males slightly smaller. Colour. — Macropterous form: ground colour blackish brown, tinged with yellowish brown; dorsal- ly dark brown on head and ventrally, wings black, brown to black beneath; anterior pronotal lobe nar- rowly yellowish on either side of midline, laterally dark but covered with silvery pubescence except me- dially, demarcated from dark collar and posterior lobe. Hemelytra black, bearing six elongate bluish- grey pruinose streaks, with two confined within the basal cells, and four others on distal third (fig. 5) Structural characters. — Only macropterous form known. Form elongate, slender (fig. 5); eyes globose, exserted beyond anterolateral pronotal angles, separ- ated by about twice the width of an eye, removed from anterior pronotal margin, with short ocular se- tae. Head moderately declivant anteriorly, recessed into pronotum, posterior margin sloping somewhat caudo-dorsally, with usual three pairs of facial tricho- bothria; gular region long, plainly visible, rostral cav- ity closed posteriorly by a strong carinate ridge. Rostrum reaching to middle of mesosternum, seg- ment I short, enclosed in rostral cavity, I and IV sub- equal in length and about three times longer than II, segment III about 8 times as long as II. Antennae very slender, very long, about 2/3 of body length. Pronotum slightly raised medially, with weak median longitudinal carina marked by a line of golden pubes- cence; collar prominent, distinctly set off by a row of deep foveae, terminating laterally; near ventral eye margins; anterior and posterior lobes set off by a transverse row of foveae, evanescent medially; entire pronotum sparsely set with fine decumbent golden pubescence; posterior lobe with numerous obscure shallow foveae, humeri moderately prominent, al- most triangular, narrowly rounded posteriorly. Thoracic venter not diagnostic, with weakly formed tubercles on either side of mesosternal midline along posterior margin opposing an unmodified metaster- num. Metasternal scent gland opening (omphalium) not visible; scent channels prominent, curving ante- rad to base of metacetabulae. Abdomen without silvery setae or other adorn- ment; prominent paired longitudinal carinae present on basal half of tergite II in macropters (visible only after removal of wings). Abdominal sternites set off from laterosternites by hair-free glabrous oval lacu- ' nae. Female connexiva expanded, plate like along ter- gite VII. Macropters with five closed cells in hemely- tra, venation of fore and hind wings similar to figs. 312-313 in Andersen (1982) except distal cell closed; fore wings set with fine semi-recumbent stiff setae on basal half; micropterous form unknown. Legs slender, long, hind legs very long; anterior fe- mur set beneath with short light setae, with ventral tubercle at distal 2/3 in males; anterior tibia of male with a fine line of minute black setae occupying 4/5 of tibial length; femora ventrally without long setae; all tarsi long (fig. 6), claws very long; both up- and down curving arolia long, slender. Male genital segments moderately large, protrud- ing, modified; proctiger unmodified (fig. 7); para- meres small, slender, symmetrical (fig. 8). Female ter- gite VIII on same plane as VII, protruding posteriorly over tergite IX; first gonocoxae small, barely exposed, plate-like; tergite IX of both sexes button-like, pro- truding posteriorly. Type-species: Aegilipsicola rapida sp. n. Remarks Etymology. — The generic name Aegzlipsicola is de- rived from Aegilips (Gr.), sheer, steep, referring to the habitat, and -cola, (L.), inhabitant. Gender feminine. Comparative notes. — In general facies Aegilipsicola most closely resembles the Neotropical genus Aegili- psivelia, however this similarity is apparently due to convergence. Although the two genera share many similar structures adapted to life on steep hygropetric habitats, such as long legs and extremely long claws, Aegilipsivelia lacks the carinate rostral cavity on the gula, the bright pruinose hemelytral markings, and the carinate female connexiva seen in Aegilipsicola. In addition, the collar in Aegzlipsivelia is set off by a sul- cus rather than a row of deep foveae, and the comb- like process on the male fore tibia is relatively short instead of almost as long as the tibia. Among Papuan genera of Microveliinae Aegilipsi- cola is apparently most closely allied to Tanyvelia gen. n. (see discussion under Tanyvelia) Distribution. — Eastern New Guinea (fig. 28). Aegilipsicola rapida sp. n. (figs. 5-8, 28) Type material: Holotype, macropterous male: Papua New Guinea, Morobe Prov., seeps along Mt. Kaindi road nr. Kunai Creek, above Wau, 17 Sept. 1983, CL 1823, D. A. and J. T. Polhemus (BPBM). — Paratypes: 39 macropterous males, 37 macropterous females, 62 immatures, same data as holotype (JTPC); POLHEMUS & POLHEMUS: Microveliinae from New Guinea 4 macropterous males, 4 macropterous females, 2 im- matures, Morobe Prov., Namie Creek, Mt. Kaindi, 17 ‘Sept. 1983RCL 1829; JT: T.-é DAs Polhemus (TPC); 1 macropterous female, Morobe Prov., Kunai River, 1500 m, 10 Oct. 1966, Illies (ZBSM). Diagnosis Size. — Macropterous male, length 2.70-2.84 mm x = 2.74, n = 10); width 0.86-0.97 mm (x= 0.94, n 0). Macropterous female, length 3.06-3.31 mm (x 16, n = 10); width 1.01-1.12 mm, (x = 1.04, n = Rea I DI 0). Colour. — Macropterous male: ground colour black, tinged with yellowish brown ventrally. Head black; two longitudinal regions between eyes and me- dial furrow, posterior margin, brown. Pronotum with anterior lobe narrowly, transversely orange brown on either side of midline, extending to inner eye margins; posterior lobe uniformly black. Hemelytra blackish brown, long, extending beyond tip of abdomen. Ab- domen brownish yellow ventrally, with a broad brown longitudinal stripe along sternal-laterosternal boundary. Legs yellowish, darker distally; antennae brown, segment I yellowish basally. Structural characters. — Macropterous male: head of moderate length, declivant anteriorly; length 0.51; width of eye/interocular space, 0.14/0.30. Pronotum long, humeri evident but not raised; length : width, 0.84 : 0.91. Abdominal venter set with very short appressed fine setae; venter VII depressed on either side of mid- line; first genital segment excavated ventrally, with three small brown sclerotized denticles distally. Legs, antennae thickly clothed with short inconspicuous se- tae, without longer setae. Legs unarmed, except fore femur set near distal 2/3 with patch of short stiff setae (fig. 6). Antennal formula I: II: III : IV; 0.40 : 0.28 : 0.74 : 0.60. Proportions of legs as follows: Femur, tibia, tarsal 1, tarsal 2 of fore leg, 0.74 : 0.60 : 0.28 : 0.0; of mid- dle leg, 1.00 : 0.98 : 0.12 : 0.33; of hind leg, 1.44 : 1.74 : 0.09 : 0.40. Paramere small, short, ovate (fig. 8). Macropterous female: Similar to male in most re- spects, but somewhat larger. Connexiva vertical, strongly raised and plate-like along tergite VII, pro- duced posteriorly along tergite VIII. ll Remarks Comparative notes. — Easily recognized among the Papuan Microveliinae by its elongate form (fig. 5); broadly pointed proctiger (fig. 7), short and broad pa- ramere (fig. 8), elongate fore tibial grasping comb 61 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 9-16. Tarsovelia gen. n. — 9. Tarsovelia alta sp. n., micropterous female, dorsal habitus; 10. Tarsovelia dani sp. n., male proc- tiger; 11-13. Male forelegs of Tarsovelia species, 11. Tarsovelia arfak sp. n.; 12. Tarsovelia dani sp. n.; 13. Tarsovelia alta sp. n.; 14-16. Male parameres of Tarsovelia species. 14. Tarsovelia dani sp. n.; 15. Tarsovelia alta sp. n.; 16. Tarsovelia arfak sp. n. 62 (fig. 6), and small tuft of black setae ventrally on the ' male fore femur (fig. 6). Biological notes. — The type series was taken from wet seeping vertical rock faces formed by road cuts along the Mt. Kaindi road. The insects were abun- dant, and ran across the wet rocks and moss with great speed and agility. Etymology. — The name ‘rapida refers to the speed with which these insects can move across vertical rock faces. Distribution. — Eastern New Guinea (fig. 28). Tarsovelia gen. n. (figs. 9-16, 29) Diagnosis Size. — Length of males 2.41-2.95 mm, females 2.77-3.45 mm, general body characteristics and size sexually dimorphic, males smaller, female abdomen more robust. Colour. — Ground colour brown to blackish brown, often blackish above, without silvery pubes- cence; winged forms dorsally black, wings black, brown to black beneath. Anterior pronotal lobe and collar usually entirely yellowish to orange brown, on- ly in 7: arfak dark laterally. Structural characters. — Form robust (fig. 9); eyes globose, exserted but usually not beyond anterolater- al pronotal angles, separated by at least twice the width of an eye, appressed to anterior pronotal mar- gin, with long ocular setae. Head declivant anteriorly, slightly recessed into pronotum, with usual three pairs of facial trichobothria; gular region moderately long, plainly visible, rostral cavity closed posteriorly. Rostrum reaching almost to metasternum, segment I short, enclosed in rostral cavity, segments I and IV subequal in length and about three times longer than II, segment III about 8 times as long as II. Antennae slender, long, about 1/2 of body length. Pronotum of micropterous form slightly raised me- dially; collar weakly formed, set off by a ragged row of small dark foveae, terminating under eyes laterally; anterior and posterior lobes set off by a transverse row of small foveae in a weak transverse sulcus, evanescent medially; anterior lobe set with stiff erect setae; poste- rior lobe with numerous shallow depressions, humeri not prominent, broadly rounded posteriorly, not modified, covering metanotum. Pronotum of alate form longer, rounded posterior margin extending far- ther caudad, almost triangular, humeri more promi- nent. Thoracic venter not diagnostic, with weakly formed tubercles on either side of mesosternal mid- line on posterior margin opposing an unmodified metasternum. Metasternal scent gland opening (om- POLHEMUS & POLHEMUS: Microveliinae from New Guinea phalium) not evident; scent channels prominent, an- gled slightly anterad to base of metacetabulae. Abdomen without silvery setae or other adorn- ment, except weak longitudinal pruinose line on mid- line of tergites IV-VII; prominent paired longitudinal carinae on tergite II of macropters (visible only after removal of wings), absent in micropterous form. Abdominal sternites set off from laterosternites by hair-free round or oval lacunae. Micropters with el- ongate white wing pads, often infuscated distally; macropters with four closed cells in dark hemelytra with white region basally; venation of fore and hind wings similar to figs. 312-313 in Andersen (1982); fe- male fore wing Sc often terminating in a triangular thickened region, resulting in a slight to pronunced sinuosity of the wing margin; fore wings set with long dark stiff erect setae on basal half, much more pro- nounced in females. Legs slender; anterior femur set beneath with short light setae, sometimes slightly modified in males (figs. 11-13); anterior tibia with a distal comb of many stiff setae; middle and anterior femora set ventrally with numerous long slender setae; middle tibia set with an evenly spaced row of about 10 erect setae beginning at basal third, very long basally, decreasing in length distally; middle tarsi extremely long (fig. 9); claws short; both up- and down curving arolia evident, but very slender. Male genital segments small, not protruding, un- modified; proctiger unmodified; parameres small, slender symmetrical (figs. 14-16). Female tergite VIII on same plane as VII, truncate posteriorly; first gono- coxae small, barely exposed, plate-like; tergite IX small, button-like, protruding posteriorly. Type-species: Tarsovelia alta sp. n. Remarks Comparative notes. — Tarsovelia is similar in gener- al facies to Microvelia, but is easily separated from this genus and all other Microveliinae by the elongate middle tarsi (fig. 9), which are adapted for rowing rat- her than running. In addition, the light colored mar- kings on the hemelytra of Tarsovelia are restricted to the basal angles, plus occasionally a weak medial fas- cia, whereas in the type-species of Microvelia (M. pul- chella Westwood) the hemelytra also bear ovate light spots in every closed cell, plus a large bright spot in the open distal cell. Biological notes. — Tarsovelia species are found along the margins of flowing pools on rocky moun- tain streams. They are swift and agile skaters if dis- turbed, and will fly readily if captured in a net. Etymology. — The generic name Tarsovelia is de- rived from tarsos (Gr.), referring to the extremely long 63 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 slender middle tarsi, and Velia, the nominate genus of the family. Gender feminine. Distribution. — New Guinea (fig. 29). Key to the species of Tarsovelia Males l. Fore femur with patch of short, stiff hairs on ventralisurface (Gps. Ul, 12) cet ene 2 — Fore femur lacking patch of short, stiff hairs on ventral surface (fig. 13) … T. alta sp. n. 2. Fore femur with patch of short, stiff hairs near middle of ventral margin (fig. 11) … ennen CROCE cene Di i Si eh T. arfak sp. n. — Fore femur with patch of short, stiff hairs near base of ventral margin (fig. 12) ..... T. dani sp. n. Tarsovelia alta sp. n. (figs. 9, 13, 15, 29) Type material: Holotype, micropterous male: Papua New Guinea, Morobe Prov., bog nr. Mt. Kaindi summit, above Wau, 17 Oct. 1983, CL 1825, D. A. and J. T. Polhemus (BPBM). — Paratypes: 13 mi- cropterous males, 14 micropterous females, 16 imma- tures, same data as holotype (ZMUC, JTPC, BPBM). Additional material examined: 2 macropterous fe- males, Papua New Guinea, Morobe Prov., Eddie Creek, 2050 m, 11 Oct. 1966, Illies leg. (ZBSM, JTPC); 1 macropterous female, Morobe Prov., Poverty Creek, Mt. Missim, 18 Sept. 1983, CL 1830, 1600 m, J. T. & D. A. Polhemus (JTPC); 9 macropterous males, 6 macropterous females, 2 immatures, West- ern Highlands Prov., rocky stream 17 km N. of Mt. Hagen on Baiyer River road, 6 and 8 Sept. 1983, CL 1780, D. A. and J. T. Polhemus (JTPC). Diagnosis Size. — Micropterous male, length 2.41-2.48 mm = 2.44, n = 10); width 0.83-0.89 mm (x = 0.85,n 10). Micropterous female, length 2.81-2.88 mm (x 2.84, n = 4); width 0.94-1.01 mm, (x = 0.99, n = 4). Macropterous male, length 2.77-2.95 mm (x = 2.87, n = 9); width 1.10-1.15 mm (x = 1.14, n = 9). Macropterous female, length 3.31-3.45 mm (x = 3.36, n = 6); width 1.19-1.33 mm, (x = 1.28, n= 6). Colour. — Micropterous male: ground colour orange brown, venter slightly lighter. Head dark orange brown, lighter beneath; rostrum luteous medi- ally, piceous distally. Pronotum with anterior trans- verse band orange brown, extending onto propleura; disc dark brown, mottled with lighter brown. Abdomen orange brown, tergal margins darker, light- er ventrally. Distal segements of legs, antennae brown, basal segments lighter. Structural characters. — Micropterous male: head x 64 of moderate length, declivant anteriorly; length 0.37; width of eye/interocular space, 0.14/0.30. Pronotum long, covering metanotum, humeri evident but not raised; length:width, 0.84 : 0.86. Abdominal tergites not shining; lengths of tergites, II : 0.14; III : 0.16; IV-VI : 0.19; VII : 0.26. Abdominal venter not modified, set with short ap- pressed setae. Legs, antennae thickly clothed with short to moderate length setae, with scattered longer setae. Legs unarmed. Antennal formula I: Il: III : IV; 0.29 : 0.33 : 0.37 : 0.42. Proportions of legs as follows: Femur, tibia, tarsal 1, tarsal 2 of fore leg, 0.65 : 0.63 : 0.28 : 0.0; of mid- dle leg, 0.98 : 0.93 : 0.42 : 0.35; of hind leg, 0.86 : 1.00 : 0.19 : 0.29. Paramere small, long, slender (fig. 15). Micropterous female: Similar to male in most re- spects, but somewhat larger; connexiva reflexed over dorsum except in gravid specimens. Macropterous male: Similar to micropterous form in most respects. Pronotum with anterior lobe entire- ly orange brown, extending onto propleura; posterior lobe uniformly dark. Hemelytra blackish brown, ba- sally entirely white except slight infuscation of Sc, white region extending beyond tip of pronotum. Macropterous female: Very similar to macropter- ous male, except lateral hemelytral margin slightly sinuate at distal thickening of Sc. Posterior lobe of pronotum with two (1+1) tufts of erect stiff black se- tae on either side of midline at about middle. Connexival margin vertical. Remarks Comparative notes. — 7. alta sp. n. may be separa- ted from 7. dani sp. n. and 7: arfak sp. n. by the un- modified male fore femur (fig. 13), brownish colora- tion in wingless forms, male paramere shape (fig.15), and length of the second segment of the hind tarsi, which is about 1.5 times as long as the first (versus only slightly longer in the other two species). In 7: arfak sp. n. the anterior femora are dark distally, and the middle and posterior femora plus all the tibiae and tarsi are dark brown to black, while in 7: alta sp. n. and 7. dani sp. n. all femora are largely light brown, and the tibiae and tarsi are dark brown but not black. In the macropters of 7. alta sp. n. and 7. arfak sp. n. the entire basal angle of the hemelytra is white (including the veins, although these are occa- sionally slightly darkened), whereas in 7: dani sp. n. the veins are dark and the light coloring is often a sor- did white. We originally considered the micropterous and macropterous specimens at hand from Papua New Guinea to represent different taxa. However with two additional new species now available from Irian Jaya * that show marked differences from the Papua New Guinea material, we have now concluded that the lat- ter represent but a single, albeit somewhat variable, species. We have thus included all of the specimens from eastern New Guinea under 7: alta sp. n., but the macropters are not named as paratypes. Macropte- rous specimens from Eddie Creek and Poverty Creek (not measured) have very long hemelytra, consider- ably exceeding the tip of the abdomen, and they are lighter in color than specimens from Mt. Hagen. Measurements of leg and antennal segments are not helpful, as they appear to vary randomly between populations, thus we defer the final decision about the specific distinctness of these populations from Papua New Guinea until more material, hopefully in- cluding both micropters and macropters in the same sample, is available. Biological notes. — Near Mt. Hagen 7. alta was found in a plunge pool below a small cascade, with individuals hiding in the crevices of a sheer rock wall. When disturbed the insects would skate rapidly over the water near the wall until disappearing into their hiding places once more. By contrast, the Mt. Kaindi specimens were taken from beneath a large, partly submerged log in a bog pool, where violent flushing was required to dislodge them. So far this species has been found only at altitudes above 1200 meters. Etymology. — The name ‘alta’ refers to the high el- evation at which this species was taken. Distribution. — Eastern New Guinea (fig. 29). Tarsovelia arfak sp. n. (figs. 11, 16, 29) Type material: Holotype, macropterous male: Indonesia, Irian Jaya Prov., Saumarin River, nr. Warkomi, Arfak Mountains, 42 km S. of Manokwari, 90 m, water temp. 25° C, 18 Oct. 1991, CL 2647, D. A. and J. T. Polhemus (USNM). — Paratypes: 67 macropterous males, 48 macropterous females, 6 immatures, same data as holotype (JTPC). Diagnosis Size. — Macropterous male, length 2.45-2.77 mm (x = 2.62, n = 10); width 1.01-1.12 mm (x = 1.06, n = 10). Macropterous female, length 2.81-3.09 mm (x = 2.98, n= 10); width 1.15-1.19 mm, (x = 1.17, n= 10). Colour. — Macropterous male: ground colour black, tinged with brown. Head black. Pronotum with anterior lobe and collar medially orange brown, extending to inner eye margins; posterior lobe uni- formly black. Hemelytra blackish brown, long, ex- tending beyond tip of abdomen, basally with broad elongate rectangular white region along outer margin, POLHEMUS & POLHEMUS: Microveliinae from New Guinea extending beyond tip of pronotum. Abdomen black, ventrally tinged with brown along connexival mar- gins and caudally. Legs, antennae brown, fore femur yellowish except distally, mid and hind femora yel- lowish ventrally. Structural characteristics. — Macropterous male: head of moderate length, declivant anteriorly; length 0.40; width of eye/interocular space, 0.12/0.33. Pronotum long, humeri evident but not raised; length:width, 0.86 : 1.00. Abdominal venter not modified, set with short ap- pressed setae. Legs, antennae thickly clothed with short to moderate length setae, with scattered longer setae; all tibia, middle and hind tarsi set with rows of long setae dorsally, in addition to long ventral setae. Legs unarmed, except fore femur ventrally slightly ex- cavate basally, set near middle with patch of short stiff setae (fig. 11). Antennal formula I: II : III : IV; 0.28 : 0.33 : 0.35 : 0.40. Proportions of legs as follows: Femur, tibia, tarsal 1, tarsal 2 of fore leg, 0.65 : 0.51 : 0.28 : 0.0; of mid- dle leg, 0.88 : 0.81 : 0.49 : 0.35; of hind leg, 0.81 : 0.95 : 0.23 : 0.28. Paramere small, long, slender (fig. 16). Macropterous female: Similar to male in most re- spects, but somewhat larger; hemelytra with lateral margin sinuate, thickened and set with long dense black setae at distal Sc. Connexiva vertical along ter- gites II-IV, slightly reflexed over tergites V-VII. Remarks Comparative notes. — Recognized by the patch of short stiff black setae on the ventral margin of the ma- le fore femur (fig. 11), and the shape of the male pa- ramere (fig. 16). For additional comparative notes see key, and discussion under 7: alta sp. n. Biological notes. — The Saumarin River at the type locality was a moderate sized stream dropping steeply to the sea off the flanks of the Arfak Mountains through a bed of boulders, cobbles and sand, heavily shaded by primary rain forest. The stream profile was punctuated by numerous small waterfalls, and 7. ar- fak was most abundant on the plunge pools below these falls. Etymology. — The name ‘arfak is a noun in appo- sition and refers to the Arfak Mountain type area. Distribution. — Western New Guinea (Vogelkop Peninsula) (fig. 29). Tarsovelia dani sp. n. (figs. 10, 12, 14, 29) Type material: Holotype, macropterous male: Indonesia, Irian Jaya Prov, swift rocky stream in 65 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 66 upper Pass Valley, 52 km NE of Wamena, 2015 m., ‘ water temp. 14°C, 23 Sept. 1991, CL 2614, D. A. and J. T. Polhemus (USNM). - Paratypes: 61 macrop- terous males, 26 macropterous females, 28 imma- tures, same data as holotype (JTPC). Diagnosis Size. — Macropterous male, length 2.55-2.95 mm (x = 2.78, n= 10); width 1.01-1.15 mm (x = 1.09, n 10). Macropterous female, length 2.99-3.45 mm (x = 3.12, n = 10); width 1.15-1.37 mm, (x = 1.22, n = 10). Colour. — Macropterous male: Ground colour orange brown, marked with black. Head blackish brown dorsally, ventrally lighter. Pronotum with ante- rior lobe and collar orange brown, extending onto propleura, collar marked only laterally, demarcated by a row of dark foveae; posterior lobe uniformly black. Hemelytra blackish brown, basally with narrow elon- gate white stripe along costal margin, extending be- yond tip of pronotum, plus a shorter sordid white re- gion basally along pronotum; with elongate grayish brown streak in middle, barely visible in some speci- mens; length variable, shortest reaching barely onto tergite VII, longest almost covering genital segments. Abdomen orange brown. Legs, antennae brown, fore femur yellowish except distally, mid and hind femora, tibia yellowish ventrally. Structural characteristics. — Macropterous male: head of moderate length, declivant anteriorly; length 0.38; width of eye/interocular space, 0.14/0.30. Pronotum long, humeri evident but not raised; length:width, 1.00 : 1.12. Abdominal venter not modified, set with short ap- pressed setae. Legs, antennae thickly clothed with short to moderate length setae, with scattered longer setae. Legs unarmed, except fore femur flattened ven- trally, curved, basally with a tuft of stiff setae (fig. 12). Antennal formula I : II : III : IV; 0.30 : 0.35 : 0.42 : 0.42. Proportions of legs as follows: Femur, tibia, tarsal 1, tarsal 2 of fore leg, 0.74 : 0.72 : 0.30 : 0.0; of mid- dle leg, 1.12 : 1.02 : 0.49 : 0.35; of hind leg, 0.98 : 1.07 : 0.26 : 0.33. Paramere small, long, slender (fig. 14). Macropterous female: Similar to male in most re- spects, but somewhat larger. Hemelytra length vari- able, shorter than in male, shortest not reaching ter- gite VII, longest reaching caudal margin of tergite POLHEMUS & POLHEMUS: Microveliinae from New Guinea VII; with lateral margin weakly sinuate, narrowly thickened and set with stiff setae at distal Sc. Connexiva vertical along tergites II-III, IV in transi- tion, strongly reflexed over tergites V-VII. Remarks Comparative notes. — Recognized by the small tuft of black setae basally on the ventral surface of the we- akly bowed male fore femur (fig. 12), and the shape of the male paramere (fig. 14). For additional compa- rative notes see key, and discussion under 7! alta sp. n. Biological notes. — The type series of 7. dani was taken from pools along a very swift, cold mountain stream in a limestone bed bordered by Nothofagus and Rhododendron forest. The insects were most common along the pool margins, skating rapidly across the wa- ter if disturbed and flying readily if taken in a net. Etymology. — The name ‘dan? is a noun in apposi- tion and refers to the native Dani people who inhabit the type area. Distribution. — Central New Guinea (fig. 29). Neusterinsifer gen. n. (figs. 17-26, 30) Diagnosis Size. — Form stout (fig. 17), length of males 1.87- 2.48 mm, females 1.91-2.77 mm, general body char- acteristics and size sexually dimorphic, males smaller, female abdomen more robust. Colour. — Ground color black, tinged with orange brown, heavily marked with silvery pubescence; ante- rior pronotal lobe orange brown on posterior half, with streak extending from near midline to middle of eyes, weakly demarcated from collar and posterior lobe. Structural characteristics. — Both apterous and macropterous forms known. Eyes globose, exserted but usually not beyond anterolateral pronotal angles, separated by about twice the width of an eye, ap- pressed to anterior pronotal margin, with moderately long ocular setae. Head strongly declivant anteriorly, recessed into pronotum, posterior margin almost ver- tical, with usual three pairs of facial trichobothria; gu- lar region short, not visible, rostral cavity open poste- riorly, with raised carinate bucculae. Rostrum reaching to middle of mesosternum, segment I short, enclosed in rostral cavity, I and IV subequal in length Figs. 17-26. Neusterinsifer gen. n. — 17. Neusterinsifer compactus sp. n., micropterous male, dorsal habitus; 18-22. Processes arising from male proctigers of Neusterinsifer species, 18. Neusterinsifer compactus sp. n.; 19. Neusterinsifer cyclops sp. n.; 20. Neusterinsifer sepik sp. n.; 21. Neusterinsifer nabire sp. n.; 22. Neusterinsifer gladius sp. n.; 23. Neusterinsifer compactus sp. n., posterior view of female abdomen showing anal plate formed by tergites VII and VIII. 24-26. Male forelegs of Neusterinsifer species: 24. Neusterinsifer sepik sp. n.; 25. Neusterinsifer gladius sp. n.; 26. Neusterinsifer compactus sp. n. 67 TiJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 and about two times longer than II, segment III about four times as long as II. Antennae slender, very long, about 2/3 of body length. Pronotum almost flat, without median carina; col- lar very weakly formed, set off by a ragged row of small dark foveae, terminating under eyes laterally; anterior and posterior lobes set off by a transverse row of foveae; anterior lobe thickly set with appressed sil- very setae except triangular median area; posterior lobe with numerous small foveae, humeri not formed, almost straight posteriorly, not modified, covering mesonotum. Metanotum almost as long as prono- tum. Thoracic venter not diagnostic, with weakly formed tubercles on either side of mesosternal mid- line on posterior margin opposing an unmodified metasternum. Metasternal scent gland opening (om- phalium) not visible, but marked by a tiny tubercle; scent channels prominent, curving slightly anterad to base of metacetabulae. Abdomen with silvery setae laterally on tergites II, VII, VIII, covering most of tergites II-VI in females, less extensive in males; prominent paired longitudinal carinae on tergite II of macropters (visible only after removal of wings), absent in micropterous form. Abdominal sternites set off from laterosternites by small hair-free glabrous oval lacunae; laterosternite VII of both sexes set with small round tuft of stiff pos- terolaterally directed setae, near connexival margin at basal 2/5 in males, much closer to base in females. Macropterous forms with four closed cells in dark hemelytra, occasionally with fifth distal cell closed; al- ways with white elongate streak basally in outer basal cell, usually with faint to pronounced light areas in other cells; venation of fore and hind wings similar to figs. 312-313 in Andersen (1982). Legs moderately stout, moderately long; anterior femur set beneath with short light setae, modified in males of some species, with a patch of stiff setae or denticles ventrally on basal half; anterior tibia of male with a comb of minute black setae occupying 2/5 to 1/2 of tibial length; all femora set ventrally with 7-12 long slender setae; all tarsi moderately long, claws moderately long; both up- and down curving arolia long, slender. ‘ Male abdominal ventrites VI, VII modified; genital segments very large, protruding, highly modified, proctiger bearing a large sword-like appendage arising from anterior margin, directed antero-laterally be- neath abdomen (figs. 18-22); segment IX, proctiger rotated by about 90°; parameres vestigial or absent. Female tergite VIII on same plane as VII basally, dis- tally sharply deflected ventrad closing abdomen pos- teriorly except for protruding tip of proctiger (fig. 23); first gonocoxae hidden. Type-species: - Neusterinsifer compacta sp. n. 68 Remarks Comparative notes. — Very similar in general facies to Microvelia (fig. 17), but separated by sword-like projection on the male proctiger (figs. 18-22), the small tuft of stiff setae on laterotergite VII, and other characters as given in the generic key. Macropters differ from Microvelia and most other genera of Microveliinae in having a well defined long slender white streak basally in the outer cell (with only ill de- fined light areas in all other cells), and lacking the dis- tal white spot that is characteristic of Microvelia and other closely related genera. The wing markings are similar to those of Tenagovelia Kirkaldy from Africa, but species in this latter genus are much larger and possess much longer legs. Biological notes. — Species of Neusterinsifer are most commonly encountered in sheltered areas along the margins of stream pools, or on the surfaces of wa- ter filled potholes in bedrock exposures adjacent to flowing sections of streams. Etymology. — The generic name Neusterinsifer is derived from neuster (Gr.), swimmer, and ensifer (L.), sword bearer. Gender masculine. Distribution. — New Guinea (fig. 30). Key to species of Neusterinsifer 1. Male process on proctiger bifurcate, short (fig. 21). Female abdomen strongly narrowed posteri- orly,, boat shaped) -:.c-12-.eeeee eee nabire sp. n. — Male process on proctiger not bifurcate, longer (- figs. 18-20, 22). Female abdomen not strongly narrowed posteriorly, not boat shaped ............ 2 2. Small species, male 1.87 mm, female 1.91 mm long. Male process on proctiger relatively short, broad, narrowly triangular distally (fig. 19 ......... decade cyclops sp. n. — Larger species, male at least 2.19 mm, female at least 2.27 mm long. Male process on proctiger long, relatively narrow, not triangular distally (- figs. 18, 20,22) ian e enr 3 3. Relatively small species, males 2.19-2.27 mm, fe- males 2.27-2.34 mm long. Process of male proc- tiger long, broad, expanded medially and distally (fig. 22); male fore femur with a row of 5-10 black spinules basally (fig. 25). Female connexival segment VII forming a small acute angle at level of caudal margin of tergite VII ...... gladius sp. n. — Larger species, males 2.27-2.48 mm, females 2.37-2.77 mm long. Process of male proctiger long, narrow or broad, not expanded medially and distally (figs. 18, 20); male fore femur with- out a basal row of 5-10 black spinules, instead with either a basal row of many short stiff black setae, or unarmed. Caudal extreme of female con nexival segment VII either forming an acute or obtuse angle, but without a small acute angle at level of caudal margin of tergite VII … … … … 4 4. Process of male proctiger long, slender, sharp dis- tally (fig. 18); male fore femur strongly bowed, bearing small black denticles basally (fig. 26). Caudal extreme of female connexival segment VII forming an obtuse angle ..... compactus sp. n. — Process of male proctiger long, broad (fig. 20); male fore femur not bowed, with a patch of black spinules basally (fig. 24). Caudal extreme of fe- male connexival segment VII slightly produced posteriorly, forming an acute angle ... sepik sp. n. Neusterinsifer compactus sp. n. (figs. 17, 18, 23, 26, 30) Type material: Holotype, apterous male: Papua New Guinea, Western Highlands Prov., Baiyer River, 53 km N. of Mt. Hagen, 6 and 8 Sept. 1983, D. A. and J. T. Polhemus (BPBM). - Paratypes: 19 apterous males, 12 apterous females, same data as holotype (TPC); 1 apterous male, 3 apterous females, Western Highlands Prov., Baiyer River Bird Sanctuary, 8 Sept. 1983, CL 1792, J. T.& D. A. Polhemus (JTPC); 1 ap- terous male, Morobe Prov., Bamboo Creek, nr. Wau, 16 Sept. 1983, CL 1817, J. T. & D. A. Polhemus (TPC); 4 apterous males, 1 apterous female, Morobe Prov., trib. to Bumbu River, N. of Lae, 20 Sept. 1983, CL 1836, J. T. & D. A. Polhemus (JTPC). Diagnosis Size. — Apterous male, length 2.37-2.48 mm (x = 2.43, n = 10); width 1.33-1.40 mm (x = 1.39, n = 10). Apterous female, length 2.63-2.77 mm (x = 2.70, n = 10); width 1.40-1.55 mm, (x = 1.50, n = 10). Colour. — Apterous male: Ground colour blackish brown, venter slightly lighter, entire dorsum, latero- tergites covered with fine appressed golden pubes- cence. Head black, orange brown beneath; rostrum fuscous medially, piceous distally. Pronotum with an- terior transverse orange brown band barely interrupt- ed medially; disc blackish brown. Abdomen blackish brown, lighter ventrally, connexiva margined with orange brown. Antennae brown; anterior legs yellow- ish, distally darkened; middle, hind legs dark brown, basally lighter; coxae, trochanters luteous. Structural characters. — Apterous male: head of moderate length, declivant anteriorly, with weak im- pressed median line; length 0.40; width of eye/inter- ocular space, 0.15/0.30. Pronotum long, covering metanotum, thickly set with obscure foveae, humeri depressed; length:width, 0.40 : 1.00. Metanotum length:width, 0.33 : 1.00. Abdominal tergites not POLHEMUS & POLHEMUS: Microveliinae from New Guinea shining; tergites II-VI subequal in length (0.14-0.16), VII longer (0.28). Abdominal venter set with short appressed setae, and scattered longer setae caudally; venter VI posteri- orly set with two (1+1) tufts of dense dark erect setae on either side of midline; venter VII broadly de- pressed medially, carinate along posterior margin. Legs, antennae thickly clothed with short to moderate length setae, with scattered longer setae. Middle and hind legs unarmed; fore femur strongly bowed, bear- ing a patch of black denticles basally, fore tibia with grasping comb extending 3/4 the length of the tibia (fig. 26). Antennal formula I: II : III: IV; 0.35 : 0.28 : 0.40 40551 Proportions of legs as follows: Femur, tibia, tarsal 1, tarsal 2 of fore leg, 0.65 : 0.60 : 0.28 : 0.0; of mid- dle leg, 0.86 : 0.86 : 0.16 : 0.30; of hind leg, 0.93 : 1.07 : 0.17 : 0.30. Proctiger produced anteriorly into a long slender sinuate ensiform process (fig. 18). Parameres vestigial or absent. Micropterous female: Similar to male in most re- spects, but somewhat larger; connexiva vertical, in- curved and slightly refexed at caudal extreme, which forms an obtuse angle. Remarks Comparative notes. — A relatively large species for the genus, recognized by the short, blunt process on the male proctiger (fig. 18), and the strongly bowed male foreleg with a patch of black denticles basally on the ventral surface (fig. 26). Similar in size and gener- al facies to N. sepik sp. n., but this latter species pos- sesses an unbowed foreleg with a large patch of black denticles (fig. 24), and a much larger and more poin- ted process on the male proctiger (fig. 20). Biological notes. — The Baiyer River at the type lo- cality was a swift, deep river flowing in a rocky bed with wet mossy banks, and partially shaded by dis- turbed upland rain forest. N. compactus was taken here skating on the surface of water filled potholes in an exposure of igneous rock along the river margin. Etymology. — The name compactus (L.) refers to the shape of this species. Distribution. — Eastern New Guinea (fig. 30). Neusterinsifer cyclops sp. n. (figs. 19, 30) Type material: Holotype, apterous male: INDONE- SIA, Irian Jaya Prov., small rocky stream and spring above Sentani, Cyclops Mountains, 325 m, 25 Sept. 1991, CL 2618, J. T. & D. A. Polhemus (JTPC, to be placed in USNM when more material is available). 69 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Diagnosis Size. — Apterous male, length, x = 1.87, n = 1; width, x = 0.79, n = 1. Apterous female, length, x = 1.91, n= 1; width, x = 0.90, n= 1. Colour. — Apterous male: Ground colour deep brown, marked with brown, venter slightly lighter, entire dorsum, laterotergites covered with fine ap- pressed golden pubescence; entire dorsum, pleura, sides of abdomen covered with moderate length erect setae. Head brown, orange brown beneath; rostrum fuscous medially, piceous distally. Pronotum with an- terior transverse orange brown band embrowned me- dially; disc blackish brown. Abdomen blackish brown, lighter ventrally, connexiva margined with orange brown. Antennae brown, base of segment I lighter; legs yellowish, distally darkened; femora dark- er posteriorly; coxae, trochanters luteous. Structural characters. — Apterous male: head of moderate length, declivant anteriorly, with weak im- pressed median line; length 0.33; width of eye/inter- ocular space, 0.12/0.30. Pronotum long, covering metanotum, thickly set with obscure foveae, humeri depressed; length:width, 0.28 : 0.79. Metanotum length:width, 0.26 : 0.79. Abdominal tergites not shining; tergites II-VI subequal in length (0.12-0.14), VII longer (0.26). Abdominal venter set with short appressed setae; venter VI posteriorly set with two (1+1) tufts of dense dark erect setae on either side of midline, between them a V-shaped depression open posteriorly; venter VIT broadly depressed medially, sharply excavated on posterior half, carinate along anterior margin of exca- vation. Legs, antennae thickly clothed with short to moderate length setae, with scattered longer setae. Middle and hind legs unarmed; foreleg lacking denti- cles or spinules, tibial comb extending for only 1/2 the length of the tibia. Antennal formula I : II : III : IV; 0.21 : 0.19 : 0.33 : 0.40. Proportions of legs as follows: Femur, tibia, tarsal 1, tarsal 2 of fore leg, 0.51 : 0.44 : 0.19 : 0.0; of mid- dle leg, 0.63 : 0.60 : 0.12 : 0.19; of hind leg, 0.74 : 0.86 : 0.12 : 0.26. Proctiger produced anteriorly into a broad ensif- orm process (fig. 19). Parameres vestigial or absent. Apterous female: Similar to male in most respects, but somewhat larger; connexiva vertical; caudal ex- treme gently curving ventrad. Macropterous forms: unknown. Remarks Comparative notes. — Separable from all other Neusterinsifer species by its small size, and short, poin- ted process on the male proctiger (fig. 19). 70 Biological notes. — The type of N. cyclops was taken from a small, clear, spring fed stream at the margin of the remaining primary rain forest on the slopes of the Cyclops Mountains above Sentani. The stream flowed in a bed of metamorphic rocks intermixed with coarse grey sand, and was partially shaded by trees felled for garden clearings. Etymology. — The name cyclops, a noun in apposi- tion, refers to the mountain range where the species was found. Distribution. — Northern New Guinea (Cyclops Mountains) (fig. 30). Neusterinsifer gladius sp. n. (fig. 22, 25, 30) Type material: Holotype, apterous male: INDONE- SIA, Irian Jaya Prov., Nabire River and swift tributary, 5 km E. of Nabire, 60 m, 14 Oct. 1991, CL 2642, J. T. & D. A. Polhemus (USNM). - Paratypes: 2 apterous males, 2 apterous females, same data as holotype (JTPC). Diagnosis Size. — Apterous male, length 2.19-2.27 mm (x = 2.24, n = 3); width 0.90-0.94 mm (x = 0.91, n = 3). Apterous female, length 2.27-2.34 mm (x= 2.30, n = 2); width 1.01-1.04 mm, (x = 1.03, n = 2). Colour. — Apterous male: Ground colour blackish brown, venter slightly lighter, entire dorsum, latero- tergites covered with fine appressed golden pubes- cence; entire body except thoracic venter covered with long erect setae. Head black, orange brown be- neath; rostrum fuscous medially, piceous distally. Pronotum with anterior transverse orange brown band sometimes interrupted medially; disc blackish brown. Abdomen blackish brown, lighter ventrally, connexiva margined with orange brown. Antennae brown; anterior femora luteous, distally darkened; anterior tibia, tarsi brown, darker distally; middle, hind legs brown dorsally, ventrally yellowish, femora basally lighter; coxae, trochanters luteous, tinged with fuscous. Structural characters.- Apterous male: head of mo- derate length, declivant anteriorly, with weak im- pressed median line; length 0.35; width of eye/inter- ocular space, 0.15/0.28. Pronotum long, covering metanotum, thickly set with obscure foveae, humeri depressed; length:width, 0.37 : 0.86. Metanotum length:width, 0.28 : 0.86. Abdominal tergites not shining; tergites II-VI subequal in length (0.12-0.14), VII longer (0.28). Abdominal venter set with short appressed setae; venter VI posteriorly set with two (1+1) widely separ- ated tufts of dense dark erect setae on either side of POLHEMUS & POLHEMUS: Microveliinae from New Guinea Fig. 27. Distribution of Tanyvelia missim, gen. n., sp. n. CE ee LE Le ae NT E PAUL Fly Ld m 3] 4 | | J N U N N A, I en ac ERGAB EN ) Ne 7 Fig. 28. Distribution of Aegilipsicola rapida, gen. n., sp. n. 2 New Guinea RES Red 192 na ue 1 midline, with shallow hair free sulcus between, ex- tending the length of ventrites V-VI; venter VII broadly and strongly depressed medially. Legs, anten- nae thickly clothed with short to moderate length se- tae, with scattered longer setae. Middle and hind legs unarmed; fore tibia bearing a row of 5-10 black spi- nules basally; fore femur with grasping comb extend- ing 1/2 the length of the tibia (fig. 25). Antennal formula I: II: III: IV; 0.35 : 0.28 : 0.35 :0.47. Proportions of legs as follows: Femur, tibia, tarsal 1, tarsal 2 of fore leg, 0.60 : 0.51 : 0.28 : 0.0; of mid- dle leg, 0.79 : 0.74 : 0.16 : 0.30; of hind leg, 0.84 : 1.02 : 0.16 : 0.30. Proctiger produced anteriorly into a long broad sinuate ensiform process, expanded medially and dis- tally (fig. 22). Parameres vestigial or absent. Apterous female: Similar to male in most respects, but somewhat larger; connexiva vertical; caudal ex- treme of VII forming an acute angle; VIII gently curving ventrad. Macropterous forms: Unknown. Remarks Comparative notes. — Similar in size to N. nabire 138 wo 142 wa e me 1 1 IT 1 st n + sp. n., but easily separable by the elongate process on the male proctiger (fig. 22), the patch of stiff, black spinules basally on the ventral surface of the male fo- re femur (fig. 25), and the relatively unmodified fe- male abdomen which is not laterally compressed or boat-shaped. Biological notes. — The Nabire River at the type lo- cality was a swift, clear, moderate sized stream flow- ing in a rocky bed, and shaded by primary lowland rain forest. The type series of N. gladius was taken downstream of a limestone waterfall, where the banks of the river included numerous boulders and bedrock exposures. The insects were taken from the surfaces of water filled potholes and on still pools amid the rocks. Etymology. — The name gladius (L.), refers to the ensiform process of the male proctiger. Distribution. — Western New Guinea (fig. 30). Neusterinsifer nabire sp. n. (figs. 21, 30) Type material: Holotype, apterous male: INDONE- SIA, Irian Jaya Prov., Nabire River and swift tributary, 5 km E. of Nabire, 60 m, 14 Oct. 1991, CL 2642, J. T. & D. A. Polhemus (USNM). - Paratypes: 98 apter- ous males, 13 macropterous males, 72 apterous fe- TA TIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 males, 35 macropterous females, 32 immatures, same data as holotype (JTPC). Diagnosis Size. — Apterous male, length 2.01-2.09 mm (x = 2.04, n = 10); width 0.79-0.83 mm (x = 0.81, n = 10). Apterous female, length 2.19-2.41 mm (x = 2.27, n = 10); width 0.79-0.90 mm (x = 0.83, n = 10). Macropterous male, length 2.23-2.34 mm (x = 2.29, n = 6); width 1.04-1.12 mm (x = 1.10, n = 6). Macropterous female, length 2.41-2.55 mm (x = 2.46, n = 6); width 1.08-1.15 mm, (x = 1.12, n = 6). Colour. — Apterous male: Ground colour blackish brown, venter slightly lighter, entire dorsum, latero- tergites covered with fine appressed inconspicuous pubescence; dorsum, body laterally covered with long erect setae. Head black, orange brown beneath; ros- trum fuscous medially, piceous distally. Pronotum with anterior transverse orange brown band broadly interrupted medially; disc blackish brown. Abdomen blackish brown, slightly lighter ventrally, connexiva margined with orange brown. Antennae brown; ante- rior femora yellowish, distally darkened, tibia brown; middle, hind legs dark brown to piceous, femora ba- sally lighter; coxae, trochanters luteous, tinged with fuscous. Structural characteristics. — Apterous male: head of moderate length, declivant anteriorly, with weak im- pressed median line; length 0.33; width of eye/inter- ocular space, 0.14/0.28. Pronotum long, covering metanotum, thickly set with obscure foveae, humeri depressed; length:width, 0.37 : 0.81. Metanotum length:width, 0.28 : 0.79. Abdominal tergites not shining; tergites II-VI subequal in length (0.12-0.16), VII longer (0.30). Abdominal venter set with short appressed setae; venter V posteriorly set with two (1+1) tufts of dense dark erect setae on either side of midline; venter VI with caudal margin medially produced, set with a large quadrate tuft of dense dark erect setae; venter VII excavated medially along posterior margin. Legs, antennae thickly clothed with short to moderate length setae, with scattered longer setae. Middle and hind legs unarmed; fore femur lacking denticles or spinules; fore tibia with grasping comb extending 1/2 the length of the tibia. Antennal formula I : II: III: IV; 0.30 : 0.21 : 0.28 : 0.44. Proportions of legs as follows: Femur, tibia, tarsal 1, tarsal 2 of fore leg, 0.51 : 0.47 : 0.21 : 0.0; of mid- dle leg, 0.70 : 0.63 : 0.12 : 0.23; of hind leg, 0.74 : 0.93 : 0.16 : 0.28. Proctiger produced anteriorly into a bifurcate pro- cess (fig. 21). Parameres vestigial or absent. Apterous female: Similar to male in most respects, but somewhat larger; abdomen narrowing posterior- ly. Connexiva vertical, touching at caudal extreme; caudally rounded; thickly set along posterior margin with long stout dark setae; tuft of stiff posterolateral- ly directed setae near connexival margin of lateroster- nite VII of females pronounced, more so than in oth- er species. Macropterous male: Similar to apterous male in many respects, but broader, humeri more pro- nounced. Hemelytra always with light streaks in each of the two basal cells, often with light areas in all closed cells. Macropterous female: Similar to apterous female in many respects, but broader, humeri more pro- nounced. Hemelytra with light areas in each of the four basal closed cells. Remarks Comparative notes. — Easily recognized by the bi- furcate process on the male proctiger (fig. 21) and the laterally compressed, boat-shaped abdomen in fema- les. This species is similar in size to N. gladius sp. n., and larger than N. cyclops sp. n., but smaller than eit- her N. compactus sp. n. or N. sepik sp. n. Biological notes. — See discussion under N. gladi- us sp. n. Etymology. — The name ‘rabire, a noun in apposi- tion, refers to the Nabire River type locality. Distribution. — Northern New Guinea (fig. 30). Neusterinsifer sepik sp. n. (figs. 20, 24, 30) Type material: Holotype, apterous male: Papua New Guinea, East Sepik Prov., Yemogu Creek, 2 km W. Tring, 12 Sept: 11983; CLN305 MAD EN Polhemus (BPBM). - Paratypes: 11 apterous males, 6 apterous females, same data as holotype (JTPC); 1 ap- terous male, 4 apterous females, 1 macropterous fe- male, East Sepik Prov., Mandi Creek, E. of Wewak, 10 Sept. 1983, CL 1797, J. T. & D. A. Polhemus (- JTPC); 4 apterous males, 1 apterous female, East Sepik Prov., Brandi Creek, nr. Wewak, 10 Sept. 1983, CL 1796, J. T. & D. A. Polhemus (JTPC). Diagnosis Size. — Apterous male, length 2.27-2.41 mm (x 2.35, n = 10); width 0.94-1.01 mm (x = 0.99, n 10). Apterous female, length 2.37-2.63 mm (x 2.48, n = 10); width 0.97-1.15 mm, (x = 1.04, n 10). Macropterous female, length, x = 2.62, n = 1); width, x= 1.15, n= 1) Colour. — Apterous male: Ground colour blackish brown, venter slightly lighter, entire dorsum and la- POLHEMUS & POLHEMUS: Microveliinae from New Guinea Fig. 29. Distribution of species of Tarsovelia gen. n. — Square = Tarsovelia dani sp. n.; Circles = Tarsovelia alta sp. n.; Triangle = Tarsovelia arfak sp. n. Figure 30. Distribution of species of Neusterinsifer gen. n. — Square = Neusterinsifer sepik sp. n.; Circles = Neusterinsifer compactus sp. n.; Triangle = Neusterinsifer cyclops sp. n.; Diamond = Neusterinsifer nabire sp. n.; Half filled square = Neusterinsifer gla- dius sp. n. e -30 i New Guinea N 274 us mo 12 wa we e mn 4 itn 4 4 4 Mm terotergites covered with fine appressed golden pu- bescence; entire body except thoracic venter covered with long erect setae. Head black, orange brown be- neath; rostrum fuscous medially, piceous distally. Pronotum with anterior transverse orange brown band barely interrupted medially; disc blackish brown. Abdomen blackish brown, lighter ventrally, connexiva margined with orange brown. Antennae brown; anterior legs yellowish, distally darkened; middle, hind legs dark brown to piceous dorsally, ventrally yellowish, femora basally lighter; coxae, tro- chanters luteous, tinged with fuscous. Structural characteristics. — Apterous male: head of moderate length, with weak impressed median lines length 0.27; width of eye/interocular space, 0.14/0.30. Pronotum long, covering metanotum, thickly set with obscure foveae, humeri depressed; length:width, 0.37 : 0.95. Metanotum length:width, 0.28 : 0.95. Abdominal tergites not shining; tergites II-VI subequal in length (0.14-0.16), VII longer (0.30). Abdominal venter set with short appressed setae; venter VI posteriorly set with two (1+1) tufts of dense dark erect setae on either side of midline, with hair free sulcus between; venter VII broadly depressed me- dially, weakly carinate along posterior margin. Legs, antennae thickly clothed with short to moderate length setae, with scattered longer setae. Middle and hind legs unarmed; fore femur bearing an elongate patch of stiff erect black spinules basally; fore tibia with a swelling distally bearing 5-7 erect short stiff se- tae (fig. 24). Antennal formula I: II: III: IV; 0.33 : 0.26 : 0.37 : 0.47. Proportions of legs as follows: femur, tibia, tarsal 1, tarsal 2 of fore leg, 0.65 : 0.53 : 0.28 : 0.0; of middle leg, 0.84 : 0.79 : 0.16 : 0.28; of hind leg, 0.93 : 1.07 0.195028. Proctiger produced anteriorly into a long broad sinuate ensiform process (fig. 20). Parameres vestigial or absent. Apterous female: Similar to male in most respects, but somewhat larger; connexiva vertical; caudal ex- treme forming an acute angle. Macropterous female: Similar to apterous female in many respects, but broader, humeri more pro- nounced. Hemelytra with light areas in each of the four basal closed cells. 78 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Remarks Comparative notes. — Recognized by the large, elongate process on the male proctiger that comes to a rounded point (fig. 20), and the elongate patch of black denticles on the inner surface of the male fore- leg (fig. 24). Biological notes. — Yemogu Creek at the type local- ity was a slow flowing stream running through a rock bed, and containing numerous large pools shaded by slightly disturbed primary rain forest. N. sepik was found here in sheltered areas along the margins of the stream pools, and in similar habitats at Brandi River and Mandi Creek, two lowland rainforest streams en- tering the sea east of Wewak. Etymology. — The name sepik, a noun in apposi- tion, refers to the district of Papua New Guinea from which the type series was taken. Distribution. — Northern New Guinea (fig. 30). ACKNOWLEDGEMENTS We are pleased to acknowledge the invaluable assistance of the following persons who aided us with permits, guidan- ce, transportation and logistics, and without whose help and patience our field surveys in Indonesia could not have been accomplished: Drs. Rosichon Ubaidillah, Dr. Soetikno Wiroatmodjo, Dr. Mohammed Amir, and other personnel from the Indonesian Institute of Sciences (LIPI), Bogor; Mrs. Moertini, Mr. Sanchoyo, the other staff members of the LIPI office in Jakarta; Mr. Benny Lesomar, Natrabu Corp., Biak; Mr. Jacob Bakabessy and Mr. Bernardus Sambery, PHPA, Manokwari; and Mr. Henk van Mastrigt, Jayapura. Special thanks are also due to the following persons who aided us with field surveys in Papua New Guinea: Stan and 74 Kris Niemi, Mt. Hagen; and especially John Ismay, former- ly of the Bureau of Primary Industry, Konedobu, and now at the Hope Museum of Entomology, Oxford. Holotypes of all new species described herein are deposit- ed in the Bishop Museum, Honolulu (BPBM), or the U.S. National Museum of Natural History, Washington, D. C. (- USNM); paratypes are held in the J. T. Polhemus collection, Englewood, Colorado (JTPC), and where numbers permit will be distributed to the institutions noted above as well as the Museum Zoologicum Bogoriense, Bogor, Indonesia (MZB), and the Universitets Zoologisk Museum, Copenhagen (ZMUC). We also thank Dr. Martin Baehr of the Zoogische Sammlung des Bayersichen Staates, Munich (ZBSM) for allowing us to examine specimens held under his care. This research was sponsored by a series of grants (2698- 83, 3053-85, 4537-91) from the National Geographic Society, Washington, D.C., and by grant BSR-9020442 from the National Science Foundation, Washington, D.C. We thank these organizations for their continued support on research into the systematics and zoogeography of aquat- ic Heteroptera. REFERENCES Andersen, N. M. 1982. The semiaquatic bugs (Hemiptera, Gerromorpha). Phylogeny, adaptations, biogeography and classification. — Scandinavian Science Press, Klampenborg, Denmark, Entomonograph 3, 455 pp. Andersen, N. M. 1989. The Old World Microveliinae (Hemiptera: Veliidae). II. Three new species of Baptista Distant and a new genus from the Oriental region. — Entomologica Scandinavica , 19: 363-380. Received: 19 November 1993 Accepted: 25 March 1994 R. ROZKOSNY' & D. KOVAC’ ! Masaryk University, Brno’ Forschungsinstitut Senckenberg, Frankfurt ADUEISANDLARVAE OF TWO PTECTICUS LOEW FROM PENINSULAR MALAYSIA (DIPTERA, STRATIOMYIDAE) Rozkosny, R. & D. Kovac, 1994. Adults and larvae of two Ptecticus Loew from Peninsular Malaysia, (Diptera, Stratiomyidae). — Tijdschrift voor Entomologie 137: 75 - 86, figs. 1-43. [ISSN 0040-7496]. Published 15 July 1994. Descriptions and redescriptions of larvae and adults of P. malayensis sp. n. and P. longipennis (Wiedemann) are given, based on material from Malaysia. The phylogenetic relationships of the two species and the systematic position of the genus based on larval and adult characters are briefly discussed, and biological notes on larvae are provided. The diagnostic characters of the larvae and imagines, including the male genitalia, are illustrated. A generic key to the larvae of the subfamily Sarginae is given. Correspondence: R. Rozkosny, Department of Environmental Studies, Faculty of Science, Masaryk University, Kotlatska 2, 611 37 Brno, Czech Republic. Key words. — Peninsular Malaysia; Stratiomyidae; larvae, key; Ptecticus new species, life histo- ry. The genus Ptecticus Loew belongs to the subfamily Sarginae and differs in the adult stage from the close- ly related genus Sargus Fabricius by the distinctly prominent pedicel on the inner side of the antenna. According to the recent catalogue by James (1975), 31 valid species have been described from the Oriental region but many of them are poorly known. The male genitalia as the most important criterion for species discrimination, for example, have not yet been examined in most Oriental species and the taxonom- ic status of many species needs a re-evaluation. The junior author, while studying insects on bam- boo (Gigantochloa scortechinii Gamble) in Peninsular Malaysia, found larvae of Stratiomyidae living inside bamboo internodes. Following our study on Camptopteromyia fractipennis de Meijere (Rozkosny & Kovac 1991) the present paper represents the sec- ond modern contribution to the taxonomy, morphol- ogy and biology of the Oriental Stratiomyidae. Two species are treated here, both reared from larvae found under semi-natural conditions. The first of them, P. longipennis (Wiedemann), is a well distinguishable species, but we decided on re-description in order to in- clude important diagnostic characters not mentioned in the original description, including the male genitalia. The second species is described here as new; it differs from all known Oriental Precticus species. Larvae and puparia of both species are described for the first time, and their descriptions enable us to propose generic characters for Ptecticus in the larval stage. Ptecticus longipennis (Wiedemann, 1824) (figs. 1-10, 26-32) Diagnosis An easily recognizable, chiefly blue black and yel- low species from the cingulatus-group with conspi- cuously long, and apically darkened wings. Measurements: Body 8.2-11.5 mm, wing 10.8- 13.0 mm. Male. — Head hemispherical, deeply concave poste- riorly, bare eyes touching on frons (figs. 1-2). Upper part of frons elongate, triangular, subshining black and sparsely covered with fairly long brownish hairs. Ocellar triangle relatively narrow and low, barely vis- ible in lateral aspect (fig. 2). Lower frons and face pro- tuberant, white to pale yellow, white frontal callus slen- der, nose-shaped. Face with dense yellow hairs. Antennae yellow, arista brown, both basal segments yellow haired. Pedicel with prominent projection on inner side. Flagellum wider than long, about as long as pedicel. Postocular margin with a complete fringe of erect hairs. Proboscis pale yellow, somewhat swollen but relatively slender, with two-segmented labellae. Thorax bright yellow in ground colour, with exten- sive dark pattern. Mesonotum chiefly shining bluish- black, leaving large anterior corners and postalar calli yellow. Scutellum dark, with margin broadly brown- ish or yellow. Pleura yellow, with large brownish to bluish-black spots on greater part of anepisternum, N UN TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 1-6. Ptecticus longipennis, adult characters. — 1, male head, dorsal view; 2, the same, lateral view; 3, female head, dorsal view; 4, dorsal part of male terminalia; 5, aedeagal complex; 6, ventral part of male terminalia. lower half of katepisternum and anterior parts of ane- pimeron and katepimeron. Laterotergite shining brown and mediotergite behind scutellum complete- ly bluish-black. Narrow subnotopleural stripe, upper margin of katepisternum, and, especially, propleuron pale, often whitish yellow. Wings slightly greyish infuscate, with yellow pte- rostigma and darkened apical part reaching almost half way between apex and discal cell. Dense micro- trichia absent in wing base, alula, subcostal cell and greater part of posterior cubital cell. Anterior cross- vein situated just at beginning of R,,, M, shortened, not reaching wing margin. Narrow squamae white to yellow, pale yellow haired. Halteres yellow, with darkened knob. Legs including coxae chiefly yellow, only hind pair partly darkened. All femora distinctly thickened in basal half. Hind coxa and femur yellow as in fore and middle legs, hind tibia and narrow base of basitarsus dark brown to black. Rest of basitarsus and all follow- ing tarsal segments white, with whitish hairs at least on dorsal side. 76 Abdomen shining bluish-brown to black, with yel- low transverse stripes and pleural membrane. Transverse stripes broad on the posterior, and narrow on the anterior margin of each tergum of segments 2- 4, apical terga completely black. Transverse stripes sometimes reduced to broad yellow sidemarkings on posterior corners of segments. Male genitalia (figs. 4- 6) rotated by 90° along longitudinal axis, so that epandrium is in right lateral position. Apart from this, hypopygium symmetrical. Epandrium almost round and compact, tightly connected with para- procts posteroventrally. Epiproct of proctiger narrow, bearing slender, rodlike cerci. Synsternum deeply emarginate, with slender and bipartite medial process shielding short, symmetrical, partly reduced aedeagal complex from ventral side. Female. — Eyes separated by shining black frontal stripe being anteriorly somewhat broader than ocellar triangle (fig. 3). White frontal callus bulbose, wider than in male. Scutellum dark, its posterior margin slightly yellowish. Dark pleural spots as in male but partly reduced on katepisternum. Abdomen shining bluish black, with transverse yellow stripes at posteri- or margins of terga 2-4 extending to anterior margins of following terga. Yellow stripes about half as wide as dark stripes between them, somewhat broader than in male. Dark pattern of hind legs as in male. Cerci with two segments, slender and relatively long. Larva (and puparium). — Subcylindrical, with pro- minent head, slightly arched lateral margins of body segments and distinctly angulate though generally ROZKOSNY & KOVAC: Ptecticus rounded anal segment. Cuticle with usual mosaic ap- pearance, some cells somewhat larger and darker, forming almost complete double rows at boundary of body segments and groups in sublateral position. General colour yellowish brown, with diffuse brown pattern consisting of broad, longitudinal, undulate bands. Anal segment yellow, with large anterodorsal brownish spot. Ventral side paler, especially in me- dian third where brownish spots are almost indistinct. Head (figs. 7-8, 26-27) moderately flat, subconical. Vestiges of mandibular-maxillary complexes about as Figs. 7-10. Ptecticus longipennis, larval characters. — 7, head and first thoracic segment, dorsal view (A, antenna; Ad, antero- dorsal setae; Cf, clypeofrontal setae; D, dorsal setae; DI, dorsolateral setae; E, eye prominence; L, lateral setae; Lb, labral se- tae; VL3, posterior ventrolateral setae); 8, the same, ventral view (SI, sublabral setae; V, ventral setae; VL1-3, ventrolateral se- tae; V3, posterior ventral setae); 9, last abdominal segments, dorsal view ( Sa, subapical setae); 10, the same, ventral view (Ap, apical setae; as, anal slit; L1 + L2, lateral setae; sp, sternal patch). TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 rp fe te > Rune Si pe BE ~ x ~ 1 1 Figs 11-19. Pre. ticu malayensis sp. n., adult characters. — 11, male head, dorsal view; 12, the same, lateral view; 13, mesono- RENEE USE A REA PAP È = > 3 i cum pattern, dorsal view; 14, male terminalia, lateral view; 15, female abdomen, dorsal view; 16, antenna, inner side; 17, dor- al part of male terminalia; 18, aedeagal complex; 19, ventral part of male terminalia. long as labrum in dorsal view, low and rounded an- tennae placed anteriorly, their apical segment about as high as broad at base. Eyes relatively little promi- nent, posterior part of head with two rounded protu- berances on each side. Three thoracic segments about same shape as ab- dominal segments 1-7, first thoracic segment with lateral vestiges of prothoracic spiracles. Thoracic seg- ment 3 and abdominal segments 1-7 with small, round, dorsolateral, probably non-functional larval spiracles. In puparia, abdominal segments 2-5 with very inconspicuous, short, rodlike pupal respiratory horns near lateral margin. Very slender and long mid- sternal patch on abdominal segment 6 consisting of 112-114 pale cells (figs. 10, 31). Rounded and much smaller sternal patch with 18-19 cells usually at poste- rior margin of abdominal segment 5 (fig. 32). Anal segment (figs. 9-10, 28) distinctly emarginate between insertions of marginal setae, with a pair of low protuberances distally. Opening of spiracular chamber with undulate posterior margin and short pennate setae (figs. 29-30). Anal slit on ventral side with somewhat swollen marginal lips, boundered proximally by usual transverse fold. Two low sym- metrical protuberances developed around insertions of posterior ventral setae. General scheme of chaetotaxy as in other known genera of Sarginae (cf. Rozkosny 1982). All setae on head short, labral setae not longer than others, spine- like sublabral setae well developed. Anterior clypeo- frontal setae inserted almost at level of eye promi- nences. Setae on thoracic and abdominal segments arranged in transverse rows (3 pairs of dorsal and 2-3 pairs of ventral setae) and lateral groups (2-4 setae). Thoracic segment 1, on each half, with 2 anterodor- sal, 3 dorsal, 1 dorsolateral and 1 ventrolateral seta. All 3 thoracic segments only with 2 pairs of ventral se- tae. Lateral group of setae on thoracic segments con- sisting of 2 setae only (1 dorsolateral and 1 ventrolat- eral). On lateral side of abdominal segments, 4 setae (1 dorsolateral, 2 lateral and 1 ventrolateral) on abdom- inal segments | and 6-7, but only 2 setae (probably | ventrolateral and | lateral) on abdominal segments 2- 5. All marginal setae on anal segment of about same length, virtually not longer than setae on preceding abdominal segments. Five pairs of ventral setae on anal segment well visible, two distal pairs inserted closely together (fig. 10). Especially dorsal and ventral setae on body segments distinctly pubescent. Length 9.3-10.6 mm, maximum width 2.6-3.0 mm (based on 4 puparial exuviae). Material examined. — Peninsular Malaysia, University of Malaya Field Studies Centre, Ulu Gombak, Selangor (3°20°N, 101°45’E, altitude 250 m), 16 from 10 May, 20 ROZKOSNY & KOVAC: Ptecticus from 12 May and 19 from 16 June 1991, all specimens ex puparia, D. Kovac leg. Material deposited in the Forschungsinstitut Senckenberg, Frankfurt (SMFD) (2d), and in the Faculty of Science, Masaryk University, Brno (MUB) (1d and 19). Remarks Distribution. — India, Malaya, Sumatra, Java, Borneo, Philippines. Discussion. — According to Brunetti (1923) this slender species with long legs and wings belongs to the cingulatus species group. It differs from the other species of this group by the darkened wing tip, a diag- nostic character shared only with P. sumatranus Enderlein. The latter species, however, possesses the normal, not elongate wings and its mesonotum is red- dish yellow, not dark with a metallic shine. Of the few Ptecticus species in which the male genitalia have been illustrated, only P. okinawensis Öuchi may belong to the same group, as shown by the compact epandrium and deeply emarginate synsternum (cf. Nagatomi 19759): Ptecticus malayensis sp.n. (figs. 11-25, 33-39) Type material. — Holotype, male: Peninsular Malaysia, University of Malaya Field Studies Centre, Ulu Gombak, Selangor (3°20’N, 101°45’E, altitude 250 m), ex puparium 15 November 1991, D. Kovac leg, SMFD. — Paratypes 3d 29: Same locality, 38 reared 15 November, 12 23 November, 1 2 20 December 1991; 1d 12 SMFD, 2d 19 MUB. Diagnosis A species of the cingulatus group with longitudinal- ly striped mesonotum and a conspicuously spindle- shaped abdomen in both sexes. Measurements: Body 11.2-12.5 mm, wing 11.8- 12.2 mm. Male (holotype). — Head hemispherical, with con- spicuously protuberant white frontal callus (figs. 11- 12). Eyes separated by yellow, sparsely black-haired frontal stripe being in its narrowest point about as wide as distance between outer margins of posterior ocelli. Postocular area narrowly protuberant in dorsal view. Postocular hairs fine and moderately long, in several rows. Antennae ochrous yellow, scape and pedicel black-haired. Pedicel with elongate process on inner side (fig. 16), annulation of flagellum barely vis- ible. Arista in subapical position, bare. Bi-partite la- bellae of proboscis distinctly swollen, white in ground colour and with short white hairs. Posterior part of head deeply concave. Thorax yellow in ground colour, with extensive 79 TI]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 20-25. Precticus malayensis sp.n., larval characters. — 20, head and first thoracic segment, dorsal view; 21, the same, ven- tral view; 22, last abdominal segments, dorsal view; 23, the same, ventral view: 24, scheme of setae on lateral wall of 6th ab- dominal segment (DI, dorsolateral seta; st, larval spiracle; VI, ventrolateral seta); 25, scheme of setae on lateral wall of abdom- inal segment 3 (ph, pupal respiratory horn; st, larval spiracle). brown to blackish pattern. Mesonotum (fig. 13) with to black spots in central part of anepisternum, anteri- 3 broad black stripes reaching its posterior margin. or side and posterior corner of katepisternum, poste- Pleura predominantly yellow with several brownish rior half of katepimeron and anterior third of latero- 80 tergite. Also posterior margin of mediotergite broadly brownish. Thoracic pubescence short and mostly ad- pressed, yellow on pleura and predominantly brown to black on mesonotum and scutellum. Wings shaded with grey, veins along anterior mar- gin including discal cell yellow, other veins brown ex- cept for bases. Anterior cross-vein placed at basis of radial fork R,, M, not reaching wing margin. Dense microtrichia covering virtually the whole wing mem- brane, only wing base, alula and posterior cubital cell mostly bare. Squamae very narrow and inconspicu- ous, with brownish margin and yellow hairs. Halteres with yellow stem and blackish knob. Legs yellow with extensive blackish pattern on hind pair. Both anterior pairs mainly yellow, including coxae, only apical half of mid-femora somewhat brownish. Hind coxae brownish on anterior side, hind femur mainly dark except for both ends. Also hind tibia chiefly brown but this darkening more pro- nounced along dorsal side and in mid-ring. Hind tar- si black but basal two thirds of basitarsus and last tar- sal segment pale yellow. Abdomen spindle-shaped, broadest behind middle, yellow with brownish black transverse bands on terga (fig. 15). First tergum yellow with a transverse brown spot in middle and following terga with relatively broad posterior margin and narrow anterior margin yellow. Venter yellow but more brownish in apical half. Genitalia (figs. 14, 17-19) brown, distinctly asymmetrical, although dorsal part only slightly so. Cerci pointed, with tips oriented ventrally (figs. 14, 17). Synsternum strongly asymmetrical in proximal half, dististyli almost hoodlike in ventral view (fig. 19). Aedeagal complex almost S-shaped in lateral as- pect (fig. 14), right paramere conspicuously reduced (fig. 18). Female. — As male but with broader wings and somewhat wider abdomen in apical half. Frontal stripe only slightly broader than in male, particularly in anterior half, white frontal callus distinctly bipar- tite and lower, antennae relatively larger. Colours of thorax, legs and abdomen virtually the same, only darkening of hind legs more extensive, leaving only bases of femora and knees yellow. Hind tarsi coloured as in male. Terminalia dark brown, cerci with two narrow black segments (fig.15). Larva (and puparium). — Larvae on average stouter and longer than those of P. longipennis, often pre- dominantly reddish brown, though dark pattern con- sisting of dark undulate longitudinal bands, especial- ly in puparia, mostly distinct. Integument rough and sclerotised, with usual mosaic surface structure. Some cuticular cells forming inconspicuous pattern as in preceding species. Head (figs. 20-21, 33-34) promi- ROZKOSNY & KOVAC: Precticus nent, strongly sclerotised, partly retracted into first thoracic segment, distinctly longer than wide and conspicuously flattened, rather subtriangular in dor- sal view. Antennae very small and low, distinctly clos- er to eye prominences than in P. longipennis, apical segment barely prominent. Eyes conspicuously pro- tuberant, posterolateral tubercles less distinct, not doubled as in P. longipennis. Tracheal system as in larva of P. longipennis, in- cluding vestiges of larval spiracles and very small lat- eral pupal respiratory horns on abdominal segments 2-5. Sternal patch in middle of abdominal segment 6 drop-like to elongate oval, consisting of 90-116 cells (fig 23). Similar but much smaller and rounded ster- nal patch at posterior margin of sternum 5 formed by 17-20 cells. Anal segment (figs. 22-23, 35) almost round posteriorly, with a small incision in middle of distal margin. Opening of posterior spiracular cham- ber with arched and finely fringed posterior margin (- figs. 36-37). Anal slit distinctly emarginate, some- what longer than one third of anal segment. General scheme of chaetotaxy as in P. longipennis although setae relatively shorter and more pubescent, especially in marginal setae of anal segment. Anterior clypeofrontal setae inserted almost between eye prominence and tip of labrum (fig. 20). Configura- tion of setae on lateral wall of abdominal segments and ventral side of anal segment identical with those of preceding species and probably confirming generic characters in this way: 4 setae on lateral wall of ab- dominal segments | and 6-7, 2 setae on abdominal segments 2-5 (figs. 24-25) and 5 pairs of ventral setae on anal segment present (fig. 23). Length 10.0-13.4 mm, maximum width 3.6-5.0 mm (based on 2 larvae, 5 puparia and 3 puparial exuviae). Discussion The new species presumably belongs to the cingu- latus species group as defined by Brunetti (1923). Three species of this group have dark longitudinal stripes on the thorax: P. brevipennis Rondani, P. in- tensivus Brunetti and P. kambangensis de Meijere. The first was redescribed by van der Wulp (1885) and Brunetti (1923). It is, in contrast to P. malayensis sp. n., characterized by the shining blue black frontal stri- pe and by much narrower and shorter stripes on the mesonotum, the black tibiae and bases of the basitar- si of the hind leg. P. intensivus is, according to the ori- ginal description by Brunetti (1927), closely related to P. brevipennis and like that species should have a black frontal stripe. Moreover, it also differs in the co- loration of the leg (the fore tarsi are partly black, only the basal third of the hind femur is darkened and the hind tarsi are completely black). P. kambangensis dis- tinctly differs by the pleura being shining blackish 81 VOLUME VOOR ENTOMOLOGIE, JDSCHRIFT | 11 o € ì a Î (=) wit aay O BR ‘ f m "MR } Y ARE an; brown with a yellow band, and by the hind basitarsus, which is completely whitish as is the basis of the fol- lowing tarsal segment. From the other Oriental Ptecticus species only P. tricolor van der Wulp displays the longitudinally striped mesonotum. Brunetti (1923) considered this species to be an intermediate form with uncertain re- lationships. According to the original description and illustration (van der Wulp in de Meijere 1904) P. tri- color differs from the new species by the colouring of the legs (hind tibia and tarsi completely black) and abdomen (segments 2-4 entirely red). The colour of the leg and partly also the male gen- italia of the new species resemble the East Palaearctic P. sinchangensis Ouchi redescribed and figured re- cently by Nagatomi (1975). However, this species does not have a striped mesonotum, and its male gen- italia, including the well developed aedeagal complex, are symmetrical. BIOLOGICAL NOTES The larvae of Ptecticus longipennis were found in felled bamboo culms whose internodes — now lying in the horizontal position on the ground — had been sawn in such a way that the upper half formed a hin- ged lid that could easily be opened for inspection and than closed again. Two lateral slits, about 2 mm wide, provided access to the interior. The internodes were partly filled with water, and the larvae stayed in the moist part above the water surface (fig. 43). They ap- peared about two months after the culms had been felled. During the six-months study period larvae of Ptecticus longipennis were found only in three out of over one hundred internodes prepared in the descri- bed manner; moreover, these three internodes were adjacent to each other, i.e. the eggs were probably laid by the same female. This indicates that bamboo in- ternodes are only an occasional habitat for the larvae of Ptecticus longipennis. Ptecticus malayensis, in contrast, was found in bam- boo more frequently. The larvae of this species were restricted to bamboo shoots that had died recently (fig. 40). Judging from the traces of insect attack found in the internodes, death of the shoots was probably caused by a weevil, Cyrtotrachelus sp. (Coleoptera, Curculionidae), adults of which had been observed on bamboo shoots. The dead shoots were 2.5 m to 6 m high. At the time of felling the lower internodes were still green while the upper part ROZKOSNY & KOVAC: Ptecticu had turned dark brown and begun to decay. In the lower part the rotten section (c. 1 m from the top) in some cases a small hole (diameter c. 1-2 mm) was found which led into the interior of the culm. Beginning at this hole a groove in the inner wall led upwards, piercing the node septa (usually c. 10-15) and gradually becoming wider and deeper (fig. 42). At several points along this depression there were holes through the internode wall to the outside which were covered by the culm sheets and therefore hidden from outside view. Just below the shoot tip there was no segmentation into internodes. In this section, which was moist and gave off a putrid smell, the culm sheets had been penetrated with a circular opening c. 1.5 cm in diameter. In one case 2 grooves and circu- lar openings were found in the same shoot. Apparently the Cyrtotrachelus larva had worked it- self up all the way to the nutritious apical meristem from the small entrance hole bored by the female, feeding and grooving along the way. After feasting on the apical meristem the fully grown larva then prob- ably broke to the outside tu pupate in the soil, as de- scribed by Witt (in Beeson 1941: 209) for Cyrtotrachelus longipes F. Ptecticus malayensis probably uses a fresh Cyrtotra- chelus emergence hole to enter the bamboo shoot. Although all shoots attacked by Cyrtotrachelus died, there was probably still some limited growth after de- parture of the weevils as indicated by a shift of the holes in the different culm sheets relative to each oth- er (fig. 41). Consequently, at the time of felling there was no longer a contiguous opening all the way to the interior of the culm. The larvae of Precticus malyensis were found in all internodes connected by holes. In the lower internodes with punctured septa water usu- ally accumulated, and dead stratiomyid larvae were found there on several occasions. Pupation of Precti- cus malayensis took place in the upper section of the shoot, behind the Cyrtotrachelus escape hole (fig. 41). Probably the freshly eclosed flies had to sqeeze them- selves past the sheets to escape from the bamboo. The adult flies left before the rotten shoot tips fell to the ground. Five out of twelve investigated bamboo shoots were occupied by Precticus malayensis, with a maximum number of 22 larvae per shoot. In two shoots larvae of Solva completa de Meijere (Xylomyi- dae, Diptera) were found together with Precticus ma- layensis. Only a few literature accounts on larvae of Ptecticus are available. Lindner (1928), who superficially de- Figs. 26-32. Ptecticus longipennis, larval structures (SEM-photographs). — 26, head and fore margin of first thoracic segment, dorsal view; 27, the same, lateral view; 28, last abdominal segments, ventral view; 29, opening to spiracular chamber, poste- rior view; 30, short pennate setae on ventral lip of spiracular chamber opening; 31, sternal patch on abdominal segment 6; 32, sternal patch on abdominal segment 5. CO Oo DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 tlayensis sp.n., larval structures (SEM-photographs). — 33, head and fore margin of first thoracic seg- ıd and two thoracic segments in lateral view; 35, last abdominal segments, ventral view; 36, open- osterior view; 37, pennate setae on lips of spiracular chamber opening; 38, sternal patches on ab- | patch on abdominal segment 5. 4 vi ROZKOSNY & Kovac: Ptecticus Figs. 40-43. Habitat of the larvae of Precticus malyensis sp.n. and P. longipennis. — 40, dead shoot tip of Gigantochloa scortechi- nii. The arrow points to the emergence hole of the weevil larva, Cyrtotrachelus sp., which was probably used by P. malayensis to enter the hollow interior of the shoot; 41, emergence hole of Cyrtotrachelus sp. after removal of several culm sheets. Two puparia of P. malayensis are visible (arrows). As can be seen, the openings of the culm sheets have shifted relative to each oth- er; 42, the hollow interior of the shoot. Arrow points to a groove made by the larva of Cyrtotrachelus sp.; 43, larva of Ptecticus scribed a larva of P. testaceus Fabricius from South America, pointed out that larvae were found in decay- ing cactuses, oranges and lemons. Engel & Cuthbertson (1939), who examined a larva of African P. posticus Wiedemann, found the larvae in the semi-liquid mass of decaying Conopharyngia johnstoni Stapf. (Apocyna- ceae) fruits together with other dipterous larvae. Larvae of North-American Peecticus trivittatus (Say) were collected from a corn compost pile, decay- ing tomatoes, garbage and fruit, from watermelon rinds, rotting paper and a fungus (McFadden 1967). Our observations suggest that P. longipennis and P. malayensis are unspecialized phytosaprophages. These findings conform to the previously reported larval mode of life within the subfamily Sarginae: sapropha- gy with some tendency to coprophagy as known in many European species. RELATIONSHIPS OF Precticus LOEW Only one reliable character separates the genus Precticus Loew from the related Sargus Fabricius, viz. he shape of the pedicel projecting into the elongate longipennis on the moist interior wall of an internode laying on the ground. process on the inner side of the antenna. However, this character may be indistinct in some species, and its generic relevance therefore seems disputable. A comparison of the structure of the male genitalia is not possible for all species of the two genera because it has not been included in all descriptions. It may ne- vertheless be stated that the differences found among some Ptecticus species seem to be more pronounced than those among the European genera of the Sarginae. This fact indicates the necessity of a revision of the generic concept of Precticus. Judging from the description of the larva of North- American P. trivittatus (Say) by McFadden (1967) and the descriptions of the two larvae presented here, Ptecticus appears, on the other hand, well separated from the genera of Sarginae in the larval stage. First of all, in addition to a long and slender sternal patch on abdominal segment 6, a small and rounded sternal patch on abdominal segment 5 is developed. Relatively short setae on the head as well as on the anal segment, the configuration of setae on the lateral wall of abdominal segments and the presence of five pairs of ventral setae on the anal segment are cosid- 85 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 ered diagnostic generic characters. So, the known lar- vae of Sarginae may be distinguished according to the following identification key: l. Setae on anal segment shorter than length of pe- AUT er 2 — At least some setae on anal segment longer than praulumare SES En DR ee 4 2. Eye prominence anterior to protruding lateral tu- bercle (according to McFadden 1967) ................ EET Merosargus Loew — Eye prominence on protruding lateral tubercle Se Abdominal segment 5 without sternal patch, four pairs of ventral setae on anal segment, 2-4 setae on lateral wall of abdominal segment 1, 6-7, and always one seta less or equal number of setae on abdominal segments 2-5 ........... Sargus Fabricius — Abdominal segment 5 with a small but distinct sternal patch in addition to sternal patch on seg- ment 6, five pairs of ventral setae on anal seg- ment, four setae on lateral wall of abdominal seg- ments |, 6-7, and only two setae on abdominal segments 2-5 (i.e. two setae less) … Ptecticus Loew 4. Anterior labral setae short, apical lobes on anal segment prominent ............ Chloromyia Duncan — Anterior labral setae strikingly long, apical lobes on anal segment less distinct .. Microchrysa Loew ACKNOWLEDGEMENTS Thanks are due to K. Fiedler and R. Klein for re- viewing the manuscript. This work is part of a re- search project funded by the Deutsche Forschungsge- meinschaft and supported by the Socio-Economic Research Unit of the Government of Malaysia and Universiti Malaya. Special thanks are due to I. Azarae for supplying the facilities of the Ulu Gombak Field Studies Centre. 86 REFERENCES Beeson, C.F.C., 1941. The ecology and control of the forest insects of India and the neighbouring countries. — Vasant Press, Dehra Dun, 1007 pp. (Reprinted 1961). Brunetti, E., 1923. Second revision of the Oriental Stratiomyidae. — Records of the Indian Museum 25: 45- 180. Brunetti, E., 1927: Notes on Malayan Diptera, with de- scriptions of new species. — Journal of the Federal Malay States Museums 13: 281-309. Engel, E.O. & A. Cuthbertson, 1939. Systematic and bio- logical notes on some brachycerous Diptera of southern Rhodesia. — Journal of the Entomological Society of South Africa 2: 181-195. James, M.T., 1975. Family Stratiomyidae. In Delfinado, M.D. & D.E. Hardy (eds.), A catalog of the Diptera of the Oriental Region. Vol.II. The University Press of Hawaii, Honolulu, 459 pp. Lindner, E. 1928. Dr. L. Ziirchers Dipteren-Ausbeute aus Paraguay: Stratiomyiiden. — Archiv fiir Naturgeschichte (A) 22 (12): 94-103 McFadden, M. W., 1967. Soldier fly larvae in America north of Mexico. — Proceedings of the United States National Museum 121: 1-72. Meijere, J. C. H. de, 1904. Neue und bekannte Siid- Asiatische Dipteren. - Bijdragen tot de Dierkunde 17-18: 83-118. Nagatomi, A., 1975. The Sarginae and Pachygasterinae of Japan (Diptera: Stratiomyidae). — Transactions of the Royal Entomological Society of London 126: 305-421. Rozkosny, R., 1982/83. A biosystematic study of the European Stratiomyidae (Dipteta). - Dr. W. Junk Publishers, The Hague, Vol.1: 1-401, Vol. 2: 1-431. Rozkosny, R. & D. Kovac, 1991. First description of the male and the larva of Camptopteromyia fractipennis de Meijere from Malaysia (Diptera: Stratiomyidae). — Entomologica scandinavica 22: 297-304. Wulp, F. M. van der, 1885. On exotic Diptera. — Notes from the Leyden Museum 7: 1-15. Received: 2 July 1993 Accepted: 9 February 1994 J. van TOL National Museum of Natural History, Leiden, The Netherlands HAE ODONATA OF SULAWESI AND ADJACENT ISLANDS Part 3. The genus Macromia Rambur (Corduliidae) Tol, J. van, 1994. The Odonata of Sulawesi and adjacent islands. Part 3. The genus Macromia Rambur (Corduliidae). — Tijdschrift voor Entomologie 137: 87-94, figs. 1-25, table 1 [ISSN 0040-7496]. Published 15 July 1994. The Macromia species of Sulawesi (Celebes) are detailed, and a key to the species is provided. One new species, Macromia celebica sp. n., and the hitherto unknown female of M. irina Lieftinck are described. The occurrence on Sulawesi of M. moorei fumata Krüger is considered doubtful. The phylogenetic relationships of the Indo-Australian species of Macromia are ana- lysed. The Sulawesi species are shown to have a sister group relationship to the monophyletic group of the Papuan species. M. chalciope from Halmahera is considered the most recently evolved taxon. These results are compared with recent reconstructions of the geological history of the region. J. van Tol, National Museum of Natural History, P. O. Box 9517, 2300 RA Leiden, The Netherlands. Key words. — Odonata; Corduliidae; Macromia; new species; phylogeny; biogeography; Indonesia; Sulawesi; New Guinea. Part 2: Zoologische Mededelingen [1987] 61 (13): 160-176. Hitherto, only two species of the rheophilic drag- onfly genus Macromia Rambur have been mentioned from Sulawesi (Celebes), viz. M. moorei fumata Krüger and M. irina Lieftinck (Lieftinck 1950: 714- 716). Specimens of Macromia that were recently col- lected in the southern part of Central Sulawesi, ap- pear to belong to a species as yet undescribed. The number of species reliably recorded from this island, however, remains two, since the only record of Macromia moorei fumata must be regarded doubtful (see below). The dragonflies of the virtually cosmopolitan ge- nus Macromia are remarkable insects in many re- spects. With more than 120 species, several of which polytypic, Macromia is one of the largest genera of the Anisoptera. They are especially speciose in the tropi- cal areas of the Old World. In Europe (one species), North America (nine species) and Australia (two spe- cies), they are far less numerous. All other species are confined to Africa and Southeast Asia. Our present knowledge of the Macromia species of Southeast Asia is mainly based on the studies of Krüger (1899) and especially Lieftinck (1929, 1950, 1971a; see also 1954). Due to their elusive behaviour, most species are poorly represented in collections. In fact, many are only known from the type series, often consisting of one specimen only. Most species are fierce flyers. The larvae seem to be quite stenotopic, so that larvae as well as adults can only be found in very restricted parts of streams. Notwithstanding their strong flying capabilities, many species have small ranges, appar- ently due to the fact that they seldom fly far from their breeding sites. In the present paper I describe the new species from Sulawesi, provide some new observations on M. irina, and give a key to species recorded from Sulawesi. Finally, after having analysed the affinities of the Sulawesi species to those of the Oriental and the Papuan regions, I propose a hypothesis for a sce- nario of the biogeographical history of this genus in the Indo-Australian region. SYSTEMATIC PART Key to the Macromia species of Sulawesi 1. Outer margin of superior appendage of male with well-developed tooth-like tubercle approximately halfway (cf. fig. 18); vertex bilobed (cf. figs. 22, 23); thorax green with metallic lustre ................. EN ISAIA RO M. moorei fumata [Occurrence on Sulawesi doubtful]. — Outer margin of superior appendage of male 87 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 1-7. Macromia celebica sp. n. — 1, Head in frontal view; 2, Anal appendages of male, dorsal view; 3, Idem, left lateral view; 4, Base of right hind wing, male; 5, Pterostigma of right fore wing; 6, Accessory genitalia of male, left lateral view; 7, Fourth segment of penis, left lateral view. somewhat curved inwards, in some specimens with very small teeth along the margin, but espe- cially in dorsal view not a conspicuous extero-lat- eral tooth (figs. 2, 9); vertex dorsally smoothly rounded, without a paired tubercle (figs. 1, 8); thoraxpreen or:browme... rer... 2 Appendix inferior distinctly shorter than superi- ors (figs. 2, 3); coloration at least on synthorax green with greenish blue metallic shine; frontal . side of head brownish black .... M. celebica sp. n. — Appendix inferior approximately as long as superiors (figs. 9, 10); coloration without any in- dication of metallic lustre, but body generally orange-brown; head orange-brown with some middle-brown parts, but without conspicuous brownish black frons ............................ M. irina tn Macromia celebica sp. n. (figs. 1-7) Type material. — Holotype d. Indonesia. Southern Central Sulawesi. Puncak Palopo. 600-1000 m. 88 October 1993. Leg. Yohan R. (JvT 6007) (MBBJ). — Paratypes: 25, same data (JvT 5930-5931) (RMNH). Description. — A medium-sized Macromia, with green thorax and dark abdomen. Male. Head with labium, labrum and clypeus orange-brown, especially clypeus without yellow markings, labrum with outer corners darker. Pyramidal processes of frons high and subacute (fig. 1); frontal side very flat, more or less triangular, with very fine transverse striae from upper-inner to lower- outer corner of each side; frontal and upper parts of frons shiny black with blue metallic lustre; frons against eyes orange-brown as lower part of face; vertex black with fine punctation, the paired dorsal tubercles hardly discernable, the lateral angles tapering into small projecting cones on top of which the lateral oc- elli are situated (shape as in irina; westwoodii has somewhat more pronounced dorsal tubercles), cov- ered with thin setae. Occiput shiny black with very fine punctation. Antennae brownish black. Thorax. Brilliant metallic green, densely covered with long and slender creamish white setae; antehu- meral stripe yellow and distinct, somewhat tapering dorsally, extending dorsally to ca. three-seventh of mesepisternum, ventrally continuing over mesinfrae- pisternum; yellow stripe over spiracle rather narrow, tapering in width dorsally and nearly reaching the upper margin of episternum. Pale stripe on latero- ventral part of metepimeron covering anterior anteri- or half (stripe absent in chalciope). Legs long and slen- der, hind femur 10 mm, reaching to base of posterior hamulus, colour brown, keel on fore leg very short, ca. 1 mm, not reaching the apex, middle tibia without keel, hind tibia with long keel from ca. 2 mm from base of tibia to less than 1 mm from its apex. Wings relatively short, fore wing with 17-18 Ax, 8 Px; hind wing with 11 Ax, 11-13 Px; fore wing with 7 Cux, hind wing with 5 Cux (incl. subtriangle); anal loop 7-9 cells without central cell; anal triangle rela- tively short (fig. 4; cf. M. chalciope, fig. 19); discoidal triangle of hind wing very narrow; pterostigma (fig. 5) reddish brown, anterior side ca. 2.0 mm. VAN Tor: Macromia of Sulawesi Abdomen brownish black with some blue-green metallic lustre especially on first few segments; yellow spot on anterior side of segment 7; general shape an- teriorly slender, especially segment 7-9 significantly widened; tergites ventro-anteriorly somewhat paler, dark or middle-brown. Spot on segment 7 consisting of a complete ring covering anterior one-fourth of segment, medio-dorsally extending posteriorly for an- other one-eighth segment length and ca. one-third the segment’s width. Segment 10 rather short, mid- dorsal process not spine-like, but consisting of a low triangular-shaped subterminal tubercle. Appendages (figs. 2-3) with superiors in dorsal view rather slender, significantly longer than inferior, basal two-fifths rather straight, then innerside slightly tapering poste- riorly and outerside strongly tapering, caudalmost part somewhat curved outwards; no trace of a lateral tubercle on superior appendage. Secondary genitalia: anterior side margin of tergite 2 with dense setae, a mixture of strong and soft setae, longer than in irina, also tuft of setae on ventral tu- Figs. 8-15. Macromia irina Lieftinck. — 8, Head in frontal view; 9, Anal appendages of male, dorsal view; 10, Idem, left later- al view; 11, Base of right hind wing, male; 12, Pterostigma of right fore wing; 13, Accessory genitalia of male, left lateral view; 14, Fourth segment of penis, left lateral view; 15, Valvula vulvae, female. 89 Ty DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 bercle of tergite longer; genital lobe directed posteri- orly, elongate, and tapering towards the top, but tip not acute; posterior hamulus glossy brown, basally coloured brown, the straight outer part pale brown, and the endhook reddish brown, curved strongly (90°) inwards, endhooks in ventral view slightly diverging, the tip reaching ca. two-thirds of genital lobe. Measurements. — Abdomen, incl. appendages 48 (paratypes 46-48) mm, hind wing 42 (paratypes 42- 43) mm Female. — Unknown. Differential diagnosis. — Differs clearly from the only other Sulawesi Macromia by the presence of green metallic lustre, the black markings on the frons, the presence of an antehumeral stripe; structurally the short inferior appendage, and shape of abdominal segment 10 are diagnostic. Both Sulawesi species can be distinguished from other Indo-Australian Macromia by the intermediate sized pterostigma, the very narrow discoidal triangle of the hind wing, and especially the rounded (not bilobed) vertex. Also, the superior appendages lack an extero-lateral tooth, but are not particularly slender. The Papuan Macromia have extremely slender superiors with or without an extero-lateral tooth (fig. 20), while the Oriental spe- cies have rather solid appendages with a more or less conspicuous extero-lateral tooth (fig. 18). Remarks. — Although this species differs in many respects from M. irina, I consider it the sister species of M. irina. This view is based on shared, presumably apomorphous, character states in the structure of ver- tex and superior appendage of male (see below). Distribution. — Only the southwestern part of Central Sulawesi. Collected between 600-900 m. Macromia trina Lieftinck (figs. 8-15) Macromia irina Lieftinck, 1950: 714-716. Holotype male ‘S Celebes, foot of Mt Lompobatang, Borong Rapao, 800 m, 12.viii. 1949, leg. A. Diakonoff in RMNH [examined]. — Lieftinck 1971b: 94 [note on type]; Askew et al. 1989: 118 [records Northern Sulawesi, Dumoga Bone National Park]. Diagnosis. — Unmistakable among Indo-Australian Macromia: head, thorax and abdomen brown without any metallic lustre. Structural characters: wing base with anal angle sharp, but not angulated (fig. 11); the anal triangle rather short (long in moorei fumata), membranula ending circa halfway between anal cross- vein and base of anal triangle; pterostigma rather short (fig. 12), circa twice as long as high (more than three times in moorei fumata); anal appendages of male with superiors with rather smooth outer margin, without any indication of tooth (figs. 9-10) (cf. moo- 90 rei fumata and westwoodii); inferior appendage ap- proximately as long as superiors (clearly shorter in ce- lebica). Description. — Little can be added to the extensive description of the male in Lieftinck (1950). Live colour of eye dark green. Since Lieftinck did not pro- vide any illustration, I have added here drawings of the head, wing base, anal appendages and secondary genitalia (figs. 8-14). The up to now undescribed fe- male was collected in Southeast Sulawesi. Female. — General coloration as the male, and structurally very similar to the male, although more robust. Head and thorax not differing from male. Wings somewhat broader, especially at base; mem- branula fully opaque white or only greyish in most narrow part; rusty brown base of wings more exten- sive than in male, in costal and subcostal area of fore and hind wing reaching to first Ax or even somewhat beyond; some less extensive rusty brown at base of medio-anal space; Arculus at Ax2 in fore wing and just beyond Ax2 in hind wing; fore wing 16-17 Ax, 9 Px, 7 Cux; hind wing 10212 Ax 71E72757.@ux: Abdomen of the coloration (‘Mikado-brown’) as male; segment 1 dorsum dark brown and lateral parts brown; segment 2 brown, somewhat darker above than laterally, as the male with two narrow transverse yellow streaks, one over transverse carina halfway ab- dominal segment and the other pair more oblique running just before posterior margin; segment 3 ante- riorly brown to transverse carina, somewhat darker above, posteriorly dark brown except for space just before segment 4; segments 4-7 similar to segment 3, but dark patch anterior to transverse carina gradually increasing in size, and posterior paler brown ring also increasing from segment 4 to 7; segment 8 brown with dorsal carina black and some darkening against dorsal carina; segment brown, dorsal carina dark; segment 10 brown; appendages middle brown, rather short, acute. Valvula vulvae short, approximately four times as wide as long, with wide triangular emargina- tion, the tips just bifid, somewhat curved inwards (fig. 15). Other material examined. — Sulawesi Tenggara. S of Sanggona: Gunung Watuwila. Sunggai Lalondu- wasi near Centipede camp. c. 1050 m. 2-4 Nov 1989 (J. van Tol) 38 12 (JvT 5932-5935); Sulawesi Tengah, ca. 10 km WNW Palopo near Tojambu. 800-1000 m, c. 2°56’S 120°07’E, Jul/Aug 1991 (Yo- han R.), 12 (JvT 5936), all in RMNH. I have also studied one of the specimens collected by Askew et al. (1989): 18, in RSME. Remarks. — I collected this species myself along a small, shaded stream in a steep valley in the Mengkoko mountain area. The insects appeared im- mediately after sunrise, and were active for approxi- VAN TOL: Macromia of Sulawesi Figs. 16-23, Macromia species. 16-18, 22, M. westwoodii, 19-21, 23, M. chalciope. — 16, 19, Base of right hind wing, male; 27, 21, Pterostigma of right fore wing; 18, 20, Anal appendages of male, dorsal view; 22-23, Vertex in frontal view (fig. 23 dorsal-most part only). mately one hour. Males as well as females flew low over the water up and down the stream, more or less in a straight line. With a rather slow and small wing beat, they were not extremely fast. The specimens could be collected at a site where a big log of more than one meter diameter was laying in the stream. Macromia specimens crossed the log at just a few cen- timeters height to return immediately to the water surface. The female was caught on a dark day at 16.30 h, Le. 1.5 hours before sunset. Macromia moorei fumata Krüger Macromia fumata Krüger, 1899: 333-335 (Java) [not exami- ned]. Macromia moorei fumata. — Lieftinck 1950: 678 (key, men- tioned Celebes); Lieftinck 1954: 119, 185 (references, geographical distribution); Lieftinck 1971a: 29 (geo- graphical distribution). Remarks. — I have been able to check the specimen (in SMFD) at which Lieftinck’s notice of this taxon on Sulawesi was based. Although this specimen can be assigned to M. moorei fumata without doubt, I con- sider its occurrence on Sulawesi still doubtful. The ‘Lompo Battau’ labels of Fruhstorfer have on many occasions proved to be incorrect (see van Tol 1987: 163). M. moorei fumata is rather common on Java, where Fruhstorfer also collected during his long jour- ney through the former Dutch Indies. Material examined. — Lompa Battau, S. Celebes, III. 1896 (H. Fruhstorfer) ‘16764 (in green ink) 16 [label in Ris’ hand] [identified by M.A. Lieftinck as Macromia moorei fumata, Jan. 1928]. PHYLOGENETIC RELATIONSHIPS Introduction The so-called Papuasian representatives of Macro- mia share at least four characters (Lieftinck 1952, 1971a), viz., an exceptionally small discoidal triangle of the hind wing, a minute pterostigma of ca. one mm, a sharply acute anal angle of the hind wing in 91 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Table 1 Character matrix for taxa of Macromia of the Indo-Australian archipelago. (1) extero-lateral tooth superior appendage (2) vertex : (3) hamulus (4) superior appendages (5) pterostigma (6) postclypeus (7) antehumeral stripe (8) anal angle (9) discoidal cell (10) general coloration (11) dorsal spine abdominal segment 10 (12) curvature superior appendages Character 1 2 3 4 5 Species amymone 1 1 2 2 3 celebica 2 7) 1 1 2 chalciope 3 1 2 2 5) Irina 2 2 1 1 2 melpomene 1 1 2 2 3 moorei fumata 1 1 1 1 1 terpsichore 3 I 2 2 5) 1 72 1 1 1 12 (present/absent) 12 (bilobed/rounded) 12 (shorter/longer lobus) 12 (normal/very slender) 123 (long/intermediate/ very short) 12 (unicolorous/yellow markings) 12 (present/absent) 12 (sharp/angulate) 12 (wide/narrow) 12 (green/brown) 123 (slender/absent/broad) 123 (curved inward/outward/straight) 6 7 8 9 10 11 12 BNP eee eb HIHNNNCEN = D = D D m Nn HNENNNNN Fr eS a & D ND = N= ND RD PW NWN N westwoodii the male, and long and slender genital hamules in the male. Lieftinck (1971a: 30-32) distinguishes three groups among the Papuan Macromia species, viz. M. terpsichore, M. melpomene, and M. chalciope, with the last one consisting of one species only. The species assigned to these groups, which the ex- ception of viridescens described from Cape York (Australia), are confined to the islands of New Guinea (astarte Lieftinck, celaeno Lieftinck, eurynome Lieftinck, melpomene Ris, terpsichore Foerster, virides- cens Tillyard (= amymone Lieftinck)), the Bismarck Archipelago (lachesis Lieftinck), Waigeu (euphrosyne Lieftinck, sophrosyne Lieftinck) and Misool (hermione Lieftinck). M. chalciope Lieftinck is restricted to Schouten Is, and the Moluccan islands of Halmahera and Bacan (Lieftinck 1971a). No species of Macromia have been mentioned from any of the oth- er Moluccan islands up to now. According to Lieftinck (1952: 460) M. chalciope can be considered ‘an immediate derivative of the eastern cluster’, i.e. that the sister-species of chalciope could be discovered among the species of New Guinea and adjacent is- lands. The Malay peninsula and the Greater Sunda Islands are inhabited by 15 species of Macromia (see Lieftinck 1971a: 29). Although Lieftinck (1929) at- tempted to distinguish species groups in the western Malesian species, this author (1950: 716) had to ad- mit that the groups had an artificial character. Variation in morphology among the species of west- 92 ern Malesia is more prominent than that among the Papuan representatives. Finally, the Philippine islands are inhabited by at least one endemic species, viz. M. negrito Needham & Gyger from Luzon. The widespread M. cincta is re- ported from Palawan (Lieftinck 1929: 92), and Tawi Tawi (Lieftinck 1974), while also M. gerstaeckeri may amymone chalciope terpsichore melpomene celebica irina westwoodii moorei fumata Fig. 24. Tree for Macromia species of the Indo-Australian archipelago. Van TOL: Macromia of Sulawesi Fig. 25. Phylogenetic relationships of Macromia species (groups) represented on the map of Malesia. occur in the Philippines (Lieftinck 1971a: 29). I present here the results of a preliminary investiga- tion of the phylogeny of the Macromia of the Indo- Australian region, mainly based on the characters used by Lieftinck to establish his Papuan species groups, or to distinguish the Oriental and Papuan groups. As outgroups I have added M. moorei fumata Krüger and M. westwoodii Selys, both considered closely allied to M. irina (Lieftinck 1950). For the present study only a limited number of the Papuan species have been included. Unfortunately, M. negri- to Needham & Gyger from Luzon, was not available for study. A more extensive phylogenetic study, in- volving more of the Oriental as well as the Papuan species, has to wait for the future. Methods and results The matrix (table 1) was analysed with PAUP, ver- sion 3.1.1 (Swofford 1993), using ‘exhaustive search’. Macromia moorei fumata was designated as outgroup, but the results appeared to be similar when M. west- woodii was selected as such. The analysis resulted in one tree of 20 steps (CI-0.800, RI=0.818). This tree (fig. 24) shows that all Papuan species form a monophyletic group, with the Sulawesi species as sister group. Ihe Papuan groups as defined by Lieftinck (1971a) are not corroborated by the present analysis. It is of interest to notice the sistergroup rela- tionship of M. terpsichore from NE New Guinea, and M. chalciope from Halmahera. Homoplasies in the present tree expecially occur in two non-structural characters, viz. character 7 (ante- humeral stripe) and to a lesser extent character 6 (postclypeus). Discussion and biogeography The results of the phylogenetic analysis partly con- firm earlier suggestions by Lieftinck on relationships among Indo-Australian Macromia species. Lieftinck (1950) considered M. irina from Sulawesi closely re- lated to M. moorei and M. westwoodii, both rather wi- despread in Southeast Asia, including Borneo. The rather solid build of the superior appendage, in com- bination with the absence of an extero-lateral tooth, ranks the Sulawesi representatives in an intermediate position between the Oriental and the Papuan spe- cies. This intermediate position can also be recogni- zed in the shape of the pterostigma. The rather distant relationships of the Sulawesi and Halmahera species are of interest in a biogeographical context. Although the geological history of Wallacea and Halmahera is still under debate (e.g. Daly et al. 1991, Honza 1991), there is general agreement that there is no shared history for the North Moluccas (Halmahera) and Sulawesi. Reconstructions prior to 10-15 million years ago (Ma) are highly uncertain be- cause of the very complicated plate movements. In the reconstruction of Daly et al. (1991) (see also Nichols & Hall 1991) Halmahera was formed at the margin of the Pacific and the Philippine plates, north of New Guinea, with a westward progression from c. 5 Ma to its present position. At least some parts of 93 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Sulawesi have a much longer (65 million years) histo- ry above sea level on a far more western position. In other words, from approximately 60 to 5 Ma, the Oriental region + Sulawesi have shared (in some re- spects) acommon history, while Halmahera shares its history mainly with New Guinea. The close contact of the Oriental and Papuan region is more recent than 5 Ma. Given these phylogenetic relationships, and our present knowledge of the geological history of the area, the following historical scenario may be outli- ned. All Macromia species in the Indo-Australian re- gion originate from an Oriental ancestor. The ance- stor of the Sulawesi + Papuan species groups possibly inhabited Sulawesi, while the ancestor of the Halmahera M. chalciope + New Guinea M. terpsicho- re species groups apparently lived in north(eastern) New Guinea. The occurrence of two relatively remo- tely allied species on Sulawesi and Halmahera can be attributed to the westward movement of Halmahera during the last ten million years. ACKNOWLEDGEMENTS Field work in Sulawesi was supported by the Zoological Museum at Bogor. The kind help of the director of the museum, drs. Mohammad Amir, and his staff, as well as the staff of Lembaga Ilmu Pengata- huan Indonesia (Indonesian Academy of Sciences) is gratefully acknowledged. The expert knowledge of Sulawesi of Mr. Baharudin (Bantimurung) was of great help for successful collecting. Dr. H. Schröder kindly arranged the loan of material in SFMD. I also thank Dr. K. Wilson (Hongkong) for his comments on a draft of this paper. REFERENCES Askew, R.R., G.G. Cleland, D.A.L. Davies & T.W. Harman, 1989. A report on a collection of Odonata from North Sulawesi, Indonesia. — Tijdschrift voor Entomolo- gie 132: 115-121, fig. 1, cables 1-2. Daly, M.C., M.A. Cooper, I. Wilson, D.G. Smith & B.G.D. Hooper, 1991. Cenozoic plate tectonics and ba- 94 sin evolution in Indonesia. — Marine and Petroleum Geology 8: 2-21, figs. 1-15. Honza, E., 1991. The Tertiary arc chain in the Western Pacific. — Tectonophysics 187: 285-303, figs. 1-5. Krüger, L., 1899. Die Odonaten von Sumatra. III. Theil. a. Familie Libelluliden (with supplement: Neue Cordulinen von Java und Borneo. — Stettiner Entomologische Zei- tung 60: 321-338. Lieftinck, M. A., 1929. A revision of the known malaysian dragonflies of the genus Macromia Rambur, with com- parative notes on species from neighbouring countries and descriptions of new species. — Tijdschrift voor Entomologie 72: 59-108, figs. 1-22. Lieftinck, M. A., 1950. Further studies on Southeast Asiatic species of Macromia Rambur, with notes on their ecolo- gy, habits and life history, and with descriptions of larvae and two new species (Odon., Epophthalmiinae). — Treubia 20: 657-716. Lieftinck, M. A., 1952. On the Papuasian representatives of the genus Macromia Ramburm with descriptions of five new species and some larval forms (Odon.). — Treubia 21: 437-468, p. 1-22. Lieftinck, M. A., 1954. Handlist of Malaysian Odonata. — Treubia 22 (Supplement): i-xiii + 1-202, map. Lieftinck, M. A., 1971a. Studies in Oriental Corduliidae (Odonata), I. — Tijdschrift voor Entomologie 114 (1): 1- 63, figs. 1-50. Lieftinck, M. A., 1971b. A catalogue of the type-specimens of Odonata preserved in The Netherlands, with a supple- mentary list of the Odonata types described by Dutch sci- entists deposited in foreign institutional collections. — Tijdschrift voor Entomologie 114 (2): 65-139 + plate 1- Je Lieftinck, M. A., 1974. Dragonflies collected by the Noona Dan Expedition in the southwestern Philippine Islands (Insecta, Odonata). — Steenstrupia 3: 111-147, figs. 1-26. Nichols, G. J. & R. Hall, 1991. Basin formation and Neogene sedimentation in a backarc setting, Halmahera, eastern Indonesia. — Marin and Petroleum Geology 8 (1): 50-61, figs. 1-9. Swofford, D. L., 1993. paup: Phylogenetic Analysis Using Parsimony. Version 3.1.1. — Illinois Natural History Survey, Champaign. Tol, J. van, 1987. The Odonata of Sulawesi and adjacent is- lands. Part 2. The genus Diplacina Brauer on Sulawesi. — Zoölogische Mededelingen 61: 160-176, figs. 14-39. Received: 2 February 1994 Accepted: 14 April 1994 JOHN S. WEAVER III & HANS MALICKY’ ! University of New Hampshire, Durham, U.S.A. * Biologische Station Lunz der Österreichischen Akademie der Wissenschaften, Lunz am See, Austria. THE GENUS DIPSEUDOPSIS WALKER FROM ASIA @RICGHOPRTERA: DIPSEUDOPSIDAE) Weaver, J. S., II & H. Malicky, 1994. The genus Dipseudopsis Walker from Asia (Trichoptera: Dipseudopsidae). — Tijdschrift voor Entomologie 137: 95-142, figs. 1-55. [ISSN 0040-7496]. Published 15 July 1994. The genus Dipseudopsis Walker from Asia is revised, and a phylogeny of Hydropsychoidea is provided showing the relationships of its family taxa and the genera of Dipseudopsidae. Presently 36 Asian species of Dipseudopsis are recognized, having a distribution including Pakistan, India, Sri Lanka, Nepal, Bangladesh, Burma, Thailand, Cambodia, Vietnam, China, Japan, Philippines, Malaysia and Indonesia, including Sumatra, Java, Borneo, and Sulawesi. Six new species are described, Dipseudopsis adiaturix sp. n., D. flinti sp. n., D. lucasi sp. n., D. mal- aisei sp. n., D. martynovi sp. n., and D. schmidi sp. n., and one new subspecies, D. robustior an- damanensis ssp. n. The identities of several problematic species are resolved, with 16 species names recognized as new junior synonyms (cf. species checklist p.102). D. nervosella Ulmer stat. n. is elevated from former subspecies status. New descriptions are provided for all species, including figures of the male hind mesoapical tibial spur (the modified spur), the male genital- ia, and other notable characteristics of the head, thorax and wings, except two species recog- nized as nomina dubia, D. onychophora Navás, and D. orientalis (Navas). New lectotypes are designated for D. contorta Banks, D. morosa Banks and D. triclavata Martynov. A pictorial at- las is provided for rapid species identification. The genus Eodipseudopsis Marlier is removed from the Dipseudopsidae. Correspondence: J. S. Weaver III, Department of Entomology, University of New Hampshire, Durham, NH, 03824, U.S. A. Key words. — Asia, Dipseudopsis, Dipseudopsidae, Hydropsychoidea, phylogeny, systematics, Trichoptera. The genus Dipseudopsis Walker exhibits a number of characteristics that are quite remarkable among the Trichoptera. The adults are the only members of the suborder Annulipalpia Martynov having siphoning mouthparts, with the lacinia forming a proboscis, a parallel modification of the galea in the higher Lepidoptera. Sexual dimorphism is conspicuous in the genus, with the males having highly contrasting colour patterns in their wings and bodies, and the fe- males less striking in comparison. Also, males have the mesoapical spur of the hind tibia modified, often twisted and/or branched. The function of the modi- fied spur is unknown, but it has been used almost ex- clusively for species determinations, to the extent that customary descriptions of the male genitalia often have been omitted from species descriptions. The convention of using the modified spur as the basis for species descriptions was established by Ulmer (1907b: 180): ‘Bei Dipseudopsis sind die Genitalien aller Species ganz gleich’ (i.e: In Dipseudopsis the gen- italia of all species are totally alike). Hence, the mod- ified spur became the primary characteristic used to recognize species; e.g. this is the main distinguishing character in every couplet of Ulmer’s (1951) key to eleven species of Dipseudopsis of the Sunda Islands. The precise orientation of the modified spur must be made apparent for species identification, because slight rotations of the tibia will show different per- spectives of the spur. Some of the previous species de- scriptions based on the modified spur were ambig- uous, because the figure of the spur did not show the remainder of the leg, and hence the actual orientation of the spur was uncertain. This problem accounts for several of the synonyms that we discovered in the course of this study. The figures of the spur herein are usually of the left leg and include the adjacent apic- olateral spur and a portion of the tibia and tarsus to provide proper comparison of other specimens in the same perspective. It seems somewhat ironic that the taxonomy of this genus, which exhibits so many fascinating character- 95 [HDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Table 1. List of character states. 0 = Plesiomorphic in respect to the outgroup Philopotamidae. 1 = Synapomorphic character. Be ar Wr Re ri I. Adult dorsal ocelli present = 0; ocelli absent = 1 (cf. Schmid 1980). Schmid 1980). 3 segment IX completely sclerotized cylinder = 0; IX jawlike with articulation between tergum & sternum = 1 (cf. 3. © segment X without lateral papillae = 0; X with lateral papillae = 1 (cf. Schmid 1980). 4. 9 segment IX normal = 0; IX long completely sclerotized cylinder = 1 (cf. Schmid 1980). 5. Lama spinneret short = 0; spinneret longer than other mouthparts = 1 (cf. Wiggins 1977). 6. © sternum VIII normal, undivided = 0; sternum VIII divided into a pair of lateral lobes = 1 (cf. Schmid 1980). 3 inferior appendages 2-segmented = 0; 1-segmented = 1 (cf. Schmid 1980). 8. Adult bases of antennae distant = 0; bases close together = 1 (cf. Crichton 1957). | zen 9. Larva tarsi nearly cylindrical, pretarsal claws long = 0; tarsi flattened, claws short = 1 (Ulmer 1957, Gibbs 1968, Wiggins 1977, Wells & Cartwright 1993). 10. Larval retreat differs from the following = 0; larva with branched tube-dwelling and filter-feeding behaviour as in Phylocentropus (Wallace et al. 1976, Wiggins 1977). 11. Adult labial palps present = 0; absent = 1 (Ulmer 1951). 12. Adult pronotum normal = 0; pronotum enlarged and with deep median fissure = 1. 13. Adult with head and thorax setous = 0; head and thorax glabrous = 1. 14.. Adult tegula knoblike = 0; tegula fattened like epaulets = 1. 15. Adult with galea short, vestigial = 0; modified into proboscis = 1. 16. Larva labrum expanded and more membranous = 1, autapomorphy for Philopotamidae (cf. Wiggins 1977). 17. Larva trochantin broad, either with blunt apical angle or truncate = 1, autapomorphy for Psychomyiidae (cf. Wiggins 1977) 18. Larva with mesopleural lobes = 1, autapomorphy for Xiphocentronidae (cf. Wiggins 1977). 19. Larva with tibiae and tarsi of legs fused = 1, autapomorphy for Xiphocentronidae (cf. Wiggins 1977). 20. Larva with ventral thoracic gills whorled = 1, autapomorphy for Hydropsychidae (cf. Wiggins 1977). 21. Larva epidermis with modified setae, especially on sternum VIII & IX and tergum = 1, autapomorphy for Hydropsychidae (cf. Wiggins 1977). Wiggins 1977). 22. Larva head with ventrolateral bands of transverse ridges used in stridulation = 1, autapomorphy for Hydropsychidae (cf. 23. Larval head and body flattened dorsoventrally = 1, autapomorphy for Ecnomidae (cf. Lepneva 1964). 24. Larva lateral fringe dense = 1, autapomorphy for Ecnomidae (cf. Lepneva 1964). istics, should have become so confused at the species level. This is chiefly because Banks, Martynov, Navas, and Ulmer from 1905 to 1937 proposed descriptions of 25 different species from Asia and the species de- scriptions by Banks and Navas are poor. Another fac- tor might be that specimens of Dipseudopsis are not particularly abundant in museum collections, even though adults are attracted to light traps. The primary objective of this work is to clarify the taxonomy of the Asian species of Dipseudopsis. To ac- complish this we have examined the types of most of the species as well as other specimens from several museums and private collections. The species descrip- tions have been expanded beyond the characteristics of the modified spur to include colouration of the body and wings and illustrations of the male genitalia. Hence, most of the redescriptions of species herein provide the first illustrations of the male genitalia. We recognize 36 species of Dipseudopsis from Asia, 30 previously described species and six new species, and one new subspecies. We also recognize the names of 16 species and one subspecies as new junior syno- nyms. Two additional Asian species, D. onychophora Navas and D. orientalis (Navás), are recognized as no- mina dubia, because both were poorly described and their types could not be located for examination. 96 DEPOSITORIES The depositories of material are abbreviated as fol- lows: BMNH, Natural History Museum (formerly the British Museum of Natural History), London; BPBM, Bernice Pauahi Bishop Museum, Honolulu; CASC, California Academy of Sciences, San Francisco; CNCI, Canadian National Collection, Ottawa; CLDD, collec- tion of D. G. Denning (now deposited at CASC); CLHM, collection of H. Malicky; CLJW, collection of J. S. Weaver III; Mczc, Museum of Comparative Zoology, Harvard University, Cambridge; MNHN, Muséum National d’Histoire Naturelle, Paris; MZBS, Museu de Zoologia, Barcelona; MZLU, Museum for Zoology, University of Lund; NHMW, Naturhistori- sches Museum Wien; NHRS, Naturhistoriska Riksmu- seet, Stockholm; NzsI, National Zoological Collec- tion, Zoological Survey of India, Calcutta; RMNH, Nationaal Natuurhistorisch Museum (formerly Rijks- museum van Natuurlijke Historie), Leiden; SOFM, National Museum of Natural History, Sofia; UOPJ, Entomological Laboratory, University of Osaka Prefecture, Sakai; USNM, Smithsonian Institution, United States National Museum of Natural History, Washington; ZRAS, Zoological Institute of the Rus- sian Academy of Sciences, St. Petersburg; ZMAN, Zoo- logisch Museum, Amsterdam; ZMHB, Zoologisches Museum der Humboldt-Universität Berlin; ZMUC, Zoologisk Museum Kobenhavn; ZMUH, Zoologisches Museum der Universität Hamburg; and ZSMC, Zoologische Staatssammlung München. METHODS Taxonomy The morphological terminology of wing venation and genitalic characters mostly follows that of Schmid (1980), and Schuster & Hamilton (1984). The species synonymies are complete listing all original species de- scriptions, but subsequent references listed in Fischer’s Trichopterorum Catalogus (1962, 1972) are not inclu- ded. Locality names in current use are provided for spe- cimens examined, where old names taken from labels are in quotes. For the type data, the term ‘holotype’ is used when the type specimen has actually been labelled or cited as such, otherwise it is listed as ‘type’. An aste- risk (*) designates types of previously described species examined by the authors. Phylogenetics Hennig86 (Farris 1988) is used for phylogenetic analysis to determine the relationship of Dipseudopsidae within the Hydropsychoidea s. str. Weaver, and the relationships of the proposed di- pseudopsid genera. Twenty four character states (ta- ble 1) are selected, and with polarity determined for twelve taxa, using Philopotamidae Stephens as an outgroup, the data matrix (table 2) is produced. These data are analyzed using the ‘mhennig* com- mand of Hennig86 for calculating trees. All charac- ters have default values, weight = 1, active and additi- ve, except weight = 2 for character 6. WEAVER & MALICKY: Dipseudopsis from Asia DIPSEUDOPSIDAE ULMER The family Dipseudopsidae is included in the sub- order Annulipalpia Martynov, in the infraorder Curvipalpia Weaver (Weaver 1984, Weaver & Morse 1986). The family name, Dipseudopsinae Ulmer (1904b), including only the nominative genus, was first proposed as a subfamily of the Hydropsychidae Curtis, but was later placed within the Polycentropo- didae Ulmer (1906), as it has been recognized by several others (Tsuda 1942, Marlier 1962, Ross & Gibbs 1973, Wiggins 1977, Schuster & Hamilton 1984). However, some workers in the past have re- cognized Polycentropodinae and the Dipseudopsinae as subfamilies of Psychomyiidae Curtis (Ross 1956, Ross & Kingsolver 1959), and more recently, some still recognize the Polycentropodinae as such (Flint 1991). The Dipseudopsidae were first given full fami- ly status by Ross (1967), and later concurred by Gibbs (1968), Wiggins (1982), Weaver (1984), Scott (1985), Weaver & Morse (1986), and Wells & Cartwright (1993). The genus Protodipseudopsis Ulmer (1909) from the Afrotropical Region was the first addition to the subfamily, a relationship later confirmed by Ross & Kingsolver (1959). Two additional genera from the Afrotropical Region were placed in the Dipseudo- psinae, Eodipseudopsis Marlier (1959) by original des- ignation, and Limnoecetis Marlier (1955) originally placed in the Leptoceridae Leach, but later moved to the Dipseudopsinae (Marlier 1961). Ross & Gibbs (1973) transferred Phylocentropus Banks (1907) from Polycentropodinae to Dipseudopsinae, based pre- dominantly on synapomorphic larval characteristics. They recognized only three genera in the Dipseudo- psinae, Dipseudopsis, Phylocentropus, and Protodipseu- dopsis, and did not consider the status of Eodipseu- dopsisand Limnoecetis. However, Schmid (1980) later Table 2. Data matrix for the families of Hydropsychoidea and genera of Dipseudopsidae, including outgroup Philopotamidae. Character Il 12345 67890 Taxon Philopotamidae 00000 00000 Psychomyiidae 11011 00000 Xiphocentronidae 11011 00000 Hydropsychidae 11100 10000 Polycentropodidae 11100 11000 Ecnomidae 11100 11000 Dipseudopsis 11101 01111 Protodipseudopsis 11101 01111 Phylocentropus 11101 01111 Limnoecetis 1110? 0112? Hyalopsyche 11101 01111 Hyalopsychella 11??? ER aaa TiD, 2222 12345 67890 00000 10000 0000 00000 01000 0000 00000 00110 0000 00000 00001 1100 00000 00000 0000 00000 00000 0011 01111 00000 0000 01110 00000 0000 00000 00000 0000 01110 20 FORT 10000 00000 0000 10000 ae PERE 97 IHDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 made the novel assignment of placing Phylocentropus in the family Hyalopsychidae Lestage, on the basis of adult characteristics shared with the genus Hyalopsyche Ulmer (1904b). Wells & Cartwright (1993) recently concurred with the position of Ross and Gibbs, and based on the morphology of the larva and female of Hyalopsyche, added this genus to the Dipseudopsidae, and accordingly proposed the suppression of the family group name Hyalopsychidae. Our phylogenetic anal- vsis tests the classifications of Ross & Gibbs and Wells & Cartwright, versus that of Schmid. Results Phylogeny. — The analysis of the data matrix via Hennig86 generated three trees with different topolo- gies, having length 28, CI 0.89, RI 0.91 (fig. 55). All three trees depict Dipseudopsidae as a monophyletic group, comprising six genera. The first tree has the fe- west number of ancestors, having 18 nodes, and is also identical to the strict consensus tree (fig. 56) of the three trees. The consensus tree depicts ((Psycho- myiidae, Xiphocentronidae Ross) ((Hydropsychidae, Ecnomidae Ulmer, Polycentropodidae) Dipseudo- psidae)), thus supporting monophyly of the Hydro- psychoidea. The clade comprising Hydropsychidae, Polycentropodidae, Ecnomidae is recognized as the sister group of the Dipseudopsidae. A few differences are noted among the three original trees: 1) trees 0 and 2 depict the trichotomy (Hydropsychidae, Ecno- midae, Polycentropodidae), which is resolved as (Hy- dropsychidae (Ecnomidae, Polycentropodidae)) in tree 1; 2) trees 0 and 1 each have a basal and a distal trichotomy among the genera of Dipseudopsidae, both of which are resolved in tree 2. Characters. — The combination of characteristics of the female genitalia as exhibited in the Dipseudo- psidae (having synapomorphy 2, lateral papillae of segment X present, but lacking synapomorphy 6, ab- dominal sternite VIII not subdivided into lateral lobes) does not represent an apomorphic state, but only a unique combination of characteristics which happens to define the family very well, but only in a taxonomic sense. Synapomorphy 7, inferior appen- dage one-segmented is a homoplasy in Polycentro- podidae and Ecnomidae. Three synapomorphies 8- 10 support monophyly of Dipseudopsidae, and the first of these, antennal bases close together, is not unique, but within the Hydropsychoidea is also present in Hydropsychinae and Macronematinae Hydropsychidae). However, since this characteristic is absent for the most part in Arctopsychinae and Diplectroninae, it is not part of the groundplan of Hydropsychidae, and thus is a homoplasy. A more ac- curate description of this condition as a homologue is 98 that the antennal scapes of Dipseudopsidae are not merely close together but rather are so enlarged that they occupy most of the space between the eyes along the anterodorsal margin of the head (fig. 13). Conclusions. — Monophyly of the Dipseudopsidae, including Dipseudopsis, Hyalopsyche, Hyalopsychella Ulmer (1930), Limnoecetis, Phylocentropus, and Protodipseudopsis, is unequivocal. Two clades are rec- ognized among these genera, (Hyalopsyche + Hyalo- psychella) supported by synapomorphy 11 (cf. table 1), and (Dipseudopsis + Protodipseudopsis + Limnoe- cetis) supported by synapomorphies 12-14. Hence, the sister group of the genus Dipseudopsis is unclear. The adult of Limnoecetis, a monotypic genus endem- ic to Lake Tanganyika, are quite remarkable, having unique adaptations to walk on water. The striking au- tapomorphies of Limnoecetis present a formidable challenge in resolving the trichotomy (Dipseudopsis + Protodipseudopsis + Limnoecetis). One obvious predic- tion based on our phylogeny, is that the unknown lar- val forms of Hyalopsychella and Limnoecetis probably share many similarities with the dipseudopsid larvae known. The phylogeny also provides evidence for re- moving Eodipseudopsis from the Dipseudopsidae (cf. final section of text). The sister group relationship of the genus Phylocentropus remains uncertain. Phylocentropus phe- netically agrees closely with Hyalopsyche, but the sim- ilarities of these two genera are predominantly plesio- morphic. Thus, the phylogenetic analysis above does not provide supporting evidence for Hyalopsychidae s. str. Schmid (1980) and in fact, infers that this tax- on might be paraphyletic. Therefore we concur with Ross and Gibbs for the most part on the classification of Phylocentropus, and Wells and Cartwright on the placement of Hyalopsyche and the suppression of the Hyalopsychidae, a taxon that presently serves no use- ful purpose in phylogenetics. Keys to genera of Dipseudopsidae Larvae The larvae of Hyalopsychella Ulmer and Limnoecetis Marlier are unknown. I. Mandibles with apicolateral teeth; head capsule short, with lateral margins nearly parallel and its total length less than its maximum width (Ulmer 1957,,Gıbbs LOGS) re RI 2 — Mandibles without apicolateral teeth; head cap- sule long, with lateral margins tapering anteriad, and with its length slightly greater than its maxi- mum width (Wiggins 1977, Wells & Cartwright 1993) nnt nennen tree SRE RR 3 2. Frontoclypeal suture V-shaped with lateral lines tapering irregularly posteriad toward the dorsal eoronalisuture (Ulmer 1957) Aven Rene octobre Dipseudopsis Walker — Frontoclypeal suture with the posterior portion U-shaped, and the anterolateral lines nearly par- allel between the eyes and then curving mesad to meet the coronal suture posteriad (Gibbs 1968) ene ia) Boe mc in. Protodipseudopsis Ulmer 3. Frontoclypeal suture sessile in respect to posterior margin, posterior 1/3 nearly regular V-shaped and without hourglass-shaped constriction (Wells & Cartwright 1993) .. Hyalopsyche Ulmer — Frontoclypeal suture with short petiole joining posterior margin, posterior 1/3 more irregular V- shaped and with hourglass-shaped constriction (Wieeins 97) Phylocentropus Banks Adults 1. Pronotum with setaceous warts conspicuous and without deep median fissure … n 2 — Pronotum with setaceous warts inconspicuous andawithideep median fissure... 4 2. Labial palps present; forewing with fork I long ANSE le BAER Phylocentropus Banks — Labial palps absent; forewing with fork I short andipetiolatetor'absent) ti: ile 3} 3. Fore and hind wings with fork I short and petio- late nee, Hyalopsyche Ulmer — Fore and hind wings with fork I absent .............. Re ete ke Hyalopsychella Ulmer 4. Tibial spurs 1, 2, 2; female with maxillary palps three-segmented; one species endemic to Lake Hansanylat nee ins Limnoecetis Marlier — Tibial spurs 3, 4, 4; female with maxillary palps five-segmented; Africa and Asia. … … … … … … 5 5. Mouthparts with maxilla modified into a probos- cis; hindwing with median cell closed; male hind mesoapical tibial spur modified, often twisted and/or branched; Asia and Africa … EEA dn Dipseudopsis Walker — Mouthparts without a proboscis; hindwing with median cell open; male hind mesoapical tibial spur normal; Africa ...... Protodipseudopsis Ulmer SYSTEMATIC PART Dipseudopsis Walker, 1852 Dipseudopsis Walker, 1852: 91. — Type species: Dipseudopsis capensis Walker (monobasic). Nesopsyche McLachlan 1866: 168. Type species: Neopsyche flavisignata McLachlan (monobasic). Esperona Navás 1915: 397. — Type species: Esperona orienta- lis Navás (monobasic). Bathytinodes Iwata 1927: 235. — Type species: Bathytinodes albus Iwata (monobasic). WEAVER & MALICKY: Dipseudopsis from Asia Dipseudopsodes Lestage 1936: 170 (invalid, type species not designated). The genus Dipseudopsis was proposed by Walker (1852) with the description of D. capensis from South Africa and placed in the family Sericostomatidae Stephens, a family designation which was common for caddisflies, exhibiting bizarre characteristics, that did not agree with the diagnoses of other families. Later when McLachlan (1863) added a second species to the genus, D. collaris from China, he placed Di- pseudopsis in the family Rhyacophilidae. Brauer (1868) placed it in the Hydropsychidae s. lat, near Cyrnus Stephens. However, after examination of the adult mouthparts, Ulmer (1904a, b) placed the genus in the Polycentropodidae, subfamily Dipseudopsinae. Accordingly, most workers have placed the genus in Polycentropodidae (Navás 1913, Martynov 1935, Schmid 1949). However, Banks (1931b) maintained a conservative approach, having never recognized the full family status of Polycentropodidae, and followed Brauer’s placement of the genus in the Hydropsychi- dae s. lat. Etymology. — Dipseudopsis: Greek di, two; pseudos, false; 0ps, eye; feminine. The actual etymology is un- certain. The basis of this name is uncertain also, since there are no noticeable characters such as wing spots present in the type species of the genus, D. capensis. Perhaps it was based on the appearance of the prono- tum, having the setal areas expanded and divided by a deep median fissure. Adult Head and thorax (figs. 12A-B, 13, 14): The head has a sparse number of short inconspicuous setae and is nearly glabrous; sometimes it is glossy and very gla- brous. The setal areas of both the head and thorax are not raised like typical setaceous warts, but are flush with other surrounding sclerites. The setae of the bo- dy and wings are mostly short sparse and inconspi- cuous, giving the imago a glabrous appearance. The pronotum is raised to the same level as the head and mesonotum and is divided by a deep median furrow, probably resulting from modifications of the lateral setal areas, which have remained separate, but have become both enlarged and glabrous. The tegula are flat and sit like epaulets on the bases of the forewings. Legs with tibial spur formula complete: 3, 4, 4. The head, thorax, and sometimes the anterior segments of the abdomen have patterns of yellowish or orangish brown contrasting with dark brown. The forewing is narrow, length ranging from 10 to 20 mm, usually mostly brown with a hyaline spot at m-cu and the ar- culus, and sometimes with a pattern of translucent 99 [iyDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 bands or spots against a dark background, or with a striped pattern of dark veins against lighter translu- cent membranes; fork I is either short and petiolate or absent, II and IV are long and sessile, III and V are long and petiolate, and the discal and medial cells are closed. The hindwing is shorter and sometimes bro- ader than the forewing, being somewhat triangular and lighter in colour, sometimes nearly translucent; forks II and V are long and sessile, forks I, III, IV ab- sent, the discal cell closed, and unlike other dipseud- opsid genera described the median cell is closed. The scent glands of sternum V are present in the female and absent in the male, their openings are rather in- conspicuous and lack exterior projections. This con- dition differs in Phylocentropus placidus (Banks) exa- mined, having the scent gland absent in both sexes. The mouthparts (fig. 12B, 13, 14) have been de- scribed by Ulmer (1904a), Cummings (1913) and Crichton (1957: fig. 53), and are quite unique, hav- ing the lacinia modified into a proboscis, functioning as a siphoning tube. The two processes which form the proboscis are not held together in preserved spec- imens, but curl away from each other distally. Each process is shaped like half of a long slender straw, with its longitudinal edges fringed with numerous short scalelike setae, having many transverse ridges along its shaft (thus being semiannulate), and having several mesal dentations at its base. The proboscis is at least as long as the first 2-4 segments of the maxillary palps, or usually about the same length as the labium, including base plus palps. The haustellum is slender, about as long as the base of the labium, and its apex is bilobed. The frontoclypeus is large and slightly bul- bous, apparently well-developed to support a pump- ing organ. Protodipseudopsis differs noticeably from Dipseudopsis by lacking a well-developed proboscis (fig. 12C); its frontoclypeus is shorter and less bul- bous and apparently the pumping organ is reduced. Male. — The male is dark brown, sometimes with contrasting patterns in the wings and body, and more striking than in the female. The mesoapical spur of the hind tibia is modified, often twisted like a cork- screw and sometimes branched into two or more pro- cesses. It is interesting to note that the same spur is al- so modified in Phylocentropus auriceps (Banks), (Schuster & Hamilton 1984), among the other dipseudopsids. being exceptional Genitalia (fig. The male genitalia are blunt, generally lacking in conspicuous characteristics, and the processes of the genitalia are retracted and ob- scured by the large blunt preanal appendages. This blunt appearance of the male terminalia, especially i in dry material, probably is why males sometimes have been mistaken as females, e.g. the types of D. flavisig- 100 nata (McLachlan, 1866) and D. benardi Navás (1930). Segment IX with articulation between tergite and sternite, resembling an open jaw in lateral view; the tergum and sternum also articulate with the pre- anal appendages. The tergum of segment IX is much smaller than the sternum and usually overhangs seg- ment X, and its apical margin may be triangular, rounded, or bilobed in dorsal view. The sternum of segment IX is heavily sclerotized and is the most mas- sive feature of genitalia, providing support for the in- ferior appendages, the phallus, and the ventral articu- lation of the preanal appendages. The sternum of segment IX is a simple structure except for the meso- superior processes that, along with segment X, form a hood that hangs over the dorsal portion of the phallus which is mostly membranous. Segment X is inferior to segment IX, and subequal in length to the phallus. The phallus is mostly membranous dorsad and heav- ily sclerotized ventrad. Hence, segment X and the phallus appear to function in concert, with segment X providing dorsal coverage and support to the phallus. The inferior appendages are thumblike, with apices directly dorsad, perhaps modified to hold segment X and the phallus together, as well as clasping the fe- male during copulation. Female. — Females are usually lighter in colour than the males, and as a result they are more dull and have fewer contrasting colour patterns in their wings and bodies. However, the contrasting colouration of an unidentified female specimen (fig. 2) may be excep- tional. In several species, such as D. nervosa (figs. 3, 4), where the males have brown wings with large translucent spots, the females have striped forewings with dark veins against a translucent background. Sometimes females have forewings with faded indis- tinct spots that are remotely similar to translucent spots in the forewings of males. However, the female forewings are most often dull brown and show no striking colour patterns. Genitalia (figs. 15-16): Sternum VIII forms a sin- gle plate, not divided into lateral lobes as in the Polycentropodidae. Segment IX has a well-sclerotized tergite, as in some species such as D. recta (fig. 15), but is more membranous in others (fig. 16); the ster- num is mostly membranous. The vaginal apparatus and the vulval scale are attached to the sternum of segment IX by sclerotized bands and membranous folds. Both the vaginal apparatus and the vulval scale are sclerotized. The length of the vaginal apparatus is variable, being long in D. recta, where it extends ante- riad reaching the anterior margin of sternite VIII (fig. 15), but is much shorter in other species (fig. 16). The genitalia are generally similar to those of Phylocentropus (Schmid 1980, Schuster and Hamil- ton 1984), except that segment IX is more heavily sclerotized. This diagnosis is provisional, because the females of most species remain undescribed. We ob- serve that the females of some species can be distin- guished from others by colour patterns of the body and wings, and characteristics of the genitalia. Larva The larva of Dipseudopsis was first described by Iwata (1927), although under a different name. A key separating the larvae of Dipseudopsis and Protodi- pseudopsis was provided by Ulmer (1957), and a simi- lar key was provided by Marlier (1962), whereby the larvae of Dipseudopsis and Protodipseudopsis were dis- tinguished by the gular sclerite, mandibles, and hind tarsal claw. Protodipseudopsis and Dipseudopsis are unique in having mandibles with apicolateral teeth, a probable synapomorphy (Ulmer 1957: f. 401; Gibbs 1968: fig. 1; Marlier 1962: fig. 50; Marlier 1979: fig. 5). The known larvae of Dipseudopsis known can be distinguished from those of other dipseudopsids by having head capsule short, lateral margins nearly pa- rallel with total length less than its maximum width, frontoclypeal suture V-shaped and the lateral lines ta- pering irregularly and posteriad toward the dorsal co- ronal suture, and mandibles with apicolateral teeth. Larval behaviour and habitat. — The larval tube- dwelling and feeding-behaviour have been described for Protodipseudopsis (Gibbs 1968) and Phylocentropus (Wallace et al. 1976, Wiggins 1977). The larvae of these genera live in the bottom substrates of streams or lakes and construct a somewhat Y-shaped silken tube composed of sand grains and detritus. The ends of the two upper branches, usually one vertical and the other inclined, are open and exposed to the stream current, the rest of the tube is located within the bottom substrate, and the end of the bottom branch is closed. The opening of the vertical branch is an intake and the inclined branch an outlet, through which a directional flow is established via rapid oscil- lations of the larva. This forced flow enables a fine capture net, within an enlarged chamber of the outlet branch, to collect fine organic food particles for the larva to eat. The larval tubes of Hyalopsyche are branched and appear to be similar to those of Phylocentropus (Wells & Cartwright 1993), as are the tube-dwellings of Dipseudopsis. Therefore, since branched tube-dwellings are characteristic for the lar- vae of Dipseudopsis, Hyalopsyche, Protodipseudopsis, and Phylocentropus, this type of larval retreat and its associated mode of feeding-behaviour is a synapo- morphy of Dipseudopsidae. The larva retreat of the hydropsychid genus Macrostemum Kolenati (Wiggins 1977, as Macronema Pictet) is generally similar, but differs in many details and is certainly a homoplasy. WEAVER & MALICKY: Dipseudopsis from Asia Ulmer’s (1957: 303) description of the larval re- treat and biology of Dipseudopsis is translated as fol- lows: ‘Larvae and pupae live in calm pools of rivers, also in lakes and lake-outflows, in the mud or fine sandy substrates; the larvae not free-living there, but construct tubes out of secretions, covered with mud etc., in the depths; these tube-dwellings are more or less bent and twisted, elastic and flexible, but never- theless fairly resistant; sometimes they are branched into 2 or 3 tubes, but only one of the branches reach- es the [substrate] surface and there it is closed, while the other lower ends are open.’ It seems obvious that Ulmer’s orientation of the tube-dwelling is upside down; his description should depict the two open ends reaching the surface, and the one closed branch completely buried in the substrate. This would agree more with the tube-dwellings described in detail for Phylocentropus and Protodipseudopsis, and the descrip- tion by Scott (1985: 334) for Dipseudopsis in south- ern Africa: ‘The larvae are bottom dwellers in stand- ing or slow-flowing water where they construct U-shaped silk-lined tubes with a transverse net in mud or silt or on submerged branches, filtering out algae and organic particles on which they feed.” The larvae and pupae of Dipseudopsis inhabit a range of freshwater habitats, including the sandy dep- ositional zones of lakes and the pool areas of rivers and streams. Marlier (1953, 1962) reports that im- matures in Africa were found in shallow water, while Tsuda (1939) reported that larvae in Japan were dis- covered at a depth of 90 m. The collections of D. diehli Malicky & Weaver, at elevations of 1,400 m in Sumatra, devoid of large rivers and lakes, suggests that some species also live in mountain streams. Distribution Dipseudopsis is known to occur generally through- out the old world tropics of Africa and Asia. The African fauna recorded from the northern Nile River Valley, including Egypt, Sudan, and throughout most of the Afrotropical Region, from Ethiopia and Kenya to Nigeria, Ghana and Sierra Leone, and to Zaire, Mozambique, South Africa, and Madagascar. The genus is represented by 42 described species, 23 from the African continent, and 19 from Madagascar. However, in light of the problems discovered in this work, we suspect that the African species are in need of a thorough taxonomic revision. The Asian fauna with 36 recognized species is known throughout most of the Oriental Region and part of the southeastern Palearctic Region, including Pakistan, India, Sri Lanka, Nepal, Bangladesh, Burma, Thailand, Cambodia, Vietnam, China, Japan, Philippines, Malaysia, and Indonesia, including Sumatra, Java, Borneo and Sulawesi. 101 [1IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137 Checklist of Dipseudopsis from Asia 102 adiaturix sp. n. \ p. 102 benardi Naväs, 1930. p. 104 langana Navás, 1930 syn. n. arculata Naväs, 1931 syn. n. ulmeri Schmid & Denning, 1979 syn. n. bicolorata Martynov, 1935. p. 107 collaris McLachlan, 1863. p- 109 stellata McLachlan, 1875 syn. n. alba (Iwata, 1927). bakeri Banks, 1916 syn. n. discors Navás, 1924 syn. n. contorta Banks, 1931b. p- 110 diehli Malicky & Weaver, 1988. p. 112 digitata Ulmer, 1907a. plu? dochleri Ulmer, 1929 (döhler:). p- 115 elongata Banks, 1920. p- 116 flavisignata (McLachlan, 1866) p- 116 flinti sp. n. p. 116 immaculata Ulmer, 1905. p. 118 moesta Banks, 1931b syn. n. indica McLachlan, 1875. p- 120 buddha Banks, 1913 syn. n. i. sindica Martynov, 1935 syn. n. infuscata McLachlan, 1875. p- 120 knappi Schmid & Denning, 1979. p. 121 lamellata Martynov, 1935. p. 121 lucasi sp. n. p- 123 maculata Ulmer, 1907a. p. 123 malaisei sp. n. p. 124 martynovi sp. n. p. 124 modesta Banks, 1911. p. 124 pallida Martynov, 1935 syn. n. morosa Banks, 1924. p- 125 nebulosa Albarda, 1881. p. 127 nervosa Brauer, 1868. p'127 luctuosa Banks, 1913. venosa Naväs, 1931 syn. n. nervosella Ulmer, 1951 stat. n. p. 128 nieuwenhuisi Ulmer, 1909. p- 130 notata (Fabricius, 1781). p. 130 horni Ulmer, 1915 syn. n. recta Martynov, 1935. p- 131 bombayana Martynov, 1935 syn. n. robustior robustior Ulmer, 1929 stat. n. palS52 akhila Schmid & I )enning, 1979 syn. n. hailandica Schmid & Denning, 1979 syn. n. junki Marlier, 1979 syn. n. robustior andamanensis ssp. n. p. 134 hmidi sp. n. p. 134 bılıs Ban 193la p. 134 batensis Malicky & Weaver, 1988. p- 135 nkın i 1921 p. 135 ata Mar ‚1935 p. 137 29. p. 138 petersorum Schmid & Denning, 1979 syn. n. voluta Ulmer, 1906. p. 138 onychophora Navás, 1935 nomen dubium p. 139 orientalis (Navás, 1913) nomen dubium p- 139 Dipseudopsis adiaturix sp. n. (fig. 18) Type material. — Holotype d : INDONESIA: Sumatra, Huta Padang, 26.1.1990, Diehl (CLHM). Male. — Head with dorsum and frontoclypeus dark brown, almost black, but posterior dorsal warts yello- wish brown. Basal segments of antennae brown (type with distal parts of antennae and mouthparts missing due to ant attack). Thorax, tergites and sternites dark brown. Coxae of fore and mid legs brown, coxa of hind legs yellowish brown. Basal % of fore femora dark brown, femora of mid and hind legs with basal % brownish; distal % of all femora, tibia of fore and mid legs, and all tarsi yellow; distal % of hind tibia brown. Modified spur (fig. 18) long and slender, more than % as long as adjacent tarsal segment, more than 2x long- er than outer spur, distal claw about X as long as spur and bent obliquely laterally, and with short lateral thorn at % length from base. Forewing 16 mm, slen- der, dark brown, but with nervation darker, having hyaline lunula at m-cu, and indistinct translucent spots between M, and Cu, and near apex of 1A and 2A. Hindwing 10 mm, dark brown, with hyaline lu- nula at m-cu. Abdomen yellowish brown. Genitalia (18A-D): Tergum IX broad and bilobate with me- dian sized cavity below in dorsal view; sternum IX with mesosuperior process fingerlike in lateral view. Dorsal edge of segment X with strong incision in lat- eral view, but apex not incised. Preanal appendages long and triangular in lateral view. Inferior appenda- ges curved dorsad in lateral view, with distal part slightly clavate and basolateral mesal margins squarish in caudal view. Phallus calyx-like in ventral view, about % as long as segment X. Etymology. — A noun in apposition: named after a Celt who lived in Austria 2000 years ago. Distribution. — Indonesia: Sumatra. Remarks. — This species is only known from the unique type. It has a modified spur similar to D. flav- isignata, being long slender, acuminate, slightly sinu- ate and bearing a short subapical lateral thorn. However, it differs in having the male modified spur with the lateral thorn closer to its apex, rather than near the base, the male forewings mostly dark brown and without a striking pattern of translucent stripes and by characteristics of the male genitalia. Pictorial atlas (p. 103-106) A pictorial atlas is provided for quick species deter- mination. The figures of the following male charac- ters are provided for each species: 1) dorsal aspect of segment IX and X, 2) shape and relative size of the left modified spur, and 3) caudal aspect of the right infe- D. nebulosa Albarda AN) Burma, Thailand, Sumatra, Malay Pen. vA D. immaculata Ulmer | LA Sumatra, Borneo, Malay Pen. D. schmidi Weaver and DI India, Bangladesh D. varians Ulmer È Thailand, Malay Pen. \ D. doehleri Ulmer | AM Assam, Burma, Thailand WEAVER & MALICKY: Dipseudopsis from Asia rior appendage (but for subspecies, only figures of the modified spur). The distributional data are listed be- low the figures. Possible identifications should be confirmed with the complete species descriptions and figures in the text. D. flavisignata (McLachlan) | A Sulawesi D. adiaturix Weaver and Malicky Sumatra D. voluta Ulmer Sulawesi D. lamellata Martynov India, Burma, N, D. indica NV India, Bangladesh 103 7,1994 [NDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 13 D. collaris McLachlan _— China, Japan, Philippines D. robustior Ulmer ue Burma, Thailand, Vietnam, Cambodia, Malay Pen. D. r. andamanensis Weaver and Malicky Andaman Is. D. infuscata McLachlan TEEN i Sumatra, Borneo, Java, Sulawesi D. notata (Fab.) PL India, Sri Lanka Dipseudopsis benardi Naväs, 1930 hgs. 10, 19) is si 1905: 96, fig. 75b [misdet]. is benardi Navás, 1930: 141, Type d*, VIETNAM Phu Lang Thuong, 1909, det. P. Navás S. IS 5 *, same data (MNHN). — Fischer D. diehli Malicky and Weaver Sumatra \ i Borneo D. knappi Schmid and Denning AWW Thailand D. nervosa Brauer Philippines 1962: 5; Fischer 1972: 3. Dipseudopsis langana Navas, 1930: 141, Type ®*, VIETNAM ‘Tonkin’: Phu Lang Thuong, 1909, det. P. Navás S. J., G. Benard (MNHN). — Fischer 1962: 10; Fischer 1972: 5. Syn. n. Dipseudopsis arculata Navás, 1931: 85-86, Holotype &*, CHINA: Guangfong: ‘Cong-tcheou’ Canton (MZBS). — Fischer 1962: 5; Fischer 1972: 2. Syn. n. D. morosa Banks Philippines D. nervosella Ulmer VES Borneo D. maculata Ulmer O En) à = \ D. nieuwenhuisi yh À D. digitata Ulmer AS Philippines Dipseudopsis moesta. — Ulmer 1951: 128 [misdet]. Dipseudopsis ulmeri Schmid & Denning, 1979: 243, fig. la- e, Holotype d*, THAILAND: Chiang Mai Prov, E fork Mae Ping, 56km N of Chiang Mai, 1300ft, at light, 24.X1.1964, Peters (CLDD). Syn. n. Specimens examined. — CAMBODIA: 1d 29, ‘Cambodge’ WEAVER & MALICKY: Dipseudopsis from Asia D. bicolorata Martynov D. contorta Banks AD Sulawesi, Borneo, Java, Malay Pen. D. martynovi Weaver and Malicky We D. tonkinensis Navas WI Erratum: D. contorta, read Sumatra for Sulawesi. Philippines Vietnam, Hainan Pavie, 1886 (MNHN). — CHINA: Fujian: 1d, ‘Foochow Fuchou, 1936-37, det. D. stellata by Mosely, M. S. Yang (- BMNH). — THAILAND: 2d, Chiang Mai, 19.V.1952, D. & E. Thurman (USNM). 1d, Doi Sutep, 1952, D. & E. Thurman (MCZC). 16, Bangkok, 5.X.1929, H. Smith (MCzc). 16, Ping River, Chiang Mai, 9.IV.1989, Malicky (CLHM). 26, Ping River near Lampun, 20.XII.1989, Malicky (CLHM). 105 TNDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 D. stabatensis Malicky and Weaver = - = Tre Sumatra, Borneo D. flinti Weaver and Malicky m ES Selangor D. malaisei Weaver and Malicky AM È JJ D. benardi Navàs AS pronti =, Thailand, Vietnam, Cambodia, China D. ee Martynov = India, Nepal 24-31.X.1988, LHM). 2d, Kanchanaburi, 5, 14.V.1988; Allen (CLHM). Zoo, light trap, ı mostly dark brown but pos- 106 D. modesta Banks India W D. triclavata Martynov , India, Sri Lanka WF D. lucasi Weaver and Malicky W India D. elongata Banks Borneo | terior warts light brown; frontoclypeus light brown but dorsal portion darker; antennal scape and pedicel light brown, flagellum dark. Pronotum, mesonotum and tegula dark brown. Forewing (fig. 10): 13.5-16.5 mm, mostly dark brown, but with translucent spots: distinct small translucent spot just distal of r; small AU Figures 1-2. — Fig. 1, Dipseudopsis bicolorata Martynov, male, dorsal. Fig. 2, Dipseudopsis sp. female, dorsal, India, Manipur, Nungha. faded spots just distal of transverse cord, hyaline lu- nula at m-cu, large translucent spot at arculus. Legs brown. Modified spur (fig. 19): basal % unmodified and apical % bifid with two short acuminate points curved, spiraling opposite each other distally, each point slightly longer than width of main shaft. Genitalia (fig. 19): Tergum IX with posterior margin curved in dorsal view, with posterior % extending over segment X in lateral view; sternum IX having mesos- uperior process with short dorsal triangular point and long slender distal point, extending above basal % of phallus in lateral view. Segment X tapering distally into slender lobe in lateral view, extending posterior- ly distal to preanal appendages about as far as inferior appendages, apex bilobed. Preanal appendages with posterior margin not distinctly incised in lateral view, having blunt dorsal point, dorsal and ventral margins converging to truncate posterior margin. Inferior ap- pendages thumblike with obtuse midventral angle in lateral view. Distribution. — Cambodia; China: Guangdong, Fujian; Thailand; Vietnam. Remarks. — This species can be recognized by hav- ing the modified spur with two short apical points WEAVER & MALICKY: Dipseudopsis from Asta spiraling opposite from each other, and the forewing longer and darker brown than most other species in southern China and Indochina. The types of D. be- nardi, D. arculata, D. langana, and D. ulmeri were ex- amined, and the latter three are recognized as junior synonyms of D. benardi. The type of D. benardi is a male, not a female as previously reported by Naväs (1930). The type of D. langana is recognized as a fe- male of D. benardi, because it is has an identical loca- lity label as the type of D. benardi and is similar in size and general appearance. Ulmer (1905: fig 75b) provided a figure of the male modified spur of an undetermined specimen of Dipseudopsis from ‘Cambodscha, Museum Paris’ and Ulmer (1951: 128) later identified this figure as D. moesta (syn. n. of D. immaculata). Ulmer’s figure is more similar to the spur of D. benardi, and is prob- ably not D. immaculata because this species is un- known from Cambodia. Also, we have examined a male specimen of D. benardi from Cambodia in the MNHN, which might be the same specimen figured by Ulmer. Dipseudopsis bicolorata Martynov, 1935 (figs. 1, 20) Dipseudopsis bicolorata Martynov, 1935: 165-166, figs. 68a- c, 69a-b, Type series: INDIA: ‘26, 2. Castle Rock, North Kanara district, Bombay Pres. October 1916. S. W. Kemp … d. Talewadi, Castle Rock, N. Kanara district, Bombay Pres. 3-10.1916. S. W. Kemp’. The type materi- al present in the collection of the Zoological Survey of India is in poor condition and too delicate to send by mail (pers. comm. Dr. K. Rai). Syntype 12 *: det. Martynov, Karnataka, N. Kanara dist., Castle Rock, in alcohol, X.1916, S. Kemp (ZRAS). — Fischer 1962: 5; Higler 1992: 80. Specimens examined. — INDIA: Karnataka: 1d, Shimoga Dist, Agumbe Ghat 2000ft, V.1974, Lucas (CLHM). Tamil Nadu: 46, Kattalaimala, 25-26.XII.1961, F. Schmid (- CNCI). 1d, Ottakada, 5.1.1962, F. Schmid (CNCI). Male. — Body and wings (fig. 1). Head dorsum and pronotum orangish brown, frontoclypeus mostly dark brown but dorsal margin orangish brown, genae yellowish, antennal scape orangish, pedicel and flagel- lum brown, palps dark brown, but stipes orangish brown, maxillary palps 3.2 mm, proboscis 1.7 mm, as long as first three segments of maxillary palps, labial palps 1.4 mm; cervical sclerites and mesonotum dark brown. Forewing 16-21 mm, almost unicoloured dark brown, but darker along anterior margin between C and R and posterior margin, only hyaline spots at m-cu and arculus. Front leg brown, middle and hind legs with coxae, trochanters and femora orangish brown, distal segments brown. Modified spur (fig. 20) having one apical point, curving in 107 psis nervosa Brauer: Fig. 3, male. Fig. 4, female. Fig. 5, D. doehleri Ulmer, male. Fig. 6, D. D. nebulosa Albarda, male. Fig. 8, D. digitata Ulmer, male. Fig. 9, D. spectabilis Banks, male male forewing. Fig. 11, D. flinti, n. sp., male forewing. Figures 12A-C. — Fig. 12A, Dipseudopsis nebulosa Al- barda, male, head and tho- rax, dorsal. Fig. 12B, D. doehleri Ulmer, male, head with mouthparts, ventral. Fig. 12C, Protodipseudopsis sjoestedti Ulmer, male, head with mouthparts, anterior. plane perpendicular to longitudinal axis of spur. Genitalia (fig. 20): Tergum IX triangular with blunt apex in dorsal view, posterior margin only slightly overhanging segment X in lateral view. Segment X ellipsoidal with length 1% X basal width in dorsal view, tapered distad and apex incised. Preanal appendages with dorsal margin irregular in lateral view, directed dorsad with apex obtuse or sometimes rounded and ventral margin broadly rounded. Inferior appendages with massive dorsal ex- tension that doubles height at middle as compared to base in lateral view, with distinct apicomesal tooth in ventral view, and with curved apicolateral flanges. Distribution. — India: Karnataka, Tamil Nadu. Remarks. — This is the largest species of Dipseudo- psis known from India and is easily recognized by its colouration of the males which superficially resemble ctenuchid moths. The orangish brown dorsum of the head and pronotum contrast strikingly with the uni- coloured, dark brown mesonotum and forewings. The modified spur bears a single curved apical point that lies in a plane perpendicular to the longitudinal axis of the spur. The specimen in fig. 2 is possibly a female of this species. WEAVER & MALICKY: Dipseudopsis from Asia Dipseudopsis collaris McLachlan, 1863 (fig. 21) Dipseudopsis collaris McLachlan, 1863: 496, pl. 19: fig. 6, Lectotype d *, selected by Kimmins (1957: 101), Hong Kong: (BMNH). Cotype 16*: Hong Kong: Soochow (MCZC 11080). Weidner (1964): Paratype 2, CHINA: Hong Kong (ZMUH). — Fischer 1962: 7; Weidner 1964: 73; Fischer 1972: 3. Dipseudopsis stellata McLachlan, 1875: 16-17, pl. 2: fig. 11, Ho- lotype d (Kimmins, 1957: 101): [CHINA: Zhejiang:] Shang- hai, McL. coll. (BMNH); 3d * [possible syntypes]: Shanghai, McL. coll. (BMNH); ‘Paratype’ 19*: ‘N China’ McL. coll (BMNH). — Fischer 1962: 15. Fischer 1972: 7-8. Syn. n. nec Dipseudopsis stellata. — Ulmer 1915: 73 (Sri Lanka); Lestage 1925: 65, 115 (Java); Banks 1931b: 401 (stellatus, Thailand); Tsuda 1942: 266 (Thailand); Schmid 1958: 35, 89 (Sri Lanka). Dipseudopsis nebulosa. — Banks 1914: 178 (Luzon) [misdet.]. Dipseudopsis bakeri Banks, 1916: 215-216, pl. 2: figs. 20, 21, Type d *, PHILIPPINES: Luzon, Laguna, Mt Maquiling and Los Baños,Baker (MCZC 11768). Only 1d at the MCZC labelled as type, although several other specimens, not labelled as types, at the MCZC and USNM are part of the type series. — Fischer 1962: 5; Fischer 1972: 3. Syn. n. Dipseudopsis discors Navas, 1924: 205. — Fischer 1962: 8. Syn. n. Dipseudopsis morosa. — Banks 1924: 450 (pro parte). Syntype 16* is actually D. collaris Luzon: Mt Maquiling, Baker (MCZC 14826). Bathytinodes alba Iwata, 1927: 209-210, pl. 4: fig. 87-91. Dipseudopsis alba (Iwata, 1927). — Fischer 1962: 217; Fischer 1972: 2. 109 IHDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Scape tentorial pit abrum i mandible lacın aania haustellum ee pap maxillary palp 13 13-14. — Fig. 13, Dipseudopsis sp., male head, anteri- modified after (Crichton 1957). Fig. 14, Dipseudopsis oehleri Ulmer, male head and prothorax, lateral. Specimens examined. CHINA: Guangdong: 36, Hushan, 3.V1.1963, Tian (CLHM). Jiangxi: 1d 2d, det. 1905) as D. stellata, “Kiang-si’, 1869, A. David Hong Kong: 1d, Foochow (MCZC). Jiangsu: 64, Fu 18.VIII.1987, Tian (CLHM). 8d 19, Jinhu, 22.V1.1987, Sun Changhai (USNM). Zhejiang: 16 19, ( kiang Hangtchéou’, Hangzhou, 1925, A. Pichon Shanghai, det. McLachlan as D. stellata, Shusan, 20.V1.1957, Tian (CLHM). JAPAN: 14, PPINES: Luzon: 8d 19 Los iM), 18 (BMNH). 16, Manila Mt Maquiling, Baker (USNM), 24 ld 14 (BMNH). 36 19, Balinag, x, wings and legs with colour- 5 e robustior. Forewing 13-15 mm, rn of 5- itor milar to D small translucent spots around discal cell, having most spots distal to cord, and spot at arculus adjoining hyaline area at apex of Cu,. Modified spur (fig. 21) apical % bifid, crescent- shaped, with two subequal curved points. Genitalia (fig. 21): Tergum IX triangular in dorsal view, slight- ly overhanging segment X in lateral view; sternum IX mesosuperior process ellipsoidal with blunt apex pointed posteriad, extending above basal % of phallus. Preanal appendages with posterior margin slightly in- cised in lateral view, dorsal margin at right angle to posterior margin, ventral margin at obtuse angle, sometimes varied with ventral margin continuous with posterior margin. Inferior appendages fingerlike with short basoventral shelf in lateral view. Distribution. — China: Guangdong, Hong Kong, Jiangsu, Jiangxi, Zhejiang; Philippines: Luzon; Japan: Honshu. Remarks. — This species is similar to D. robustior, but differs by having inferior appendages more broad and fingerlike in lateral view (although this is vari- able) and modified spur with two apical points both crescent-shaped and subequal in length. In D. robust- ior the inferior appendage is more slender and angled, its modified spur bears two points, one point slender- er, more curved and more than 2x longer than other. Previous records of D. collaris, as D. stellata, from Sri Lanka, Java and Thailand are probably in error. Dipseudopsis contorta Banks, 1931 (fig. 22) Dipseudopsis sp. — Ulmer, 1905: 96, fig. 75c (not fig. 75e as listed by Fischer 1962). Dipseudopsis infuscata. — Ulmer 1905: 96, fig. 75d; Ulmer 1951: 128 [prob. misdet.]. Dipseudopsis contorta Banks, 1931b: 400-401, fig. 20, lecto- type by present designation d*: MALAYSIA: West Ma- laysia: Selangor, Kuala Lumpur, 29.1.1924, at light, H. M. Pendlebury (BMNH); paralectotype 1d *, without hind legs: Pahang, Kuala Tahan, 300ft at light, 23.X1.1921, H. M. Pendlebury (MCZC 16480). — Fischer 1962: 8; Malicky & Weaver 1988: 3-4, fig. 3a-e. Specimens examined. — 2d, ‘Borneo’ det. Ulmer 1905 as D. infuscata (MNHN). INDONESIA: Kalimantan: 1d, Pontianak, VII.1907, F. Muir (USNM). Java: 16, [Muller?] (RMNH). Sumatra: 1d, Dolok Merangir, 19.VII-20.VI- 11.1971, Diehl (CLHM). 16, same data, V-X.1980, Diehl (CLHM). 1d, Ost-Aceh 26-30.VIII. 1972, Diehl (CLHM). 1 d, NE Sumatra, 16.1-18.11.1979, Diehl (CLHM). Male. — Head dorsum dark brown, except posteri- or warts orangish brown, frontoclypeus dark brown, antennae orangish brown; cervical sclerites dark brown, pronotum light orangish brown and mesono- tum dark brown. Maxillary palps 2.2 mm, brown; Figures 15-16. — Fig. 15, Dipseudopsis recta Marty- nov, female: 15A, genitalia, lateral; 15B, genitalia, ven- tral; 15C, abdominal seg- ment VI, lateral. Fig. 16, Dipseudopsis sp. female, ‘Manipur Nungha’ (CNC): 16A, genitalia, lateral; 16B, genitalia, ventral. proboscis 1.6 mm, as long as first two segments of maxillary palps. Forewing 10-12.5 mm, brown with row of lighter spots from stigma to Cu, just distal to cord, and large translucent spot at arculus, hyaline spot at m-cu. Legs mostly light brown, except front and middle coxae dark brown. Modified spur (fig. 22) distal 4 branched into two subequal curved points, with points diverging basally but converging distally, sometimes with rounded subapical shelf. Genitalia (fig. 22): Tergum IX semicircular in dorsal view, with posterior margin having almost no exte- nsion overhanging segment X in lateral view; sternum IX with mesosuperior process short, slender, acumi- nate having apex curved dorsad in lateral view. Segment X ellipsoidal in dorsal view, long trapezoidal in lateral view, extended distad beyond apex of inferi- WEAVER & MALICKY: Dipseudopsis from Asia or appendage. Preanal appendages variable, either D- shaped with lobe extending distally as far as inferior appendages in lateral view, or with posterior margin incised and with short dorsoapical blunt point. Inferior appendages with basoventral notch and apic- oventral obtuse angle in lateral view, apical % variable, either equal in height to base and truncate, or some- times more expanded and clavate; with apex and apic- olateral margin curved in ventral view, but bearing subapicomesal obtuse tooth. Distribution. — Indonesia: Kalimantan, Sumatra, Java; Malaysia: West Malaysia. Remarks. — This species is similar to D. collaris, but differs by having the inferior appendage capitate, with 111 THDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 tergite IX preanal appendage on X | re inferior appendage — — Figure 17. — Fig. 17, Dipseudopsis sp., male genitalia, lateral. a basodorsal notch and an apicoventral obtuse angle in lateral view, and the modified spur with apical points directed more laterally. Slight variation is no- ticeable in the male modified spur and genitalia (cf. Malicky & Weaver 1988). Dipseudopsis diehli Malicky & Weaver, 1988 (fig. 23) Dipseudopsis diehli Malicky & Weaver, 1988: 8, fig. Ga-f. Holotype d*: INDONESIA: Sumatra: Prapat, 1400m, Diehl (CLHM); paratypes 27 d : Prapat, Bukit Tinggi, Aek Farum, Sitahoan, Dolok Merangir (CLHM, CLJW). Specimens examined. — INDONESIA: Sumatra: Pematang Siantar: 5d, 1.VII.84-13.IV.1985; 34, 1.VI-12.XII.1985; 3d, 22.IV-11.VII.198G; 56 , 1.IX.88-15.1.1989; 16, 9.III- 10.X.1989, Diehl (CLHM). Male. — Head dorsum yellowish brown with dark semicircular spot at anterior margin. Frontoclypeus brown. Scape yellowish brown, distal portion of an- renna brown. Maxillary palps 3.0 mm, proboscis 2.4 mm, extending as far as 4% segment of maxillary palp. Cervical sclerites and pronotum yellowish brown. Mesonotum and tegulae brown, with scutellum yello- wish brown. Forewing 16-20 mm, brown with four translucent areas: 1) long stripe adjacent to costal margin; 2) ellipsoidal spot between stems of R and M; 3) transverse band distal to cord, between R, and M,, vith two V-shaped incisions along distal margin, and having narrow connection with anterior margin; 4) riangular spot at arculus. Hindwing unicoloured brown, but with hyaline comma at m-cu. Legs yello- vish brown, with light brownish tibiae, apparently 112 caused by more dense covering of fine brown setae. First segment of hind tarsi uniquely bent mesad about X of its length to permit free motion of broad modified spur. Modified spur (fig. 23) with apex bi- fid with two sinuate points nearly parallel and slight- ly unequal in length, both curved mesad, but apices curved distad. Genitalia (fig. 23): tergum IX with ap- ical 4 hanging over base of segment X in lateral view; broad and short with posterior margin bilobed or slightly triangular in dorsal view, anterior margin slightly bilobed. Sternum IX mesosuperior process extending to end of phallus. Segment X ellipsoidal in dorsal view. Preanal appendages triangular in lateral view with dorsal margin slightly inclined dorsad and ventroposterior margin curved steeply dorsad. Inferior appendages fingerlike and with ventral mar- gin slightly concave in lateral view, but dorsal margin serrate in middle; with U-shaped basomesal notch in ventral view, distal arms slightly diverging apically and each tapering into small apical lobe. Distribution. — Indonesia: Sumatra. Remarks. — This species is easily recognized by hav- ing the male modified spur extremely broad and bi- furcate, with points curved a wide whorl, and by hav- ing forewing with large translucent spots. Some specimens are darker brown, having brown parts that are usually yellowish brown in other species. Dipseudopsis digitata Ulmer, 1907 (figs. 8, 24) Dipseudopsis digitata Ulmer, 1907a: 39-40, figs. 56-57, Holotype d *: MALAYSIA: Banguey Is, Staudinger (ZMHB). — Fischer 1962: 8; Fischer 1972: 3. Specimens examined. — PHILIPPINES: Balabac: 2d, Dalawan Bay, 5.X.1961; 18, 9.X.1961; 38, 12.X.1961 (ZMUC). Busuanga: 5d, 4 km N San Nicolas, 20- 29.V.1962, H. Holtmann (BPBM). Culion: 68 19, 6km W Culion, 7-11.VI.1962, light trap, H. Holtmann (BPBM). Palawan: 16, Brooke’s Point, Macagua, 75m, 1-4.IV.1962 (BPBM). 196 19, Tarumpitao Point, H. E. Milliron (BPBM). 176 79, Irawan, 14km W Puerto Princessa, 5m, D. R. Davis (USNM). 1d, Litso, Amoyan Ck., 57km N Puerto Princessa, 10m, 10.XII.1965, D. R. Davis (USNM). 16, Mainit, 11km NW Brooke’s Point, 18.XI.1965, D. R. Davis (USNM). 1d, Chromite Mine, 28km W Puerto Princessa, 400m, 1-7.XI1.1965, D. R. Davis (USNM). 306, Brooke’s Point, Uring Uring, 14-23.VIII.1961 (ZMUC). Male. — Body and wings (fig. 8). Head, thorax and forewing with pattern of light and dark bands. Head dorsum with dark brown median band widening an- teriad and narrowing posteriad, lateral areas light brown and posterior warts lighter. Frontoclypeus mostly dark brown, but with yellowish V-shaped band pointed dorsad. Antennae light brown. Maxillary palps 2.3 mm, proboscis 1.4 mm, as long as WEAVER & MALICKY: Dipseudopsis from Asia Figures 18-21. — Dipseudopsis, A-H, male genitalia: A, lateral; B, preanal appendages, segments IX and X, dorsal; C, inferior appendages, ventral; D, phallus, lateral; E, phallus, ventral; F, segments IX, X, dorsal; G, ventral; H, left inferior appendage, ventral. I-J, male modified apical spurs of hind tibia, ventral. Fig. 18, D. adiaturix, n. sp. Fig. 19, D. benardi Navas. Fig. 20, D. bicolorata Martynov. Fig. 21, D. collaris McLachlan, 21H, type of D. stellata McLachlan. 1113) SCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 23A ehleri Ulmer. first three segments of maxillary palp. Cervical scler- ites dark brown ventrally, light brown dorsally. Pronotum with raised areas light brown, median fur- row and lateral sides dark brown. Mesonotum and te- gula dark brown. Forewing 10-14 mm, with striped pattern of dark brown veins and lighter cells, notice- ably darker along thicker veins, R, and Cu,. Legs with coxae dark brown and trochanters and distal seg- ments light brown, except front coxa bicoloured, dark brown ventrad and lighter dorsad. Modified spur (fig. 24) with three minute apical points and one long acu- minate subapical point curved laterally, all points dis- tal to blunt subapical shelf. Genitalia (fig. 24): Tergum IX short and truncate in lateral view, posteri- or margin not overhanging segment X, rectangular in dorsal view; sternum IX small, with massive mesos- uperior process having large broad dorsal lobe and slender posteroventral process extending above basal % of phallus. Segment X with pair of minute basodor- sal lateral lobes, dorsal margin angled obliquely ven- trad in lateral view, but with apex turned dorsad, ven- tral margin straight but also turned dorsad apically; about % as wide as tergum IX and about twice as long as wide in dorsal view, lateral sides parallel but ta- pered into triangular apex, and apex slightly incised. Preanal appendages broad and not incised in lateral view, somewhat triangular with dorsal margin and ventral margin curved and converging at blunt dorsal apex. Inferior appendages capitate in lateral view, dorsal margin concave, ventral margin straight with short basal shelf, apex truncate with corners slightly curved, extended dorsad. Distribution. — Malaysia: Banguey Is; Philippines: Balabac, Busuanga, Culion, Palawan. Remarks. — This species is easily recognized, having the male segment X with a pair of minute dorsolateral lobes at its base, and modified spur with three minute apical points, one long acuminate subapical point curved laterally, and all points distal to a squarish sub- apical shelf. Dipseudopsis doehleri Ulmer, 1929 (figs. 5, 12B, 14, 25) Dipseudopsis döhleri Ulmer, 1929: 189-190, fig. 27, Holotype d *: INDIA: Meghalaya: Khasi Hills, McL. coll (BMNH). — Fischer 1962: 8. Dipseudopsis doehleri Ulmer. — Fischer 1972: 4; Higler 1992: 80 Specimens examined. — BURMA: 16, Washaung, 20km E of Myitkyina, 200m, 4.VII.34, Malaise (NHRS). 1d, Myitkyina, 175m, 1934, Malaise (BMNH). INDIA: Assam: 1d, 8mi E Ledo, 125m, 27.34N 95.34E, 13.X.1961, Ross & Cavagnaro (CASC). 29, Abhoypur For., Naphuk, 360m, 12.X.1961, Ross & Cavagnaro (CASC). THAILAND: 9d, WEAVER & MALICKY: Dipseudopsis from Asia Nakhon Nayok Prov., Khao Yai National Park, 700m, 29.IX-6.X.1984, Karsholt, Lomholdt & Nielsen (ZMUC). 7d, Nam Nao, Pet Cha Boon, 5.V.1989, Allen (CLHM). 2064, Puk Hieo, Chayapoom, 6.V.1989, Allen (CLHM). Male. — Body and wings (fig. 5). Head and all of its appendages dark brown, except proboscis yellowish. Mouthparts (fig. 12B, 14): Maxillary palps 3.6 mm; proboscis 2.7 mm, extended to midlength of apical segment of maxillary palp; haustellum 0.8 mm; labial palps 1.7 mm. Cervical sclerites and prothorax ivory. Mesonotum, tegula and metanotum dark brown. Forewing 16-17 mm, dark brown with three distinct translucent spots: 1) basal triangle between R, and Cu, from base to fork I of Rs, 2) subapical spot distal to cord with apical margin incised at forks II, and III) ellipsoidal spot at arculus, similar as in D. nebulosa ex- cept for translucent spot at arculus. Legs dark brown, except front coxa with lateral side ivory. Hindwing with large translucent triangle as in front wing, apical region brown, anal region translucent, Cu, dark brown. Modified spur (fig. 25) similar to D. immacu- lata, unbranched, slightly longer than adjacent spur, basal 7 unmodified, apical % twisted 180° into short spiral, apical point directed mesad. Abdomen with anterior sternites II-V ivory, giving translucent trian- gles of forewings a white background so that contrast- ing pattern is displayed when wings are folded, ster- nites VI-VIIT dark brown. Genitalia (fig. 25): Tergum IX triangular with posterior margin rounded in dorsal view, apical % overhanging segment X in lat- eral view. Segment X long slender in lateral view, with short base having parallel sides in dorsal view, and ap- ical portion cordate and apex deeply incised. Preanal appendages abbreviated with slender dorsoapical point in lateral view, but truncate and triangular in dorsal view. Inferior appendages long and thumblike, without incisions or points, curved slightly dorsad in lateral view. Distribution. — Burma; India: Assam, Meghalaya; Thailand. Remarks. — This species is most similar to D. neb- ulosa, but differs by having the head entirely dark brown and pronotum ivory, forewing with translu- cent round spot at the arculus, and modified spur rel- atively shorter, with less than % of apical portion twisted. It also differs by having male genitalia with segment X and the preanal appendages more slender in lateral view, and segment X also bears a unique pair of basolateral humps in dorsal view. A teneral male specimen from Hainan Island, possibly an unde- scribed species near D. doehleri, was observed in the collection at the MCZC. 115 THDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Dipseudopsis elongata Banks, 1920 (fig. 26) Dipseudopsis OBEN Banks, 1920: 361-362, pl. 1: fig. 13, Type Z*. MALAYSIA: Sabah, Sandakan, Baker (MCZC 10882). - Fischer 1962: 8-9; Fischer 1972: 4. Specimens examined. — MALAYSIA: Sabah: ld, Sandakan Bay, NW Sepilok For. Res., 1-10m, DGOS ZT: Gressitt (BPBM). 14, ‘North Borneo’ McL. coll (BMNH). Male. — Head dorsum glossy dark brown, but pos- terior warts light brown, frontoclypeus glossy dark brown, genae yellowish. Cervical sclerites, pronotum, mesonotum tegula dull dark brown. Antenna brown, but scape light brown. Palps brown, maxillary palps 3.0 mm; proboscis 2.0 mm, as long as first 3% seg- ments of maxillaty palp. Forewing 16 mm, uniformly dark brown, wich small hyaline spots at nygma and m-cu, and large translucent spot at arculus. Front and middle leg with coxa, trochanter and basal % of femur dark brown, and distal portion of leg light brown; hind leg light brown. Modified spur (fig. 26) long and truncate, about 3X longer than adjacent spur, with apex more than 2X as wide as base, and minute apical points curved mesad. Genitalia (fig. 26): Tergum IX short and slightly overhanging segment X in lateral view, posterior margin rectangular with lat- eral edges curved in dorsal view; sternum IX with el- ongate mesosuperior process, extending above % of phallus in lateral view. Segment X acuminate with blunt apex in lateral view; broad ellipsoidal with apex slightly incised in dorsal view. Preanal appendages tri- angular with corners rounded and dorsal margin slightly concave in lateral view, ventroposterior mar- gin curved dorsad, apex broad and blunt. Inferior ap- pendages capitate with curved truncate apex in lateral VIEW. Distribution. — Malaysia: Sabah. Remarks. — This species was previously only known from the unique type from Sabah. Two addi- tional male EEDE: were found in the material bor- rowed from BPBM and BMNH. The male is distin- guished by SRE modified spur long and truncate, about 3% longer than adjacent spur, with apex more than 2x wider than base, and with minute apical points curving mesad. D li a (McLachlan, 1866) fig Mel Lach! an, 1866: 269, pl. 17: fig. 6, IDONESIA: Sulawesi: Ujung ‚ Wallace coll. ( (BMNH). The separate slides by Mosely En rs 1.1 cle D randang, cient ViaCassar mounted on two (1933), one slide with wings mounted dry, and the sec- ond slide with body mounted in balsam, but prothorax absent. Dipseudopsis flavisignata (McLachlan, 1866). — Mosely 1933: 499-502, figs. 1-7. Fischer 1962: 9. Fischer 1972: 4, Male. — Head dorsum dark brown, yellowish brown between antenna, frontoclypeus dark brown. Antenna base of scape yellowish brown and remain- der dark brown. Maxillary palps 2.3 mm dark brown, labial palps 0.8 mm yellowish. Proboscis 1.7 mm about as long as first four segments of maxillary palp. Mesothorax and abdomen black. Forewing 12 mm, similar to that of D. spectabilis (fig. 9), dark brown background with large yellowish spots: 1) irregular band on anterior margin tapering toward base, 2) subapical C-shaped spot having distal side open and filled with dark narrow stripe along M, (but in de- scription by Mosely (1933: fig. 1) the latter is depict- ed as an irregular O-shaped spot), 3) semicircular ir- regular band through anal region, running from base, distad through anal loops and thyridial cell to arculus. Front leg with coxa, trochanter and basal portion of femur dark, apical portion yellowish; Middle and hind leg with similar colour pattern but lighter. Modified spur (fig. 27) nearly as long as first tarsal segment, basal X unmodified, distal slender twisted blade, bearing a small tooth at base, a ridge at mid- length and apical % flattened like a blade. Genitalia (fig. 27): Tergum IX posterior margin broad at base in dorsal view, but abruptly tapering into slender fin- gerlike extension, apical 4 overhanging segment X in lateral view. Segment X long ovoid projection taper- ing slightly posteriad in dorsal view. Preanal appenda- ges with dorsal margin nearly horizontal, meeting curved ventroposterior margin at small dorsoapical lobe in lateral view. Inferior appendages similar as in D. spectabilis, heavily sclerotized, and darker brown than other appendages, basal 4 angled dorsally, and apical % fingerlike and angled more posteriorly in lat- eral view. Distribution. — Indonesia: Sulawesi. Remarks. — This species is only known from the unique type. It is a sibling species to D. spectabilis, but differs by having the forewing with a yellowish C- shaped subapical spot, almost closed except for a nar- row dark strip along M,, and by having the modified spur with one short subapical point distant from api- cal point. Dipseudopsis flinti sp. n. (figs. 11, 28) Type material. — Holotype d : MALAYSIA: West Malaysia: WEAVER & MALICKY: Dipseudopsis from Asia 29E 29C Figures 26-29. — Dipseudopsis, A-G, male genitalia: A, lateral; B, preanal appendages, segments IX and X, dorsal; C, inferior appendages, ventral; D, phallus, lateral; E, phallus, ventral; F, segments IX, X, dorsal; G, ventral. I-J, male modified apical spurs of hind tibia, ventral. Fig. 26, D. elongata Banks. Fig. 27, D. flavisignata (McLachlan). Fig. 28, D. flinti, n. sp. Fig. 29, D. immaculata Ulmer. 117 TIIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Selangor, Kuala Lumpur, 21.1V.1931, H. M. Pendlebury, (- MCZC 33382 Male. — Head completely glossy dark brown, al- most black, antennae and mouthparts dark brown. Maxillary palps 2.3 mm, proboscis 1.3 mm, as long as first three segments of maxillary palp. Pronotum dull dark brown, except anterior margin light, mesono- tum and tegula dark brown and metanotum light brown. Forewing 12 mm, dark brown except for slen- der light brown stripe along posterior margin, run- ning from cell 3A and expanded at arculus; cells be- low M, and Cu, light brown distally. Front leg with coxa, trochanter and basal X of femur dark brown, distal portions of leg light brown; middle leg wich coxa dark brown and distal segments light brown; hind leg light brown. Modified spur (fig. 28) about 2x longer than adjacent spur, having less than apical % bifid with two short apical points parallel and angled distad toward tarsus. Genitalia (fig. 28): Tergum IX posterior margin broadly bilobed in dorsal view; IX sternum base rectangular in lateral view, mesosuper- ior process with blunt dorsal lobe and slender posteri- or projection with apex curved anteriad. Preanal ap- pendages trapezoidal with dorsal margin slightly inclined dorsad toward blunt distal apex in lateral view, ventral margin curved gradually into posterior margin. Segment X about % as wide as tergum IX in dorsal view, somewhat rectangular with posterior margin broadly bilobed. Inferior appendages capitate with apical } ellipsoidal and about 3x as thick as base in lateral view, lateral arms slender and fingerlike with truncate apex in ventral view. Etymology. — Named after Oliver S. Flint, Jr. Distribution. — Malaysia: West Malaysia. Remarks. — This species is only known from the unique type. The modified spur of the male is similar to that of D. stabatensis, but it differs by having two apical points parallel and angled toward the tarsus, the adjacent spur is relatively shorter, about % as long as the modified spur; in D. stabatensis the apical points of the modified spur are not parallel but are di- rected in opposite directions, and the adjacent spur is relatively longer, % as long as the modified spur. Also, D. flinti differs by having the inferior appendage cap- tate in lateral view, with the apical / ellipsoidal and bout 3x as broad as the base. Dipseudopsis immaculata Ulmer, 1905 ) y 9 ita Ulmer, 1905: 92-93, 96, figs. 68b, | HN). — Fischer 1962: DOTI in alcohol (Mr 9; Weidner 1964: 73; Fischer 1972: 4; Malicky & Weaver 1988: 4, fig. la-e. Dipseudopsis moesta Banks, 1931b: 400, 401, fig. 9, Type d*: MALAYSIA: West Malaysia: Johor, Lubok, Kedondong, NW Gunung Ledang (Mt Ophir), Lalang & Belukar, H. C. Abraham (MCZC 16481). — Fischer 1962: 11. Syn. n. nec Dipseudopsis moesta. — Ulmer 1951: 128 [reidentifica- tion of Ulmer 1905, fig. 75b sub nom. D. stellata) [= D benardi]. Specimens examined. — INDONESIA: Sumatra: 26 , Karny, Pakan Baroe, at light, 23.X.1925 (RMNH). 1d, Huta Padang, 20.1.1990, Diehl (CLHM). 1d, Dolok Merangir, 15.V1.82-4.1.1983, Diehl (CLHM). MALAYSIA: Sarawak: 14, Merirai Valley near Kapft, 180m, 28.VII.-6.VIII.1958, T. C. Maa (BPBM). West Malaysia: 1d 19, Pahang, Rompin Mining Co. railway track, 31km, Petoh Swamp, 11.XII.1960 (BPBM). Male. — Head dorsum, cervical sclerite and prono- tum light orangish brown; frontoclypeus mostly dark brown, but lateral margins yellowish brown; genae yellowish brown except dark brown ventral spike. Antennae light brown. Mouthparts brown, maxillary palps 2.2 mm; proboscis 1.2 mm, about as long as first three segments of maxillary palps. Mesonotum and tegula dark brown, about as dark as eyes. Legs light brown. Modified spur (fig. 29) slighdy longer than adjacent spur, apical 4 modified into twisted point with setose basal pocket; apex almost truncate, but lateral side curved and other side with short point directed toward tarsus. Tegula dark brown. Forewing 11-12 mm, almost uniformly brown, except for hya- line spots at m-cu and arculus. Genitalia (fig. 29): Tergum IX triangular in dorsal view, with about % of posterior overhanging segment X in lateral view. Segment X thumblike in lateral view, extending pos- teriad as far as inferior appendage; broadly cordate in dorsal view with apex slightly incised. Preanal appen- dages rhomboidal or sometimes simply rounded in lateral view. Inferior appendages oblong and truncate in lateral view, ellipsoidal in ventral view, but api- comesal margin incised with blunt subapicomesal tooth. Distribution. — Indonesia: Sarawak, West Malaysia. Sumatra; Malaysia: Remarks. — The male of this species has a modified spur which is similar to that of D. doehleri and D. nebulosa, but can be distinguished from these by its almost uniformly dark brown forewing. It also differs by having the male genitalia with more than % of ter- gum IX overhanging segment X, and the inferior ap- pendage with a blunt subapicomesal tooth in ventral view. A lighter phase is noted in male specimens from Borneo, having apical region of the forewing brown, remainder nearly translucent except for several dark WEAVER & MALICKY: Dipseudopsis from Asia 32A 33E Figures 30-33. — Dipseudopsis, A-G, male genitalia: A, lateral; B, preanal appendages, segments IX and X, dorsal; C, inferior appendages, ventral; D, phallus, lateral; E, phallus, ventral; F, segments IX, X, dorsal; G, ventral. I-J, male modified spurs ap- ical spurs of hind tibia, ventral. Fig. 30, D. indica McLachlan. Fig. 31, D. infuscata McLachlan. Fig. 32, D. knappi Schmid and Denning. Fig. 33, D. lamellata Martynov. 119 TNDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 veins. A slight variation also is observed in males from Sumatra, having forewing with narrow lighter strip along anterior margin and indistinct translucent area in outer X of wing, and preanal appendage more no- ticeably incised distally. Regarding specimen dete- rmined as D. stellata by Ulmer (1905: fig. 75b) and later determined as D. moesta by Ulmer (1951), cf. ‘Remarks’ of D. benardi above. Dipseudopsis indica McLachlan, 1875 | fig. 30) Dipseudopsis indica McLachlan, 1875: 18-20, pl. 2: figs. 13- 14 (indicus), Type ‘Habitus in India orientali.’ Lectotype ‘Dia, designated by Kimmins (1957: 101). — Fischer 1962: 10; Fischer 1972: 4; Higler 1992: 80. Dipseudopsis buddha Banks, 1913: 239, pl. 23: fig. 8, pl. 24: fig. 19, Type INDIA: West Bengal, Chapra, Mackenzie, no date (MCZC 11757). — Fischer 1962: 5; Higler 1992: 80. Syn. n. Dipseudopsis indica sindica Martynov, 1935: 159, 206, fig. 59a-b, Type: Pakistan: ‘8. Bubak, Sind, from pools. 12.X1.1927. B. Prashad and B. N. Chopra.’ — Fischer 1962: 10; Fischer 1972: 4. Syn. n. Specimens examined. — BANGLADESH: 18d 369, Chicknagul, 10mi N Sylhet, 10m, 30.IX.61, Ross & Cavagnaro (CASC). 5d 62, Cormilla, 50ft, 23.IX.61, Ross & Cavagnaro (CASC). INDIA: Bihar: 1d, Pusa, at light, 22.- V111.08 (MCZC). 1d, Pusa, IV.1919, M. A. Husain (BMNH). 1d, Pusa, 2.VIII.1924, Mukerjee (BMNH). Orissa: 26 , det. Martynov, Balighai, near Puri, 16-20.VIII.11 (ZRAS). 16, Cuttack, 6.111.1944, at light (BMNH). 2d, Bhubaneswar, 11.1983, J. Olah (CLHM). West Bengal: 1d, Chapra, Mackenzie (MCZC). 64 322, Malda 110m, 28.X.61, Ross 2d 29, 10mi SE Asansol 175m, 6.X1.61, Ross & Cavagnaro (CASC). Pakistan: Sind: 16, Makli, near Thatta, 22.IX.1976, Hevel & Dietz (USNM). 28 1%, Makli, near Hyderabad, 24.IX.1976, Hevel & Dietz (- & Cavagnaro (CASC). Male. — Colour dull brown, with eyes dark brown. Pronotum lighter brown than head and mesonotum. Maxillary palps 2.0 mm; proboscis 0.8 mm, as long as first 34 segments of maxillary palp. Legs light brown. Modified spur (fig. 30) with apical 4 modified, subdi- vided into two acuminate branches, first branch bifid with apical points unequal and both directed distally, econd branch curved, directed laterally with minute point at its base. Forewing 11-14 mm, mostly uni- dull brown, but costal margin lighter brown, with arc of faint translucent spots distad of cord, cen- spot at m-cu, and two translucent spots at arcu- Genitalia (fig. 30): Tergum IX only slightly rhanging segment X in lateral view, posterior mar- in dorsal view, but middle slightly rnum IX short, with mesosuperior in lateral view, extending above egment X dorsal margin inclined rad with blunt apex lobiform in lateral view, el- 120 lipsoidal and long in dorsal view. Preanal appendages with dorsal margin straight and horizontal in lateral view, ventroposterior margin curved apicodorsal an- gle squarish. Inferior appendages capitate in lateral view, with short ventrobasal shelf, narrow middle sec- tion and broad squarish apex; lateral arms with api- comesal tooth in ventral view. Distribution. — Bangladesh; India: Bihar, Orissa, West Bengal; Pakistan: Sind. Remarks. — This species is easily recognized by the complex male modified spur, having two branches, one branch directed distally and bifid, and the other branch curved laterally and acuminate, but with a minute point at its base. Examination of the type of D. buddha revealed that it is identical to D. indica. Based on its original description, the subspecies D. i. sindica is recognized as a junior synonym of D. indi- ca. Dipseudopsis infuscata McLachlan, 1875 (fig. 31) Dipseudopsis infuscata McLachlan, 1875: 15-16, pl. 2: figs. 6-10, Holotype d*: INDONESIA: Sulawesi, Ujung Pandang ‘Piepers Celebes Macassar’ (RMNH). — Fischer 1962: 10; Fischer 1972: 4; Malicky & Weaver 1988: 3, fig.4a-e. nec Dipseudopsis infuscata. — Ulmer 1905: 96, fig. 75d; Ulmer 1951: 128; Malicky 1984: 214, 216 Specimens examined. — INDONESIA: Java, Jakarta ‘Bata- via: 1d, 11.1908; 16, VIII.1907; 16, IX.1907 (RMNH); 1d, 1.1908, Jacobson (ZMAN). Kalimantan: 16 39, Telang X.1881 (MCZC). 1d, West Java, [Piepers?], McL. coll (BMNH). Sumatra: 2d, Palembang, 13.V.1972, Diehl (CLHM). Male. — Head and antennae brown. Maxillary palps 1.6-1.8 mm, proboscis 0.5-1.0 mm as long as first two or three segments of maxillary palp. Antenna yellowish brown with fine brown setae that become less dense apically so that distal part appears more yel- lowish. Forewing 12 mm, brown with small translu- cent spot distal to r, hyaline spots at m-cu and arcu- lus. Modified spur (fig. 31) slightly longer than adjacent spur, bifid with two acuminate apical points, longer point straight directed distally and almost % as long as long as spur, shorter point curved with apex directed mesad. Genitalia (fig. 31): Tergum IX with posterior margin rounded in dorsal view, with slight protrusion in middle, short and not extended over segment X in lateral view. Segment X with basal % of dorsal margin inclined ventrad apical portion direct- ed posteriad, having apex truncate, in lateral view; dorsal and ventral margins narrowing toward apex, with apex truncate and slightly emarginate in dorsal view. Preanal appendages with posterior margin in- cised in lateral view. Inferior appendages with dorsal and ventral margins nearly parallel and curved dorsad in lateral view, having thumblike apex; base with me- sal margin squarish in ventral view, distal portion long with mesal margin slightly convex, apicolateral margin rounded, and with apicomesal point. Distribution. — Indonesia: Java, Kalimantan, Sulawesi, Sumatra. Remarks. — This species is similar to D. robustior, but differs by having the two apical points of the modified spur with the longer point straight and slightly shorter than the unmodified base of the spur, and the shorter point curved with its apex directed mesad. All specimens determined as D. infuscata by Malicky (1984) from South Andaman Island are now considered to be D. robustior andamanensis Weaver & Malicky, n. ssp. The specimen referred to by Ulmer (1905: fig, 75d, 1951) is D. contorta, not D. infuscata (cf. D. contorta above). Dipseudopsis knappi Schmid & Denning, 1979 (fig. 32) Dipseudopsis knappi Schmid & Denning, 1979: 246, 248, fig. 5a-e, Holotype d : THAILAND: Tha Phra, Khom Kaen, 17.V.1968, blacklight trap, F. W. Knapp (CASC). Male. — Body brown. Forewing 13 mm, brown with scattered translucent spots. Modified spur (fig. 32) trifid with long curved acuminate point and slightly shorter fingerlike lobe bearing minute subap- ical point. Genitalia (fig. 32): Tergum IX broadly rounded in dorsal view, short in lateral view; sternum IX with mesosuperior process triangular in lateral view. Segment X slender with apex distinctly bilobed in dorsal view, lateral margins sinuate, base of dorsal margin slightly inclined dorsally distad in lateral view, apical % horizontal with ventral margin concave, apex slender and fingerlike. Preanal appendages trapezoid- al with corners rounded and dorsal margin inclined dorsad in lateral view, having blunt apicodorsal lobe. Inferior appendages slightly clavate with basodorsal shelf in lateral view, middle portion constricted, and having clavate apex. Distribution. — Thailand. Remarks. — This species is only known from the type series from northeastern Thailand. The modified spur of D. knappi is trifid and somewhat similar to that of D. spectabilis, but differs by having second longest process of the spur lobiform, rather than pointed. The male genitalia differ by having segment WEAVER & MALICKY: Dipseudopsis from Asia X with a bilobed apex. The diagnosis of the male giv- en above is based on the original species description (Schmid & Denning 1979). Dipseudopsis lamellata Martynov, 1935 (fig. 33) Dipseudopsis lamellata Martynov, 1935: 160-161, figs. 60a- b, 61, Type series: BURMA: Myitkyina Dist.: 1d 129, Lonton, W shores Indawgyi Lake, U. Burma, 18- 31.X.1926, B. N. Chopra; Syntype 19*: Kamaing, U. Burma, X.1926, B. N. Chopra (NZSI); INDIA: Assam: Mangaldai, 26.26N 92.02E, 16-18.X.1910, S. W. Kemp. Fischer 1962: 10; Higler 1992: 80. Specimens examined. — BANGLADESH: 1d, Chicknagul, 10mi N Sylhet, 10m, 30.IX.61, Ross & Cavagnaro (CASC). 2d 49, Harbang For. 6mi N Chiringa, 50ft, 18.IX.61, Ross & Cavagnaro (CASC). 1d, Umgebung Dacca, 1- 15.V.1976, Dietz (CLHM). INDIA: Assam: 26 39, Kohara, Kaziranga, 110m, 26.36N 93.28E, 7-16.X.61, Ross & Cavagnaro (CASC). Orissa: 28, Bhubaneswar, 11.1983, J. Oläh (CLHM). Male. — Colour brown. Forewing 13-15 mm, colouration similar as in D. indica, mostly brown, but with arc of five faint translucent spots distad of cord, central hyaline spot at m-cu, and broad translucent spot at arculus. Modified spur (fig. 33) 2x longer than adjacent spur, bearing two distal processes, longer process slender twisted and fingerlike, shorter process slender acuminate and bent laterad at right angle. Genitalia (fig. 33): Tergum IX broad triangular in dorsal view, apex slightly overhanging segment X in lateral view; sternum IX short, with mesosuperior process having blunt dorsal lobe and slender ventr- oposterior process in lateral view. Segment X dorsal margin concave, ventral margin convex, apical part narrow and curved dorsad in lateral view; lateral mar- gins slightly convergent at basal %, distal portion elon- gate ellipsoidal, with apex slightly incised in dorsal view. Preanal appendages slender and somewhat trap- ezoidal in lateral view, with posterior margin strongly incised, bearing two blunt apical points. Inferior ap- pendages capitate in lateral view, with base constrict- ed, having short slender basal projection and widen- ing into rhomboid apex; basomesal margin with deep curved incision in ventral view, distal portion wid- ened, having apex with lateral margin curved and with short apicomesal point. Distribution. — Bangladesh; Burma; India: Assam, Orissa. Remarks. — This species differs from other species of Dipseudopsis by having a unique modified spur, bi- fid, with a long slender fingerlike twisted process, and with an acuminate process angled laterad. 121 vooR ENTOMOLOGIE, VOLUME 137, 1994 e genitalia: A, lateral; B, preanal appendages, segments IX and X, dorsal; C, inferior ; È, phallus, ventral; F, segments IX, X, dorsal; G, ventral. I-J, male modified spurs ap- D, lucasi, n. sp. Fig. 35, D. maculata Ulmer. Fig. 36, D. malaisei, n. sp. Fig. 37, D. Dipseudopsis lucasi sp. n. (fig. 34) Type material. — Holotype d: INDIA: Karnataka: Shimoga Dist., Agumbe Ghat, 2000ft, 1974, Lucas (CLHM). Paratypes: 7d 39, Someshwar, 27.1.1959, F. Schmid (CNCI). 14, Nagodi, 28.1.1959, F. Schmid (CNCI). 1d, Kerala: Sangalipalam, 20-21.XI1.1961, F. Schmid (CNCI). Male. — Head brown, eyes darker brown, fronto- clypeus brown, antennae brown but ventral sides lighter brown, palps brown, maxillary palps 2.8 mm, labial palps 1.0 mm, proboscis 1.2 mm, almost as long as first three segments of maxillary palp. Pronotum dark brown, but lateral sides lighter, mes- onotum and tegula dark brown. Forewing 14-17 mm unicoloured dark brown except small hyaline spots at m-cu and arculus. Legs with coxae dark brown, distal segments light brown. Modified spur (fig. 34) slight- ly longer than adjacent spur, apical % modified with short curved subapical shelf, and long slender acumi- nate curved point directed distad. Genitalia (fig. 34): Tergum IX with posterior % overhanging X in lateral view, posterior margin broadly rounded in dorsal view; sternum IX mesosuperior process with middle constricted and apex curved abruptly dorsad in lateral view. Segment X thumblike and directly posteriad in lateral view; elongate ovoid, with apex slightly incised in dorsal view. Preanal appendages short, extending to middle of segment X in lateral view, posterior mar- gin broadly rounded, almost semicircular but with blunt apicodorsal angle. Inferior appendages basal % thick and broadened toward middle in lateral view, apical 4% tapered to truncate apex; lateral margin rounded and mesal margin almost straight in ventral view, but with obtuse apicomesal angle and minute basomesal angle. Etymology. — Named after the collector J.A.W. Lucas. Distribution. — India: Karnataka, Kerala. Remarks. — The male genitalia of D. /ucasi are sim- ilar to those of D. doehleri and D. nebulosa, having ab- breviated preanal appendages. However, D. lucasi dif- fers by having forewing almost unicoloured dark brown, and lacking the contrasting pattern, as in the forewings of D. doehleri and D. nebulosa (figs. 5, 7). Dipseudopsis maculata Ulmer, 1907 (fig. 35) Dipseudopsis maculata Ulmer, 1907a: 37-39, fig. 55a-b, Type d*: MALAYSIA: Sabah: Sandakan (BMNH). Fischer 1962: 11. Fischer 1972: 5. WEAVER & MALICKY: Dipseudopsis from Asia Specimens examined. — MALAYSIA: Sabah: 14, Tawau, Quoin Hill, Cocoa Res. Sta., 1.X.1962, Y. Hirashima (BPBM). 16, Tawau, Quoin Hill, 15-20.VIl.1962, Y. Hirashima (BPBM). 1d, Sepilok For. Res., Sandakan Bay, 1- 10m, 27.X.1957, J. L. Gressitt (BPBM). Male. — Head dorsum with anterior 4 glossy dark brown, almost black, and posterior warts light brown, frontoclypeus glossy dark brown; antennal scape and pedicel light brown, flagellum darker, palps dark brown, maxillary palps 3.6 mm, proboscis 2.2 mm, as long as first four segments of maxillary palp. Pronotum, mesonotum and tegula dark brown, but metathorax and abdomen yellowish brown. Forewing 17-18 mm, veins dark brown against brown and translucent background, large translucent triangular spot between R, and Cu,, also translucent spots at ar- culus, discal cell, and area distal to cord, and dark band distal to hyaline spot at m-cu. Fore leg coxa to basal % of femur dark brown, distal % of femur and other distal leg segments yellowish brown; mid leg coxa dark brown, distal leg segments yellowish brown; hind leg yellowish brown. Modified spur (fig. 35) 2x longer than adjacent spur, apical 4 modified, into two unequal short nearly parallel points curved laterally away from tarsus. Genitalia (fig. 35): Tergum IX distal 4 overhanging segment X and with broad apex in lateral view, posterior margin broadly rounded and bilobed in dorsal view. Sternum IX me- sosuperior process triangular with blunt dorsal angle and acute posterior angle in lateral view. Segment X ventral margin straight and horizontal, dorsal margin slightly concave and inclined ventrad, apex rounded; ellipsoidal about 4 as wide as tergum IX in dorsal view. Preanal appendages broad and triangular with corners rounded in lateral view, dorsal and ventropo- sterior margins slightly irregular, apicodorsal apex blunt. Inferior appendages inclined dorsad and wid- ening apically, having apex truncate with rounded corners in lateral view, dorsal margin straight and longer than ventral margin; ventral margin with basal % having protruding shelf extended as short free lobe, oblique angle at midlength, and distal 4 inclined dor- sad; distal process thumblike in ventral view, mesal margin with deep narrow curved basolateral notch, and apex rounded. Distribution. — Malaysia: Sabah. Remarks. — The male of this species has a unique modified spur with a prominent apical point extend- ing laterally from a blunt apical shelf, and a shorter subapical point curved laterally, having both points curved distad. D. maculata and D. nieuwenhuisi ap- pear to be closely related based on similar colouration in the male forewing (cf. fig. 6 of D. nieuwenhuisi), and similar characteristics in the male genitalia. D. 123 [1}DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 maculata differs by being smaller in size, and by hav- ing male modified spur with only two apical points, rather than having three as in D. nieuwenhuisi. Dipseudopsis malaisei sp. n. (fie. 36) Type material. — Holotype d': BURMA: Washaung, 600ft, 13.VII.1934, Malaise (NHRS). Paratypes: 3d, same data, Malaise (NHRS). Male. — Head dorsum and frontoclypeus dark brown, nearly black, posterior warts brown. Antennae, mouthparts, legs and sclerotized parts of thorax unicoloured brown. Maxillary palps 2.2-2.5 mm; proboscis 1.2-1.8 mm, as long as first four seg- ments of maxillary palps. Forewing length 13-15 mm, dark brown except for small hyaline comma at m-cu. Modified spur (fig. 36) slender, slightly curved and longer than adjacent spur, with two small spiral- ling apical points. Abdomen yellowish brown. Genitalia (fig. 36): Tergum IX broadly rounded in dorsal view, with distal 4 overhanging segment X in lateral view; sternum IX with mesosuperior process fingerlike and % as long as segment X in lateral view. Segment X long and fingerlike, tapering slightly dis- tad and apex incised in dorsal view. Preanal appenda- ges oblong, with posterodorsal margin ellipsoidal in lateral view. Inferior appendages curved dorsad and broad in lateral view; broad with lateral and mesal margins parallel, and apices slightly bent mesad in ventral view. Etymology. — Named after the collector R. Malaise. Distribution. — Burma. Remarks. — This species is only known from the type series. The male has segment X long and slender and modified spur with two small apical spiralling points similar as in D. benardi. However, D. malaisei differs by having male with preanal appendages ob- long and ellipsoidal in lateral view, inferior appenda- ges broad and flat, and dark brown forewings without large translucent spots, similar to D. benardi (fig. 10). Dipseudopsis martynovi sp. n. he. 37) Type material Holotype d: PHILIPPINES: Sulu: larawaka NI 12.11.1957 Yoshio Kondo (BrBM). s: Tawitawi, Tarawakan, north of Batu Batu, na Dan Exp. (ZMUC): 108, 21.X.1961; 46, 20.X.1961; 26, 13.XI.1961; 23: 14.X1.1961. Male. — Head dark brown, but posterior warts light 124 brown; frontoclypeus with ventral % light brown, dor- sal % dark brown; antennae light brown, palps light brown, maxillary palps 2.3 mm, proboscis 1.2 mm, as long as first three segments of maxillary palp. Pronotum light brown. Mesonotum and tegula dark brown. Forewing 13 mm, dark brown with C and small light brown stigma and small hyaline spots at m-cu and arculus. Legs with coxae dark brown, distal segments light brown. Modified spur (fig. 37) short and stout, about 4X, as long as basal width apical % modified with three broad points clustered together. Genitalia (fig. 37): Tergum IX posterior % extended over segment X in lateral view, broad and triangular in dorsal view; sternum IX mesosuperior process straight rodlike with apex curved dorsad in lateral view. Segment X dorsal margin slightly concave and inclined ventrad in lateral view, ventral margin curved ventrad at middle and curved dorsad distad, apex blunt; ellipsoidal and about % as wide as tergum IX in dorsal view. Preanal appendages slightly shorter than segment X in lateral view, dorsal margin slightly sinuate at base, with broad dorsoapical curve, ventral margin nearly straight inclined dorsad, posterior mar- gin with shallow incision. Inferior appendages finger- like and inclined dorsad in lateral view, with narrow basoventral shelf, squarish with distal arms parallel and straight in ventral view, lateral apex rounded, ba- somesal margin with square-angled basal shelf, and lateral notch and apicomesal angle both square. Etymology. — Named after Andrei Vasilievich Martynov. Distribution. — Philippines: Sulu, Tawitawi. Remarks. — This species is superficially similar to D. recta and D. triclavata, having modified spur tri- fid. However, it differs by having the spur short and stout, about 4X longer than basal width, with apical 4 of spur modified, and inferior appendage fingerlike in lateral view. Dipseudopsis modesta Banks, 1911 (fig. 38) Dipseudopsis modesta Banks, 1911: 105-106, pl. 6: fig. 2, Type d*: INDIA: Bihar: Pusa, 21.IX.08, T. N. J. (MCZC 11756). — Fischer 1962: 11; Fischer 1972: 5; Higler 1992: 80. Dipseudopsis pallida Martynov, 1935: 166-167, fig. 70a-b, Type series: 1d 12 in copula, Pusa, 31.111.1911-4.IV, F. H. Gravely (probably NZS1). — Fischer 1962: 14; Higler 1992: 80. Syn. n. Specimens examined. — INDIA: Assam: 36 39, Kohara, Kaziranga, 110m, 26.36N 93.28E, 16.X.61, Ross & Cavagnaro (CASC). 1d 39, 13mi SW Kochugaon, 50m, 17.X.61, Ross & Cavagnaro (CASC). 2d 29, Meleng For., Mariani, 110m, 26.39N 94.18E, 10.X.61, Ross & Cavagnaro (CASC). 2d 39, Garampani Res. 10mi S Gologhat, 110m, 9.X.61, Ross & Cavagnaro (CASC). Assam: Id 22, ‘Doom Dooma Dum Duma, 27.34N 95.34E, VI.1943, E. Hardy (USNM). ‘U.D.M.N.C.H. [United District of Mikir and North Cachae Hills]: 16, Kalanga, 1.V.1960, F. Schmid (cncı). Bihar: 48 29, Pusa (MCZC). 2d, Pusa (BMNH). Orissa: 24, Bhubaneswar, 21.11.1985, J. Olah (CLHM). West Bengal: 34 , Chapra, Mackenzie (MCZC). Male. — Head glossy dark brown; frontoclypeus, an- tennae and palps dark brown; maxillary palps 2.6 mm, proboscis 1.0 mm, almost as long as first three seg- ments of maxillary palp. Pronotum, mesonotum and tegula dark brown. Forewing 12-15 mm, dark brown, except for hyaline lunula at m-cu and small translu- cent spot at arculus. Legs light brown slightly darker basad. Modified spur (fig. 38) slightly shorter than ad- jacent spur with two minute apical points (cf. ‘Remarks’ below). Genitalia (fig. 38): Tergum IX tri- angular with acute apex slightly inclined dorsad and distal 4 overhanging segment X in lateral view; poste- rior margin broadly rounded in dorsal view; sternum IX mesosuperior process with short broad dorsal trian- gle, posterior process narrow having dorsal margin ir- regular and slightly concave, and ventral margin curved dorsad in lateral view, extending above basal %of phallus. Segment X long slender and apex bilobed with deep mesal notch in dorsal view; fingerlike and slightly tapered in middle, extending posteriad distal to inferior appendages in lateral view. Preanal appen- dages trapezoidal, posterior margin slightly incised in lateral view. Inferior appendages slender and long in lateral view, extending dorsad toward apex of preanal appendage, dorsal margin curved and concave, ventral margin irregular; lateral margins curved and apicome- sal margin straight with apical point in ventral view, mesal margin with curved notch in basolateral corner. Distribution. — India: Assam, Bihar, Orissa, West Bengal. Remarks. — This species can be distinguished by its unique male modified spur, having a short inconspic- uous apical point directed distally and a long slender point curving halfway around the apex; the straight point is minute and lightly sclerotized, and without clear magnification it appears to be a small lobe. This species is similar to D. benardi, but it differs by hav- ing male modified spur with smaller apical points and inferior appendages narrower in lateral view. We have observed some variation in the shape of the preanal appendages. Comparison of the type of D. modesta with the original description of D. pallida, reveals that these species are conspecific, and thus the latter is rec- ognized as a junior synonym. WEAVER & MALICKY: Dipseudopsis from Asia Dipseudopsis morosa Banks, 1924 (fig. 39) Dipseudopsis morosa Banks, 1924: 450, pl. 3: fig. 31. — Fischer 1962: 11; Fischer 1972: 5. nec Dipseudopsis morosa. — Schmid 1958: 13, 24, 27, 89, pl. 15: figs. 12-13. Lectotype designation. — d*: PHILIPPINES: Mindanao: Dapitan, Baker (MCZC 14826). 1d”: Basilan: probably syntype but without type label, Baker (USNM). The type series: 3d (MCZC) is mixed and comprises three different species; in addition to the lectotype, the d * syntype from Luzon is D. colla- ris, and the d * syntype from ‘Borneo’ Kalimantan is D. stabatensis. Specimens examined. — PHILIPPINES: Mindanao: 16, Lanao, 4.8 km E of Dansalan, 750m, 11.V1.1958, jungle along stream, H. E. Milliron (BPBM). Male. — Head brown with black triangular median stripe, broad along anterior margin and tapering to middle of posterior margin, frontoclypeus glossy black and genae brown, antennae and palps brown, maxillary palps 2.3 mm, proboscis 1.0 mm, almost as long as first three segments of maxillary palp. Pronotum black with posterior dorsal margin brown, mesonotum and tegula almost black. Forewing 9-10 mm, uniformly dark brown, but small hyaline spots at m-cu and arculus. Legs with coxae of fore and mid legs dark brown and distal segments yellowish, hind leg yellowish. Modified spur (fig. 39) 14x longer than adjacent spur, apex truncate with two apical points, both points curved and one point nearly in plane per- pendicular to longitudinal axis of spur, and other longer point with apex directed more distad. Genitalia (fig. 39): Tergum IX not extended over base of segment X in lateral view, slightly inclined dorsad; sternum IX mesosuperior process short and blunt in lateral view, with square dorsoanterior angle, posteri- or lobe inclined dorsad and extended above basal % of phallus. Segment X dorsal and ventral margins nearly parallel and inclined ventrad in lateral view, but api- cal portion of ventral margin horizontal, apex round- ed; triangular and with rounded posterior apex in dorsal view. Preanal appendages broad and somewhat D-shaped with apex extended dorsoposteriad in late- ral view. Inferior appendages oblong with distinct subapical ventral tooth in lateral view, distal portion capitate in ventral view with apex and basomesal notch curved. Distribution. — Philippines: Basilan, Mindanao. Remarks. — D. morosa has been confused with D. triclavata, due to similarities of the male modified spurs. However, these species are not closely related and are distinguished from each other by characteris- 125 VOOR ENTOMOLOGIE, VOLUMI nitalia: A, lateral; B, preanal appendages, segments IX and X, dorsal; C, inferior phallus, ventral; F, segments IX, X, dorsal; G, ventral. I-J, male modified spurs ap- D. modesta Banks. Fig. 39, D. morosa Banks. Fig. 40, D. nebulosa Albarda. Fig. 41, tics of the male genitalia, D. morosa having the inferi- or appendage broader and with a distinct blunt ven- tral tooth in lateral view, and by having modified spur with only two apical points. The type series of D. mo- rosa at the MCZC is mixed, comprising three syntypes, each a different species. Records of D. morosa from Sri Lanka by Schmid (1958) are actually D. triclava- La. Dipseudopsis nebulosa Albarda, 1881 (figs. 7, 12A, 40) Dipseudopsis nebulosa Albarda, 1881: 19, pl. 5: fig. 4, Type ®, INDONESIA: ‘in insula Sumatra’. — Fischer 1962: 11- 12; Fischer 1972: 6; Malicky & Weaver 1988: 4, fig. 2a- fi nec Dipseudopsis nebulosa. — Banks 1914: 178 (Luzon). Specimens examined. — BURMA: 2d, Tenasserim, Sukli, 75km E of Moulmein, 600m, 27-31.X.1934, Malaise (BMNH). 1d, Mekane, 20 km W Myawaddy, Amherst Dist, 700ft, 3.XI.1934, Malaise (NHRS). INDONESIA: Sumatra: 1d, W Morton (CNCI). 1d, Tandjong Morawa Serdang (RMNH). 1d, Loeboek Sikaping, 450m (RMNH). 1d, Pematang Siantar, 1.VI-12.XII.1985, Diehl (CLHM). 126, Huta Padang, Diehl (CLHM). MALAYSIA: West Malaysia: Pahang: 36, Kuala Tahan, at light, Pendlebury (BMNH). 14, Sungei Tembeling, 18.X1.1922, Pendlebury (MCZC). 1d, Kuala Teku, 500ft, 5.XII.1921, Pendlebury (MCZC). Selangor: 123 62, Kuala Lumpur, 15-31.X11.1958, Quate (BPBM). THAILAND: 36, upper Pran River, 13-15.1V.1926 (MCZC 18, BMNH 2d). 2d, det. Mosely as D. doehleri, same data, 15.IV.1927, Landell (BMNH). 1d, Trang, 26.VI.1924, Evans (MCZC). 28 19, Trang Prov., Khaophappha Khaochang, 200-400m, 12-13.1.1964, Samuelson (BPBM). 5d, Krachong For. near Trang, 100m, 2.VII.62, Ross & Cavagnaro (CASC). 18, 20 km E of Krabi, 10-20.II.1962, Friedel (CLHM). 1d, Trang, Banchang, 16.V.1924, at light, Evans (BMNH). Male. — Body and wings (fig. 7). Head dorsum, an- tennae, cervical sclerites and pronotum orangish brown, and contrasting with dark brown mesono- tum, tegula and forewings (fig. 12A). Frontoclypeus and mouthparts brown. Fore and mid legs with coxae and trochanters brown, femora with basal portions dark brown and apical portions orangish brown, tib- iae and tarsi orangish; hind leg dark brown. Forewing 12-14 mm, dark brown with translucent spots: long triangular spot below R, above Cu, from base of Rs to hyaline lunula at m-cu, translucent oblong spot distal to cord with apical margin deeply incised, and small translucent spot at arculus. Modified spur (fig. 40) with twisted glabrous apical point, about % as long as spur. Genitalia (fig. 40) similar to D. schmidi. Tergum IX with posterior margin triangular with apex curved and lateral sides slightly depressed in dor- sal view, apical 4 of dorsum overhanging segment X in lateral view. Sternum IX extended anteriad, with mesosuperior process globular in lateral view. WEAVER & MALICKY: Dipseudopsis from Asia Segment X short stout and ellipsoidal in lateral view, dorsal and ventral margins nearly parallel and hori- zontal, apex bluntly rounded, broad and cordate in dorsal view, length and width subequal, with apex slightly incised. Preanal appendages abbreviated in length, with posterior margin incised. Inferior appen- dages broad, somewhat ellipsoidal in lateral view, but dorsal margin slightly concave. Distribution. — Burma; Indonesia: Java, Sumatra; Malaysia: West Malaysia; Thailand. Remarks. — This species is similar to D. doehleri, but differs by having a translucent spot of forewing with faded edges, rather than a well defined oval- shaped spot at the arculus, and genitalia with sternum IX, segment X, and preanal appendages slightly broader in lateral view. The modified spur of D. neb- ulosa is similar to that of D. immaculata and D. doeh- leri, with a twisted glabrous apical point, having a se- tose pocket at its base, but it differs by having a longer glabrous point, about % as long as spur. We have ac- cepted the description of the male provided by Ulmer (1951), who associated it with the female type from middle Sumatra. We have observed that a male spec- imen from Luzon was misidentified by Banks (1914) and actually is D. collaris. Dipseudopsis nervosa Brauer, 1868 (figs. 3, 4, 41) Dipseudopsis nervosa Brauer, 1868: 267-268, Type © : PHILI- PPINES: (Semper) [cf. ‘Remarks’ below]. — Fischer 1962: 12; Fischer 1972: 6. Dipseudopsis luctuosa Banks, 1913: 178, pl. 8: fig. 3, Type G*: Luzon: Los Banos, P. L. Baker (MCZC 11769). — Ulmer 1930: 380, as syn. — Fischer 1962: 12; Fischer 1972: 6. Dipseudopsis venosa Navás, 1931: 86, Type © ‘Filipinas: Mindanao, Kolambogan, 11.VI.1914. Col. m.’ [We were unable to locate the female type of D. venosa for examina- tion, but are relatively certain that this species is synony- mous with D. nervosa, because only two species of Dipseudopsis are known from Mindanao, D. modesta and D. nervosa, and the size of the female type of D. venosa only agrees with the female of D. nervosa.) — Fischer 1962: 16. Syn. n. Specimens examined. — PHILIPPINES: Leyte: 36, Palo (- BPBM). 1d, Tacloban (USNM). Luzon: 36 39, Los Banos (- MCZC). 28 59, Los Baños (USNM). 46 39, Mt Maquiling (CNCI, USNM). 16, det. Ulmer (1905), bez. Bilneao, prov. Laguna, 1876, Laglaize (MNHN). Mindanao: 16, Kolambugan (BMNH). 16 (‘Paratypus’ see discussion above) Mindanao, 1867 (NHMW). 2d, Agusan Los Arcos (BPBM). 19, Agusan Esperanza (BPBM). 36, Agusan Jabonga Mayogda (MCZC). 26, Bukidnon, Dalongdong, 40km NW Maramag, Talakag, Urwaldrand, 800m, 7°53’N 124°40°E (- ZMHB). 16 29, Butuan (MCZC, USNM). 1d, Cotabato, Kalaong (BPBM). 2d, Sapamoro, Curnan dist. (ZMUC). 127 [iDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Mindoro: 1d, Laguna de Nanjan Soldanski (CLHM). Negros: 2d 39, L. Balinsasayao (BPBM). 1d, Sibulan BPRM). 28 19, Victorias (MCZC, USNM). Panay: 1d, Pr. Capiz, Libacao (ZSMC). Sibuvan: 1d, Magdiwang- Tampayan, Ga-ong, Pawala River (CLHM). Male. — Body and wings (fig. 3). Head dorsum similar to D. morosa, having dark brown median acute triangle, tapering posteriad, and lateral sides vellowish brown; frontoclypeus with dark brown me- dian stripe tapering dorsad and lateral sides yellowish brown; antennae and palps brown, maxillary palps 2.8 mm, proboscis 1.7 mm, as long as first 3% seg- ments of maxillary palp. Cervical sclerites mostly dark brown, but dorsal part yellowish brown. Pronotum vellowish brown, median furrow and lateral sides dark brown; propleuron yellowish brown. Mesonotum, metanotum and tegula dark brown. Mesothorax, metathorax and abdomen brown dorsad and yellowish brown ventrad. Forewing 14-17 mm, dark brown with three translucent spots: 1) basal stripe berween R, and Cu, 2) triangular spot distal to cord, and 3) round spot at arculus. Fore leg coxa bi- coloured, basal 4 dark brown and distal % yellowish brown. Pleuron bicoloured, cervical sclerites and side of thorax with lower portion dark brown and upper part yellowish brown. Modified spur (fig. 41) trifid, with long apical point spiraling distad, minute apical point, and subapical lobe. Genitalia (fig. 41): Tergum IX inclined dorsad in lateral view, minute portion of apex overhanging segment X, posterior margin broadly curved in dorsal view; sternum IX mesosupe- rior process oblong in lateral view, dorsal margin curved. Segment X dorsal margin slightly concave, strongly inclined ventrad to blunt apex in lateral view; base broad and narrowing toward oval-shaped apex in dorsal view. Preanal appendages irregular triangle, with dorsal margin irregular and curved in lateral view, ventroposterior margin steeply inclined dorsad, dorsoposterior apex squarish. Inferior appendages fingerlike and curved dorsad in lateral view, lateral margin curved in ventral view, extended arms finger- like with middle of mesal margin having small semi- circular incision. Female. — Body and wings (fig. 4). Distribution. Philippines: Leyte, Luzon, Mindanao, Mindoro, Negros, Panay, Sibuvan. Remarks. — This species is similar to D. digitata, but differs by having unique colouration of the male head and forewings, and by characteristics ofthe male hind spur and genitalia. | Brauer (1868) described this species after a female from the Philippines, and the species epithet, nervosa, was probably derived in references to the colouration pattern of the female forewing, having dark veins against a translucent background. The type which should be at the NHMW is missing. Instead a male specimen with a hand-written type label reading ‘Paratypus’ is present, but this specimen was not mentioned by Brauer and thus cannot be regarded as a syntype. The red ‘paratype! label was probably add- ed to this specimen after the species was described. This specimen cannot be the type because the type description was clearly based on a female specimen; also the forewing length of the type should be 19 mm, but that of the male ‘paratype’ is only 17 mm. Therefore, we conclude that the type is missing. We accept the male association of this species by Ulmer (1929). Furthermore, we are reasonably sure that the identity of this species is correct because there are no other species as large as D. nervosa known from the Philippines, i.e. the other four Philippine species, D. collaris, D. digitata, D. martynovi, and D. morosa, are much smaller. Dipseudopsis nervosella Ulmer, 1905 stat. n. (fig. 42) Dipseudopsis sp — Ulmer, 1905: 96, fig. 75e. Dipseudopsis morosa. — Ulmer 1951: 128 [misdet.]. Dipseudopsis morosa nervosella Ulmer, 1951: 141-143, pl. 7: figs. 140-142, Type d* ‘Borneo’ 1886, F. Baczes (NHMW). — Fischer 1972: 5. Specimens examined. — MALAYSIA: Sarawak: 1d, at foot of Mt Dulit, junction of rivers Tinjar & Lejok, 19.VI- 11.1932 (BMNH). 19 , same data, 30.VIII.1932 (BMNH). 16, R. Kapah trib. of R. Tinjar, 25.X.1932, Oxford Univ. Exp. (BMNH). Male. — Dorsum of head with curved M-shaped pattern marked by division between glossy dark brown, almost black, anterior triangle pointed poste- riad and dark brown lateral margins, and yellowish brown area, filling in two triangles of ‘M’ and with acute angle of each triangle pointed toward base of antenna. Frontoclypeus glossy, almost black, ante- nnae yellowish brown, cervical sclerites dark brown. Maxillary palps 3.0 mm, dark brown; proboscis brown, 1.8 mm, almost a long as first four segments of maxillary palp; labial palps 1.2 mm. Pronotum dark brown with slender yellowish brown posterior edges along median fissure and lateral margins, pro- pleuron light brown; mesonotum almost black, mes- opleuron dark brown except epimeron light brown; metathorax and abdomen yellowish brown. Forewing 15-16 mm, dark brown with golden hairs, and three large translucent spots, similar to D. nervosa and D. nebulosa (cf. figs. 3, 7): 1) long band between R, and Cu, 2) equilateral triangle at arculus, and 3) subapical area distal to cord with six translucent gaps between WEAVER & MALICKY: Dipseudopsis from Asia Figures 42-45. — Dipseudopsis, A-G, male genitalia: A, lateral; B, preanal appendages, segments IX and X, dorsal; C, inferior appendages, ventral; D, phallus, lateral; E, phallus, ventral; F, segments IX, X, dorsal; G, ventral. I-J, male modified spurs ap- ical spurs of hind tibia, ventral. Fig. 42, D. nervosella Ulmer. Fig. 43, D. nieuwenhuisi Ulmer. Fig. 44, D. notata (Fabricius). Fig. 45, D. recta Martynov. 129 LIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 dark apical veins, and in discal ceil. Fore leg and mid- dle leg with coxae, trochanter and basal portion of fe- mur dark brown, distal portion of leg yellowish; hind 14x longer than adjacent spur, apex truncate with two apical points, both points curved in plane perpendicular to longitu- dinal axis of spur (cf. ‘Remarks’ below). Genitalia (fig. 42): Tergum IX with posterior 4 overhanging seg- ment X in lateral view, posterior margin bilobed with two blunt lateral angles separated by a obtuse mesal notch in dorsal view; sternum IX dorsomesal process slender horizontal, with pointed apex curved dorsad in lateral view. Segment X dorsal margin slightly con- cave in lateral view, distal portion narrow; ellipsoidal and with apex slightly incised in dorsal view. Preanal appendages large lobate and extended dorsoposteriad in lateral view. Inferior appendages thumblike and in- clined dorsad in lateral view, ventrobasal shelf with short free apical projection; distal arms ellipsoidal in ventral view, without sharp mesal tooth, mesal mar- gin with deep narrow curved basolateral notch. leg yellowish. Modified spur (fig. 42) Distribution. — Malaysia: Sarawak. Remarks. — This species was originally described as a subspecies of D. morosa. Comparison of the male type of D. nervosella with that of D. morosa reveals that each is a distinct species. The modified spur in D. nervosella is similar to that of D. morosa, but more robust and ac- tually a mirror image of the other, having apical points curved in opposite directions. This species also differs by having the male forewing with a more contrasting colour pattern, and inferior appendage having mesal margin with deep curved basolateral notch in ventral view. Also, D. nervosella is only known from Sarawak, and D. morosa is only known from the islands of Mindanao and Basilan of the Philippines. Dipseudopsis nieuwenhuisi Ulmer, 1909 figs. 6, 43) Dipseudopsis nieuu Sn Ulmer, 1909: 128-129, fig.4, [ype d*: Borneo Ex Dr. Nieuwenhuis, L. Dengay, X.1894 (RMNH). — F We 1962: 12; Fischer 1972: 6. xamined. — MALAYSIA: Sarawak, Mt Dulit, forest, at light, Oxford Univ. Exp., B. M. ¢ (BMNH): 1d, 22.IX.1932; 5d, 19- 17.1X.1932. LAS D 4 DOG and wings (fig. 6). Head dorsum langle along anterior margin ex- of posterior margin, posterior gins yellowish, frontoclypeus ellowish brown, palps dark 0 mm, proboscis 2.8 mm, otum dark brown except posterior lateral corners light brown, propleuron light brown, mesonotum and tegula dark brown, meta- thorax light brown. Forewing 19-23 mm, dark brown, veins dark brown with several translucent stripes between veins, similar to that D. digitata, hav- ing three translucent areas, basal stripe in thyridial cell, round marginal spot at arculus and subapical translucent stripes distal to cord. Front and middle legs with coxae dark brown, trochanters to apex yello- wish brown; hind leg mostly yellowish. Modified spur (fig. 43) 2x as long as adjacent spur and 2x as thick, with three apical points, one minute point, and two large points parallel and curved, one large point subapical. Genitalia (fig. 43): Tergum IX posterior margin broadly bilobed, having shallow mesal notch in dorsal view. Segment X originating from relatively shallow cavity of tergum IX in lateral view, apex blunt, ventral margin slightly concave; ellipsoidal and posterior margin broadly bilobed in dorsal view. Preanal appendages triangular and rounded with dor- sal margin horizontal, broadly curved and slightly emarginate in lateral view, ventroposterior margin broadly curved and strongly inclined dorsad. Inferior appendages oblong with ventrobasal shelf in lateral view; angulate with distal arms thumblike in ventral view, mesal margin with deep narrow curved basola- teral notch. Distribution. — Malaysia: Sarawak. Remarks. — This is the largest Asian species of Dipseudopsis known. It appears to be closely related to D. digitata and D. nervosella, but differs by its larger size and by characteristics of the male modified spur and genitalia. Dipseudopsis notata (Fabricius, 1781) (fig. 44) Phryganea notata Fabricius, 1781: 390, Type d*: America’ [mislabelled, cf. ‘Remarks’ below], (BMNH). — Fischer 1962: 13. Dipseudopsis horni Ulmer, 1915: 46-47, fig. 13. Type series: SRI LANKA: ‘2d, [N. Cen. Prov: Mahl Iluppallama] Maha Illupalama, Ceylon, Juli 1912, J. C. F. Fryer leg., im Museum Cambridge [MCZC]; ferner 9 Exemplare (d, 2), (Cen. Prov.]: Kandy, resp. [E. Prov: Kantalai?] Kanthaley, resp. (Cen. Prov.: Matale] Matala, Ceylon, Dr. W. Horn leg, im Deutsch. Entom. Museum Berlin- Dahlem’ [ZMUH]. — Weidner (1964: 73): ‘Ceylon, Matala, Paratypoid’ of D. horni, (ZMUH). — Fischer 1962: 9; Weidner 1964: 73; Fischer 1972: 4. Syn. n. Specimens examined. — INDIA: Karnataka: 88, Bhadra- vati (MCZC). 2d , Shimoga, R. Tunga 1865ft (MCZC). Tamil Nadu: 3d, Tanjore dist. ‘Thanjavur, Nedungadu (MCZC). SRI LANKA: N. Prov: 2d, Marichchukkaddi (BMNH). N. Cen. Prov: 66, Moradan Maduwa, Wilpattu Natl. Park, 23mi W of Anuradhapura, 80m (MZLU). 1d, Kontalai, 60m (MZLU). Anuradhapura Dist: 7d 89, Wildlife Soc. Bungalow, Hunuwilagama, Wilpattu Natl. Park, 200ft (- USNM). 104 109, Irrigation Bungalow, [Padawiya Tank] ‘Padaviya’, 180ft (USNM). 88 29, Padaviya, 180ft (USNM). Polonnaruwa Dist: 2d 7%, Pinburettawa, 13mi S Mannampitiya (USNM). 1d, Polonnaruwa (SOFM). N. W. Prov: 14 2®, Marai Villu (BMNH). 3d, Smi NNE of Puttalam, 5m (MZLU). S. Prov: 14, Hikkaduwa, 11mi NW of Galle, 10m (MZLU). 28 29, Badagiriya Tank, 6.5mi N Hambantota, 75ft (CLJW). 14 19, Hambantota (BMNH). 16, Tangalla (MCZC). 1d 19, Weligama (BMNH). W. Prov.: 106 119, Colombo (BMNH). 5d, 18mi NE of Colombo, Dambuwa Estate, 30m (MZLU). Male. — Head dorsum mostly brown, posterior margin lighter, frontoclypeus, antennae, and palps brown. Maxillary palps 2.3 mm; proboscis 0.9 mm, as long as first 2% segments of maxillary palp. Pronotum brown, slightly lighter in colour than head and mesonotum. Mesonotum and tegula brown. Forewing 12-14 mm, mostly brown, similar to that of D. robustior, with arc of small translucent spots distad of cord, hyaline spot at m-cu and big spot at arculus. Legs with coxae brown and distal segments lighter. Modified spur (fig. 44) slightly longer than adjacent spur, bifid with two unequal sinuate points directed distally, longer point almost % as long as spur. Genitalia (fig. 44): Tergum IX apical % overhanging segment X in lateral view, posterior margin with broad obtuse angle in dorsal view; sternum IX meso- superior process triangular and extended above basal % of phallus in lateral view. Segment X dorsal margin slightly concave with blunt apex in lateral view, oval- shaped almost % as wide as IX in dorsal view. Preanal appendages trapezoidal with dorsal margin horizontal in lateral view, ventral margin inclined dorsad and posterior margin slightly incised. Inferior appendages angled dorsad in lateral view, dorsal margin concave and nearly parallel with ventral margin, middle of ventral margin with obtuse angle, apex rounded; dis- tal arms diverging in ventral view, lateral margins with base and apex rounded, mesal margin with squa- rish basomesal angle, and short apicomesal point. Distribution. — India: Karnataka, Tamil Nadu; Sri Lanka. Remarks. — This species has male modified spur bi- fid and somewhat similar to that of D. robustior, but differs by having spur with the shorter point nearly straight and directed distad. Until now the identity of D. notata has been enigmatic, because the locality stated by Fabricius (1781) as ‘America boreali was very doubtful. McLachlan (1864) placed this species in Dipseudopsis, and suggested that the original loca- lity information by Fabricius was in error, since at the time the genus was known only from South Africa WEAVER & MALICKY: Dipseudopsis from Asia and China. Our examination of the male type of D. notata reveals that it is identical to Ulmer’s (1915) original description of D. horni, only known from southern India and Sri Lanka, and we therefore con- cur with McLachlan that the type of D. notata is mis- labelled. Dipseudopsis recta Martynov, 1935 (figs. 15, 45) Dipseudopsis recta Martynov, 1935: 163-164, fig. 65a-b, Type, INDIA: ‘d. Sanjai River, Chakradharpur, Chota Nagpur. 8-10.ii.1918. N. Annandale and F. H. Gravely’ (probably Nzst). — Fischer 1962: 14; Higler 192: 80. Dipseudopsis bombayana Martynov, 1935: 164-165, figs. 66a-b, 67, Type, INDIA: ‘d. Gokak, Belgaum district, Bombay Presidency (at light). June 1928 B. Prashad and H. S. Rao.’ [Type (NZSI) in poor condition; not able to borrow for examination, pers. comm. Dr. K. Rail. — Fischer 1962: 5; Higler 1992: 80. Syn. n. Specimens examined. — INDIA: Andhra Pradesh: 16, Lamnasinghi Ghats, 75km SW Vishak, 850m, 10.V11.1982, D. R. Davis (USNM). Bihar: 1d 59, Namkum Ranchi, 675m, 8.XI.61, Ross & Cavagnaro (CASC). Karnataka: 12d 1®, Haveri, 2.XII.1961, F. Schmid (CNCI). 16, Jannapura, 22.1.1959, F. Schmid (CNCI). 2d, Kakankote, 10.1.1959, F. Schmid (CNCI). 46 89, Yellapur, 30.1.1959, F. Schmid (CNcI). 28, Shimoga, R. Tunga, 1865ft, III-12.VIII, P.S. Nathan (MCZC). Madhya Pradesh: Satanwara, 376 179, 28.XI.1961; 1d 39, 29.X1.1961, F. Schmid (CNCI). 166, Phalghat, 6.1.1959, F. Schmid (CNCI). Maharashtra: 3d 19, (Bombay) Sykes, 5.1.1959, F. Schmid (CNCI). Tamil Nadu: 36, Varaiyattu Tittu, 21.XII.1958, F. Schmid (CNCI). 26, Maraiyur, 19.XII.1958, F. Schmid (CNCI). 16, Sathuparai, 1.XII.1958, F. Schmid (CNCI). NEPAL: 16, Chitwan Dist., Sauraha S.J. Camp, 31.X.1985, W. N. Mathis (USNM). Male. — Head brown, occipital margin light brown, frontoclypeus light brown except dorsal margin with dark band, antennae with dorsal portions brown, and ventral areas light brown, palps light brown, maxil- lary palps 3.7 mm, proboscis 1.5 mm, almost as long as first three segments of maxillary palp. Pronotum brown, hind margin light brown, lateral sides light brown; mesonotum brown, with tegula and forew- ings light brown. Forewing 10-12 mm long, faded ar- ea distal to cord, small hyaline spots at m-cu and ar- culus. Legs light brown. Modified spur (fig. 45) about 1.5x long as adjacent spur, apical % with crown of three minute points, having two larger exterior points spiraling around minute middle point, middle point slightly curved, directed distad, inconspicuous and more lightly sclerotized than other points. Genitalia (fig. 45): IX tergum with apical % overhanging X in lateral view, posterior margin slightly incised in dorsal view; IX sternum mesosuperior process with blunt dorsal extension and slender posterior lobe overhan- ging basal % of phallus in lateral view. Segment X 131 [HDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 slighdy longer than preanal appendage, with lateral margins paralle | and apex deeply bilobed in dorsal view. Preanal appendages dorsal margin straight in lateral view, ventral margin curved dorsad, posterior margin truncate, apicodorsal angle square, ventropo- sterior angle obtuse. Inferior appendages similar as in D. triclavata, basal % twice as thick as apical portion in lateral view, with dorsal margin almost straight, ven- tral margin having distal % incised, distal 4 fingerlike but slightly acuminate; basal shelves separated by V- shaped notch in ventral view, posterior arms with blunt apicomesal point, lateral margin straight with curved apical and basal angles. Female. — Tergite of abdominal segment VIII (fig. 15) with ventral submarginal recessed groove. Genitalia (fig. 15). Distribution. — India: Andhra Pradesh, Bihar, Karnataka, Madhya Pradesh, Maharashtra, Tamil Nadu; Nepal. Remarks. — D. recta is a sibling species to D. tri- clavata, but differs by having male modified spur with smaller apical points. The specific differences shown in the original species descriptions of D. recta and D. bombayana can be observed by rotating the modified spur. Hence, D. bombayana is considered to be a jun- ior synonym of D. recta. Dipseudopsis robustior robustior Ulmer, 1929 stat. n. (fig. 46) Dipseudopsis robustior Ulmer, 1929: 185-186, fig. 22, Type series: M LA A: West Malaysia: ‘Material in Coll. McLachlan: Perak (etwas verletzt).’ Coty pe d *: ab- domen missing, Perak (BMNH). — Fischer 1962: 14. Dipseudopsis akhila Schmid & Denning, 1979: 243-244, fig. 2, Holotype d*: THAILAND: Chiang Mai Prov., Mae 1000ft, 10.XI.1964, Peters (CLDD); Tha Phra, Khon Kaen. Syn. n. thailandica Schmid & Denning, 1979: 246, fig. : Tha Phra, Khon Kaen, 24.XII.1968, Nong Pora Kit bamroog, Roi-et, and Ping, Chiang Mai, Paratypes”, Dipseudo pst +, Holoty pe Kr nap Pi Paratypes”, Kalasin. Syn. n. Dip do is junki Marlier, 1979: 1-10, figs. 1-7, Type se- ries: 25 79, 4 larvae, 1 pupa, Bung Borapet, lake in cen- tral Thailand, 5.IX.1971, Dr. W. Junk. Syn. n. 1d, Rangoon, McL. coll 4: 63 42, Phnom Penh (USNM, ZMHB): West Malaysia: Kedah: 13, Kuala Ketil (BMNH). Bumbong Lima, light trap (USNM). THAILAND: 26 19, Bangkok NM, MCZC). Chiang Mai: 25¢ 162 (USNM); ] M ; g Mai, Ban-tin-doi, 310m (- Klong Rang Sit, at ‘light (BMNH, NHMW). 1d, 1). 2d 43, det. Banks nin. Siam’, Patalung, 1.V.1924, I. H. N. Evar MCZC, BMNH). 16, Patalung, Paknam 12, Chian Jorapet (USNM) Lampam (BMNH). 8d, Tha Phra, Khon Kaen (CLDD). 76, Chiang Mai, River Ping (CLHM, ZMUC). 10d, Chiang Mai Zoo, light trap (CLHM). 27d, Tap-Tan (CLHM). Chantaburi, Khao Soi Dao, 400m (vopj). 16, Kanchanaburi Dist., Sai Yok Natl. Park, 400m (CLHM). 1d, Nam Nao Natl. Park (CLHM). 34 , Puk Hieo, Chayapoom (- CLHM). 5d, Ayutthaya (CLHM). 26 , Bangsaen (CLHM). 1d, Tham Than Lod Natl. Park (CLHM). 1d, Phuket, Kao Prataew Natl. Park (CLHM). VIETNAM: 1d, Long Xuyen (- MNHN). 2d, Hue (MNHN). Male. — Head mostly brown, posterior warts light brown, frontoclypeus with ventral % light brown and dorsal part brown, antennae mostly yellowish brown but dorsal surface dark brown, palps yellowish brown, maxillary palps 2.1 mm, proboscis 1.1 mm, as long as first three segments of maxillary palp. Pronotum and tegula light brown, mesonotum dark brown. Forewing 11-12 mm, similar as in D. collaris, with 5- 7 small translucent spots in circular pattern around discal cell, having most spots distal to cord. Legs with coxae dark brown, distal segments yellowish brown. Modified spur (fig. 46) bifid, with two apical points, shorter point stout and angular, longer point slender, semicircular, and distant from other point. Genitalia (fig. 46): Tergum IX with short apical lobe extending over X in lateral view, posterior margin triangular in dorsal view; sternum IX mesosuperior process with short rounded dorsal lobe and long slender fingerlike posterior lobe in lateral view. Segment X dorsum in- clined steeply ventrad toward blunt acuminate apex in lateral view, apex directed posteriad, having middle of ventral margin curved ventrad, and apicoventral margin slanted dorsad, ellipsoidal in dorsal view. Preanal appendages trapezoidal in lateral view, dorsal margin straight, ventral margin slanted dorsoposteri- ad, apicodorsal angle obtuse, apicoventral apex curved and extended more posteriad than apicodorsal angle. Inferior appendages with dorsal and ventral margins parallel and apical % bent dorsad in lateral view, basal % with thin basal shelf, and apical % trape- zoidal with apex slighted rounded, distal arms almost fingerlike in ventral view, lateral margins with apical and basal ends curved, basoventral shelf short, and mesal margin with squarish basal angle. Distribution. — Burma; Cambodia; Malaysia: West Malaysia; Thailand; Vietnam. Remarks. — D. robustior is the most common spe- cies of Dipseudopsis in Thailand, and is somewhat dif- ficult to distinguish from D. collaris and D. infuscata. Possibly these species are merely variants of a widely distributed species. WEAVER & MALICKY: Dipseudopsis from Asia Figures 46-50. — Dipseudopsis, A-G, male genitalia: A, lateral; B, preanal appendages, segments IX and X, dorsal; C, inferior ap- pendages, ventral; D, phallus, lateral; E, phallus, ventral; F, segments IX, X, dorsal; G, ventral. H, see specific descriptions below. I], male modified apical spurs of hind tibia, ventral. Fig. 46, D. robustior Ulmer. Fig. 47, D. robustor andamanensis, n. ssp. 47 I, modified spur, isolated spur depicted at left. Fig. 48, D. schmidi, n. sp., 48H. left inferior appendage, ventral, variation with ap- icomesal tooth. Fig. 49, D. spectabilis Banks. Fig. 50, D. stabatensis Malicky & Weaver; 50H, segment IX, dorsal. 133 T1IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Dipseudopsis robustior andamanensis ssp. n. { fie. 47) Dipseudopsis infuscata. — Malicky 1984: 214, 216 [misdet.]. Holotype d: INDIA: South Andaman Island: 20.X11.1976, Starmühlner (CLHM). — Paratypes: 1d 59, same data (CLHM); 26 159, Bimblton, 22.XII.1976, Starmühlner (CLHM). Type data. Mongelutonge, Male. — Head, body, wings and genitalia similar to D. robustior. Modified spur (fig. 47) with two apical subequal in length and size. Remarks. — This subspecies of D. robustior has unique, pincher-like modified spur (fig. 47) with two subequal points, but in D. robustior (fig. 46) the api- cal points of the modified spur are unequal, having a long slender curved point and a short broad point. It differs from D. infuscata (fig. 41) which has the mod- ified spur with a long straight point directed distad and a shorter curved point. In D. collaris the points are shorter, only % as long as the spur. This subspecies represents the only member of the genus known from the Andaman Islands. All previous records of D. in- fuscata from South Andaman Island (Malicky 1984) are now recognized as this subspecies. Etymology. — This subspecies is named after the Andaman Islands. Distribution. — India: South Andaman Island. Dipseudopsis schmidi sp. n. ( fig. 48) Type material. — Holotype d: INDIA: Manipur: Kaiphundai, 20.V.1960, F. Schmid (CNCI). Paratypes: 94, same data, F. Schmid (cncı). 16, Kambiron, 24.V.1960, F. Schmid (CNCI). Assam: 66 29, Garampani Res., 10mi S Golaghat, 110m, 9.X.61, E. S. Ross & D. Cavagnaro (- C). 1d, Kaziranga, 75m, 26.36N 93.28E, 7-9.V.1976, Wittmer (CLHM). ‘U.D.M.N.C.H’ [United District of Mikir and North Cachae Hills]: 16, Langtrang, 30.1V.1960, F. Schmid (CNCI). BANGLADESH: 1d 29, Lawa Chera For., Srimangal, 110m, 27.IX.61, E. S. Ross & D. Cavagnaro (CASC). Male. — Head dorsum, pronotum and cervical sclerites light orangish brown, but head slightly dark- er. Frontoclypeus dark brown except for small oran- gish brown triangular area between bases of antennae; genae yellowish brown except ventral sides dark brown. Scape mostly orangish brown, but apical por- tion and remainder of dark brown. Mouthparts, with palps dark brown, maxillary palps IS antenna mm, proboscis 1.5 mm, as long as first four seg- nts of maxillary palp. Mesonotum and tegula dark ng 11-13 mm, dark brown with indis- tinct translucent spots, large basal triangular translu- cent spot bordered by Sc, Cu, and first branches of Rs and M, another translucent spot distal to cord. Hindwing lighter brown than forewing, with translu- cent basal triangular area between Sc and Cu,. Legs with colouration similar as in D. nebulosa, mostly dark brown, but apical % of front and middle femora lighter brown, as well as tibiae and tarsi. Hind leg mostly dark brown, tarsi slightly lighter. Modified spur (fig. 48) with basal 4 unmodified, distal 4 with setose pocket at base of straight acuminate apex pointed distad. Genitalia (fig. 48): Tergum IX broad and triangular in dorsal view, long with apical % overhanging segment X in lateral view. Segment X broad, extending distad slightly further than inferior appendages in lateral view, almost twice as long as broad and somewhat ellipsoidal, with dorsal and ven- tral margins inclined ventrad but extended posteriad at apex, apex broadly curved; elongate and cordate in dorsal view, with lateral margins tapered distad and apex slightly incised. Preanal appendages broad and rhomboid in lateral view. Inferior appendages broad and somewhat rectangular with curved ventral mar- gin in lateral view, dorsal margin with short broad ba- sal hump, maximum height about % total length. Etymology. — Named after its collector Fernand Schmid. Distribution. — Bangladesh; India: Assam, Mani- pur. Remarks. — This species is easily recognized by hav- ing the male modified spur acuminate and nearly straight. The male is somewhat similar to D. nebulo- sa, but differs by having light areas of forewing with faded borders and less distinct, and sternum IX with mesosuperior process not short and globular, but long and slender, extending above basal % of phallus. Dipseudopsis spectabilis Banks, 1931 (figs. 9, 49) Dipseudopsis spectabilis Banks, 1931a: 69, pl. 5: fig. 1, Type d'*: MALAYSIA: Sabah: Sandakan, Baker (MCZC 16421). — Fischer 1962: 15. Male. — Head dorsum with anterior portion dark brown, posterior warts and lateral margins yellowish brown, frontoclypeus dark brown, genae and anten- nal scape yellowish brown, but flagellum dark brown, palps dark brown; maxillary palps 1.8 mm, labial palps 0.6 mm, proboscis 1.4 mm, longer than first four segments of maxillary palp. Pronotum, mesono- tum and tegula dark brown. Wings (fig. 9): Forewing 11 mm, fork I absent, conspicuous pattern of lighter bands on dark brown background, similar as in D. flavisignata, with five translucent spots: 1) long slen- der stripe along anterior margin, % as long as Sc; 2) small oval-shaped spot at base of Rs; 3) cordate spot in discal and medial cells; 4) truncate V-shaped band running basad from subapex to crossvein r-m, then distal to cord to base of M,, and to submargin through fork IV including M, 5) long irregular semi- circular band running from base of anal margin, ta- pering at m-cu, and widening toward margin at arcu- lus. Modified spur (fig. 49): trifid, with three apical points, first point longest, slightly curved at base and % as long as spur, second point % as long as spur, and third point minute and adjacent to second point and located in curved notch between first and second points. Genitalia (fig. 49): Tergum IX with posterior margin slightly extended over segment X in lateral view, posterior margin broadly rounded in dorsal view; sternum IX mesosuperior process acuminate, dorsal margin irregular, ventral margin nearly straight, extending above basal % of phallus. Segment X dorsal and ventral margins nearly parallel in lateral view; thumblike with lateral margins parallel in dorsal view. Preanal appendages trapezoidal in lateral view, posterior margin slightly incised and lateral side with ridge parallel to anterior margin. Inferior appendages heavily sclerotized, fingerlike with dorsal and ventral margins tapering slightly distad in lateral view, having short ventrobasal shelf, and dorsal margin with some minute bumps; squarish in ventral view, with lateral arms fingerlike with lateral margins converging slightly distad and mesal margin finely serrate. Distribution. — Malaysia: Sabah. Remarks. — This species is only known from the unique type; it is a sibling species to D. flavisignata and differs from it in having the male forewing with a translucent subapical truncated V-shaped band. It al- so differs by characteristics of the male modified spur and genitalia. Dipseudopsis stabatensis Malicky & Weaver, 1988 (fig. 50) Dipseudopsis stabatensis Malicky & Weaver, 1988: 4-5, fig. Sa-e, Holotype d*: INDONESIA: Sumatra: Dolok Merangir, 2-18.V.1971, Diehl (CLHM). Paratype d*: Stabat 23.V1.1974 (CLJW). Dipseudopsis morosa. — Banks 1924: 450 (pro parte) [misdet. syntype 1d*: INDONESIA, Kalimantan, ‘Borneo’, Telang, X.1881 (MCZC)]. Male. — Head dorsum with dark brown median stripe tapering posteriad, lateral sides and posterior warts yellowish brown, frontoclypeus dark brown, ge- nae light brown, antennae light brown, palps light WEAVER & MALICKY: Dipseudopsis from Asia brown, maxillary palps 2.4 mm, proboscis 1.2 mm, as long as first three segments of maxillary palp. Prono- tum with raised areas yellowish brown, but deep me- dian furrow and lateral sides dark brown. Mesono- tum and tegula dark brown. Metanotum light brown. Forewing 13 mm, brown with translucent apical and anal spots, hyaline spots at m-cu and arculus. Fore and mid legs with coxae dark brown and distal seg- ments yellowish brown, hind leg yellowish brown. Modified spur (fig. 50) 1%x longer than adjacent spur, apex bifid with two minute curved apical points, each about % length of spur. Genitalia (fig. 50): Tergum IX posterior margin bilobed, with lobes separated by squarish mesal notch in dorsal view, minute portion of posterior margin overhanging seg- ment X in lateral view; sternum IX mesosuperior pro- cess similar to that of D. flinti, but slightly slenderer in lateral view. Segment X thumblike with lateral sides parallel and sinuate in dorsal view. Preanal ap- pendages ellipsoidal and extended dorsoposteriad in lateral view. Inferior appendages capitate with trape- zoid apex extended dorsad in lateral view; lateral arms fingerlike with lateral margins straight and parallel in ventral view, and mesal margins with broad curved basolateral incisions. Distribution. — Indonesia: Kalimantan, Sumatra. Remarks. — This species is similar to D. flinti, but differs by having inferior appendage slenderer in late- ral view, and the modified spur with apical points shorter. Dipseudopsis tonkinensis Navas, 1921 (fig. 51) Dipseudopsis tonkinensis Naväs, 1921: 79-80, Holotype d *: VIETNAM: “Tonkin, Chogahn, 1919, P. de Joannis’ (- MZBS). — Fischer 1962: 16; Fischer 1972: 8. Specimens examined. — CHINA: Hainan: 16, Ta Hau, 4.V11.35, J. L. Gressitt (MCZC). Male. — Head dorsum brown, posterior margin lighter brown, frontoclypeus brown, antennae brown but ventral side lighter brown, palps brown, maxillary palps 2.2 mm, proboscis 1.0 mm, as long as first three segments of maxillary palp. Pronotum brown, but posterior margin and lateral sides yellowish brown. Mesonotum and tegula brown. Forewing 12.5 mm, faded brown with small hyaline spots at m-cu and ar- culus. Legs with coxae brown and distal segments yel- lowish. Modified spur (fig. 51) bifid, with two robust acuminate points, first point directed obliquely distad and second point curved mesad. Genitalia (fig. 51): Tergum IX posterior margin only slightly overhan- ging segment X in lateral view, posterior margin 135 [IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 “san 54B | | i if 5 54 n, ip: Bohn A-G, male genitalia: A, lateral; B, preanal appendages, segments IX and X, dorsal; C, inferior ph illus, lateral; E, phallus, ventral; F, segments IX, X, dorsal; G, ventral. I-J, male modified spurs ap- tral. Fig. 51, D. tonkinensis Naväs. Fig. 52, D. triclavata Martynov. Fig. 53, D. varians Ulmer. Fig. 136 broad triangular in dorsal view; sternum IX mesos- uperior process in lateral view with blunt dorsal pro- jection and acuminate posterior process extended above basal % of phallus. Segment X dorsal margin slightly concave and slanted ventrad having blunt apex in lateral view, ellipsoidal with apex slightly in- cised in dorsal view. Preanal appendages dorsal mar- gin straight and ventral margin inclined dorsad distad with posterior margin broadly incised in lateral view. Inferior appendages, clavate, being slightly broadened in lateral view, bent dorsad with oblique angle in middle, dorsal and ventral margins parallel, apex truncate; apical arms with apicolateral margins rounded and slightly broadened in ventral view, api- comesal angles squarish, mesal margins with square basolateral notch and square basomesal shelf. Distribution. — China: Hainan; Vietnam. Remarks. — This species is similar to D. collaris, but differs by having male modified spur with apical points shorter, and by having the male genitalia with segment X broader in dorsal view. Dipseudopsis triclavata Martynov, 1935 (fig. 52) Dipseudopsis triclavata Martynov, 1935: 161-162, figs. 62a- b, 63a-b, 64. New Lectotype d*: INDIA: Maharashtra: Khanapur, Belgaum dist, Bombay Presidency, 18- 19.X1.1928, B. P. & H. S. R. (nzs1). Remainder of type series: 19, same data as lectotype. 16 [doubtful det.], Meghalaya: The Peak, Shillong, 8,400ft, 25.34N 91.53, 12.X.1914. S. W. Kemp. — Fischer 1962: 16; Higler 1992: 80. Dipseudopsis morosa. — Schmid 1958: 13, 24, 27, 89, pl. 15: figs. 12-13 [misdet.]. Specimens examined. — INDIA: Karnataka: 34, Uppinangadi (CNCI). 14, Patan (CNCI). 146 49, Shimoga, R. Tunga, 1865ft (MCZC). 26 109, Bhadravati (MCZC). 84 49, Kakankote (CNCI). SRI LANKA: E. Prov.: Amparai Dist.: 1d, Inginiyagala, 250ft (USNM). Cen. Prov.: 1d det. Ulmer as D. stellata, Peradeniya (BMNH). Kandy Dist.: 114 69, Hasalaka, Dam Spillway, 300ft (USNM). Matale Dist.: 24 29, Sigiriya, 800ft (USNM). Nuwara Eliya Dist.: 24, Milk Board Dairy, 4.2mi SW Nuwara Eliya, 6200ft (USNM). N. Prov.: Vavuniya Dist.: 8d 29, Parayanalankulam Irrigation Canal, 25mi NW of Medawachchiya, 100ft (USNM). N. Cen. Prov.: Anuradhapura Dist.: 48 29, Wildlife Soc. Bungalow, Hunuwilagama, Wilpattu Natl. Park, 200ft (- USNM). NW. Prov.: Puttulam Dist.: 18, Wilpattu Natl. Park, 100ft, Tala Wila (USNM). Sabaragamuwa Prov.: 14, Niriella (USNM). Ratnapura Dist: 156, Walawe Ganga, Embilipitiya, 100ft (USNM). S. Prov.: Galle Dist: 18 19, Hiniduma (USNM). Hambantota Dist: 16, Palatupana, 10ft (USNM). Uva Prov.: Badulla Dist.: 18 19, Dunhinda Falls, 1300ft (USNM). Monaragala Dist: 96, Sella Karagama, Menik Ganga, 150ft (USNM). W. Prov.: Colombo Dist.: 16, Malewana (USNM). WEAVER & MALICKY: Dipseudopsis from Asia Male. — Head dorsum brown, except narrow yello- wish brown posterior margin; frontoclypeus mostly yellowish brown with brown stripe along dorsal mar- gin, antennae dorsum brown and ventral sides yello- wish brown; palps light brown, maxillary palps 2.5 mm, proboscis 1.3 mm, as long as first three segments of maxillary palp. Pronotum mostly brown, but late- ral sides yellowish brown, propleuron and cervical sclerites yellowish brown. Mesonotum and tegula brown. Forewing 11-14 mm, dark brown with trans- lucent stigma and hyaline spots at m-cu and arculus. Fore and mid legs with coxae to mid femora dark brown and apical % of femora and distal leg segments light brown; hind leg yellowish brown. Modified spur (fig. 52) trifid, apex of spur appears truncate, but wich three distinct apical points spiraling in plane perpen- dicular to longitudinal axis of spur, having one min- ute point and two longer points, long points each % as long as spur. Genitalia (fig. 52): Tergum IX with minute portion of apex overhanging segment X in lat- eral view, posterior margin truncate with broadly rounded lateral corners in dorsal view. Sternum IX mesosuperior process with rounded triangular dorsal process and smaller posterior lobe extended above ba- sal % of phallus in lateral view. Segment X dorsal mar- gin slightly concave and inclined steeply ventrad in lateral view, apex acuminate with rounded end, ellip- soidal with apex incised in dorsal view. Preanal ap- pendages trapezoidal with dorsoposterior apex broad- ly rounded and dorsal and posterior margins concave in lateral view. Inferior appendages similar to D. rec- ta, base 2x as thick as fingerlike apical % in lateral view, ventral margin of apical portion incised; basal shelves with broad V-shaped medial notch in ventral view, apical arms with middle of mesal margin having oblique incision, lateral apical margin rounded, bas- olateral margin squarish and apicomesal angle squa- rish with short blunt tooth. Distribution. — India: Karnataka, Maharashtra; Sri Lanka. Remarks. — This species is closely related to D. rec- ta, but it differs by having male modified spur with longer apical points. There has been some confusion regarding the identity of this species, due to an erro- neous redescription of ‘D. morosa’ by Schmid (1958), based on a specimen from Sri Lanka that was actually D. triclavata. D. morosa is only known from the southern islands of the Philippines. The male modified spurs of these species are similar, however the spur of D. triclavata (fig. 52) differs by being tri- fid, but that of D. morosa (fig. 39) is bifid. The record of the male syntype from Shillong, Meghalaya by Martynov (1935) is doubtful and should be con- firmed. 137 [HIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Dipseudopsis varians Ulmer, 1929 > er (fig. 25) Dipseudopsis varians Ulmer, 1929: 191-192, figs. 28-30, 1 ype series: MALAYSIA: West Malaysia: ‘Material in Coll. McLachlan: 3€, Perak.’ Type 1d *: Perak (BMNH), con- dition poor, abdomen and hind legs absent, no other type material available at Mus. (pers. comm. P. C. Barnard). — Fischer 1962: 16. petersorum Schmid & Denning, 1979: 345, fig. 3a-e, Holotype 3, THAILAND: Chiang Mai Prov., east fork, Mae Ping at junction of small stream, 59 km north Chiang Mai, 1350ft, at light, 30.X1.1964, W. L. & J. G. Peters. Syn. n. - , Dipseudopsis Specimens examined. — BURMA: Tenasserim: 14, Mekane, 90km E Moulmein, 200m (NHRS). MALAYSIA: West Malaysia: 1€ 222, Melaka, det. Ulmer 1905 as Dipseudopsis sp., de Malacca’ Presqu'ile, 1899, Errington de la Croix et P. Chapé (MNHN). 1d, Johor: Kota Tinggi (BMNH). 1d, Negeri Sembilan: Kuala Pilah, at light (BMNH). 36 , Pahang: Kuala Tahan, at light, (MCZC, BMNH, BPBM). 2d, Kedah: Bumbong Lima, light trap (USNM). 1d, Selangor: Kuala Lumpur (BMNH). THAILAND: 1d, Trang (MCZC). 1d 19, 40mi SE Prachuab, 75m (casc). 1d, det. Denning as D. pe- tersorum, Krabinburi, Thung Pho, 5.X.1970, Balmer (- CLDD). 1¢, Phuket, Tonesai waterfall (SOFM). 16, Khao Soy Dan (CLHM). 1d, Chantaburi, Phliu, 100m (uop)). 2d, Loei, Phu Rua, 800m (uoPJ). 1d, Nakorn Nayok Prov., Khao Yai Natl. Park, 700m (zMuc). 236 79, Chiang Mai Zoo, light trap (CLHM). Male. — Head, cervical sclerites, and pronotum orangish brown. Mesonotum, tegula, frontoclypeus, palps and antennae dark brown, but genae lighter brown. Maxillary palps 2.4 mm, proboscis 1.3 mm, as long as first three segments of maxillary palps. Forewing 12 mm, colouration similar to that of D. nebulosa but borders of translucent spots more indis- tinct, dark brown background, large triangular spot between bases of R, and Cu, and translucent subapi- cal spot proximal to fork I, similar to bilobed spot of D. nebulosa but second lobe below R, less distinct. Legs with coxae dark brown, distal segments light brown. Modified spur (fig. 53) shorter than adjacent spur, with modified apical portion % as long as spur, having spiraling point with tuft of setae in basal pock- et. Genitalia (fig. 53): Tergum IX with apical % or nore overhanging segment X in lateral view, triangu- with posterior margin rounded in dorsal view; ternum IX mesosuperior process slender and acumi- in lateral view. Segment X oblong with dorsal nd ventral margins roadly rounded wi parallel in lateral view, apex ith venter slightly incised; cordate | with lateral sides tapering distad in dorsal | lightly incised. Preanal appen- th obtuse dorsal angle in lateral r appendages oblong with irregular dor- | s in lateral view, ventral margin D-shaped wi >” we al bump, dorsal margin irregu- lar with conspicuous tooth in middle, distal arms fin- gerlike in ventral view, with lateral margins diverging slightly distad and basal shelves broad. Distribution. — Burma; Malaysia: West Malaysia; Thailand. Remarks. — This species is somewhat similar to D. immaculata, but differs by having male with modified spur shorter and twisted like a short corkscrew, and by having the male genitalia with inferior appendages with a mid dorsal tooth in lateral view and without an apicomesal tooth in ventral view. Dipseudopsis voluta Ulmer, 1906 (fig. 54) Dipseudopsis voluta Ulmer, 1906: 87-88, 110, fig. 89, Type d*: INDONESIA: Sulawesi, ‘Makassar’, McL. coll, without abdomen and hind legs (BMNH). — Fischer 1962: 17; Fischer 1972: 8. Specimens examined. — INDONESIA: Sulawesi: 1d, with- out hind legs, ‘Makassar’, McL. coll (BMNH). Tenggara: 1d, Desa Aopa, 50m, 27.X.1989, at light, R. de Jong & J. Huisman (RMNH). 1d, Moramo, Sg Sena, 50m, 15.X1.1989, at light, R. de Jong & J. Huisman (RMNH). Male. — Head dorsum mostly dark brown, posteri- or warts light brown; frontoclypeus brown with dor- sal portion dark brown, genae light brown, antennal scape dark dorsad and light brown ventrad. Maxillary palps 2.3 mm, apical segments 1-3 yellowish brown, but basal segments 4-5 dark brown; proboscis 1.1 mm, as long as first three segments of maxillary palps, labial palps 0.5 mm, second and third segments min- ute. Pronotum mostly brown, posterior margin light brown with a few darker muscle scars, and propleu- ron yellowish brown. Mesothorax with notum and te- gula dark brown, epimeron dark brown, episternum and sternum yellowish brown. Forewing 14-15 mm, dark brown with golden setae, and some translucent spots between R,., and R,,,, M, , and M, and hyaline spots at m-cu and arculus. Front leg with anterior portion of coxa dark brown, remainder of leg yello- wish brown, mid and hind legs yellowish brown, but hind tibia with brown ring at subapical spur. Modified spur (fig. 54) 2 longer than adjacent spur, bifid with long subapical point spiraling around long- er straight point, about % as long as spur. Genitalia (fig. 54): Tergum IX broad with posterior margin blunt triangular in dorsal view; sternum IX dorsome- sal process with blunt posterior lobe in lateral view. Segment X similar as in D. tonkinensis, but base of ventral margin nearly straight and apex curved dorsad in lateral view. Inferior appendages basal % with slen- der squarish ventral shelf, distal 4 of ventral margin bent dorsad, dorsal margin slightly concave, apex broadened slightly and obtuse. Distribution. — Indonesia: Sulawesi. Remarks. — This species is easily recognized by hav- ing the modified spur with a long straight apical point and another more slender point spiraling completely around its base. The genitalia of D. voluta are similar as in D. tonkinensis, but differs by having inferior ap- pendages with short obtuse subapical mesal tooth in ventral view, and with apical portion slightly tapered in lateral view. This species was previously only known from the male type which lacks the abdomen and is in rather poor condition. Two male speci- mens were collected recently, making it possible to provide a description of the male genitalia for the first time. NOMINA DUBIA Dipseudopsis onychophora Navas, 1935 nomen dubium Dipseudopsis onychophora Navás, 1935: 104-105, Type d: INDIA: ‘Lonawla (Bombay), 9-X-1934. H. Benavent, S. J.’ [location of type unknown]. — Fischer 1962: 14; Higler 1992: 80. Remarks. — The distributional data of other species of Dipseudopsis in India and the figure provided in the original description of D. onychophora by Naväs (1935) of the modified spur of the male type, suggests that this species is probably synonymous with D. tri- clavata. However, since we are unable to locate and examine the type, the status of this species remains uncertain. Dipseudopsis orientalis (Naväs, 1913) nomen dubium Esperona orientalis Naväs, 1913: 12-13, fig. 2a-c, Type, sex unknown: VIETNAM: ‘Hanoï (Tonkin). Un échantillon envoyé et cédé généreusement pour ma collection par M. Laboissiére.’ [location of type unknown]. — Fischer 1962: 14. Remarks. — We were not able to locate and ex- amine the type of this species, hence its identity re- mains uncertain. D. orientalis could be synonymous with one of the three species of Dipseudopsis known from Vietnam, D. benardi, D. robustior, and D. tonk- inensis. According to the original species description (Naväs 1913) the forewing of the type of D. orientalis is 12.4 mm, suggesting that this species is probably synonymous with either D. robustior or D. tonkine- nsis. WEAVER & MALICKY: Dipseudopsis from Asia POLYCENTROPODIDAE ULMER Eodipseudopsis Marlier 1959 nomen dubium Eodipseudopsis Marlier 1959: 117. Type species: Eodipseudo- psis tomensis Marlier (original designation). The genus Eodipseudopsis is removed from the Dipseudopsidae and is provisionally placed in the Polycentropodidae. Three of its characteristics sup- port its removal from Dipseudopsidae: 1) Female genitalia as figured in the original description have sternum VIII bearing a pair of ventrolateral processes (Marlier 1957: fig. 10A). 2) The antennae (cf. Marlier 1959: fig. 48A) are slender with their bases distant from each other. 3) The long slender lobe of the sent gland of sternum V (Marlier 1962: fig 10B) is uncharacteristic of dipseudopsid females examined, Le. Dipseudopsis, Hyalopsyche, Phylocentropus. tree (0 O Philopotamidae Hiel e Psychomyiidae 16=2 Xiphocentronidae m Ecnomidae 145=4 Polycentropodidae L3 Hydropsychidae 17 me Phylocentropus List ho Hyalopsyche 13-—11 Hyalopsychella E i= Dipseudopsis 12:—7 Protodipseudopsis 9 Limnoecetis tree?! eo Philopotamidae po] fF Psychomyiidae 17-—2 Xiphocentronidae m Ecnomidae —12-—4 Polycentropodidae [Est Hydropsychidae 18 as Phylocentropus Lie ee Hyalopsyche el14=11 Hyalopsychella m Dipseudopsis —=133=7 Protodipseudopsis 9 Limnoecetis tree 2 : Ta Philopotamidae 20] | el Psychomyiidae 182=2 Xiphocentronidae m Ecnomidae r16 = Polycentropodidae | 3 Hydropsychidae 19 8 Phylocentropus —17 we: Hyalopsychella L mV Hyalopsyche 15 —9 Limnoecetis 13 E Dipseudopsis Li 2-=7 Protodipseudopsis Figure 55. Total tree output calculated by Hennig 86 for Hydropsychoidea, including outgroup and genera of Dipseudopsidae. 139 IHDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Psychomiidae X iphocentronidae = + Ecnomidae LE Polycentropodidae = Hydropsychidae nen Hyalopsyche Hyalopsychella Dipseudopsis Protodipseudopsis Limnocoetis TE [pes FE Figure 56. Consensus tree of the families of the Hydro- psychoidea s str. Weaver, showing phylogenetic relation- ships of the genera of Dipseudopsidae. ACKNOWLEDGEMENTS We are grateful to the following institutions and curators for transacted loans for material examined: British Museum of Natural History, London, Stephen J. Brooks and Peter C. Barnard; Bishop Museum, Honolulu, Scott E. Miller, Gordon M. Nishida and Keith Arakaki; California Academy of Sciences, San Francisco, Wojcieck J. Palawski, Paul H. Arnaud, Jr., and Vincent F. Lee; Canadian National Collection, Ottawa, Fernand Schmid; per- sonal collection, D. G. Denning; Entomological Laboratory, University of Osaka Prefecture, S. Moriuti; Instituut voor Taxonomische Zoologie Amsterdam, L. Botosaneanu; Museum of Comparative Zoology, Harvard University, David G. Furth; Muséum National d'Histoire Naturelle, Paris, Drs. Jean Legrand, and Jeanine Casewitz-Weulersse; Museu de Zoologia, Barcelona, O. Escolä; Museum for Zoology, University of Lund, L. Cederholm; Naturhistorisches Museum Wien, A. Kaltenbach and U. Aspöck; Nationaal Natuurhistorisch Museum, Leiden, Jan van Tol and Jolanda Huisman; Naturhistoriska Riksmuseet, Stockholm, G. Hallin; Zoological Institute of the Russian Academy of Sciences, St. Petersburg, Vladimir D. Ivanov and Lidija A. Zhiltsova; Smithsonian Institution, United States National Museum of Natural History, Washington, Oliver S. Flint, Jr; Zoologisches Museum der Humboldt-Universität Berlin, W. Mey; Zoologisk Museum Kobenhavn, ©. Karsholt; Zoological Museum Sofia, K. Kumanski and P. Beron; Zoological Survey of India, Calcutta, K. Rai K. Ghosh: and Zoologische Staatssammlung Mi pi de E. Burmeister. re also grateful to the following colleagues for luable material from their personal collec- ns: M. Allen, P. Chantaramongkol, E. Diehl, J. A. W. Lucas, R. Müller, J. Oláh, V. Puthz, W. Speidel, 140 F. Starmühlner, Tian Li-xin, and W. Wittmer. Also, we thank Arturs Neboiss, Museum of Victoria, for providing an editorial review of the manuscript and transferring borrowed material from RMNH, John F. Burger, University of New Hampshire, for reviewing the manuscript, and Matthew Kelly for most of the il- lustrations. The support of the National Science Foundation (grant BSR8907401) is gratefully ac- knowledged. REFERENCES Albarda, H., 1881. Neuroptera. — In: J. P. Veth, ed., Midden-Sumatra 4 (2), Natuurlijke Historie. 5: 1-22, pls. 1-6, Leiden. Banks, N., 1907. Descriptions of new Trichoptera. — Proceedings of the entomological Society of Washington 8: 117-33, pls. 8-9. Banks, N., 1911. Notes on Indian neuropteroid insects. — Proceedings of the entomological Society of Washington 13: 99-106, pl. 6. Banks, N., 1913. Synopsis and descriptions of exotic Neuroptera. — Transactions of the American entomolog- ical Society 39: 201-42, pls. 23-6. Banks, N., 1914 (1913). On a collection of neuropteroid in- sects from the Philippine Islands. — Proceedings of the en- tomological Society of Washington 15: 170-80, pls. 8-9. (22 January 1914). Banks, N., 1916. Neuropteroid insects of the Philippine Islands. — Philippine Journal of Science 11: 195-217, pls. 1-2. Banks, N., 1920. New neuropteroid insects. — Bulletin of the Museum of Comparative Zoology 64: 299-362, pls. 1-7. Banks, N., 1924. Descriptions of new neuropteroid insects. — Bulletin of the Museum of Comparative Zoology 65: 421-55, pls. 1-4. Banks, N., 1931a. Some Oriental neuropteroid insects. — Psyche 38: 56-70, pl. 5. Banks, N., 1931b. Some neuropteroid insects from the Malay Peninsula. — Journal of the Federated Malay Museums 16: 377-409. Brauer, F., 1868. Neue von Herrn Dr. G. Semper gesa- mmelte Neuropteren. — Verhandlungen der zoologisch- botanischen Gesellschaft in Wien 18: 263-268, 2 pls. Crichton, M. I., 1957. The structure and function of the mouth parts of adult caddis flies (Trichoptera). — Philosophical Transactions of the Royal Society of London (Series B) 241: 45-91. Cummings, B. F., 1913. Apropos of the first maxillae in the genus Dipseudopsis Walk. (Trichoptera). — Annals and Magazine of Natural History 11: 308-312. Fabricius, J. C., 1781. Species Insectorum. Vol. I, 552 p. — Hamburg. Farris, J. S., 1988. Hennig86, version 1.5. Program and doc- umentation. Port Jefferson, New York. Fischer, F. C. J., 1962. Polycentropodidae and Psychomyi- dae. Trichopterorum Catalogus. Nederlandsche Entomo- logische Vereeniging, Amsterdam 3: 236 p. Fischer, F. C. J., 1972. Supplement to volumes III and IV. Trichopterorum Catalogus. Nederlandsche Entomologi- sche Vereniging, Amsterdam13: 172 p. Flint, O. S., Jr. 1991. Studies of Neotropical caddisflies, XLV: The taxonomy, phylogeny, and faunistics of the Trichoptera of Antioquia, Colombia. — Smithsonian Contributions to Zoology, no. 520: 1-113. Gibbs, D. G., 1968. The larva, dwelling-tube and feeding of a species of Protodipseudopsis (Trichoptera: Dipseudo- psidae). — Proceedings of the Royal entomological Society London (Series A) 43: 73-79. Higler, L. W. G., 1992. A checklist of the Trichoptera re- corded from India and a larval key to the families. — Oriental Insects 26: 67-106. Iwata, M., 1927. Trichopterous larvae from Japan. — Annotationes zoologicae japonenses 11: 203-233. Kimmins, D. E. 1957. Lectotypes of Trichoptera from the McLachlan Collection now in the British Museum (Natural History). — Bulletin of the British Museum (Natural History) 6(4): 93-126. Lepneva, S. G., 1964. Trichoptera (Rucheiniki), Larvae and pupae of Annulipalpia. In: Pavlovskii, E. N., ed, Fauna of the U. S. S. R., 2 (1): 560 p. (Translated into English, by Program for Scientific Translations, Jerusalem 1970). Lestage, J. A., 1925. Notes trichoptérologiques (7me Note). — Bulletin et Annales de la Société royal entomologique de Belgique 65: 35-44. Lestage, J. A., 1936. Notes trichoptérologiques. XIV. — Les composantes de la faune sud-africaine et la dispersion transafricaine de quelques espèces. — Bulletin et Annales de la Société royal entomologique de Belgique 76: 165- 192. McLachlan, R., 1863. On Anisocentropus, a new genus of ex- otic Trichoptera, with descriptions of five species, and of a new species of Dipseudopsis. — Transactions of the ento- mological Society of London 1(3): 492-6, pl. 19. McLachlan, R., 1864. On the types of Phryganidae de- scribed by Fabricius from the Banksian Collection. — Transactions of the entomological Society of London 1: 656-659. McLachlan, R., 1866. Descriptions of new or little-known genera and species of exotic Trichoptera, with observa- tions on certain species described by Mr. F. Walker. — Transactions of the Royal entomological Society of London 5: 247-75, pls. 17-9. McLachlan, R., 1875. Descriptions de Plusieurs Névro- pteres-Planipennes et Trichoptères nouveaux de l’île de Celebes et de quelques espèces nouvelles de Dipseudopsis avec considérations sur ce genre. — Tijdschrift voor Entomologie 18: 1-21, pls. 1, 2. Malicky, H., 1984. The caddisflies (Trichoptera) collected by the Austrian-Indian Mission in 1976 on the Andaman Islands. — Annalen des Naturhistorischen Museums in Wien 86: 213-8. Malicky, H. & J. S. Weaver III, 1988. Die von Dr. E. W. Diehl in Sumatra gesammelten Dipseudopsis-Arten (Trichoptera: Dipseudopsidae). — Opuscula Zoologica Fluminensia 24: 1-9. Marlier, G., 1953 (1952). Études hydrobiologiques dans les rivières du Congo oriental. B. Trichopteres - Polycentro- podidae (Dipseudopsinae). — Annales du Musée royal du Congo belge, Série 8° (Sciences Zoologiques) 21: 51-65. Marlier, G., 1955. Un Trichoptère pélagique du Lac Tanganika. — Revue de zoologie et de botanique af- ricaines 52: 150-155. Marlier, G., 1959. Mission du Museum dans les Iles du Golfe de Guinée. Entomologie XII (1). Trichoptères de Sao Tomé. — Revue Française d’entomologie 63: 108- 120. WEAVER & MALICKY: Dipseudopsis from Asia Marlier, G., 1961. La nymphe et la position systematique de Limnoecetis tanganicae Marlier (Trichoptera). — Bulletin Institut royal des Sciences naturelles de Belgique 38: 1-6. Marlier, G., 1962. Genera des Trichoptères de l’Afrique. — Annales du Musée Royal de l’Afrique Centrale, Série 8° (Sciences Zoologiques) no. 109: 1-261. Marlier, G., 1979. Une nouvelle espèce de Dipseudopsis de Thailande (Trichopteres: Polycentropidae). — Bulletin Institut royal des Sciences naturelles de Belgique 51: 1- 11. Martynov, A. [V.], 1935. On a collection of Trichoptera from the Indian Museum. — Records of the Indian Museum 37: 93-209. Mosely, M. E., 1933. Synonyms in the order Trichoptera. — Annals and Magazine of Natural History (series 10) 12: 497-503. Naväs, L., 1913. Espèces nouvelles de Néveroptères exo- tiques. — Annales de |’Assacociation des naturalistes de Levallois-Perret 19: 10-13. Naväs, L., 1915. Neuröpteros nuevos o poco conocidos (cuarta serie). — Memorias de la Real Academia de cien- cias y artes de Barcelona 11(23): 373-98. Navas, L., 1921. Insectos exöticos neuvos, criticos 0 poco conocidos. — Revista de la Academia de Ciencias exactas fisico-quimicas y naturales de Zaragoza 6: 61-81. Navás, L., 1924. Neue Trichopteren. — Konowia, Wien 3: 204-9. Navas, L., 1930. Insectos del Museo de Parie (6° serie). — Brotéria, Série zoolögia 26(3): 120-44. Navas, L., 1931. Decadas de insectos neuvos. — Revista de la Academia de ciencias exactas fisico-quimicas y naturales de Madrid 26: 60-86. Navas, L., 1935. Décadas de insectos nuevos. — Brotéria, Série Ciencias naturais 4(31): 97-107. Ross, H. H., 1956. Evolution and classification of the mountain caddisflies. Univ. Illinois Press, Urbana, 213 p. Ross, H. H., 1967. The evolution and past dispersal of the Trichoptera. — Annual Review of Entomology 12: 169- 206. Ross, H. H. & J. M. Kingsolver, 1959. The Madagascan fauna of Dipseudopsis with speculations about the phylog- eny and dispersal of the genus (Trichoptera). — Mémoires de l’Institut Scientifique de Madagascar, Série E 11: 103- 128. Ross, H. H. & D. G. Gibbs, 1973. The subfamily relation- ship of the Dipseudopsinae (Trichoptera, Polycentropo- didae). — Journal of the Georgia entomological Society 8: 312-6. Schmid, F., 1949. Les Trichopteres de la collection Navas. — Eos, Madrid 25: 305-426. Schmid, F., 1958. Trichoptéres de Ceylan. — Archiv fiir Hydrobiologie 54: 1-173, 34 pls. Schmid, F., 1980. Les insectes et arachinides du Canada, partie 7: Genera des Trichoptéres du Canada et des Etats adjacents. Agriculture Canada. pub. 1692. Ottawa. 296 p- Schmid, F. & D. G. Denning, 1979. Descriptions of new Annulipalpia (Trichoptera) from southeastern Asia. — Canadian Entomologist 111: 243-9. Schuster, G. A. & S. W. Hamilton, 1984. The genus Phylocentropus in North America (Trichoptera: Polycen- tropodidae). — Proceedings International Symposium on Trichoptera 4 (Series Entomologica 30): 347-62. Scott, K. M. F., 1985. Order Trichoptera (caddis flies). In: C.H. Scholtz & E. Holm, eds., Insects of Southern 141 T11DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Africa. p. 327-40. Butterworths, Durban. Tsuda, M., 1939. Metamorphose von drei Köcherfliegen, lanna falcata Ulmer, Tinodes sauteri Ulmer und Dipseudopsis stellata MacLachlan. — Annotationes Zoolo- 14, gicae Japonenses 18(3): 207-212. Tsuda, M., 1942. Japanische Trichopteren, pt. 1. Systematik. — Memoirs of the College of Science, Kyoto Imperial University, Series B, 17(1): 239-339. Ulmer, G., 1904a. Über einige Trichopteren mit rüsselfor- migen Kopfanhängen. — Zoologischer Anzeiger 28: 56- 59. Ulmer, G., 1904b. Uber westafrikanische Trichopteren. — Zoologischer Anzeiger 28: 353-359. Ulmer, G., 1905. Neue und wenig bekannte außereuro- päische Trichopteren, hauptsächlich aus dem Wiener Museum. — Annalen des Naturhistorischen Museums in Wien 20: 59-98. Ulmer, G., 1906. Neuer Beitrag zur Kenntnis aussereuro- päischer Trichopteren. — Notes from the Leyden Mu- seum 28: 1-116. Ulmer, G., 1907a. Neue Trichopteren. — Notes from the Leyden Museum 29: 1-53. Ulmer, G., 1907b. Trichoptera. In: P. Wytsman, ed., Gene- ra Insectorum. Verteneuil & Desmet, Bruxelles, Fasc. 60, 259 p., 41 pls. Ulmer, G., 1909. Einige neue exotische Trichopteren. — Notes from the Leyden Museum 31: 125-42. Ilmer, G., 1915. Trichopteren des Ostens, besonders von Ceylon und Neu-Guinea. — Deutsche entomologische Zeitschrift, Berlin pp. 41-75. Ulmer, G., 1929. Über einige, hauptsächlich asiatische, Ephemeropteren und Trichopteren aus der Sammlung R. McLachlan. — Deutsche entomologische Zeitschrift (Ber- lin) 3: 161-195. Ulmer, G., 1930. Trichopteren von den Philippinen und von den Sunda-Inseln. — Treubia 11: 373-498. m Ulmer, G., 1951. Köcherfliegen (Trichopteren) von den Sunda-Inseln. Teil I. — Archiv für Hydrobiologie. Suppl. 19: 1-528, 28 pls. Ulmer, G., 1957. Köcherfliegen (Trichopteren) von den Sunda-Inseln. Teil III. — Archiv für Hydrobiologie. Suppl. 23: 109-470, 31 pls. Walker, F., 1852. Catalogue of the specimens of neuropte- rous insects in the collections of the British Museum, pt. I, 135 p. Wallace, ti B., W. R. Woodall & A. A. Staats, 1976. The lar- val dwelling-tube, capture net and food of Phylocentropus placidus (Trichoptera: Polycentropodidae). — Annals of the entomological Society of America 69: 149-154. Weaver, J. S., III., 1984. The evolution and classification of Trichoptera, part I: The groundplan of Trichoptera. — Proceedings International Symposium Trichoptera 4 (Series Entomology 30): 413-419. Weaver, J. S., III & J. C. Morse., 1986. Evolution of feed- ing and case-making behavior in Trichoptera. — Journal of the North American Benthological Society 5: 150-8. Weidner, H., 1964. Die entomologischen Sammlungen des Zoologischen Staatsinstituts und Zoologischen Museums Hamburg. — Mitteilungen aus dem Hamburgischen zoo- logischen Museum und Institut 62: 55-100. Wells, A. & D. Cartwright, 1993. Females and immatures of the Australian caddisfly Hyalopsyche disjuncta Neboiss (Trichoptera), and a new family placement. — Transactions of the Royal Society of South Australia 117: 97-104. Wiggins, G. B., 1977. Larvae of the North American Caddisfly genera (Trichoptera). Univ. Toronto Press, 401 p. Wiggins, G. B., 1982. Trichoptera. — In: S. P. Parker, ed., Synopsis and classification of living organisms. McGraw- Hill, New York, vol. 2, p. 599-612. Le Sie = hee À Urs A i \ { qu Ten 42 T } I n 4 il | | ; Dt n ‘ = er D È + h 1 n B Ta i ae » 7 2 à h x Paz vi i" i Em Tijdschrift voor Entomologie Volume 137, no. I Articles 27 57 75 87 95 © A. J. de Boer The taxonomy and biogeography of the loriae group of the genus Baeturia Stal, 1866 (Homoptera, Tibicinidae). J. Patocka Die Puppen der Spanner Mitteleuropas (Lepidoptera, Geometridae): Charakteristik und Bestimmungstabelle der Gattungen. J. T. Polhemus & D. A. Polhemus Four new genera of Microveliinae (Heteroptera) from New Guinea. R. Rozkosny & D. Kovac Adults and larvae of two Ptecticus Loew from Peninsular Malaysia (Diptera, Stratiomyidae). J. van Toi The Odonata of Sulawesi and adjacent islands. Part 3. The genus Macromia Rambur (Corduliidae). J. S. Weaver & H. Malicky The genus Dipseudopsis Walker from Asia (Trichoptera: Dipseudopsidae). Nederlandse Entomologische Vereniging, Amsterdam Published 15 July 1994 ISSN 0040-7496 ‚Volume 137, no. 2, 1994 |. | ISSN 0040-749 Tijdschrift voor — Entomologie A journal of systematic and evolutionary entomology since 1858 Published by the Nederlandse Entomologische Vereniging Tijdschrift voor Entomologie A journal of systematic and evolutionary entomology since 1858 Scope The ‘Tijdschrift voor Entomologie’ (Netherlands Journal of Entomology) has a long tradition in the publication of original papers on insect taxonomy and systematics. The editors particularly invite papers on the insect fauna of the Palaearctic and Indo-Australian regions, especially those including evolutionary aspects e.g. phylogeny and biogeography, or ethology and ecology as far as meaningful for insect taxonomy. Authors wishing to submit papers on disciplines related to taxonomy, e.g. descriptive aspects of morphology, ethology, ecology and applied entomology, are requested to contact the editorial board before submitting. Usually, such papers will only be published when space allows. Editors E. J. van Nieukerken (elected 1986) and J. van Tol (1985) Co-editors A. W. M. Mol (1990) and R. T. A. Schouten (1990) Advisory board M. Brancucci (Basel), N. E. Stork (London) and M. R. Wilson (Cardiff). The ‘Tijdschrift voor Entomologie’ is published in two issues annually by the ‘Nederlandse Entomologische Vereniging’ (Netherlands Entomological Society), Amsterdam. Editorial address c/o National Museum of Natural History, Postbus 9517, 2300 RA Leiden, The Netherlands. Correspondence regarding membership of the society, subscriptions and possibilities for exchange of this journal should be addressed to: Nederlandse Entomologische Vereniging c/o Zoölogisch Museum Amsterdam Plantage Middenlaan 64 1018 DH Amsterdam The Netherlands Subscription price per volume Hfl. 300,— (postage included). Special rate for members of the society. Please enquire. Instructions to authors Published with index of volume 137 (1994). Graphic design Ontwerpers B.V, Aad Derwort, s-Gravenhage M. ASCHE & M. D. WEBB? ' Marburg, Germany &° The Natural History Museum, London, UK REVIEW) @E WTAE SOUTHERN PALAEARCTIC AND BPABABOTROPICAL LEAFHOPPER GENUS HENGCHUNIA VILBASTE (HOMOPTERA, CICSDELEIDAE) Asche, M. & M. D. Webb, 1994. Review of the southern Palaearctic and Palaeotropical leaf hopper genus Hengchunia Vilbaste (Homoptera, Cicadellidae). — Tijdschrift voor Entomologie 137: 143-154, figs. 1-63 (issn 0040-7496). Published 15 December 1994. The leafhopper genus Pseupalus Remane & Asche (Deltocephalinae: Paralimnini) is synonym- ized with Hengchunia Vilbaste syn. n. Four new species are described: H. helleri sp. n. from Cameroun and Zaire, H. indica sp. n. from India, H. pakistanica sp. n. from Pakistan and H. javana sp. n. from Indonesia: Java. Hengchunia now contains nine species which are probably all associated with the grass genus Zmperata. The zoogeographical distribution of the species and their potential phylogenetic relationships are briefly discussed. Additions and amendments to Webb & Heller’s (1990) check-list of Afrotropical and Oriental Paralimnini are made and in- clude Paralimnellus Emeljanov (1972), oldest available name for Bubulcus Dlabola (1961) (jun- ior homonym of Bubulcus Bonaparte 1854, Aves) and Paralimnellus vittata (Matsumura) comb. n. from Henschia. Correspondence: Dr Manfred Asche, Am Schwanhof 2A, D-35037 Marburg, Germany. Key words. — Homoptera; Auchenorrhyncha; Cicadellidae; Paralimnini; Hengchunia; Pseupalus; review; new species; South Palaearctic and Palaeotropical Region. This paper concerns a group of paralimnine leaf- hoppers (Cicadellidae: Deltocephalinae) apparently associated with the Palacotropical grass /mperata, es- pecially the species Z. cylindrica. This association was reported by Remane & Asche (1980) who erected the genus Pseupalus for a single paralimnine species, P. graecanarus from the Canary Islands (type locality) and Greece. Recently, Pseupalus and related taxa were reviewed by Webb & Heller (1990) who added three more species: P. gaiseri from Ivory Coast, P. digitus from Thailand, and P. murtus from the Philippines: Luzon. Webb & Heller (l.c.) discussed the validity of the tribe Paralimnini sensu Emeljanov (1962), Remane & Asche (1980), Ossiannilsson (1983), and Emeljanov & Kirilova (1989) and Hamilton’s ‘Deltocephalina’ (Hamilton 1975), and confirmed the tribal placement of Pseupalus. In addition, Webb & Heller (l.c.) provided a checklist of Afrotropical and Oriental paralimnine genera. Since then, the sec- ond author has examined specimens of a species from Taiwan, described as Thamnotettix koshunensis Matsumura, for which Vilbaste (1969) had estab- lished the genus Hengchunia. A comparison revealed not only the placement of Hengchunia in the Oriental Paralimnini, but also its generic synonymy with Pseupalus. Moreover, four new species of this genus were discovered in samples from Africa, India, and Indonesia, described below, together with an uniden- tifiable female specimen from Australia. The genus Hengchunia now contains nine species: one in the Southwestern Palaearctic, two in the Afrotropical Region, and 6 in the Oriental Region. The following amendments and additions apply to the Afrotropical and Oriental paralimnine check-list (Webb & Heller 1990), noted above: Paralimnellus Emeljanov (1972) (type species: cingulatus Dlabola) oldest available name for Bubulcus Dlabola (1961) (type species: cingulatus Dlabola) (junior homonym of Bubulcus Bonaparte, 1854, Aves); Paralimnellus vittata Matsumura comb. n. (Henschia); Pteropyx ac- utus (Singh) (Jassargus) Emeljanov, 1972: 66; Pteropyx maculatus Emeljanov, 1972: 67 (Malı); Khasia prima Rao, 1989: 82 (India). Depositories Abbreviations for depositories of material used in this paper: M. Asche & H. Hoch, Marburg, private collection (AH); Agricultural University Wageningen, The Netherlands (auwn); The Natural History Museum, London, United Kingdom (BMNH); Musee 143 DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Webb & Heller) ‚ paratype d from Ivory Coast; Figs. 6, 10: 6 from Sierra Leone. — 1, male genital plate, dorsal view; 3, same, dorsolateral view; 4, subgenital plates, ventral view; nd connective, ventrocaudal view; 8, aedeagus, left lateral view; 9, 10, same, left la- : 0.1 mm. Royal de l’Afrique Centrale, Tervuren, Belgium (MRAC); Institut Royal des Sciences Naturelles de Belgique, Brussels, Belgium (isNB); Entomological Institute, Hokkaido University, Sapporo, Japan (E1- HU); Staatliches Museum für Naturkunde Stuttgart, Germany (smns); University of Agricultural Sciences, Bangalore, India (UASB). TAXONOMY Hengchunia Vilbaste Hengchunia Vilbaste, 1969: 8. — Type species by original de- signation: Thamnotettix koshunensis Matsumura. Pseupalus Remane & Asche, 1980: 88, syn. n. - Type species by original designation: Pseupalus graecanarus Remane & Asche. For relationship and diagnosis of Hengchunia see the corresponding paragraphs for Pseupalus in Webb & Heller (1990: 2, 3). Key to the species of Hengchunia (males) (after Webb & Heller 1990, modified and supplemented) 1. Preatrium of aedeagus with a pair of long and slenderiprocesses,(e.g., figs. 7, 8)... ? — Preatrium of aedeagus without a pair of processes (Be RL nl H. pakistanica sp. n. 2. Aedeagal shaft with four spinose processes apical- WE OS) te: H. koshunensis (Matsumura) — Aedeagal shaft with two or less spinose processes apically or subapically (e.g., fig. 7) . … eee 3 3. Aedeagal shaft with a single spinose process or lat- Éraldanses Subapicallys...................1!. 4 — Aedeagal shaft with two spinose processes apical- 1 or gula el 5 4. Aedeagal shaft with a horn-shaped subapical pro- cess directed ventrad, dorsal side of shaft at mid- length with a bifurcate process and shortly basad of it with a single short process (figs. 32-34) … RR Mi neri H. indica sp.n. — Aedeagal shaft with a subapical flange on each side, devoid of spinose processes (see Webb & Peer MIO OO She. 19)... SIOE al ON H. graecanarus (Remane & Asche) 5. Subgenital plates with distal margin slightly con- cave (fig. 14); aedeagal shaft medially on dorsal side with a group of 3 short spines (fig. 18) .... 6 — Subgenital plates with distal margin convex (e.g., figs. 2, 22); aedeagal shaft medially on dorsal side devoid of short spines (e.g., figs. 8, 26, 61) ..... 8 7. Aedeagal shaft with apical processes almost di- rected laterad; mediodorsal processes compara- tively distant from each other; style almost evenly tapering to apex without a subapical lobe (see ASCHE & WEBB: Review of Hengchunia Webb: 82. Heller, 19902 6, fies: 1451 Lo) rans RISARCIRE POTRA H. digitus Webb & Heller — Aedeagal shaft with apical processes directed ven- trobasad; mediodorsal processes near to each oth- er (fig. 18); style with a subapical lobe (fig. 16) .. ERTL H. murtus Webb & Heller 8. Aedeagal shaft with processes apically, directed dorsocaudad, not reflected (figs. 60-62) ............ Cri i er Re H. javana sp. n. — Aedeagal shaft with processes subapically or me- dially, directed laterobasad (e.g figs. 7, 25) … 9 9. Aedeagal shaft with processes located subapically (fps 710) mann H. gaiseri Webb & Heller — Aedeagal shaft with processes located mediolate- ral (fies 25,20) en cate cece H. helleri sp. n. 1. Hengchunia koshunensis (Matsumura) (fig. 60) Thamnotettix koshunensis Matsumura, 1914: 178. Lectotype 3, Taiwan (ErHU), here designated [examined]. Hengchunia koshunensis: Vilbaste 1969: 8, fig. 6: A-J. Material examined. — Lectotype d, ‘Formosa/Mat- sumura’ and ‘5.vii.1906/Koshun’ on reverse; 19, 1 ? (sex unknown). — Paralectotypes, originally on same mount as lectotype (re-mounted) (EIHU). Fig. 11. Hengchunia murtus (Webb & Heller), d from Sulawesi. — 11, habitus. Scale: 1.0 mm. 145 T1NDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 er Ener vier ee (Webb & Heller), d from Sulawesi. — 12, male genitalia, ventral view; 13, same, left lateral iew; 14, left subgenital plate, dorsal view; 15, same, dorsolateral view; 16, left style, dorsal view; 17, aedeagus and connecti- ve, ventrocaudal view; 18, aedeagus, dorsal view; 19, same, left lateral view. Scales: 0.1 mm. 146 Remarks. — H. koshunensis can be separated from the other Hengchunia species by the crown-shaped apex of the aedeagal shaft consisting of two pairs of spines: a pair of longer spines directed ventrad and a pair of shorter spines directed dorsad (not vice versa as noted in Vilbaste 1969: 8). Distribution. — This species is only known from Taiwan (Matsumura 1914; Vilbaste 1969). 2. Hengchunia graecanarus (Remane & Asche) comb. n. Pseupalus graecanarus Remane & Asche, 1980: 90, fig. 16 a- I. Holotype d, Canary Is, (Remane coll.) [examined]. — Webb & Heller 1990: 5, fig. 13... Material examined (additional to the type material of Remane & Asche 1980). - 2d, 5%, S Greece, Peloponnessos, Skala nr Gythion, on Imperata cylindrica, 6.x.1981, M. Asche & H. Hoch. - 9d, 269, NW Greece, Corfu I., Aghios Stephanos, coastal biotopes with /mperata cylindrica, 14.x.1981, M. Asche & H. Hoch. - 3d, 12, NW Greece, 27 km S. Igoumenitsa, S. Kastri, on /mperata cylin- drica, 10.viii.1983, M. Asche & H. Hoch (an). Remarks. — Within the genus, A. graecanarus dis- plays a unique configuration of the aedeagus with subapical flanges at the shaft instead of spinose pro- cesses. Distribution. — This species is known from the type-locality in the Canary Islands: Gran Canaria, and from several localities in Greece (Remane & Asche 1980). According to the distribution of its host plant, /mperata cylindrica, a wider distribution at least in the Mediterranean Region is possible. 3. Hengchunia digitus (Webb & Heller) comb. n. Pseupalus digitus Webb & Heller, 1990: 4, figs. 14-18. Holotype d, Thailand (BMNH) [examined]. Remarks. — No material additional to the type ma- terial has been examined. H. digitus closely resembles H. murtus in the shape of the subgenital plates (including the toothed area) and aedeagus but the aedeagal shaft is straighter and has the three mediodorsal spines less closely adjacent and the style apex narrower. Distribution. — This species has been reported on- ly from Thailand (Webb & Heller 1990). 4. Hengchunia gaiseri (Webb & Heller) comb. n. (figs. 1-10) Pseupalus gaiseri Webb & Heller, 1990: 4, figs. 1-12. Holotype à , Ivory Coast (sMNs) [examined]. ASCHE & WEBB: Review of Hengchunia Remarks. — No material additional to the type ma- terial has been examined. H. gaiseri is similar (and probably closely related) to H. helleri (see below). It differs from this species in the distally more pointed subgenital plates and in the position of the lateral spinose aedeagal processes which are located subapically in H. gaiseri and in midlength of shaft in H. helleri. Distribution. — Although this species was described from Ivory Coast it is likely that a similar male spec- imen from Sierra Leone is conspecific (Webb & Heller 1990: 4). 5. Hengchunia murtus (Webb & Heller) comb. n. (figs. 11-19) Pseupalus murtus Webb & Heller, 1990: 6. Holotype à, Philippines (BMNH) [examined]. Material examined (additional to the type material of Webb & Heller 1990). — Indonesia: Sulawesi: 1d, Sulawesi Utara, Dumoga Bone National Park, rainforest near base- camp Toraut, 19.x.1985; 1d, ibid., banks of Toraut river; 14, ibid., light trap near basecamp, 7.xii.1985; 1d, 2°, ib- id, trail to Tumpah river, 23.xi.1985; 1d, ibid., trail to Tumpah river uphill to ca. 400 m, 20.x.1985; 12, ibid. banks of Tumpah river; 1d, Molosso I., opposite Lolok (N.-coast), 11.xi.1985; all: on Jmperata grass, Project Wallace Expedition, M. Asche & H. Hoch (AH, BMNH). Remarks. — H. murtus closely resembles A. digitus in the shape of the male genitalia (see under H. digi- tus). Distribution. — This species was originally descri- bed from Luzon I. in the Philippines (Webb & Heller 1990). Its occurrence in Sulawesi extends its range to the south. 6. Hengchunia helleri sp. n. (figs. 20-26) Type material. — Holotype d: Zaire, Lubumbasha (= Elisabethville), 30.iii1.1939, at light, H.-J. Brédo (isn). — Paratypes: Zaire: 196, 439, 2 ? (sex unknown), same data as holotype (various dates) (ISNB, MRAC, BMNH); 30, 32, Libenge, Savane Liki-Bembe, 23-28.ii.1948 (one without date) (R. Cremer & M. Neuman, (isns); 12, Kivu, Kinchassa, Waelbroeck, 24.ii and 3.iv.1899 (IsNB); 18,12, Maka, Lualaba, 25.1.1939, H.-J. Bredo (tsNB). - Cameroun: 14, Bertoua, 14.ix.1960, F.J. Breteler (auwn). Description. — Habitus, bodily proportions and colouration as in generic description (see Webb & Heller 1990). Male genitalia: Subgenital plate (figs. 20-23) distal- ly broadly rounded; toothed area rather long, extend- ing from apex to about half of the length of the sub- genital plate mediobasad, apical tooth and basal teeth 147 THDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 20-26. Hengchunia helleri sp.n., paratype d from Cameroun. — 20, male genitalia, ventral view; 21, same, left lateral view; 22, left subgenital plate, dorsal view; 23, same, dorsolateral view; 24, left style, dorsal view; 25, aedeagus and connecti- ve, ventrocaudal view; 26, aedeagus, left lateral view. Scales: 0.1 mm. 148 AscHE & WEBB: Review of Hengchunia Figs. 27-35. Hengchunia indica sp.n., holotype 3 from India. — 27, male pygophore, left lateral view; 28, left subgenital pla- te, dorsal view; 29, same, dorsolateral view; 30, subgenital plates, ventral view; 31, left style, dorsal view; 32, aedeagus and connective, ventrocaudal view; 33, aedeagus, left lateral view; 34, same, dorsal view; 35, apex of aedeagus, left dorsocaudal view. Scales: 0.1 mm. strongest. Style (fig. 24) with short inner and long outer apophysis, the latter continuously tapering to apex, distal margin with about 7 tooth-like projec- tions. Aedeagus (figs. 25, 26) with shaft short, evenly curved dorsad, on both sides at midlength a spinose process which is directed basad; phallotreme apically, exposed to the ventrocaudal side; processes of preatri- um long and slender, straight. Length: d: 3.5-3.6 mm, 2: 3.8 mm. Remarks. — H. belleri is similar (and probably closely related) to A. gaiseri; it can be distinguished mainly by the shape of the subgenital plates which are apically less pointed than in H. gaiseriand which bear a more extended toothed area, as well as by the posi- tion of the aedeagal processes which arise at mid- length of the shaft (not subapically as in H. gaiseri). Distribution. — The occurrence of A. helleri in Zaire and Cameroun may indicate a wider distribu- tion, at least in Central- and West Africa. 149 [1]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 36-39. Hengchunia pakistanica sp.n., holotype d from NE Pakistan. — 36, head and thorax, dorsal view; 37, head, fron- tal view; 38, left tegmen; 39, left hind wing. Scales: 0.5 mm. 7. Hengchunia indica sp. n. (figs. 27-35) Type material. — Holotype d: India, Bihar, Pusa, 14.v.1931, T. B. Fletcher (BMNH). Description. — Habitus, bodily proportions, and colouration as in generic description (see Webb & Heller 1990). Male genitalia: Subgenital plate (figs. 27-30) rather stout with distal margin very little convex, almost truncate; toothed area with a triangular portion at margin and an arm-like extension directed mediobas- ad. Style (fig. 31) comparatively short and stout, de- void of a mediodistal apophysis, subapically with a short lobe-like projection. Aedeagus (figs. 32-35) with shaft evenly curved dorsad; ventrobasad of the apical phallotreme with a gently curved horn-shaped process which arises on a broad base; medially on dor- sal side of shaft a longer, apically bifurcate process, and basad of it a shorter single process; processes of 150 preatrium slender, slightly curved ventrad. Length: d: 3.0 mm. Remarks. — Within the genus, H. indica is most similar to A. pakistanica (see below) with which it shares the short and distally truncate subgenital plates and a single horn-shaped aedeagal process basad of the phallotreme on the ventral side. However, it dif- fers considerably from this species in the shape of the toothed area of the subgenital plates and in details of the aedeagus, especially by the presence of paired pro- cesses of the preatrium which are entirely missing (probably reduced) in H. pakistanica. Distribution. — Only known from the Bihar- Region in North India. 8. Hengchunia pakistanica sp. n. (figs. 36-50) Type material. — Holotype d: Pakistan, Gujrat (= Guja- ASCHE & WEBB: Review of Hengchunia Figs. 40-50. Hengchunia pakistanica sp.n., holotype & from NE Pakistan: Figs. 45, 48: paratype d from SE Pakistan. — 40, male genitalia, ventrocaudal view; 41, same, left lateral view; 42, left subgenital plate, dorsal view; 43, same, dorsolateral view; 44, 45, left style, dorsal view; 46, aedeagus and connective, ventrocaudal view; 47, 48, aedeagus, left lateral view; 49, same, dorsal view; 50, apex of aedeagus, left dorsocaudal view. Scales: 0.1 mm. 151 TIIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 51-54. Hengchunia javana sp.n., holotype d from Java. — 51, head and thorax, dorsal view; 52, head, frontal view; 53, left tegmen; 54, left hind wing. Scales: 0.5 mm. rat), Nausari, 16.1.1981, C.A. Viraktamath (BMNH). - Paratypes: 42, same data as holotype (UASB, BMNH); 1d, 12, Pakistan, Hyderabad, Patanduru, vii-ix.1980, Bernays & Woodhead (BMNH). Description. — Habitus, bodily proportions, and colouration as in generic description (see Webb & Heller 1990). Male genitalia: Subgenital plate (figs. 40- 43) short, distally truncate, toothed area separated in an apical and a median field of teeth which are connect- ed by a ridge. Style (figs. 44, 45) distally with a short median edge and a long outer apophysis with a short subapical lobe. Aedeagus (figs. 46-50) with shaft short, slightly curved dorsad; a single long, horn- shaped, ventrobasad directed process arising basad of phallotreme on the ventral side; shaft submedially with a pair of lateral processes; preatrium devoid of el- ongate paired processes. Length: d : 3.0 mm, ©: 3.1-3.3 mm. \N N Remarks. — H. pakistanica can readily be distin- guished from all other congeners by the lack of pro- cesses of the aedeagal preatrium. The aedeagal shaft with its single horn-shaped process on the ventral side is similar to that of H. indica but it differs in other genital characters, as noted above. Slight differences could by found in the genital structures (especially ae- deagal spines and styles) between the males from North and South Pakistan but they seem too small to warrant species or even subspecific status. Distribution. — Only known from the Gujrat- Region in the North-East and from the Sind-Region in the South-East of Pakistan. 9. Hengchunia javana sp. n. (figs. 51-62) Type material. — Holotype d : Indonesia: Java, Bogor, on Imperata, 24.xi.1989, M.R. Wilson (BMNH). - Paratype: 12, same data as holotype (BMNH). AscHE & WEBB: Review of Hengchunia N Figs. 55-63. Hengchunia species. — 55-62. Hengchunia javana sp.n., holotype & from Java; 55, male genitalia, ventrocaudal view; 56, same, left lateral view; 57, left subgenital plate, dorsal view; 58, same, dorsolateral view; 59, left style, dorsal view; 60, aedeagus and connective, ventrocaudal view; 61, aedeagus, left lateral view; 62, apex of aedeagus, ventrocaudal view. Scales: 0.1 mm. 63; H. koshunensis (Matsumura), aedeagus, redrawn and modified from Vilbaste, 1969, Fig. 6 E. 153 INDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Description. — Habitus, bodily proportions, and colouration as in generic description ( see Webb & Heller 1990). Male genitalia: Subgenital plate (figs. 55-58) distal- ly convex, toothed area devided in an apical and a me- dian portion, both portions with prominent teeth. Style (fig. 59) wich a short inner edge and an elongate outer apophysis with undulate margin. Aedeagus (figs. 60-62) with shaft short, in middle almost rec- tangularly bent dorsad; phallotreme apically, dorsally on each side flanked by a pair of short, fork-like spi- nose processes directed dorsocaudad; paired processes of the preatrium conspicuously surpassing the aedea- gal shaft, almost straight. Length: d: 3.4 mm; 9: 3.6 mm. Remarks. — Within the genus, H. javana is rather isolated. It can readily be separated from the other congeners by the configuration of the aedeagal shaft with its fork-like, dorsocaudad directed ae reflect- ed) spinose processes at apex. Distribution. — Only known from Bogor in West Java. 10. Hengchunia sp. We have examined one female from Australia, Queensland, Cairns (ridge at Edge Hill, rainforest), collected on /mperata grass, on 10.vi.1987 by M. Asche & H. Hoch (AH). Externally, this specimen re- sembles H. murtus, but its specific identity can only be clarified after a corresponding male becomes avail- able; however, this finding extends the geographic range of the genus Hengchunia to NE Australia. ACKNOWLEDGEMENTS For the loan of material we would like to thank the following: Dr. C. Viraktamath (uass), Dr. P.W.F. de Vrijer (auwn), Dr. P. Grootaert (isNB), Dr. H. Andre (MRAC) and Dr. S. Takagi (emu). We also wish to thank Dr. H. Hoch, Museum fiir Naturkunde, Berlin, Germany, and Dr. M. R. Wilson, National Museum of Wales, Cardiff, UK, for helpful com- ments on the manuscript. REFERENCES Dlabola, J., 1961. Die Zikaden von Zentralasien, Dagestan und Transkaukasien (Homopt. Auchenorrhyncha). — Acta Entomologica Musei Nationalis Pragae 34: 241- 358. Emeljanov, A. F., 1962. New tribes of leafhoppers of the subfamily Euscelinae (Auchenorrhyncha, Cicadellidae). — Entomological Review 41: 236-240. Emeljanoy, A. F., 1972. New Palearctic leafhoppers of the subfamily Deltocephalinae (Homoptera, Cicadellidae). — Entomological Review 51: 63-68. Emeljanov, A. F. & V. I. Kirilova, 1989. Trends and modes of karyotype evolution in the Cicadina (Homoptera) I. (Cicadelloidea). — Entomologicheskoe Obozrênie 68: 587-603. Hamilton, K. G. A., 1975. Review of the tribal classification of the leafhopper subfamily Aphrodinae (Deltocephalinae of authors) of the Holarctic region (Rhynchota: Homoptera: Cicadellidae). — Canadian Entomologist 107: 477-498. Ossiannilsson, F., 1983. The Auchenorrhyncha (Homop- tera) of Fennoscandia and Denmark. Part 3: The family Cicadellidae: Deltocephalinae. — Fauna Entomologica Scandinavica 7: 594-979. Matsumura, S., 1914. Die Jassinen und einige neue Acocephalinen Japans. — Journal of the College of Agri- culture, Tohoku Imperial University 5: 165-240. Rao, R., 1989. Descriptions of some new leafhoppers (Homoptera: Cicadellidae) with notes on some synony- mies and imperfectly known species from India. — Hexa- poda 1: 59-84. Remane, R. & M. Asche, 1980. Neue Zikaden-Taxa aus dem Tribus Paralimnini Distant, 1908, aus dem Mittelmeergebiet (Homoptera Cicadina Cicadelloidea, Cicadellidae). — Marburger Entomologische Publikatio- nen 1 (4): 67-166. Webb, M. D. & F. R. Heller, 1990. The leafhopper genus Pseupalus in the Old World Tropics, with a check-list of the Afrotropical and Oriental Paralimnini (Homoptera: Cicadellidae: Deltocephalinae). — Stuttgarter Beiträge zur Naturkunde, Serie A (Biologie), 452: 1-10. Vilbaste, J., 1969. On some East-Asiatic leafhoppers de- scribed by Professor S. Matsumura (Homoptera: Cica- dinea: lassidae). — Insecta Matsumurana, Supplement 6. 12pp. Received: 15 February 1994 Accepted: 28 April 1994 D. B. BAKER Hope Entomological Collections, University Museum, Oxford, United Kingdom. A NEW GENUS OF NOMADINE BEES FROM NORTH AFRICA (HYMENOPTERA: APOIDEA, ANDROLHORIDAE) Baker, D. B., 1994. A new genus of nomadine bees from North Africa (Hymenoptera: Apoidea, Anthoporidae). — Tijdschrift voor Entomologie 137: 155-159, figs. 1-4. [1ssN 0040- 7496]. Published 15 December 1994. Aethammobates prionogaster gen. et sp. n. Anthophoridae: Nomadinae) is described from a unique male collected in Egypt. The new genus is referred to the Holcopasitini and its relation- ships with other holcopasitine bees are discussed. D.B. Baker, Hope Entomological Collections, University Museum, Oxford ox1 3pw, U.K. Keywords. — Hymenoptera; Anthophoridae: Holcopasitini; Aethammobates prionogaster, new genus; new species; North Africa. Among a number of epeoline bees received many years ago from the late Prof. Dr. H. Priesner was a single male of a strange nomadine bee that could not be referred to any described genus. A description and figures were prepared, and the specimen was then set aside in the expectation that further specimens, in- cluding possibly the female, would sooner or later come to light. After more than 40 years, this expecta- tion has not been fulfilled, although the locality where the bee was obtained (Gebel el Asfar, near Cairo) was one much visited by both pre-War and more recent local collectors. Since the locality is now being devastated by construction works (C.G. Roche, in liet, 6 September, 1992), making the recovery of further material less likely, and since also the system- atics of the Nomadinae have recently come under re- view by several authors, notably Alexander (1990) and Roig-Alsina (1991), it seems desirable to place the new genus and species on record. Aethammobates gen. nov. Description. — Head in frontal aspect transverse, the eyes rather small, separated by 1.2 times their length, their inner margins convex, subparallel; clype- us short, carinate laterally along epistomal suture, paraocular area adjacent to carina slightly concave and narrowly impunctate; labrum elongate (length to breadth ration 1.44 : 1), apically entire, lacking discal tubercles; antennal sockets at mid-level of eyes; inter- antennal carina divided, forming a V-shaped protub- erance, the open end of the V upward; facial foveae absent; vertex longer than inter-ocellar distance; pre- occipital ridge carinate; malar area rudimentary. Antennae 13-segmented, scape long (length to breadth ratio 3 : 1); pedicel free; first flagellar segment long (length to breadth ratio 3 : 1), equal to segments 2+3; segments 2-10 passing from transverse to quad- rate. Mandibles simple, with weak basal protuberance opposed to lateral termination of clypeal carina; only posterior articulation in contact with eye. Stipital comb absent; maxillary palpi 5-segmented; blade of galea moderately, uniformly sclerotized, acuminate, extreme apex very narrowly rounded. Mesosoma broad, dorso-ventrally compressed. Pronotum with well developed dorsal surface limited anteriorly by an uninterrupted carina; anterior surface broad, flat, rounded abruptly into collum. Mesoscutum little convex; parapsidal lines well marked, about as long as tegula; axillae not protu- berant; scutellum extending laterad slightly beyond axillae, with distinct dorsal and posterior vertical sur- faces separated by a weak, irregular (interrupted by punctation) carina, slightly overhanging subvertical, coplanar surfaces of metanotum and propodeum; omaulus continuous across mesepisternum, forming prominent angles on either side of a median, semicir- cular inflexion; mesepisternum anterior to intermed- jate coxae strongly transverse, transversely concave. Metanotum with prominent lateral lobes. Wings (fig. 4) short, forewing about 0.6 body length. Forewing bare basally, becoming pubescent apically, especially beyond cells; marginal cell broad, rather broadly rounded at apex, apex separated from wing-tip by less than cell length; distal abscissa of Rs absent; two sub- marginal cells, the second much shorter than the first, the first receiving 1 m-cu near its apex, the second 2 m-cu beyond the middle. Hindwing with jugal lobe 155 [IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 very short; second abscissa of M + Cu about 1,5 times longer than cu-v, not quite half as long as M. Legs of normal proportions. Anterior coxae proximate, trans- verse, the trochanters widely separated; tarsus sub-bi- pectinate, basitarsus and segments two to four apical- ly each with a few strong setae on either side; basitarsus flattened and slightly expanded apicad. Intermediate coxae widely separated, short, much shorter than distance from their summits to posterior wing bases; femur broad; tibia without anterior setose area, coarsely spiculate. Posterior coxae large, broad, inwardly weakly, externally strongly carinate; tibiae coarsely spiculate. Arolia present, normally devel- oped. Metasoma broad, depressed, greatest width at apex of segment 2. Tergum 1 with vertical anterior and horizontal dorsal surfaces separated by a deeply arcu- ate carina, in dorsal aspect with prominent lateral an- gles; marginal areas of terga abruptly and progressiv- ely more strongly depressed caudad, the margins of the discal areas passing from subserrate on tergum 1 to strongly denticulate on terga 4-6; pygidial plate strong, rostriform, rounded apically, not constricted basally, considerably exceeding apical margin of seg- ment; sternum 6 laterally dentate. Vestiture generally squamiform, on the terga form- ing basal and marginal fasciae, not maculae; simple, erect hairs largely confined to mandibles and disc of labrum; anterior angles of malar areas not penicillate; eyes bare; sterna without subapical fimbriae of modi- fied hairs, but sterna 5 and 6 mesially with fine, dense, erect pubescence, on 6, towards the lateral teeth, becoming longer and directed mesad. Integument black with extensive pale areas, the ap- pendages and metasoma predominantly pale; moder- ately strongly, moderately densely to subreticulately punctate, microsculpture not evident at 37,53. Distribution. — N.E. Africa. Host. — Not known. Type species. — Aethammobates prionogaster, sp. n. Derivatio nominis. — Gr. anOns, strange, + Ammobates, nom. propr. Hymenoptera Aethammobates is separated from other nomadines by a variety of autapomorphies, but for purposes of recognition its general habitus and the modifications of the pronotum and first metasomal tergum are im- mediately diagnostic (see Discussion). Aethammobates prionogaster sp. n. figs. 1-4 [ype material. — Holotype d: ‘Coll. A. Mochi / 27.V.37 Geb. Asfar / Egitto” (in coll. Baker ) Description Male. — Structural characters: See generic descrip- tion. POL = oor. Hamuli 8. Length 7.5 mm, forewing 4.75 mm. Vestiture. The face, dorsal surface of pronotum, peripheries of mesoscutum and scutellum, mesepi- sterna, and outer ventral surfaces of posterior coxae, all more or less densely clothed with white squami- form or subsquamiform pubescence, that on the pro- notum apparently particularly dense (matted in type); elsewhere on head and mesosoma sparser, the individ- ual hairs becoming longer and less squamiform. Terga with broader, white, medially emarginate, lat- erally expanded, basal fasciae and narrow, entire, api- cal fasciae, filling the depressed marginal areas, of squamiform pubescence; tergum 5 nearly completely covered. Disc of pygidial plate with similar pubes- cence. Sterna 2-4 with weak submarginal fasciae, ster- na 5 and 6 medially with fine, dense, erect pubes- cence (as noted in generic description - possibly homologous with pre-genital brush in Nomada). Integument of head and mesosoma black with ex- tensive pale areas. Clypeus, anterior extremities of paraocular areas, malar areas, labrum, small maculae adjacent to summits of eyes, larger maculae on genal areas adjacent to mandibles, extending narrowly up- wards almost to summits of eyes, labrum and all head appendages (except the mandibles apically) castane- ous, the flagellum darker. Lateral lobes of pronotum, axillae, tegulae, sclerites of wing bases, wing veins (the costa darker), and legs (the coxae basally darker) cas- taneous. Metasoma castaneous, terga 2-5 discally pro- gressively, irregularly (possibly the result of post-mor- tem decomposition) darker; sterna 2 and 3 discally largely, 4 and 5 almost wholly, dark. Labrum smooth, moderately strongly, irrergularly punctate; anterior face of pronotum smooth, impunctate; mesoscutum and scutellum smooth, moderately densely, simply punctate; terga smooth, medially densely, laterally re- ticulately, punctate, the punctation coarsest on ter- gum 2, on tergum 1 about equal to that of mesoscu- tum, on terga 3-5 progressively finer. Margins of pygidial plate reflexed. d ; Derivatio nominis. — Gr. TPLOV — WÖNS / WTOS, serrated, + gaster. Discussion Both Alexander (1990) and Roig-Alsina (1991) have recently published cladistic analyses of nomadi- ne bees. Roig-Alsina’s analysis was based primarily (15 out of 22 characters) on characters exclusive to the female, and is consequently of limited use in the present context. Alexander’s analyses were based on (1) larval characters only, (2) on adult characters of those genera where larvae were known, (3) on both larval and adult characters of these genera, and (4) on the adult characters of all genera known to him. It is this last analysis, (4), that is relevant to the placement Figs. 1-4. Aethammobates prionogaster sp.n. 157 [1IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Table 1 Data matrix for characters as listed by Alexander, Appendix 6; codings in accordance with Alexander, Appendices 3 and 5. Characters 1 2 4 5 I dolce pasutes l 0 0 0 Schmiedeknechtia | 0 0 0 Aethammobates 0 0 1 0 12 13 14 15 Holocopasites 0 0 0 0 Schmiedeknechtia 0 0 0 0 Aethammobates 1 1 0 0 20 21 e f Holocopasites l 0 Il 0 Schmiedeknechtia 1 0 1 0 Aethammobates 1 0 1 ? Notes on matrix a 6 7 8 9 LO 1 0 0 1 0 0 0 al 0 0 0 0 0 0 0 0 = 0 1 0 2 WG as jb c ig al 19 0 1 1 0 ie 1 0 0 1 1 0 if 1 0 0 0° 0 I 1 1 1 g 22 MR 25 ANZN 2 0 0 2 1 0 1 0 0 0 2 1 0 1 0 0 0 2 ie 0 0 0° 1) character a: Antennal sockets below mid-level of eyes, but not far below; no code. — (2) character 7: ‘Length/width of first flagellomere’ = 3: no code. — (3) character 17: No ‘patches of appressed squamiform setae’: Alexander presumably refers to discrete metasomal maculations of Holocopasites and Schmiedeknechtia. — (4) character 18: Incorrectly coded ‘0’ by Alexander: vein Ist r-m of forewing is absent (two submarginal cells). — (5) character g: ‘arising from’ in Alexander’s Appendix 5 presum- ably means ‘rising from’ (i.e., in lateral aspects bends upward from ventral plane of mesosoma immediately before intermedi- ate coxae). — (6) character 23: a weak, short, dorso-basal carina present on posterior coxa. — (7) character 32: Coded ‘0’ since apex narrowly rounded, subacuminate, but galea is more than weakly sclerotized. of Aethammobates. Alexander's data matrix (his Appendix 6) employed 46 characters, of which six ap- plied solely to the female, leaving 40 in which Aethammobates could potentially be compared. In practice, since the holotype and only known speci- men of Aethammobates has not been dissected (cf. Roig-Alsina’s comment, 1991: 25, on the pre-epis- ternal sulcus in Rhopalolemma), comparison has been limited to 33 characters. These have been covered in the description given above, and are now given in co- ded form, following Alexander’s Appendices 3 and 5, as an extension of his data matrix, in Table 1. This ta- ble reproduces Alexander’s codings, but substitutes observed values for six characters marked with a quae- re in his matrix. It should be noted that Alexander’s codings, here and elsewhere, cannot always be accepted because he did not see certain described and undescribed taxa that would have necessitated different coding or the adoption of additional codes; however, a complete re- vision of his data matrix, which might well result in significant changes in his cladogram, is beyond the scope of the present paper. Important characters that should be taken into account in future work on the phylogeny and classification of the Nomadinae in- clude the presence of facial foveae in some Nomada (a group of, chiefly, Panurgus parasites); the develop- ment of an anal truncation, with associated modifica- tions of the apical sterna, among the pasitines (Bischoff 1923: 585, Analstutz): ‚and modifications of the antenna in such genera as Morgania and Pasitomachthes, where the scape and pedicel may be coalescent, forming a single functional unit (Baker 1971: 7, footnote). Inspection of the extended matrix confirms 4 prio- ri impressions that Aethammobates shares a preponde- rance of characters with Holcopasites and Schmie- deknechtia and must be placed in the Holcopasitini. While obviously most nearly related to Holcopasttes [Nearctic, revised by Linsley (1943, as Neopasites) and by Hurd and Linsley (1972)] and Schmiedeknechtia [western palaearctic, revised by Popov (1933)], and sharing such derived characters as the lateral process- es of the metanotum and certain venational details, Aethammobates is not close to either. It differs from both in, among other characters: The dorso-ventrally compressed form of mesoso- ma and metasoma. 2. The mid-dorsally exposed pronotum with sharply differentiated dorsal, densely pubescent and anter- ior, glabrous surfaces. (In lateral aspect, the prono- tum resembles that of Pae (Sphecoidea) as illus- trated by Bohart & Menke (1976: 15, fig. 3A),] In Holcopasites and Schmiedeknechtia the pronotum is, as in most other bees, recessed beneath the me- soscutum with only its lateral lobes and, when the head is deflexed, the collum conspicuous in dorsal aspect. 3. The V-shaped interantennal carina. 4. The strongly expanded intermediate femora. 5. The pronounced basal truncation of the mesoso- ma. 6. The form of the first metasomal tergum, which is broadly, deeply emarginate, with prominent ante- ro-lateral angles (see fig. 3) and the anterior and dorsal surfaces sharply, rectangularly separated. 7. The presence of basal and marginal tergal fasciae and the absence of of discrete spot- or bar-like markings. Of these, characters 1-6 appear as autapomorphies. Aethammobates differs further, from male Holcopasites, in the 13-segmented antennae, and from Schmiedeknechtia in the form of the pygidial plate, which is not narrowed basad; in not having the inner orbits convergent below; in not having the anterior mandibular articulations in contact with the eyes; and in the venation: in Schmiedeknechtia either both 1 m- cu and 2 m-cu are received by second Rs, or (Cyrtopasites, a single species with convex mesoscu- tum and fine and sparse punctation) 1 m-cu is inter- stitial with or slightly precedes 1 r-m. Biology It is interesting to speculate on the possible host of Aethammobates. Known hosts of Holcopasites and Schmiedeknechtia are panurgines. Camptopoeum and Meliturgula are possible candidates, but the former. which is normally abundant where it does occur, and which appears not to have been recorded from the Cairo area (it is not represented in recent, extensive Egyptian collections), has species of Parammobatodes as its regular parasites. The somewhat flattened shape of Aethammobates suggests some correlation with BAKER: A new genus of nomadine bees habits or host. Both sexes of Meliturgula have a cons- picuously broad or flattened metasoma: regional spe- cies are few in number and rare in collections; not- hing is known of their biology. REFERENCES Alexander, B., 1990. A cladistic analysis of the nomadine bees (Hymenoptera: Apoidea). — Systematic Entomology 5121-1522 Baker, D. B., 1971. A new Pasitomachthes from Rhodesia (Hymenoptera, Apoidea). — Novos Taxa entomológicos, no. 98: 1-8. Bischoff, H., 1923. Zur Kenntnis Schmarotzerbienen. - Deutsche Zeitschrift 1923: 585-603. Bohart, R. M. & A. S. Menke, 1976. Sphecid wasps of the world: a generic revision: ix + 695 pp. — University of California Press, Berkeley. Hurd, P. D. & E. G. Linsley, 1972. Parasitic bees of the ge- nus Holcopasites Ashmead (Hymenoptera: Apoidea). — Smithsonian Contributions to Zoology, no. 114: 1-41. Linsley, E. G., 1943. A revision of the genus Neopasites (Hymenoptera: Nomadidae). — Transactions of the American Entomological Society 69:119-140. Popov, V. B., 1933. Notes on the parasitic bees allied to the genus Biastes Panz. (Hymenoptera, Nomadidae). — Trudy zoologicheskogo Instituta Akademy Nauk SSSR 2: 91279: Roig-Alsina, A., 1991. Cladistic analysis of the Nomadinae s.str. with description of a new genus (Hymenoptera: Anthophoridae). — Journal of the Kansas Entomological Society 64: 23-37. afrikanischer entomologische Received: 7 April 1993 Accepted: 30 December 1993 159 x im houden tech os Al en aar aA EN h we nl a TE Diez | pay We vrai da jit [EL f ‚wahr af (Wash LA) diet N i i] | rar Ca) k Pr fi bar (MENTO. ty pur 1 PI I iw we ALL ELLE } i = 9 i { ' 4 ko i 4 = = Ù ' Li 4 e en N ll n i 2 j { DENT iù LL ar I) 7 Re A 1 A ) è Oe q =» 49 f 4 j aan wt Le) À , ; A N Im à if f pi ze. . id \ ia { FR | \ MA di Li 7 Br | M A me a tà € ‘ è È 8 wer. i A ha és de à bo vi | | 4 # 7 vo PR x 4 N © b n 8 de x " di hars » Mi Gi ' = + A | . 5 D 0 t in Pasi pg ctl eet mer = 4) Er © arr ig AN Wil, te Pail RITA et atlete i ira gl ank eid. der. ! vi lati e In raken Kut raf hi EEE Ania Alpini LIN yen) Ta, Jr isd gn Nieyy SL IE he Vi rss ld CART Rah è r Wir wirt n J ) Ih Lt 1 iv} i i CRUI | Î un La va Fa ent u ih DA \ qe 1. rt | lo tel gela ı TANT 183 À i NE ‘ i ‘ i u a i j ‘ Wu ‘i is RENT LITIO | et IAA A. J. DE BOER Institute for Systematics and Population Biology (Zoölogisch Museum), University of Amsterdam BOURSPEEIES ADDED TO THE BAETURIA NASUTA GROUP, WITH NOTES ON TAXONOMY AND BIOGEOGRAPHY (HOMOPTERA, TIBICINIDAE) Boer, A. J. de, 1994. Four species added to the Baeturia nasuta group, with notes on taxonomy and biogeography (Homoptera, Tibicinidae). — Tijdschrift voor Entomologie 137: 161-172, figs. 1-39. [ISSN 0040-7496]. Published 15 December 1994. A new concept is proposed for the nasuta group of the cicada genus Baeturia Stäl, 1866. Four species are added to this group, bringing the total number of its species to twelve. One of these species (B. guttulipennis Blöte, 1960) is redescribed and three (B. gibberosa, B. splendida, and B. retracta) are described as new. Implications for the biogeographic pattern of the group are, that the nasuta group can no longer be seen as a typical central New Guinea group; most of the spe- cies can apparently be found in northern New Guinea. A.J. de Boer, Institute for Systematics and Population Biology (Zoölogisch Museum), University of Amsterdam, P.O. Box 94766, 1090 GT Amsterdam, The Netherlands Key words. — Baeturia, nasuta group, taxonomy, biogeography, New Guinea. The nasuta group was recognized as a monophylet- ic group consisting of eight New Guinean species of the genus Baeturia Stäl (de Boer 1982). The mono- phyly of that group is based on two presumed apo- morphies: a narrow and angularly swollen postclype- us and four sclerotized ridges on the tymbals. 5. guttulipennis Blöte was not included in the nasuta group, since it does not share these apomorphies. However, now that most of the remaining species of Baeturia have been revised and placed in monophy- letic species groups, it appears that 2. guttulipennis is more closely related to the nasuta group than to any other species group. It is proposed here to change the concept of the nasuta group and to include B. guttu- lipennis in that group. Furthermore, three new species can be added to the nasuta group, these species do share the original apomorphies of that group. Species of the nasuta group, as previously defined, were mainly distributed in the central mountain ranges of New Guinea. Its distribution pattern has been compared to that of the cicada genus Cosmopsaltria Stal, which also centres in the montane parts of central New Guinea (de Boer 1982; Duffels 1986; Duffels & de Boer 1990) and it was suggested that the similarities between these patterns might find a common cause in the paleogeology of New Guinea. The present additions to the nasuta group greatly al- ter its general distribution pattern and necessitate a reconsideration of its paleogeographic history. MATERIAL AND METHODS The material examined for this study is deposited in the following collections: AMNH: American Museum of Natural History, New York; AMS: Australian Museum, Sydney; BMNH: Natural History Museum (formerly: British Museum (Natural History)), London; BPBM: Bernice P. Bishop Museum, Honolulu; Moul: Personal collection Mr M.S. Moulds, Sydney; RMNH: Nationaal Natuur his- torisch Museum (formerly: Rijksmuseum van Natuurlijke Historie), Leiden; sMN: Staatliches Museum für Naturkunde, Stuttgart; ZMA: Institute for Systematics and Population Biology (Zoölogisch Museum), Amsterdam. The following geographical sources have been used: Atlas van tropisch Nederland (1938), The Times Atlas of the World (1968) and the ‘List of New Guinea localities’ published by the Bishop Museum (1966). After overnight softening, male genitalia were ex- amined by pulling out the pygofer with a sharp nee- dle inserted between pygofer and 8th abdominal seg- ment. The aedeagus was pulled out at the same time, by inserting the needle between the claspers. Measurements are based on all available specimens. PHYLOGENY Baeturia belongs to a larger group of genera, the 161 ÎTIDSCHRIFT VOOR I NTOMOLOGIE, VOLUME 137, 1994 Baeturia and related genera complex’, as defined ear- lier (de Boer 1990). A phylogenetic analysis of this ge- nus complex, with the oriental Prasiini as defined by De Jong (1985) as outgroup, is in preparation. The ensuing phylogenetic discussions are based on the preliminary results of that analysis. The monophyly of the nasuta group The monophyly of the nasuta group is based on three presumed apomorphies. Males of the nasuta group are easily identified by the almost transparent lateral sides of the abdomen, usually restricted to seg- ments 3-7. Only B. marmorata does not show this character. A similar transparency of the male abdo- men was found in a monophyletic group of three spe- cies of the genus Gymnotympana (unpublished), but not in other related taxa. A laterally transparent abdo- men is regarded as apomorphous for the nasuta group (1 in fig. 1b) and the similar transparency in three Gymnotympana species must be explained by paralle- lism. A strongly curved aedeagus, with lateral lobes at its basal curve, is regarded apomorphous for the genus Baeturia. The aedeagus of the species of the nasuta nasuta group viridis group conviva group guttulinervis group bloetei group exhausta group loriae group laminifer retracta bipunctata gibberosa arabuensis mamillata marmorata parva splendida intermedia nasuta guttulipennis Cladogram of Baeturia, numbers ıssed in the section on the phyloge- nasuta group; b. Cladogram of the ‘ta group, numbers refer to characters discussed he monophyly and ingroup phylogeny of group differs from that of other species by a longer and more slender apical part, distally of the lateral lobes (fig. 11), which is regarded apomorphous (2 in fig. 1b). A very similarly elongate, but stouter, aedea- gus was found in B. loriae and B. pigrami of the lori- ae group (de Boer 1994a). The elongation of the ae- deagus in these two groups cannot be regarded as homologous in the most parsimonious solution. The male pygofer is characterized by a strongly bent and dorsally flattened caudodorsal beak. This shape of the caudodorsal beak is shared by all species and regarded apomorphous for the nasuta group (3 in fig. 1b). In other species, which have a similarly strongly bent beak, the beak is convexly curved dor- sally (comp. the viridis group, de Boer 1992). The phylogenetic position of the nasuta group Baeturia is amonophyletic genus, for which a fron- tally rounded and protruding postclypeus, a triangle- shaped medial thorn on the fore femur, and lobate la- teral crests on the aedeagus are the presumed apomorphies (1, 2, 3 in fig. la). Baeturia can be sub- divided into seven monophyletic species groups: the bloetei group (de Boer 1989), the conviva group (de Boer 1986), the exhausta group (de Boer 1994b) the guttulinervis group (de Boer 1994c), the loriae group (de Boer 1994a), the viridis group (de Boer 1992), and the nasuta group (de Boer 1982). The nasuta group is presumed to form a monophy- letic group with the conviva, guttulinervis, and viridis groups. The large, often conically-shaped, distinctly posteriorly projecting protuberance on the lateral lobe of pygofer is regarded synapomorphous for these four species groups together (4 in fig. la). Furthermore, these four groups share a very stout and strongly bent caudodorsal beak. The occurrence of a similarly bent caudodorsal beak in several Melanesian species of the bloetei group must be explained by par- allel development. The nasuta and viridis groups are sister groups, sharing an angularly bent caudodorsal beak as sup- posed synapomorphy (5 in fig. la). In the species of these two groups the dorsal margin of the pygofer is not continuously rounded with the dorsal margin of the caudodorsal beak. Furthermore, these species share a generally truncate, but sometimes bluntly rounded or bicuspidate, caudodorsal beak. Species of other groups of Baeturia generally have a more nar- rowly rounded or pointed beak, while a truncate beak only sporadically occurs. The conviva and guttuliner- vis groups are sister groups based on a shared apical swelling of the clasper (this swelling is almost globu- lar in the guttulinervis group) and a dorsally strongly rounded pygofer (6 in fig. 1a), but the relative posi- tion of the remaining species groups of Baeturia is not clear. Ingroup phylogeny The species of the nasuta group are very similar in general appearance. The species differ mainly in the shape of the claspers and some clasper characters indi- cate phylogenetic relationships. Most other differen- ces observed are either species specific characters, or characters that also occur outside the nasuta group. It is not possible to construct a fully resolved and un- ambiguous cladogram for the group, based on the li- mited number of characters now available. Nevertheless, some remarks on ingroup phylogeny can be made (see fig. 1b). B. guttulipennis presumably forms the sister taxon of all other species of the nasuta group; these other species share four tymbal ridges (three in B. parva) and a more angularly swollen postclypeus, which are the original apomorphies for the nasuta group (4, 5 in fig. 1b). B. guttulipennis and B. nasuta however, share similar brown markings on the tegmina and an api- cally rounded male caudodorsal beak, and could be sister species. B. nasuta and B. intermedia share an al- most identical clasper and are presumably either sister species or part of a paraphyletic subgroup. B. arabuensis, B. bipunctata, B. laminifer, B. ma- millata, B. retracta and B. gibberosa share a distinct dorsal protrusion on the clasper as presumed synapo- morphy (6 in fig. 1b). That B. nasuta and B. interme- dia should also be included in this group, as was sug- gested before (de Boer 1982) seems unlikely, since the weakly developed dorsal crest of the claspers of B. na- suta and B. intermedia is very similar to that found in many other Baeturia species. The remaining species of the nasuta group share a dorsally smoothly round- ed clasper. DE BOER: The Baeturia nasuta group B. arabuensis and B. mamillata share a posteriorly directed and almost finger-shaped protrusion, and are possibly sister species. B. bipunctata, B. gibberosa, B. laminifer and B. retracta have a much broader dorsal protrusion on the clasper, which is regarded as synap- omorphous (8 in fig. 1b). B. laminifer and B. retracta share a laminiform and square-shaped dorsal protru- sion on the clasper as presumed synapomorphy (9 in fig. 1b). B. bipunctata and B. gibberosa have a more triangular, hump-shaped, protrusion; these are either sister species or part of a paraphyletic subgroup. Females of most of the species that have a dorsal protrusion on the clasper, as far as they could be iden- tified, generally have a truncate caudodorsal beak, which character is unique for these species and pre- sumably apomorphous at that level (7 in fig. 1b). Only the females that presumably belong to B. ara- buensis and the females of B. gibberosa have a pointed beak as those in B. guttulipennis, B. nasuta, and B. parva. The female of B. splendida has a bluntly round- ed, almost truncate, caudodorsal beak. BIOGEOGRAPHY The nasuta group is distributed in New Guinea and some adjacent islands, some species reach to the Bismarck Archipelago or the Admiralty Islands. Study of the New Guinean cicadas indicates that the various genera or monophyletic species groups show different patterns of distribution. The genera seem concentrated with most, and often endemic, species in different parts of the island. These parts of New Guinea, thus recognized as areas of endemism (Duffels & de Boer 1990), largely coincide with frag- Pr rain Fig. 2. Distribution of Baeturia gibberosa, B. gut- tulipennis, B. retracta, and B. splendida. B guttulipennis © retracta A splendida Y gibberosa 136° 140° 144° 148° 152° TI]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 ments of a historic island arc, that collided at different times and in various places with the northern craton of the Australian continent, to form present day New Guinea (for geological reviews see Daly et al. 1991; Rangin et al. 1990a, 1990b; Pigram & Davies 1987). It is supposed that the various genera of New Guinean cicadas evolved on isolated island arc frag- ments previous to their collision with the Australian continent (de Boer 1994c). The nasuta group was regarded as a typical central New Guinea group (de Boer 1982) and, as such, has been compared to the genus Cosmopsaltria Stal (Duffels 1986; Duffels & de Boer 1990). Cosmopsaltria and the B. nasuta group both appeared to be concentrated in the central mountain ranges of New Guinea. The distributions of B. arabuensis, B. mamillata and B. nasuta, from the Wissel Lakes to well into the Papuan Peninsula, are indeed very simi- lar to that of many species of Cosmopsaltria (Duffels 1983). A similar historical explanation for the distri- bution patterns found in these two groups seemed re- quired. Such an explanation can be found in suppos- ing that both Cosmopsaltria and the nasuta group evolved on the Sepic Arc terrane as defined by Pigram & Davies (1987), which is the first of the island arc fragments that collided with the Australian continent, and which developed into the central mountain rang- es of New Guinea. However, the phylogenetic level of the groups (Cosmopsaltria as sister group of Diceropyga, Rhadinopyga, Aceropyga, and Moana (Duffels 1986, 1993), and the nasuta group as a sub- group of Baeturia) possibly indicates a considerable age difference, which contradicts to a contemporal evolution on the Sepic Arc fragment. With the four species here added to the nasuta group, this group as a whole no longer has a typical central New Guinea distribution. In fact, the number of species that occurs in northern New Guinea about equals that of central New Guinea. Only four species (B. arabuensis, B. intermedia, B. mamillata, and B. nasuta) can be said to have a central mountain range distribution, though two (B. mamillata, and B. nasu- ta) also occur in the Papuan Peninsula. The distribu- tions of B. laminifer and B. retracta are very similar; they are recorded from southern New Guinea (un- published data for /aminifer), the most eastern part of northern New Guinea, and the Papuan Peninsula. B. laminifer is recently recorded from New Britain (un- published) and B. retracta is recorded from the Admiralty Islands. These two species are absent from the western parts of the central mountain ranges. B. gibberosa is possibly restricted to the Papuan Peninsula and Normanby Island. B. splendida and, possibly, B. guttulipennis are endemic to northern (it is not certain that the female allotted pecies really belongs to B. guttulipennis). New Guin to the latter 164 Also B. parva is distributed in northern New Guinea, but its distribution extends to the Vogelkop Peninsula, Misool, Biak, the Bismarck Archipelago, and Admiralty Islands. The two remaining species (B. bipunctata and B. marmorata), both from Araucaria camp and Rattan camp on the northern slopes of the central mountains of western New Guinea, should presumably be regarded more as northern, than as central New Guinean: several other species groups with a predominantly northern New Guinea distri- bution (e.g. the viridis group and the guttulinervis group) also have species in that area. The distribution pattern of the nasuta group in its present concept is more in agreement with that of other groups of Baeturia, most of which seem to con- centrate in northern New Guinea, while similarities with Cosmopsaltria have become less obvious and a paleobiogeographical comparison with that genus seems no longer relevant. ‘TAXONOMY Description of the zasuta group Species of the nasuta group are all very small sized. The specimens are shorter than 2.5 cm and generally do not exceed 2.0 cm. Males are easily recognized by their laterally transparent abdomen (this character is absent in B. marmorata). Body generally yellowish brown or reddish brown coloured and covered with many irregularly shaped brown speckles. Females ge- nerally darker coloured and more densely speckled (B. splendida unspeckled with greenish head and tho- rax). Head (fig. 3a): Narrow, 2.3-2.8 x as wide as long. Vertex narrower than in most other Baeturia species, ocelli close together. Distance between lateral ocelli less than 2 x as wide as, and sometimes narrower than, frontal ocellus, and generally about as long (0.7-1.2 x) as distance between lateral ocellus and eye. Vertex 1.6-2.5 x as wide as postclypeus. Postclypeus in dorsal view narrow (1.1-2.2 x as wide as long) and triangu- larly protruding beyond vertex lobes, though with rounded anterior margin. Postclypeus distinctly swol- len ventrally, its anterior margin (lateral view) often angularly bent (fig. 3b). Legs: Fore femur (Fig. 6) with row of three erect and sharply pointed spines, diminishing in length to- wards tibia. The most distal spine often bicuspidate, forming a very small 4th spine. Most proximal spine slightly shorter than distance to middle spine. Middle spine triangular, very broad at its base. Tegmina and wings: Hyaline, though tegmina sometimes with darkly suffused spots along veins. Tegmina with 8, wings with 6 apical areas. Hyaline border along hind margin of tegmen distinct, though narrow. This border slightly broader in wing. Veins ochraceous, sometimes tinged with red, and set with many long setae. Tymbals: With 3-5, generally 4, parallel sclerotized ridges from dorsal to ventral tymbal margin. Often traces of red between dorsal parts of tymbal ridges. Opercula: Male operculum fairly small, its distal part not covering tymbal cavity in ventral view and nearly always shorter than meracanthus, generally not reaching to anterior margin of 2nd abdominal seg- ment. Operculum in ventral view reaching to under, or medially of, meracanthus. Female operculum smaller than that of male, sickle-shaped and erect. Abdomen: Male abdomen conspicuous by almost transparent sides, its dorsal and distal parts darkened and brown speckled, ventral parts unspeckled. Lateroventral row of dark spots generally very dis- tinct. Hind margins of abdominal segments often reddish. First and second sternites adjacent. First ter- gite very short and often partly hidden under metano- tum (in B. arabuensis distinctly longer than in other species). Second tergite medially with almost straight anterior margin. Auditory capules well developed, globularly swollen. Female abdomen more robust than that of male, darker coloured, and more densely speckled. Ovipositor sheaths sometimes reaching be- yond apex of caudodorsal beak. Female caudodorsal beak often truncate at apex. Male genitalia: Caudodorsal beak strongly bent, making an almost right angle with dorsal margin of pygofer. Dorsal margin of pygofer not continuously rounded into dorsal margin of beak, and weakly con- cave near apex of beak. Caudodorsal beak very broad and straight, dorsally flattened, and with bluntly rounded or truncate, often weakly concave, apex. Lateral lobe of pygofer with strongly swollen, often conically shaped, protuberance, projecting posterior- ly beyond margin of pygofer. Claspers fairly long, parallel, and generally directed posteriad. Apical part of clasper hardly bent down, with large and sharply edged ventral hollow. Aedeagus S-curved with lateral lobes at basal curve. Apical part of aedeagus extreme- ly long and slender (compared to other species of Baeturia). Aedeagal pore narrow and oval. Baeturia guttulipennis Blöte (figs. 2-13) Baeturia guttulipennis Blöte, 1960: 69, fig. 15; de Boer, 1994c. Material examined: Bewani R. territ., 1200 m, 1939, W. Stüber, d holotype, RMNH; Bivak 39A, Star Range, 1150 m, 6.vii.1959, 12 Baeturia guttulipennis? det. H.C. Blöte, RMNH. B. guttulipennis is easily separated from other spe- DE BOER: The Baeturia nasuta group 2%) \ S N Figs. 3-6. Baeturia guttulipennis Blöte, 1960. — 3a, head in dorsal view, holotype; 3b, head in lateral view, holotype; 4, female caudodorsal beak in dorsal view; 5, female genital segment in lateral view; 6, fore femur, holotype. cies of the nasuta group by the dark spots along the veins of tegmina and by its five tymbal ridges. Similar dark spots in tegmina were found in several other spe- cies of Baeturia, e.g. B. guttulinervis Blöte (de Boer 1994c), B. pigrami De Boer and B. silveri De Boer (de Boer 1994a), and, though less distinct, in B. nasuta Blöte (de Boer 1982). Description Body of male light ochraceous brown and weakly speckled with brown. Female darker, red-brown and more densely speckled. Female head and thorax larger than in male. Tegmina of males 1.2 x as long as body length, of females 1.5 x. Male abdomen 1.3 x as long as head and thorax, of females 1.1 x. Head (fig. 3a): Ochraceous, brown-stained mainly on vertex lobes. Postclypeus bluntly rounded anteri- orly, weakly protruding in male, more strongly in fe- male. Postclypeus 1.4-2.0 x as wide as long. Postclypeus in lateral view (fig. 3b) distinctly swollen, anterior margin forming an almost right angle at 1/2 its length, and concave to anteclypeus. Thorax: Pronotum ochraceous with dark speckling predominantly in medial band and on medial parts of 165 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 20 21 Figs. 7- 1 3. Baeturia guttulipennis Blöte, 1960: 7, pygofer in lateral view, holotype; 8, clasper, holotype; 9, male operculum, holotype; 10, male caudodorsal beak in dorsal view, holotype; 11, aedeagus from behind, holotype; 12, aedeagus in lateral view, holotype; 13, female operculum. 14-22. Baeturia splendida sp. n.: 14, pygofer in lateral view; 15, male caudodorsal en i a Boy, 16, clasper; 17, male operculum; 18, head in lateral view; 19, female operculum; 20, female caudodorsal pains 0) a Pte aen in lateral view; 22, aedeagus from behind. Lettering: bp = basal part of operculum; cb = cau- ol a = Gd = wei margin of operculum; di = = distal margin of pygofer; dmc = distomedial corner of operculum; do = dorsal margin of pygofer; dp = distal part of operculum; he = clasper heel; ho = clasper hollow; | = lateral margin of opercu- lum; m = medial margin of ope pe n = medial margin of operculum; p = protuberance on lateral lobe of pygofer; ve = ventral margin of pygofer. 166 pronotal lobes, unspeckled in and along its fissures. Mesonotum greenish brown, weakly brown speckled in male, but densely speckled in female. Male with two brown spots in front of cruciform elevation. Legs: Ochraceous, often reddish towards tarsi, and brown speckled. Fore femur (Fig. 6) with row of three or four sharply pointed and erect spines. Tegmina and wings: Tegmina with dark brown markings concentrated along veins. Wings hyaline. Veins ochraceous brown, densely set with setae. Tymbal organ: Five darkly sclerotized parallel ridg- es spanning the tymbal from dorsal to ventral margin. Four intercalary ridges form a lateral band across tym- bal. Opercula: Male operculum (fig. 9) very small and not covering tymbal cavity. Distal part of operculum erect and reaching to about 2/3 the length of mera- canthus, but not reaching to abdomen. Lateral mar- gin of distal part short, concavely bent into crest around distolateral corner of basal part, and convexly bent into long and almost straight distal margin. Distomedial corner rounded, medial margin straight. Female operculum (fig. 13) almost as large as that of male, broadly sickle-shaped and curved to abdomen, with almost continuously rounded distal margin. Abdomen: Male abdomen weakly inflated, yellow brown and brown speckled, though ventrally un- speckled. Tergites 3-6 almost transparent laterally. A distinct lateroventral row of brown spots on tergites 3-7. First tergite very short and medially partly hid- den under metanotum. Female abdomen more ro- bust than that of male, dark reddish brown, and densely speckled. Lateroventral spots on tergites 3-7 almost lost in speckling. Ovipositor sheaths not reaching to apex of caudodorsal beak (fig. 5). Female caudodorsal beak in dorsal view (fig. 4) sharply point- ed at apex. Male genitalia: Pygofer in lateral view as in fig. 7. Dorsal margin weakly convex, strongly bent into al- most straight caudodorsal beak. Distal margin con- cave, gradually bent into straight margin of beak. Protuberance on lateral lobe of pygofer rounded and conically projecting posteriad. Pygofer lobe forming an angular corner just above this protuberance. Ventral margin almost straight, but angularly bent just under lateral protuberance. Ventral margins con- verge to sharp angle at base of pygofer opening. Caudodorsal beak in dorsal view (fig. 7) rounded at apex. Clasper (fig. 8) almost straight and directed posteriad, with long and slender, dorsally smoothly rounded, apical part. Basal part of clasper with weak- ly developed clasper heel. Clasper hollow fairly long. Aedeagus (fig. 12) with very long and slender apical part and well developed lateral lobes. Aedeagus from behind (fig. 11) very slender, aedeagal pore narrow, oval shaped and truncate at apex. De BOER: The Baeturia nasuta group Measurements: Body length d : 18.6 mm, 2: 18.2 mm; tegmen length d: 22.9 mm, 2: 28.0 mm; head length d: 1.4 mm, 2: 1.8 mm; pronotum length d: 2.4 mm, 2: 2.8 mm; mesonotum length d: 3.8 mm, 2: 4.8 mm; head width d: 3.7 mm, 9: 4.3 mm; width of pronotal collar d: 5.3 mm, 9: 6.5 mm. Distribution (fig. 2). — The male holotype of B. guttulipennis comes from Bewani in northeastern Irian Jaya; a female that possibly belongs to the spe- cies comes from the Star Range in central eastern Irian Jaya. Baeturia splendida sp. n. (figs. 2, 14-22) Holotype male: ‘NEW GUINEA NE / Torricelli Mts. / Mokai Vill. 750 m / xii-16-31-1958° (print); “W.W. Brandt / Collector / Bishop’ [print], BPBM. — Paratypes: same data as holotype 59, BPBM; same data but 8-15.xii.1958, 39; 1- 23.1.1959, 19, all BPBM. Only one male and several females are known of this species. B. splendida is easily recognized by its greenish colour, with a pattern of brown markings. Due to this colour pattern the females too are easily identified. The colour pattern is unique within the nasuta group. Description Head and thorax olive green, abdomen yellowish brown. Tegmina of male 1.2 x as long as body length, of females 1.3-1.4 x. Male abdomen 1.2 x as long as head and thorax, of females 0.9-1.1 x. Head: Olive-green. Vertex lobe with dark brown spot between eye and lateral ocellus. Postclypeus dis- tinctly protruding, bluntly rounded anteriorly, and 1.1-1.8 x as wide as long. Postclypeus in lateral view (fig. 18) distinctly and angularly swollen, anteriorly almost globularly inflated. Thorax: Pronotum greenish and unmarked. Mesonotum green with two large semi-oval parame- dian brown spots at pronotal margin, reaching to about half-length of mesonotum. No dark spots in front of cruciform elevation. Legs: Greenish, yellow-brown towards tarsi. Fore femur with row of four sharply pointed and erect spines. Tegmina and wings: Hyaline. Veins greenish, and set with setae. Tymbal organ: Four weakly sclerotized parallel ridges spanning the tymbal from dorsal to ventral margin. Three intercalary ridges form a lateral band across tymbal. Opercula: Male operculum (fig. 17) very small and not covering tymbal cavity. Distal part of operculum erect and reaching to about half-length of meracan- 167 TuDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 168 thus, not reaching to abdomen. Lateral margin of dis- tal part short, concavely bent into crest around disto- lateral corner of basal part, and convexly bent into long and almost straight distal margin. Distomedial corner rounded, medial margin straight. Female op- erculum (fig. 19) almost as large as that of male, broadly sickle-shaped and erect, with almost continu- ously rounded distal margin. Abdomen: Male abdomen hardly inflated, yellow brown, darker brown in distal part. Third and 4th tergites almost transparent laterally. A distinct latero- ventral brown spot on 3rd tergite and a slightly small- er lateral spot on 8th tergite. First tergite very short and medially partly hidden under metanotum. Female abdomen more robust than that of male, light brown, with a dark brown lateroventral spot on 3rd tergite. Ovipositor sheaths not reaching to apex of caudodorsal beak. Female caudodorsal beak in dorsal view (fig. 20) weakly convex, almost truncate, at apex. Male genitalia: Pygofer in lateral view as in fig. 14. Dorsal margin weakly convex, angularly bent into al- most straight caudodorsal beak. Distal margin con- cave, almost continuously bent with margin of beak. Protuberance on lateral lobe of pygofer bluntly rounded and weakly projecting posteriad. Pygofer lobe forming an angular corner just above this protu- berance. Ventral margin angularly bent at about 1/3 its length and weakly concave towards base. Ventral margins converge to sharp angle at base of pygofer opening. Caudodorsal beak in dorsal view (fig. 15) al- most truncate at apex, but ending in very small and sharply pointed medial thorn. Clasper (fig. 17) curv- ing upwards with long and slender, dorsally smoothly rounded and recurving, apical part. Basal part of clasper without clasper heel. Clasper hollow fairly long. Aedeagus (fig. 21) with very long and slender apical part. Basal part of aedeagus with small and an- gular dorsolateral protuberances. Lateral lobes small, hardly projecting in lateral view. Aedeagus from be- hind (fig. 22) very slender. Aedeagal pore narrow, oval-shaped and truncate at apex. Measurements: Body length d: 16.5 mm, 9: 16.0-19.1 mm (x 18.0 mm + 0.9); tegmen length d: 20.6 mm, 2: 22.0-25.5 mm (x 24.6 mm + 1.0); head length d: 1.5 mm, 9: 1.6-1.9 mm (x 1.8 mm); pro- notum length d: 2.3 mm, 2: 2.8-3.2 mm (x 2.9 mm); mesonotum length d: 4.0 mm, 2: 4.0-5.2 mm (x4.8 mm); head width d: 3.3 mm, 2: 4.0-4.8 mm (&4.6 mm); width of pronotal collar &: 5.1 mm, 2: DE BOER: The Baeturia nasuta group 5.7-7.0 mm (X 6.6 mm). Distribution (fig. 2). — B. splendida is only known from the Torricelli mountains in northern Papua New Guinea. Etymology. — Splendidus (Latin) refers to the beau- tiful colour pattern of this species. Baeturia gibberosa sp. n. (figs. 2, 23-31) Holotype: ‘New Guinea SE / Popondetta / 25 m vi.66° (- print); ‘Shanahan- Lippert / Light Trap / Bishop Museum’ (print), d, BPBM. — Paratypes: NEW GUINEA: Popondetta, 60 m, 1-4.ix.1963, J. Sedlacek, 2d, BPBM; Popondetta, Inbora Plant’n, 23.1x.1963, P. Shanahan, 26, BPBM; NORMANBY: Wakaiuna, Sewa Bay, 5-9.x1.1956, W.W. Brandt, 19, BPBM; same data but 1-10.xi1.1956, 19; 11-20.x11.1956, 19, both BPBM; Wamula, 400-500 m, 31.xii.1988, R. de Keyzer, 36, Moul. B. gibberosa is easily recognized by the rectangular dorsal corner of its clasper. Description Body ochraceous brown, sometimes tinged with red, and densely speckled with brown. Tegmina of males 1.1-1.2 x as long as body length, of females 1.3 x. Male abdomen 1.3-1.5 x as long as head and tho- rax, of females 1.0 x. Head: Ochraceous, brown speckled. Postclypeus weakly protruding and bluntly rounded anteriorly, 1.7-2.2 x as wide as long. Postclypeus in lateral view distinctly swollen with straight anterior margin, con- cave towards anteclypeus. Thorax: Pronotum ochraceous, brown speckled. Mesonotum greenish brown, slightly brown speckled. Two dark spots in front of cruciform elevation. Legs: Ochraceous and weakly speckled. Fore femur with row of four sharply pointed and erect spines. Tegmina and wings: Hyaline. Veins ochraceous, moderately set with setae. Tymbal organ: Four weakly sclerotized parallel ridges spanning the tymbal from dorsal to ventral margin. Three intercalary ridges form a lateral band across tymbal. Opercula: Male operculum (fig. 30) small and on- ly partly covering tymbal cavity. Distal part of oper- culum oval shaped, slightly erect and reaching to about 2/3 of length of meracanthus, in some speci- mens reaching to anterior margin of 2nd abdominal Figs. 23-31. Baeturia gibberosa sp. n. — 23, pygofer in lateral view; 24, pygofer from aslant; 25, clasper; 26, female caudodor- sal beak in dorsal view; 27, aedeagus from behind; 28, aedeagus in lateral view; 29, male caudodorsal beak in dorsal view; 30, male operculum; 31, female operculum Figs. 32-39. Baeturia retracta sp. n. — 32, female caudodorsal beak in dorsal view; 33, pygofer in lateral view; 34, aedeagus from behind; 35, aedeagus in lateral view; 36, pygofer from aslant; 37, male caudodorsal beak in dorsal view; 38, female op- erculum; 39, male operculum. 169 [i] DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 segment. Lateral margin of distal part fairly long and directed mesiad, concavely bent into crest around dis- tolateral corner of basal part, and forming an obtuse angle wich straight distal margin. Medial part of oper- culum angularly rounded, extending medially of mer- acanthus. Female operculum (fig. 31) short, sickle- shaped and erect. Abdomen: Male abdomen not inflated, ochraceous brown and brown speckled, though ventrally un- speckled. Segmental hind margins reddish. Tergites 3-6 almost transparent laterally. A distinct lateroven- tral row of brown spots on tergites 3-8. First tergite very short and medially partly hidden under metano- tum. Female abdomen red-brown and densely speck- led. Female caudodorsal beak (fig. 26) sharply point- ed at apex. Male genitalia: Pygofer in lateral view as in fig. 23. Dorsal margin weakly concave, but convexly bent into almost straight caudodorsal beak and concave near apex of beak. Distal margin concave, and continuous- ly bent into margin of beak. Protuberance on lateral lobe of pygofer bluntly rounded and weakly projecting posteriad. Pygofer lobe forming a small angular corner just above this protuberance. Ventral margin almost straight, but convexly bent just under lateral protuber- ance. Ventral margins converge to sharp angle at base of pygofer opening (fig. 24). Caudodorsal beak in dor- sal view (fig. 29) truncate at apex. Clasper (fig. 25) al- most straight and directed posteriad. Apical part of clasper forming a distinct and angularly protruding dorsal crest, giving the clasper a humped appearance. Clasper hollow fairly short. Basal part of clasper grad- ually widening, not forming an angular clasper heel. Aedeagus (fig. 27) with fairly long and slender apical part, but much shorter than in most other species of this group, and well developed lateral lobes. Aedeagus from behind (fig. 28) very slender, aedeagal pore nar- row, oval-shaped and truncate at apex. Measurements: Body length d: 14.7-17.8 mm (x 15.9 mm + 1.0), 2: 16.5-17.7 mm (& 17.2 mm + 0.5); regmen length d: 17.1-19.1 mm («18.1 mm + 0.8), 2: 21.9-22.3 mm (x 22.1 mm + 0.9); head length d: 1.2-1.5 mm (x 1.3 mm), 2: 1.5-1.9 mm (x 1.7 mm); pronotum length d: 1.9-2.3 mm (x 2.1 mm), 2: 2.6-2.9 mm (x2.8 mm); mesonotum length 3: 2.9-3.5 mm (x 3.2 mm), 2: 4.1-4.6 mm (x 4.4 mm); head width d: 2.5-3.7 mm (x3.3 mm), 2: 4.0- 4.3 mm (%4.2 mm); width of pronotal collar d: 4.3- 5.2 mm (x4.6 mm), 9: 5.9-6.4 mm (x6.2 mm). Distribution (fig. 2). — B. gibberosa is recorded from Popondetta on the Papuan Peninsula and from Normanby Island. Etymology. — Gibberosus (Latin) humpbacked re- fers to the hump-shaped dorsal protrusion of the clasper. 170 Baeturia retracta sp. n. (figs. 2, 32-39) Holotype: ‘PAPUA NEW GUINEA / Tabubil, Western Province / 5°15’ S 140° 13 E / 13 oct. 1992 / R.B. Lachlan’ (print, day and month written); ‘On loan from / M.S. Moulds / Sydney’ [print, violet label], 5, AMS). — Paratypes: same data as holotype but 17.x.1992, 1d, Moul; same data but 29.xi.1992, 16, Moul; Garaina, 11-14.vii.1969, J.L. Gressitt, 1d, BPBM; Gurakor, Wampit R. Valley, 45 mi from Lae, 670 m, 8.v.1959, L.J. Brass, Sixth Archbold Exped. to Papua New Guinea, 19, AMNH; Gurakor, 7.vii.1965, H. Pyka, 1d, SMN; Iongai, 1700-1900 m, 9.xi.1965, J. & M. Sedlacek, 19, BPBM; Iongai, 10 km E of Mt. Albert Edward, 1450-1800 m, 8-10.xi.1965, J. Sedlacek, 16, BPBM; Kiunga, 18.vii.1970, O.K. McCaw, 16, AMS; Kiunga, 5° 15’ S 141° 05° E, 2280 m, 25.v.1970, O.K. McCaw, 19, Moul; Kiunga, Fly River, 35 m, vili.1969, J. Sedlacek, 19, BPBM; Koitaki, 1500 ft, x- xi.1928, Pemberton, 14, BPBM; Wau, Morobe Distr., 1050- 1100 m, 15.xii.1961, J., J.H. & M. Sedlacek, 16, BPBM; Yagaum, Madang, 3-5.vi.1965, H. Pyka, 1 6 , SMN; MANUS: Manus isl., 24.1.1966, 1d, SMN. B. retracta can be recognized by its erect and re- tracted claspers, which lie for the greater part con- cealed within the pygofer. The species is closely relat- ed to B. laminifer Blöte, sharing a very similar clasper with a similarly square, laminiform, dorsal protru- sion. This protrusion, however, is completely con- cealed within the pygofer, and therefore hardly visible in B. retracta. B. retracta is regarded as a separate spe- cies, since the retraction of the clasper appears to be a constant character. Furthermore, B. retracta tends to have a more strongly swollen postelypeus, while males have a slightly shorter operculum and a shorter, more rounded, protuberance on the lateral lobe of the py- gofer. Description Body of males light yellow-brown or greyish brown and distinctly speckled with brown. Females more red-brown and more densely speckled. Female head and thorax larger than in male. Tegmina of males 1.1- 1.3 x as long as body length, of females 1.2-1.4 x. Male abdomen 1.2-1.5 x as long as head and thorax, of females 1.0-1.3 x. Head: Ochraceous, densely brown speckled. Postclypeus weakly protruding, almost oblong- shaped and bluntly rounded anteriorly. Postclypeus 1.4-2.0 x as wide as long. Postclypeus in lateral view distinctly swollen with strongly convex anterior mar- gin. Thorax: Pronotum ochraceous or greyish brown, with brown speckling concentrated in, often some- what reddish, medial band and on medial parts of pronotal lobes. Mesonotum greenish brown, densely brown speckled. Two brown spots in front of cruci- form elevation. Legs: Ochraceous, often reddish towards tarsi, and weakly speckled. Fore femur with row of three or four sharply pointed and erect spines. Tegmina and wings: Hyaline. Veins ochraceous, and densely set with setae. Tymbal organ: Four weakly sclerotized parallel ridges spanning the tymbal from dorsal to ventral margin. Three intercalary ridges form a lateral band across tymbal. Opercula: Male operculum (fig. 39) very small and not covering tymbal cavity. Distal part of operculum erect and reaching to about half-length of meracan- thus, but not reaching to abdomen. Lateral margin of distal part short, concavely bent into crest around dis- tolateral corner of basal part, and angularly bent into long and concave distal margin. Distomedial corner rounded, medial margin straight. Male from longai with larger, more oval-shaped, operculum. Female operculum (fig. 38) almost as large as that of male, broadly sickle-shaped and erect, with almost continu- ously rounded distal margin. Abdomen: Male abdomen not inflated, ochraceous brown and brown speckled, though ventrally un- speckled. Segmental hind margins sometimes red- dish. Tergites 3-6 almost transparent laterally. A dis- tinct lateroventral row of brown spots on tergites 3-8. First tergite very short and medially almost complete- ly hidden under metanotum. Female abdomen more robust than that of male, dark reddish brown, and densely speckled. Lateroventral spots on tergites 3-8 very distinct. Ovipositor sheaths not reaching to apex of caudodorsal beak. Female caudodorsal beak in dor- sal view (fig. 32) truncate at apex (pointed in Iongai specimen). Male genitalia: Pygofer in lateral view as in fig. 33. Dorsal margin concave, but convexly bent into al- most straight caudodorsal beak and concave near apex of beak. Distal margin almost straight, but concavely bent into straight margin of beak. Protuberance on lateral lobe of pygofer bluntly rounded and weakly projecting posteriad. Pygofer lobe sometimes forming a small rounded corner just above this protuberance. Ventral margin almost straight, but convexly bent just under lateral protuberance and weakly concave near base. Ventral margins converge to sharp angle at base of pygofer opening (fig. 36). Caudodorsal beak in dorsal view (fig. 37) weakly bicuspidate at apex. Clasper strongly upcurved, reaching into basal part of caudodorsal beak. Dorsal part of clasper with lamini- form and square-shaped protrusion. Clasper hollow fairly short. Aedeagus (fig. 35) with long apical part, but shorter and broader than in most other species of this group, and with well developed lateral lobes. Basal part of aedeagus very stout. Aedeagus from be- hind (fig. 34) long and slender, though fairly broad between its lateral lobes. Aedeagal pore narrow, oval- DE Boer: The Baeturia nasuta group shaped and truncate at apex. Measurements: Body length d: 14.9-19.6 mm (x 17.2 mm + 1.4), 2: 16.3-18.1 mm (x 17.5 mm + 0.7); tegmen length d: 18.2-21.9 mm (x 20.6 mm + 1.1), 2: 21.4-25.9 mm (x 23.0 mm + 1.7); head length d: 1.3-1.5 mm (x1.4 mm), 9: 1.6-1.8 mm (x 1.7 mm); pronotum length d: 2.1-2.5 mm (x 2.2 mm), 2: 2.4-2.7 mm (x2.6 mm); mesonotum length d: 3.2-4.1 mm (x 3.7 mm), 9: 3.9-4.8 mm (x 4.2 mm); head width d: 3.4-3.8 mm (x3.7 mm), 9 : 3.9- 4.2 mm (x 4.1 mm); width of pronotal collar d: 4.7- 5.6 mm (x5.1 mm), 9: 5.7-6.3 mm (x5.9 mm). Distribution (fig. 2). — B. retracta is widely distrib- uted in Papua New Guinea, and also recorded from Manus Island of the Admiralty Islands. Etymology. — Retracta refers to the retracted clasp- ers of this species. ACKNOWLEDGEMENTS For the loan of material I am indebted to: Dr R.T. Schuh (AMNH); Mr. B.J. Day (AMS); Dr W.J. Knight and Mr. M.D. Webb (BMNH); Mr. G.M. Nishida and Mr. K. Arakaki (BPBM); Mr. J. van Tol (RMNH); Mr. F. Heller (SMN); and to Mr. M.S. Moulds for the loan of specimens from his private collection. I would like to thank Mr. G. Verlaan for technical assistance and and Mr. D.A. Langerak for preparing the map (fig. 2). I am indebted to Prof. Dr F.R. Schram and Dr J.P. Duffels (Institute for Systematics and Population Biology, University of Amsterdam) for their critical reading and comments on the manu- script. REFERENCES Anonymous, 1938. Atlas van tropisch Nederland: [i-ix], maps 1-31b, legenda 1-31b, 1-17 (Koninklijk Neder- lands Aardrijkskundig Genootschap & Topografische Dienst in Nederlandsch-Indié, Amsterdam/Batavia). Anonymous, 1966. List of New Guinea localities: 1-19 (Bernice P. Bishop Museum, Honolulu). Anonymous, 1968. The Times atlas of the world: i-xliii, 1- 272, pls. 1-123. (Times Newspapers Ltd., London). Blöte, H. C., 1960. The genus Baeturia Stal as represented in New Guinea (Homoptera, Cicadidae). — Zoologische Mededelingen, Leiden 37: 61-80. Boer, A. J. de, 1982. The taxonomy and biogeography of the nasuta group of the genus Baeturia Stal, 1866 (Homoptera, Tibicinidae). — Beaufortia 32 (4): 57-78. Boer, A. J. de, 1986. The taxonomy and biogeography of the conviva group of the genus Baeturia Stal, 1866 (Homoptera, Tibicinidae). — Beaufortia 36 (7): 167-182. Boer, A. J. de, 1989. The taxonomy and biogeography of the bloetei group of the genus Baeturia Stal, 1866 (Homoptera, Tibicinidae). — Beaufortia 39 (1): 1-43. Boer, A. J. de, 1990. Aedeastria, a new cicada genus from New Guinea, its phylogeny and _ biogeography 171 [I] DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Homoptera, Tibicinidae), preceded by a discussion on the taxonomy of the New Guinean Tibicinidae. — Beaufortia 40 (3): 63-72. Boer, A. J. de, 1992, The taxonomy and biogeography of the group of the Baeturia Stal, 1866 Homoptera, Tibicinidae). — Bijdragen tot de Dierkunde 61 (3): 163-183. Boer, A. J. de, 1994a. The taxonomy and biogeography of the /oriae group of the genus Baeturia Stal, 1866 Homoptera, Tibicinidae). — Tijdschrift voor Entomolo- gie 137: 1-26. Boer, A. J. de, 1994b. The taxonomy and biogeography of the exhausta group of the genus Baeturia Stäl, 1866 (Homoptera, Tibicinidae). — Beaufortia (in press). Boer, A. J. de, 1994c. The taxonomy and biogeography of the guttulinervis group of the genus Baeturia Stal, 1866 (Homoptera, Tibicinidae). — Bijdragen tot de Dierkunde 64: 87-100. Daly, M. C., M. A. Cooper, I. Wilson, D. G. Smith, & B. G. D. Hooper, 1991. Cenozoic plate tectonics and basin ev- olution in Indonesia. — Marine and Petroleum Geology (MPGD) 8 (1): 1-21. Duffels, J. P., 1983. Taxonomy, phylogeny and biogeogra- phy of the genus Cosmopsaltria, with remarks on the his- toric biogeography of the subtribe Cosmopsaltiaria (Homoptera: Cicadidae). — Pacific Insects Monographs 39: 1-127. Duffels, J. P., 1986. Biogeography of Indopacific Cicadoidea, a tentative recognition of areas of endemism. — Cladistics 2: 318-336. Duffels, J. P., 1993. The systematic position of Moana ex- pansa (Homoptera: Cicadidae), with reference to sound organs and the higher classification of the superfamily VITIAIS genus Cicadoidea. — Journal of Natural History 27: 1223-1237. Duffels, J. P. & A. J. de Boer, 1990. Areas of endemism and composite areas in East Malesia. In: P. Baas, C. Kalkman & R. Geesink (eds.), The plant diversity of East Malesia: Proceedings of the Flora Malesiana Symposium com- memorating Professor Dr. C.G.G.J. van Steenis, Leiden, August 1989: 249-272 (Kluwer Academic Publishers, Dordrecht). Duffels, J. P. & P. A. van der Laan, 1985. Catalogue of the Cicadoidea (Homoptera, Auchenorhyncha) 1956-1980. — Series Entomologica 33: i-xvi, 1-414. Jong, M.R. de, 1985. Taxonomy and biogeography of orien- tal Prasiini 1: The genus Prasia Stäl, 1863 (Homoptera, Tibicinidae). — Tijdschrift voor Entomologie 128: 165- 191. Pigram, C. J., & H. L. Davies, 1987. Terranes and the accre- tion history of the New Guinea orogen. B. M. R. Journal of Australian Geology & Geophysics 10: 193-212. Rangin, G., L. Jolivet, & M. Pubellier, 1990a. A simple model for the tectonic evolution of the southeast Asia and Indonesia region for the past 43 my. — Bulletin de la Société géologique de France (8) 6: 889-905. Rangin, G., M. Pubellier, J. Azema, A. Briais, P. Chotin, H. Fontaine, P. Huchon, L. Jolivet, R. Maury, C. Muller, J.- P. Rampnoux, J.-F. Stephan, J. Tournon, N. Cottereau, J. Dercourt, & L. E. Ricou, 1990b. The quest for Tethys in the western Pacific. 8 paleogeodynamic maps for Cenozoic time. — Bulletin de la Société géologique de France (8) 6 (6): 907-913 Received: 13 January 1994 Accepted: 11 April 1994 ROSSER W. GARRISON ., Los Angeles County Agricultural Commissioner Office, U.S.A. A REVISION OF THE NEW WORLD GENUS ERPETOGOMPHUS HAGEN IN SELYS (ODONATA: GOMPHIDAE) Garrison, R. W., 1994. A revision of the New World genus Erpetogomphus Hagen in Selys (Odonata: Gomphidae). — Tijdschrift voor Entomologie 137: 173-269, figs. 1-230, tabs. 1-4 [issn 0040-7496]. Published 15 December 1994. This revision of adult Erpetogomphus includes a phylogenetic assessment of all 21 species using outgroup comparison and parsimony algorithm, descriptive biogeography, keys to both sexes, synonymies, descriptions, type designations, and illustrations, including distribution maps of all species. Six new species are described: £. agkistrodon, E. leptophis, E. elaphe, E. liopeltis, E. bothrops, E. heterodon. Erpetogomphus coluber is considered a junior synonym of £. compositus, E. natrix is considered a subspecies of E. lampropeltis, and a neotype is designated for £. cophias. Phylogenetic assessment of 41 mostly somatic characters shows Erpetogomphus to be partitioned into three monophyletic groups: 1) six dark green species (E. constrictor, E. sabaleticus, E. tristani, E. agkistro- don, E. schausi, E. ophibolus) with mostly allopatric or parapatric distributions along the eastern coast of Mexico south into northern Colombia and Venezuela, 2) two species (E. leptophis, E. eu- tainia) with distributions from southern Texas south through Mexico and up the west coast to Michoacan states, and 3) 13 remaining species (E. elaphe, E. elaps, E. liopeltis, E. bothrops, E. vi- perinus, E. designatus, E. sipedon, E. lampropeltis, E. crotalinus, E. heterodon, E. compositus, E. boa, E. cophias) with distributions in the central United States south through Mexico to Costa Rica. Derived characters states were gleaned mostly from primary and secondary genitalic characters (head structure, penis, hamules, caudal appendages, vulvar lamina); but satisfactory resolution of terminal clades, especially of the third group, is difficult due to apparent character reversals. Dr. R. W. Garrison, Los Angeles County Agricultural Commissioner’s Office, 3400 La Madera Ave., El Monte, CA 91732, USA Key words. — Odonata; Gomphidae; Erpetogomphus systematics; keys; cladistics; phylogeny; biogeography. CONTENTS Erpetogomphus agkistrodon sp. n. nnn 199 Erpetogomphus schausi Calvert .................... 200 Introductory part Erpetogomphus eutainia Calvert .................. 201 litroductionge nant... user 174 Erpetogomphus leptophis sp. n. ...... 204 USO een a 174 Erpetogomphus elaphe sp. n. ...... i 205 The status of the name Erpetogomphus elaps Selys … … … n 207 Erpetogomphus menetriesii (Selys) ................ 174 Erpetogomphus liopeltis sp. n. LL... 210 Biologa iii 176 Erpetogomphus bothrops sp. n. ..................... 212 Phylogenetics and biogeography Erpetogomphus viperinus Selys ..................... DIS) Phylogenene analysis ................01:.csn002:- 177 Erpetogomphus designatus Hagen in Selys … 217 CAS ESS RARE 180 Erpetogomphus sipedon Calvert .................... 221 Distribution patterns... 181 Erpetogomphus lampropeltis Kennedy .......... 225 Descupuvebiogeography,............... 182 Eslalampropeltis Kennedy}®..2.2.1225 Character wegen lé se 182 E .l. natrix Williamson and Erpetogomphus species groups …nnen annen 185 Wallamsonstatin sn 226 Systematic part Erpetogomphus crotalinus (Hagen in Selys) .. 227 DEROO 185 Erpetogomphus heterodon sp. n. ................... 230 Genus Erpetogomphus Hagen in Selys, 1858 186 Erpetogomphus compositus Hagen in Selys ... 232 Rei tn 188 Erpetogomphus boa Selyss........222..2....22....,290 Erpetogomphus constrictor Ris ...................... 193 Erpetogomphus cophias Selys ........... 238 Erpetogomphus sabaleticus Williamson ........ 195 Acknowledgements eee en 240 Erpetogomphus tristani Calvert … 196 References ch... see 241 Erpetogomphus ophibolus Calvert ................. 197 Mlustrauonsp aan. kee ee 247 175 I) DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 INTRODUCTORY PART Introduction 3 No other group of Middle American Gomphidae has been in need of revision as much as the genus Erpetogomphus Hagen in Selys. The last species were described by Williamson and Williamson (1930), and although several odonatists have since received species which are possibly undescribed, there has been a hesitancy to describe them because so many species are rare or poorly known. Several years ago, I collec- ted a few specimens of one species I thought new, but I realized chat an examination of most of the types would be necessary before I could resolve the pro- blem. This paper is the result of those investigations. My purpose here is to: 1) describe and figure all species currently going under the generic name Erpetogomphus, 2) select lectotypes and establish type localities, if possible, and 3) construct a cladogram of all the known species so that phylogenetic relation- ships and distribution patterns can be adduced for the genus. I have examined types for all taxa except for two species, £. cophias (type lost), and E. ophibolus. The last species is distinct, and no confusion exists about its identity or type deposition. À neotype is se- lected for E. cophias. I also provide keys for all species. Twenty-three names (table 1) have been included in Erpetogomphus, but three, E. severus Selys, Herpetogomphus rupinsulensis Walsh, and H. pictus Needham (= Ophiogomphus rupinsulensis (Walsh)) have been transferred to Ophiogomphus. Another, E. menetriesii (Selys) I consider to be a nomen dubium, for reasons detailed below; and six new species are de- scribed. One species thought lost, £. boa Selys, has been rediscovered, and £. viperinus auctorum is really an undescribed species. I suspect that several new spe- cies will be found in Mexico and Central America, and I hope this paper will make it easier for others to describe those new species. The difficulties I have encountered during this study have mostly been due to the paucity of speci- mens of various species which have been considered rare. I have successfully associated both sexes of all current species except the male of E. agkistrodon, which is unknown. My task of recording aspects of intraspecific variation has been made considerably easier by various museum authorities and individuals listed under acknowledgements. History The genus Erpetogomphus was described (as a sub- genus) by Selys (1858) to receive two previously de- scribed species, Gomphus menetriesii Selys, 1850, and Ophiogomphus crotalinus Hagen in Selys (the latter originally described from a pair in 1854), as well as 174 two new species, Erpetogomphus elaps Selys, and E. cophias Selys. In the ‘Appendice’ of the same paper, Hagen described £. compositusand E. designatus. Selys described Æ. boa in 1859 and E. viperinus in 1868. Some of these species were briefly redescribed or dis- cussed by Selys and Hagen in 1859, 1869, 1873, and 1878. In 1879, Selys redefined the subgenus Herpetogomphus (an unjustified emendation of Erpetogomphus, see Cowley, 1934), under which were included eight species: E. compositus, E. designatus, E. viperinus, E. menetriesii, E. elaps, E. boa, E. cophias, and E. crotalinus. Calvert (1899) redescribed E. vipe- rinus (now E. bothrops sp. n.) based on material col- lected in Tepic, and included a diagnosis of the fema- les of E. viperinus and E. elaps. Kirby (1890), in a synonymic catalogue of the world Odonata, designa- ted Herpetogomphus crotalinus as the type species of the genus. Calvert (1905) included a synopsis for all Erpetogomphus from Middle America and described E. eutainia, E. ophibolus, E. sipedon, and E. diadophis. Calvert (1912a) added £. tristani. Ris (1917) descri- bed E. constrictor and redescribed what he thought to be the second known male of E. boa. In 1918 Williamson described E. sabaleticus from a pair from Colombia, and Kennedy (1918) described £. lampro- peltis from California. Calvert (1919) described £. schausi from a male from Guatemala. Finally, Williamson and Williamson (1930) described two new species from Baja California, E. coluber and E. natrix. In the same paper, they presented a synopsis of what was known for all species. Their summary mak- es clear that only a few common species were repre- sented in collections, and that many others were poorly known and/or described from insufficient ma- terial. The status of the name Erpetogomphus menetriesii (Selys) This name was first applied by Selys to an incom- plete male supposedly from Brazil. Selys (1850) brief- ly compared it to Paragomphus genei, as follows: Its colouration yellow, almost without spots, prevents confusion of this species with any other in Europe, but it resembles in this respect G. pallidus (Ramb.) of Southern America whose stature is stronger, and Gomphus menetriesii (new) from Brazil, but the feet of the last are mostly black, the d abdomen is bordered by black on the two sides, etc...’. The same specimen was redescribed it in 1854 as follows: ‘Head yellow, except for space between ocelli, that between the eyes and ciliated margin of the occiput black. Front of thorax yellow, with 2 median, contiguous stripes, al- most obliterated, pale red, antehumeral and humeral stripe equidistant. 3 Appendages lacking. From: Brazil? The type is said to be in the ‘Musée de St- Petersburg. Table 1. Species-group names used in the genus Erpetogomphus Name agkistrodon boa bothrops coluber compositus constrictor cophias crotalinus designatus diadophis elaphe elaps eutainia heterodon lampropeltis leptophis liopeltis menetriesii montanus natrix ophibolus pictus rupinsulensis sabaleticus schausi SEVETUS sipedon tristani viperinus Original Genus Erpetogomphus Erpetogomphus Erpetogomphus Erpetogomphus Erpetogomphus Erpetogomphus Erpetogomphus Ophiogomphus Erpetogomphus Erpetogomphus Erpetogomphus Erpetogomphus Erpetogomphus Erpetogomphus Erpetogomphus Erpetogomphus Erpetogomphus Ophiogomphus Herpetogomphus Erpetogomphus Erpetogomphus Herpetogomphus Herpetogomphus Erpetogomphus Erpetogomphus Ophiogomphus Erpetogomphus Erpetogomphus Erpetogomphus Original Reference This paper Selys 1859 This paper Williamson & Williamson 1930 Hagen in Selys 1858 Ris 1917 Selys 1858 Hagen in Selys 1858 Hagen in Selys 1854 Calvert 1905 This paper Selys 1858 Calvert 1905 This paper Kennedy 1918 This paper This paper Selys 1854 Selys 1878 Williamson & Williamson 1930 Calvert 1905 Needham 1897 Walsh 1862 Williamson 1918 Calvert 1919 Hagen 1874 Calvert 1905 Calvert 1912 Selys 1868 Type locality Jalapa, Mexico Vera Cruz, Mexico Rio Otapa, Veracruz State, Mexico San José de Comandu, Baja Calif., Mexico Pecos River, Texas; corrected to vicinity of Roswell, NM' Misantla, Veracruz, Mexico Mexico Mexico Pecos River, Texas; corrected to vicinity of Roswell, NM' Texas Costa Rica Mexico Guerrero, Mexico Aragon, New Mexico Fillmore, California Blue Creek, Belize La Estanzuela, Nuevo Leon, Mexico Brazil? Yellow Town, Montana San Jose de Comandu, Baja Calif., Mexico Atoyac, Veracruz, Mexico Ithaca, New York Rock Island, Illinois Cristalina, Colombia Purulta, Guatemala Colorado Guadalajara, Mexico Oricuajo, Costa Rica Orizaba, Veracruz, Mexico GARRISON: Revision of Erpetogomphus Location of type UNAM (8) IRSN (d) USNM (8) UMMZ (8) MCZC (9) SMF (d) MNHP (8) MCZC (8) MCZC (8) BMNH (8) FSCA (8) MNHP (8) BMNH (d) USNM (8) USNM (d) FSCA (8) FSCA (8) Lost IRSNG (8) UMMZ (8) BMNH (8) CUIC (8) Lost UMMZ (8) ANSP (8) Lostd (d, 2) BMNH (2) ANSP (ó) IRSN (8) _ See remarks under E. designatus.; ? Transferred to Ophiogomphus by Needham (1899); “_As Herpetogomphus; ° Transferred to Ophiogomphus by Selys (1879) Type status Holotype Lectotype Holotype Holotype Holotype Lectotype Neotype Lectotype Lectotype Holotype Holotype Holotype Holotype Holotype Holotype Holotype Holotype Holotype Holotype Holotype Holotype Holotype Holotype Holotype Holotype Lectotype Reference for first placement in Erpetogomphus This paper Selys 1859 This paper Williamson & Williamson 1930 Hagen in Selys 1858 Ris 1917 Selys 1858 Hagen in Selys 1858 Hagen in Selys 1858 Calvert 1905 This paper Selys 1858 Calvert 1905 This paper Kennedy 1918 This paper This paper Selys 1858 Selys 1879 Williamson & Williamson 1930 Calvert 1905 Needham 1897 Walsh 1862 Williamson 1918 Calvert 1919 Selys 1878° Calvert 1905 Calvert 1912 Selys 1868 ° Transferred to Ophiogomphus by Hagen (1874) 175 Present placement E. agkistrodon E. boa E. bothrops E. compositus E. compositus E. constrictor E. cophias E. crotalinus E. designatus E. eutainia E. elaphe E. elaps E. eutainia E. heterodon E. I. lampropeltis E. leptophis E. liopeltis Nomen dubium Ophiogomphus Severus montanus E. lampropeltis natrix E. ophibolus Ophiogomphus rupinsulensis Ophiogom phus rupinsulensis E. sabaleticus E. schausi Ophiogomphus severus E. sipedon E. tristani E. viperinus A\ [ITDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 In 1858, Selys discussed £. menetriesir under £. cro- talinus (in translation): ‘N.B. In the Synopsis des Gomphines, I described under the name of Ophiogomphus Menetriesii an in- complete male individual, which was communicated long time ago by the knowledgeable M I am inclined to believe that this is not a species distinct from crotalinus. One reason that made me suppose the species dif- ferent, was its occurrence in Brazil, but it was still necessary that this fact be duly confirmed and that the missing anal appendages be known This ¢ is a little smaller than our specimens [the presently described males and female of £. crotalinus], the space around the ocelli is distinctly black, the ap- pearance of the reddish antehumeral stripes more closely approaches the median [stripes], the humeral is more distinct and more elongated, the base of the abdomen is less swollen, the dorsal yellow stripe of 3rd, 4th, and 5th segments seems modified: it con- sists of spots of three lobes [each], whose posterior isn't visibly narrowed into a little round head. (This note is based on diagnosis and a description made a long time ago, so that I have not been able to compare the specimen since I have had crotalinus at hand.) A female, same indication of origin [Brazil] from the Mus. St. Petersburg, communicated to M. Hagen by M. Menetries differs very slightly from crotalinus by the following: 6 cells under the pterostigma.; costa yellower, tarsi black, the first article [tarsomere] of the posterior [leg] yellow, black tibiae bilineated with yel- low outside; femora yellow with a double black exter- nal line shorter and finer at the ends. Abdomen thick- er, marked with black; 9th segment black above with a dorsal round spot, touching the posterior margin. (Abdominal segment] 10 and appendages light red, the end of these last [structures] pale not marked with Selys’]. Vertex more black. Dimensions a little more robust. Not being certain of its identity, I fear to cause confusion later by putting the name E. menetriesii in true crotalinus from Mexico, even though I have al- ready published this name [menetriesii] in speaking of another species (page 102 Revue des Odonates, 1850; and Synopsis des Gomphines, 1854). If its identity is confirmed, it would be necessary to replace the name of E. crotalinus with that of E Menetrie il. . Finally, Selys (1878) described Herpetogomphus menetriesii based on ‘un mâle in complet [in MNHP]’ 1 female [in 1RSN] from Guatemala. He states: ‘I mee it identical with the d example described in Synopsis No. 20 [1854] and reported with doubt as rotalinus in the Monographie [1858]. Calvert pointed out the inconsistencies of the three descriptions and left open the question of what £. to me Menetries. Now, black [italics are 1905 176 menetriesii really is. According to the original description (Selys, 1854), E. crotalinus cannot be E. menetriesii. A supplemen- tary male of £. crotalinus described in 1858 (in ZMHB, examined) is unusual in having faint antehumeral stripes, but no £. crotalinus I have examined fits the description of E. menetriesii of 1854 (with middorsal, antehumeral and humeral stripes). The 1854 description is too brief to indicate what E. menetriesii is, and I cannot associate it with any known species. According to Hagen (1861), it was destroyed. I have been unable to trace the 1858 fe- male described by Selys, but I was able to examine the ‘pair’ described in 1878. The two specimens, which have determination labels by Selys, are really two fe- males of E. eutainia Calvert. However, I cannot be certain that these females are conspecific with Selys’ original description of E. menetriesii. Selys and others apparently had difficulty in associating specimens of E. eutainia, because a male from Texas in the IRSN is given the manuscript name of Erpetogomphus berus, another female is identified as Cyanogomphus ?mexic- anus (by R. Martin), and, of course, the two females, one in the IRSN, the other in the MNHP, are labelled as E. menetriesit. Thus, the original description of £. menetriesii may refer to E. eutainia, but because no one can ever be sure of this, I propose that Æ. menetriesii should be considered a nomen dubium. Biology Little is known of the biology of the genus, though their habits as recorded in the literature mirror the be- haviour of other Gomphidae. Adults are most com- monly found near shores of streams and rivers, but they may also be found in agricultural stubble or on tree branches near streams. Many species are seldom encountered and only a few species appear to be com- mon in collections. Available notes taken from the literature, personal correspondence, and personal observations cited under various species accounts indicate that female Erpetogomphus oviposit by swiftly tapping the water’s surface or by hovering motionless over moderately swift water systems. Details of copulation and copula- tion time are largely unknown, although the bizarre male and especially female morphological adaptations for achieving the tandem position in £. tristani have been detailed by Calvert (1912a). Larvae of E. designatus, E. compositus, and E. lam- propeltis were keyed by Needham and Westfall (1955), and illustrations and full descriptions of E. crotalinus and E. lampropeltis natrix were given by Novelo and González (1991). The larva of £. sabalet- icus was recently described and illustrated by Belle (1992). GARRISON: Revision of Erpetogomphus Table 2. Characters used in cladistic analysis of Erpetogomphus adults. d (male) or ? (female) in parentheses indicates sex to which character pertains. An asterisk (*) indicates an ordered multistate character. Numbers in parentheses following char- acter states refer to figure numbers. Number/character Primitive state 1. Postocciput width (9) tumid, convex 2. Postocciput (2) not visible from above 3. Occiput (9) semicircular 4. Vertex (9) with long median trough 5. Postocellar ridge (9) incomplete 6. Vertex: sexually dimorphic 7. Frontoclypeal suture with black 8. Hind margin of metepimeron with black 9. Hind wing 3-celled anal loop present . Anterior hamule* (4) . Posterior hamule* (4) . Tip of posterior hamule* (4) . Posterior lobes of penis segm. 1 (4) . Penis: segment 3 dorsally (4) . Penis: segment 4 (4) . Prepuce of penis* (4) . Lateral lobe of penis-shape* (4) . Membranous hood of penis (4) states 1-5 (fig. 208) states 1-4 (fig. 209) states 1-5 (fig. 210) small with pair of tubercles about twice as long as wide (fig. 211) states 1-3 (fig. 212) states 1-5 (fig. 213) not overlapping distally 19. Shape of cornua of penis* (4) states 1-6 (Fig. 214) 20. Development of cornua of penis* (4) reduced (1), absent (2) 21. Length of vulvar lamellar plates (9) about 0.25 length of sternite 22. Medial margins of vulvar lamina (9) folded under 23. Postlamellar ridge (2) Y-shaped 24. Postlamellar ridge with juncture (2) posterior to plates (e.g. figs. 184, 195) 25. Ventral base of cercus (4) with large blunt tooth 26. Ventral base of cercus (4) with no carina 27. Distal part of cercus ventrally (4) with no carina 28. Dorsal surface of cercus (G ) with a tubercle 29. Ventral surface of cercus (G ) linear or concave 30. Curvature of epiproct with tips (4) about parallel to base 31. Tips of epiproct () spatulate or bidentate 32. Penis segment 3 (d) short and stocky 33. Penis guard (¢) quadrate in cross-section 34. Epiprocts (4) approximate and curved at distal 0.50 35. Spermatheca (9) states 1-2 (fig. 215) 36. Dorsal surface of epiproct (4) with dorsal spine (fig. 95) 37. Ventral base of cercus (4) with pebble-like carina 38. Membranous hood of penis (4) partially (1) or wholly (2) exposing subcuticular membrane 39. Ventral base of cercus (4) with carina ending in tooth (fig. 116) 40. Postocciput (2) with lateral depressions (fig. 165) 41. Occiput (9) with a posteriorly directed medial spine fig. 151 Derived state(s) linear visible from above transverse with no trough complete not sexually dimorphic with no black with no black absent state 0 (fig. 208) state 0 (fig. 209) state 0 (fig. 210) well developed, divided lobes lacking tubercles about as long as wide (fig. 211) state 0 (fig. 212) state 0 (fig. 213) overlapping state 0 (fig. 214) present (fig. 214) 2 0.50 length of sternite not folded under semicircular at or before margin of plates (fig. 192) with no tooth (fig. 94). with a carina with a carina without a tubercle convex at about 90° to base pointed longer than wide circular in cross-section not approximatebarely curved state 0 (fig. 215) unarmed otherwise (fig. 97) completely covering subcuticular membrane with carina (if present) not ending in tooth with no lateral depressions without a spine PHYLOGENICS AND BIOGEOGRAPHY Phylogenetic analysis I have attempted to construct a cladogram based on an algorithm which infers phylogenies based on the presence of shared derived characters. Using the outgroup comparison method of Watrous and Wheeler (1981), I consider Ophiogomphus to be the outgroup to Erpetogomphus, following Carle (pers. comm.) in his assessment of gomphid dragonfly clas- sification. In this analysis, I have used the most austr- al species, O. severus Hagen. The genus Erpetogomphus has never been easily de- fined (Calvert 1905, Needham and Westfall 1955), these authors using a series of characters to character- ize the genus. A synapomorphy apparently unique to Erpetogomphus is the condition of the epiprocts of the males: all have gently (at least 110°) to strongly dor- sally curved, non-divergent epiprocts. Carle (1986), in his treatment of the higher classifi- cation of the Gomphidae, included Erpetogomphus in the tribe Onychogomphini of the subfamily Onychogomphinae. In his key to the eight subfami- lies of Gomphidae, he characterized the Onychogomphinae as follows: 1) hind femur shorter than head, 2) anal triangle typically 4-celled with small rectangular cell along inner margin, 3) anterior hamuli with shoulder and end hook forming lateral u-shaped notch, and, 4) female sternum 9 with large U- or V-shaped basal membranous area. Within this subfamily, the tribe Onychogomphini (including Erpetogomphus) is distinguished from the tribe Crenigomphini by the following characters (antago- nistic characters for Crenigomphini in parentheses): postgenal suture present (absent), male tibial laminae 177 TI]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 present (absent), penile segment one (= peduncle) without cuplike thin-walled hood (with bilobed cup- like thin-walled hood), female sternum 9 with dome- shaped membranous area not extending to distal half of sternum (this structure extending to distal half of sternum). None of these characters is stated by Carle (1986) to be uniquely derived for any of the above categories. Carle (pers. comm.) and I agree that Ophiogomphus is the most likely outgroup to Erpetogomphus, and we agree that both should be placed in the Onychogomphini. Three characters seem unique to the New World Onychogomphini, which include only Ophiogomphus and Erpetogomphus. 1) rami of male epiprocts contiguous along all or most of their length, 2) female sternum 8 with a well-defined costate postlamellar ridge separat- ing the soft membranous area from the posteriorly sclerotized area (this ridge reduced in Ophiogomphus), 3) presence of a lateral lobe on the ventrolateral mar- gin of the fourth penile segment (fig. 213) (reduced in Ophiogomphus, and not specialized in the E. ophib- olus group). This last character seems to be a unique synapomorphy for Ophiogomphusand Erpetogomphus. Carle (1992) further separated Erpetogomphus into three subgenera, two of which were new: Calogomphus (type species E. eutainia) and Erpetocyclops (type species E. ophibolus). Both Carle (pers. comm.) and I consider Erpetogomphus to be composed of three monophyletic groups as discussed below. While several earlier versions of my cladogram using series of characters and different states generally support a conservative grouping of the genera into three monophyletic groups, I do not think it neces- sary to provide subgeneric rank to those groups. A suite of 41 characters, 7 of which are multistate (table 2), was analyzed using the HENNIG86 comput- er program (Farris, 1988), with the following op- tions: The character set was first run using the ie* (guaranteed to find the most parsimonious trees) op- tion where all characters were ordered and each was equally weighted (default options). All autapomor- phies were included from this analysis. The final set of 41 characters was finally chosen from larger sets, each with different coding sequences, because this set seemed to be the most heuristic. The resultant analy- sis yielded six equally parsimonious trees. A Nelsen (consensus) tree (length 106, consistency index 0.62) was then generated from the six trees. The resultant characters, tree, and matrix are shown in figs. 208- 216, and tables 2-3. As this manuscript was in review, Jan van Tol kindly ran the same data set on pauP 3.1.1 for the Macintosh (Swofford 1993) using unor- dered as well as ordered sequence of multistate char- acters. His results, which he kindly communicated to me, resulted in 61 semistrict consensus trees each with a length of 94 for unordered multicharacter 178 states. His resulting cladogram is similar to that in fig. 216 except that the entire £. crotalinus group (except for three sister groups: £. elaphe and E. elaps, E. sipe- don and E. lampropeltis and E. boa and E. cophias) collapses to a polytomious assemblage. (I recorded a similar cladogram when multistate characters were listed as unordered for HENNIG86.) When all multis- tate characters were run as ordered on PAUP 3.1.1, the resulting cladogram was the same as shown for fig. 216. I have not been totally successful in resolving sever- al of the terminal clades, due to high degree of appar- ent homoplasies. The best characters are those of the secondary genitalia of the male (hamules, penis); but, despite a thorough examination of these structures, I am still uncertain of the correct interpretation of character polarity for several of these complex, mult- istate character sets (figs. 208-210, 213-214). Little difficulty was encountered in assigning states to members of the £. ophibolus and E. eutainia groups, and their generalized tracks also independently sup- plement in understanding their phylogeny; but the same characters in males of the £. crotalinus group show, if my cladogram is correct, a bewildering com- plex of character reversals for some characters. The generalized tracks within the £. crotalinus group (fig. 226) also provide few clues to possible speciation events, except for that between the £. elapsand E. cro- talinus (s.s.) groups. Individual clades are numbered as shown in fig. 216, each of which is discussed below. Clade 1.1. — E. ophibolus group (six species: E. constrictor, E. sabaleticus, E. tristani, E. ophibolus, E. agkistrodon, E. schausi) = subgenus Erpetocyclops Carle Character 3. — female occiput semicircular (parallel development in Clade 1.22112, E. liopeltis) Character 5. — female postocellar ridge incomplete (parallel development in Clade 1.211, E. leptophis) Character 6. — vertex sexually dimorphic (state not known for male £. agkistrodon, but probably follows for other members of this clade) Character 11, state 1. — male with posterior hamule triangular (fig. 209) Character 12, state 1. — male with tip of posterior hamule acuminate (fig. 210) Character 16, state 1. — male with prepuce well-de- veloped (fig. 212) Character 17, state 1. — male with lateral lobe ves- tigial, broadly angulate with no serrations (fig. 213) Clades 1.11 (E. constrictor, E. sabaleticus, E. tristani, E. ophibolus), 1.12 (E. agkistrodon), 1.13 (E. schausi). Clade 1.11 is defined by the following synapomorphies: Character 20, state 2 - male with loss of cornuae of penis (fig. 214). Most Gomphidae (including the outgroup, Ophiogomphus severus), have a pair of long, strap-like cornuae. I have hypothesized the character transformation for this structure throughout the ge- nus (fig. 214). Character 25. — ventral base of cercus of male with large blunt tooth (fig. 94) Character 28. — dorsal surface of cercus of male with a tubercle (fig. 94) I have found no synapomorphy for uniting 1.12 (E. agkistrodon) and 1.13 (E. schausi). The former is known only from the holotype female, the latter by two males and a teneral female. Clades 1.111 (E. constrictor), 1.112 (E. sabaleticus), 1.113 (E. tristani) Character 29. — curvature of epiproct of male with tips completely recurved so as to lie parallel to base Clade 1.21. — E. eutainia group (two species: E. eutainia, E. leptophis) = subgenus Calogomphus Carle Character 10, state 1. — anterior hamule with equally divided arms (fig. 52-53, 208) at distal 0.25 Character 11, state 2. — posterior hamule digit- shaped (fig. 209) Character 14. — dorsal surface of third segment of penis of male with two tubercles Character 16, state 2. — prepuce of male reduced, largely hidden by frill-like lateral lobe (fig. 212) Character 17, state 2. — lateral lobe of penis of male frill-like (fig. 213) Character 35, state 1. — spermatheca of female a single tube (fig. 215). I was only able to study this structure in E. eutainia, as only the allotype of E. lep- tophis is known. Due to other synapomorphies listed above, I hypothesize that £. /eptophis will possess sin- gle, rather than paired, spermatheca. Clade 1.22. — E. crotalinus group (13 species: E. elaphe, E. elaps, E. liopeltis, E. bothrops, E. viperinus, E. designatus, E. sipedon, E. lampropeltis, E. crotalinus, E. heterodon, E. compositus, E. boa, E. cophias) = subgenus Erpetogomphus Hagen in Selys Character 16, state 3. — prepuce of penis of male lacking Character 35, state 2. — spermatheca of female pai- red, each forming a long, narrow sac (fig. 215) Clade 1.221. — E. elaps subgroup (five species: E. elaphe, E. elaps, E. liopeltis, E. bothrops, E. viperinus) Character 17, state 3. — unique shape of lateral lobe (fig. 213) GARRISON: Revision of Erpetogomphus Clade 1.2211 (E. elaphe, E. elaps, E. liopeltis, E. bothrops) No satisfactory synapomorphy has been found for this clade. One character, the spatulate tip of the epi- proct (fig. 129) separates £. viperinus from its sister species. Clade 1.22111 (E. elaphe, E. elaps, E. liopeltis) No synapomorphy has been found, but the trans- verse condition of the female postocciput in £. both- rops and in E. viperinus (fig. 166) split £. bothrops from the other three sister species. Clade 1.221111 (£. elaphe, E. elaps) Character 22. — vulvar lamina of female with mar- gin folded under ventrally (figs. 180- 181) Clade 1.222 (£. designatus, E. sipedon, E. lampropeltis, E. crotalinus, E. heterodon, E. compositus, E. boa, E. cophias) No satisfactory synapomorphy has been found to unite the £. crotalinus group (sensu stricto [s.s.]). The one aberrant species in the group is £. designatus. It alone has the following characters which separate it from the other 7 species: 1) tip of posterior hamule lacking a tooth (recurrence of this state from the £. eutainia group) and 2) unique shape and position of the lateral lobe (fig. 82). The shape of the anterior ha- mule of £. designatus (fig. 60) is, however, like that of E. sipedon, E. lampropeltis, E. crotalinus, and E. com- positus, which thus unites these species into a clade. The widely distributed £. designatus has been diffi- cult to place cladistically in this study. It combines characters of the £. elaps group (for example, linear condition of lateral lobe and acuminate condition of the cornua) and even apomorphic characters com- mon to the £. crotalinus (s.s.) group as noted above. Clade 1.2222 (£. sipedon, E. lampropeltis, E. crotalinus, E. heterodon, E. compositus, E. boa, E. cophias) Character 12, state 5. — tip of posterior hamule of male with anteapical tooth Character 17, state 5. — unique shape of lateral lobe of fourth segment of penis of male (fig. 213). The small, semicircular condition of the lateral lobe present in Clade 1.2222 is similarly developed in Clade 1.12 (E. schausi). Clades 1.22221 (E. sipedon), 1.22222 (E. lampropeltis), and 1.22223 (E. crotalinus, E. heterodon, E. compositus, E. boa, E. cophias) No satisfactory apomorphy has been found to di- chotomize these clades. The distinctive shape of the cornuae of E. sipedon and E. lampropeltis (figs. 83-85) has separated these two species from clade 1.22223. 179 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Table 3. Character matrix used in producing Nelsen (consensus) tree shown in Fig. 216. Character No. 1 2 3 4 5 6 7 8 9 NO Mees 12 STATS 16 7 alts} Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ( )p} », severus 0 0 0 0 0 0 0 0 I 0 0 0 0 0 0 0 0 0 constrictor l l l l 1 1 1 1 0 0 1 1 Il 0 0 1 Il 1 sabaleticus ] ] l ] ] ] 1 1 0 0 1 l I (0) 0 1 1 1 tristant l l ] l l 1 1 1 0 0 1 Il 1 0 0 1 Il 1 ophibolus l l | 0 1 l 1 1 0 0 1 Il 1 0 0 1 1 1 IDR AO OR IT IT Ae OWN NOPE HETE Al RO AO ip, SE schausi | lé, 0200 1 1 hu: CAO Ora A 1 1. COM 1 Sala eutainia 0 1 1 0 0 0 1 IE 0 1 2 2 Il Il 0 2 2 1 lepto phis l l 0 0 1 0 1 1 0 Il 2 2 1 1 0 2 2 0 elaphe 1 0 0 0 0 0 0 0 0 3 3 2 1 0 1 3 3 1 elaps 1 0 0 0 0 0 0 0 0 2 3 5 1 0 1 3 3) 1 liopeltis 1 ] l 0 0 0 0 0 0 2 3 4 1 0 1 3 3 1 bothrops 0 1 0 0 0 0 0 0 0 2 5) 5 1 0 1 3 3 1 viperinus 0 l 0 0 0 0 0 0 0 2 9) 3 1 0 Il 3 3 1 designatus 0 l 0 0 0 0 0 0 0 4 3 9) 0 0 1 3 4 0 sipedon 0 1 0 0 0 0 0 0 0 4 3 5 1 0 1 3 5 1 lamp ropeltis 0 1 0 0 0 0 0 0 0 4 3 5 1 0 1 3 5 1 crotalinus 0 1 0 0 0 0 0 0 0 4 3 5 1 0 1 3 5 1 heterodon 0 l 0 0 0 0 0 0 0 4 3 5 1 0 1 3 5) Il compositus 0 0 0 0 0 0 0 0 0 4 3 5 1 0 1 3 5 1 boa 1 1 0 0 0 0 0 0 0 5 4 5 1 0 1 3 5) 1 cophias 1 0 0 0 0 0 0 0 0 5 4 5 I 0 1 3 5 1 Only the holotype 2 is known for this species. Male characters were scored as ? (unknown), on 0? or 1?, based on af- finity with rest of the E. ophibolus group. * Character prevalent in this species. * Basal carina weak in E. boa. * This structure not examined for this species. Plesiomorphy/apomorphy hypothesized by examination of spermathe- ca in closely related species, and placement of these taxa in cladogram due to other synapomorphies. Clades 1.222231 (E. crotalinus) and 1.222232 (E. heterodon, E. compositus, E. boa, E. cophias) No satisfactory apomorphy has been found for cla- because of the unique shape of its cornuae (fig. 86). Clades 1.2222321 (E. heterodon), 1.2222322 (E. compositus), 1.2222323 (E. boa, E. cophias) The unique shape of the cornua (figs. 87-89) has segregated £. heterodon and E. compositus from clade Apomorphies for the last clade, repre- senting the two closely related species, E. boa and E. cophias, are: DE du de Vba I è Character 10, state 5. — unique shape of the anteri- or hamule of male (fig. 208) Character 11, state 4. — unique shape of the poste- rior hamule of male (fig. 209) Character 19, state 6. — unique shape of the cornua of male (fig. 214) The clades mentioned above have problems pri- marily within the £. crotalinus group (sensu lato, [s./]), and most dichotomies are based on one or two highly complex character states of the penis. I realize that my interpretation of the morphoclines of this character is open to reevaluation, which could result in a somewhat different shuffling of some species 180 within the larger group. Only two species in this group, E. boa and E. cophias, are easily characterized as amonophyletic group. A further analysis to resolve some of the difficulties described above will have to await further specimens and a proper analysis of their larvae. As stated earlier, I have not assigned formal names to subordinate clades, because further investigation may involve transfer of some monophyletic or paraphyletic groups to other clades, nor can I see any purpose in a prolife- ration of infrageneric names. Cladospecies Of the 21 species described here, seven (33%) can be defined by autapomorphies and can be termed cla- dospecies (Ackery and Vane-Wright, 1984). These cladospecies with their uniquely derived character sta- tes are shown below. Clade 1.112. E. ophibolus. Character 36, dorsal sur- face of epiproct of male with a dorsal spine (fig. 95) Clade 1.211. £. leptophis. Character 41, female oc- ciput with a posteriorly directed medial spine (fig. 151) Clade 1.212. E. eutainia. Character 37, ventral base of cercus of male with a pebble-like carina. Clade 1.2211112. E. elaphe. Character 10, state 3, N Oo N SS i) Nn N (i N I N co N NO © (©) DD ISS) N GARRISON: Revision of Erpetogomphus wv v v u 0 0 1 I 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 SITISISISTSESTSIT EIGEN Oro LS) (SSIS) pda a Ka ri rl ar rei cr u eg je je << Cn _ © mi © © HH HAWK WWW hi mi Hi © © HMHmHooooo00- DO ST _S _S rr I-r-r-D° OON OO nm nm © © -iereketele}-{-K-kKko-k-li-cKolekoKoKcW{_W_W sedi} =O O10 Of attirati 33 34 35 36 37 38 39 40 4] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 Il I 0 0 0 0 0 0 0 1 I 0° 0 0 0 0 0 0 1 1 0° I 0 0 0 0 0 1 1 0 0 0 0 0 0 0 1 1 0? 0 0 0 0 0 0 1 1 0° 0 1 0 0 0 0 1 1 Il 0 0 0 0 0 1 1 1 = 0 0 0 0 0 0 1 1 OF 0 0 0 0 0 0 1 il 2 0 0 0 0 0 0 1 1 De 0 0 0 0 0 0 1 1 2 0 0 0 0 0 0 1 1 2 0 0 0 0 0 0 1 1 2 0 0 2 0 0 0 1 1 2 0 0 0 0 0 0 Il Il 2 0 0 0 0 0 0 Il 1 2 0 0 0 0 0 0 1 Il 2 0 0 1 0 0 0 1 Il 2 0 0 0 0 0 0 Il 1 Di 0 0 0 1 1 0 unique shape of anterior hamule of male (fig. 208) Clade 1.22221. E. sipedon. Character 38, state 2, membranous hood of penis of male wholly exposing subcuticular membrane. Clade 1.2222322. E. compositus. Character 38, state 1, membranous hood of penis of male partially exposing subcuticular membrane. Clade 1.22223232. E. cophias. Character 39, ven- tral base of cercus in male with carina ending in tooth (fig. 116), and Character 40, postocciput of female with lateral depression (fig. 165). These two charac- ters are probably correlated: I hypothesize that the postoccipital depressions of the female receive the in- ferior carinal tooth of the cerci of the male during copulation. I have been unsuccessful in finding autapomor- phies for the 14 other species, and they can be termed paraspecies (Ackery and Vane-Wright, 1984). A fur- ther, more detailed analysis of these species may yield autapomorphies. The cladist may question the status of species recognized by single or combinations of plesiomorphic characters, but I agree with Ackery and Vane-Wright (1984: 10), who cogently argue: ‘the traditional groupings often (but by no means always) turn out to be good. The cladist must, therefore, al- ways retain respect for the ‘Gestalt’ approach (espe- cially when practised by field workers), unless armed with concrete evidence for some contrary arrange- ment... Distribution patterns Table 4 details the distributional patterns of all New World Gomphidae according to five political re- gions: Canada/Alaska, U. S. A., Mexico, Central America, and South America. All speciose (i.e., N > 3 species) gomphid genera in the New World have pro- portionally more species in the United States and in South America than Erpetogomphus. Erpetogomphus has the most species in Mexico (76%). Central America follows with 48%, the United States with 24%. No other genus is proportionally as speciose in Mexico. The Gomphus complex of genera (Arigomphus, Gomphus and subgenera, Stylurus), and Ophiogomphus are dominantly North American, and several genera (Agriogomphus, Aphylla, Archaeogomphus, Cyanogomphus, Gomphoides, Neogomphus, Phyllocycla, Phyllogomphoides, Progomphus, and Zonophora) are South American. Epigomphus is most speciose in Central America, where 46% of all described species occur, closely fol- lowed by South America. Figure 217 shows the density and approximate overall distribution of species of Erpetogomphus as 181 T1NDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 shown by 150 km squares. Of the five species found in the United States, most are found in the south- western states. The most northerly distribution for the genus is Grant County, Washington (46°50’N), for E. compositus (Paulson and Garrison, 1977). The greatest concentration in species is in south-central Mexico between approximately 18°-20°N with eight species (31%) of all che total number of species in the genus found in southern Veracruz alone. Descriptive biogeography In this section I describe the current distribution range (= tracks) of various monophyletic groups ba- sed on my cladogram. Unfortunately, the numerous problems associated with the cladistic analysis conco- mitant with the poorly known or spotty distributions (most of which may be due to inadequate collecting in parts of Mexico) preclude a discussion of vicarian- ce biogeography. However, I do discuss possible spe- ciation sequences within the £. ophibolus and E. eu- tainta groups. The cladogram (fig. 216) for Erpetogomphus di- vides the 21 known species into three monophyletic groups. The generalized track for the genus is shown in fig. 218. The first two groups, the E. ophibolus and E. eutainia groups, have generalized tracks encom- passing south Texas, eastern Mexico south into northern Colombia and Venezuela (figs. 219-220). The E. ophibolus group comprises six closely related species united by seven derived character states. Figure 221 shows that the distribution of all six spe- cies except E. schausi are largely allo- or parapatric. Little can be said of E. agkistrodon or E. schausi due to paucity of records. Erpetogomphus ophibolus occupies an area between two disjunct populations of E. con- strictor. Further collecting may show E. constrictor to be sympatric or parapatric with E. ophibolus. Aside from E. agkistrodon and E. schausi, the track of E. ophibolus was probably the first to separate from the E. ophibolus group, a hypothesis consistent with the results of my cladogram (fig. 216). The three remain- ing species, E. constrictor, E. tristani, and E. sabaleti- cus, are closely related sister taxa whose tracks are largely parapatric. Presumably their differentiation by allopatric speciation?) took place from a common ancestor which occupied part of the present range of all three species. Ihe tracks of the second monophyletic group com- prises just two species, the relatively widespread £. latter, known only from the holotype and allotype from Belize, is sympatric with E. ophibolus, but its ge- nealogical affinities are unquestionably with E. eutai- nia (fig. 216). Erpetogomphus leptophis probably arose s a peripheral isolate from E. eutainia, or from a common ancestor, since it is a more apomorphic spe- 182 cies than £. eutainia. The remaining 13 species comprise two monophy- letic subgroups, one predominantly austral, the other more boreal. Their generalized tracks are shown in figs. 223-226. The E. elaps subgroup contains five species, whose somewhat limited distributions occu- py the western and eastern lowlands of Mexico south along the mountainous areas into Costa Rica. Erpetogomphus elaps and E. elaphe are closely related sister species which are parapatric in Guatemala. A slight but consistent southern variant of E. elaps occu- pies the area between more northerly typical popula- tions of E. elapsand E. elaphe (see discussion under E. elaps): it may be an incipient species. These two (or three?) species, like those for £. constrictor, E. tristani, and E. sabaleticus, probably arose from an E. elaps-like ancestor or from one another. Interpretations of the tracks of the remaining three species is problematical due to the uncertain position of E. liopeltis in my cla- dogram. It possesses characters common to E. both- rops and E. viperinus. No convincing apomorphy unites E. /iopeltis with the sister taxa E. elaps and E. elaphe, so its genealogical relationship must remain in doubt. The two species with the greatest ranges (£. elapsand E. bothrops) are broadly sympatric; E. liopel- tisand E. viperinus are allopatric. The E. crotalinus (s.s.) subgroup comprises the bo- real group of eight species, with at least one represen- tative occurring throughout most of the southern half of the United States. Their complex, mostly overlap- ping distributions (fig. 226) and questionable phylo- genetic sequence in the cladogram preclude any meaningful discussion on their biogeography. Erpetogomphus boa and E. cophias stand out because of three synapomorphies. Their limited distribution and close proximity indicate speciation, possibly by allopatric means. Erpetogomphus sipedon and E. heter- odon make up the only other allopatric species pair in this complex subgroup. Although distribution of E. heterodon and E. compositus overlap, I have seen no evidence of these two species collected together. It is obvious that a more thorough knowledge of the phylogeny of the genus will be necessary before interpretation of biogeographical events is possible. Of particular interest will be a methodological inter- pretation of the biogeography of the genus, for then we may be able to determine which species tracks are the result of dispersal and gene flow or vicariant events. Characters The best diagnostic characters for males are found in the caudal appendages, accessory genitalia, and overall body maculation. Most species are easily iden- tified by the caudal appendages. The cerci (superior appendages) assume a wide variety of shapes (figs. 92- GARRISON: Revision of Erpetogomphus Table 4. Distribution of species of New World Gomphidae. Number under each region = area of region in square miles; % directly under area = % of total land area; N = total number of species; number under each region = total number of species; number in parentheses () = percent of total number of species in genus. Phylogenetic classification is after Carle (1986). Genera are arranged alphabetically under each tribe. Canada/Alaska U.S.A. Mexico Cent. Amer.” South Amer. 4,438,221 mi’ 3,022,261 mi 761,530 mi’ 208,800 mi’ 6,597,386 mi Genus N 29.5% 20.2% 5.2% 1.1% 43.9% Hageniinae Hagenini Hagenius 1 1 (100) 1 (100) Octogomphinae Hemigomphini Neogomphus 3 3(100) Octogomphini Lanthus 1 (50) 2 (100) Octogomphus 1 1 (100) 1 (100) 1 (100) Stylogomphus I 1 (100) 1 (100) Gomphinae Gomphini Arigomphus 7 3 (43) 7 (100) Dromogomphus 3 1 (33) 3 (100) Gomphus 38 13 (34) 38 (100) 1 (3) Stylurus 13 12 (92) 4 (31) Epigomphinae Epigomphini Epigomphus 26 6 (23) 12 (46) 10 (38) Austrogomphinae Archaeogomphini Archaeogomphus 5 1 (20) 1 (20) 5 (100) Cyanogomphini Agriogomphus 4 1125) 1(25) 3 (75) Cyanogomphus 6 6(100) Tibiagomphus 2 2(100) Onychogomphinae Onychogomphini Erpetogomphus 21 5 (24) 16 (76) 10 (48) 1 (5) Ophiogomphus 18 8 (44) 18 (100) Lindeniinae Gomphoidini | Aphylla 19° 3 (16) 2 (11) 3 (16) 18 (95) Gomphoides 3 3 (100) Idiogomphoides 2 2 (100) Peruviogomphus 2 2 (100) Phyllocycla 31 3 (10) 4 (13) 27 (87) Phyllogomphoides 43 2 (5) 12 (28) 7-(17) 27 (63) Lindeniini Cacoides 1 1 (100) Melanocacus 2 2 (100) Mitragomphus 1 1 (100) Progomphini Progomphus 59° 4 (7) 9 (15) 7 (12) 52 (77) Zonophorini Desmogomphus 2 1 (50) 2 (100) Diaphlebia DI 2 (100) Perigomphus 1 1(100) 1 (100) Zonophora 10 10 (100) TOTAL 329 35 (10.6) 97 (29.4) 50 (15.2) 47 (14.3) 180 (54.7) i Excluding Hawaii; * Belize, Guatemala, El Salvador, Honduras, Nicaragua, Costa Rica, Panama; ° Includes subgenera Gomphurus, Gomphus, Phanogomphus, Stenogomphurus; * One species, A. caraiba, is endemic to Hispaniola and Cuba, not in- cluded in N; ° Three species, P. integer, P. serenus, P. zephyrus, are endemic to the Greater Antilles, not included in N. 183 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 116), but they are never longer than segments 9 and 10 combined. All males of the E. ophibolus group, ex- cept for £. schausi (and possibly the unknown male of E. agkistrodon) have a dorsal appendage on the distal third of the cercus. The dorsal surface of the cercus may be straight as in £. elaps, strongly curved as in £. schausi and E. viperinus, angulate as in most species in the E. crotalinus group, or concave as in the E. boa and £. cophias group. The inferior third of the cercus forms a strong carina in E. cophias (weakly so in £. boa), in the £. crotalinus group, and in E. eutainia. In the lighter coloured species of the E. crotalinus group, the carina does not contrast with the rest of the ap- pendage. A unique series of pebble-like structures characterizes the basal carina in £. eutainia. The infe- rior carina was used by Calvert (1905), although he wrongly placed £. eutainia among individuals lacking such a carina. Instead of a basal carina, the £. ophibo- lus group, E. bothrops and E. viperinus have an inferi- or carina on the distal third of the appendage. This structure is weak in £. elaps. The epiprocts (inferior appendages) are not as var- ied as the cerci. In most of the Æ. ophibolus group, they are strongly curved so that their distal third is nearly parallel to the basal 0.30. The epiproct of E. ophibolus is unique in having an anteriorly directed projection on its dorsal surface near the middle. The epiprocts are remarkably similar throughout the re- maining groups: they form a gende curve with the tips at approximately 90° to the basal third of the ap- pendage. The penes of most species possess a semi-hyaline pair of flagella or cornuae best seen in dorsal view. The cornuae are considerably reduced in the £. ophibolus group (except for E. schausi). Their shape provides a useful character to differentiate among £. designatus, E. compositus, and E. lampropeltis. The tip of each cornua is bluntly rounded in the £. crotalinus group (except for E. crotalinus). In all others, the tip forms a sharp point. The spiny lateral lobes of the penis are well devel- oped and frill-like in E. eutainia and E. leptophis, pri- marily knife-like or pointed in E. viperinus, E. liopel- tis, E. bothrops, E. elaps, and E. elaphe, and a protruding semicircle in most members of the E. cro- talinus and E. cophias groups. When the penis is viewed laterally, the lateral lobe is on edge, because this structure rotates posteriorly with its flat portion placed mediodistally. The lateral lobes are poorly de- veloped in E. constrictor, E. tristani, and E. sabaleticus (possibly also in £. agkistrodon), but those species and E. schausi have a unique ventral, arcuate prepuce. The anterior hamules are always divided, but the posterior arm in the E. tristani group forms more ofa shoulder than a branch. The bifurcation occupies the distal 0.25 of the hamule in the E ophibolus, E. elaps, 184 and £. eutainia groups, the distal 0.50 in the £. cro- talinus group, and the entire length of the appendage in E. boa and È. cophias. The posterior hamules are always more prominent than the anterior pair. They are broadly triangular in the £. ophibolus group, but they possess an anterior shoulder in the £. crotalinus group. In E. eutainia and E. leptophis, the posterior hamule forms a small fin- ger-like appendage. All known species except E. desig- natus, E. elaphe, E. leptophis, and E. eutainia possess a black apical tooth. This distal tooth is rotated lateral- ly in E. viperinus and in some E. bothrops. The best structural characters for separating the fe- males are those associated with the vertex, occiput, postocciput (note: my use of the term, ‘postocciput’ is not equivalent to the structure immediately surround- ing the occipital foramen, but, instead refers to the medioposterior side of the head behind and below the occipital crest), and vulvar lamina. Calvert (1905) relied on the curvature of the posterior border of the occiput, but it is variable within species. I do not use it here. Better characters are the shapes of structures on the vertex. In £. constrictor, E. tristani, and E. sab- aleticus, the median ocellus is placed within a deep longitudinal trough which accommodates the strong- ly curved epiproct of the male. In all other species may be found a transverse trough (figs. 145-146) or a pair of anterolateral pits (fig. 137) anterior to the me- dian ocellus. These pits correlate with the shape of the tip of the male epiproct, and these depressions are al- so present on the vertex of males. Their shape is use- ful in distinguishing between the closely related pairs, E. viperinusand E. bothrops, and between E. composit- usand E. lampropeltis. A complete postocellar ridge is present only in all members of the £. tristani group and £. leptophis. In all others, the middle part of the ridge is lacking, leaving only postocellar tubercles. Females of a few species have autapomorphic charac- ters, such as the raised median area of the occiput of E. designatus (figs. 156-157), the posteriorly directed pointed protuberance on the occiput of £. leptophis (fig. 151) or the postoccipital depressions of E. cophi- as (fig. 165). The vulvar laminae vary in the shape of the plates. The simple horizontal plates are bisected by a small V-shaped cleft in the E. ophibolus group. The plates are corrugated and more membranous in £. boa and E. cophias, often succumbing to much post mortem distortion. In the E. crotalinus group, each plate is a large, rectangular, planar structure which meets the other only at the anterior angle. The resulting median cleft is only a third or less the width of each plate; but the cleft is wider and U-shaped in £. designatus. A fur- ther widening of the cleft is evident in the E. elaps group. The plates form two small, widely separated, digit-like structures in £. eutainia and E. leptophis. A Y-shaped postlamellar ridge is found posterior to the vulvar lamina. The posterior stem of the Y is ab- sent in E. eutainia and E. leptophis, so that the ridge forms a semicircle (figs. 178-179). I have found no structural differences between the females of E. heterodon and E. sipedon. Body pattern must be used to differentiate females of those species. Erpetogomphus tristani and E. sabaleticus, and E. elaphe and E. elaps are also closely related pairs, and their females are distinguishable only by locality. In well-preserved specimens, body maculation pro- vides useful characters. Dark thoracic maculation pre- dominates in the £. tristani group, in E. leptophis, and in southernmost specimens of E. eutainia. All of those have a well-defined dark stripe bordering the posteri- or margin of the metepimeron. Erpetogomphus cophi- as, E. crotalinus, and most E. elaps have nearly immac- ulate body patterns. Thoracic maculation is usually stable, but there is variability in E. lampropeltis, E. compositus, E. designatus, E. elaps, and E. eutainia. Two formerly named species, E. natrixand E. coluber, I consider subspecies of £. lampropeltisand E. compos- itus, respectively: I have found no morphological characters which separate those forms. Thoracic pat- terns in £. elaps are often nearly absent (fig. 13), and more extensive dark markings characterize popula- tions of E. eutainia in the north, less extensive in the south (figs. 7-9). Abdominal patterns are characterized by a combi- nation of light and dark banding patterns separated dorsally by pale longitudinal stripes. The least amount of black occurs in £. cophias and E. crotalinus, the most in £. agkistrodon. A sharp, prominent, middorsal posterior spine on abdominal segments 8 and 9 is found in many males of E. ophibolus, E. bothrops, E. liopeltis, and E. viperi- nus. The spine is variable and cannot be used reliably to distinguish among species. For example, many £. bothrops have no indication of a middorsal spine, but it is well developed in others. The spines are also found on abdominal segment 9 in some females of the E. ophibolus group. Erpetogomphus species groups I divide the 21 species into five groups. I believe the unknown male of one species (E. agkistrodon) will fol- low those characters listed for that sex discussed be- low. 1. E. ophibolus group (= subgenus Erpetocyclops Carle, 1992). Males (probably so for E. agkistrodon, when discovered): With a distinct curved prepuce, posterior hamule forming a triangle culminating in a sharp black tooth. Females: Vulvar lamina simple, small, relatively undifferentiated postocellar ridge complete, though curved in £. tristani, E. sabaleticus, and E. constrictor. Six species: E. tristani, E. sabaleti- GARRISON: Revision of Erpetogomphus cus, E. constrictor, E. ophibolus, E. agkistrodon, E. schausi. 2. E. eutainia group (= subgenus Calogomphus Carle, 1992). Males: Prepuce small, obscured by lat- eral lobes of penis; lateral lobes of penis strongly de- veloped, forming a spiny frill; posterior hamule fin- ger-like, lacking an apical tooth. Reduced prepuce present. Females: Vulvar lamina small, finger-like, widely separated; post-laminar ridge semicircular, not Y- shaped. Two species: £. leptophis, E. eutainia. 3. E. elaps group. Males: Lateral lobe of penis knife- like. Females: Vulvar lamina connected anteriorly, but separated by a cleft 1.0 or more the width of each plate. Five species: £. bothrops, E. liopeltis, E. viperi- nus, E. elaphe, E. elaps. 4. E. crotalinus group. Males: Superior surface of cercus angulate (except for E. compositus); anterior ha- mule divided at distal half. Females: Vulvar lamina large, planar, rectangular, meeting at anterior margins only, resultant cleft one third the width of each plate (except for E. designatus). Six species: E. crotalinus, E. lampropeltis, E. compositus, E. heterodon, E. sipedon, E. designatus. 5. E. cophias group. Males: Anterior hamule talon- like, divided along its full length; posterior hamule broadly spatulate, surmounted by a small, black, curved anteapical tooth. Females: Vulvar lamina di- agonally corrugated; distal and lateral margins with a strongly raised semicircular ridge. Two species: £. boa, E. cophias. SYSTEMATIC PART Introduction Under each species, I have endeavoured to include a complete synonymy (some minor references may be lacking for E. designatus, a common species in the eas- tern United States), a full description, type data, diag- nosis, remarks, biology, distribution, and material examined. For new species, I have designated alloty- pes when the opposite sex was available, but I have not established allotypes for syntypic material. Under comments, I have quoted older authors (Ris, Selys, Hagen), when this information contributes to the his- tory of the species. For £. agkistrodon, I include a dis- cussion of what the male should be like, based on my cladogram. Unless otherwise stated, measurements are taken from 20 males and 20 females, when availa- ble, and from as many localities as possible. Abdominal measurements include caudal appenda- ges. Full locality data are given for all species, except for the following species where abundant material was available: E. eutainia, E. elaps, E. viperinus, E. de- signatus, E. |. lampropeltis, E. |. natrix, E. crotalinus, and E. compositus. For these species, I have listed sta- tes, counties, brief data, and repository. Full locality 185 Tij DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 data for these specimens are available from me. Maps represent distribution records from collections. For some common and well known species (such as £. bothrops, E. compositus, E. crotalinus, E. designatus), I include records from published sources I consider re- liable. Abbreviations for figure legends and synonymies are: abd. = illustration(s) of abdomen; app. = illustra- tions(s) of caudal appendages; cat. = listed in cata- logue, descr. = description of male, female, larva, or all, may include keys; distr. = documents or mentions distribution of taxon; ept. = illustration(s) of epi- proct; sep. = [page number of] separate (see Cowley (1937) for pagination of reprint of Selysian mono- graphs and synopses of Odonata); thx. = illustra- tion(s) of thorax; vl. = illustration(s) of vulvar lami- nae. Abbreviations in descriptions are as follows: anx = antenodal crossveins, pnx = postnodal crossveins, cs = crossveins. New names proposed in this paper follow the ophidian example set by Selys, Hagen, Calvert, and Williamson. All should be considered as nouns in ap- position. Most illustrations are from type material, which I believe will aid further researchers when describing new species. All illustrations were executed with the aid of a camera lucida with a Wild M-8 stereoscopic microscope. Erpetogomphus Hagen in Selys, 1858 Erpetogomphus Selys, 1858 (Type species Ophiogomphus cro- talinus (Hagen in Selys], 1854; designated by Kirby 1890: 61). — Selys 1858: 329 (69 sep.) (characters of genus); Selys 1859: 535 (9 sep.) (addition of generic characters); Hagen 1861: 98 (brief descr. of genus); Walsh 1863: 253 (comparison with other American gomphid genera); Brauer 1868: 372 (in key to genera of Gomphidae); Karsch 1890: 371 (mentioned in Gomphidae classifica- tion); Calvert 1905: 147 (key to Middle American Gomphidae); Calvert 1909: 468 (distr.); Calvert 1912a: 289 (distr. of genus); Kennedy 1917a: 544 (comments on larvae); Tillyard 1917: 296 (distr.); Ris 1921: 343 (com- parison with Mesogomphus [= Paragomphus)); Garman 1927: 125 (note on genus); Needham and Heywood 1929: 78 (descr.); Byers 1930: 52 (descr.); Cowley 1934: 241 (nomenclature of Erpetogomphus); Tinkham 1934: 218 (comments on distr. in Tex.); Fraser 1940: 544 (pos- sible affinities with Onychogomphus based on penis mor- phology); Needham 1941: 240 (comments on larva); Needham 1944: 172 (verification table, larva); Wright and Peterson 1944: 152 (in key to larvae of nearctic Gomphidae); Needham & Westfall 1955: 139 (treat- ment of genus); Pritchard & Smith 1956: 114 (key to lar- vae of nearctic Gomphidae, key to genera of Gomphidae); Ferguson-Beatty 1956: 369 (proventriculus of larva); Musser 1962: 13 (diagnosis of larvae); Gloyd 1963: 147 (molar structure of adult); Westfall 1984: 158 (in key to larvae of nearctic Gomphidae); Chao 1984: 79 186 (uncertainty of ÆErpetogomphus in subfamily Onychogomphinae); Carle 1986: 313 (characters of ge- nus within Onychogomphini); Belle 1988: 99 (in key to neotropical genera of Gomphidae); Bridges 1991: 11.17 (cat.); Novelo-G. & Pena-O. 1991: 129 (origin of Erpetogomphus in Mexico); Gonzalez-S. & Novelo-G. 1991: 97 (adults in Mexico); Novelo-G. & Gonzalez-S. 1991: 150 (larvae in Mexico); Belle & Quintero 1992: 93 (in key to adults and larvae of Panamanian Gomphidae); Carle 1992: 148 (key, Erpetocyclops subg. n. [type = E. ophibolus Calvert), Calogomphus subg. n. [type = E. eutai- nia Calvert]). Herpetogomphus Walsh, 1862: 388 (unjustified emendation of Erpetogomphus). — Hagen 1875a: 42 (cat.); Selys 1879: 63 (2 sep.) (modification of generic characters); Kirby 1890: 60 (cat.); Carpenter 1897: 452 (distr.); Needham 1897: 166 (in key to nearctic Gomphidae); Needham 1899: 234 (venation); Calvert 1899: 386 (comments on Selys’ classification); Needham & Hart 1901: 53 (in key to larvae of North American Gomphidae); Förster 1914: 73 (comparison with Ammogomphus); Seemann 1927: 19 (key to larvae, adults of western U. S. Gomphidae); Needham 1940: 389 (key, verification table). Medium sized to small neotropical Gomphidae of slender build. Head. — Eyes widely separated, characters of the mouth parts, clypeus, frons typical of the Gomphidae. Vertex variable, often interspecifically and, in the £. ophibolus group, sexually dimorphic. Anterior margin of vertex with a transverse trough, or a pair of transverse pits anterior or anterolateral to median ocellus, median ocellus anterior to lateral oc- elli, or in some members of the £. ophibolus group, this structure at posteriormost part of deep, longitu- dinal trough, thus causing median ocellus to be pos- terior to lateral ocelli (figs. 145-146); median surface of vertex at vicinity of lateral ocelli and postoccipital tubercle mostly planar, but members of E. ophibolus group with a complete postocellar ridge, which may be bilobed (E. agkistrodon, fıg. 148) or, in remainder of group, vestigial; occiput highly variable according to species and sex, E. ophibolus group with occiput small, semicircular, in other groups roughly trape- zoidal, with posterior part broader than anterior part, its dorsal surface roughly planar to nearly vertically inclined, some species with slight to strongly tumid area medially, one species (female of E. leptophis) with a posteriorly projected medial spine; rest of occiput variable, semicircular, linear, weakly to strongly notched medially, or sinuate; condition of crest often variable within a few species; postocciput variable, most in the E. ophibolus group convexly arcuate, in others slightly convex to linear, postocciput mostly not visible from above, but conspicuously so in others (e.g., female of E. compositus); lateral margins of one species (females of E. cophias) with transverse depres- sions. Thorax. — Prothorax and synthorax typical of the Gomphidae, but with great differences in maculation and colouration due to species and age differences. Pale colour pale green, apple green to vivid blue green, usually lighter ventrally and often with a slight dusting of white pruinosity ventrally and around cox- ae. Synthorax immaculate to heavily patterned with full complement of thoracic stripes. Legs moderately short, with femora slightly swollen and usually pale basally, with dark brown to black distally; tibiae in al- most all species black, prothoracic tibiae of males with small keel occupying distal 0.15 of mesal ventral margin. Wings (figs. 227-230) variable inter- and intraspe- cifically. Fore wing moderately narrow, with 11-21 antenodal crossveins, 5-16 postnodal crossveins, no subcostal crossvein; pterostigma prominent, about 4 times as long as wide, widest medially with its poste- rior margin slightly convex, surmounting 4-8 cross- veins, the proximal a brace vein; anterior margin of costa in many species pale, becoming dark at pteros- tigma; arculus at vicinity of second antenodal, the sectors not stalked, separated at base by distance greater than width of each sector; supratriangular, tri- angular, and subtriangular crossveins absent; paranal cells variable in number, often numbering 5-7, small- er species usually with zero or few supplementary marginal cells behind paranal cells, other species with irregular row of up to 6-7 supplementary cells; no ap- ical or medial planates, discoidal field of two rows. Hind wing with 7-14 antenodal crossveins, 7-14 postnodal crossveins; anal triangle of 2-4 cells, often with one of them in the latter condition small; anal loop absent (some specimens with a vestigial anal loop of 3 cells, but this condition rare); one paranal cell at Y interspace, two paranal cells at X interspace. Wing colouration entirely hyaline or flavescent at bases in some species. Hind wing length 21-36 mm. Abdomen cylindrical, narrowest medially in males, transverse carinae on segments 3-7; auricles well de- veloped in male, inner margin denticulate; auricle vestigial in female; male posterior margin of tergites 8 and especially 9 in a few species prolonged into a me- dial point, but this condition variable intraspecifical- ly; segments 8-9 moderately clubbed, with moderate foliate extensions; female with apical abdominal seg- ments hardly widened, foliar expansions small. Colour pattern and maculation variable between spe- cies and within sexes; all species with a pale green to pale orange middorsal stripe, sometimes extending full length of segment, but progressively less so on more posterior segments; dark brown to black dorso- lateral stripe present on first 7 segments, usually con- stricted or separated medially, giving most of abdo- men a serial dark and pale appearance; posterior part of segment 7, all of segments 8-10 tan to dark red brown, especially middorsally and ventrolaterally. GARRISON: Revision of Erpetogomphus Length of abdomen including appendages: d : 29-40 mm, 9: 29-41 mm. Abdominal appendages of male with cerci of great diversity according to species, but these structures al- ways as long as segment 10, cercus usually gently de- cumbent at apical 0.30; rarely with a dorsal append- age. Ventral carina present basally, or distally, or absent, depending on species, epiprocts divided at base but contiguous along most of their length, gent- ly to strongly curved at distal 0.50 to 0.30, depending on species; this structure lacking any accessory spines, or protuberances (except for £. ophibolus, fig. 95); tips of epiprocts assuming a wide variety of shapes and forms, and often variable intraspecifically (eg, E. bothrops, figs. 127-128). Vulvar lamina of female variable interspecifically, but never more than 0.50 length of segment 9, vulvar plates small, digit-like to forming large, contiguous, triangular or quadrangular plates, a moderate to large V- or U-shaped notch medially; postlamellar ridge al- ways present, semicircular in only two species (E. eu- tainia, E. leptophis), in all others forming an inverted Y; stem of Y originating before or often at level of vul- var plates, some species with a well defined semicircu- lar depression on each side of posterior stem. Male genitalia variable among species: anterior lamina not prominent; anterior hamule smaller than posterior hamule, usually with the distal end hidden from view by posterior lamina; anterior lamina dark, divided at various heights according to species, but with posterior branch usually less conspicuous, small- er, and anterior branch more prominent; posterior lamina usually pale, linear in two species (£. eutainia, E. leptophis); in all others with a moderately wider base, its tip usually adorned with an apical or anteap- ical hook directed cephalad or cephalolaterad. Penis with basal (first) segment prominent, with a semicir- cular rim, especially so laterally, posteriorly with a pair of prominent lobes, or these lobes vestigial; sec- ond segment moderately long, third short, with a pair of small, nipple-like sclerotizations dorsally in two species (E. eutainia, E. leptophis); fourth segment highly variable specifically, with or without a pre- puce, most species with a well defined lateral lobe; in most of the E. ophibolus group, lateral lobe small and pointed; membranous hood short to long and over- lapping; a pair of flattened membranous cornuae of various lengths and shapes present on apical part of segment; in most of the £. ophibolus group, cornuae absent or vestigial; penis guard short, quadrate in cross-section. A more complete discussion of the male and female genitalia is given under discussions of the species group accounts and in the cladistics section. 187 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Keys The keys should work for most specimens, but the worker may have difficulty when trying to deter- mine poorly preserved material. I have used morp- hological characters for both sexes wherever possi- ble, but some species pairs (eg, females of E. heterodon and E. sipedon, or E. tristani and E. sabale- ticus) show little or no morphological differentia- tion. For them, I have had to rely on colour pattern. Abdomens of some females, particularly those late- rally compressed, may need to be relaxed so that the vulvar lamina can be seen. Key to Erpetogomphus males 1. (1). Cercus with a prominent superior tooth at about 0.75 of appendage length (figs. 92-94) Cercus with dorsal surface convexly angulate (fig. 112), smoothly curved (fig. 103), straight (fig. 100), or with a concavity (fig. OES) ened ae 5 Epiproct smoothly curved, with tip pointing dorsally or posterodorsally, a small, sharp, anteriorly directed superior tooth at middle of epiproct (fig. 95). Southern Mexico, Belize absenten ee. ophibolus Epiproct strongly curved so that distal 0.30 is parallel to basal 0.30, tip of epiproct pointing anteriorly, superior surface of epi- proctiwith:nojtooth;;. i. rt 3 Ventral margin of cercus smoothly concave (fig. 92). Northeastern Mexico to Costa Rice e i Ma NI ai constrictor Ventral margin with a large ventral tubercle at 0.50-0.75 length of cercus (figs. 93-94). Costa Rica south to Colombia and Venezuela Sees as m atd 4 In lateral view, superior tooth of cercus as long as rest of appendage, so that appendage appears to end in two equal branches (fig. 93). Panama, Colombia, Venezuela ..... ERS a ben il oe pate ca sabaleticus In lateral view, superior tooth of cercus less than 0.50 as long as remainder of appendage (fig. 94). Costa Rica, northern Panama LD A ee tristani Metathoracic tibiae either entirely yellow ex- ternally (fig. 169), or yellow with median longitudinal black line. Western Mexico, southeastern Arizona ................... crotalinus Metathoracic tibiae entirely brown or black externally; never with any yellow (e.g, figs. 168, 170) Basal 0.25 to 0.30 of lower margin of cercus with a distinct longitudinal carina which 188 7 (6). 11 (10). 12 (10). may end in a small ventral tooth (e.g., E. co- phias fig;.116)...... teen. o 7 Basal part of cercus with no ventral carina, though a small tubercle may be present at basal 0.15-0.20 of appendage (e.g., figs. 99, TOD)... era EEE 14 With a dark stripe bordering posterior mar- gin of metepimeron (figs. 8-9). Southern Mexico, Belize through Costa Rica ............. aen nie pdl ai Nn eutainia (in part) Without a dark stripe on posterior margin of Metepimeron sneren 8 In lateral view, ventral margin of cercus with carina terminating in a distinct ventral tooth at basal 0.30 of appendage (fig. 116); thorax almost immaculate, at most barely an indi- cation of a dark antehumeral stripe (fig. 27). Michoacan and Morelos states of western Mexiconi.. nti RE cophias Ventral margin of cercus with longitudinal carina not terminating in a ventral tooth; thorax with well-defined dark antehumeral stripe, though it may be an elongated, isolat- ed spot (e.g., fig. 23) Dorsal surface of cercus distinctly angulate in lateral view (figs. 110, 112) Dorsal surface of cercus smoothly curved (fig. 112) Sides of thorax almost entirely green, with only a small, ill-defined dark humeral stripe; second lateral stripe incomplete, dark only below metathoracic spiracle, or absent; and third lateral stripe present only on upper 0.30 of suture (fig. 23), or absent (fig. 19)... Sides of thorax green with usual comple- ment of dark thoracic stripes; second lateral stripe may be incomplete, but humeral and third lateral stripes always present (figs. 17, 20-21) sehen ne ee ee 12 Dark antehumeral stripe incomplete, not touching humeral stripe (fig. 23); tip of cer- cus acute, but not gradually narrowing pos- teriorly (figs. 110-111); cornuae of penis without mesal lobes (fig. 87). New Mexico, western Texas, northern Mexico ................. Dark antehumeral stripe complete, usually connected to humeral stripe (fig. 19); tip of cercus drawn out into a narrow cylindrical point (figs. 105-106); cornuae with mesal lobes (fig. 83). Durango, Jalisco, Morelos, and Puebla states of central Mexico rt Talia alta ARA od aa Gn sipedon Tip of cercus strongly acuminate (fig. 104), dark antehumeral stripe not connected to 13 (9). 14 (6). 17 (16). collar (fig. 17) (except in some Mexican in- dividuals), base of wings flavescent (except in some Mexican individuals, fig. 18), me- dian area of occiput with a strongly raised tubercle (fig. 156). Eastern U.S. to Arizona, northern Mexico ........................ designatus Tip of cercus blunt, not acuminate (figs. 107-108); dark antehumeral stripe always connected to collar (ssp. natrix, fig. 21), or largely connected to humeral stripe (ssp. lampropeltis, fig. 20); base of wings hyaline, median area of occiput only slightly raised (fig. 160). Southern California, Arizona, New Mexico, western Texas, Baja California, Durango state, Mexico … … … … NE beb qu rit aie, corey lampropeltis Ventral carina at base of cercus black, con- trasting with remainder of appendage; this carina usually denticulate; thorax blue-green in life; sides of posterior hamules parallel, linear (fig. 52). Southern Texas to southern Mexico ame eutainia (in part) Ventral carina at base of cercus pale, the same colour as rest of appendage, this carina smooth; thorax pale green in life; sides of posterior hamules converging toward up, not linear (fig. 66). Southwestern United States Baja{California:. snaren: compositus Distal 0.5 of cercus concave dorsally, this ar- ea covered with thick, long bristles (fig. 115). Southern Veracruz state, Mexico … … Distal 0.5 of cercus straight or convex (e.g., figs. 99, 103) . Ventral margin of cercus straight (figs. 99, 100) or forming a gentle concave curve (fig. 98) so that appendage appears linear ..... 16 Apical 0.30 of ventral margin of cercus CURVEIAOWNWATE 2.222521. uses 18 . Thorax entirely yellow-green (fig. 13), or with only an antehumeral and faint indica- tion of a dark middorsal and antehumeral stripe (figs. 11-12); face entirely pale. Mexico south to Costa Rica. .................. 17 Thorax blue-green with full complement of dark thoracic stripes (fig. 10); face blue- green with extensive dark brown frontocly- pealisinipesBelize}. Arien leptophis Distal branch of anterior hamule thick, as high as gap separating the two branches (fig. 54); posterior hamules with no distal tooth (fig. 54). Guatemala south to Costa Rica .. EE ee «eek ete dan ae elaphe Distal branch of anterior hamule narrow, not as high as gap separating the two branches (figs. 55-56); posterior hamules 18(15). 19(18). 0 (19). GARRISON: Revision of Erpetogomphus with a distal tooth (fig. 55). Mexico to Guatemala IO, elaps Thorax blue-green, with full complement of dark stripes, including second lateral stripe and stripe on posterior margin of metepime- ron (fig. 6); face (fig. 142) predominantly dark, with a well defined blue green spot on frons. Guatemala, Costa Rica .......... schausi Thorax blue-green or yellow-green; sides lacking complete dark second lateral stripe (except in some viperinus, incomplete in some bothrops) and without dark stripe on posterior margin of metepimeron; face pre- dominantly pale, with dark areas limited to SULUTES AE N en 19 Tip of epiproct in posterior view terminat- ing in a blunt point (fig. 129); southern Veracruz state, Mexico................... viperinus Tip of epiproct in posterior view spatulate (fig. 126), broadly truncate (fig. 127), or bi- dentarei(ties] 28) see ee. en 20 Larger species (hind wing 30-31 mm); epi- procts in lateral view curved, extending 0.50 or less the length of cerci (fig. 101), epi- procts in posterior view truncate or slightly bidentate; dark antehumeral and humeral stripes combined or nearly so (fig. 14); pos- terior hamular tooth prominent, directed la- terocephalad or cephalad. Nuevo Leon, Michoacän, San Luis Potosi states, Mexico. Kerl) ee liopeltis Smaller species (hind wing 21-28 mm); epi- procts in lateral view curved, extending 0.75- 0.90 the length of cerci (fig. 102), epi- procts in posterior view bidentate or trun- cate (fig. 128); dark antehumeral and hu- meral stripes separate (fig. 15); posterior hamular tooth not as prominent, more strongly directed cephalad. Tamaulipas, Mexico, south to El Salvador (fig. 199) vnd ER PEST ibi. bothrops Key to Erpetogomphus females le Medial area of occiput with a posteriorly pointed protuberance (fig. 151) … leptophis Medial area of occiput variously a but never with a posteriorly directed protube- LANCER at RE sehe 2 Occiput with a shallow oval depression on each side posterior to elevated rim (fig. 165). Michoacán and Morelos, Mexico .... cophias Occipital area posterior to elevated rim en- tire, with no shallow depressions Outer surfaces of metathoracic tibiae entire- ly yellow, or yellow with a median longitudi- nal black line (fig. 169). Western Mexico, 189 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 5 (4). 10 (9). southeastern Arizona … crotalinus Outer surfaces of metathoracic tibiae entire- ly black (e.g., figs. 168, 170) .................... 4 Median ocellus posterior to lateral ocelli, the former lying within a deep longitudinal trough (figs. 145-146) Median ocellus at level of or anterior to lat- eral ocelli, no longitudinal trough ............ 6 Dorsal surface of vertex with sides of trough roughly parallel, or only slightly converging anteriorly (fig. 145). Northeastern Mexico, South to. Gostay canne cane constrictor Dorsal surface of vertex with sides of trough strongly converging anteriorly ......(fig. 146) Costa Rica, western Panama ........... tristani Central Panama, Colombia, Venezuela … CARRI RI CORI ee EE ebr sabaleticus Vertex with transverse ocellar ridge bilobed behind median ocellus (figs. 148-149) ..... 7 Vertex with transverse ocellar ridge entire (fig. 147), low and almost vestigial (fig. 150), or absent, its lateral ends forming oval tubercles posterior to lateral ocelli (e.g., fig. 150) Occiput transversally narrow, forming a shallow semicircle (fig. 148). Jalapa, Mexico REN REN IE EN Mai agkistrodon Occiput wide, forming a full semicircle (fig. 149). Guatemala, Costa Rica .......... schausi Vertex with transverse ocellar ridge entire, prominent, and concave posteriorly (fig. 147). Southern Veracruz, Mexico, Belize .... u en AO REEL. IND Bear ophibolus Vertex with median area bare, or at most on- ly slightly raised (e.g., fig. 150) Vulvar lamina followed on segment 9 by dis- tinct and prominent semicircular ridge, nev- er with a posteriorly directed arm (fig. 178). Southern Texas, eastern Mexico, Belize to Guatemala and Costa Rica ............ eutainia Vulvar lamina followed on segment 9 by a Y-shaped ridge (e.g., fig. 195) ................ 10 Vulvar lamina with each plate strongly cor- rugated diagonally, distal and lateral mar- gins of each plate forming a strongly raised semicircular ridge (figs. 195-196); a median tumid swelling on postocciput immediately behind occipital ridge (fig. 167) [includes /i- opeltis, but properly goes to 10’). Southern Veracruz: Mexico RER en. boa Vulvar lamina with each plate entire (fig. 185) or bisected diagonally by only one groove (fig. 186); distal and lateral margins of each plate planar, not forming a strongly raised semicircular ridge; postocciput not differentiated, evenly curved (fig. 166) [tu- 190 11 (10). IA): 13:02): 14 (13). 15:5). 16 (15). midsnvkopeltisihe....... es ener: = 11 Median surface of occiput with a strongly raised tubercle (fig. 156). Eastern United States west to Arizona, northern Mexico ..... EN RE NE designatus Median surface of occiput planar or at most slightly raised (fig. 152) Lobes of vulvar lamina separated by an al- most U-shaped interval 3 to 4 times as wide as either lobe (figs. 180-181) Mexico south to Guatemala Guatemala to Costa Rica … elaphe Lobes of vulvar lamina separated by a trian- gular or semicircular interval 0.5 to 1.0 the width of each lobe (e.g., figs. 182, 187) ...13 With a second complete dark lateral stripe on thorax (e.g., figs. 20-21, 24) .............. 14 Second dark lateral stripe completely lacking or vestigial, at most extending from base to just above metathoracic spiracle (e.g., fig. 36) Occiput in dorsal view narrow, its width less than width between median ocellus and oc- ciput; postoccipital area easily visible (fig. 163); base of wings with flavescent infusion between Sc and R at least up to first anx and cubitoanal area; top of abdomen with light areas ivory or with a tinge of orange, usually not different from white on sides. Southwestern United States, Baja California ee LR compositus Occiput in dorsal view wide, its width al- most equal to width between median ocellus and occiput; postoccipital area not visible (fig. 160); base of wings hyaline; top of ab- domen with light areas much darker than white on sides. Southwestern United States, Baja California, Durango state, Mexico ...... redes EN lampropeltis Occiput narrow, vestigial, with only a small medially arched area (fig. 153); post-occiput with a median tumid area (as in fig. 167). Nuevo Leon, Michoacan, San Luis Potosí states, MEXICO... RE liopeltis Occiput not especially narrow, extending full width (eg, figs. 154-155, 159, 162); post-occiput planar or, at most, only slightly convex medially (as in fig. 166) ............. 16 Stem of postlamellar ridge with a well de- fined circular or oval depression on either side; junction of Y of postlamellar ridge usu- ally posterior to hind margin of vulvar lami- nariplatesi(fig:-183)\ seen... 17 Stem of postlamellar ridge with only a slight depression or, more often, without any cir- cular or oval depression on either side; junc- 17 (16). 8 (16). tion of Y of postlamellar ridge at or anterior to hind margin of vulvar laminar plates (figs. FANS MOD). AE er. 18 A deep pit at anterior margin of frons ante- rolateral to median ocellus (fig. 155); cleft between vulvar laminar plates wide, forming an obtuse arc greater than 100°; dark hu- meral and antehumeral stripes combined or nearly so (fig. 16); larger species (hind wing 31-34 mm). Southern Veracruz, Mexico … INR EA EREN viperinus Anterior margin immediately anterior to median ocellus forming a narrow V-shaped trough with base of postfrons (fig. 154); cleft between vulvar laminar plates narrow, form- ing a V- shaped notch of not more than 90° (fig. 183); dark humeral and antehumeral stripes separate (fig. 15); smaller species (hind wing 17-31 mm). Tamaulipas, Mexico, south through El Salvador … … …. Ec leten deed ke bothrops Dark antehumeral stripe complete, usually connected at its upper end to narrow hu- meral stripe (fig. 19). Jalisco, Morelos, and Puebla Mexico in... sipedon Dark antehumeral stripe vestigial, forming an isolated elongate spot, not connected to vestigial humeral stripe (fig. 23). Southwestern United States, northern Nex Come PENS Mer rh heterodon Key to Erpetogomphus males based primarily on characters of abdominal segments 2 and 3 I include this key for male specimens lacking the posterior abdominal segments. The key should allow a high probability of identification, though some spe- cies (E. ophibolus, E. sabaleticus, E. tristani, E. constric- tor, for example) appear inseparable by penis struc- ture and are distinguished only by the caudal appendages. For those species, I have added other so- matic characters and/or geographic data with which to separate them. The key does not include E. agkis- trodon, the male of which is unknown. 1. Ental surface of penis with a prepuce (e.g., fig. 76), this structure reduced and partially hidden by frill-like lateral lobe in leptophis (fig. 76) and eutainia (fig. 75); thoracic pale colouration in life deep blue green, dark tho- racic stripes always present on sides (e.g., fig. UO oc PAE EEL MERTEN 2 Ental surface of penis lacking a prepuce (e.g., fig. 90); thoracic pale colour in life yellow green (except for viperinus and probably /io- peltis); dark lateral thoracic stripes often (but not always — figs. 16, 18) reduced or absent GARRISON: Revision of Erpetogomphus Posterior hamule digit-shaped (figs. 52-53), its tip broadly rounded; anterior hamule di- vided at upper 0.30 to 0.25, its posterior branch well defined, its tip almost meeting larger anterior branch (figs. 52-53); lateral lobe of penis composed of large spinulose frill (figs. 52-53); cornuae well developed, theinapicesipointede nen Me 3 Posterior hamule triangular, its tip armed with a spine (fig. 47); anterior hamule with only a posterior shoulder (e.g., fig. 47); later- al lobe of penis reduced to a small, semicir- cular, spinulose frill (fig. 74) or vestigial (fig. 70); cornuae vestigial (fig. 74) or present (fed) NAS N Danni ne 4 Membranous hood of penis about 3 times as long as wide, apices overlapping (fig. 76) [Note: this character is based on the only known specimen, the holotype; other speci- mens may show variability similar to eutai- nia.); dark second and third lateral stripes connected at upper ends, forming an isolat- ed pale spot below subalar carina; dark later- al stripe along posterior margin of metepim- eron well defined (fig. 10); caudal appendages as in fig. 98; Belize …. leptophis Membranous hood of penis reduced, less than 2 times as long as wide, though apices may overlap (fig. 75); dark second and third lateral thoracic stripes not connecting (fig. 7-9); dark lateral stripe along Eee mar- gin of metepimeron absent (fig. 7) or present primarily in populations of central Mexico (fig. 8) south through Costa Rica (fig. 9); caudal appendages as in fig. 97 Lbs, SCAN ALE CREER dn Nei, eutainia Cornuae of penis present; their tips diver- gent (fig. 74); lateral lobe a small, circular, spinulose frill (fig. 74); Guatemala and Costa Rica mr Ar En a schaust Cornuae vestigial (fig. 71); lateral lobe vesti- gial, forming a small, acute point (fig. 71) … Appendages as in fig. 93; Central Panama south to Colombia and Venezuela EA EAR ROM. ROES sabaleticus Mexico south through Costa Rica and nonthernsBanam ane EN MIER Aa 6 Epiproct smoothly curved, with tip pointing dorsally or posterodorsally, a small, sharp, anteriorly directed superior tooth at middle of epiproct (fig. 95). Southern Mexico, Bere tat AN ophibolus Epiproct strongly curved so that distal 0.30 is parallel to basal 0.30, tip of epiproct 191 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 9 (8). 10 (8). 11 (10). pointing anteriorly, superior surface of epi- proct with no OO ER 7 Appendages as in fig. 92; Mexico soutl through Costa Rica .................... constrictor Appendages as in fig. 94; northern Costa Rica through northern Panama ...... tristani Anterior hamule divided at base; its branch- es resembling a claw; posterior hamule broadly spatulate (figs. 68-69) Anterior hamule divided at lower 0.30 to upper 0.25 posterior arm short and truncate or acute; posterior hamule broadly triangu- Tan re LEE 10 Dark antehumeral and humeral stripes unit- ed, forming a broad, dark stripe over first lateral suture (fig. 26); southern Veracruz, Mexiconsan he st ea. PA boa Thorax almost entirely pale, with only a ves- tigial antehumeral stripe (fig. 27); Michoacán and Morelos states of Mexico … Mt Ale teal Fe gle cooler dle ide cu cophias Lateral lobe of penis linear, at least 3 times longer than wide; its posterior margin spinu- losei(Gies M/S 82Merc) Pet. 11 Lateral lobe of penis round, about as long as wide, its margin spinulose (fig. 86) ........ 16 Lateral lobe of penis almost recumbent against ental margin of fourth segment (fig. 82); anterior hamule divided at distal 0.5 of segment; posterior hamule lacking a distal tooth (fig. 60); posterior margin of first seg- ment (peduncle) of penis strongly cleft, each lobe on either side of cleft protruding well beyond posterior margin of segment; south- ern United States west to Arizona, northern Meier ha lee designatus Lateral lobe of penis prominent, directed al- most perpendicularly to ental margin of fourth segment; anterior hamule divided at distal 0.30 of segment, posterior hamule with a distal tooth (except for elaphe from Guatemala south through Costa Rica); pos- terior margin of first segment of penis weak- ly bilobed, lobes on either side of emargina- tion only extending beyond posterior margin of segment; northern Mexico south through Costa Rica ......... 12 slightly . Distal branch of anterior hamule thick, as tall (high) as gap separating the two branch- es (fig. 54); posterior hamule with no distal tooth (fig. 54); Guatemala south to Costa RCA. portes ee teu elaphe Distal branch of anterior hamule narrow, not as tall (high) as gap separating the two branches (figs. 55-56); posterior hamule withaldistalitoocher Sr ti a 13 13 (12). 14 (13). 15 (14). 16 (10). 17 (16). 18 (17). Distal tooth of posterior hamule bent anteri- orly so that it lies in same direction as the planar surface of hamule (figs. 55-56); Mexico south to Guatemala ............... elaps Distal tooth of posterior hamule bent later- ally or anterolaterally so that it points distal- ly and perpendicularly from planar surface of hamule (fig. 59) [Note: This condition appears universally only in £. viperinus, but variation in this condition occurs in E. both- ropsand E. liopeltis some specimens of these last two species may not be separable from E. elaps by this character]; Mexico south through Guatemala Base of postfrons medially with a pair of deep pits, each anterolateral to median ocel- lus (fig. 137). Southern Veracruz, Mexico .. HE REEL se dors ee NÉS viperinus Base of postfrons medially with a transverse arcuate trough immediately anterior to me- dian ocellus (hg. 130) 15 Smaller species (hind wing 21-28 mm); dark antehumeral and humeral stripes separate (fig. 15); posterior hamule with apical tooth shorter, less prominent, more strongly curved laterally over tip of hamule. Tamaulipas, Mexico, south through EI Salvador (fig. 190) presse bothrops Larger species (hind wing 30-31 mm); dark antehumeral and humeral stripes connected (fig. 14); posterior hamule with apical tooth longer, more linear, not as strongly curved laterally over tip of hamule. Nuevo Leon, Michoacan, San Luis Potosi, Mexico .......... ER aie nee liopeltis Cornuae of penis with tips long, narrow, and attenuate; medial lobe present (fig. 86); extensor surfaces of tibiae yellow, or yellow with a median longitudinal black line; south-eastern Arizona south through central Mexico: cae crotalinus Cornuae of penis with tips bluntly rounded (figs. 87-88); medial lobe present (figs. 83- 84) or absent (figs. 87-88); extensor surfaces of tibiae black, never with yellow … … … 17 Tips of cornuae bluntly rounded; medial lobe absent (figs. 87-88) Tips of cornuae with a well defined medial lobe (figs. 83-84) Median part of base of postfrons with a transverse arcuate trough immediately ante- rior to median ocellus (as in fig. 138); west- ern United States. ee compositus Base of postfrons medially with a pair of deep pits, each anterolateral to median ocel- lus (as in fig. 138); western Texas, western 19 (17). Penis in dorsal view with membranous hood short, exposing a subcutaneous membra- nous hood which forms the posterior margin of segment [be sure to examine closely: rota- tion of penis to dorsolateral view will reveal a raised hood anterior to subcutaneous membrane], (fig. 83); Jalisco, Morelos, and ively las Mexico... sipedon = Penis in dorsal view with membranous hood long, completely covering subcutaneous membrane, thus forming posterior margin of segment (figs. 84-85); western Texas, southern and central New Mexico and Arizona, southwestern California, Baja California, Durango state, Mexico … … … … SRO AA EROI see indies lampropeltis Erpetogomphus constrictor Ris (figs. 1-thx, 28-abd, 97-hamules, 70-penis, 92-app, 145-@ vertex, 171-vl, 198-distr) Erpetogomphus constrictor Ris, 1917: 154 (‘Nach 3d, 19 von Misantla, Vera Cruz, Mexiko, VI 1911, die ich dem schweizerischen Sammler W. Gugelmann verdanke.’). — Williamson and Williamson 1930: 12 (summary of stat- us); Montgomery 1973: 239 (derivation of name); Paulson 1982: 255 (Mex.); Davies and Tobin 1985: 27 (cat.); Tsuda 1986: 87 (cat); Dunkle 1988: 46 (Honduras); Maes et al. 1988: 36 (Nicaragua); Bridges 1991: VII.51 (cat.); Tsuda 1991: 95 (cat.). Description Male. — Labium grey, becoming dark grey medial- ly, labrum entirely brown to grey green with dark brown margin and large inverted medial triangular spot; base of mandibles grey green, anteclypeus grey green; postelypeus brown except for small triangular green spots above lateral lobes; ventral margin of frons brown joining brown of postclypeus, thus for- ming a large brown frontoclypeal stripe; remainder of frons green; extreme base of frons, vertex, and occiput dark brown; area immediately anterior to median ocellus with a shallow longitudinal trough, postocel- lar tumid areas prominent, a lower incomplete trans- verse ridge posterior to median ocellus but not conti- nuous with each postocellar tumid area laterally; occiput planar, its crest costate, linear, fringed with dark brown hairs; postocciput brown, linear to gently concave medially; rear of head brown. Prothorax brown, paler dorsally, anterior and pos- terior lobes green, some specimens with two small green midlateral spots (almost touching) on median lobes; synthorax (fig. 1) with prominent, well defined dark stripes on green background as follows: middor- sal stripe narrowing anteriorly at collar, costal mar- GARRISON: Revision of Erpetogomphus gins of antealar sinus and area along this structure connecting with humeral stripe; antehumeral stripe connecting with humeral basally and dorsally, but its upper end often not touching brown below antealar crest, or, if so, then isolating small area of green; sec- ond and third lateral stripes connecting at upper 0.25, isolating small green spot, second lateral stripe swollen in vicinity of metaspiracle and, in some spec- imens, enlarged and touching third lateral stripe; pos- terior margin of metepimeron, metasternum, and venter of thorax grey green to tawny. Venter of profe- mora grey green, remainder of femora dark brown, becoming black distally; tibiae, tarsi, armature black. Wings hyaline, venation and pterostigma black. Venational statistics. Fifth (rarely fourth or sixth) antenodal thickened; number of marginal cells be- hind fore wing paranal cells: 0-1/0-1; anx: fore wing 13-17/14-17, hind wing 10-12/10-13; pnx: fore wing 10-14/9-14, hind wing 9-12/9-13; cs under pteros- tigma.: fore wing 5-7/4-7, hind wing 4-7/5-7; anal triangular cells: 3- 5/3-4. Hind wing 25-30 mm. Abdomen with segment 1 brown, green laterally and with a green middorsal stripe; segment 2 brown except for green auricles, ventral margin, and narrow middorsal stripe, annulus black; segment 3 primarily black with green middorsal stripe extending from black annulus of segment 2 and ending at distal annu- lus, a pale grey green anterolateral spot becoming nar- rower along ventral margin and disappearing at 0.25 to 0.75 of segment length; segments 4-6 similar to segment 3 but with pale middorsal stripe ending at posterior 0.10 to 0.25 of segment, segment 6 in some specimens with anterolateral white connecting dorsal- ly with pale green middorsal stripe; segment 7 grey green on anterior 0.50 to 0.60, except for dark trans- verse carina, posterior part of segment brown to black; segments 8-10 red brown, becoming darker dorsally and ventrally, especially along foliate mar- gins, posterior medial margin of segment 9 bluntly pointed or slightly so. Abdomen 31-35 mm. Cercus (fig. 92) grey green, becoming dark brown posteriorly, arcuate with a prominent laterally com- pressed dorsal tubercle on distal 0.75 of cercus, ex- treme base with a black ventral tooth (mostly hidden by posterior margin of abdominal tergite 10), a ven- tral carina on distal 0.25 of cercus, tip of cercus with a black tooth; epiprocts dark brown, strongly curved at distal 0.50 so that distal 0.5 of appendage is par- allel, distal 0.30 sulcate medially on exterior surface. Accessory genitalia. Hamules (fig. 47) small; anterior hamule brown, divided at distal 0.30, posterior branch a well developed shoulder; posterior hamule roughly triangular, its tip with a spine; penis (fig. 70) with cornuae slightly bilobate or transverse at base, a well developed prepuce, lateral lobe a small tooth. Female. — Head markings similar to male but with 193 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 dark median spot on labrum larger, often connecting with ventral brown, thus separating green into two lateral spots; some specimens with an entirely brown labrum; brown on frontoclypeal suture more exten- sive laterally, so that entire postclypeus may be brown; brown at base of antefrons with a median ex- tension which, in some specimens, connects with brown frontoclypeal suture, thus dividing frontal green into two spots. Morphology of vertex and occi- put differs from male as follows: a large longitudinal trough on vertex with median ocellus recessed poste- rior to lateral ocelli (fig. 145); bottom of median fur- row with longitudinal convex area; lateral margins of trough formed by median ocellar protuberances con- necting posteriorly forming a semicircular ridge; OCCI- put small, forming a narrow semicircle, crest covered with long brown hairs; postocciput brown, convex. Pro- and synthorax as in male, but with dark areas more extensive. Venational statistics (n = 15). Fifth antenodal thickened in all wings; number of marginal cells be- hind fore wing paranal cells: 0-5/0-3; anx: fore wing 14-17/14-17, hind wing 11-12/11-12; pnx: fore wing 11-14/11-14, hind wing 10-14/11-13; cs under pte- rostigma.: fore wing 5-7/5-7, hind wing 5-7/5-7. Hind wing 28-31 mm. Abdomen (fig. 28) with segment 1 brown, a green spot posterolaterally and with a complete green mid- dorsal stripe; segment 2 brown except for following green areas: wide middorsal stripe narrowing posteri- orly, longitudinal lateral stripe connecting just anteri- or to auricle and narrowing slightly to posterior mar- gin of segment, a thin pale margin at venter of segment; segment 3 similar to segment 2 but lateral pale stripe ending posteriorly at transverse carina, middorsal green narrowly acuminate posteriorly, forming a mere line on posterior 0.25 of segment; segments 4-6 similar to segment 3 but with middor- sal green progressively shorter posteriorly, lateral pale stripe white and forming a small quadrangular spot at anteroventral margin of segment; segment 7 with an- terior 0.50 pale green, interrupted by black transverse carina, posterior 0.50 becoming red brown; segments 8-10 all red brown, often darker dorsoposteriorly; cerci pale grey green. Vulvar lamina (fig. 171) with plates broadly connected at basal 0.75, relatively un- specialized, cleft obtusely V- or U-shaped, about as wide as each lobe; postlamellar ridge with Y-juncture posterior to plates, base of stem well defined and with oval depression laterally. Abdomen 30-35 mm. Diagnosis Males of E. constrictor are easily separated from its closest allies, £. tristani and E. sabaleticus, by the sha- pe of the cercus. This structure is concave ventrally in È. constrictor, not convex as in the other two species. 194 It superficially resembles Æ. schausi in body coloura- tion and is diagnosed under that species. Females seem to be separable from the same two species by the width of the median trough of the ver- tex. This structure is slightly divergent anteriorly in £. constrictor (figs. 145), but is largely parallel in the oth- er two species (fig. 146). Remarks Variation. — Venational details of lectotype male of E. constrictor. No marginal cells behind fore wing pa- ranal cells; anx: fore wing 15/14, hind wing 11/11; pnx: fore wing 11/12, hind wing 12/13; cs under pte- rostigma: fore wing 7/7, hind wing 7/6; number of anal triangular cells: 4/3. Hind wing 29 mm. Material I examined indicates that specimens from more southerly parts of its range (Guatemala, Honduras, Nicaragua) are more melanic than those from eastern Mexico. For example, black markings on the labrum, postclypeus, and thoracic striping are generally more extensive, further reducing or isolating green areas. Two females, one from San Luis Potosi (El Salto Falls) and the paralectotype are unusual in having 5/3 and 4/3 marginal cells behind fore wing paranal cells in right and left fore wings, respectively. All other fe- male specimens ranged from 0-1 marginal cells. Biology. — E. González (in litt.) found this species to be abundant at the Rio Huichihuayan. Males perched on leaves, vines, and snags at the edge of the river. Judging from specimens in collections, E. con- strictor is the northern counterpart of E. tristani, and appears to be common in certain areas. Erpetogomphus constrictor appears to be a lowland species of wet tropical forest habitats along the east coast of Mexico south through Costa Rica (Guanacaste Prov.: Miravalles, 26 April 1991 (C. Esquivel), 39, A. Ramirez, in lit). Altitudes range from near sea level (70 m, San Luis Potosi, km 410) to about 550 m (Nuevo Leon, entrance to Horsetail Falls). Collection dates range from 26 April to 4 October. One specimen was collected in an ultravio- let light trap. Distribution (fig. 198). — This species occurs along the eastern Mexican states of Nuevo Leon, Tamaulipas, San Luis Potosi, Veracruz, south through Guatemala, Honduras, and Nicaragua, to Guanacaste Prov., Costa Rica. It is apparently parapa- tric or allopatric with £. tristant in Costa Rica. Material Type data. — Lectotype d by present designation with the following data in Ris’ hand: ‘15098’ in green ink in an unknown hand, Erpetogomphus | n. sp. 3 / constrictor Ris 1917/ Misantla, Vera Cruz / Mexico VI 1914 / W. Gugelmann. Two paralectotype males and one female with same data. One male is labelled ‘15092’ in green ink, another male in poor condition (head partially fragmented, abdomen broken in sever- al places) is labelled ‘Type’ probably by Ris. The left pair of wings is missing and probably served for figu- re 6 in the original description. The female is labelled ‘15096’ in green ink. In smr. Other material (534, 159, including lectotype d and paralectotypes). — Mexico: Nuevo Leon: Cola Caballo, SW Monterrey, 6 July 1960 (W. B. Cutts), 19 (ummz); Rio Ramos, 3 km S of Allende, 500 m, 7 Sept. 1963 (T. W. Donnelly), 16 (rw); south of Monterrey, Rio Elizondo, 19-20 June 1965 (O. S. Flint, Jr.), 64 (USNM, RWG); stream at entrance to Horsetail Falls, 550 m, 25 June 1965 (D. R. Paulson), 16 (pre); Tamaulipas: Gomez Farias, Rio Frio, 4 Oct. 1985 (F. Arias, L. Cervantes, M. Garcia), ‘colectado en trampa U.V. en la noche’, 1d (rwe); San Luis Potosi: El Salto, 4 June 1967 (O. S. Flint, Jr.), 19 (usnm); Palitla, N of Tamazunchale, 25 June 1965 (O. S. Flint, Jr.), 96 (UsNM, RWG); Huichihuayan, rte 85, km 399, 7 Aug. 1966 (O. S. Flint, Jr.), 5d (usnM, cc, TwD); stream crossing Hwy 85 (1000); 6 mi N of Tamazunchale, 4 Sept. 1957 (G. H. Beatty, III), 16 (rsca); km 410, ‘Sam Brown’ Hacienda, 250 ft, 27 Sept. 1938 (L. J. Lipovsky), 16 (ummz); Route 85, km 399, Huehuetlän, 26 June 1965 (O. S. Flint, Jr.), 45,1% (usnm); Rio Huichihuayan, Mpio de Huchuetlan, 100 m, 26 May 1987 (E. González-S, 25 (rwe); upstream Cascadas Micos just S of town N of Ciudad Valles off Hwy 85, 27 June 1990 (J. S. Daigle), 1d (RwG); GUATEMALA: Suchitepequez Dept., Cuyotenango, Finca San Rafael Olimpo, 1700 ft, 10-20 June 1966 (O. S. Flint, Jr., M. A. Ortiz-B.), 88 (USNM, RWG); HONDURAS: Comayagua Dept., Rancho Chiquito, 11.8 mi SE Villa de San Antonio, 2-3 Aug. 1967 (O. S. Flint, Jr., M. A. Ortiz-B.), 38,7? (usnm, CC, RWG); Francisco Morazan Dept, 30 km E of Tegucigalpa, 13 June 1982 (Fernandez, Garcia), 19 (swp); [no date], (Carlos Jaramillo), 19 (swp); EL SALVADOR: Ahuachapan Dept., Loma de Paya, Bosque El Impossible, San Francisco Menendez, 24 July 1987 (A. Campus), 14 (va); 20 Aug. 1986 (V. Hellebuyck), 19 (vx) 4 Aug. 1987 (V. Hellebuyck), 19 (vri); NICARAGUA: Chobales Dept., La Flor, rte 7, km 159, 4 mi W of Acoyapajit, 29 July 1967 (O. S. Flint, Jr, M. A. Ortiz- B), 45 (usnm); Managua Dept. stream along Pan American Hwy, 7 km S of Nandaime, 21 June 1962 (T. W. Donnelly), 1d (twp). Erpetogomphus sabaleticus Williamson (figs. 2-thx, 48-hamules, 71-penis, 93-app, 172-vl, 198-distr) Erpetogomphus sabaleticus Williamson, 1918: 1 (descr. holo- type d, allotype 9). — Williamson and Williamson 1930: 12 (summary of status); Paulson 1982: 255 (Panama, South America); Davies and Tobin 1985: 28 (cat.); Tsuda 1986: 87 (cat.); De Marmels 1990: 338 (Venez.); Bridges 1991: VII.184 (cat.); Tsuda 1991: 95 (cat.); Belle 1992: 32 (descr. larva); Donnelly 1992: 85 (Panama); Belle and Quintero 1992: 99 (key, Panama). Description A thorough comparative description of the holoty- pe male and allotype female is given by Williamson GARRISON: Special codes for printers (1918). I illustrate the thorax (fig. 2), caudal appen- dages (fig. 93) and details of the accessory genitalia (fig. 48) of the holotype, and the vulvar lamina of the allotype (fig. 172). Male (n=6, including holotype). — Venational sta- tistics. Fifth (rarely sixth) antenodal thickened; num- ber of marginal cells behind fore wing paranal cells: 0/0; anx: fore wing 13-18/15-17, hind wing 10-12/9- 12; pnx: fore wing 11-13/15-16, hind wing 11- 14/11-13; cs under pterostigma: fore wing 5-6/6-7, hind wing 5-7/6-7; anal triangular cells 3-4/3-4. Hind wing 26-27 mm. Abdomen 30-33 mm. Female (n = 4, including allotype). — Venational statistics . Fifth or sixth (rarely fourth) antenodal thickened; number of marginal cells behind fore wing paranal cells: 0-1/0-1; anx: fore wing 15-17/15-16, hind wing 11-12/9-12; pnx: fore wing 11-15/13, hind wing 10-12/11-13; cs under pterostigma: fore wing 6-7/6-7, hind wing 6-7/6-7. Hind wing 28-29 mm. Abdomen 31-32 mm. Diagnosis Erpetogomphus sabaleticus is diagnosed under E. tristani. Remarks Variation. — Venational details of holotype male: number of marginal cells behind fore wing paranal cells: 0/0; anx: fore wing 15/16, hind wing 11/11; pnx: fore wing 13/13, hind wing 12/11; cs under pte- rostigma: fore wing 6/6, hind wing 6/6; anal triangu- lar cells: 3/3. Hind wing 26 mm. Venational details of allotype female: number of marginal cells behind fore wing paranal cells: 1/0; anx: fore wing 16/15, hind wing 12/11; pnx: fore wing 12/13, hind wing 10/11; cs under pterostigma: fore wing 6/6, hind wing 6/7. Hind wing 29 mm. The living colours of the holotype male were re- corded by Williamson (1918). Biology. — Williamson (1918) collected the allo- type female and holotype male along the upper San Juan near Maraquita in Colombia. The male was tak- en along the Quebrada Sabaleticus as it rested on a flat leaf about five feet above the water. Only one oth- er specimen of this species was seen (but not collect- ed) during their stay in Colombia. A description of the type locality is given by Williamson (1918). Michael May collected one male and three female larvae of the species at the Quebrada Juan Grande, along Pipeline Road in the Panama Canal Zone on 28 January 1975 and 23 January 1977. These were brought back to Gainesville, Florida, where the adults emerged about three months later. This species, like E. tristani, is apparently rare. Other odonatists and I have collected along Pipeline Road in Panama, but have failed to find the species. 195 TlIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Distribution (fig. 198). — Erperogomphus sabaleticus is the most austral of the genus and is currently known from one locality in northwestern Venezuela, one locality in northern Colombia, and Panama. It is apparently the only member of the genus found in the first two countries. Its northern limit is unknown, but it is probably parapatric with the more northerly E. tristani. Collection dates range from 3 February (Maraquita, Colombia) through 24 June (Panama Canal Zone). Material Type data. — Holotype male: CoLomBia: Dept. Antioquia: Cristalina, 28 km on railroad above Puerto Berrio, 19 Feb. 1917 (J. H. and E. B. Williamson); allotype female: CoromBia: Dept. Tolima: Maraquita, 3 Feb. 1917 (J. H. and E. B. Williamson). Both specimens in UMMZ. Other material (6d, 49, including holotype 4 and allotype 2). — PANAMA: Canal Zone. Quebrada Juan Grande, Pipeline Road, 28 Jan. 1975 (M. L. May), 19 larva, emerged 19 June 1975 (FscA); same data but 23 Jan. 1977 (rsca), 16, 22 larvae, male emerged 16 May 1977, females emerged 20, 26 May 1977; Pipeline Road, 1.7-4.8 mi NW of Gamboa, 2nd bridge, 24 June 1970 (E. S. Morton), 1d (rwp); VENEZUELA: Tachira Dept: Tachira, 11 April 1920 (J. H. and E. B. Williamson, W. H. Ditzler), 38 (24 in UMMZ, ld in ESCA). Erpetogomphus tristani Calvert (figs. 3-thx, 49-hamules, 72-penis, 94-app, 120-ept, 146-vertex, 173,174-vl, 198-distr) Erpetogomphus tristani Calvert, 1912a: 290 (descr. of d, 2). — Calvert 1912b: 384 (mentions mating adaptations); Williamson 1918: 1 (comparison with £. sabaleticus); Ris 1917: 154 (Panama); Ris, 1918: 154 (Panama); Calvert 1920b: 339 (mentions mating adaptations); Williamson and Williamson 1930: 12 (summary of status); Paulson 1982: 256 (Costa Rica, Panama); Davies and Tobin 1985: 28 (cat); Tsuda 1986: 87 (cat); Bridges 1991: VIL212 (cat.); Tsuda 1991: 95 (cat.); Belle and Quintero 1992: 100 (key, Panama). Description Male. — A thorough description of the holotype male and allotype female is given by Calvert (1912). I illustrate the caudal appendages of the holotype (fig. 94) and vulvar lamina of the allotype (fig. 173). In addition to Calvert’s description, I add the following: À strong ventral tooth at base of cercus (plainly seen in lateral view, fig. 94), which in most specimens is partially hidden by lateroposterior margin of tergite 10; tips of epiprocts narrow in posterior view (fig. 120); anterior hamule black, divided at apical 0.25, lower branch a rounded shoulder; anterior (upper) 196 branch well developed; posterior hamule pale, acutely triangular, with a prominent apical tooth; penis with well developed prepuce, which, when viewed lateral- ly, is hidden by ventrolateral margin of segment 4; la- teral lobe small, rudimentary, barely visible, and with spinulose margin; cornua rudimentary, forming an ental membranous plate with a quadrate margin (fig. 72). Abdomen 31-33 mm. Venational statistics. — (n = 5, including holotype). Fifth antenodal thickened; number of marginal cells behind fore wing paranal cells: 0-1/0; anx: fore wing 13-15/14-15, hind wing 10-12/10-11; pnx: fore wing 11-13/11-13, hind wing 11-13/10-12; cs under pte- rostigma: fore wing 5-6/5-6, hind wing 5-6/5-6; anal triangular cells: 3/3. Hind wing 25-27 mm. Female. — Vulvar lamina (fig. 174, allotype; 173) broadly connected, relatively unspecialized, and with a gently U-shaped notch; basal plate poorly devel- oped; juncture of Y-shaped postlamellar ridge posteri- or to posterior margin of lamina; well defined arcuate depression on either side of central stem. Abdomen 31-33.5 mm. Venational statistics. — (n = 4, including allotype). Fifth antenodal thickened; number of marginal cells behind fore wing paranal cells: 0-1/0; anx: fore wing 15-16/15-17, hind wing 11-12/10- 12; pnx: fore wing 12-14/11-12, hind wing 12-13/12-13; cs under pterostigma: fore wing 5-7/6-7, hind wing 6-7/5-6. Hind wing 29 mm. Diagnosis Erpetogomphus tristani, E. sabaleticus, and E. cons- trictor form a compact group which is easily separable from all other congeners by the unique shape of the epiprocts of the male (figs. 92-94, 120) and by the morphology of the vertex of the female (figs. 145- 146). Males of E. tristani differ from the other two species only in the shape of the cerci. The ventral margin is distinctly convex at the apical 0.60-0.80 in E. tristani (fig. 94) and E. sabaleticus (fig. 93), not smoothly concave as in E. constrictor (fig. 92). All th- ree species have a prominent dorsal tooth near the distal end, but the tooth in £. constrictor is broader in lateral view and more vertical than in £. tristani and E. sabaleticus. Males of E. tristani and E. sabaleticus approach one another closely. Williamson (1918: 4-5) found males of both species inseparable except by the male caudal appendages. Williamson sent a pair of E. sabalecticus to Calvert, who also supported the statement of dif- ferences between the males of both species. Calvert was unable to separate the females of these two spe- cies. I have found the females of these three species (E. constrictor, E. tristani and E. sabaleticus) to be ex- tremely similar, more so than their males. The sides of the longitudinal trough of the vertex of £. constric- tor are more parallel (fig. 145), than those of the oth- er two species. Like Calvert and Williamson, I have not found any differences between females of the small series of £. tristani and E. sabaleticus. Remarks Variation. — Little variation is expressed among the small series (males, n = 5; females, n = 4, including holotype and allotype). Venational details of holotype male: number of marginal cells behind fore wing paranal cells: 0/0; anx: fore wing 14/14, hind wing 11/11; pnx: fore wing 12/12, hind wing 13/12; cs under pterostigma: fore wing 5/5, hind wing 5/5. Hind wing 27 mm. Venational details of allotype female: number of marginal cells behind fore wing paranal cells: 0/0; anx: fore wing 15/15, hind wing 11/11; pnx: fore wing 14/12, hind wing 13/12; cs under pterostigma: fore wing 7/7, hind wing 7/6. Hind wing 29 mm. Biology. — The species is relatively rare in collec- tions, despite intensive collecting in Costa Rica in re- cent years. Collection dates are February, June, and July. Nothing else is known of its biology. Distribution. (fig. 198). — Erpetogomphus tristani is thus far known from only two northwestern provinc- es of Costa Rica and Panama (Lino, Chiriqui Prov. at about 1600 m, 8°48’N, 82°26W, 16, 1 ©, Ris 1917). It is replaced to the north by £. constrictor (fig. 198) and to the south by Æ sabaleticus. Erpetogomphus tristani will probably be found in southern Nicaragua, but its southernmost distribu- tion is unknown. Material Type data. — Holotype male with following label data (all handwritten by P. P. Calvert unless otherwi- se noted): ‘Oricuajo,/ Costa Rica./ July, 1911/ J. F. Tristan’, ‘Erpetogomphusl tristanil 8 | Calvert/ TYPE. , ‘Needham/ fig [in Needham’s hand?], red label: ‘9244 Tyre [printed] 944/ Erpetogomphusl tristani | Calvert’, ‘2nd ham-/ ules/ sheath of/ penis.’ Allotype female with same label as holotype except for follo- wing: Labium &/ Maxillae.’, “Erpetogomphusl trista- ni 2/ Calvert’. Both specimens were originally pin- ned, but each has been placed in a clear envelope. Both specimens in ANSP. Other material (54, 49, including holotype d and allotype 9). — Costa Rica: Guanacaste Prov.: Nicoga, Feb. 1912 (J. F. Tristan), 14 (ummz); Quebrada Azul, 2.5 mi W of Tilaran, 24 July 1967 (O. S. Flint, Jr., and Ortiz B), 16, 19 (usnm); Rio Santa Rosa, 3.7 mi E of Las Canas, 25 July 1967 (O. S. Flint, Jr., and Ortiz B.), 12 (USNM); Puntarenas Prov.: 8 mi WNW of Esparta, 26 June 1967 (O. S. Flint, Jr., and Ortiz B.), 1d (usnm); stream 8.1 mi WNW of GARRISON: Revision of È rpetogomphus Esparta, 26 June 1967 (O. S. Flint, Jr., and Ortiz B.), 14 (pre); San Jose Prov.: Res. Biol. El Rodeo, 7 km W of Villa Colon, 9°54’N 84°16°W, 800 m, 10-13 July 1990 (T. W. Donnelly), 19 (rwp). Erpetogomphus ophibolus Calvert (figs. 4-thx, 29-abd, 50-hamules, 73-penis, 95-app, 121-ept, 147-vertex, 175-vl, 198-distr) Erpetogomphus ophibolus Calvert, 1905: 163 (Mexico: Atoyac in Veracruz, H. H. Smith, 2d). — Calvert 1909: 489 (distr., Mex.); Muttkowski 1910: 87 (cat.); Calvert 1912a: 294 (comparison with £. tristani); Ris, 1917: 155 (comparison with EZ. constrictor); Williamson and Williamson 1930: 489 (summary of status); Kimmins 1969: 296 (type in BMNH); Montgomery 1973: 239 (der- ivation of name); Paulson 1982: 255 (Mex.); Davies and Tobin 1985: 27 (cat.); Tsuda 1986: 87 (cat.); Bridges 1991: VII.154 (cat.); Tsuda 1991: 95 (cat.). Erpetogomphus (Erpetocyclops) ophibolus. — Carle 1992: 148 (key to subgenera, type species of Erpetocyclops subg. n.). Description Male. — Labium, labrum, mandibles, anteclypeus brown; postelypeus entirely brown, or with small la- teral green triangular spot; frons green except for brown along frontoclypeal suture, and at base of an- tefrons; vertex dark brown; area anterior to median ocellus undifferentiated, postocellar tubercles well de- veloped, connected medially by a low transverse rid- ge, slightly tuberculate medially; occiput planar, tra- pezoidal, hind margin linear or slightly concave medially, crest covered with long brown hairs; post- occiput brown, transversely concave; rear of head brown. Prothorax brown, anterior lobe green, posterior lobe green brown; synthorax (fig. 4) with prominent well defined dark stripes on green background as fol- lows: middorsal stripe slightly narrowing dorsally, an- tealar sinus and area along this structure connecting to humeral stripe, antehumeral stripe of same width as humeral connecting ventrally and (but not always) dorsally, and in some specimens broadly so, antehu- meral not or just touching antealar sinus; second and third lateral stripes connecting at upper 0.75, isolat- ing round green spot below subalar carina; posterior margin of metepimeron, metasternum and venter of thorax grey brown; venter of profemora grey green, remainder of femora dark brown, becoming black distally; tibiae, tarsi, armature black. Wings hyaline, venation and pterostigma black. Venational statistics. Fifth (rarely fourth or sixth) antenodal thickened; number of marginal cells be- hind fore wing paranal cells: 0/0; anx: fore wing 13- 15/13-16, hind wing 10-12/10-12; pnx: fore wing 10-13/8-12, hind wing 9-12/9-12; cs under pteros- tigma: fore wing 4-7/5-8, hind wing 5-6/4-7; anal tri- 197 [1IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 angular cells: 3/3. Hind wing 25-27 mm. Abdomen (fig. 29) wich segment 1 brown, green posterolaterally and with a green middorsal stripe; segment 2 brown except for green auricles, ventral margin, and narrow middorsal stripe; annulus dark brown; segment 3 primarily dark brown, becoming black distally with green middorsal stripe extending from annulus of segment 2 and narrowing posterior- ly to end, a pale grey anterolateral spot becoming nar- rower along ventral margin and disappearing at 0.25 to 0.50 of segment length; segments 4-6 similar to segment 3 but with pale middorsal stripe ending at posterior 0.10 to 0.75 of segment and anterolateral pale spot ending at transverse carina; segment 7 grey green on anterior 0.50 to 0.60, except for dark trans- verse carina, posterior part of segment brown to black; segments 8-10 red brown, becoming darker dorsally and ventrally, especially along foliate mar- gins; posterior medial margin of segment 9 bluntly pointed or slightly so. Abdomen 30-33 mm. Cercus (fig. 95) ivory brown, becoming brown dis- tally, arcuate with a prominent laterally compressed dorsal tubercle on distal 0.75, extreme base with a black ventral tooth (mostly hidden by posterior mar- gin of abdominal segment 10), though not as promi- nent as in £. constrictor, a ventral carina on distal 0.30 of cercus, tip of cercus with a black tooth; epiprocts brown, gently curved at distal 0.50, tips slightly di- verging, each forming a blunt point (fig. 121), a well defined anterodorsally projecting tooth at middle of medial margin of epiproct. Accessory genitalia. Hamules (fig. 50) small; ante- rior hamule brown, divided at distal 0.30, posterior branch a well developed shoulder; posterior hamule roughly triangular, its tip with a spine; penis (fig. 73) with cornuae slightly bilobate or transverse at base; a well developed prepuce, lateral lobe a small tooth. Female. — Head markings similar to male but with some green on mediolateral areas of labrum, brown at base of antefrons with a median extension connecting with frontoclypeal suture, thus dividing frontal green into two spots. Morphology of vertex and occiput similar to male with following differences: transverse postocellar ridge (fig. 147) complete, slightly arcuate medially, with a slight shallow notch laterally before resuming at postocellar tubercle; occiput a narrow planar semicircle, crest covered with long brown hairs; postocciput tumid, brown; rear of head brown. Pro- and synthorax as in male, with dark markings in some specimens less extensive. Venational statistics (n = 11). Fifth (rarely fourth or sixth) antenodal thickened in all wings; number of marginal cells behind fore wing paranal cells: 0-1/0-2; anx: fore wing 14-16/14-17, hind wing 10-12/10-13; pnx: fore wing 10-13/9-13, hind wing 10-13/10-14; s under pterostigma: fore wing 5-8/5-8, hind wing 198 5-7/4-7. Hind wing 26-29 mm. Abdomen with segment 1 brown, a green spot pos- terolaterally and a complete green middorsal stripe; segment 2 brown with following areas green: wide middorsal stripe narrowing posteriorly, incomplete lateral stripe extending from just anterior to auricle to black posterior annulus, a thin pale margin at venter of segment; segment 3 similar to segment 2 but brown becoming darker; middorsal stripe narrowing posteriorly and forming a mere line on posterior 0.25 of segment; lateral pale stripe ending posteriorly at transverse carina; segments 4-6 similar to segment 3 but middorsal green becoming increasingly smaller on successive segments, so that only a small middor- sal spot occupies basal 0.10-0.50 of segment 6; seg- ment 7 with anterior 0.50 pale green white, interrupt- ed by brown transverse carina, posterior 0.50 becoming red brown to almost black; segment 8-10 all red brown to dark brown, darker dorsoposteriorly; cerci pale grey green. Vulvar lamina (fig. 175) with plates broadly connected at basal 0.75; relatively un- specialized, cleft obtusely V- or U-shaped, about as wide as each lobe; postlamellar ridge with Y- juncture posterior to plates, base of stem well defined and with oval depression laterally. Abdomen 30-33 mm. Diagnosis Males of E. ophibolus are unique in possessing an anterodorsal spine on the mediodorsal surface of the epiproct (fig. 95). The cercus of this species is most si- milar to that of E. constrictor (fig. 92). Morphology of the vertex and occiput serve to distinguish female £. ophibolus from similarly marked species. The median ocellus is in line with the lateral ocelli (fig. 147), sepa- rating this species from E. sabaleticus, E. tristani (fig. 146), and £. constrictor (fig. 145). Female £. ophibolus more closely resemble females of £. schausi, E. agki- strodon, and E. eutainia. However, E. schausi and E. agkistrodon have a transverse postocellar ridge which is medially bilobed (figs. 148-149), while £. eutainia lacks any postocellar ridge and has postocellar tuber- cles only. The postocellar ridge in E. ophibolus is arcu- ate and entire medially (fig. 147). Erpetogomphus op- hibolus and E. eutainia are often sympatric, and the females superficially resemble one another. The se- cond and third thoracic stripes in £. eutainia are al- ways separate (figs. 7-9), but are connected at their upper 0.30 in E. ophibolus (fig. 4). Finally, females of E. eutainia lack the basal stem of the postlamellar rid- ge (fig. 178) present in E. ophibolus (fig. 175). Remarks Variation. — Most specimens of E. ophibolus that I examined are from southern Veracruz, and they show little variability in maculation. The single male from Chiapas has slightly more extensive black thoracic markings: the pale area between the dark humeral and antehumeral stripes is a narrow line. The two females from Belize have the apical dark brown on abdominal segment 7 laterally extending anteriorly separating the lateral pale area into one anterodorsal and one an- teroventral spot. The acuminate condition of the pos- terior medial margin of abdominal segment 9 varies intraspecifically. Biology. — This small green species often occurs with £. eutainia at small rivulets along the lowland Gulf of Mexico drainage systems of eastern Mexico. In 1976 I collected many by flushing them from agri- cultural stubble bordering trees and shrubs along the Rio Otapa in central Veracruz. Adults, when flushed, did not fly far and were consequently easy to collect. Six other gomphids collected at the same site under the same circumstances included Phyllocycla brevi- phylla Belle, P. volsella (Calvert), Phyllogomphoides suasus (Selys), P. duodentatus Donnelly, Progomphus clendoni Calvert, Erpetogomphus eutainia, and E. bothrops. Raúl López (in lit), from information given to him by Enrique González, writes that Æ. ophibolus is the most common gomphid in Veracruz, found year- round in the vicinity of the Los Tuxtlas Biological Station near Catemaco. Altitudinal range is from sea level to 615 m (Chiapas). Collection dates range from 8 June (Belize) to 13 September (Veracruz). Distribution (fig. 198). — This species is thus far known from southern Veracruz (southeast of 19°N, 97°W), central Chiapas, Guatemala, and Belize. Tineke Boomsma (in litt.) collected a pair of this spe- cies in the Mountain Pine Ridge area of Belize (Cayo Distr.: Privassion Creek, 24 July 1993). Material Type data. Holotype male: [Mexico] Vera Cruz, Atoyac, May (H. H. S[mith])/ Erpetogomphus ophi- bolus Calv. Type d. P. P. Calvert det. 1905. B. C. A. Neur., p. 164. Original of pl. 7, figs. 30-32, 46 (Kimmins 1969). In BMNH. Other material (434, 122). — Mexico: Veracruz: Rio Otapa, 8 km S of La Tinaja, 90 m, 13 Aug. 1976, (R. W. and J. A. Garrison), 183, 39, (Rwa); 20 Aug. 1976, 128, 12 (Rwe); Playa Escondida, 20-24 June 1981, (E. González), 19, (UNAM); Arroyo cerca de Playa Escondida, 17-24 July 1981, (E. González), 19, (UNAM); arroyo nr. Playa Escondida, about 30 km NE of Catemaco, 18°34-36°N, 95°04-09°W, 23 July-6 Aug., 9-13 Aug. 1982, (R. W. Garrison), 56, (RwG); Los Tuxtlas, arroyo despues de Laguna Escondida, 25 May 1980, (E. González), 1d, (swp); 11 June 1980, (R. Novelo), 18, (UNAM); Los Tuxtlas, stream nr. Jicacal Beach, 10 July 1979, (Gerado Jimenez), 19, (cc); Colonia Apachital, 16 km S, 10 km E of Tierra Blanca, 10 Sept. 1965, (T. W. Donnelly), 24, 19, GARRISON: Revision of Erpetogomphus (TWD, CC, UNAM); Rio Hondo, on rd. to ‘Colonia la Apachital’ 50 ft [15 m], 6 mi E of Hwy from Tierra Blanca to Cd. Aleman, 25-26 Aug. 1957, (G. H. Beatty III), 1d, (esca); 3 km N of Santiago Tuxtla, 13 Sept. 1965, (T. W. Donnelly), 19, (cc); Chiapas stream 20.1 mi N Ocozocoautla, 2000 ft [615 m], 25 Aug. 1967, (D. R. and M. L. Paulson), 1d, (rsca); BeLize: Toledo Distr: Blue Creek Village, EARTH- WATCH Belize Expedition, 1981, 8 June 1981, (D. H. Messersmith, W.E. Steiner, et al), 19, (USNM); Cayo Distr.: Mountain Pine Ridge, Rio Frio at Augustine, 16°58’N, 88°59’W, 500 m, 22-25 July 1983, (T. W. Donnelly), 19, (rwp); GUATEMALA: El Progreso Dept.: 6.5 km N of Est. de la Virgén, 29 Aug. 1965, (T. W. Donnelly), 18,2%, (rwp). Erpetogomphus agkistrodon sp. n. (figs. 5-thx, 30-abd, 148-vertex, 176-vl, 198-distr) Type material. — Holotype 9, Mexico: Veracruz State, Parque Javier Clavijero, Jalapa, 23 Aug. 1982, 1300 m (R. López) in UNAM. Description Male. — Unknown. Holotype female. — Labium pale grey, becoming dark grey medially; labrum, clypeus, base of mandible dark brown washed with dark olive green along me- dian area of labrum, ante-clypeus, and above lateral lobes; pale green spot along lateral margin of labium; broad green stripe above frontoclypeal suture contin- uing to near base of vertex; antennae, vertex, occiput dark brown; transverse postocellar ridge (fig. 148) complete, emarginate medially; rear of occiput broad- ly semicircular; rear of head dark brown. Prothorax entirely brown except for green anterior lobe; synthorax (fig. 5) with dark brown thoracic stripe well developed on green background; triangular middorsal stripe interrupting inverted green ‘7’ at base, upper end of middorsal stripe connecting with antehumeral and humeral stripes, these two united leaving only pale spot below antealar sinus, and small green streak below; second lateral stripe connecting with humeral and third lateral stripes above and be- low, this stripe also connecting with third lateral be- low antealar sinus, thus isolating a green spot; meta- spiracle black; posterior margin of metepisternum with a dark brown stripe; metasternum dark brown- grey. Femora red-brown, becoming black distally; re- mainder of legs and armature black. Wings hyaline, venation and pterostigma black. Venational details. Sixth antenodal thickened in all wings; number of marginal cells behind fore wing par- anal cells: 3/2; anx: fore wing 15/17, hind wing 12/11; pnx: fore wing 14/16, hind wing 15/15; cs under pt: fore wing 6/5, hind wing 5/6. Hind wing 31 mm. 199 [1IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Abdomen (fig. 30) predominantly black with ves- tiges of a narrow green middorsal stripe on segments 2-6; anterior 0.25 of segment 7 green, becoming white laterally; broad brown midlateral stripe on seg- ment 2 becoming darker on segment 3 and more ex- tensive on succeeding segments, thus encircling small white anterolateral spots; auricle on segment 2 green; segments 8-10 entirely black, cercus white. Vulvar lamina as shown in fig. 176. Abdomen 31 mm. Diagnosis Erpetogomphus agkistrodon belongs to the E. trista- ni group because of the simple condition of the vulvar lamina and the complete postocellar ridge. This spe- cies is distinguished from females of all other species by the emarginate condition of the postocellar ridge. That of E. ophibolus is entire. The female of E. agki- strodon is similar to the only known female of E. schausi, and is diagnosed under the latter species. Remarks I suspect the male of £. agkistrodon will have a pe- nis similar to that of E. ophibolus or E. schausi; it should have a prepuce and the lateral lobes should be small, semi-circular and with or without spinules. The thickened sixth antenodal in this species is an unusual condition: the fifth antenodal is usually thickened throughout the genus. However, this may be an anomaly of the holotype. No conspicuous depressions or pits are present an- terior to the median ocellus. This area, instead, forms a shallow transverse V. Based on female morphology, I suspect the male epiprocts will be relatively thick, gently curved at posterior 0.5, forming a 90° angle, tip broadly spatulate. Biology. — The female was collected along a shaded creek in company with Cordulegaster diadema godma- ni McLachlan. Erpetogomphus boa were also taken at the locale, though they were flushed from low vegeta- tion about 50 m from the stream. Distribution (fig. 198). — Known only from the type locality. Erpetogomphus schausi Calvert (figs. 6-thx, 31-abd, 51-hamules, 74-penis, 96-app, 142-face, 149-vertex, 177-vl, 198-distr) Erpetogomphus schausi Calvert, 1919: 33 (desc. holotype d). Calvert 1920a: 113 (note on type); Williamson and Williamson 1930: 13 (summary of status); Paulson 1982: 255 (Guat.); Davies and Tobin 1985: 28 (cat.); Tsuda 1986: 87 (car); Bridges 1991: VII.187 (cat); Tsuda 1991: 95 (cat.). Description Male. — A thorough description of the holotype male accompanies the original description. I illustrate 200 the thorax (fig. 6), abdomen (fig. 31), accessory geni- talia (fig. 51), penis (fig. 74), head (fig. 142), and cau- dal appendages (fig. 96) of the holotype. The speci- men has been skewered with a bristle. The cercus of the holotype (fig. 96) is strongly curved at distal 0.30, the ventral margin has a remnant of an inferior carina at its extreme base, but is well developed again on the distal 0.50. Venational statistics (based on holotype d and one d from Costa Rica): number of marginal cells behind fore wing paranal cells: 3/2-3; anx: fore wing 16- 18/16-20, hind wing 12-14/12-13; pnx: fore wing 15/14-15, hind wing 13-14/13-15; cs under pteros- tigma: fore wing 5-6/5-6, hind wing 6/5-6; anal trian- gular cells: 4/4. Hind wing 29-30 mm. Abdomen 32- 33 mm. Female. — Labium grey; labrum brown with indica- tion of green mediolaterally, ante- and postclypeus brown with possible green medially on postclypeus; frons green with brown along frontoclypeal suture and base of antefrons, especially medially, with off- shoot almost touching frontoclypeal suture, thus al- most separating green; base of antefrons with a slight longitudinal raised area with a slight concavity on each side; antennae, vertex brown; transverse post- ocellar ridge (fig. 149) complete, emarginate medial- ly; occiput brown, semicircular crest covered with long brown hairs; postocciput brown, convex, rear of head brown. Prothorax entirely brown, paler on anterior and posterior lobes; synthorax as in holotype male (fig. 6); femora grey brown, tibiae and tarsi darker, armature black. Wings hyaline, venation and pterostigma black. Venational statistics (n = 1). Fifth antenodal thick- ened in all wings; number of marginal cells behind fore wing paranal cells: 2/0; anx: fore wing 16/16, hind wing 12/11; pnx: fore wing 15/15, hind wing 14/14; cs under pterostigma: fore wing 6/5, hind wing 6/6. Hind wing 31 mm. Abdomen compressed laterally with markings mostly unrecognizable, but probably similar to E. ag- kistrodon (fig. 30). Vulvar lamina simple (fig. 177), similar to that of £. agkistrodon. Abdomen about 32 mm. Diagnosis Erpetogomphus schausi is unique in possessing a combination of characters of the penis (with well de- veloped prepuce) and caudal appendages (with gently decumbent cercus (fig. 96)). Although the morpholo- gy of its penis allies this species with E. sabaleticus, E. tristani, and E. constrictor, the penes of the other three species lack well developed cornuae and a spinulate lateral lobe, typical of E. schausi. The only known fe- male differs from all other known species except E. agkistrodon in possessing a complete transverse posto- cellar ridge which is notched medially. The only morphological difference that I have noted between these two species is the broader, more semicircular oc- ciput (figs. 149) of E. schausi, compared with the nar- rower, more broadly semicircular condition found in the holotype of E. agkistrodon (fig. 148). The sixth antenodal crossvein is thickened in all four wings of E. agkistrodon, while the fifth is thickened in this fe- male of £. schausi, but, as stated under E. agkistrodon, I suspect the condition in the holotype is atypical. Remarks Variation. — The male sex of this species is known from only two specimens, the holotype male descri- bed in detail by Calvert (1919), and another male fr- om Costa Rica collected by J. Belle and loaned to me for inclusion in this paper. The Costa Rican male is slightly smaller (HW = 29 mm) than the holotype (HW = 30 mm) and differs as follows (condition for holotype in parentheses): small green triangle just la- teral to medial area of anteclypeus (all dark brown); round green spot below antealar sinus of mesepister- num smaller (bigger); anterior pale abdominal spots extending well beyond transverse carina on segments 3-7 (these spots smaller and confined to anterior 0.25 of segments 4-6, to 0.5 of segment 7); sides of abdo- minal segments 8-10 light brown (dark brown); later- al lobe of penis more angulate (more rounded); se- cond thickened antenodal in fore wing 5-5 (6-7) and in hind wing 5-6 (6-6). The holotype appears to be a fully mature male: traces of pruinosity are present on the venter of its thorax. The Costa Rican male seems to be less ma- ture: it lacks pruinosity, and the dark brown areas of the holotype are pale in this specimen. The green ar- eas of the thorax of the Costa Rican male are slightly more restricted than for the holotype. The teneral female described above seems to repre- sent the female of this species. The combination of characters allying it with the £. tristani group and the occurrence of a male from Puntarenas Province, lead me to suspect that I have associated this specimen correctly. Before the discovery of the two E. schausi from Costa Rica, I considered the possibility that E. agkis- trodon might be the female of E. schausi: morpholog- ical considerations in light of the characters for other members of the E. tristani group (e.g., vulvar lamina, shape of occiput) supported this view. However, the geographic distance (about 1700 km), coupled with the local occurrence of both forms, tend to discount this. It is possible that the morphological differences I detect in the occiput of female E. agkistrodon and E. schausi, though major, may be due to geographic vari- ation, which would render E. agkistrodon a junior GARRISON: Revision of Erpetogomphus synonym of E. schaust. Biology. — Nothing is recorded of the circumstanc- es of capture of the holotype. Jean Belle records the following for the Costa Rican male: “This gomphid was secured on the grass-covered bank of an almost impassable mountain rivulet where the insect alighted after a clash with a Hetaerina specimen. The colors when alive are as follows: compound eyes dark brown; head and thorax dark brown with light green; middorsal pale spots of abdominal segments 3 to 6 and pale spot of abdominal 7 yellow.’ Altitudinal data range from 1220 m to 1500 m; the female was collected 8-9 April, the Costa Rican male on 30 August. Material Type data. — Holotype male: GUATEMALA: Dept. Purulhä, Baja Vera Paz, 1220 m, forest stream (W. Schaus, J. Barns), 7 July [no date], in ansp. Other material (23, 19, including holotype d ). — Costa Rica: Puntarenas Prov.: Monte Verde, 1500 m, rivulet, 30 Aug. 1986 (J. Belle), 1d (RNHL); Rio Bellavista, ca. 1.5 km NW of Las Alturas (8.951 N, 82.846 W), elev. 1400 m, 8-9 April 1987 (Holzenthal, Hamilton, Heyn), 19 (usnm). Distribution (fig. 198). — Known only from the type locality in Guatemala and Puntarenas Province, Costa Rica. It probably occurs in intervening Central American republics. Erpetogomphus eutainia Calvert (figs. 7, 8, 9-thx, 32, 33-abd, 52-hamules, 75-penis, 97-app, 122, 123, 124, 125-ept, 143, 144- face, 150-vertex, 178-vl, 199-distr) Herpetogomphus menetriesii. — Selys, 1878: 429 (24 sep.) (‘Guatemala. (Un mâle [szc, female] incomplet. Musée de Paris, une femelle. — Coll. Selys.’)). Erpetogomphus eutainia Calvert, 1905: 162 (1d, Rio Papagayo in Guerrero, Mexico). — Calvert 1919: 35 (comparison with £. schaust); Kimmins 1969: 293 (type in BMNH); Paulson 1982: 255 (U.S., Mex., Guat.); Davies and Tobin 1985: 27 (cat.); Tsuda, 1986: 87 (cat.); Bridges 1991: VII.72 (cat.); Tsuda 1991: 95 (cat.). Erpetogomphus diadophis Calvert, 1905: 167 (26, Texas). — Muttkowski 1910: 87 (cat.); Needham and Heywood 1929: 79 (descr.); Needham and Westfall 1955: 147 (descr.); Borror 1963: 104 (common name); Montgomery 1968: 133 (distr.); Kimmins 1969: 293 (type in BMNH); Paulson, 1982: 266 (synonymy of E. di- adophis and E. eutainia); Bridges 1991: VII.60 (cat.). Erpetogomphus ? diadophis. — Calvert 1919: 36 (Guat., pos- sible conspecificity with E. diadophis) Herpetogomphus diadophis. — Byers 1928: 5 (larva un- known). Erpetogomphus (Calogomphus) eutainia. — Carle 1992 (key to subgenera, type species of Calogomphus subg. n.). 201 TI]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Description Male. — Labium grey to dark grey, labrum all green with a small medial spot on anterior margin (fig. 144) to entirely brown with large central green spot (fig. 143), base of mandibles green, tips becoming black, anteclypeus all green to largely brown; postelypeus gr- een with brown medially along frontoclypeal suture and descending laterally along lateral lobes, thereby isolating medial green spot; in more southerly speci- mens, brown expanding so that green confined to la- teral margins; frons green with brown along fronto- clypeal suture and at extreme base of antefrons; vertex brown with medial transverse area at postocellar tu- bercles green, to entirely brown; occiput brown, transverse triangular pits anterolateral to medial ocel- lus, postocellar tubercles low, incomplete medially; occiput trapezoidal, largely planar except for medial swelling; crest costate, linear to slightly concave, rim- med with long brown hairs; postocciput green me- dially to all brown, concave; rear of head brown. Prothorax brown along middle lobe, remainder green or becoming entirely brown except for pale an- terior lobe; synthorax (figs. 7-9) with prominent, well-defined dark brown stripes on green background as follows: middorsal stripe expanding toward collar, divided by pale middorsal thoracic carina, not touch- ing collar, medial 0.5 or entire antealar crest, linear antehumeral, not touching upper margin; humeral, second and third lateral stripes, medial area of mesin- fraepisternum; more southerly specimens with these stripes becoming thicker (figs. 8-9) so that antehu- meral and humeral connect at lower and upper ends; a vestigial stripe along posterior margin of metepime- ron; venter of thorax and metasternum pale grey green to grey. Coxae, trochanters pale green, venter of profemora green, remainder of profemora black; met- afemora dark brown to black except for pale area ven- trally; basal 0.40 to 0.60 of metafemora pale green to dark grey, becoming black apically; tibiae, tarsi and armature black. Wings hyaline, venation and pterostigma black; more northerly specimens with pale, narrow line along costa disappearing proximal to pterostigma. Venational statistics. Fifth, rarely fourth, antenodal thickened in all wings; number of marginal cells be- hind fore wing paranal cells: 0-1/0-1; anx: fore wing | 1-15/11-15, hind wing 8-11/8-10; pnx: fore wing 8- 1/8-12, hind wing 8-11/8-11; cs under pterostigma: fore wing 4-6/4-6, hind wing 4-6/4-6; anal triangular cells: 3-4/3-4. sut wing length 23-25 mm. en (fe.3 2) with segment | green and a pale brown dorsolater zi stripe on each side; segment 2 brown except for green auricles, ventrolateral: margin, ind middorsal stripe; segment 3 with lateral brown i interrupted medi ally just posterior to trans- verse carina, and with an irregular pale spot within 202 anterior brown spot, posterior brown more promi- nent posteriorly and meeting dorsally on posterior 0.10 of segment; more southerly specimens with lat- eral brown not interrupted medially and with pale an- terolateral spot reduced or almost absent; incomplete middorsal stripe blue green, pale areas becoming white laterally; segments 4-6 similar to segment 3 but with dark brown more prominent posteriorly and with anterolateral white spot confluent with ventral margin; segment 7 pale blue green on anterior 0.50- 0.60, except orange brown transverse carina, posteri- or part of segment dark red brown; segments 8-10 red brown, becoming darker dorsally and ventrally, espe- cially along foliate margins; posterior medial margin of segment 9 smoothly carinate, or with a moderately blunt point. Abdomen 29-32 mm. Cercus (fig. 97) ivory, becoming brown distally, gently arcuate at distal 0.30, with a small black tooth, a ventral costate carina on basal 0.35-0.40 of cercus, often ending in a small, isolated, pebble-like tooth; in other specimens, the posterior part of this carina may be pebble- like; epiprocts (figs. 122-123) brown, moderately curved along posterior 0.50, tip of epi- proct.in lateral view thick, bluntly pointed, with lat- eral costate margin; tips in posterior view (figs. 124- 125) planar, roundly divergent. Accessory genitalia. Hamules (fig. 52) small, ante- rior hamule brown, distal 0.25 divided, the superior branch meeting the inferior branch, thus enclosing an oval space; posterior hamule small, digit-shaped, no apical hook; penis (fig. 75) with long pointed cornu- ae; dorsal membranous hood prominent, its ends par- allel; lateral lobes large, frill-like, their margins with spinules; dorsal area of membranous third segment with a pair of mediolateral blunt chitinized tubercles. Female. — Head as in male with following differ- ences: brown at base of antefrons projecting anterior- ly, partially isolating dorsal green; vertex morphology similar to male, postoccipital tubercles small, isolated medially, occiput trapezoidal as in male, planar with faintly convex arcuate crest; postocciput brown, pla- nar. Thorax as in male. Venational statistics. Fifth, rarely sixth, antenodal thickened in all wings; number of marginal cells be- hind fore wing paranal cells: 0-2/0-3; anx: fore wing 13-16/13-16, hind wing 9-11/8-11; pnx: fore wing 8- 12/8-11, hind wing 8-12/8-13; cs under pterostigma: fore wing 4-7/4-7, hind wing 4-7/4-7. Hind wing 25- 28 mm. Abdomen (fig. 33) with segment 1 green with a brown dorsolateral stripe; segment 2 similar to seg- ment 1, but with wash of brown along ventral margin of tergite; segment 3 with dorsolateral stripe darker, expanding posteriorly and connecting above at poste- rior 0.10 to 0.20 or separated by thin hairline of green, ventral margin with brown interrupted by white below brown dorsolateral stripe; in more north- erly specimens, connecting above with middorsal green stripe; posterior part of brown connecting with dorsolateral stripe posteriorly, transverse carina dark brown; segments 4-6 similar to segment 3 but brown darker, more definite, connecting posterodorsally on 0.25-0.50 on progressive segments, lateral white con- fined to spots, one at anteroventral margin, another ventromedially, these two spots sometimes connected by narrow line of white ventrally; more northerly specimens with medioventral white connecting with middorsal green stripe; segment 7 with anterior 0.50 white except for darkened transverse carina, posterior 0.50 dark brown; segments 8-10 dark brown; cerci white, their tips black. Vulvar lamina (fig. 178) with plates small, almost digit-shaped; U-shaped cleft about as wide as or wider than each plate; postlamel- lar ridge with a costate semicircular ridge. Abdomen 29-33 mm. Diagnosis The characteristically shaped posterior hamules (fig. 52) and penile structures (short prepuce, wide spinulate lateral lobe [fig. 75]) separate males of this species from all others except E. leptophis. Erpetogomphus eutainia possesses a straight or pebble- like ventral carina on the basal 0.25 of the cercus (fig. 97); no such carina is present in E. leptophis (figs. 98). Females of £. eutainia and E. leptophis are unique in possessing a semicircular postlamellar ridge; in all other congeners, this structure is a Y-shaped ridge. The posteriorly pointed occipital protuberance in £. leptophis (fig. 151) is distinct from the relatively un- modified condition present in E. eutainia (fig. 150). Remarks Variation. — Venational details of holotype male of E. eutainia. No marginal cells behind fore wing pa- ranal cells; anx: fore wing 14/14, hind wing 10/10; pnx: fore wing 10/9, hind wing 10/10; cs under pte- rostigma: fore wing 5/5, hind wing 5/6; number of anal triangular cells: 4/4. Hind wing 24.5 mm. Venational details of holotype female of £. diado- phis number of marginal cells behind fore wing para- nal cells: 0/1; anx: fore wing 13/14, hind wing 10/10; pnx: fore wing 10/9, hind wing 10/11; cs under pte- rostigma: fore wing 6/7, hind wing 6/5. Hind wing 28 mm. This small species has a widespread distribution and, consequently, exhibits some degree of clinal vari- ation north to south. Specimens from southeastern Texas and northern Mexico are the palest: the dark transverse markings of the face are narrower, and green is often present medially on the vertex and oc- ciput. The labrum is mostly green (fig. 144), the dark GARRISON: Revision of Erpetogomphus thoracic stripes are narrow, and none is present on the posterior margin of the metepimeron (fig. 7). Medial bands of white encircle abdominal segments 3-6 in both sexes. The dark somatic markings are darker, broader, and more definite in specimens from south- ern Veracruz. Vestiges of a stripe on the posterior margin of the metepimeron can be scen (fig. 8), and the pale white medial abdominal bands are obscured, entirely or almost entirely interrupted by the dorso- lateral brown stripe. Specimens from Costa Rica are the most melanic (figs. 9, 143); no green is present on the vertex or occiput, and the middorsal green stripe of abdominal segments 3-6 is always separate from the medioventral white. The ventral carina of the cercus is also variable. Some specimens have a continuous costate carina with a small, isolated, pebble-like tooth distally; but others may have several antepenultimate pebble-like teeth in addition to the longer ultimate one. The tips of the epiproct may also be variable. In most speci- mens, these structures are posteriorly flattened (fig. 122) with the outer margin slightly to strongly con- cave (fig. 123). The male from EI Progreso, Guatemala, has the epiprocts similar to the condition described above, but they are narrower (fig. 125). The clinal differences described above probably led Calvert (1905) to interpret the female of E. diadophis as a separate species from £. eutainia. Calvert (1919) later received another female from Guatemala, which he postulated may be the same as the paratype of £. diadophis he made no mention of E. eutainia. The large series of males and females examined from various parts of its range convince me that Paulson (1982) was correct in synonymizing the two names. The status of the two females (not male and female) described by Selys (1878) as Herpetogomphus menetriesii is discussed under the status of £. mene- triesii. I also examined a female labelled ‘Cyanogomphu® | mexicanusl n. sp. / coll. R. Martin / 2’, from Honduras, in the MNHP. Biology I collected this small, colourful species in southeas- tern Texas over several years. They were flushed from stubble bordering agricultural fields next to the Gonzales River. They were sympatric with the larger E. designatus, and were about as common. However, the former species behaved more like a damselfly, for they never flew far, rested on tips of stubble or barbs of barbed-wire fences, and were always easy to take with a net. I collected one male which was the prey of a rob- ber fly (Asilidae); another, I rescued from an orb-we- aver spider web. I was able to photograph a pair in co- pula in the late afternoon. In southern Veracruz at the Rio Otapa, £. eutainia had similar habits, as described under E. ophibolus. 203 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 The female from El Salvador (Rio Palio) has the fol- lowing by V. Hellebuyck: ‘on bushes along a fast run- ning river, clear water, rocky bottom & shady banks’. Altitudinal ranges are from near sea level (20 m, Medina Bank, Belize) to 615 m (Oaxaca). Collection dates range from 19 May (Costa Rica) through Oc- tober (Veracruz, La Gloria; Guerrero, Rio Papagaio). Distribution (fig. 199). — This species occurs southeast of the Edwards Plateau in southern Texas, south along the lowlands and foothills of the Sierra Madre Oriental in northern and central Mexico. The farthest western locality is El Sabino in Michoacan state. Specimens have been collected sporadically in Belize, Guatemala, Honduras, El Salvador, and northern Costa Rica. Material Type data. — Holotype male of £. eutainia [labels all printed unless otherwise stated]: “R. Papagayo, / Guerrero, 1200 ft. / Oct. H. H. Smith’, ‘Brit. Mus. / 1911-339., ‘ERPETOGOMPHUS/ eutainia Calvert [written]/ P. P. Calvert, det. 1905 [written]/ B. C. A. Neur., p. 162 [written]/ Original of PI. 7, f. 24-27, 39 [all written)’, small round label with red border: ‘Holo-/ type’, in BMNH. Holotype female (not male, as erroneously stated by Kimmins, 1969) of £. diad- ophis. [labels all printed unless otherwise stated}: ‘Texas [written by an unknown hand)’, pale violet la- bel: ‘McLachlan Coll./ B. M. 1938-674.’, ‘ERPETOG- omPHus 2 [written]/ diadophis Calv. Tyre [written]/ P. P. Calvert, det. 1905 [written]/ B. C. A. Neur., p. 167 [written]/ original of PI. VII, ff. 35, 47 [all writ- ten)’, in BMNH. Material examined (746 , 429, including holotypes of E. eutainia and E. diadophis). — U. S. A: Texas: Caldwell Co.: San Marcos R. at Luling, (rsca, cc); Gonzales Co.: Guadalupe River, 4 mi S of Gonzales, (cc, Rw@; near Otting, Palmetto State Park, (rsca); Palmetto State Park, 6 mi S of Luling, (RwG); Mexico: Michoacan: El Sabino, 20 mi SSE of Uruapan (UMMZ, RwG); Oaxaca: La Escondida, route 190, km 727, 2000’, (usnm); San Luis Potost. Tamazunchale, (USNM); Tamaulipas. 3 mi S of Ciudad Victoria, (USNM); Veracruz. Cardel, La Gloria, (USNM); pond 2.7 mi S of La Tinaja, 300 ft., (DRP); Rio La Palma, 25 km N of Catemaco, (Rw6); Rio Otapa, 8 km S of La Tinaja, 90 m, (rwo); Sontecomapan, Coscoapan, (UNAM); BELIZE: Toledo Dist.: Medina Bank, 20 m, (TB); GUATEMALA: El Progreso Dept.: San Agustin Ac., (FSCA); Zacapa Dept.: Zacapa, (FSCA); un- known locality and date (‘Rodig.’ [illegible]) [labelled as Herpetogomphus menetriesii), (IRSN); unknown locality, date, and collector [labelled ‘E. menetriesir Selys’], (MNHP); Honpuras: Unknown locality, date, and collector [labelled ‘Cyanogomphus?/ mexicanusl n. sp./ (Coll. R. Martin/ 2’), (MNHP); EL SaLvapoRr: La Libertad Dept.: Rio Palio, San Juan Opico, 400 m, (vH); Costa Rica: Alajuela Prov.: Quebrada Mina 0.2 mi S Hwy. 11 on Hwy, (rsca); Guanacaste Prov.: Hda. Taboga, 100 ft, (pre); Rio Santa Rosa, 3.5 mi N of Cañas, 300 ft (pr). 204 Erpetogomphus leptophis sp. n. (figs. 10-thx, 34, 35-abd, 53-hamules, 76-penis, 98- app, 151-vertex, 179-vl, 200-distr) Type material. — Holotype male. BeLize: Toledo District, I. Z. E. field station, Blue Creek, 25 June 1983 (M. L. May) (rsca). Allotype female. BELIZE: Toledo District, Blue Creek Village (Earthwatch) Belize Expedition 1981, 8 June 1981 (D. H. Messersmith, W. H. Steiner, et al.) (USNM). Description Holotype male. — Labium light grey, becoming dark grey medially; labrum green with dark brown margin and medial line almost separating green into two spots; base of mandibles grey-green, anteclypeus green, postclypeus brown except for small green spots above lateral lobes; ventral margin of frons brown, joining brown of postclypeus, thus forming a large frontoclypeal stripe; remainder of frons green; a large transverse furrow at base of antefrons, this area with a pair of deeper pits, each anterolateral to median ocel- lus; postocellar ridge complete, but with medial part lower and at level of lateral ocelli; extreme base of frons, entire vertex, and occiput dark brown, occiput tumid medially, hind margin mostly straight but pe- aked medially so that hind margin seems slightly an- gularly convex; rear of head dark brown. Prothorax entirely brown except for following green areas: anterior lobe, and two small midlateral spots (almost touching) on median lobe; synthorax (fig. 10) with dark brown thoracic stripes well devel- oped on green background; dark middorsal stripe tri- angular, its wide base touching collar but not con- necting with antehumeral; antehumeral and humeral connected along upper 0.25, thus isolating small green spot; second lateral as shown in fig. 10, a poste- rior branch connecting with third lateral along upper 0.30, isolating a green spot; posterior margin of met- epimeron brown, all thoracic stripes connected by brown along antealar sinus; metasternum grey- brown. Venter of profemora grey-green, remainder of legs dark brown; tibiae, tarsi, armature black. Wings hyaline, venation and pterostigma black. Venational details. Fifth antenodal thickened in all wings; no supplementary marginal cells behind fore wing paranal cells; anx: fore wing 14/13, hind wing 10/10; pnx: fore wing 10/11, hind wing 10/10; cs under pterostigma: fore wing 6/5, hind wing 5/5; anal triangular cells: 4. Hind wing 29 mm. Abdomen as in fig. 34. Segment 1 brown, green laterally and with a green middorsal spot; segment 2 brown except for green auricles, ventral margin, and narrow middorsal spot with a cordate expansion in the middle; segment 3 black with green middorsal stripe ending at basal 0.30 of segment and grey ante- rolateral spot; segment 4 black with anterior 0.2 white laterally, becoming green dorsally; segments 5- 6 like segment 4; segment 7 with anterior 0.5 white laterally, becoming green dorsally, posterior 0.5 of segment dark red-brown; segments 8-10 red-brown, posterior medial margins of segments 8-9 slightly pointed. Abdomen 31 mm. Cercus (fig. 98) light green, simple, with no ventral carinae, slightly arcuate, tip black; epiprocts brown, gently curved at distal 0.5, tips viewed posteriorly with parallel sides, forming bluntly rounded tips. Accessory genitalia. Hamules (fig. 53) small; ante- rior hamule brown, distal 0.25 divided, the superior branch meeting the inferior branch, thus enclosing an oval space; posterior hamule small, finger-like, no ap- ical hook; penis (fig. 76) with long, pointed cornuae, dorsal membranous hood long, its ends crossing (fig. 76); lateral lobes large, frill-like, their margins with spinules. Allotype female. — Head as in male except on la- brum, brown medial line separating green into two spots; base of antefrons as in male (fig. 151) with a raised bilobed ridge between lateral ocelli; posterior margin of occiput with a prominent, posteriorly di- rected, pointed protuberance, its base tumid so that entire raised area assumes shape of a plumbob. Thorax as in male. Wings hyaline, similar to male. Venational details. No supplementary marginal cells behind fore wing paranal cells; anx: fore wing 14/16, hind wing 9/11; pnx: fore wing 11/13, hind wing 10/11; cs under pterostigma: fore wing 5/4, hind wing 5/6. Hind wing 25 mm. Abdomen (fig. 35) with segment 1 brown, becom- ing dark green-brown laterally; segment 2 brown with a narrow green middorsal stripe and green lateral stripe encircling auricle and extending entire length of segment; segment 3 like segment 2, but brown be- coming darker and green lateral stripe abbreviated, forming an elongate spot along basal 0.5 of segment; segments 4-6 all black with white basal band, extend- ing ventroposteriorly along ventral margin of tergites; segment 7 with anterior 0.30 white, remainder dark red-brown; segments 8-10 dark red-brown, cercus white. Vulvar lamina (fig. 179) with plates connected anteriorly, plates wider than in £. eutainia; postlamel- lar ridge semicircular as in E. eutainia. Abdomen 30 mm. Diagnosis The structure of the hamules and penis ally E. lep- tophis with E. eutainia, but the superior appendages easily distinguish the two species (figs. 97-98). Erpetogomphus eutainia has a pebble-like ventral cari- na on the basal 0.25 of the cercus; but no such struc- ture is present in £. leptophis. The posteriorly directed GARRISON: Revision of Erpetogomphus pointed occipital protuberance easily identifies the fe- male of E. leptophis. The female shares with £. eutai- nia the semicircular postlamellar ridge, but £. leptop- his possesses a bilobed ridge between the lateral ocelli. This surface is planar in E. eutainia. Remarks Although the types were taken by different collec- tors two years apart, the female characters indicate it is the female of E. leptophis. Biology. — Michael May collected the male along a creek in the forest. Distribution (fig. 200). — Known only from the type locality. Erpetogomphus elaphe sp. n. (figs. 11-thx, 54-hamules, 77-penis, 99-app, 180-vl, 200-distr) Erpetogomphus elaps. — Calvert 1907: 398 (Costa Rica, fig- ures anterior hamules). Erpetogomphus sp. n. near elaps. — Dunkle 1988: 46 (Honduras). Description Holotype male. — Labrum pale grey green; antecly- peus, labrum, base of mandibles pale green, tips of mandibles dark brown; postclypeus and frons pale gr- een, a slight wash of brown on lateral margins of fron- toclypeal suture, antefrons pale green, slightly darker at base, vertex dark brown; a deep transverse trough in front of median ocellus; pedicel, scape, and flagel- lum brown; occiput trapezoidal, pale green, mostly planar, slightly tumid medially, crest slightly promi- nent, slightly emarginate medially and covered with long brown hairs; postocciput green, transverse when viewed dorsoposteriorly; rear of head yellow brown. Prothorax green, becoming brown on middle lobe laterally and anteromedially. Synthorax (fig. 11) ap- ple green with following dark areas: poorly defined middorsal stripe not extending to collar and inter- rupted medially by middorsal thoracic carina and an- tealar crest, well defined antehumeral stripe narrow- ing dorsally but touching antealar crest and extending ventrally to mesinfraepisternum, a small posterior oft- shoot from antehumeral stripe at upper 0.10 connect- ing with vestigial first lateral stripe which is abbreviat- ed to upper 0.50; venter of synthorax green, coxae and trochanters grey green, femora pale green becom- ing dark brown on distal extensor surfaces, these dark markings occupying distal 0.80 of profemora, about distal 0.60 of mesofemora, and distal 0.30 of metafe- mora; tibiae and tarsi black; armature black. Wings hyaline, anterior margin of costa yellow to pterostigma, thereafter black; yellow at base of costa darkening to brown at costal triangle, pterostigma 205 TI]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 brown, veins bordering it black. Venational details. Fifth antenodal thickened in all number of marginal cells behind fore wing paranal cells: 1/1; anx: fore wing 13/13, hind wing 10/10; pnx: fore wing 9/9, hind wing 9/10; cs under pt: fore wing 6/5, hind wing 4/6; anal triangular cells: 4/3. Hind wing 26 mm. Abdomen with segment 1 yellow green with a wash of brown dorsol: aterally; segment 2 yellow green dor- sally and ventrally, w ith a brown idle stripe ex- panding ventrally behind auricle, anterior 0.5 of this stripe poorly defined, annulus dark brown; segment 3 light yellow green dorsally interrupted laterally by a longitudinal brown stripe well marked at transverse carina, expanding again at posterior 0.25 of segment and connecting dorsally at black annulus; inferior margin of tergite probably ivory (post mortem preser- vation has partially obscured this area); segments 4-6 similar to segment 3, with anterior 0.10 of dark mid- lateral interrupted and with gradual dorsolateral ex- pansion of stripe so that inverted dorsal pale green wedge-shaped middorsal stripe disappears at distal 0.40 of segment 6; segment 7 pale green brown be- coming tawny on posterior 0.10 of segment; seg- ments 8-10 red brown becoming darker dorsolateral- ly on segment 8; foliate expansion and denticles black. Abdomen 33 mm. Cercus (fig. 99) pale ocher becoming dark brown at tip; linear, gradually narrowing distally, tips round- ly pointed, a weakly defined ventral carina on distal 0.50 of cercus; epiprocts brown, gently curved at 90° angle; tips spatulate when viewed posteriorly. Accessory genitalia (fig. 54). Anterior hamule green basally, becoming dark brown distally, branched at distal 0.50; superior (upper) branch greatly enlarged, as long as stem of hamule with resulting interval berween branches small, semicircular; posterior ha- mule (fig. 54) almost digit-shaped, wider at base, with poorly developed anterior shoulder, tip broadly rounded, without an apical tooth; penis with a prom- inent knife-like serrated lateral lobe; membranous hood well developed but lobes not overlapping, cor- nua pointed but with median shoulder as in fig. 77. Allotype female. — Similar to male in markings and colouration. Head with transverse trough anterior to median ocellus as in male, but occipital region differ- ing as follows: occiput small, erect, barely visible in dorsal view, convex medially; postocciput with me- dian tumid area. Thoracic, leg, and abdominal pat- terns as in male, but dark areas more restricted; mid- dorsal and antehumeral stripes faint, first lateral stripe vestigial, confined to upper 0.20 of thorax. Venational details. Fifth antenodal thickened in all wings; Haie D marginal cells behind fore wing par cells: anx: fore wing 14/15, hind ne U 10/11, hind wing 11/12; c Ww Ings; T ); pnx: fore wine 206 under pterostigma: fore wing 6/5, hind wing 5/6. Hind wing 29 mm. Vulvar lamina (fig. 180) with basal plate promi- nent, lobes narrow with medial margins distinctly concave, their bases not touching; distal part of lobe with a ventral flap bent under; its costate rim visible externally and disappearing medially near base of lobe (fig. 180); with resultant cleft forming a wide, U- shaped interval; Y-shaped juncture of postlamellar ridge not extending beyond posterior margin of plate, stem with diagonally ovaloid depression on each side. Abdomen 34 mm. Diagnosis Erpetogomphus elaphe males differ from males of the more northerly £. elaps by only two morphologi- cal characters (contrasting characters for E. elaps in parentheses): 1) The superior branch of the anterior hamule is robust and is as high as the basal 0.5 of the hamule so that the resulting gap between upper and lower branches is small (fig. 54), (superior branch less robust, thinner; resultant gap between hamular bran- ches wider [figs. 55-56]), and 2) The tip of the poste- rior hamule is rounded, (tip of posterior hamule with a cephalad directed tooth). The posterior hamule of E. elaphe appears to be more digit-shaped and is remi- niscent of those of E. eutainia and E. leptophis. The same structure appears more triangular in Æ. elaps. However, this last character is subtle and is best de- tected when specimens of both species are in hand: I do not consider it a good diagnostic character. I have found no significant differences in the penis or caudal appendages between the closely related and allopatric E. elaphe and E. elaps. I have been unable to find any differences between females of the two species. Differences between E. elaphe and other similar species parallel those of E. elaps and are discussed under that species. Remarks Variation. — Erpetogomphus elaphe does not show as great variation in thoracic markings as does E. elaps. The holotype male is the most boldly marked of the series. All paratype males (n = 15), two of which are teneral, have an antehumeral stripe, although it is ob- scure in one male from Costa Rica. The humeral stri- pe in all but one of the paratypes is restricted to the upper 0.20 of the humeral suture, and the middorsal stripe is vestigial or absent. Female paratypes of E. elaphe are similar in macu- lation to the allotype, although one female from Costa Rica has a poorly defined antehumeral stripe. The small humeral stripe is reduced to the upper 0.20 and barely visible. Venational statistics for paratype males (n = 18); number of marginal cells behind fore wing paranal cells: 0-2/0-3; anx: fore wing 12-15/12-15, hind wing 9-11/9-11; pnx: fore wing 8-11/9-11, hind wing 9- 11/9-11; cs under pterostigma: fore wing 3-6/4-7, hind wing 4-6/4-7; anal triangular cells: 3-5. Hind wing 25-28 mm. Abdomen 30-34 mm. Venational statistics for paratype females (n = 6); number of marginal cells behind fore wing paranal cells: 0-4/0-3; anx: fore wing 11-17/14-16, hind wing 10-12/10-12; pnx: fore wing 9-12/10-12; hind wing 9-12/9-13; cs under pterostigma: fore wing 4-6/4-5, hind wing 5-6/4-6. Hind wing 27-30 mm. Abdomen 33-34 mm. Biology. — Little is known of the biology of this species, though its habits probably mirror those of E. elaps. The female from Cartago Province was collect- ed in a ‘marshy area by grassy hillside’. Altitudinal gradients for E. elaphe range from about 900 m (nr. Camotän, Guatemala) to about 1700 m (SE of Cartago, Costa Rica). Collection dates range from 2 June (Agua Caliente, Guatemala) to 5 August (San Jose, Costa Rica). Distribution (fig. 200). — Erpetogomphus elaphe is known only from Guatemala, Honduras, and Costa Rica, but it probably exists in El Salvador and Nicaragua. Its distribution is allopatric (or possibly parapatric) with the form of E. elaps with thin anteri- or hamules. Material Type data. — Holotype male: GUATEMALA, Guatemala Dept., El Fiscal, 6 June 1909 (E. B. Williamson). Allotype female: same data, but 4 June 1909. Both in ummz. Paratypes examined (186, 79): GUATEMALA: Chiquimula Dept.: Alda dos Quebradas nr. Camotän, 900 m (T. W. Donnelly) , 19 (twp); Guatemala Depa Acuna Caliente, 2 June 1909 (E. B. Williamson), 19; El Fiscal, 4 June 1909 (E. B. Williamson); 1d, 19; 5 June 1909 (E. B. Williamson), 16; 6 June 1909, 24 (all ummz); Finca El Rosario, 36.5 km SE of Guatemala City, 15 June 1975 (J. E. Hafernik, Jr.), 2d (RwG); HONDURAS: Francisco Morazan Dept: 30 km ESE Tegucigalpa, 29 June 1985 (Pinto and Ranch), 19 (swp); 24 July 1983 (L. Cordoba), 16 (swp); costa RICA: [no local- ity data], 1920 (Paul Serre), 36 (MNHP); Cartago Prov., 5 mi SE Cartago, 5500 ft, 14 June 1963 (F. G. Thompson), 12 (pre); San Jose Prov.: (all collected by H. Schmidt), San José, 18 June [no year stated], 16 (H. Kahl Coll’n, Acc. 12676); 24 June, 14; 4 July, 16; 5 July, 28; 17 July, 16,19; 18 July, 19; 21 July, 16; 5 August, 1d (rsca). Erpetogomphus elaps Selys (figs. 12, 13-thx, 55, 56-hamules, 78-penis, 100- app, 152-vertex, 181-vl, 200-distr) GARRISON: Revision of Erpetogomphus Erpetogomphus elaps Selys, 1858: 330 (70 sep.) (4 descr. ‘Le Mexique, d’après un exemplaire unique du Museum de Paris, rapporté par M. Salflle.”). — Selys 1859: 538 (12 sep.) (descr. d ); Selys 1869: 175 (12 sep.) (descr. d, 2); Selys 1873b: 519 (75 sep.) (list); Calvert 1905: 163 (lo- calities in Mex.); Calvert 1907: 399 (comparison with E. boa); Calvert 1909: 481 (distr. in Mex.); Muttkowski 1910: 87 (cat.); Ris 1917: 153 (mentions Calvert’s [1907] comments on comparison with £ boa); Williamson and Williamson 1930: 13 (summary of stat- us); Montgomery 1973: 239 (derivation of name); Paulson 1982: 255 (Mex., Guat., Costa Rica); Davies and Tobin 1985: 27 (cat); Tsuda 1986: 87 (cat.); Bridges 1991: VII.68 (cat.); Tsuda 1991: 95 (cat.). Gomphus elaps. — Hagen, 1861: 100 (descr. d) Herpetogomphus elaps. — Hagen 1875a: 42 (cat.); Selys 1879: 64 (note on classification); Kirby 1890: 60 (cat.); Needham 1897: 182 (name mentioned); Calvert 1899: 386 (descr. d, comparison with E. viperinus auct.), 415 (gizzard); Higgins 1901: 128 (mentions Calvert, 1899). Description Male. — Labrum grey white with wash of brown around margin of median lobe; anteclypeus, labrum, base of mandibles pale grey green, tips of mandibles dark brown; postclypeus and frons light green, some specimens with brown at base; vertex dark brown; a deep transverse trough in front of median ocellus, a small tubercle behind each lateral ocellus; pedicel, scape and flagellum brown; occiput trapezoidal, light green, mostly planar; crest slightly prominent, straight, slightly notched medially, and covered with long brown hairs; postocciput green, transverse when viewed dorsoposteriorly; rear of head yellow brown, paler laterally. Prothorax and median lobe red brown becoming pale apple green medially, especially on anterior and posterior lobes. Synthorax ranging from entirely apple green (fig. 13) with slight hint of dark antehumeral stripe to having following brown areas (fig. 12): poor- ly defined middorsal stripe divided by pale middorsal thoracic carina, well-defined antehumeral stripe just touching antealar crest dorsally and not touching me- sinfraepisternum ventrally; a vestigial first lateral stripe confined to upper 0.30-0.40 of suture and sometimes with a small anterior offshoot touching an- tehumeral; venter of synthorax, coxae, trochanters, grey green; femora pale green, becoming dark brown on distal extensor surfaces; these dark markings occu- pying distal 0.80 of profemora, distal 0.50 of mesofe- mora, and distal 0.20 to 0.40 of metafemora; tibiae and tarsi dark brown to black; armature black. Wings hyaline, anterior margin of costa yellow to pterostigma, thereafter black; yellow at base of costa darkening to brown costal triangle; pterostigma dark brown, veins bordering it black. Venational statistics. Fifth antenodal (occasionally fourth or sixth) thickened in all wings; number of marginal cells behind fore wing paranal cells: 0-2/0-3; 207 [1] DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 anx: fore wing 11-15/12-15, hind wing 9-11/8-11 pnx: fore wing 8-11/8-11, hind wing 8-11/8-11; cs under pr: fore wing 4-6/4-7, hind wing 4-6/4-7; anal triangular cells: 3-5. Hind wing 24-29 mm. Abdomen with segment 1 yellow green with a wash of brown dorsol aterally; segment 2 2 yellow green dor- sally with a large midlateral brown spot dorsoposteri- orly to auricle, or, in well- marked specimens, a dis- tinct dark brown midlateral stripe expanding ventrally behind auricle, annulus dark brown; seg- ment 3 broadly light yellow green dorsally interrupt- ed laterally by a longitudinal brown stripe well- marked at transverse carina and flaring again at posterior 0.25 of segment to annulus; in palely marked specimens, brown midlateral stripe interrupt- ed behind transverse carina; in well-marked speci- mens, midlateral stripe broad and meeting dorsally on posterior 0.10 of segment; inferior margin of tergite ivory, in heavily marked specimens limited to a trian- gular spot an anterior 0.10 of segment; segments 4-6 similar to segment 3, but some specimens with dark midlateral stripe more prominent and not interrupted behind transverse carina; in heavily marked speci- mens, midlateral stripe almost black and covering en- tire segment, except for basal pale annulus at basal 0.10 of segment and dorsal inverted pale green wedge- shaped middorsal stripe gradually narrowing posteriorly and disappearing at distal 0.20 of segment 4; these dorsal pale areas lacking on segment 5 and 6; segment 7 pale green, becoming tawny on posterior 0.10 of segment, or with posterior 0.20 dark brown, becoming black dorsally, transverse carina brown; segments 8-10 generally red brown becoming darker dorsally; in darkly marked specimens becoming al- most black dorsally; foliate expansion and denticles black. Abdomen 29-38 mm. Cercus (fig. 100) pale ocher to light brown, linear, gradually narrowing distally, the tips roundly point- ed, a we: akly ite ventral carina on distal 0.50 of cercus; epiprocts brown, gently curved at a 90° angle, tips when viewed posteriorly slightly spatulate, broad- ly rounded, or obliquely truncate. ‘Accessory genitalia (fig. 55). Anterior hamule dark brown to black, branched at distal 0.30, superior branch larger than inferior branch; superior branch moderately thickened in specimens from northern Mexico south to Chiapas, Mexico; specimens south of Chiapas (fig. 56) with superior arm of hamule more slender (see remarks); posterior hamule (figs. 55-56) pale, triangular, with an anterior CAMES tip with a tooth pointed cephalad; penis h prominent, knife-like serrated lateral lobe; mem- branous hood well developed but lobes not overlap- t ping, cornua pointed but with median shoulder as shown in fig. 78. roughly Female. — Similar to male in markings and colour- 208 ation; head with transverse trough anterior to median ocellus as in male, but occipital region differing as fol- lows: occiput small, planar, encompassed posteriorly by arcuate (convex) crest (fig. 152); postocciput with median tumid area. Thoracic, leg, and abdominal patterns as in male, but dark areas more restricted, es- pecially on femora and abdomen; cerci pale. Abdomen 30-36 mm. Venational statistics. Fifth (rarely sixth) antenodal thickened in all wings; number of marginal cells be- hind fore wing paranal cells: 0-3/0-3; anx: fore wing 12-16/13-16, hind wing 9-12/9-12; pnx: fore wing 8- 11/9-12, hind wing 8-12/9-12; cs under pt: fore wing 5-6/5-7, hind wing 5-7/5-7. Hind wing 26-30 mm. Vulvar lamina (fig. 181) with basal plate promi- nent, lobes narrow with medial margin distinctly concave, their bases not touching, with resultant cleft forming a wide, U-shaped interval; apical 0.30 of plate with mesal margin folded ventrally, beneath main plate, the costate rim sometimes visible along lateral margin of plate (fig. 181); Y-shaped juncture of postlamellar ridge not extending beyond posterior margin of plate, stem with diagonally ovaloid depres- sion on each side. Diagnosis Erpetogomphus elaps and E. elaphe are closely rela- ted species which are diagnosed under E. elaphe. In males, the linear cercus of £. elaps (fig. 100) easily dis- tinguishes it from the decumbent cercus (fig. 102) of E. bothrops. Body colouration is similar in E. elapsand E. bothrops, but all E. bothrops have well-marked ante- humeral and humeral stripes (fig. 15). Many speci- mens of £. elaps have these markings poorly develo- ped or nearly absent (fig. 13). With the exception of £. elaphe, whose females ap- pear indistinguishable, females of E. elaps are most similar to £. bothrops. The differences in the vulvar lamina as discussed by Calvert (1899) are an easy way to distinguish them. Secondly, the occiput of £. elaps is small and convexly arcuate (fig. 152); in E. both- rops, this structure is wider (fig. 154) and not nearly as arcuate. Remarks Variation. — The description is based on 49 males and 21 females. The holotype is in fair condition, alt- hough most of dorsum of the prothorax and part of the mesepisternum have been eaten away by dermest- ids. It has a completely pale thorax and the black mid- lateral stripes on abdominal segments 3-6 are inter- rupted only on the basal 0.10 of each segment. The hamules (fig. 55) and cerci (fig. 100) show that this specimen is the apple-green species ranging from northern Mexico south through Guerrero and into southern Veracruz. Venational details of holotype male: number of marginal cells behind fore wing paranal cells: 0/0; anx: fore wing 12/12, hind wing 9/10; pnx: fore wing 9/8, hind wing 8/9; cs under pt: fore wing 4/4, hind wing 5/5; anal triangular cells: 3/4. Hind wing 24 mm. Abdomen 31 mm. I consider E. elaps to be the most variable of any species in the genus. Calvert (1907) figured three forms of the anterior hamules of this species when he examined a male from San Jose, Costa Rica (= E. elaphe). His fig. 31 (Guerrero, Rincon) and fig. 32 (Morelos, Cuernavaca) correspond to the holotype of E. elaps (fig. 55), which is characterized by possessing a moderately thickened anterior branch whose tip al- most overlaps the posterior branch when viewed lat- erally. Calvert’s fig. 30 (Guerrero, Rincon) is typical of specimens I have seen from Chiapas, Mexico, and northern Guatemala (fig. 56). The anterior branch forms a slender hook which rarely overlaps the poste- rior branch when viewed posteriorly. Figures 33 and 34 of Calvert (1907) represent E. elaphe and are dis- cussed under that species. Calvert (1907) stated that he was unable to find any other characters which var- ied correlatively, and I have found no other characters to separate E. elaps and the narrow hamular form. I have not found any intermediate conditions in the shape of the anterior hamule. Their distributions ap- pear to be largely allopatric, although Calvert (1907) figures both forms from Rincon in Guerrero. Interestingly, the distribution of the form with the thin anterior hamular branch lies between typical E. elaps and E. elaphe. Because I have been unable to find any other diagnostic characters, I am inclined to treat the Rincon male and others from Chiapas and north- ern Guatemala as a variant of £. elaps. The divergent morphology of southernmost £. elaps may represent character displacement between E. elaps and E. elaphe. Both males and females show remarkable range in size and maculation patterns throughout its range. The smallest specimens examined are from Morelos (HW: males 23 mm, females 26 mm), the largest from the Rio Metlac region of Veracruz (HW: males 29 mm, females 31 mm). Specimens collected at 1220 m in Sinaloa State are melanic, characterized by pale areas of abdominal seg- ments 4-6 reduced to basal 0.10 laterally; pale areas on dorsum of segments 3-6 reduced laterally, this pale area a thin hairline on segment 5, a basal ring on seg- ment 6. I find no correlation of presence or absence of the dark thoracic stripe to locality. A series of specimens from Palapita, Nayarit, ranges from an entirely pale thorax to possessing antehumeral and humeral stripes. All specimens from Veracruz that I have ex- amined lack dark thoracic stripes, but all southern variants with the thin anterior hamular branch from GARRISON: Revision of Erpetogomphus Chiapas and northern Guatemala possess them. Biology. — This small species has habits similar to those described for £. bothrops. The following field notes accompany specimens collected by E. B. Williamson: Nov. 6 (Tepic): “2 Gomphine — from brushy vegetation between road and river at rapids above mill’; Nov. 7: ‘Gomphine d — Have seen these only in bushes along bank of Rio de Tepic — usually not over foot above ground — resting or flying. One 9 was ovipositing in rather still water below the rapids’; Nov. 14 (Jalisco: San Diego Rancho): ‘Very hard to catch in net. Sits tight on rock and lets it [net] pass over and then flies. Easily approached and I finally began catching them by dropping the net over their rock & letting them rise into the bag.’; and Nov. 16 (Jalisco: San Diego Rancho): ‘[d, 2] caught in grass in banana patch early in AM, sluggish’. Löpez (in litt.) records it as one of the most com- mon members of the genus in Mexico, having a wide tolerance for diverse stream habitats. I have seen no records of this species taken near sea level, unlike £. bothrops. Altitudinal records range from 670 m (Nayarit, Palapita) to about 1500 m (Cuernavaca), dates of capture from 9 June (Veracruz, Teocelo) to 23 November (Jalisco, San Diego Rancho). Distribution (fig. 200). — Both this species and £. bothrops appear to be the most widespread members of this genus in Mexico. Erpetogomphus elaps occurs in mesic areas on either side of the Sierra Madre Oriental and Occidental. It has been collected as far north as Nuevo Leön in the east, and southern Sinaloa in the west, southward through Jalisco, Morelos, Guerrero, and Veracruz. Material Type data. — Holotype male with following data: small round white label with handwritten ‘2/44; rec- tangular manila label with handwritten (in R. Martins hand) ‘Ophiog. elaps/ d De Selys/ Mexique? a renvoyer’; large green label with black border with handwritten ‘E. elaps, Selys/ Mexique’; small white label with printed “MUSEUM PARIS’; nar- row red label with printed ‘HOLOTYPE'; white printed label with ‘E. elaps HOLOTYPE/ Det. J. Belle, 1977/ No. 2, 1844. Insects [sic] de/ [reverse side] differents Ordres envoyes/ de Mexique par/ Mr. Ghresbreght ! and in lower right hand corner ‘ne pas/ Salll]e! In MNHP. The caudal appendages and accessory genitalia are shown in figs. 55, 78, 100. Other material (498, 219). — Mexico: Chiapas Pacific slope, 800-1000 m), (UsNM); stream 20.1 mi NE Tapanatepec, Oaxaca, 2700 ft, (esca); San Geronimo, Volcan Tacana, 450 m (UMMz); Distrito Soconusco, Finca Juarez, (UMMZ); Guerrero: Chilpancingo, 4600’, (esca, BMNH); Tepetlapa, 3000 ft, (BMNH); Jalisco: San Diego Rancho nr. Cocula, 209 T1JDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 (ummz); 15 mi on road to Tequila, (uMMZ); Michoacan: El Sabino, 20 mi SSE of Uruapan, (ummz); Morelos: Alpuyeca, S of Cuernavaca on Rt. 95, (rsca); Cerro de Higuerön, Jojutla, (UNAM); Cuautla, (ummz); Cuernavaca, ca. 1200-1500 m, (UMMZ, USNM, RWG); 5 mi S of Cuernavaca, (UMMZ); 16 mi S of Cuernavaca, (uMMZ); Itzamatitlan, 5 km SW of Yautepec, (Rwa); Nayarit: Acaponeta, (BMNH); Jumatän, (UNAM); Tepic, (UMMZ, USNM); Jalisco: Palapita, 670 m (unam); Nuevo León: Municipio de Monterrey, Cerro El Mirador, (rsca); Sinaloa: stream 27.9 mi NE of Concordia, 4000 ft., (DRP, ESCA, CC, RWG); Veracruz: Barranca de Cayoapa, Teocelo, (UNAM); Jalapa, (USNM); 4.9 km N of Coscomatepec by Mex. Hwy 156 (rwe); Rio Metlac, 2 km WNW of Fortin, 900 m, (rw); GUATEMALA: Suchitepequez Dept.: Finca Moca Grande, Rio Bravo, (rsca). Erpetogomphus liopeltis sp. n. (figs. 14-thx, 36-abd, 57-hamules, 79-penis, 101- app, 117-cercus, 126-ept, 140, 141-occiput, 153- vertex, 182-vl, 201-distr, 227-wings) Erpetogomphus elaps. — Novelo & Penia, 1991: 130 (misiden- tification). Type material. — Holotype male: Mexico: Nuevo Leon State, Municipio de Monterrey, La Estanzuela, 2 July 1987 (M. J. Westfall, Jr., H. Quiréz, A. Contreras). In rsca. Allotype female: Nuevo Leon State, Monterrey, El Diente, 5 mi SE, 2 July 1960 (W.B. Cutts). In UMMZ. Paratypes examined (154,39, including holotype 3 and allotype ©). — Mexico: Hidalgo: Pemuxtitla, Rio Zacuala, 1000 m, 22 April 1984 (R. Novelo), 1 4 (RWG); Michoacán: 12 km W Apatzingän, 369 m (1200 ft), 4th Hoogstraal Mexican Biological Expedition, 12 Aug. 1941 (Harry Hoogstraal), 19 (ummz); Nuevo Leon: Chipingue, 12 July 1960 (W.B. Cutts), 12 (rw); Municipio de Monterrey, Cerro El Mirador, 9 Aug. 1984 (A. Contreras), 26 (tori); same data as holotype, 3d (rsca); Municipio de Santiago, El Cercado, Arroyo Dolores, 17 Aug. 1984 (A. Contreras), 24 (roRI); Municipio de Santiago, Rancho Los Pinos, 3 July 1987 (M. J. Westfall, Jr., Jr., G. Luna, A. Contreras), 56 (rsca, RWG); San Luis Potosf: La Conchita (Camino a Xilitla), 7 May 1950 (no collector), 16 (cc). i Description Holotype male. — Entire head pale green, labrum dull grey green, base of mandible, anteclypeus dull gr- een, tip of mandible black; basal 0.25 and all of vertex brown, a transverse trough at anterior margin of ver- tex at juncture of base with antefrons. this trough 210 with two slightly deeper pits, one each anterolateral to median ocellus; occiput with slightly medial tumid area, occipital crest distinctly emarginate medially (figs. 140-141), dark brown, rimmed with long dark brown hairs; postocciput green, becoming black late- rally, concave medially; rear of head brown, beco- ming yellow brown laterally; antennae dark brown. Prothorax brown, anterior and posterior lobes and small dorsolateral area of median lobe pale green. Thorax (fig. 14) entirely green (probably blue green in life) with following dark brown: obscure middorsal stripe ending before collar, becoming darkest near middorsal thoracic carina, lateral margins of this stripe gradually fading to green on mesepisternum; posterolateral rim of antealar crest; lateral 0.40 of mesepisternum except for dorsal emargination, with no indication of pale green separating coalesced hu- meral and antehumeral stripes; this stripe narrowing slightly ventrally but connecting with lateral margin of collar and anterior 0.50 of mesinfraepisternum; rim below subalar carina at obsolete second lateral su- ture; narrow partial third lateral stripe extending ven- trally to 0.50 of suture before merging with green. Metasternum pale grey brown, tawny anteriorly. Coxae, trochanters grey; base of femora brown, be- coming black distally; tibiae and armature black. Wings hyaline, anterior margin of costa dark brown at base, becoming yellow distally to proximal level of pterostigma, remainder of venation and pte- rostigma black. Venational details. Fifth antenodal thickened in all wings; no supplementary marginal cells behind fore wing paranal cells; anx: fore wing 14/15, hind wing 10/10; pnx: fore wing 8/9, hind wing 9/9; cs under pterostigma: fore wing 7/6, hind wing 6/5; anal trian- gular cells: 4/3. Hind wing 31 mm. Abdomen (fig. 36) with segment 1 green dorsally, paler laterally with a midlateral stripe of brown ex- panding anteriorly at base and darkening posteriorly at articulation point; segment 2 similar to segment 1 but with a ventral stem of brown behind auricle from dark midlateral stripe, midlateral brown becoming wider and darker along posterior 0.50 of segment, an- nulus black; segment 3 with pale green middorsal stripe acuminate and disappearing at posterior 0.10 of segment, black midlateral stripe expanding slightly at transverse carina and again widening at posterior 0.10 of segment, this area with a ventral offshoot ter- minating anteriorly at 0.20 of segment; segments 4-6 similar to segment 3, but with middorsal pale green stripe shorter and narrower on each succeeding seg- ment, thus black midlateral stripe connecting dorsal- ly at posterior 0.40 of segment 6, these midlateral stripes interrupted at anterior 0.10 of each segment with small longitudinal wedge-shaped white area along ventral margin of tergite behind transverse cari- na; segment 7 with basal 0.60 pale grey green, except for irregular diffuse black spot at transverse carina, re- mainder of segment red brown, becoming black dor- sally; segments 8-10 red brown, becoming darker dorsally; carinae and lateral foliations black, posterior medial margin of segment 9 pointed. Cercus (fig. 101) pale ocher, decumbent at posteri- or 0.30, largely parallel in lateral view except for dis- tal 0.20, tip with a shiny black tooth; black inferior carina on distal 0.25; epiproct brown, about 0.45 as long as cercus, U-shaped in lateral view, tip of epi- proct when viewed posteriorly (fig. 126) broadly truncate. Abdomen 40 mm. Accessory genitalia (fig. 57). Anterior hamule dark brown, branched at distal 0.30, superior branch larg- er than inferior branch; posterior hamule pale, rough- ly triangular with a well developed anterior shoulder, distal 0.75 curved posteromedially with a well devel- oped black tooth twisted laterally from broad axis of hamule; penis with cornuae acuminate as shown in fig. 79, lateral lobe prominent, knife-like, serrated posteriorly. Allotype female. — Similar to male with following differences: labrum slightly darker (possibly due to post mortem effects); transverse trough at juncture of antefrons and vertex well defined with slightly larger and deeper pit anterolateral to median ocellus; occi- put strongly reduced to a small, gently convex, large- ly perpendicular medial arch; postocciput tumid me- dially, dark middorsal stripe vestigial, with only a wash of brown on either side of lower arms of antealar sinus, a narrow poorly defined incomplete green stripe separating dark brown antehumeral and hu- meral stripes up to dorsal 0.75. Abdominal segment 3 with acuminate black spot posterolateral to black dor- solateral stripe projecting anteriorly to black trans- verse carina; dorsolateral black stripes on segments 4- 7 more extensive; pale middorsal stripe reduced to basal 0.75 (segments 4-5) to basal 0.50 (segment 7), and anteriorly touching annulus on each segment; segments 8-9 black dorsally, red brown laterally. Abdomen 38 mm. Vulvar lamina (fig. 182) small, cleft larger than each plate, each plate connected basally, each with a raised shelf occupying anterior 0.5, postlamellar ridge with Y-shaped juncture posterior to hind margin of lamina, central stem with a well defined circular de- pression on either side. Venational details. Fifth antenodal thickened in left fore wing; the sixth in remaining wings; Number of marginal cells behind fore wing paranal cells: 2/3; anx: fore wing 16/17, hind wing 13/11; pnx: fore wing 10/12, hind wing 12/11; cs under pterostigma: fore wing 6/7, hind wing 6/7. Hind wing 34 mm. GARRISON: Revision of Erpetogomphus Diagnosis This remarkable species approaches E. viperinus in size, colour, and maculation, but is most closely rela- ted to £. bothrops in morphology. Males of E. hopeltis and £. bothrops are similar in possessing epiprocts with broadly truncate tips (figs. 126-128); but those of £. bothrops are usually, but not always, bidentate (fig. 128). In lateral view, the epiprocts of E. liopeltis are more strongly curved and extend about 0.5 or less the length of the cercus; the same structures in £. bothrops extend 0.75 or more the length of the cercus (fig. 102). The apical tooth of the posterior hamule of £. /iopeltis is longer and not as tightly recurved as in E. bothrops (figs. 57-58). Abdominal segments 1-3 in E. hopeltis (fig. 36) are more heavily marked than in £. bothrops (fig. 37); and the dark thoracic markings are also mo- re extensive (fig. 14) than in £. bothrops (fig. 15). In size and overall colouration, £. liopeltis seems in- distinguishable from £. viperinus but differences in the tips of the epiproct (figs. 126, 129) and structure of the area anterior to the vertex in both sexes (figs. 136 [as for E. bothrops|, 137) easily distinguish the two species. Subtle differences between these two spe- cies exist in the morphology of the distal 0.5 of the cercus. In E. liopeltis, these structures are thicker and more cylindrical when viewed dorsally (fig. 101) and mediodorsally (fig. 117). The cercus of £. viperinus is more laterally compressed (fig. 119). The tip of the cercus in £. liopeltis terminates in a sharp black tooth; this tooth is lacking in E. viperinus. Females of E. liopeltis and E. bothrops can be separ- ated by overall size (£. liopeltis hind wing 32-34 mm; E. bothrops hind wing 28-31 mm) and thoracic mac- ulation and colouration, as stated for males. Structural differences of the occiput and postocciput easily distinguish these species. The occiput of E. /io- peltis is rudimentary, with only a gently rounded ver- tical medial crest. The medially tumid postocciput is easily visible in dorsal view (fig. 153). In E. bothrops, the occiput is narrow, but the anterior and posterior margins are roughly parallel except for the angulate lateral margins (fig. 154). In E. bothrops, the crest is vertical and the exposed postocciput is easily visible; but its posterior margin is planar or only slightly tu- mid. Receptacles for the tips of the male cerci differ slightly between the females. The transverse trough located anterior to the median ocellus narrows lateral- ly in E. bothrops (fig. 136); but the lateral area of the greater U-shaped trough tends to curve lateroposteri- orly around the anterior margin of the median ocellus in E. liopeltis (as in fig. 139). Like males, females of E. liopeltis and E. viperinus can show striking similarities, but greater morpholog- ical differences manifest themselves in structures of the head and vulvar laminae. The trough anterior to the median ocellus of £. liopeltis is unlike the antero- > [1IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 lateral pits of E. viperinus (fig. 137). The occiput of E. viperinus (fig. 155) is much broader, and the anterior margin of the occiput and curvilinear occipital crest are roughly parallel. The median postoccipital swell- ing of £. liopeltis is absent in E. viperinus. The vulvar laminae of the two species are similar, but the Y- shaped postlamellar ridge generally arises at or just before the posterolateral margins of the lamellar plates in £. liopeltis the postlamellar ridge usually sur- passes the vulvar lamina in £. viperinus. The median cleft between the vulvar lamellae is more obtuse in £. viperinus (fig. 184) than in £. liopeltis (fig. 182). Erpetogomphus liopeltis and E. viperinus are allopat- ric and £. viperinus generally has more dense vena- tion, despite wing lengths (£. /iopeltis hind wing 29- 31 mm [n = 13 males, 3 females], E. viperinus hind wing 29-31 mm [n = 20 males, 20 females]). The number of marginal cells behind fore wing paranal cells in male E. /iopeltis ranges from 0-2 (only one wing has 3), compared with 1 (in 4 wings only) to 4 in E. viperinus. Both sexes of £. liopeltisand E. elaps are easily diag- nosed in the key by differences in body colouration, shape, and position of the posterior hamular tooth and shape of the cercus (males) and shape of the vul- var laminae (females). The occiput of females of E. elaps, E. elaphe, and E. liopeltis are similar in being re- duced to a perpendicularly raised medial area (figs. 152-153) and a medial tumid area on the postocci- put. Remarks Variation. — I found little variation among the 12 paratype males and 2 paratype females. The hind margin of the occiput (crest) is medially concave in the holotype (fig. 140), but ranges to almost linear (fig. 141) in other specimens. A poorly defined nar- row green stripe is present between the coalesced dark antehumeral and humeral thoracic stripes in 7 males, but this condition varies from a short, narrow mark to one occupying the medial 0.50 of the lateral margin of the mesepisternum. -Wing variation among paratypes: Males: anx: fore wing 13-15/13-16, hind wing 9-12/10-12; pnx: fore wing 8-10/8-10, hind wing 9-13/9-12; cs under pte- rostigma: fore wing 5-7/4-6, hind wing 4-6/5-7; anal triangular cells: 3-4/3-4. Hind wing 29-31 mm. Abdomen 37-40 mm. Females: Anx fore wing 16/14-15, hind wing 10/10; pnx fore wing 8-10/9-10, hind wing 11/9-11; cs under pterostigma fore wing 6-7/6, hind wing 6- _8. Other aspects of wing venation variability are char- ıcterized under the diagnosis for E. liopeltis and E. vi- Derinus. Some (n = 5) males have no marginal cells behind 212 fore wing paranal cells, while two others (fig. 227) have one marginal cell in only one wing. All females have at least two such cells, except for the left fore wing of the Chipingue female. The fifth antenodal is almost always thickened; but the sixth is thickened in two males from La Estanzuela and in the right fore wing of the Apatzingan female. One male from La Estanzuela has aberrant hind wing venation and was not included in the measured samples. Most specimens, including the holotype, had been preserved in acetone; and their resultant preservation leads me to believe that the pale coloura- tion is similar to the blue green of £. viperinus. Biology. — Novelo (pers. comm.) misidentified this species as E. elaps in Novelo and Pefia (1991). They collected £. /iopeltis in Hidalgo state from 22 April to 25 July (Rio Zacuala, Pemuxtitla) and 27 July (Calnali). They were taken as they perched on rocks along narrow, shaded, shallow, rocky streams. Erpetogomphus liopeltis was collected at Cerro El Mirador with E. elaps. Data for the female collected in Michoacan indicate that it was collected on ‘side of stream’. Collection dates range from 22 April (Rio Zacuala) to 17 August (Arroyo Dolores). Distribution (fig. 201). — Erpetogomphus liopeltis is only known from the states of Hidalgo, Michoacan, Nuevo Leon, and San Luis Potosi in northeastern Mexico. Erpetogomphus bothrops sp. n. (figs. 15-thx, 37-abd, 58-hamules, 80-penis, 102- app, 118-cercus, 127, 128-ept, 136-base of postfrons, 154-vertex, 183-vl, 201-distr, 228-wings) Erpetogomphus viperinus. — Calvert 1899: 385 (described as E. viperinus Selys); Calvert 1905: 163 (described as £. vi- perinus Selys); Calvert 1909: 35 (misidentified as E. vi- perinus, compared with E. schausi); Muttkowski 1910: 87 (in part, cat.); Williamson & Williamson 1930: 11 (in part, refers to true E. viperinus but specimens in E. B. Williamson Coll. misidentified); Calvert 1947: 608 (stat- us of Tepic specimens); Paulson 1982: 256 (in part, as £. viperinus from Mex., Guat.); Davies and Tobin 1985: 28 (in part, cat); Tsuda, 1986: 87 (in part, cat); Maes et al. 1988: 36 (as E. viperinus from Nicaragua); Tsuda 1991: 95 (in part, cat.). Type material. — Holotype male: Mexıco: Veracruz: Rio Otapa, 8 km S of La Tinaja, elev. 90 m, 13 Aug. 1976 (R. W. Garrison). Allotype female: same data, but 20 Aug. 1976. In usNM. Paratypes examined (618, 289). — Mexico: Guerrero: Dos Arroyos, 1000 ft., Sept. 1888 (H. H. Smith), 19 (BmnH); Jalisco: San Diego Rancho near Cocula, 14 Nov. 1923 (J. H. Williamson), 16 (BMNH); Michoacan: El Sabino, 20 mi SSE of Uruapan, 27 July, 1 Aug. 1936 (H. Devlin Thomas), 86, 39 (ummz, Fsca); Morelos: Puente de Ixtla, 3 July 1900 (C.C. Deam), 19 (ummz); Cerro del Higuerón-Jojutla, July 1983 (C. Deloya), 1d (UNAM); Nayarit. Acaponeta, 1-2 Nov. 1923 (J. H. Williamson), 9d, 12 (ummz); Tepic, 7 Nov. 1923 (J. H. Williamson), 16 (ummz); Jumatän, 10 Sept. 1980 (G. Jiménez), 2d (UNAM); San Luis Potosí, Cascadas Micos nr. aqueduct, 27 June 1990 (K. J. Tennessen), 1d (KJT); El Salto, 400 m, 6 Sept. 1963 (T. W. Donnelly), 1d, 19 (rw); Huichihuayan, km 410, S of Valles, ‘Sam Brown’ Hacienda, 25 Sept. 1938 (L. J. Lipovsky), 15,2? (ummz); Tamaulipas: Rio Corona, 20 mi N of Ciudad Victoria, nr. Mex. Hwy 101, 1000 ft., 26 July 1968 (R. W. Garrison), 16, 29 (rwc); Veracruz Atoyac, 400 m, (Schumann), 19 (BMNH); 16.5 mi S of Catemaco, by Hwy 180, 25 June 1985, (I. S. Askevold), 2d (cc); Cordoba 12-25 July 1964 (E. Fisher, D. Verity), 19 (Lacm); Isla, 17 July 1969 (R. Wind), 1d (cc); K375-390 Cordoba Rd., 11 Aug. 1961 (R. and K. Dreisbach), 16 (usnm); 4.9 km N of Coscomatepec, by Mex. Hwy 156, 11 Aug. 1976 (R. W. and J. A. Garrison), 19 (rw); Rio Otapa, 8 km S of La ina 90m 013, 20 Aug. 1976 (R. W. and J. A. Garrison), 104, 22 (rwe); Rio Hondo, on road to ‘Colonia la Apachital’ (50°), Soteapan, 500 m, July- Aug. 1990, (collector unknown), 68,4? (pM, RWG); 6 mi E of Hwy from Tierra Blanca to Ciudad Aleman, 25-26 Aug. 1957 (G. H. Beatty, III), 1d (esca); Salto Eyipantla, 8 km S of San Andres Tuxtla and Mex. Hwy 180, 15 Aug. 1976 (R. W. and J. A. Garrison), 28 (Rw); Tierra Colorada, nr. Veracruz, 17 July 1932 (H. M. Smith), 19 (ummz); Oaxaca: Candelaria Loxicha, 500 m, 7 Sept. 1973, (E. C. Welling-M.), 1d (cc); 5 July 1974, (E. C. Welling- M.), 16 (swp); Chiapas river 26.1 mi NE Tapanatepec, 2100 ft., 2 Aug. 1965 (D. R. Paulson), 14 (DRP); stream 15.4 mi NE Arriaga on Mex. 195, 2300 ft., 24 July 1965 (D. R. Paulson), 26 (pre); El Aguacero, nr. Ocozocoautla, 26 Oct. 1986 (E. Fisher), 16 (RwG); GUATEMALA: Baja Vera Paz Dept.: San Geronimo, 1879-80 (G. C. Champion), 1d (BMNH); Chiquimula Dept: streams vic. Tierra Colorado, 800 m, 20 July 1962, (T. W. Donnelly), 14,29 (rwp); small stream vic. Veguitas, 600 m, 10 July 1962 (T. W. Donnelly), 18, 12 (rwp); Dept. Zacapa: Gulan, 16-17 June 1909 (E. B. Williamson), 26 (ummz); La Union, 850 m, 31 Aug. 1972 (E. C. Welling M.), 19 (ummz); EL SALVADOR: Anuachopan Dept.: (all collected by V. Hellebuyck): Bosque El Imposible, San Francisco Mendez, 16 Aug. 1987, 35 (vx, RwG); May 1987, 1d, 12 (va); 12 July 1987, 26 (va); 25 July 1987, 18, 12 (vm); 16 Aug. 1987, 16 (vx); 3 Sept. 1987, 36,1 (1 pair in copula) (va); El Coyolar, El Imposible, 2 Aug. 1987, 2d (vn); Valle de la Puerta, Bosque El Imposible, 3 Sept. 1987, 34 (vu). GARRISON: Revision of Erpetogomphus Description Holotype male. — Labium grey-white with wash of brown around margin of median lobe; entire face in- cluding vertex and occiput yellow green, a transverse trough in front of median ocellus, lateral ends slight- ly deeper than medial area; a small tubercle behind each ocellus; pedicel, scape, flagellum brown; occiput yellow green, mostly planar, slightly tumid medially, crest green, slightly prominent, straight, slightly notched medially and covered with long brown hairs; postocciput green, transverse when viewed dorsopos- teriorly: rear of head yellow brown, paler laterally; la- teral margins of labrum and base of mandible ocher; rear of head yellow brown. Prothorax yellow green except for brown on anteri- or margin of median lobe and posterior area of ante- rior lobe. Synthorax (fig. 15) entirely yellow green ex- cept for following brown areas: slight vague stripe lateral to middorsal carina; antehumeral stripe; spot on upper end of humeral suture connected anteriorly to antehumeral, and gradually disappearing basally toward mesinfraepisternum; line bordering antealar crest; metasternum pale yellow green. Coxae, femora pale yellow green, becoming dark brown dorsodistal- ly; tibiae, tarsi and armature black. Wings hyaline, anterior margin of costa yellow, re- mainder of venation black, pterostigma brown. Venational details. Fifth antenodal thickened in all wings; number of marginal cells behind fore wing paranal cells: 1/3; anx: fore wing 15/15, hind wing 11/11; pnx: fore wing 11/13, hind wing 13/13; cs under pterostigma: fore wing 6/6; hind wing 7/6; anal triangular cells: 4. Hind wing 27 mm. Abdomen (fig. 37) with segment 1 yellow green with a wash of brown dorsally; segment 2 yellow green with diffuse brown spot posterodorsally to auri- cle; segment 3 broadly light green dorsally, interrupt- ed laterally by a longitudinal brown stripe flaring at transverse carina, inferior margin of tergite ivory; seg- ments 4-6 similar to segment 3, but lateral brown stripes progressively more extensive and each meeting its neighbor dorsally at posterior end of segment; seg- ment 7 with anterior 0.66 pale green white except for black transverse carina, posterior 0.30 of segment red brown; segments 8-10 red brown, darker dorsally, carinae and lateral foliations black; posterior medial margin of 9 pointed. Abdomen 35 mm. Cercus (fig. 102) pale ocher, gently decumbent at posterior 0.30, gradually narrowing toward end, tip with a shiny black tooth; epiproct brown, gently curved as is characteristic for the genus, tip of epi- proct when viewed posteriorly (fig. 128) bidentate. Accessory genitalia. Anterior hamule dark brown, branched at distal 0.30, superior branch larger than inferior branch; posterior hamule (fig. 58) pale, roughly triangular with an anterior shoulder, tip with 213 T1IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 a black tooth twisted laterally from broad axis of ha- mule; penis with cornuae pointed but each with me- dian shoulder as shown in fig. 80; lateral lobe promi- nent, knife- like, serrated posteriorly. Allotype female. — Similar to male with following differences: synthorax with definite dark triangular middorsal stripe, its base not touching collar; broad, diffuse brown dorsolateral stripe on abdominal seg- ment 2, brown lateral stripes on segments 3-6 with their anterior ends ending at basal 0.20 of each seg- ment so that lateral white connects with dorsal yel- low-white middorsal stripe; cercus pale. Vulvar lamina (figs. 183) small, cleft about as large as each plate; each plate connected basally, each with a raised shelf occupying anterior 0.5; postlamellar ridge with Y-shaped juncture posterior to hind mar- gin of lamina; central stem with a well defined circu- lar or oval depression on each side. Abdomen 32 mm. Venational details. Fifth antenodal thickened in all wings; no marginal cells behind fore wing paranal cells; anx: fore wing 12/13, hind wing 9/9; pnx: fore wing 8/9, hind wing 10/10; cs under pterostigma: fore wing 4/5, hind wing 5/5. Hind wing 27 mm. Diagnosis Selys (1868) description of Erpetogomphus viperi- nus was too brief and inexact for Calvert (1899) to de- tect a specific difference between his specimens of E. viperinus (= E. bothrops) and the true E. viperinus. Examination of the lectotype male and female of E. viperinus shows that the Erpetogomphus viperinus of Calvert and all subsequent authors is referable to the new species E. bothrops. E. bothrops is most similar to E. liopeltis and is diag- nosed under that species. Erpetogomphus bothrops also resembles £. viperinus, but in life, the pale coloura- tion of E. bothrops is yellow green (deep green in £. viperinus). Erpetogomphus viperinus is generally larger (hind wing male 29-31 mm) and darker. In males, the cercus of E. bothrops is pale with a distinct black tooth; while in £. viperinus this structure is dark red brown with no apical tooth. In E. bothrops, the distal 0.5 of the cercus gradually narrows, so the distal 0.5 of the cercus in lateral view is not as thick or robust as the base (fig. 102). In E. viperinus, the distal 0.5 of the appendage is robust and is as thick as its base (fig. 103). In dorsoposterior view, the posterior 0.30 of the cercus of E. bothrops gradually narrows (fig. 118), and is not as laterally compressed as in E. viperinus (fig. 119). The inferior distal margin of the cercus of E. viperinus is strongly carinate, not so in E. bothrops. [he tip of each epiproct of E. bothrops in posterior view is bidentate (fig. 128) or truncate (fig. 127), not bluntly pointed as in £. viperinus (fig. 129). Other characters are given in the key. Females of these species are easily distinguished by overall body colouration in life, and thoracic macula- tion. Two structural characters separate £. bothrops from E. viperinus the anterior margin of the vertex of E. viperinus contains two well defined pits, each dor- solateral to the median ocellus. Each pit is the recep- tacle for the bluntly pointed tip of the male epiproct when in copulation. In £. bothrops, these pits are re- placed by a continuous groove deepest medially, which accommodates the explanate tips of the male epiprocts. Secondly, the V-shaped notch of the vulvar lamina in £. bothrops is more acute (fig. 183) than the more widely divergent notch of £. viperinus (fig. 184). The shape of the vulvar lamina of £. bothrops (fig. 183) distinguishes it from similarly marked fe- males of E. sipedon (figs. 187, 188) and E. elaps (fig. 181). Remarks Variation. — Venational variation among the para- type series of males (fig. 228): number of marginal cells behind fore wing paranal cells: 0-3; anx: fore wing 14-16/13-16, hind wing 9-12/10-12; pnx: fore wing 9-11/8-13, hind wing 10-13/10-13; cs under pterostigma: fore wing 5-7/5-7, hind wing 5-7/5-7; number anal triangular cells: 3-4. Hind wing 21-28 mm. Variation exists in the direction of the apical tooth of the posterior hamule. It is present in all males and, in most, is twisted so that its tip is directed laterally (Ge, distally) to the direction of the planar surface of the hamule. In a few males, the tooth, though rotat- ed, points in a more lateroanterior direction. The tip of the epiproct, when viewed posteriorly, is planar and bidentate (e.g., fig. 128); but in some, the distal and mesal teeth are reduced. In rare cases, the tip is almost straight (fig. 127); variations link these extremes. I have found no correlation of the posterior hamular condition with the epiproct condition; nor do these conditions seem to vary according to altitude or locality. For example, of 5 males from Nayarit, Jamatan, only one has hamular teeth pointing pre- dominantly cephalad. This same specimen has a bi- dentate condition of tips of the epiprocts. Another male has a bidentate condition, another has an inter- mediate condition, and two others have truncate tips. Of another series of 9 males (including holotype) from the Veracruz, Rio Otapa, two have the hamular tooth pointing predominantly cephalad. These two specimens have bidentate epiprocts, although the right epiproct of one male is intermediate. The re- maining 7 males have bidentate epiprocts, but one has an intermediate condition of its right epiproct. The membranous hood of the penis is variable among and within populations. Most specimens, in- cluding the holotype, possess a short hood, exposing a subcutaneous membrane (fig. 80); but a male from San Luis Potosi, Cascadas Micos, has a long, acute, non-overlapping hood. Another two males, one from Guatemala (Dept. Chiquimula, small stream, vic. Veguitas), and all specimens I examined from El Salvador (Dept. Anuachopan, Bosque El Imposible, San Francisco Mendez), have long, overlapping hood membranes similar to that illustrated for the holotype male of £. leptophis (fig. 76). With the exception of the San Luis Potosi male, specimens with long, acuminate hoods seem to typify most southerly specimens. I can find no other characters differentiating these speci- mens from others. Venational variation among the paratype series of females (n = 18): number of marginal cells behind fore wing paranal cells: 1-3; anx: fore wing 12-16/12-16, hind wing 9-11/8-11; pnx fore wing 8-12/9-12, hind wing 9-13/9-13; cs under pterostigma: fore wing 4- 7/5-7, hind wing 5-8/5-7. Hind wing 27-31 mm. Biology. — I have often seen this species along mar- gins of cut agricultural fields bordering streams and canals. Its apple green colouration renders it difficult to detect among vegetation, but when disturbed, it does not fly far. The type locality is a wide, shallow stream bordered by trees in extensively cultivated farmland. I collected Æ. bothrops next to cut fields along with six other gomphids: Erpetogomphus eutai- nia, E. ophibolus, Phyllocycla breviphylla Belle, P. vol- sella (Calvert), Phyllogomphoides duodentatus Donnelly, and Progomphus clendoni Calvert. Twenty- five other species of Odonata were collected there. R. Novelo (pers. comm.) observed a pair in copula in Morelos state (Cerro del Higuerön) at 1250 m at 18.20 hr on 14 Sept. 1988. Williamson records the following on two Acaponeta males: ‘Easfilly caught while sitting in brushy weed patch on bank of river, sun behind cloud,’ and “Quite common flying over water just above the real rapids and lighting on willow-like stems and twigs along the bank. Saw no 2 gomphine today.’ López (in litt.) writes that they perch along sides of streams or on exposed rocks in the middle of streams. He states that they do not remain long at a spot, mov- ing constantly along the length of the stream. Elevation gradients range from near sea level (50 m, Veracruz, Agua Caliente) to 1250 m (Morelos, Cerro del Higuerön). Collection dates range from May (El Salvador) to November (Mexico, Jalisco). Distribution (fig. 201). — This species and £. e/aps are the most widely distributed species in Mexico and Guatemala. Erpetogomphus bothrops ranges from Tamaulipas in eastern Mexico (24°N) south to El Salvador (14°N). It also occurs in mesic areas in west- ern Mexico from Nayarit (22°N) south. Its distribu- tion indicates an avoidance of the xeric areas in north- western Mexico and the high mountain plateau in northern Mexico. GARRISON: Revision of Erpetogomphus Erpetogomphus viperinus Selys (figs. 16-thx, 38-abd, 59-hamules, 81-penis, 103- app, 119-cercus, 129-pet, 137-base of postfrons, 155-vertex, 166-postocciput, 184-vl, 201-distr, 229- wings) Erpetogomphus viperinus Selys, 1868: 68 (3 sep.) (descr. of 3, ?, D'Orizaba’). — Selys 1869: 176 (13 sep.) (rede- scription of d and ® from Orizaba); Selys 1873b: 519 (75 sep.) (list); Muttkowski 1910: 87 (in part, cat.); Williamson & Williamson 1930: 14 (summary of status); Paulson 1982: 256 (Mex.); Davies and Tobin 1985: 28 (cat.); Tsuda 1986: 87 (cat.); Bridges 1991: VII.220 (cat.); Tsuda 1991: 95 (cat.). Herpetogomphus viperinus. — Hagen 1875a: 42 (cat.); Selys 1879: 64 (2 sep.) (status of classification); Kirby 1890: 60 (cat.). Description Male. — Entire head pale green (blue green in life), slightly darker on anterior of ante- and postclypeus and labrum, lateral margins of labrum ocher, base of mandible and labium dull grey green; basal 0.25 of postfrons and all of vertex brown, a pair of pits at an- terior margin of vertex at juncture of base of antefrons (fig. 137), one each anterolateral to median ocellus; antennae dark brown; occiput with well developed medial tumid area, green, occipital crest slightly emarginate medially, dark brown to black, rimmed with long dark brown hairs; postocciput brown, gr- een medially, transverse, with a slight vertical depres- sion medially; rear of head yellow brown. Prothorax brown, anterior and posterior lobes green. Synthorax (fig. 16) entirely green (blue green in life) with following dark brown: obscure middorsal stripe ending before collar, becoming darkest at base of antealar crest and extending as narrow wash of brown below rim of antealar crest and joining wide stripe along lateral 0.30 of mesepisternum; this stripe coalescing with normal humeral stripe; often with a narrow isolated stripe of green separating these two stripes; combination antehumeral and humeral stripe reaching collar; all of mesinfraepisternum and ventral part of thorax; rim below subalar carina, often with small extension on obsolete second lateral suture, which in some specimens may form an indistinct sec- ond lateral stripe; narrow but well defined third later- al stripe. Metasternum pale grey green. Coxae, tro- chanters grey, becoming brown exteriorly; base of femora brown, becoming black distally; tibiae and ar- mature black. Wings (fig. 229) hyaline, anterior margin of costa dark brown at base, becoming yellow distally to prox- imal level of pterostigma, remainder of venation and pterostigma black. Venational statistics. Fifth (occasionally sixth) an- tenodal thickened in all wings; number of marginal cells behind fore wing paranal cells: 1-4/1-4; anx: fore TI) DScHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 wing 13-18/13-18, hind wing 9-12/10-13; pnx: fore wing 8-13/10-12, hind wing 10-14/10-13; cs under pterostigma: fore wing 5-7/4-7, hind wing 5-7/5-7; anal triangular cells: 4 (rarely 5). Hind wing 29-31 mm. Abdomen (fig. 38) with segment 1 green with brown dorsolaterally except for green on posterior dorsal 0.50 of segment; segment 2 with middorsal green stripe; brown dorsolaterally surrounding green auricle; ventrolateral area pale green, annulus black; segments 3-6 primarily black with following pale are- as: basal 0.10 of each segment, dorsally forming an in- complete blue green middorsal stripe extending to posterior 0.20 of each segment; ventrolateral wedge shaped white spot at middle of each segment, con- necting in some specimens to basal pale ring; segment 7 with dorsal 0.50 light green, posterior 0.50 red brown, black along transverse carina; segments 8-10 predominantly red brown, darker dorsally, denticu- lated posterior margin of these segments black; poste- rior margin of segment 9 often forming a posteriorly directed point. Abdomen 36-40 mm. Cercus (fig. 103) pale ocher, slightly darker ven- trally, robust and strongly curved at posterior 0.30, inner margin of posterior 0.30 of cercus planar (fig. 119); same structure in posterodorsal view narrow, inferior carina along posterior 0.30 of cercus; epiproct brown, becoming black posteriorly, gently curved as is characteristic for the genus; tip of epiproct, when viewed posteriorly (fig. 129) forming a narrow, blunt point. Accessory genitalia. Anterior hamule (fig. 59) black, divided at distal 0.30; smaller posterior branch almost as large as anterior branch, its pointed tip al- most meeting tip of anterior (larger) branch; posteri- or hamule (fig. 59) pale, roughly triangular, swollen at distal 0.25 to 0.50 of appendage, tip with a black tooth twisted laterad from broad axis of hamule; penis with cornuae pointed but each with well developed median shoulder (or median lobe in some specimens), as shown in fig. 59; lateral lobe prominent, knife-like, arcuate laterally, its outer margin strongly serrated. Female. — Similar to male, with following differ- ences: Vertex (fig. 155) without postocellar ridge, a well defined pit anterolaterad to median ocellus as in male; occiput green, narrow, mostly planar, or with only a slight tumid area medially, hind margin slight- ly sinuate, barely emarginate at middle; postocciput green, transverse; synthorax with no vestige of second lateral stripe; abdomen with dorsolateral dark stripe on segment Ì reduced or vestigial, middorsal dark stripe on segments 2-6 narrower, anteriorly touching black annulus, widened at transverse carina, con- stricted near center and widening at distal 0.25 of each segment, ventral white narrow but expanded near center of segment, sometimes a longitudinal 216 wedge-shaped spot isolated by narrow posterior and anterior lateral offshoots of black dorsolateral stripe, white lateral wedge- shaped spots largest on segments 2 and 3, dividing or partially dividing dorsolateral brown stripe; segment 7 similar to segment 6, but dorsal pale orange green connecting anteriorly with lateral white at basal 0.10 of segment; segments 8-10 dark brown, becoming black dorsally and posterolat- erally to all black with ill-defined white lateral spot on each segment; cercus, paraproct brown. Abdomen 37-41 mm. Vulvar lamina (fig. 184) small, cleft as large as each plate; each plate broadly connected basally, each with a well developed raised shelf occupying anterior 0.50; medial margin of each plate slightly concave, its tip falcate; postlamellar ridge with Y-suture posterior to hind margin of lamina; central stem short with a well- defined circular or oval depression on each side. Venational statistics. Anx: fore wing 15-20/15-21, hind wing 10-14/11-14; pnx: fore wing 10-13/10-15, hind wing 10-15/9-14; cs under pterostigma: fore wing 5-8/5-8, hind wing 5-8/6-7. Hind wing 31-34 mm. Diagnosis Erpetogomphus viperinus is most closely related to E. bothrops, and is diagnosed under that species. It is also superficially similar to E. /iopeltis and is diagno- sed under that species. The thoracic pattern of E. vi- perinus is superficially similar to that of £. sipedon, but is easily separated by characters listed under that spe- cies and in the key. Remarks Variation. — The description is based on 49 males and 38 females from Veracruz State, Mexico. The lectotype and paralectotype are in reasonably good condition, but post mortem preservation is poor and would yield an in- adequate description. Venation characters for this pair are as follows: Lectotype male: number of marginal cells behind fore wing paranal cells: 3/1; anx: fore wing 16/17, hind wing 12/12; pnx: fore wing 13/12, hind wing 11/13; cs under pterostigma: fore wing 6/6, hind wing 6/5; anal triangu- lar cells: 4/4. Paralectotype female: number of marginal cells be- hind fore wing paranal cells: 4/3; anx: fore wing 15/18, hind wing 12/11; pnx: fore wing 10/12, hind wing 12/13; cs under pterostigma: fore wing 7/8; hind wing 716. Examination of 20 males shows variability is some ve- national characters. All specimens have marginal cells behind the fore wing paranal cells, ranging from 1-4 cells. The fifth antenodal is most often thickened, but rarely the fourth (one wing) or sixth (7 wings). The sixth antenodal is thickened on both wings of the lectotype. Little variation exists in body colouration. The sec- ond lateral thoracic stripe is incomplete or vestigial in most males. It is complete on only one male from Rio Metlac. Females show a greater tendency toward a dense venation than do males. Marginal cells behind the fore wing paranal cells were present in all 20 fe- males I examined, ranging from 1 (one wing) to 8 (two wings). The fifth antenodal is generally thick- ened, but it is the fourth on the left wing of the para- lectotype, sixth in 17 wings (including right wing of paralectotype) and seventh in four wings. Biology. — Although of restricted distribution, £. viperinus can be common during certain years. I col- lected a few specimens of both sexes in the mostly shaded understory at the Rio Metlac. The locality consists of a fast running river in a steep canyon. The narrow valley contained many coffee plants. I flushed various specimens, and they alit on low vegetation, where their deep blue green colour made them diffi- cult to detect unless the eye followed closely where each specimen landed. Once approached they were easy to take with a net. Other Odonata taken at the same site were Hetaerina cruentata Rambur, Argia sp. nr. fissa Selys, Argia extranea Hagen in Selys, Aeshna psilus Calvert, Brechmorhoga pertinax (Hagen), B. vi- vax Calvert, Cannaphila vibex Hagen, Erythrodiplax fusca (Rambur), and E. umbrata (Linnaeus). The species has been collected more frequently far- ther north at Teocelo under similar circumstances. The Odonata assemblage there is more varied and consists of Palaemnema sp. n., Paraphlebia zoe Selys, Argia extranea, Argia sp. n., Brechmorhoga pertinax, B. rapax Calvert, B. tepeaca Calvert, Cannaphila vibex, Libellula herculea Karsch. This species has been taken at elevations of 800- 1300 m (Teocelo and vicinity). Collection dates range from 6 June through 18 September (Teocelo and vicinity). I have abstracted the following biological notes from López (in lit): Tenerals and adults travel far from the stream, hiding in vegetation in open areas, where they feed by making short sallies from perches. They capture primarily small flies and butterflies (Lycaenidae) passing by. Copulation takes place in open fields away from streams, as males grab females which fly by. The copulating pair then retires to the dense forest. Females oviposit while flying above wa- ter, dropping their eggs from about 30 cm. At dusk, males return to feeding sites, often in considerable numbers, where threat display patterns among males and females have been observed when two or more individuals try to occupy the same perch. Specimens are often preyed upon by spiders (Araneidae). Distribution (fig. 201). — Erpetogomphus viperinus is apparently restricted to central Veracruz. As stated under the species account for E. bothrops, all pub- Teocelo, (UNAM); Santa Rosa, Teocelo, GARRISON: Revision of Erpetogomphus lished records of this species since Calvert’s redescrip- tion of E. viperinus refer to the widespread £. both- rops. I examined one male from the Paris Museum with a handwritten label, ‘N. Carolina’, but I have seen no specimens of E. viperinus from the United States. I believe that locality is in error. Material Type data. — Lectotype male by present designa- tion with following data: small green handwritten la- bel ‘Mex/B[ouchard]’; white handwritten label “Herp. / viperinus /S.[elys]/ 3’; two manila coloured labels each with ‘26’ handwritten in pencil in an unknown hand; rectangular red label with printed ‘LECTOTY- PE /and handwritten: “Evpetogomphusl viperinus Selys 1869/ G /des. [printed] R. W. Garrison 1984’. The caudal appendages are shown in fig. 103. Paralectotype female: small green handwritten label ‘Mex./Blouchard]’; white handwritten label ‘Herp./ viperinusl Slelys)/ 2’, two manila coloured labels, each with ‘28’ handwritten in pencil in an unknown hand; rectangular white label with printed ‘LECTOTY- PE / and handwritten ‘Erpetogomphusl viperinus Selys 1869/ @/ des. [printed] R. W. Garrison 1984’. Both specimens in IRSN. Other material (48d , 329, including lectotype d and paralectotype 2). — Mexico: Veracruz nr. Municipio Teocelo at Puente Teocelo, ca. 1150 m, (UNAM, RWG); Barranca de Cayoapa, Teocelo, 750- 890 m, (UNAM, JB, CE, PSM, DALD); Barranca de Cayoapa, Tejeria-Teocelo, 800 m, (UNAM); road to Monte Blanco, Teocelo, (UNAM); El Trapiche, Teocelo, 1100 m (UNAM); road to Santa Rosa, (UNAM); Cascada de Xico, Teocelo, 1100 m, (UNAM); Cascada Texolo, (UNAM); Parque Javier Clavijero, Jalapa, 1300 m (UNAM); Fortin de las Flores, 1010 m (RwG); Rio Metlac, ca. 3.5 km WNW of Fortin de las Flores, 900 m (rwc); km 327 W of Cordoba, (rsca); Cordoba, (USNM). Erpetogomphus designatus Hagen in Selys (figs. 17, 18-thx, 60-hamules, 82-penis, 104-app, 130, 131-ept, 156, 157-vertex, 185, 186-vl, 202- distr) Erpetogomphus designatus Hagen in Selys, 1858: 661 (401 sep.) (descr. d, 2, “Pecos River, Texas’). — Hagen in Selys 1859: 536 (10 sep.) (descr. d, 2); Selys 1873b: 519 (75 sep.) (list); Calvert 1899: 386 (mentioned); Calvert 1905: 166 (Ohio, Ind., Mexico); Muttkowski 1910: 86 (cat.); Calvert 1912a: 289 (Mex., distr.); Hine 1913: 96 (Ohio); Williamson 1914b: 447 (Tex.); Kennedy 1917a: 544 (larva, notes); Williamson 1917: 8 (Ind.): Kennedy 1917b: 137 (Kans.); Kennedy 1918: 298 (notes); Williamson 1923: 8 (Kentucky); Montgomery 1925: 386 (Ind., habits); Montgomery 1927: 289 (Ind.); TIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Kennedy 1928: 373 (seasonal distribution); Montgomery 1929: 340 (Ind.); Needham & Heywood 1929: 80 (key, descr.); Byers 1930: 53 (Fla. Key, descr., habits); Williamson & Williamson 1930: 12 (summary of status); Bvers 1931: 51 (Tenn.); Bird, 1932: 51 (Okla.); Williamson 1932: 23 (Mo., habits); Tinkham 1934: 216 (Tex.); Montgomery 1935: 234 (Ind.); Borror 1935: 453 (Ohio); Borror 1937: 186 (Ohio); Montgomery 1937: 207 (Ind., habits); Ahrens 1938: 11 (Mo., predation); Wright 1938: 27 (Tenn.); Ferguson 1940: 5 (Tex.); Montgomery 1940: 289 (S. C.); La Rivers 1940b: 63 (Nev.); Montgomery 1941: 230, 238 (Ind.); Ferguson 1942: 146 (Tex.); Montgomery 1947: 165 (distr.); Bick 1951: 179 (Okla.); Montgomery 1951: 207 (Ind.); Needham and Westfall 1955: 146 (key, descr.); Cross 1955: 11 (S. C.); Montgomery 1955: 133 (Ind.); Cross 1956: 4 (Fla.); Bick and Bick 1957: 2 (Okla.); Kormondy 1957: 108 (Tenn.); Bick 1957: 80 (La.); Gloyd 1958: 8 (Tex.); Bick and Bick 1958: 240 (Okla.); Bick 1959: 131 (Ark.); Donnelly 1961: 7 (Md. Va., D. C.); Alrutz 1961: 23 (Ohio); Borror 1963: 104 (common name); Macklin and Cook 1967: 120 (Ky.); Montgomery 1967: 127 (distr.); Roback & Westfall 1967: 114 (water quality data); Montgomery 1968: 134 (distr.); Pinhey 1969: 189 (tandem linkage); Resener 1970: 37 (Ky.); Huggins et al. 1976: 16 (Kans.); Harp & Rickett 1977: 50 (Ark.); Huggins 1978: 2 (Kans.); Young & Bayer 1979: 90 (lar- va, key, Tex.); Carle 1979: 322 (status in Virginia); White et al. 1980: 26 (S. C.); Carle 1982: 339 (W.Va.); Dunkle & Westfall 1982: 32 (status in Fla.); Paulson 1982: 255 (U.S., Mex.); Huggins & Brigham 1982: 4: 39 (N. C., S. C.); Harp & Rickett 1985: 132 (Ark.); Davies & Tobin 1985: 27 (cat.); Tsuda 1986: 87 (cat., attributes authorship to Selys); Bridges 1991: VII.60 (cat.); Tsuda 1991: 95 (cat.); Dunkle 1992: 39 (Fla). Gomphus designatus. — Hagen 1861: 99 (descr. d, 9). Gomphus sp. — Cabot,1872: 4 (descr. larva as Gomphus sp. No. 6, Poles Creek, Tex.). Herpetogomphus designatus. — Walsh 1862: 389 (compared with Ophiogomphus rupinsulensis), Hagen 1875a: 42 (cat.); Selys 1879: 64 (2 sep.) (characters of genus); Hagen 1885: 255 (larva of Cabot identified); Kirby 1890: 60 (cat., attributes authorship to Selys); Banks 1892: 179 (Kans.); Calvert 1899: 386 (characters of genus); Adams 1900: 622 (Ark.); Williamson 1903: 226 (Tenn.); Van der Weele 1906: 177 (mentioned); Tucker 1907: 79 (Kans.); Tucker 1908: 99 (Tex.); La Rivers 1938: 76 (Nev., habits); La Rivers 1940a: 112 (Nev.). Description Male. — Entire face pale green, more vivid on post- clypeus and frons, vertex with wash of brown around ocelli or entirely dark brown, scape dark brown, pedi- cel light brown, flagellum dark brown; anterior mar- gin of vertex forming a trough at junction of ante- frons, this trough with a pair of deeper pits anterolateral to median ocellus; occiput green, wide, tumid medially, crest slightly sinuate, barely emargi- nate medially or straight, its hind margin covered with long brown hairs; postocciput not visible dorsal- ly, green; rear of head brown, darker toward occipital foramen. Prothorax predominantly brown, becoming green 218 dorsally; anterior and posterior lobes green; synthorax (fig. 17) green with following dark brown: well de- fined middorsal stripe widening to collar, its upper end extending along antealar sinus connecting with narrow humeral stripe; an isolated antehumeral stripe not touching dorsal or ventral margin (except in spec- imens from Durango and Chihuahua — see remarks); lower part of humeral stripe extending posteriorly to form a fragmented second lateral stripe, though in some specimens a vestige of a complete stripe exists; narrow third lateral stripe. Coxae, trochanters pale green, femora green becoming brown posterolaterally at distal 0.20; distal 0.05 of metafemora green with narrow brown line extending distally from subapical brown area; tibiae dark brown with yellow along lat- eral carinae of basal 0.5 of mesotibiae, a vestige of yel- low or none on metatibiae; tarsi dark brown, arma- ture black. Wings hyaline with wash of yellow basally, vena- tion dark brown, basal wing venation light brown, es- pecially anteriorly, anterior margin of costa yellow, except for anterior margin of pterostigma; pterostig- ma dark brown, veins bordering it black. Venational statistics. Fifth (rarely fourth) anteno- dal thickened in all wings; number of marginal cells behind fore wing paranal cells: 2-5/1-4; anx: fore wing 11-14/11-14, hind wing 8-11/8-11; pnx: fore wing 7-10/8-11, hind wing 7-11/8-12; cs under pte- rostigma: fore wing 5-7/5-7, hind wing 4-7/4-7; anal triangular cells: 3-5/3-4. Hind wing 28-31 mm. Abdomen. Segment 1 green with dorsolateral brown stripe occupying basal 0.75 of segment; seg- ment 2 similar to segment 1, but dark brown expand- ing ventrolaterally behind auricle, contracting again and connecting with dark brown annulus; segment 3 similar to segment 2 but dark brown dorsolateral stripe interrupted at anterior 0.25 and again at distal 0.50 to 0.75 of segment, dark brown prominent at transverse carina, posterior 0.25 to 0.30 of segment with dorsolateral brown expanding laterally to encir- cle segment at posterior 0.10 of segment; segments 4- 6 similar to segment 3 but dorsolateral brown inter- rupted only at anterior 0.25 to 0.30 of segment and expanding abruptly at transverse carina; segment 7 with narrow dark brown on transverse carina and poorly defined brown on lateral 0.50 of segment; seg- ments 8-10 yellow brown with dorsolateral red brown occupying basal 0.40 to 0.75 of segments; pale areas of segments 2-6 orange green dorsally, becom- ing white below. Abdomen 35-37 mm. Cercus (fig. 104) yellow brown, strongly angulate near middle; tip of cercus beyond angulation narrow, drawn out into a fine point; ventral margin with infe- rior carina at basal 0.45. Epiprocts yellow brown, gently curved dorsally at 90°, tips obliquely truncate (fig. 130) or spatulate (fig. 131). Accessory genitalia (fig. 60). Anterior hamule black, divided at upper 0.50; lower branch forming a small, posteriorly directed truncate appendage; poste- rior hamule pale green, triangular, with anterior basal shoulder, tip evenly round, black; penis with long, decumbent lanceolate lateral lobe, its posterior end serrated; cornuae separated, tips narrowly rounded, with no medial lobe; membranous hood moderately long but usually with ends not overlapping. Female. — Head as in male, but median swelling of occiput more prominent; prothorax and synthorax as in male; wings with yellow at base more extensive, of- ten with markings at base of arculus, to cubitoanal crossvein; abdomen with definite pale orange dorsal- ly, white to orange white laterally, these areas separat- ed by brown dorsolateral stripes as in male; these stripes on each segment may be narrower, especially so between transverse carina and posterior 0.20 to 0.30 of segment; segments 8-9 mostly dark brown dorsally, sides tawny, annulus pale; segment 10 yel- low brown, cercus ocher. Abdomen 33-36 mm. Vulvar lamina (figs. 185-186) with prominent ba- sal plates; quadrangular laminar plates acuminate posteriorly, meeting at medioanterior margin; cleft between plates broadly U-shaped; juncture of Y- shaped postlamellar ridge at or before level of posteri- or margin of lamina. Abdomen 33-36 mm. Venational statistics: number of marginal cells be- hind fore wing paranal cells: 2- 5/1-5; anx: fore wing 12-15/11-15, hind wing 8-12/8-11; pnx: fore wing 7- 12/7-12, hind wing 8-12/7-12; cs under pterostigma: fore wing 5-7/5-7, hind wing 4-8/5-7. Hind wing 28- 32 mm. Diagnosis Little difficulty should be encountered in identi- fying either sex of this species. The strongly angulate cercus superficially allies it with Æ. sipedon, E. hetero- don, and E. lampropeltis, but the structure of the occi- put and the penile characters easily separate males of these species. The broadly defined occiput, with the large medial tumid area characteristic of E. designatus, is not present in the other three species. Male £. lam- propeltis have a broad occiput, as do male E. designa- tus, but the medial area of the occiput is only slightly tumid. The long, recumbent lateral lobe of the penis is unique to this species (fig. 82). In E. lampropeltis and E. sipedon, this structure forms a spinulose semicircle (figs. 83-85). As in males, females of E. designatus are distin- guished from all other similar species by the tumid median area of the occiput (fig. 156). The vulvar lam- ina in £. designatus (figs. 185- 186) more closely ap- proaches that of E. bothrops (fig. 183) than of E. sipe- don (figs. 187-188), E. heterodon (fig. 193), or E. GARRISON: Revision of Erpetogomphus lampropeltis (figs. 189-190). Each lamellar plate of £. designatus is smaller and the notch in the medial mar- gin is more concave than in the other species. Remarks Variation. — Calvert (1907) described a specimen from San Pedro, Coahuila, Mexico, which differed fr- om typical £. designatus in the following characters (typical condition in parentheses): 1) dark antehu- meral stripe reaching down to mesinfraepisternum (lower part of stripe isolated), and 2) darker, more pronounced markings on abdominal segments as fol- lows: segment 7 with a black spot on each side of pos- terior 0.50 of segment (like area red brown, blending anteriorly with ocher pale area); segment 8 with a black stripe on each side for almost its entire length (this area red brown); segment 9-10 with black at ba- sal dorsolateral 0.50 to 0.75 (this area red brown with some black on dorsal 0.50 of segment). Calvert (1907) considered the San Pedro male to represent £. designatus. This specimen is indicated as in the ANSP, but it could not be found (Azuma in Üitt., 1984). However, I have seen three males (Durango and Chihuahua) and a female (Durango) whose markings agree with Calvert’s San Pedro male. I also find the following other differences: 1) no wash of yellow pre- sent on wing bases, 2) dark thoracic stripes thicker and more pronounced (fig. 18), 3) a black spot anter- oventrally to auricle of abdominal segment 2, and 4) tip of epiproct, viewed posteriorly, is spatulate (fig. 131), not obliquely truncate (fig. 130) as is rypical of other specimens. The single female has a more arcu- ate occiput (fig. 157) than typical £. designatus (fig. 156), but the dark anteroventral spot on segment 2 is small and inconspicuous. Specimens from Durango, Chihuahua, and Coahuila could be a well-defined subspecies charac- terized by overall darker colouration and slight mor- phological modification of the female occiput (if typ- ical for this form) and male epiproct. I detect no morphological differences in the penes, hamules, or cerci and, like Calvert, I am inclined to treat these specimens as £. designatus. Further evidence to sup- port their conspecificity comes from a male from Nuevo Leon (in foothills of the northernmost range of mountains) in which the antehumeral stripe ex- tends to the mesinfraepisternum but is in all other characters typical of £. designatus. One male from McLennan Co., Texas, has the tips of the posterior hamules with a small, anteriorly re- curved hook. This male and another from the same locality have vestiges of an anteroventral dark spot on abdominal segment 2. I have not seen sufficient material of this species throughout eastern parts of its range to render any conclusions about geographic variability. Numerous 219 [lIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 specimens, primarily from Gonzales and Williamson counties, Texas, show little variability. Although the type locality for £. designatus and many other species collected by Capt. John Pope of the Smithsonian Institution and described by Hagen, is said to be the Pecos River in western Texas (32°N, 104°W), Needham and Cockerell (1903), citing Scudder and Cockerell (1902), argue that the type lo- cality was probably at or near Roswell, Chaves Co., New Mexico, as follows (Needham and Cockerell 1903: 138-139): ‘In Proc. Davenport Acad. Sci. vol. 9 (1902), p. 51, it is inferred that Pope collected his material at very different localities above the river, since he obtained both Melanoplus bivittatus and M. differentialis, which inhabit different life-zones. However, in 1902 it was observed that these two grasshoppers do actually occur together at Roswell. As it is not very likely that their ranges overlap very much, it thus seems likely that Pope’s collections, if all from one place, were from the vicinity of the present town of Roswell, rather than from Lat. 32° (the present boundary between New Mexico and Texas), as Hagen indicates. In all probability, howev- er, the collections were made in several places.’ No year is indicated on the data of any of the type material, nor have I found any reference to the year of Capt. Pope’s expedition. However, a public display on the recent human history of the caverns at the vis- itor center at Carlsbad Caverns National Park, Eddy Co., New Mexico, states that Pope’s expedition oc- curred in 1854-1855. Venational details of lectotype male: number of marginal cells behind fore wing paranal cells: 1/2; ; anx: fore wing 12/13, hind wing 10/10; pnx: fore wing 8/9, hind wing 9/9; cs under pterostigma: fore wing 6/6, hind wing 7/6; anal triangular cells: 3/4. Biology. — The synonymy indicates that more is known about this species than any other congener. Erpetogomphus designatus seems to be a species of mes- ic deciduous forests where there are clear streams and rivers. Montgomery (1925) writes of its habits in Indiana: “This species was common in a well-cultivat- ed erchard near Vincennes on August 7, 1924, rest- ing on the ground with wings spread out against the soil but taking flight so readily that only three speci- mens were taken in two hours although I saw 20 or more. In Hamilton Co., Indiana, Montgomery 1937) records it as … rather numerous at ripples in the river, flying just above the surface of the water, and alighting on rocks. However, it was exceedingly difficult to capture... Williamson (1932) observed his species flying back and forth near the surface over ripples in Missouri. Others alighted on vegetation or he flight of D,.L ihranacat i aiıtanagal Valley, Nevada, as ‘low, swift, and change of position rapid... Ahrens (1938) describes the cap- ture of a male along with a male Hagenius brevistylus Selys near Lebanon, Missouri: ‘The H. brevistylus had struck and seized the designatus with such force that both were carried into the water. Both were dipped from the river an instant after they struck the surface.” Dunkle and Westfall (1982) assign £. designatus as ‘threatened’ in Florida, because it is thus far known only from Liberty and Gadsden counties in the pan- handle. In 1975 and 1977, I found this species abun- dant along cultivated stubble bordering trees next to the Guadalupe River in Gonzales County, Texas. It was more abundant than £. eutainia; and both species were easily flushed from low or cut vegetation. Members of both sexes did not fly far, and I could easily photograph and capture specimens. Collection dates range from 6 May in Texas (Williamson, 1914b) to 5 October in Tennessee (Wright, 1938). The larva has been described by Cabot (1872), Needham and Heywood (1929) and Needham and Westfall (1955). Distribution (fig. 202). — Erpetogomphus designatus and £. compositus are the most boreal of the genus, with £. designatus generally replacing the western £. compositus in the eastern United States. Erpetogomphus designatus occurs as far north as Maryland and Washington, D. C., west through West Virginia, Ohio, Indiana, Missouri, Kansas, southeastern Colorado, eastern New Mexico, into western Texas. Sidney Dunkle (772 Zitt., 1991) men- tions one female from South Dakota (Fall River Co., 9.7 mi SW Hot Springs, 29 July 1976, G. and ]. Bick, L. Hornuff), erroneously published as Stylurus intricatus (Selys) (Bick et al., 1977: 151), which is the northernmost record for this species. It occurs gener- ally throughout the southeastern United States, but does not penetrate peninsular Florida (Dunkle and Westfall, 1982). Farthest western records include Hiko Springs, Lincoln Co., Nevada (La Rivers, 1938) and Arizona (Maricopa Co., Salt River at Corn Bluff Campground, 15 mi NE of Mesa). Erpetogomphus de- signatus is found as far south as Nuevo Leon (24- 26°N, 100°W) and west to Coahuila (Calvert, 1907), Durango, and Chihuahua, Mexico. It is partially sympatric with Æ. compositus in Arizona, Nevada, New Mexico, and western Texas. I collected it with E. eutainia in Gonzales Co., Texas. Material Type data. — Lectotype male by present designa- tion with following data: ‘July/ 16° [handwritten], ‘designatus [handwritten], ‘Hagen’ [printed], small white label with printed ‘Type’ with red label glued to lower edge with handwritten ‘1864’; red rectangular label with printed ‘LECTOTYPE/ and written ‘Erpetogomphus designatusl Hagen d/ des. [printed] R. W. Garrison 1984’. Another d, a paralectotype, has the handwritten labels, ‘Erp. designatus! d / Pecos River’ [in Hagen’s hand] and ‘Pecos River/ Uhler 1860. in an unknown hand. A female paralectotype, collected ‘July/ 15’ has the additional label: [all prin- ted unless otherwise stated] ‘ERPETOGOMPHUS/ DE- SIGNATUS HAGEN [stamped] / P. P. Calvert, det [writ- ten] 1905/ B. C. A. Neur., p. [written] 166.’ Type locality herein restricted to vicinity of Roswell, Chaves Co., New Mexico (see remarks). Other material (848, 539, including lectotype d and d ? paralectotypes). — U. S. A: FLORIDA: Gadsden Co., Apalachicola River, Aspalaga Landing, Hwy I-10 (jp, Rwa); GEORGIA: Houston Co. Ocmulgee River at 96, (RWG); VIRGINIA: Louisa Co., South Anna River, Rte 657, (csuc); North Anna River, Rte 601, (csuc); Kansas: Chatauqua Co., Big Caney River, 1 1/2 mi W of Elgin, (csuc); TEXAS: Caldwell Co, Plum Creek at US Hwy 90 (rwe); Goliad Co., 3 mi N of Goliad, (RwG); Gonzales Co., Guadalupe River, 4 mi S of Gonzales (RWG); Palmetto State Park, 6 mi S of Luling, nr. US Hwy 183, (rwg); Hidalgo Co, Bentsen, Rio Grande State Park, (csuc); McLennan Co., Hog Creek, (RWG); Real Co, Nueces River at Tex Hwy 55, S of Barksdale, (RwG); Reeves Co., roadside irrigation ditch W of Balmorhea (rw); Williamson Co., 3 mi SW of Taylor, (RWG); Mustang Creek by Carlos G. Parker Blvd (= Loop 427), Taylor, (RwG); Cororapo: Las Animas Co., Purgatory River, Bent Cyn nr. Delphi (csuc, RWG); New Mexico: Guadalupe Co., Santa Rosa, 4600 ft. (ummz); ARIZONA: Maricopa Co., Salt River at Corn Bluff Campground, 15 mi NE Mesa, (- rsca); Mexico: Nuevo Leon: in foothills of the north- ernmost range of mountains, (UMMZ); Apodaca, just NE of Monterrey, near course on Instituto farm, (- UMMZ, RWG); Chihuahua: La Cruz, (Rwa); Naica (ummz); Durango: 3.7 mi NW of Gomez Palacio on Hwy 49, 3700 ft., (ummz); 8.2 mi N of Gomez Palacio on Hwy 49, (ummz); Rio Nazas at Mex. Hwy 49, 3.4 mi S Ciudad Leon Guzman, 4200 ft. (DRP). Erpetogomphus sipedon Calvert (figs. 19-thx, 39, 40-abd, 61-hamules, 83-penis, 105, 106-app, 158, 159-vertex, 168-leg, 187, 188- vl, 203-distr) Erpetogomphus sipedon Calvert, 1905: 165 (descr. 9 Jalisco: Guadalajara). — Calvert 1907: 399 (2? ‘Las Bocas in Durango...’); Calvert 1908c: xxx (Erpetogomphus sipedon, type 9, Guadalajara, found by Schumann); Calvert 1909: 481 (Cuernavaca, Mexico); Williamson and Williamson 1930: 13 (summary of status); Kimmins 1969: 297 (type in BMNH); Paulson 1982: 256 (Mex.); Davies and Tobin 1985: 28 (cat.); Tsuda 1986: 87 (cat.); Bridges 1991: VII.195 (cat.); Tsuda 1991: 95 (cat.). GARRISON: Revision of Erpetogomphus Description Male. — Entire face light green; this colour more vi- vid on frons, post- and anteclypeus; base of mandible, labrum light green, vertex with brown around and on ocellar prominences, remainder light green; antennal pedicel dark brown dorsally, light green ventrally; se- cond segment and flagellum dark brown; occiput light green, crest slightly emarginate medially, its hind margin covered with long brown hairs; rear of head brown, darker toward occipital foramen, beco- ming light green laterally. Prothorax predominantly dark brown, becoming light green dorsally; anterior and posterior lobes light green. Synthorax entirely light green (probably a viv- id apple green in life) with following dark brown: an- tehumeral stripe ending dorsally before antealar sinus, narrow stripe along humeral suture, this stripe widen- ing at dorsal 0.30 of suture and with anterior margin just touching upper part of antehumeral stripe, ante- rior 0.50 of mesinfraepisternum. Coxae, trochanters ivory, femora light green with black along external surfaces, more so on mesofemora; most of lateral sur- face of mesofemora dark brown, these markings more restricted and limited to apical 0.30 of metafemora; tibiae black with small lateral ivory stripes along basal 0.30; tarsi and armature black. Wings hyaline, venation dark brown, basal wing venation brown, anterior margin of costa yellow; pte- rostigma brown, black along margins. Venational statistics (n = 13). Fifth antenodal thickened in all wings; number of marginal cells be- hind fore wing paranal cells: 1; anx: fore wing 12- 13/12-15, hind wing 9-10/9-10; pnx: fore wing 7- 10/7-9, hind wing 9-11/8-10; cs under pterostigma: fore wing 5-6/5-6, hind wing 5-6/6; anal triangular cells: 4. Hind wing 28-31 mm. Abdomen (fig. 39). Segment 1 light green with dorsolateral brown stripe; segment 2 similar to seg- ment 1 but dark brown expanding posteriorly behind auricle, contracting again and interrupted just before brown annulus; segment 3 similar to segment 2 but with dark brown at transverse carina, anterior end of dorsolateral not on or just touching segment 2, its posterior end broadly connecting with dark brown annulus; segments 4-6 similar to segment 3 with pos- terior 0.40 wider than anterior 0.60; segment 7 large- ly pale except for brown along transverse carina, be- coming red brown posteriorly, especially so laterally; segment 8 largely red brown with ill-defined light yel- low brown middorsal and ventrolateral stripes; seg- ment 9 similar to segment 8; segment 10 red brown basally, becoming pale yellow brown distally. Abdomen 33-37 mm. Cercus (figs. 105-106) ivory, strongly angulate near middle; tip of cercus beyond angulation narrow, drawn out to a fine point; ventral margin with inferi- 221 [1IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 or carina at basal 0.30. Epiprocts yellow brown, typi- cal of genus. Accessory genitalia (fig. 61). Anterior hamule black, divided at upper 0.30, lower branch forming a small, truncate appendage; posterior hamule white, triangular, with anterior basal shoulder, tip with small, anteriorly recurved tooth; penis with small ser- rated lateral lobes, membranous hood short, exposing a subcutaneous membrane which forms posterior margin of segment (fig. 83); cornuae (fig. 83) moder- ately long, widely separated, medial lobe present, usu- ally 0.5 to 1.0 length of outer lobe. Holotype female. — Face entirely pale except for small black area at base of mandible, occiput (fig. 159) dark brown around ocelli, remainder pale. Prothorax mostly pale, darker along pleura. Synthorax (fig. 19) mostly pale except for brown mar- gin around antealar sinus and incomplete antehumer- al stripe, its upper end joined with narrow humeral stripe. Coxae, trochanters, and femora pale with api- cal 0.30 black, especially mesofemora, metathoracic legs with black less extensive (fig. 168), forming black streaks externally; tibiae entirely black except for pale line laterally, tarsi and armature black. Abdomen (fig. 40) similar to male but with dark brown dorsolateral stripe of uniform width on seg- ment 2, segments 8-9 mostly black dorsally, sides tawny, annulus pale, cerci lacking. Abdomen 35 mm. Venational details: number of marginal cells be- hind fore wing paranal cells: 2/2; anx: fore wing 13/13, hind wing 9/11; pnx: fore wing 9/9, hind wing 10/12; cs under pterostigma: fore wing 6/7, hind wing 7/6. Hind wing 31 mm. Vulvar lamina (fig. 187) with broadly rectangular flaps meeting at anterior margin; juncture of Y- shaped postlamellar ridge at level of posterior margin of lamina. Diagnosis Males of Erpetogomphus sipedon are most similar to E. heterodon but are easily distinguished by the struc- ture of the penis and cercus. The cornua in E. sipedon has a medial lobe (fig. 83), which is lacking in E. he- terodon (fig. 87), and the tip of the cercus in E. sipe- don are slender and finely attenuate (figs. 105-106), not robust and short as in E. heterodon (figs. 110- 111). Erpetogomphus heterodon and E. sipedon are the only two species of the E. designatus group which sha- re a similar thoracic design. However, E. sipedon lacks any dark thoracic stripes on the second and third la- teral sutures; a vestigial incomplete second lateral stri- pe and complete narrow third lateral stripe exist on £. heterodon. The thoracic pattern of E. boa is also simi- ar to E. sipedon, but E. boa is well differentiated by genital characters given in the keys. Although the dif- ference in somatic patterns between E. lampropeltis | 999 and £. sipedon is great, these two species are separable by the structure of the penis. In £. sipedon, the hood is abbreviated, exposing a subcutaneous membrane, which latter forms the posterior margin of the seg- ment. In £. lampropeltis, the hood is of normal length and it forms the posterior margin of the segment. Although females of £. sipedon are easily differen- tiated from all other similar species by differences in colour and pattern, their separation from the same species using only structural characters is more diffi- cult. Species morphologically similar to £. sipedon in- clude E. crotalinus, E. compositus, E. viperinus, E. bothrops, and E. heterodon. The occiput of E. sipedon is straight or, at most, weakly notched, (notch in E. crotalinus not clearly seen in the paralectotype female, [fig. 161]). The lon- gitudinal distance of the occiput (hereafter called width) is narrower in E. sipedon than in E. crotalinus, and the posterolateral ocellar tubercle is less pro- nounced in £. sipedon than in E. crotalinus. The occipital crest of E. compositus is sinuous (mostly linear in E. sipedon) and the postocciput is easily visible in dorsal view (fig. 163); the same struc- ture is barely visible in Æ. sipedon (fig. 159). The structure of the postlamellar ridge provides the best means of separation between £. sipedon and sister taxa E. bothropsand E. viperinus. In E. sipedon, this Y- shaped structure is short: the juncture of the lateral arms meets at or near the hind margin of each vulvar plate. The longitudinal stem also does not possess any circular or oval depressions on either side (figs. 187- 188). In E. bothrops and E. viperinus, the juncture of the Y-stem is well posterior to the hind margin of each vulvar plate, and a well-defined oval or circular depression exists on either side of the longitudinal ridge (figs. 183-184). I have been unable to find any morphological dif- ferences between females of E. sipedon and E. hetero- don. Body pattern (figs. 19, 23) and distribution (figs. 201-204) will serve to separate females of these close- ly related species. Remarks Variation. — The male description is based on 11 males from Jalisco, Morelos, and Durango states, Mexico. The four from Durango are slightly larger (hind wing 30-31 mm); and their thoracic pattern is similar to that of the holotype female (fig. 19). The other seven males are smaller (hind wing 28-29 mm), and the lower 0.60 of the dark humeral stripe is nar- rower. The Durango males, though larger, have a sparser venation. For example, three have no margin- al cells behind fore wing paranal cells, one has one marginal cell in both wings; and all other specimens have one marginal cell, except for the left fore wing of one Acatlipa male, which has no marginal cells. Two of the Durango males have a 3-celled anal triangle in both hind wings, while all others possess 4 cells. The cerci of the Durango males are less angulate and the tips more robust (fig. 106) than the other males, but other characters of the body and penis leave no doubt that they represent £. sipedon. Variability for females is less pronounced than for males, and I can find no overall differences in vena- tional patterns among the 12 females examined. Female venational statistics (n = 12, including hol- otype): number of marginal cells behind fore wing paranal cells: 1-3/0-3; anx: fore wing 12-14/12-15, hind wing 9-12/9-11; pnx: fore wing 8-10/8-10, hind wing 9-11/8-12; cs under pterostigma: fore wing 5- 715-7, hind wing 5-7/4-6. Hind wing 30-34 mm. Abdomen 36-37 mm. When Calvert (1905) described this species, he had only 6 females available; and he noted differences in the condition of the occiput between the holotype and a paratype from Matamoros, Puebla (figs. 158- 159). Both forms were represented in his key. I can- not consider the differences noted by Calvert as use- ful, as much variation seems to exist in the shape of the occipital ridge in females of this and other species. At the time of the original description, Calvert (1905) postulated that the female of E. sipedon could be £. boa, which was then known only from one in- complete teneral male. The discovery of more speci- mens of both species, and especially the capture of both males and females of E. sipedon in Durango, leaves no doubt that both are valid species. Biology. — González (in litt.) collected this species at Palo Bolero, where adults perched on herbaceous vegetation on the banks of the Rio Sabinos. Its known distribution suggests that it is more widely distributed, but specimens are rare in collections. Specific ecological requirements may restrict its oc- currence to certain sites, but a more likely answer to its apparent rarity is a lack of thorough collecting within its range. Collection dates range from 21 May (Cuernavaca) to 13 August (Nombre de Dios). Distribution (fig. 203). — Erpetogomphus sipedon is known from northern Durango (Las Bocas [Calvert, 1907] settlement on the Rio Florida in extreme northern Durango near Villa Ocampo, approx. 26°25°N 105°25’W, [Selander and Vaurie 1962]) south through Morelos and Puebla (approx. 18°50°N, 99°W) and seems restricted to the highland area of central and western Mexico from altitudes of 1000 m (Palo Bolero) to about 2073 m (Las Bocas). González (in litt.) collected this species in Mexico state (Tonatico) during July, 1992. Material Type data. — Holotype female: white printed label: [Mexico] Guadalajara,/Jalisco,/July. Schumann:’, GARRISON: Revision of Erpetogomphus white, partially printed label: “Erpetogomphus ® / sipe- don Calv. Tyre /P. P. Calvert, det. 1905/B. C. A. Neur., p. xxx, 166/orig. of Pl. VIII, ff. 34, 40° with ‘(form/a) handwritten on left side; printed label ‘Brit. Mus./1911-339.’; small round label with red margin: ‘Holo-/type’. Other material (11d, 129, including holotype 2). — Mexico: Durango: Nombre de Dios, just SE of Durango on Mex. Hwy 45 (23°51’N, 104°14’W), 5900 ft. (1800 m), 13 Aug. 1947 (W. Gertsch, M. Cazier), 48, 72 (AMNH, RWG); Morelos: Cuernava- ca, 21 May 1898 (no collector), 19 (curc); Cuerna- vaca, 7 July 1900 (C.C. Deam), 22 (ummz); Palo Bolero, km. 18.3, route 95, 5 km. S. Acatlipa, 1000 m (approx. 18°45’N, 99°15’W), 26 June 1985 (E. Gonzalez), 25 (UNAM); 18 June 1986 (E. Gonzalez), 26 (UNAM), 12 July 1986 (E. González, V. Garcia), 36 (rwe); Morelos: Pueblo Cocyotla, Hwy 421, about 30 km SW of Cuernavaca, 23 July 1992 (J. Daigle), 16 (jp); Puebla: Matamoros, no date, (Otis W. Barrett), 12 (ANsp). Erpetogomphus lampropeltis Kennedy Description Male. — Face entirely pale with dark markings as follows: along basal margin of labium, along fronto- clypeal suture (less developed in £. / natrix), base of antefrons; vertex all dark brown; antennae dark brown; anterior margin of vertex with an arcuate pit, one each anterolaterad to median ocellus; area be- tween these pits tumid (fig. 138), occiput white green, wide, slightly tumid medially, crest entirely black, slightly raised, linear or slightly convex posteri- orly, some with a small median notch, its hind mar- gin covered with pale brown hairs; postocciput slight- ly concave, pale; rear of head dark brown. Prothorax largely brown with pale green or grey green areas on anterior and posterior lobes, and with two small median spots on median lobe; contrasting dark/pale areas more prominent in nominotypic sub- species; pale colours of synthorax grey or grey green (E. I. lampropeltis) to green (E.l. natrix) with follow- ing dark brown areas: middorsal stripe, widening ba- sally to collar, its upper end covering antealar sinus, extending posteriorly and connecting with broad, well defined antehumeral and humeral stripes; well defined sinuate second lateral stripe; third lateral stripe. Coxae, trochanters pale grey green with slight wash of brown posteriorly; tibiae pale grey green with defined superior surfaces brown, becoming dark brown distally; tibiae black with narrow strip of pale grey green along basal 0.30 to 0.50 of lateral margin; base and armature black. Wings hyaline, basal wing venation brown; anteri- or margin of costa yellow up to pterostigma; pteros- 223 T1]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 tigma black; fifth antenodal thickened in all wings (wich rare exceptions). Abdomen. Abdominal segment 1 pale grey green to green with basal 0.25 brown and with dorsolateral brown stripe; segment 2 with dorsolateral stripe send- ing a ventral stem posterior to auricle and (often) an isolated spot on anteroventral margin of segment, dorsolateral stripe often connecting with black annu- lus or ending just before; segment 3 with isolated black dorsolateral spot surrounding lateral carina; its anterior end tapering and sometimes touching anteri- or margin of segment, distal 0.25 of segment with dorsolateral black increasing in width posteriorly and touching dorsally near black annulus; segment 4 sim- ilar to segment 3, but two black spots broadly con- nected forming a continuous dorsolateral stripe occu- pying distal 0.80 of segment, this stripe often constricted in middle; segments 5 and 6 similar to segment 4, but with middorsal pale areas becoming darker; segment 7 with anterior 0.50 pale, except for black transverse carina, becoming entirely tawny to black posteriorly; segments 8-10 yellow brown, be- coming black dorsally in most specimens; foliar ex- tensions dark brown to black. Cercus (figs. 107-108) ocher, in lateral view slight- ly concave dorsally near base, posterior 0.25 moder- ately angulate, the short tips slightly recumbent; cer- cus enlarged ventrally at basal 0.40 to 0.50 with inferior carina along same area, remainder slightly concave with tip forming a blunt point. Epiprocts yellow brown, gently curved dorsally at 90°, tips (fig. 132) dorsoventrally flattened and divergently bluntly pointed. Accessory genitalia (figs. 62-63). Anterior hamule dark brown, divided at basal 0.50, lower branch forming a small, posteriorly directed truncate ap- pendage; posterior hamule grey white, triangular, with anterior basal shoulder, tip bluntly rounded and armed with a small anteriorly directed tooth on pos- terior border; penis with lateral lobe small, roughly semicircular, serrate; cornuae (figs. 84-85) moderate- ly widely separated, tips broadly rounded with well developed median lobes. Female. — Head as in male, but with crest of occi- put slightly more erect and convexly arcuate; posteri- or margin of postocciput slightly concave; prothorax and thorax as in male, but with pale areas generally more extensive; wings hyaline or with saffron infu- sion at base in some £./. natrix, abdomen in well pre- served specimens with pale colours primarily white vith dull orange dorsally; black dorsolateral stripes and spots more reduced than in male, these stripes sometimes separated or greatly constricted on seg- ments 4 and 5; cercus ocher. \ Yulvar lamina (figs. 189-190) with prominent ba- sal plates, quadrangular laminar plates acuminate Dì N HN posteriorly, meeting at anteromedial margin; cleft between plates forming a 90° angle; V juncture of Y- shaped postlamellar ridge at level of posterior margin of lamina. Diagnosis This species is most similar to its congeneric relati- ves, E. sipedon and E. compositus, and is diagnosed un- der those species. Remarks The original description of E. lampropeltis is brief and is based on an unspecified number of males and females from Sespe Creek, Fillmore, Ventura County, California. Williamson and Williamson (1930) originally described E. natrix in detail from 21 males and two females from Baja California, Mexico. An abstract of their diagnosis of males E. natrix from E. lampropeltis is as follows: ‘The head and thorax of lampropeltis are duller and paler than in natrix.... the rear of the head is brown in lampropel- tis and black in natrix, the thoracic brown is paler in lampro- peltis and the pale areas are grayer, giving much less contrast in the thoracic pattern; the two lateral dark stripes are always joined, often very broadly in lampropeltis, and are not joined in.....natrix,....the femora are brown in /ampropeltis.....black in natrix, and the inferior dilared edge of abdominal seg- ments 8 and 9 is black in lampropeltis and, in natrix, is col- ored light vivid brown like the sides of the segment adjacent to the edge..... The apparent differences in morphology described above fall well within the latitude of individual and geographic variation and do not, in my opinion, con- stitute specific differences. Further, several specimens of E. natrix from various parts of its range have mac- ulation patterns more closely approaching those of £. lampropeltis. For example, the antehumeral and hu- meral stripes are briefly connected, the dorsum of ab- dominal segments 7-10 have black as in £. lampropel- tis, and the inferior dilated margin of abdominal segments 8 and 9 is black. I can also detect no differ- ences in the shape of the vulvar lamina between fe- males of E. lampropeltis and E. natrix. The forms are allopatric, separated by the Colorado Desert. The large series of E. lampropeltis (mostly topotypes) that I examined does show consis- tent differences in pale thoracic colouration and mi- nor differences in thoracic maculation. I consider the forms to represent subspecies separable as follows: 1. Pale thoracic colouration grey green, antehumer- al and humeral stripes broadly connected on dor- sal 0.25, after isolating a small pale spot below an- tealar sinus; second lateral stripe about twice as wide on dorsal 0.5 and often connecting with narrow third lateral stripe; dark thoracic stripes, especially second and third, often not well de- fined (fig. 20); Ventura, San Diego, Los Angeles, and western San Bernardino counties, California E. lampropeltis lampropeltis — Pale thoracic colouration vivid green, antehumer- al and humeral stripes not connected, or only narrowly so; second lateral stripe narrower than in E. L lampropeltis, not connecting with third lateral stripe (fig. 21); central Baja California, Arizona, New Mexico, western Texas, south throught Durangoy Mexico... tn Erpetogomphus lampropeltis lampropeltis Kennedy (figs. 20-thx, 62-hamules, 84-penis, 107-app, 132- ept, 138-base of postfrons, 160-vertex, 189-vl, 204- distr) Erpetogomphus lampropeltis Kennedy, 1918: 297 (descr. d Sespe Creek, Fillmore, Ventura Co., California). — Needham and Heywood 1929: 80 (descr. d, 9); Williamson and Williamson 1930: 13 (summary of stat- us); Needham and Westfall 1955: 147 (descr.); Pritchard and Smith 1956: 116 (key); Musser 1962: 14 (notes on larva); Borror 1963: 104 (common name); Montgomery 1968: 133 (distr.); Paulson and Garrison 1977: 157 (Calif.); Paulson 1982: 255 (U.S.); Davies and Tobin 1985: 27 (cat.); Tsuda 1986: 87 (cat.); Bridges 1991: VII.115 (cat.). Herpetogomphus lampropeltis. — Byers 1928: 5 (larva un- known). Erpetogomphus lampropeltis lampropeltis. — Tsuda 1991: 95 (cat.). Description Male. — Colouration and maculation as given un- der species account, key and in fig. 20. Venational statistics: number of marginal cells be- hind fore wing paranal cells: 0-3/1-3. — anx: fore wing 9-14/10-14, hind wing 8-10/8-10; pnx: fore wing 7- 9/5-10, hind wing 7-12/7-12; cs under pterostigma: fore wing 4-5/4-5, hind wing 3-5/3-6; number of anal triangular cells: 3-4/3-5. Hind wing 28-31 mm. Abdomen 34-37 mm. Female. — Venational statistics (n = 6): number of marginal cells behind fore wing paranal cells: 0-4/1-3; anx: fore wing 13-15/12-16, hind wing 9-11/9-11; pnx: fore wing 9-11/9-10, hind wing 9-12/9-12; cs under pterostigma: fore wing 5-6/5-6, hind wing 5- 6/5-6. Hind wing 32-36 mm. Abdomen 33-36 mm. Diagnosis See account under £. compositus and key to subspe- cies of E. lampropeltis. Remarks Variation. — Venational details of holotype male: number of marginal cells behind fore wing paranal GARRISON: Revision of Erpetogomphus cells: 1/2; anx: fore wing 12/12, hind wing 9/9; pnx: fore wing 8/9, hind wing 9/9; cs under pterostigma: fore wing 5/5, hind wing 5/4; number of anal trian- gular cells: 4/5. Hind wing 28 mm. The large series of specimens primarily from the type locality shows little somatic colour and/or pat- tern variation, due, possibly, to its limited distribu- tion. The fifth antenodal in the fore wings is general- ly thickened, but 7 of 40 wings of 20 males examined have the fourth antenodal thickened; and one wing has the sixth antenodal thickened. A 3-celled anal tri- angle is common in this subspecies, as almost half (19/40) of the wings possess this condition. The female from Matillija Hot Springs has flaves- cent wing bases. Biology. — At the type locality, this species was abundant during early September, 1982. I collected 54 males within about two hours. All preferred to sit on exposed rocks and were easily collected with a net. Hundreds of cast skins were attached to the exposed rocks. I did not find the species in smaller rivulets emptying into Sespe Creek. The only female I saw and collected was found dead in the water. I returned to the locality in September, 1984, but the entire area had been fenced off. Five males were collected in the general vicinity, below Matillija Hot Springs, but they were rare and more wary. This is a late summer species: the specimens from San Bernardino Co. collected in June are all teneral. Flight dates for specimens from Ventura Co. range from 10 July to 8 September. Distribution (fig. 204). — The nominotypic sub- species has been collected only at three localities in Ventura County one in Sar Bernardino County, one in Los Angeles County, and one in San Diego County, California. It has the most restricted distri- bution of any Odonata form in southern California. Material Type data. — Holotype male: caLIF: Ventura Co., Sespe Creek, Fillmore) 7 Aug: 11915 (Gn: Kennedy); in usnm. I have illustrated the thorax (fig. 20), caudal appendages (fig. 107), and penis (fig. 84) of the holotype. Other material (706, 69, including holotype ¢). — U.S. A.: CALIFORNIA: Los Angeles Co., San Gabriel Mtns., jct of N and W Fork of San Gabriel River, E of Hwy 39, 11 Aug. 1986 (D. Swinney), 16 (RWG), Ventura Co., Sespe Creek, 4.0 mi N of Fillmore, 600 ft., 8 Sept. 1971 (D. Paulson), 2d (cc, RwG); 5 Sept. 1982 (R. W. and J. A. Garrison), 546, 12 (Rwe); Matillija Hot Springs, 14 Aug. 1927 (T. Craig), 1d, 12 (rsca); just below Matillija Hot Springs, 1400 ft., 2 Sept. 1984 (R. W. and J. A. Garrison), 5d (RWG); Wheeler Hot Springs, 10 July 1927 (T. Craig), 46, (rsca); San Bernardino Co, Deep Creek Public ho ho UN T1IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Camp, 15 June 1957 (A. Menke, L. Strange), INS) A x (LACM); San Diego Co., La Jolla, La Jolla Indian Res. Campground, San Luis Rey River, 25 Aug. 1991 (W. F. Mauffray), 1d (RWG). Erpetogomphus lampropeltis natrix Williamson and Williamson stat. n. (figs. 21-thx, 63-hamules, 85-penis, 108-app, 1990- vl, 204-distr) Erpetogomphus natrix Williamson & Williamson,1930: 19 (- descr. 6, 2 Baja Calif. Sur, Purissima, 12 October 1923). — Byers 1939: 50 (quotes Williamson & Williamson, 1930); Fraser 1940: Pl. 6 (penis); Needham & Westfall 1955: 148 (descr. &); Borror 1963: 104 (common name); Montgomery 1968: 133 (distr.); Paulson 1982: 255 (U.S., Mex.); Davies and Tobin 1985: 28 (car.); Tsuda 1986: 87 (cat.). Erpetogomphus lampropeltis. — Tinkham 1934: 215 (Tex.), Needham & Westfall 1955: 147 (Tex.); Gloyd 1958: 8 (Tex.); Kormondy 1960: 122 (Ariz.); Gonzälez & Novelo 1991: 97 (Mex.); Novelo & Gonzälez 1991: 154 (descr. larva, Mex.). Erpetogomphus lampropeltis natrix. — Bridges 1991: VII.144 (cat.); Tsuda 1991: 95 (cat.). [Demotion of E. natrix to subspecies status was advised by me in litt. to Bridges and Tsuda.] Description Male. — Colouration and maculation as given un- der species account, key, and fig. 21. Venational statistics: number of marginal cells be- hind fore wing paranal cells: 0-3/0-2; anx: fore wing 11-15/11-14, hind wing 9-11/9-11; pnx: fore wing 6- 11/7-10, hind wing 8-14/7-12; cs under pterostigma: fore wing 4-6/4-6, hind wing 4-6/4-6; number of anal triangular cells: 3-4/2-4. Hind wing 24-33 mm. Abdomen 30-37 mm. Female. — Venational statistics (n = 17, including allotype): number of marginal cells behind fore wing paranal cells: 0-6/0-5; anx: fore wing 12-15/12-15, hind wing 9-11/9-11; pnx: fore wing 7-11/7-12, hind wing 9-12/9-12; cs under pterostigma: fore wing 4- 6/4-7, hind wing 4-7/5-7. Hind wing 30-35 mm. Abdomen 36-41 mm. Diagnosis See comments under E. compositus and key to sub- species of E. lampropeltis. Remarks Venational details of holotype male: number of marginal cells behind fore wing paranal cells: 2/1; anx: fore wing 13/13, hind wing 9/9; pnx: fore wing 9/9, hind wing 10/9; cs under pterostigma: fore wing 4/4, hind wing 5/4; number of anal triangular cells: 4/3. Hind wing 30 mm. Venational details of allotype female: number of marginal cells behind fore wing paranal cells: 2/1; anx: fore wing 13/13, hind wing 9/10; pnx: fore wing 7/8, hind wing 10/9; cs under pterostigma: fore wing 6/5, hind wing 5/5. Hind wing 30 mm. Variation. — Intrapopulational variation of body maculation appears to be as great as interpopulation- al variation. The antehumeral and humeral stripes may be connected or separate in any small series from a given locality: 1 of 3 specimens from Cane Springs, Mojave Co., 3 of 12 specimens from Cave Creek, Maricopa Co. (4 have these stripes barely connected), and 2 of 11 specimens from 15 mi E of Douglas, Cochise Co., Arızona, all have connected antehumer- aland humeral stripes. Similar degrees of variation ex- ist for the dorsal and dorsolateral darkening of ab- dominal segments 7-10. Some specimens have little black on these segments (as noted by Williamson and Williamson, 1930, for their type series of E. natrix), but others have various degrees of darkening, and some specimens are as dark as I have observed for no- minotypic E. lampropeltis. A male I collected at Las Parras, Baja California, has the posterior 0.5 of the dorsum of abdominal seg- ments 7 and all of segments 8-9 with black. The male from ‘Corralitos’ is unusual in having narrow second and third lateral thoracic stripes. The upper 0.5 of the second lateral stripe is also lacking, but in all other as- pects, this specimen is E. lampropeltis natrix by mor- phology and overall body colouration and macula- tion. Biology. — Williamson and Williamson (1930) de- scribed the type locality. González (in litt.) collected this subspecies with £. crotalinus in Durango state (La Michiliá). I collected one male of the species at Las Parras in September, 1985 (Garrison, 1986) as it sat upon an exposed rock in the middle of the small stream. It was the only individual I saw. This subspecies can be common at certain sites in Arizona and New Mexico. Jo A. Garrison and I col- lected E. lampropeltis natrix on the East Fork of the Gila River at Grapevine Campground, where it was equally as common as E. heterodon. Adults landed on exposed rocks, exposed bleached logs in the center of the stream, or on the gravel shores. I have also taken it on sandy-bottomed desert streams, often in compa- ny with £. compositus. Novelo (pers. comm.) found this subspecies with the more common £. crotalinus in Durango state (La Michiliá). I observed oviposition on 1 August 1992 at the Agua Fria River, Yavapai County, Arizona. Females appeared over the stream surface in the late afternoon (about 6: 15 p.m.), and flew rapidly over the water, dispersing their eggs by dipping the abdomen briefly into the water. This behaviour continued until sun- down (about 7: 30 p.m.). A few males were still present, sitting on emergent rocks near oviposition sites. No copulations were observed. Altitudinal gradients for the species range from near 923 m (Tangle Creek, Yavapai Co., Ariz.) to 1754 m in New Mexico and southeastern Arızona (W of Portal). Like £. / lampropeltis, E. 1. natrix is pri- marily a late summer to autumn species. Flight dates range from 27 June to 10 October. Distribution (fig. 204). — Though originally de- scribed from Baja California, E. lampropeltis natrix is widely distributed throughout the southwestern United States. It has been collected as far east as west- ern Texas (Tinkham, 1934, Gloyd, 1958) and Durango and Chihuahua, Mexico. The most north- erly records are from northcentral Arizona. | have seen no specimens from central or eastern New Mexico, although it probably occurs in those areas. The species apparently does not penetrate north of the Mogollon Rim in Arizona northwest into the Colorado Desert. Material Type data. — Holotype male: Mexico: Baja California Sur: Purissima, 12 Oct. 1923 (J. H. Williamson); allotype female, same data but 6 Oct. 1923 (both examined), in ummz. I have illustrated the thorax (fig. 21), hamules (fig. 63), caudal appendages (fig. 108), and penis (fig. 85) of the holotype, and vulvar lamina (fig. 190) of the allotype. Other material (1016, 229, including holotype d and allotype 9). — U. S. A.: ARIZONA: Cochise Co.: Bear Creek, Rte 61, 6 mi SE Parker Lake, Coronado National Forest (csuc); mouth Miller Canyon, Huachuca Mtns., (rsca, UMMZ), pond N of Slaughter Ranch, 15 mi E of Douglas, 3800 ft, (RWG); springs just E of Slaughter Ranch, 17 mi E of Douglas, (RWG); Bear Creek, Rte 61, 6 mi SE Parker Lake, Coronado Nat! Forest, (csuc), San Pedro River at Hwy 90 (rsca); San Pedro River, jet. Charleston Rd., (Byuc); San Pedro River, 9 mi SW of Tombstone (Byuc); San Pedro River at Hereford Rd., (RwG); Southwestern Research Station, 5 mi W of Portal at John Hands Picnic Grounds, 5400 ft, (esca); Gila Co., Tonto Natural Bridge (RwG); Webber Creek at Camp Geronimo, 8 mi ENE of Payson, (RwG); Maricopa Co., Cave Creek by Ocotillo Rd, Cave Creek, (RWG); Camp Creek by Cave Creek Rd., 12 mi NE of Scottsdale Rd., 2700 ft., Tonto Nat'l. Forest, (RWG); Sycamore Cyn., 3.3 mi WNW of Hwy 87 (RwG); Mojave Co., irrigation ditches E of Cave Springs, (DRP); Pima Co., Sabino Cyn, N of Tucson (rsca); Lower Sabino Cyn, (RWG); Upper Sabino Cyn, ca. 3 mi N of visitor center, N of Tucson (RWG); Santa Cruz Co., Sonoita Creek Ranch (on way to Salero Cyn), along Santa Cruz River, along Ariz. Hwy 82, 2 mi SW of Patagonia, (Rw); White Rock Campground, Pena Blanca Lake, 4200 ft, (RwG, JD); Yavapai Co., Agua Fria River at Black Canyon City, (RwG); Tangle Creek, nr Tangle Creek cabin, ca 33 mi N of Carefree, TIN-RSE, sec. 1, 2800 ft (RWG); on limestone rim of Montezuma’s (sic) Well (National Monument) (RwG); Sycamore Canyon, Atasco Mtns (Lacm); New Mexico: Catron Co, Gila Cliff Dwellings, (rw); West Fork Gila River at Gila Cliff Dwellings Nat'l Mon., 5700 ft, (RwG); Grant Co., Gila River at Cliff (rsca); Mangus Springs Creek, GARRISON: Revision of Erpetogomphus Hwy 180, Mangus Springs, (RWG); Grapevine Campground, East Fork of Gila River at N. Mex. Hwy 15, 5600 ft, (RwG); Texas: Jeff Davis Co., Ft. Davis, (rsca); Limpia Creek, 2-3 mi N of Ft. Davis, (ummz); Musquiz Creek, 6 1/2 mi SE Ft. Davis (rsca); Mexico: Chihuahua: Cascada de Basaseachic National Park, (csuc, Byuc); San Diego Canyon, (Byuc); Durango: Arroyo El Temazcal (ca- mino a El Aleman), Reserva de la Biosfera La Michiliä, (UNAM); km 21 Camino Suchil-Reserva de la Biosfera La Michiliá, ‘Corralitos’, (UNAM); Sonora: Maycoba River, (syuc); Maycoba River, W of Maycoba, (Byuc); 8 mi W of Maycoba River, (csuc); Maycoba River, Hwy 16, 13 mi W of Maycoba at San Francisco Bridge, (csuc); Rio Magdalena, Hwy 15 S of Immuris, (Byuc, csuc); small riv- er W of Vecora, Rte 15, (csuc, Byuc); Baja California Sur: Purissima (rscA); Las Parras, about 18 km W of Hwy 1 on route to San Javier, SW of Loreto, (RWG). Erpetogomphus crotalinus (Hagen in Selys) (figs. 22-thx, 41-abd, 64-hamules, 86-penis, 109- app, 133-ept, 161-vertex, 169-leg, 191, 192-vl, 205- distr) Ophiogomphus crotalinus Hagen in Selys, 1854: 40 (21 sep.) (3, 2 descr. from Mexico). Erpetogomphus crotalinus. — Hagen in Selys 1858: 332 (72 sep.) (descr. d, 9, relationship with E. menetriesii); Walsh 1863: 253 (mentioned); Selys 1869: 174 (11 sep.) (Mex., attributes authorship to Selys); Selys 1873b: 519 (75 sep.) (list, attributes authorship to Selys); Kirby 1890: 61 (cat., attributes authorship to Selys); Calvert 1899: 38 (Tepic, Mex.); Calvert 1905: 165 (Mex.); Calvert 1907: 399 (Mex.), Calvert 1909: 481 (distr. in Mexico); Muttkowski 1910: 86 (cat.); Ris 1917: 153 (comparison with £. boa); Kennedy 1918: 298 (comparison with £. lampropeltis); Byers 1939: 50 (quotes Calvert 1907); Williamson and Williamson 1930: 12 (summary of sta- tus); Montgomery 1968: 133 (distr.); Paulson 1982: 255 (cat.); Davies and Tobin 1985: 27 (cat.); Tsuda 1986: 87 (cat., attributes authorship to Selys); Bridges 1991: VII.54 (cat.); Tsuda 1991: 95 (cat.); Cannings and Garrison 1991: 478 (Mex.); González and Novelo 1991: 97 (Mex.); Novelo and Gonzalez 1991: 150 (descr. larva, Mex.). Herpetogomphus crotalinus. — Hagen 1875a: 43 (cat); Selys 1879: 64 (2 sep.) (characters of genus); Kirby 1890: 61 (cat.). Description Male. — Labium grey white, base of mandibles pale green becoming red brown distally, lips black, re- mainder of face pale green except for narrow wash of brown at base of antefrons, vertex mostly dark brown with green medially behind postoccipital tubercles; antennae dark brown; anterior margin of vertex with two lanceolate pits, each anterolateral to median ocel- lus; occiput bright green, broad, gently tumid medial- ly, crest black, linear to emarginate medially, its rim covered with row of long brown hairs; postocciput gr- een, smoothly concave or somewhat angulate medial- ly, rear of head light green with brown near occipital foramen. 227 TIIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Prothorax predominantly green, most of middle lobe and posterior of frontal lobe marked with dark brown, a small middorsal green spot longitudinally divided by narrow dark brown line medially on mid- dle lobe. Synthorax (fig. 22) entirely green with brown along margins of antealar crest in a few speci- mens wich a hint of a dark brown antehumeral stripe. Coxae, trochanters pale green, femora green with streak of dark brown becoming wider distally, these streaks restricted to lateral surface of femora (fig. 169) and brown reduced to apical 0.50 of metafemora; a supplementary narrow, linear streak of dark brown more posteriorly placed on femora and, in some spec- imens, touching lateral streaks of brown distally; tib- iae ranging from all yellow green except dark brown to all brown with valle) on lateral carinae; carsi and armature black or with some yellow on exte- rior surfaces of second and third tarsomeres. Wings hyaline, venation dark brown, anterior mar- gin of costa yellow to base of pterostigma, this struc- ture brown, veins bordering it black. Venational statistics. Fifth, rarely fourth or sixth, antenodal thickened in all wings; marginal cells be- hind fore wing paranal cells: 1-3/1-3; anx: fore wing 11-13/11-13, hind wing 8-9/8-10; pnx: fore wing 7- 10/7-10, hind wing 8-10/8-10; cs under pterostigma: fore wing 4-6/4-7, hind wing 4-7/4-6; anal triangular cells: 3-4/3-5. Hind wing 26-31 mm. Abdomen (fig. 22). Samen: 1 green, slightly fus- cous basally, w ith small line of black at posterior artic- ulation point; segment 2 all green with narrow dorso- lateral streak of black above auricle and with poorly defined dorsolateral brown spot on posterior 0.25 of segment, dorsal 0.75 to 0.50 of annulus black; seg- ment 3 pale green, becoming white ventrally with black along transverse carina and elongate black dor- solateral spot on posterior 0.25 of segment connect- ing to black annulus; segments 4-7 similar to segment 3, but with narrow anterior offshoot of black laterally on transverse carina and with posterior midlateral black occupying posterior 0.50 of segments, middor- sal pale area of these segments becoming progressive- ly-more tawny posteriorly, especially so on posterior 0.50 of segment 7; segments 8-9 ocher yellow with well defined dark brown dorsolateral stripe; segment 10 ocher becoming red brown at 0.50. Abdomen 31-36 mm. laterally, basal Cercus (fig. 109) yellow, strongly to moderately angulate dorsally at distal 0.50-0.30; tip of cercus be- yond angulation attenuate, drawn out to a blunt point, ventral margin with inferior carina at basal 0.40 to 0.50. Epiprocts yellow brown, gently curved dorsally at 90°, extending to distal 0.25 or almost to level of tip of cercus, tips forming divergent points, dorsoventrally flattened (fig. 133). genitalia (fig. 64). A ACCESSOry Anterior hamule 228 black, divided at upper 0.50, lower branch forming small posteriorly directed truncate appendage; poste- rior hamule pale green, triangular, with anterior basal shoulder, tip rounded with a small anteriorly directed tooth on posterior margin; penis with semicircular spinose lateral lobe, cornuae separated; with medial lobe, and a long narrow flap at outer distal margin, membranous hood present but not overlapping. Abdomen 31-37 mm. Female. — Head as in male but with crest of occiput (fig. 161) more emarginate medially, in some speci- mens with lateral margin arcuate; prothorax and syn- thorax as in male; legs with black areas more reduced than in male; wings with slight wash of yellow at base (more so in tenerals); abdomen as in male but with black areas more reduced on abdominal segments 1- 3; dorsolateral brown stripes almost complete on seg- ments 4-7 except for basal 0.05 of each segment, these stripes complete on segments 8-9; cercus ocher. Abdomen 30-36 mm. Vulvar lamina (figs. 191-192) with broadly rectan- gular flaps meeting at anterior margin, juncture of Y- shaped postlamellar ridge at level of posterior margin of lamina. Venational statistics: number of marginal cells be- hind fore wing paranal cells: 1-4/1-3; anx: fore wing 11-13/11-13, hind wing 8-10/8-10; pnx: fore wing 7- 10/7-9, hind wing 8-10/8-10; cs under pt: fore wing 4-7/4-6, hind wing 4-6/4-6. Hind wing 26-32 mm. Diagnosis Erpetogomphus crotalinus is unique in having the extensor surfaces of the tibiae yellow (fig. 169). Some specimens have largely brown tibiae, but the lateral costate ventral margins are always yellow; all other congeners have dark brown to black tibiae. The al- most immaculate synthorax is shared by only two ot- her species: E. cophias (fig. 27) and some E. elaps (fig. 13), but the male cerci of these two species (figs. 100, 116) are never angulate as in £. crotalinus (fig. 109). The cornua of the penis (fig. 186) has a well develo- ped median lobe and a long, acuminate tip, a condi- tion unique to this species. This species is superficial- ly similar to E. heterodon and is diagnosed under that species. Females of E. crotalinus differ from other species by three characters of the head, the presence of anterolat- eral pits at the base of the frons, the relatively broad occiput, and the medial notch of the crest. The vulvar lamina (figs. 191-192) provides no discriminatory characters, and is similar to those of E. lampropeltis (figs. 189-190), E. compositus (fig. 194), E. heterodon (fig. 193), and E. sipedon (figs. 187-188). Remarks The description is based on 91 males and 43 fema- les. The lectotype male is reasonably well preserved, although it lacks the accessory genitalia. I illustrate the cercus (fig. 109) and right hind leg (fig. 169) of the lectotype, and the vertex, occiput (fig. 161), and vulvar lamina (fig. 191) of the paralectotype. Venational details for lectotype male: number of marginal cells behind fore wing paranal cells: 2/3; anx: fore wing 13/13, hind wing 9/8; pnx: fore wing 10/10, hind wing 10/10; cs under pt: fore wing 6/6, hind wing 6/6; anal triangular cells: 4/5. Hind wing 31 mm. Abdomen 36 mm. Venational details for paralectotype female: num- ber of marginal cells behind fore wing paranal cells: 1/1; anx: fore wing 11/11, hind wing 9/9; pnx: fore wing 9/9; hind wing 9/9; cs under pt: fore wing 5/4, hind wing 4/5. Hind wing 29 mm. Abdomen 33 mm. Variation. — Two teneral males of E. crotalinus reared in Arizona are superficially similar to E. heter- odon in the blunt shape of the cercus, the vestige of an isolated antehumeral stripe in one specimen, and the fuscous condition of the extensor surfaces of the met- afemora. However, the penis structure of both speci- mens is unmistakably that of £. crotalinus. Because the two Arizona specimens are teneral, I cannot deter- mine if dark thoracic stripes similar to those of £. het- erodon would have developed, or whether the macula- tion would have been the same as observed when mature. The male described by Hagen in Selys (1858) has the darkest and most distinct antehumeral stripes that I have seen for any specimen of this species. The tib- iae of a majority of specimens I examined are entirely dark brown with the lateral costae yellow; others, in- cluding the lectotype, have the entire ventral margin yellow. Biology. — Specimens are apparently common at favored localities. Boris Kondratieff (in litt., 1990) states that adults of this species were ‘on every [ex- posed] rock [in Rio Macoba] — hundreds and hun- dreds of them.’ During a two-day stay at Ajijic near Laguna Chapala, Oliver S. Flint, Jr., and M. A. Ortiz collected 28 males and 23 females. Calvert (1907) records the following concerning its habits: ‘Progomphus borealis] and Erpetogomphus were found on sand- or mud-banks, hardly above water-level, along sides of an irrigating ditch into which the water from the baths of Santa Rosalia empties. On alighting on these banks, P. obscurus borealis held its abdomen slanting upward to form an angle of 45°-60° with the bank, while E. crotalinus held it nearly horizontal.’ The male collected at Lago de Patacuaro in Michoacan is noted as taken on ‘stream flowing into lake’, the two males and one female from the pyra- mids at S. Juan Teotihuacan as ‘on high grassy plains far from water’. Gonzälez and Novelo (1991) record GARRISON: Revision of Erpetogomphus the following about £. crotalinus (in translation): “This is the most abundant gomphid in the area [the Biosphere Reserve of La Michiliä, Durango]. Like other members of the family, the adults of E. crotali- nus have a seasonal flight pattern that, although it has not been exactly established, apparently is restricted to the months of May to October. The individuals of this species are characteristic inhabitants of both sea- sonal and permanent streams, easily recognized in the field by the bright green colour of the thorax. Males have the habit of resting on emergent rocks, a posi- tion they defend aggressively from other conspecific males.” Erpetogomphus crotalinus has been collected with £. designatus, E. lampropelti, and E. heterodon. Collection dates range from May through October (Calvert 1905). Distribution (fig. 205). — Erpetogomphus crotalinus is restricted to the higher xeric areas of central and western Mexico. The species was reared from larvae collected by Minter Westfall, Jr., from the Slaughter Ranch, 15 mi W of Douglas, Cochise Co., Arizona; but no other collectors have encountered it there. I have also seen a pair from New Mexico (Eddy Co., Sitting Bull Falls), which may be strays from Mexico; its occurrence in the southwestern United States ap- pears sporadic, and other odonatists have failed to find the species during their work in Arizona and New Mexico. Altitudinal data show a range of 1225 m (Mexico: Chihuahua: Santa Rosalia Springs) to 2290 m (Mexico: Mexico: Pyramids, S. Juan Teotihuacan). In the Paris Museum is a male labelled, ‘Veracruz’, which is probably in error. Material Type data. — Lectotype male by present designa- tion with following data: white handwritten label by Hagen, ‘G. lineatus | Mexico’; white handwritten la- bel, ‘crotalinus in an unknown hand; white, black- bordered printed label: ‘Mus. Berol.’; two small prin- ted labels, ‘Hagen’; red label with “Type [printed]/12334 (written by N. Banks); rectangular label [all printed unless otherwise stated]: ‘ERPETOGOMPHUS / CROTALINUS HAGEN [stamped]/ P. P. Calvert, det. 1905 [written]/ B. C. A. Neur. p. 165 [written]. In mczc. The penis and hamules are missing and probably served as the basis for the free- hand illustrations of these structures for Hagen in Selys (1858). Paralectotype female with following da- ta: green printed label: ‘Mexico/ Deppe.’; red printed label “Typus’; yellow printed label, ‘Zool. Mus./ Berlin’, on reverse side written in ink, ‘cat nr./ 2314; white label: ‘syntypus © von: / Ophiogomphus | crot- alinus Hagen’ [written in an unknown hand]; yellow label (handwritten except as noted), ‘PARALECTOTYPE [printed] 9/ Ophiogomphus | crotalinus Hagen/ des. 229 Li [1NDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 [printed] R. W. Garrison 1984’. In ZMHB. There is a male in the zMHB with the same data as the paralectotype female, except as follows: green la- bel, ‘Mexico Deppe’; green label, ‘crotalinus | Hag.’ [both handwritten by Selys); printed label, ‘2314’; and ‘Syntypus d von / Ophiogomphus | crotalinus Hagen’. Although the specimen is labelled as a type, I believe it is not a syntype and, in fact, that it repre- sents the second male described by Hagen in Selys, 1858. I decided that the male in the mczc and the fe- male in the zMHB represent the original syntypes for the following reasons: The original (1854) descrip- tion of Ophiogomphus crotalinus is brief and was based on a male and female. Hagen gave measurements for both sexes (abdomen: male 36 mm, female 33 mm; hind wing male 31 mm, female 29 mm). These meas- urements and the description of the male (‘Front of thorax yellow, with indication of two median stripes and a pale red humeral stripe...’) correspond to the meze male and the ZMHB female. Hagen in Selys (1858) described this species in greater detail and had access to a second, smaller male (abdomen 33 mm, hind wing 29 mm). He stated, ‘In the other specimen (fe, the second male], the stripes are brown and straight, quite visible...” The original male is again characterized as follows, "Thorax yellow green with- out spots and...one can scarcely see the appearance of two median stripes and a thick antehumeral stripe...’, which tallies with the earlier description of 1854. The male from the zMHB corresponds with the smaller of the two measurements given by Hagen in 1858; and it also has a pair of easily visible antehumeral stripes, an apparently rare feature for this species. The ZMHB male is unique in this respect. Even though the origi- nal description by Hagen gives ‘Musee de Berlin’ as the repository of the types, I strongly suspect that Hagen took the male with him when he was invited to join the faculty at the Museum of Comparative Zoology in 1867. Other material (97d, 449, including lectotype d and paralectotype 2). — U. S. A.: Arizona: Cochise Co., San Bernardino Ranch, 15 mi E of Douglas (rsca); New Mexico: Eddy Co, Sitting Bull Falls, Guadalupe Mtns., RWG, wM); Mexico: Chihuahua: Chihuahua Creek, 4 mi SW of Col. Juarez, (csuc); Rio Pacheco, (csuc); Santa Rosalia Springs, 1219 m, (anse); Durango: Durango City, NSP); Rio Dalita and marshes, 6 mi E of Durango, 1890 Jalisco: Jopopau, 1667 m, (Ansp); Guadalajara, MCZC); Laguna de Chapala (nr. Mex. Hwy 15), 75 mi SE of Guadalajara, (RwG); Ajijic nr. Laguna Chapala (usNM, \jijic, (rsca); Rio La Sanguijuela, Degollado, 15, km 206, San Lorenzo, 9 mi W c); Jalisco [no other locality], (BMNH); » W of La Piedad, (unam); Lago de RWG); route ro, 2.6 mi W of Quiroga, 2042 m, (rsca); Route 15, San Lorenzo, 8 mi W of Tuxpan, (UsNM); Route ( ; USNM); Route 15, km 291, near Nacional Insurgente Morelos, (USNM); 30 stream 0.5 mi S of Tuxpan, 1798 m, (pre); Hidalgo: ditch- es around Tasquillo, 1700 m, (pre); Rio Tula at Puente Tasquillo, (UsNM); Tecozautla, Rio San Juan, 1710 m, (RWG); Tecozautla, Rio Tecozautla, 1750 m (Rw); Tecozautla, arroyo San José del Desierto, 1730 m (RwG); Mexico: Route 15, km 125, El Salitre, (UsNM); Pyramids, S. Juan Teotihuacan, 2290 m, (rsca, cuic); Morelos: Cuernavaca (BMNH, ANSP, CUIC); Puebla: ditch just NW of Tehuacan, 1700 m, (DRP, TwD); Veracruz: Veracruz [prob- ably erroneous locality], (MNHP); Mexico [no other local- ity], (ZMHB). Erpetogomphus heterodon sp. n. (figs. 28-thx, 42, 43-abd, 65-hamules, 87-penis, 110, 111-app, 162-vertex, 170-leg, 193-vl, 206- distr, 230-wings) Type data. — Holotype male. U. S. A: New Mexico: Catron Co., Tularosa River just E of Aragon on N. Mex. Hwy 12, 28 July 1984 (R. W. and J. A. Garrison). Allotype female. Same data as male. Both in USNM. Paratypes (369,22). - New Mexico: Catron Co., same data as holotype and allotype, 27-28 July 1984, 22d, 19 (Rwe); same data, 23 August 1964 (Clifford Johnson), 38, 12 (pre, Fsca); Gila Cliff Dwellings, 30 July 1983 (John E. Hafernik, Jr.), 4d (RWG); Grant Co, Grapevine Campground, East Fork of Gila River at N. Mex. Hwy 15, 5600 ft., 27 July 1984 (R. W. and J. A. Garrison), 118 (Rwo); Texas: Jeff Davis Co, Limpia Canyon in Davis Mtns. State Park, 23 June 1958 (M. J. Westfall, Jr.), 16 (rsca); Limpia Creek at Tex. Hwy 17, 3.4 mi NE of Ft. Davis, 4500 ft., 24 August 1977 (Dennis Paulson, Susan Hills), 1d (pre); Reeves Co., Balmorhea State Park, 13 Sept. 1983 (S. W. Dunkle), 1d (swp); Mexıco: Chihuahua: 6 mi NNE of Boquilla, 16 July 1960 (Scheibner), 16 (Rw); Chihuahua Creek, 4 mi S of Col. Juárez, 25 Aug. 1986 (B. C. Kondratieff), 14 (csuc); ‘Bosuchel [probably Basuchi]’, 18 Aug. 1950 (R. F. Smith), 28 (AMNH, RWG). Description Holotype male. — Entire face pale light green, slightly darker on postclypeus and anterior of frons; labrum, base of mandibles, labium ivory; vertex and antennae black; base of antefrons with transverse fur- row, tumid medially, laterally with a pair of pits; po- stocellar tubercles small, no postocellar ridge; occiput pale green, slightly tumid medially, hind margin frin- ged with long hairs, crest with black along lateral margin, slightly emarginate medially; postocciput gr- een, transverse; rear of head dark brown, becoming pale green laterally. Prothorax light green with dark brown on middle lobe extending laterally to pleura, two adjacent later- odorsal small green spots on medial lobe. Synthorax (fig. 23) light green with following dark brown: in- verted wedge-shaped middorsal stripe, its ventral end not touching collar, its dorsal end forming a narrow stripe below antealar carina and meeting narrow hu- meral stripe; small isolated antehumeral stripe; vesti- gial second lateral stripe extending from metacoxa to metaspiracle; upper end of third lateral stripe. Coxae, trochanters ivory, femora ivory with black external surfaces spreading to lateral areas at distal 0.25 (fig. 170), tibiae black with small lateral ivory stripes along basal 0.5; tarsi and armature black. Wings (fig. 230, paratype) hyaline, venation dark brown, basal wing venation brown; anterior margin of costa yellow; pterostigma brown, darker around margin. Venational details. Fifth antenodal thickened in all wings; number of marginal cells behind fore wing | paranal cells: 2; anx: fore wing 14/12, hind wing 9/9; pnx: fore wing 9/8, hind wing 10/10; cs under pteros- tigma: fore wing 4/6, hind wing 6/4; anal triangular cells: 4. Hind wing 33 mm. Abdomen (fig. 42). Segment 1 pale orange-yellow dorsally with narrow black lateral stripe, white lateral- ly; segment 2 similar to segment 1 but with black along anterior margin of segment laterally; segment 3 similar to segment 2 but with black along transverse carina and posterior annulus; segments 4-7 similar to segment 3 but black dorsolateral stripe incomplete anteriorly so that dorsal orange-yellow merges with lateral white and with black along transverse carina, expanding ventrally, and with a similar mark posteri- orly just before black annulus; segments 8-9 predom- inantly yellow-orange with black denticles dorsally, incomplete black dorsolateral stripe extending from anterior margin of segment but ending just before posterior margin; segment 10 yellow-orange posteri- orly, orange-brown anteriorly. Abdomen 40 mm. Cercus (fig. 110) yellow brown, strongly angulate near middle, ventral margin with inferior carina at ba- sal 0.30. Epiprocts curved at distal 0.75, tips dark- ened, bluntly pointed, slightly divergent. Accessory genitalia (fig. 65, paratype). Anterior ha- mule black, divided at upper 0.30, lower branch form- ing a small, truncate appendage; posterior hamule white, triangular, with anterior basal shoulder, tip with small, anteriorly recurved tooth; penis with small serrated lateral lobes, cornuae (fig. 87, paratype) mod- erately long, widely separated tips broadly rounded. Allotype female. — Overall colouration similar to male but with more extensive pale areas on abdomen and legs; furrow anterior to medial ocellus as in male; vertex (fig. 162) with postocellar tubercles not con- nected but with small remnant of ridge extending medially; occiput simple, posterior margin raised with row of hairs; postocciput green, transverse. Venational details. Fifth antenodal thickened in all GARRISON: Revision of Erpetogomphus wings; number of marginal cells behind fore wing paranal cells: 3/2; anx: fore wing 13/12, hind wing 10/10; pnx: fore wing 7/8, hind wing 9/10; cs under pterostigma: fore wing 6/6, hind wing 6/6. Hind wing 36 mm. Abdomen (fig. 43) as in male, but with black dor- solateral stripe reduced, with black extending anteri- orly before transverse carina. Abdomen 37 mm. Vulvar lamina (fig. 193) with broadly planar rec- tangular flaps meeting at anterior margin; juncture of Y-shaped postlamellar ridge at level of posterior mar- gin of lamina. Diagnosis This species is related to E. crotalinus and E. sipe- don. Erpetogomphus crotalinus lacks the dark thoracic stripes present in E. heterodon (fig. 22); the extensor surfaces of the tibiae of E. heterodon are black (fig. 170) (yellow in E. crotalinus); and the cornuae of the penis in E. heterodon are broadly rounded (fig. 87) (long and acutely pointed in £. crotalinus [fig. 86]). Males of E. heterodon are most similar to E. sipedon, but they differ in thoracic maculation shape of the epiprocts, and structure of the penis. In E. sipedon, the antehumeral stripe is long and joined with the hu- meral stripe (fig. 19); in £. heterodon, this stripe is ves- tigial and disjunct. In E. sipedon, the tips of the super- ior appendages are slender and acuminate (figs. 105-106), not obtusely acuminate as in E. heterodon (figs. 110-111). The cornua of the penis in E. sipedon (fig. 83) possess a mesal lobe absent in £. heterodon (fig. 87). Structurally, the females of E. crotalinus, E. hetero- don, and E. sipedon approach one another closely. The latter two species have a relatively narrow occiput with the width across the crest about 4 times as long as its greatest length. The occiput in E. crotalinus is wider, about 3 times the width. I have been unable to detect morphological differences between females of E. heterodon and E. sipedon. Body maculation must be used to separate females. Remarks Variation. — Slight variation occurs in males within the type series. The dark dorsolateral stripe on abdo- minal segment 3 may be broken in the middle. In the male from Mexico (6 mi NNE Boquilla), this stripe is reduced to an isolated spot on the basal 0.30 of the segment. Similarly, the extent of black markings on abdominal segments 8-10 may be reduced. Venational statistics for type series. Males: Number of marginal cells behind fore wing paranal cells: 0-3; anx: fore wing 11-14/11-15, hind wing 8-11/8-11; pnx: fore wing 7-11/7-11, hind wing 8-12/8-11; cs under pterostigma: fore wing 4-7/4-6, hind wing 4- 7/4-7; anal triangular cells: 3-4. Hind wing 32-34 231 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 mm. Abdomen 36-40 mm. Females: Number of marginal cells behind fore wing paranal cells: 1-4; anx: fore wing 13/12-13, hind wing 8-10/9-10; pnx: fore wing 7-9/8-9, hind wing 8-10/8-10; cs under pterostigma: fore wing 5-6/5-6, hind wing 5-6/5-6. Hind wing 34-36 mm. Abdomen 36-37 mm. Biology. — Adult males were taken along sandy margins of the Tularosa River in New Mexico. They often sat on the bank facing the river and were easy to approach. I collected one female as it hovered over the river ovipositing. Another teneral female was flushed from nearby vegetation. At the East Fork of the Gila River, males sat on emergent rocks in the stream or by the river’s edge. They occasionally made swift sorties, usually to chase another male. At this lo- cale, £. heterodon was sympatric with E. lampropeltis natrix. Altitudinal data show £. heterodon to occupy high, mostly xeric areas of the southwestern United States and northern Mexico. Collection dates are from 23 June to 13 September at elevations of 1370 m (Limpia Creek, Texas) to 1700 m (Grapevine Campground, New Mexico). Distribution (fig. 206). — Erpetogomphus heterodon is known only from western Texas, New Mexico, and northern Mexico. It has been collected with £. crotal- inus in at least one locality in Mexico (Chihuahua state, Rio Pacheco, 27 Aug. 1986, B. Kondratieff, S. Dunkle, pers. comm.), and is apparently allopatric with £. sipedon. Erpetogomphus compositus Hagen in Selys (figs. 24, 25-thx, 66, 67-hamules, 88, 89-penis, 112, 113-app, 145-ept, 139-base of postfrons, 163-vertex, 194-vl, 206-distr) Erpetogomphus compositus Hagen in Selys, 1858: 660 (400 sep.) (2, ‘Rivière Peros [sic Pecos] (Texas occidental’). — Hagen in Selys 1859: 536 (10 sep.) (descr. 9); Hagen in Selys 1873a: 740 (12 sep.) (descr. &, 9, ‘Le nord de la Californie”); Selys 1873b: 519 (75 sep.) (list); Calvert 1905: 166 (Ariz.); Calvert 1908a: 45 (Ariz.); Muttkowski 1910: 86 (cat.); Calvert 1912a: 289 (mentioned); Williamson 1914a: 226 (Ariz.); Kennedy 1917a: 544 (Calif.); Seemann 1927: 22 (Calif.); Byers 1928: 51 (lar- va unknown); Needham and Heywood 1929: 80 (descr.); Williamson and Williamson 1930: 12 (summary of stat- us); Tinkham 1934: 215 (W. Tex.); Ahrens 1938: 11 (Arız., Utah); Fraser 1940: pl. 5 (penis); Ferguson 1940: 5 (Tex.); Needham and Westfall 1955: 144 (descr.); Pritchard and Smith 1956: 116 (key); Gloyd 1958: 8 (W. Tex.); Musser 1961: 54 (larva, Utah); Musser 1962: 14 larva, Utah); Borror 1963: 104 (common name); Cruden 1964: 81 (Calif.): Montgomery 1968: 133 (distr.); Paulson and Garrison 1977: 151 (Washington); Molnar and I ivigne 1979: 130 (Wyo.); Paulson 1982: 255 (distr.); Paulson 1983: 67 (Wash.); Davies and Tobin 1985: 27 (cat.); Tsuda 1986: 87 (cat attributes authorship to Selys); Bick 1990: 3 (Idaho); Bridges 1991: 232 VII: 49 (cat.); Tsuda 1991: 5 (cat.). Gomphus compositus. — Hagen 1861: 99 (desc. 2, Tex.). Gomphus (Herpetogomphus) ? viperinus. — Hagen, 1873 (mis- identification, Yellowstone). Herpetogomphus compositu. — Hagen 1874: 597 (Yellowstone, Tex.); Hagen 1875a: 42 (Tex.); Hagen 1875b: 918 (N. Mex.); Selys 1879: 64 (2 sep.) (characters of genus); Kirby 1890: 60 (cat., attributes authorship to Selys); Calvert 1899: 386 (characters of genus); Currie 1903: 303 (Ariz.); Osburn 1905: 186 (mentioned); Byers 1928: 5 (larva unknown); La Rivers 1938: 85 (Nev.); La Rivers 1940a: 112 (Nev.); La Rivers 1941: 177 (Nev.). Erpetogomphus coluber Williamson and Williamson, 1930: 17 (d, San José de Comandu, Baja Calif. Sur). — Needham and Westfall 1955: 143 (desc.); Borror 1963: 104 (common name); Montgomery 1968: 133 (distr.); Montgomery 1973: 239 (comment on name); Paulson 1982: 255 (distr.), 266 (synonymy with £. compositus); Bridges 1991: VII: 49 (cat.). Description Male. — Entire face pale grey green with dark mar- kings as follows: along basal margin of labium and median extension ending before distal margin (almost absent in some specimens), frontoclypeal suture, base of antefrons; vertex all dark brown or with pale me- dian area extending posteriorly from median ocellus to anterior margin of occiput; antennae dark brown; anterior margin of vertex forming a trough at junc- tion of postfrons, this trough with a pair of deeper pits anterolateral to median ocellus (fig. 139); occiput white green, wide, planar with slight tumid median area, crest barely convex, slightly to broadly emargi- nate medially, its hind margin covered with pale brown hairs; postocciput white green, transverse, not visible dorsally, rear of head dark brown. Prothorax pale green with brown dorsolaterally on middle lobe; synthorax (fig. 24) predominantly pale green with following dark brown: well defined mid- dorsal stripe widening to collar, its upper end extend- ing along antealar sinus connecting with well defined humeral stripe; antehumeral stripe widened dorsally and connected basally at mesinfraepisternum; lower part of humeral stripe extending posteroventrally but not connecting with well defined sinuate second lat- eral stripe; lower 0.50 of this stripe encompassing metaspiracle, constricted along posterior margin at 0.50 and abruptly widening dorsally before turning anterodorsally toward subalar carina; complete third lateral stripe ending behind posterior margin of met- acoxa. Pale colour of thorax light green with lighter tone almost becoming white in areas between antehu- meral and humeral stripes, and between second and third lateral stripes. Coxae, trochanters white grey with slight wash of brown, tibiae pale grey green with defined superior surfaces black, basal 0.40 of metafe- mora with streaks of grey green or with basal 0.60 of this area largely pale, tibiae black with narrow stripe of pale grey green along basal 0.30 of lateral margin in some specimens; tarsi and armature black. Wings hyaline with slight wash of yellow at ex- treme base (more so on teneral specimens), basal wing venation brown, especially anteriorly, anterior margin of costa yellow white except for anterior margin of pterostigma, costa beyond pterostigma white or dark- ened, but contrasting with black of pterostigma; pte- rostigma black. Venational statistics. Fifth antenodal thickened in all wings; marginal cells behind fore wing paranal cells: 0-4; ; anx: fore wing 10-15/10-15, hind wing 7- 11/7-11; pnx: fore wing 6-11/6-10, hind wing 7- 11/7-11; cs under pterostigma: fore wing 3-6/3-6, hind wing 3-6/3-6; anal triangular cells: 2-5. Hind wing 25-32 mm. Abdomen. Segment 1 pale grey green with broad basal semicircle of brown laterally, its posterior end touching lateral articulated area; segment 2 white with dorsolateral brown stripe sending a ventral stem posterior to auricle, and connecting with black annu- lus, a small brown spot at anteroventral margin of segment; segment 3 with isolated black dorsolateral spot surrounding lateral carina, distal 0.25 of segment with dorsolateral black increasing in width posterior- ly and touching dorsally near black annulus; segment 4 similar to segment 3, but isolated black of transverse carina longer and touching or connecting with distal black, thus forming an incomplete midlateral stripe occupying distal 0.75 of segment and isolating ventral longitudinal wedge of white; segments 5 and 6 like segment 4, but with middorsal white becoming fus- cous; segment 7 with anterior 0.5 white, except for tawny transverse carina, becoming entirely tawny posteriorly, some specimens with varying degrees of lateral black on posterior 0.25 of segment or in speci- mens from Baja California (E. coluber) with tawny ar- eas becoming black; segments 8-10 yellow brown with darker red brown dorsally, in Baja California specimens this dark red brown replaced with well de- fined black on segments 8 and 9, and dark brown on segment 10. Abdomen 32-39 mm. Cercus (figs. 112-113) ocher, in lateral view slight- ly concave dorsally near base and with tips slightly re- cumbent, cercus enlarged ventrally at basal 0.45 with inferior carina along this same area, remainder slight- ly concave with tip forming a blunt point. Epiprocts yellow brown, gently curved dorsally at 90°, tips (fig. 134) dorsoventrally flattened and truncate or obliquely truncate. Accessory genitalia (figs. 66-67). Anterior hamule dark brown, divided at basal 0.5; lower branch form- ing a small, posteriorly directed truncate appendage; posterior hamule grey white, triangular; with anterior basal shoulder, tip bluntly rounded and armed with a small anteriorly directed tooth on posterior border; penis with lateral lobe small, roughly semicircular, GARRISON: Revision of Erpetogomphus serrate; cornuae (figs. 88-89) moderately long, widely separated, tips broadly rounded. | Female. — Head as in male but pale areas more ex- tensive, especially on vertex, where dark brown may be reduced to areas around ocelli and postocellar pro- tuberances; occiput narrower than in male, with erect curvilinear occipital crest separating easily visible postocciput (fig. 163), posterior margin of postocci- put slightly concave; prothorax and thorax as in male, but with pale areas more extensive; wings with yellow at base more extensive, often with markings within second series of antenodals to level of arculus and cu- bitoanal area in both wings. Abdomen in well preserved specimens with pale colours primarily white with pale orange dorsally, black dorsolateral stripes reduced and separated in their middle on segments 3-6; segment 7 similar to male but with posterior 0.25-0.30 darker laterodor- sally; segments 8-10 tawny, with varying amounts of dark brown, especially dorsally, on segments 8-9; cer- ci ocher. Abdomen 31-37 mm. Vulvar lamina (fig. 194) with prominent basal plates, quadrangular laminar plates acuminate poste- riorly, meeting at anteromedial margin; cleft between plates forming a 90° angle; stem of Y-shaped postla- mellar ridge at level of posterior margin of lamina. Venational statistics: number of marginal cells be- hind fore wing paranal cells: 1-6/0-5; anx: fore wing 12-15/12-14, hind wing 8-11/9-11; pnx: fore wing 7- 11/8-10, hind wing 9-11/8-11; cs under pterostigma: fore wing 3-5/4-6, hind wing. Hind wing 30-32 mm. Diagnosis This common, distinctive species is easily separated from its nearest ally, £. lampropeltis, by several struc- tural and maculation characters. Males of £. composi- tus differ from E. lampropeltis in lacking the characte- ristic middorsal angulation of the cercus. The tip of the epiproct in E. compositus is truncate (fig. 134), but this structure in E. lampropeltis is obliquely and bluntly pointed (fig. 132). Depressions in male and female anterolateral to median ocellus will also sepa- rate the species: in Æ. compositus, an arcuate trough has more pronounced pits anterolateral to the median ocellus (fig. 139). In E. lampropeltis, the anterolateral pits are deeper and the intervening area directly ante- rior to the median ocellus is not as depressed (fig. 138). This structure in £. lampropeltis more closely approaches the condition found in £. viperinus (fig. 137). Erpetogomphus compositus, though a boldly marked species, is lighter than £. lampropeltis. The occipital crest in E. compositus is largely pale, though there may be some darkening along the lateral 0.30 of margins in specimens from Baja California. In Æ lampropeltis, the occipital crest is entirely dark brown and provides TI]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 a striking contrast to the pale occiput. The thoracic patterns of the two species are different (figs. 20-21, 24-25), and the dorsolateral black stripes of abdomi- nal segments 4-6 of E. lampropeltis are more exten- sive. In che field, pale thoracic colouration will easily separate both species. Erpetogomphus compositus is unique in having a pale green thorax with intervening grey white areas between the antehumeral and hu- meral stripes and second and third lateral stripes. In E. lampropeltis, the thoracic colouration is grey green (E. lampropeltis lampropeltis) or dark green (E. lam- propeltis natrix). Females of £. compositus differ from E. lampropeltis in the shape of the occiput. In E. compositus, the occi- put is narrow and the postocciput is clearly visible in dorsal view (fig. 163). In E. lampropeltis, the occiput is broad and the postocciput is not visible dorsally (fig. 160). Secondly, the area in front of the median ocellus mirrors the condition stated for males above. The vulvar laminae of the two species appear indistin- guishable. Remarks Williamson and Williamson (1930) described in detail Erpetogomphus coluber, comprising 31 males fr- om San José de Comandu, Baja California. In com- paring E. coluber with other species, they stated that E. coluber would, in Calvert's (1905) key to the genus run to AA, page 160, and if the individual be one with the facial dark markings reduced, it will run to H. compositus. Drs. Calvert and Kennedy regard colu- ber as distinct from any described species.’ The Williamsons sent two males to Calvert, who com- mented, ‘No special reason for thinking it diadophis [= £. eutainia); like a small compositus, differs in size; has broader metepisternal dark stripe; brown on [ab- dominal segments] 3-7 more extended; brown on 8- 10 much darker; darker lines or stripes on the fronto- clypeal and clypeo-labral sutures’. Finally, the Williamsons commented on the affinities of E. colu- berto E. compositus as follows: ‘Compositus is certain- ly its closest relative and the derivation of coluber fr- om compositus, through geographical isolation in Baja California, is almost certain and is a case exactly pa- rallel, so far as speciation goes, to that of certain spe- cies, of several genera, endemic in Florida. Coluber is separated from compositus by its darker color, espe- cially of the last four abdominal segments and by ve- national characters, especially the two-celled anal triangle and the single row of cells posterior to A in the front wing.’ I have seen the holotype and 15 paratypes of E. co- luberand believe, as did Paulson (1982), that they re- present diminutive, melanic examples of E. composi- tus. The specific differences ascribed to E. coluber are i) Qo HN those of venation only, and I believe these to be cor- related to their smaller size. I have found no differen- ces in body morphology, including the accessory ge- nitalia. Variation. — Two specimens from San Diego County appear intermediate to £. coluber and E. com- positus s.s. These males have more pronounced dark thoracic stripes, but the dorsum of abdominal seg- ments 8-10 lacks dark brown markings typical of £. coluber. However, other E. compositus have varying amounts of dark brown on these segments. The hind wing anal triangle in 3 of 16 specimens of £. coluber has 3 anal triangular cells instead of 2, and a 2-celled anal triangle, though rare, does exist in E. compositus (s.s.). For example, of 21 males I measured from California, Arizona, and another from northern Baja California, one of the San Diego males and another from Riverside, California, had one wing each with two cells. These two specimens had shorter hind wings (26 mm and 28 mm, respectively) than is typi- cal for E. compositus (s.s.). Statistics for the number of anal triangular cells show it to be variable within Z. compositus (s.s.). The number of wings with range of stated conditions is (number of wings follows in pa- rentheses): 3 (19), 4 (19), 5 (2). Hind wing lengths from the San Diego males are 25-26 mm, well within the range for E. coluber. The next smallest specimen is a male from Riverside (28 mm), the same male with two anal triangular cells in one hind wing. Most wings of E. coluber which I have examined have no marginal cells behind the fore wing paranal cells, but 6 of 32 (19%) did. Eight of 42 (19%) wings of E. compositus (s.s.) had no marginal cells. Williamson and Williamson (1930) warned of fal- libility of some of their characters among the para- types they examined: “The face markings are very pro- nounced in some and scarcely discernible in others; there is considerable variation in the extent of the dark thoracic markings [fig. 25] and this variation oc- curs independently on the mesepisternum and mete- pisternum, so an individual may have the dark hu- meral and antehumeral relatively extensive and the two dark lateral stripes relatively reduced and vice ver- sa. With the exception of specimens from Baja California and the two specimens from San Diego, body colouration appears remarkably constant throughout its range. The male from Baja California Norte (Guadalupe Hot Springs) is more like typical E. compositus, but it has no marginal cells behind the fore wing paranal cells. The hind wing length (29 mm) and other venational characters indicate a closer relationship to more northerly populations. The female holotype was collected with the type se- ries of E. designatus and was probably collected in the vicinity of Roswell, New Mexico in 1854-1855. See remarks under E. designatus for further details. Venational details. Holotype female of £. composit- us number of marginal cells behind fore wing paranal cells: 2/4; anx: fore wing 13/13, hind wing 9/10; pnx: fore wing 8/9, hind wing 9/8; cs under pterostigma: fore wing 5/6, hind wing 5/6. Holotype male of £. co- luber. number of marginal cells behind fore wing par- anal cells: 0/0; anx: fore wing 12/12, hind wing 9/9; pnx: fore wing 7/7, hind wing 9/9; cs under pterostig- ma: fore wing 2/2, hind wing 4/4. Biology. — This species and Progomphus borealis McLachlan in Selys are the two most conspicuous gomphid elements along most desert streams and irri- gation ditches in the southwestern United States. In Arizona, this species occurs commonly in the Lower Sonoran Zone, and in ecotonal fingers into the Upper Sonoran Zone. Kennedy (1917a), who collected £. compositus at certain sites in the northern Central Valley of California, writes: ‘One female, a teneral, was taken on the irrigating ditch across the river from Oroville.’ Collecting along the lower Truckee River south of Pyramid Lake, Nevada, Kennedy (1917a) mentions that “This widely spread species occurred sparingly on the riffles of the lower Truckee. The males appeared to be more nervous and more touchy, flying farther for conflict with passing males than the males of Ophiogomphus morrisoni nevadensis.’ The two males collected at the Boyce Thompson Southwestern Arboretum in Pinal County, Arizona, had the following annotation by H. K. Gloyd: ‘[in] Creek, arboretum. Canyon, after rain and run-off al- most over. between 5 & 7: 00 p.m., water low, few dragonflies seen’. Williamson and Williamson (1930) provided a lengthy account of the type locality and as- semblage of Odonata present with Erpetogomphus co- luber. In August, 1972, I collected several E. compos- itus with Stylurus plagiatus (Selys) at Riverside Park, Yuma, Arizona. Adults were taken in the late after- noon on grassy areas shaded by trees. Adults of this species and Progomphus borealis were abundant along Big Chico Creek in Chico, Butte County, California in June and July of 1974, 1976, and 1978. They sat on exposed sand bars near the edge of the creek, and both were difficult to approach. Progomphus borealis was the more wary. In the late afternoon, I have col- lected E. compositus away from water, resting on dry desert scrub. Collection dates range from 24 May (Nevada: Clark Co.) to 13 Dec. (Calif: San Bernardino Co.). Distribution (fig. 206). — Erpetogomphus composit- us is a desert species found commonly in the south- western United States. It has been taken as far east as Dallas (Hagen, 1875a), but most records from Texas are from the western portion of the state. It occurs in eastern New Mexico and is common in low desert re- gions of Arizona, especially southwest of the GARRISON: Revision of Erpetogomphus Mogollon Rim. It is common in the arid regions of southern California and penetrates north through the Central Valley to Chico. Paulson and Garrison (1977) list it from south central Washington. The species was listed from Oregon by Hagen (1875a, 1875b) with no specific locality); but specimens were recently collected from the John Day River, Deschutes Cos, Oregon (Valley 1995): Erpetogomphus compositus occupies river systems in the Great Basin of Nevada, but its most easterly dis- tribution is unknown. Bick (1990) lists a specimen from Owyhee Co., Idaho. Hagen (1874) records it from the Yellowstone, but it was probably taken in a more arid environment around what is present-day Yellowstone National Park. Molnar and Lavigne (1979) also cite it without specific locality from Wyoming, based on previous records given by Needham and Heywood (1929), Needham and Westfall (1955), and Pritchard and Smith (1956). All of these records probably originated from the original Hagen (1874) citation. In Utah, the species apparent- ly penetrates only the southwesternmost part of the state. The most southerly distribution for E. compos- itus is also unknown. Except for E. coluber, the only records I have observed from mainland Mexico are two males collected by J. H. Williamson in Hermosillo, Sonora. The distribution of £. composit- usin Texas, New Mexico, and Arizona indicates that it must be further widespread in northern Mexico. Material Type data. — Of E. compositus Holotype female with following label data: ‘Aug/ 16° [handwritten], ‘Hagen’ printed [two of these labels attached], ‘com- positus written in an unknown hand; red label ‘Ho- LOTYPE [printed] / Erpetogomphus compositus | Hagen 1857 [sic, should be 1858] ‘2’ handwritten by rwe. In mezc. Of E. coluber. Holotype male dry in envelo- pe with following data: ‘E. coluber written in pencil by E. B. Williamson, ‘Mexico/ State of Baja California/ San Jose de Comandu/ J. H. Williamson/ Oct 10 1923’ [all stamped]/ 6214.’ handwritten by EBW, ‘Type handwritten in pencil by EBW on side of label. In ummz. Type locality of E. compositus herein restricted to vicinity of Roswell, Chaves Co., New Mexico (see re- marks under £. designatus). Other material (1516, 469, including holotype ? E. compositus and holotype & E. coluber). — U. S. A: ARIZONA: Cochise Co., Miller Cyn., Huachuca Mtns., (cc); pond at Slaughter Ranch, San Bernardino Valley, 15 mi E of Douglas, 1169 m, (rwc); 5 mi E of Hereford (cc); San Pedro River, 9 mi SW of Tombstone, (Byuc); San Pedro River, at Hereford Rd., (rw); Coconino Co, Havasu Canyon, 3 mi N of Supai, 923m, (rwG); Supai, (CDFA); Graham Co., 3 mi SE of Bylas, valley of the Gila, 2000 fr, (ummz); Roper Lake, 5 mi S of Safford (Rw); Maricopa D (SS) UN [ij DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Co., Granite Reef Dam, (umMZ); Mesa, (UMMZ); Phoenix, (rw); slough ponds by Verde River, by Ariz. Hwy 87, Ft. McDowell Indian Reservation (RWG); Tempe (UMMZ); Mohave Co., spring, Hwy 15 bridge, Littlefield, (Byuc, rwG); Virgin River, Big Bend, (Byuc); Pima Co., Organ Pipe Cactus Natl. Mon., Quitobaquito, (Long Beach State Univ.); Quitobaquito Springs, 15 mi S of Lukeville, Organ Pipe Cactus Natl. Mon., (RwG); Pinal Co, Boyce Thompson Southwestern Arboretum, 4 1/2 mi SW of Superior, (UMMz); 8 mi NW of Florence, (UMMZ); Santa Cruz Co., Santa Cruz River, Sonoita Creek Ranch (on way to Salero Cyn.), ca. 2 mi W of Patagonia, by Ariz. Hwy 82, (RwG); Yavapai Co., Oak Creek at Cornville, 1077 m, (csuc); Yuma Co. Ave 3E at Co. 14 St., SE of Yuma, (RWG); Riverside Park, nr. Colorado River, Yuma, (RwG); N. R. Adair Park, McPhaul Bridge, by Gila River, ca. 15 mi ENE of Yuma, (RWG); CALIFORNIA: Butte Co., Bidwell City Park, by Big Chico Creek, Chico, 61 m, (RWG); Oroville, (uUMMZ); Fresno Co., Friant, San Joaquin River, (UMMZ); Imperial Co. Calexico, (Byuc); canal 8 mi E of Holtville, sea level, (bre); ditch 4.7 mi E of Bond’s Corner, (DRP); Hot Mineral, (LACM); irrigation canal 7.3 mi E of Holtville, (pre); irriga- tion canal at Winterhaven, (pre); /ryo Co., Hunter Cyn., Salino Valley, (LACM); Laws, Owens River, (ummz); Lone Pine, Owens River, (uMMZ); Los Angeles Co., Tanbark Flat, (cpFa); Mono Co., 5 mi N of Benton Station, 5300 ft., (UMMZ, LACM); Riverside Co. Blythe, (cc); Coachella Valley Preserve, NE of Thousand Palms, (J. Cole); Palm Springs, (LacmM); Whitewater River nr. Salton Sea, (DRP); San Bernardino Co., Colorado River at Moabi Rd., S of Needles, (Rw6); Lost Palm Cyn., Joshua Tree N. Mon., (Long Beach State Univ.); Parker Dam, (rwe); San Diego Co: T14S, RSE, (casc); San Felipe Creek, 13.8 mi E of Julian, 615 m, (pre); San Felipe Creek at Scissors Crossing, 11.9 mi E of Julian, 738 m, (DRP, RWG); Sentenac Cyn., (LACM); Vallecito, (Lacm); Yolo Co., Cache Creek at William H. ‘Bill’ Davis Memorial Picnic Area, by Calif. Hwy 16, 6 mi N of Rumsey, 132 m, (RWG); Nevapa: Clark Co., Logandale, (syuc); Elko Co., Carlin, Humboldt River, (ummz); 8.5 mi NW of Currie, (ummz); Humboldt Co., Can Spring, 3 mi S of Pahuhe Meadows, (ummz); Lincoln Co., 6 mi S of Alamo, (umMZ); Nye Co., Amargosa R., 2 mi below Beatty, (ummz); Pershing Co., Lovelock, Humboldt River, (ummz); Washoe Co., Pyramid Lake, Truckee River, (ummz); White Pine Co., 2 mi NW of Preston, (ummz); Texas: Brewster Co., Big Bend Nar! Park, 1 mi N of Rio Grande Village, (rw); Boquillas Cyn., Big Bend Natl. Park, (csuc); Rio Grande Village, Big Bend Natl. Park, (csuc); Maverick Co. Quemado, (LACM); Uran: Washington Co., St. George, (cc); Beaver Dam Wash, Terry Ranch, (Byuc); Beaver Dam Wash, Lytle Ranch, (Byuc); Gunlock, W of Veyo, (Byuc); Virgin River, Virgin, (Byuc); Mexico: Baja Calif: Norte. Guadalupe Hot Spgs., Guadalupe Cyn., Sierra Juarez, (csuc); Baja Calif. Sur. San José de Comandu, 10 Oct. 1923 (J. H. Williamson), 154 (all paratypes of E. coluber) (ummz, RWG, FSCA, USNM); Sonora: Hermosillo, (UMMZ). Erpetogomphus boa Selys (figs. 26-thx, 44, 45-abd, 68-hamules, 90-penis, 114, 115-app, 135-ept, 164-vertex, 167- postocciput, 195, 196-vl, 207-distr) Erpe togomphus boa Selys, 1859: 37 (11, sep.) (descr. d ‘Vera Cruz, Mexique. Par M. Salle. (Collect. Selys.)’). — Walsh 1863: 253 (mentioned); Selys 1873b: 519 (75, sep.) (list); 236 Calvert 1905: 165 (notes); Calvert 1907: 399 (possible identity with £. elaps); Muttkowski 1910: 86 (cat.); Ris 1917: 153 (notes and descr. of 2nd known d); Williamson and Williamson 1930: 11 (summary of stat- us); Paulson 1982: 255 (Mex.); Davies and Tobin 1985: 27 (cat.); Tsuda 1986: 87 (cat.); Bridges 1991: VII.31 (cat.); Tsuda 1991: 95 (cat.). Erpetogomphus crotalinus, nec Hagen in Selys, 1854. — Selys 1859: 537 (11, sep.) (9 descr. “Vera Cruz, Mexique. Par M. Salle. (Collect. Selys.)’ = E. crotalinus Hagen in Selys, 1858. Gomphus boa. — Hagen 1861: 100 (descr. from Selys). Herpetogomphus boa. — Walsh 1862: 389 (mentioned); Hagen 1875a: 42 (cat.); Selys 1879: 64 (2, sep.); Calvert 1899: 386 (list, English translation of Selys, 1879); Calvert 1908b: 693 (mentioned). Description Male. — Labrum, clypeus, and frons light blue gr- een, paler along sides of labrum and lateral lobes of postclypeus; base of mandibles pale green, tips black; base of antefrons anterior to median ocellus with nar- row, arcuate, V-shaped sulcus, shallower medially; vertex dark brown with usual postocellar tubercles, occiput tumid medially, pale green brown, its posteri- or margin ciliated, transverse to slightly concave, slightly emarginate in the middle, or more rarely smoothly curved, postocciput green, tumid; rear of head red brown. Prothorax red brown, synthorax (fig. 26) predomi- nantly green, area around middorsal carina washed with brown, often appearing as a faded middorsal stripe; antehumeral and humeral stripes dark brown, united into one large stripe; most of this stripe occu- pying posterior 0.30 of mesepisternum, metastigma black, a wash of brown between metepisternum and metepimeron often forming an obscure, narrow tho- racic stripe, subalar carina darkened with brown, me- sinfraepisternum and metinfraepisternum brown; metasternum green-brown. Coxae pale grey- brown with some green on exterior surfaces, trochanters grey brown, femora tawny at base, becoming dark brown distally, tibiae and tarsi black. Venational statistics (n = 13, including lectotype d). Fifth antenodal (occasionally fourth, sixth, or seventh) thickened in all wings; marginal cells behind fore wing paranal cells: 0- 4/0-4; anx: fore wing 13- 15/13-16, hind wing 9-11/10-12; pnx: fore wing 9- 12/9-11, hind wing 10-13/10-13; cs under pterostig- ma: fore wing 5-7/5-7, hind wing 5-7/5-7; anal triangular cells: 4-5. Hind wing 33-34 mm. Abdomen (fig. 44). Segment 1 brown dorsally, be- coming green laterally; segment 2 with blue green middorsal stripe, with dorsolateral stripe of brown surrounding green auricle, lateral margin near ha- mules pale greenish white; segment 3 with green mid- dorsal stripe bounded on sides by dorsolateral stripes of brown, lateral areas of tergites pale green becoming white posteriorly, anterior transverse carina edged with black expanding laterally interrupting white lat- eral areas of segment, thus forming two pale spots, posterior transverse carina also black, expanding later- ally so that posterior 0.2 of segment is black; segments 4-6 similar to segment 3, but with middorsal brown stripe narrowed anteriorly so that whitish green side almost touches middorsal green; segment 7 with ante- rior 0.5 dull white, conspicuously traversed by black anterior transverse carina, posterior 0.5 of segment red brown; segments 8-10 red brown with ventral and posterior borders of each segment black. Appendages red brown with apices becoming black. Abdomen 37- 40 mm. Cercus (fig. 115) nearly straight, swollen anterior- ly, apical 0.5 abruptly concave dorsally and covered with thick series of strong black bristles. Epiprocts about 0.5 the length of superiors, strongly curved dorsally; tips slightly divergent, each a thick, blunt point (fig. 135). Accessory genitalia (fig. 68). Anterior hamule black, glabrous, deeply forked with posterior branch slightly longer than anterior, the whole structure re- sembling a talon; anterior branch of hamule with a small semi-oval planar surface on outer side, this area covered with small hairs. Posterior hamule pale green or white, becoming brown near tip; in lateral view, tip obtusely pointed with posterior margin gently con- vex. Penis (fig. 90) with lateral lobes poorly developed posteriorly, almost circular, its margin serrated. Female. — Overall colouration as in male, but with pale areas on abdomen and legs more extensive. On abdomen (fig. 45), brown on terminal segments ex- tending anteriorly so that green middorsal and dorso- lateral brown stripes become ill-defined. Lateral mar- gins of abdominal segments mostly white and similar to male except that white extends posteriorly to seg- ment 8. abdomen 37-40 mm. Vertex with slightly tumid, oval areas posterior to lateral ocelli; posterior margin of occiput (fig. 164) el- evated and nearly straight; postocciput green, tumid medially (fig. 167). Venational statistics. Fifth antenodal (occasionally fourth, sixth, or seventh) thickened in all wings; anx: fore wing 13-16/13-16, hind wing 10-12/10-11; pnx: fore wing 9-13/9-12, hind wing 10-13/10-14; cs under pt: fore wing 5-7/5-8, hind wing 5-8/5-8. Hind wing 34-37 mm. Vulvar lamina (fig. 195) membranous, diagonally corrugated, outer margin of plate with strong costate ridge, area immediately mesal to ridge furrow-like, with a smaller tumid area occupying the center, medi- al area tumid, posterior margin of each plate a thin, membranous ridge, postlamellar ridge long, greatly surpassing hind margin of lamina, its Y-shaped junc- ture at distal 0.30 of sternum. Abdomen 37-40 mm. GARRISON: Revision of Erpetogomphus Diagnosis Erpetogomphus boa is most similar to £. cophias, but is easily distinguished by the thoracic pattern and the male caudal appendages. A well-defined combined antehumeral and humeral stripe is present in £. boa, but no such stripe is present in E. cophias. The supe- rior appendages of E. cophias possess a large ventral ridge which ends in a prominent tooth 0.30 the length of the appendage (fig. 116). This structure is lacking in E. boa (fig. 115). Dorsally, the apical 0.30 of the superior appendages of E. cophias contains a few strong bristles, not as many or as stout as in £. boa. The females of these two species are similar mor- phologically, but may be distinguished by the thorac- ic pattern. The postocciput of E. boa (fig. 164) does not have the posterior depressions on each side found in E. cophias (fig. 165). Females of £. boa superficially resemble females of E. viperinus, but the latter have a pair of circular de- pressions on abdominal sternite 9 posterior to the juncture of the postlamellar ridge which are absent in E. boa. The vulvar lamina of E. viperinus (fig. 184) is totally unlike that of E. boa (figs. 195-196). Remarks This species has been known from only two speci- mens: Selys’ type and an incomplete male described by Ris (1917). The lectotype is a teneral male with the apical 0.5 of the cerci missing (fig. 114), but it was apparently complete at the time of the original description. The original description of the male is as follows (translated from the French): ‘Prerostigma light yellow brown. Head and thorax yellow, femora yellow, with an external brown stripe, short, the ante- rior four tarsi brownish black. ‘[male]. Superior appendages swollen at their base, with an obtuse tooth above at the end of the swelling; their points rounded, slightly bent inwards, hairy. Inferior appendage divided, branching a little dis- tance, attaining 1/2 length of the superior. Occiput nearly straight. Tibiae brown.’ Selys’ (1859) description of the female led Calvert (1905) to believe that it was really a specimen of E. crotalinus, and my examination of this female con- firms Calvert’s assertion. Erpetogomphus crotalinus ap- pears to be restricted to the Pacific side of Mexico, and I doubt that the locality (Vera Cruz’) is correct for this specimen. In the supplement to the Biologia (Calvert, 1907), Calvert received drawings of the in- complete abdominal appendages of the type which led him to remark that Æ boa seemed hardly different from E. elaps. Ris (1917) provided a history of the type male of E. boa and remarked (translated from the German): ‘I examined a long time ago in the Hamburg Museum an old male that was different 237 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 from elaps and corresponds with the description of boa, that its identity [with boa) seemed likely. The lo- cality is, unfortunately, uncertain, the label reads ‘Agua Caliente’ without further information. Its pres- ervation is fair, the tip of the inferior appendages are missing about the distal fourth as shown in the fig- ure...” Examination of the hamules of the lectotype veri- fies the identity of a series of Æ. boa all collected in southern Veracruz; and the Agua Caliente specimen of Ris is also Æ boa. Calvert (1905) postulated that £. sipedon might be the female of £. boa. Although I have not seen a pair of E. boa in copula or in tandem, I am confident in as- cribing these females to Æ. boa. The similarity of the vulvar laminae in £. boa and E. cophias (whose males are closely related) further strengthens my conviction that Æ. sipedon does not represent the female of E. boa. López (in litt.) has observed tandem pairs, con- firming the female sex of this restricted species. The male epiprocts, which are robust and form a sharp point, are apparently responsible for structural damage to the vertex of many females during mating. Of the 26 females I examined, 14 (54%) had two holes medially to the postoccipital tubercles. Some of these specimens had encrusted haemolymph sur- rounding the wounds. Dunkle (1984, 1991) reviews cases of traumatic mating among other anisopteran Odonata, but their negative effect, if any, is un- known. Of the remaining 12 females, which were un- injured, 5 were teneral and had not yet mated. Venational details of lectotype male: fifth anteno- dal thickened in fore wings, fourth in hind wings; marginal cells behind fore wing paranal cells: 12/2; anx: fore wing 14/14, hind wing 10/10; pnx: fore wing 10/10, hind wing 10/11; cs under pterostigma: fore wing 5/6, hind wing 5/5; anal triangular cells: 4. Hind wing 34 mm. Biology. — Most specimens were obtained at a city park near Jalapa. Individuals were found by flushing them from low vegetation about 50 m from a small stream. Adults spent most of their time resting and were difficult to detect among the green foliage. Its habits are similar to those reported for E. viperinus. Adults prefer parts of small, low volume sandy-bot- tomed streams where banks are often covered by trees. Reproductive maturation continues through July. Copulation apparently occurs primarily in open areas near streams. Ensuing pairs fly some distance from these sites. Some males settle at the edge of the stream, but passing females are unreceptive there. Females, like other members of this genus, oviposit unattended, flying over the water, stopping briefly to drop eggs from a height of about 40 cm. Löpez has observed E. boa captured by spiders (Araneidae) and robber flies (Asilidae). 238 Collection dates are July and August. Altitudes of capture range from 1150 m (Orizaba) to 1300 m (Jalapa). Distribution (fig. 207). — Erpetogomphus boa has been taken only in central Veracruz between 18°50’N and 19°30’N along the eastern foothills of the Sierra Madre Oriental. Material Type data. — Lectotype male by present designa- tion with following data: small green label “Vera Cruz/Salle’, yellow rectangular label: ‘ERPET. BOA / 3’ [this last label probably added by Selys during or after 1879], all in Selys’ hand; two pencilled labels on red tags: ‘No. 100’; and two pencilled white labels: ‘28’, all in an unknown hand. The caudal appendages and accessory genitalia are shown in fig. 114. The female specimen originally described as the female of £. boa is actually E. crotalinus and possesses the following la- bels: small green label: “Vera Cruz/Salle’; and white label: “Herp. boa S. / 2’, both in Selys’ hand. Both specimens in IRSN. Other material (136, including lectotype d; 262). — Mexico: Veracruz. Parque Javier Clavijero, Jalapa, 1300 m, 21 June 1980 (Raul López), 16 (RWG); (same data), 11 Aug. 1980 (A. Garcés), 1d (UNAM); (same data), 13 Aug. 1980 (R. López), 1 d (UNAM); (same data), 16 Aug. 1980 (A. Garcés), 16 (UNAM); (same data), 20 Aug. 1980, 19 (unam); (same data), 5 Aug. 1981, (R. López), 25,6? (RwG); (same data), 6 Aug. 1981, 1d (UNAM); (same data), 1 Aug. 1982, 3d, 19 (RwG); (same data), 9 Aug. 1982, 1® (UNAM); (same data), 19 Aug. 1982, 3? (UNAM); (same data), Bosque Mesofilo de Montana, 12 June 1981 (R. López), 19 (UNAM); Barranca de Cayoapa, Teocelo, (no collector), 12 (rwG); 4.8 mi N of Coscomatopec, (M. A. Ortiz, O. S. Flint), 1d, 29 USNM); 2.5 mi S of Huatusco, 23 July 1966 (M. A. Ortiz, ©. S. Flint), (USNM); 4.7 mi N of Huatusco, Puente Ruiz Cortines, 31 July 1966 (M. A. Ortiz, O. S. Flint), 3? (usnm); Orizaba, Ojo de Agua, 3800 ft. (D. R. Paulson), 18,5? (DRP, FSCA). Erpetogomphus cophias Selys (figs. 27-thx, 46-abd, 69-hamules, 91-penis, 116- app, 165-vertex, 197-vl, 207-distr) Erpetogomphus cophias Selys, 1858: 332 (72 sep.) (descr. d ‘Le Mexique, d’après un mâle du Museum de Paris’). — Selys 1859: 537 (11 sep.) (descr. d ); Selys 1869: 175 (12 sep.) (descr. 9); Selys 1873b: 519 (75 sep.) (list); Calvert 1899: 386 (mentioned); Calvert 1905: 164 (descr. 3, 2); Calvert 1907: 398 (mention of new figure); Calvert 1909: 481 (seasonal distr.); Muttkowski 1910: 86 (cat.); Williamson and Williamson 1930: 12 (summary of stat- us); Montgomery 1973: 239 (comment on name); Paulson 1982: 255 (Mex.); Davies and Tobin 1985: 27 (cat.); Tsuda 1986: 87 (cat.); Bridges 1991: VII.52 (cat.); Tsuda 1991: 95 (cat.). Gomphus cophias. — Hagen 1861: 100 (descr. ). Herpetogomphus cophias. — Walsh 1862: 389 (mentioned); Hagen 1875a: 42 (cat.); Selys 1879: 64 (2 sep.) (charac- ters of genus); Kirby 1890: 60 (cat). Description Neotype male. — Entire face pale green; base of mandibles pale green, tips black; antefrons pale green with wash of brown at base; vertex brown, with some green on tubercle behind each lateral ocellus; two prominent cone-shaped pits anterolateral to median ocellus; scape and pedicel dark brown, flagella mis- sing; occiput broad, its dorsal surface evenly convex, especially medially, its posterior margin slightly arcu- ate; crest with long brown hairs; postocciput light gr- een, evenly concave, rear of head red brown. Prothorax primarily brown, with light green medi- ally on anterior, median, and posterior lobes; synthor- ax entirely pale green with wash of brown ventrally above coxae. Coxae and trochanters pale grey green, femora pale grey green becoming tawny distally, a dark brown streak on lateroextensor surfaces of femo- ra occupying almost all of profemora, about 0.60 of mesofemora, and distal 0.30 of metafemora; tibiae, tarsi, and armature black. Wings hyaline, venation dark brown basally, black distally; anterior margin of costa pale yellow to proximal end of pterostigma; pte- rostigma brown, veins bordering it black. Venational details. Fifth antenodal thickened in all wings; no marginal cells behind fore wing paranal cells; anx: fore wing 13/13, hind wing 9/9; pnx: fore wing 7/7, hind wing 7/9; cs under pterostigma: fore wing 5/5, hind wing 5/4; anal triangular cells: 4 . Hind wing 29 mm. Abdomen predominantly pale with following dark brown markings: vestige of midlateral stripe on seg- ment 1; this stripe more defined on segment 2, espe- cially behind auricle; annulus; midlateral stripe on seg- ment 3, transverse carina and darker annulus; segments 4-6 similar to segment 3 but stripes darker, each beginning at a little beyond beginning of seg- ment, enlarging at transverse carina, narrowing and then widening, especially posteriorly, but not touch- ing dorsally; segment 7 with transverse carina and pos- terior 0.40 of segment laterally; segment 8 pale yellow brown with ill-defined dorsolateral dark stripe; seg- ment 9 similar to segment 8, but with a longitudinal lateral middorsal spot on posterior 0.75 of segment; segment 10 pale ocher with dark red brown along an- terior 0.20 of segment. Pale areas of segments 1-7 pri- marily pale olive dorsally and ventrally; pale areas of segments 8-10 red brown dorsally and ventrally. Abdomen 35 mm. Cercus yellow brown, linear, slightly concave dor- sally beyond basal 0.30, tip smoothly rounded, ventral GARRISON: Revision of Erpetogomphus margin with inferior carina at basal 0.30, terminating in a distinct ventral tooth; dorsal concave area of cerci with scattered thick bristles. Epiprocts yellow brown, distal 0.50 curved dorsally at 90°, as is typical of ge- nus, tips slightly divergent, each with a thick, blunt point. Accessory genitalia. Anterior hamule black, gla- brous, deeply forked, with posterior branch slightly longer than anterior, the whole structure resembling a talon; distal 0.20 of anterior branch with a longitudi- nal, obtuse V-shaped area. Posterior hamule spatulate, pale green, becoming brown near tip; in lateral view, tip obtusely pointed with a small, blunt cephalad di- rected tooth on rear margin. Peduncle of penis with prominent rounded foliate erect lateral lobes, its pos- terior margin not prominent, forming a gentle bilobed area; penis with lateral lobes poorly developed posteri- orly, forming a serrated semicircle; membranous hood not overlapping; cornuae well developed, parallel sid- ed, their tips evenly rounded. Female. — Overall colouration as in male, head with prominent conical pits anterolateral to median ocellus as in male; vertex with tumid oval area posterior to lat- eral ocelli; occiput narrow, crest prominent, straight to slightly sinuate along medial 0.30, lateral arms bent posteriorly; prominent transverse postoccipital pit (fig. 165) immediately behind lateral arms of crest, medial area of postocciput tumid. Pro- and synthorax as in male, femora with brown areas reduced with little brown on extensor surface of metafemora. Venational statistics (n = 2): number of marginal cells behind fore wing paranal cells: 2-4/1-3; anx: fore wing 13/13, hind wing 9-11/9-10; pnx: fore wing 9/8- 9, hind wing 9-10/10; cs under pterostigma: fore wing 5/4-5, hind wing 4-6/4-5. Hind wing 32-34 mm. Abdomen with brown dorsolateral stripe more re- duced than male, more prominent and expanded around lateral carinae; segments 8-9 tawny, cercus pale ocher. Vulvar lamina (fig. 197) membranous, di- agonally corrugated; posterior margin of each plate gently curved, a prominent ridge along its border, a prominent long diagonal depression immediately an- terior to posterior margin; plates meeting medially at tumid area; cleft between plates a small notch; junc- ture of Y-shaped postlamellar ridge well posterior to vulvar laminar plates; area on each side of central stem of postlamellar ridge darkened, with a slight de- pression. Abdomen 35-37 mm. Diagnosis Erpetogomphus cophias is most similar to E. boa and is diagnosed under that species. The characteristic ventral tooth of the cercus in the male and the post- occipital depressions in the female are autapomorphic characters. 239 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Remarks This is apparently a rare species. Williamson and Williamson (1930) stated that there were only 10 known specimens. Only one other specimen from Michoacan, collected in 1941, has come to my atten- tion. I have examined only 3 males and 4 females, of which one male and one female are teneral and in poor condition. I can see no noticeable differences among the few specimens examined. Variation. — Venational details of one additional male: number of marginal cells behind fore wing par- anal cells: 3/2; anx: fore wing 11/12, hind wing 9/8; pnx: fore wing 9/8, hind wing 9/10; cs under pteros- tigma: fore wing 5/5, hind wing 5/5; number of anal triangular cells 3/3. Hind wing 30 mm. Biology. — Nothing is known of the biology of this species. Distribution (fig. 207). — Like E. boa, E. cophias has a restricted distribution and apparently replaces E. boa in the highlands of west-central Mexico south of 20°N and west of 100°W. Records indicate an ele- vational gradient of 1525 m (Cuernavaca) to 2438 m (Omilteme) and flight during June and July. Material Type data. According to Dr. J. Legrand ( in itt. 11 May 1984), the holotype male from ‘Le Mexique, d’apres un mäle du Museum de Paris’, is missing, the- re remaining only a large rectangular green label with [handwritten] “E. cophias, Selys’ which was originally attached to the specimen. Dr. Legrand was kind enough to send a male of which he states, … we have from ex Martin's collection a specimen called O. cop- hias Selys (determined by Selys himself, according to Martin...), which he suggested I designate as neoty- pe. This I now do. The neotype male contains the fol- lowing data: white label in R. Martin's hand: ‘Ophiogomphus | cophias Selys/ Det. De Selys/ Mexique’, small green label printed: “MUSEUM PARIS / Coll. r. MARTIN 1920’, and red rectangular label handwritten: “Erpetogomphus | cophias Selys à / NEO- TYPE / des./ R. W. Garrison 1986’. The abdomen is detached and is in a triangular envelope pinned bene- ath the specimen. Other material (3d, 49, including neotype 4). — Mexico: Guerrero: Omilteme, 8000 ft., July 1888 (H. H. Smith), 12, (BMNH); Michoacan: Tancitaro, 6000 ft. (1846 m), 4th Hoogstraal Mexican Biological Expedition, 28 July 1941 (H. Hoogstraal), 1d (ummz); Morelos. Cuernavaca, June 1897 (O. W. Barrett), 16, 19 (anse); June 1888 (H. H. Smith), 16 (BMNH); 8 July 1900 (C.C. Deam), 29 (ummz). ACKNOWLEDGEMENTS Completion of this work would have been impossible 240 without the help of the curators of many institutions and many individuals who placed many specimens at my dispo- sal. I thank Donald Azuma of the Academy of Natural Sciences, Philadelphia (ansp); Stephen Brooks, British Museum (Natural History) (BMNH); Dr. Paul Dessart, Institut Royal des Sciences Naturelles de Belgique, Brussels (irsn); Dr. David Furth, Museum of Comparative Zoology, Harvard University (mczc); Dr. Jean Legrand, Muséum National d’Histoire Naturelle, Paris (MNHP); Mr. Mark O’Brien and Mrs. Leonora Gloyd, University of Michigan, Museum of Zoology, Ann Arbor (ummz); Dr. Oliver S. Flint, Jr, National Museum of Natural History, Smithsonian Institution, Washington, D. C. (usNm); Prof. Dr. Kurt Günther, Zoologisches Museum, Humboldt Universität, Berlin (zMHB), for allowing me the privilege of examining type material in their great institutions. Dr. Jean Belle, Velp, The Netherlands, kindly interceded on my be- half in obtaining the syntypes of Erpetogomphus constrictor from the Senckenberg Museum, Frankfurt-am-Main (SMF) so that I could designate a lectotype. Examination of these specimens allowed me to correct errors in earlier descrip- tions and to make correct association of at least one species with its name. The following individuals provided specimens, records, and other information which added considerably to this paper: Frank Carle, Warren, NJ, for material he borrowed from the American Museum of Natural History, NY (AMNH); Donald Azuma, Academy of Natural Sciences, Philadelphia (Ansp); Stephen Brooks, British Museum (Natural History); Dr. Richard Baumann, Brigham Young University, Provo, UT (Byuc); Dr. Paul Arnaud, California Academy of Sciences, San Francisco (cAsc); Carl Cook, Center, KY (cc); Mrs. Tineke Boomsma, Orange Walk, Belize (TB); Jerrell Daigle, Tallahasse (jp); Dr. Alan Hardy, California Department of Food and Agriculture, Sacramento (cpra); M. en C. Carlos Esquivel, San José, Costa Rica (ce); Dr. Clifford Johnson, University of Florida, Gainesville (cj); Dr. Boris Kondratieff, Colorado State University, Fort Collins (csuc); Dr. James Liebherr, Cornell University, Ithaca, NY (curc); Dr. D. A. L. Davies, Cambridge, United Kingdom (parp); Dr. Dennis Paulson, Burke, Washington (pre); Dr. Minter J. Westfall, Jr. and William Mauffray (wm), Florida State Collection of Arthropods (rsca) and International Odonata Research Institute (ort), University of Florida, Gainesville; Dr. Jean Belle, Velp, The Netherlands (JB) [collection now in National Museum of Natural History, Leiden, The Netherlands (RMNH)]; Dr. Ken Tennessen, Florence, AL (KT); the late Dr. Charles Hogue, Los Angeles Co. Natural History Museum, Los Angeles (LACM); Prof. Dr. Paul Dessart, Institut Royal des Sciences Naturelles, Brussels; Jeffrey Cole, Los Angeles; Dr. David Furth, Museum of Comparative Zoology, Harvard University; Dr. Michael May, Rutgers University, NJ; Dr. Jean Legrand, Muséum National d’Histoire Naturelle, Paris; Paul Miliotis, Westford, MA (psm); Philippe Machet, St.-Cloud, France (PM); Larry Muller, El Segundo, CA; Dr. Sidney Dunkle, Collin County Community College, Plano, TX (swp); Dr. Thomas W. Donnelly, State University of New York, Binghamton (rwp); Mark O’Brien and Leonora Gloyd, University of Michigan, Museum of Zoology, Ann Arbor; M. en C. Enrique González S. and Lic. Raul Armando López P., Universidad Nacional Autónoma de México, Mexico City (IBUNAM) and Museo Zoologica ‘Alfonso Herrera’, Facultad Ciencias (UNAM); M. en C. Rodolfo Novelo Gutierrez, Instituto de Ecologia, Xalapa, Veracruz, Mexico; Dr. Oliver S. Flint, Jr., United States National Museum, Smithsonian Institution, Washington, D. C.; Victor Hellebuyck, Sherbrooke, Quebec (vH); Prof. Dr. Kurt Günther, Zoologisches Museum, Humboldt Universität, Berlin, Clark Shiffer, State College, PA; Dick Swinney, Glendora, CA. Specimens in my collection are la- belled rwe. Versions of this paper benefited from critical readings by Drs. J. Belle, T. W. Donnelly, S. W. Dunkle, M. en C. Enrique González S., and Lic. Rodolfo Novelo G. Lic. Raul López P., and Lic. Rodolfo Novelo G. of Universidad Nacional Autónoma de México very kindly made available to me notes on the biology and reproductive behaviour of certain Mexican Erpetogomphus Lic. Alonso Ramirez, Universidad Nacional, Heredia, Costa Rica, supplied me with records of E. constrictor from Costa Rica; and Dr. Jerry Louton, United States National Museum, Smithsonian Institution, was helpful in reviewing the cladistic analysis. I also benefited from discussions on the philosophy of cladis- tics from Don Frack, Covina, CA. My wife, Jo, typed the paper and has given freely of edi- torial advice, encouragement, and help in making time for me to work on this project. Dr. Jan van Tol, National Museum of Natural History, Leiden, The Netherlands (RMNH) provided editorial advice and supplemented my cladistic analysis and interpretation as this paper was going to press. To all these persons, I express my sincere thanks and appreciation for all the help they have rendered me. REFERENCES Ackery, P.R. & R. I. Vane-Wright, 1984. Milkweed butter- flies their cladistics and biology being an account of the natural history of the Danainae, a subfamily of the Lepidoptera, Nymphalidae. — Cornell University Press, Ithaca: ix + 425. Adams, C. C., 1900. Odonata from Arkansas. — Entomological News 11: 621-622. Ahrens, C., 1938. A list of dragonflies taken during the sum- mer of 1936 in western United States (Odonata). — Entomological News 49: 9-16. Alrutz, R. W., 1961. Notes and records of Ohio dragonflies and damselflies (Odonata). — Ohio Journal of Science 61 (1): 13-24. Ander, K., 1929. Über die Nymphe von Mesogomphus Hageni Selys (Odonata). — Konowia 8 (2): 159-162. Banks, N., 1892. A synopsis, catalogue, and bibliography of the neuropteroid insects of temperate North America. — Transactions of the American Entomological Society 19: 327-373. Belle, J., 1972. An unknown gomphid larva from Surinam, possibly Progomphus geijskesi Needham, 1944 (Odonata). — Odonatologica 1 (2): 113-116. Belle, J., 1988. A synopsis of the species of Phyllocycla Calvert, with descriptions of four new taxa and a key to the genera of neotropical Gomphidae (Odonata, Gomphidae). — Tijdschrift voor Entomologie 131: 73- 102. Belle, J., 1992. Studies on ultimate instar larvae of neotrop- ical Gomphidae, with the description of Tibiagomphus gen. nov. (Anisoptera). — Odonatologica 21 (1): 1-24. Belle, J. & D. Quintero, 1992. Chapter 6. Clubtail dragon- flies of Panama (Odonata: Anisoptera: Gomphidae), pp. 91-101. — In D. Quintero & A. Aiello, ed. Insects of GARRISON: Revision of Erpetogomphus Panama and Mesoamerica: selected studies. Oxford University Press: xxii + 692 pp. Bick, G. H., 1951. Notes on Oklahoma dragonflies. — Journal of the Tennessee Academy of Science 26: 178- 180. Bick, G. H., 1957. The Odonata of Louisiana. — Tulane Studies in Zoology 5 (5): 71-135. Bick, G. H., 1959. Additional dragonflies (Odonata) from Arkansas. — The Southwestern Naturalist 4 (3): 131-133. Bick, G. H., 1990. Unpublished records in Florida State Collection of Arthropods (rsca). — Argia, the News Journal of the Dragonfly Society of America 2 (1-4): 3-4. Bick, G. H. & J. C. Bick, 1957. The Odonata of Oklahoma. — The Southwestern Naturalist 2 (1): 1-18. Bick, G. H. & J. C. Bick, 1958. The ecology of the Odonata at a small creek in southern Oklahoma. — Journal of the Tennessee Academy of Science 33 (3): 240-251. Bick, G. H., J. C. Bick and L. E. Hornuff, 1977. An anno- tated list of the Odonata of the Dakotas. — Florida Entomologist 60 (3): 149-166. Bird, R. D., 1932. Dragonflies of Oklahoma. — Publications of the University of Oklahoma Biological Survey 4 (1-2): 50-57. Borror, D. J., 1935. New records of Ohio dragonflies (Odonata). — Ohio Journal of Science 35 (6): 451-456. Borror, D. J., 1937. An annotated list of the dragonflies (Odonata) of Ohio. — Ohio Journal of Science 37 (3): 185- 196. Borror, D. J., 1963. Common names for Odonata. — Proceedings North Central Branch, Entomological Society of America 18: 104-107. Brauer, F., 1868. Verzeichniss der bis jetzt bekannten Neuropteren im Sinne Linne’s. — Verhandlungen der zoologisch-botanischen Gesellschaft in Wien 18: 359- 416, 711-742. Bridges, C. A., 1991. Catalogue of the family-group, genus- group and species-group names of the Odonata of the world. — Available from the author, 502 W. Main St., Urbana, Illinois, U. S. A., 61801: xiv + 704 pp. Byers, C. F., 1928. The unknown nymphs of North American Odonata. — Canadian Entomologist 60: 4-6. Byers, C. F., 1930. A contribution to the knowledge of Florida Odonata. — University of Florida Publication. Biological Science Series 1 (1): 1-137. Byers, C. F., 1931. Dixie dragonflies collected during the summer of 1930 (Odonata). — Entomological News 42: 113-119. Byers, C. F., 1939. A study of the dragonflies of the genus Progomphus (Gomphoides) with a description of a new species. — Proceedings of the Florida Academy of Sciences 4: 19-85. Cabot, L., 1872. The immature state of the Odonata. Part I. — Subfamily Gomphina. — Illustrated Catalog of the Museum of Comparative Zoology, at Harvard College 5: 1-17. Calvert, A. S. & P. P. Calvert, 1917. A year of Costa Rican natural history. — Macmillan Co., New York: xix + 577 pp. Calvert, P. P., 1895. The Odonata of Baja California, Mexico. — Proceedings of the California Academy of Sciences (2) 4: 463-558. Calvert, P. P., 1899. Odonata from Tepic, Mexico, with supplementary notes on those of Baja, California. — Proceedings of the California Academy of Sciences (3) 1: 371-418. Calvert, P. P., 1905. Odonata, in Biologia Centrali- TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Americana: Insecta Neuroptera. — R. H. Porter and Dulau Co., London: pp. 145-212. Calvert, P. P., 1907. Odonata, in Biologia Centrali Americana: Insecta Neuroptera, — R. H. Porter and Dulau Co., London: pp. 309-404. Calvert, P. P., 1908a. List of Odonata taken by Dr. Henry Skinner in Carr Canyon, Huachuca Mountains, Arizona. — Entomological News 19: 45. Calvert, P. P., 1908b. The present state of our knowledge of the Odonata of Mexico and Central America. — Science 28 (724): 692-695. Calvert, P. P., 1908c. Introduction to the Odonata, in Biologia Centrali-Americana: Insecta Neuroptera. — R. H. Porter and Dulau Co., London: pp. i-xxx. Calvert, P. P., 1909. The composition and ecological rela- tions of the odonate fauna of Mexico and Central America. — Proceedings of the Academy of Natural Sciences of Philadelphia 60: 460-491. Calvert, P. P., 1912a. Studies on Costa Rican Odonata. IV. Erpetogomphus in Costa Rica, with descriptions of a new species having complex structural mating adaptations. — Entomological News 23 (7): 289-295. Calvert, P. P., 1912b. [Note on Erpetogomphus tristani.]. — Entomological News 23: 384. Calvert, P. P., 1919. Odonata Anisoptera from Guatemala. — Entomological News 30: 72-78. Calvert, P. P., 1920a. [Note on Guatemalan Odonata]. — Entomological News 31: 113. Calvert, P. P., 1920b. The Costa Rican species of Epigomphus and their mutual adaptations (Odonata). — Transactions of the American Entomological Society 46: 323-354. Calvert, P. P., 1942. Increase in knowledge of the odonate fauna of Mexico, Central America, and the West Indies since 1908. — Proceedings: Eighth American Scientific Congress, Biological Sciences: Zoology 3: 323-331. Calvert, P. P., 1947. The Odonate collections of the California Academy of Sciences from Baja California and Tepic, Mexico, of 1889-1894. — Proceedings of the California Academy of Sciences (4) 23: 603-609. Cannings, R. & R. W. Garrison, 1991. Sympetrum signifer- um, a new species of dragonfly (Odonata: Libellulidae) from western Mexico and Arizona. — Annals of the Entomological Society of America 84 (5): 474-479. Carle, F. L., 1979. Environmental monitoring potential of the Odonata, with a list of rare and endangered Anisoptera of Virginia, United States. - Odonatologica 8 (4): 319-323. Carle, F. L., 1982. Ophiogomphus incurvatus: a new name for Ophiogomphus carolinuss Hagen (Odonata: Gomphidae). — Annals of the Entomological Society of America 75 (3): 335-339. arle, F. L., 1986. The classification, phylogeny and bio- geography of the Gomphidae (Anisoptera). 1. Classification. — Odonatologica 15 (3): 275-326. Zarle, F. L., 1992. Ophiogomphus (Ophionurus) australis spec. nov. from the gulf coast of Louisiana, with larval and adult keys to American Ophiogomphus (Anisoptera: Gomphidae). — Odonatologica 21 (2): 141-152. arle, F. L. & C. Cook, 1984. A new Neogomphus from South America, with extended comments on the phylog- eny and biogeography of the Octogomphini trib. nov. (Anisoptera: Gomphidae). — Odonatologica 13 (1): 55- 70. ZA AA Ps FA varpenter, G. H., 1897. The geographical distribution of bo "N ho dragonflies. — Scientific Proceedings of the Royal Dublin Society 8 (N. S.) (5) 55: 439-468. Chao, H-f., 1984. Reclassification of Chinese gomphid dragonflies, with the establishment of a new subfamily and the descriptions of a new genus and species (Anisoptera: Gomphidae). — Odonatologica 13 (1): 71- 80. Cowley, J. C., 1934. Notes on some generic names of Odonata. — Entomologists Monthly Magazine 70: 240- 247. Cowley, J. C., 1937. The pagination of the reprints of the Selysian monographs and synopses of Odonata. — Journal of the Society for the Bibliography of Natural History 1 (3): 73-81. Cross, W. H., 1955. Anisopteran Odonata of the Savannah River Plant, South Carolina. — Journal of the Elisha Mitchell Scientific Society 71 (1): 9-17. Cross, W. H., 1956. Dragonflies in the Tallahassee region. — Florida Entomologist 39 (1): 9-16. Cruden, R. W., 1964. Notes on Brechmorhoga mendax (Hagen): Odonata. — Entomological News 75 (3): 79-82. Currie, R. P., 1903. The Odonata collected by Messrs. Schwarz and Barber in Arizona and New Mexico. — Proceedings of the Entomological Society of Washington 5 (4): 298-303. Davies, D. A. L. & P. Tobin, 1985. The dragonflies of the world: a systematic list of the extant species of Odonata. Vol. 2 Anisoptera. — Societas Internationalis Odona- tologica Rapid Communications (Suppl.) 5: xi + 151 pp. De Marmels, J., 1990. An updated checklist of the Odonata of Venezuela. — Odonatologica 19 (4): 333-345. Donnelly, T. W., 1961. The Odonata of Washington, D. C., and vicinity. — Proceedings of the Entomological Society of Washington 63 (1): 1-13. Donnelly, T. W., 1992. Chapter 5. The Odonata of Central Panama and their position in the neotropical odonate fauna, with a checklist, and descriptions of new species, pp. 52-90. In D. Quintero and A. Aiello, ed. Insects of Panama and Mesoamerica: selected studies. — Oxford University Press: xxii + 692 pp. Dunkle, S. W., 1984. Head damage due to mating in Ophiogomphus dragonflies (Anisoptera: Gomphidae). — Notulae Odonatologicae 2 (4): 63-64. Dunkle, S. W., 1988. A list of the Odonata of Honduras. — Ceiba 29 (1): 41-49. Dunkle, S. W., 1991. Head damage from mating attempts in dragonflies (Odonata: Anisoptera). — Entomological News 102 (1): 37-41. Dunkle, S. W., 1992. Distribution of dragonflies and dam- selflies (Odonata) in Florida. — Bulletin of American Odonatology 1 (2): 29-50. Dunkle, S. W. & J. J. Belwood, 1982. Bat predation on Odonata. — Odonatologica 11 (3): 225-229. Dunkle, S. W. & M. J. Westfall, Jr., 1982. Order Odonata, pp- 32-45. InR. Franz, ed. Invertebrates, vol. 6. Rare and endangered biota of Florida. — University Presses of Florida: xx + 131 pp. Farris, J. S., 1988. HENNIG86 version 1.5 [Phylogenetic pro- gram for PC’s]. — Port Jefferson, NY. Ferguson, A., 1940. A preliminary list of the Odonata of Dallas County Texas. — Field and Laboratory. Journal of the Graduate Research Center, Southern Methodist University 8 (1): 1-10. Ferguson, A., 1942. Scattered records of Texas and Louisiana Odonata with additional notes on the Odonata of Dallas County. — Field and Laboratory. Journal of the Graduate Research Center, Southern Methodist University 10 (2): 145-149. Ferguson-Beatty, A., 1956. An inquiry into the significance of the larval proventriculus in the taxonomy of Odonata. — Proceedings Tenth International Congress of Entomology 1: 367-372. Förster, F., 1914. Beiträge zu den Gattungen und Arten der Libellen. III. — Archiv für Naturgeschichte (A) 80 (2): 59- 83. Fraser, F. C., 1940. A comparative study of the penes of the family Gomphidae (Order Odonata). — Transactions of the Royal Entomological Society of London 90 (20): 541-550. Garman, P., 1927. Guide to the insects of Connecticut. Part V. The Odonata or dragonflies of Connecticut. — Bulletin. Connecticut State Geological and Natural History Survey 39: 1-331. Garrison, R. W., 1986. In the wilds of Baja. — Selysia 15 (1): 15-17. Gloyd, L. K., 1958. The dragonfly fauna of the Big Bend re- gion of trans-Pecos, Texas. — Occasional Papers of the Museum of Zoology, University of Michigan 593: 1-23. Gloyd, L. K., 1963. A movable molar in the Odonata — Proceedings, North Central Branch Entomological Society of America 18: 147-149. Gonzälez-S., E. & R. Novelo-G., 1990. Dos nuevas especies de Phyllogomphoides Belle 1970 (Odonata: Gomphidae) del estado de Morelos, Mexico. — Folia Entomologica Mexicana 79: 33-43. Gonzalez-S., E. & R. Novelo-G., 1991. Odonata de la Reserva de la Biosfera Michilia, Durango, Mexico. Parte I. Imagos. — Folia Entomologica Mexicana 81: 67-105. Hagen, H. A.., 1861. A synopsis of the Neuroptera of North America. Smithsonian Miscellaneous Collections, Washington, D. C.: xx + 347 pp. Hagen, H. A., 1873. Odonata from the Yellowstone. /n F. V. Hayden. Sixth annual report of the United States Geological Survey of the Territories, embracing portions of Montana, Idaho, Wyoming, and Utah; being a report of progress of the explorations for the year 1872. — Report ofthe United States Geological Survey of the Territories 6: 727-729. Hagen, H. A., 1874. Report on the Pseudo-neuroptera col- lected by Lieut. W. L. Carpenter in 1873 in Colorado. Jn F. V. Hayden. Annual report of the United States Geological and Geographical Survey of the Territories, embracing Colorado, being a report of progress of the ex- ploration for the year 1873. — Report of the United States Geological Survey of the Territories 7: 571-606. Hagen, H. A., 1875a. Synopsis of the Odonata of America. — Proceedings of the Boston Society of Natural History 18: 20-96. Hagen, H. A., 1875b. Chapter 14. Report upon the collec- tions of Neuroptera and Pseudo-neuroptera, made in portions of Colorado, New Mexico, and Arizona during the years 1872, 1873, and 1874. In Volume 5. — Zoology. Report upon geographical and geological explo- rations and surveys west of the one hundredth meridian, in charge of First Lieut. Geo. M. Wheeler, corps of engi- neers, U.S. Army, under the direction of Brig. Gen. A. A. Humphreys, chief of engineers, U.S. Army, pp. 909-922. Hagen, H. A., 1885. Monograph of the earlier stages of the Odonata. Sub-families Gomphina and Cordulegastrina. — Transactions of the American Entomological Society GARRISON: Revision of Erpetogomphus 12: 249-291. Harp, G. L. & J. D. Rickett, 1977. The dragonflies (Anisoptera) of Arkansas. — Arkansas Academy of Science Proceedings 31: 50-54. Harp, G. L. & J. D. Rickett, 1985. Further distributional records for Arkansas Anisoptera. — Arkansas Academy of Science Proceedings 39: 131-135. Higgins, H. T., 1901. The development and comparative structure of the gizzard in the Odonata Zygoptera. Proceedings of the Academy of Natural Sciences of Philadelphia 126-141. Hine, J. S., 1913. Additions and corrections to the Odonata of Ohio. — Ohio Naturalist 13 (5): 94-96. Huggins, D. G., 1978. Additional records of Kansas Odonata. — Technical Publications of the State Biological Survey of Kansas 6: 1-35. Huggins, D. G. & W. U. Brigham, 1982. Chap. 4, Odonata, pp. 4.1-4.100. /n A. R. Brigham, W. U. Brigham & A. Gnilka, eds. Aquatic insects and oligo- chaetes of North and South Carolina. Midwest Aquatic Enterprises, Mahomet, IL: 837 pp. Huggins, D. G., P. M. Liechti and D. W. Roubik, 1976. Species accounts for certain aquatic macroinvertebrates from Kansas (Odonata, Hemiptera, Coleoptera and Sphaeriidae). /n J. Caldwell, ed. New records of the fau- na and flora of Kansas for 1975, Technical Publications of the State Biological Survey of Kansas 1: 13-77. Karsch, F., 1890. Ueber Gomphiden. — Entomologische Nachrichten 16: 370-382. Kennedy, C. H., 1917a. Notes on the life history and ecolo- gy of the dragonflies (Odonata) of central California and Nevada.. — Proceedings of the United States National Museum 52: 483-635. Kennedy, C. H., 1917b. The dragonflies of Kansas. The Odonata of Kansas with reference to their distribution. — Bulletin of the University of Kansas Biological Series 11: 127-143. Kennedy, C. H., 1918. New species of Odonata from the southwestern United States. Part II. — Canadian Entomologist 50 (9): 297-299. Kennedy, C. H., 1928. Evolutionary level in relation to geo- graphic, seasonal and diurnal distribution of insects. — Ecology 9 (4): 367-379. Kimmins, D. E., 1969. A list of the type-specimens of Odonata in the British Museum (Natural History) Part II. — Bulletin of the British Museum Natural History 23 (7): 287-314. Kirby, W. F., 1890. A synonymic catalogue of Neuroptera Odonata, or dragonflies, with an appendix of fossil spe- cies. — Gurney and Jackson, London: ix + 202 pp. Kormondy, E. J., 1957. New knowledge of the Odonata of Tennessee. — Journal of the Tennessee Academy of Science 32 (2): 106-115. Kormondy, E. J., 1960. New North American records of an- isopterous Odonata. — Entomological News 71 (5): 121- 130. La Rivers, I., 1938. An annotated list of the Libelluloidea (Odonata) of southern Nevada. — Pomona College Journal of Entomology and Zoology 30: 73-85. La Rivers, I., 1940a. A preliminary synopsis of the dragon- flies of Nevada. — Pan-Pacific Entomologist 16 (3): 111- 123; La Rivers, I., 1940b. Some dragonfly notes from northern Nevada. Pomona College Journal of Entomology and Zoology 32: 61-68. 243 Tij DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 La Rivers, I., 1941. Additions to the list of Nevada dragon- flies (Odonata). — Entomological News 52: 126-130, 155-157. Larsen, W. P., 1952. The dragonflies (Anisoptera) of Utah. — Unpublished Master's Thesis, Department of Zoology, University of Utah: 95 pp. Macklin, J. A. & C. Cook, 1967. New records of Kentucky Odonata. — Proceedings North Central Branch, Entomological Society of America 22: 120-121. Maes, J-M., J-M. Desmedt & V. Hellebuyck, 1988. Caralogo de los Odonata de Nicaragua. Revista nicara- guense entomologica 4: 29-43. Molnar, D. R. & R. J. Lavigne, 1979. The Odonata of Wyoming (dragonflies and damselflies). — University of Wyoming Agricultural Experiment Station Scientific Monograph 37: 1-142. Montgomery, B. E., 1925. Records of Indiana dragon- flies-1. — Proceedings of the Indiana Academy of Sciences 34: 383-389. Montgomery, B. E., 1927. Records of Indiana dragon- flies-II. — Proceedings of the Indiana Academy of Sciences 36: 287-291. Montgomery, B. E., 1929. Records of Indiana dragonflies, III. — 1927-1928. Proceedings of the Indiana Academy of Sciences 1929: 335-343. Montgomery, B. E., 1935. Records of Indiana dragonflies, VII. 1934. — Proceedings of the Indiana Academy of Sciences 44: 231-235. Montgomery, B. E., 1937. Records of Indiana dragonflies, IX. 1935-1936. — Proceedings of the Indiana Academy of Sciences 46: 203-210. Montgomery, B. E., 1940. The Odonata of South Carolina. — Journal of the Elisha Mitchell Scientific Society 56 (2): 283- 301. Montgomery, B. E., 1941. Records of Indiana dragonflies, X. 1937-1940. — Proceedings of the Indiana Academy of Sciences 50: 229-241. Montgomery, B. E., 1947. The distribution and relative sea- sonal abundance of Indiana species of five families of dragonflies (Odonata: Calopterygidae, Petaluridae, Cordulegasteridae, Gomphidae and Aeshnidae). — Proceedings of the Indiana Academy of Sciences 56: 163- 169. Montgomery, B. E., 1951. Notes and records of Indiana Odonata, 1941-1950. — Proceedings of the Indiana Academy of Sciences 60: 205-210. Montgomery, B. E., 1955. Notes and records of Indiana Odonata, 1953-54. — Proceedings of the Indiana Academy of Sciences 64: 131-135. Montgomery, B. E., 1967. Geographical distribution of the North Central States. — Proceedings North Central Branch, Entomological Society of America 22: 121-129. Montgomery, B. E., 1968. The distribution of western Odonata. — Proceedings North Central Branch, Entomological Society of America 23 (2): 126-136. Montgomery, B. E., 1973. Why snakefeeder? Why dragon- fly? Some random observations on etymological entomol- ogy. — Proceedings of the Indiana Academy of Sciences 82: 235-241. Musser, R. J., 1961. Some noteworthy dragonfly records from Utah (Odonata: Anisoptera). — Entomological News 72 (2): 53. Musser, R. J., 1962. Dragonfly nymphs of Utah (Odonata: Anisoptera). — University of Utah Biological Series 12 (6): vii + 74 pp. 244 Muttkowski, R. A., 1910. Catalogue of the Odonata of North America. — Bulletin of the Public Museum of the City of Milwaukee 1 (1): 1-207. Needham, J. G., 1897. Preliminary studies of N. American Gomphinae. — Canadian Entomologist 29: 164- 168, 181-186. Needham, J. G., 1899. Ophiogomphus. — Canadian Entomologist 31 (9): 233-238. Needham, J. G., 1911. Notes on a few nymphs of Agrioninae (Order Odonata) of the Hagen collection. — Entomological News 22: 342-345. Needham, J. G., 1940. Studies on neotropical Gomphine dragonflies. (Odonata). — Transactions of the American Entomological Society 65: 363-394. Needham, J. G., 1941. Life history studies on Progomphus and its nearest allies (Odonata: Aeschnidae). — Transactions of the American Entomological Society 67: 221-245. Needham, J. G., 1943. Notes on some Gomphine dragon- flies from Venezuela and Guatemala. — Boletin de Entomologia venezolana 2 (4): 197-206. Needham, J. G., 1944. Further studies on neotropical Gomphinae (Odonata). — Transactions of the American Entomological Society 69: 171-224. Needham, J. G. & M. H. Anthony, 1903. The skewness of the thorax in Odonata. — Journal of the New York Entomological Society 11: 117-124. Needham, J. G. & T. D. A. Cockerell, 1903. Some hitherto unknown nymphs of Odonata from New Mexico. — Psyche 10: 134-139. Needham, J. G. & C. A. Hart, 1901. The dragonflies (Odonata) of Illinois. Part I. Petaluridae, Aeschnidae, and Gomphidae. — Bulletin of the Illinois State Laboratory of Natural History 6: 1-94. Needham, J. G. & H. B. Heywood, 1929. A handbook of the dragonflies of North America. — Charles C. Thomas, Springfield: viii + 372 pp. Needham, J. G. & M. J. Westfall, Jr., 1955. A manual of the dragonflies of North America (Anisoptera). — Univ. of Calif. Press, Berkeley: xii + 615 pp. Novelo-G., R. & E. Gonzälez-S., 1991. Odonata de la Reserva de la Biosfera La Michiliá, Durango, Mexico. Parte II. Náyades. — Folia Entomologica Mexicana 81: 107-164. Novelo-G., R. & J. Pena-O., 1991. Odonata from the northern mountain range of Hidalgo State, Mexico. — Notulae Odonatologicae 3 (8): 129-131. Osburn, R. C., 1905. The Odonata of British Columbia. — Entomological News 16: 184-196. Paulson, D. R., 1982. Odonata, pp. 249-277. In S. H. Hurlbert & A. Villalobos-Figueroa, eds. Aquatic biota of Mexico, Central America and the West Indies. San Diego State University, San Diego, CA: xv + 529 pp. Paulson, D. R., 1983. A new species of dragonfly, Gomphus (Gomphurus) lynnae spec. nov., from the Yakima River, Washington, with notes on the pruinosity in Gomphidae (Anisoptera). — Odonatologica 12 (1): 59-70. Paulson, D. R. & R. W. Garrison, 1977. A list and new dis- tributional records of Pacific Coast Odonata. — Pan- Pacific Entomologist 53 (2): 147-160. Pinhey, E., 1969. Tandem linkage in dichoptic and other Anisoptera (Odonata). — Occasional Papers of the National Museums of Rhodesia 4 (28B): 137-207. Pritchard, A. E. & R. F. Smith, 1956. Odonata, pp. 106- 153. In R. L. Usinger, ed. Aquatic insects of California. — University of California Press, Berkeley: ix + 508 pp. Resener, P. L., 1970. An annotated check list of the dragon- flies and damselflies (Odonata) of Kentucky. — Transactions of the Kentucky Academy of Science 31 (1- 2): 32-44. Ris, F., 1908. /nL. Schultze, Forschungsreise im westlichen und zentralen Südafrika, ausgeführt in den Jahren 1903- 1905. Denkschriften der medizinisch-naturwissenschaft- lichen Gesellschaft 13: 303-346. Ris, F., 1909. Abessinische Libellen, gesammelt von Dr. Eduard Rüppell. — Bericht über die Senckenbergische na- turforschende Gesellschaft, Frankfurt am Main 40: 21- Dijks Ris, F., 1917. über drei Arten Erpetogomphus (Odonata). — Archiv für Naturgeschichte (A) 82 (3): 152-158. Ris, F., 1918. Libellen (Odonata) aus der Region der ame- rikanischen Kordilleren von Costarica bis Catamarca. — Archiv für Naturgeschichte (A) 82 (9): 1-197. Ris, F., 1921. The Odonata or dragonflies of South Africa. — Annals of the South African Museum 18 (3): 245-452. Roback, S. S. & M. J. Westfall, Jr., 1967. New records of Odonata nymphs from the United States and Canada with water quality data. — Transactions of the American Entomological Society 93: 101-124. Scudder, S. & T. Cockerell, 1902. A first list of the Orthoptera of New Mexico. — Proceedings of the Davenport Academy of Sciences 9: 1-60. Seemann, T. M., 1927. Dragonflies, mayflies and stoneflies of Southern California. — Journal of Entomology and Zoology (Pomona College) 19 (1): 1-69. Selander, R. B. & P. Vaurie, 1962. A gazetteer to accompa- ny the ‘Insecta’ volumes of the ‘Biologia Centrali- Americana’. — American Museum Novitates 2099: 1-70. Selys-Longchamps, Edm. de, 1850. Revue des Odonates ou Libellules d'Europe. — Mémoirs de la Société royale des Sciences Liège 6: xxii + 408 pp. Selys-Longchamps, Edm. de, 1854. Synopsis des Gomphines. — Bulletin de [Académie royale de Belgique 21 (2): 23-112 (3-93 separate). Selys-Longchamps, Edm. de, 1858. Monographie des Gomphines. — Mémoirs de la Société royale des Sciences Liège 11: 257-720 (1-460 separate). Selys-Longchamps, Edm. de, 1859. Additions au synopsis des Gomphines. — Bulletin de lAcadémie royale de Belgique (2) 7: 530-552 (1-26 separate). Selys-Longchamps, Edm. de, 1868. Note sur quelques Odonates nouveaux du Mexique. — Comptes Rendus de la Société entomologique de Belgique 11: Ixvi-Ixxi (1-6 separate). Selys-Longchamps, Edm. de, 1869. Secondes additions au synopsis des Gomphines. — Bulletin de l’Académie royale de Belgique (2) 28: 168-208 (1-45 separate). Selys-Longchamps, Edm. de, 1873a. Troisiemes additions au synopsis des Gomphines. — Bulletin de l’Académie royale de Belgique (2) 35: 732-774 (1-46 separate). Selys-Longchamps, Edm. de, 1873b. Appendices aux troisiemes additions et liste des Gomphines, descrites dans le synopsis et ses trois additions. — Bulletin de l’Académie royale de Belgique (2) 36: 492-531 (47-87 separate). Selys-Longchamps, Edm. de, 1878. Quatrièmes additions au synopsis des Gomphines. — Bulletin de l’Académie royale de Belgique (2) 46: 408-698 (1-106 separate). Selys-Longchamps, Edm. de, 1879. Revision des Ophiogomphus et descriptions de quatre nouvelles GARRISON: Revision of Erpetogomphus Gomphines Americaines. - Comptes Rendus de la Socié- té Entomologique de Belgique 22: Ixii-Ixx (1-8 separate). Swofford, D. L. 1993. PAUP: Phylogenetic analysis using parsimony, version 3.1. — Computer program distributed by the Illinois Natural History Survey, Champaign, Illinois. Tennessen, K. J. & J. A. Louton, 1984. The true nymph of Gomphus (Gomphurus) crassus Hagen (Odonata: Gomphidae), with notes on adults. — Proceedings of the Entomological Society of Washington 86 (1): 223- 227. Tillyard, R., 1917. The biology of dragonflies (Odonata or Paraneuroptera). — Cambridge University Press: xii + 396 PP- Tinkham, E. R., 1934. The dragonfly fauna of Presidio and Jeff Davis counties of the Big Bend region of trans-Pecos, Texas. — Canadian Entomologist 66: 213-218. Tsuda, S., 1986. A distributional list of world Odonata. Preliminary edition. — Privately published, Osaka: viii + 246 pp. Tsuda, S., 1991. A distributional list of world Odonata. — Privately published, Osaka: 362 pp. Tucker, E. S., 1907. Some results of desultory collecting of insects in Kansas and Colorado. — Kansas University Science Bulletin 4 (2): 51-112. Tucker, E. S., 1908. Incidental captures of Neuropterous insects at Plano, Texas: — Psyche 15: 97- 100. Valley, S., 1993. D[ragonfly ]S[ociety of]A[merica] meeting in Bend, Oregon. — Argia 5 (2): 3-6. Van der Weele, H. W., 1906. Morphologie und Entwicklung der Gonapophysen der Odonaten. — Tijdschrift voor Entomologie 49: 99-198. Walsh, B., 1862. List of the pseudoneuroptera of Illinois contained in the cabinet of the writer, with descriptions of over forty new species, and notes on their structural af- finities. — Proceedings of the Academy of Natural Sciences of Philadelphia 14: 361-402. Walsh, B., 1863. Notes [on Pseudoneuroptera], pp. 182- 272. In Observations on certain N. A. Neuroptera, by H. Hagen, M. D., of Koenigsberg, Prussia; translated from the original French MS., and published by permission of the author, with notes and descriptions of about twenty new N. A. species of Pseudoneuroptera. — Proceedings of the Entomological Society of Philadelphia 2 (1): 167-272. Watrous, L. E. & Q. D. Wheeler., 1981. The out-group comparison method of character analysis. — Systematic Zoology 30(1): 1-11. Westfall, M. J., Jr., 1984. Odonata, pp. 126-176. In R. W. Merrittand K. W. Cummins, eds. An introduction to the aquatic insects of North America (2nd ed). — Kendall/Hunt, Dubuque: xiii + 722 pp. Westfall, M. J., Jr., 1987. Order Odonata, pp. 95-117. /zF. W. Stehr, ed. Immature insects. Kendall/Hunt, Dubuque: xiv + 754 pp. White, T. R., K. J. Tennessen, R. C. Fox, P. H. Carlson, 1980. The aquatic insects of South Carolina. Part I: Anisoptera (Odonata). — Station Bulletin, South Carolina Agricultural Experiment Station, Clemson University, Clemson: 632: 1-153. Williamson, E. B., 1902. Additions to the Indiana list of dragonflies, with a few notes. No. IL. — Proceedings of the Indiana Academy of Sciences 119-127. Williamson, E. B., 1903. The dragonflies (Odonata) of Tennessee, with a few records for Virginia and Alabama. — Entomological News 14: 221-229. Williamson, E. B., 1914a. September dragonflies about TijpsCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Mesa, Arizona (Odon.). Entomological News 25: 225- 226. Williamson, E. B., 1914b. Dragonflies (Odonata) collected in Texas and Oklahoma. — Entomological News 25: 411- 415, 444-454. Williamson, E. B., 1917. An annotated list of the Odonata of Indiana. — Miscellaneous Publications of the Museum of Zoology, University of Michigan 2: 1-13. Williamson, E. B., 1918. Results of the University of Michigan-Williamson expedition to Colombia, 1916-17. I. Two interesting new Colombian Gomphines (Odonata). — Occasional Papers of the Museum of Zoology, University of Michigan 52: 1-14. Williamson, E. B., 1923. Odonatological results of an auto trip across Indiana, Kentucky and Tennessee. — Entomological News 34: 6-9, 37-40. Williamson, E. B., 1932. Dragonflies collected in Missouri. — Occasional Papers of the Museum of Zoology, University of Michigan 240: 1-40. Williamson, E. B. & J. H. Williamson, 1930. Five new Mexican dragonflies (Odonata). — Occasional Papers of the Museum of Zoology, University of Michigan 216: 1- 34. Wright, M., 1938. A review of the literature on the Odonata of Tennessee. — Tennessee Academy of Science 13 (1): 26-33. Wright, M. & A. Peterson, 1944. A key to the genera of an- isopterous dragonfly nymphs of the United States and Canada (Odonata, Suborder Anisoptera). — Ohio Journal of Science 44 (4): 151-166. Young, W. C. & C. W. Bayer, 1979. The dragonfly nymphs (Odonata: Anisoptera) of the Guadalupe River Basin, Texas. — Texas Journal of Science 31 (1): 85-98. Received: 19 November 1993 Accepted: 10 August 1994 GARRISON: Revision of Erpetogomphus 6. schausi d 14. liopeltis d Figs. 1-14. Synthorax of Erpetogomphus species, lateral view (males, except given otherwise). — 1, constrictor (Honduras); 2, sa- baleticus (holotype); 3, tristani (Costa Rica); 4, ophibolus (Mexico: Veracruz State); 5, agkistrodon (female holotype); 6, schau- si (holotype); 7, eutainia (Texas: Gonzales Co); 8, eutainia (Mexico: Veracruz State); 9, eutainia (Costa Rica); 10, leptophis (holotype); 11, elaphe (holotype); 12, elaps (Mexico: Sinaloa State); 13, elaps (Mexico: Morelos State); 14, liopeltis (holotype). Scale line: 3 mm. 1994 [1IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 13 17. designatus d Itis lampropeltis 3 wee 2 22. crotalinus d i, > 23. heterodon 3 21. lampropeltis natrix d an N at 27. cophias 7. Synthorax of Erpetogomphus species, lateral view (males, except given otherwise). — 15, bothrops (holotype); 16, is (Mexico, Veracruz State); 17, designatus (lectotype); 18, designatus (Mexico, Durango State); 19, sipedon (female ho- ype); 20, £ lampropeltis (holotype); 21, lampropeltis natrix (holotype); 22, crotalinus (lectotype); 23, heterodon (holotype); 24, compositus (California, Yolo Co.); 25, compositus (paratype of coluber); 26, boa (Mexico: Veracruz State); 27, cophias Mexico: Morelos State). Scale line 3 mm. 248 28. 29. 30. Sl. 32: 33. 34. 33. 36. constrictor 9 ophibolus d agkistrodon 9 schausi d eutainia d eutainia 9 leptophis d leptophis 9 liopeltis d GARRISON: Revision of Erpetogomphus anf ae E à 4, A Haas aiar Een) Figs. 28-36. Abdomen, lateral view. — 28, constrictor 2 (Honduras); 29, ophibolus 8 (Mexico: Veracruz State); 30, agkistro- don ® (holotype); 31, schausi 3 (holotype); 32, eutainia 8 (Texas: Gonzales Co.); 33, eutainia 2 (idem); 34, leptophis 3 (ho- lotype); 35, leptophis 2 (allotype); 36, liopeltis 3 (holotype). Scale line 10 mm. 249 [I]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 37. bothrops d 38. viperinus d 39. sipedon d 40. sipedon ? 41. crotalinus d ET 43. heterodon © en nl jr ges 11: D à suey n 44. boa d 46. cophias d \bdomen, lateral view. — 37, bothrops 3 (holotype); 38, viperinus & (Mexico: Veracruz State); 39, sipedon 3 s State); 40, sipedon 2 (holotype); 41, crotalinus & (Mexico: Jalisco State); 42, heterodon & (holotype); 43, VI 44, boa 3 (Mexico: Veracruz State); 45, boa 2 (idem); 46, cophias & (Mexico: Morelos State). Scale GARRISON: Revision of Erpetogomphus 48. sabaleticus 52. eutainia 53. leptophis 56. elaps 58. bothrops Figs. 47-58. Anterior and posterior hamules. First view is anterolateral view showing profile of anterior hamule; second view (when shown) is lateral view showing profile of posterior hamule. — 47, constrictor (Mexico: San Luis Potosi State); 48, saba- leticus (holotype); 49, tristani (Costa Rica); 50, ophibolus (Mexico: Veracruz State); 51, schausi (holotype); 52, eutainia (Texas: Gonzales Co.); 53, leptophis (holotype); 54, elaphe (Costa Rica); 55, elaps (holotype); 56, elaps (variant) (Mexico: Chiapas State); 57, liopeltis (paratype); 58, bothrops (paratype). Scale line 2 mm. 57. liopeltis (N°) N a [1JDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 62. lampropeltis 63. lampropeltis natrix 64. crotalinus lampropeltis 65. heterodon 68. boa 69. cophias nd posterior hamules. First view is anterolateral view showing profile of anterior hamule; second view is lateral view showing profile of posterior hamule. — 59, viperinus (Mexico: Veracruz State); 60, designatus (lec- ipedon (Mexico: Morelos State); 62, /. lampropeltis (California: Ventura Co.); 63, £ natrix (holotype); 64, crot- O: Miche acan State); 65, heterodon (New Mexico: Catron Co.); 66, compositus (Arizona: Pima Co.); 67, compo- pe of coluber); 68, boa (lectotype); 69, cophias (Mexico: Morelos State). Scale line 2 mm. LD) GARRISON: Revision of Erpetogomphus N 79. liopeltis 80. bothrops Figs. 70-80. Penis segment 4, left: dorsal view, right: lateral view. — 70, constrictor (Mexico: San Luis Potosi State); 71, saba- leticus (holotype); 72, tristani (Costa Rica); 73, ophibolus (Mexico: Veracruz State); 74, schausi (holotype); 75, eutainia (Texas: Gonzales Co.); 76, leptophis (holotype); 77, elaphe (Costa Rica); 78, elaps (holotype); 79, liopeltis (paratype); 80, bothrops (Mexico: Tamaulipas State). Scale line 1 mm. TIIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 91. cophias Figs. 81-91. Penis segment 4, left: dorsal view, right: lateral view. — 81, viperinus (Mexico: Veracruz State); 82, designatus Texas: Gonzales Co.); 83, sipedon (Mexico: Morelos State); 84, L lampropeltis (California: Ventura Co.); 85, L natrix Arizona: Cochise Co.); 86, crotalinus (Mexico: Michoacan State); 87, heterodon (New Mexico: Catron Co.); 88, compositus CA: Yolo Co.); 89, compositus (paratype of coluber); 90, boa (lectotype); 91, cophias (Mexico: Morelos State). Scale line 1 mm. GARRISON: Revision of Erpetogomphus 93. sabaleticus 96. schausi 97. eutainia 98. leptophis 99. elaphe 100. elaps u + 101. liopeltis 102. bothrops 103. viperinus Figs. 92-103. Caudal appendages of male, above = lateral view, below = dorsal view of cerci. — 92, constrictor (Mexico: Tamaulipas State); 93, sabaleticus (holotype); 94, tristani (holotype); 95, ophibolus (Mexico: Veracruz State); 96, schausi (ho- lotype); 97, eutainia (Texas: Gonzales Co.); 98, leptophis (holotype); 99, elaphe (Costa Rica); 100, e/aps (holotype); 101, Zo- peltis (holotype); 102, bothrops (holotype); 103, viperinus (lectotype). Scale line 5 mm. 255 IDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 107. lampropeltis lampropeltis 108. lampropeltis natrix 112. compositus A nen er 117. hopeltis 118. bothrops 119. viperinus 115. boa Fig. 104-116. Caudal appendages of male, above = lateral view, below = dorsal view of cerci. — 104, designatus (lectotype); 105, sipedon (Mexico: Morelos State); 106, sipedon (only lateral view) (Mexico: Durango State); 107, £ lampropeltis California: Ventura Co.); 108, 1 natrix (holotype); 109, crotalinus (lectotype); 110, heterodon (holotype); 111, heterodon Mexico: Durango State); 112, compositus (California: Yolo Co.); 113, compositus (paratype of coluber); 114, boa (only lateral v) (lectotype); 115, boa (Mexico: Veracruz State); 116, cophias (Mexico: Morelos State). Figs. 117-119. Right cercus, me- lateral view. — 117, liopeltis (Mexico: Nuevo Leon State); 118, bothrops (Mexico: Nayarit State); 119, viperinus (Mexico: cruz State). Scale line 5 mm. GARRISON: Revision of Erpetogomphus 121. ophibolus 122. eutainia 123. eutainia 125. eutainia 126. liopeltis 127. bothrops 128. bothrops 130. designatus 131. designatus 132. lampropeltis lampropeltis 133. crotalinus 134. compositus 135. boa Figs. 120-135. Erpetogomphus. — 120, tip of epiproct of male, posterior view, tristani (Costa Rica); 121, idem, ophibolus (Mexico: Veracruz State); 122, epiproct, lateral view, eutainia (Texas: Gonzales Co.); 123, idem, dorsolateral view; 124-135, tip of epiprocts of male, posterior view, 124, eutainia (Texas: Gonzales Co.); 125, eutainia (Guatemala); 126, Lopeltis (Mexico: Nuevo Leon State); 127, bothrops (Mexico: Nayarit State); 128, bothrops (holotype); 129, viperinus (Mexico: Veracruz State); 130, designatus (Texas: Gonzales Co.); 131, designatus (Mexico: Durango State); 132, / lampropeltis (California: Ventura Co.); 133, crotalinus (Mexico: Jalisco State); 134, compositus (Arizona: Coconino Co.); 135, boa (Mexico: Veracruz State). Scale line 2 mm (figs. 122-123), 1 mm (figs. 120-121, 124-135). TIDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 @ CC) 138. lampropeltis lampropeltis d ( (È 136. bothrops \ } 139. compositus 3 ANY | 147. ophibolus istani 146 tr iy / if yh if IP [IS] 150. Stia + | ) 151. leptophis = Figs. 136-153. Erpetogomphus. — 136-139. Anterior part of frons, dorsal view. 136, bothrops © (Mexico: San Luis Potosi State); 137, viperinus 2 (Mexico: Veracruz State); 138, / lampropeltis 8 (California: Ventura Co.); 139, compositus 3 (AZ: Maricopa Co.). — Figs. 140-141. Crest outline of occiput, dorsal view. 140, Liopeltis (holotype), 141, liopeltis (Mexico: Nuevo Leon State). — Figs. 142-144. Head, anterior view. 142, schausi (holotype); 143, eutainia (Costa Rica); 144, eutainia (Texas: Gonzales Co.). — Figs. 145-153.Vertex, occiput of female, dorsal view. 145, constrictor (Honduras); 146, tristani (Costa Rica); 147, ophibolus (Mexico: Veracruz State); 148, agkistrodon (holotype); 149, schausi (Costa Rica); 150, eutainia (Texas: Gonzales Co.); 151, leptophis (allotype); 152, elaps (Mexico: Morelos State); 153, liopeltis (allotype). Scale line 4 mm (figs. 142-144); 1 mm (136-141, 145-153). 153. liopeltis 258 GARRISON: Revision of Erpetogomphus 154. bothrops ji Auen 157. designatus 4 RE VIE rtl 60. lampropeltis | 161. crotalinus lampropeltis N 166. viperinus 167. boa 165. cophias Figs. 154-165. Vertex, occiput of female, dorsal view. 154, bothrops (allotype); 155, viperinus (Mexico: Veracruz State); 156, designatus (paralectotype); 157, designatus (Mexico: Durango State); 158, sipedon (form ‘b’) (Mexico: Puebla State); 159, s- pedon (holotype); 160, I. lampropeltis (California: Ventura Co); 161, crotalinus (paralectotype); 162, heterodon (allotype); 163, compositus (holotype); 164, boa (Mexico: Veracruz State); 165, cophias (Mexico: Guerrero State). — Figs. 166-167. Crest and part of rear of head (postocciput) of female, dorsal view. 166, viperinus (Mexico: Veracruz State); 167, boa (Mexico: Veracruz State). Scale line 1 mm. 259 T1JDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 ti » 4 2, Ser / AO OI RON A SOS OF ge pce ir Werke ee 171. constrictor 172. sabaleticus 173. tristani 176. agkistrodon 177. schausi 179. leptophis 180. elaphe 181. elaps 182. liopeltis Fig. 168-170. ‘ig. Merathoracic leg, lateral view. — 168, sipedon © (holotype); 169, crotalinus d (lectotype); 170, heterodon à (holoty pe). — Figs. 171-182. Vulvar lamina, dorsal view. — 171, constrictor (Honduras); 172, sabaleticus (allotype); 173, tri- ini allotype); 174, tristani (Costa Rica); 175, ophibolus (Mexico: Veracruz State); 176, agkistrodon (holotype); 177, schausi Costa Rica) | distorted on right side); 178, eutainia (Mexico: Veracruz State); 179, leptophis (allotype); 180, elaphe (Costa Rica); 181, elaps (Mexico: Morelos State); 182, liopeltis (Mexico: Nuevo Leon State). Scale line 4 mm (figs. 168-170), 2 mm 71-182). 260 GARRISON: Revision of Erpetogomphus 183. bothrops 184. viperinus 187. sipedon 189. lampropeltis 190. lampropeltis lampropeltis natrix 191. crotalinus . heterodon À NE ik, NA 9 195. boa 196. boa 197. cophias Figs. 183-197. Vulvar lamina, dorsal view. — 183, bothrops (allotype); 184, viperinus (Mexico: Veracruz State); 185, designa tus (paralectotype); 186, designatus (Texas: Gonzales Co.); 187, sipedon (holotype); 188, sipedon (paratype); 189, l. lampro- peltis (California: Ventura Co.); 190, /. natrix (allotype); 191, crotalinus (paralectotype); 192, crotalinus (Mexico: Jalisco State); 193, heterodon (allotype); 194, compositus (Arizona: Maricopa Co.); 195, boa (Mexico: Veracruz State); 196, boa (Veracruz State); 197, cophias (Mexico: Guerrero State). Scale line 2 mm. 261 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 @ = constrictor \ < A = ophibolus sabaleticus @ = narrow hamular | — form of elaps A- elaphe À = leptophis B = liopeltis ®- bothrops A= viperinus 202 =) Figs. 198-202. Distribution patterns. — 198, E. ophibolus group; 199, E. eutainia group; 200, E. elaps group (in part); 201, E. elaps group (in part); 202, E. designatus. 262 GARRISON: Revision of Erpetogomphus 7 PL LI GR tr xo] dr; x Se ® = compositus À = heterodon | Figs. 203-207. Distribution patterns. 203, E. sipedon; 204, E. lampropeltis 205, E. crotalinus, 206, E. heterodon and E. com- positus, 207, E. boa and E. cophias. 263 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 O. severus constr. eut. elaps elaphe lampro. boa u) è = | \ A 4 zoal D 1 2 3 4 5 \ = i i } i À O. severus CONStr. eut. elaps cophias > N > Î > (| > fl 209 1 2 3 4 O. severus sabalet. viper. liop. lampro. Sw ss > > > > 210 0 2 3 4 5 Ophiogomphus 211-b 211-c 211-d 211-e severus tristani viperinus compositus boa Figs. 208-211. 208, morphocline of anterior hamule, character states 0-5; 209, morphocline of posterior hamule, character states 0-4; 210, morphocline of tip of posterior hamule, character states 0-5; 211, dorsal view of fourth segment of penis (S4); È 211 = es = e 0 = 1 figs. 21 1a and 211b have 4th segment about as long as wide, figs. 211c-e about twice as long as wide. Scale line 2 mm (figs. 208-209); 1 mm (figs. 210-212). 264 GARRISON: Revision of Erpetogomphus Ophiogomphus ophibolus eutainia crotalinus severus group group group > ua > C 5 > prepuce lacking > ai (state 3) 3 ca prepuce prepuce prepuce (hidden by lateral lobe: 212 (state 0) (state 1) state 2) crotalinus Ophiogomphus ophibolus eutainia elaps subgroup Severus group group group designatus (except designatus) 213 € NX, boa, cophias Spiubolus compositus \\ | PS È ! [| group (part) = pi | lea 0 A crotalinus i N 60 2 À lampropeltis dé (i : o schausi SE Ee elaps ae subgroup Ve eutainia le IN Oph. sev. = oe | KA Li 00 i 214 Re io a8 BR Ophiogomphus severus eutainia viperinus 215 (state 0) (state 1) (state 2) Figs. 212-215. — 212, morphocline of prepuce of penis, character states 0-3; 213, morphocline of shape of lateral lobe, char- acter states 0-5; 214, morphocline of cornua: character states 0-6 for character 19 and character states 0-2 for character 20; 215, morphocline of spermatheca in dorsal (internal) view character states 0-2 (figures show dorsal (internal) view of sternum 9 (including postlamellar ridge). Spematheca is dorsad of vulvar lamina). Scale line 1 mm (figs. 212, 215); 0.5 mm (figs. 213- 214). VOLUME [I]DSCHRIFT VOOR ENTOMOLOGIE, Zoe sejydos B0q snjjsodu09 UOpo.1339Y snu/[81019 spodoiduwue] uopadjs - - sn) eudjsop -- snuj19dja sdo1yjoq spiodo] aydeyja ae sde jo dno13 snu1 [81019 dno13 ejujggna a Bees sjydo]da] uopo11S]498 jsney>s dno13 snjogjydo snjogıydo 1039]1]}SUOI sn>pjaleges JUBIS]A] sn1949$ ‘ydo cececcec | . eze2222 1 leceeece | zeezeze, EEE o _ leeceee | edel LEZE 222221 leeee | NEEN AE ZIE zz, FE Libliee | zz: Lia ELL 211 TAN ELLI LL nn ANN LEEL'L ceed | Leed’ l ciec't Lee L Lel cec | Lee | cc | PE cl 7 / 66 GARRISON: Revision of Erpetogomphus 0 species s 1 species > RE 2 species 3 species 4 species 5 species 8 species Fig. 217. geographic distribution of Erpetogomphus collected in the U. S. A., Mexico, and Central America. Each square is approximately 150 kilometers square. 267 TI]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 agkistrodon ophibolus = ”) N eutainia N SE _ schausi N» EX Jt Ao je 4 Ja leptophis constrictor TA 39 | a EN À — tristani — 221 sabaleticus } 22D 4 (ss na = e e | . IO u: DIE NG € \ 7 =erotalinus group le [a] = elaps group > vrt C À > 7 =A 29 SIC 223 IR 224 gs. 218-224. — 218, Generalized track for genus Erpetogomphus, 219, generalized tracks for Erpetogomphus ophibolus | eu- ind £. elaps | crotalinus groups; 220, generalized tracks for Erpetogomphus ophibolus and E. eutainia groups; 221, tracks ded GI I / E . erde A within the Erpetogomphus ophibolus group; 222, tracks of species within the Erpetogomphus eutainia group; 223, ge- acks for Erpetogomphus crotalinus and E. elaps groups; 224, tracks of species within the Erpetogomphus elaps group 268 GARRISON: Revision of Erpetogomphus cophias“ 229. viperinus | 230. heterodon ER Figs. 225-230. — 225, tracks of species within the Erpetogomphus elaps group (in part); 226, tracks of species within the Erpetogomphus crotalinus group; 227, pair of wings of liopeltis & (Mexico: Nuevo Leon State); 228, idem, bothrops È (Mexico: Nayarit State); 229, idem, viperinus & (Mexico: Veracruz State); 230, idem, heterodon 8 (N. Mexico: Grant Co.) 269 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 “ah 270 N i en il wor 2 4 . F d x Air ©) FEW Kove Jen tar È Li Adi cabal 7 oc dr He EME Mir? aval A) ovens das TE HERMAN DE JONG Institute for Systematics and Population Biology, Amsterdam HAE PEMLOGENY OF THE SUBGENUS TIPULA (SAVTSHENKTA) (DIPTERA: TIPULIDAE), WITH SPE@GIAHREEERENCGE TO THE WESTERN MEDITERRANEAN FAUNA In memory of E. N. Savchenko (12.VH.1909 - 8.1.1994) Jong, H. de, 1994. The phylogeny of the subgenus Tipula (Savtshenkia) (Diptera: Tipulidae), with special reference to the western Mediterranean fauna. — Tijdschrift voor Entomologie 137: 271-323, figs. 1-202, tabs. 1-5, appendices A-C. [ıssn 0040-7496]. Published 15 December 1994. The phylogeny of the currently recognized 87 species and subspecies of the subgenus Tipula (Savtshenkia Alexander) (Diptera, Tipulidae) is discussed in reference to 87 morphological characters of the adults. The distribution of the species and subspecies of Savtshenkia is briefly outlined. Taxonomic novelties include the description of three new western Palaearctic species, viz. alpha, omega, and trinacria; the establishment of the synonymy of broweri Alexander under fragilina Alexander, insignifica Alexander under invenusta Riedel, nebulipennis Alexander under alpium Bergroth, packardi Alexander under invenusta Riedel, and perparvula Alexander under ignobilis Loew; the elevation of the subspecies subsignata haennii Dufour to species rank; and the removal of the species convexifrons Holmgren from Savtshenkia. H. de Jong, Department of Entomology, Institute for Systematics and Population Biology (Zoological Museum), Plantage Middenlaan 64, 1018 DH Amsterdam, The Netherlands. Key words. — Diptera, Tipulidae, Savtshenkia, phylogeny, western Mediterranean. As part of a research project on the historical bio- geography of the western Mediterranean region, the phylogeny of the subgenus Tipula (Savtshenkia Alexander) is discussed. Savtshenkia as defined here, contains 87 described species and subspecies, of which 22 have a restricted distribution in the western Mediterranean. Special emphasis is laid on the estab- lishment of the phylogenetic position of these western Mediterranean endemics, as they offer potential clues to a better understanding of the historical biogeogra- phy of the area. The species constituting the subgenus Savtshenkia were formerly grouped in the marmorata species group of the genus Tipula Linnaeus (Alexander 1919a, 1919b et seqq.). The same assemblage of spe- cies has also been known as the fragilis group (Alexander 1934 et seqq.) and obsoleta group (Lackschewitz 1936). Usually, these species groups were referred to the subgenus Oreomyza Pokorny (Alexander 1934, Edwards 1931). Savchenko (1961), misinterpreting Riedel (1913) regarding the type spe- cies of Pterelachisus Rondani, synonymized Oreomyza under the latter taxon. Savchenko’s lapsus resulted in the improper use of the name Pterelachisus for the species at present placed in Savtshenkia. Although Savchenko’s reasoning was fallacious, Mannheims (1962a) showed that the synonymy of Oreomyza and Pterelachisus sensu Rondani in itself was valid and in- troduced the name Savtshenkia to replace Pterelachisus sensu Savchenko. As Mannheims omit- ted to denote the type species of this new taxon, Savtshenkia remained a nomen nudum until Alexander (1965a) designated rufina Meigen as the type species. (In the literature, a paper by Alexander published in the 1965 volume of the Philippine Journal of Science is usually cited as the relevant ref- erence for the authorship of the name Savtshenkia. As the pertaining publication was actually issued August 9, 1966, Alexander’s paper that appeared in Pacific Insects of June 20, 1965, should be accepted as the correct reference for the type designation.) Savtshenkia has an essentially Holarctic distribu- tion (fig. 1), with 65 species and subspecies occurring in the Palaearctic Region and eight in the Nearctic Region. Both regions have two species in common. In addition to the Holarctic members of the subgenus, five Oriental and 12 Afrotropical species and subspe- cies are known. The Oriental fauna of Savtshenkia has a single species in common with the Palaearctic fau- na, the Afrotropical fauna shows no species overlap 271 TI]DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Fig. 1. Global distribution of the subgenus 7ipula (Savtshenkia). with that of any other biogeographical region. At the end of this paper a brief account of the distribution of the species of Savtshenkia is provided. I will present more detailed information on the distribution of the western Mediterranean species and their closest rela- tives in a forthcoming paper on the historical bio- geography of the western Mediterranean. In Appendix A, three western Palaearctic species of Savtshenkia are described as new, five synonymies are established, while one subspecies is raised to species rank and one species is removed from the subgenus. The resulting list of species and subspecies of Savtshenkia as recognized in this paper is given in ta- ble 1. Savchenko (1961, under Pterelachisus) and Theowald (1957, under Oreomyza, 1967, 1973, 1978) arranged the (western) Palaearctic members of Savtshenkia in a number of species groups. Theowald’s 1957 and 1967 arrangements were based on characters of the preadult stages of western Palaearctic species and contained four species groups (table 2). Savchenko (1961) recognized five species groups plus a miscellaneous group within the fauna of the former USSR (table 3). Theowald’s 1973/1978 classification of the adults of Savtshenkia in 14 species groups basically is a refinement of Savchenko’s 1961 concept (table 3). The arrangements of both authors will be compared with the results of the present study in the Discussion of adopted phylogeny’, below. DD MATERIAL, METHODS AND TERMINOLOGY During the course of this investigation material re- presenting 78 species and subspecies of Savtshenkia was examined (table 1). Most specimens used for the present study originated from the Institute for Systematics and Population Biology (Zoological Museum), Amsterdam. Supplementary material was studied during a short stay at the United States National Museum of Natural History, Smithsonian Institution, Washington D.C. In addition to this, material was borrowed from the following institu- tions: Academy of Natural Sciences, Philadelphia; Musée d'Histoire Naturelle, Neuchâtel; Natural History Museum, London; Staatliches Museum für Naturkunde, Stuttgart; United States National Museum of Natural History, Smithsonian Institution, | Washington D.C; Zoologiska Institutionen, Systematiska avdelningen, Lund. The majority of specimens consisted of pinned ma- terial; in addition to this, material preserved in alco- hol was studied from the collections in Amsterdam, Lund, and Neuchâtel. Preparations of the male and female terminalia were made by removing these parts and clearing them in a nearly boiling 10% KOH solution for about five minutes. After rinsing with water and 70% alcohol, the terminalia were transferred to glycerol. Examination of the specimens and their terminalia H. DE Jong: Phylogeny of Tipula (Savtshenkia) Table 1. List of species and subspecies of Tipula (Savtshenkia) as recognized in this paper. Abbreviations: +: examined; -: not examined; u: unknown; Afro: Afrotropical Region; Nea: Nearctic Region; Or: Oriental Region; Pal: Palacarctic Region; Th., Duf. & Oost: Theowald, Dufour & Oosterbroek. d ? Region aberdareica Alexander, 1956 + u Afro a. ulugurica Alexander, 1962 UAL akeleyi Alexander, 1956 + u Afro alpha sp. n. + + Pal alpium Bergroth, 1888 + + Nea, Pal asbolodes Speiser, 1909 TREIA TO aspromontensis Theowald, 1973 a + Pal aster Theischinger, 1983 + u Pal atlas Pierre, 1924 + + Pal baltistanica Alexander, 1936 + Or Pal benesignata Mannheims, 1954 + + Pal boreosignata Tjeder, 1969 Bus all breviantennata Lackschewitz, 1933 + + Pal caligo Alexander, 1956 + + Afro cheethami Edwards, 1924 Pal chrysocephala Mannheims, 1958 + u Afro confusa Van der Wulp, 1887 +0 +0 Pal corsosignata Th., Duf. & Oost. 1982 + + Pal cyrnosardensis Th., Duf. & Oost. 1982 + + Pal draconis Alexander, 1964 + u Afro eleonorae Theischinger, 1978 + u Pal elgonensis Alexander, 1956 + u Afro eugeni Theowald, 1973 +. + Pal fragilina Alexander, 1919 + + Nea fragilis Loew, 1863 + + Nea gimmerthali Lackschewitz, 1925 Da g. mattheyi Theowald & Dufour, 1983 - u Pal g. pteromaculata Th., Duf. & Oost. 1982+ u Pal glaucocinerea Lundström, 1915 HN Pal goriziensis Strobl, 1893 En ee Pal graciae Alexander, 1947 + u Ne grisescens Zetterstedt, 1851 AU Pal haennii Dufour, 1991 stat.n. SEP a hancocki Alexander, 1956 + + Afro hartigiana Th., Duf. & Oost. 1982 ee al holoptera Edwards, 1939 + + Pal ignobilis Loew, 1863 + + Nea imperfecta Riedel, 1914 = - Afro interserta Riedel, 1913 ae Pal invenusta Riedel, 1919 ee REN eas al 1. microinvenusta Dufour, 1990 er Pal i. subinvenusta Slipka, 1950 + + Pal jeekeli Mannheims & Theowald, 1959 + + Pal kiushiuensis Alexander, 1925 END koreana Alexander, 1934 fo RDA was carried out using a Wild stereomicroscope with a magnification of up to 100 times. Drawings were made with the aid of a drawing tube attached to the microscope. Illustrations of the genital structures were made from macerated specimens. For perma- nent storage, the terminalia were transferred to a mi- crovial containing a drop of glycerol. The microvial was pinned with the relevant specimen. The parsimony program HENNIG86, version 1.5 (Farris 1988) was used to analyse the phylogeny. d ¢ Region letifera Alexander, 1951 È u Pal limbata Zetterstedt, 1838 n + Pal lundbladi Mannheims, 1962 + + Pal macaronesica Savchenko, 196] + u Pal mannheimst Theowald, 1973 + u Pal minuscula Savchenko, 1971 - = Pal mohriana Alexander, 1954 + + Pal multipicta Becker, 1908 + + Pal nephrotemoides Alexander, 1924 + + Afro n. invariegata Alexander, 1956 u + Afro nielseni Mannheims, & Theowald, 1959+ + Pal nivalis Savchenko, 1961 - Pal obsoleta Meigen, 1818 + + Pal odontostyla Savchenko, 1961 Dal omega sp. n. + u Pal ornata Theowald & Oosterbroek, 1987 + + Pal pagana Meigen, 1818 + + Pal pechlaneri Mannheims & Theowald, 1959 HO eta Pal persignata Alexander, 1945 + u Pal p. tofina Alexander, 1945 NET Pal phoroctenia Alexander, 1919 + + Nea postposita Riedel, 1919 Te ar Pal productella Alexander, 1928 = u Nea rufina Meigen, 1818 1 eee al r. maderensis Lackschewitz, 1936 Fo Pal sardosignata Mannheims & Theowald, 1959 el sciadoptera Alexander, 1964 nn Pooh serrulifera Alexander, 1942 el signata Staeger, 1840 + + Pal simulans Savchenko, 1966 + PA sordidipes Alexander, 1961 MOT staegeri Nielsen, 1922 + + Pal subalpium Savchenko, 1961 = u Pal subnodicornis Zetterstedt, 1838 si Pal subsignata Lackschewitz, 1933 + + Pal s. cazorla Dufour, 1991 a eal subvafra Lackschewitz, 1936 u ei tetragramma Edwards, 1928 = Um (Ofr trinacria sp. n. + Pal tulipa Dufour, 1983 u st Bell venerabilis Alexander, 1936 ui. Ol: villeneuvii Strobl, 1909 tie Heal More information on the calculation procedures fol- lowed is given in “Discussion of adopted phylogeny’. The terms for the structures of Savtshenkia as used in the character discussion are in general in accor- dance with the terms employed by McAlpine (1981), with a few additions for particular features of the Tipulidae of the study group. The terms are ex- plained in figs. 3-6, 12, and 58. It should be noted that I consider the structure lateral of the gonapoph- ysis in Savtshenkia homologous with the fragmentum TiIJjDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Lunatipula Mediotipula alpha grisescens hartigiana breviantennata ignobilis baltistanica postposita lundbladi atlas - goriziensis - cheethami — subnodicornis - tulips — vilteneuvil ——— simulans gp glaucocinerea multipicta rufina maderensis nephrotomoldes elgonensis draconis aberdareica asbolodes hancocki akeleyl caligo chrysocephala alplum macaronesica Jeekeli obsoleta holoptera pagana | | 72084 | K S7 0720 [es] | +69 +5,70 59 di Il dis | | 7,1502 O67 x82 PUN QE | m un — +43,68 +85 oe 023 REV 4 | | 716,58 | +21 as RE | D9 Tass m2 Dsssr pes $47 02,9 fn xa 11032 Iss D24 153,83 14,82 Fig. 2. Strict consensus tree of 14 equally parsimonious trees derived after successive weighting for the species and subspecies of Tipula (Savtshenkia). —: synapomorphy; in the sense of Rees & Ferris (1939) and Frommer (1963) and name it accordingly (figs. 3, 4). Mannheims (1951) denoted the same sclerite A9s (appendage of sternite 9). Suture x is employed in the sense of Neumann (1958). A recent overview of the morphology of the terminalia of Tipulidae was given by Tangelder (1985) to which paper the reader is re- ferred to for further details. Acronyms used throughout the text are the follow- ing: BMNH: Natural History Museum, London, England; BsNH: Boston Society of Natural History, Boston, Massachusetts, U.S.A; cncı: Canadian National Collection, Biosystematics Research Centre, Ottawa, Ontario, Canada; mczc: Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts, U.S.A.; MNHN: Muséum National d'Histoire Naturelle, Paris, France; MRAC: Koninklijk Museum voor Midden Afrika, Tervuren, Belgium; usNM: United States National Museum, Smithsonian Institution, Washington D.C., U.S.A; zias: Zoological Institute, Russian Academy of Sciences, St. Petersburg, Russia; zman: Institute for Systematics and Population Biology, Department of Dl: homoplasy; X: reversal. Entomology, Zoological Museum, Amsterdam, the Netherlands. BIOLOGY In general, the members of Savtshenkia are adapted to either mountainous, high latitudinal, or autumnal conditions. Within these limits, adults of the respec- tive species of Savtshenkia show a wide variety of pre- ferred biotopes. They can be found in alpine Rbododendron assocations, the heath zone at about 3300 m, mountainous wet or dry Pinus forests and mixed woods, low woods, low moorlands, marshes, bogs, and gardens. The species may occur in the ve- getation near the sources and on the banks of streams, on boulders in streams, on dripping rocks, and in the splash zone of cascades (Alexander 1961, Audcent 1932, Brunhes & Dufour 1992, Bryce 1956, Coulson 1959, 1962, Dufour 1983, 1986, 1990, 1991, Mannheims & Pechlaner 1963, Stubbs 1992, Theowald 1973, Theowald, Dufour & Oosterbroek 1982). The larvae of the species of Savtshenkia are associat- ed with mosses and liverworts. They are recorded immerthali pteromaculata 3 £ tr ri E E Bo = £ L 7 E E o © © immerthali mattheyi confusa interserta ornata graci sordidipes venerabilis eugeni serrulifera aster phoroctenia odontostyla fragilina limbat subvafra pechlaneri 0 aspromontensis OD frragllls rs a enusta nvenusta subinv H. DE Jong: Phylogeny of Tipula (Savtshenkia) ata cazorla nvenusta invenusta nvenusta microinvenusta staegeri sardosignata subsignata bsignat persignata tofina t — persignata persignata — corsosignata 5 O n 2 | | | I A er Ee | | K61X71| ise A 77,78 K67 | ESSA nd (730,811-14,34 | 72 | RS: 3 | | Les — Eu +13,18,63,64 DI, 80 12 +62 from mosses on stones and logs in and along cold streams, from wet moorland mosses and mosses of cold bogs of tundra and high mountains, as well as from relatively dry mosses on walls, stones and trees (Gelhaus 1986). As in most Tipulidae, the majority of species of Savtshenkia have a one-year life cycle. A few are, or appear to be, bivoltine (viz. alpium, breviantennata, rufina rufina, and rufina maderensis Dufour 1986, Mannheims 1951, Theowald 1973). On the other hand, Lantzov (1982) suggested a long term larval de- velopment for the arctic species glaucocinerea of up to six years. [Multi-annual life cycles are known of other species of Tipulidae with a northerly distribution; see Pritchard (1983) for an overview of life cycles in Tipulidae.] Details on the copulation of a few species of Savtshenkia were given by Hemmingsen (1954, 1962; viz. limbata, signata, staegeri, and subsignata). Hemmingsen (1952) provided a short account on oviposition of the species confusa (as marmorata) and rufina. SYSTEMATIC POSITION The species and subspecies of Savtshenkia can be distinguished from other Tipulidae by the following derived character states: (1) Adult, male terminalia: sp2 medially blackish sclerotized (character 53, be- low); (2) Adult, female terminalia: dorsal margin of hypogynial valve blackish sclerotized (character 83, below); (3) Larva: eight anal papillae arranged in four pairs. The number of anal papillae in other groups of Tipulidae is either six, four, two, or zero (Gelhaus 1986, Theowald 1957, 1967). As both adult charac- ters occur also in other genera and subgenera of Tipulidae, the larval character seems to offer the best criterion to distinguish Savtshenkia from the remain- der of Tipulidae. The systematic position of Savtshenkia within the Tipulidae is not yet fully established. Savchenko (1966, 1979, 1983) placed Savtshenkia next to the subgenus Tipula (Mediotipula Pierre), a relationship substantiated by the presence in the female pupa of a laterally placed small tubercle at the apex of the sheath of the hypogynial valve (see Theowald 1957, 1967). In species of the genus Dolichopeza Curtis, the sheath of the hypogynial valve is also provided with an extension, but here the extension is much bigger than in Mediotipula and Savtshenkia, and is placed near midlength of the sheath (Byers 1961, Theowald 1957, 1967, Wood 1952). Dolichopeza probably is not closely related to Mediotipula and Savtshenkia, and the extension of the sheath of the hypogynial ho NI un TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 st7 midv area 188 tg9 tg10 st10 o gonst i gonst aed gd gon fragm comp apod aed ant apod endoph post apod spmth gen fk st8 st9 hyp vlv Figs. 3-6. Tipula (Savtshenkia) rufina rufina. — 3, male terminalia, lateral view; 4, aedeagal guide with right gonapophysis and fragmentum, posteromedial view; 5, sperm pump, lateral view; 6, female terminalia, lateral view. Abbreviations: aed: aedeagus; aed gd: aedeagal guide; ant apod: anterior apodeme; cerc: cercus; comp apod: compressor apo- deme; endoph: endophallus; fragm: fragmentum; gen fk: genital fork; gon: gonapophysis; goncx: gonocoxite; hyp vlv: hypo- gynial valve; i gonst: inner gonostylus; midv area: midventral area: o gonst: outer gonostylus; post apod: posterior apodeme; pret: proctiger; sp2: lateral part of genital bridge; spm pmp: sperm pump; spmth: spermatheca; st7 etc.: sternite 7 etc.; sut x: suture x; tg7 etc.: tergite 7 etc. valve in Dolichopeza supposedly is not homologous to that of Mediotipula and Savtshenkia. Theowald (1978) accepted Savchenko’s views regarding a sister group relationship of Mediotipula and Savtshenkia. Another character state which could be indicative of a close phylogenetic relationship of Mediotipula and Savtshenkia is the shape of wing cell m1. In both sub- genera this cell is ‘bell-shaped’, narrowing towards the wing margin, whereas it usually widens towards the wing margin in other Tipulidae (character 4, below). The genital fork in both Mediotipula and Savtshenkia is broadest at its anterior part, a presumed apomor- phy within Tipulidae (character 82, below). 276 Mediotipula at present contains 12 recognized species with a primarily Mediterranean distribution (Oosterbroek & Theowald 1992). Conflicting with a presumed sister group relation- ship of Mediotipula and Savtshenkia is the presence of a separate sclerite lateral of the gonapophysis in Savtshenkia. As stated above (section Material, meth- ods and terminology’), I consider this sclerite homol- ogous with the so-called fragmentum as found in oth- er Tipulidae. The presence of a fragmentum is a derived character state within the Tipulidae and could substantiate a phylogenetic relationship of Savtshenkia with the complex of Lunatipula and allies (here after referred to as the Lunatipula group). The fragmentum is absent in Mediotipula. The Lunatipula group contains about 1200 species of Tipulidae, that is circa one fourth of the total number of species of the family. Besides the subgenus Zunatipula Edwards, the Zunatipula group includes the following subgene- ra of Tipula: Beringotipula Savchenko, Eremotipula Alexander, Eumicrotipula Alexander, Hesperotipula Alexander, Labiotipula Alexander, Lindnerina Mannheims, Odonatisca Savchenko, Pectinotipula Alexander, Pterelachisus Rondani, Ramatipula Alexander, Serratipula Alexander, Seritipula Alexander, Triplieitipula Alexander, and Vestiplex Bezzi. The phylogenetic relationships of the species of the Lunatipula complex are still largely unresolved, but it is nevertheless clear that not all of the subgene- ra in their present delimitation represent monophy- letic taxa (e.g., Lunatipula and Pterelachisus). In the discussion of the characters, I will refer to the species of the Zunatipula group and Mediotipula as outgroups of Savtshenkia. CHARACTER DISCUSSION In this section I will discuss the characters used in the phylogenetic analysis of Savtshenkia. Preceding a more detailed discussion of each character, a couplet shortly defines the alternative character states and gi- ves their codes. The character state matrix is presen- ted in table 4. Although I did not examine the male of gimmerthali mattheyi, this subspecies is, according to its authors, very similar to the nominotypical form, differing only in minute discriminating characters. Therefore, I feel safe to copy the score of the male characters of gimmerthali gimmerthali for this subspe- cies. In the discussion of the characters below, the re- sulting strict consensus tree (fig. 2) will often be men- tioned to point out the species which show a particular character state. I have done this in order to save space and for convenience, rather than presen- ting a list of species for each of the distinguished sta- tes. The results of the phylogenetic analysis are dis- cussed in the next section. Species and subspecies of Savtshenkia which for various reasons are not inclu- ded in the character discussion below, are dealt with in Appendix B. A list of autapomorphies of the spe- cies and subspecies of Savtshenkia is presented in Appendix C. Antenna 1. Flagellum: (0) well developed in male; (1) short, about the length of head or shorter. The male antenna of most species of Savtshenkia is well developed and relatively long, with the flagel- lomeres being about 2.5 times as long as their diame- ter at the greatest width, or longer (figs. 7, 11, 171). H. DE Jong: Phylogeny of Tipula (Savtshenkia) A similar state represents the presumed groundplan condition of the Lunatipula group and Mediotipula. A few members of Savtshenkia are characterized by a conspicuously short flagellum in the male. The flagel- lomeres of the pertaining species are about two times as long as their diameter or shorter (fig. 8). The spe- cies with exceptionally short male antennae are atlas, breviantennata, cyrnosardensis, and villeneuvii. 2. — Flagellum: (0) male flagellomeres well developed; (1) male flagellomeres longer than standard length. As noticed under the previous character, the flagel- lomeres of the males of Savtshenkia are usually about 2.5 times as long as their diameter at greatest width or slightly longer. Flagellomeres distinctly longer than this are found in alpha, grisescens, holoptera, pagana, and most Afrotropical species (the clade elgonensis to chrysocephala; figs. 9, 10, 152). In the pertaining spe- cies, the male flagellomeres are about 5 to 6.5 times as long as wide. The only Afrotropical species of Savtshenkia of which the male antennae are not con- spicuously long are hancocki and nephrotomoides (and the unexamined imperfecta which doubtfully belongs to Savtshenkia, see Appendix B). 3. — Flagellum: (0) male flagellomeres with slightly enlarged base, remainder cylindrical; (1) flagellomeres more or less reniform. In the genus Tipula sensu lato, the flagellum is usu- ally composed of eleven segments of which the inter- mediate ones have a slightly bulbous base and a cylin- drical apical part (figs. 7-10, 152, 171). This character state probably represents the groundplan condition of the Lunatipula group and Mediotipula. The same state is also found in most species of Savtshenkia. The species lundbladi and multipicta dif- fer from the other species of Savtshenkia by having the intermediate flagellomeres more or less reniform (fig. 11). In the Lunatipula group, a similar form of the flagellomeres is present in most species of the subge- nus Vestiplex. These species appear, however, to rep- resent a more derived group within Vestiplex. Wing 4. — Cell ml shape: (0) widening towards wing margin; (1) constricted before wing margin. In the Tipulidae veins M1 and M2, enclosing cell ml, usually diverge towards the wing margin (fig. 12). This situation probably represents the ground- plan condition of the Lunatipula group, which con- tains only a few species that have cell ml constricted near the margin of the wing. In most species of Mediotipula, cell ml is bell-shaped or paunchy (‘bauchig’, Mannheims & Pechlaner 1963; Theowald 1973) with veins M1 and M2 approaching towards DIT TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 È = EG a Figs. 7-15. — 7-11, male antenna; 7, Tipula (Savtshenkia) hartigiana; 8, T. (S.) atlas 9, T. (S.) grisescens 10, T. (S.) elgonensis, 11, 7. (5) lundbladi, 12-15, tip of wing; 12, T. (S.) grisescens, 13, T. (S.) goriziensis 14, T. (S.) cheethami, 15, T. (S.) akeleyi. Abbreviations: M1 etc.: medial vein 1 etc.; R4 etc.: radial vein 4 etc.; r-m: radial-medial crossvein. the margin of the wing, a state which probably repre- sents the groundplan condition of Mediotipula. The majority of species of Savtshenkia also show this latter character state (figs. 13-15). Only a few species of Savtshenkia have veins M1 and M2 diverging towards the wing tip, viz. alpha, glaucocinerea, grisescens, and postposita. The species subnodicornis usually has cell ml narrowing, but specimens of this species are known which show a widening cell ml. 278 5. — Cells of wing tip: (0) membrane covered with microtrichia only; (1) membrane partly covered with macrotrichia. Most species of Tipulidae have the membrane of the wing tip covered with microtrichia only, appear- ing bare under low magnification (figs. 12, 13, 15). This situation is found in the species of the Lunatipula group, Mediotipula, and in most species of Savtshenkia. The species cheethami and simulans are H. DE Jong: Phylogeny of Tipula (Savtshenkia) 25 SORTE | 3 31 Figs. 16-31. — 16, Tipula (Savtshenkia) grisescens, aedeagal guide with right gonapophysis and fragmentum, posteromedial view; 17, 7. (S.) hartigiana, aedeagal guide with right gonapophysis and fragmentum, posteromedial view; 18, 7. (S.) atlas, aedeagal guide with right gonapophysis, posteromedial view; 19, 7: (S.) draconis, aedeagal guide with gonapophyses, ventral view after microscopic slide holotype; 20, 7: (S.) aberdareica, aedeagal guide with right gonapophysis, een view; 21, T. (S.) obsoleta, aedeagal guide with right gonapophysis and fragmentum, posteromedial view; 22, 7: (S.) fragilis, aedeagal gui- de with right gonapophysis and fragmentum, posteromedial view; 23, 7. (S.) limbata, aedeagal Le ch right gonapophy- sisand fragmentum, posteromedial view; 24, T° (S.) pechlaneri, aedeagal guide with right gonapophysis and fragmentum, pos- teromedial view; 25-27, 7. (S.) haennii, 25, aedeagal guide, tip, posterior view; 26, right gonapophysis, medial view: 7, right fragmentum, posteromedial view; 28-30, 7: (S.) corsosignata, 28, aedeagal guide, tip, posterior view; 29, right ar sis, medial view; 30, right fragmentum, posteromedial view; 31, 7. (S.) subsignata subsignata, right gonapophys sis, medial view. unique within Savtshenkia by having macrotrichia on the membrane of the wing tip. The macrotrichia cov- er the wing to a larger extent in simulans than in chee- thami. The density of the macrotrichia is also higher in the former species. Macrotrichia are usually re- stricted to cells r4, r5, and ml in cheethami (fig. 14), while in simulans cell m2 and the tip of cell m3 are al- so provided with macrotrichia. 6. — Crossvein r-m: (0) apical section directed towards wingtip, or at most transverse to longitudinal axis of wing; (1) apical section directed towards base of wing. In Tipulidae crossvein r-m usually lies in an oblique position relative to the longitudinal axis of the wing, with its apical part directed towards the wing tip. This situation is found in the species of the Lunatipula group, Mediotipula, and in the majority of species of Savtshenkia (figs. 12-14). Three species of Savtshenkia have the apical end of r-m directed to- wards the base of the wing, viz. akeleyi, caligo, and chrysocephala (fig. 15). TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Male terminalia 7.— Aedeagal guide, apex: (0) tapering towards tip; (1) at tip with flaring rim. As is usual in the Tipulidae, the apex of the aedea- gal guide is the narrowest part of this structure in the species of the Lunatipula group, Mediotipula, and the majority of species of Savtshenkia (figs. 4, 16, 17, 19, 20, 22-25, 28, 162, 182, 194). The species atlas, breviantennata, and obsoleta have the apex of the ae- deagal guide bent outwards as a flaring rim (figs. 18, 21). 8. — Aedeagal guide, apex: (0) gradually tapering towards tip; (1) subapically abruptly narrowed, terminating in a slender point. The aedeagal guide in Tipulidae normally gradual- : ly narrows towards its apex, as it does in almost all species of Savtshenkia (caudal view, fig. 25). The apex of the aedeagal guide of corsosignata, subsignata sub- signata, and subsignata cazorla differs from that of the other species of Savtshenkia by the abruptly constrict- ed apex which terminates in a slender tip (fig. 28). 9. — Aedeagal guide, gonapophysis: (0) well developed; (1) very short or almost completely absent. Gonapophyses are usually well developed in the Tipulidae, a state which probably represents the groundplan condition of the Zunatipula group and Mediotipula. The majority of species of Savtshenkia also have well developed gonapophyses which occur in a wide range of sizes and shapes (figs. 4, 17-24, 26, 29, 31, 182, 194). The gonapophyses are very short or almost completely absent in the species alpha, alpı- um, boreosignata, grisescens, interserta, and macarones- ica (figs. 16, arrow, 162). 10. — Aedeagal guide, gonapophysis: (0) dorsal margin evenly fused with base of aedeagal guide; (1) anterodorsally slightly extended, thus with free-lying dorsal tip. In general, the dorsal margin of the gonapophysis in the Tipulidae gradually merges with the base of the aedeagal guide. This situation probably represents the groundplan condition of the Lunatipula group and Mediotipula. The gonapophyses of most species of Savtshenkia also show this character state (figs. 16, 21-24, 162, 182, 194). The species of the clade har- tigiana to simulans are characterized by their typically shaped gonapophyses, which differ from the gona- pophyses of the other species of Savtshenkia by the presence of a free anterodorsal extension (figs. 17, ar- row, 18). 11. — Aedeagal guide, gonapophysis: (0) pubescent or short haired; (1) long haired. 280 The covering of the gonapophyses in the Tipulidae usually consists of pubescence or short hairs, as is the presumed groundplan state of the Lunatipula group and Mediotipula. A similar condition is found in the majority of species of Savtshenkia (figs. 4, 17-19, 21- 24). A number of Afrotropical species of Savtshenkia are distinguished from the remainder of the subgenus by the dense and elongate hairing of the gonapophy- ses, viz. aberdareica, akeleyi, asbolodes, caligo, chrysoce- phala, and hancocki (fig. 20). 12. — Aedeagal guide, gonapophysis: (0) present, variously shaped; (1) a large and elongate structure. The gonapophyses of the Tipulidae show a wide range of shapes. When present, they are usually well developed and rather slender. This is the state that oc- curs in most of the species of Savtshenkia and is the presumed groundplan condition of its outgroups (- figs. 4, 17-21, 23, 24, 182, 194). The Nearctic species fragilis and phoroctenia differ from the majority of species of Savtshenkia by their extremely large and el- ongate gonapophyses (fig. 22). The only other species of Savtshenkia with similarly enlarged gonapophyses is the western Mediterranean cyrnosardensis. 13. — Aedeagal guide, gonapophysis: (0) a relatively slender structure; (1) a large blade. As noted under the previous character, the plesio- morphous condition of the gonapophyses in Savtshenkia is probably represented by a well devel- oped, rather slender structure (figs. 4, 17-21, 23, 24, 182, 194). Large mediolaterally flattened gonapophy- ses are found in corsosignata, haennii, subsignata sub- signata, and subsignata cazorla (figs. 26, 29, 31; see al- so next character). 14. — Aedeagal guide, gonapophysis: (0) at apex at most moderately enlarged; (1) dorsally and ventrally extended and thus widening towards broad apex. The two subspecies subsignata subsignata and sub- signata cazorla are characterized by their expanded gonapophyses (fig. 31). This character state corre- sponds with character 6 of Dufour (1991). 15. — Aedeagus: (0) short, directly curving from sperm pump to aedeagal guide, anteriorly reaching as far as abdominal segment 8; (1) elongate, anteriorly running to segment 7 or beyond. A relatively short aedeagus which loops in the me- dian sagittal plane from the sperm pump through ab- dominal segment 8 and from there to the aedeagal guide, is the presumed groundplan state of this fea- ture in the Lunatipula group. The species of Mediotipula also show this character state, as do a number of species of Savtshenkia (figs. 3, 155). Other species of Savtshenkia have a longer aedeagus. The ae- deagus in these species reaches anteriorly at least to segment 7 and often lies coiled inside the abdomen (- figs. 173, 185). The species involved constitute the clades confusa to aster and fragilis to omega. Extremely long aedeagi were found in corsosignata, ornata, and staegeri, species in which the aedeagus reaches abdom- inal segment 1. The species pair boreosignata and ele- onorae is characterized by a peculiarly modified aedea- gal complex (character 52). Although belonging to the clade fragilis to omega, they have a short aedeagus. 16. — Fragmentum: (0) either absent, or posterior margin convex or slightly concave; (1) posterior margin subapically with sharp emargination. The fragmenta of the species of the Zunatipula group and Savtshenkia occur in a diversity of forms. Usually, the posterior margin near the dorsal tip of the structure is convex. This is the presumed ground- plan condition of the Zunatipula group and the com- mon character state in Savtshenkia (figs. 4, 16, 17, 22- 24, 27, 30, 162, 182, 194). (The fragmentum is absent in the species of Mediotipula.) The fragmenta of the species holoptera, obsoleta, and pagana differ from the standard situation of Savtshenkia by the presence of a distinct emargination just below the dorsal apex (fig. 21, arrow). 17. — Fragmentum: (0) either absent, or anterior part sclerotized as the remainder of the structure; (1) anterior part dorsally blackish sclerotized. In general, the fragmentum is yellowish-brown in colour and uniformly sclerotized in the species of the Lunatipula group and Savtshenkia (figs. 4, 16, 17, 21, 22, 24, 162, 182, 194). A unique state is found in the species fragilina, limbata, and subvafra, where the dorsal margin of the anterior part of the fragmentum is blackish sclerotized (fig. 23). 18. — Fragmentum: (0) either absent, or anterior part a rounded lobe; (1) anterior part a separate anterocaudally flattened extension with slender tip. The fragmentum consists of a single undivided sclerite in the species of the Lunatipula group and most species of Savtshenkia (figs. 4, 16, 17, 21-24, 162, 182, 194). In corsosignata, haennii, subsignata subsignata, and subsignata cazorla, the anterior part of the fragmentum is characteristically modified in a slender tipped structure which stands apart from the posterior part (figs. 27, 30). The few hairs of the brush on top of the anterior part of the fragmentum are very short in these four taxa. 19. — Gonocoxite, midventral area: (0) variously shaped, usually tumescent; (1) with cone-shaped membranous extension. The midventral area in between the gonocoxites is H. DE Jone: Phylogeny of Tipula (Savtshenkia) represented by a tumescent structure in the presumed groundplan state of the Lunatipula group and Mediotipula. The majority of species of Savtshenkia also have a tumescent midventral area (fig. 32), but a number of species show a modification of this part of the terminalia. The species atlas, breviantennata, chee- thami, hartigiana and simulans have the midventral area provided with a ventrally directed cone-shaped membranous projection. The projection is relatively low in atlas and breviantennata (fig. 34), in the other species listed it is longer and more slender (figs. 33, 35). The structure is placed near the posteroventral margin of the gonocoxites in cheethami (fig. 35), in the four other species it has a more anterior position (figs. 33, 34). Other species of Savtshenkia with a projection to the area in between the gonocoxites are alpium, aspro- montensis, macaronesica, ornata, phoroctenia, and a few Afrotropical species. As the actual shapes of the projections differ notably among these species, a number of different character states are distinguished. In the species alpium and macaronesica, the append- age is membranous, dorsoventrally flattened and midventrally placed (character 21; fig. 39), in ornata it is swollen, sclerotized and posteriorly projecting (fig. 40), in phoroctenia it is slender, sclerotized and posteriorly projecting (fig. 41). The modifications of the midventral area in ornata and phoroctenia are characteristic of the respective species and represent autapomorphies (see Appendix C). Males of the spe- cies aspromontensis originating from Greece (Peloponnisos, Crete) have the midventral area poste- riorly terminating in a small and low extension. A similar extension is absent in specimens of aspro- montensis from the type locality in southern Italy. I could not find other distinguishing characters between male specimens of the populations from Greece and from Italy and consider them conspecific. The Afrotropical species of Savtshenkia with a modi- fied midventral area include akeleyi, caligo, chrysoce- phala, and nephrotomoides. These species are charac- terized by a relatively large and bulbous extension to the midventral area (next character). 20. — Gonocoxite, midventral area: (0) posterior part not bulbous; (1) bulbous. As noted under the previous character, the midven- tral area in Savtshenkia usually is tumescent. The Afrotropical species caligo, chrysocephala, nephrot- omoides, and, to a lesser extent, akeleyi, are distin- guished by a posteroventrally projecting lobe near the posterior margin of the midventral plate. The lobe is large in the first three species listed (fig. 36), in akeleyz it is relatively low (fig. 38). 21. - Gonocoxite, midventral area: (0) without 281 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 32-45. — 32-41, male sternite 9, left gonocoxite and midventral area, lateral view; 32, Tipula (Savtshenkia) grisescens, 33, T. (S.) hartigiana; 34, T. (S.) atlas, 35, T. (S.) cheethami, 36, T. (S.) nephrotomoides, 37, T. (S.) aberdareica; 38, T. (S.) ake- leyi, 39, T. (S.) alpium; 40, T. (S.) ornata; 41, T. (S.) phoroctenia; 42-45, male sternite 9, gonocoxites and midventral area, ventral view; 42, 7. (S.) baltistanica; 43, T. (Sì) lundbladi; 44, T. (Sì) nielseni: 45, T. (Sì) haennii. Abbreviations: midv area: midventral area; sut x: suture x. flattened appendage; (1) at about midlength with dorsoventrally flattened appendage. As stated above, the midventral area in between the gonocoxites is tumescent in most species of Savtshenkia. The species alpium and macaronesica are distinguished from the other species of Savtshenkia by the presence of a dorsoventrally flattened membra- 282 nous appendage at about midlength of the midventral area. The appendage is long in alpium (fig. 39), in macaronesica it is considerably shorter. (The unexam- ined subalpium has a small tongue-shaped appendage to the midventral area and is probably closely related to alpium and macaronesica; see Appendix B). H. DE Jon: Phylogeny of Tipula (Savtshenkia) Figs. 46-51. Male tergite 9-10 and dorsal part of left gonocoxite plus appendages, lateral view. — 46, Tipula (Savtshenkia) har- tigiana; 47, T. (S.) tulipa; 48, T. (S.) haennii, 49, T. (S.) boreosignata; 50, T. (S.) benesignata; 51, T. (S.) persignata tofina. 22. — Gonocoxite, midventral area: (0) broad and usually separated from gonocoxites by membranous zone; (1) narrow and fused with gonocoxites. The midventral area is probably broad in the groundplan states of the Lunatipula group and Mediotipula. The majority of species of Savtshenkia also have a broad midventral area in between the gon- ocoxites (figs. 42-44). It either is membranous, or partly or entirely sclerotized. In most Savtshenkia spe- cies with a sclerotized midventral area, a membranous zone (partly) separates the plate from the gonocoxites. The species pechlaneri and the members of the clade haennii to omega have a slender parallel sided and sclerotized midventral area which is largely fused with the ventromedial margins of the gonocoxites (fig. 45). 23. — Gonocoxite, posterodorsal margin: (0) with scattered short setae; (1) with concentration of long and strong setae. The gonocoxites of most species of Tipulidae are covered with scattered setae, a situation which prob- ably represents the groundplan condition of this char- 283 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 ant pt È: lat rdg post pt LIO 55 Figs. 52-63. Left inner gonostylus. — 52, Tipula (Savtshenkia) grisescens, dorsal view; 53, 7. (S.) villeneuvii, dorsal view; 54, 7. (S.) goriziensis, dorsal view; 55, 7. (S.) rufina rufina, lateral view; 56, 7. (S.) aberdareica, lateral view; 57, T. (S.) akeleyi, lateral view; 58, 7. (S.) confusa, lateral view; 59, 7. (S.) ornata, lateral view; 60, 7. (S.) serrulifera, lateral view; 61, 7. (S.) aspromon- tensis, lateral view; 62, 7. (S.) fragilis, medial view; 63, 7. (S.) odontostyla, medial view. Abbreviations: ant pt: anterior part; lat rdg: lateral ridge; post pt: posterior part. acter for the Lunatipula group and Mediotipula. It is also found in most species of Savtshenkia (figs. 3, 46, 48, 49, 155). Within Savtshenkia, the species of the clades tulipa to simulans and benesignata to omega are distinguished by a concentration of strong setae along the posterodorsal margin of the gonocoxites (figs. 47, arrow, 50, 51, 173, 185). The setae are placed in two or three irregular rows. They extend dorsally towards the membranous suture separating the gonocoxite from tergite 9-10 in the species of the clade tulipa to simulans and in benesignata (figs. 47, 50). In the spe- cies of the clade persignata to omega, the setae are con- fined to a more ventral position (fig. 51, 173, 185). 24. Gonocoxite, posterodorsal margin: (0) posterodorsal corner rounded off; (1) with posterior projection. The posterodorsal corner of the gonocoxites is pre- sumably rounded off in the groundplan condition of the Zunatipula group, as it is in the majority of species of Savtshenkia. In the species of Mediotipula and a number of species of Savtshenkia, the posterodorsal margin of the gonocoxite is extended into a more or less well developed projection (fig. 48, arrow). The species of Savtshenkia which show this character state 284 are those of the range fragilina to subsignata cazorla. 25. — Gonocoxite, suture x: (0) angle between suture x and the median line about 45° or less; (1) angle between suture and median line much wider, up to 90°. Suture x usually stands oblique to the median line of the body under an angle of about 45° or less in the Tipulidae. This probably is the groundplan condition of the character for the Zunatipula group and Mediotipula, and is the general state in Savtshenkia (- figs. 42, 44, 45). In the species lundbladi, multipicta, rufina, and the Afrotropical members of the subge- nus, the angle is much wider, often reaching a magni- tude of about 90° (fig. 43). (The direction of suture x could not be adequately studied in the celluloid prep- aration of the holotype of elgonensis, the only known specimen of this species.) 26. — Gonocoxite, suture x: (0) narrow; (1) broad. Suture x usually is rather narrow in the Tipulidae, a condition considered the groundplan state of the Lunatipula group and Mediotipula. A narrow suture x is also found in the majority of species of Savtshenkia (figs. 3, 32-39, 41-44). The species and subspecies corsosignata, haennii, subsignata subsignata, and sub- signata cazorla difter from the other members of Savtshenkia by their broad suture x (fig. 45, arrow). The suture is straight in these four taxa. Other species of Savtshenkia with a broad suture x are fragilis and ornata. In fragilis, suture x is sigma-shaped, in ornata it is, as many other genital features in this species, very different from that of the remainder of species of Savtshenkia. In ornata, suture x encompasses an iso- lated sclerite (fig. 40, arrow; see Appendix C). . 27. — Inner gonostylus, anterior part: (0) ventral surface smooth, or at most rugged; (1) serrate. The ventral surface of the anterior part of the inner gonostylus usually is smooth in Tipulidae. It is con- sidered the groundplan state of the Lunatipula group and Mediotipula, and is found in the majority of spe- cies of Savtshenkia (all figures that depict the inner gonostylus in side view, except 57). Three Afrotropical members of Savtshenkia are character- ized by a partly serrate lower margin of the anterior part of the inner gonostylus, viz. akeleyi, caligo, and chrysocephala (fig. 57, arrow). 28. — Inner gonostylus, anterior part: (0) crest absent on posterodorsal margin of anterior part; (1) crest present. The anterior part of the inner gonostylus probably lacks a crest in the groundplan state of the Zunatipula group. All species of Mediotipula are devoid of a crest on this part, as are most species of Savtshenkia (the majority of figures depicting the inner gonostylus). Three species of Savtshenkia have a well developed membranous crest on the posterodorsal margin of the anterior part of the inner gonostylus, viz. aspromonte- nsis, aster, and serrulifera. The crest is dorsally round- ed in aster and serrulifera (fig. 60), but it is grotesque- ly widening towards its slightly concave dorsal margin in aspromontensis (fig. 61). 29. — Inner gonostylus, anterior part: (0) anteriorly terminating in various ways; (1) anteriorly terminating in a prolonged nipple. The anterior part of the inner gonostylus shows a wide range of forms in the Zunatipula group, Mediotipula, and Savtshenkia. In Savtshenkia it is of- ten short and terminates anteroventrally in a more or less acute tip (the majority of figures showing the in- ner gonostylus). The species aspromontensis, aster, and serrulifera have a uniquely shaped anterior part which gradually tapers towards an anteriorly extended slen- der nipple-like projection (figs. 60, 61). The exten- sion is longest in aspromontensis (fig. 61) and aster. 30. — Inner gonostylus, anterior part: (0) tip single; (1) tip bifid. H. DE JONG: Phylogeny of Tipula (Savtshenkia) The anterior part of the inner gonostylus termi- nates in a single tip in the majority of Tipulidae, a sit- uation representing the presumed groundplan condi- tion of the Lunatipula group and Mediotipula. The majority of species of Savtshenkia also have the tip of the anterior part of the inner gonostylus undivided (- figs. 55-63, 68, 70-75, 161, 181, 193). The species corsosignata, nielseni, and the representatives of the clade fragilina to gimmerthali differ from the other species of Savtshenkia in having a bifid anterior tip of the inner gonostylus. The tip in most of these species shows a relatively broad dorsal and finer lateroventral point (figs. 64-67). In corsosignata (fig. 69), the small- er tip lies in a more lateral position compared with that of the other species with a bifid tip. 31. - Inner gonostylus, anterior part: (0) narrowing towards pointed tip; (1) abruptly bent, almost truncate at tip. The majority of species of Savtshenkia are charac- terized by a gradually curved and pointed tip of the anterior part of the inner gonostylus (lateral view; see figs. 55-72). The same situation probably is the groundplan condition of the Lunatipula group and Mediotipula. Within Savtshenkia, the species of the clade persignata to omega are distinguished from the remainder of the subgenus by the broad, almost par- allel-sided anterior part of the inner gonostylus which terminates in a rather steep front end (figs. 73-75, 181, 195): 32. — Inner gonostylus, base: (0) posteriorly rounded; (1) posteromedially terminating in pointed extension. Usually, the base of the inner gonostylus is round- ed at its posterior margin in the Tipulidae, a situation representing the presumed groundplan condition of the Lunatipula group and Mediotipula. It is also found in most species of Savtshenkia (dorsal view, fig. 52, arrow). A number of species of Savtshenkia are distinguished by the presence of an acute point at the posteromedial corner of the base of the inner gonos- tylus (figs. 53, arrow, 54). The species concerned con- stitute the clade hartigiana to simulans. 33. — Inner gonostylus, lateral ridge, ventral part: (0) without extensions; (1) with one or two acute blackish pointed extensions. The ventral part of the lateral ridge of the inner go- nostylus gradually merges with the main body of the inner gonostylus in most species of Savtshenkia. The three species cheethami, goriziensis, and simulans differ from the remainder of species of Savtshenkia by the presence of one or two acute blackish points at the end of the lower part of the lateral ridge of the inner gonostylus. The points are located near midheight of TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 64-75. Left inner gonostylus, lateral view. — 64, Tipula (Savtshenkia) limbata; 65, T. (S.) pechlaneri, 66, T. (S.) gim- merthali gimmerthali, 67, T. (5.) gimmerthali pteromaculata; 68, T. (S.) haennii, 69, T. (S.) corsosignata; 70, T. (S.) subsignata subsignata; 71, T. (S.) boreosignata; 72, T. (S.) benesignata; 73, T. (S.) persignata tofina, 74, T. (S.) signata 75, T. (S.) man- nheimsi. the body of the inner gonostylus. In goriziensis, a sin- gle point is present (fig. 54, arrow), while cheethami and simulans have two points, the posterior of which is bifid in simulans. 34. — Inner gonostylus, lateral ridge: (0) ventrally directed and terminating at about midlength of base of inner gonostylus; (1) in ventral part angularly 286 bent and anteriorly extended, terminating in front of base of inner gonostylus. The lateral ridge of the inner gonostylus is ventral- ly directed in the majority of species of Savtshenkia. The lateral ridge of subsignata subsignata and subsig- nata cazorla is present as a sharp edge directed to- wards the anterior margin of the inner gonostylus. The ridge makes a distinct angle in its posteroventral section near the base of the inner gonostylus (fig. 70). Dufour (1991, character 3) recognized the pres- ence of a ridge, on what he referred to as the median part of the inner gonostylus, as a synapomorphy of the subspecies subsignata subsignata and the then newly described subsignata cazorla and subsignata haennii. Furthermore, Dufour (1991, character 4) considered the continuity of the ventral and posterior parts of the ridge, as found in subsignata subsignata and subsignata cazorla, a synapomorphy of these two subspecies. Regarding the shape of the ridge in other species of Savtshenkia, it appears, however, that a con- tinuous ridge should be judged a plesiomorphy in subsignata subsignata and subsignata cazorla. Consequently and contrary to Dufour, I regard the separation of the posterior and ventral parts of the ridge in haennii an autapomorphy of that species (fig. 68; see Appendix C). 35. — Inner gonostylus, posterior part: (0) variously shaped; (1) posterolaterally extended, cup-shaped, laterodorsal margin blackened and serrate. The groundplan condition of the posterior part of the inner gonostylus in the Lunatipula group and Mediotipula is probably represented by a relatively simple and single structure. The posterior part is rounded and little developed in most species of Savtshenkia. The species confusa, eugeni, and jeekeli differ from the other species of Savtshenkia by the configuration of the posterior part. In these three spe- cies, it projects posterolaterally and carries acute pro- cesses along its blackish sclerotized laterodorsal mar- gin (fig. 58). The species confusa has a serrate and blackened rim running medially of and parallel to the posterolateral margin (fig. 58; see Appendix C). 36. — Inner gonostylus, posterior part: (0) variously shaped; (1) a posteriorly projecting, often rather acute, structure. As stated above, the posterior part of the inner go- nostylus is moderately developed and rounded off in most species of Savtshenkia. The members of the clade odontostylato omega are characterized by a more caudal development of the posterior part. In these species, the posterior part either is a rather slender and often somewhat acute structure (fig. 63), or a more enlarged and posteriorly produced extension (figs. 68-75, 181, 193; see also next character). A cau- dally produced posterior part of the inner gonostylus is also present in the species ornata (fig. 59). 37. — Inner gonostylus, posterior part: (0) a relatively small projection; (1) posteriorly extended, large. As mentioned under the previous character, the posterior part of the inner gonostylus is usually little developed in the species of Savtshenkia. The species H. DE Jone: Phylogeny of Tipula (Savtshenkia) benesignata, mannheimsi, omega, persignata, signata, and trinacria are distinguished among the species of the subgenus by a strikingly large posterior part which is of about the same length as the anterior part (figs. 72-75, 181, 193). The posterior part in these species is vertically orientated and lies more or less in line with the anterior part. 38. — Inner gonostylus, posterior part: (0) dorsal surface even; (1) dorsal surface corrugated. The dorsal surface of the posterior part of the inner gonostylus is even in the majority of species of Savtshenkia, as is the probable groundplan condition of the Lunatipula group and Mediotipula. Within Savtshenkia the dorsal surface of this part of the inner gonostylus is corrugated in both persignata persignata and persignata tofina (fig. 73). 39. — Inner gonostylus, posterior part: (0) variously shaped; (1) anterior section anterodorsally extended towards anterior part of inner gonostylus. As noted under character 37, a number of species of Savtshenkia have a caudally produced and large posterior part of the inner gonostylus. The extended posterior part of the inner gonostylus is low in its an- terior half in benesignata, persignata persignata, and persignata tofina (figs. 72, 73), as it is in the other spe- cies of Savtshenkia. In the species mannheimsi, omega, signata, and trinacria it is anterodorsally protruding (figs. 74, 75, 181, 193). 40. — Inner gonostylus: (0) variously shaped; (1) similarly shaped, with high anterior part and blackish sclerotized triangular projection posterolaterally. The inner gonostylus shows a wide variety of forms throughout Savtshenkia. The two subspecies rufina rufina and rufina maderensis display a remarkable similarity in the structure of the inner gonostylus which differs most notably from that of other species of Savtshenkia by the shape of the anterior part and the location of the posterolateral projection (fig. 55). 41. — Inner gonostylus: (0) variously shaped; (1) similarly shaped, anterior part anterodorsally straightened, gradually narrowing towards bifid tip. Throughout Savtshenkia a wide variety of forms of the inner gonostylus can be found. The three recog- nized subspecies of gimmerthali can be separated from the other species of the subgenus by the unique shape the inner gonostylus, of which the anterior part grad- ually narrows towards a bifid tip (figs. 66, 67). 42. — Inner gonostylus: (0) variously shaped; (1) similarly shaped, with long sloping caudodorsal margin. TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 76-99. — 76-88, left outer gonostylus, lateral view; 76, Tipula (Savtshenkia) grisescens 77, T. (S.) hartigiana; 78, T. (Sì) lundbladi; 79, T. (S.) multipicta; 80, T. (Sì) jeekeli, 81, T. (Sì) aspromontensis 82, T. (S.) fragilis 83, T. (S.) gimmerthali gim- merthali, 84, T. (S.) gimmerthali pteromaculata; 85, T. (S.) nielseni; 86, T. (Sì) sardosignata; 87, T. (Sì) signata; 88, T. (Sì) man- nheimsi, 89-99, sperm pump and appendages; 89, 7. (S.) subnodicornis, lateral view; 90, 7. (S.) baltistanica, lateral view; 91- 93, 7. (S.) alpium, 91, lateral view; 92, compressor apodeme, posterior view; 93, anterior view; 94-96, T. (S.) jeekeli, 94, lateral view; 95, compressor apodeme, posterior view; 96, anterior view; 97, T. (S.) confusa, lateral view; 98, T. (S.) graciae, la- teral view; 99, 7. (S.) boreosigata, lateral view. The species boreosignata and eleonorae are distin- guished among the species of Savtshenkia by their uniquely shaped and very similar inner gonostyli. The overall structure is rather long, with a well developed anterior part and a long caudodorsal margin which gradually slopes towards the low posterior part (fig. Zul): 43. — Outer gonostylus: (0) anterior margin convex 288 or slightly and evenly concave; (1) concave, with distinct bend at about two-fifth of length from base. The anterior margin of the outer gonostylus is ei- ther entirely convex or evenly concave in its lower half in the presumed groundplan condition of the Lunatipula group and Mediotipula, and in the major- ity of species of Savtshenkia (figs. 76, 78-88, 160, 180, 192). The three species atlas, breviantennata, and hartigiana have the anterior margin of the outer gonostylus angularly concave near two-fifth of its height (fig. 77, arrow). 44. — Outer gonostylus: (0) upright structure; (1) inclined forward. The outer gonostylus in Tipulidae usually is an up- right structure, as is the presumed groundplan condi- tion of the Lunatipula group and Mediotipula. The majority of species of Savtshenkia are also distin- guished by the presence of a vertically orientated out- er gonostylus (figs. 76, 77, 80, 82-88, 160, 180, 192). The outer gonostylus of the species aspromontensis, lundbladi, and multipicta leans over anteriorly (figs. 78, 79, 81). The actual shapes of the outer gonostyli of the three pertaining species differ notably. The base of the outer gonostylus is low and elongate in lundbladi (fig. 78) and multipicta (fig. 79), whereas it is rather short in aspromontensis (fig. 81). 45. — Outer gonostylus: (0) anterior part rather broad; (1) slender and elongate. In general, the outer gonostylus is present as a sin- gle oblong and rather broad structure in the Tipulidae, a situation which probably represents the groundplan condition of this character for the Lunatipula group and Mediotipula. A broad outer go- nostylus is found in about half of the species of Savtshenkia (figs. 76-81, 160). The species of the clade fragilis to omega are distinguished by the pres- ence of a rather slender and elongate anterior sub- structure of the outer gonostylus (figs. 82, 83, 85-88, 180, 192). A similar outer gonostylus with slender anterior section is found furthermore in the species eugeni and ornata. The subspecies gimmerthali pte- romaculata, which is included in the clade fragilis to omega, has a relatively wide anterior part of the outer gonostylus (fig. 84). 46. — Outer gonostylus: (0) base variously shaped, usually relatively narrow; (1) broad, laterally concave and blackish sclerotized. As noticed under the previous character, the outer gonostylus generally is an oblong structure in the Tipulidae. The base of the outer gonostylus is usually relatively narrow compared with the length of the structure, a situation present in the majority of spe- cies of Savtshenkia (fıgs. 76, 77, 80, 82-84, 87, 88, 160, 180, 192). The species nielseni (fig. 85) and sar dosignata (fig. 86) differ from the remainder of Savtshenkia by the presence of a large base of the out- er gonostylus. The base is laterally concave, partly blackish sclerotized and carries a strong posterior pro- jection. 47.- Sperm pump, endophallic lumen: (0) bulbous, ventrally expanded; (1) small. H. DE Jong: Phylogeny of Tipula (Savtshenkia) Throughout the Tipulidae, the sperm pump has a rather large and bulbous endophallic lumen, a situa- tion which probably represents the groundplan con- dition of the Lunatipula group and Mediotipula. Within Savtshenkia a similar condition is found in the species alpha, grisescens, ignobilis, subnodicornis, and the species of the clade hartigiana to simulans (figs. 89, 163). The majority of species of Savtshenkia, however, have an endophallic lumen that is hardly ventrally expanded (figs. 5, 90, 91, 94, 97, 98, 183, 195). The species concerned include the members of the clade baltistanica to omega. 48. — Sperm pump, compressor apodeme: (0) a flattened blade; (1) posteriorly provided with U- shaped or V-shaped brace. The groundplan condition of the compressor apo- deme of the Lunatipula group and Mediotipula prob- ably consists of a flattened blade with a more or less distinct dorsomedial emargination. Within Savtshenkia a similar compressor apodeme is found in the species that range from alpha to macaronesica (- figs. 92, 93, 164). The other species of Savtshenkia are distinguished by the presence of a U-shaped or V- shaped support on the posterior side of the compres- sor apodeme (fig. 95, arrow, 96, 184, 196). The spe- cies involved constitute the clade jeekeli to omega. 49. — Sperm pump: (0) ventral wall convex; (1) ventral wall flattened or concave. In Tipulidae, the body of the sperm pump usually is a bulbous structure which has a convex ventral wall. This situation probably reflects the groundplan con- dition of the Lunatipula group and Mediotipula and is found in about two-fifth of the species of Savtshenkia (the range alpha to macaronesica, frontal view, figs. 93, 164). Within Savtshenkia the bulb of the sperm pump has a flat or concave ventral wall in most spe- cies of the clade jeekeli to omega (figs. 96, 184, 196). There are, however, a few species within this latter clade which have a convex ventral wall of the sperm pump, viz. aspromontensis, eugeni, and interserta. 50. — Sperm pump, posterior apodemes: (0) well developed and posteriorly or posterodorsally directed; (1) elongate and anteriorly extended. In Tipulidae, the posterior apodemes of the sperm pump are generally well developed, rather broad and posteriorly or posterodorsally directed. This character state probably represents the groundplan condition of the Lunatipula group and Mediotipula. The majority of species of Savtshenkia also display this character state (figs. 5, 89-91, 94, 163, 183, 195). In contrast, the members of the clade confusa to aster are distin- guished by the presence of elongate and anteriorly ex- tended posterior apodemes. The tips of the posterior 289 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 100-112. — 100, 101, sp2 and part of tergite 9-10, posteromedial view; 100, Tipula (Savtshenkia) odontostyla, 101, T. (S} limbata; 102-112, male sternite 8, posterior margin; 102, 7. (S.) grisescens, ventral view; 103, 7. (S.) lundbladi, ventral view; 104, 7. (S.) rufina rufina, posterior view; 105, 7. (S.) alpium, ventral view; 106, 7. (S.) obsoleta, ventral view; 107, 7. (S.) holoptera, ventral view; 108, T. (S.) confusa, ventral view; 109, T. (S.) aster, ventral view; 110, 7. (S.) limbata, ventral view; 111, T. (5) pechlaneri, ventral view; 112, T. (S.) invenusta invenusta, ventral view. Abbreviations: sp2: lateral part of genital bridge; tg9-10: tergite 9-10. apodemes in these species lie alongside the compres- sor apodeme (figs. 97, 98). 51. — Sperm pump, posterior apodemes: (0) well developed, broad; (1) slender. As noticed under the previous character, the poste- rior apodemes of the majority of species of Savtshenkia are well developed and broad (figs. 5, 89- 290 91, 94, 98, 163, 183, 195). Three species are distin- guished from the remainder of the subgenus by their long and slender posterior apodemes, viz. confusa, interserta, and ornata (fig. 97). The posterior apo- demes of these three species are anteriorly extended (see previous character). 52. — Sperm pump: (0) general shape of the standard tipulid type; (1) general shape different, partly heavily sclerotized. The sperm pump in Tipulidae usually consists of a bulbous main body, which is surrounded in the hori- zontal plane by a pair of anterior and a pair of poste- rior apodemes, and perpendicular and dorsal to these a single compressor apodeme. The anterior and pos- terior apodemes are attached to the main body of the sperm pump at about midheight and the aedeagus leaves the sperm pump in between the anterior apo- demes. This configuration doubtlessly represents the groundplan condition of the Lunatipula group and Mediotipula. The majority of species of Savtshenkia also show a sperm pump of this form (figs. 5, 89-98, 163, 164, 183, 184, 195, 196). However, the species boreosignata and eleonorae possess a peculiarly built sperm pump, which has the heavily sclerotized main body lying just dorsal of (boreosignata, fig. 99) or on (eleonorae) a membranous disc-like structure which contains the remnants of the anterior and posterior apodemes. The heavily sclerotized compressor apo- deme seems to be fused with the main body of the pump. In boreosignata, the short aedeagus appears to be attached to the ventral side of the main body, after which it makes a loop at the level of the disc-like structure containing the anterior and posterior apo- demes. Compared with the sperm pump of the other species of Savtshenkia, the whole structure is small in both boreosignata and eleonorae. 53. — Sp2: (0) uniformly and moderately sclerotized; (1) medially blackish sclerotized. Throughout the Tipulidae, sp2 generally is uni- formly sclerotized. This situation is found in the Lunatipula group and in Mediotipula and represents the groundplan state of this character for both taxa. All species of Savtshenkia as recognized in this paper have the medial margin of sp2 blackish sclerotized (figs. 100, 101, 159, 179, 191). The only other Tipulidae in which a similar situation was studied are species of the genus Nephrotoma Meigen and of the subgenus Tipula (Schummelia Edwards). In Nephrotoma, a partly blackish sclerotized sp2 can be found in for instance the Palaearctic aurantiocincta Alexander, biarmigera Alexander, medioproducta Alexander and the Oriental citricolor Alexander, inte- gra Alexander, progne Alexander, subumbonis Alexander, and umbonis Alexander. In the pertaining species of Nephrotoma, the general shape of sp2 differs from that of the species of Savtshenkia in that sp2 is short and broad, anteriorly notched, and rather deep- ly concave. All species of Schummelia studied (viz. ah- rensi Savchenko, butzi Edwards, variicornis Schummel, yerburi Edwards, zernyi Mannheims, and zonaria Goetghebuer) have the anteromedial margin of sp2 more or less darker sclerotized than the re- H. DE JONG: Phylogeny of Tipula (Savtshenkia) mainder of the structure. The general shape of sp2 in the species of Schummelia is similar to the elongate form as found in the majority of species of Savtshenkia. The phylogenetic positions of the per- taining species of Nephrotoma and Schummelia with- in the respective taxa are still unestablished. Except for the similar modification of sp2, there are at the moment, however, no indications to suggest that ei- ther Nephrotoma or Schummelia, or both combined, represent(s) the closest relative of Savtshenkia (see al- so next character). 54. — Sp2: (0) dorsally separate from tergite 9-10; (1) dorsally fused with ventromedial part of tergite 9-10. Sp2 and tergite 9-10 are separate from each other in the greater majority of Tipulidae, a situation con- sidered the groundplan condition of the Zunatipula group and Mediotipula. More than half of the species of Savtshenkia also show this character state (fig. 100, arrow). In the species of the clade fragilina to omega, sp2 is fused with the ventromedial sclerotization of tergite 9-10 (fig. 101). Both structures are firmly con- nected over their entire width. Outside Savtshenkia, I found a similar character state in species of the subge- nus Zipula (Schummelia) only. In the examined spe- cies of Schummelia (see previous character), sp2 and the ventral sclerotization of tergite 9-10 are approxi- mated. Some of these species show a partial fusion of sp2 and tergite 9-10 (viz. ahrensi, variicornis, and yer- buri). In zernyi, both structures are more firmly con- nected. The studied species of Schummelia all have a western Palaearctic distribution and represent only a minor fraction of the subgenus which contains about 75 species distributed further in the Afrotropical, Nearctic and especially Oriental Regions. Fusion of tergite 9-10 with sp2 probably evolved independent- ly in Schummelia. As stated under the preceding char- acter, I do not consider Schummelia a potential sister group of Savtshenkia. 55. — Sternite 8, posterior margin: (0) unmodified; (1) caudally extended. An unmodified sternite 8 in Tipulidae consists of a relatively short structure of which the posterior mar- gin is evenly convex or slightly concave and lacks any trace of spinosity. This situation probably represents the groundplan condition of this character for the Lunatipula group and Mediotipula. In a number of species of Savtshenkia, a modification of the posterior margin of sternite 8 is found, which generally consists of a posterior elongation of the structure (figs. 103, 105-1105 112-117, 173, 177; 4178"185, 18952190): Elongation of sternite 8 is usually combined with the presence of spines or strong setae along the posterior margin (character 57). A caudally produced posterior 291 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 113-128. — 113-117, male sternite 8, posterior margin; 113, 7. (S.) cyrnosardensis, posterior view; 114, 115, 7. (S.) ha- ennii, 114, posterior view; 115, ventral view; 116, 7. (S.) signata, posterior view; 117, T. (S.) mannheimsi, posterior view; 118- 120, male sternite 9, anteroventral appendage, dorsal view; 118, 7: (S.) sordidipes 119, T. (S.) eugeni, 120, T. (S.) limbata; 121-128, male tergite 9-10; 121, 122, T. (S.) grisescens, 121, dorsal view; 122, posterior view; 123, 7. (S.) hartigiana, dorsal view; 124, 7. (S.) atlas, dorsal view; 125, T. (S.) goriziensis, dorsal view; 126, T. (S.) cheethami, dorsal view; 127, T. (S.) ig- nobilis, dorsal view; 128, T. (S.) subnodicornis, dorsal view. margin of sternite 8 is found in the species lundbladi, multipicta, rufina, and those of the clade alpium to omega. Within the latter clade a short sternite 8 is present in aspromontensis, gimmerthali, interserta, pechlaneri, sordidipes, and venerabilis (see fig. 111). 56. — Sternite 8, posterior margin: (0) variously 292 shaped; (1) with V-shaped ventral carina. Sternite 8 is present as an evenly curved sclerite in the presumed groundplan condition of the Lunatipula group and Mediotipula. Within Savtshenkia a variety of forms of the posterior margin of sternite 8 is found. The two subspecies rufina rufi- na and rufina maderensis are distinguished from the remainder of the subgenus by the presence of a mid- ventral V-shaped carina in the posterior part of ster- nite 8 (posterior view, fig. 104). 57. — Sternite 8, posterior margin: (0) unarmed; (1) provided with strong setae or black spines. In most Tipulidae, sternite 8 is unarmed, a situa- tion considered to represent the groundplan state of this character for the Lunatipula group and Mediotipula. About half of the species of Savtshenkia have the posterior margin of sternite 8 provided with strong setae or short spines (figs. 105-109, 113-117, 177, 178, 189, 190). The number of spines or setae and their location varies among the different species of Savishenkia. The species with an armed posterior margin of sternite 8 include baltistanica, multipicta, and most species of the clade alpium to omega. Within the latter clade a number of species show an unarmed sternite 8, viz. interserta, macaronesica, orna- ta, and the species of the clade fragilina to invenusta (see figs. 110-112). 58. — Sternite 8, posterior margin: (0) variously shaped; (1) with medial tongue-like extension, posteriorly concave. As stated above, a number of species of Savtshenkia have a caudally produced posterior margin of sternite 8. The species holoptera, obsoleta, and pagana are dis- tinguished among these by the peculiarly shaped pos- terior extension of sternite 8, which appears as a tongue-like blade which is caudally more or less emar- ginate (figs. 106, 107). In most specimens of pagana, the caudal margin of the posterior extension is nearly straight, in the species holoptera and obsoleta it is dis- tinctly concave. 59. — Sternite 8, posterior margin: (0) unarmed or armed with more than 10 spinous setae; (1) number of spinous setae less, ranging from seven to one. The number of strong setae or spines on the poste- rior margin of sternite 8 in the species of Savtshenkia which have this structure armed usually is (consider- ably) more than 10 (figs. 105, 113-117, 177, 178, 189, 190). In the species obsoleta, there are about 13 to 15 spines on the posterior margin of sternite 8 (fig. 106). The species holoptera and pagana have less than 10 spines along the posterior margin of sternite 8. Sternite 8 of pagana usually has about four to eight spinous setae, in holoptera the number varies between one and three (fig. 108; see Eiroa, 1987). 60. — Sternite 8, posterior margin: (0) medial membranous area absent, small or, if large, without rim of longish setae; (1) membranous area broadly V-shaped, anteriorly extending, its margins set with long slender setae. H. DE Jone: Phylogeny of Tipula (Savtshenkia) As noticed above, the groundplan condition of ster- nite 8 for the Lunatipula group and Mediotipula prob- ably consists of an unmodified sclerite. The posterior margin of the sternite in this situation is uniformly sclerotized. In Savtshenkia, sternite 8 generally has no membranous area at its posterior margin or the mem- branous area is restricted to a narrow zone (figs. 102, 103, 105-107 dels 125158170189 "Ihe three species fragilina, limbata, and subvafra are distin- guished from the remainder of Savtshenkia by a large and V-shaped membranous area which is bordered by a fringe of thickly set longish slender setae (fig. 110). The species aster and confusa are the only other species of Savtshenkia that also have a relatively large membra- nous area on the posterior margin of sternite 8. In as ter, a V-shaped membranous area is present in between the caudolateral extensions, the extensions themselves are provided with a few black spines (fig. 109; see Appendix C). The species confusa has a broad and rounded membranous zone (fig. 108; see Appendix C). In both species, the margin of the mem- branous area is surrounded by scattered setae only. 61. — Sternite 8, posterior margin: (0) ventrally directed medial extension absent; (1) unarmed medial extension present rostrally of lateral extensions, or at this position with sclerotized edge. In the majority of species of Savtshenkia, the poste- rior margin of sternite 8 medially either is unmodified or produced into a rather broad lobe provided with spines or strong setae (figs. 102, 103, 105-111, 158). An unarmed medial extension rostrally of a pair of lateral extensions on the posterior margin of sternite 8 can be found in the three subspecies of invenusta and in the species of the clade nielseni to omega. The size, shape and sclerotization of the medial extension var- ies among the species involved. In the subspecies of invenusta, and in mannheimsi, omega, and trinacria the medial extension is relatively small and moderate- ly sclerotized (figs. 112, 117, 177, 178, 189, 190), it is large and partly membranous in zielseni, whereas it is more pronounced and heavily sclerotized in most other species involved (figs. 114, 116). It should be noted, however, that the species benesignata, boreosig- nata, and eleonorae lack a protruding medial exten- sion. In these three species, a sclerotized rim marks the position of the medial extension. 62. — Sternite 8, posterior margin: (0) extension(s) absent or dorsoventrally flat; (1) lateral extensions appearing as bulbous structures, densely set with spines. The extensions of sternite 8 are relatively flat and follow the curve of sternite 8 when seen in posterior view in about half of the species of Savtshenkia with a modified sternite 8. Most members of the clade cyr- 295 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 PITT Eli ery 6 AAN Sa) Ss = ee Figs. 129-141. Male tergite 9-10.— 129, 7. (S.) baltistanica, dorsal view; 130, 7. (S.) elgonensis, posterior margin, dorsal view, after celluloid preparation holotype; 131, 132, 7. (S.) aberdareica, 131, posterior view; 132, dorsal view; 133, T. (S.) sordidi- pes, dorsal view; 134, 7. (S.) serrulifera, posterior view; 135, 7. (S.) aspromontensis, posterior view; 136, 7. (S.) invenusta in- venusta, dorsal view; 137, 7. (S.) nielseni, dorsal view; 138, 7. (S.) boreosignata, dorsal view; 139, 7. (S.) persignata tofina, pos- terior view; 140, 7. (S.) signata, posterior view; 141, 7. (S.) mannheimsi, posterior view. nosardensis to omega are characterized by the presence of bulbously swollen lateral extensions of sternite 8. In the majority of these species, the lateral extensions are thickly set with black spines (figs. 113-116), but in the species boreosignata only a few spines on the lat- eral lobes are found. Moreover, the three species mannheimsi, omega, and trinacria have a relatively flat pair of lateral extensions (figs. 117, 178, 190; charac- ter 65). 294 63. — Sternite 8, posterior margin: (0) lateral extensions absent, or, if present, dorsally not connected; (1) dorsally connected by a sclerotized bridge. In most species of Savtshenkia which carry a pair of lateral extensions on the posterior margin of sternite 8, the extensions are dorsally separated by a membra- nous area (figs. 113, 116, 117, 178, 190). The species and subspecies corsosignata, haennii, subsignata subsig- nata, and subsignata cazorla are distinguished by the presence of a sclerotized dorsal bridge that connects the lateral extensions of sternite 8 (fig. 114). 64. — Sternite 8, posterior margin: (0) variously shaped; (1) in between medial and lateral extensions a pair of medially directed projections. The ventral margin of sternite 8 in between the medial and lateral extensions is straight or evenly curved in most species of Savtshenkia with a three- lobed sternite 8. As an extra feature, corsosignata, haennii, subsignata subsignata, and subsignata cazorla have an acute and dorsomedially directed sclerotized projection on each side of the medial extension (fig. 115, arrow). 65. — Sternite 8, posterior margin: (0) variously shaped; (1) lateral extensions dorsoventrally flattened, posterior margin straight, spines confined to posterior margin. The bulbously swollen lateral extensions of sternite 8 in most species of the clade cyrnosardensis to omega are more or less circular in cross section and covered with spines on their posterior half, dorsally as well as ventrally (figs. 113-116). The species mannheimsi, omega, and trinacria have the lateral extensions of sternite 8 dorsoventrally flatter and straight along their posterior margins (figs. 117, 177, 178, 189, 190). The spines in these three species are concentrat- ed along the posterior margin of the sternite. 66. — Sternite 9, midventral anterior appendage: (0) apically rounded; (1) apically bifid. Sternite 9 often carries a more or less developed an- terior midventral appendage in the Tipulidae. When present, this appendage has a rounded anterior mar- gin, as it has in the majority of species of Savtshenkia (figs. 118, 120). In the species aspromontensis, aster, eugeni, and serrulifera the anterior appendage has a bi- fid apex (fig. 119). 67. Sternite 9, midventral anterior appendage: (0) a plain structure; (1) dorsally provided with medial carina. In the presumed groundplan condition of the Lunatipula group and Mediotipula, the ventromedial part of sternite 9, including the anterior appendage, is a dorsoventrally flattened structure. The majority of species of Savtshenkia show an anterior midventral appendage of sternite 9 which looks like a simple tongue-like extension without further modifications (fig. 118). A different form of this structure is present in the species of the clade fragilina to omega, where the appendage is provided with a medial carina on its dorsal side (fig. 120). Although the height of the cari- na varies among the pertaining species, it is usually H. DE JONG: Phylogeny of Tipula (Savtshenkia) well developed and clearly visible. A similar anterior appendage was also found in the species hartigiana. Within the clade fragilina to omega only pechlaneri, sardosignata, and subsignata cazorla lack a distinct me- dial carina. 68. — Tergite 9-10, posterior margin: (0) extensions variously shaped; (1) narrowing toward caudally produced tip. In Tipulidae, tergite 9-10 often has a pair of low and rounded extensions at its posterior margin, a sit- uation that probably represents the groundplan con- dition of the Lunatipula group and Mediotipula. Within Savtshenkia, only a few species show similar extensions (figs. 121, 122, 125, 126, 127, 156, 157). The species concerned lie in the range alpha to ignob- ilis, with the exception of the species atlas, brevianten- nata, and hartigiana. In the three latter species, the extensions are caudally produced and constricted be- fore their tips. The medial margins of the extensions in these species are slightly concave in dorsal view (figs. 123, 124). 69. — Tergite 9-10, posterior margin: (0) posteroventral sclerotization uninterrupted along posterior margin; (1) separated into two parts by medial incision. Tergite 9-10 in Tipulidae often has a sclerotized rim running along the ventral side of the posterior margin. An uninterrupted rim probably represents the groundplan state of this character for the Lunatipula group and Mediotipula. The majority of species of Savtshenkia also have a complete and broad posteroventral sclerotized rim along the posterior margin of tergite 9-10 (figs. 121, 123, 125-129, etc., after macerated specimens). Seen in dorsal view, this rim seems to be intersected by the narrow and deep medial incision of tergite 9-10 in the species atlas and breviantennata (fig. 124). 70. — Tergite 9-10, posterior margin: (0) tips of extensions broadly separate; (1) close together, separated by a narrow gap. A broad emargination or medial notch usually sep- arates the extensions on the posterior margin of ter- gite 9-10 in Tipulidae, a situation reflecting the pre- sumed groundplan condition of the Lunatipula group and Mediotipula. In the majority of species of Savtshenkia, the posterior extensions of tergite 9-10 are separated by a broad V-shaped or U-shaped emar- gination (figs. 121-125, 127, 129, etc.). In cheethami and simulans, the extensions approach each other me- dially and are separated by a rather deep and narrow fissure (fig. 126). 71. — Tergite 9-10, posterior margin: (0) extensions TijpsCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 148 Figs. 142-151. — 142, 143, genital fork, dorsal view; 142, Tipula (Savtshenkia) hartigiana, 143, T. (S.) atlas, 144, 145, ovi- positor, lateral view; 144, T° (S.) grisescens 145, T. (S.) hartigiana; 146, 147, spermatheca; 146, 7. (S.) hartigiana, 147, T. (S.) villeneuvit, 148, 149, female sternite 8 and hypogynial valves, ventral view; 148, 7. (S.) grisescens 149, T. (S.) ignobilis, 150, 151, female sternite 9, medial part, ventral view; 150, 7. (S.) jeekeli; 151, T. (S.) obsoleta. posteriorly directed and rounded at tip; (1) laterally curved and pointed at tip. As noted above, the groundplan condition of ter- gite 9-10 for the Lunatipula group and Mediotipula probably shows a pair of broad and rounded exten- sions on the posterior margin of the tergite. The ma- jority of species of Savtshenkia, however, have the tips of the extensions of tergite 9-10 pointed and out- wardly curved when seen in dorsal view (figs. 128- 130, 137, 139-141, 174, 175, 186, 187). 72. — Tergite 9-10, posterior margin: (0) provided with black spines; (1) spines absent. The section of the family Tipulidae to which Savtshenkia and its outgroups belong, is, among oth- er things, characterized by the presence of black spines on the posterior margin of tergite 9-10. Although such spines are absent in the majority of species of the Lunatipula group, there are several spe- cies within the group which have the posterior mar- gin of tergite 9-10 armed with spines. For this reason I, putatively, consider this latter state the groundplan 296 condition of the Zunatipula group. All species of Mediotipula and the majority of species of Savtshenkia also show a spinose posterior margin of tergite 9-10 (figs. 121-128, 133-136, 139, 140, 187). Within Savtshenkia, total absence of spines on the posterior margin of tergite 9-10 is found in the spe- cies alpha (figs. 156, 157) and the species of the clade baltistanica to chrysocephala (figs. 129-132). The spe- cies glaucocinerea has a few knob-like protuberances on the lower margin of the extensions of tergite 9-10, but lacks distinct spines. Spines along the ventral margin of the extensions of tergite 9-10 are absent in the species mannheimsi and omega, but in both species a few small spines are present in the dorsal part of the medial cleft (figs. 141, 175). 73. Tergite 9-10, posterior margin: (0) extensions variously shaped, usually long and laterally directed; (1) extensions present as short spineless points. The laterally directed extensions of the species of the clade subnodicornis to omega usually terminate in slender and elongate points (figs. 128-130, 137-141, 174, 175, 186, 187). A number of Afrotropical spe- cies of Savtshenkia are characterized by the presence of short and acute extensions, viz. the species aberda- reica, akeleyi, asbolodes, chrysocephala, draconis, and hancocki (figs. 131, 132). 74. — Tergite 9-10, posterior margin: (0) spines confined to the extensions; (1) spines present above medial notch. As is usual in Tipulidae which have tergite 9-10 provided with spines, the spines are concentrated on the posterior margin of the tergite in the majority of species of Savtshenkia (figs. 121-128, 139, 140, 175, 187). Spines are present also in the area above the me- dial notch of tergite 9-10 in the species aspromonte- nsis, aster, eugeni, odontostyla, serrulifera, sordidipes, and venerabilis (figs. 133-135). In the latter five spe- cies, there are about 10 spines in this area (figs. 133, 134), while aspromontensis and aster carry 20 to 30 spines above the notch (character 78). 75. — Tergite 9-10, posterior margin: (0) variously shaped; (1) broadly V-shaped emarginate with rounded corners. The Kashmirian species sordidipes and venerabilis differ from all other species of Savtshenkia by the shape of the posterior margin of tergite 9-10. Tergite 9-10 in both species has a broad V-shaped emargina- tion with a small median notch, while the posterior corners are broadly rounded off (fig. 133). The spe- cies sordidipes and venerabilis resemble each other very much, including in details of the male terminalia. Differences are found in the colour of the legs and ab- domen and in the height of the anterior part of the in- ner gonostylus. 76. — Tergite 9-10, posterior margin: (0) single pair of extensions present; (1) two pairs present. As in most other Tipulidae, the standard number of extensions on the posterior margin of tergite 9-10 in Savtshenkia is two (figs. 121-127, 129-133, 136- 141, 156, 157, 174, 175, 186, 187). The species ser- rulifera, aspromontensis, and aster differ from this by having a medial and a sublateral pair of extensions. Both pairs of extensions are provided with black spines in aster and serrulifera (fig. 134), in aspro- montensis the sublateral pair is unarmed (fig. 135). 77.—Tergite 9-10, posterior margin: (0) dorsal surface of extensions rounded; (1) flattened. In general, the extensions on the posterior margin of tergite 9-10 are rounded on their dorsal surface in Tipulidae, a situation found also in the majority of species of Savtshenkia. As a special trait, the species as- promontensis and aster have the medial pair of exten- sions of the posterior margin of tergite 10 flattened. H. DE song: Phylogeny of Tipula (Savtshenkia) The dorsal surface of the medial extensions of both species is glossy black. 78. — Tergite 9-10, posterior margin: (0) dorsal surface in between extensions unarmed or armed with up to 10 spines; (1) dorsal surface in between extensions armed with 20 to 30 spines. As discussed under character 74, a few species of Savtshenkia have the area above the medial notch of tergite 9-10 armed with up to 10 spines (fig. 133, 134). The number of spines is higher in the species aspromontensis and aster. In aspromontensis, there are about 30 spines above the medial notch (fig. 135), in aster there are about 20. 79. — Tergite 9-10, posterior margin: (0) extensions present; (1) margin straightened, with small medial notch. As discussed above, the majority of species of Savtshenkia have a pair of lateral extensions on the posterior margin of tergite 9-10, either low and rounded or elongate, diverging and pointed (figs. 128-130, 134, 135, 137-141, 174, 175, 186, 187). The species interserta, the three subspecies of invenus- ta, and the species subvafra are characterized by a straightened posterior margin of tergite 10 (fig. 136). In subvafra, a pair of small posteroventral projections flank the narrow medial notch; such projections are absent in interserta and invenusta. 80. — Tergite 9-10: (0) broader than long, or about as long as broad; (1) elongate, distinctly longer than broad. In Tipulidae, tergite 9-10 usually is a relatively short and broad structure, a situation which probably reflects the groundplan condition of the Lunatipula group and Mediotipula. The majority of species of Savtshenkia also show a similar character state (figs. 121, 123-129, 132, 133, 136, 156). In the pertaining species, tergite 9-10 is broader than long and in most of the species shows a broad medial membranous zone. The species and subspecies of the clade zielsen? to omega differ from this in having an elongate tergite 9-10 which is markedly longer than broad (fig. 137, 138, 174, 186). The medial membranous area of ter- gite 9-10 in these taxa is restricted to a narrow zone, with the exception of the species boreosignata which has a relatively wide membranous area (fig. 138). 81. — Tergite 9-10, posterior margin: (0) extensions variously shaped; (1) rather broad, ventral margin straightened, terminating in an acute lateral tip. As noted above (character 71), most species of Savtshenkia have the extensions on the posterior mar- gin of tergite 9-10 laterally directed and pointed at the tip. In the majority of these species, the ventrome- 297 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 dial margin of the extensions is distinctly concave when seen in posterior view (fig. 139). The species corsosignata, mannheimsi, omega, signata, and trina- cria are characterized by rather broad extensions of tergite 9-10 of which the ventral margin appears to be almost straight in posterior view (figs. 140, 141, 175, 187). Female terminalia 82. — Genital fork: (0) broadest width in posterior half: (1) broadest width in anterior half. In general, the genital fork in Tipulidae is a rather slender structure that widens in its apical posterior part. This situation probably represents the ground- plan condition of the Lunatipula group. Within Mediotipula, the genital fork is slender throughout, with che widest portion lying in the anterior part. The general structure of the genital fork is rather uniform in the species of Savtshenkia, being broad with its greatest width lying in the anterior half (figs. 142, 169, 201; lower side in illustrations). The species at- las and breviantennata are the only studied represen- tatives of Savtshenkia which have the broadest part of the genital fork situated in the posterior section (fig. 143; upper side in illustration). 83. — Hypogynial valve, dorsal margin: (0) not blackish sclerotized; (1) blackish sclerotized. The hypogynial valves are usually uniformly sclero- tized in Tipulidae and this condition probably repre- sents the groundplan state of this character for the Lunatipula group and Mediotipula. All species of Savtshenkia of which the female was examined have the dorsal margin of the hypogynial valve blackish sclerotized. 84. — Ovipositor: (0) well developed, elongate; (1) short. The ovipositor in Tipulidae generally is well devel- oped and has elongate cerci and hypogynial valves, a situation which represents the groundplan condition of the Lunatipula group and Mediotipula. The same state occurs in the majority of species of Savtshenkia (figs. 145, 165, 197). Cerci and hypogynial valves are short in the females of grisescens, holoptera, obsoleta, pagana, and subnodicornis (fig. 144). 85. — Spermathecae: (0) globular; (1) elongate. Theowald’s (1957, 1967) arrangement of species groups of Tipula (Savtshenkia) based on larvae and pupae (1957), and pu- Table 2 pae (1967). 1957 rufina group cheethami rufina subnodicornis group goriziensis (not mentioned) subnodicornis subvafra group atlas (as breviantennata) signata group alpium confusa nielseni obsoleta (not mentioned) (not mentioned) signata staegeri Table 3 1967 cheethami rufina (not filed) grisescens subnodicornis atlas (as breviantennata) alpium confusa (not mentioned) obsoleta odontostyla pagana signata staegeri Division of the Palaearctic species and subspecies of Tipula (Savtshenkia) into species groups according to Savchenko (1961) and Theowald (1973, 1978). Theowald’s group numbering is given in parentheses, a ‘?’ denotes a species placed by Theowald under restriction in the pertaining group. Notes: 1: placed in miscellaneous group by Savchenko, under restriction in convexifrons group by Theowald; 2: referred to by Theowald only; 3: placed in rufina group by Savchenko, in pagana group by Theowald; 4: placed in miscellaneous group by Savchenko, under restriction in pagana group by Theowald; 5: placed in cheethami group by Savchenko, in alpium group by Theowald; 6: placed in rufina group by Savchenko, under restriction in obsoleta group by Theowald; 7: referred to by >» Savchenko only. 298 Savchenko 1961 subnodicornis group limbata group rufina group cheethami group marmorata group miscellaneous group Theowald 1973, 1978 subnodicornis group (1) convexifrons group (2) grisescens group (3) invenusta group (4) limbata group (5) pagana group (6) gimmerthali group (7) rufina group (8) alpium group (11) atlas group (9) obsoleta group (10) marmorata group (12) staegeri group (13) signata group (14) H. DE Jone: Phylogeny of Tipula (Savtshenkia) Species and subspecies postposita subnodicornis convexifrons interserta 5 ped ? venerabilis goriziensis i grisescens (as macrocera in Savchenko 1961) villeneuvii” (as goriziensis villeneuvii) invenusta invenusta invenusta subinvenusta subvafra limbata baltistanica È glaucocinerea holoptera (as pagana holoptera) ? kiushiuensis * ? koreana’ ? minuscula’ (as nana) nivalis pagana (as pagana pagana) gimmerthali pechlaneri È rufina rufina rufina maderensis alpium subalpium macaronesica ° baltistanica ” odontostyla ° atlas breviantennata cheethami macaronesica ” aspromontensis 2 lundbladi” ? mohriana multipicta ° obsoleta ? odontostyla ° serrulifera confusa (as marmorata) eugeni = jeekeli boreosignata ° nielseni ® sardosignata staegeri subsignata benesignata mannheimsi persignata persignata persignata tofina signata koreana È letifera venerabilis 299 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 The shape of the (three) spermathecae in Tipulidae is usually spherical to somewhat ovoid. The ground- plan condition of this character for the Lunatipula group and Mediotipula is probably represented by this state. The majority of females of Savtshenkia exam- ined also show this spermathecal form (figs. 146, 170, 202). Only the species cheethami, goriziensis, simu- lans, and villeneuvii are characterized by the presence of elongate spermathecae (fig. 147). It should be not- ed that the female of tulipa is not yet known. 86. — Sternite 8: (0) ventromedial membranous area near base of hypogynial valves V-shaped, encompassing base of medial margins of hypogynial valves; (1) posteriorly terminating in between hypogynial valves. The ventromedial membranous area of sternite 8 broadens posteriorly and encompasses the bases of the hypogynial valves medially in the species of the Lunatipula group and Mediotipula, thus probably representing the groundplan condition of this charac- ter for the pertaining taxa. The same state is found in the species of Savtshenkia which lie in the range alpha to simulans (fig. 148). The membranous area lies in between the bases of the hypogynial valves in the spe- cies of the clade ignobilis to omega (fig. 149). In these species, the valves are usually broadly separate at the posterior end of the membrane. 87. — Sternite 9: (0) with long and slender medial projection; (1) with broad and short medial projection. The medial projection of sternite 9 usually takes the form of an elongate slender and pointed structure in Tipulidae. This situation, which probably reflects the groundplan condition of the Zunatipula group and Mediotipula, is present in the majority of species of Savtshenkia (figs. 150, 168, 200). The species ben- esignata, eugeni, holoptera, obsoleta, and pagana have a differently built medial projection, which consists of a rather broad structure that is narrowed near mid- length (fig. 151). In eugeni, the cavities next to the medial projection are restricted to the anterior part of sternite 9, whereas they run along the whole length of the medial projection in the four other species. DISCUSSION OF ADOPTED PHYLOGENY The character state matrix as given in table 4 was used to establish the phylogeny of the species and subspecies of Savtshenkia. The character states are bi- nary coded, polymorphies are coded ‘—”, absent data ?. Initially all characters employed had the same weight 1. The character state matrix of table 4 was run under the command ‘mh*; bb*; of the parsimo- ny program HennıG86. The first part of the com- 300 mand (mh*;) constructs several trees to which it ap- plies branch-swapping, retaining one tree for each ini- tial tree, while the second part of the command (bb*;) applies extended branch-swapping to each of the in- put trees, retaining all most parsimonious trees. Although it does not guarantee to find all trees of mi- nimal length, applying ‘mh*; bb*;’ to extensive data sets offers a good alternative for the time-consuming implicit enumeration options of HENNIG86 that de- finitely result in finding all trees of minimal length (Farris 1988). The procedure followed resulted in 28 equally most parsimonious trees with length 157, consistency index 55, and retention index 88. Subsequently successive weighting was carried out under the command ‘mh*; bb*; xs w; cc;’. This proce- dure uses the fit of the characters to the input trees to calculate the weights as the product of the character consistency and character retention indices, which are then scaled in the range 0-10 (Farris 1988). Successive weighting resulted in 14 equally most par- simonious trees with length 723, consistency index 84, and retention index 96. This result was obtained after a single weighting run and did not improve upon further repeats. The calculated weights for the characters are given in table 5, the strict consensus tree of the 14 equally most parsimonious trees is shown in fig. 2. The strict consensus tree of the 14 trees resulting of successive weighting (fig. 2) differs from the strict consensus tree of the 28 trees without successive weighting in the presence of the clade bal tistanica to chrysocephala in the former. The internode leading to this clade is not recognized in the consen- sus tree of the 28 trees without weighting, lowering the grouping baltistanica to chrysocephala one level relative to its position in fig. 2. For the remainder the topologies of both consensus trees are identical. The results of the phylogenetic analysis will be discussed with reference to fig. 2. At the base of the cladogram a trichotomy of the clades alpha and grisescens, hartigiana to simulans, and ignobilisto omega is found. The species pair alpha and grisescens probably are sister species as exemplified by two homoplasies and one reversal [characters 2 (length of male antenna), 9 (size of gonapophysis), and 4 (shape of cell ml), respectively]. The second clade seems to be better substantiated by the characteristic shape of the gonapophysis (char- acter 10) and the acute posterior corner of the base of the inner gonostylus (character 32). The first lineage within this clade leads to a triplet containing the spe- cies hartigiana as the sister species of the pair atlas and breviantennata. In its original description, hartigiana was referred to the atlas group sensu Theowald (table 3) (Theowald, Dufour & Oosterbroek 1982). Besides the species atlas, breviantennata, and hartigiana, the atlas group sensu Theowald also included the species cheethami. As the present analysis shows, the latter species seems to be more closely related to the species of the second group within the clade hartigiana to simulans. This group, the clade tulipa to simulans, consists of the species tulipa, villeneuvii, goriziensis, and the pair cheethami and simulans. Dufour (1983), on describing the species tulipa, placed it in the grises- cens group sensu Theowald which by then contained the species goriziensis, grisescens, and villeneuvii (table 3). Dufour (1983) suggested that grisescens stands apart from the other three species, including villeneu- vii, thus countering Theowald & Oosterbroek’s (1981) contention that villeneuvii should be regarded a subspecies of grisescens. The results of the present analysis endorse Dufour’s views. The third basal clade in fig. 2 contains the remain- der of Savtshenkia. It opens with the successive branching off of the species ignobilis and subnodicor- nis, followed by the remainder of species of Savtshenkia which constitute the monophyletic group baltistanica to omega. The species subnodicornis was placed either near goriziensis, grisescens, or postposita by Savchenko (1961) and Theowald (1957, 1967, 1973, 1978) (tables 2, 3). The species listed belong to the phylogenetically more basal species of Savtshenkia, but none of them appears to be the actu- al sister species of subnodicornis. The clade baltistanica to chrysocephala contains the majority of species of Savishenkia with a non-spinous posterior margin of tergite 9-10 (character 72). The only other species of Savtshenkia in which the spines on the extensions of tergite 9-10 are absent, are the phylogenetically basal species alpha and the derived species mannheimsi and omega. The basal trichotomy of the clade baltistanica to chrysocephala is constituted of the species baltistanica, the species pair glaucocine- reaand postposita, and the clade /undbladi to chrysoce- phala. The clade lundbladi to chrysocephala contains almost all strictly Atlantic (viz. lundbladi, multipicta, and rufina maderensis), and all Afrotropical species of Savtshenkia (viz. nephrotomoides to chrysocephala). (The only remaining Atlantic species of Savtshenkia is macaronesica from the Azores, a species closely related to alpium, see below.) As table 3 shows, the Palaearctic species of the clade baltistanica to chrysoc- ephala were placed in various groupings by Savchenko (1961) and Theowald (1973). (The phy- logeny of the Afrotropical species presented here should be regarded as preliminary only. Most species are known of just a few specimens, thus giving the strong impression that the Afrotropical fauna is poor- ly known. Furthermore, the species are in general very similar to each other and homoplasy in this group seems to be rather common. As the present study is primarily concerned with the western Mediterranean species of Savtshenkia, this issue is not further ex- H. DE JONG: Phylogeny of Tipula (Savıshenkia) plored here.) Modifications of the posterior margin of the male sternite 8 (characters 55, 57) suggest that the remain- der of species of Savtshenkia, the clade alpium to ome- ga, constitute a monophyletic group. The contours of this group were recognized by Theowald (1957, 1967) in his signata group (table 2), by Savchenko (1961) in his marmorata group, and by Theowald (1973) in his obsoleta, alpium, marmorata, staegeri, and signata groups (table 3). Within the clade alpium to omega, the basal lineage is composed of the species pair alpium and macaronesica. Theowald (1973) cor- rectly grouped both species in his a/pium group, but Savchenko (1961) placed macaronesica in his cheetha- mi group (table 3). The species subalpium is probably closely related to alpium and macaronesica (see Appendix B). The sister group of alpium and macar- onesica is substantiated by apomorphies of the sperm pump (characters 48, 49). The basal phylogeny of this clade (jeekeli to omega) could not be fully re- solved. Four lineages constitute a polytomy, with the first containing the single species jeekeli, the second the clade obsoleta to pagana, the third the clade confu- sa to aster, and the fourth the clade fragilis to omega. The species of the clade obsoleta to pagana are most readily distinguished from the other species of Savtshenkia by the peculiar form of sternite 8 (charac- ter 58). Notwithstanding this, Savchenko (1961) as well as Theowald (1973) placed obsoleta in a species group different from holoptera and pagana (table 3). [I here retain holoptera as a taxon separate from paga- na. Both taxa are usually distinguished by the length of the wing and build of the legs in the female (mac- ropterous with slender legs in holoptera, brachypte- rous with stoutish legs in pagana), and the shape of the posterior margin of male sternite 8 and the num- ber of setae it carries (more deeply emarginate with one seta on tubercle in holoptera, less emarginate to truncate with about five to seven setae on tubercles in pagana) (Coe 1950; Edwards 1939). However, num- bers of setae on the posterior margin of sternite 8 intermediate to those of typical holoptera and pagana can be found in certain specimens (Eiroa 1987; see al- so Hutson & Vane-Wright 1969). The occurrence of intermediate stages of fully winged and brachypterous forms has been recorded within Savtshenkia for the fe- male of gimmerthali (Dufour & Brunhes 1984). As long as a similar variability of wing length is not re- corded for females of the pair holoptera and pagana, I prefer to keep them apart. This the more so as the studied male specimens of holoptera show a slight dif- ference with those of pagana in the presence of a small acute and blackish sclerotized extension at the poste- rior end of the dorsal edge of the inner gonostylus.] The clade confusa to aster, which is substantiated by the elongate aedeagus (character 15) and the presence 301 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Table 4. Character state matrix of Tipula (Lunatipula), T. (Mediotipula), and the species and subspecies of 7. (Savtshenkia). Character 1 2 5 0 0 0 Lunatipula 00000 00000 00000 00000 00000 00000 00000 Mediotipula 00010 00000 00000 00000 00010 00000 00000 aberdareica 01010 00000 10000 00000 00001 00000 00000 akeleyi 01010 10000 10000 00001 00001 01000 00000 alpha 01000 00010 . 00000 00000 00000 00000 00000 alpium 00010 00010 00000 00000 10000 00000 00000 asbolodes 01010 00000 10000 00000 00001 00000 00000 aspromontensis 00010 00000 00001 000-0 00000 00110 00000 aster 00010 00000 00001 00000 00000 00110 00000 atlas 10010 01001 00000 00010 00000 00000 01000 baltistanica 00010 00000 00000 00000 00000 00000 00000 benesignata 00010 00000 00001 00000 01100 00000 00000 boreosignata 00010 00010 00000 00000 01000 00000 00000 breviantennata 10010 01001 00000 00010 00000 00000 01000 caligo 01010 10000 10000 00001 00001 01000 00000 cheethami 00011 00001 00000 00010 00100 00000 01100 chrysocephala 01010 10000 10000 00001 00001 01000 00000 confusa 00010 00000 00001 00000 00000 00000 00001 corsosignata 00010 00100 00101 00100 01010 10001 00000 cyrnosardensis 10010 00000 01001 00000 00010 00000 00000 draconis 01010 00000 00000 00000 00001 00000 00000 eleonorae 00010 00000 00000 00000 01000 00000 00000 elgonensis 01010 00000 00000 00000 0000? 00000 00000 eugeni 00010 00000 00001 00000 00000 00000 00001 fragilina 00010 00000 00001 01000 00010 00001 00000 fragilis 00010 00000 01001 00000 00000 10000 00000 gimmerthali 00010 00000 00001 00000 00010 00001 00000 g. mattheyi 00010 00000 00001 00000 00010 00001 00000 g. pteromaculata 00010 00000 00001 00000 00010 00001 00000 glaucocinerea 00000 00000 00000 00000 00000 00000 00000 goriziensis 00010 00001 00000 00000 00100 00000 01100 graciae 00010 00000 00001 00000 00000 00000 00000 grisescens 01000 00010 00000 00000 00000 00000 00000 haennii 00010 00000 00101 00100 01010 10000 00000 hancocki 00010 00000 10000 00000 00001 00000 00000 hartigiana 00010 00001 00000 00010 00000 00000 01000 holoptera 01010 00000 00000 10000 00000 00000 00000 ignobilis 00010 00000 00000 00000 00000 00000 00000 interserta 00010 00010 00001 00000 00000 00000 00000 invenusta 00010 00000 00001 00000 00010 00000 00000 i. microinvenusta 00010 00000 00001 00000 00010 00000 00000 i. subinvenusta 00010 00000 00001 00000 00010 00000 00000 jeekeli 00010 00000 00000 00000 00000 00000 00001 limbata 00010 00000 00001 01000 00000 00001 00000 lundbladi 00110 00000 00001 00000 00001 00000 00000 macaronesica 00010 00010 00000 00000 10000 00000 00000 mannheimsi 00010 00000 00001 00000 01100 00000 10000 302 H. DE zong: Phylogeny of Tipula (Savtshenkia) 7 8 0 0 00000 00000 00000 00000 00000 00 00000 00000 00000 00000 00000 00000 00000 01000 00 01000 00100 00000 00000 00000 11100 00000 02??? 01000 00100 00000 00000 00000 11100 00000 02222 22 00000 00100 00000 00000 00000 01000 00000 01100 00 01000 00101 01000 00000 00000 10000 00000 01100 10 01000 00100 00000 00000 00000 11100 00000 0222? 2? 01101 00100 01000 00000 10000 00010 11100 01100 10 01111 00101 01000 00000 10000 00010 11100 02??? 2? 00000 00100 00000 00000 00110 00000 00000 00100 00 01000 00100 01000 00000 00000 11000 00000 01100 10 01110 00111 01000 11000 01000 10000 00001 01100 11 01010 01111 01000 11000 01000 00000 00001 Or 22 00000 00100 00000 00000 00110 00000 00000 00100 00 01000 00100 00000 00000 00000 11000 00000 01100 10 00000 00100 00000 00000 00001 00000 00000 01101 00 01000 00100 00000 00000 00000 11100 00000 0222? 2? O1111 10101 01000 00000 00000 10000 00000 01100 10 01110 00111 01000 11110 01000 10000 00001 11100 10 01110 00111 01000 01000 01000 10000 00000 01100 10 01000 00100 00000 00000 00000 11100 00000 02222 22 01010 01111 01000 11000 01000 10000 00001 02??? 01000 00100 00000 00000 00000 11000 00000 0222? 2? 01101 00101 01000 00000 10000 10010 00000 01100 11 01110 00111 00001 00000 01000 10000 00000 01100 10 01110 00101 01000 00000 00000 10000 00000 01100 10 01110 00110 00000 00000 01000 10000 00000 01100 10 01110 00110 00000 00000 01000 10000 00000 (Ba 01110 00110 00000 00000 01000 10000 00000 ER E 01000 00100 00000 00000 00000 11000 00000 01100 10 00000 00100 00000 00000 00000 00000 00000 01101 00 01111 00101 01000 00000 00000 00000 00000 Va 22 00000 00100 00000 00000 00000 00000 00000 01110 00 01110 00111 01000 11110 01000 10000 00001 rara de 01000 00100 00000 00000 00000 11100 00000 08223 2? 00000 00100 00000 00000 01100 00000 00000 01100 00 01110 00101 01110 00000 00000 10000 00000 01110 11 00000 00100 00000 00000 00000 00000 00000 01100 10 01101 10100 00000 00000 00000 00000 00010 01100 10 01110 00111 00000 10000 01000 00000 00010 Oe rer? “ue 01110 00111 00000 10000 01000 00000 00010 01100 10 01110 00111 00000 10000 01000 10000 00010 01100 0 01110 00101 01000 00000 00000 10000 00000 01100 10 01110 00111 00001 00000 01000 10000 00000 01100 10 01000 00101 00000 00000 00000 11000 00000 01100 10 01000 00101 00000 00000 00000 10000 00000 0 2? 01110 00111 01000 11001 01000 11000 00001 Lessee TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Table 4. (continued) Character state matrix of Tipula (Lunatipula), T. (Mediotipula), and the species and subspecies of 7. (Sav Character 1 2 3 0 0 0 multipicta 00110 00000 00001 00000 00001 00000 00000 nephrotomoides 00010 00000 00000 00001 00001 00000 00000 nielseni 00010 00000 00001 00000 00010 00001 00000 obsoleta 00010 01000 00001 10000 00000 00000 00000 odontostyla 00010 00000 00001 00000 00010 00000 00000 omega 00010 00000 00001 00000 01100 00000 10000 ornata 00010 00000 00001 00000 00000 10000 00000 pagana 01010 00000 00000 10000 00000 00000 00000 pechlaneri 00010 00000 00001 00000 01010 00001 00000 persignata 00010 00000 00001 00000 01100 00000 10000 p. tofina 00010 00000 00001 00000 01100 00000 10000 phoroctenia 00010 00000 01001 00000 00000 00000 00000 postposita 00000 00000 00000 00000 00000 00000 00000 rufina 00010 00000 00000 00000 00001 00000 00000 r. maderensis 00010 00000 00001 00000 00001 00000 00000 sardosignata 00010 00000 00001 00000 00010 00000 00000 serrulifera 00010 00000 00001 00000 00000 00110 00000 signata 00010 00000 00001 00000 01100 00000 10000 simulans 00011 00001 00000 00010 00100 00000 01100 sordidipes 00010 00000 00001 00000 00000 00000 00000 staegeri 00010 00000 00001 00000 00010 00000 00000 subnodicornis 000-0 00000 00000 00000 00000 00000 00000 subsignata 00010 00100 00111 00100 01010 10000 00010 s. cazorla 00010 00100 00111 00100 01010 10000 00010 subvafra 00010 00000 00001 01000 00010 00001 00000 trinacria 00010 00000 00001 00000 01100 00000 10000 tulipa 00010 00001 00000 00000 00100 00000 01000 venerabilis 00010 00000 00001 00000 00000 00000 00000 villeneuvii 10010 00001 00000 00000 00100 00000 01000 of elongate and anteriorly directed posterior apo- demes of the sperm pump (character 50), contains as its first lineage the triplet confusa, interserta, and orna- ta. Although differing in most genital features, these three species share the presence of the quite remark- able slender elongate and anteriorly extended posteri- or apodemes of the sperm pump (character 51). In fig. 2, the remainder of the clade confusa to aster is de- picted as a monophyletic group which is substantiat- ed by a single reversal only. The reversal concerns the redirection and reduction of the extensions of the posterior margin of tergite 9-10 (character 71). The extensions are rather differently shaped in the species of this grouping and its actual monophyly is open to question. The species eugeni, which is included in this grouping, has the extensions of tergite 9-10 well de- veloped and laterally extended. Theischinger (1983) correctly considered his spe- cies aster to be the closest relative of aspromontensis. Theowald (1973) placed the species confusa (as 304 marmorata), eugeni, and jeekeli in his marmorata group on account of the peculiar shape of their inner gonostyli (character 35). When other characters are taken into consideration, the three species appear not to be closely related, forcing the conclusion that the characteristic shape of the inner gonostylus must be due to homoplasy. Next in fig. 2 is the clade fragilisto omega, in which the Nearctic species pair fragilis and phoroctenia branch off first, followed by the species odontostyla. The remainder of species of Savtshenkia constitute the monophyletic group fragilina to omega of which the first lineage leads to the clade fragzlina to invenus- ta microinvenusta. This latter clade contains species which are characterized by the presumably secondary absence of spines on the posterior margin of male sternite 8 (character 57). It is composed of species which Theowald (1973) arrayed in his gimmerthali, invenusta, and limbata groups (table 3), together with the Nearctic species fragilina (the species mohriana H. DE Jong: Phylogeny of Tipula (Savtshenkia) 7 8 0 0 00000 00000 11000 00000 01100 10 01000 00100 00000 00000 00000 11000 00000 01100 10 11110 00111 01000 11000 01000 10000 00001 01100 10 01110 00101 01100 00000 00000 10000 00000 01110 1] 01110 00101 01000 00000 00000 10010 00000 0228 22 01110 00111 01000 11001 01000 11000 00001 12222 2? 01111 10101 00000 00000 00000 10000 00000 01100 10 01110 00101 01110 00000 00000 10000 00000 01110 1] 01110 00110 00000 00000 00000 10000 00000 01100 10 01110 00111 01000 11000 01000 10000 00001 0212520 22 01110 00111 01000 11000 01000 10000 00001 OR 2? 01110 00101 01000 00000 00000 10000 00000 01100 10 01000 00100 00000 00000 00000 11000 00000 02222 22 01000 00101 10000 00000 00000 11000 00000 01100 0 01000 00101 10000 00000 00000 11000 00000 01100 0 11110 00111 01000 11000 00000 10000 00001 01100 10 01111 00101 01000 00000 10000 00010 10000 01100 10 01110 00111 01000 11000 01000 10000 00001 11100 10 00000 00100 00000 00000 00001 00000 00000 01101 00 01111 00100 01000 00000 00000 00011 00000 01100 10 01110 00111 01000 01000 01000 10000 00000 01100 10 00000 00100 00000 00000 00000 10000 00000 01110 10 01110 00111 01000 11110 01000 10000 00001 01100 10 01110 00111 01000 11110 00000 10000 00001 Oscar Ets 01110 00111 00001 00000 01000 10000 00010 01100 10 01110 00111 01000 11001 01000 10000 00001 11100 10 00000 00100 00000 00000 00000 00000 00000 OH ae O1111 00100 01000 00000 00000 00011 00000 MEERE 00000 00000 00000 00000 01101 and productella probably also belong here; see variability of this character within gimmerthali it has Appendix B). The clade fragilina to invenusta microinvenusta contains two trichotomies. The exact relationships of the species fragilina, limbata and subvafra as well as those between the three subspecies of gimmerthali are hard to establish. The subspecies gimmerthali pterom- aculata was described as differing from the nomino- typical subspecies by its darker colour, spotted wings, and relatively short second flagellomere (Theowald, Dufour & Oosterbroek 1982). Examination of mate- rial of the nominotypical form of gimmerthali, how- ever, revealed that the length of the second flagel- lomere varies considerably within this subspecies. The relative length of the second flagellomere thus of- fers no arguments to distinguish gimmerthali pterom- aculata from the nominotypical subspecies. [The length of the second male flagellomere has also been used to distinguish gimmerthali from pechlaneri (Mannheims & Pechlaner 1963). Considering the proved to be non-applicable here. The species gim- merthali and pechlaneri are more easily distinguished on account of the structure of the inner gonostylus and the shape of male tergite 9-10 (see also Appendix C). The monophyly of the three subspecies of gim- merthali could be demonstrated by the shape of the inner gonostylus only (character 41)]. Useful differ- ences between gimmerthali gimmerthali and gimmer- thali pteromaculata are found in the autapomorphous larger outer gonostylus and relatively plump anterior tip of the inner gonostylus of gimmerthali pteromacu- lata (cf. figs. 66 and 67, and 83 and 84). The subspe- cies gimmerthali mattheyi, known of the male holo- type only (not examined), apparently difters from the other subspecies by the autapomorphous absence of a posterior projection on the outer gonostylus and the reduced armature of the posterior margin of tergite 9- 10 (Theowald & Dufour 1983). [The females of both gimmerthali mattheyi and gimmerthali pteromaculata 305 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Table 5. Weights of characters after single weighting run using the command ‘mh*; bb*; xs w; cc; of Hennig86. Character Weight Character Weight Character Weight l 1 31 10 61 4 2 ] 32 10 62 10 3 10 33 10 63 10 4 1 34 10 64 10 5 10 35 0 65 10 6 10 36 4 66 10 7 2 37 10 67 1 3 10 38 10 68 10 9 1 39 10 69 10 10 10 40 10 70 10 11 10 41 10 71 Il 12 2 42 10 72 2. 13 10 43 10 73 4 14 10 44 2 74 4 15 1 45 2. 75 10 16 10 46 10 76 10 17 10 47 10 Wi 10 18 10 48 4 78 10 19 3 49 2 79 1 20 3 50 10 80 10 21 10 51 10 81 3 22 4 52: 10 82 2 23 4 DI 10 83 10 24 22 54 10 84 1 25 10 55 1 85 10 26 2 56 10 86 10 27 10 57 Il 87 Il 28 10 58 10 29 10 59 10 30 2 60 10 remain unknown. Those of gimmerthali gimmerthali are usually brachypterous, but populations of the no- minotypical form of gimmerthali are known from Switzerland which have female wing lengths ranging from the brachypterous to the fully winged condition (Dufour & Brunhes 1984). Dufour (1990) further- more recorded hemipterous females of gimmerthali from the eastern side of the Sierra Nevada, Spain.) The species invenusta presently contains three sub- species. Differences between the subspecies can be found in details of the male terminalia and in the ven- tral distance berween the eyes. The examined speci- mens of invenusta invenusta conform to Riedel’s (1919) description of the species as having rounded posterolateral corners on tergite 9-10 [‘Lam. term. sup. (...) mit abgerundeten Ecken’]. However, illus- trations of the terminalia of ‘invenusta invenusta by Savchenko (1961) and Theowald (1973) depict ter- gite 9-10 with acute posterior extensions. Examination of material present in ZMAN showed that Theowald actually figured a specimen that can only be identified as invenusta subinvenusta. However, considering Savchenko’s illustration there still is a possibility that the structure of the posterior margin of tergite 9-10 varies within invenusta invenusta. [All 306 three described subspecies bear spines on the postero- ventral margin of tergite 9-10. Having examined sev- eral of its paratypes, I can not corroborate Dufour’s (1990) observation that spines are absent from the posterior margin of tergite 9-10 in his invenusta mi- croinvenusta.) Another complicating factor in main- taining the three recognized subspecies is the observa- tion that Tipulidae occurring at high geographical latitudes or high altitudes usually have smaller eyes than their relatives living under more moderate eco- logical conditions. Therefore, the size of the eyes in specimens of invenusta might well be an ecotypic dif- ferentiation in response to local environmental condi- tions and thus would be of little value in separating the subspecies. The complicated nature of the taxon- omy of invenusta is furthermore illustrated by the fact that Alexander described invenusta invenusta under three different names, viz. docilis (correctly synonym- ized with invenusta by Savchenko 1961), and insig- nifica and packardi (synonymized in this paper; see Appendix A, synonymy). Hancock (1988) suggested that yet another subspecies might be introduced to accommodate a morphologically slightly deviating population from Andorra (Pyrenees). I here would like to stress the need to study the variability of this taxon over its immense Holarctic range before even considering to introduce new formal names for single isolated populations. [The most recently published map showing the Palaearctic distribution of invenusta is the one given by Dufour (1992).] The clade cyrnosardensis to omega contains the spe- cies of Theowald’s (1973) staegeri and signata species groups (table 3). Most species of this clade are easily recognized by the presence of the bulbous and spi- nous extensions on the posterior margin of sternite 8 (character 62). The clade opens with a trichotomy consisting of the species cyrnosardensis and staegeri in combination with a clade which includes the remain- der of the species. Both cyrnosardensis and staegeri are species distinguished by a number of autapomorphies but apparently lack any further character states that could enable the resolution of their exact phylogenet- ic position. The species nielseni and sardosignata are depicted here as sister species as is suggested by their broad based outer gonostyli, although the overall shape of this structure differs remarkably in both species (char- acter 46). The two species are, moreover, distin- guished from each other by the autapomorphous form of the inner gonostylus and the shape of sternite 8. Their tentative sister species relationship could not be substantiated by other apomorphies. Although on quite different grounds, the clade haennii to subsignata cazorla contains the four taxa which were recognized as a monophyletic group by Dufour (1991). The characters employed in the present analysis result in a sequence of relationships between these taxa which differs from the solution provided by Dufour. Compared with Dufour’s clado- gram, corsosignata and haennii have changed places in the present phylogeny, necessitating the elevation in rank of haennii from subspecies of subsignata to spe- cies (see Appendix A). Dufour substantiated his con- cept ofa monophyletic group containing corsosignata, haennii, subsignata subsignata, and subsignata cazorla by the degree of sclerotization and shape of the ven- tromedial appendage of male sternite 8 and the de- gree of sclerotization of female sternite 9 (Dufour 1991, characters 1 and 2). I can see no essential differ- ences between the sclerotization and shape of the ven- tromedial extension of male sternite 8 in this clade of four taxa and that of other derived members of Savtshenkia such as persignata, sardosignata, and sig- nata, and prefer to consider it a plesiomorphy at this level. The presence of an unsclerotized female sternite 9 as a synapomorphy for the members of the clade haennii to subsignata cazorla has not been employed in this paper because it is not always clear whether sternite 9 should be valued as sclerotized or not in species of Savtshenkia. Moreover, the female of sub- signata cazorla remains unknown. Dufour considered H. DE song: Phylogeny of Tipula (Savtshenkia) haennii the sister species of the pair subsignata subsig- nata and subsignata cazorla because of the presence of a ridge on what he denoted as the median part of the inner gonostylus (Dufour 1991, character 3). As dis- cussed under character 34, I consider the presence of the anteriorly extended part of the lateral ridge of the inner gonostylus an apomorphy of the pair subsignata subsignata and subsignata cazorla. Of the three apo- morphies of Dufour supporting the monophyly of subsignata subsignata and subsignata cazorla, two ap- pear to be invalid (Dufour 1991, characters 4 and 5). The continuity of the posterior and ventral parts of the lateral ridge of the inner gonostylus in these two subspecies (Dufour 1991, character 4) probably is a plesiomorphy (see discussion under character 34). The relatively broad base of the outer gonostylus in subsignata subsignata and subsignata cazorla (Dufour 1991, character 5) probably also represents a plesio- morphy considering the situation in most related spe- cies. The width of the base of the outer gonostylus furthermore varies within the species and subspecies and differs only slightly from that in haennii. Dufour’s third argument for the monophyly of the pair subsignata subsignata and subsignata cazorla (Dufour 1991, character 6, shape of the gonapoph- ysis) is employed in this paper as well (character 14). Following on his description of eleonorae, Theischinger (1978) suggested that this species is most closely related to benesignata. I am, however, convinced that the sister species of eleonorae is boreo- signata, which is substantiated by the almost identical shape of the inner gonostylus (character 42) and the bizarre modification of the aedeagal complex (charac- ter 52). Both boreosignata and eleonorae are known of their holotypes only, with the type localities lying about 2500 kilometers apart, that of boreosignata in northern Sweden and that of eleonorae in western Rumania! The clade benesignata to omega contains the species which Theowald (1973) included in his signata group, with the addition of the species omega and tri nacrıa described herein as new. DISTRIBUTION In this section I will give a brief account on the dis- tribution of the species and subspecies of Savtshenkia. As noted above, more detailed information on the distribution of the western Mediterranean species and subspecies and their closest relatives will be given in a forthcoming paper on the historical biogeography of the western Mediterranean area. A survey of the dis- tribution of the species and subspecies of Savtshenkia is given for the Afrotropical Region by Hutson (1980, under Pterelachisus), the Nearctic Region by Alex- ander (1965b), the Oriental Region by Alexander & 307 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Alexander (1973), and the Palaearctic Region by Oosterbroek & Theowald (1992). The sequence of taxa in the summary below follows that of fig. 2. The two species of the pair alpha and grisescens dif- fer considerably in their ranges. The first species is known of its type locality in the Rodna Mountains of northern Rumania only, while grisescens is distributed over the central belt of Europe eastwards to the Altai Mountains in central Asia. The species of the clade hartigiana to simulans are confined to the western Palaearctic, with most of them having a restricted range. The species hartigiana is known from Corsica and Sardinia only, atlas is re- corded from Algeria and Morocco, and breviantenna- ta from Spain, southern Switzerland, Corsica, Italy, including Sardinia, and northern Algeria. The species tulipa was described from two localities in southern Switzerland, villeneuvii is known from central Spain, while goriziensis, cheethami and simulans are more widely distributed over the mountainous areas of cen- tral Europe. The species ignobilis has an eastern Nearctic distribution between about 35° and 50°N latitude. The species subnodicornis has a wide range extending from the Faeroes and Ireland in the west over central Europe towards Lake Baykal in the east. The clade baltistanica to chrysocephala contains spe- cies which cover a wide geographic area. The species baltistanica occurs in the Caucasus, Tadzhikistan and Kashmir, glaucocinerea has an arctic Siberian distribu- tion, as does its presumed sister species postposita, which is furthermore known from northern Finland, the Altai Mountains and Mongolia. The species lundbladi is an endemic of Madeira, the species mul tipicta is presently recorded from the Canary Islands only. The nominotypical form of rufina is wide- spread, ranging from Iceland and north-west Africa towards Tadzhikistan in the east, while the subspecies rufina maderensis is endemic to Madeira. The major- ity of species of the grouping nephrotomoides to chry- socephala is distributed in the mountainous areas sur- rounding Lake Victoria in eastern Africa. The only member of this group that does not occur in this re- gion is the species draconis from South Africa (Natal). The species alpium has a wide distribution which includes most of Europe. It also occurs along the east- coast of the Nearctic Region between about 42° and 55° N latitude (see Appendix A, synonymy, under nebulipennis). The species macaronesica is endemic to the Azores. As is presently known, the phylogenetically rather isolated species jeekeli has a disjunct distribution throughout the Mediterranean, ranging from Galicia in north-west Spain to Cyprus in the east. The species obsoleta is distributed over most of Europe, extending its range towards Georgia in the south-east, while the pair holoptera and pagana is con- 308 fined to the central belt of Europe, with holoptera having a disjunet distribution in south-west England, north-west Spain, and Czechia. Of the species of the clade confusa to aster, confusa occurs over most of Europe, including the Faeroes and Ireland in the north-west. The species interserta has a distribution ranging from the central European mountains in the west towards the Tuva in the east. The beautiful ornata was described on the basis of a single male from northern Israel, but a series of this species, including females, was collected recently on Cyprus by Pjotr Oosterbroek and Cita Hartveld. It seems that the species graciae still is known from its type locality in southern California only. Both sordi- dipes and venerabilis are restricted to Kashmir. The species eugeni is presently known from Corsica, Sicily, and southern Italy. The species serrulifera has a more extensive distribution, occurring in mountainous are- as from the Pyrenees in the west to the Caucasus in the east. Theowald (1973) described aspromontensis from southern Italy (Aspromonte), but more recently it was collected also in Greece (Peloponnisos, Crete). The species aster still is known of its holotype only, originating from the Greek island Thasos in the northern Aegean Sea. The presumed sister species fragilis and phoroctenia have a Nearctic distribution which extends across the continent in the region of the 50th degree of latitude. The species odontostyla is known from the north-east- ern coastal areas of the Black Sea and the Greek island Thasos. The species fragilina has a Nearctic distribution, being recorded from the north-western part of the continent, ranging from Alaska to Colorado, as well as from Maine on the east coast (under its synonym broweri). The species limbata is widespread in the mountainous areas of Eurasia, subvafra seems to be confined to continental western Europe. The mem- bers of the clade pechlaneri to gimmerthali pteromacu- lata are distributed in the mountainous areas of the western Palaearctic Region, with pechlaneri having a disjunet distribution from Andorra in the Pyrenees to northern Iran, gimmerthali gimmerthali ranging from the Sierra Nevada of Spain in the south towards northern Scandinavia and eastwards to the Caucasus, gimmerthali mattheyi being restricted to southern France, and gimmerthali pteromaculata to Corsica. The three subspecies of invenusta differ consider- ably in the extent of their distribution area. The sub- species invenusta microinvenusta is presently known only from southern Spain (Sierra Nevada), the sub- species invenusta subinvenusta has a wider range in the mountains of central Europe, while the nominotypi- cal subspecies disjunctly ranges across Eurasia and al- so occurs in eastern North America (see Appendix A, synonymy, under insignifica and packardı). The species cyrnosardensis is endemic to Corsica and Sardinia, staegeri occurs in Europe from the Faeroes to northern Spain in the west towards the coastal areas of the Baltic Sea in the cast. Both species of the pair nielseni and sardosignata have a restricted range, with nielseni being confined to the Alps, and sardosignata occurring on Corsica and Sardinia, as well as in southern France. The four members of the clade haennii to subsigna- ta cazorla are also restricted to Europe, with haennii being recorded from the French and Spanish Pyrenees, corsosignata occurring on Corsica, subsigna- ta cazorla in northern Spain, and subsignata subsigna- ta being more widely distributed in the mountains of central Europe. As mentioned above, both boreosig- nata and eleonorae are known of their holotypes only, with the type localities lying in northern Sweden and western Rumania respectively. The distribution area of the species benesignata consists of widely separated populations ranging from the Alps towards northern Scandinavia and eastwards to the Tien Mountains. The original descriptions of the two subspecies persig- nata persignata and persignata tofina were based on material originating from North Korea, to which area the distribution range of the nominotypical form seems to be restricted. Later records of persignata tofi- na, however, showed that this subspecies ranges across all of northern Eurasia. The species signata is widespread in Europe, cover- ing most of the area. The triplet trinacria to omega is restricted to Italy, with trinacria occurring on Sicily, mannheimsi in the north of Italy, and omega on Sardinia. ACKNOWLEDGEMENTS For the loan of material of Savtshenkia I am most indebted to the following persons: Roy Danielsson (Lund), Christophe Dufour (Neuchätel), Jon Gelhaus (Philadelphia), Brian Pitkin (London), Hans-Peter Tschorsnig (Stuttgart), and Holly Williams (Was- hington D.C.). Lali Eiroa (Santiago de Compostela, Spain) kindly donated material of Savtshenkia holopte- ra to the collection of the Zoological Museum, Amsterdam. I would like to thank Wayne Mathis and Holly Williams for their much appreciated help du- ring my short visit to the United States National Museum of Natural History, Washington D.C. A draft version of this paper was read and com- mented upon by Pjotr Oosterbroek, Hans Duffels, Fred Schram, and an anonymous reviewer. Their crit- icism has led to a number of substantial improve- ments of the text. The investigations were supported by the Life Science Foundation (SLW), which is sub- sidized by the Netherlands Organization for Scientific Research (NWO). H. DE JONG: Phylogeny of Tipula (Savtshenkia) REFERENCES Alexander, C. P., 1919a New Nearctic (Rhyphidae and Tipulidae) part VII. Entomologist 51: 162-172. Alexander, C. P., 1919b.The crane-flies of New York. Part L Distribution and taxonomy of the adult flies. — Cornell University Agricultural Experiment Station, Memoir 25: 76-993. Alexander, C. P., 1924. New or little known crane-flies from New England. -- Occasional Papers of the Boston Society of Natural History 5: 115-118. Alexander, C. P., 1925. Undescribed species of Japanese crane-flies. Part V. — Annals of the Entomological Society of America 17: 431-448. Alexander, C. P., 1926. Undescribed species of crane-flies from the United States and Canada, part II. — Insecutor Inscitiae Menstruus 14: 114-122. Alexander, C. P., 1928. New or little-known species of the genus Tipula from Labrador (Tipulidae, Dipt.). — Canadian Entomologist 60: 95-101. Alexander, C. P., 1934. New or little-known Tipulidae from Eastern Asia (Diptera), XXI. — Philippine Journal of Science 55: 19-60, pls. 1-3. Alexander, C. P., 1940. Undescribed species of crane-flies from the eastern United States and Canada (Dipt.: Tipulidae). Part VI. — Entomological News 51: 83-85, 99-103. Alexander, C. P., 1942. See Alexander 1966b (reprint). Alexander, C. P., 1951. New or little-known Tipulidae (Diptera). - XCI. Oriental-Australasian species. — Annals and Magazine of Natural History (12) 6: 1072-1102. Alexander, C. P., 1953. The Oriental Tipulidae in the col- lection of the Indian Museum. Part III. — Records of the Indian Museum 50; 321-357. Alexander, C. P., 1954. The Tipulidae of Mount Ishizuchi and the Omogo Valley, Iyo, Shikoku, Japan. — Trans- actions of the Shikoku Entomological Society 4: 23-31. Alexander, C. P., 1955. Records and descriptions of Japanese Tipulidae (Diptera), part IV. The crane-flies of Shikoku, IV. — Philippine Journal of Science 83: 263- 306, pls. 1-4. Alexander, C. P., 1956. Tipulidae. — Ruwenzori Expedition 1934-35 1: 129-380, pls. XXVII-XXVIIL. Alexander, C. P., 1961. New or little-known Tipulidae from Eastern Asia (Diptera), L. — Philippine Journal of Science 90: 397-445, pls 1-3. Alexander, C. P., 1962. Mission zoologique de l'I.R.S.A.C. en Afrique orientale. (P. Basilewsky et N. Leleup, 1957) LXXX. - Diptera Tipulidae. — Annales Musée Royal de l’Afrique Centrale, Série 8vo (Zool.) 110: 343-362. Alexander, C. P., 1964. New or little-known Tipulidae from Eastern Asia (Diptera), LIL. — Philippine Journal of Science 93: 77-130, pls. 1-6. Alexander, C. P., 1965a. New subgenera and species of crane-flies from California (Diptera: Tipulidae). — Pacific Insects 7: 333-386. Alexander, C. P., 1965b. Family Tipulidae. — In: Stone, A. et al. (eds.) A catalog of the Diptera of America north of Mexico. Agriculture Handbook, Agriculture Research Service 276: 16-90. Alexander, C. P., 1966a. New or little-known Tipulidae from Eastern Asia (Diptera), LVI. — Philippine Journal of Science 94: 235-286, pls. 1-5. Alexander, C. P., 1966b. Family Tipulidae. — In: Crampton, G. C. et al. (eds.) Guide to the insects of Connecticut. cranc-flics Canadian 309 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Part VI. The Diptera ot true flies of Connecticut. First fascicle. Bulletin of the Connecticut State Geological and Natural History Survey 64: 196-486b. (Reprint of Alexander 1942). i Alexander, C. P. & M. A. Alexander, 1973. Tipulidae. — In: Delfinado, M. D. & D. E. Hardy (eds.) A catalog of the Diptera of the Oriental Region. Volume I. Suborder Nematocera: 10-224. Honolulu. Audcent, H., 1932. British Tipulinae (Diptera, Tipulidae). — Transactions of the Entomological Society of the South of England 8: 1-34, pls. III. Bergroth, E. 1888. Oesterreichische Tipuliden, gesammelt von Professor J. A. Palmen im Jahre 1870. — Verhandlungen der Zoologisch-Botanischen Gesellschaft in Wien 38: 645-656. Brindle, A., 1958. A field key for the identification of Tipula larvae (Dipt.: Tipulidae). — Entomologist’s Gazette 9: 165-182. Brindle, A., 1959. Notes on the larvae of the British Tipulinae (Dipt., Tipulidae) part 7. - The larvae of the Tipula marmorata group. — Entomologists Monthly Magazine 95: 204-205. Brindle, A., 1960. The larvae and pupae of the British Tipulinae (Diptera: Tipulidae). — Transactions of the Society for British Entomology 14: 63-114. Brunhes, J. & C. Dufour, 1992. Etudes structurales et dyna- miques sur les écosystèmes de Tourbières acides. III - le peuplement des Tipulides (Diptera, Tipulidae). — Bulletin d’Ecologie 23: 17-26. Bryce, D., 1956. Notes on the life-history of Tipula cheetha- mi Edw. (Dipt., Tipulidae). — Entomologist's Monthly Magazine 92: 104-106. Byers, G. W., 1961. The crane fly genus Dolichopeza in North America. — University of Kansas Science Bulletin 42: 665-924. Coe, R. L., 1950. Family Tipulidae. — Handbooks for the Identification of British Insects IX 2: 1-66. Coulson, J. C., 1959. Observations on the Tipulidae (Diptera) of the Moor House Nature Reserve, Westmoreland. — Transactions of the Royal Entomological Society of London 111: 157-174, pls. I, II. Coulson, J. C., 1962. The biology of Tipula subnodicornis Zetterstedt, with comparative observations on Tipula paludosa Meigen. — Journal of Animal Ecology 31: 1-21. Dufour, C., 1983. Tipula (Savtshenkia) tulipa sp.n. from xenothermic valleys of the Swiss Alps (Diptera: Tipulidae). — Mitteilungen der Schweizerischen Entomologischen Gesellschaft 56: 275-281. Dufour, C., 1986. Les Tipulidae de Suisse (Diptera, Nematocera). — Documenta Faunistica Helvetiae 2: 1- 187, + 1-149. Dufour, C. 1990. Tipula (Acutipula) nevada sp.n. and Tipula (Savtshenkia) invenusta microinvenusta ssp.n. from the heights of Sierra Nevada in Spain (Diptera: Tipulidae). — Mitteilungen der Schweizerischen Entomologischen Gesellschaft 63: 227-232. Dufour, C., 1991. Tipula (Savtshenkia) subsignata haennii ssp.n. and Tipula (S.) subsignata cazorla ssp.n. from the Iberian Peninsula (Diptera, Tipulidae). — Mitteilungen der Schweizerischen Entomologischen Gesellschaft 64: 243-249. Dufour, C., 1992. High altitude Tipulidae in Switzerland (Diptera, Nematocera). — Acta Zoologica Cracoviensia 35: 113-134. 310 Dufour, C. & J. Brunhes, 1984. Les Tipulidae brachyptères de la région paléarctique occidentale avec les descriptions des 22 holoptères de Tipula (Savtshenkia) gimmerthali Lackschewitz et de Tipula (Platytipula) luteipennis agilis ssp.n. (Diptera, Tipulidae). — Mitteilungen der Schwei- zerischen Entomologischen Gesellschaft 57: 133-151. Edwards, F. W., 1928. Some Nematocerous Diptera from Yunnan and Tibet. — Annals and Magazine of Natural History (10) 1: 681-703, pl. XIX. Edwards, F. W., 1931. Some suggestions on the classifica- tion of the genus Tipula (Diptera, Tipulidae). — Annals and Magazine of Natural History (10) 8: 73-82. Edwards, F. W., 1939. Additions to the list of British crane- flies. — Entomologist’s Monthly Magazine 75: 241-249. Eiroa Alvarez, M. E., 1987. Estudio de los Tipúlidos (O. Dipteros) en Galicia. — Thesis, Facultad de Biologia, Universidad de Santiago de Compostela: 1-422 + un- numbered pages. Farris, J. S. 1988. HENNIG86, version 1.5 + reference. — Port Jefferson Station, New York. Frommer, S. I., 1963. Gross morphological studies of the re- productive system in representative North American crane flies (Diptera: Tipulidae). — Kansas University Science Bulletin 44: 535-626, pls. I-XX. Gelhaus, J. K., 1986. Larvae of the crane fly genus Tipula in North America (Diptera: Tipulidae). — Kansas University Science Bulletin 53: 121-182. Hancock, E. G., 1988. A cranefly, Tipula (Savtshenkia) in- venusta Riedel (Diptera: Tipulidae), new to the British Isles. — Entomologist 107: 96-98. Hemmingsen, A. M., 1952. The oviposition of some crane- fly species (Tipulidae) from different types of localities. — Videnskabelige Meddelelser fra Dansk Naturhistorisk Forening i Kjobenhavn 114: 365-430. Hemmingsen, A. M., 1954. The function of the peculiar processes of the 8th sternite in the males of Tipula (Oreomyza) staegeri Peder Nielsen and Tipula (Oreomyza) signata Staeger. — Videnskabelige Medde- lelser fra Dansk Naturhistorisk Forening i Kjobenhavn 116: 411-418. Hemmingsen, A. M., 1962. Copulatory adaptations of male hypopygium to female tergal ovipository valves (cerci) in certain crane-flies (Tipulidae). — Videnskabelige Med- delelser fra Dansk Naturhistorisk Forening i Kjobenhavn 124: 135-163, figs. 1-18. Hutson, A. M., 1980. Tipulidae. — In: Crosskey, R. W. (ed.) Catalog of the Diptera of the Afrotropical Region: 47-91. London. Hutson, A. M. & R. I. Vane-Wright, 1969. Corrections and additions to the list of British Nematocera (Diptera) since Kloet and Hincks’ ‘A check list of British insects’ (1945), part 1. — Entomologist’s Gazette 20: 231-256. Kandybina, M. N., V. I. Lantzov & E. N. Savchenko, 1987. A catalog of the type-specimens in the collection of the Zoological Institute, Academy of Sciences of the USSR, Insecta, Diptera, N3. Families Tanyderidae, Trichoceridae, Limoniidae, Tipulidae: 1-68 + inside cov- ers. — Nauka, Leningrad. (In Russian). Lackschewitz, P., 1936. A new species of Tipula from the Azores (Diptera). — Proceedings of the Royal Entomo- logical Society of London (B) 5: 118. Lantzov, V. I., 1982. Ecology and preimaginal developmen- tal stages of Tipula gleucocinerea (Diptera, Tipulidae). — Zoologicheskii Zhurnal 61: 1913-1916. Loew, H., 1863. Diptera Americae Septentrionalis indigena. Centuria quarta. — Berliner Entomologische Zeitschrift 7: 275-326. Mannheims, B., 1951. Tipulidae. — Fliegen der Palacark- tischen Region 15, Lieferung 167: 1-64, pls. I-VII. Mannheims, B., 1962a. Zur Synonymie der europäischen Tipuliden (Dipt.) VII. — Bonner Zoologische Beiträge 13: 193-195. Mannheims, B., 1962b. Die Tipuliden Madeiras (Dipt. Tipulidae). — Notulae Entomologicae 42: 130-136. Mannheims, B. & E. Pechlaner, 1963. Die Tipuliden Nordtirols (Dipt.). — Stuttgarter Beiträge zur Naturkunde 102: 1-29. McAlpine, J. F., 1981. Morphology and terminology - adults. — In: McAlpine, J. F. et al. (eds.) Manual of Nearctic Diptera. Volume 1. Research Branch, Agriculture Canada, Monograph 27: 9-63. Neumann, H., 1958. Der Bau und Funktion der männlichen Genitalapparate von Trichocera annulata Meig. und Tipula paludosa Meig. (Dipt. Nematocera). — Deutsche Entomologische Zeitschrift, Neue Folge 5: 235-298. Oosterbroek, P. & B. Theowald, 1992. Family Tipulidae. — In: Soös, A., L. Papp & P. Oosterbrock (eds.) Catalogue of Palearctic Diptera 1: 56-178. Budapest. Pritchard, G., 1983. Biology of Tipulidae. — Annual Review of Entomology 28: 1-22. Rees, B. E. & G. F. Ferris, 1939. The morphology of Tipula reesi Alexander (Diptera: Tipulidae). — Microentomology 4: 143-178. Riedel, M. P., 1913. Die paläarktischen Arten der Dipteren- (Nematocera polyneura-) Gattung Tipula L. (Dipt.). — Abhandlungen des Vereins für Naturwissenschaftliche Erforschung der Niederrheins- (Bezirksgruppe des deuts- chen Lehrervereins für Naturkunde.) I. Band 1913: 1- 122, pls. II. Riedel, M. P., 1914. Nematocera polyneura. — In: Voyage de Ch. Alluaud et R Jeannel en Afrique orientale (1911- 1912). Résultats scientifiques. Diptera, III: 69-100. Schulz, Paris. Riedel, M. P., 1919. Résultats scientifiques de l'expédition des frères Kuznecov (Kouznetzov) à l’Oural arctique en 1909, sous la direction de H. Backlund. 8. Nematocera polyneura. — Zapiski Rossiiskoi Akademii Nauk 28 (8): 1-10. Savchenko, E. N., 1961. Crane flies (Fam. Tipulidae), sub- fam. Tipulinae: genus Tipula L. (part 1). — Fauna SSSR (N.S. 79) Two-winged insects II (3): 1-488. (In Russian). Savchenko, E. N., 1966. Crane flies. — Fauna Ukraini 14(1): 1-552. (In Ukrainian). Savchenko, E. N., 1968a. On a new species of the crane-fly (Diptera, Tipulidae) from Kamchatka. — Zoologicheskii Zhurnal 47: 1567-1570. (In Russian, English summary). Savchenko, E. N., 1968b. Crane-flies (Diptera, Tipulidae) new for the fauna of the USSR. Dopovidi Akademii Nauk Ukrainskoi RSR 5: 469-472. (In Ukrainian, Russian and English summaries). Savchenko, E. N., 1969. On some little-known species of crane flies (Diptera, Tipulidae) of Transcaucasica. — Zbirnyk Prats Zoolohichnoho Muzeyu 33: 51-60. (In Ukrainian, English summary). Savchenko, E. N., 1979. Phylogenie und Systematik der Tipulidae. — Tijdschrift voor Entomologie 122: 91-126. (German translation of Savchenko 1966: 63-88). Savchenko, E. N., 1983. Crane flies of the family Tipulidae. General part and beginning of systematic part. Subfam. H. DE JONG: Phylogeny of Tipula (Savtshenkia) Dolichopezinae; subfam. Tipulinae (part). — Fauna SSSR (N.S. 127) Two-winged insects II (1-2): 1-585. (In Russian). Stubbs, A. E., 1992. Provisional atlas of the long-palped craneflies (Diptera: Tipulinae) of Britain and Ireland: 1- 134. — Biological Records Centre, NERC Institute of Terrestrial Ecology, Monks Wood, Huntingdon, UK. Tangelder, I. R. M., 1985. Phylogeny of the Nephrotoma dorsalis species-group (Diptera, Tipulidae), mainly based on genital characters. -Beaufortia 35: 135-174. Theischinger, G., 1978. Tipula (Savtshenkia) eleonorae sp.nov. aus Rumänien (Diptera, Tipulidae). — Entomologische Berichten 38: 45-46. Theischinger, G., 1983. Über eine Aufsammlung von Savtshenkia Alexander, 1965, aus dem östlichen Mittelmeerraum (Diptera: Tipulidae). — Entomologische Berichten 43: 26-28. Theowald, B. 1957. Die Entwicklungsstadien der Tipuliden (Diptera, Nematocera), insbesondere der west-palaeark- tischen Arten. — Tijdschrift voor Entomologie 100: 195- 308. Theowald, B. 1967. Familie Tipulidae. (Diptera, Nema- tocera). Larven und Puppen. — Bestimmungsbücher zur Bodenfauna Europas 7: 1-100. Theowald, B., 1973. Tipulidae. — Fliegen der Palaeark- tischen Region 15, Lieferung 300: 321-404. Theowald, B., 1978. Tipulidae. — Fliegen der Palaeark- tischen Region 15, Lieferung 318: 405-436. Theowald, B. & C. Dufour, 1983. Tipula (Savtshenkia) gimmerthali mattheyi ssp.n. from the French Pyrenees. — Mitteilungen der Schweizerischen Entomologischen Gesellschaft 56: 283-284. Theowald, B., C. Dufour & P. Oosterbroek, 1982. The zoogeography of the western Palaearctic Tipulidae (Diptera). Part IV: the Tipulidae of Corsica and Sardinia with a note on Dolichopeza fuscipes Bergroth. — Mitteilungen der Schweizerischen Entomologischen Gesellschaft 55: 317-332. Theowald, B. & P. Oosterbroek, 1981. Zur Zoogeographie der westpalaearktischen Tipuliden II. Die Tipuliden der iberischen Halbinsel (Diptera, Tipulidae). — Beaufortia 30: 179-192. Wood, H. G., 1952. The crane-flies of the South-West Cape (Diptera, Tipuloidea). — Annals of the South African Museum 39: 1-327. Received: 2 June 1994 Accepted: 13 September 1994 APPENDIX A In this appendix taxonomic changes pertaining to Savtshenkia are dealt with. The changes include the description of three new western Palaearctic species of the subgenus, the establishment of five synonymies, the elevation of one subspecies to species rank, and the removal of one species from Savtshenkia. Descriptions of new species Tipula (Savtshenkia) alpha sp. n. (figs. 152-170) Type material. — Holotype dg: Rumania, Mara- all TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 152-164, Tipula (Savtshenkia) alpha sp. n. — 152, male head and appendages, lateral view; 153, male head, dorsal view, antennae and mouthparts omitted; 154, female antenna, basal segments; 155, male terminalia, lateral view; 156, male tergi- te 9-10, dorsal view; 157, male tergite 9-10, posterior view; 158, male sternite 8, posterior margin, ventral view; 159, right sp2, dorsal view; 160, left outer gonostylus, lateral view; 161, left inner gonostylus, lateral view; 162, aedeagal guide with right gonapophysis and fragmentum, posteromedial view; 163, sperm pump and appendages, lateral view; 164, sperm pump and appendages, anterior VIEW. mures, Muntii Rodnei, Borsa, subalpine lake ‘Stiolul’, main sources of river Bistrita Aurie, 1700- 1800 m, 31.V.1974, L. Botosaneanu (ZMAN). — Paratypes 5d, 19, same data as holotype (ZMAN). Material preserved in alcohol. Description Body length 10-12 mm (d), 14 mm (2); wing length 12-13.5 mm (dg), 13.5 mm (2); antennal length 5-5.5 mm (3), ? mm (9; antenna broken, fig. 154). 312 Colour. — Ground colour greyish brown. Head and thorax dark greyish. Antenna with scapus dark grey- ish, remainder of segments dark brown. Palpus dark brown. Prescutum with single broad medial and pair of lateral dark stripes. Wing veins brown; pterostigma brown, distinct; membrane unpatterned. Haltere yel- lowish brown, including knob. Trochanters of same dark greyish colour as coxae, femora yellowish brown in basal section, from about half length darkening to- wards tip, remainder of legs dark brown. Abdomen dark brown, terminalia slightly darker. (Colour de- Af fc ‘off GEO P 169 170 H. DE JONG: Phylogeny of Tipula (Savtshenkia) Figs. 165-170. Tipula (Savtshenkia) alpha sp. n., female. — 165, ovipositor, lateral view; 166, sternite 8 and hypogynial valves, ventral view; 167, sternite 8 and hypogynial valves, dorsal view; 168, sternite 9, medial part, ventral view; 169, genital fork, dorsal view; 170, spermathecae. duced by approximation, material preserved in alco- hol). Head (figs. 152-154). — Rostrum shorter than re- mainder of head, nasus well developed; eyes small, dorsally and ventrally separated by about four times diameter of scapus; occiput perpendicular to lateral sides of head (dorsal aspect, fig. 153); male antenna about four times as long as head, longest verticils about one and a half times as long as diameter of fla- gellomeres at base; female antenna shorter (flagella of both antennae broken beyond basal flagellomeres in female paratype, fig. 154). Thorax. — Wing fully developed in both sexes, squama devoid of macrotrichia; cell ml gradually widening towards wing margin. Tarsal claws tooth- less in both sexes. Male terminalia (figs. 155-164). - Compactly built, tergite 9-10 and gonocoxites separated by membra- nous zone (fig. 155). Tergite 9-10 (figs. 156, 157) broad and short, with medial membranous area occu- pying about one third of total width; posterior margin widely V-shaped, with shallow emargination near me- dian line; margin ventrally set with scattered hairs, without spines; lateral corners broadly rounded. Posterior margin of sternite 8 unmodified (fig. 158). Gonocoxites ventrally separated by membranous area, membrane with pair of weakly sclerotized bands ex- tending anteriorly from base of aedeagal guide. Suture x narrow, about one fourth length of gonocoxite (fig. 155). Sp2 (fig. 159) long, medial blackish sclerotized margin moderately serrate; dorsally separate from ter- gite 9-10; connexion between sp2 and posterior mar- gin of foramen of gonostyli short (fig. 159, arrow). Outer gonostylus (fig. 160) broad, its tip anteriorly in- clined, posterior margin only moderately blackened. Inner gonostylus (fig. 161) with relatively slender blackish sclerotized anterior part; lateral rim in its ven- tral part directed towards base of gonostylus; postero- laterally near base with blackened rim (fig. 161, ar- row). Aedeagal guide (fig. 162) a short and plump blackish sclerotized structure, gonapophysis reduced to a sclerotized strip (fig. 162, arrow). Fragmentum (fig. 162) a rounded lobe with a few long hairs dorso- medially. Sperm pump (figs. 163, 164) with anterior apodemes short, posterior apodemes well developed and obliquely placed, compressor apodeme dorsally emarginate, blackish; lumen of endophallus well de- veloped; ventral wall convex. Aedeagus short, anterior- ly reaching segment 8 (fig. 155), tubular throughour, gradually narrowing towards tip. Female terminalia (figs. 165-170). — Cercus and hypogynial valve well developed, robust, relatively short (fig. 165). Sternite 8 dorsolaterally near base of hypogynial valve with concavity (fig. 165, arrow); ventral membrane in posterior part of sternite 8 en- closing medial base of hypogynial valves (fig. 166). Dorsomedial margin of hypogynial valve only moder- ately blackened (fig. 167). Sternite 9 with slender me- dial projection (fig. 168). Genital fork elongate, ante- riorly broader (fig. 169). Three spermathecae, globular to oblong, of about same size, moderately sclerotized (fig. 170). Etymology. — The name of this species was prompted by its phylogenetic position within Savtshenkia and is the pendant of Tipula (Savtshenkia) omega sp. n. A noun in apposition. Remarks. — Tipula (Savtshenkia) alpha differs from its presumed sister species 7. (S.) grisescens by the TiJDsCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 171-176. Tipula (Savtshenkia) omega sp. n., male. — 171, head and appendages, lateral view; 172, head, dorsal view, an- tennae and mouthparts omitted; 173, terminalia, lateral view; 174, tergite 9-10, dorsal view; 175, tergite 9-10, posterior view; 176, left sclerotization proctiger, lateral view. colour of the coxae (dark greyish in alpha, yellowish brown in grisescens), the less elongate male antenna (see figs. 9 and 152), the separated gonocoxites (fused in grisescens), the broader outer gonostylus (see figs. 76 and 160), the more slender anterior part of the in- ner gonostylus and the presence of a sclerotized rim posterolaterally at the base of the same structure, the shallowly emarginate compressor apodeme of the sperm pump (deeply emarginate in grisescens), the rounded body of the sperm pump (with conical ante- roventral extension in grisescens). The type locality of alpha lies above the tree line on the subalpine plain at an altitude of 1700-1800 m. 314 The locality has a characteristic vegetation of Pinus mugo Turra and the grass Nardus stricta Linnaeus. At the end of May thaw has set in, but the surface of the area is still extensively snow- and ice-covered. Under sunny conditions alpha emerges in large numbers as soon as holes begin to appear in the ice-covering. Although fully winged, the insects tend to crawl around on the ice and surrounding grasses. Under cloudy conditions alpha shows no sign of activity, suggesting that the behaviour of the species is largely influenced by the environmental temperature. In these aspects, the biology of alpha resembles that of the trichopteron Chionophylax czarnohoricus Dziedz., H. DE JONG: Phylogeny of Tipula (Savtshenkia) Figs. 177-184. Tipula (Savtshenkia) omega sp. n., male. — 177, sternite 8, posterior margin, ventral view; 178, sternite 8, pos- terior margin, posterior view; 179, right sp2, dorsal view; 180, left outer gonostylus, lateral view; 181, left inner gonostylus, lateral view; 182, aedeagal guide with right gonapophysis and fragmentum, posteromedial view; 183, sperm pump and ap- pendages, lateral view; 184, sperm pump and appendages, anterior view. a species which occurs in the same locality (Dr Botosaneanu, pers. comm.). Tipula (Savtshenkia) omega sp. n. (figs. 171-184) Type material. — Holotype d: Italy, Sardinia, Fiume Taloro, 670 m, 9°31/40°06, 21.X.1981, H. Malicky (zman). — Paratypes: 14, Italy, Sardinia, Rio Aratu, 970 m, 9°15/40°02, 22.X.1981, H. Malicky (ZMAN). Material preserved in alcohol. Description Body length 12-14.5 mm (4); wing length 14- 15.5 mm (d); antennal length 5-6 mm (¢). Colour. — Ground colour yellowish brown. Vertex of head and dorsal and lateral sides of thorax darker, probably slightly greyish pruinose in dry specimens. Antenna yellowish brown, scapus and pedicel some- what lighter coloured than flagellum. Palpus yellow- ish brown. Presutum with confluent medial and pair of lateral dark stripes indicated. Wing veins yellowish brown; pterostigma yellowish brown, distinct; mem- brane vaguely patterned. Halteres yellowish brown, including knob. Legs yellowish brown, femora and tibiae darkened at extreme tip, tarsi darkening to- wards tip. Abdomen yellowish at base, darkening to- wards tip. (Colour deduced by approximation, mate- rial preserved in alcohol). Head (figs. 171, 172). — Rostrum about as long as remainder of head, nasus well developed; eyes well developed, dorsally separated by about two and a half times diameter of scapus, ventrally by about two times diameter of scapus; occiput gradually narrow- ing towards cervix (dorsal aspect, fig. 172); male an- tenna just less than two times length of head, longest verticils about as long as flagellomeres (fig. 171). Thorax. — Wing fully developed, squama devoid of macrotrichia; cell ml elongate bell-shaped, narrow- ing towards wing margin. Tarsal claws with small me- dial teeth in male. Male terminalia (figs. 173-176). — Large, tergite 9- 315 Ty DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 185-191. Tipula (Savtshenkia) trinacria sp. n., male. — 185, terminalia, lateral view; 186, tergite 9-10, dorsal view; 187, tergite 9-10, posterior view; 188, left sclerotization proctiger, lateral view; 189, sternite 8, posterior margin, ventral view; 190, sternite 8, posterior margin, posterior view; 191, right sp2, dorsal view. 10 and gonocoxites separated by membranous zone (fig. 173). Tergite 9-10 (figs. 174, 175) elongate, with narrow medial membranous area; posterior mar- gin with pair of broad laterally diverging extensions; extensions not blackish sclerotized, devoid of spines along ventral margin. Proctiger with rather diffuse and narrow lateral sclerotization (fig. 176). Sternite 8 (figs. 173, 177, 178) posteriorly with diverging broad dorsoventrally flattened lobes, each lobe carrying about 35-40 black spines. Gonocoxites ventrally fused with narrow medial carina; at level of insertion of inner and outer gonostyli with concentration of 316 long and strong setae (fig. 173). Suture x narrow, about two-fifth length of gonocoxite (fig. 173). Sp2 (fig. 179) medially protruding, blackish sclerotized medial margin entire; sp2 fused with tergite 9-10; connexion between sp2 and posterior margin of fora- men of gonostyli long. Outer gonostylus (fig. 180) with elongate slender anterior part, posterior blackish sclerotized part relatively broad, with short poste- rodorsal extension, slightly concave. Inner gonostylus (fig. 181) with anterior and posterior parts about as long, posterior part caudally terminating in a densely pubescent lobe. Aedeagal guide (fig. 182) long and H. DE Jone: Phylogeny of Tipula (Savtshenkia) Figs. 192-196. Tipula (Savtshenkia) trinacria sp. n., male. — 192, left outer gonostylus, lateral view; 193, left inner gonosty- lus, lateral view; 194, aedeagal guide with right gonapophysis and fragmentum, posteromedial view; 195, sperm pump and appendages, lateral view; 196, sperm pump and appendages, anterior view. slender, blackish sclerotized, gonapophysis short, dorsally directed. Fragmentum (fig. 182) anteriorly with sclerotized bar which dorsally carries a bundle of long hairs, posteriorly weakly sclerotized. Sperm pump (figs. 183, 184) with large anterior apodemes, posterior apodemes caudally directed, compressor ap- odeme dorsally deeply emarginate, wings broad; lu- men endophallus small, body wall ventrally concave. Aedeagus elongate (fig. 173), anteriorly reaching seg- ment 3, tubular throughout. Etymology. — The phylogenetic position of this species within Savtshenkia suggested its name. A noun in apposition. Remarks. — The species omega very much resembles the species mannheimsi and trinacria sp. n. It differs from these species by the number of spines on the posterior lobes of male sternite 8 (ca. 35-40 in omega to ca. 50 in mannheimsi and ca. 65 in trinacria), the absence of spines on the ventral margin of male ter- gite 10 in omega (spines absent also in mannheimsi, present in #rinacria), and in the shape of the inner and outer gonostyli. The female of omega remains unknown. Tipula (Savtshenkia) trinacria sp. n. (figs. 185-202) Type material. — Holotype d: Italy, Sicily, Nebrodi, 10 km north of Capizzi, Fiume Troina, 1200-1400 m, 11-14.X.1993, P. Oosterbroek & C. Hartveld (zman). — Paratypes: 138,3, same data as holotype; 4d, Italy, Sicily, Nébrodi, 5 km north of Monte Soro, Biviere di Cesaro, 1300 m, 12.X.1993, P. Oosterbroek & C. Hartveld; 2d, Italy, Sicily, Nebrodi, 6 km east of Serra del Re, Foresta Vecchia, Fiume di Saracena, 1100 m, 13.X.1993, P. Oosterbroek & C. Hartveld; 46, 39, Italy, Sicily, Madonie, 6 km north of Monte San Salvatore, Torrente Vicaretto, 1300 m, 15-21.X.1993, P. Oosterbroek & C. Hartveld, (ZMAN). Description Body length 13.5-16 mm (dé), 15.5-17 mm (2); wing length 16-19 mm (|), 16.5-18 mm ( ©): anten- nal length 6-7 mm (dg), 5-5.5 mm (2). Colour. — Ground colour yellowish brown. Excepting rostrum, head greyish pruinose. Antenna with yellowish scapus and pedicel, flagellum almost uniformly brownish, basal flagellomeres somewhat lighter. Palpus yellowish brown. Dorsal and lateral sides of thorax slightly greyish pruinose. Presutum 317 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Figs. 197-202. Tipula (Savtshenkia) trinacria sp. n., female. — 197, ovipositor, lateral view; 198, sternite 8 and hypogynial val- ves, ventral view; 199, sternite 8 and hypogynial valves, dorsal view; 200, sternite 9, medial part, ventral view; 201, genital fork, dorsal view; 202, spermathecae. with pair of approximating slender medial and pair of lateral dark stripes, medial stripes anteriorly lighter coloured. Wing veins brown; pterostigma brown, dis- tinct; membrane vaguely patterned and iridescent. Halteres with yellowish brown pedicel, knob brown. Legs yellowish brown, femora and tibiae darkened at extreme tip, tarsi largely dark brown. Abdomen yel- lowish at base, darkening towards tip. Head. — Very much as chat of omega sp. n. (see figs. 171, 172; see for description of head under that spe- cies, above). Thorax. — Wing fully developed in both sexes, squama devoid of macrotrichia; cell m1 elongate bell- shaped, narrowing towards wing margin. Tarsal claws with small medial teeth in the male, female claws toothless. Male terminalia (figs. 185-196). — Large, tergite 9- 10 and gonocoxites separated by membranous zone (fig. 185). Tergite 9-10 (figs. 186, 187) elongate, 318 with narrow medial membranous area; posterior mar- gin with pair of broad laterally diverging extensions; extensions blackish sclerotized along ventral margin, ventral margin set with small black spines. Proctiger with rather well demarcated slender lateral sclerotiza- tion (fig. 188). Sternite 8 (figs. 189, 190) posteriorly with diverging broad flattened lobes, each lobe carry- ing about 65 slender black spines. Gonocoxites ven- trally fused with narrow medial carina. Suture x nar- row (fig. 185). Sp2 (fig. 191) medially protruding, blackish sclerotized medial margin entire; sp2 fused with tergite 9-10; connexion between sp2 and poste- rior margin of foramen of gonostyli long. Outer go- nostylus (fig. 192) with elongate slender anterior part, posterior blackish sclerotized part elongate, upcurved. Inner gonostylus (fig. 193) with anterior and posteri- or parts about as long, posterior part anteriorly ex- tended, posteriorly terminating in cup-shaped struc- ture. Aedeagal guide (fig. 194) long and slender, blackish sclerotized, gonapophysis short, dorsally di- rected. Fragmentum (fig. 194) anteriorly with sclero- tized bar which dorsally carries a bundle of long hairs, posteriorly weakly sclerotized. Sperm pump (figs. 195, 196) with large anterior apodemes, posterior ap- odemes caudally directed, at tip somewhat upturned, compressor apodeme dorsally deeply emarginate, with narrow cleft, wings broad; lumen endophallus small, wall ventrally concave. Aedeagus elongate (fig. 185), anteriorly reaching segment 4, tubular through- out. Female terminalia (figs. 197-202). — Cercus and hypogynial valve well developed, cercus approximate- ly straight (fig. 197). Ventral membrane in posterior part of sternite 8 terminating in between bases of hy- pogynial valves (fig. 198). Dorsomedial margin of hy- pogynial valve only moderately blackened (fig. 199), ventrally at base with concavity (figs. 197, 198). Sternite 9 with relatively narrow and slender medial projection, concave lateral of this structure (fig. 200). Genital fork short and broad, broadest at anterior side (fig. 201). Three spermathecae, globular, of about same size, blackish sclerotized (fig. 202). Etymology. — Trinacria is an old name of Greek or- igin for Sicily. A noun in apposition. Remarks. — The species trinacria very much resem- bles mannheimsi and omega sp. n. For distinguishing characters see under omega. The type specimens of trinacria were captured along rivers and a lake in de- ciduous forests. The specimens collected in the peri- od from 15-21 October were taken from the moss covered river banks of the Torrente Vicaretto. Among these latter specimens was one pair in copula. Synonymy insignifica Alexander, 1924 = invenusta invenusta Riedel, 1919 syn. n. Alexander (1924) described insignifica as a new species based on a single male originating from Mount Washington, New Hampshire, U.S.A. Alexander (loc. cit.) recorded the collection of BSNH as the depository of the holotype (holotype not exam- ined). I have studied specimens labelled by Alexander as metatypes, i.e. specimens compared with the type specimen(s), originating from Mount Katahdin, Maine and Mount Madison, New Hampshire (- USNM). Examination of these specimens revealed only slight differences between the Nearctic insignifica and the Palaearctic invenusta invenusta in the distance of the eyes on the ventral side of the head (see also ‘Discussion of adopted phylogeny’). No other struc- tural differences between insignifica and invenusta in- venusta were found and I therefore consider insignifi- H. DE sone: Phylogeny of Tipula (Savtshenkia) ca a junior synonym of invenusta invenusta. |Contrary to Alexander’s (1942, 1966b) assertion, the posterior margin of tergite 9-10 is posteroventrally armed with small black spines in insignifica, as it is in invenusta). broweri Alexander, 1940 = fragilina Alexander, 1919 syn. n. Alexander (1940) described broweri based on 12 males and two females from Mount Katahdin, Maine, U.S.A. (types in usum). Despite Alexander’s claim that broweri should be entirely distinct from its relatives, examination of type material of broweri showed that it is morphologically identical with frag- ilina. I therefore consider broweri a junior synonym of fragilina. nebulipennis Alexander, 1919 = alpium Bergroth, 1888 syn. n. Alexander (1919a) introduced nebulipennisas a new species based on two males from Battle Harbour, Labrador, Canada (holotype in Brooklyn Museum, New York, U.S.A., paratype in USNM). Later, nebuli- pennis was also recorded from Quebec and New Hampshire (Alexander 1965b). Examination of mate- rial of nebulipennis in the Alexander collection (usNM) revealed no significant morphological differences between these specimens and the European species al pium. I therefore consider nebulipennis a junior syno- nym of alpium. The larvae of alpium were recorded from terrestrial mosses (Brindle 1958, 1959, 1960). Theowald (1957, 1967) recorded the larvae further- more from the litter layer beneath oaks. Possibly a/pi- um has been introduced into North America in a way similar to the European species Tipula (Tipula) palu- dosa Meigen (see Alexander 1965a). packardi Alexander, 1928 = invenusta invenusta Riedel, 1919 syn. n. Alexander (1928) described Tipula packardi based on a single male from Hopedale, Labrador, Canada. The holotype of packardi is preserved in cncı (holo- type not examined). An additional male specimen originating from the type locality of packardi is present in USNM. This latter specimen was compared by Alexander with the holotype of packardi and con- sidered conspecific. Examination of the metatype of packardi in USNM showed that it has the eyes on the ventral side of the head slightly more approximate than the examined specimens of invenusta invenusta (see also ‘Discussion of adopted phylogeny’, and z7- significa, this section). No other structural differences were found and I therefore consider packardi a junior synonym of invenusta invenusta. perparvula Alexander, 1926 = ignobilis Loew, 1863 syn.n. (partim?) TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Loew (1863) described the species ignobilis based on the female sex from material captured by Osten- Sacken in the District of Columbia, U.S.A. (holotype in MCZC, not examined). According to its labels, Alexander compared a male specimen captured by Osten Sacken in Catskill, New York, with the type specimen of ignobilisand considered the two conspe- cific. Examination of this inferred male specimen of ignobilis and the paratype male of the species perpar- vula (both in USNM) revealed that these two speci- mens belong to the same species. Alexander (1926), describing perparvula based on two males originating from Aweme, Manitoba, Canada (holotype in CNCI, paratype in USNM), explicitly stated that perparvula has a relatively stout nasus. Examination of the para- type of perparvula shows that this structure is entirely absent in the specimen, as it is in the species ignobilis. Eicher the holotype and paratype of perparvula be- long to two different species, with the paratype being conspecific with zgnobilis, or both type specimens are conspecific and belong to ignobilis. Examination of the holotype of perparvula should settle this question. Subspecies raised to species rank haennii Dufour, 1991 This taxon was originally described as a subspecies of subsignata (Dufour 1991). The tip of the aedeagal guide of haennii does not show the abrupt constric- tion which probably represents a synapomorphy of a monophyletic group containing corsosignata, subsig- nata subsignata and subsignata cazorla (character 8, above). Because of its phylogenetic position next to this clade, haennii is elevated from subspecific to spe- cific rank. Species removed from Savtshenkia convexifrons Holmgren, 1883 Savchenko (1961) placed the species convexifrons Holmgren in his subnodicornis group of Pterelachisus sensu Savchenko, nec Rondani. Savchenko (1961: 223) based its membership of the subnodicornis group on the shape of the inner gonostylus, and its incorpo- ration in Prerelachisus sensu Savchenko on the proxi- mal position of the fork of vein M3+4. At the same ti- me, Savchenko underlined the isolated position of convexifrons within this grouping. As noticed in the introduction to the present paper, the name Savtshenkia replaces Pterelachisus sensu Savchenko. Theowald (1973) followed Savchenko in considering convexifrons a species of Savtshenkia, but split the sub- nodicornis group sensu Savchenko into two species groups (table 3). I remove convexifrons from Savtshenkia because of the absence of the defining ap- omorphies 53 and 83 of Savtshenkia. Furthermore, 320 the species convexifrons difters from the species of Savtshenkia by the shape of male tergite 9-10, the ventral structure of the fused gonocoxites, the shape of the elongate aedeagal guide and its gonopophyses, and, most noteworthy, by the presence of a complete genital bridge (a plesiomorphy; the medial part of the genital bridge is absent in all species of Savtshenkia). The fork of vein M3+4 is located proximally in many genera and subgenera of Tipulidae other than Savtshenkia. The resemblance of the inner gonostylus of convexifrons to the inner gonostyli of plesiomorp- hous species of Savtshenkia is only superficial. At the moment I am unable to establish the actual phyloge- netic relationships of convexifrons and can not allocate it to any other subgenus or species group of Tipulidae. APPENDIX B In this appendix are listed the species and subspe- cies of Savtshenkia which, because of deficiency of da- ta, were not incorporated in the character state matrix given in table 4. Incompleteness of data resulted of several causes. For some of the pertaining taxa no ma- terial was examined, for others the material examined was incomplete as regards the structures of the male terminalia, for others again only the female is known, thus excluding their incorporation in a phylogeny which is largely based on characters of the male ter- minalia. Where possible, the presumed phylogenetic position of the species and subspecies concerned is in- dicated. For the deposition of the type material of species described by Savchenko see Kandybina, Lantzov & Savchenko (1987). Unexamined imperfecta Riedel, 1914. — Description based on single brachypterous male specimen from Kenya (Mount Kenya) and single brachypterous female spe- cimen, possibly belonging to the same species, from Tanzania (Mount Kilimanjaro). Types in MNHN. Species doubtfully belongs to Savtshenkia. Distribu- tion: Kenya, Tanzania. letifera Alexander, 1951. — Description based on single male from China (Tibet, Shugden Gompa a.k.a. Hsiu-teng). Type in BMNH. Inner and outer gonostyli quite similar to those of Kashmirian sordidipes and venerabilis. Differences with these species in structures of posterior margins of tergite 9-10 and sternite 8. Shape of sternite 8 suggests relationship with Nearctic graciae and Palaearctic kiushiuensis and koreana. Alexander (1953) illustrated details male terminalia of letifera. Distribution: China (south-east Tibet). minuscula Savchenko, 1971. Originally described as Tipula (Savtshenkia) nana by Savchenko (1968a) based on two males and single female from USSR (Kamchatka). Types in zıas. Savchenko, probably correctly, considered minuscula closely related to Nearctic ignobilis, with which it should differ in colour characteristics. Details male terminalia of mi- nuscula illustrated by Savchenko (1968a). Distribution: Russia (Kamchatka). nivalis Savchenko, 1961. — Description based on three males and four females from USSR (Stavropolskiy kray and Gruzinskaya SSR). Types in zıas. Savchenko (1961) compared nivalis with glau- cocinerea. Theowald (1973) considered both very closely related and suggested that nivalis might be considered a subspecies of glaucocinerea. Details male terminalia of nivalis illustrated by Savchenko (1961) and Theowald (1973). Distribution: Russia (Stavropolskiy kray), Georgia. productella Alexander, 1928. — Description based on single male from Canada (Labrador). Type in cenci. According to Alexander (1928) most closely al- lied to Nearctic fragilina. Structure of sternite 8 of productella of same build as found in clade fragilina to subvafra of present paper; productella probably be- longs to this clade; synonymy with one of its species possible. Distribution: Canada (Labrador). subalpium Savchenko, 1961. — Description based on single male from USSR (Krasnodarskiy kray, Glavniy Mountains). Type in zıas. Savchenko (1961) compared subalpium with alpium, with which it shares same general colour and structure of inner and outer gonostyli and to which it is probably closely re- lated. Differs quite remarkably from alpium in struc- ture of posterior margin of sternite 8: whereas poste- rior margin of sternite 8 is U-shaped emarginate and carries spines along its edge in alpium (fig. 105), emargination is filled up with membrane covered with short setae in subalpium. According to Savchenko, there is a short pubescent extension in between what he denotes sternites 8 and 9. This structure could well be homologous with midventral extension found in alpium and macaronesica, which in dry material is often hardly visible or seems to orig- inate from posterior margin of sternite 8 (see fig. 39). Savchenko (1969) depicted male terminalia of sub- apium in lateral and posterior view. Distribution: Russia (Krasnodarskiy kray), Georgia. tetragramma Edwards, 1928. — Description based on single male from China (Yunnan). Type in BMNH. Edwards (1928) was of the opinion that tetragramma must bear a rather strong resemblance to kiushiuensis. Distribution: China (Yunnan). Incomplete male terminalia kiushiuensis Alexander, 1925. — Description based on single male and single female from Japan (Kyushu, Mount Kirishima). Types in usnm. Glass slides with terminalia male holotype and additional male speci- men (USNM) do not allow study of all relevant struc- H. DE Jone: Phylogeny of Tipula (Savtshenkia) tures. Related to species in range jeekeli to aster of cla- dogram fig. 2. Posteriorly produced unarmed medial part of posterior margin male sternite 8 suggests rela- tionship with koreana. Savchenko (1968b) illustrated details male terminalia of kiushiuensis. Distribution: Russia (Primorskiy kray), Japan (Kyushu, Shikoku). koreana Alexander, 1934. — Description based on single male and four females from North Korea (Mount Kongo). Types in usnm. Glass slide with ter- minalia male holotype does not allow study of all rel- evant structures. Related to species in range jeekeli to aster of cladogram fig. 2. Posteriorly produced un- armed medial part of posterior margin male sternite 8 suggests relationship with kiushiuensis. Alexander (1934) illustrated details male terminalia and wing of koreana. Distribution: Russia (Sakhalin), North Korea, Japan (Shikoku). mohriana Alexander, 1954. — Description based on single male and single female from Japan (Shikoku, Mount Ishizuchi). Types in usnm. Glass slide with terminalia male holotype does not allow study of all relevant structures. Most probably sister species of clade fragilina to subvafra of present paper as implied by fused sp2 and tergite 9-10 (character 54), slender anterior part outer gonostylus (character 45), bifid tip of anterior part of inner gonostylus (character 30), and widely emarginate bilobed and unarmed posteri- or margin of male sternite 8 (cf. character 60). Differing most notably from species of clade fragilina to subvafra by presence of spinous gonapophysis (au- tapomorphy) and absence of blackened dorsal margin of fragmentum (cf. character 17). Details male termi- nalia and wing of mohriana illustrated by Alexander (1955). Distribution: Japan (Shikoku). sciadoptera Alexander, 1964. — Described as gurtu- lifera by Alexander (1961) based on two males from India (Himalayas). Types in usnm. Glass slides with terminalia male holotype and paratype do not allow study of all relevant structures. Probably related to species in range baltistanica to chrysocephala of clado- gram in fig. 2. Alexander (1961) illustrated details male terminalia of sciadoptera (as guttulifera). Distribution: India (Sikkim). Known in female sex only aberdareica ulugurica Alexander, 1962. — Description based on single female from Tanzania (Uluguru Mountains). Type in MRAC, not examined. According to Alexander (1962) closely related to no- minotypical form, sharing several colour characteris- tics with chrysocephala. Distribution: Tanzania. nephrotomoides invariegata Alexander, 1956. — Description based on single female holotype from Uganda (Ruwenzori Range). Type in BMNH. Agreeing in its dark colour with the nominotypical form from Zaire and Uganda. Distribution: Uganda. 321 TijDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 APPENDIX C Autapomorphies of the species and subspecies of Savtshenkia aberdareica: no autapomorphy recognized. aberdareica ulugurica: no autapomorphy recog- nized (known of female holotype only). akeleyi: inner gonostylus with dorsal margin anteri- or part undulating. alpha: posterior rim of inner gonostylus ventrally blackish sclerotized (fig. 161, arrow). alpium: shape of posterior margin of male tergite 9- 10; shape of posterior margin of male sternite 8 (fig. 105); shape of extension of midventral area (long; fig. 39); asbolodes. inner gonostylus with anterior part rath- er slender. aspromontensis shape of inner gonostylus (crest, fig. 61); shape of male tergite 9-10 (two pairs of pointed extensions, fig. 135). aster. shape of aedeagal guide; gonapophysis slen- der, downcurved and pointed; shape of male tergite 9-10; shape of posterior margin of male sternite 8 (fig. 109). atlas. extensions of male tergite 9-10 mediolaterally compressed. baltistanica: arrangement of setae on posterior mar- gin of male sternite 8. benesignata: shape of inner gonostylus (fig. 72). boreosignata: broad membranous zone of tergite 9- 10 (fig. 138); reduction of spines on lobes on posteri- or margin male sternite 8; fragmentum with low an- terior part; shape of aedeagal complex (fig. 99). breviantennata: anterior part of inner gonostylus widening towards apex, lateral ridge of inner gonosty- lus short and angled. caligo: shape of posterior margin of male tergite 9- 10. cheethami: midventral area posteriorly with long conical extension (fig. 35). chrysocephala: anterior part of inner gonostylus constricted before tip (lateral view). confusa: inner gonostylus with two blackish sclero- tized dentate ridges on lateral side of posterior part (fig. 58); male tergite 9-10 with truncate extensions close together; shape of posterior margin of male ster- nite 8. corsosignata: shape of inner (fig. 69) and outer go- nostyli. cyrnosardensis inner gonostylus caudodorsally with posteriorly produced pubescent lobe; outer gonosty- lus with slender extension on posterior margin at about two-thirds from base; apex of elongate frag- mentum medially curved. draconis. shape of inner gonostylus; reduced nasus. eleonorae. no autapomorphy recognized. 322 elgonensis no autapomorphy recognized. eugeni. extensions of male tergite 9-10 upcurved (lateral view); shape of inner gonostylus; anterior ap- odemes sperm pump elongate and dorsally directed, compressor apodemes mediolaterally flattened in api- cal part. fragilina: inner gonostylus with acute sclerotized extension at ventral tip of posterior ridge. fragilis. aedeagal guide elongate and slender (fig. 22); suture x broad, sigma-shaped. gimmerthali: inner gonostylus with anterior part gradually sloping towards bifid apex (fig. 66). gimmerthali mattheyi. outer gonostylus without posterior sclerotized appendage (present in both oth- er subspecies). gimmerthali pteromaculata: inner gonostylus with relatively broad anterior part (fig. 67); outer gonosty- lus broadened above midlength (fig. 84). glaucocinerea: shape of extensions of tergite 9-10. goriziensis. shape of inner gonostylus (fig. 54). graciae. posterior margin of male sternite 8 medial- ly with rounded extension set with long and strong setae. grisescens. extensions of male tergite 9-10 dorsoven- trally flattened (figs. 121, 122). haennii: fragmentum 3-lobed (fig. 27); inner go- nostylus with ridge separated in posterior and ventral part, ventral part ridge extended over posterior part of inner gonostylus, and dorsal crest anteriorly abruptly terminating (fig. 68). hancocki: male antenna reduced in length. hartigiana: shape of inner gonostylus (long anterior part, strongly protruding lateral ridge); extension of midventral area gonocoxites (fig. 33). holoptera: posterior margin of male sternite 8 with a few strong setae (fig. 107). ignobilis shape of outer gonostylus; shape of sp2. imperfecta: wings reduced in both sexes. interserta: shape of inner gonostylus; outer gonos- tylus large, posteriorly inclined. invenusta: shape of posterior margin of male tergite 9-10 (fig. 136). invenusta microinvenusta: inner gonostylus anteri- orly more curved forward than in both other forms; differing primarily in smaller size and darker colour from both other forms. invenusta subinvenusta: inner gonostylus relatively slender. jeekeli: shape of inner gonostylus; shape of outer gonostylus (fig. 80); shape of posterior margin of male sternite 8; sperm pump with sclerotized sperm duct; apex of aedeagus with spines. kiushiuensis. shape of posterior margin of male ster- nite 8. koreana: shape of posterior margin of male sternite 8. letifera: no autapomorphy recognized (probably very similar to sordidipes and venerabilis). limbata: anterior part of inner gonostylus produced into long and bifid extension (fig. 64). lundbladi: shape of inner gonostylus (elongate, with short crest); shape of outer gonostylus (fig. 78); shape of posterior margin of sternite 8 (fig. 103); surface of sp2 anterodorsally with cone-shaped extension. macaronesica: shape of extension of midventral area (short). mannheimsi: outer gonostylus with incision ventral of posterior sclerotized spur. minuscula: no autapomorphy recognized (probably very similar to ignobilis). mobriana: gonapophysis as slender curved spine. multipicta: shape of inner gonostylus (high anterior part with long and slender anterior apex); shape of outer gonostylus (oblong). nephrotomoides. blackish species. nephrotomoides invariegata: no autapomorphy rec- ognized (known of female holotype only). nielsent. shape of inner gonostylus; shape of outer gonostylus (fig. 85); shape of medial extension of male sternite 8. nivalis no autapomorphy recognized (probably very similar to glaucocinerea). obsoleta: shape of inner and outer gonostyli; shape of posterior margin of male sternite 8 (fig. 106); spi- nous extensions on posterodorsal part fragmentum (fig. 21). odontostyla: shape of male tergite 9-10; inner go- nostylus dorsally with long and dense pubescence (fig. 63); apical part of fragmentum bent and posteri- orly directed; gonapophysis twisted. omega: inner gonostylus with hump in anterior half of posterior part (fig. 181). ornata: wing cell a2 almost symmetrical in distal tip; posterior extensions male tergite 9-10 dorsoven- trally flattened, laterally curved, ventrally carrying long thin spinous setae; shape of posterior margin of male sternite 8; midventral anterior extension of male sternite 9 elongate, membranous; shape of appendage of midventral area gonocoxites (fig. 40); suture x broad, enclosing isolated sclerite; female tergite 9 me- dially reduced to strip. pagana: female brachypterous. pechlaneri: male tergite 9-10 with pair of low extra extensions lateral of usual extensions on posterior margin; shape of inner gonostylus (fig. 65). persignata: posterior part of inner gonostylus corru- gated (fig. 73). persignata tofina: differing in details of male termi- nalia from nominotypical form. phoroctenia: shape of male tergite 9-10; shape of posterior margin of male sternite 8; lateral ridge of in- ner gonostylus posterodorsally angular; suture x of H. DE JONG: Phylogeny of Tipula (Savtshenkia) gonocoxite Y-shaped; shape of extension of midven- tral area gonocoxites (fig. 41). postposita: shape of outer gonostylus. productella: no autapomorphy recognized. rufina: extensions of tergite 9-10 extremely long and slender; shape of inner gonostylus; tip of gona- pophysis medially curved; posterior margin of male sternite 8 bulbously curved. rufina maderensis. larger than nominotypical spe- cies. sardosignata: shape of inner gonostylus; shape of outer gonostylus (fig. 88); shape of posterior margin of male sternite 8. sciadoptera: shape of inner gonostylus. serrulifera: shape of inner gonostylus (fig. 60); shape of outer gonostylus; shape of posterior margin of male sternite 8. signata: posterior part of inner gonostylus anterior- ly rounded (fig. 74); midventral extension on posteri- or margin male sternite 8 large, with dorsal sclerite (fig. 116). simulans. inner gonostylus with lateral ridge dorsal- ly produced into pointed extension; shape of exten- sion of midventral area. sordidipes. legs extensively blackened. staegeri: extensions of male tergite 9 widely separ- ate, mediolaterally flattened; male sternite 8 midven- trally conical; posterior extensions of male sternite 8 elongate; aedeagal guide posteriorly curved; shape and length of gonapophysis; shape of inner gonosty- lus; female sternite 8 dorsally extended; base of hypo- gynial valve laterally bulging. subalpium: shape of posterior margin of male ster- nite 8. subnodicornis. tergite 9-10 with extension in between lateral pair (fig. 128). subsignata: together with next subspecies distin- guished by lateral ridge being acutely bent and anteri- orly produced in ventral part (fig. 70); differing from subsignata cazorla in details of inner gonostylus. subsignata cazorla: differing from nominotypical subspecies by shape of inner gonostylus (anterior part dorsally abruptly narrowed, lateral view). subvafra: shape of male tergite 9-10; shape of inner gonostylus. tetragramma: Wing pattern. trinacria: outer gonostylus with extremely long posterior sclerotized appendage (fig. 192); sclerotiza- tion of proctiger well demarcated (fig. 188). tulipa: inner gonostylus with relatively short lateral ridge; midventral area gonocoxites broad and long pubescent. venerabilis no autapomorphy recognized (very similar to sordidipes). villeneuvii: small and posteriorly curved outer go- nostylus; gonocoxite posterodorsally acutely angled. 925 E, VOLUME 137, 1994 Agr RIINA | it Î j Ren peeler | Hr vasi toy 30 puita Feu ’ NC uti) wo AL; di AAN ' y hy vn Au î i LÉ mk bir a LE 1) Ce IMA MN LUE byt de ai EU rn An EN RET AT rik nerd be A ur * | u à à 4 ‘ ] i a RAT CITE tot] MUR, à [ch Ti, ve L , Li austere ŒIUL I RITIRO | tet t { NT mad { ik 4) LATE u In Lu 7 Mb ie u «| ty er Med Gee tu i i a i | 1 L Ci DI i 1e 1 1 t | 1 a ER i N war à a i HN dir i a LA ik Hr CH (ROM tome ‘ { Chi ‘ I I { | € | A bs dal of fi iù (ni # in} i A À i { un | | ol i a } hi At LEET i | | LAY x [ IER f b d TRISTI ® ni ì ii vi (E, : i Tas ir! hi Re 3 i Gat ia (| n i A si bet ! Ù Ù D. f 7 ARI | i i ra mia yn da P 4 i hy } x { f ri ey Sees rk i 3 Na Pal = u 6 ind Ban. | ‘a ; f | \ jh HU DO. 174 ms % motu Ma I (RTE i it, La 1 Y Ki a N ß me) iv i ti ‘ Mal IE Ce È 4 x | i ilo j + ANT EE tp \ al | re | À qu UN DT Won if en LARA, SAH ENGE DI 7 y LOT e } ine) ki fl v i Lu ait i ki; f Pelt 4 i : $ p 0 ili u Mi y wen la Kar am na jar Gr shy 0 MR KR | ver i? tetek, A 1 | né ’ iu ne te AM id ti Da a pil le if à 4 } fi rc Arns men | ju AN Mal Aa, EN | (mea tig iid 11 2h LI Mme dii iridata dl. ceil man a eri) ri à vo hik Ki MN shel te bi. Kae; ade 3 ha Î PR ars: i Lalygne lano AR le ET der tee NT RATE verbe tr, i I mn À Tei à i TAGE: fl J. KVERANDSEN Museum of Zoology, University of Bergen, Norway ie NEW SPECIES OF PSEUDEXECHIA TUOMIKOSKI FROM TANZANIA AND THAILAND (DiEERA: MYCE TOPHILIDAE) Kjerandsen, J., 1994. Three new species of Pseudexechia Tuomikoski from Tanzania and Thailand (Diptera: Mycetophilidae). — Tijdschrift voor Entomologie 137: 325-330, figs. 1-4. [1ssN 0040-7496]. Published 15 December 1994. Pseudexechia longistylus sp. n. and P. lanceostylus sp. n. from Tanzania and P. inthanonensis sp. n. from Thailand are described, based on adult males. The terminalia are figured. The presen- ce of a pair of medium sized bristle-like dorsocentral setae on scutum in the two African species and an aberrant wing venation in P. inthanonensis sp.n. is commented on. J. Kjærandsen, Museum of Zoology, University of Bergen, Muséplass 3, N-5007 Bergen, Norway. Key words. — Diptera; Mycetophilidae; Pseudexechia, Tanzania; Thailand; taxonomy. Pseudexechia Tuomikoski, 1966 constitutes one of 14 genera in the tribus Exechiini, subfamily Mycetophilinae (Tuomikoski 1966). The genus is characterised by the absence of large discal setae on the mesoscutum, by the ovate clypeus and on charac- ters in the male terminalia such as the bud-like stern- al process. Chandler (1978) treated all eight holarctic species known at that time. Three palaearctic species were added later; viz. P. trilobata Ostroverkhova, 1979, P. ussurensis Zaitzev, 1982 and P. altaica Zaitzev, 1988. Matile (1970) revised the Afrotropical Pseudexechia comprising four species. A fifth species, P. tanganyikae (Lindner, 1958) originally described as Exechia, was later referred to this genus as well (Matile 1980). No species of Pseudexechia are hither- to known from the oriental region. Thus, 16 species of the genus Pseudexechia were known: 9 palaearctic, 2 nearctic and 5 afrotropical. Of only one of the five Afrotropical species both sexes were described, the four other were all described on single females. Female holotypes make it difficult to describe new species from the region as certain association of the sexes is not easy to establish without large samples or reared specimens. However, based on differences in external characters as coloration and wing venation the present specimens can not belong to any of the described species. MATERIAL The material consists of two males collected in the West Usambara Mountains in Tanzania in 1990, and one male collected in the Doi Inthanon mountain in Thailand in 1991. The holotypes are deposited in the Museum of Zoology, Bergen (ZMBN). METHODS AND TERMINOLOGY The specimens were cleared and slide mounted in Canada balsam. The general terminology follows McAlpine (1981). Wing measurements (fig. 1): Total wing length is measured from the extreme base of the distal median plate. A = distance between the points where R, and R,,,; reach wing margin. B = distance between the points where R,, and M, reach wing margin. C = width of M-fork at wing margin. D = width of Cu- fork at wing margin. E = distance from extreme base of the distal median plate to base of M-fork. F = dis- tance from extreme base of the distal median plate to base of Cu-fork. Other measurements are self-expla- natory and indicated in Fig. 1. The tibial spur formula is the length of each spur in relation to the apical diameter of tibia in the follo- wing order: fore tibial spur; the two mid tibial spurs; the two hind tibial spurs. SYSTEMATIC PART Pseudexechia longistylus sp. n. (figs. 1 - 2) Type material. — Holotype male, TANZANIA: Tanga re- gion, W. Usambara Mts, Mazumbai, 1440 m asl, 22.XI.1990, G. E. E. Soli, sweep net (ZMBN No. 179). Diagnostic characters. — The male imago is charac- 325 Ty DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 = WING LENGTH Fig. 1. Pseudexechia longistylus sp. n., right wing showing measuring points. — Total wing length is measured from the extre- me base of the distal median plate to tip of wing. A = distance between the points where R, and R,.; reaches wing margin. B = distance between the points where R,,, and M, reaches wing margin. C = width of M-fork at wing margin. D = witdh of Cu-fork at wing margin. E = distance from extreme base of the distal median plate to point of M furcation. F = distance fr- om extreme base of the distal median plate to point of Cu furcation. M-p. = M-petiole. Fig. 2. Pseudexechia longi- ’ \ | | stylus sp. n., male termina- \ (\ lia. — A, ventral view; B, in- AN INI ternal face of the NN RUN) | gonostylus. Scale 0.50 mm. N NN ti zit SNS 4 — : SSS AL ee == NN DEN IE Gn 47 = RSS == a terised by the enlarged terminalia in which the dorsal lobe of gonostylus ends in a very elongated lobe with a setose cushion basally, and by the shape of the bi- furcate internal process. The ventral lobe of the gono- stylus also is enlarged compared to other species. Etymology. — From Latin, longus, long, and stylus, style, referring to the elongated gonostylus. A noun in apposition. Description of male holotype. — Length of thorax 0.86 mm. Wing length 2.62 mm. Wing length / length of thorax 3.1. Wing length / length of fore fe- mur 3.1. Coloration. Antenna with scape, pedicel and basal part of flagellum yellow, gradually becoming darker and more greyish dusted towards tip. Head yellow. Maxillary palp greyish yellow. Thorax mainly yellow; scutum with narrow pale yellow or somewhat silvery shining stripe along lateral margin, delimited by dar- ker margins below and above; scutellum darker yel- low. Wing unmarked, yellow tinted, paler towards tip and hind margin. Wing veins yellow with dark setae. Halter greyish yellow. Legs mainly yellow; coxae paler yellow, hind coxa posteriorly with darker stripe. Abdominal tergites mainly yellowish brown; tergite 1 and 2 with pale lateral markings; tergite 3 to 5 with pale posterolateral markings, almost reaching middle; tergite 6 almost entirely dark. Terminalia mainly yel- low, darker apically on gonocoxite, on sternal process and on parts of dorsal lobe of gonostylus. Head. Total length of flagellum 1.35 mm. First fla- gellomere 1.5 times as long as second flagellomere. Second flagellomere 2.2 times as long as wide. Five large orbital setae. Small round median ocellus pre- sent. Diameter of lateral ocellus / diameter of median ocellus 1.7. Clypeus bearing 31 setae; width / length 0.8. Fourth palpomere about 1.5 times as long as third palpomere. Thorax. Pronotum with vertical row of 3 erect setae. Scutum with pair of medium sized bristle-like dorsocentral setae posteromedially in front of scutel- lum (prescutellar setae); anterior and lateral margin with large setae; otherwise uniformly clothed with small setae (but larger and fewer than in other Pseudexechia except in P. lanceostylus sp. n.). Scutellum with pair of strong scutellar setae and pair of very short incurved setae basad of them. Proepisternum with 1 strong, 1 medium sized and 5 small setae. Laterotergite with 6-7 large and about 50 small setae. Wing (fig. 1). Crossvein h with 1 ventral seta. R, with 11 ventral setae apically. R,,, with 26-27 ventral setae apically. Wing length / length of R, 2.6. Wing length / length of R,., 1.9. R,,, nearly straight. Length of r-m / length of M-petiole 1.7. A/ B 1.4. E/ length KJAERANDSEN: New species of Pseudexechia of M, 0.7. E/ length of M,0.8. Base of Cu—fork well beyond base of M-fork. M-fork veins distaly diver- gent. Cu-fork veins divergent. F / E 1.3. F / length of CuA, 1.7. F / length of CuA, 2.5. C / D 1.4. Vein CuP reaching as far as base of Cu-fork. Vein A, weak, shorter. Legs. Length of fore basitarsus / length of fore tibia 1.3. Tibial spur formula 3.4; 8.8, 5.6; 4.6, 4.6. Fore tibia with 1 ad, 36 p and 3 v setae. Mid tibia with 40 a, 5 pd and 11 p setae. Hind tibia with 8 a, 5 pd and 3 p setae. Posterior sensillae placodea on basal part of tibia 3; 353. Terminalia (fig. 2A, B). Length of gonocoxite 0.60 mm. Sternal process narrow with about 40 tiny setae; length / width 1.9. Dorsal lobe of gonostylus 0.84 mm long with very elongated lobe, basally with strongly setose cushion. Sclerotized internal process bifurcate with both prongs widening apically. Ventral lobe of gonostylus large, bearing 6 fan-tipped setae basally. One lobe of the inner lamellate parts of the gonostylus apically with a slightly curved, spine-like seta thus forming a hook. Cerci thin and slender, re- aching to about apical level of gonocoxite, with two short but strong apical setae (not figured). Pseudexechia lanceostylus sp. n. (fig. 3) Type material. — Holotype male, TANZANIA: Tanga re- gion, W. Usambara Mts., Mazumbai, 1530 m asl, 02.-03.XI.1990, G. E. E. Soli, Malaise tent (ZMBN No. 180). Diagnostic characters. — The male imago is charac- terised by the shape of the dorsal lobe of gonostylus which ends in a slender lobe, and by the shape of the bifurcate internal process. Etymology. — From Latin, /ancea, a small light spe- ar, and stylus, style, referring to the slender lance-like dorsal lobe of gonostylus. Description of male holotype. — Length of thorax 0.98 mm. Wing length 3.02 mm. Wing length / length of thorax 3.1. Wing length / length of fore fe- mur 3.1. Coloration. Antenna with scape, pedicel and basal part of flagellum yellow, gradually becoming darker and more greyish dusted towards tip. Head and maxillary palp yellow. Thorax mainly yellow; scutum uniformly brownish yellow with pale yellow or so- mewhat silvery shining stripe along lateral margin, delimited by darker margins below and above; scutel- lum brownish yellow with faint paler median stripe. Wing unmarked, yellow tinted, slightly paler towards hind margin. Wing veins yellow with dark setae. Halter pale yellow with three darker yellow stripes. 92. TiJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Fig. 3. Pseudexechia lanceostylus sp. n. male terminalia. — A, ventral view; B, internal face of the gonostylus. Scale 0.50 mm. Legs mainly yellow; coxae paler yellow, fore coxa an- teriorly with darker stripe. Abdominal tergites mainly yellowish brown; tergite 1 to 3 with pale lateral mar- kings; tergite 4 and 5 with pale posterolateral mar- kings, almost reaching middle; tergite 6 almost enti- rely dark. Terminalia mainly yellow, darker on sternal process and on parts of dorsal lobe of gonostylus. Head. Total length of flagellum 1.53 mm. First fla- gellomere 1.7 times as long as second flagellomere. Second flagellomere 2.3 times as long as wide. Six lar- ge orbital setae. Tiny median ocellus present. Diameter of lateral ocellus / diameter of median ocel- lus 2.4. Clypeus bearing about 60 setae; width / length 0.8. Fourth palpomere about 1.5 times as long as third palpomere. Thorax. Pronotum with vertical row of 3 erect setae. Scutum with pair of medium sized bristle like dorsocentral setae present posteromedially in front of scutellum (prescutellar setae); anterior and lateral margins with large setae; otherwise uniformly clothed with small setae (but larger and fewer than in other 328 Pseudexechia except in P. longistylus sp. n.). Scutellum with pair of strong scutellar setae and 2 pairs of very short posteriorly curved setae basad of them. Proepisternum with 1 strong, 2 medium sized and 8 small setae. Laterotergite with 8-10 large and about 60 small setae. Wing. Crossvein h with 1-2 ventral setae. R, with 5-8 ventral setae apically. R,,, with 3-4 ventral setae basally and 36-39 ventral setae apically. Wing length / length of R, 2.6. Wing length / length of R,,, 1.8. R,.. nearly straight. Length of r-m / length of M-pe- tiole 1.6. A / B 1.5. E / length of M, 0.7. E / length of M,0.8. Base of Cu-fork well beyond base of M-fork. M-fork veins distaly divergent. Cu-fork veins diver- gent. F / E 1.3. F / length of CuA, 1.5. F / length of CuA, 2.3. C/D 1.1. Vein CuP reaching as far as ba- se of Cu-fork. Vein A, strong, shorter. Legs. Length of fore basitarsus / length of fore tibia 1.8. Tibial spur formula 3.7; 9.8, 7.5; 5.4, 5.3. Fore tibia with 1 ad, 39 p and 3 v setae. Mid tibia with 41 a, 5 pd and 7 p setae. Hind tibia with 9 a, 5 pd and 4 KJAERANDSEN: New species of Pseudexechia Fig. 4. Pseudexechia inthanonensis sp. n. male terminalia. — A, ventral view; B, internal face of the gonostylus. Scale 0.50 mm. p setae. Posterior sensillae placodea on basal part of ti- bia 3; 4; 3. Terminalia ( Fig. 3A, B). Length of gonocoxite 0.57 mm. Sternal process rounded with about 55 tiny setae; length / width 1.4. Dorsal lobe of gonostylus 0.51 mm, ending in slender lobe. Sclerotized internal process bifurcate, the the prongs pointing in almost opposite directions. Ventral lobe of gonostylus broad and rounded, bearing 6 fan-tipped setae. Inner lamel- late parts of gonostylus with lobe ending in strong seta—like hook. Cerci thin, not reaching apical level of gonocoxite, with two strong apical and one strong subapical setae (not figured). Pseudexechia inthanonensis sp. n. (fig. 4). Type material. — Holotype male, THAILAND: Chiang Mai province, Doi Inthanon, about 2200 m a.s.l., 12.1V.1991, J. Kjærandsen, sweep net (ZMBN No. 181). Diagnostic characters. — The male imago is separa- ble from other species by the shape of the dorsal lobe of gonostylus, which has a smooth rounded hook at the tip and a large internal process with 25 teeth, by the shape of the sternal process and by the short stem of the cubital fork. Etymology. — From Doi Inthanon, a mountain in northern Thailand. A noun in genitive case. Description of male holotype. — Length of thorax 1.20 mm. Wing length 3.82 mm. Wing length / length of thorax 3.2. Wing length / length of fore fe- mur 3.3. Coloration. Antenna with scape, pedicel and first flagellomere for half of its length pale yellow, rest of flagellum greyish yellow. Head mainly brown, frons and sides yellowish. Maxillary palp yellow. Pronotum yellowish brown. Scutum brown with three yellow stripes, median stripe broadening to fore margin, and narrow yellow lateral margins. Scutellum brown with yellowish lateral margins. Pleura yellowish brown. Wing unmarked, greyish yellow tinted, paler on basal half posterior of R-stem. Wing veins yellowish brown with dark setae. Halter pale yellow with three greyish yellow stripes. Legs mainly yellow; coxae paler yellow, each coxa with yellowish brown stripe. Abdominal tergites mainly brown; tergite 2 to 5 with pale poste- rolateral markings, reaching middle; tergite 6 dark brown. Terminalia mainly yellow, darker on sternal process and on parts of dorsal lobe of gonostylus. Tip of dorsal lobe of gonostylus whitish. Head. Total length of flagellum 1.84 mm. First fla- gellomere 1.7 times as long as second flagellomere. 329 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Second flagellomere 2.0 times as long as wide. Four to five large orbital setae. Small median ocellus pre- sent. Diameter of lateral ocellus / diameter of median ocellus 2.1. Clypeus bearing about 40 setae; width / length 0.8. Fourth palpomere about 1.4 times as long as third palpomere. Thorax. Pronotum with vertical row of 2 strong and 2 medium sized upcurved setae. Anterior and la- teral margin of scutum with setae of different sizes; scutum otherwise uniformly clothed with small dark setae. Scutellum with pair of strong scutellar setae and pair of short posteriorly curved setae basad of them. Proepisternum with 1 large, 2 medium sized and 24 small setae. Anepisternum with 6 small setae. Laterotergite with 12-14 large and about 200 tiny setae. Wing. Crossvein h without ventral setae. R, with 9 ventral setae apically. R,,, with 4-5 ventral setae basal- ly and 58-65 ventral setae apically. Wing length / length of R, 2.3. Wing length / length of R,,, 1.7. R,., distinctly downcurved. Length of r-m / length of M- petiole 1.0. A/B 1.9. E/ length of M, 0.7. E / length of M, 0.8. Base of Cu-fork not beyond base of M-fork. M-fork veins only slightly divergent apically. Cu-fork large, veins divergent. F / E 0.98. F / length of CuA, 1.2. F / length of CuA, 1.7. C / D 0.9. Vein CuP reaching well beyond base of Cu-fork. Vein A, strong, shorter. Legs. Length of fore basitarsus / length of fore tibia 1.2. Tibial spur formula 3.5; 6.7, 6.4; 4.6, 4.5. Fore tibia with 6 ad, 45p and 8 v setae. Mid tibia with 49 a, 5 pd and 4 p setae. Hind tibia with 6 a, 5 pd and 8 p setae. Posterior sensillae placodea on basal part of ti- bals, Terminalia ( Fig. 4A, B). Length of gonocoxite 0.67 mm. Sternal process narrow without setae; length / width 1.8. Dorsal lobe of gonostylus 0.67 mm long with smooth rounded hook at tip and rela- tivly large sclerotized internal process with 25 teeth. Ventral lobe of gonostylus bearing 4-5 fan-tipped ventral setae and one strong dorsal setae apically. Inner lamellate parts of gonostylus with 5 small setae. Cerci thin and slender, reaching to about apical level of gonocoxite, without strong setae (not figured). Discussion The presence of a pair of medium sized bristle-like dorsocentral setae posteromedially above scutellum in Pseudexechia longistylus sp. n. and P. lanceostylus sp. n. is not in accordance with the generic description. Tuomikoski (1966) used absence of ‘discal bristles’ as one of the main diagnostic characters for Pseudexechia. The other setae on the scutal disc in these two species are also relatively strong, erect and bristle-like when compared with other species in the 330 genus, and they are fewer in numbers. Hence, this character should be used with care. However, other generic characters clearly place these species within Pseudexechia. P. inthanonensis sp. n. is evidently closely related to the Palaearctic P. trisignata (Edwards, 1913) based on the structure of the male terminalia. However, P. in- thanonensis sp. n. is rather different in coloration (yel- low thoracic stripes on brown ground) and shows an aberrant and interesting wing venation. The short Cu-stem places the base of Cu-fork slightly before ba- se of M-fork (F / E = 0.98). A long Cu-stem (short Cu-fork) has been used to group the species of Exechia s.l. (including Exechiopsis and Pseudexechia ) (e.g. Edwards 1925). These findings strengthen the view of Tuomikoski (1966) that Pseudexechia is more closely related to Allodiopsis and Allodia than to Exechta and Exechiopsis. ACKNOWLEDGEMENTS I am indebted to G. E. E. Soli for giving me the op- portunity to examine the material collected during ZMBN's Tanzania expedition in 1990, and for critical- ly reading the manuscript. The Tanzania expedition was funded by the Norwegian Research Council (NAVE). REFERENCES Chandler, P. J., 1978. Notes on the Holarctic species of Pseudexechia Tuomikoski (Diptera: Mycetophilidae), with the description of a new British species. — Entomologist’s Record and Journal of Variation 90: 44-51. Edwards, F. W., 1925. British fungus-gnats (Diptera, Mycetophilidae). With a revised generic classification of the family. — Transactions of the Royal Entomological Society of London 1924: 505-670. Matile, L., 1970. Pseudexechia de la région éthiopienne (Dipt. Mycetophilidae). — Bulletin de la Société entomo- logique de France 75: 209-214. Matile, L., 1980. 15. Family Mycetophilidae. — In: Crosskey, R. W., Catalogue of the Diptera of the Afrotropical region. — British Museum (Natural History), London., pp. 216-230. McAlpine, J. F., 1981. Morphology and terminology - Adults. — In: J. F. McAlpine et al. (eds.), Manual of Nearctic Diptera. Volume 1. — Research Branch Agriculture Canada, Monograph no. 27. Ottawa, Ontario. Pp. 9-63. Tuomikoski, R. 1966. Generic taxonomy of the Exechiini (Dipt., Mycetophilidae). — Annales entomologici fennici 32(2): 159-194. Received: 24 March 1994 Accepted: 8 August 1994 JOHN T. POLHEMUS', NICO NIESER’ & STEVEN L. KEFFER’ 1 University of Colorado Museum, Englewood, Colorado, © Tiel, The Netherlands & James Madison University, Harrisonburg, Virginia SYNONYMICAL NOTES ON THE NEPA CINEREA BINNAEUS COMPLEX (NEPIDAE: HETEROPTERA) Polhemus, J. T., N. Nieser & S. L. Keffer, 1994. Synonymical notes on the Nepa cinerea Linnaeus complex (Nepidae: Heteroptera). — Tijdschrift voor Entomologie 137: 331-336. [1ssN 0040-7496]. Published 15 December 1994. The Nepa cinerea Linnaeus complex is distributed over Europe, North Africa, the Middle East and northern Asia. Within this complex we recognize three species, N. cinerea, N. sardiniensis Hungerford and N. anophthalma Décu, Gruia, Keffer & Sarbu. The following are considered to be junior synonyms of N. cinerea Linnaeus, 1758: Nepa rubra Linnaeus, 1758; Nepa scor- pioaquaticus De Geer, 1773; Nepa cinerea var. minor Puton, 1886, syn. n.; Nepa seurati Bergevin, 1926, syn. n.; Nepa cinerea var. major Bergevin, 1926, syn. n.; Nepa dollfusi Esaki, 1928, syn. n.; Nepa cinerea var. orientalis Esaki, 1928, syn. n.; Nepa rubra meridionalis Poisson, 1961, syn. n.; Nepa remyi Poisson, 1961, syn. n.; Nepa cinerea poissoni Tamanini, 1973. Correspondence: Dr. John T. Polhemus, University of Colorado Museum, 3115 S. York St., Englewood, Colorado 80110. Key words. — Heteroptera; Nepidae; Nepa cinerea; synonymy. The Nepa cinerea Linnaeus complex is a tightly knit and very widespread species group in the Old World, distributed throughout Europe, the Mediterranean region, and as far east as northwest China and the Amur region of Siberia. Nepa cinerea was one of the earliest water bugs to be studied (Hoefnagel 1592, Moufet 1634, Frisch 1728, Swammerdam 1737-38), and the morphology has been extensively investigated, with an excellent treat- ment and summary of earlier studies by Hamilton (1931). Despite of this wealth of knowledge, the spe- cies group nomenclature concerning this insect re- mains unsettled. A definitive synonymical treatment is needed for the Catalogue of Palearctic Heteroptera (B. Aukema & C. Rieger, editors) now being prepa- red for publication. The early nomenclatural dilemma regarding the priority and synonymy of Nepa cinerea versus Nepa rubra Linnaeus, 1758, first addressed in detail by Esaki (1926), then by Tamanini (1973), was finally resolved by the ICZN (Opinion 1335/1985) in re- sponse to a petition by Kerzhner (1981). Tamanini (1973) has declared the type locality to be Sweden, because the first specimens were collected there; he ci- tes inclusion in Linnaeus’ Fauna Svecica (1761) as evidence for this. Over the years however, a number ofadditional taxa have been proposed, as species, sub- species and varietal forms, that are either synonymous or very closely allied with N. cinerea. Esaki (1928), Poisson (1961) and Tamanini (1973) addressed this species-group problem, however, far from resolving the status of these species-group taxa, each of these authors exacerbated the situation by adding at least one additional variety. The following is a brief summary of the taxa and their supposed differences from Nepa cinerea sensu stricto, with comments on the validity of the charac- ters used for differentiation. Nepa cinerea var. minor Puton, 1886 and Nepa ci- nerea var. major Bergevin, 1926 were established sole- ly on the basis of size differences, which is not a diffe- rentiating characteristic, as it varies widely within single populations. Jaczewski (1934) examined speci- mens of N. cinerea minor from Algeria, and stated that ‘they seem to be conspecific with the typical N. cinerea L. We therefore synonymize both varieties. Bergevin (1926) established Nepa seurati n. sp. fr- om Tunisia on the basis of a number of somatic diffe- rences, which he compared, in a table, with Nepa ci- nerea vat. minor Puton. Esaki (1928) provided a photograph of the type of seurati which does indeed have a slightly different shape than the cinerea speci- “mens he illustrates, with the lateral abdominal mar- gins straight along much of the basal part, widening posteriorly, and abruptly incurved beyond, as noted by Poisson (1961: 631). Also the fore femur basally is wider than the N. cinerea specimens illustrated. This taxon would therefore seem to be separable from Nepa cinerea cinerea, however in long series taken fr- om single populations in Morocco (nr. Tangier bor- 331 Ty DSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 der, 8.VIII.1955, D. R. Lauck, Polhemus Coll.) and Portugal (nr. Portinas, Riba la Banho, 15.1V.1976, N. Nieser, Nieser and Polhemus Collns.) all of these character states are present along with intermediates. If considered as a valid subspecies, then many of the populations from north Africa would probably be at- tributable to this taxon. Esaki (1928) described Nepa dollfusi from a single female collected in Morocco. Some of the differences from JV. cinerea he cited may be attributable to alary polymorphism (see Larsén 1949, 1955), as the flight- less forms, such as dollfusi, have an altered morpholo- gy, e. g. straighter hind margin of the pronotum, and reduced hemely tral membrane. Esaki provided a pho- tograph of the type of dollfusi which appears to repre- sent a malformed specimen, quite possibly resulting from damage during emergence from the last nymp- hal stage. The specimen also has reduced antennae, discussed below, that may have also resulted from a traumatic ecdysis. The red abdominal dorsum noted for European specimens of N. cinerea in contrast to dollfusi is unreliable, as it varies from grey brown to yellowish to pink or red in individual series from Europe and the Mediterranean region. Theiry (1981) noted the great variation in size and coloration (of both ground colour and abdominal tergites) in the 100+ specimens of Nepa collected over a wide range of elevation in Morocco, all of which he assigned to N. cinerea, noting that the parameres and antennae of all forms were of the ‘type cinerea.’ Lindberg (1929) considered N. dollfusi to be only an aberrant speci- men of N. cinerea. Nepa cinerea var. orientalis Esaki, 1928 was foun- ded on the basis of size and slight differences in an- tennal and head morphology, all unreliable. We have not been able to study a long series from the Far East to determine if there is any discernable difference in the male genitalia from the western European stock, but for the present the available evidence leads us to synonymize this form with N. cinerea. Hungerford (1928) described Nepa sardiniensis from the island of that name. He separated it from re- lated taxa by the lack of a prolongation on the second antennal segment and slender fore femur. These alo- ne are questionably enough to justify the status of a separate species for the populations of Sardinia and Corsica, however Mazza (1971, 1978) has given addi- tional evidence for separate species status (see below). Nepa sardiniensis is the only geographically isolated taxon in the N. cinerea complex, aside from the uni- que spatially isolated cavernicolous N. anophthalma Décu et al (in press). Linnavuori (1960) reported this species from Israel, Transcaspica and Turkestan, but according to Tamanini (1973) these records refer in- stead to N. cinerea cinerea. For the Palearctic region, Stichel (1955) listed as 392 species Nepa seurati Bergevin, Nepa dollfusi Esaki, Nepa sardiniensis Hungerford, and N. cinerea (as Nepa rubra), with minor Puton, major Bergevin and orientalis Esaki clearly designated as forms of the lat- ter. Linnavuori (1960) claimed separation of N. cinerea and N. sardiniensis on the basis of the short second antennal segment, broader sternite VII, and differen- tly formed male clasper of the latter. He stated that N. cinerea does not occur in Palestine. He also stated that the anterior femora of Nepa seurati are much broader basally than in N. sardiniensis. Poisson (1961) gave figures of the antennae, fema- le subgenital plate and male genitalia of N. dollfusi Esaki, N. cinerea cinerea (as N. rubra rubra), N. cine- rea meridionalis (as N. rubra meridionalis), and Nepa remyi Poisson, 1961 from Morocco. N. remyi was se- parated on the basis of slight differences in the shapes of the male subgenital plate (‘opercule génital”), male paramere, and the base of the fore femur, all of which have been seen to be somewhat variable within popu- lations. We consider this variety to fall within the va- riability of N. cinerea, thus a synonym. Poisson also gave a key to separate all of the forms and species known in 1961 (including Nepa apiculata Uhler from North America and Nepa hoffmanni Esaki from China), based on differences in the antennae, the re- spiratory siphon, shape of the anterior femur, overall body length. Seidenstücker (1963) investigated the status of N. dollfusi, N. sardiniensis and N. seurati, relying mainly on the antennal structure to conclude that Nepa seur- ati is a good species, and that Nepa sardiniensis Hungerford, 1928 (June) is a junior synonym of Nepa dollfusi Esaki, 1928 (April), and occurs in Turkey as well as Corsica and Sardinia. If Seidenstiicker’s analysis were accepted, the N. cinerea complex would include N. anophthalma, N. cinerea, N. dollfusi (= N. sardiniensis), and N. seurati. Later authors have not accepted the synonymy of N. sardi- niensis (see below). Mazza (1968) studied the variability of the number of respiratory horns of 5081 eggs of N. cinerea from Italy. He noted that the modal number of horns do- cumented by Hinton (1961) for a British population (7) was different than the modal number (6) for the Italian population from La Spezia Province. Later he analyzed many specimens and concluded that the ge- ographical forms were impossible to delimit, and that they form a continuum (Mazza 1974). This evidence suggests clines within N. cinerea. Mazza (1971) noted constant differences in the number of respiratory horns (without giving details) between N. sardiniensis collected on Corsica and N. cinerea collected in Toscana (Italy), and further stated that these two populations were reproductively isola- ted, as cross breeding produced sterile Fl hybrids. Later he (Mazza, 1978) examined about 1000 speci- mens of each species, and using several somatic char- acters in first order equations he showed consistent separation, therefore N. sardiniensisand N. cinerea are presently accepted as distinct species. Tamanini (1973) discussed all of the previously mentioned forms except Nepa remyi Poisson, 1961, and proposed yet another subspecies, Nepa cinerea poissoni Tamanini from the Pyrenees Mountains of France and Spain, based on differences in pronotal morphology and antennae. He compared his new taxon to N. sardiniensis and N. cinerea cinerea. The differences he cites are not sufficient to separate this taxon from the latter, and it was synonymized with the latter by Nieser & Montes (1984). Tamanini claims that N. sardiniensis is endemic to Corsica and Sardinia, and states that N. cinerea dollfusi and N. ci- nerea meridionalis are of ‘uncertain systematic rank.’ For all of Europe and North Africa, Nieser (1978) lis- ted only Nepa sardiniensis Hungerford and Nepa cine- rea, and recognized as subspecies of the latter N. c. poissoni Tamanini and N. c. seurati Bergevin. Later Nieser & Montes (1984) synonymized N. c. poissoni with N. cinerea, without comment. Nepa anophthalma Décu, Gruia, Keffer & Sarbu (in press) is separable from all other species of the ci- nerea complex by five characters; 1. A yellow ground color (vs. dark), 2. Eyes absent (vs. present), 3. Hemelytra brachypterous, posterior margin sinuate (vs. macropterous, posterior margin rounded ), 4. Midlongitudinal groove of mesosternum absent (vs. present), 5. Paraterga hirsute (vs. glabrous except for hairs ringing stigmata). This is clearly the most an- nectant species of the complex, yet the male genitalia are very similar to the other members of the group. It is endemic to the Movile cave system in Romania. As discussed above, most of the characters used by various authors to separate species-group taxa in the Nepa cinerea complex are as variable within popula- tions as they are between populations, thus must be considered unreliable for the separation of taxa, espe- cially when dealing with single specimens or small se- ries. The antennae have been extensively cited as a se- parating characteristic notwithstanding their variability. Tamanini (1973) has shown that the an- tennal morphology is quite variable, even in a single specimen, thus the antennal morphology is not a completely reliable specific character in this genus, although their form may indicate trends, certainly va- ry between some populations, and seem to have a norm within certain populations (vide Seidenstiicker, 1963; Tamanini, 1973). Esaki (1928) illustrated the antennae of a topotypic specimen of N. seurati, the type of N. dollfusi, and of N. cinerea (no provenance given). In dollfusi the prolongation of the second seg- POLHEMUS ET AL: Synonymy of Nepa cinerea ment is lacking and segments two and three are fused and short; in this regard it is very similar to N. sardi- niensis, in fact so similar that Seidenstiicker (1963) proposed the synonymy of the two taxa. In seurati the prolongation of the second segment is moderately long, and segments two and three are separated and moderately long, very similar to cinerea except in the latter the prolongation of the second segment is slightly longer. Tamanini (1973) studied many speci- mens from various localities, and illustrated the an- tennal forms most frequently encountered as well as aberrant forms rarely seen. The range of morphology within the species essentially covers the spectrum of differences used to separate the three species group taxa discussed above, although the ‘normal’ form of each conforms roughly to those illustrated by Esaki (1928). Mancini (1936), who tentatively identified specimens from Libya as N. cinerea minor, questioned whether the shape of the antennal segments was a va- lid character. One of us (SLK) has studied the male genitalic structure of several populations of the Nepa cinerea complex from Finland, Morocco, Sardinia, Romania, and Iran and can find only one significant genitalic difference in any of the taxa (see below), although the genitalia of the cinerea complex are considerably dif- ferent than the genitalia of the apiculata group which contains Nepa apiculata and Nepa hoffmanni (Keffer et al. 1990). However, this general lack of male geni- talic difference should not be taken as evidence that species or subspecies designations within the cinerea complex are invalid. Keffer (1991 and personal obser- vation) has found that male genitalia are often of lit- tle taxonomic value at the species level in the Nepidae. For example, in his revision of the New World waterscorpion genus Curicta, Keffer (1991) found male genitalia to be diagnostic for the genus but largely invariant throughout the range of the ge- nus. Similarly, Keffer (in prep.) has studied a majori- ty of the species across the family Nepidae and found that male genitalic characters often have generic or species group significance but are often of little value for species determination. One genitalic character, paramere shape, has been used extensively by two nepid taxonomists of the recent past, Jose De Carlo and Raymond Poisson, in their numerous new species descriptions. Keffer, however, has found that parame- res may vary either little or else randomly across spe- cies groups, or even genera, and thus are not general- ly valid as a species specific character. Keffer’s observations with Curicta and other nepid genera are affirmed in the genus Nepa. Male genitalia do separate the cinerea and apiculata species groups (Keffer et al. 1990). However, within the cinerea complex phallic structures, with one exception, are either invariant, or they vary randomly, across all 333 TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 taxa. For example, parameres appear to vary random- ly and are therefore of no value in making species and subspecies determinations. Taminini (1973, figs. 61- 78) figured parameres for N. cinerea cinerea, N. cine- rea poissoni, N. cinerea seurati, and N. sardiniensis. We have observed N. cinerea cinerea specimens from Finland with paramere shapes similar to those figured by Taminini for N. cinerea poissoni and we have ob- served N. sardiniensis parameres that are similar to those figured by Taminini for N. cinerea cinerea. Further, Decu et al. (in press) have shown that N. anophthalma has parameres similar to those of N. ci- nerea poissont. Only one genitalic character, the shape of the se- condary struts, appears to be of value for separating some cinerea group taxa. During copulation, the ma- le’s secondary struts and ventral diverticulum lodge in the female bursa copulatrix (Larsen 1938, fig. 26; se- condaryasistrutsi or an Griffel bursans niv: Vaginaltasche’; Keffer 1991, p.3) where they may ha- ve a holdfast and/or stimulatory function. If, in fact, the struts are stimulating the female during copula- tion, and thus subject to sexual selection, they could exhibit species specificity. Taminini (1973: 233) no- ted that, in ventral view, the secondary struts (‘pro- cessi a spatola della coniunctiva’) of N. cinerea cinerea specimens from Sicily and continental Italy converge toward the midline distally, whereas N. sardiniensis secondary struts are nearly straight. We would add to Taminini’s description two of our own observations. First, converging secondary struts, as in cinerea cine- rea, are also found in the cave dwelling cinerea group taxon, N. anophthalma and in apiculata group taxa. Second, in both N. cinerea cinerea and N. sardiniensis the secondary struts are dorsoventrally flattened whe- reas in N. anophthalma and in the apiculata group taxa the struts are tubular in shape. In sum, it appears that secondary struts can be used to separate N. sard- iniensis (secondary struts nearly straight, not conver- ging distally) and N. anophthalma (secondary struts tubular, not dorsoventrally flattened) from other ci- nerea group taxa. The status of four species group taxa must be deci- ded; they are N. cinerea, N. dollfusi, N. sardiniensis and N. seurati. The first question to be settled is the proposed synonymy of dollfusi and N. sardiniensis, by Seidenstücker (1963). This synonymy seems to rest on the strength of the similarity of the antennae of the two taxa. If this synonymy is accepted, and the modal antennal form is accepted as a valid discriminating character, then the range of N. dollfusi, the senior sy- nonym, is Morocco, Tunisia, Corsica and Sardinia. Tamanini (1973) rejected this synonymy, however, and also rejected the citation of N. sardiniensis for Israel by Linnavuori (1960) and for Turkey, Transcaspica and Turkestan by various authors, as be- 334 ing narrowly based on antennal form only. In this contention he is supported by Mazza (1971, 1978), thus current opinion supports the recognition of N. sardiniensis as a valid species endemic to Corsica and Sardinia, with N. dollfusi relegated to synonymy un- der N. cinerea. Tamanini (1973) contends that N. cinerea cinerea occurs only as far south as northern Italy, and that N. cinerea seurati is distributed from southern Italy through Sicily to Libyia and Tunisia (we would pos- sibly add Morocco in part). He shows a gradation of characters on a latitudinal cline through Italy, which suggests a single species rather than subspecies. If the subspecies status of seurati were to be accepted, then dollfusi should be considered as a synonym with aber- rent antennae. This would leave N. cinerea cinerea as a variable subspecies with a range throughout nort- hern Europe and as far south as Morocco nearest Spain, across the Balkans, through the Middle East, and as far cast as Siberia. We here reject the subspecies concept for seurati, however, as based on insufficient evidence and unre- liable characters, and synonymize it under cinerea. Mayr & Ashlock (1991:43) give the following de- finition of the subspecies: ‘A subspecies is an aggrega- te of phenotypically similar populations of a species inhabiting a geographic subdivision of the range of that species and differing taxonomically from other populations of that species’. We accept this defini- tion, and all of the species group taxa of the N. cine- rea complex that fail to conform to this definition are placed in synonymy of N. cinerea below. We have seen no convincing evidence that the characters upon which these taxa were founded are anything more than individual variations within and between popu- lations, scattered more or less randomly, and without significant separation as required for subspecies char- acterization. It is indeed possible that sufficient evi- dence may be presented in the future to reestablish one or more of the synonymized taxa as valid subspe- cies, but for the present such clear evidence is lacking. We therefore recognize only three species, without subspecies, in the cinerea complex; Nepa anophthalma (endemic to the Movile Cave system of Romania), N. sardiniensis (endemic to Corsica and Sardinia), and the widespread N. cinerea.The results of our investi- gations are summarized as follows: Nepa anophthalma Décu, Gruia, Keffer & Sarbu, in press. Nepa anophthalma Decu, Gruia, Keffer & Sarbu, in press. Holotype: d, Movile Cave, Romania. (ERSI) [examined]. Distribution. — Europe, Romania, nr. Mangalia, Movile Cave. Endemic. Nepa cinerea Linnaeus, 1758 Nepa cinerea Linnaeus 1758: 440. Syntypes, d, ®, Sweden (type locality restricted to Sweden by Tamanini 1973: 226). (uzıu) [not examined]. Nepa rubra Linnaeus 1758: 440. Holotype, 2, locality unk- nown. (uziu) [not examined]. (syn. of Nepa cinerea Linnaeus 1758; see Opinion 1335/1985). Nepa scorpioaquaticus De Geer 1773: 361. Syntypes, 5, sex unknown, type locality unknown. (coll De Geer, NHRS) [not examined]. Unneccessary new name for N. cinereaL. (see Retzius 1783: 90). Nepa cinerarea P. L. S. Müller 1774: 472. Incorrect subse- quent spelling. Nepa cinerea var. minor Puton 1886: 8. Syntypes, sex unk- nown, Tunisia, Oued Tessa, ‘Sidi-Mohamed-ben-Ali’, Oued Bateha, Oued Eddedj, Gafsa, Gabés. (Depository unknown) [not examined] (syn. Jaczewski 1934, suspec- ted). Syn. n.. Nepa seurati Bergevin 1926: 290. Holotype, d, Tunisia, Kebili. (MNHN) [not examined]. Syn. n. Nepa cinerea var. major Bergevin 1926: 294. Holotype, à, Morocco, Taza. (Depository unknown; coll. Bergevin; MNHN?) [not examined]. Syn. n. Nepa dollfusi Esaki 1928: 434 (April). Holotype, 9, Morocco, Oued Djenanimes pres Khénifra. (MNHN) [not examined]. (syn. Lindberg 1929: 9; restored by Seidenstücker 1963: 315, and named senior synonym of Nepa sardiniensis, synonymy with sardiniensis rejected by inference by Nieser 1978: 283). Syn. n. Nepa cinerea var. orientalis Esaki 1928: 436. Syntypes, 39, Russia, Amur (BMNH) [not examined]. Syn. n. Nepa rubra meridionalis Poisson 1961: 635. Syntypes, d, 9, France, Tunisia, ‘Iberian Peninsula’. (coll. Poisson, USNM) [examined]. Syn. n. Nepa remyi Poisson 1961: 636. Syntypes, 16, 29, Morocco, region d’Ifrane, Tagbaloute. (coll. Poisson, USNM) [examined]. Syn. n. Nepa cinerea poissoni Tamanini 1973: 239. Holotype, à, France, Pyrenées-Orientales, Collioure. (LTRC) [not exa- mined]. (syn. Nieser & Montes 1984: 38). Distribution. — Europe, North Africa, Middle East, northern Asia. Note. — Nepa annulipes (non Laporte 1833): Kolenati 1857: 481 is a misidentification referring to a Laccotrephes species, not collected in the Caucasus; see Kiritshenko 1918: 171; see also Esaki 1928: 434. Nepa sardiniensis Hungerford, 1928 Nepa sardiniensis Hungerford, 1928: 120 (June). Syntypes, 14,19, Italy, Sardinia. (semc) [examined]. (syn. with dollfusi by Seidenstücker 1963: 322; considered as valid species by Mazza 1971: 539, 1978: 85, Tamanini 1973: 241). Distribution. — Europe (Corsica, Sardinia) Depositories Abbreviations (codens) for depositories follow Arnett, Samuelson & Nishida (1993), with the addi- tion of: ERSI (Emil Racovità Speleological Institute, POLHEMUS ET AL: Synonymy of Nepa cinerea Bucharest); ırrc (Livio Tamanini Rovereto, Italy). Collection, ACKNOWLEDGMENTS We thank the following for helpful comments: I. Lansbury, Oxford; E. Kanyukova, Vladivostok; I. M. Kerzhner, St. Petersburg. We are grateful to F. Faraci, Bardolino and A. Carapezza, Palermo, Italy, for assi- stance with literature and specimens. REFERENCES Arnett, R. H., Jr., G. A. Samuelson & G. M. Nishida, 1993. The insect and spider collections of the world. — Sandhill Crane Press, Gainesville, vi + 308 pp. Bergevin, E. de, 1926. Description d’une nouvelle espèce de Nepa (Hémiptère Nepidae) de sud Tunisien. — Bulletin de la Société Histoire Naturelle de Afrique Nord 17: 290- 294, 1 fig. Decu, V., M. Gruia, S. L. Keffer & S. M. Sarbu, #7 press. A stygiobiotic waterscorpion, Nepa anophthalma n. sp. (Heteroptera: Nepidae), from a sulfurous cave in Romania. — Annals of the Entomological Society of America. De Geer, C., 1773. Memoirs pour servir a l’histoire des Insectes. — Hosselberg, Stockholm, Vol. 3, 2 + 696 pp, 44 Is. Esaki, T., 1926. Remarks on the Linnaean species of Nepa and Laccotrephes (Heteroptera:*Nepidae). — Bulletin of the Brooklyn Entomological Society (N. S.) 21: 177-181. Esaki, T., 1928. Contribution to the knowledge of the ge- nus Nepa (Hemiptera; Nepidae). — Annals and Magazine of Natural History (10) 1: 434-441, 1 pl. Frisch, J. L., 1728. Beschreibung von allerley Insecten in Teutschland, nebst nützlichen Anmerkungen und nöthi- gen Abbildungen von diesem kriechenden un fliegenden inländischen Gewürme, zur Bestätigung und Fortsetzung der gründlichen Entdeckung, so einige von der Natur dieser Creaturen herausgegeben, und zur Ergänzung und Verbesserung der andern. — Nicolai, Berlin, Teil 7, pp. 8 + 31, figs. I-XXII. Hamilton, M. A., 1931. Morphology of the water scorpion Nepa cinerea Linn. (Rhynchota, Heteroptera). — Proceedings of the Linnaean Society of London 1931: 1067-1136. 6 pls., 22 figs. Hinton, H. E., 1961. The structure and function of the egg- shell in the Nepidae (Hemiptera). — Journal of Insect Physiology 7: 224-257. Hoefnagel, G., 1592. Archetypa Studiaque Patris Georgii Hoefnagelii. — Jacobus F. Francoforti ad Moenum, 4 pts., each with 12 pls. and title page, 52 pp. Hungerford, H.B., 1928. A new Nepa (Hemiptera - Nepidae). — Bulletin of the Brooklyn Entomological Society (N. S.) 23: 119-123, 1 pl. Jaczewski, T., 1934. Notes on some palaearctic aquatic and semiaquatic Heteroptera, chiefly from south-eastern Europe. — Annales Musei Zoologici Polonici 10: 267- 288. Keffer, S. L., 1991. Taxonomic revision of the Neotropical genus Curicta Stal (Insecta: Heteroptera: Nepidae). — PhD Dissertation, Southern Illinois University, Carbondale, xvii + 264 pp. 35 Oo TIJDSCHRIFT VOOR ENTOMOLOGIE, VOLUME 137, 1994 Keffer, S. L., J. E. McPherson & I. T. Polhemus, 1990. What is Nepa hoffmanni (Heteroptera: Nepidae)? Male genitalia hold the answer, and delimit species groups. — Journal of the New York Entomological Society 98 (2): 154-162. Kerzhner, I. M., 1981. Nepa cinerea Linnaeus, 1758 (Insecta, Heteroptera, Nepidae): proposed conservation under the plenary powers. Z. N. (S.) 2144. — Bulletin of Zoological Nomenclature 38: 138-141. Kiritshenko, A. N., 1918. Hemiptera-Heteroptera of the fauna of the Caucasian Region. I. (Hemipt. Het. Faunae Caucasicae). — Mémoires du Musée du Caucase, Tiflis (A) 6: 1-177. (in Russian) Kolenati, F. A., 1857. Meletemata Entomologica. Fasc. VI. Hemipterorum Heteropterorum Caucasi. — Bulletin de la Société Impériale des Naturalistes de Moscou 29: 419- 491, 1856. Larsen, O., 1938. Untersuchungen über den Geschlechtsapparat der Aquatilen Wanzen. — Opuscula Entomologica, Lund, Suppl. 1, 338 pp., 151 figs. Larsen, O., 1949. Die Ortsbewegungen von Ranatra linearis L. Ein Beitrag zur vergleichende Physiologie der Lokomotionsorgane der Inseckten. — Lunds Universitets Arsskrift N. F. Avd. 2, 45: 1-82. Larsen, O., 1955. Der Bau des Flugapparats bei Nepa cine- rea L. Ein Vergleich zwischen der flugfähigen Imago und der gewöhnlichen Form mit reduzierten Flugorganen. — Opuscula Entomologica 20: 170-173. Lindberg, H., 1929. Inventa entomologica itineris Hispanici et Maroccani, quod a. 1926 fecerunt Harald et Hakan Lindberg. I. Hemiptera Heteroptera Hydrobiotica. = Commentationes Biologicae. Helsingfors 3: 1-12, 1 pl. Linnaeus, C., 1758. Systema naturae per regna tria naturae, secundum classes, ordines, genera, species, cum characte- ribus, differentiis, synonymis, locis. Editio decima, refor- mata. — L. Salvii, Holmiae, v + 824 pp. Linnaeus, C., 1761. Fauna Svecica sistens animalia sveciae regni: Mammalia, Aves, Amphibia, Pisces, Insecta, Vermes, distributa per classes & ordines, genera & spe- cies. — Laurentii Salvii, Stockholmiae, frontis. + xlvi + 578 pp-, 2 pls. Linnavuori, R., 1960. Hemiptera of Israel I. — Annales Zoologici Societas Zoologicae-Botanicae Fennicae, Vanamo, Helsinki 22: 1-71. Mancini, C., 1936. Contributo alla conoscenza degli emit- teri della Tripolotania e della Cirenaica. — Annali del Museo Civico di Storia Naturale di Genova 59: 195-201. Mayr, E. & P. D. Ashlock, 1991. Principles of sytematic .zoology. Second Edition. — McGraw Hill, New York, xx + 475 pp- Mazza, M., 1968. Osservazioni sulla variabilita del numero dei prolungamenti respiratori nelle uova di Nepa rubra (Rincote Eterottero). — Bollettino di Zoologia, Roma 35: 448-449. Mazza, M., 1971. Ricerche su Nepa sardiniensis. — Bollettino di Zoologia, Roma 38: 359. Mazza, M., 1974. Variabilità ed anomalie negli scorpioni d’acqua euromediterranei (Heteroptera: Nepidae). — Atti della Società Toscana di Scienze Naturali, Pisa, Memorie (Serie B) 81: 209-247, 1 pl.,10 tables. Mazza, M., 1978. Caratteristiche discriminanti in scorpioni 336 d’acqua euromediterranei (Heteroptera: Nepidae). — Atti XI Congresso nazionale italiano Entomologica 1976: 85- 89. Moufet, T., 1634. Insectorum sive minimorum animalium theatrum. — T. Cotes, London, 9 + 326 + 2 pp. Müller, P. L. S., 1774. Des Ritters Carl von Linné Königlich Schwedischen Leibarztes k. k. vollständiges Natursystem nach der zwölften latineischen Ausgabe und nach Anleitung des Holländischen Houttuynischen Werks, mit einer ausführlichen Erklärung. — Gabriel Nicolaus Kaspe, Nürnberg, Vol. 5, part 1, 4 + 8 + 758 pp. Nieser, N., 1978. Heteroptera. pp. 280-285. In: Illies, J., Ed., Limnofauna Europaea. — G. Fischer Verlag, Stuttgart, New York; Swets & Zeitlinger B. V., Amsterdam. Nieser, N. & C. Montes, 1984. Listas faunistica y bibliogra- fica de los Heteropteros Aquaticos (Nepomorpha & Gerromorpha) de Espafia y Portugal. — Lista de la Flora y Fauna las aguas continentales de la Peninsula Ibérica, Publ. No. 1, 69 pp. Opinion 1335/1985: Nepa cinerea Linnaeus, 1758 (Insecta, Heteroptera): Conserved. — Bulletin of Zoological Nomenclature 42: 241-243. Poisson, R., 1961. A propos d’une nouvelle espece Palearctique du genre Nepa L. 1758 (Heteroptera, Nepidae). — Vie et Milieu 11: 628-640, 1960. Puton, A., 1886. Enumération des Hémiptères recueillis en Tunisie in 1883 et 1884. In: MM. Valery Mayet et Maurice Sedillot, Exploration Scientifique de la Tunisie. — Paris, ii + 24 pp. Retzius, A. J., 1783. Caroli Lib. Bar. De Geer, genera et spe- cies insectorum e generosissimi auctoris scriptis extaxit, digessit, latine quoad partem reddidit, et terminologiam insectorum Linneanam addidit. — Siegfried Lebrecht Crusium, Lipsiae, 220 pp. Seidenstücker, G., 1963. Zur Aufklärung von Nepa dollfusi (Heteroptera). Reichenbachia 1: 315-322. Stichel, W., 1955-56. Illustrierte Bestimmungstabellen der Wanzen. II. Europa. (Hemiptera - Heteroptera Europae). — Selbstverlag, Berlin-Hermsdorf, Vol. 1, 168 pp. (pp. 1- 32: 14.11.1955; 33-64: 25.vi.1955; 65-96: 1.viii.1955; 97-128: 1.x1.1955; 129-160: 15.x11.1955; 161-168: 1.11.1956.) Swammerdam, J., 1737-1738. Biblia naturae; sive Historia Insectorum, in classes certas redacta, nec non exemplis, et anatomico variorum animaculorum examine, aeneisque tabulis illustrata. — Severin etc., Leydae, 2 vols, 910 + 56 + 36 + 124 pp., 53 pls. Tamanini, L., 1973. Priorità e sinonimia di Nepa cinerea Linneo e Nepa rubra Linneo. Regione tipica e valore del- le razze europee di Nepa cinerea Linneo, 1758 (Hemiptera Heteroptera, Nepidae). — Studi Trentini di Scienze Naturali (B) 50: 222-259. Thiery, A., 1981. Contribution a la connaissance des Hétéroptères du Maroc: Les Hétéroptères aquatiques du Haut Atlas Occidental. — Bulletin de Institut Scientifique, Rabat 1981: 13-34. Received: 15 July 1994 Accepted: 10 August 1994 OS di i | } PN DI Is} Et u MA oa In ra we) j ] 4 N va = are Ny i Di ARI LE Lo. rt un i fa op tf fet Asp ried a On asthme TRE tai} did N i ' F yl R DI O pic i LI tt a 1 î { if N ; i ‘ my Ù UO Ud 4 hh "fi no i ie x i \ i { | Isa FU i "i i i x Mi Tijdschrift voor Entomologie Volume 137, no. 2 Articles 143 M. Asche & M. D. Webb Review of the southern Palaearctic and palaeotropical leafhopper genus Hengchunia Vilbaste (Homoptera, Cicadellidae). 155 D. B. Baker A new genus of nomadine bees from North Africa (Hymenoptera: Apoidea, Anthophoridae). 161 A. J. de Boer Four species added to the Baeturia nasuta group, with notes on taxonomy and biogeography (Homoptera, Tibicinidae). 173 R. W. Garrison A revision of the New World genus Erpetogomphus Hagen in Selys (Odonata: Gomphidae). 271 H. de Jong The phylogeny of the subgenus Tipula (Savtshenkia) (Diptera: Tipulidae), with special reference to the western Mediterranean fauna. 325 J. Kjærandsen Three new species of PseudexechiaTuomikoski from Tanzania and Thailand (Diptera: Mycetophilidae). 331 J. T. Polhemus, N. Nieser & S. L. Keffer Synonymical notes on the Nepa cinerea Linnaeus complex (Nepidae: Heteroptera). © Nederlandse Entomologische Vereniging, Amsterdam Published 15 December 1994 ISSN 0040-7496 PET 4 2 z 38 ontents on inside back cover |