Bulletin of the British Museum (Natural History). British Museum (Natural History) Converted as part of the ABLE project by Dauvit King London : BM(NH) Continued as: Bulletin of the Natural History Museum. Entomology series Vol.1 (1950) - vol.61 (1992) 2 4 This document has been converted to TEI XML as part of the ABLE project to make it more widely available to biodiversity researchers in a useful format. eng text No corrections have been made in the text. The original source has not been regularized or normalized. Quotation marks have not been processed. They are as in the original DjVu XML document. Hyphens, including end-of-line hyphens, have not been processed. They are as in the original DjVu XML document. The text has been segmented based purely on layout based on page breaks. No language level segmetation, such as sentences, tone-units or graphemic, has been applied. Additional mark up using taXMLit has been applied to the TEI XML based on analysis of the original source through the uBio and OpenCalais web services. (Add comment for fuzzy matching once this has been brought into the final workflow too.) Bulletin of the British Museum (Natural History) ^36 PRESENTED ' A review of the Miletini(Lepidoptera: Lycaenidae) J. N. Eliot Entomology seriesVol53 Nol 25 September 1986 The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in fourscientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology,and an Historical series. Papers in the Bulletin are primarily the results of research carried out on the unique andever-growing collections of the Museum, both by the scientific staff of the Museum and byspecialists from elsewhere who make use of the Museum's resources. Many of the papers areworks of reference that will remain indispensable for years to come. Parts are published at irregular intervals as they become ready, each is complete in itself,available separately, and individually priced. Volumes contain about 300 pages and severalvolumes may appear within a calendar year. Subscriptions may be placed for one or more ofthe series on either an Annual or Per Volume basis. Prices vary according to the contents ofthe individual parts. Orders and enquiries should be sent to: Publications Sales, British Museum (Natural History),Cromwell Road, London SW75BD,England. World List abbreviation: Bull. Br. Mus. nat. Hist. (Ent.) Tustees of the British Museum (Natural History), 1986 The Entomology series is produced under the general editorship of the Keeper of Entomology: Laurence A. Mound Assistant Editor: W. Gerald Tremewan ISBN 565 06019 8ISSN 0524-6431 British Museum (Natural History)Cromwell RoadLondon SW7 5BD Entomology seriesVol53Nolppl-105 Issued 25 September 1986 (NATURAL HISTORY) 19 PRESENTEDGENr Bulletin of the British Museum (Natural History) Entomology series Vol 53 1986 British Museum (Natural History)London 1986 Dates of publication of the parts No 1 25 September 1986 No 2 . . . . 30 October 1986 No 3 . . . . . . . . . . . 30 October 1986 No 4 27 November 1986 No 5 18 December 1986 ISSN 0524-6431 Printed in Great Britain by Henry Ling Ltd., at the Dorset Press, Dorchester, Dorset ContentsEntomology Volume 53 No 1 A review of the Miletini (Lepidoptera: Lycaenidae). J.N.Eliot 1 No 2 Australian ichneumonids of the tribes Labenini and Poecilocryptini. I. D. Gauld & G. A. Holloway ... 107 No 3 The tribe Pseudophloeini (Hemiptera: Coreidae) in the Old World tropics witha discussion on the distribution of the Pseudophloeinae.W. R. Dolling 151 No 4 The songs of the western European grasshoppers of the genus Omocestus inrelation to their taxonomy (Orthoptera: Acrididae).D. R. Ragge 213 No 5 The structure and affinities of the Hedyloidea: a new concept of the butterflies. M.J. Scoble. 251 A review of the Miletini (Lepidoptera: Lycaenidae) J. N. Eliot Upcott House, Bishop's Hull, Taunton, Somerset TA4 1AQ Contents Synopsis 1 Introduction 1 Acknowledgements 2 Checklist of the Miletini 3 Tribal characters 6 Key to the genera of Miletini 7 Genus AllotinusC.&R.Felder 7 Genus Logania Distant 57 Genus Lontalius gen. n 74 Genus Miletus Hiibner 75 Genus Megalopalpus Rober 84 References 86 Index 104 Synopsis This review of the Miletini is based mainly on characters of the male genitalia. Keys to, and descriptions of,the genera, subgenera, species and subspecies are provided. One genus, one subgenus, eight species andseven subspecies are newly described. One species is relegated to subspecies, and three subspecies areraised to species; 41 new synonyms are established. Introduction The butterflies which are here considered to comprise the Oriental section of the tribe MiletiniReuter, 1897 (see Corbet, 1939b) were originally given family-group status (as the Gerydinae)by Doherty (1886), mainly on the basis of their distinctive eggs. Reuter (1897: 263) included theAfrican genus Megalopalpus Rober, 1886, in the group, treating the latter as a tribe, theMiletidi, which is distinguished from all other Lycaenidae by the peculiar labial palpi bearinglong hairs on the inner surface of the basal segment (Basalfleck). Corbet (19396) used the familygroup name Miletinae in place of Gerydinae because Gerydus Boisduval, 1836, is an objectivesynonym of Miletus Hiibner, 1819; his action is valid under Article 40b of the International Codeof Zoological Nomenclature (1985). Eliot (1973) used Miletinae in a wider sense than previousauthors, placing the Miletini (Miletinae sensu Corbet) as one of four tribes which he included inthe subfamily. Only two complete analyses of the Oriental Miletini (sensu Eliot) have so far been attempted,both by Fruhstorfer who based his studies in part on a number of genitalia preparations made byReverdin. In the first (Fruhstorfer, 1913-14) it is evident that the new taxa he described hadthree separate origins. Many descriptions are based on specimens in his own collection, some ofwhich bear type and/or determination labels attached either at that time or at some later date.Others relate to specimens in the collections of the British Museum (Natural History) andWalter Rothschild at Tring, to which no type or determination labels were attached; thesecollections were studied by Fruhstorfer during a visit to England. Finally, a few descriptions arebased on records and figures by other authors of specimens which he had not seen; inFruhstorfer's first analysis such taxa were, in most cases, based on a work by Semper (1889), sothose specimens listed by Semper may be considered to comprise Fruhstorfer's type-series. Incertain cases, however, it is clear that Fruhstorfer applied a name to a single specimen figured by Bull. Br. Mus. not. Hist. (Ent.) 53 (1): 1-105 Issued 25 September 1986 2 J. N. ELIOT Semper, and in such cases the original of the figure is automatically the holotype; in other cases,however, it has been necessary to designate a lectotype. In his second analysis (Fruhstorfer,1916) it is evident that he had in the meantime examined Semper's collection, and he describedfurther material to which type and determination labels were attached. There is some evidenceto suggest that during this examination some of Semper's original labels were inadvertentlytransferred from one specimen to another. Such instances are mentioned later under therelevant taxon. As Fruhstorfer seldom selected types, I have designated lectotypes when it is not clear that ataxon was described from a single specimen. I have also designated lectotypes for some taxanamed by other authors when I have been able to recognise primary type-material in the BritishMuseum (Natural History). When the original type-series are in museums which I have notvisited, I have generally left the selection of lectotypes to future workers, except in a few caseswhere, to avoid confusion, I have designated figured specimens. The following abbreviations have been used for the actual or supposed location of type-material. BMNH British Museum (Natural History), London CM Carnegie Museum, Pittsburgh MNHU Museum fiir Naturkunde der Humboldt-Universitat, Berlin RNH Rijksmuseum van Natuurlijke Historic, Leiden SM Senckenberg Museum, Frankfurt am Main SMT Staatliches Museum fiir Tierkunde, Dresden ZSI Zoological Survey of India, Calcutta Here it may be noted that there are in the BMNH collection a number of specimens bearingred type labels as well as, in some cases, labels in Corbet's handwriting identifying them asholotypes or allotypes which were not cited as such in the original descriptions. Where it isreasonable to assume that these were part of the original type-series I have normally designatedthem as lectotypes in preference to any other syntypes. In some cases, however, it is quite clearthat the specimens cannot possibly be part of the original type-series; such specimens I haverejected as syntypic material and labelled them accordingly. These 'types' are listed, however, inthe section dealing with species and subspecies since it is possible that they may have misledauthors in the past or might do so in the future. Acknowledgements This revision is based mainly on material in the BMNH, and I am grateful to the Trustees for access to thecollections and library, to Mr R. I. Vane-Wright and Mr P. R. Ackery for their help and encouragement,and to the Photographic Unit for providing the photographs for Figs 55-104. I thank especially MrVane- Wright and Mrs S. M. North for undertaking a review of the African genus Megalopalpus, of which Ihave no practical knowledge in the field, and for preparing the draft incorporated in this paper. I am alsovery grateful to Dr Heinz Schroder for the loan of important material from the Semper collection in theSenckenberg Museum, without which it would have been impossible to carry out a satisfactory review. Ithank Dr R. de Jong, RNH, and Professor S. Murayama, who sent material for examination. Mr G. C.Treadaway placed the whole of his extensive Philippine material at my disposal and sent me thephotographs for Figs 105-108. I also received valuable Philippine material through the generosity of MrYusuke Takanami and Dr A. Ballantine, including new taxa which I have named in their honour. Major A.Bedford Russell lent me material from Vietnam and Sulawesi, and presented the holotype of Allotinussamarensis russelli to the BMNH, and Dr E. Diehl sent me material from Simeulue. In addition thecollections of the late Mr W. A. Fleming (kindly placed at my disposal by his widow), Lt-Col C. F. Cowan,Dr T. Norman and my own have been examined. THE MILETINI Checklist of the Miletini Genus ALLOTINUS C. & R. Felder, [1865] Subgenus ALLOTINUS C. & R. Felder, [1865]fallax C.&R. Felder, [1865]fallaxfallaxC. & R. Felder, [1865]sabazus Fruhstorfer, 1913 syn. n.zaradrus Fruhstorfer, 1916fallax aphacus Fruhstorfer ,1913ancius Fruhstorfer, 1913 syn. n.artinus Fruhstorfer, 1916fallax eryximachus Fruhstorfer, 1913fallax dotion Fruhstorfer, 1913fallax tymphrest us Fruhstorfer, 1916fallax audaxH. H. Druce, 1895fallax apusde Niceville, 1895 michaelis Eliot, 1959albifasciatus Eliot, 1980subvioIaceusC. & R. Felder, [1865]subvioIaceussubviolaceusC. & R. Felder,[1865] alkamah Distant, 1886 syn. n.manychus Fruhstorfer, 1913 syn. n.kallikrates Fruhstorfer, 1913 syn. n.silarus Fruhstorfer, 1916 syn. n.subviolaceusmirusVan Eecke, 1914agnolia sp. n.nicholsi Moulton ,1911 nicholsi nicholsi Moulton, 191 1nicholsi battakanus Fruhstorfer, 1913major C. & R. Felder, [1865] depictus Fruhstorfer, 1913 syn. n.kdldwarus Ribbe, 1926max/musStaudinger, 1888 stat. n. Subgenus FABITARAS subgen. n.fabius (Distant & Pryer, 1887)fabiusfabius (Distant & Pryer, 1887)caudatus Grose-Smith, 1893 syn. n.pamisus Fruhstorfer, 1914 syn. n.fa biusarrius Fruhstorfer, 1914borneensis Moulton, 1911 elioti Corbet, 1939 syn. n.punctatus (Semper, 1889) anaxandridas Fruhstorfer, 1916 syn. n.caesemius Fruhstorfer, 1916 syn. n.nigritus (Semper, 1889) eretria Fruhstorfer, 1916strigatus Moulton, 1911 strigatus strigatus Mouhon, 1911strigatus malayanus Corbet , 1939denalus Corbet, 1939 syn. n.brooks! sp. n.bidiensissp. n.faras(Doherty, 1889) panormis (Elwes, 1893) syn. n.sarrastes Fruhstorfer, 1913 stat. n.mendava Riley, 1944 syn. n.porriginosus Toxopeus, 1932 syn. n. portunus (de Niceville, 1894) portunus port unus (de Niceville, 1894) narsares Fruhstorfer, 1913 syn. n.portunus maitus Fruhstorfer, 1914 fruhstorferi Corbet, 1939 syn. n.portunus pyxus (de Niceville, 1894) wdterstradti H. H. Druce, 1895 Subgenus PARAGERYDUS Distant, 1884Miletographa Rober, 1892horsffeldi (Moore, 1857)horsfieldi horsfieldi (Moore , 1857)horsfieldi permagnus Fruhstorfer, 1913infumata Fruhstorfer, 1913nessus Corbet, 1939horsfieldi siporanus Riley, 1944horsfieldi satelliticus Fruhstorfer, 1913leogoron Fruhstorfer, 1916 leogoron leogoron Fruhstorfer, 1916 intricata Fruhstorfer, 1913 (unavailable name)vadosus Corbet, 1939lindus Corbet, 1939leogoron normani Eliot, 1967leogoron batuensis Eliot, 1967leogoron plessis Eliot, 1967melos (H. H. Druce, 1896) sp. rev.reverdini Fruhstorfer, 1916 syn. n.talu Eliot, 1967 syn. n.samarensissp. n.samarensis samarensis subsp. n.samarensis russelli subsp. n.macassarensis (Holland, 1891)macassarensis macassarensis (Holland, 1891) damodar Fruhstorfer, 1913macassarensis menadensis Eliot, 1967luzonensis Eliot, 1967 stat. n.albatusC. & R. Felder, [1865] albatusalbatusC. & R. Felder, [1865];i//>;ifi/,v/iiriH/;/vsiihsp. n.apries Fruhstorfer, 1913 apriesapries Fruhstorfer, 1913 eupalion Fruhstorfer, 1914 syn. n.apries dositheus Fruhstorfer, 1914apries ristus subsp. n.corbeti Eliot, 1956unicolorC. & R. Felder, [1865] unicolorunicolorC. & R. Felder, [1865]eurytanus Fruhstorfer, 1913 syn. n.dilutus Corbet, 1939 syn. n.unicolor continental Fruhstorfer, 1913 dtddnus Fruhstorfer, 1916unicolor rekkia Riley & Godfrey, 1920unicolor mooreiH. H. Druce, 1895 rebilus Fruhstorfer, 1913 syn. n.unicolor aphocAaKheil, 1884 myridndus Fruhstorfer, 1913 syn. n.unicolor posidion Fruhstorfer, 1913 molionides Fruhstorfer, 1913 syn. n. J. N. ELIOT niceratus Fruhstorfer, 1913 syn. n.bajanus Fruhstorfer, 1913 syn. n.enatheus Fruhstorfer, 1913suka Piepers & Snellen, 1918enganicus Fruhstorfer, 1913 syn. n.unicolor georgius Fruhstorfer, 1913 leitus Fruhstorfer, 1916unicolor zitema Fruhstorfer, 1916paetus(de Niceville, 1895)parapus Fruhstorfer, 1913nivalis (H. Druce, 1873)nivalis nivalis (H. Druce, 1873)nivalis felderi Semper , 1 889substrigosus (Moore, 1884)substrigosus substrigosus (Moore, 1884) magaris Fruhstorfer, 1913substrigosus lenaia Fruhstorfer, 1913substrigosus sibyllinus Riley, 1944substrigosus ballantinei subsp. n.substrigosus yusukei subsp. n.da vidis Eliot, 1959(I mini In (Moore, [1866])drumila drumila (Moore, [1866])multistrigatus de Niceville, 1886insignis (Staudinger, 1888)drumila aphthonius Fruhstorfer, 1913grisea Riley & Godfrey, 1920 Genus LOGANIA Distant, 1884Malais Doherty, 1889malayica Distant, 1884malayica malayica Distant, 1884malayica subura Fruhstorfer, 1914nehalemia Fruhstorfer, 1914 stat. rev.waltraudae sp. n.regina (H. Druce, 1873)regina regina (H. Druce, 1873) evora Fruhstorfer, 1916 syn. n.regina sriwa Distant, 1886paluana sp. n.marmorata Moore, 1884marmorata marmorata Moore, 1884marmorata damis Fruhstorfer, 1914marmorata hilaeira Fruhstorfer, 1914obscura Distant & Pryer, 1887 (nom. preocc.)nada Fruhstorfer, 1914stenosa Fruhstorfer, 1916 syn. n.cineraria Fruhstorfer, 1916 syn. n.sora Fruhstorfer, 1916 syn. n.marmorata lahomius(K.hei\, 1884)marmorata diehJi subsp. n.marmorata munichya Fruhstorfer, 1914marmorata javanica Fruhstorfer, 1914 glypha Fruhstorfer, 1914marmorata palawana Fruhstorfer, 1914 distanti Staudinger, 1889 (nom. preocc.)marmorata samosata Fruhstorfer, 1914marmorata faustina Fruhstorfer, 1914 obscura Rober, 1886 martinus (Fruhstorfer, 1913) syn. n.donussa Fruhstorfer, 1916 syn. n.distanti Semper, 1889 distanti distanti Semper , 1889apsines Fruhstorfer, 1914turdeta Fruhstorfer, 1916 syn. n.distanti massalia Doherty, 1891 stat. n. luca de Niceville, 1894 syn. n.distanti drucei Moulton ,1911distanti staudingeriH. H. Druce, 1895hampsoni Fruhstorfer, 1914 meeki Rothschild, 1915 syn. n.masana Fruhstorfer, 1916 syn. n.watsoniana de Niceville, 1898subfasciata Tytler, 1915 Genus LONTALIUS gen. n.eltussp. n.eltus eltus subsp. n.eltus treadawayi subsp. n. Genus MILETUS Hubner, 1819Symetha Horsfield, 1828Gerydus Boisduval, 1836Archaeogerydus Fruhstorfer, 1916chinensisC. Felder, 1862 chinensischinensisC. Felder, 1862 chinensislearchusC. & R. Felder, [1865]irroratusH. Druce, 1874kelantanus Corbet, 1938 chinensis assamensis (Doherty, 1891)milvius (Fruhstorfer, 1913) chinensis longeana (de Niceville, 1898)croton (Doherty, 1889) croton croton (Doherty, 1889)tavoyana (Evans, 1932) croton corus Eliot, 1961 croton karennius (Evans, 1932)mallus (Fruhstorfer, 1913) mallusmallus (Fruhstorfer, 1913) mallus gethusus (Fruhstorfer, 1913) mallus shanius (Evans, 1932)gaesa (de Niceville, 1895) gaesagaesa (de Niceville, 1895) gaesa carrinas (Fruhstorfer, 1916)nymphis (Fruhstorfer, 1913) n ymphis n ymphis (Fruhstorfer, 1913) nymphis porus Eliot , 1 96 1 nymphis fictus Corbet, 1939 nymphis eneus Eliot, 1961zinckeniiC. & R. Felder, [1865] zinckeniizinckeniiC. & R. Felder, [1865] zinckenii improbus (H. H. Druce, 1895)gopara (de Niceville, 1890) goparagopara (de Niceville, 1890)denticulata (Fruhstorfer, 1913) gopara pardus Eliot , 196 1 gopara eustatius (Fruhstorfer, 1913) THE MILETINI gopara artaxatus (Fruhstorfer, 1913) oichalia (Fruhstorfer, 1913)valeus (Fruhstorfer, 1913) pallaxopas (Fruhstorfer, 1913)gaetulus (de Niceville, 1894) gaet ulus gaet ulus (de Niceville, 1894) gaetulus innocens(H. H. Druce, 1895) gaetulus aphytis (Fruhstorfer, 1913)boisduvali Moore, 1857 boisduvali boisduvali Moore, 1857vincula (H. H. Druce, 1895)heraeon (Fruhstorfer, 1916)courvoisieri (Fruhstorfer, 1915)oxylus (Fruhstorfer, 1916)lombokianus (Fruhstorfer, 1913)acragas (Doherty, 1891)buruensis (Holland, 1900)ceramensis Ribbe, 1889dossemus (Fruhstorfer, 1913)stygianus Butler, 1884adeus (Fruhstorfer, 1913) boisduvali diotrophes (Fruhstorfer, 1913) boisduvali avitus (Fruhstorfer, 1916)drucei (Semper, 1889) druceidrucei (Semper, 1889)philippus (Staudinger, 1889)jacchus (Fruhstorfer, 1913)palanius (Fruhstorfer, 1913)epidurus (Fruhstorfer, 1913) drucei metrovius (Fruhstorfer, 1913)phradimon (Fruhstorfer, 1915)biggsii (Distant, 1884) biggsii biggsii (Distant, 1884)atomaria (Fruhstorfer, 1913)xeragis (Fruhstorfer, 1916)hyllus (Fruhstorfer, 1916)sebethus (Fruhstorfer, 1916)extraneus (Toxopeus, 1929) biggsii natunensis (Fruhstorfer, 1916) biggsii niasicus (Fruhstorfer, 1913)batunensis (Fruhstorfer, 1913) biggsii albotignula (Van Eecke, 1914) simalurensis (Toxopeus, 1928)cellarius (Fruhstorfer, 1913)symethus (Cramer, 1779) symethus symethus (Cramer, 1779)pandu (Horsfield, 1828) symethus petronius (Distant & Pryer, 1887)diopeithes (Fruhstorfer, 1913)bangkanus (Fruhstorfer, 1914)hieropous (Fruhstorfer, 1916) symethus solitariusOkubo, 1983 symethus acampsis (Fruhstorfer, 1913) symethus nuct us Eliot, 1961 symethus perlucidus (Fruhstorfer, 1913)megaris (Fruhstorfer, 1913) symethus vespesianus (Fruhstorfer, 1913) symethus batuensis (Fruhstorfer, 1914) symethus edonus (Fruhstorfer, 1913)symethus philopator (Fruhstorfer, 1914)symethus hierophantes (Fruhstorfer, 1916)symethus phant us subsp. n. gallus (de Niceville, 1894)gallus gallus(de Niceville, 1894)gallus leucocyon (Toxopeus, 1940)heracleion (Doherty, 1891) heracleion heracleion (Doherty, 1891)heracleion arion Eliot, 1961a/icon (Doherty, 1889)ancon ancon (Doherty, 1889)ancon gigas(H. H. Druce, 1895) anconides (Fruhstorfer, 1913)archilochus (Fruhstorfer, 1913) archilochus archilochus (Fruhstorfer, 1913)archilochus siamensis (Godfrey, 1916)gigantes(de Niceville, 1894)atimonicus Murayama & Okamura, 1973celinus Eliot, 1961takanamiisp. n./eos(Guerin-Meneville, 1830)leosteos (Doherty, 1891)leos fforensis (Fruhstorfer, 1913) eulus (Fruhstorfer, 1913)leostellus (Fruhstorfer, 1913)leos catoleucos (Fruhstorfer, 1913)leos maximus (Holland , 1 89 1 )divisa (Fruhstorfer, 1913)sarus (Fruhstorfer, 1913)leosvaneecki (Toxopeus, 1930)leos mangolicus (Fruhstorfer, 1913)/eos/eos(Guerin-Meneville, 1830)boisduvalii (Butler, 1884)meronus (Fruhstorfer, 1913)amphiarus (Fruhstorfer, 1913)gardineri (Fruhstorfer, 1914)leos virtus (Fruhstorfer, 1913) pentheus (Fruhstorfer, 1913)leosaronicus (Fruhstorfer, 1914)nineyanus (Fruhstorfer, 1914)acrisius (Fruhstorfer, 1914)melanion C. & R. Felder, [1865] melanionmelanionC. & R. Felder, [1865] albiguttatus f. n. (infrasubspecific name)melanion euphranor (Fruhstorfer, 1914)bazilan us (Fruhstorfer, 1913) stat. n. vitelianus (Fruhstorfer, 1913) syn. n. Genus MEGALOPALPUSRober, 1886 angulosus Griinberg, 1910mete/eucusKarsch, 1893simplex Rober, 1886 bicoloria (Capronnier, 1889) similis (Kirby, 1890) gigas Bethune-Baker, 1914zymna (Westwood, 1851) pallida Aurivillius, 1922 6 J. N. ELIOT Tribal characters The characters of Miletini (sensu Eliot, 1973) are as follows. Eyes smooth. Antennae withnarrow, gradually incrassate club and with the nudum extending down the shaft to the base orvery nearly so. Labial palpi asymmetrical, but there is no constancy in which palpus is the longer.In most species the palpi are unusually long and thin, more so in females than in males, andprotrude well beyond the head; they are also unusual in that the 'Basalfleck' of Reuter is clothedwith hairs of unknown function. Proboscis long, bearing many sensilla throughout its length.Legs more or less abnormal, with the tarsi very long, flattened and blade-like in Miletus andMegalopalpus, cylindrical but very long and thin in Allotinus and with the tibiae outwardlyswollen or incrassate in Logania and Lontalius. The mid- and hind-tibiae lack the usual pair ofterminal spurs. The male fore-tarsus is reduced to a single segment ending abruptly but with asmall pointed process directed downwards, except in the typical species of Logania in which thetarsus tapers to a down-curved point. The claws on the mid- and hind-tarsi of both sexes aresmall, and minute on the fore-tarsi of females. The abdomens of males are long and protrudewell beyond the hindwings, except in Allotinus major. There is a double hair tuft, sometimesinconspicuous but sometimes large and erectile, on the sternum of the eighth segment; it is notknown whether it plays any part in courtship. The tergum of the eighth segment is unusually longto accommodate the peculiar genitalia, and bears a long apophysis at its proximo-ventral edge.The wing venation shows a high degree of individual variation, and does not often provide goodcharacters for separating genera and species. The forewing always has 11 veins (vein R 4missing), and veins Sc and RI are separate throughout their lengths. In the males of most speciesthe basal portion of vein M 3 is slightly swollen and devoid of normal cover scales, but bears smallto very small specialised (?scent) scales. The hindwing has a well-developed humeral vein inMegalopalpus and Lontalius, but in the other genera it is only weakly developed or absent. Themale genitalia are highly characteristic. The uncus and tegumen are in the form of enormouspaired plates, which are attached to the vinculum only narrowly in the dorsal region, so that theyare capable of considerable freedom of movement . Articulating brachia are always present . Thevinculum bears on either side two more or less triangular processes, one directed proximad andthe other distad. The latter process is comparatively weakly sclerotised and overlies the sides ofthe tegumen. The valvae are small, with the outer dorsal portion (sometimes referred to as theampulla) bearing a dense hair fringe. The juxta is present as a furca whose arms are united by aband above the phallus just distad of the ductus. Eliot (1973: 386) incorrectly called this band aform of transtilla. The female genitalia have not been investigated. The early stages are very imperfectly known. According to Doherty (1889) the eggs of Miletusand Allotinus are much flattened and disc-like, but in Logania are stouter and scarcely more thantwice as wide as high. They bear between two and five lateral carinae which are either simple orbroken into short teeth placed one above the other, giving the appearance of a cogged wheel.The larvae, so far as known, are wholly aphytophagous, feeding on Homoptera. They are moreor less cylindrical and have a particularly thick cuticle, and apparently lack the 'honey gland' onthe seventh and paired eversible tubercles on the eighth abdominal segments which are presentin the majority of Lycaenidae. Their relationship with the ants attending the Homopteraappears to be one of neutrality, and from this it appears at least possible that the larvae may befurnished with small glands on a number of segments, as in many other Lycaenidae (Cottrell,1984), and that these secrete some substance which inhibits ant aggression. In one species,Miletus boisduvali, the larva pupates inside ants' nests and the pupa has attractant glands and isattended by ants; the emerging adult is clothed with fugitive scales, as in Liphyra brassolis,which confuse attacking ants (Roepke, 1918). However, in M . chinensis, the larva pupates in theopen and is attached by the cremaster with or without a weak girdle (Kershaw, 1905) , and in thisspecies, as well as in Allotinus subviolaceus, there is no evidence to suggest that fugitive scalesare present in the adult (Piepers & Snellen, 1918). The adults feed on the excretions ofHomoptera and do not visit flowers. The tribe comprises four Oriental genera and one African genus (the latter erroneouslydescribed from Borneo). The Oriental genera, totalling 69 mainly Sundanian species, have been THE MILETINI 7 subjected to a full taxonomic revision. No comparable attempt has been made to revise the fourspecies of Megalopalpus , but a review of the genus, based on a draft kindly prepared by Mr R. I.Vane-Wright and Mrs S. M. North, is included. In the keys and descriptions which follow, the system of veins and spaces is as in Fig. 49 (p. 74). Key to the genera of Miletini 1 Tarsi with first segment more or less cylindrical 2 - Tarsi with first segment flattened and blade-like 4 2 Tibiae swollen or incrassate 3 - Tibiae not swollen; legs long, thin, cylindrical ALLOTINUS (p. 7) 3 Hindwing without a humeral vein. Legs comparatively short LOGANIA (p. 57) - Hindwing with a humeral vein . Legs long and thin LONTALIUS (p. 74) 4 Hindwing without a humeral vein. Oriental MILETUS (p. 75) Hindwing with a humeral vein. African MEGALOPALPUS (p. 84) Genus ALLOTINUS C. & R. Felder Allotinus C. & R. Felder, [1865]: 285. Type-species: Allotinus fallax C. & R. Felder, [1865]: 285, pi. 35, fig.24, by designation of Scudder, 1875: 107. Gender masculine. Legs long, thin, cylindrical. Labial palpus with third segment longer than half second segment. The wingvenation shows a high degree of infra-specific individual variation. Hindwing cilia elongated into short tuftsat vein endings in the male, the margin more or less crenulate and tufted in the female. Undersidecharacteristic, white to pale buff, densely striated with small striae or spots, with heavier spots arranged inthe usual lycaenid pattern. Forewing usually with very small white costal flecks at ends of veins Sc, R\, R2,/? 3 and, in one species-group, R 5 also. The genus ranges from north India to Sundaland, the Philippines and Sulawesi. Fruhstorfer (1913-15; 1916) divided the genus into two subgenera: Allotinus for species allegedlywithout a sex stripe in the male and Paragerydus for species with a sex stripe. I maintain these subgenera,though not in the arrangement adopted by Fruhstorfer, and add a third subgenus. Key to the subgenera of Allotinus 1 cf valva ending in an apical point; abdominal hair tufts poorly developed, not protruding except when genitalia are extruded. Hindwing usually with a weak humeral vein 2 - Cf valva with costa abruptly truncate (except in A . davidis) , but ending in a ventral , more or less pointed process; abdominal hair tufts prominent. Hindwing without a humeral vein PARAGERYDUS (p. 30) 2 Antenna barely longer than half the forewing costa, with less than 50 segments, cf forewing with vein A/3 not, or only very briefly, swollen ALLOTINUS (p. 7) - Antenna nearly two-thirds length of forewing costa, with 60 or more segments, cf forewing vein M 3 prominently swollen, clothed with specialised scales for one-third of its length or longer FABITARAS (p. 20) Subgenus ALLOTINUS C. & R. Felder The principal characters of the subgenus are as given in the key. Nineteenth century authors separated Allotinus from Paragerydus by the former's possession of a shortupper discocellular vein, but this character does not occur in A. (A.) agnolia, wherein veins R 5 and M l havea short common stalk. Typical species, in which the males have vein M 3 unswollen, have vein /? 3 long,arising closer to the cell apex than to the wing apex; but in the two species in which the males have vein A/ 3weakly swollen, vein /? 3 is shorter, arising close to or opposite the end of vein R 2 . On the basis of the male structure the subgenus can be divided into two species-groups: the/fl/fax-groupin which the abdomen is of normal length for the tribe and the uncus/tegumen plates are long and narrow;and the major-group in which the abdomen is much shorter and the uncus/tegumen plates are morerounded. The subgenus has a restricted distribution in Sundaland, the Philippines and Sulawesi, and comprisesseven species. 8 J. N. ELIOT Key to the species of subgenus Allotinus 1 cf abdomen longer than hindwing dorsum. Underside of hindwing with postdiscal spot in space 6 more or less below the spot in space 7 and remote from the spot in space 5 (fallax- group) 2 Cf abdomen same length as hindwing dorsum. Underside of hindwing with postdiscal spot inspace 6 more or less equidistant from the spots in spaces 7 and 5 (major-group) 6 2 Upperside with white or grey-blue areas 3 Upperside plain brown nicholsi (p. 16) 3 Upperside with white areas 4 Upperside with greyish-blue areas subviolaceus (p. 14) 4 cf upperside of hindwing brown or diffusely sullied with white scales 5 Cf upperside of hindwing with a clearly defined white band albifasciatus (p. 13) 5 Forewing with veins R 5 and MI just separate at their origins, cf forewing vein M 3 not swollen nor clothed with specialised scales. Underside with postdiscal series of spots clearly marked fallax (p. 8) - Forewing with veins RI and M\ stalked, cf forewing vein M 3 briefly swollen and clothed with specialised scales. Underside with postdiscal series not apparent on forewing and barelydiscernible on hindwing agnolia (p. 15) 6 Smaller, forewing 13-5-19-0 mm. cf upperside of forewing all brown or with a small white or whitish spot comprising at most a small sullied area at base of space 2 and a larger white areaup to 3-0 mm wide in space Ib. $ with a larger white area filling basal third of space 2 and abouthalf of space Ib, but not reaching vein Cu\ nor entering the cell; white patch may be reduced,sullied or absent major (p. 17) - Larger, forewing 20-0-21-0 mm. cf upperside of forewing with a white patch filling basal third of space 2 and about half of space Ib which is basally grey. $ with white patch larger, extendingabove vein Cu\ and into lower edge of cell maximus (p. 19) Allotinus (Allotinus) fallax C. & R. Felder (Figs 1-3, cf genitalia)Allotinus fallax C. & R. Felder, [1865]: 285, partim. The Felders confused two distinct species under this name. In pi. 35, fig. 24 they figured as a male what iscurrently treated as the female of A. fallax, and at figs 25, 26 as the female a different species which Idescribe later as the female of A. (Paragerydus) albatus mendax. Semper (1889: 163) correctly pointed outthat the Felders' 'male' was in fact a female. However, he incorrectly identified their female as a male Fig. 1 Allotinus (Allotinus) fallax fallax C. & R. Felder; Luzon. Male genitalia. Above, lateral view ofright half of armature ; below, dorsal view of phallus. THE MILETINI 9 variety of fallax, having a white area on the upperside of the hindwing such as he had seen only in a fewexamples from Luzon; these may have been either males or females ofmendax. Apart from these supposedvariants from Luzon he correctly stated that the male of fallax has a plain brown hindwing in thePhilippines. The species is distinguished by lacking a swelling of vein A/ 3 in the male and by the presence of a whitepatch on the upperside of the dark brown forewing. The hindwing of the male is brown, except in Borneowhere it is usually dusted with white scales. Except in Sumatra and the Malay Peninsula, where thehindwing is brown, the female has a white discal area (rarely vestigial or absent) which shows a high degreeof individual variation, so that it is an unreliable character by which to distinguish subspecies. On the basis of the male genitalia A. fallax can be divided into three subspecies-groups. Type (a) (Fig. 1): valva with short apical hook and ventro-distal edge not strongly exarcate; phalluscomparatively short, broad and abruptly tapered at distal end; Philippine. Type (b) (Fig. 2): valva with short apical hook and ventro-distal edge strongly exarcate; phalluscomparatively long and thin and gradually tapered at distal end; Philippine. Fig. 2 Allotinus (Allotinus) fallax aphacus Fruhstorfer ; Mindanao. Male genitalia. Type (c) (Fig. 3): valva with long apical hook; phallus like type (c), but still slimmer; Sundanian. So far as my experience goes type (a) occurs exclusively in Luzon, Masbate, Bohol, Samar and Leyte,and type (b) exclusively in Mindanao and Bazilan. Both occur in Mindoro, each type having a distinctivephenotype. I have not been able to dissect males from Cebu, Panaon, Camiguin de Mindanao and the SuluIs. , whence Fruhstorfer named subspecies, but I think it certain that type (a) will be found in Cebu and type(b) in Camiguin de Mindanao and Sulu Is. , whilst in Panaon either might occur. The occurrence of type (a)in Bohol, Samar and Leyte is rather unexpected; usually butterflies from these islands are associated withMindanao rather than with Luzon. The dichotomy in Mindoro can be paralleled in the lycaenopsidAcytolepis puspa which occurs in Mindoro in subsp. cagaya and the very different-looking subsp. bazilanain equality and without evidence of intergradation (Eliot & Kawazoe, 1983: 183). Conceivably types (a)and (b) may represent distinct species which evolved in the northern and southern groups of islandsrespectively and which may now have a wider overlap in distribution than is at present known; otherwise itis difficult to see how type (b) could have reached Mindoro without also becoming established in theintervening islands. A. fallax flies at low to moderate elevations, and is common and widespread in the Philippines, but rarein Sundaland. It has not been found in Palawan, but is likely to occur there. Key to the subspecies of A. (A.) fallax1 cf genitalia of type (c) . Sundanian - cf genitalia of types (a) or (b). Philippine 2 cf genitalia of type (b). $ upperside of hindwing with white area normally confined to basaltwo-thirds of spaces 4 and 5 , but may be vestigial or obsolete 10 J. N. ELIOT Cf genitalia of type (a). $ upperside of hindwing with white area normally extending below veinA/3, often as far as vein A! fallax fallax (p. 10) 3 cf upperside of forewing with white patch reaching vein Cu\ 5 Cf upperside of forewing with white patch usually not above mid-space 2; if reaching vein Cui, upper part narrow and sullied 4 4 cf white patch comparatively large, reaching into space 2. Underside of forewing with central area above dorsum sparsely striated fallax aphacus (p. 11) Cf sullied white patch smaller, usually confined to space Ib. Underside of forewing with centralarea above dorsum as densely striated as rest of wing fallax eryximachus (p. 11) 5 Underside ground colour pale buff. $ upperside of hindwing with a small white patch fallax dotion (p. 12)Underside ground colour whitish buff, with pale markings. $ hindwing white patch obsolete fallax tymphrestus (p. 12) 6 Upperside of hindwing with a dusting of white scales usually present in cf and always present in $ fallaxaudax (p. 12) Upperside of hindwing plain brown fallax apus (p. 12) Allotinus (Allotinus) fallax fallax C. & R. Felder(Fig. 1, cf genitalia) Allotinus fallax C. & R. Felder, [1865]: 285, partim 'cf ', recte $ [nee $ , pi. 35, fig. 24 'cf']; Semper, 1889:163, partim, pi. 31, figs 23cf, 26$. LECTOTYPE $, PHILIPPINES: Luzon (BMNH), here designated[examined]. Allotinus fallax fallax Felder; Fruhstorfer, 1913: 343; 1916: 809. Allotinus fallax sabazus Fruhstorfer, 1913: 343; 1916: 809. LECTOTYPE cf, PHILIPPINES: Bohol [notlocated] . Syn. n. Allotinus fallax zaradrus Fruhstorfer, 1916: 809. Syntypes, PHILIPPINES: Cebu [not located]. In the male the white patch on the forewing almost always extends across vein Cu\ into space 3. In thefemale the white patch on the hindwing is variable in extent, but almost always extends dorsad into space 3and often as far as space Ib. The ground colour of the underside is pale buff, richer in the male than in thefemale. The Felders did not specify a type. Their 'male', as pointed out by Semper (1889: 163), is a female, whiletheir female belongs to another species described later as A. (Paragerydus) albatus mendax. In BMNHthere is a male ex Felder coll. bearing a BMNH red type label, but as the original figure and descriptiondoes not apply to this sex it seems better to reject this unpublished type selection and to designate aslectotype a female in BMNH labelled /79/Luzon Lorquin [round blue]/Felder Coll. /Rothschild Bequest1939-17. When naming sabazus, Fruhstorfer, evidently referring to the male, stated that the white patch on theforewing was reduced compared with subsp. fallax and that the underside was darker and more closelystriated. The male from Bohol figured by Semper has the white patch entirely below vein Ci, as inoccasional Luzon examples, but in a series of both sexes from Bohol in coll. Treadaway the males have thewhite patch extending above vein Cu\ and in neither sex can I detect any consistent differences from Luzonfallax. I therefore consider that sabazus cannot be maintained as a valid subspecies. There is in SM a femalelabelled /Bohol/Coll. C. Semper/2 11/Typus [red]/, but in my view it cannot be accepted as the type ofsabazus since Fruhstorfer's description did not apply to this sex. I therefore designate as lectotype the malefigured by Semper (pi. 31, fig. 23), which unfortunately has not been recognised in SM. Semper recorded fallax from Cebu, but did not figure it. It is not clear whether Fruhstorfer (1916) wasmerely attaching a name to Semper's record or actually saw specimens from Cebu. No examples from Cebucan now be found in SM, but according to Treadaway (pers. comm.) it is thought that some may possiblyexist but be temporarily mislaid. The only specimen that I have seen from Cebu is a female in coll.Treadaway. This has the underside ground colour somewhat whiter than usual, but the difference fromLuzon females seems too slight to justify retaining zaradrus as a valid subspecies, and I thereforeprovisionally synonymise it with /a/tax. Examples from Mindoro and Sibuyan Is. do not differ from normal fallax, but two males from Masbatein coll. Treadaway have a more whitish underside, and a female from Panay is almost pure white andrelatively weakly striated below and has an exceptionally large white patch on the upperside of thehindwing. I provisionally place these under fallax . Males from Samar and Leyte seem never to have the forewing white patch extending above vein Cui and THE MILETINI 11 in both sexes the underside ground colour is darker buff than in examples from Luzon and Bohol. Thesepossibly constitute a distinct subspecies to which Fruhstorfer's name artinus, given by him to the populationin Panaon, may apply. Unfortunately I have seen no males from Panaon and Semper's figure of a Panaonmale (pi. 31, fig. 24) shows only the underside, from which it is not possible to establish whether or not itresembles Samar and Leyte examples. The single female from Panaon which I have seen does not differfrom Mindanao females, and I therefore provisionally synonymise artinus with aphacus. DISTRIBUTION. Luzon, Mindoro, Sibuyan, Cebu, Bohol. In slightly different forms also in Masbate, Panay,Samar and Leyte. I have seen no examples from Negros, but it must occur there as well as in other islands. Allotinus (Allotinus) fallax aphacus Fruhstorfer(Fig. 2, cf genitalia) Allotinus fallax C. & R. Felder; Semper, 1889: 163, partim, pi. 31, figs 24 cf , 25 $ . Allotinus fallax aphacus Fruhstorfer, 1913: 343; 1916: 809. Holotype $, PHILIPPINES: Camiguin de Mindanao [not located].Allotinus fallax ancius Fruhstorfer, 1913: 343; 1916: 809. LECTOTYPE cf , PHILIPPINES: Mindanao (BMNH), here designated [examined]. Syn. n.Allotinus fallax artinus Fruhstorfer, 1916: 809. LECTOTYPE $, PHILIPPINES: Panaon (SM), here designated [examined]. On the upperside the forewing white patch of the male is barely half as large as in subsp. fallax, seldom fullycrossing space 2, and when it does so the upper part is sullied with brown scales. The female has the whitepatch on the hindwing almost always restricted to the basal two-thirds of spaces 4 and 5, and often it isalmost obsolete. On the underside of the forewing there is an almost unstriated central area above thedorsum. The male genitalia are type (b). Fruhstorfer named aphacus from Semper's record and figure 25 of a female from Camiguin deMindanao. The original of this figure, which I have not been able to locate, is therefore automatically theholotype. I have seen a single male labelled /Cam. de Mind./211/Coll. C. Semper/ which does not differfrom Mindanao males. I designate as lectotype of ancius a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Mindanao/Mindanao Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/Allotinus fallax ancius [in Corbet'shand]/. A female labelled /Type [red]/Type [Fruhstorfer orange]/Mindanao/Mindanao Fruhstorfer/Fruh-storfer Coll. B.M. 1933-131/fallax ancius Fruhst. [in Fruhstorfer's hand]/ becomes a paralectotype. I designate as lectotype of artinus a female in SM labelled /Coll. C. Semper/Panaon/211/Typus [red]/,which does not differ from Mindanao examples. DISTRIBUTION. Mindanao, Camiguin de Mindanao, Panaon. I also provisionally place under this subspeciesa female in BMNH from Talaud I. Allotinus (Allotinus) fallax eryximachus Fruhstorfer Allotinus fallax eryximachus Fruhstorfer, 1913: 343; 1916: 809. LECTOTYPE cf , PHILIPPINES: Mindoro(BMNH), here designated [examined]. This is the darkest subspecies. As pointed out by Fruhstorfer in his original description, the forewing whitepatch in the male is even more insignificant (unbedeutenderen) than in subsp. aphacus, being always moreor less sullied with brown scales and sometimes restricted to a mere streak below vein Cu 2 . On theunderside the central area above the forewing dorsum, which is not, or is comparatively weakly, striated inthe other subspecies, is striated all over. The male genitalia are type (b). Fruhstorfer stated that the type was in BMNH. However, there are no males agreeing with hisdescription labelled as types, but a pair of subsp. fallax have been placed in the type drawer above the nameeryximachus. The male, with a large white forewing patch, is labelled /Mindoro Philippine Is. Dr.Platen/Godman-Salvin Coll. 1908-168/A. fallax 107 Var. ?/, and the female is labelled like the male butwith an additional label /Allotinus 'cf ' fallax Felder/. Agrees with figure of type F. A. H. 8. x. 09/. Neitherspecimen bears a BMNH red type label, presumably because they cannot be reconciled with Fruhstorfer'sdescription of eryximachus. Although Fruhstorfer may have seen this pair in BMNH they are not to beconsidered as types, and I designate as lectotype a male from a series which Fruhstorfer may have seen atTring and which he may have confused with what he saw in BMNH; it is labelled /Mindoro Platen/Rothschild Bequest 1939-1/. It is interesting that Platen should have caught both subspecies in Mindoro. DISTRIBUTION. Mindoro, where it appears to outnumber sympatric subsp. fallax. 12 J. N. ELIOT Allotinus (Allotinus) fallax dotion Fruhstorfer Allotinus fallox dotion Fruhstorfer, 1913: 343; 1916: 809, pi. 141h cf $ as 'dolion\ LECTOTYPE cf,PHILIPPINES: Bazilan (BMNH), here designated [examined]. On the upperside both sexes resemble subsp. fallax, with the white patch on the forewing of the maleunsullied and reaching vein 1*2 broadly. On the underside the ground colour is a paler, more whitish buffthan in subspp. sabazus and aphacus. The male genitalia are type b. I designate as lectotype a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/PhilippinenBazilan II-III. 98 Doherty ex coll. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/fallax dotion Fruhst. [inFruhstorfer's hand]/. A female labelled similarly, except that the final label reads /dotion [in Fruhstorfer'shand]/, is labelled paralectotype. DISTRIBUTION. Bazilan. Allotinus (Allotinus) fallax tymphrestus Fruhstorfer Allotinus fallax tymphrestus Fruhstorfer, 1916: 809. LECTOTYPE cf, PHILIPPINES: Sulu Is. (SM), heredesignated [examined] . I have seen only the original pair from which Fruhstorfer described the subspecies. The male exactlyresembles subsp. dotion on the upperside, but on the underside the ground colour is slightly more whitishand the markings less well-defined. The female lacks the usual white patch on the upperside of thehindwing, but it may be that a white patch is sometimes present. The subspecies is doubtfully separate fromsubsp. dotion. I designate as lectotype the male in SM labelled /Coll. C. Semper/Sulu Inseln Meyerink/211/Typus[red]/. The female is labelled paralectotype. DISTRIBUTION. Sulu Is. Allotinus (Allotinus) fallax audaxH. H. Druce Allotinus audox H. H. Druce, 1895: 564, pi. 31, figs 11 cf , 12 $>. LECTOTYPE cf , BORNEO: Mt Kina Balu (BMNH), here designated [examined].Allotinus fallax audax Druce; Fruhstorfer, 1913: 343; 1916: 809; Corbet, 1939ft: 66. In both sexes the forewing white patch is more extensive than in the Philippine subspecies, almost reachingthe wing base in space Ib where it is dusted with grey scales. The hindwing is more or less lightly dusted withwhite scales in the male, but heavily dusted over most of the wing in females. The subspecies was described from an unspecified number of specimens in coll. H. H. Druce andStaudinger. I designate as lectotype a male in BMNH labelled /Kina Balu Waterstr. A. audax co-typeH. H. Druce/ex coll. Hamilton Druce 1919/Type PT [green]/Joicey Bequest Brit. Mus. 1934-120/. Afemale labelled $ co-type, but otherwise as the male, is a paralectotype. DISTRIBUTION. Borneo; apparently only known from Mt Kina Balu. Allotinus (Allotinus) fallax apus de Niceville(Fig. 3, cf genitalia) Allotinus apus de Niceville, 1895: 27, pi. S, fig. 17 $; Fruhstorfer; 1913: 343; 1916: 809. Syntypes $, SUMATRA (? ZSI).Allotinus fallax apus de Niceville; Corbet, 1939ft: 66; Eliot, 1962: 218; 1978: 240; 1980: 145, figs 7 cf , 14 cf genitalia; Fleming, 1975; 22, pi. 50, fig. L50 cf .Allotinus fallax michaelis Eliot, 1959: 376, partim cf nee $ , pi. 10, fig. 8 cf . Holotype cf , WEST MALAYSIA (BMNH) [examined]. [Synonymised by Eliot, 1962: 218.] Differs from subsp. audax in the plain brown hindwing in both sexes. A. apus was described from two females from north-east Sumatra, which should be in ZSI. There is inBMNH a female labelled /Type [red]/Battak Mts. N.E. Sum. 11. 94 Dr Martin/ Allotinus apus de Nicevilleand, [on reverse] , Collect Dr. Martin/Rothschild Bequest B.M. 1939-1/. As there is no evidence to suggestthat this female was one of de Niceville's original two I reject it as a type. DISTRIBUTION. Sumatra and West Malaysia, at elevations of about 500 m upwards. THE MILETINI 13 Fig. 3 Allotinus (Allotinus) fallax apus de Niceville; Malay Peninsula. Male genitalia. Allotinus (Allotinus) albifasciatus Eliot (Fig. 4, cf genitalia) [Allotinus fallax michaelis Eliot, 1959: 376 (partim, $ nee cf), pi. 10, fig. 9 <j>. Misidentification.][Allotinus fallax audax H. H. Druce sensu Eliot, 1978: 241. Misidentification.] Allotinus albifasciatus Eliot, 1980: 143, figs 5 d", 6 <j>, 14 cf genitalia. Holotype cf, SUMATRA (BMNH)[examined]. The species is distinctive in possessing a clear white band crossing the hindwing in both sexes. On theunderside the ground colour is off-white, with the postdiscal markings larger and more blotchy than in A.fallax. DISTRIBUTION. The species is extremely rare, and has so far only been found in the Malay Peninsula andSumatra at elevations of about 1000 m. Fig. 4 Allotinus (Allotinus) albifasciatus Eliot; Sumatra. Male genitalia. 14 J. N. ELIOT Allotinus (Allotinus)subvioIaceusC. & R. Felder (Fig. 5, cf genitalia)Allotinus subviolaceus C. & R. Felder, [1865]: 286. The greyish blue areas on the upperside render this species unmistakable. As in the white areas of A. fallaxthese blue areas are individually variable in extent, especially on the hindwing where they may be absent orcover almost the whole wing. On average they are more extensive on both wings in the female. The species has the widest distribution in the subgenus, occurring from Assam to Java and thePhilippines. Key to the subspecies of A. subviolaceus 1 Upperside greyish blue. Upperside of forewing in cf with space 3 all or almost all black subviolaceus subviolaceus (p. 14) Upperside pale grey, with only a slight blue tint , becoming whitish in lower half of forewing disc.Upperside of forewing in cf with basal half of space 3 and base of space 4 bluish grey subviolaceus mirus (p. 15) Allotinus (Allotinus) subviolaceus subviolaceus C. & R. Felder (Fig. 5 cf genitalia) Allotinus subviolaceus C. & R. Felder, [1865]: 286, pi. 35, figs 27, 28 cf ; Piepers & Snellen, 1918: 17, pi. 20,figs 20a cf , 20b $ , 20c larva. LECTOTYPE cf , JAVA (BMNH), here designated [examined]. Allotinus alkamah Distant, 18860: 452, pi. 44, 'cf ' recte <j>. Holotype <j>, WEST MALAYSIA (not located).Syn. n. Allotinus subviolescens [sic] Felder; Swinhoe, 1910: 196, pi. 616, figs 1, la $, Ib cf. Allotinus subviolaceus alkamah Distant; Fruhstorfer, 1913: 342; 1916: 808; Corbet, 1939a: 66; Eliot, 1978:240; Fleming, 1975: pi. 57, fig. L37 $. Allotinus subviolaceus subviolaceus C. & R. Felder; Fruhstorfer, 1913: 342; 1916: 808, pi. 141g . Allotinus subviolaceus manychus Fruhstorfer, 1913: 342; 1916: 808; Evans, 1932: 212; Cantlie, 1967: 28.LECTOTYPE <j>, BURMA (BMNH), here designated [examined]. Syn. n. Allotinus subviolaceus kallikrates Fruhstorfer, 1913: 342; 1916: 808. LECTOTYPE cf, PHILIPPINES:Mindanao (BMNH), here designated [examined]. Syn. n. Allotinus subviolaceus silarus Fruhstorfer, 1916: 808. LECTOTYPE cf , BORNEO (BMNH), here desig-nated [examined]. Syn. n. Males have a broad black border on the forewing which often fills the whole of space 3 and the cell, but insome examples the blue may enter the lower half of space 3 and fill the lower two-thirds of the cell. Usuallythe hindwing has only a few blue scales. In the female the forewing border is much narrower, and the hindwing varies individually from whollyfuscous to nearly all blue. In Palawan the greyish blue colour tends to be paler in both sexes than in other areas, and this does notappear to be due to season, as in the still paler examples flying in Burma and Thailand during the dryseason. In addition the forewing border tends to be wider in females, measuring an average of 2-5 mm atvein Cu 2 compared with an average of 2-0 mm in normal examples. These differences seem hardly sufficientto justify erecting another subspecies. Females from Borneo and Bangka also have, on average, theforewing border similar to Palawan examples. The Felders described subviolaceus only from the male, but did not designate a type. I thereforedesignate as lectotype a male in BMNH labelled /Type [red]/Java Coll. . . . [illegible]/[circular blue]/Allotinus subviolaceous Feld. /Felder COLLN. /subviolaceus n. /Rothschild Bequest B.M. 1939-1/. The holotype of alkamah is the 'male' (recte female) figured by Distant, which has a fuscous hindwing. InBMNH there is a Sumatran female labelled /Type [red]/alkamah Dist. /Sumatra Forbes/RothschildBequest B.M. 1939-1/alkamah Dist. $ Allotype [in script believed to be by Corbet]/. As the holotype is afemale the allotype cannot be another female , and I reject it as such . Fortunately Corbet does not appear tohave published this type selection. Fruhstorfer did not designate a type of manychus. He had no examples from Burma in his collection, butsaid that he had examined examples in BMNH from Pegu and Rangoon. There is a female in BMNHlabelled /Type [red]/Moore Coll. 1908-208 Pegu Magaree/May be taken as type of Allotinus subviolaceus THE MILETINI 15 Fig. 5 Allotinus (Allotinus) subviolaceus subviolaceus C. & R. Felder; Malay Peninsula. Male genitalia. manychus Fruh. [in script believed to be by Corbet]/. As it is likely that Fruhstorfer saw this specimenduring his visit to BMNH I designate it as lectotype. I designate as lectotype of kallikrates a male in BMNH labelled /Type [red]/Type [Fruhstorferorangej/Mindanao Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/subviolaceus kallikrates Fr. [in Fruhstor-fer's hand]/. A female labelled /Type [red]/Type [Fruhstorfer orangej/Mindanao/Mindanao Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/Allotinus subviolaceus kallikrates Fruh. [in Corbet's hand]/ is a paralec-totype. I designate as lectotype of silarus a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Nordborneo Brunei Waterstradt 1890/Fruhstorfer Coll. B.M. 1933-131/Allotinus subviolaceus silarusFruh. [in Corbet's hand]/. A female in BMNH labelled /Type [red]/W. Borneo Sintang 16. III. 10/Fruhstorfer Coll. B.M. 1933-131/subviolaceus silarus Fr. [in Fruhstorfer's hand]/ is labelled paralectotype. DISTRIBUTION. Manipur; Burma; Thailand; West Malaysia; Singapore; Sumatra; Bangka I.; Java; Borneo;Palawan; Mindanao. Allotinus (Allotinus) subviolaceus mirus van Eecke Allotinus subviolaceus mirus van Eecke, 1914: 247; Fruhstorfer, 1916, 808. LECTOTYPE cf , SIMEULUE I.(Simalur) (RNH), here designated [examined]. The subspecies is much paler than the palest dry season examples from Burma. Above, both sexes are pale bluish grey, becoming whitish on the forewing in the middle of space Ib andbasal half of space 2. In the male the forewing border is inwardly rather irregular and narrow, leaving thebasal half of space 3 , the base of space 4 and the lower three-quarters of the cell grey ; it is narrowest at veinCu 2 , where it is just over 2-0 mm wide. The hindwing is mostly grey, with the discocellular veins darkenedand with an inwardly diffuse fuscous marginal and costal border 2-3 mm wide. The female differs as usualin having more extensive bluish grey areas on both wings. I designate as lectotype a male in RNH labelled /60. 49/E. Jacobson Sinabang. Sim. Sum. 7. 1913/Museum Leiden Allotinus subviolaceus mirus Det. R. v. Eecke/Type/. A further male and a femalelabelled /60. 50/ and /60, 51/ respectively, but otherwise as the lectotype, are paralectotypes. DISTRIBUTION. Simeulue I. (Simalur). Allotinus (Allotinus) agnolia sp. n. (Figs 6 cf genitalia and scales; 55 9) Cf forewing length 12-5 mm. Forewing with apex rounded, as in females of the genus; veins M\ and R 5 witha short common stalk and vein R 3 shorter than in the three preceding species, arising only a little before end 16 J. N. ELIOT Fig. 6 Allotinus (Allotinus) agnolia sp. n.: Sumatra. Male genitalia. Lower left, androconial scale fromfore wing vein M 3 ; below, normal cover scale from the same area. of vein R 2 . Upperside dark brown, with basal quarter of forewing a paler grey brown. Forewing with arhomboidal white patch about 3-5 mm broad resting on vein A\ and reaching into lower angle of cell andbase of space 3. Basal quarter of vein A/ 3 rather weakly swollen and clothed with very pale buff specialisedscales about one-sixth size of white cover scales. Underside with a whitish area above mid-dorsum offorewing , but rest of both wings pale buff very densely striated with buff-brown striae ; usual postdiscal andsubmarginal markings of the genus small and only made out with difficulty. 9 . Apart from lacking the swelling of vein M 3 of forewing, similar in all respects to male. MATERIAL EXAMINED Holotype cf , Sumatra: 'Battak Mts., N.E. Sum., II. 94 (Dr. Martin)' (BMNH). Paratypes. Sumatra: 1 $ (allotype), 'VIII. 94', otherwise as holotype (BMNH); 1 9 'N.O. SumatraMartin Coll. H. Fruhstorfer', 'XL 94', 'Fruhstorfer Coll. B.M. 1933-131' (BMNH). The specific name is an anagram of Logania, given because the holotype was found in BMNH among aseries of Logania distanti massalia bearing a manuscript label (in an unknown hand) reading Loganiamassalia Doherty. Allotinus (Allotinus) nicholsi Moulton (Figs 7, cf genitalia; 56 cf ; 57 $)Allotinus nicholsi Moulton, 1911: 83. The species differs from all the preceding in being reddish brown without white or blue areas. On theforewing vein R 3 is as short as in Paragerydus, arising opposite the end of vein R 2 . On the hindwing a weakhumeral vein is present. In the male, vein M 3 of the forewing is weakly swollen in its basal quarter (a factnot noticed by Moulton and Corbet) , and the specialised scales are slightly smaller than those of A . agnolia .The male genitalia possess a feature unique in the tribe, viz. a prominent cornutus in the phallus, consistingof a longer central and shorter outer bundles of minute spicules. The species is known only from Borneo and Sumatra, and appears to be very rare. Provisionally Irecognise two subspecies, but more material may show that no subdivision is necessary. Key to the subspecies of A. (A.) nicholsi 1 Underside markings faint and comparatively broad nicholsi nicholsi (p. 17) - Underside markings narrow, darker and sharply defined nicholsi battakanus (p. 17) THE MILETINI 17 Fig. 7 Allotinus (Allotinus) nicholsi battakanus Fruhstorfer; Sumatra. Male genitalia. Allotinus (Allotinus) nicholsi nicholsi Moulton Allotinus nicholsi Moulton, 1911: 83; Fruhstorfer, 1913: 343; 1916: 809; Corbet, 19396: 66, fig. 11 cT genitalia. Holotype cf , BORNEO: Sarawak (BMNH) [examined].Allotinus nicholsi nicholsi Moulton; Eliot, 1967: 71. The subspecies is known only from the male holotype, which is in worn condition. On the forewing veinsMI and R 5 are just separate. On the underside the markings are rather faint and comparatively wide, andresemble those of A. subviolaceus, as pointed out by Moulton. Allotinus (Allotinus) nicholsi battakanus Fruhstorfer (Figs 7 cf genitalia; 56 cf ; 57 9) Allotinus taros battakanus Fruhstorfer, 1913: 370 (partim); 1916: 813, ? pi. 141g $ (partim). LECTOTYPE Cf , SUMATRA (BMNH), here designated [examined].Allotinus nicholsi battakanus Fruhstorfer; Eliot, 1967: 71. The short series of one male and four females in BMNH have veins M^ and R 5 of the forewing connate , andon the underside the markings are narrower, darker and more sharply defined than in subsp. nicholsi. Although Fruhstorfer drew attention to the absence of the usual forewing brand of the male, part of histype-series, including two female syntypes in BMNH, comprise A. (Fabitaras) sarrastes, which is acommon species in Sumatra. No doubt it was on this account that Fruhstorfer stated that the species wastaken in great numbers by Dr Martin. It is not possible to say whether the figure of a female in Seitzrepresents battakanus or sarrastes. I designate as lectotype the male in BMNH, unfortunately lacking its head, which is labelled /Type[red]/Type [Fruhstorfer orange]/CMB II. 95/Fruhstorfer Coll. B.M. 1933-131/Allotinus battakanus Frhst.[in Fruhstorfer's hand]/. DISTRIBUTION. Known only from the Battak Mts. Allotinus (Allotinus) major C. & R. Felder sp. rev.(Fig. 8, Cf genitalia) Allotinus major C. & R. Felder, [1865]: 286, partim cf nee $, pi. 35, fig. 29 cf. LECTOTYPE cf,SULAWESI (BMNH), here designated [examined]. 18 J- N. ELIOT Allotinus fallax major Felder; Fruhstorfer, 1913: 343; 1916: 809. Allotinus fallax depictus Fruhstorfer, 1913: 343; 1916: 809 [misspelled depista], pi. 141h cf [as major]. LECTOTYPE cf , SULAWESI (BMNH), here designated [examined]. Syn. n.Allotinus kalawarus Ribbe, 1926: 91. Syntypes, SULAWESI (probably in SMT). The most remarkable feature of this species and A. maximus is the short abdomen, which in the male doesnot protrude beyond the hindwings; this is due to all the abdominal segments being relatively shortcompared with those of the other species of the genus. Both species can also be recognised by thearrangement of the postdiscal series on the underside of the hindwing, wherein the spot in space 6 ismidway between those in spaces 7 and 5. A. major shows much individual variation in size and pattern. The male is typically of medium size(forewing 17-0 mm) with a small, sullied, white patch on the forewing astride the base of vein Cu 2 .Occasionally the white area below vein Cu 2 is unsullied and up to 3-0 mm wide; but in more than half of theexamples examined the white patch is entirely absent, and such examples from east and south Sulawesiwere treated as a distinct subspecies, depictus, by Fruhstorfer. However, similar males occur throughoutSulawesi, so that depictus can only be used as a varietal name. Females have, on average, a larger whitepatch than males. In the commonest form the ovate white patch is about 4-0 mm wide and fills the centre ofspace Ib, the basal half of space 2 and may enter space 3. Unmarked brown females appear to be very rare,as there is only one example in BMNH, from west central Sulawesi, without a trace of white. Other femalesfrom the same area and also from east Sulawesi form a complete connecting series leading up to the normalwhite-patched form. I have not seen Ribbe's type-series of A. kalawarus, so cannot be positive that the name is applicable toA. major. Ribbe described his species from at least four examples which he did not sex. He wrote that theupperside agreed with the figure of major (Fruhstorfer, 1916: pi. 141h) which is cf-var depictus and that theunderside was nearest to the figure of pyxus [A. portunus pyxus] (Fruhstorfer, 1916: pi. 141i), but differedin having a light whitish ground colour and heavier marginal flecks and other markings. He queriedwhether kalawarus might not be the same as damodar (= A. macassarensis macassarensis), but said that theonly specimen which he had as damodar was a female with a crenulate hindwing termen (which is indeed acharacter of macassarensis female), whilst in kalawarus the termen was not crenulate. A non-crenulatehindwing is a feature of both sexes of A . major, so that if any of Ribbe's specimens were females this wouldconfirm that they were major. If they were males the absence of any mention of a brand (conspicuous in A .macassarensis) also suggests that they were major. I therefore feel reasonably confident that kalawarus is asynonym of major. Fig. 8 Allotinus (Allotinus) major C. & R. Felder; south-west Sulawesi. Male genitalia. Lower left,uncus, tegumen and brachium of an example from north-east Sulawesi: Minahassa. THE MILETINI 19 In the male genitalia the uncus/tegumen blades are shorter and more rounded than in any other Orientalspecies of Miletini. Judging by the few preparations I have made they are roundest in south Sulawesi andmost elongated in Minahassa, those from east-central Sulawesi being intermediate. I designate as lectotype of major a male in BMNH labelled /Celeb Lorquin type [blue circular]/Allotinusmajor cf Fd./Type [red]/FELDER COLLN. /Rothschild Bequest 1939-1/. The female figured by theFelders is an example of the next species. I designate as lectotype of depictus a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Sud-Celebes Dr. Martin Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/major fa depicta Fr. [in Fruhstor-fer's hand]/. A female labelled/Type [red]/Type [Fruhstorfer orange]/O. Celeb. Fruhstorfer/depictus Fruh.[in Corbet's hand]/ is a paralectotype; it is a typical example with a white patch on the forewing. DISTRIBUTION. The species occurs at low levels throughout Sulawesi and in Banggai I. and Sangihe I. Allotinus (Allotinus) maximus Staudinger stat. n.(Fig. 9, cf genitalia) [Allotinus major C. & R. Felder, [1865]: 286, partim $ nee cf , pi. 35, figs 30, 31 $. Misidentification.]Allotinus albatus Felder var. maximus Staudinger, 1888: 269, pi. 94. Holotype cf, SULAWESI (coll. Staudinger, probably in MNHU).Allotinus fallax major Felder $-f. albatus Felder [= maximus}; Fruhstorfer, 1913: 343; 1916: 809 [misspelled albadus and magnimus}. Both Fruhstorfer and Staudinger confused this species with Allotinus albatus, a distinct species in subgenusParagerydus (p. 30). The former recognised its kinship with A. major, regarding it as a montane varietyfrom Minahassa. I think it is more likely that major and maximus are distinct sibling species. The latter maywell be wholly montane, since of four males in BMNH the only two with complete data came from the Peakof Bonthain, 1000-2000 m. These examples from the extreme south-west of Sulawesi do not appear todiffer from Staudinger's holotype male from Minahassa, so that it is probable that the species occursunchanged throughout the mountains of Sulawesi. A. maximus is larger than A. major, and in the male there is a clear white patch on the forewing filling thebasal half of space 2 and the central half of space 1 b , which is grey at the wing base . The only female I haveseen is the misidentified female syntype of major figured by the Felders, which has a large white patch onthe forewing extending into the base of space 4 and the lower edge of the cell. It is labelled /Type[red]/Celebes Lorquin type [blue circular]/Allotinus major n $ Fd. /major n/FELDER COLLN. /Roth-schild Bequest 1939-1/. The male genitalia (Fig. 9) of the single example dissected differ slightly from those of A. major in havinga narrower, more pointed phallus and longer, more rectangular uncus blades. DISTRIBUTION. Confined to Sulawesi. Fig. 9 Allotinus (Allotinus) maximus Staudinger; Sulawesi (no further data). Male genitalia. 20 J. N. ELIOT FABITARAS subgen. n. Type-species: Paragerydus fabius Distant & Pryer, 1887: 266. Gender masculine. Antenna nearly two-thirds length of forewing costa, with about 62-66 segments. In males, abdominal hairtufts short, not protruding except when genitalia are everted. Forewing venation particularly subject toindividual variation, with apical angle of cell sometimes sharply acute, sometimes almost a right angle.Veins M l and R 5 usually stalked, but sometimes connate from cell apex, most often in A. taras and A.strigatus, while in A. bldiensis they may be just separate. Vein RT, also very variable, usually arising fromabout middle of vein R 5 , but sometimes as short as in subgen. Paragerydus, sometimes as long as in typicalspecies of subgen. Allotinus. In males vein M 3 always basally swollen and clothed with small specialisedscales. Hindwing usually with a weakly developed humeral vein extending less than half-way across space8, but sometimes absent or vestigial, most often in females. Underside characterised by a series ofsubmarginal blackish dots which, on the forewing, are outwardly white-edged in females but not, or moreindistinctly so, in males; indeed, even the blackish dots may be absent in males of A. strigatus and A.portunus. Male genitalia not unlike those of subgen. Allotinus, with valva ending in an apical hook butedges folded inwards more strongly in the form of a trough. The species included in Fabitaras were placed in Paragerydus by earlier authors up to and includingFruhstorfer (1916) because of the swollen vein M 3 of males and the common origin of veins M } and R 5 inboth sexes; but on the basis of the male genitalia and the frequent presence of a humeral vein it is certainthat Fabitaras is more closely allied to Allotinus than to Paragerydus. The subgenus is found from central Burma to Sundaland and the Philippines, and comprises ten species. Key to the species of subgenus Fabitaras 1 Underside of hindwing with postdiscal spot in space 6 placed below, or just inside, that in space 7. Female with hindwing produced at vein M 3 (faWus-group) 2 Underside of hindwing with postdiscal spot in space 6 placed well outside spot in space 7. Female hindwing rounded (faras-group) 4 2 Underside of forewing with postdiscal series more or less parallel to termen except that spot in space 5 is moved a little outwards. Postdiscal spots above vein M 3 on forewing and vein M\ on hindwing not enlarged nor blotchy 3 Underside of forewing with postdiscal series inclined towards apex; spots in spaces 4 and 5moved out of line, the latter very close to termen. Postdiscal spots above vein M 3 on forewingand vein MI on hindwing enlarged and blotchy fabius (p. 21) 3 cf upperside of forewing with vein M 3 swollen for three-quarters of its length; visual brand prominent. Underside of forewing with apical region shaded reddish brown; hindwing withsubmarginal black spot in space 6 at most barely larger than other submarginal spots borneensis (p. 22) Cf upperside of forewing with vein M 3 swollen for half its length; visual brand narrow andobscure. Underside of forewing with apical region not shaded reddish brown; hindwing withsubmarginal black spot in space 6 triangular and larger than other submarginal spots punctatus (p. 23) 4 Cf 9 upperside of hindwing with tornal area brown 5 Cf 9 upperside of hindwing with tornal quarter white brooksi (p. 26) 5 Underside of hindwing with submarginal black spot in space 6 at most barely larger than other submarginal spots, cf upperside of forewing with at least indications of a visual brand astride swollen portion of vein M 3 6 Underside of hindwing with submarginal black spot in space 6 triangular and larger than othersubmarginal spots, cf upperside of forewing without a visual brand. 9 upperside of hindwingmay have a brownish cream border nigritus (p. 24) 6 Underside of hindwing with postdiscal spot in space 6 placed roughly midway between the spots in-spaces 7 and 5 7 Underside of hindwing with postdiscal spot in space 6 placed much further from spot in space 7so that it is almost on an even arc with the spots in spaces 2 to 5 8 7 Underside ground colour white, cf forewing with vein M 3 swollen for only one-third its length; visual brand ill-defined and less than 1 -0 mm wide bidiensis (p. 26) Underside ground colour pale buff, cf forewing with swollen portion of vein M 3 about half its length; visual brand about 2-0 mm wide strigatus (p. 24) 8 cf forewing with vein M 3 swollen for more than half its length ; visual brand at least 1-0 mm wide 9 THE MILETINI 21 O" forewing with vein M 3 swollen for half its length; visual brand very narrow, inconspicuous. Underside ground colour greyish white ;forewing apex shaded reddish brown turns (p. 27) 9 Underside ground colour greyish white , not darkened towards forewing apex; black submargin-al spots outwardly white-edged; postdiscal series slightly closer to termen at dorsum than atcosta. cf upperside of forewing with vein M 3 swollen for two-thirds its length; visual brand up to 2-0 mm wide, comparatively well defined sarrastes (p. 28) - Underside ground colour in cf buff, becoming darker towards forewing apex; striation dense,black marginal spots absent or obscure, not white-edged. $ ground colour greyish white withforewing apex more or less shaded with brown; submarginal white-edged black spots presenton forewing. Postdiscal series parallel to termen. cf upperside of forewing with vein M 3swollen for nearly four-fifths its length; visual brand ill-defined, just over 1-0 mm wide portunus (p. 29) Allotinus (Fabitaras) fabius (Distant & Pryer) (Fig. 10, cf genitalia)Paragerydus fabius Distant & Pryer, 1887: 266. The species is readily recognised by the underside markings. On the forewing the postdiscal spots in spaces4 and especially 5 are shifted towards the termen, whilst the spots in spaces 6 and 7 are placed much furtherbasad; in addition the spots above vein M 3 are more or less enlarged and blotchy. On the hindwing thepostdiscal spots in spaces 6 and 7 are placed one above the other and are large and blotchy. The female hasthe hindwing weakly caudate at vein M 3 . The male has vein M 3 of the forewing swollen for aboutthree-quarters of its length and the visual brand is about 1-5 mm wide. The species has a restricted distribution in Sundaland, and has not been found in Java, Palawan nor theislands off the west coast of Sumatra. There are two subspecies. Key to the subspecies of A. (F.) fabius 1 $ upperside of hindwing outwardly white fabius fabius (p. 21) $ upperside of hindwing brown fabius arrius (p. 21) Allotinus (Fabitaras) fabius fabius (Distant & Pryer) Paragerydus fabius Distant & Pryer, 1887: 266; Cowan, 1966: 5. Holotype $ , BORNEO: Sandakan (BMNH) [examined].Allotinus caudatus Grose-Smith, 1893: 34. Holotype 'cf' recte $, BORNEO: Mt Kina Balu (BMNH) [examined]. Syn. n. Paragerydus caudatus (Grose-Smith); H. H. Druce, 1895: 563, pi. 31, figs 7, 8 cf -Allotinus fabius fabius (Distant & Pryer); Fruhstorfer, 1914: 22; 1916: 814; Corbet, 19396: 74.Allotinus fabius caudatus Smith; Fruhstorfer, 1914: 22; 1916: 814; Corbet, 19396: 74.Allotinus fabius pamisus Fruhstorfer, 1914: 22; 1916: 814; Corbet, 19396: 74 [misspelt pamiscus]. Holotype $, BORNEO: south-east (BMNH) [examined]. Syn. n. Fruhstorfer (1915) perpetuated Grose-Smith's mistake over the sex of caudatus by incorrectly stating thatboth sexes of the race from Kina Balu have the outer part of the hindwing white. In fact it is only femaleswhich have the hindwing partly white, and examples from Kina Balu do not differ significantly from thosefrom the rest of Borneo nor from examples from south-west Sumatra. DISTRIBUTION. Borneo, including Pulo Laut; south Sumatra (1 cf , 3 $ , Lebong Tandai (C. J. Brooks)). Allotinus (Fabitaras) fabius arrius Fruhstorfer(Fig. 10, cf genitalia) [Paragerydus panormis Elwes, 1893: 619 (partim), pi. 43, fig. 9 $. Misidentification.] Allotinus fabius arrius Fruhstorfer, 1914: 22; 1916: 814, pi. 1411, cf $; Corbet, 19396: 74, fig. 4 cf genitalia; Fleming, 1975: 21, pi. 58, fig. L45 cf; Eliot, 1978: 240. LECTOTYPE cf. SUMATRA (BMNH), here designated [examined].[Allotinus fabius panormis (Elwes); Fruhstorfer, 1916: 814. Misidentification.] 22 J. N. ELIOT Fig. 10 Allotinus (Fabitaras) fabius arrius (Fruhstorfer); Malay Peninsula. Male genitalia. Differs from the nominate subspecies only in the all brown female. I designate as lectotype of arrius a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/CMBII. 94/Sumatra Monies Batlak ex coll. Fruhslorfer/Fruhstorfer Coll. B.M. 1933-131/fabius arrius Frhsl. [inFruhslorfer's hand]/. A similarly labelled female is a paraleclolype. DISTRIBUTION. Malay Peninsula; north Sumatra. Allotinus (Fabitaras) borneensis Moulton(Figs 11 Cf genitalia; 58 cf) Allotinus borneensis Moullon, 1911: 81; Fruhslorfer, 1913: 371; 1916: 814. Hololype cf , BORNEO: Sarawak (BMNH) [examined]. Allotinus borneensis borneensis Moullon; Corbel, 19396: 76, fig. 3, cf genitalia.Allotinus borneensis elioti Corbet, 19396: 76; Fleming, 1975: 22, pi. 58, fig. L49 cf ; Eliol, 1978: 240. Hololype cf , WEST MALAYSIA (BMNH) [examined]. Syn. n. On Ihe upperside bolh sexes resemble A. fabius arrius, but are more reddish brown and, in Ihe male, Ihe Fig. 11 Allotinus (Fabitaras) borneensis Moulton; Malay Peninsula. Male genitalia. THE MILETINI 23 swollen portion of vein M 3 is very slightly longer and the visual brand is very slightly wider. The underside isgreyish white to pale buff-white, with the apical area of the forewing shaded with reddish brown. On theforewing the postdiscal series is parallel to the termen except that the spot in space 5 is shifted outwards, butnot nearly as much as in A. fabius. On the hindwing the spot in space 6 is usually placed just inside that inspace 7. DISTRIBUTION. Borneo; Malay Peninsula; Sumatra; Bangka. Allotinus (Fabitaras) punctatus (Semper)(Figs 12, cf genitalia; 59 9) Paragerydus punctatus Semper, 1889: 165, pi. 31, figs 16 cf, 17 $. LECTOTYPE cf, PHILIPPINES:Mindanao (SM), here designated [examined]. Allotinus punctatus (Semper); Fruhstorfer, 1913: 371; 1916: 814; Corbet, 1939b: 74, fig. 1, cf genitalia. Allotinus anaxandridas Fruhstorfer, 1916: 814. Holotype 9, PHILIPPINES: Mindanao (SM) [examined].[$-morph.] Syn. n. Allotinus caesemius Fruhstorfer, 1916: 814. LECTOTYPE $, PHILIPPINES: Mindanao (SM), here desig-nated [examined]. [$-morph.] Syn. n. On the upperside the male has vein M 3 swollen for half its length and the visual brand is rather narrow andill-defined. The underside is pale buff with the postdiscal markings showing great individual variation inwidth, in some examples being as much as 2-0 mm wide. In females the ground colour is greyish white, andthe postdiscal markings are much narrower, seldom as much as 1-0 mm in width. The white-edged blacksubmarginal spots are well marked, that in space 6 of the hindwing being triangular and larger than theothers. As in A. fabius and A. borneensis the postdiscal spots in spaces 7 and 6 of the hindwing are placedmore or less one above the other, and the female has the hindwing weakly caudate at vein M 3 . The species is chiefly remarkable for the polymorphism of the female, which was recognised by Semperwho described three morphs. Fruhstorfer, however, considered that the three morphs were distinct species- a view for which there is no supporting evidence. The typical female is plain brown on both wings. In $-f.caesemius Fruhstorfer the forewing has a large white patch while the hindwing is brown; it seems to be thecommonest morph. In $-f. anaxandridas Fruhstorfer the forewing resembles that of caesemius but thehindwing also has a large white patch; it appears to be much the rarest morph. Another 9 -morph wasdescribed by Fruhstorfer as 9-f- eretria, but I think that the butterfly in question was a female of A. nigritus(seep. 24). Semper did not designate a type of punctatus, whilst Fruhstorfer named caesemius from two females andanaxandridas from a single female in Semper coll., which Semper had designated merely as varieties ofpunctatus. I now designate as lectotype of punctatus one of two males ex Semper coll. in SM labelled /Coll.C. Semper/Sibulan/215/Parag. punctatus typ. Semper/297c/Typus [red]/. A second male labelled /215/Coll.C. Semper/ is a paralectotype. I designate as lectotype of caesemius a female ex Semper coll. in SM labelled/Sibulan/297a/215/. The single female of anaxandridas is automatically the holotype and has been solabelled. Fig. 12 Allotinus (Fabitaras) punctatus (Semper); Mindanao. Male genitalia. 24 J. N. ELIOT DISTRIBUTION. The species has hitherto been recorded only from Mindanao. There is a single female in coll.Treadaway from Leyte (Catmon 450 m, 10. v. 1977) which differs from f. caesemius in having the whiteforewing patch reduced, with the dusky scaling at the wing base and above the dorsum almost as darkbrown as the marginal and costal borders, whilst the underside is yellowish white with the markings heavierand more reddish than in females from Mindanao. It is probable that a distinct subspecies flies in Leyte, butI hesitate to name one on the basis of a single female. Allotinus (Fabitaras) nigritus (Semper)(Figs 13, Cf genitalia, 60 $) Paragerydus nigritus Semper, 1889: 164, pi. 31, fig. 15 cf . LECTOTYPE cf , PHILIPPINES: Mindanao (SM), here designated [examined]. Allotinus nigritus (Semper); Fruhstorfer, 1914: 22; 1916: 814.Allotinus punctatus (Semper) $-f. eretria Fruhstorfer, 1916: 814. Holotype $. PHILIPPINES: Mindanao (not located). In the male vein M 3 is swollen for a little under half its length and there is no visual brand. The underside ispale brownish ochreous with darker brown markings. The black submarginal spots resemble those of A.punctatus, from which A. nigritus can be separated by the postdiscal spot in space 6 of the hindwing beingplaced midway between the spots in spaces 7 and 5, or a little closer to the latter, as well as by its richer,more ochreous appearance. The only female I have seen (Mindanao, Mt Apo, 15. ii. 1983, (A. Ballantine)) is dark brown above withan inwardly diffuse, sullied whitish border 2-0 mm wide on the hindwing from the tornus to vein R 5 . Thehindwing termen is barely dentate at the end of vein M 3 , as in all species of the subgenus dealt withsubsequently. Fruhstorfer named as A. punctatus $ -f . eretria a female with the underside bright ochreous bearing thickbrown markings and black marginal spots. The underside of females of A. punctatus is greyish white, sothat I suspect that eretria really applies to A nigritus. Fruhstorfer did not mention the upperside of eretria,which was presumably unmarked brown, as in the typical female morph of A. punctatus, and if mysupposition of the true identity of eretria is correct it would appear that the female of A. nigritus isdimorphic - with or without a whitish border on the hindwing. I designate as lectotype of nigritus a male in SM labelled /Coll. C.Semper/Ost Mind./214/297b/15/Parag.nigritus typ. Semper/Typus [red]/. A second male labelled /Ost Mind./214/Typus [red]/is a paralectotype. DISTRIBUTION. Mindanao. Fig. 13 Allotinus (Fabitaras) nigritus (Semper); Mindanao. Male genitalia. Allotinus (Fabitaras) strigatus Moulton (Figs 14, cf genitalia; 61 cf )Allotinus strigatus Moulton, 1911: 80.The species can be recognised by the fact that on the underside of the hindwing the postdiscal spot in space THE MILETINI 25 6 is placed midway between those in spaces 7 and 5, while on the forewing the postdiscal series is muchcloser to the termen near the tornus than near the costa. The underside ground colour is a uniform palebuff. On the upperside the male has vein A/ 3 swollen for half its length, and the visual brand is quadrate,about 2 mm wide and not sharply outlined. The species is strictly Sundanian, and has not yet been found in Java, Palawan or the islands off the westcoast of Sumatra. There are two subspecies. Key to the subspecies of A. (F.) strigatus 1 On the underside of the forewing the blackish submarginal dots are inconspicuous and not outwardly white-edged strigatus strigaius (p. 25) - On the underside of the forewing the blackish submarginal dots are outwardly white-edged in $ and in the apical half of the wing in cf strigatus malayanus (p. 25) Allotinus (Fabitaras) strigatus strigatus Moulton(Fig. 61 cf) Allotinus strigatus Moulton, 1911: 80. Holotype cf , BORNEO: Pulo Laut (BMNH) [examined].Allotinus strigatus strigatus Moulton; Fruhstorfer, 1914: 22; 1916: 813; Corbet, 1939ft: 75, fig. 5, cfgenitalia. On the underside the submarginal blackish dots are not white-edged, and the usual markings arecomparatively broad and well defined. DISTRIBUTION. Borneo, including Pulo Laut. Allotinus (Fabitaras) strigatus malayanus Corbet(Fig. 14, cf genitalia) Allotinus strigatus malayanus Corbet, 1939ft: 75; Fleming, 1975: 22, pi. 58, fig. L48 cf ; Eliot, 1978: 240. Holotype cf , WEST MALAYSIA (BMNH) [examined].Allotinus strigatus denalus Corbet, 1939ft: 75. Holotype cf , SUMATRA: Battak Mts (BMNH) [examined]. Syn. n. On the underside of the forewing the blackish submarginal dots are outwardly edged with whitish in thefemale, but in the male less distinctly and only in the apical half of the wing. The markings are narrower andless well defined than in the nominate subspecies. In the male the swelling of vein M 3 is slightly shorter,being just under half its length, and the visual brand is usually a little narrower. DISTRIBUTION. Malay Peninsula, including Singapore; Sumatra. Fig. 14 Allotinus (Fabitaras) strigatus malayanus Corbet; Malay Peninsula. Male genitalia. 26 J. N. ELIOT Allotinus (Fabitaras) brooksisp. n. (Figs 15, cf genitalia; 63 cf ) Cf forewing length 15-0 mm. Venation normal for the subgenus, with veins M j and R 5 of forewing having amoderately long common stalk. Upperside brown, with tornal quarter of hindwing white, very sparselydusted with brown scales and with adjoining cilia white chequered with dark brown at vein endings.Forewing with vein M 3 swollen for slightly less than one-third of its length; visual brand very narrow,obscure. Underside very pale greyish white, sparsely striated; postdiscal spots small , more or less parallelto termen; hindwing with postdiscal spot in space 6 a little closer to that in space 7 than to that in space 5;cell-end bars unusually heavy, especially on hindwing; forewing with submarginal blackish dots outwardlywhite-edged. Genitalia similar to those of A. strigatus, A. taras and A. sarrastes, but apical process of valvaslightly larger. $ forewing length 15-0 mm. Differs from male in that inwardly diffuse white area on hindwing a littlelarger, occupying one-third of wing. At first sight this sex might be mistaken for A. fabius fabius , but therounded hindwing termen separates it readily. MATERIAL EXAMINED Holotype cf , Borneo: 'Bau Feb Feb 10', 'Sarawak C. J. Brooks', '192', 'C. J. Brooks Bequest. B.M.1953-173' (BMNH). Paratype. Borneo: 1 $ (allotype), 'Bau Dec. 09', '192', 'C. J. Brooks Bequest B.M. 1953-173' (BMNH). Fig. 15 Allotinus (Fabitaras) brooksi sp. n. ; Borneo. Male genitalia. Allotinus (Fabitaras) bidiensissp. n. (Figs 16, cf genitalia; 62 cf ) Cf forewing length 15-0-16-0 mm. In the two examples examined forewing venation differs from rest ofsubgenus in that veins MI and R 5 are just separate at their origins; vein /? 3 rather short, arising just beforeend of vein R 2 . Upperside brown, with basal third of vein M 3 swollen; visual brand narrow, obscure.Underside with markings generally arranged as in sympatric A. strigatus, but differing in the pale greyishwhite ground colour, the smaller, more rounded postdiscal spots and the prominently white-edged blackishsubmarginal dots on the forewing. Male genitalia distinguished by valva, in which the lower processextends beyond the short apical hook, and comparatively short, stout phallus. 9 forewing length 14-0 mm in the single example seen, wherein veins MI and R 5 of forewing connatefrom cell apex. Upperside brown. Underside similar to male. MATERIAL EXAMINED Holotype cf , Borneo: Sarawak, '647 20/5/08', '190f , 'C. J. Brooks Bequest. B.M. 1953-173' (BMNH). Paratypes. Borneo: 1 $ (allotype), Sabah, 'N. Born.', 'Madai 3.2.92', 'Joicey Bequest. Brit. Mus.1934-120' (BMNH); cf , 'Bidi Sarawak 1907 C. J. Brooks', '190f , 'C. J. Brooks Bequest. B.M. 1953-173'(BMNH). THE MILETINI 27 Fig. 16 Allotinus (Fabitaras) bidiensis sp. n. ; Borneo. Male genitalia. Allotinus (Fabitaras) taras (Doherty) (Fig. 17, Cf genitalia) Paragerydus taras Doherty, 1889: 437, pi. 23, fig. 10 cf . Syntypes, BURMA: Tenasserim (not located).Parageryduspanormis Elwes, 1893: 619, partim 'cf ' recte , nee $ , pi. 43, fig 8 'cf '. Holotype 'cf ' recte $ , BURMA: East Pegu (BMNH) [examined]. Syn. n.Allotinus taras (Doherty) Bingham, 1907: 300; Swinhoe, 1910: 199, pi. 617, figs 2, 2b cf , 2a, 2c $; Evans, 1932: 212; Cantlie, 1963: 27.Allotinus panormis (Elwes) Bingham, 1907: 301; Swinhoe, 1910: 197 partim, pi. 616, figs 3, 3b 'cf' recte $, nee $. Allotinus taras taras (Doherty); Fruhstorfer, 1913: 370; 1916: 813; Corbet, 19396: 74, partim.Allotinus fabius panormis (Elwes); Fruhstorfer, 1916: 814, partim; Evans, 1932: 212.Allotinus panormis panormis (Elwes); Corbet, 19396: 74; Cantlie, 1963: 27. On the upper surface of the forewing the male has vein M 3 swollen for half its length, and the visual brand isvery narrow and inconspicuous; indeed Doherty, in his original description, gave absence of the brand as Fig. 17 Allotinus (Fabitaras) taras (Doherty); Burma. Male genitalia. 28 J. N. ELIOT one of the characters of the species. The underside of both sexes is greyish white turning to reddish browntowards the forewing apex and, occasionally, at the hindwing apex also. The postdiscal series of spots israther lightly marked, and though these spots were mentioned by Doherty they are wanting in the examplechosen for his figure. Such examples cannot be regarded as typical, only two out of 28 males in BMNHbeing similar to Doherty's figure. The 'male' (recte female) of Paragerydus panormis is a quite typical female of A taras, and I am at a lossto understand why Corbet regarded it as a different species conspecific with A. portunus. The male genitalia of A. taras and A. sarrastes are similar, so there are grounds for regarding these twotaxa as conspecific. However, they overlap in south Burma between Tavoy and Mergui over a distance ofsome 160 km without any evidence of intergradation, so that it seems highly probable that their period ofisolation, when there was a sea barrier in the region of the Isthmus of Kra, was sufficiently prolonged forinterbreeding to be impossible when they again met. I have seen none of Doherty's type-series and as the figured male is atypical I do not designate it aslectotype. DISTRIBUTION. Burma, from the Karen Hills to Mergui. Allotinus (Fabitaras) sarrastes Fruhstorfer stat. n.(Fig. 18, cf genitalia) Allotinus taras sarrastes Fruhstorfer, 1913: 370; 1916: 813; Corbet, 19396: 74; Eliot, 1961: 71; 1978: 240; Fleming, 1975: 21, pi. 58, fig. L46 cf . LECTOTYPE cf , BORNEO (BMNH), here designated [examined].[Allotinus taras battakanus Fruhstorfer, 1913: 370, partim; 1916: 813 partim, ? pi. 141g $. Misidentifica- tion.] Allotinus porriginosus Toxopeus, 1932: Ixxvii. Holotype cf , JAVA: south (not located). Syn. n.[Allotinus taras taras (Doherty); Corbet, 19396: 74, fig. 2, cf genitalia. Misidentification.]Allotinus taras mendava Riley, 1944: 253, pi. 2, fig. 30 cf, 31 $. Holotype cf, MENTAWAI Is.: Sipora (BMNH) [examined]. Syn. n. This Sundanian species is obviously very closely related to A. taras, but it is best regarded as specificallydistinct. It differs from A. taras on the underside by the absence of reddish brown shading towards theforewing apex. In addition, on the upperside of the male forewing the swelling of vein M 3 extends overabout two-thirds of its length and the visual brand is prominent and usually about 1-5 mm wide. On theunderside the white edges to the blackish submarginal spots are particularly well marked, especially in thefemale. I designate as lectotype a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Kina Balu excoll. H. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/sarrastes Frhst. [in Fruhstorfer's hand]/. A femalelabelled /Type [red]/Type [Fruhstorfer orange]/Nord-Borneo Brunei Waterstradt 1890/Fruhstorfer Coll.B.M. 1933-131/sarrastes Frhst. [in Fruhstorfer's hand]/ is a paralectotype. Fig. 18 Allotinus (Fabitaras) sarrastes Fruhstorfer; Malay Peninsula. Male genitalia. THE MILETINI 29 DISTRIBUTION. Burma, from Tavoy southwards; Malay Peninsula; Sumatra; Mentawai Is.; Borneo; Java;Mindanao (SM, 1 cf labelled /Ost Mind./298a/213/Coll. C. Semper/re verdini/; this specimen has the visualbrand obscure and ill-defined, but this may be due partly to its rubbed condition.) Allotinus (Fabitaras) portunus (de Niceville) (Fig. 19, cf genitalia)Paragerydus portunus de Niceville, 1894: 27. Males are easily recognised by the long swelling of vein M 3 of the forewing, extending over nearlyfour-fifths of its length, and the rather diffuse and narrow visual brand, as well as by the buff or pale rufousunderside on which the usual blackish submarginal spots are absent or, if present, are not outwardlywhite-edged. Females most nearly resemble that sex of A. taras on the underside, but the subapicaldarkening on the forewing is less developed and brownish ochreous without the red tinge of taras. In bothsexes the postdiscal series of spots on the forewing is more exactly parallel to the termen than in the otherspecies of the subgenus. As already pointed out (p. 28) Corbet (19396) subordinated A. portunus and its subspecies under A.panormis, which is a synonym of the Burmese species A. taras (Doherty). The species is purely Sundanian, not extending to Palawan or the islands off the west coast of Sumatra,and, at least in the Malay Peninsula, flies at higher average elevations than the other species of thesubgenus, being seldom found below 800 m. I recognise three weak subspecies based on a mean of differences. Key to the subspecies of A. (F.) portunus 1 cf underside of forewing with blackish submarginal spots normally absent 2 - cf underside of forewing with blackish submarginal spots normally present... portunus pyxus (p. 30) 2 cf underside of forewing with postdiscal series ill-marked. $ underside usually pale greyish buff, with forewing apex shaded with ochreous portunus portunus (p. 29) Cf underside of forewing with postdiscal series narrow but usually clearly defined. $ undersidegreyish white with forewing apex only narrowly and faintly shaded portunus maitus (p. 29) Allotinus (Fabitaras) portunus portunus (de Niceville) Paragerydus portunus de Niceville, 1894: 27, pi. 5, fig. 14 cf . Syntypes, JAVA (?ZSI). Allotinus taras narsares Fruhstorfer, 1913: 370; 1916: 813; Corbet, 19396: 74, partim. LECTOTYPE $, JAVA (BMNH), here designated [examined]. Syn. n. Allotinus portunus portunus (de Niceville) Fruhstorfer, 1914: 22; 1916: 813.[Allotinus strigatus dositheus Fruhstorfer, 1914: 22 partim, $ nee cf .][Allotinus taras (Doherty); Piepers & Snellen, 1918: 14, pi. 19, fig. 16 $. Misidentification.]Allotinus portunus (de Niceville); Piepers & Snellen, 1918: 15, pi. 20, figs 18a cf , 18b $.Allotinus panormis portunus (de Niceville); Corbet, 19396: 75. On the underside the ground colour is variable in both sexes. In the male it varies from pale buff to rufousbuff, and generally the postdiscal markings are ill-defined and may be absent as in the example figured byde Niceville. In the female the ground colour may be as buff as in the male, but in some examples, as in thelectotype of narsares, which apparently was the model for the figure of A. taras in Piepers & Snellen (1918),it is pale greyish. The blackish submarginal spots on the underside of the forewing are usually missing in themale but present in the female and at most only weakly edged with white. I designate as lectotype of narsares a female in BMNH labelled /Type [red]/Java Occident. Sukabumi2000' ex coll. H. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/narsares Frhst. [in Fruhstorfer's hand]/. DISTRIBUTION. Java. Allotinus (Fabitaras) portunus maitus Fruhstorfer(Fig. 19, Cf genitalia) Allotinus portunus maitus Fruhstorfer, 1914: 21; 1916: 813. LECTOTYPE cf, SUMATRA (BMNH), heredesignated [examined]. 30 J. N. ELIOT Fig. 19 Allotinus (Fabitaras) portunus maitus Fruhstorfer; Malay Peninsula. Male genitalia. Allotinus panormis fruhstorferi Corbet, 19396: 74, fig. 7, cf genitalia; Fleming, 1975: 22, pi. 58, fig. L47 $ ; Eliot, 1978: 240. Holotype cf , WEST MALAYSIA (BMNH) [examined]. Syn. n.Allotinus panormis maitus Fruhstorfer; Corbet, 19396: 74. In the male the postdiscal markings on the underside are usually more clearly defined than in subsp.portunus and the submarginal blackish spots are missing; in the female the ground colour is more greyishwhite. I designate as lectotype of maitus a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/CMB111. 94/Fruhstorfer Coll. B.M. 1933-131/portunus maitus Frhst. [in Fruhstorfer's hand]/. A female labelled/Type [red]/Type [Fruhstorfer orange]/Sumatra Monies Battak ex coll. H. Fruhstorfer/Fruhstorfer Coll.B.M. 1933-131/allotype Allotinus portunus maitus Fruh. [in Corbet's hand]/ is a paralectotype. DISTRIBUTION. Sumatra; West Malaysia. Allotinus (Fabitaras) portunus pyxus (de Niceville) Paragerydus pyxus de Niceville, 1894: 27, pi. 5, fig. 2 cf . Holotype cf , BORNEO: Kina Balu (? ZSI).Paragerydus waterstradti H. H. Druce, 1895: 562, pi. 31, figs 1 cf, 2 $. LECTOTYPE cf, BORNEO (BMNH), here designated [examined]. [Synonymised by Fruhstorfer, 1914: 22.]Paragerydus waterstradti ab. absens H. H. Druce, 1895: 562. Allotinus portunus pyxus (de Niceville); Fruhstorfer, 1914: 22; 1916: 813, pi. 141i, cf $.Allotinus panormis pyxus (de Niceville); Corbet, 19396: 75. Females appear to be inseparable from the nominate subspecies but in males the blackish submarginalspots on the underside of the fore wing are more often present. De Niceville, who named the subspecies from a single male, differentiated it from subsp. portunus by themore rufous tone of the upperside and the pale rufous instead of pale ochreous colour of the underside.However, the ground colour is very variable in all three subspecies and is not a reliable character. DeNiceville's figure is misleading, as it does not show the visual brand which is mentioned in his description. H. H. Druce described waterstradti from syntypes in coll. Staudinger and in his own collection. Idesignate as lectotype a male in BMNH labelled /Kina Balu Waterstr./P. waterstradti co-type H. H.Druce/ex coll. Hamilton Druce/Joicey Bequest Brit. Mus. 1934-120/. As there is no specimen in BMNH ofab. absens H. H. Druce, it is likely that the name applies to a specimen in coll. Staudinger. DISTRIBUTION. Borneo, apparently only known from Mt Kina Balu. Subgenus PARAGERYDUS Distant Paragerydus Distant, 1884: 207. Type-species: Miletus horsfieldi Moore, 1857: 19, pi. la, fig. 1, by designation of Kirby, [1885]: 191.Miletographa Rober, 1892: 277. Type-species: Miletus drumila Moore, [1866]: 777, pi. 41, fig. 12, by monotypy. [Synonymised by Fruhstorfer, 1913: 371.] THE MILETINI 31 Antenna a little over half length of costa, slightly longer than in subgen. Allotinus and shorter than insubgen. Fabitaras. Antennal segments number about 40 to 60. Shaft segments of the smaller speciesrelatively longer, consequently fewer in number; thus in the nivalis-group and in A. corbeti there are about40 segments, whereas in the larger species, such as A. horsfieldi and A. apries, there are about 58 segments.Species of intermediate size have an intermediate number of segments, for example in A. unlcolor thereare usually about 45. In dwarf individuals, occurring most frequently in A. horsfieldi and A. unicolor, thereis no diminution in the number of segments. Venation shows less individual variation than in Fabitaras.Hindwing without humeral vein. Forewing veins MI and R 5 usually connate or briefly stalked, but in thenivalis-group they may be just separate at their origins. In males, vein M 3 of forewing basally swollen,clothed with small specialised scales, those of the nivalis-group being relatively large. In males, abdominalhair tufts prominent, permanently extruded and, in the genitalia, the valva has a truncate apex, except inA. davidis, and its ventral edge prolonged into a more or less pointed process. With only a few exceptions,the interspecific differences in the valva are very slight and perhaps inconstant; in general the genitalia givelittle assistance in identification. Early authors, up to Fruhstorfer (1913; 1916), used Paragerydus in a wider sense, either as a genus orsubgenus (Artengruppe) to include the species here placed in Fabitaras. The subgenus ranges from north India to Sundaland, the Lesser Sunda Is., Philippines and Sulawesi(including Sula Is.). It comprises 16 species. Key to the species of subgenus Paragerydus 1 Underside mottled with brown specks and striae; postdiscal series not outlined by darker lines nor catenulate. Smaller, forewing 9-0-23-0 mm 2 - Underside mottled with brown spots ringed with pale buff; postdiscal series catenulate and outlined by darker lines. Larger, forewing usually over 23-0 mm drum Ha (p. 56) 2 Underside of forewing without a white fleck at end of vein R 5 (horsfieldi-group) 3 - Underside of forewing with a white fleck at end of vein R 5 (nivalis-group) 14 3 Underside with postdiscal series more or less the same size throughout. Upperside of hindwing unmarked brown 4 Underside with postdiscal spots below vein M 3 on forewing much larger than those above, andspots above vein MI on hindwing much larger than those below. Upperside of hindwingpartly white in cf $ or in $ alone 13 4 Upperside of forewing brown; white patch or streak present in 9 of two species. Hindwing brown, of same shade as forewing; termen crenulate (except in A . apries) 5 Upperside of forewing white to wing base, with dark brown margin and costa. Hindwinguniform pale buff-brown; termen not crenulate parapus (p. 51) 5 $ upperside brown, cf with visual brand, if present, not extending into cell basad of origin of vein Cui 6 - $ upperside of forewing with white patch or streak, cf (where known) with visual brand extending obscurely into cell basad of origin of vein Cu 2 samarensis (p. 37) 6 Underside of forewing with postdiscal spots in spaces 4, 5 and 6 more or less in line and equidistant , or with the spot in space 5 nearer to , and often touching, the spot in space 6 7 Underside of forewing with postdiscal spots in spaces 4, 5 and 6 on an uneven curve, that inspace 5 overlapping that in space 4 and well separated from that in space 6 macassarensis (p. 39) 7 Forewing veins MI and R 5 connate or with a common stalk not more than 1-0 mm long. Underside of forewing with postdiscal spot in space 2 nearer termen than the spots in spaces 3and Ib (if present), cf $ with hindwing cilia elongated into tufts at vein endings, termencrenulate in $ - Forewing veins MI and R 5 with a stalk more than 1-0 mm long. Underside of forewing with postdiscal spots in spaces 3, 2 and Ib (if present) in line and parallel to termen. cf $ withhindwing cilia barely longer at vein endings and termen not crenulate in $ apries (p. 42) 8 cf upperside of forewing with a visual brand , vein M 3 swollen for about half its length or longer 9Cf upperside of forewing without a visual brand, vein M 3 swollen for only one-quarter of its length.Small, forewing 10-0-12-0 mm corbeti (p. 44) 9 Forewing veins M l and R 5 usually with a very short common stalk, cf valva with terminal process not rising above costa ; tip of uncus not , or only a little , produced 10 - Forewing veins M 1 and R 5 usually connate, cf valva with terminal process curved upwards, ending just above costa; tip of uncus produced , rostriform horsfieldi (p. 32) 32 J. N. ELIOT 10 cf forewing with vein A/ 3 swollen for at least three-fifths of its length; visual brand comparative- ly wide 11 - cf forewing with vein M 3 swollen for only half its length or slightly over; visual brandcomparatively narrow, less than 1-5 mm wide (except in subspp. continentalis , and mooreisometimes, of A. unlcolor) 12 11 cf valva with terminal process comparatively long and narrow (Fig. 21). Underside of sympatric taxa greyish white; on hindwing postdiscal spot in space 6 usually more or less below that in space 7 leogoron (p. 34) Cf valva with terminal process short and broad (Fig. 22). Underside ground colour very palebuff; on hindwing postdiscal spot in space 6 nearly always inside the spot in space 7 andsometimes nearly mid-way to the end-cell bar melos (p. 36) 12 Underside ground colour greyish white to pale buff. Comparatively small, with forewing 10-5-18-0 mm. cf valva (Figs 29, 30) comparatively broad, with terminal process in centre line unicolor (p. 45) Underside chalky white. Larger, forewing 18-0-20-0 mm. cf valva narrower (Fig. 31), withterminal process nearly in line with ventral edge paetus (p. 50) 13 cf upperside greyish brown. $ with a whitish discal patch on forewing and an obscure whitish streak in space 5 on hindwing. cf 9 discocellular veins on hindwing not darkend luzonensis (p. 40)Cf $ upperside partly white on both wings. Discocellular veins blackened albatus (p. 41) 14 Underside of hindwing with central spot in space 7 at least partly blackened. Smaller, forewing 9-0-14-5 mm 15 Underside of hindwing with central spot in space 7 not blackened (may be darker brown thanother spots). Larger, forewing 12-5-15-5 mm nivalis (p. 51) 15 cf valva with costa incised shortly before apex (as in all preceding species). Underside of hindwing with postdiscal spot in space 7 not blackened on its inner edge, except sometimes inthe dry season form in Burma; ground colour greyish white, except in Philippines where it is pale buff substrigosus (p. 53) Cf valva with costa entire. Underside of hindwing with postdiscal spot in space 7 blackened onits inner edge; ground colour pale buff. Not found in Burma or Philippines davidis (p. 55) Allotinus (Paragerydus) hors field! (Moore) (Fig. 20, cf genitalia)Miletus horsfieldi Moore, 1857: 19. This and three succeeding species, leogoron, melos and macassarensis , form a confusing group difficult toseparate by superficial characters. In the males of all four species the swelling of vein M 3 and the visualbrand on the forewing vary according to locality. Venation is individually variable, but in A. horsfieldi veinsMI and R 5 of the forewing are nearly always connate , whereas in the other three species they usually share avery short common stalk. In both sexes the hindwing cilia are elongated into short tufts at the vein endingsand females have the hindwing termen crenulate. In the female of A. horsfieldi the hindwing is particularlystrongly crenulate. In A. leogoron the crenulations are much less pronounced, while in A. melos and A.macassarensis they are intermediate. On the underside the markings show much individual variability intheir density and in the position of the postdiscal series. On the forewing this series is dislocated at vein M 3 ,with the stria in space 3 moved basad and often forming with the stria in space 2 an irregular, oblique stripe;but the degree of dislocation is variable and usually slight in A. macassarensis. In addition the stria in spaceIb, if present, is moved basad in relation to that in space 2. Males of A. horsfieldi can always be identified with certainty by the genitalia (Fig. 20), wherein theventral tip of the uncus is much produced and the terminal process of the valva is long and curved up abovethe costa. In both sexes A. horsfieldi shows great variability in size. The smaller males, with forewing length as littleas 14 mm, have the forewing apex and termen more rounded and the visual brand relatively narrow, butthey are connected by a complete range of intermediates to the largest males, with forewing length up to 23mm, which have the forewing apex more pointed and the termen straighter. A. horsfieldi has a restricted distribution confined to Sundaland excluding Palawan. Where it occurs it isusually the commonest species of the subgenus apart from A. unicolor. THE MILETINI 33 Key to the subspecies of A. (P.) horsfieldi 1 Underside pale buff, postdiscal series not quadrate 2 - Underside greyish white, postdiscal series comprising large, dark, quadrate spots horsfieldi siporanus (p. 34) 2 Upperside brown without a reddish tinge . Underside usually lightly marked 3 - Upperside brown with a reddish tinge , especially strong in 9 and in basal half of forewing in cf . Underside more strongly marked horsfieldi permagnus (p. 33) 3 cf with visual brand broad, for about half its length contiguous with vein Cu\. 9 with forewing disc not conspicuously paler horsfieldi horsfieldi (p. 33) - cf with visual brand narrower, only its basal quarter touching vein Cu\. 9 with forewing disc conspicuously paler horsfieldi satelliticus (p. 34) Allotinus (Paragerydus) horsfieldi horsfieldi (Moore) Miletus horsfieldi Moore, 1857: 19, pi. la, fig. 2 cf . LECTOTYPE cf , JAVA (BMNH), here designated [examined]. Allotinus horsfieldi horsfieldi (Moore) Fruhstorfer, 1913: 367; 1916: 812; Corbet, 1939ft: 73.Allotinus horsfieldii [sic] (Moore); Piepers & Snellen, 1918: 12, pi. 19, figs 12a Cf , 12b 9, 14a cf , 14b $. In the male the visual brand is broad, and touches vein Cu^ for about half its length. The female is dullbrown without a reddish tinge, and the forewing disc is only a little paler. On the underside, especially infemales, the usual lycaenid markings are generally light and may be faded out, as in the extremes shown inPiepers & Snellen (1918: pi. 19, figs 14a, 14b), and on the hindwing the postdiscal spot in space 6 is usuallyplaced well inside the spot in space 7. I designate as lectotype a male in BMNH labelled /Type [red]/60-15 E. E.G. /Miletus horsfieldi cfM/GENITALIA Slide No ASC 23 Allotinus/. A female labelled /Type [red]/60-15 E.E.C./ is a paralecto-type. DISTRIBUTION. Java. Allotinus (Paragerydus) horsfieldi permagnus Fruhstorfer (Fig. 20, cf genitalia)Paragerydus horsfieldi (Moore); Distant, 1884: 207, pi. 20, fig. 7 '9' recte cf . Fig. 20 Allotinus (Paragerydus) horsfieldi permagnus Fruhstorfer; Rhio Archipelago: Great Karimon I.Male genitalia. 34 J. N. ELIOT Allotinus horsfieldi permagnus Fruhstorfer, 1913: 366; 1916: 812. LECTOTYPE d", SUMATRA (BMNH), here designated [examined].Allotinus horsfieldi nessus Corbet, 19396: 72, fig. 15, cf genitalia; Eliot, 1978: 240. Holotype cf , WEST MALAYSIA (BMNH) [examined]. [Synonymised by Eliot, 1967: 66.][Allotinus leogoron lindus Corbet, 19396: 73, partim $ nee d*. Misidentification.]Allotinus horsfieldi permagnus Fruhstorfer; Eliot, 1967: 66, fig. 1, cf genitalia; Fleming, 1975: 21, pi. 58, fig. L43 O". The subspecies is best distinguished by the female which is reddish brown with the forewing disc only veryslightly paler. The female named by Fruhstorfer as f . infumata is well within the normal range of individualvariation. In the male the visual brand is comparatively narrow, only touching vein Cui at its base, and thewing base of the forewing has a more reddish tinge than the darker apical region. I designate as lectotype of permagnus a male in BMNH labelled /Type [red]/Type [Fruhstorferorange]/W. Sumatra H. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/horsfieldi permagnus Fruhst. [inFruhstorfer's hand]/. The specimen, which is rather more lightly marked on the underside than usual, haslost its abdomen. A female labelled /Type [red]/Sumatra Montes Battak ex coll. H. Fruhstorfer/^horsfieldi Selesseh 19. xii. 94/Fruhstorfer Coll. B.M. 1933-131/horsfieldi permagnus Fruhst. [in Fruhstor-fer's hand]/ is a paralectotype. I designate as lectotype of infumata a female in BMNH labelled /Type [red]/Type [Fruhstorferorange]/N. Oekor viii. 94/Fruhstorfer Coll. B.M. 1933-131/infumata Frhst. [in Fruhstorfer's hand]/. DISTRIBUTION. Malay Peninsula, including Singapore; Sumatra; Bangka; Borneo; Peninsular Thailand(Pinratana). Three females in BMNH from Batu Is. are provisionally included in this subspecies; they aresmall with the underside ground colour whiter and the postdiscal markings rather dark and heavy, showingan approach to subsp. siporanus. They are also rather similar to A. leogoron batuensis, with which Eliot(1967) originally confused them and from which they can be distinguished by their more crenulatehindwingtermen. Allotinus (Paragerydus) horsfieldi siporanus Riley Allotinus horsfieldi siporana Riley, 1944: 253. Holotype $. MENTAWAI Is.: Sipora (BMNH) [examined]. The subspecies was described from a single large female in extremely battered condition; the whole of theouter half of the left hindwing and the tornal quarter of the right hindwing below vein Cu\ are missing. Thegreyish white ground colour of the underside and heavy dark brown markings suggest that it might pertainto A. leogoron, but such crenulations as remain on the termen of the right hindwing suggest A. horsfieldi,and it seems best to leave Riley's combination unchanged pending the discovery of the male. DISTRIBUTION. Mentawai Is. Allotinus (Paragerydus) horsfieldi satelliticus Fruhstorfer Allotinus horsfieldi satelliticus Fruhstorfer, 1913: 366; 1916: 812; Eliot, 1967: 66. LECTOTYPE $,ENGANO I. (BMNH), here designated [examined]. In the male the visual brand is like that of subsp. permagnus, but in other respects the subspecies morenearly resembles the nominate subspecies, especially in the female which is without a reddish tinge on theupperside. However, this sex differs from Javanese females by having a very prominent paler discal patchon the forewing. I designate as lectotype a female in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/EnganoApril-Juli Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/horsfieldi satelliticus Frhst. [in Fruhstorfer'shand]/. DISTRIBUTION. Engano I. Allotinus (Paragerydus) leogoron Fruhstorfer (Fig. 21, cf genitalia)Allotinus leogoron Fruhstorfer, 1916: 811. This species has been much confused in the past with A. horsfieldi, from which the male is readily separatedby the genitalia, wherein the ventral tip of the uncus is not produced and the terminal process of the valva is THE MILETINI 35 fairly long and narrow but not upturned. Females are best distinguished by the hindwing termen, which isless crenulate than in horsfieldi, but the cilia bear slightly longer and narrower tufts at the vein endings. Onthe forewing veins MI and R 5 usually have a short common stalk and, except in Java, the ground colour ofthe underside is whiter with darker, more contrasted markings than in horsfieldi.A. leogoron has the same distribution as A. horsfieldi. I recognise four subspecies. Key to the subspecies of A. (P.) leogoron 1 cf visual brand narrow, only its base touching vein Ci 2 - cf visual brand wide, touching vein Cu\ throughout half its length leogoron leogoron (p. 35) 2 Underside greyish white with dark markings 3 - Underside pale buff with markings smaller and paler brown leogoron plessis (p. 36) 3 d" visual brand with only its extreme base touching vein Cu\. Underside of hindwing with postdiscal spot in space 6 below that in space 7, as in the nominate subspecies leogoron nor muni (p. 36) - cf visual brand a little wider, touching vein Cu\ for a quarter of its length . Underside of hindwing with postdiscal spot in space 6 placed inside that in space 7, sometimes almost half-way tocell-end bar leogoron batuensis (p. 36) Allotinus (Paragerydus) leogoron leogoron Fruhstorfer (Fig. 21, cf genitalia) Allotinus horsfieldi permagnus $-f. intricata Fruhstorfer, 1913: 366; 1916: 812. LECTOTYPE $, SUMATRA (BMNH), here designated [examined]. [Unavailable name.] [Synonymised by Eliot, 1967: 68.]Allotinus leogoron Fruhstorfer, 1916: 811. LECTOTYPE cf, SUMATRA (BMNH), here designated [examined].Allotinus continental vadosus Corbet, 19396: 72. Holotype cf , WEST MALAYSIA (BMNH) [examined]. [Synonymised by Eliot, 1967: 69.]Allotinus leogoron leogoron Fruhstorfer; Corbet, 19396: 73; Eliot, 1967: 68, fig. 5, cf genitalia; 1978: 240; Fleming, 1975: 21, pi. 58, fig. L44 cf .Allotinus leogoron lindus Corbet, 19396: 73, partim cf nee $. Holotype cf, WEST MALAYSIA (BMNH) [examined]. [Synonymised by Eliot, 1967: 69.] The male can be separated from that sex of sympatric A. horsfieldi by the wider, more sharply definedvisual brand which touches vein Cu\ throughout its basal half, and the female by the less crenulate hindwingtermen. In addition, the underside is whiter and the markings are darker brown. Fig. 21 Allotinus (Paragerydus) leogoron leogoron Fruhstorfer; Malay Peninsula. Male genitalia. 36 J. N. ELIOT I designate as lectotype of leogoron a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/CMB viii. 94/A. leogoron Fr. [in Fruhstorfer's hand]/ and of intricata a female in BMNH labelled /Type[red]/Type [Fruhstorfer orange]/Sumatra Montes Battak ex coll. H. Fruhstorfer/Fruhstorfer Coll. B.M.1933-131/$ horsfieldi CMB viii. 94/intricata [in Fruhstorfer's hand]/. DISTRIBUTION. Malay Peninsula; Sumatra; Bangka; Peninsular Thailand (Pinratana). Allotinus (Paragerydus) leogoron normani EliotAllotinus leogoron normani Eliot, 1967: 69. Holotype cf , BORNEO (BMNH) [examined]. Differs from the nominate subspecies only in that the male has the visual brand much narrower, onlytouching vein Cui at its extreme base. DISTRIBUTION. Borneo. Allotinus (Paragerydus) leogoron batuensis EliotAllotinus leogoron batuensis Eliot, 1967: 69. Holotype cf , BATU Is. (BMNH) [examined]. A small subspecies, with a slightly more reddish brown tinge than in the two preceding subspecies. In themale the visual brand is intermediate between that of leogoron and that of normani, touching vein Cuialong its basal quarter. It differs on the underside of the hindwing in that the postdiscal spot in space 6 isplaced inside the spot in space 7. DISTRIBUTION. Batu Is. Allotinus (Paragerydus) leogoron plessis Eliot Allotinus leogoron plessis Eliot, 1967: 69. Holotype cf , JAVA (BMNH) [examined]. In the male the visual brand is narrow, as in subsp. normani, whereas the brand of sympatric A horsfieldihorsfieldi is as wide as in subsp. leogoron. The subspecies is further distinguished by the pale buff undersideon which the markings are smaller and less contrasted. DISTRIBUTION. Java. Allotinus (Paragerydus) melos (H. H. Druce) (Figs 22, cf genitalia; 64 cf ) [Paragerydus horsfieldi (Moore) sensu Semper, 1889: 164 partim, pi. 31, figs 19 cf, 20 $. Misidentifica- tion.]Paragerydus melos H. H. Druce, 1896: 652. LECTOTYPE cf , PHILIPPINES: Cagayan Sulu (BMNH), here designated [examined] Allotinus melos (Druce) Fruhstorfer, 1913: 369.Allotinus horsfieldi leos [sic] (Druce); Fruhstorfer, 1916: 812.Allotinus horsfieldi reverdini Fruhstorfer, 1916: 812. LECTOTYPE cf , PHILIPPINES (SM), here designated [examined]. Syn. n. melos (H. H. Druce); Eliot, 1967: 68.re verdini Fruhstorfer; Eliot, 1967: 68.talu Eliot, 1967: 68. Holotype cf , BORNEO: Pulo Laut (BMNH) [examined]. Allotinus macassariensis [sicAllotinus macassariensis [sicAllotinus macassariensis [sicSyn. n. This species replaces A. horsfieldi, with which it has been confused in the past, in the southern Philippines.Both species are found in Borneo, but A. melos seems to be the commoner throughout most of the island.Sympatric examples of the two species are very difficult to separate by external characters, but males maybe separated by the genitalia without difficulty; in melos the terminal process of the valva is short and broadand the tip of the uncus is not produced. Females may be impossible to separate with certainty, but on theunderside of the hindwing of melos the postdiscal spot in space 6 is usually placed inside the spot in space 7,sometimes almost half-way to the end-cell bar, whereas in horsfieldi these spots usually overlap. In additionthe hindwing termen of melos is very slightly less crenulate. Eliot (1967) treated melos as conspecific with the allopatric A. macassarensis , but in view of constantsmall differences in the male genitalia and in the arrangement of the postdiscal spots in spaces 4, 5 and 6 onthe underside of the forewing (see p. 39), I no longer maintain this combination. THE MILETINI 37 Fig. 22 Allotinus (Paragerydus) melos (H. H. Druce); Mindanao. Male genitalia. Males have vein M 3 swollen for three-fifths of its length and the visual brand is clearly defined and about1-75 mm wide. The underside shows very great individual variation in the intensity of the markings and inthe placing of the postdiscal series. The postdiscal series is quite well separated from the wing margins intypical examples but, especially in north-east Borneo, examples occur in which this series is placed close tothe termen, as in A. macassarensis . On the hindwing the postdiscal spot in space 6 is usually placed insidethat in space 7, sometimes as much as mid-way to the end-cell bar, and this character may be of assistance inseparating Bornean females from sympatric A. horsfieldi permagnus . H. H. Druce stated that the types of melos were in Mus. Cator and Druce. There is in coll. Cator (now inBMNH) a long series of both sexes, none of which has been labelled type, and in the main BMNH coll. apair ex H. H. Druce coll. labelled as paratypes. Of the latter I designate the male as lectotype; it is labelled/Paratype [yellow]/Cagayan 3. 6. 94/Paragerydus melos d" co-type H. H. Druce/ex coll. Hamilton Druce1919/Joicey Bequest Brit. Mus. 1934-120/. The female, labelled /Paratype [yellow]/Cagayan 2. 6. 94/P.melos $ co-type H. H. Druce/ex Coll. Hamilton Druce 1919/Joicey Bequest Brit. Mus. 1934-120/, as wellas the series in coll. Cator, are paralectotypes. When naming reverdini, Fruhstorfer referred to large males in coll. Semper. I designate as lectotype amale in SM labelled /19/Bohol/298b/213/Coll. C. Semper/Original of Semper PI. 31, fig. 19 from Mindanao.? loc. label changed. Det. as A. melos reverdini Fruh. C? J. N. Eliot, ix. 1982/. The male figured by Semperon pi. 31, fig. 19 was said by him to have come from Mindanao. I think it more likely that the label 'Bohol'was inadvertently transferred from the male shown in fig. 18 (which is the holotype of A. posidion georgiusand which is without a locality label - see p. 49) than that Semper made a mistake in the legend to his plate31. DISTRIBUTION. Cagayan Sulu; Mindanao; Palawan; Balabac I.; Borneo (many localities, including PuloLaut). There is a male in BMNH labelled /Sula Mangoli, Oct. '97, W. Doherty/, which I suspect may bewrongly labelled although Doherty certainly collected there at that date. Examples from Mindanao tend tobe rather heavily marked on the underside. Males from Palawan and Balabac tend to have the visual brandslightly longer and narrower. Examples from Pulo Laut are slightly richer buff on the underside thantypical examples from Cagayan Sulu, but the difference is hardly sufficient to justify maintaining talu as adistinct subspecies. There is in coll. Cator a series of seven males and one female from Melikop andSapagaya (small islands off the north coast of Sabah) which Cator had placed separately from his series ofmelos as an unrecognised species; they are small, darker brown above and with a greyish white undersidebearing darker markings. Superficially they resemble A. leogoron normani, but the male genitalia provethem to be A. melos, of which they could well be regarded as a microsubspecies. Allotinus (Paragerydus) samarensis sp. n. (Figs 23, cf genitalia; 65, 66 $, 105 cf ) Upperside of both sexes blackish brown, almost as dark as the shade of A. macassarensis. Male, onlyknown from the nominate subspecies, distinctive in that visual brand extends obscurely as a paler streak 38 J. N. ELIOT Fig. 23 Allotinus (Paragerydus) samarensis samarensis sp. n. ; Samar. Male genitalia. Phallus with juxtaattached. into fore wing cell basad of origin of vein Cu 2 . Female distinguished by the possession of a white area onforewing. Underside marked like A melos. Male genitalia hardly differ from those of A. leogoron, of which it may conceivably be a subspecies;because of the unusual visual brand and shorter swelling of vein M 3 in the male, and the distinctivewhite-marked female, I think it is best regarded as a distinct species. The species is only known from a very few examples from the southern Philippines and Sulawesi inslightly different subspecies. Key to the subspecies of A. (P.) samarensis 1 Underside pale greyish white. $ upperside of forewing with white area extending into cell well basad of origin of vein Cu 2 samarensis samarensis (p. 38) Underside very pale buff. $ upperside of forewing with white area not, or only just, entering cell samarensis russelli (p. 38) Allotinus (Paragerydus) samarensis samarensis subsp. n. (Figs 23, cf genitalia; 65 $, 105 cT) Cf upperside blackish brown, with vein M 3 swollen for just over half its length, visual brand about 1 -25 mmwide; a pale streak above cubitus extending a little basad of origin of vein Cu 2 appears to be an extension ofthe brand. Underside marked as in sympatric A. melos, but with ground colour more greyish white. $ upperside blackish brown. Forewing with a creamy white streak in spaces 4 and 3 ending 2-5 mm fromtermen and extending into cell along cubitus to beyond origin of vein Cu 2 . Underside similar to male. MATERIAL EXAMINED Holotype cf , Philippines: Samar, Bagacay, 2000', 31. v. 1979 (C. G. Treadaway) (coll. Treadaway). Paratypes. Philippines: 1 $ (allotype), west Samar, Hinabangan, 1000 m, 5.ii.l984 (coll. Treadaway); 2Cf , data as holotype (coll. Treadaway). Excluded from type-series. Mindanao: 1 $, Surigao Sur, Tandag, viii. 1981 (ex Takanami coll.)(BMNH), with the white streak wider and extending into space 2. Allotinus (Paragerydus) samarensis russelli subsp. n. (Fig. 66 $) Cf unknown. $ forewing 19-0-21-0 mm. Upperside dark brown with a white discal patch divided by dark veins onforewing, measuring 4-0 mm in space 2 and 8-0 mm in space 3, entering base of space 4 and lower angle ofcell where it is sullied. In a second specimen the white patch is smaller, not extending above vein M 3 norentering cell. Underside very pale buff with brown markings arranged as in nominate subspecies. THE MILETINI 39 MATERIAL EXAMINEDHolotype $?> Sulawesi: east-central, north-west of Morowali, Kabalo, 450 m, 10.iii.1980 (A. Bedford Paratype. 1 $ , data as holotype (coll. Bedford Russell). Allotinus (Paragerydus) macassarensis (Holland) (Figs 24 cf , genitalia; 67 cf , 68 )Paragerydus macassarensis Holland, 1891: 70. The species name has been consistently misspelled macassariensis in subsequent literature. Previously (Eliot, 1967) I united this species and A. melos, but in view of small but apparently constantdifferences in facies and male genitalia it seems best to treat them as separate allopatric species. The upperside in both sexes is slightly blacker brown than in the other species of the subgenus. On theunderside the postdiscal markings, which are usually heavy and dark chocolate brown on an off-whiteground, are placed closer to the termen than in the preceding species. A distinguishing character is that onthe forewing the postdiscal spots in spaces 4, 5 and 6 are on an irregular curve, with the spot in space 5overlapping that in space 4 and well separated from that in space 6. In the male genitalia the valvae differfrom those of A. melos in two respects; the terminal process is narrower and slightly longer from whicheverangle it is viewed and, in lateral view, there is a distinct concavity (indicated in Fig. 24 by an arrow) abouttwo-thirds from the base. The species is confined to Sulawesi and its satellite islands and is represented by two subspecies. Key to the subspecies of A. (P.) macassarensis 1 cf vein M 3 of forewing swollen for over two-thirds of its length, visual brand usually less than 1 mm wide .................................................................. macassarensis macassarensis (p. 39) - Cf vein M 3 of forewing swollen for only half its length, visual brand more than 1 mm wide macassarensis menadensis (p. 40) Allotinus (Paragerydus) macassarensis macassarensis (Holland) (Figs 24, cf genitalia; 67 cf , 68 ) Paragerydus macassarensis Holland, 1891: 70, pi. 4, fig. 5 $ . Holotype $ , SULAWESI (not located, probably in CM).Allotinus horsfieldi macassariensis [sic] (Holland) Fruhstorfer, 1913: 368; 1916: 812, pi. 141h. Fig. 24 Allotinus (Paragerydus) macassarensis macassarensis (Holland); Sulawesi. Male genitalia. Lowerleft, right valva enlarged. 40 J. N. ELIOT Allotinus unicolor damodar Fruhstorfer, 1913: 369; 1916: 811. LECTOTYPE cf , SULAWESI (BMNH), here designated [examined]. [Synonymised by Eliot. 1967: 68.]Allotinus macassariensis [sic] (Holland); Corbet, 19396: 72, fig. 14 cf genitalia.Allotinus macassariensis macassariensis [sic] (Holland); Eliot, 1967: 68. In the male the swelling of vein A/ 3 , extending just over two-thirds of its length, is the longest in thesubgenus and the visual brand is typically very narrow. But in east-central Sulawesi there is a tendency forthe brand to become wider and less well-defined; in one extreme example in coll. Bedford Russell (Fig. 64)there is a diffuse lighter patch which reaches across vein Cu\ more than half-way to vein Cu 2 . The femalehas the forewing disc at most only slightly paler. I designate as lectotype of damodar a small but otherwise normal male in BMNH labelled /Type[red]/Type [Fruhstorfer orange]/S. Celebes Tonus 27.xi.06/Fruhstorfer Coll. B.M. 1933-131/unicolordamodar Frhst. [in Fruhstorfer's hand]/. A female labelled /Type [red]/S. Celebes Samangi 17.xi.06/Fruhstorfer Coll. B.M. 1933-13 1/damodar Frhst. [in Fruhstorfer's hand]/ is a paralectotype. DISTRIBUTION. South to central Sulawesi; Banggai I. Allotinus (Paragerydus) macassarensis menadensis Eliot Allotinus macassariensis [sic] menadensis Eliot, 1967: 68. Holotype cf, SULAWESI: north (BMNH),[examined]. Differs from the nominate subspecies in that the swollen portion of vein M 3 in the male is only half its lengthand the visual brand is quite well-defined and from 1-25 to 2-0 mm wide. In both sexes the forewing discmay be lightened by a scattering of white scales, extreme examples showing an approach to A. samarensisrusselli. DISTRIBUTION. North Sulawesi as far south as Paloe Bay (Lat. 35' S). A single male from Bangka I. (offthe tip of the Minahassa Peninsula) is more lightly marked beneath and the ground colour has a buff tint. Allotinus (Paragerydus) luzonensis Eliot stat. n. (Figs 25, cf genitalia; 69 cf , 70 9) Allotinus macassariensis [sic] luzonensis Eliot, 1967: 68, fig. 3 cf genitalia. Holotype cf, PHILIPPINES:Luzon (BMNH) [examined]. The male is greyish brown above; on the forewing vein M 3 is swollen for half its length and the visual brandis rather diffuse. The underside is buff, closely striated, and the postdiscal markings in spaces 2 and 3 on the Fig. 25 Allotinus (Paragerydus) luzonensis Eliot; Luzon. Male genitalia. THE MILETINI 41 forewing and 6 and 7 on the hindwing are broad and rather blotchy. The female has a small white patch onthe forewing disc and an obscure whitish streak in space 5 of the hindwing. In these white patches it showsan approach to the female of A. albatus mendax, but they are less developed than in that species, and thediscocellular veins on the hindwing are not blackened. The species must be closely related to A . albatus because of the same arrangement of the markings on theunderside. The male genitalia of the only example dissected are intermediate between those of A. albatusand A. melos. DISTRIBUTION. Luzon. Allotinus (Paragerydus) albatus C. & R. Felder (Figs 26, cT genitalia; 71 , 72 c?, 73 )Allotinus albatus C. & R. Felder, [1865]: 287. The species is readily recognised by the presence in both sexes of white areas on both wings and by theheavily blackened discocellular veins on the hindwing. The underside is white marked more or less as in A.luzonensis, and like this species veins MI and R 5 of the forewing usually have a very short common stalk.The male genitalia most nearly resemble those of A leogoron, with the terminal process of the valva ratherlonghand slender.The species appears to be very rare and to be confined to Wallacea. Key to the subspecies of A. (P.) albatus 1 $ upperside with larger white areas, on forewing extending from space Ib to base of space 5, on hindwing from outer part of space 6 to dorsum albatus albatus (p. 41) - C? $ upperside with smaller white areas, on forewing not below mid-space Ib and not entering spaces 4 and 5, on hindwing usually not below vein 4 albatus mendax (p. 41) Allotinus (Paragerydus) albatus albatus C. & R. Felder (Fig. 71$) Allotinus albatus C. & R. Felder, [1865]: 267. Holotype $, SULAWESI (BMNH) [examined].[Allotinus fallax major Felder f. albatus Felder (= maximus Staudinger) sensu Fruhstorfer, 1913: 343,partim; 1916: 809, partim, asalbadus [sic].] The subspecies is known only from the female holotype, which the authors stated was from 'CelebesLorquin'. It was therefore surprising to find the holotype bearing labels reading /Halmaheira Lorquin[round blue]/Allotinus albatus Feld./Type [red]/FELDER COLLn. /albatus n./Rothschild Bequest B.M.1939-1/. As no Allotinus species is known from east of Weber's Line I feel confident that the locality'Halmaheira' is the result of a wrongly acquired label and that the specimen came from north Sulawesi,where Lorquin is known to have collected. The forewing is white dusted with brown basally and along the dorsum below vein A\, and with anirregular dark brown border narrowest at vein Cu2, where it measures 2-5 mm; above vein Cu-i it curves tothe wing base through the upper third of the cell. The hindwing is white, becoming sullied towards thedorsum, except for most of the cell and spaces 8, 7 and the basal half of space 6 which are brown; thediscocellular veins are heavily blackened. The underside is white with heavy brown markings. Fruhstorfer confused A. albatus with large examples of A. major with extensive white areas on theforewing which were named f . maximus by Staudinger. DISTRIBUTION. Sulawesi. Allotinus (Paragerydus) albatus mendax subsp. n. (Figs 26, c? genitalia; 72 a", 73 $) [Allotinus fallax C. & R. Felder, 1865: 285, partim, pi. 35, figs 25, 26 $; Semper, 1889: 163, partim.Misidentifi cations . ] Cf forewing length 20-0 mm. Upperside blackish brown. Forewing with a discal white patch 7-0 mm wide atthe bases of spaces 3, 2 and upper part of Ib and entering the lower angle of the cell; the inner and lowerpart of the cell and lower part of space Ib paler brown than the marginal and costal border. Vein M 3 swollen 42 J. N. ELIOT : .OP Fig. 26 Allotinus (Paragerydus) albatus mendax subsp. n.; Luzon. Male genitalia. Lower left, internalview of left valva enlarged; lower centre, lateral view of phallus. for just under three-fifths of its length. Hindwing with a white streak filling most of space 5, the upper basalpart of space 4 and just entering the cell; discocellular veins heavily blackened. Underside white, withreddish brown markings arranged as in sympatric A luzonensis. 9 similar to the male, except that the white streak on the hindwing is wider and enters space 6. MATERIAL EXAMINED Holotype cf, Philippines: Luzon, bearing labels /Allotinus mendax Bd. Manille/Ex Musaeo D risBOISDUVAL/ex Oberthur Coll. Brit. Mus. 1927-3/ Allotinus albatus Feld. [in Corbet's hand]/. Paratypes. Philippines: 1 $ (allotype) Luzon, labelled /N. Luzon, Whitehead. 94/Rothschild BequestB.M. 1939-1/ (BMNH); 1 cf , Luzon, Bicol Nat. Park, 29.viii.1980 (Y. Takanami) (BMNH); 1 cf , Luzon,Quezon Nat. Park, 27.iv.1983 (Y. Takanami) (BMNH); 1 cf, Luzon, Banahao Ridge, v.1982 (A.Ballantine) (coll. Ballantine); 1 <j>, Luzon, Quezon Nat. Park, 1000', Altimonan Rd, 20.vi.1954 (coll.Treadaway); 1 $, Quezon Nat. Park, Altimonan area, 28. iv. 1969 (coll. Treadaway). Excluded from type-series. Marinduque: 1 $ , Nu Boac, xi.1980 (coll. Treadaway) with the white area onthe hindwing more extensive and nearly reaching the dorsum and the discocellular veins less heavilyblackened. Samar: 1 $, east, Borongan, 100', 10.viii.1979 (coll. Treadaway), resembling the allotypeabove, but beneath with the postdiscal markings diffuse and obscure. Allotinus (Paragerydus) apries Fruhstorfer (Figs 27 Cf genitalia; 74, 106 cf )Allotinus horsfieldi apries Fruhstorfer, 1913: 344, partim. Fruhstorfer 's type-series probably comprised more than one species; in addition he confused apries with A.strigatus. I use apries here for the species identified as such by Corbet (19396). The species has several characters which enable it to be recognised with comparative ease. On theforewing veins M a and R 5 share a long common stalk averaging 1-5 mm. On the underside of the forewingthe postdiscal series is not, or only very slightly, dislocated at vein M 3 and the spots in spaces 4,3,2 and Ib(if present) are small, rounded and well separated from, and parallel to, the termen. In the female thehindwing termen is almost evenly rounded and the cilia are a little longer, but not tufted, at the veinendings. In the male genitalia the comparatively slender phallus is distinctive. The species occurs throughout Sundaland. Key to the subspecies of A. (P.) apries 1 Underside pale buff, tending to become darker towards forewing apex in cf, with darker buff-brown markings, cf visual brand 2-0 mm wide, with half its lower edge touching vein Cui 2 Underside greyish white, with darker greyish brown markings, cf visual brand 1-5 mm wide,with only basal fifth of its lower edge touching vein Cui apries ristus (p. 44) 2 Upperside reddish brown apries apries (p. 43) - Upperside brown without a reddish tint apries dositheus (p. 44) THE MILETINI 43 Allotinus (Paragerydus) apries apries Fruhstorfer (Figs 27, Cf genitalia; 74 cf ) [Allotinus horsfieldi (Moore) sensu Swinhoe, 1910: 198, partim, pi. 617, fig. 1 cf . Misidentification.]Allotinus horsfieldi apries Fruhstorfer, 1913: 344, partim; 1916: 812, partim, pi. 141g cf nee $. LECTO- TYPE cf , BORNEO (BMNH), here designated [examined].Allotinus strigatus eupalion Fruhstorfer, 1914: 22; 1916: 813. LECTOTYPE cf , SUMATRA (BMNH), here designated [examined]. Syn. n.Allotinus apries apries Fruhstorfer; Corbet, 1939ft: 70.Allotinus apries eupalion Fruhstorfer; Corbet, 1939ft: 70; Fleming, 1975: 21, pi. 58, fig. L42 cf ; Eliot, 1978: 240. In both sexes the underside is nearly always pale buff, with darker buff-brown markings, but occasionally itis more greyish white with only a slight buff tint. In the female the postdiscal series is lightly marked andmay be obsolete on the forewing. The male has vein A/ 3 swollen for a little over three-fifths its length, andthe visual brand is clearly defined and 2-0 mm wide. The female is more reddish brown above, with theforewing disc a little paler. Fruhstorfer described A. horsfieldi apries from several males from Sintang (south-west Borneo) and 10females from north Borneo. He wrote that the underside was bluish white speckled with pale brown in themale and thicker grey-brown in the female. This description does not accord well with the taxon heretreated as apries. However, there is a male from Sintang in BMNH labelled apries by Fruhstorfer and withthe underside paler and greyer than usual, which presumably formed part of the type-series. In additionthere are seven males of A. horsfieldi permagnus from Sintang ex Fruhstorfer coll. , and it seems likely thatthese also formed part of Fruhstorfer's type-series even though none had been labelled by him as apries.The only other males of apries ex coll. Fruhstorfer in BMNH are one labelled by Fruhstorfer as A. strigatusMoulton and one which was apparently regarded by Corbet as the holotype of apries and is labelled /Type[red]/Type [Fruhstorfer orange]/Kina Balu ex coll. H. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/apries/. As it did not come from Sintang it cannot have formed part of the type-series, and I reject it as atype. In its place I designate as lectotype of apries the male from Sintang, which is labelled /Sintang DrMartin H. Fruhstorfer/21.IV.10/Fruhstorfer Coll. B.M. 1933-131/apries Fr. [in Fruhstorfer's hand]/A.apries apries Det. by Dr. A. S. Corbet [in Corbet's hand]/. There are also two females of apries in BMNH which have been treated as syntypes. One is labelled/Type [red]/Type [Fruhstorfer orange]/Kina Balu Borneo/Fruhstorfer Coll. B.M. 1933-131/horsfieldiapries Frhst. [in Fruhstorfer's hand]/ and is a paralectotype. The other, labelled /Type [red]/Type[Fruhstorfer orange]/Sintang 10.IV.10/Fruhstorfer Coll. B.M. 1933-131/ is rejected as a paralectotype as itdid not come from north Borneo and therefore cannot have formed part of the type-series. Fig. 27 Allotinus (Paragerydus) apries apries Fruhstorfer; Malay Peninsula. Male genitalia. Lower left,lateral view of phallus. 44 J. N. ELIOT I designate as lectotype of eupalion a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/CMB II.94/Fruhstorfer Coll. B.M. 1933-131/strigatus eupalion Frhst. [in Fruhstorfer's hand]/GENITA-LIA Slide No. ASC 9 Allotinus/. Here it may be mentioned that the female figured by Fruhstorfer (1916: pi. 141g) as apries is a differentspecies, either A. horsfieldi or A. melos, and also that the male and female figured by Swinhoe (1910: pi.617, figs 1, la) as A. horsfieldi are in BMNH and are in fact Bornean specimens of A. apries and A. melosrespectively. DISTRIBUTION. Borneo, including Pulo Laut; Malay Peninsula; Sumatra. Allotinus (Paragerydus) apries dositheus Fruhstorfer Allotinus strigatus dositheus Fruhstorfer, 1914: 22; 1916: 813. LECTOTYPE cT, JAVA (BMNH), here designated [examined]. [Allotinus strigatus Moulton sensu Piepers & Snellen, 1918: 22, pi. 20, fig. 17 cf Misidentification.]Allotinus apries dositheus Fruhstorfer; Corbet, 19396: 72. The male does not differ from that sex of the nominate subspecies. The female differs in being brownwithout a reddish tint, in this respect showing parallel variation with the female of A. horsfieldi. I designate as lectotype of dositheus a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Java Occident. Sukabumi 2000' ex coll. H. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/strigatus dositheusFrhst. [in Fruhstorfer's hand]/GENITALIA Slide No. ASC 8 Allotinus/. A female, similarly labelled,which has been treated as a syntype, is a misidentified example of A. portunusportunus. DISTRIBUTION. Java. Allotinus (Paragerydus) apries ristus subsp. n. (Fig. 106 cf ) Cf on the upperside differs from subsp. apries in having a narrower forewing visual brand 1-5 mm wide. Theunderside differs by its greyish white ground colour and darker greyish brown markings. The striations arefine and the postdiscal spots small. MATERIAL EXAMINEDHolotype cf , Philippines: central Palawan, Languan, i.1981 (Treadaway Coll.). Allotinus (Paragerydus) corbeti Eliot (Fig. 28, cf genitalia) [Allotinus dilutus Corbet sensu Corbet, 19406: pi. 1, figs 13 cf , 14 $. Misidentification.]Allotinus corbeti Eliot, 1956: 34; 1984: 100. Holotype cf , WEST MALAYSIA (BMNH) [examined]. Fig. 28 Allotinus (Paragerydus) corbeti Eliot; Borneo: Pulo Laut. Male genitalia. THE MILETINI 45 [Allotinus felderi felderi Semper sensu Eliot, 1967: 70, partim. Misidentification.] Allotinus felderi corbeti Eliot; Eliot, 1967: 70, fig. 2 cf genitalia; 1978: 240; Fleming, 1975: 21, pi. 21, fig.L41cf. This very small species has an uniform reddish brown upperside in both sexes, and the underside marked asin A. leogoron leogoron. In the male the forewing apex is rounded, and the swelling of vein M 3 is confinedto the basal quarter of the vein; there is no visual brand. The female has the hindwing termen more stronglycrenulate than A leogoron. Although smaller, it is possible that A corbeti might be confused with A nicholsi, since the males of bothspecies are similar on the upperside. Females are also a similar shade of uniform reddish brown, but thoseof corbeti are readily separated by their crenulate hindwing with the cilia elongated into tufts at the veinendings. In addition, corbeti has no humeral vein on the hindwing, this vein being present in all sixexamples of nicholsi which I have examined. DISTRIBUTION. West Malaysia; Singapore; Sumatra; Pulo Laut; Mindanao; Peninsular Thailand (Pinrata-na). The species appears to be extremely rare, but is perhaps often overlooked. Allotinus (Paragerydus) unicolorC. & R. Felder (Figs 29, 30, O" genitalia)Allotinus unicolor C. & R. Felder, [1865]: 286. This, the most wide-ranging species of the genus, shows great individual variation in size, wing shape andunderside pattern. In general the smaller males, with forewing length as little as 11-0 mm, have a morerounded forewing apex and termen and a relatively shorter brand, whilst the bigger males have a moreacute apex and straighter termen and a brand which is slightly longer and narrower in relation to wing span.Examples, such as the holotype of unicolor, in which the usual lycaenid markings on the underside areheavy, are comparatively sparsely striated. In examples in which the underside is more densely striatedthere is usually a reduction in the size and darkness of these markings. On the forewing the postdiscal seriesis typically on an almost even curve, but examples in which it is dislocated at vein M 3 to a greater or lessextent are equally frequent. The best characters for identifying the species are the male genitalia (Figs 29, 30) with a broad valva withthe terminal process almost in the centre line, the short stalk of veins MI and R 5 of the forewing and therelatively short swelling of vein M 3 in the male. Fruhstorfer grouped the complex into three sympatric 'species': unicolor, aphocha andposidion. Underunicolor he grouped comparatively small specimens with a whitish ground colour on the underside andheavy postdiscal markings, and under aphocha specimens of similar size and wing shape but with lesspronounced markings. He reserved posidion for large specimens in which the male forewing was moreproduced. Corbet (1939ft), though at first inclined to agree with Fruhstorfer's arrangement, finally decidedthat all three constituted a single, variable species, but at the same time erected a new 'species', A dilutus, Fig. 29 Allotinus (Paragerydus) unicolor continentalis Fruhstorfer; Burma. Male genitalia. 46 J. N. ELIOT Fig. 30 Allotinus (Paragerydus) unicolor zitema Fruhstorfer; Sulawesi. Male genitalia. which he later (1956) relegated to subspecies, for very small examples from the Malay Peninsula. Thevarious sizes, wing shapes and patterns are connected iby intermediates and, like Corbet, I can find nodifferences in the male genitalia of the largest and smallest specimens. Therefore I also consider that thereis only one species, to which I now add the taxon continentalis , which has previously been treated as asubspecies of A. horsfieldi or as a distinct species. It is extraordinarily difficult to decide what constitutes a valid subspecies because of the great range ofphenetic variation. When large series are available for comparison, as in coll. BMNH, it is possible todetect differences between the average phenotype of different geographical areas, but because of theoverlap of phenetic characters it may often be impossible to ascribe individual specimens, if deprived oftheir locality labels, to any particular country of origin. I have opted to retain as subspecies those groups ofpopulations which can be distinguished by a mean of differences and which are contained within generallyaccepted faunal areas in preference to lumping together into just two or three polytypic subspecies thepopulations of widely separated geographical areas which in all probability differ genetically to aconsiderable degree. In using the key below this limitation should be borne in mind. The species ranges from Assam to the Lesser Sunda Is. , Philippines and Sulawesi (including the Sula Is.) ,and is most abundant in low level primary or secondary forest. Key to the subspecies of A. (P.) unicolor 1 Underside of hindwing with the postdiscal spot in space 6 well inside that in space 7 and often mid-way to the end-cell spot 2 Underside of hindwing with the postdiscal spot in space 6 much closer to the spot in space 7 thanto the end-cell spot 7 2 Underside ground colour greyish white . cf vein M 3 of fore wing swollen for not more than half its length. upperside of forewing with disc not conspicuously paler 3 Underside ground colour pale buff, cf vein M 3 swollen for a little over half its length. $ forewingdisc conspicuously paler and often sullied whitish unicolor continentalis (p. 47) 3 cf forewing brand 1 -0 mm or more wide. Continental Asia and Borneo 4 Cf brand less than 1-0 mm wide. Philippines and Sulawesi 6 4 9 upperside reddish brown 5 9 upperside brown without a reddish tint unicolor rekkia (p. 48) 5 cf forewing brand about 1-0 mm wide unicolor unicolor (p. 47) Cf brand comparatively short and broad, about 1-75 mm wide in large specimens. Underside rather lightly marked unicolor moorei (p. 48) 6 Underside with postdiscal markings usually heavy and sharply defined, cf vein A/ 3 swollen for just under half its length, brand sharply defined unicolor georgius (p. 49) Underside with postdiscal markings usually rather light and tending to be blurred, cf vein M 3swollen for half its length and brand usually rather inconspicuous unicolor zitema (p. 50) THE MILETINI 47 7 Upperside hindwing with postdiscal spot in space 6 often below and conjoined to the spot in space 7. $ upperside brown without a reddish tint, forewing disc paler . . . unicolor postilion (p. 48)Underside hindwing with the postdiscal spot in space 6 nearly always inside the spot in space 7. 9 upperside with a slight reddish tint, forewing disc only slightly paler... unicolor aphocha (p. 48) Allotinus (Paragerydus) unicolor unicolor C. & R. Felder Allotinus unicolor C. & R. Felder, [1865]: 286. Holotype 'cf' recte $, SINGAPORE (BMNH) [examined]. [Allotinus posidion myriandus Fruhstorfer, 1913: 368 (partim); 1916: 811 (partim).] Allotinus posidion eurytanus Fruhstorfer, 1913: 368; 1916: 811. LECTOTYPE cf , BORNEO (BMNH), here designated [examined]. Syn. n.Allotinus unicolor unicolor Felder; Fruhstorfer, 1913: 369; 1916: 809, pi. 141i; Corbet, 19396; 68, pi. 1, figs 3 $ holotype, 4 cf; 1956: 269, pi. 44, fig. 153 cf; Eliot, 1978: 240, pi. 20, figs 5 cf, 6 $.[Allotinus aphocha aphocha (Kheil); Fruhstorfer, 1913: 370 (partim); 1916: 810 (partim).]Allotinus unicolor eurytanus Fruhstorfer; Corbet, 19396: 70, pi. 1, figs 9 cf 'holotype of eurytanus' (in error), 10 cf 'holotype of eurytanus f. rebilus' (in error). Allotinus dilutus Corbet, 19396: 70. Holotype cf , WEST MALAYSIA (BMNH) [examined]. Syn. n.Allotinus unicolor dilutus Corbet; Corbet, 1956: 269; Cantlie, 1967: 27; Fleming, 1975: 21, pi. 57, fig. L40cf. The female holotype, figured by Corbet (19396) and mistaken by the Felders for a male, is an atypicalspecimen with heavy markings and a whiter than usual ground colour. Such specimens occur most often inSingapore, where they are connected by intermediates to normal phenotypes. The male figured by Corbet(19396), which matches the holotype fairly well but has a slightly greyer ground colour on the underside (asusual in males), would probably have been identified by Fruhstorfer as A. posidion myriandus because ofits size and wing shape. In general Bornean examples have the lycaenid markings smaller than in those from continental Asia,but compensate by being slightly more densely striated. This tendency is most extreme in examples fromsouth-west Borneo, named eurytanus by Fruhstorfer, which are particularly densely striated but can bematched by occasional examples from other areas. I designate as lectotype of eurytanus a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Sintang 19.IV.10/Fruhstorfer Coll. B.M. 1933-131/posidion eurytanus Fr. [in Fruhstorfer's hand]/. It wasfigured by Corbet (19396: fig. 10) in error as the holotype of A. unicolor eurytanus f. rebilus. The smallerspecimen figured by Corbet at fig. 9 in error as the holotype of eurytanus is labelled /Type [red]/Type[Fruhstorfer orangeJ/Sintang 26.IV.10/aphocha rebilus Frhst. [in Fruhstorfer's hand]/GENITALIA SlideNo. ASC 14 Allotinus/. Despite Fruhstorfer's identification label it cannot be accepted as lectotype ofrebilus, since this taxon was described from North Borneo; it probably formed part of Fruhstorfer's originaltype-series of four males and one female of A. posidion eurytanus, and how it obtained its label as rebilus isa mystery. DISTRIBUTION. This very variable subspecies is found in south Burma, where it has a zone of intergradationwith subsp. continentalis between Rangoon and Tavoy, and in Peninsular Thailand, West Malaysia,Singapore, Lingga Is., Natuna Is. and throughout Borneo, except in the Kina Balu area where it mergesinto subsp. moorei. Allotinus (Paragerydus) unicolor continentalis Fruhstorfer (Fig. 29, cf genitalia) [Paragerydus horsfieldi (Moore) sensu de Niceville, 1890: 26, pi. 36, fig. 156 cf . Misidentification.][Allotinus horsfieldi (Moore) sensu Bingham, 1907: 287, 299, fig. 73 cf ; sensu Swinhoe, 1910: 198 (partim, nee pi. 617, figs 1, la, Ib, Ic). Misidentifications.]Allotinus horsfieldi continentalis Fruhstorfer, 1913: 344; 1916: 812; Evans, 1932: 212; Cantlie, 1963: 27. Holotype cf , BURMA: Bhamo (probably in ZSI).[Allotinus posidion subsp.; Fruhstorfer, 1913: 368.]Allotinus posidion atacinus Fruhstorfer, 1916: 811; Evans, 1932: 212. Holotype $. BURMA (BMNH) [examined]. [Synonymised by Eliot, 1967: 70.] Allotinus unicolor atacinus Fruhstorfer; Corbet, 19396: 68, pi. 1, fig. 2 $ holotype; Cantlie, 1963: 27.Allotinus continentalis continentalis Fruhstorfer; Corbet, 19396: 72.Allotinus continentalis Fruhstorfer; Eliot, 1967: 70, fig. 4 cf genitalia. 48 J. N. ELIOT This is the most distinctive and largest subspecies of A. unicolor, with the forewing length of the maleusually 18-19 mm. On the underside both sexes are pale buff. On the upperside the male brand is longerand wider than in the other subspecies, usually 2-0 mm wide and touching vein Cu } at its origin, while theswelling of vein M 3 is a little over half its length. The female is distinguished by the prominently paler discalarea on the upperside of the forewing, which may be sullied whitish in the dry season. DISTRIBUTION. Assam; Burma as far south as Tavoy; north-west Thailand. Around the latitude of Rangoonit intergrades with subsp. unicolor. Allotinus (Paragerydus) unicolor rekkia Riley & Godfrey Allotinus posidion rekkia Riley & Godfrey, 1921: 180, pi. 6, figs 1 d", 2 $. Holotype cf , THAILAND: east (BMNH) [examined].Alliotinus unicolor rekkia Riley & Godfrey; Eliot, 1967: 70. Females are brown without the reddish tint of subsp. unicolor, but otherwise the subspecies does not differand is of doubtful validity. DISTRIBUTION. Only known from eastern Thailand, but probably also occurs in Cambodia, Laos andVietnam. Allotinus (Paragerydus) unicolor moorei (H. H. Druce) Paragerydus moorei H. H. Druce, 1895: 562, pi. 31, figs 5 cf , 6 $ . Syntypes, BORNEO: Mt Kina Balu (coll. Staudinger, probably in MNHU). Allotinus paetus moorei (Druce) Fruhstorfer, 1913: 369: 1916; 811.Allotinus aphocha rebilus Fruhstorfer, 1913: 370; 1916: 810. LECTOTYPE $, BORNEO (BMNH), here designated [examined]. Syn. n.Allotinus moorei (H. H. Druce); Corbet, 19396: 68. Druce's figure of the male shows a comparatively large specimen weakly marked beneath and with thebrand on the upperside of the forewing rather short, broad and diffuse. Judging by material in BMNH suchmales are rare, and are connected by intermediates to smaller males which differ little, if at all, fromnominate unicolor. The subspecies is therefore of doubtful validity. Fruhstorfer's taxon rebilus is applicable to smaller examples. I designate as lectotype a female in BMNHlabelled /Type [red]/Kina Balu/ex coll. H. Fruhstorfer/ Allotinus aphocha rebilus Fr. [in Corbet's hand]/. DISTRIBUTION. The subspecies flies on Mt Kina Balu, and might conceivably be a local modificationoccurring at higher elevations than normal examples referable to subsp. unicolor. Allotinus (Paragerydus) unicolor aphocha Kheil Allotinus aphocha Kheil, 1884: 28, pi. 5, fig. 30 $. Holotype $, NIAS (probably in MNHU). Allotinus posidion myriandus Fruhstorfer, 1913: 368; 1916: 811. Lectotype cf, SUMATRA (BMNH), by designation (as holotype) of Corbet, 19396: 68 [examined]. Syn. n. Allotinus aphocha aphocha Kheil; Fruhstorfer, 1913: 370 (partim); 1916: 810 (partim), pi. 141g cf 9-Allotinus unicolor myriandus Fruhstorfer; Corbet, 19396: 68, pi. 1, figs 5 cf holotype, 6 9 allotype.Allotinus unicolor aphocha Kheil; Corbet, 19396: 70, pi. 1, fig. 11 $. Differs from subsp. unicolor only in that on the underside of the hindwing the postdiscal spot in space 6 isusually placed closer to the spot in space 7. Many specimens are inseparable from unicolor, so thesubspecies is of doubtful validity. Fruhstorfer did not designate a type of myriandus, but Corbet (19396) figured its 'holotype' male and'allotype' female, and this action constitutes a valid lectotype selection. DISTRIBUTION. Sumatra; Bangka I.; Batu Is; Mentawai Is; Nias I. Allotinus (Paragerydus) unicolor posidion Fruhstorfer Allotinus posidion posidion Fruhstorfer, 1913: 368; 1916: 811. Lectotype cf, JAVA: west (BMNH), by designation (as holotype) of Corbet, 19396: 68 [examined].Allotinus posidion molionides Fruhstorfer, 1913: 368; 1916: 811. Lectotype cf, BALI (BMNH), by designation (as holotype) of Corbet, 19396: 70 [examined]. Syn. n. THE MILETINI 49 Allotinus posidion niceratus Fruhstorfer, 1913: 368; 1916: 812. LECTOTYPE cf, SUMBAWA (BMNH), here designated [examined]. Syn. n.Allotinus unicolor enganicus Fruhstorfer, 1913: 369; 1916: 811. Lectotype d", ENGANO I. (BMNH), by designation (as holotype) of Corbet, 19396: 70, pi. 1, fig. 12 [examined]. Syn. n.Allotinus unicolor bajanus Fruhstorfer, 1913: 369; 1916: 811; Corbet, 19396: 70. LECTOTYPE cf, LOMBOK (BMNH), here designated [examined]. Syn. n.Allotinus aphocha enatheus Fruhstorfer, 1913: 370; 1916: 810. LECTOTYPE $, JAVA (BMNH), here designated [examined]. [Synonymised by Corbet, 19396: 70.] Allotinus horsfieldii [sic] f. posidion Fruhstorfer; Piepers & Snellen, 1918: 12, pi. 19, figs 14a cf , 14b $.Allotinus suka Piepers & Snellen, 1918: 13 (partim), pi. 27, fig. 181 cf . LECTOTYPE cf , JAVA (probably in RNH), here designated. [Aberration reduced to infrasubspecific status by Corbet, 19396: 70.]Allotinus unicolor Felder; Piepers & Snellen, 1918: 14, pi. 19, figs 13a cf , 13b ?.[Allotinus aphocha Kheil; Piepers & Snellen, 1918: 15, pi. 19, figs 15a cf , 15b $.]Allotinus unicolor posidion Fruhstorfer; Corbet, 19396: 68, pi. 1, figs 7 cf holotype, 8 $ 'allotype' of enatheus. Allotinus unicolor molionides Fruhstorfer; Corbet, 19396: 70, pi. 1, fig. 13 cf holotype.Allotinus unicolor bajanus Fruhstorfer; Corbet, 19396: 70. The female lacks the reddish tint of subsp. myriandus and usually the forewing disc is paler. On theunderside of the hindwing the postdiscal spot in space 6 is more often directly below, and conjoined to, thespot in space 7; this character occurs most often in examples from the Lesser Sunda Is. , which also have thepostdiscal spots on average heavier than in Javanese examples. Fruhstorfer did not designate types of his taxa, but Corbet (19396) figured the 'holotype' males ofposidion, molionides and enganicus, and this action constitutes valid lectotype selections. He also figured afemale which he described as A. unicolor posidion f. enatheus Fruh. $ allotype. I now designate this femaleas lectotype of enatheus; it is labelled: /Type [red]/Type [Fruhstorfer orange]/Java Occident. Sukabumi2000' ex coll. H. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/aphocha enatheus Fr. [in Fruhstorfer'shand]/. I designate as lectotype of niceratus a male in BMNH labelled /Type [red]/Sumbawa/Allotinus posidionniceratus Fruh. Type [in Corbet's hand]/Adams Bequest B.M. 1912-399/. It is to be presumed thatFruhstorfer saw this specimen during his visit to BMNH prior to publication of his 1913 paper, as there areno specimens from Sumbawa from his collection in BMNH. I designate as lectotype of bajanus a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Pringabaja April 1896 H. Fruhstorfer/unicolor bajanus Frhst. [in Fruhstorfer's hand]/. Piepers & Snellen (1918) figured two different species as Allotinus suka: a female on pi. 19, fig. 12b towhich they had previously referred as Allotinus horsfieldii and a male on pi. 27, fig. 181. When relegatingsuka to the status of a form of A. unicolor posidion, in which the usual striations of the underside areabsent, Corbet (19396: 70) referred only to the latter figure which he said, in error, represented a female.Corbet's action could, perhaps, be taken to restrict the name suka to the species represented in fig. 181 , butto make absolutely certain that there should be no confusion in the application of the name I now designatethis male as lectotype. It should be in RNH. DISTRIBUTION. Java; Bali; Lombok; Sumbawa; Engano I. Allotinus (Paragerydus) unicolor georgius Fruhstorfer [Paragerydus horsfieldi (Moore) sensu Semper, 1889: 164, partim. pi. 31, fig. 18 cf ] Allotinus posidion georgius Fruhstorfer, 1913: 368, partim; 1916: 812. Holotype cf, PHILIPPINES: Bohol (SM) [examined].Allotinus unicolor leitus Fruhstorfer, 1916: 811. Holotype $, PHILIPPINES: Mindoro (coll. Staudinger, probably in MNHU). Fruhstorfer named georgius from Semper's records and figures of 'horsfieldi' from Bohol and Mindanao,which he subsequently (1916) realised represented two species. Thereupon he restricted georgius to Bohol,so that this name applies to the male figured by Semper at fig. 18, which automatically becomes theholotype; it is labelled /18/213/Coll. C. Semper/reverdini/original of Semper PI. 31, fig. 18 cf from Bohol.Holotype of Allotinus posidion georgius Fruh. det. J. N. Eliot ix.1982/. It is heavily marked below much asin the holotype of unicolor, but on the upperside the brand is a little shorter and narrower than in thatsubspecies.Fruhstorfer named leitus from a single female with a yellowish discal area on the forewing and heavy 50 J. N. ELIOT markings on the underside. The latter character suggests that it pertains to the same subspecies as georgius,but as I have seen no examples from Mindoro it is provisionally placed in synonymy. DISTRIBUTION. A unicolor must be very rare in the Philippines, as it is not represented in coll. Treadaway,and in BMNH there are only two females, one of which is labelled 'Philippines Pryer'. This appears to be anaberration; it is lightly striated on the underside, as in examples from north Borneo, and the postdiscalmarkings are elongated into longitudinal streaks. It may have come from the Sulu Is. , which were visited byPryer, or may be a mislabelled example from Borneo, where Pryer did nearly all his collecting. Allotinus (Paragerydus) unicolor zitema Fruhstorfer(Fig. 30, cf genitalia) Allotinus aphocha zitema Fruhstorfer, 1916: 810. LECTOTYPE $, SULAWESI (BMNH), here designated[examined]. The subspecies is distinguished by the male brand, which is a little narrower and more obscure than in othersubspecies. I designate as lectotype a female in BMNH labelled: Type [red]/Type [Fruhstorfer orange]/Nord-Celebes Toli Toli Nov. - Dez. 1895 H. Fruhstorfer/aphocha zitema Fr. [in Fruhstorfer's hand]/FruhstorferColl. B.M. 1933-131/. I have seen no males ex Fruhstorfer coll. DISTRIBUTION. Sulawesi and the Sula Is. Allotinus (Paragerydus) paetus (de Niceville)(Fig. 31 cf genitalia) Paragerydus paetus de Niceville, 1895: 269, pi. O, fig. 12 cf . Syntypes, SUMATRA: north-east (probably in ZSI). Allotinus paetus paetus (de Niceville) Fruhstorfer, 1913: 369; 1916: 811, pi. 141i <j>.Allotinus paetus (de Niceville); Corbet, 19396: 68, pi. 1, fig. 1 cf . This species bears a fairly close resemblance to examples of A. unicolor which are strongly markedbeneath, as in the holotype of unicolor. But it is larger, with forewing length averaging 18-19 mm in males,the forewing brand and swelling of vein A/ 3 are slightly longer than in all unicolor subspecies apart fromsubsp. continental, being just over half the length of the vein, and on the underside the ground colour ismore chalky whitish. The male genitalia are rather similar to those of A. unicolor, but the phallus is stouterand the distal portion of the valva narrower, with the terminal process not curved up so strongly towardsthe centre line. DISTRIBUTION. The species is only known from Sumatra, where it appears to fly in the Barisan Range fromthe Battak mountains in the north to the extreme south, where it was taken in numbers by Doherty. Fig. 31 Allotinus (Paragerydus) paetus (de Niceville); Sumatra. Male genitalia. THE MILETINI 51 Fig. 32 Allotinus (Paragerydus) parapus Fruhstorfer; Borneo. Male genitalia. Lower right, ventral viewof valvae and phallus. Allotinus (Paragerydus) parapus Fruhstorfer (Fig. 32, cf genitalia) Allotinus parapus Fruhstorfer, 1913: 343; 1916: 809, pi. 141h cf; Corbet, 19396: 66, fig. 8 cf genitalia.LECTOTYPE cf , BORNEO (BMNH), here designated [examined]. The sexes are alike in wing shape and in having a rounded hindwing termen with the cilia inconspicuouslyelongated at the vein endings. In the male vein M 3 is swollen for just under half its length and clothed withthe usual specialised scales. The white forewing b< >rdered with dark brown and the paler brown hindwingrender the species unmistakable. Both Fruhstorfer and Corbet, whose figure of the male valva is completely misleading, stated that veinM 3 was not swollen in the male, and on this account placed the species in the /a//ajt-group (ArtengruppeAllotinus). In fact the swelling is as well developed as in some other species of the subgenus, but becausethe surrounding area is white no visual brand is apparent. I designate as lectotype a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/FruhstorferColl. B.M. 1933-131/Nord-Borneo/ Allotinus parapus Frhst. [in Fruhstorfer's hand]/. A female, similarlylabelled, is a paralectotype. DISTRIBUTION. The species is montane and is known only from Mt Kinabalu in Sabah. Allotinus (Paragerydus) nivalis (H. Druce) (Figs 33, 34, cf genitalia; 75 cf )Miletus nivalis H. Druce, 1873: 348. This and the next two taxa form a natural group of small species distinguished by the possession of a whitefleck at the end of vein R 5 at the apex of the forewing in addition to the usual white flecks at the ends of veinsSc, RI, R 2 and R 3 . In males the swelling of vein M 3 is weak and short, and there is no visual brand. In A. nivalis the forewing termen is almost regular in both sexes, and on the underside of the hindwingthe central spot in space 7 is not, or only very little darkened. On average it is larger than the allied A.substrigosus and A. davidis. The species is confined to Borneo and the Philippines. Key to the subspecies of A. (P.) nivalis 1 Underside of forewing with a more or less developed submarginal brownish blotch about 1 -0-1 -5mm wide astride vein M 3 ; postdiscal series strongly dislocated at veins M 3 and Cu 2 nivalis nivalis (p. 52) 52 J. N. ELIOT Fig. 33 Allotinus (Paragerydus) nivalis nivalis (H. Druce); Borneo. Male genitalia. - Underside without a brownish blotch astride vein M 3 ; postdiscal series placed nearer the termen and only a little dislocated at veins A/ 3 and Cu 2 nivalis felderi (p. 52) Allotinus (Paragerydus) nivalis nivalis (H. Druce) (Fig. 33, cf genitalia) Miletus nivalis H. Druce, 1873: 348, 'cf' recte <j>. Holotype $, BORNEO (BMNH) [examined].Allotinus nivalis nivalis (Druce) Fruhstorfer, 1913: 370; 1916: 810, pi. 141g.Allotinus nivalis (H. Druce); Eliot, 1967: 71. The characters of the subspecies are given in the key. The figure by Fruhstorfer (1916: 141g) is very poorand shows an undersized specimen. DISTRIBUTION. Throughout Borneo, including Pulo Laut. Allotinus (Paragerydus) nivalis felderi Semper (Fig. 34 cf genitalia; 75 cf ) Allotinus felderi Semper, 1889: 163, pi. 31, fig. 22 $ ; Fruhstorfer, 1913: 370; 1916: 810. LECTOTYPE cf ,PHILIPPINES (SM), here designated [examined]. Fig. 34 Allotinus (Paragerydus) nivalis felderi Semper; Mindanao. Male genitalia. Lower left, ventralview of valvae and phallus of another specimen. THE MILETINI 53 Allotinus felderi felderi Semper; Eliot, 1967: 70, partim.Allotinus niv alls felderi Semper; Eliot, 1984: 100. The differences from the nominate subspecies are given in the key. Semper described the taxon from two males from Mindanao and two females from Luzon. I designate aslectotype one of the males in SM, labelled /Coll. C. Semper/Ost Mind./212/All. felderi typ. Semper/. Eliot (1967) confused felderi with A. corbeti from a pair oicorbeti and one exceptionally small female oftrue felderi from Mindanao, all of which had been placed under the latter name in coll. BMNH. DISTRIBUTION. Throughout the Philippines. In addition to examples from Luzon, Sibuyan and Mindanao inBMNH I have examined examples in coll. Treadaway from Marinduque, Negros and Samar. Allotinus (Paragerydus) substrigosus (Moore) (Figs 35, cf genitalia; 76, 77 cT)Logania substrigosa Moore, 1884: 22. Until separated by Eliot (1967) all authors treated this species as, or as a subspecies of, A. nivalis, fromwhich it differs as follows. On average it is smaller, with forewing length occasionally as little as 9-0 mm. Inboth sexes the forewing termen is distinctly crenulate, becoming more exaggerated in the dry season formfrom Burma. On the underside of the hindwing the central spot in space 7 is at least partially blackened andthere is often a black subbasal spot in space Ib, the corresponding spots in nivalis not being blackened. Themale genitalia are very similar, but in substrigosus the phallus is narrower than in sympatric nivalis. The species ranges from central Burma and Thailand to Sundaland and Mindanao. In Borneo it appearsto be rarer than A. nivalis. Key to the subspecies of A. (P.) substrigosus 1 Underside of hindwing with the spot mid-space 7 blackened only in upper half of space, and not more than 1 -0 mm wide 2 - Underside of hindwing with the spot mid-space 7 solidly black, nearly 2-0 mm wide and extending right across space substrigosus ballantinei (p. 55) 2 Underside greyish- white. Forewing crenulate 3 - Underside pale buff. Forewing only very weakly crenulate substrigosus yusukei (p. 55) 3 On underside of forewing the white fleck at end of vein 7 barely enters space 6 4 - On underside of forewing the white fleck at the end of vein R 5 is continued as an oblique white streak half-way across space 6 substrigosus substrigosus (p. 53) 4 cf with vein M 3 swollen for only one-quarter of its length substrigosus lenaia (p . 54) - cf with vein M 3 swollen for one-third of its length, as in subsp. substrigosus substrigosus sibyllinus (p. 54) Allotinus (Paragerydus) substrigosus substrigosus (Moore) (Fig. 35, cf genitalia) Logania substrigosa Moore, 1884: 22; 1886; 39, pi. 3, fig. 8 $. Holotype $, BURMA: Mergui Archipelago (probably in ZSI). [Paragerydus nivalis (H. Druce) sensu Distant, 1884: 207, pi. 22, fig. 11 $. Misidentification.][Allotinus nivalis (H. Druce) sensu de Niceville, 1890: 30 partim, pi. 36, fig. 159 $ holotype of substrigosa; sensu Bingham, 1907: 301; sensu Swinhoe, 1910: 197, pi. 616, figs 2, 2b cf, 2a $; sensu Piepers & Snellen, 1918: 16, pi. 20, fig. 19 cf . Misidentifications.]Allotinus nivalis magaris Fruhstorfer, 1913: 370; 1916: 810. LECTOTYPE cf , SUMATRA (BMNH), here designated [examined]. [Synonymised by Eliot, 1967: 71.]Allotinus nivalis substrigosa (Moore) Fruhstorfer, 1916: 810; Evans, 1932: 212; Corbet, 19396: 68; Cantlie, 1963: 28. Allotinus substrigosa substrigosa (Moore); Eliot, 1967: 71; 1978: 240.Allotinus substrigosus substrigosus (Moore); Fleming, 1975: 21, pi. 57, fig. L38 $ In the male the swelling of vein M 3 extends to one-third of its length. On the underside there is a prominentwhite streak at the forewing apex, and on the hindwing the central spot in space 7 is strongly blackened.From central Burma as far south as Tavoy a distinct dry season form occurs. The forewing termen is more 54 J. N. ELIOT Fig. 35 Allotinus (Paragerydus) substrigosus yusukeisubsp. n.; Mindanao. Male genitalia. Lower left, A.substrigosus substrigosus (Moore), Malay Peninsula; right valva and phallus. strongly crenulate; on the underside the ground colour becomes more greyish; on the forewing the apexand margin to a depth of 2-0-3-0 mm are shaded with brown, so that the oblique apical white streak standsout more conspicuously; and on the hindwing there is a similar brown area in spaces 3, 4 and 5, and theouter postdiscal spot in space 7, and sometimes that in space 6 also, may be blackened on their inner edges -a feature otherwise only found in A. davidis. This dry form, with its mottled appearance and crenulatewings, gives the impression of a Logania species. I designate as lectotype of magaris a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Selesseh 15.VII.94/Fruhstorfer Coll. B.M. 1933-131/Sumatra Monies Battak ex coll. Fruhstorfer/nivalismagaris Frhst. [in Fruhstorfer's hand]/. The locality Selesseh lies a little above sea level in north-eastSumatra and is more likely to be the correct locality than the Battak Mts. DISTRIBUTION. Burma, as far north as east Pegu; Thailand; West Malaysia; Sumatra; Borneo; Java (Piepers& Snellen, 1918). Allotinus (Paragerydus) substrigosus lenaia Fruhstorfer Allotinus nivalis lenaia Fruhstorfer, 1913: 370; 1916: 810. LECTOTYPE cf, NIAS (BMNH), here designated [examined].Allotinus substrigosa lenaia Fruhstorfer; Eliot, 1967: 72. In the single male which I have seen the swelling of vein M 3 is confined to the basal quarter of the vein. Onthe underside of the forewing the white streak at the end of vein R 5 at most barely enters space 6, and infemales the markings are on average lighter than in the nominate subspecies. I designate as lectotype a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Nias ex coll.Fruhstorfer/nivalis lenaia Frhst. [in Fruhstorfer's hand]/. DISTRIBUTION. Nias I. Allotinus (Paragerydus) substrigosus sibyllinus Riley Allotinus nivalis sibyllina Riley, 1944: 254, pi. 2, fig. 27 cf . Holotype cf , MENTAWAI Is.: Sipora (BMNH)[examined]. The upperside agrees with the nominate subspecies. On the underside the white fleck at the forewing apexis as redu'ced as in subsp. lenaia. On the hindwing the central spot in space 7 may be heavily blackened, as inthe nominate subspecies, or only lightly blackened. In his original description Riley wrote that this spot isnot blackened, but in seven out of the eight specimens in the type-series it is blackened to a greater or lessdegree. The eighth specimen, a worn male, is so rubbed that it is possible some black scales were originallypresent.The subspecies seems to be only doubtfully separable from subsp. lenaia. DISTRIBUTION. Mentawai Is. THE MILETINI 55 Allotinus (Paragerydus) substrigosus ballantinei subsp. n. (Fig. 76 cf ) Cf forewing 13-5 mm. Wing shape, as in subsp. substrigosus, with forewing termen crenulate and apexrather pointed. Swelling of vein M 3 inconspicuous, extending a little under one-third of its length.Underside very pale buff, rather densely striated but with postdiscal spots small or evanescent, as in someindividuals of subsp. substrigosus. Forewing with white speck at end of vein R 5 not extending into space 6.Hindwing with spot in mid-space 7 exceptionally large, rather quadrate, nearly 2-0 mm wide and extendingright across space. MATERIAL EXAMINEDHolotype cf , Philippines: Palawan, Port Barton, ix.1983 (A Ballantine) (BMNH). The subspecies is named after Dr Alistair Ballantine, who caught the unique holotype and presented it toBMNH. Allotinus (Paragerydus) substrigosus yusukei subsp. n. (Figs 35 cf genitalia; 77 cf ) Cf forewing 11-5 mm. Termen more rounded and apex less pointed than in foregoing subspecies. Swellingof vein M 3 short, extending over one-quarter of vein, overlying specialised scales pale so that it stands outprominently. Underside ground colour pale buff; forewing with white fleck at end of vein R 5 extending asan oblique white streak half-way across space 6. Hindwing with central black spot in space 7 narrow butwell-defined; a small subbasal black spot in space la. MATERIAL EXAMINEDHolotype cf , Philippines: Mindanao, Tandag, Surigao, xii.1982 (ex Takanami coll.) (BMNH). The subspecies is named after Mr Yusuke Takanami, who generously presented the holotype and manyother rare Philippine lycaenids to BMNH. Allotinus (Paragerydus) davidis Eliot(Fig. 36, cf genitalia) Allotinus davidis Eliot, 1959: 377, pi. 10, figs 6 cf , 7 $ , text-fig, cf valva; 1978: 240; Fleming, 1975: 21, pi.57, fig. L39 $. Holotype cf , SINGAPORE (BMNH) [examined]. The wing shape is that of A. nivalis, with the forewing termen barely perceptibly crenulate. On theupperside both sexes are reddish brown. In the male vein M 3 of the forewing is swollen only in its basalquarter. On the underside the ground colour is pale buff; on the forewing the white fleck at the apex doesnot cross vein R 5 ; on the hindwing the central spot in space 7 is blackened and the postdiscal spot is alsonarrowly blackened on its inner edge. In the male genitalia the valva differs from all others in the subgenus by not having the costa truncatebefore the terminal process, which curves upwards in the manner of subgenus Allotinus. In all other Fig. 36 Allotinus (Paragerydus) davidis Eliot; Singapore. Male genitalia. 56 J. N. ELIOT respects the genitalia are typical of Paragerydus, and the absence of a humeral vein on the hindwing helpsto confirm its position in this subgenus. DISTRIBUTION. West Malaysia; Singapore; Sumatra (coll. Diehl); Peninsular Thailand (Pinratana). Thespecies appears to be extremely rare, but is perhaps sometimes overlooked because of its small size andresemblance to A. substrigosus . Allotinus (Paragerydus) drumila (Moore) (Fig. 37, cf genitalia)Miletus drumila Moore, [1866]: 777. This large species stands rather far apart from the remainder of the subgenus in its markings and in itspronounced seasonal variation. On the other hand the male genitalia, presence on the forewing of aswollen vein M 3 and absence of a humeral vein on the hindwing, are characteristic of Paragerydus, so that itseems unnecessary to retain for it the separate subgenus Miletographa. On the underside the appearance is of a Miletus species, with the catenulate markings characteristic ofthat genus; the difference is that in A. drumila the ground between the usual lycaenid markings is speckledwith small, irregular pale-edged spots. The upperside and especially the seasonally dimorphic wing shaperecall the smaller, sympatric Miletus chinensis. The wet season form is brown, with a similar curved seriesof postdiscal spots on the forewing, paler brown in the male, larger and sullied white in the female. In thedry season the forewing acquires a sharp point at the apex and prominent lobe at the tornus, and the mainlywhite female is broadly similar to the dry season M. chinensis longeana. The main point of dissimilarity isthat in A. drumila the most extreme dry form occurs in the Himalayas and Assam, whereas in M. chinensisthe most extreme dry form occurs in Burma. The species is Indo-Burmese, submontane and apparently not rare where it occurs, but with a restricteddistribution. There are two rather doubtfully valid subspecies. Key to the subspecies of A. (P.) drumila 1 $ dry season form white with black costal and marginal border on the forewing and blackish costal area on the hindwing drumila drumila (p. 56) $ dry season form with the white areas sullied with buff scales drumila aphthonius (p. 57) Allotinus (Paragerydus) drumila drumila (Moore) (Fig. 37, cf genitalia) Miletus drumila Moore, [1866]: 777, pi. 41, fig. 12 $. Holotype 'cf ' recte <j>, INDIA: Sikkim (not located). [Dry season form.]Gerydus drumila (Moore) Moore, 1883: 521. [Wet season form.] Fig. 37 Allotinus (Paragerydus) drumila drumila (Moore) wet season form multistrigatus de Niceville;Sikkim. Male genitalia. THE MILETINI 57 Allotinus multistrigatus de Niceville, 1886: 253, pi. 11, figs 11 o" , 2 $ ; de Niceville, 1890: 29, pi. 26, figs 157Cf, 158 $; Bingham, 1907: 298; Swinhoe, 1910: 195, pi. 615, figs 2, 2b cf , 2a, 2c $ ; Fruhstorfer, 1913:371; 1916: 815, pi. 141h cf $. LECTOTYPE cf , INDIA: Sikkim (BMNH), here designated [examined][Wet season form.] [Synonymised by Cantlie, 1963: 26.] Miletus insignis Staudinger, 1888: 269, pi. 94 'cf' recte $. Holotype $, INDIA: Sikkim (probably MNHU).[Synonymised by de Niceville, 1890: 28.] Allotinus drumila (Moore) de Niceville, 1890: 28; Bingham, 1907: 297; Swinhoe, 1910: 194, pi. 615, figs 1,Ib Cf, la, Ic $; Fruhstorfer, 1913: 371; 1916: 815, pi. 141i $; Evans, 1932: 211. Miletographa drumila (Moore) Rober, 1892: 277. Allotinus multistrigatus multistrigatus de Niceville; Evans, 1932: 242. Allotinus drumila drumila (Moore) with wet season form multistrigatus de Niceville; Cantlie, 1963: 26, pi.26, fig. H. 6. 2. Moore originally described the dry season female in error as the male. Later (1883) he realised his mistakeand correctly described the male from the wet season form, which de Niceville later redescribed as aseparate species, A. multistrigatus. Judging by material in BMNH, intermediate forms, with the wing shape of the dry form drumila, fly inthe autumn and winter, while drumila, with its mostly white female, is really a spring form flying fromFebruary to May (one female ex Fruhstorfer coll. is labelled June, but I feel sure this is an error, as thetype-series of multistrigatus was caught in this month). Moore's type of drumila cannot be traced, but there are two wet season males in BMNH from Darjilingand the Khasi Hills from which Moore described the male, both of which bear BMNH type-labels. But asthey did not form part of the original type-series they cannot be accepted as types. The figures accompanying de Niceville's original description of multistrigatus were taken from examplesin coll. Moller, and I designate as lectotype a male in BMNH, presumably from the original type-series,labelled /Type [red]/Sikkim Moller/ Allotinus multistrigatus de Niceville cf TYPE/Rothschild BequestB.M. 1939-1/. DISTRIBUTION. Kumaon, in the central Himalayas, to Assam. Allotinus (Paragerydus) drumila aphthonius Fruhstorfer Allotinus aphthonius Fruhstorfer, 1913: 371; 1916: 815. LECTOTYPE cf, BURMA (BMNH), here designated [examined]. [Intermediate form.]Allotinus drumila grisea Riley & Godfrey, 1921: 180. Holotype $, THAILAND (BMNH) [examined]. [Dry season form.] [Synonymised by Pinratana, 1981: 31.]Allotinus multistrigatus apthonius [sic] Fruhstorfer; Evans, 1932: 242Allotinus drumila apthonius [sic] Fruhstorfer; Cantlie, 1963: 26; Pinratana, 1981: 31, pi. 4, fig. 19 cf , pi. 5, figs 10 Cf, 11 $. The subspecies is doubtfully valid, differing only in that the most extreme dry form of female has the whiteareas always sullied with buff scales, as in the holotype of grisea and as in Pinratana's figure of the female.The wet season and intermediate forms do not differ from the corresponding forms of subsp. drumila. In BMNH there are one male and two females labelled as types of aphthonius. I designate as lectotypethe male labelled /Type [red]/Type [Fruhstorfer orangeJ/Tenasserim Tandong 4000' Mai Fruhstorferleg. /Fruhstorfer Coll. B.M. 1933-1/Allotinus aphthonius Frhst. [in Fruhstorfer's hand]/. Two females,similarly labelled, except that one bears a final label /Allotinus aphthonius Fruh. [in Corbet's hand]/ inplace of the label in Fruhstorfer's hand, are paralectotypes. The male has a pointed forewing and ratherdentate hindwing, and is marked as in the wet season form. Fruhstorfer stated that it was a dry season form,but in fact it is an intermediate form nearer to the wet than to the dry season form. Of the two females, oneis intermediate, as in the male, and the other of the normal wet season form. DISTRIBUTION. Throughout Burma, except in the extreme south of Tenasserim; Thailand. Genus LOGANIA Distant Logania Distant, 1884: 197, 208. Type-species: Logania malayica Distant, 1884: 208, pi. 22, fig. 21 $, by monotypy. Gender feminine.Malais Doherty, 1889: 414, 415, 436. Type-species: Loganiasriwa Distant sensu Doherty, 1889 [= Logania marmorata Moore, 1884], by designation of Corbet, 1940a: 111. Gender feminine. [Synonymised by de Niceville, 1890: 32.] Eyes smooth. Antennae half the length of the forewing costa, with about 36 segments in the type-species; 58 J. N. ELIOT shorter, with under 30 segments in L. waltraudae; rather longer, with 40-45 segments in the marmorata-group (= Malais}. Nudum extending widely to the base of the shaft. Legs much shorter than in Allotinus,with the tibiae outwardly swollen and the fore-tarsi, except in L. waltraudae, gradually incrassate. In L.malayica and L. waltraudae the male fore-tarsus ends in a rather long, tapered, down-curved point, but inthe marmorata-group the fore-tarsus ends abruptly in a rounded pad from which a minute point is directeddownwards, as in Allotinus. Labial palpi shorter than in Allotinus, with the third segment usually shorterthan one-half of the second segment in males, but may be slightly longer than half in females. Males of allspecies have a small double hair tuft on the sternum of the eighth abdominal segment. The type-species hasthe forewing apex produced to a sharp point and veins MI and R 5 have a long common stalk; all theremaining species have a rounded or square apex and, except in L. nehalemia, the stalk of veins Mj and R 5is absent or short. There is no trace of a humeral vein on the hindwing. Males of the ma/ay/ca-group havevein A/3 of the forewing unswollen; in the marmorata-group the basal portion of vein M 3 is briefly swollenand clothed with small, specialised scales which are about the same size as those of the nivalis-group ofParagerydus. In L. regina, and probably also in L. paluana, the swelling is inconspicuous and partly hiddenby normal cover scales. The underside pattern is generally similar to that of Allotinus, but the groundcolour is seldom uniform, being mottled in shades of black, brown and white, for which reason Evanscoined the popular name 'Mottles' for the genus. On the forewing the usual lycaenid markings may bedifficult to make out, but on the hindwing they are usually apparent and the postdiscal series may becatenulate. In some species there is very great individual variation in the extent of whitish scaling on theupperside of the hindwing, which may be absent or cover almost the whole of the wing. The male genitaliaare of the usual miletine type and are rather constant in appearance, except that L. waltraudae shows somecharacters suggestive of Allotinus. Doherty (1889: 414, 415) erected the generic name Malais and wrote that 'it will include L. marmorataand L. sriwa (probably the same species) and one or two rare kinds undescribed'. He had before him only asingle female fronrMergui which he treated with some doubt as the same as the taxon named Logania sriwaDistant, 1886. De Niceville (1890: 33), who had custody of the type of L. marmorata Moore, 1884, statedthat he had examined Doherty's female of 'sriwa' and found that 'it appears to differ from L. marmorataonly by the greater prominence of all the markings of the underside'. As the females of L. sriwa and L.marmorata differ so strongly it is inconceivable that Doherty's female was really sriwa, and highly probablethat de Niceville was correct in assuming that it was L. marmorata. I can find no convincing evidence that L.sriwa has ever been taken in Burma, although it was recorded by Evans (1932) and Cantlie (1963) fromMergui, probably on the basis of Doherty's misidentified female. Soon afterwards Doherty (1891o: 29) doubted if Malais was distinct from Logania; and though he did notformally synonymise the two he thereafter used Logania for species which would fall naturally into Malais.Bingham (1907: 302), Swinhoe (1910: 200), Evans (1932: 199) and Cantlie (1963: 2) all treated Malais as asubjective synonym of Logania, as did Corbet (1940a: 111) who stated that L. sriwa was its type-species.However, as Doherty's 'sriwa' was misidentified, Corbet's statement cannot, under Article 70 of the Code,be held to constitute a valid type selection. Fruhstorfer (1914; 1915), whilst using Logania as the genericname, employed Malais in a subgeneric sense for the 'Artengruppe' with banded legs and roundedforewing apex which includes L. marmorata and L. sriwa. Hemming (1960: 11), apparently unaware ofCorbet's action, designated Logania malayica Distant, 1884, as type-species of Malais. His action isinvalid, since malayica was not one of the species originally included in Malais by Doherty; indeed, thelatter (1889: 437) had specifically excluded malayica from his new genus, stressing that it was a trueLogania. Malais remains, therefore, a valid and available genus-group name, which can be used as asubgenus, largely in the sense in which it was employed by Fruhstorfer, by those authors who consider thatthe differences between Logania and Malais are greater than those between species-groups. Malais is,however, still without a properly established type-species, and any author intending to use it will have torefer the matter to the International Commission on Zoological Nomenclature, as required under Article70 of the Code, with a recommendation that Logania marmorata Moore, 1884, be designated astype-species. The genus ranges from Peninsular India through the Archipelago to New Guinea and the Bismarcks, andcomprises 10 species. Key to the species of Logania 1 cf upperside of forewing with vein M 3 unswollen and clothed with normal cover scales, cf fore-tarsus, so far as known, ending in a tapered, down-curved point 2 Cf upperside of forewing with basal part of vein M 3 swollen and clothed with specialised scales. Cf fore-tarsus ending abruptly , but with a small point directed downwards from its lower edge 4 2 Forewing apex not produced 3 THE MILETINI 59 Forewing apex produced to a sharp point malayica (p. 59) 3 Upperside of hindwing all white waltraudae (p. 61) Upperside of hindwing with outer half black and inner half white nehalemia (p. 60) 4 Legs banded, with a specially prominent, broad, brown band on outer half of tibiae 5 - Legs freckled, not clearly banded 8 5 $ upperside with grey to white areas except in marmomta diehli 6 - $ upperside entirely brown obscura (p. 68) 6 Forewing very weakly crenulate . Underside of hindwing with a white streak or patch bearing few if any striae 7 - Forewing termen crenulate. Underside of hindwing without an unstriated area... marmorata (p. 64) 7 Upperside of hindwing pale grey to whitish regina (p. 62) - Upperside of hindwing white with a broad, black border paluana (p. 63) 8 Forewing termen only very weakly crenulate 9 - Forewing termen crenulate watsoniana (p. 73) 9 Underside mottling more or less ochreous distant! (p. 69) - Underside mottling dark brown; no ochreous tinge hampsoni (p. 72) Logania malayica Distant(Figs 38, 39, cf genitalia)Logania malayica Distant, 1884: 208. The species is instantly recognisable by the pointed and produced forewing apex. The sexes are alike,above white with a blackish forewing border expanding from less than 1-0 mm at the tornus to nearlymid-costa. Underside white densely mottled with reddish brown striae. The legs are buff-brown freckledwith whitish scales, and the male fore-tarsus ends in a comparatively long, tapered, down-curved point. The species occurs from Peninsular Thailand to Malaya, Sumatra, Borneo and the Philippines. Thereare two subspecies with somewhat different male genitalia. Key to the subspecies of L. malayica 1 Underside irregularly mottled and blotchy, with, some of the usual lycaenid markings apparent malayica malayica (p. 60) - Underside regularly and densely mott; , with none of the lycaenid markings apparent malayica subura (p. 60) Fig. 38 Logania malayica malayica Distant; Malay Peninsula. Male genitalia. 60 J. N. ELIOT Logania malayica malayica Distant (Fig. 38, cf genitalia) Logania malayica Distant, 1884: 208, pi. 22, fig. 21 $, text-fig. 61 hind-leg. Holotype $, WEST MALAYSIA: Sungei Ujong (not located).Logania malayica malayica Distant; Fruhstorfer, 1914: 23; 1916: 805, pi. 141f cf ; Corbet, 1940a: 111, fig. 1 Cf valva; Fleming, 1975: 22, pi. 58, fig. L51 $; Eliot, 1978: 241, pi. 20, fig. 8 cf . On the underside the mottling is irregular and coalesced into blotches in places, and covers both wingsexcept for a white area above the forewing dorsum in spaces la and Ib. The usual lycaenid markings can bepartly made out with difficulty. DISTRIBUTION. Peninsular Thailand; West Malaysia; Sumatra; Borneo, including Pulo Laut. Logania malayica subura Fruhstorfer(Fig. 39, cf genitalia) Logania malayica Distant; Semper, 1889: 160, pi. 31, fig. 3 cf. Logania malayica subura Fruhstorfer, 1914: 23; 1916: 805. Holotype cf , PHILIPPINES: Mindanao (SM). On the underside the mottling of striae is regular and without blotches; none of the usual lycaenid markingscan be made out. On the forewing the white area above the dorsum extends into space 2. The male genitaliadiffer more than usual in subspecies, and it may be that subura has achieved species status. Fruhstorfer named subura from Semper's figure, so the male depicted therein is automatically theholotype. DISTRIBUTION. Probably throughout the southern Philippines, but I have only seen examples fromMindanao and Samar. Fig. 39 Logania malayica subura Fruhstorfer; Samar. Male genitalia. Logania nehalemia Fruhstorfer stat. rev.(Figs 40, cf genitalia; 78 cf ) Logania nehalemia Fruhstorfer, 1914: 25; 1916: 808. Holotype '$' recte cf, NEW GUINEA (BMNH) [examined].Logania hampsoni nehalemia Fruhstorfer; D'Abrera, 1971: 384. Above, the forewing is white with a black border which curves in above the cell to the wing base; thehindwing has the basal half white and the outer half black. The underside is white, with dense, dark brownstriae more or less corresponding with the areas which are black on the upperside.In the unique male holotype, which has hitherto been regarded as a female, the legs are missing except THE MILETINI 61 Fig. 40 Logania nehalemia Fruhstorfer; New Guinea. Male genitalia. for the femur and tibia of one hindleg with the scales rubbed off; these hardly differ from the femur andtibia of L. malayica. The male genitalia are chiefly distinguished by the phallus, which is considerablystouter than that of its congeners except for L. waltraudae. The holotype is labelled /Type [red]/New Guinea. Hewitson Coll. 79-69 Miletus 1/9 holotype Loganianehalemia Fruhst. [in Corbet's hand]/. Given its early date of capture, the type-locality is likely to lie in thenorth-western part of Irian Jay a. DISTRIBUTION. New Guinea. Logania waltraudae sp. n. (Figs 41, cT genitalia; 107 cf ) C? forewing 10-0 mm. Generally similar in appearance to the sympatric L. malayica subura, and like it withthe basal portion of vein M 3 unswollen and clothed with normal cover scales; but differing in having theforewing apex rounded and veins MI and R 5 connate. Fig. 41 Logania waltraudae sp. n. ; Samar. Male genitalia. 62 J. N. ELIOT Upperside white; forewing with a blackish brown apical border running from just below vein Cu 2 on thetermen to just above the cell apex on the costa; hindwing with a dark brown marginal hairline. Cilia darkbrown. Underside pale brown very densely mottled with darker brown striae except on the forewing inmost of spaces Ib and 2 and in the basal part of space 3, which are white. A dislocated series of postdiscalspots can just be made out on the forewing. Antennae just under half the length of the forewing costa, thinner than in L. malayica, with probably28-30 segments (both antennal clubs are broken off just before the tip after 26 segments). The middle shaftsegments are just over twice as long as wide (in the remaining Logania species these segments are nearly aswide as long). The nudum extends widely to the base of the shaft, which is brown on the upper surface witha central buff patch on each segment. The palpi are clothed with brown and a few buff adpressed scales andare exceptionally short, not protruding beyond the head, with the third segment comparatively stout andonly a quarter the length of the second segment. The legs are about as long as those of L. malayica, whichthey resemble in having the fore-tarsus ending in a tapered, down-curved point; but they differ in severalrespects. The femora, tibiae and tarsi are subequal; the femora are broader and somewhat flattened; theforetibiae are narrow and cylindrical, and the middle tibiae are slightly swollen, the swelling being greatestin the upper half. The hind legs are missing. Body dark brown, slightly paler beneath; the abdominal hairtufts on the eighth sternum are smaller than in the other species of Logania. The male genitalia are broadlyof Logania type, but the phallus is distinctive, while the triangular vinculum flap and strut running parallelto the lower edge of the uncus plate recall those structures in Allotinus. $ unknown. MATERIAL EXAMINED Holotype cf , Philippines: Samar, 18.viii.l980(C. G. Treadaway) (coll. Treadaway, but will be depositedin due course in SM). The species, which is named after Mrs Treadaway, occupies an isolated position in the genus, differingfrom the remainder in palpi, antennae, legs and male genitalia, and possibly deserves to be placed in aseparate subgenus. Logania regina (H. Druce) (Fig. 42, cf genitalia)Miletus regina H. Druce, 1873: 348. This and subsequent species differ from the foregoing species in the male fore-tarsus, which is not taperedbut ends abruptly in a rounded pad from the lower side of which a small, short point is directed downwards,as in Allotinus. The legs, in this and the next three species, are banded, most prominently on the tibiae,with whitish and brown, while in the remaining species they are freckled and sometimes have obscurelongitudinal streaks. A distinguishing character of L. regina is the white ground colour of the underside of the hindwing,which is almost devoid of striae in a streak-like area running from the wing base to the termen through theupper part of the cell and space 6. On the upperside the male has a fuscous border on the forewing tending to run narrowly along the costato the wing base, and a fuscous costal area above vein 6 on the hindwing. The rest of the wings are whitish tobluish grey. The female has a narrower forewing border and the pale areas on both wings are whiter than inthe male. In the female of this species and of L. paluana (infra) the abdomen is longer than in the other Loganiaspecies, and extends just beyond the hindwing tornus. The species has a restricted distribution in Sundaland, excluding Java, Palawan and the islands off thewest coast of Sumatra, but has reached the Sulu Is. where it must be a recent immigrant. There are twosubspecies. Key to the subspecies of L. regina 1 cf upperside of forewing with black border comparatively wide, filling whole of space 5. $forewing border reaching dorsum; underside of forewing with a white area above dorsumusually reaching vein M 2 regina regina (p. 63) - cf upperside of forewing with black border narrower, not reaching base of space 5. $ forewingborder fades out at, or just before, tornus; underside of forewing without a white area regina sriwa (p. 63) THE MILETINI 63 Fig. 42 Logania regina regina (H. Druce); Borneo. Male genitalia. Logania regina regina (H. Druce) (Fig. 42, cf genitalia) Miletus regina H. Druce, 1873: 348, pi. 32, fig. 4 cT. Holotype cT, BORNEO (BMNH) [examined].Logania regina regina (H. Druce); Fruhstorfer, 1914: 23; 1916: 806. Logania evora Fruhstorfer, 1916: 806. Holotype $, PHILIPPINES: 'Sula Is.' recte Sulu Is (SM) [examined]Syn. n. In the male most, if not all, the pale areas on the upperside are clothed with bluish grey scales. The female ismuch whiter, with a narrower forewing border which is usually about 1-5 mm wide at the tornus. On theunderside of the forewing there is sometimes a narrow whitish area along the dorsum, the correspondingarea in the female being much wider and usually extending into the basal half of space 4. DISTRIBUTION. Borneo; Sulu Is (only known from unique holotype). In BMNH there is a single male fromPulo Laut, in which the fore wing border is even narrower than in subsp. sriwa, while the pale areas arewhite, so that the general appearance is of a female; it probably represents a further subspecies. Logania regina sriwa Distant Logania sriwa Distant, 1886a: 531; 18866: 452, pi. 44, fig. 16 $; Evans, 1932: 212. Holotype $, WEST MALAYSIA (not located).Logania regina sriwa Distant; Fruhstorfer, 1914: 23; 1916: 805, pi. 141f $; Corbet, 1940: 112, fig. 2 cf valva; Cantlie, 1967: 28; Fleming, 1975: 22, pi. 58, fig. L52 ?; Eliot, 1978: 241, pi. 20, fig. 9 cf . The male is usually whiter, with a narrower border, than nominate regina. In the female the forewingborder usually fades out at or before the tornus. On the underside both sexes lack a white area above theforewing dorsum, and on the hindwing the white streak is narrower. DISTRIBUTION. West Malaysia; Sumatra; peninsular Thailand (Pinratana, 1981). Logania paluanasp. n. (Fig- 79 9) This taxon, at present only known from two females, appears to replace L. regina, of which it may be asubspecies, in Sulawesi. It is larger, with forewing length 17-0 mm, compared with an average of 14-0 mm inregina, and differs additionally as follows. On the upperside the forewing border is narrower, endingbeyond the middle of the costa. On the hindwing there is a black border measuring nearly 4-0 mm at thedorsum, expanding to 4-5 mm at vein Cu 2 and thence decreasing to 1-0 mm at the apex whence it iscontinued as a blackish line along the costa to the wing base. The underside is generally marked as in 64 J. N. ELIOT regina, but on the hindwing the comparatively well-defined white streak ofregina is replaced by a wider andmore obscure, because more heavily striated, white discal patch extending to vein Cu\ and the lower edgeof the cell. The postdiscal markings are olive-brown and there is a well-defined marginal olive-brown line0-5 mm wide on the forewing and between veins Cu 2 and M 2 on the hindwing. MATERIAL EXAMINED Holotype 9, Sulawesi: labelled /G. Rangkoean, Paloe, West Celebes, 900', Nov. 1936 (J. P. A. Kalis)(BMNH). Paratype. 1 9 , data as holotype (BMNH). Logania marmorata Moore (Fig. 43, cf genitalia)Logania marmorata Moore, 1884: 22. The species can be recognised by its crenulate forewing termen and banded legs, which are shorter andstouter than in any other Logania species. Except in the subspecies from Simeulue, both sexes always havea pale area on the forewing, which is more extensive in the female, but in the male of some subspecies it isreduced to a small discal patch. The hindwing of the male is normally brown, at least in the wet season,while that of the female bears some grey scales; but in Nias and south Sumatra the hindwing is partly grey inthe male and nearly all white in the female. The species ranges from central Burma to Vietnam and throughout Sundaland, the Lesser Sunda Is. andthe Philippines into north Sulawesi. The dividing line between subspecies is difficult to draw because of thehigh degree of individual variation; 10 are provisionally recognised. Key to the subspecies of L. marmorata 1 9 upperside of forewing with basal half whitish 2 - 9 upperside of forewing unmarked brown marmorata diehli (p. 66) 2 (if upperside of hindwing plain brown 3 - cf upperside of hindwing partly white or grey 11 3 cf upperside of forewing with some bluish grey scaling reaching wing base 4 - cf upperside of forewing with brown wing base 6 4 cf upperside of forewing with pale area outwardly white but wing base and space la rather dark, having many brown scales intermixed with the grey. 9 upperside of hindwingsometimes plain brown 5 - cf upperside of forewing with pale area outwardly white becoming light bluish grey at wing base and in space la. $ upperside of hindwing with at least some grey scaling below vein 6 marmorata damis (p. 65) 5 cf upperside of forewing with white extending fully across space Ib. $ underside of forewing without a whitish area beyond cell marmorata hilaeira (p. 65) - cf upperside of forewing with white area not below midspace Ib. $ underside of forewing with a whitish area beyond cell extending from vein A\ to veins M 2 or MI dry season form of marmorata javanica (p. 67) 6 9 underside of forewing without a prominent whitish area beyond the cell 7 9 underside of forewing with a broad whitish area beyond the cell stretching from vein A\ to vein M 2 . Cf upperside of forewing with an ovate discal white patch about 2-5 mm wide at base ofspaces 4,3,2 and sometimes just entering space Ib marmorata munichya (p. 66) 7 cf upperside of forewing with discal white patch at least 2-0 mm wide, reaching and often crossing vein Cu 2 . $ upperside of forewing with base more or less grey-scaled 8 - cf upperside of forewing with discal patch about 1-25 mm wide and not below mid-space 2. 9 upperside of forewing with base brown, bearing only a trace of grey scaling wet season form of marmorata javanica (p. 67) 8 Cf 9 underside comparatively paler and browner ; base and disc of forewing not blackish 9 Cf 9 underside dark; base and disc of forewing blackish 10 9 cf upperside of forewing with white patch usually not below vein Cu 2 . Underside comparative- ly dark. Continental wet season form of marmorata marmorata (p. 65) Cf upperside of forewing with white patch usually crossing vein Cu 2 . Underside comparativelylight. Palawan marmorata palawana (p. 67) THE MILETINI 65 10 9 upperside of hindwing with some grey scaling marmorata samosata (p. 67) - 9 upperside of hindwing brown marmorata Faustina (p. 68) 11 cf upperside of hindwing brown streaked with grey below vein MI. 9 upperside of hindwing mostly dirty whitish below vein MI dry season form of marmorata marmorata (p. 65) - cf upperside of hindwing white below vein M\ except for an inwardly diffuse brown border 1 '0-1 -5 mm wide. 9 upperside of hindwing nearly all white marmorata lahomius (p. 66) Logania marmorata marmorata Moore Logania marmorata Moore, 1884: 22; 1886: 39, pi. 3, fig. 7; de Niceville, 1890: 33, frontispiece, fig. 128 $holotype; Bingham, 1907: 303; Swinhoe, 1910: 200, pi. 618, figs 1, la 'cf' recte <j>; Evans, 1932: 213.Holotype $, BURMA: Mergui (ZSI). [Malaissriwa (Distant) sensu Doherty, 1889: 436. Misidentification.] Logania marmorata marmorataMoore; Fruhstorfer, 1914: 23; 1916: 806, pi. 141f cf $ ; Corbet, 19400: 112;Cantlie, 1963: 29. The subspecies occurs in two seasonal forms, at least in the northern part of its range. In the wet seasonform both wings are brown, with a small white patch on the forewing above vein Cu 2 in the male and a muchlarger white area which is greyish basally in the female. In the dry season the white areas of the forewing aremuch enlarged in both sexes, the hindwing is more or less overlaid below vein MI with grey or whitishscales, especially in the female, and on the underside of the forewing there is no whitish area beyond thecell. Intermediate season examples may be indistinguishable from subsp. damis. DISTRIBUTION. Burma, from Karen Hills to Mergui; Thailand; Vietnam. Logania marmorata damis Fruhstorfer Logania massalia damis Fruhstorfer, 1914: 24; 1916: 807. LECTOTYPE cf , SINGAPORE (BMNH), here designated [examined].Logania marmorata damis Fruhstorfer; Corbet, 1940a: 111; Fleming, 1975: 22, pi. 58, fig. L53 9; Eliot, 1978: 241, pi. 20, fig. 10 cf. In the wet and only seasonal form the pale areas on the forewing are whiter and more extensive than in anyother subspecies except lahomius. In the male the pale area is outwardly white, inwardly rather pale bluishgrey, reaches the dorsum and fills the cell; the hindwing is brown. In the female the pale area is whiter andbroader, and the hindwing always bears at least some grey scales below vein M\. I designate as lectotype of damis a male in BMNH labelled /Type [red]/Singapora II. 95/almost certainlytype of Logania massalia damis Fruhst. [in Corbet's hand]/. DISTRIBUTION. Peninsular Thailand; West Malaysia; Singapore; east coastal region of Sumatra. Logania marmorata hilaeira Fruhstorfer Logania obscura Distant & Pryer, 1887: 266. Syntypes, BORNEO: Sandakan (not located). [Secondary homonym of Logania obscura (Rober, 1886).]Logania marmorata hilaeira Fruhstorfer, 1914: 23; 1916: 806; Corbet, 1940a: 112. LECTOTYPE cf, SUMATRA (BMNH), here designated [examined].Logania marmorata stenosa Fruhstorfer, 1914: 23 (nomen nudum); 1916: 806; Corbet, 1940a: 112. LECTOTYPE 9, BORNEO (BMNH), here designated [examined]. Syn. n.Logania marmorata obscura Distant & Pryer; Fruhstorfer, 1914: 23.Logania massalia nada Fruhstorfer, 1914: 24; 1916: 807. LECTOTYPE cf, SUMATRA (BMNH), here designated [examined]. [Synonymised by Corbet, 1940a: 112.]Logania marmorata cineraria Fruhstorfer, 1916: 806; Corbet, 1940a: 112. [Replacement name for Logania obscura Distant & Pryer, 1887.] Syn. n.Logania massalia sora Fruhstorfer, 1916: 807. LECTOTYPE cf, BORNEO (BMNH), here designated [examined]. Syn. n.Logania marmorata sora Fruhstorfer; Corbet, 1940a: 112. Above, the male has a slightly wider forewing border than subsp. damis, and the base of the wing is darkerwith brown scales intermixed with the grey. The female generally has a darker hindwing which is oftenwithout any grey scales. I designate as lectotype of hilaeira a male in BMNH labelled /Type [red]/CMB IV.94/Fruhstorfer Coll. 66 J. N. ELIOT B.M. 1933-131/marmorata hilaeira Frhst. [in Fruhstorfer's hand]/. A female labelled /Type [red]/Type[Fruhstorfer orange]/Sumatra Montes Battak ex coll. Fruhstorfer/marmorata Selesseh 15.VII.94/Fruh-storfer Coll. B.M. 1933-131/$ Allotype of Logania marmorata hilaeira Fruh. [in Corbet's hand]/ is aparalectotype. In his earlier work (1914) Fruhstorfer did not list or describe Logania marmorata stenosa, either throughan oversight or lapsus calami, but he mentioned stenosa twice by comparison with his sub-spp. javanica andsamosata. In 1916 he gave a brief description and type-locality, so that the name dates from 1916. Idesignate as lectotype of stenosa a female in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Sintang Dr. Martin H. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/Marmorata stenosa Fr. [in Fruhstor-fer's hand]/. I designate as lectotype of nada a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/CMBX.94/Fruhstorfer Coll. B.M. 1933-131/massalia nada Frhst. [in Fruhstorfer's hand]/. I designate as lectotype of sora a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Nord-Borneo ex coll. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/massalia sora Fr. [in Fruhstorfer'shand]/. DISTRIBUTION. Northern and western Sumatra at least as far south as Padang; Borneo, including Pulo Laut;Cagayan Sulu. Logania marmorata lahomius (Kheil) Miletus lahomius Kheil, 1884: 27, pi. 5, figs 28, 29 cf . Syntypes, NIAS (not located).Logania regina lahomius (Kheil) Fruhstorfer, 1914: 23.Logania lahomius (Kheil); Fruhstorfer, 1916: 806.Logania marmorata lahomius (Kheil); Corbet, 1940a: 113. Above, both sexes are greyish white with the forewing border narrower than in the preceding subspecies.The hindwing of the male has a diffuse fuscous margin about 1-5 mm wide and the costal area is fuscousabove vein MI, but in the female the margin is vestigial or absent and there is only a little fuscous dustingbelow the costa. DISTRIBUTION. Nias I. There is a pair in BMNH taken by Doherty at Liwa, in the extreme south-west ofSumatra, which differ only that in the female the fuscous scaling below the hindwing costa is solid as far asvein MI; they are provisionally placed under lahomius. Logania marmorata die/i/i subsp. n. Cf upperside brown; forewing with a small, sullied, circular whitish patch 2-0 mm wide surrounding swollenportion of vein M 3 in spaces 2, 3 and 4. Underside generally pale, with postdiscal markings ill-defined;forewing without a trace of a white or paler discal area, as in the otherwise rather similar subspecies fromJava. $ upperside entirely brown. Underside like male. MATERIAL EXAMINED Holotype cf , Simeulue (Simalur): 16-17. ii. 1984 (E. Dieht) (BMNH).Paratype. 1 $ (allotype), data as holotype (BMNH). The subspecies is named in honour of the captor, Dr Edvard Diehl. It is instructive that the two extremes of geographical variation in this species are found in theneighbouring islands of Nias and Simeulue, the former having the most extensive white markings, the latterthe least; a good example of the haphazard course of evolution in small, isolated populations. Logania marmorata munichya Fruhstorfer Logania massalia munichya Fruhstorfer, 1914: 24; 1916: 807. LECTOTYPE cf , JAVA (BMNH), here designated [examined].Logania marmorata javanica Fruhstorfer; Corbet, 1940a: 112, partim. In this and the remaining subspecies the males, at least in the dry season, have a brown forewing bearing awhite discal patch and without grey scales at the wing base. In munichya the white patch is 2-5 mm wide andlies at the bases of spaces 4, 3 and 2 and just enters space Ib. The female resembles subsp. hilaeira on theupperside but is distinctive in possessing, on the underside of the forewing, a white area 3-0-4-0 mm widebeyond the cell stretching from vein A to vein M 2 . THE MILETINI 67 It seems likely that this subspecies and the east Javanese subsp. javanica may represent the ends of acline, in which case Corbet's action in synonymising munichya with javanica would be justified. I designate as lectotype of munichya a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Java Occident. Sukabumi 2000' ex coll. Fruhstorfer/massalia munichya Frhst. [in Fruhstorfer's hand]/. DISTRIBUTION. West Java. Logania marmorata javanica Fruhstorfer Logania marmorata javanica Fruhstorfer, 1914: 23; 1916: 806; Corbet, 1940a: 112. LECTOTYPE cf , JAVA (BMNH), here designated [examined].Logania massalia glypha Fruhstorfer, 1914: 23; 1916: 807. LECTOTYPE cf, JAVA (BMNH), here designated [examined]. [Synonymised by Corbet, 1940: 112.]Logania marmorata Moore; Piepers & Snellen, 1918: 18, pi. 20, fig. 21 cf[Logania massalia Doherty; Piepers & Snellen, 1918: 18, pi. 20, figs 22a cf , 22b $. Misidentification.] The male differs from subsp. munichya in the smaller white patch on the forewing, which is 1-25 mm wideand does not descend below mid-space 2. The female differs in having the base of the forewing darker and,on the underside, in lacking the prominent white patch beyond the cell, this area being only slightly palerthan the rest of the wing. I designate as lectotype of javanica a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/Ostjava Lawang 1897 ex coll. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/Logania javanica Frhst. [inFruhstorfer's hand]/. Its left forewing is much discoloured, the basal two-thirds of the wing having an oilybluish sheen; the right forewing is undamaged. The figure in Piepers & Snellen (1918), showing only the lefthalf of a male with a basally bluish forewing, was obviously made from the lectotype; Piepers said that hehad himself seen no Javanese examples of L. marmorata. There are also in BMNH two females of theoriginal type-series labelled /Logania javanica Frhst. [in Fruhstorfer's hand]/ which are paralectotypes. I designate as lectotype of glypha a male in BMNH labelled /Type [red]/Ostjava Lawang 1897 ex coll.Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/massalia glypha Frhst. [in Fruhstorfer's hand]/. All the above lectotypes and paralectotypes appear to be wet season forms. There is in BMNH a series of4 cf , 7 $ from Sumbawa labelled /Sumbawa Doherty Sept '91/ which must represent the dry season formand which I provisionally place under subsp. javanica. The males have a white patch on the forewing similarto that of subsp. munichya, but the base of the wing is covered with rather dark bluish grey scales. Thefemales also are rather similar on the upperside to that sex of subsp. munichya, but the hindwing is palerbrown and may bear some grey scales, while on the underside of the forewing the whitish area beyond thecell is narrower and more sullied. There is also in BMNH a single female labelled /S. Flores xi.96. Dry s.Everett/ which, despite the label appears to be a wet season form as might be anticipated from the date ofcapture at the change of seasons. Above, it differs only slightly from females of subsp. javanica in havingthe base of the forewing browner, with only a very few overlying grey scales, while on the underside of theforewing the whitish patch beyond the cell is vestigial. Provisionally I attach it also to subsp. javanica. DISTRIBUTION. East Java; Lesser Sunda Is. (Sumbawa, Flores). Logania marmorata palawana Fruhstorfer Allotinus (Logania) distanti Staudinger, 1889: 93, pi. 1, fig. 3 $. Synt'ypes, PALAWAN (? MNHU). [Secondary homonym of Logania distanti Semper, 1889.]Logania marmorata palawana Fruhstorfer, 1914: 23; 1916: 806. [Replacement name for Logania distanti (Staudinger, 1889).] The male resembles subsp. munichya, but the forewing patch is a little larger and faintly bluish grey. Thefemale is most like subspp. marmorata and hilaeira, especially on the underside, but on the upperside of theforewing the border is a little wider at the tornus. Single females from Luzon in BMNH and Marinduque (coll. Treadaway) appear to belong to thissubspecies. DISTRIBUTION. Palawan; Balabac I. (coll. Treadaway); Luzon; Marinduque. Logania marmorata samosata Fruhstorfer [Logania obscura Distant & Pryer sensu Semper, 1889: 160 partim, pi. 31, fig. 4 $. Misidentification.]Logania marmorata samosata Fruhstorfer, 1914: 23; 1916: 806. Holotype cf , PHILIPPINES: Cebu (SM),[examined]. 68 J. N. ELIOT I have seen only two females and no males from Cebu. The former, which have the lower part of thehindwing lightly grey-scaled, differ additionally from subsp. palawana by a much darker underside with theforewing mostly blackish. A female from Mindoro is similar. Fruhstorfer named the subspecies from Semper's fig. 4, so the specimen depicted therein is automaticallythe holotype. DISTRIBUTION. Cebu; probably Mindoro. Logania marmorata Faustina Fruhstorfer(Fig. 43, cf genitalia) [Logania obscura Distant & Pryer sensu Semper, 1889: 160 partim, pi. 31, fig. 5 $. Misidentification.]Logania marmorata faustina Fruhstorfer, 1914: 23; 1916: 806. Holotype 9, PHILIPPINES: Mindanao (SM),[examined]. All the females I have seen from Mindanao, Samar and Leyte have the upperside of the hindwing plainblackish brown; otherwise they do not differ from subsp. samosata. The males resemble subsp. palawanaon the upperside, but are readily separable by their much darker blackish undersides. Fruhstorfer described the subspecies from Semper's fig. 5, so the female from Mindanao depictedtherein is automatically the holotype. DISTRIBUTION. Mindanao; Leyte; Samar (coll. Treadaway); Sulu Is.: Tawi Tawi (Tite, 1969). Fig. 43 Logania marmorata faustina Fruhstorfer; Mindanao. Male genitalia. Logania obscura (Rober)(Fig. 44, cf genitalia; 80 cT) Allotinus obscurus Rober, 1886: 52, pi. 4, fig. 8 cf . Syntypes, SULAWESI (? SMT). Allotinus martinus Fruhstorfer, 1913: 371; 1916: 814, pi. 141h $. Holotype $, SULAWESI: Buton I.(BMNH) [examined]. Syn. n. Logania donussa Fruhstorfer, 1914: 24. LECTOTYPE $, SULAWESI (BMNH), here designated [ex-amined]. Syn. n. Logania distanti donussa Fruhstorfer; Fruhstorfer, 1916: 807. Logania obscura (Rober) Fruhstorfer, 1916: 807. Logania marmorata obscurus (Rober); Corbet, 1940a: 112. In the past L. obscura has been thought to replace L. marmorata in Sulawesi and its satellite islands, and itwas treated as a subspecies thereof by Corbet. However, the very recent discovery of L. marmorata innorth Sulawesi in a still undescribed subspecies indicates that L. obscura is a distinct species. Its status assuch is further confirmed by the following characters. The female is all brown, whereas the male has a whiteforewing patch - a reversal of the usual sexual differences in Logania marmorata wherein the female has THE MILETINI 69 Fig. 44 Logania obscura Rober; Sulawesi. Male genitalia. more extensive white areas than the male (except in Simeulue). The forewing termen and apex are morerounded and the crenulations are weaker. The legs are longer and thinner. The wing span is considerablylarger. Finally, there is a small difference in the male valva. The holotype of martinus has no head, legs nor right forewing. It bears a label /damaged in shelter byburst pipe, G.E.T./. The remaining wings show no significant difference to normal L. obscura. I designate as lectotype ofdonussa a female in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/S.Celebes Bua-Kraeng 5000' Febr. 1896 H. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/Logania donussaFr. [in Fruhstorfer's hand]/. A single male from Banggai I. in coll. Bedford Russell (Fig. 80) has the forewing patch sullied all overwith fuscous scales, so that it hardly shows up, and the underside is darker and more densely striated than intypical examples. A male from the Toekan Besi Is. in BMNH resembles the Banggai male on theupperside, but on the underside is paler than typical examples. These two males probably represent furtherminor subspecies. DISTRIBUTION. Sulawesi. Logania distant! Semper(Figs 45, 46, cf genitalia; 81 cf )Logania distanti Semper, 1889: 161. In this species I combine the subspecies which have hitherto been treated as pertaining to two distinctspecies, L. distanti and L. massalia. Fruhstorfer, and also Corbet in part, confused the massaliasubspecies-group with L. marmorata, but the two species are easily distinguished by several characters.Firstly, in L. distanti the legs are longer, thinner and not banded, but speckled and streaked with brown ona whitish ground. Secondly, the forewing termen is barely crenulate. Thirdly, the underside is moreochreous and the striations are usually denser. In the male genitalia the phallus is much less strongly bent. The male is brown above, normally with a whitish or bluish grey discal patch on the forewing, but in thePhilippines and Malay Peninsula (and probably elsewhere) the upperside is sometimes unmarked. On theforewing vein M 3 is briefly swollen and clothed with specialised scales. Females show very great individualvariation, the forewing white or very pale greyish blue with a broad blackish border, while the hindwing,except in subsp. staudingeri, varies from brown to mainly whitish in any one area. The species flies from India to the Malay Peninsula, Sumatra, Borneo and the Philippines. It isapparently absent from the islands off the west coast of Sumatra and from Java, records of L. massalia fromthe latter island applying to misidentified L. marmorata. Key to the subspecies of L. distanti 1 cf with upperside of forewing brown to brownish at base. $ with upperside of hindwingvariable, but in the most lightly marked examples always with a fuscous border 1-0-2-0 mmwide .. 2 70 J. N. ELIOT Cf upperside of fo rawing with pale bluish grey area reaching wing base . $ upperside of hindwingall white except for sparse fuscous dusting above vein A/i distant! staudingeri (p. 72) 2 Cf upperside of forewing with discal patch, if present, pale bluish grey 3 Cf upperside of forewing with discal patch , if present , white distant! massalia (p . 70) 3 Philippine distant! distanti (p. 70) Bornean distanti drucei (p. 71) Logania distanti distanti Semper (Fig. 81 Cf ) Logania distanti Semper, 1889: 161, pi. 31, figs 6, 7 $, 21 cf. Holotype <j>, PHILIPPINES: Cebu (SM) [examined]. Logania distanti distanti Semper; Fruhstorfer, 1914: 24; 1916: 807.Logania distanti apsines Fruhstorfer, 1914: 24; 1916: 807. LECTOTYPE cf , PHILIPPINES: Mindanao (SM), here designated [examined].Logania turdeta Fruhstorfer, 1916: 807. Holotype 'cf' recte $, PHILIPPINES: Cebu (SM) [examined]. Syn. n. Semper originally described L. distanti from three examples: a male and female of large size fromMindanao, which he figured at figs 21 and 7 respectively, and a small female from Cebu figured at fig. 6.Although he labelled the male as type he unfortunately did not specify it as such, and this left the way openfor Fruhstorfer later to restrict nominate distanti to Cebu and to name the Mindanao pair as subsp. apsines.Semper's female from Cebu is, therefore, automatically the holotype of distanti. Fruhstorfer 's action, taken at a time when it is evident that he had not even seen Semper's specimens, isdoubly regrettable, since a year later, after examining Semper's collection, he named the Cebu female,which he then inexplicably mistook for a male, as a new species, L. turdeta, and remarked that Semper hadconfused it with L. distanti. At the same time he still recorded nominate distanti from Cebu and maintainedapsines for Mindanao examples. Semper's original female from Cebu, which is the holotype of both distantiand turdeta, is labelled /CEBU/Coll. C. Semper/206/1004/No 6/Typus [and on reverse] Logania turdeta(Fruh.)/. The small ticket reading 'No 6' obviously refers to Semper's fig. 6. The specimen has extensivewhitish dusting on the upperside of the hindwing and agrees fairly well with the butterfly figured by deNiceville (1894: pi. 2, fig. 13) as Logania luca. Semper's pair from Mindanao may or may not be a valid subspecies; but as size, by which aloneFruhstorfer separated them as subsp. apsines, is an unreliable character. I treat apsines provisionally as asynonym of distanti. Fruhstorfer did not designate a type, so I now designate the male as lectotype; it islabelled /Log. Distanti typ. Semper/206/1004/16/Typus [red]/. The small ticket reading '16' should refer tothe figure in Semper's pi. 31; but in fact the specimen is shown at fig. 21, while fig. 16 representsParageryduspunctatus. Possibly the labels got transferred by mistake, as in other cases already mentioned(p. 37) during Fruhstorfer's examination of the collection. Above, the specimen is unmarked brown, butas it is the only Mindanao male I have seen it would be premature to assume that an all-brown forewing is anunvarying character of the male in that island. I have seen no males from Cebu, but have examined single males from Luzon and Negros kindly sent meby Dr Alistair Ballantine and Mr Yusuke Takanami respectively, both of which have been deposited inBMNH. The former is as large as the lectotype of apsines, but has a small diffuse bluish grey discal patch onthe forewing. The latter is smaller, and has a more extensive bluish grey area which nearly reaches the baseof the forewing. Judging by the three males I have seen I expect that it will be found that throughout thePhilippines males most often have a pale discal patch on the forewing, but that unmarked males may occuranywhere from time to time, just as they do in the Malay Peninsula. Apart from the two females in Semper's type-series the only females I have seen are two from Samar incoll. Treadaway. Both have some grey dusting on the upperside of the hindwing, but in neither example is itas extensive as in the holotype of distanti/ turdeta. DISTRIBUTION. Probably throughout the Philippines, where it appears to be everywhere rare. Examplesseen from Luzon, Negros, Cebu, Samar and Mindanao. Logania distanti massalia Doherty stat. n. (Fig. 45, cf genitalia) Logania massalia Doherty, 18916: 37 '$' recte cT; Bingham, 1907: 304; Swinhoe, 1910: 202, pi. 618. figs 3,3b cf , 3a, 3c '$' recte cf ; Evans, 1932: 213. Holotype cf , INDIA: Assam (BMNH) [examined]. THE MILETINI 71 Fig. 45 Logania dlstantl massalia Doherty; Malay Peninsula. Male genitalia. Logania luca de Niceville, 1894: 28, pi. 2, fig. 13 <j>: Evans, 1932: 213. 4 $ syntypes, SUMATRA and WEST MALAYSIA (probably ZSI). Syn. n. Logania luca luca de Niceville; Fruhstorfer, 1914: 24; 1916: 806. Logania massalia massalia Doherty; Fruhstorfer, 1914: 24; 1916: 807, pi. 141f cf ; Corbet, 1940a: 112.Logania massalia luca de Niceville; Corbet, 1940a: 112; Fleming, 1975: 22, pi. 58, fig. L54 cf $; Eliot, 1978: 241, pi. 20, fig. 10 cf. Usually the male has a circular white discal patch more or less dusted with grey scales on the forewing, butvery occasionally the patch extends to the wing base where it is heavily dusted with bluish grey and fuscousscales, as in single males in BMNH from south India and Burma and in a male from the Malay Peninsula incoll. Fleming. Sometimes the white patch is wanting or vestigial, most often in examples from the MalayPeninsula. The hindwing is blackish brown. Females show much individual variation, and cannot be reliably separated from that sex of the nominatesubspecies. The hindwing may be unmarked blackish brown, but nearly always there is at least some greyscaling. De Niceville described and figured as L. luca females from Perak and Sumatra in which thehindwing was more than half whitish, and even more extreme examples with paler undersides occur inSumatra and in Assam and Burma during the dry season. In BMNH there is a female labelled /Type [red]/Perak Jan. -Feb. '90 W. Doherty/Elwes Coll. 1915-207/L. malais Doh. so named in Elwes Coll. /Perak 2c Malais sp. undescribed /Probably not a type but agreesexactly with fig. of luca Nic. in orig. desc. [in Corbet's hand]/. In fact this female, which may conceivably beone of de Niceville's four female syntypes, does not agree closely with de Niceville's figure, since thehindwing has hardly any whitish scaling. DISTRIBUTION. South India (cf , Thantipandal, 60 m. NNW. of Madras); Nepal; north-east India (Sikkim toManipur); Burma; peninsular Thailand; West Malaysia; Sumatra. Logania distant! drucei Moulton Logania drucei Moulton, 1911: 85, fig. 9 cT; Fruhstorfer, 1914: 24; 1916: 807. Holotype cf , BORNEO: Sarawak (BMNH) [examined].Logania massalia drucei Moulton; Corbet, 1940a: 112. Probably inseparable from subsp. distanti, but as I have seen few specimens of either subspecies I maintainit provisionally as distinct. The only males I have seen have a circular bluish grey discal patch on the forewing and the females have abrown or very nearly plain brown hindwing. I have seen no females similar to the figure of luca. DISTRIBUTION. Borneo, excluding Mt Kina Balu where it is replaced by subsp. staudingeri. 72 J. N. ELIOT Logania distant! staudingeri H. H. Druce(Fig. 46, cf genitalia) Logan iastaudingeriH. H. Druce, 1895: 565, pi. 31, figs 13 cf , 14 $ . Syntypes, BORNEO: Mt Kina Balu (coll. Staudinger, probably in MNHU). Logania luca staudingeri Druce; Fruhstorfer, 1914: 24; 1916: 806.Logania marmorata staudingeri H. H. Druce; Corbet, 1940a: 112. The male differs from subsp. drucei in that the bluish grey patch extends to the base and dorsum of theforewing. The female is very pale bluish grey, with the usual blackish apical and marginal border, while thehindwing may be unmarked except for some fuscous scaling above vein Af t and a marginal blackishhairline, or there may be a diffuse border up to 1-0 mm wide. DISTRIBUTION. Only known from Mt Kina Balu. Fig. 46 Logania distanti staudingeri H. H. Druce; Borneo: Kina Balu. Male genitalia. Logania hampsoni Fruhstorfer (Fig. 47 cf genitalia) Logania hampsoni Fruhstorfer, 1914: 25 'cf ' recte $ ; 1916: 807. LECTOTYPE $ , NEW GUINEA (BMNH), here designated [examined].Malais meeki Rothschild, 1915: 387. LECTOTYPE cf, NEW GUINEA: Dampier I. (BMNH), here designated [examined]. Syn. n.Logania masana Fruhstorfer, 1916: 808. Holotype cf, NEW GUINEA (coll. Staudinger, probably in MNHU). Syn. n. Logania hampsoni masana Fruhstorfer; D'Abrera, 1971: 384, figs cf $.Logania hampsoni hampsoni Fruhstorfer; D'Abrera, 1971: 384. The species replaces L. distanti to the east of Weber's Line and is rather doubtfully distinct. It differs chieflyin the browner underside markings and striae, which lack the characteristic ochreous tone of distanti.Individual variation, so conspicuous a feature of L. distanti, is virtually non-existent. The male is plainbrown above, and the female has a white forewing with a broad brown marginal and apical border and abrown hindwing. Fruhstorfer described L. hampsoni from the female, which he mistook for the male, and gave astype-locality Kumusi River in Papua New Guinea. As there were no specimens in his collection he musthave described the species from memory of a series of females from that locality which he had seen in coll.Rothschild. I therefore designate as lectotype a female in BMNH labelled /Lectotype [purple]/Kumusi R.N. E. Brit. N. Guin. low elev. VIII-IX.07 (A. S. Meek)/Rothschild Bequest B.M. 1939-1/Lectotype L.hampsoni Fruhstorfer 'cf ' recte $ , designated by J. N. Eliot Nov. 1983/. Fruhstorfer's error over the sex of L. hampsoni was, no doubt, responsible for the mis-statement by THE MILETINI 73 Fig. 47 Logania hampsoni Fruhstorfer; Papua New Guinea. Male genitalia. D'Abrera (1971) that the male of L. hampsoni hampsoni has a white basal area on the forewing, but canhardly explain his further mis-statement that the type-locality was north-western West Irian. I designate as lectotype of M. meeki a male in BMNH labelled /Type [red]/Dampier Isl. Feb. & March1914 (Meek's Expedition)/Malais meeki Type Rothsch./. DISTRIBUTION. New Guinea, including Dampier I.; New Britain; North Moluccas: Obi I. Logania watsoniana de Niceville sp. rev.(Fig. 48 cf genitalia) Logania watsoniana de Niceville, 1898: 143, pi. Z, figs 17 cf , 18 $; Bingham, 1907: 303; Swinhoe, 1910:201, pi. 618, figs 2, 2b cf , 2a, 2c $ [intermediate form]; Fruhstorfer, 1914: 23. Syntypes, BURMA: NorthShan States (probably ZSI). [Dry season form.] Logania subfasciata Tytler, 1915: 120. Holotype cf, INDIA: Manipur (BMNH) [examined]. [Wet seasonform.] [Synonymised by Cantlie, 1963: 29.] Logania marmorata watsoniana de Niceville; Fruhstorfer, 1916: 806; Corbet, 1940a: 112; Cantlie, 1963: 29. Logania watsoniana watsoniana de Niceville; Evans, 1932: 213. Logania watsoniana subfasciata Tytler; Evans, 1932: 213. Structurally L. watsoniana is closer to L. distanti than to L. marmorata, with which, probably because of itscrenulate forewing, it has sometimes been confused. It can easily be separated from the latter by its legs,which are even longer than those of L. distanti and are buff-brown streaked and speckled with whitish. Fig. 48 Logania watsoniana de Niceville; Burma. Male genitalia. 74 J. N. ELIOT In the wet season form, subfasciata, the male is blackish with a small bluish grey discal patch in spaces 3and 4 of the forewing, just extending into the upper part of space 2 and the base of space 5. The hindwingmay have a few grey scales. The female has a more extensive pale patch, outwardly white and inwardly darkbluish grey, which may extend to the wing base and has its outer edge right-angled at vein M 3 . Theunderside is much like that of L. marmorata. In the dry season form, watsoniana, the pale patch on the forewing of the male extends to the wing baseand is outwardly white and angled on its outer edge, and the hindwing is dusted with grey below vein M\ . Inthe female the pale areas are more extensive and mostly white, and on the underside of the forewing thereis a large white area beyond and below the cell. DISTRIBUTION. Manipur; Burma as far south as Tavoy; Thailand. Genus LONTALIUS gen. n. Type-species: Lontalius eltus sp. n. Gender masculine. The generic name is an anagram oi Allotinus. The genus is described from single females from Pulo Laut and Samar. Eyes glabrous. Antennae similar to those of Paragerydus, with slender shaft and gradually incrassateclub; segments number about 55; nudum tapering almost to base of shaft. Labial palpi comparatively short,resembling those of Logania rather than Allotinus, with third segment barely longer than one-quarter ofsecond segment. Legs (Fig. 49) unlike those of other miletine genera, about as long as those of Allotinus,generally thin, but with tibiae slightly swollen; long first segment of fore-tarsus arched, clothed with longhairs in place of normal scales, while the scales on remaining tarsal segments much longer than those onfemur and tibia but in other respects of normal type. Wings crenulate, with a particularly long tooth onforewing at end of vein M 3 . Venation (Fig. 49) especially distinguished by a strong humeral vein onhindwing which reaches a little over half-way across base of space 8. The single species is moderately large,brown on the upperside and marked on the underside much as in Allotinus (Fabitaras) fabius . This monotypic genus appears to fall between Allotinus and Logania, having the antennae, pattern andsize of the former, while the crenulate wings and short labial palpi are characteristic of the latter, but itdiffers from both in the peculiar fore-tarsi and strong humeral vein. Fig. 49 Lontalius eltus treadawayi subsp. n., $; Samar. Left, venation, indicating system of veins andspaces used in the text; centre, mid-leg; right, fore-leg. THE MILETINI 75 Lontalius eltus sp. n. (Figs 49, venation and legs; 82 9 ; 108 9)The characters are those of the genus. The female sex only is known. Key to the subspecies of L. eltus 1 Upperside reddish brown. Underside with normal lycaenid markings present and partly catenulate on hindwing eltus eltus (p. 75) Upperside dull brown without reddish tinge. Underside with lycaenid markings barely de-veloped eltus treadawayi (p. 75) Lontalius eltus eltus subsp. n. (Fig. 82 9) 9 forewing length 20-0 mm. Upperside reddish brown; cilia pale buff but dark brown at vein endings,broadly so on forewing and narrowly so on hindwing. Underside greyish white, freckled with chocolate-brown specks and striae which are more or less coalesced to give lycaenid-type markings. On forewingthese comprise a series of broad postdiscal spots in spaces 4, 5 and 6, a broad cell-end bar and two spots incell; on hindwing a complete postdiscal series of which the spots in spaces 2, 3 and 4 are inwardlydarker-edged and catenulate, as well as a large spot mid-space 7 and a smaller spot in inner half of space Ib.Cilia as on upperside. Venation differs from that shown in Fig. 49 - on the forewing veins Cu v and M 3 are connate, on thehindwing the humeral vein is straight, while the forewing termen is more strongly dentate at vein Cu 2 . Antennae with 55 segments; ringed on dorsal side of shaft with dark brown and whitish; club whitish inbasal third, remainder dark brown except for unsealed tip; nudum reddish brown tapering gradually towithin four segments of base of shaft. Labial palpi freckled with adpressed dark brown and whitish scales.Legs uniformly pale buff. Body brown above and buff below. MATERIAL EXAMINED Holotype 9, Borneo: Pulo Laut, vi.1891 (W. Doherty) (BMNH). Unique. The specimen is labelled /Type [red]/Puro Laut Borneo June 1891 Doherty/Borneo Allotinus n.sp. 9/Elwes coll. 1915-207/Sp. inc. so named in Elwes coll. /Body loose and fixed 7. v. 1948. A.S.C./Allotinus eltus Cbt. 9 H.T./. Shortly after affixing the last two labels Corbet died, and the name eltus wasnot published. Lontalius eltus treadawayi subsp. n. (Figs 49, venation and legs; 108 9) 9 forewing length 19-5 mm. Upperside dark brown without a reddish tinge. Cilia mainly worn away, butappear to be similar to those of subsp. eltus. Underside greyish white freckled with brown as in subsp. eltus,but lycaenid markings not apparent, the specks being only slightly denser to give an indication of postdiscalspots in spaces 4, 5 and 6 on forewing and on hindwing in spaces 6 and 7 as well as a suggestion of a spot inbasal half of space 7. MATERIAL EXAMINED Holotype 9> Philippines: central Samar, Bagacay, 900 ft, ll.viii.1979 (G. C. Treadaway) (coll.Treadaway). Regrettably the specimen is worn and in tattered condition, with the antennae broken after 38segments and one middle leg and both hind legs missing. < Genus MILETUS Hiibner Miletus Hiibner, 1819: 71. Type-species: Papilla symethus Cramer, 1779, by designation of Doubleday, Westwood & Hewitson, 1852: 502. Gender masculine. Symetha Horsfield, 1828: 59. Type-species: Symetha pandu Horsfield 1828, by monotypy.Gerydus Boisduval, 1836: pi. 23, fig. 2. Type-species: Papilio symethus Cramer, 1779, by original designation.Archaeogerydus Fruhstorfer, 1916: 816. Type-species: Gerydus croton Doherty, 1889, by designation of Hemming, 1960: 9. 76 J. N. ELIOT The genus is instantly recognisable by three external characters. The first segment of the tarsi is long,flattened and blade-like. There is no humeral vein on the hindwing. The pattern on the undersidecomprises normal catenulate lycaenid markings on a ground which is not striated or freckled. The malegenitalia are typical of the tribe, with the arms of the juxta particularly strongly conjoined above thephallus. The twin abdominal hair tufts on the sternum of the eighth abdominal segment are usuallypermanently extruded. The genus is distributed from India to south China and through the Malay Archipelago and Philippinesto New Guinea. It was revised by Eliot (1961), and I have no reason to alter the arrangement proposedtherein with the exception of a few amendments detailed below. However, as there are now a fewadditional species, I have thought it best to include herein a fresh key to the species. Opportunity has beentaken to figure the new species as well as a few previously unfigured taxa, viz. M. nymphis eneus (Fig. 85,Cf), M. cellarius (Figs 86, cf , 87, $), M. heradeion (Figs 94, 96, cf , 95 $), M. mallus mallus (Fig. 83, Cf)and M. celinus (Fig. 104, cf). Key to the species of Miletus 1 cf valva with distal half not tapering distad, dorsal edge not narrowly folded inwards, cf forewing vein M 3 unswollen or basally swollen and clothed with specialised scales 2 Cf valva more or less tapering distad, dorsal edge narrowly folded inwards (Fig. 50D). cfforewing with vein M 3 always swollen and clothed with specialised scales (symethus-group) 2 cf distal third of valva not trough-shaped . cf forewing vein M 3 swollen or unswollen Cf distal third of valva more or less rectangular, with edges curved inwards to form a U-shaped trough (Fig. 50C) . cf forewing vein M 3 swollen (boisduvali-group) 3 cf distal third of valva subspatulate, with dorsal margin less convex than ventral margin, and with a terminal hook at apex (Fig. 50A) (cMnens/s-group) Cf distal third of valva spatulate, with terminal hook very small and in centre line (Fig. SOB)(zinclcenii-group) 4 cf forewing vein M 3 not or only weakly swollen , partly covered with normal scales Cf forewing vein M 3 strongly swollen and clothed with specialised scales M. chinensis 5 Cf $ upperside of forewing with white to whitish markings, which may be heavily sullied in wet season forms, comprising a patch beyond end-cell and smaller and usually separate spots in spaces 2 and Ib. cf forewing vein M 3 not swollen Cf 9 upperside of forewing not so marked . cf forewing vein M 3 weakly swollen or unswollen .... 6 Larger, forewing 20-24 mm. Underside of hindwing rather dark, variegated brown, usually becoming more or less blackish on disc, with reddish brown markings. Underside of forewing without a whitish streak above dorsum M. croton Smaller, forewing 14-22 mm. Underside of hindwing greyish to buff-brown, with markingsonly slightly more reddish than ground colour. Underside of forewing with a more or less developed whitish streak above dorsum M. mallus 1 Cf $ upperside of forewing with a continuous and more or less even white band M. nymphisCf $ upperside of forewing usually unmarked brown, but markings similar to those of M.croton and M. mallus are occasionally faintly discernible (Fig. 84), most often in $ M. gaesa 143 11 C D Fig. 50 Valvae of species-groups of Miletus. A, M. chinensis C. Felder; B, M. zinckenii C. & R. Felder;C, M. boisduvali Moore; D, M. symethus (Cramer). THE MILETINI 77 8 cf vein M 3 strongly swollen and clothed with specialised scales 9 - cf vein M 3 unswollen and clothed with normal scales 10 9 Cf $ upperside of forewing with white band outwardly more or less straight and oblique from vein A i to vein Af 3 , thence curved basad. $ hindwing strongly produced at vein M 3 M. gopara - Cf $ upperside of forewing with white band outwardly angled at vein Cui , below which it is at right angles to dorsum. 9 hindwing normal, barely toothed at vein M 3 M. zinckenii 10 cf $ upperside of hindwing unicolorous brown M. valeus - Cf $ upperside of hindwing mostly white M. gaetulus 11 Cf upperside of forewing with white or whitish markings, except in smokey brown M. drucei sometimes 12 - cf upperside of forewing brown with a sepia tinge and unmarked M. boisduvali 12 cf $ upperside of forewing with white band not extending basad of outer quarter of cell 13 - Cf $ upperside of forewing with broad white band filling outer half of cell. Underside markings very clearly defined (Figs 86, 87) M. cellarius 13 cf very variable; on upperside of forewing white markings may be absent, or form a diffuse whitish spot beyond end-cell in spaces 3-5, or may be clear white and extend into spaces 2and Ib commencing at base of space 2. $ usually with a more or less circular white patch beyond end-cell, but a white band similarto that of well-marked males may be present. Philippines and North Borneo M. drucei - cf less variable; white markings, at their most extensive, do not reach base of space 2 M. biggsii 14 cf outer half of valva tapering more or less evenly and ending in a blunt point or narrowly truncate (Figs 50D, 51) 15 - Cf outer half of valva tapering unevenly, apex broadly truncate or broadly rounded (Figs 52, 53) 24 15 Underside of hindwing with postdiscal markings in spaces 4 and 5 equidistant from end-cell and termen , or very nearly so 16 - Underside of hindwing with postdiscal markings in spaces 4 and 5 much closer to termen than end-cell 23 16 Forewing apex without a protruding point 17 - Forewing apex produced to a short point, terrhen concave in space 6 (Figs 102, 103) M. takanatnii 17 Upperside of hindwing with discocellular veins not darkened , without bluish grey scaling 18 - Upperside of hindwing with discocellular veins more or less darkened, with bluish grey scaling (except in some Philippine subspecies) (Figs 88-9 1 ) M. symethus 18 Upperside forewing base and all hindwing brown (but hindwing may be sullied with white scales in dry season M. ancon) 19 - Upperside forewing base white and hindwing almost all pure white 22 19 Upperside of forewing with white band (if present) constricted and may be completely divided below basal half of vein Cui 20 - Upperside of forewing with white band not constricted below vein Cu\ 21 20 Upperside blackish brown. Forewing with a broad white band from vein Sc to dorsum which is constricted and may be narrowly divided below vein Cu\, at least one-third of space 2white M. ancon - Upperside more reddish brown. Forewing with white markings much reduced (absent in cf of nominate subspecies), with lower part of band widely separated from upper part; in space 2at most a round white spot M. archilochus 21 Underside of hindwing with markings in cell and spaces 3 and 4 darker than other markings. Smaller, forewing 17-18 mm M. gallus - Underside of hindwing with all markings of more or less same intensity. Larger, forewing 19-23 mm M. heracleion 22 Underside of forewing with a black discal area which more or less completely blacks out cell markings M. gigantes - Underside of forewing with cell markings not blacked out M. atimonicus 23 Forewing termen and apex normal M. leos - Forewing apex produced to a short point, termen strongly convex (Fig. 104) M. celinus 24 cf valva with apex broadly truncate, slightly concave (Fig. 52) M. melanion - Cf valva with apex broadly rounded (Fig. 53) M. bazttanus 78 J. N. ELIOT Miletus mallus (Fruhstorfer) (Fig. 83 cT)Gerydus croton mallus Fruhstorfer, 1913: 310. In my original analysis of Miletus (Eliot, 1961) I treated M. mallus as a good species, but later (1967: 72) Isuggested that it was conspecific with, and a subspecies of, M. gaesa. At the same time I erroneouslyrecorded a montane form or microsubspecies of M. gaesa (Fig. 84) from the mountains of the MalayPeninsula as M. gaesa mallus. I now consider that M. mallus and M. gaesa should be treated as distinctallopatric species. Until recently I was only able to examine four specimens of M. mallus from Vietnam, namely the maleholotype of mallus from south Annam, which was taken in the dry season, and the male holotype and twofemales of gethusus, which were taken in Tonkin in the wet season. I presumed, mistakenly, that theserepresented seasonal forms of a single taxon. I have now been able to examine a series of mallus taken byBedford Russell at Dalat in south Vietnam during the wet season. The males differ hardly at all from theholotype of mallus, but differ from gethusus in several particulars, notably in having quite well-developedwhitish postdiscal markings on the upperside of the forewing and a blackish area in and beneath the cell onthe underside. In consequence I consider that gethusus should be reinstated as a valid subspecies of M.mallus flying in north Vietnam. Miletus symethus (Cramer)(Figs 88 $; 89, 90 cf; 91$)Papilio symethus Cramer, 1779: 84. In my analysis (1961) I suggested that the dark subspecies philopator from Mindoro probably occurred inLuzon and other Philippine islands. I have now seen females from Luzon and both sexes from Marinduque,which bear little resemblance to philopator and constitute a new subspecies. Miletus symethus phantus subsp. n. (Fig. 88$) Cf broadly similar to subsp. edonus from Palawan, from which it differs by the reduction of the greyish bluescaling on the hindwing, so that the lower half of the wing looks greyish brown. $ differs from subsp. edonus in the same way as the male, namely in the considerably browner hindwing,which usually has a white streak in the basal three-quarters of spaces 4 and 5 and in the extreme end of thecell. The subspecies approaches subsp. hierophantes (Fig. 89) from the Sulu Is and, in a form with a muchdarker underside (Figs 90, 91), from Mindanao, in which the usual bluish grey scaling on the hindwing isentirely absent, so that the non-white area of the wing is dark brown. MATERIAL EXAMINED Holotype cf , Philippines: Marinduque, iv.1980 (coll. Treadaway). Paratypes. Philippines: 1 $ (allotype), Marinduque, vii.1979 (ex coll. Takanami) (BMNH); 1 $ , data asholotype (coll. Treadaway). Excluded from type-series. Philippines: 2 $ , Luzon, Bicol National Park, 28-29. viii.l 980 (Y. Takanami)(BMNH). These differ from Marinduque females by a larger white area on the underside of the forewing;in addition, in one example (Fig. 88) the white streak on the hindwing is obsolescent. Miletus ? symethus solitarius Okubo Miletus ancon solitaria Okubo, 1983: 176, figs 27, 28 cf . Holotype cf , WEST MALAYSIA: Tioman I. (coll.Okubo). I have not seen the unique holotype from Tioman, an island some 30 miles off the east coast of the MalayPeninsula. Okubo's figures show a butterfly in which the white markings on the forewing comprise atriangular patch beyond the end of the cell which is separated by a broad dark bar from a smaller, elongated THE MILETINI 79 spot in space 2, a larger more or less contiguous spot in space Ib and a narrow white streak in space la abovethe centre of the dorsum. I do not think it can be a subspecies of M. ancon since it possesses severaldiscordant characters. It is smaller than any ancon I have seen, the forewing length being only 18-0 mm; thebase of the forewing and all the hindwing below vein MI are paler; the discocellular veins of the hindwingare darkened; and on the underside of the forewing the markings in the cell are not blacked out. In fact, thebutterfly accords much better with M. symethus than with any other species, and broadly resembles afrequent male variety of the nominate subspecies in Java, in which the white markings are similarlyreduced. The only two males of M. symethus from Tioman which I have examined differ from subsp. petroniusfrom the nearby mainland in having the basal area of the forewing darker and more extensive, so that thewhite markings are reduced though not divided. On these grounds the local Tioman race appears to be agood subspecies. I suspect that the holotype ofsolitarius is just a variety of this race, in which the tendencyfor a darker and enlarged basal area is much increased. I therefore treat the name solitarius conditionally asthat of the symethus subspecies flying in Tioman I. Miletus archilochus siamensis (Godfrey) Gerydus ancon siamensis Godfrey, 1916: 134. Holotype cf , THAILAND: east (BMNH) [examined].Miletus archilochus (Fruhstorfer); Lewis, 1974: pi. 178, fig. 29 cf . In my analysis of 1961 1 retained siamensis as a subspecies of M. ancon, as I thought that the two taxa wereallopatric. Since then I have seen examples of nominate ancon from throughout Thailand apparently flyingsympatrically with siamensis. So wide an overlap argues against conspecificity, and I now think thatsiamensis must be a subspecies of the sibling species M. archilochus. In siamensis the rather sullied whitemarkings on the forewing decrease in extent from west to east. These markings are absent in the male ofnominate archilochus and present, but obscure, in the female. I have seen no examples from the territorybetween Thailand and Tonkin (type-locality of archilochus}, and I anticipate that these will prove to beintermediate and show that variation is clinal. Miletus gigantes (de Niceville)Gerydus gigantes de Niceville, 1894: 23. In my analysis (1961) I considered this species to be a subspecies of M. ancon because of the absence of anyknown overlap in their distribution. Later (1978: 237) I restored gigantes to specific status. The phenotype,with the upperside mainly white, differs greatly from that of M. ancon and, in addition, there is anapparently constant difference in the male genitalia. In continental and Bornean ancon there is only asingle spine-like cornutus in the phallus, whereas in gigantes there are two spines as in most other species ofthe genus. Miletus atimonicus Murayama & Okamura stat. n. (Figs 51, cf genitalia; 92 cf ; 93 $) Miletus sumethus [sic] atimonicus Murayama & Okamura, 1973: 23, figs 45, 46 'cf' recte 9- Holotype 'cf'recte $, PHILIPPINES: Luzon (coll. Murayama) [examined]. Hitherto only known from the female holotype, which the authors mistook for a male of a M. symethussubspecies. Recently I have been sent a male from Negros and a female from Luzon by Mr YusukeTakanami. Both sexes bear a superficial resemblance on both surfaces to M. gaetulus (de Niceville, 1894),and on the upperside to M. gigantes. In the male the basal third of vein M 3 of the forewing is swollen andclothed with specialised scales; this character immediately separates it from M. gaetulus wherein it isabsent. It also differs from the latter on the underside by a browner ground colour and more reducedmarkings in the forewing cell. The underside is not at all like that of M. gigantes, lacking the prominentblackish discal areas found on the forewing of that species. The male genitalia indicate that M. atimonicus belongs to the symethus-group. However, the valva istruncate just before the apex and shows a slight approach to the valva of M. melanion. The phallus, with asingle spine-like cornutus, recalls that organ in M. ancon. 80 J. N. ELIOT Fig. 51 Miletus atimonicus Murayama & Okamura; Negros. Male genitalia. Miletus heracleion (Doherty) (Figs 94, 96 cf; 95$)Gerydus heracleion Doherty, 1891ft: 36. This species has not been recorded from Sumatra, but there is in BMNH a male (West Sumatra, LebongTandai, March 1923, C. J. Brooks) which is somewhat larger but otherwise hardly differs from thenominate subspecies. There is also a male from Sarawak (Bidi, February 1909, C. J. Brooks) (Fig. 94)which is similar to the Sumatran male except that on the underside of the forewing the white band crossedby dark-dusted veins is little more than half as wide. It, and a female from Pulo Laut (Fig. 95) previouslyrecorded as subsp. arion (Fig. 96) (Eliot, 1961: 171), but which has the white band on the forewingconsiderably narrower on the upperside but wider on the underside than in that subspecies, possiblyrepresent a further minor subspecies, but they are provisionally placed under subsp. heracleion. Miletus melanionC. & R. Felder(Figs 52, cf genitalia; 97, 98 C?)Miletus melanion C. & R. Felder, 1865: 284. It has been found that melanion, as treated by all authors up to and including Eliot (1961), comprises twosuperficially similar species with different male genitalia, namely M. melanion and M. bazilanus (Fruhstor-fer, 1913). The former occurs throughout the Philippines (excluding Palawan) at least as far south asMindanao, whilst the latter is at present known only from Mindanao and Bazilan. This distributionsuggests that the two species evolved in the northern and southern island groups respectively, and thatmelanion was later able to spread south into Mindanao, where to-day it is a common species, withoutinterbreeding with the endemic bazilanus. M. melanion exhibits a good deal of individual variation, but broadly the males can be broken down intotwo dimorphs, which at first sight look as though they should represent distinct species. The male genitalia,however, reveal no significant differences, and the females cannot be separated into two groups. In thetypical male (Fig. 97), figured by the Felders (1865: pi. 35, figs 32, 33), the brown forewing has a short,more or less sullied white subtornal streak in space Ib and the swollen portion of vein M 3 is not surroundedby a white area, although a thin band of whitish scales below and a sullied white spot above the swelling maybe present. On the underside of the forewing there are sullied white subtornal spots in spaces Ib and 2 and asullied whitish streak beyond the cell. The dimorph, cf-f. albiguttatus f. n. (fig. 98) (holotype cf , Luzon,Lorquin, ex Felder coll. (BMNH)) is generally smaller and more blackish brown above; there is anadditional whitish subtornal spot in space 2 and a prominent, usually clear white, more or less triangularspot surrounding the swollen portion of vein M 3 , with its base extending from mid-space 3 to the cell apex. THE MILETINI 81 On the underside of the forewing there is a broad white area crossed by dark-dusted veins, extending fromthe dorsum to vein MI. The female generally resembles cf-f. albiguttatus, but the white spots are usually larger. I do not thinkthey are ever conjoined into a continuous white band, such as occurs frequently in M. bazilanus. In the male genitalia the valva is slightly upturned and truncate before the apex and its outer edge isslightly concave. Key to the subspecies of M. melanion 1 cf dimorphic. $ upperside of forewing with white markings comprising a sullied streak in spacela, a double subtornal spot in space Ib, an overlapping spot in space 2 separated by a dark barfrom a much larger spot beyond the cell which extends above vein R 5 , often as far as vein Sc melanion melanion (p. 81) Cf monomorphic, apparently only f. melanion being present. $ upperside of forewing withsubtornal white markings normally reduced to an often sullied white streak in upper part ofspace Ib, with the white spot beyond the cell smaller, not extending above vein R 5 melanion euphranor (p. 82) Miletus melanion melanion C. & R. Felder(Figs 52 cf genitalia; 97, 98 cf ) Miletus melanion C. & R. Felder, [1865]: 284, pi. 35, figs 32, 33 cf . LECTOTYPE cf , PHILIPPINES: Luzon (BMNH), here designated [examined]. Gerydus melanion (C. & R. Felder); Semper, 1889: 161, pi. 31, figs 13, 14 $.Gerydus melanion melanion (Felder); Fruhstorfer, 1913: 246; 1916: 822, pi. 14d cf .Miletus melanion melanion C. & R. Felder; Eliot, 1961: 175, partim. The characters of the subspecies are given in the key. Judging by material in BMNH, in Luzon f. melanion nearly always has a small, often sullied, white spotabove the swollen portion of vein M 3 . In the southern group of islands this white spot is seldom present,nearly all males resembling Fig. 97. It seems that fully marked f. albiguttatus is rare in Luzon, and manyexamples are intermediate to f. melanion; elsewhere the two forms remain more sharply distinct. In BMNH there are three males from the Felders' type-series, one of which is f. albiguttatus alreadydesignated as holotype. The other two are f. melanion with a sullied white spot above vein M 3 . 1 designateas lectotype the one which most nearly resembles the example figured by the Felders and has the white spotvery small and sullied; it is labelled /80/Luzon Lorquin [round blue]/Melanion n. /Felder Colin, [roundwhite]/Rothschild Bequest B.M. 1939-1/. DISTRIBUTION. Throughout the Philippines, excluding Mindoro, Palawan and the Sulu Is. Fig. 52 Miletus melanion melanion cf-f. albiguttatus f. n.; Negros. Male genitalia. Left, above, latero-dorsal view of interior of left valva and, below, ventral view of both valvae. 82 J. N. ELIOT Miletus melanion euphranor (Fruhstorfer) Gerydus melanion euphranor Fruhstorfer, 1914: 60; 1916: 822. LECTOTYPE cf , PHILIPPINES: Mindoro (BMNH), here designated [examined].Miletus melanion melanion C. & R. Felder; Eliot, 1961: 175, partim, fig. 24 cf genitalia. The characters of this dark subspecies are given in the key. In BMNH there are two male and one female syntypes. I designate as lectotype one of the males,labelled: /Syntype [blue]/Baco Dist. Mindoro 6.5.09/Adams Bequest B.M. 1912-399/; its genitalia werefigured by Eliot (1961: fig. 24). The other male and female are paralectotypes. DISTRIBUTION. Mindoro. Also Leyte, whence some examples in BMNH tend to be even darker than thosefrom Mindoro whilst others (Fig. 97) are inseparable from Luzon examples; it might have been expectedthat subsp. melanion alone would have occurred there. Miletus bazilanus (Fruhstorfer) stat. n.(Figs 53 cf , genitalia; 99 cf ; 100, 101 $) Gerydus melanion bazilanus Fruhstorfer, 1913: 246; 1916: 822. LECTOTYPE cf , PHILIPPINES: Bazilan (BMNH), here designated [examined].Gerydus melanion vitelianus Fruhstorfer, 1913: 246; 1916: 822. LECTOTYPE $, PHILIPPINES: Mindanao (BMNH), here designated [examined]. Syn. n.Miletus melanion vitelianus (Fruhstorfer) Eliot, 1961: 175, partim.Miletus melanion bazilanus (Fruhstorfer) Eliot, 1961: 175. The male is very similar to sympatric M. melanion f. melanion, but on the upperside of the forewing thewhite streak in space Ib is usually more prominent and about 3-5 mm long; the underside is slightly darkerand the postdiscal markings are wider. However, to be absolutely certain of identification some males mayneed dissection. The females generally have the white markings more extensive than M. melanion, and in typicalexamples (Fig. 100) there is a continuous white band on the forewing extending from the dorsum to vein Sc.However, some examples (Fig. 101) resemble normal melanion females, except by a slightly darkerunderside with wider postdiscal markings, and these are connected by intermediates to the typical form. The male genitalia differ from those of M. melanion in three respects. The valva is distally broader, withthe tip rounded; the paired, spine-like cornuti in the phallus are more than twice as large; the unci have amore elongated tip. I designate as lectotype of bazilanus a male in BMNH labelled /Type [red]/Type [Fruhstorfer orange]/ Fig. 53 Miletus bazilanus (Fruhstorfer); Mindanao. Male genitalia. Left, above, latero-dorsal view ofinterior of left valva and, below, ventral view of both valvae. THE MILETINI 83 Philippinen Bazilan II-III 98 Doherty ex coll. Fruhstorfer/Fruhstorfer Coll. B.M. 1933-131/melanionbazilanus Frhst. [in Fruhstorfer's hand]/. I designate as lectotype of vitelianus a female in BMNH labelled /Type [red]/Type [Fruhstorferorangej/Fruhstorfer Coll. B.M. 1933-131/melanion vitelianus Frhst. [in Fruhstorfer's hand]/; it has veinCui dark-dusted across the forewing white band. There are no males from Mindanao ex Fruhstorfer coll. inBMNH, but his description of the male accords reasonably well with bazilanus. DISTRIBUTION. Mindanao; Bazilan. Miletus takanamiisp. n. (Figs 54, cf genitalia; 102 cf ; 103 ) Cf forewing length 19-0 mm. Body and labial palpi dark brown. Legs buff-brown. Wing shape similar tothat of the Celebesian M. celinus (Fig. 104), with the forewing concave in space 6 below a pointed apex, andwith the basal quarter of vein M 3 swollen and clothed with the usual minute wedge-shaped specialisedscales. Upperside blackish brown ; forewing with a white patch 3-0 mm wide surrounding swollen portion ofvein M 3 , extending from mid-space 2 across spaces 3 and 4 and into space 5 as a narrow projection, and awhite streak 5-5 mm long and 1-0 mm wide just below centre of vein Cu 2 . Underside paler brown becomingreddish near middle of forewing termen; usual markings reddish brown, outlined by black lines except onouter side of postdiscal series of hindwing. Forewing with an irregular white area extending from just abovedorsum to vein R 5 , 5-0 mm wide in space Ib, increasing to 6-5 mm in space 2, then decreasing to 3-5 mm inspace 6. Genitalia with valva tapering to a rounded point, as in most species of the symethus-group, phallusshort, much as in M. melanion and M. bazilanus, but lacking cornuti. 9 forewing 18-0 mm. Forewing termen more convex than in male, hindwing slightly dentate at vein M 3 ,as usual in the genus. Upperside blackish brown; forewing with an irregular, oblique white band runningfrom vein A l , where it is 5-0 mm wide, to vein /? 1? where it is 2-0 mm wide, and entering outer part of cell;veins Cui and Cu 2 are dark-dusted where they cross this band. Underside generally similar to male, but onhindwing lower half of area between discal and postdiscal spots dusted with black scales. MATERIAL EXAMINED Holotype cf , Philippines: Mindanao, Tandag, Surigao, 1981-1982 (ex coll. Takanami) (BMNH).Paratype. 1 $ (allotype), ii.1982, data otherwise as holotype (BMNH). Apart from its distinctive wing shape the male most nearly resembles M. melanion f. albiguttatus, fromwhich it differs in having a larger white patch astride the swollen portion of vein M 3 and in lacking a smallwhite subtornal spot in space 2 and, on the underside, in a darker and more irregular ground colour and C P Fig. 54 Miletus takanamii sp. n. ; Mindanao. Male genitalia. Lower left, latero-dorsal view of interior ofleft valva. 84 J- N. ELIOT wider white area on the forewing. The female is much like that sex of M. bazilanus, but on the undersidethe ground colour is much darker and the white central area of the forewing is wider.The species is named after Mr Yusuke Takanami, who generously presented the types to the BMNH. Genus MEGALOPALPUS Rober Megalopalpus Rober, 1886: 51. Type-species: Megalopalpus simplex Rober, 1886: 51, pi. 4, fig. 4, byoriginal designation. Gender masculine. [See also Opinion 566, 1959.] This genus appears to be closely allied to Miletus, of which it should perhaps be treated as a subgenus,having similar long legs with the foretarsi flattened and blade-like, and the undersurface of the wingssimilarly patterned with catenulate markings, although these are sometimes obsolete. The only significantdifference lies in the presence of a strongly developed humeral vein in Megalopalpus. Males also lacksecondary sexual characters, but this condition is matched in some species belonging to the Orientalgenera. The male genitalia are of the normal pattern for the tribe, but with some distinctive characters: theuncus/tegumen plates are triangular and bear a broad, lobe-like process directed ventrad, and the brachiaare curved (see illustrations in Bethune-Baker, 1914; Stempffer, 1967; Eliot, 1973). The genus is restricted to the forested regions of West Africa, from Liberia south to Angola and east toUganda. The larvae are predators of Jassidae and Membracidae (Cottrell, 1984). Gilbert (1976) recordsand illustrates an adult feeding from the 'nectary' of a lycaenid larva. Species limits within Megalopalpus are very imperfectly understood. Aurivillius (1922), followed byStempffer (1967) and Cottrell (1984), accepted four species: metaleucus, simplex, zymna and angulosus.Of these, Stempffer (1967) dissected specimens of the first three, and stated that 'The male genitalia of M.zymna and M. metaleucus are very similar to those of simplex'. D'Abrera (1980) and Carcasson (1981),however, following Peters (1952), recognised only three species, zymna (to include simplex), metaleucusand angulosus - the last, representing the species not dissected by Stempffer, they regarded as doubtful,and possibly only a synonym of zymna. Berger (1981) also recognised only three species but, in contrast,these were zymna, simplex and metaleucus - no mention was made of angulosus (confounded by Bergerwith metaleucus - see below). In the BMNH as apparently curated by Ms S. J. May, all four species listed by Aurivillius and Stempfferare purportedly represented. However, it is very difficult to see how simplex differs consistently fromzymna, and under the former the following note appears: 'Probably some simplex in zymna series, as thesewere hitherto considered to be synonymous. SJM.'. This note is doubly puzzling because not only have thetwo often been considered to be separate, there is no evidence of comprehensive dissection of thespecimens segregated as simplex (and therefore, presumably, their distinguishing features must have beenopen to inspection). The basis of this separation, if indeed it is valid, is thus unclear or unknown. Carcasson(1981) implies, effectively, that there are just two species: large (metaleucus) and small (zymna).Unfortunately, the type of zymna is of intermediate size! However, there seems little doubt that there areat least three species, corresponding to the arrangement in Peters (1952) and illustrated by D'Abrera(1980). All four nominal species are grossly sympatric in some areas (e.g. Cameroun, Zaire), althoughClench (1965) noted a 'strong correlation with geography' for the four forms (including ? simplex) that hedescribed under zymna. Detailed field observations and reared material are probably essential for thesolution of this problem. In the following key, simplex and zymna are separated following Aurivillius (1922) and Berger (1981),but this is not very reliable if the identifications in the BMNH are correct; see also Clench (1965). Key to the species of the genus Megalopalpus 1 Larger species, forewing length 17-23 mm; black border of forewing upperside does not extend broadly quite as far posterior as vein A\\ catenulate markings of hindwing underside darker than general ground colour, not solely forming a series of concentric bands 2 Smaller species, forewing length 11-18-5 mm; black border of forewing upperside extendsbroadly and fully to vein A\; catenulate markings of hindwing underside not appreciablydarker than ground colour, outlined by pale scales, forming a series of roughly concentricbands 3 2 Hindwing upperside very narrowly bordered (up to 1 mm) with dark scales along posterior margin; postdiscal catenulate band of hindwing underside disrupted medially; hindwing underside pattern with overall appearance of veined marble metaleucus (p. 85) - Hindwing upperside usually more broadly bordered (2-3 mm) with dark scales, affecting entire THE MILETINI 85 outer margin; postdiscal catenulate band of hindwing underside reticulate medially; hindwing underside pattern with a more complex marbling effect angulosus (p. 85) 3 Hindwing upperside with dark marginal band 3-5 mm in width, not interrupted in middle zymna (p. 85) Hindwing upperside with dark marginal band no more than 3 mm in width (often much less),medially very narrow or completely interrupted simplex (p. 85) Megalopalpus angulosus Griinberg Megalopalpus angulosus Grunberg, 1910: 478; D'Abrera, 1980: 469 (illustr.). 3 cf, 3 $ syntypes, EQUATORIAL GUINEA: Makomo and Alcu (depository unknown).[Megalopalpus metaleucus Karsch; Berger, 1981: pi. 197, fig. 5. Misidentification.] This butterfly, judging by the underside and if correctly identified in the BMNH, is a good species and notmerely a maculated or seasonal form of metaleucus . M. angulosus is known from Cameroun, EquatorialGuinea and Zaire. Specimens in the BMNH have a forewing length of 20-23 mm. Megalopalpus metaleucus Karsch [Allotinus zymna (Westwood); Grose-Smith & Kirby, 1891: figs 1, 2. Misidentification.]Megalopalpus metaleucus Karsch, 1893: 217; Clench, 1965: 326, figs 198-201; D'Abrera, 1980: 470(illustr.); Berger, 1981: pi. 197, figs 3, 4. 2 cf syntypes, TOGO: Bismarckburg (7MNHU) [not examined]. M. metaleucus is considered to occur from Liberia and Ivory Coast south to Cameroun and Zaire, and eastto Uganda. Specimens in the BMNH have a forewing length range of 17-22 mm. Clench (1965) discusses avariation of the species. Megalopalpus simplex Rober Megalopalpus simplex Rober, 1886: 51, pi. 4, fig. 4. $ syntype(s), EQUATORIAL AFRICA: 'Borneo' (patria falsa) (?SMT) [not examined].Liptena bicoloria Capronnier, 1889: 121. Holotype (sex?), ZAIRE: Lopori Shoven (Martini) (depository unknown). [Synonymy from Stempffer, 1967.]Allotinus similis Kirby, 1890: 262. Syntype(s), (s'ex?), CAMEROUN: Barombi (Preuss.) (MNHU) [not examined]. [Synonymy from Aurivillius, 1922; Stempffer, 1967.]Megalopalpus gigas Bethune-Baker, 1914: 335, pi. 58, figs 9, 9a. cf syntypes, CAMEROUN/GABON (BMNH) [examined]. [Synonymy from Stempffer, 1967.] Material in BMNH identified as simplex is very variable with respect to the hindwing dark border (see alsozymna below). As already discussed, the basis of the separation of simplex from zymna is elusive, and maynot be real. The identification of specimens illustrated by Berger (1981) to represent both species should beregarded with caution. The forewing length of the BMNH material varies from 14-0-18-5 mm. Thedistributional range is thought to include Ghana, Liberia, Nigeria, Cameroun, Gabon, Zaire and Uganda. Megalopalpus zymna (Westwood) Pentila zymna Westwood, 1851: pi. 76, fig. 7; 1852: 503. 1 cf syntype, NIGERIA: Ashanti (ex Wesleyan Missionary Society) (BMNH) [examined].Megalopalpus zymna f. pallida Aurivillius, 1922: 362. Holotype cf , UGANDA: Ruwenzori (?Stockholm) [not examined]. As already noted, it is doubtful if simplex can be reliably separated from zymna. However, Clench (1965)recognised four more or less distinct forms under zymna, including one apparently corresponding tosimplex, and the possibility that zymna s.l. represents a species complex must be considered. Specimensidentified in the BMNH as zymna have a forewing length of 11-18 mm, with the lone syntype at 18-5 mm(despite this, the syntype is definitely of the correct general facies - it is not angulosus or metaleucus ascurrently understood). The distribution of zymna encompasses that given for simplex, and is considered toextend also to Macias Nguema (Fernando Poo), Equatorial Guinea, Angola, southern Sudan andZimbabwe. 86 J. N. ELIOT References Aurivillius, P. O. C. 1908-1925. The African Rhopalocera (translated by L. B. Prout). In Seitz, A., Die Gross-Schmetterlinge der Erde (2) 13: 11-613, 80 pis. Stuttgart.Berger, L. A. 1981. Les Papillons du Zaire. 543 pp. (incl. 213 pis). Bruxelles.Bethune-Baker, G. T. 1914. Notes on the taxonomic value of the genital armature in Lepidoptera. Transactions of the Entomological Society of London 1914: 314-335, 11 pis. Bingham, C. T. 1907. The fauna of British India. Butterflies, 2. viii + 480 pp., 20 pis, 104 figs. London.Boisduval, J. B. 1836. In Roret, Suite a Buffon, Histoire naturelle des Insectes. Species general des Lepidopteres 1: xii + 690 pp., 24 pis. Paris.Butler, A. G. 1884. The Lepidoptera collected during the recent expedition of H. M.S. Challenger, Pt. II. Annals and Magazine of Natural History (5) 13: 183-203. Cantlie, K. 1963. The Lycaenidae portion (except the Arhopala group) of Brigadier Evans' The Identifica-tion of Indian Butterflies, 1932 (India, Pakistan, Ceylon, Burma), vi + 156 pp., 5 pis. Bombay.Capronnier, J. B. 1889. Liste des Lepidopteres captures au Congo par Messieurs Thys, Legat, Martini et Machado en 1887. Bulletin de la Societe Entomologique de Belgique, Bruxelles 33: cxviii-cxxvi.Carcasson, R. H. 1981. Collins handguide to the butterflies of Africa, xix + 188 pp. London.Clench, H. K. 1965. Superfamily Lycaenoidea. In Fox, R. M., Lindsey, A. W. Jr, Clench, H. K. & Miller, L. D. , The butterflies of Liberia. Memoirs of the American Entomological Society: 267-403.Corbet, A. S. 1938. Observations on certain species of Rhopalocera from the Malay Peninsula. Journal of the Federated Malay States Museums 18: 242-262. - 1939a. A revision of the Malayan species of Miletus Hiibner (= Gerydus Boisduval) (Lepidoptera:Lycaenidae). Proceedings of the Royal Entomological Society of London (B) 8: 25-31, 19 figs. - 19396. A revision of the Malayan species oiAllotinus Felder & Felder (Lepidoptera: Lycaenidae).Transactions of the Royal Entomological Society of London 89: 63-78, 1 pi., 21 figs. 19400. A revision of the Malayan species of Logania Dist. (Lepidoptera: Lycaenidae). Proceedings ofthe Royal Entomological Society of London (B) 9: 111-112, 4 figs. 19406. A revision of the Malayan species of Poritiinae (Lepidoptera: Lycaenidae). Transactions of theRoyal Entomological Society of London 90: 337-350, 1 pi., 21 figs. 1956. In Corbet, A. S. & Pendlebury, H. M., The Butterflies of the Malay Peninsula, edn 2. xi + 537 pp., 55 pis, 159 figs. Edinburgh.Cottrell, C. B. 1984. Aphytophagy in butterflies: its relationship to myrmecophily. Zoological Journal of the Linnean Society, London 80: 1-57.Cowan, C. F. 1966. Some type specimens of W. L. Distant (Lep., Lye.). Proceedings of the Royal Entomological Society of London (B) 35: 5-6. Cramer, P. 1779. De uitlandsche Kapellen; Papillons exotiques, 2:2 + 151 pp., pis 97-192. Amsterdam.D'Abrera, B. 1971. Butterflies of the Australian Region. 415 pp. Melbourne.- 1978. Ibidem, edn 2: 415 pp. Melbourne. 1980. Butterflies of the Afrotropical Region, xx + 593 pp. Melbourne.de Niceville, L. 1886. On some new Indian butterflies. Journal of the Bombay Natural History Society 5: 249-255, 1 pi. - 1890. The butterflies of India, Burmah and Ceylon. 3: xi + 503 pp., 6 pis. Calcutta. - 1894. On new and little-known butterflies from the Indo-Malayan Region. Journal of the AsiaticSociety of Bengal (2) 63: 1-59, 5 pis. - 1895. On new and little-known Lepidoptera from the Indo-Malayan region. Journal of the BombayNatural History Society 10: 13-40, 3 pis. 1898. On new and little-known butterflies from the Indo-Malayan and Australian Regions. Journal of the Bombay Natural History Society 12: 131-160, 4 pis.de Niceville, L. & Martin, L. 1895. A list of the butterflies of Sumatra with especial reference to the species occurring in the north-east of the island. Journal of the Asiatic Society of Bengal (2) 64: 357-555.Distant, W. L. 1882-1886^. Rhopalocera Malayana (1882-1886). 16 + 481 pp., 41 pis. London. 18866. Contributions to a knowledge of Malayan entomology, Pt. V. Annals and Magazine of Natural History (5) 17: 530-532.Distant, W. L. & Pryer, W. B. 1887. On the Rhopalocera of northern Borneo, Pt. 2. Annals and Magazine of Natural History (5) 19: 264-275.Doherty, W. 1886. A list of butterflies taken in Kumaon. Journal of the Asiatic Society of Bengal (2) 55: 103-140.- 1889. Certain Lycaenidae from Lower Tenasserim. Journal of the Asiatic Society of Bengal (2) 58: 409^40, 1 pi. THE MILETINI 87 1891a. A list of the butterflies of Engano. Journal of the Asiatic Society of Bengal (2) 60: 4-32, 4 figs. 18916. New and rare Indian Lycaenidae. Journal of the Asiatic Society of Bengal (2) 60: 32-38, 4 figs. 1891c. The butterflies of Sumba and Sumbawa, with some account of the Island of Sumba. Journal of the Asiatic Society of Bengal (2) 60: 141-197, 1 pi.Druce, H. 1873. A list of the collections of diurnal Lepidoptera made by Mr Lowe in Borneo. Proceedings of the Zoological Society of London 1873: 337-361, 2 pis.1874. List of 79 species of butterflies collected by L. Layard in Siam. Proceedings of the Zoological Society of London 1874: 102-109, 1 pi.Druce, H. H. 1895. A monograph of the Bornean Lycaenidae. Proceedings of the Zoological Society of London 1895: 556-627, 4 pis. 1896. Further contributions to our knowledge of the Bornean Lycaenidae. Proceedings of the Zoological Society of London 1896: 650-683, 3 pis. Eliot, J. N. 1956. New and little-known Rhopalocera from the Oriental Region. Bulletin of the Raffles Museum 27: 32-38, 3 figs.1959. New or little-known butterflies from Malaya. Bulletin of the British Museum (Natural History) (Entomology) 7: 371-391, 1 pi., 7 figs.- 1961. An analysis of the genus Miletus Hiibner (Lepidoptera: Lycaenidae). Bulletin of the Raffles Museum 26: 154-177, 24 figs. 1962. On some Lycaenidae and Hesperiidae from Malaya. Entomologist 95: 217-225, 1 pi., 2 figs. 1967. Revisional notes on Oriental butterflies, with special reference to Malaya, Pt. 2. Entomologist 99: 66-72, 5 figs.1973. The higher classification of the Lycaenidae (Lepidoptera): a tentative arrangement. Bulletin of the British Museum (Natural History) (Entomology) 28: 373-505, 6 pis, 162 figs.1978. In Corbet, A. S. & Pendlebury, H. M., The Butterflies of the Malay Peninsula, edn 3: xiv + 578 pp., 36 pis, 146 + 438 figs. Kuala Lumpur. 1980. New information on the butterflies of the Malay Peninsula. Malayan Nature Journal 33: 137-155, 16 figs. 1984. On some butterflies from Malaysia and Peninsular Thailand. Malayan Nature Journal 38: 99-1 11, 8 figs. Eliot, J. N. & Kawazoe, A. 1983. Blue butterflies of the Lycaenopsis group. 309 pp. , 560 figs. London.Elwes, H. J. 1893. On butterflies collected by Mr W. Doherty in the Naga and Karen Hills and in Perak. Proceedings of the Zoological Society of London 1892: 617-644, 2 pis. Evans, W. H. 1932. The identification of Indian butterflies, edn 2: x + 454 pp., 32 pis. Madras.Felder, C. 1862. Verzeichniss der von den Naturforschern der K. K. Fregatte 'Novara' gesammelten Macrolepidopteren. Verhandlungen der Zoologisch-Botanischen Gesellschaft in Wien 12: 473^96.Felder, C. & Felder, R. 1865. Reise der Osterreichischen Fregatte Novara 2 (Rhopalocera): 6 + 548 pp. , 137 pis. Wien. Fleming, W. A. 1975 . Butterflies of West Malaysia and Singapore 2: ix + 93 pp. , pis 35-90. Kuala Lumpur.Fruhstorfer, H. 1913-1915. Ubersicht der Gerydinae und Diagnosen neuer oder verkannter Formen. Zeitschrift fur Wissenschaftliche Insektenbiologie 9: 242-247; 307-310; 341-344; 367-371; 10: 20-25; 59-63; 11: 267-269.1916. In Seitz, A. , Die Gross-Schmetterlinge der Erde 9, Gattungen Gerydus, Allotinus, Logania, pp. 803-823. Stuttgart.Gilbert, L. E. 1976. Adult resources in butterflies: African lycaenid Megalopalpus feeds on larval nectary. Biotropica, Washington 8: 282-283. Godfrey, J. 1916. The butterflies of Siam. Journal of the Natural History Society of Siam 2: 106-147.Grose-Smith, H. 1893. Descriptions of two new species of butterflies from Kina Balu, North Borneo, in the collection of H. Grose-Smith, captured by Mr Everett. Annals and Magazine of Natural History (6) 12: 34-35.Grose-Smith, H. & Kirby, W. F. 1891. Lycaenidae (African). Plate 12. Rhopalocera Exotica 1: 49-51, 1 pi. London.Grunberg, K. 1910. Neue westafrikanische Lepidopteren. Sitzungsberichten der Gesellschaft Naturfors- chender Freunde zu Berlin 10: 469-480.Guerin-Meneville, F. E. 1830. In Duperry, L. I. etal., Voyage autour du Monde, execute par ordre du Roi, surla Corvette desa Majeste, la Coquille, pendant 1822-1 825 ... I (Zoologie) 2, pp. 9-320, 21 pis. Paris.Hemming, A. F. 1960. Annotationes lepidopterologicae, Pt. 1, 36 pp. London.Holland, W. J. 1891. Asiatic Lepidoptera. List of the diurnal lepidoptera taken by Mr William Doherty of Cincinnati, in Celebes, June and July 1887, with descriptions of some apparently new forms. Proceedings of the Boston Society of Natural History 35: 52-82, 3 pis. 88 J. N. ELIOT 1900. The Lepidoptera of Buru. Part 1, Rhopalocera. Novitates Zoologicae 7: 5485. Horsfield, T. 1828. Descriptive catalogue of the lepidopterous insects contained in the Museum of the Honourable East-India Company 80 pp., 4 pis. London. Hiibner, J. 1816-27. Verzeichniss bekannter Schmetterlinge 431 + 72 pp. Augsburg.Inoue, S. & Kawazoe, A. 1966. Riodinidae, Curetidae and Lycaenidae (Lepidoptera: Rhopalocera) from South Viet-Nam. Nature and Life in Southeast Asia 4: 317-394, 20 pis, 55 figs.Karsch, F. 1893. Die Insecten der Berglandschaft Adeli im Hinterlande von Togo (Westafrika). 1. Abtheilung: Apterygota, Odonata, Orthoptera Saltatoria, Lepidoptera Rhopalocera. Berliner Entomo- logische Zeitschrift 38: 1-266, 6 pis, 1 map.Kershaw, J. C. W. 1905. The life history of Gerydus chinensis Felder. Transactions of the Entomological Society of London 1905: 1-4, 1 pi. Kheil, N. M. 1884. Die Rhopalocera der Insel Nias 38 pp. , 5 pis. Berlin.Kirby, W. F. 1885. The Zoological Record 21 (Insecta): 191. - 1890. Descriptions of new species of African Lycaenidae, chiefly from the collections of DrStaudinger and Mr Henley Grose Smith. Annals and Magazine of Natural History (6) 6: 261-274. Lewis, H. L. 1974 Butterflies of the World xvi + 104 pp., 208 pis. London. Moore, F. 1857. Catalogue of the lepidopterous insects in the Museum of the Honourable East IndiaCompany 5 + 278 + 4 + 11 pp., 12 + 6 pis. London. 1866. On the lepidopterous insects of Bengal. Proceedings of the Zoological Society of London 1865:755-823, 3 pis. - 1883. Descriptions of new Asiatic diurnal Lepidoptera. Proceedings of the Zoological Society ofLondon 1883: 521-535, 2 pis. - 1884. Descriptions of some new Asiatic diurnal Lepidoptera; chiefly from specimens in the IndianMuseum, Calcutta. Journal of the Asiatic Society of Bengal (2) 53: 16-52. 1886. List of the Lepidoptera of Mergui and its Archipelago collected for the Trustees of the Indian Museum, Calcutta, by Dr John Anderson F.R.S., Superintendent of the Museum. Journal of theLinnean Society of London (Zoology) 21: 29-60, 2 pis. Moulton, J. C. 1911. A list of the butterflies of Borneo with descriptions of new species (3). Journal of theStraits Branch of the Royal Asiatic Society 60: 73-177, 1 pi. Murayama, S. & Okamura, H. 1973. Butterflies of Luzon Island in the Philippines, with descriptions ofnew species and subspecies. Tyo to Ga 24: 10-25, 48 figs. Okubo, K. 1983. Butterflies of Tioman Island, West Malaysia, with the description of new species. Tyo toGa 33: 168-184, 47 figs. Opinion 566 1959. Suppression under the Plenary Powers of the generic name Tingra Boisduval, 1847, anddesignation under the same Powers for the genera Liptena Westwood, [1851], and Pentila Westwood,[1851], of type species in harmony with accustomed usage (Class Insecta, Order Lepidoptera). Opinionsand Declarations Rendered by the International Commission on Zoological Nomenclature, London 20:377-390. Peters, W. 1952. A provisional check-list of the butterflies of the Ethiopian region 201 pp. Feltham, Middx. Piepers, M. C. & Snellen, P. C. T. 1918. The Rhopalocera of Java Erycinidae, Lycaenidae. xiv + 114 pp. , 9pis. The Hague. Pinratana, A. 1981. Butterflies in Thailand 4: vii + 215 pp., 36 pis, 47 figs. Bangkok. Reuter, E. 1897. Uber die Palpen der Rhopaloceren. Acta Societatis Scientiarum Fennicae 22: xi + 588 pp. ,6 pis. Ribbe, C. 1889. Beitrage zur Lepidopteren-Fauna von Gross-Ceram. Deutsche Entomologische Zeit-schrift, Iris 2: 187-265. - 1926. Neue Lycaenenformen hauptsachlich von Celebes. Entomologische Mitteilungen 15: 78-91.Riley, N. D. 1944. Spolia Mentawiensia: Rhopalocera, Lycaenidae and Riodinidae. Transactions of the Royal Entomological Society of London 94: 247-271, 2 pis. Riley, N. D. & Godfrey, E. J. 1921. Some undescribed Rhopalocera from Siam. Journal of the NaturalHistory Society of Siam 4: 167-190, 4 pis. Rober, J. 1886. Neue Tagschmetterlinge der Indo-Australischen Fauna. Deutsche Entomologische Zeit-schrift, Iris 1: 45-72, 4 pis. - 1892. In Staudinger, O. & Schatz, E., Exotische Schmetterlinge 2. Die Familien und Gattungen derTagfalter (6), pp. 225-284, pis 43-50. Furth. Roepke, W. 1918. Zur Myrmekophilie von Gerydus boisduvali Moore. Tijdschrift voor Entomologie 61: 1-16, 3 pis, 2 figs.Rothschild, W. 1915. On the Lepidoptera in the Tring Museum sent by Mr A. S. Meek from the Admiralty Islands, Dampier and Vulcan Islands, Pt. 2. Novitates Zoologicae 22: 387-402. THE MILETINI 89 Scudder, S. H. 1875. Historical sketch of the generic names proposed for butterflies; a contribution to systematic nomenclature. Proceedings of the American Academy of Arts and Sciences 10: 91-293.Semper, G. 1889. Die Schmetterlinge der Philippinischen Inseln 1. Tagfalter (1886-1892). 380 + 14 pp., 49 + 2 pis. Wiesbaden.Staudinger, O. 1888. In Staudinger, O. & Schatz, E., Exotische Schmetterlinge 1. Exotische Tagfalter, 1. 333pp., 100 pis., 1 map. Furth. 1889. Lepidopteren der Insel Palawan. Deutsche Entomologische Zeitschrift, Iris 2: 3-180, 2 pis.Stempffer, H. 1967. The genera of the African Lycaenidae (Lepidoptera: Rhopalocera). Bulletin of the British Museum (Natural History) (Entomology) Supplement 10: 322 pp.Swinhoe, C. 1910. Lepidoptera Indica 7: x + 286 pp., pis 551-639. London.Tite, G. E. 1969. Lycaenidae (Lepidoptera) of the Noona Dan Expedition to the Philippines, Bismarcks and Solomons. Entomologiske Meddelelser37: 47-69.Toxopeus, L. J. 1928. Tijdschrift voor Entomologie 71 (Verslag); xxii. 1929. Beschreibung einiger Schmetterlinge (Riodinidae und Lycaenidae) von Pulu Weh bei Sumatra. Tijdschrift voor Entomologie 72: 204-214. 1930. Die soort als functie vanplaats en tijd . . . xii + 198 pp., 4 pis, 17 figs. Amsterdam. 1932. Over eenige nieuwe of weinig bekende vlindersoorten. Verslagen van de Vergaderingen der Afdeeling Nederlandsch-Oost-Indie van de Nederlandsche Entomologische Vereeniging 1: Ixvii-lxxvii.1940. On a new Gerydus from West-Java. Entomologische Mededeelingen van Nederlandsch- Indie 6: 15-16.Tytler, H. C. 1915. Notes on some new and interesting butterflies from Manipur and the Naga Hills, Pt. III. Journal of the Bombay Natural History Society 24: 119-155, 2 pis.Van Eecke, R. 1914. Studien iiber Indo-Australische Lepidopteren, Fauna Simalurensis. Notes from the Ley den Museum 36: 193-258, 1 pi.Westwood, J. O. 1851, 1852. In Doubleday, E. & Westwood, J. O., The Genera of diurnal Lepidoptera 2: 251-534, pis 31-80 + suppl. pi. London. 90 J. N. ELIOT Fig. 55 Allotinus (Allotinus) agnolia sp. n., $ paratype; Sumatra. Fig. 56 Allotinus (Allotinus) nicholsi battakanus Fruhstorfer, cf lectotype; Sumatra. Fig. 57 Allotinus (Allotinus) nicholsi battakanus Fruhstorfer, $ paralectotype; Sumatra. Fig. 58 Allotinus (Fabitaras) borneensis Moulton, cf ; Borneo. Fig. 59 Allotinus (Fabitaras) punctatus $-f. caesemius Fruhstorfer; Mindanao. Fig. 60 Allotinus (Fabitaras) nigritus Semper, $ ; Mindanao. Fig. 61 Allotinus (Fabitaras) strigatus strigatus Moulton, d" ; Borneo. THE MILETINI 91 61 92 J. N. ELIOT Fig. 62 Allotinus (Fabitaras) bidiensis sp. n. , C? holotype; Borneo. Fig. 63 Allotinus (Fabitaras) brooksi sp. n. , cf holotype; Borneo. Fig. 64 Allotinus (Paragerydus) melos (H. H. Druce), cf topotype; Cagayan I. Fig. 65 Allotinus (Paragerydus) samarensis samarensis sp. n. , $ ; Mindanao. Fig. 66 Allotinus (Paragerydus) samarensis russelli subsp. n., $ holotype; Sulawesi. Fig. 67 Allotinus (Paragerydus) macassarensis macassarensis (Holland), cf ; Sulawesi. Fig. 68 Allotinus (Paragerydus) macassarensis macassarensis (Holland), $ ; Sulawesi. THE MILETINI 93 64 |S5fsw?B<fM r;-'- ; V'! l/ l^"^.i. 94 J. N. ELIOT Fig. 69 Allotinus (Paragerydus) luzonensis Eliot, cf holotype; Luzon. Fig. 70 Allotinus (Paragerydus) luzonensis Eliot, 9 allotype; Luzon. Fig. 71 Allotinus (Paragerydus) albatus albatus C. & R. Felder, $ holotype; Sulawesi. Fig. 72 Allotinus (Paragerydus) albatus mendax subsp. n., cf paratype; Luzon. Fig. 73 Allotinus (Paragerydus) albatus mendax subsp. n., 9 allotype; Luzon. Fig. 74 Allotinus (Paragerydus) apries apries Fruhstorfer, cf ; Borneo. Fig. 75 Allotinus (Paragerydus) nivalis felderi Semper, cf ; Mindanao. Fig. 76 Allotinus (Paragerydus) substrigosus ballantinei subsp. n. , cf holotype; Palawan. THE MILETINI 95 70 74 75 96 J. N. ELIOT Fig. 77 Allotinus (Paragerydus) substrigosus yusukei subsp. n. , cf holotype; Mindanao. Fig. 78 Logania nehalemla Fruhstorfer, cf holotype; New Guinea. Fig. 79 Logania paluana sp. n., $ holotype; Sulawesi. Fig. 80 Logania obscura Rober subsp. , cf ; Banggai I. Fig. 81 Logania distanti distanti Semper, cf ; Luzon. Fig. 82 Lontalius eltus eltus sp. n. , $ holotype; Pulo Laut. Fig. 83 Miletus mallus mallus (Fruhstorfer), cf ; Vietnam: south. Fig. 84 Miletus gaesa gaesa (de Niceville) , cf var. ; West Malaysia. Fig. 85 Miletus nymphis eneus Eliot, cf ; Sumatra: south-west. THE MILETINI 97 78 82 85 98 J. N. ELIOT Fig. 86 Miletus cellarius (Fruhstorfer), cf ; Borneo: Kina Balu. Fig. 87 Miletus cellarius (Fruhstorfer), $ ; Borneo: Kina Balu. Fig. 88 Miletus symethus near subsp. phantus subsp. n. , 9 5 Luzon. Fig. 89 Miletus symethus hierophantes (Fruhstorfer), cf ; Sulu Is. Fig. 90 Miletus symethus near subsp. hierophantes (Fruhstorfer), cf ; Mindanao. Fig. 91 Miletus symethus near subsp. hierophantes (Fruhstorfer), $ ; Mindanao. Fig. 92 Miletus atimonicus Murayama & Okamura, cf ; Negros. Fig. 93 Miletus atimonicus Murayama & Okamura, 9 ; Luzon. THE MILETINI 99 100 J. N. ELIOT 94 Figs. 94-98 94, Miletus heradeion near subsp. heradeion (Doherty), cf ; Sarawak; 95 Miletus heradeionnear subsp. heradeion (Doherty), $ ; Borneo: Pulo Laut; 96, Miletus heradeion arion Eliot, cf ; Borneo:Kina Balu; 97, Miletus melanion melanion cf-f. melanion C. & R. Felder, cf; Leyte; 98, Miletusmelanion melanion cf-f. albiguttatus f. n., cf holotype; Luzon. THE MILETINI 101 99 102 100 103 104 Figs. 99-104 99, Miletus bazilanus (Fruhstorfer) , O"; Mindanao; 100, Miletus bazilanus (Fruhstorfer), $paralectotype; Bazilan; 101, Miletus bazilanus (Fruhstorfer), $ var.; Mindanao; 102, Miletus takanamiisp. n. , cf holotype; Mindanao; 103, Miletus takanamii sp. n. , $ allotype; Mindanao; 104, Miletus celinusEliot, cf ; Sulawesi. 102 J. N. ELIOT 106 Fig. 105 Allotinus (Paragerydus) samarensis samarensis sp. n. , cf holotype; Samar.Fig. 106 Allotinus (Paragerydus) apries ristus subsp. n., d" holotype; Palawan. THE MILETINI 103 107 Fig. 107 Logania waltraudae sp. n., C? holotype; Samar. Fig. 108 Lontalius eltus treadawayi subsp. n., $ holotype; Samar. 104 J. N. ELIOT Index Principal references are in bold; synonyms, infrasubspecific and unavailable names are in italics absens 30acampsis 5acragas 5acrisius 5adeus5 agnolia 3, 8, 15albatus3,32,41albifasciatus3, 8, 13albiguttatus 5 , 80albotignula 5alkamah 3, 14Allotinus3, 7amphiarus 5anaxandridas 3, 23ancius3, 11ancon5, 77anconides 5angulosus 5, 85aphacus3, 10, 11aphocha3,48aphthonius 4, 56, 57aphytis 5 apries3,31,42, 43apsines 4, 70apus3, 10, 12Archaeogerydus 4, 75archilochusS, 77arionS, 80aronicus 5arrius3, 21artaxatus 4artinus 3, 11assamensis 4 atimonicus 5, 77, 79atomaria 5audax3, 10, 12avitus 5 bajanus 4, 49ballantinei 4, 53, 55bangkanus 5battakanus3, 16, 17batuensis, Allotinus 3, 35, 36batuensis, Miletus 5batunensis 5bazilanus5,77,82bicoloria5, 85bidiensis3,20, 26biggsiiS, 77boisduvali Moore 5, 77boisduvalii Butler 5borneensis3,20, 22brooksi3,20, 26buruensis 5 celinus 5, 76, 77ceramensis 5chinensis 4, 76cineraria 4, 65continentalis 3, 46, 47corbeti3,31,44corus 4courvoisieri 5croton 4, 76 damis 4, 64, 65 damodar3, 40 davidis 4, 32, 55 denalus3,25 denticulata 4 depictus 3, 18 diehli 4, 64, 66 dilutus3,47 diopeithes 5 diotrophes 5 distant! Semper 4, 59, 69, 70 distanti Staudinger 4, 67 divisa 5 donussa 4, 68 dositheus3,29,42,44 dossemus 5 dotion 3, 10, 12 drucei, Logania 4, 70, 71 drucei, Miletus 5, 77 drumila4, 31, 56 edonus 5elioti3,22 eltus 4, 74enatheus 4, 49eneus 4, 76enganicus 4, 49epldurus 5eretria 3, 24eryximachus 3, 10, 11 ,23carrinas 4catoleucos 5caudatus 3, 21 eupalion 3, 43euphranor5,81,82eurytanus 3, 47eustatius 4evora 4, 63extraneus 5 Fabitaras3,7, 20fabius3,20,21fallax3,8, 10faustina4, 65,68felderi 4, 52fictus 4florensis 5fruhstorferi3, 30 gaesa 4, 76gaetulusS, 77gallus5,77gardineri 5georgius 4, 46, 49 Gerydinae 1 Gerydus 4, 75 gethusus4, 78 gigantes5 ; 77,79 g/gos Bethune-Baker 5, 85 gigas H. H. Druce 5 glypha 4, 67 gopara 4, 77 grisea 4, 57 hampsoni 4, 59, 72heracleion5,76, 77,80heraeon 5hilaeira4, 64, 65hierophantes 5, 78hieropous 5horsfieldi3,31,32,33hyllus 5 improbus 4infumata3,34innocens 5ins ignis 4, 57intricata3, 35irroratus 4 jacchus 5javanica 4, 64, 67 kalawarus 3, 18kallikrates 3, 14karennius 4kelantanus 4 Iahomius4, 65, 66learchus 4/e/fws 4, 49Ienaia4, 53, 54leogoron 3, 32, 34, 35Ieos5,77leucocyon 5Iindus3,34,35Logania 4, 7, 57lombokianus 5longeana 4Lontalius 4, 7, 74luca 4, 71Iuzonensis3, 32,40 macassarensis 3, 31, 39magaris 4, 53maitus3,29major 3, 8, 17Malais 4, 57malayanus3, 25malayica 4, 59, 60mallus 4, 76, 78mangolicus 5manychus 3, 14marmorata 4, 59, 64, 65martinus 4, 68 masana 4, 72massalia 4, 70maximus, Allotinus 3, 8, 19maximus, Miletus 5meeki 4, 72Megalopalpus 5, 7, 84megaris 5 melanion5,77,80, 81melos3, 32, 36menadensis 3, 39, 40mendava 3, 28mendax 3, 41meronus 5metaleucus 5, 84, 85metrovius 5michaelis 3, 12Miletographa 3, 30Miletus 4, 7, 75milvius 4minis 3, 14, 15molionides 3 , 48moorei 3, 46, 48multistrigatus 4, 57munichya 4, 64, 66myriandus 4, 48 , 65 narsares 3, 29natunensis 5nehalemia 4, 59, 60ness us 3, 34niasicus 5nicer atus 4, 49nicholsi3,8, 16, 17nigritus3,20, 24nineyanus 5nivalis 4, 32,51,52normani3, 35,36nuctus 5nymphis 4, 76 obscura Distant & Pryer 4, 65obscuraR6ber4,59,68oichalia 5oxylus 5 paetus 4, 32, 50 INDEX paianlus 5pallaxopas 5pallida5,8Spaluana 4, 59, 63palawana 4, 64, 67pamisus 3, 21pandu 5panormis 3, 27Paragerydus 3, 7, 30parapus4, 31, 51pardus 4pentheus 5perlucidus 5permagnus 3, 33petronius 5phantusS, 78philippus 5philopator 5phradimon 5plessis3, 35, 36porriginosus3, 28portunus3, 21, 29porus 4posidion 3, 48punctatus3,20, 23pyxus3,29,30 rebilus 3, 48regina 4, 59, 62, 63rekkia 3, 46, 48reverdini3, 36ristus3,42,44russelli3, 38 sabazus3, 10samarensis 3, 31, 37, 38samosata4, 65,67sarrastes3, 21, 28sarus 5 satelliticus3, 33, 34sebethus 5shanius 4siamensisS, 79sibyllinus4, 53, 54silarus 3, 14simalurensis 5similis 5, 85 105 simplex 5, 85siporanusS, 33,34solitarius5,78sora 4, 65sriwa 4, 62, 63staudingeri 4, 70, 72stenosa 4, 65strigatus3,20, 24, 25stygianus 5subfasdata 4, 73substrigosus 4, 32, 53subura 4, 60subviolaceus 3, 8, 14suka 4, 49Symetha 4, 75symethusS, 77, 78 takanamii 5, 77, 83 talu3,36 taras3,21,27 tavoyana 4 tellus 5 teos5 treadawayi 4, 74 turdeta 4, 70 tymphrestus 3, 10, 12 unicolor 3, 32, 45, 46, 47 vadosus3,35valeus5,77vaneeckei 5vespesianus 5vincula 5virtus 5vitelianus 5 , 82 waltraudae 4, 59, 61waterstradti 3, 30watsoniana 4, 59, 73 xeragis 5yusukei4, 53, 55 zaradrus3, 10zinckenii 4, 77zitema 4, 46, 50zymnaS, 85 Occasional Papers on Systematic Entomology New Series The economic importance of insects, and the enormous number of species, have resulted in avast literature written in many languages; that which is of direct economic importance andrecently published can, to an increasing extent, be searched by using computerized data bases,but a great amount of more general information is unlikely to be available so readily in thenear future. The objective of this new occasional series is to make available in hard copy someof the basic data that is essential to the preparation of comprehensive accounts of the worldinsect fauna. The papers have been fully researched bibliographically and consist of checklistsof nominal taxa, and faunal lists with information on host plants and localities, based mainlyon the collections and libraries of the British Museum (Natural History). No. 1. A checklist of Neotropical arctiine and pericopine tiger moths. A. Watson & D. T. Goodger 72pp. inc. 4 colour plates 27 February 1986 No. 2. An annotated checklist of the Carabidae (including Cicindelinae, Rhysodinae and Paussinae) recorded from Borneo. N. E. Stork 26pp. , 1 map 29 May 1986 Catalogues of BM (NH) Bulletins and Books free on request to Publication Sales. Titles to be published in Volume 53 A review of the Miletini (Lepidoptera: Lycaenidae) By J.N.Eliot Australian ichneumonids of the tribes Labenini and Poecilocryptini By I. D. Gauld & G. A. Holloway The tribe Pseudophloeini (Hemiptera: Coreidae) in the Old World tropics with a discussion onthe distribution of the Pseudophloeinae By W. R. Dolling The songs of the western European grasshoppers of the genus Omocestusin relation to theirtaxonomy (Orthoptera: Acrididae) ByD. R. Ragge Photoset by Rowland Phototypesetting Ltd, Bury St Edmunds, SuffolkPrinted in Great Britain by Henry Ling Ltd, Dorchester PRrGENE Bulletin of the British Museum (Natural History) Australian ichneumonids of the tribesLabenini and Poecilocryptini I. D. Gauld & G. A. Hollo way Entomology series Vol53 No 2 30 October 1986 The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in fourscientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology,and an Historical series. Papers in the Bulletin are primarily the results of research carried out on the unique andever-growing collections of the Museum, both by the scientific staff of the Museum and byspecialists from elsewhere who make use of the Museum's resources. Many of the papers areworks of reference that will remain indispensable for years to come. Parts are published at irregular intervals as they become ready, each is complete in itself,available separately, and individually priced. Volumes contain about 300 pages and severalvolumes may appear within a calendar year. Subscriptions may be placed for one or more ofthe series on either an Annual or Per Volume basis. Prices vary according to the contents ofthe individual parts. Orders and enquiries should be sent to: Publications Sales, British Museum (Natural History),Cromwell Road, London SW75BD,England. World List abbreviation: Bull. Br. Mus. not. Hist. (Ent.) Trustees of the British Museum (Natural History), 1986 The Entomology series is produced under the general editorship of the Keeper of Entomology: Laurence A. Mound Assistant Editor: W. Gerald Tremewan ISBN 565 06020 1ISSN 0524-6431 British Museum (Natural History)Cromwell RoadLondon SW7 5BD Entomology seriesVol 53 No 2 pp 107-150 Issued 30 October 1986 D r? *" Australian ichneumonids of the tribes Labenini and Poecilocryptini I. D. Gauld Department of Entomology, British Museum (Natural History), Cromwell Road, LondonSW7 5BD G. A. Holloway Division of Invertebrate Zoology, Australian Museum, 6-8 College Street, Sydney, NewSouth Wales, Australia Contents Synopsis 107 Introduction : 107 Material examined 108 Checklist of Australian Labenini and Poecilocryptini 108 Subfamily Labeninae 109 Key to tribes of Labeninae 110 Tribe Labenini 110 Key to genera of Labenini occurring in Australia Ill Labena Cresson Ill Key to species of Labena occurring in Australia 112 Certonotus Kriechbaumer 117 Key to species of Certonotus occurring in Australia 119 Tribe Poecilocryptini 137 Key to genera of Poecilocryptini 138 Alaothrys Gauld 138 Poecilocryptus Cameron 139 Key to species of Poecilocryptus 139 Urancyla Gauld 141 Acknowledgements 142 References 142 Index to hosts 149 Index to Ichneumonidae 149 Synopsis The Australian species of the labenine tribes Labenini and Poecilocryptini are revised and keys provided tothe five genera and 36 species occurring on the continent. Twenty new species are described, but one is notformally named as its status requires further investigation. The remaining 16 species are redescribed andtheir diagnostic features emphasized. Asperellus Townes is newly placed as a junior synonym of CertonotusKriechbaumer, Certonotus tasmaniensis Turner is treated as a junior synonym of C. nitidulus Morley andPoecilocryptus straminea Morley placed as a junior synonym of P. nigromaculatus Cameron. Lectotypesare designated for five species. Details of known hosts are given together with notes about the geographicaldistribution of each species. A brief introductory section discusses the systematic position of the subfamilyand outlines possible relationships between species. A checklist of Australian species, and indexes to hostsand parasitoids complete the work. Introduction Amongst the most important of the natural enemies of insect pests are the parasitic Hymenop-tera, a very large group of animals whose larvae develop at the expense of other insects (Askew,1971). Under normal circumstances, the populations of many injurious insects are severely Bull. Br. Mus. nat. Hist. (Ent.) 53 (2): 107-149 Issued 30 October 1986 108 I. D. GAULD & G. A. HOLLOWAY limited by the attacks of Parasitica, and in several countries, including Australia, the ravages ofaccidentally imported pests have been curtailed by the introduction of one or more hymenopter-ans (Muldrew, 1967; Taylor, 1978). Other introduced pests have been severely attacked bynative Australian parasitoids (Tryon, 1900). During the past 30 years there has been an upsurgein interest in using Hymenoptera and other organisms for purposes of pest control (Wilson,1960; Huffaker & Messenger, 1976) as an alternative to costly, ineffective and environmentallydestructive chemical methods (Bosch, 1978). However, for biological control programmes to besuccessful an intimate knowledge is necessary of the life history and interactions of the pest andits parasites. A sound taxonomic basis is vital for the development of such knowledge (Hardy,1982), for such work permits the accurate identification of an organism and hence providesconstancy and universality in the usage of names, a prerequisite for the national and internation-al communication of information. The present work is a taxonomic study of two very distinct tribes of one of the apparently mostprimitive extant ichneumonid subfamilies, the Labeninae. These two tribes, the Labenini andPoecilocryptini, are virtually confined to the southern hemisphere with most species occurring inAustralia and South America. The generic phylogeny of the Labeninae and their geographicaldistribution has recently been studied (Gauld, 1983) but no keys are currently available tofacilitate identification of the species. This paper is an attempt to provide a means of determiningthe Australian species of this interesting group. The terminology in this work follows that of Gauld (1984). Material examined The study is based on examination of almost all specimens available in collections of Australianichneumonids. Special attention was paid to collections in agricultural institutions that contain alarge number of reared specimens. Examination of these collections has been supplemented byextensive collecting, particularly in Tasmania and the south-east. Although the resulting sampleis thought to be fairly representative of the fauna of the more humid eastern part of thecontinent, relatively little material has been examined from the west. The following abbreviations have been used for museums containing Australian material.AM Australian Museum, Sydney, New South Wales, Australia ANIC Australian National Insect Collection, Canberra, Australian Capital Territory, Australia BMNH British Museum (Natural History), London, U.K. BPBM Bernice P. Bishop Museum, Honolulu, Hawaii, U.S.A. MNHN Museum National d'Histoire Naturelle, Paris, France MNHU Museum fur Naturkunde der Humboldt-Universitat, Berlin, D.D.R. NMV National Museum of Victoria, Melbourne, Victoria, Australia NSWDA New South Wales Department of Agriculture, Rydalmere, New South Wales, Australia QM Queensland Museum, Brisbane, Queensland, Australia TC Townes Collection , Ann Arbor , Michigan , U . S . A . WAD A Western Australian Department of Agriculture, Perth, Western Australia, Australia Checklist of Australian Labenini and Poecilocryptini LABENINI CERTONOTUS KriechbaumerAsperellus Townes syn. n.andrt-wisp. n.annulatus Morleyapicalis Morleyiivi/u.vsp. n.ce/eussp. n.cestussp. n.farrugiaisp. n.geniculatus Morleyhinnuleus Krieger comb. rev. humeralifer Krieger ixion sp. n. leeuwinensis Turner comb. rev. mogimbensis Cheesman comb. rev. monticola Morley nitidulus Morley tosmaniensis Turner syn. n.palunia sp. n.pineussp. n.rufescens Morleysisyphus sp. n.talus sp. n. AUSTRALIAN LABENINI & POECILOCRYPTINI 109 toolangisp. n. POECILOCRYPTINI zebrus sp. n. ALAOTHYRIS Gauld species A elongissimus Gauld LABENA Cresson POECILOCR YPTUS Cameron annulata (Brulle) coloratussp. n. chadwickii (Parrott) galliphagus sp. n. grand is sp. n. nigripectus Turner & Waterston jacunda sp. n. nigromaculatus Cameron keira sp. n. straminea Morley syn. n. malecasta sp. n. URANCYLA Gauld pudenda sp. n. fulva Gauld Nomen dubium Certonotus varius Kriechbaumer From Kriechbaumer's original description (1889: 308) it is apparent that this species is one of the largerAustralian Certonotus. As we have seen no species that agrees with the description, nor can the type belocated, we have no option but to treat this as an unrecognized taxon. Subfamily LABENINAE The Labeninae is a moderately large subfamily. Amongst the ichneumonid subfamilies it isunique in being a Gondwanic group. Most genera and species are confined to Australasia and/orSouth America; only a dozen or so taxa occur outside this area (Table 1). All other ichneumonidsubfamilies are well represented, if not actually most diverse, in the Holarctic region. Phy-logenetic analysis suggests that the Labeninae radiated in West Gondwanaland, prior to theseparation of Australia, Antarctica and South America, some 55 mya (Gauld, 1983). The groupis only represented by some highly derived species in India and South Africa, suggesting that theGondwanic radiation occurred after the separation of these continents (c. 100 mya). Manylabenines are associated with the southern temperate rain forests, the areas floristicallydominated by Nothofagus, Araucaria and Podocarpus. One labenine genus is known to ovipositinto the seeds of an Araucaria in Australia. Structurally labenines exhibit many of the plesiomorphic features of ichneumonids, and thelarvae are particularly primitive (Short, 1978). Biologically they also show many of the so-calledprimitive features of ichneumonids - they are ectoparasitoids and many are known to parasitizewood-boring insects. It is not clear what the sister-group to the labenines is. The more primitive taxa (Labenini)closely resemble some Pimplinae, but all the similarities can, we believe, be interpreted as eithersymplesiomorphies, or parallel adaptations to the wood-boring habit. We suggest that theshared specializations shown by labenines and rhyssines are parallelisms rather than synapomor-phies because of the striking differences in oviposition behaviour between the groups. Labe-nines manipulate the ovipositor with a specialized guide on the antero-medial surface of the hindcoxa; rhyssines achieve similar control with specialized claspers on the ventral surface of thegaster. It is not impossible that these taxa share a close common ancestor (a wood-borer with nospecialization to control the ovipositor), but there is no conclusive phylogenetic evidence tosupport such speculation. We suggest that the labenines are a primitive offshoot of the earliestichneumonid radiation, and that at an early stage in their evolutionary history they becameisolated on a southern continent prior to undergoing a modest radiation, that has in somerespects paralleled that of other ectoparasitic ichneumonids (see Gauld, 1983; 1984). The Labeninae comprises four morphologically distinctive tribes, Labenini, Groteini, Poeci-locryptini and Brachycyrtini, all of which are well represented in Australia (Table 1). Both theGroteini and Brachycyrtini include some genera for which relatively insufficient material is athand to justify undertaking a taxonomic revision at present. Quite good series of some species ofLabenini and Poecilocryptini are available for systematic study, and as there are far more speciesthan the few described the opportunity has been taken to monograph these taxa. 110 I. D. GAULD & G. A. HOLLO WAY Table 1 The distribution of the Labeninae. Key to tribes of Labeninae 1 Mandible long and slender, labrum very large, almost as long as clypeus (Fig. 1); female with ovipositor barely projecting beyond apex of gaster , dorsoventrally depressed GROTEINI - Mandible rather short and not particularly slender, labrum from moderately small to virtually concealed (Fig. 2); female with ovipositor projecting beyond apex of gaster by at least 0-5times length of hind tibia 2 2 Fore wing with abscissa of Cu\, between Im-cu and C la at least 1-4 times as long as Cu ib (Fig. 3); hind coxa of female with a furrow on anterior surface internally; base of valvula 3 bearingdistinct sclerotized lobe (Fig. 4) LABENINI(p. 110) - Fore wing with abscissa of Cui between Im-cu and Cw la from subequal to, to conspicuously shorter than Cw lb (Figs 5, 6); hind coxa of female without a furrow on anterior surfaceinternally; base of valvula 3 without a distinct sclerotized lobe 3 3 Fore wing with 2m-cu with one long bulla, sometimes with an indistinct trace of a vein centrally (Fig. 5); occipital carina dorsally absent POECILOCRYPTINI (p. 137) - Fore wing with 2m-cu with two short bullae widely separated from each other (Fig. 6) ; occipital carina dorsally complete BRACHYCYRTINI The Groteini and Brachycyrtini are not discussed further in this paper. Tribe LABENINI Labenines are characterized by having a distinct sclerotized lobe at the base of the third valvula.Most species also have fine file-like teeth at the distal apex of the ovipositor. The tribe containsfour genera: Labena, Certonotus, Apechoneura and Xenothyris. The last two are exclusivelyNeotropical, and Apechoneura may well be a specialized species-group of Certonotus. Labenaand Certonotus are most commonly found in Australia and South America, and there is evidenceto suggest that the group is an ancient southern one that originated in the cool temperate forestsof Gondwanaland (Gauld, 1983). As far as is known, the native hosts of the group are AUSTRALIAN LABENINI & POECILOCRYPTINI 111 coleopterous larvae boring in wood (Townes, 1969; Gauld, 1984), though one species hasadapted to parasitizing introduced siricids (Hocking, 1967). Structurally, labenines are amongst the most primitive ichneumonids. Virtually all specieshave clearly developed parapsidal furrows, a primitive hymenopterous feature found widelyamongst Symphyta and some lower Aculeates. An apparently unique feature of the group is thepresence of a distinct sclerotized lobe at the base of the third valvula (Fig. 4). In otherichneumonids a slightly broadened, rounded area is present in a homologous position and thismay well represent a precursor of the valvular lobe. Quite what this structure is, at present isunclear, as recent morphological treatises make no mention of such a structure (e.g. Smith,1970; Matsuda, 1976). Short (1978) comments on the primitive structure of labenine larvae. Key to genera cf Labenini occurring in Australia 1 Mesoscutum punctate or puncto-striate; occipital carina mediodorsally complete or onlynarrowly interrupted; apex of fore tibia with a conspicuous long curved spine on outer distalmargin; female with third fore tarsal segment produced apically into a lobe that reachesbeyond apex of fourth segment (Fig. 7) LABENA Cresson (p. Ill) - Mesoscutum with transverse rugae; occipital carina dorsally entirely absent; apex of fore tibiawith a short tooth on outer distal margin; female with fore tarsus simple CEtfrCWOritf Kriechbaumer (p. 117) LABENA Cresson Labena Cresson, 1864: 399. Type-species: Cryptus grallator Say, by subsequent designation, Viereck, 1914: 80.Caryoecus Walsh, 1866: 30. Type-species: Mesochorus fuscipennis Brulle (= Cryptus grallator Say), by monotypy. Microtritus Kriechbaumer, 1889: 307. Type-species: Microtritus apicalis Kriechbaumer, by monotypy.Dyseidopus Kriechbaumer, 1890: 489. Type-species: Dyseidopus sericeus Kriechbaumer, by monotypy.Dysidopus Schulz, 1906: 103. [Unjustified emendation.]Neonotus Parrott, 1955: 230. Type-species: Neonotus chadwickii Parrott, by original designation. Moderately large to large insects, fore wing length 6-20 mm; clypeus small, flat or concave, very thin withmargin arcuate; labrum barely projecting; mandible tapered, twisted 30 with upper tooth slightly thelonger; outer mandibular surface with a median longitudinal groove bearing hairs; malar space shorterthan basal mandibular width. Occipital carina complete or narrowly interrupted centrally; eye marginslightly indented opposite antennal socket. Antenna almost cylindrical, that of female with a small flatsensillum on extreme distal apex. Mesoscutum punctate or striate transversely, with notauli vestigial;notaular crests well developed. Propodeum rather short, convexly rounded with spiracle elliptical;propodeal areae usually defined, area superomedia usually larger than area petiolaris; gaster inserted highon propodeum, above level of hind coxae. Fore tibia of female inflated (so it often collapses in driedspecimens) with a large curved spine on outer distal margin; fore tarsus with segment 3 lobed, the lobereaching nearly to centre of segment 5, segment 4 reduced; hind coxa of female flattened internally with ashort basal groove; tarsal claws large, simple. Fore wing with cu-a more or less opposite base of Rs&M;3r-m present, areolet large, rhombic or with very short anterior side; 2m-cu sinuous with two bullae. Hindwing with distal abscissa of Cu\ present, sometimes not joining to first abscissa of Cu\, if joined, then firstabscissa of Cu\ is shorter than cu-a; basal cell slender; Sc with one or two hamuli. Gaster quite long,cylindrical; tergite 1 with spiracles a little before centre; sternite 1 usually reaching to or slightly beyondlevel of spiracles; laterotergites 2-4 membranous; last visible tergite with a small, weakly defined dorsalplate differentiated. Ovipositor projecting beyond apex of gaster by 1-8-4-4 times length of hind tibia;upper valve with apex bearing blunt serrations, lower valve apically almost enclosing the upper, usuallywith series of close file-like teeth and a fine coriaceous patch just proximal to these teeth, rarely with teethquite strongly developed. REMARKS. Labena is a large genus with most species occurring in the Neotropics. Two species occur inNorth America, and in this work seven are recognized from Australia. Labena species are easily separated from other Labenini by the specialized fore tarsus and the possessionof a sensillum on the extreme apex of the antenna. It resembles the Neotropical genus Xenothyris Townesin having a relatively smooth mesoscutum, though in the latter taxon rugae are present on the anterior partof the median lobe. Unlike Xenothyris, Labena species have simple claws. 112 I. D. GAULD & G. A. HOLLO WAY The relationships of the Australian species. As the vast majority of the Neotropical species areundescribed and virtually unstudied it is impossible as yet to state whether the Australian Labenaconstitute a holophyletic group, so no formal cladistic analysis is presented here. However, the sevenAustralian species are clearly divisible into two monophyletic species-groups, each of which is charac-terized by several apomorphic features. The annulata-group contains four taxa (L. annulata, grandis,chadwickii and malecasta). All these species have Cu\ in the hind wing incomplete and have the whiteflagellar band at the distal apex of the antenna. The keira-group contains three taxa (L. keira, jacunda andpudenda). These species have virtually a smooth mesopleuron, and the male genitalia are specialized inthat the apex of the gonosquama is flattened or impressed and surrounded by a fringe of very long hairs(Figs 9, 10). The interrelationship of the taxa within these species-groups is less clear. Within the keira-group, keiraandpudenda share two apparently apomorphic features, a specialized mid tibia and a medially interruptedyellow band on the central female gastral segments. The unspecialized mid tibia found in L. jacunda mayrepresent a reversal to the plesiomorphic condition unless the apomorphic state has been derivedindependently in both the keira and annulata lineages (and probably also in various Neotropical groups). Itis therefore arguable that the possible sister-relationship between keira and pudenda is only supported bythe colour character. Contradicting this relationship are two features shared by pudenda and jacunda - thecoarse ovipositor teeth (Fig. 21) and the large impressed area on the apex of the male gonosquama. Thelatter is quite clearly an apomorphic condition but the polarity of the former is controversial as thepossession of coarse teeth is probably plesiomorphic for ichneumonids. Other Labenini all have fine,file-like ovipositor teeth and it is considered that this condition is an apomorphy of the tribe (Gauld, 1983)and consequently plesiomorphic for Labena species. Thus the possession of coarse teeth can be regarded asa further apomorphic development, albeit a reversal, supporting the group pudenda and jacunda. Within the annulata-group the three species grandis, chadwickii and malecasta seem to form a subgroupcharacterized by possession of transverse striations on the mesoscutum. This subgroup apparently has asister-relationship to L. annulata (Fig. 65). Key to species of Labena occurring in Australia 1 Hind wing with distal abscissa of Cu\ present in membrane, not confluent with Cu\ and cu-a (Fig. 11); mesopleuron quite closely punctate or punctocoriaceous, often submatt; male gonos-quama not distally strongly flattened, with a fringe of fine hairs (Fig. 8) or a single tuft of longhairs ( a/inu/afa-group) 2 - Hind wing with distal abscissa of Cu^ complete (Fig. 12); mesopleuron smooth and polished, at most with sparse punctures; male gonosquama distally flattened or impressed, with this areafringed with long hairs (Figs 9, 10) (Iteira-group) 5 2 Tergites 2-4 of gaster coarsely and very closely punctate, matt (Fig. 13); ovipositor 2-0 or less times as long as hind tibia annulata (Brulle) (p. 113) - Tergites 2-4 of gaster smooth and polished , at most with very fine punctures (Fig . 14) ; ovipositor at least 2-5 times as long as hind tibia 3 3 Posterior end of gaster with mediodorsal margin thickened and produced into a rounded prominence (Fig. 15); pronotum with a median lateral conical tubercle; metapleuron stronglyvermiculate grandis sp. n. (p. 1 14) - Posterior end of gaster with mediodorsal margin thin, slightly concave, not produced (Fig. 16); pronotum, at most, with a median lateral convex swelling, usually virtually flat; metapleuronpunctate or puncto-striate 4 4 Female with ovipositor at least 3-5 times as long as hind tibia; occipital carina mediodorsally complete; male with a tuft of long hairs on distal apex of gonosquama malecasta sp. n. (p. 116) - Female with ovipositor at most 3 times as long as hind tibia ; occipital carina (of female at least) mediodorsally, narrowly interrupted; male with gonosquama distally evenly rounded anduniformly hirsute chadwickii (Parrott) (p. 1 13) 5 Mid tibia slender, subcylindrical (Fig. 17); tergites 4-5 of gaster with broad posterior yellow marginal band; hind coxa of female with ovipositor guide barely differentiated jacunda sp. n. (p. 115) - Mid tibia distally swollen, proximally narrowed and at least slightly flattened (Fig. 18); tergites 4-5 of gaster with posterior marginal yellow band centrally interrupted or yellow bandindistinct ; hind coxa of female with ovipositor guide well developed 6 6 Apex of fore wing infumate; mid tibia with a longitudinal row of 3-6 spine-like bristles on outer surface (Fig. 18); submetapleural carina evenly and weakly broadened anteriorly (Fig. 19);ovipositor about 3-0 times as long as hind tibia keira sp. n. (p. 115) AUSTRALIAN LABENINI & POECILOCRYPTINI 113 - Apex of fore wing very uniformly hyaline; mid tibia without obvious spine-like bristles on outersurface; submetapleural carina anteriorly abruptly expanded into a rounded lobe (Fig. 20);ovipositor about 4-0 times as long as hind tibia pudenda sp. n.(p. 117) Labena annulata (Brulle)(Figs 8, 11, 13) Ephialtes annulatus Brulle, 1846: 86. Holotype 9, TASMANIA (MNHN) [examined].Labena annulata (Brulle) Townes etal. , 1961: 112. Female. Moderately large species, fore wing length 7-13 mm. Malar space 0-4 times as long as basalmandibular width. Occipital carina mediodorsally narrowly interrupted, ventrally joining hypostomalcarina and continuing to mandibular base as a distinct carina, not developed into a flange. Pronotumlaterally almost flat; mesothorax laterally and ventrally bearing dense, short, white pubescence; mesoscu-tum closely and coarsely punctate; horizontal mesopleural furrow indistinct; metapleuron closely andcoarsely punctate; submetapleural carina anteriorly expanded into a rounded lobe. Propodeum withanterior transverse carina present centrally separating area superomedia from area basalis. Fore wing with3r-m and 2r-m converging, widely separated at junction with Rs. Hind wing with distal abscissa of Cuipresent in membrane, not confluent with Cu\ and cu-a. Mid tibia with a longitudinal row of spine-likebristles on outer surface, proximally slender, distally swollen with a trace of groove in inner surface. Gasterwith sternite 10-7 times as long as the hind coxa, reaching to level of spiracle; tergites 2-4 coarsely and veryclosely punctate, matt; apex of terminal plate of tergite 8 simple. Ovipositor projecting beyond apex ofgaster by 1-8-2-0 times length of hind tibia, the apex bearing very fine teeth. Coloration. As for L. pudenda although some specimens with fore and mid legs entirely yellow. Male. Similar to female but with fore wing length 7-8 mm; malar space 0-4 times basal mandibular width;gaster with sternite 1 0-7-0-8 times as long as hind coxa. Gonosquama with an apical tuft of fine, long hairs. REMARKS. L. annulata is one of the most distinctive species of Australian Labena and can easily berecognized by its very close, coarse punctured microsculpture, particularly on the tergites of the gaster.The ovipositor is shorter than that of any other species. This species is widespread in Australia though itseems to be most common in the south and west. HOST RECORDS. Cerambycidae: Ur acanthus strigosus Pascoe (NMV). MATERIAL EXAMINED Holotype $, Tasmania: no further data (MNHN). Queensland: 1 $ , 14 km W. by N. Hope Vale Mission (15-16S 144-59E), x.1980 (Cardale) (ANIC). NewSouth Wales: 1 $ , 2 d", Kinchega NP, ix.1980 (Rodd) (AM). Victoria: 1 $ , Baxter, parasite on Uracanthusstrigosus (Dixon) (NMV); 1 $, Healesville, xi.1983 (NMV). Northern Territory: 2 , Areyonga, 600 m,viii (TC). Tasmania: 2 $, Hobart, ix.1921 (Cole) (NMV); 1 c?, Ridgeway ix.1943 (Cole) (NMV); 1 $,1867(Sichel) (TC) (compared with type). Western Australia: 1 $, 13 km WNW. Northampton (28-18S114-31E), x.1981 (Naumann & Cardale) (ANIC); 1 $ , no further data (BMNH). Labena chadwickii (Parrott)(Fig. 14) Neonotus chadwickii Parrott, 1955: 230. Holotype $, NEW SOUTH WALES (NSWDA) [examined].Labena chadwickii (Parrott) Townes & Townes, 1960: 531. Female. Moderate to large species, fore wing length 6-17 mm. Malar space 0-5 times as long as basalmandibular width. Occipital carina mediodorsally narrowly interrupted at least in female, ventrally joininghypostomal carina and continuing to base of mandibular base as distinct carina, not developed into aflange. Pronotum laterally with a median swelling; mesothorax laterally and ventrally bearing sparse,short, white pubescence; mesoscutum punctate, at least transversely striate in centre, in larger specimensuniformly striate; horizontal mesopleural furrow distinct; metapleuron punctate; metapleural carinaexpanded anteriorly into a small rounded lobe. Propodeum with anterior transverse carina presentcentrally, separating area superomedia from area basalis. Fore wing with 3r-m and 2r-m converging,almost joining at junction with Rs. Hind wing with distal abscissa of Ci present in membrane, notconfluent with Cu\ and cu-a. Mid tibia with a longitudinal row of spine-like bristles on outer surface,proximally slender, distally swollen and with indication of groove on inner surface. Gaster with sternite 10-8 times as long as hind coxa, reaching from opposite to well behind level of spiracle; tergites 2-4 smooth 114 I. D. GAULD & G. A. HOLLOW AY and polished, at most with very fine punctures; apex of terminal plate of tergite 8 simple. Ovipositorprojecting beyond apex of gaster by 2-2-3-0 times length of hind tibia, the apex bearing very fine teeth. Coloration. Black. Palps, labrum, clypeus, face (except median longitudinal area), genae, orbits,anterolateral margin of mesonotum, tegula, posterior 0-3 of subalar process, posterior 0-5 of scutellum,postscutellum, posterior 0-3 of propodeum, posterior margins of tergites 1-7, distal end of each coxa, foreand mid legs with trochanters, distal 0-2 of femur, tibia and tarsus, hind legs with distal 0-1 of femur andtibial spurs yellow. Median area of frons, mesonotum, mesopleuron, metapleuron, upper surface of hindcoxa partly deep red. Antenna with disruptive mark subapical, last 2-3 flagellar segments black. Male. Similar to female but with fore wing length 6-9 mm; malar space 0-6 times basal mandibular width;gaster with sternite 1 0-8 times as long as hind coxa. Gonosquama distally evenly rounded and uniformlyhirsute. Coloration as for female with less deep red areas. REMARKS. L. chadwickii is the commonest and most widespread eastern species in Australia. Of the specieswith an apical white flagellar band, complete yellow posterior margined gastral tergites and incomplete Cu\in the hind wing only chadwickii has a simple terminal plate, smooth gastral tergites and incompleteoccipital carina. The females are rather similar to L. malecasta but differ in having a shorter ovipositor andthe genal orbits more broadly yellow and confluent with the yellow on the face. L. malecasta has onlynarrow yellow marks on the genae and these are not confluent with the facial yellow. HOST RECORDS. Buprestidae: Ethon affine Laporte & Gory (Parrott, 1955). Several individuals have beenreared in Tasmania from logs of Pinus radiata infested by Sirex. MATERIAL EXAMINED Holotype $, New South Wales: Middlecove Point, Willoughby, Sydney (Chadwick) (NSWDA). Queensland: 1 $,Eungella,xi(TC); 1 $,6mN. of Kuranda, ll.i.1961 (McAlpine&Holloway) (AM); 1$, Mt Cootha, iv (TC); 1 $, Mt Glorious, vi.1977 (Miller) (BMNH); 1 $, Mt Glorious, xii.1976 (Boucek)(BMNH); 2 $ , Mt Glorious, i (TC); 2 $ , Mt Nebo, ii (TC); 1 $ , Mt Tambourine, iii.1950 (Burns) (NMV);4 $, Mt Tambourine, xii (TC); 1 $, O'Reilly's Guest House, iii.1980 (BMNH); 1 cf, Palm Is., nearTownsville, x (TC); 1 cf, Stanthorpe, 1982 (Sedlacek) (TC). New South Wales: 1 $, Eucambene Dam,i.1961 (Leipa) (ANIC); 3 cf (paratypes), same data as holotype (AM, ANIC, NSWDA); 1 $, Monga,ix.1957 (ANIC); 6 cf, Mt Toman, ix.1982 (Rodd) (AM); 1 $, Mt Tomah, xi.1978 (Rodd) (AM); 1 $,Wentworth Falls, Blue Mtns, xi.1982 (McAlpine) (AM). Australian Capital Territory: 1 <j>, Lees Ck,xi.1977 (Daniels) (AM). Victoria: 1 $ , 6 km S. Aberfeldy, xi.1976 (Colder) (NMV); 1 $ , Blackwood Ra.,x.1953 (Neboiss) (NMV); 2 $, Ferntree Gully, x.1951 (Oke) (NMV); 1 $>, Gippsland (AM); 1 cf,Kinglake W., x.1954 (Neboiss) (NMV); 1 $, Lai Lai, i.1954 (Neboiss) (NMV); 1 $, 1 cf, 12 km SE.Merrijig, Howqua R., xi.1971 (Neboiss) (NMV); 1 $, Mitta Mitta R., x.1973 (NMV); 1 $, 1 cf, MtDandenong, 200 m, ii (TC); 1 cf, Mt Drummer, xii. 1956 (Riek) (ANIC); 1 $, Warburton, xii. 1972(Sedlacek) (TC); 1 $, Woori Yallock, x.1932 (Burns) (NMV). Tasmania: 1 cf , Bronte Park, i (TC); 1 $,Cambridge, i.1965, ex Pinus radiata log (Taylor) (ANIC); 1 $, Cambridge, xii. 1964, exPinus radiata logs(Taylor) (BMNH); 1 $, 2 cf , Hobart, xi.1922 (Cole) (NMV); 1 $, Hobart, xii.1915 (Cole) (NMV); 2 cf ,Launceston, x.1927 (Cole) (NMV); 1 $, Leven R., xi.1975 (Neboiss); 1 $, Ridgeway, ix.1942 (Cole)(NMV); 2 $, Roseberry, i (TC); 1 $, Strahan, iii (TC). Labenagrandissp. n. (Fig. 15) Female. Large species, fore wing length 17 mm. Malar space 0-6 times as long as basal mandibular width.Occipital carina mediodorsally complete, ventrally joining hypostomal carina and continuing to mandibu-lar base as distinct carina, not forming a flange. Pronotum laterally with a median conical tubercle;mesothorax laterally and ventrally bearing sparse, short, white pubescence; mesoscutum strongly,transversely striate; horizontal mesopleural furrow distinct, at least posteriorly; metapleuron vermiculate;submetapleural carina strongly broadened into a triangular striate lobe. Propodeum with anteriortransverse carina present centrally, separating area superomedia and area basalis. Fore wing with 2r-mconverging towards 3r-m, almost joining it at junction with Rs. Hind wing with distal abscissa of Cu\present on membrane, not confluent with Cu\ and cu-a. Mid tibia with longitudinal row of spine-likebristles on outer surface, the distal end swollen somewhat, proximally slender and with a diagonal internalfurrow present. Gaster with sternite 10-5 times as long as hind coxa, reaching to level of spiracle; tergites2-4 smooth and polished ; apex of terminal plate of tergite 8 specialized. Ovipositor projecting beyond apexof gaster by 2-8 times length of hind tibia, the apex bearing very fine teeth. Coloration. Black. Palps, labrum, clypeus, inner orbits, outer orbits, antero-lateral margin of mesoscu- AUSTRALIAN LABENINI & POECILOCRYPTINI 115 turn , tegula , posterior 5 of subalar process , posterior 4 of propodeum , posterior margin of tergites 1-7 ,fore and mid legs (except inner surface of femur and tibia), hind leg with distal tip of coxa, trochanter,femur, proximal end of tibia and tarsus yellow.Male. Unknown. REMARKS. Only a single specimen of this species is known. It is immediately distinguished by possession of auniquely specialized tergite 8, having the metapleuron strongly vermiculate and having a swollenpronotum. The hind coxae are shorter and stouter than those of other species and the petiole isconsiderably more robust. MATERIAL EXAMINEDHolotype $, Western Australia: Yallingup, Cape Naturaliste, ix-x.1913 (Turner) (BMNH). Labenajacunda sp. n. (Fig. 17) Female. Large species, fore wing length 13-14 mm. Malar space 0-6 times as long as basal mandibularwidth. Occipital carina mediodorsally complete, ventrally joining hypostomal carina far from base ofmandible. Pronotum laterally weakly convex; mesothorax laterally and ventrally bearing sparse pubes-cence; mesoscutum with superficial punctures; horizontal mesopleural furrow indistinct; metapleuronfinely sparsely punctate; submetapleural carina broadened anteriorly into ribbed lobe. Propodeum withanterior transverse carina complete so area superomedia is distinctly delineated. Fore wing with 3r-mslightly inclined, 2r-m strongly convergent anteriorly. Hind wing with distal abscissa of Cui complete. Midtibia slender, cylindrical, with scattered spines on outer surface. Gaster with sternite 1 0-6-0-7 times as longas hind coxa, reaching to level of spiracle, the membranous portion unusual in being a pair of well-developed crests; tergites 2-4 smooth and polished; apex of terminal plate of tergite 8 simple. Ovipositorprojecting beyond apex of gaster by about 3-5 times length of hind tibia, apex bearing 8 distinct strongteeth. Coloration. Black. Face, frontal and genal orbits narrowly, tegula, small mark on subalar prominence,scutellum, posterior part of propodeum and posterior margins of all tergites yellow; mesopleuron and baseof petiole reddish; anterior two pairs of legs predominantly yellow; hind legs black, distal apex of coxa andfemur and proximal end of tibia and basitarsus yellow. Male. Similar to female but with fore wing length 13-14 mm; malar space 0-3 times basal mandibularwidth; gaster with sternite 11-0 times as long as hind coxa. Gonosquama distally flattened, impressed, theimpressed area surrounded by a fringe of long hairs. Black with face, frontal and genal orbit, tegula,subalar prominence, scutellum, postscutellum, hind margin of propodeum, broad posterior bands ongastral tergites, anterior 2 pairs of legs, hind trochanter and trochantellus, proximal 0-5 of tibia, entirebasitarsus and extreme proximal apex of second tarsal segment yellow. Flagellum black, with a subapicalwhitish mark that is not a complete band. REMARKS. The female of this species is immediately recognizable by the slender mid tibia and ventralpetiolar prominences which are unique features of this species amongst Australian Labena. The broadyellow bands on the gaster distinguish the male ofjacunda from pudenda and keira, the only other specieswith ornamented male gonosquamae. MATERIAL EXAMINEDHolotype $, Victoria: Chiltern (NMV).Paratypes. Victoria: 2 cT, Mt Buffalo, 4,500', 13.U955 (Neboiss) (NMV). Labena keira sp. n. (Figs 9, 12, 18, 19) Female. Large species, fore wing length 11-16 mm. Malar space 0-5 times as long as basal mandibularwidth. Occipital carina mediodorsally complete, ventrally joining hypostomal carina and forming a flange.Pronotum laterally with a slight median swelling; mesothorax laterally and ventrally bearing no whitepubescence; mesoscutum smooth, with sparse fine punctures; horizontal mesopleural furrow distinct andwidening posteriorly; metapleuron smooth and polished; submetapleural carina slightly broadenedanteriorly, not produced into a lobe. Propodeum with anterior transverse carina present centrally,separating area superomedia from area basalis. Fore wing with 3r-m and 2r-m convergent, sometimesalmost joining at junction with Rs. Hind wing with distal abscissa of Ci complete. Mid tibia with alongitudinal row of spine-like bristles on outer surface, proximally slender and slightly flattened, distally 116 I. D. GAULD & G. A. HOLLOWAY swollen. Gaster with sternite 1 0-7 times as long as hind coxa, just reaching to slightly behind level ofspiracle ; tergites 2-4 smooth and polished ; apex of terminal plate of tergite 8 simple . Ovipositor projectingbeyond apex of gaster by 3 times length of hind tibia, the apex bearing very fine teeth. Coloration. Black to deep red. Palps, labrum, clypeus, face, outer orbits, frons (except ocellar triangleand vertex), scutellum, scutellar ridges, postscutellum, postscutellar ridges, tegulae, subalar process, smallmesopleural macula, posterior 0-5 of propodeum, posterior margin of tergites 1-8 (often dividedmedially), fore and mid legs except sometimes distal 0-5 of mid femur, hind trochanters, basal 0-5 of tibiaand basal tarsal segment yellow. Disruptive mark of antenna medially placed, apex of fore wing infumate. Male. Similar to female but with fore wing length 8-13 mm; malar space 0-4 times basal mandibularwidth; gaster with sternite 1 0-9 times as long as hind coxa. Gonosquama distally flattened and fringedpartially by very fine hairs. Coloration. As for female, sometimes with less yellow on gaster and no whitedisruptive marks on antenna. REMARKS. L. keira is immediately recognizable by the infumate mark on the distal apex of the fore wing.Structurally it is rather similar to L. pudenda from which it may be separated by the submetapleural carinabeing barely expanded, spinose mid tibia and centrally complete anterior transverse carina of thepropodeum. This is one of the most widely distributed species of Labena and is recorded from Victorianorth to central Queensland. It is also known to occur on Lord Howe Island. MATERIAL EXAMINED Holotype $, New South Wales: Mt Tomah, 28.iii.1980 (Rodd) (AM). Paratypes. Queensland: 1 $, Bunya Mtns, i.1938 (Geary} (AM); 3 cf , Eungella, ix.1923 (NMV); 2 $,Eungella, xi (TC); 1 $ , 9 cf , Montville, ix.1955 (Bums) (NMV); 1 $ , 9 cf , Mt Glorious, xi & i (TC); 4 $ , 1Cf , Mt Tambourine, x. 1977 (Galloway) (BMNH); 7 $ , 8 cf, Mt Tambourine, xi-xii (TC); 1 $ , Mt Tip Tree(17-02S 145-37E), x.1980 (Cardale) (ANIC); 1 cf, Westwood, xi.1924 (Burns) (NMV); 1 cf , Wilson'sPeak, Killarney, 9.U977 (Boucek) (BMNH). New South Wales: 1 cf , Lord Howe Is., (AM); 1 cf , LordHowe Is., 30.xi.1955 (Paramonov & Leipa) (ANIC); 1 cf , Lord Howe Is. xii.1977 (Liepa) (ANIC); 1 cf ,Mooney Mooney Creek, near Gosford, xi.1975 (McAlpine & Schneider) (AM). Australian CapitalTerritory: 1 $, Mt Gingera, i.1957 (Riek) (ANIC). Labena malecasta sp. n. (Fig. 16) Female. Large species, fore wing length 13-15 mm. Malar space 0-8 times as long as basal mandibularwidth. Occipital carina mediodorsally complete, ventrally joining hypostomal carina and continuing tomandibular base as a distinct carina, not developed into a flange. Pronotum laterally with a medianswelling; mesothorax laterally and ventrally bearing sparse, long, white pubescence; mesoscutum finelypunctate with indications of weak transverse striae; horizontal mesopleural furrow indistinct; metapleuronindistinctly puncto-striate; submetapleural carina anteriorly abruptly expanded into an almost quadratelobe. Propodeum with anterior transverse carina present centrally, separating area superomedia from areabasalis, latter with transverse striae. Fore wing with 3r-m converging towards 2r-m, almost joining atjunction with Rs. Hind wing with distal abscissa of Ci present in membrane, not confluent with Cu\ andcu-a. Mid tibia with a longitudinal row of spine-like bristles on outer surface, proximally slender, distallyswollen, with an oblique furrow on inner surface. Gaster with sternite 1 0-7 times as long as hind coxa,reaching to well behind level of spiracle; tergites 2-4 smooth and polished with sparse, fine punctures; apexof terminal plate of tergite 8 simple . Ovipositor pro j ecting beyond apex of gaster by at least 3 5 times lengthof hind tibia, the apex bearing distinct teeth. Coloration. Black to red. Palps, clypeus, inner orbit of face, emargination of orbit opposite antennalsocket, thin line on outer orbit, antero-lateral corner of mesoscutum, tegula, subalar process, posterior 0-3of scutellum, postscutellum, posterior 0-3 of propodeum, posterior margin of tergites 1-7, fore leg (exceptcoxa and distal part of femur), mid leg with distal end of femur, tibia and tarsal segments 1-4, hind leg withdistal end of tibia and basal tarsal segment yellow. Antennal disruptive mark at distal end except for last1-2 segments which are blackish. Male. Similar to female but with fore wing length 12 mm; malar space 0-5 times basal mandibular width;gaster with sternite 10-9 times as long as hind coxa. Gonosquama with an apical tuft of long hairs.Coloration similar to female. REMARKS. This large species is structurally rather similar to L. chadwickii. The female has a subtly differentcolour pattern and a longer ovipositor whilst the male is distinct in bearing a tuft of long hair on thegonosquama. L. malecasta is only known from Tasmania, Victoria and the southern alps of New SouthWales. AUSTRALIAN LABENINI & POECILOCRYPTINI 117 MATERIAL EXAMINED Holotype $, Tasmania (BMNH). Paratypes. New South Wales: 1 cf, Mt Kosciusko, Dainer's Gap, x.1929 (Musgrave) (AM). Victoria:1 9, Warburton, ii-iii (TC). Tasmania: 1 $ (BMNH). Labena pudenda sp. n. (Figs 10, 20, 21) Female. Large to very large species, fore wing length 13-20 mm. Malar space 0-7 times as long as basalmandibular width. Occipital carina mediodorsally complete, ventrally joining hypostomal carina andforming flange. Pronotum laterally with a median convex swelling; mesothorax laterally and ventrallybearing sparse, short white pubescence; mesoscutum smooth with sparse fine punctures; horizontalmesopleural furrow distinct; mesopleuron smooth and polished; metapleuron smooth and polished;submetapleural carina anteriorly expanded with a large rounded striate lobe. Propodeum with anteriortransverse carina absent centrally so areae superomedia and basalis are confluent. Fore wing with 3r-mslightly inclined, 2r-m strongly converging towards 3r-m, both widely separated at junction with Rs. Hindwing with distal abscissa of Ci complete. Mid tibia without obvious spine-like bristles on outer surface,proximally slender and slightly flattened, distally swollen. Gaster with sternite 1 0-7-1-0 times as long ashind coxa, reaching behind level of spiracle; tergites 2-4 smooth and polished; apex of terminal plate oftergite 8 simple. Ovipositor projecting beyond apex of gaster by 4-0-4-4 times length of hind tibia, the apexbearing strong teeth. Coloration. Black to deep red. Palps, labrum, clypeus, inner and outer orbits, ventral corner ofpropleuron, tegula, subalar process, scutellum, postscutellum, posterolateral corner of propodeum,posterior margin of tergites 1-7 (may be divided medially), fore and mid leg with distal part of coxa,trochanter, femur, distal and proximal ends of tibia and tarsal segments 1-4, hind leg with distal end of eachsegment including tarsal segment 1 and entire tarsal segments 2-4 yellow. Male. Similar to female but with fore wing length 10-11 mm; malar space 0-9 times basal mandibularwidth; gaster with sternite 1 0-9 times as long as hind coxa. Gonosquama distally flattened and fringedalmost entirely by very fine hairs. Coloration similar to female. REMARKS. This large, rather slender species is easily recognized by the very long ovipositor of the female.The tip of this organ is distinctive in having much coarser teeth than are usually found in species of Labena.Males have the most specialized gonosquamae of any Australian Labena species. L. pudenda seems to be asouthern temperate forest species and has been recorded from New South Wales, Victoria and theAustralian Capital Territory. MATERIAL EXAMINED Holotype $, Victoria: BogongHigh Plains, xii.1931 (Kubala) (NMV). Paratypes. New South Wales: 1 cf, Dainer's Gap (36-12S 148-43E), xi.1973 (Morrow) (ANIC); 1 $,Kosciusko, xii.1922 (Goldfinch) (AM); 1 $, Mt York, x.1930 (NMV). Australian Capital Territory: 1 $,Lees Springs, xi.1953 (Riek) (ANIC). Victoria: 1 ? , 1 cf , 'Alps', xii. 1910 (NMV); 1 $ , Mt Buffalo, 1600 m,ii (TC); 1 ? , 2 cf , 'Victoria', ii.1901 (French) (BMNH); 1 $ , Yarra Falls, S. Warburton, i.1907 (Barnard)(NMV). CERTONOTUS Kriechbaumer Certonotus Kriechbaumer, 1889: 308. Type-species: Certonotus varius Kriechbaumer, by monotypy.Asperellus Townes in Townes et al., 1961: 471. Type-species: Certonotus hinnuleus Krieger, by originaldesignation. Syn. n. Small to very large insects, fore wing length 4-17 mm; clypeus flat, transverse, margin thin, evenly arcuate;labrum barely projecting; mandible short, stout but tapered, twisted 25-30, with upper tooth the longer;outer mandibular surface with a groove bearing hairs; malar space trans-striate, usually a little longer thanbasal mandibular width. Occipital carina dorsally absent; eye not indented next to antennal socket.Antenna slightly clavate, apically pointed, without a flat sensillum. Mesoscutum with transverse rugae,notauli weak, notaular crests very weak. Propodeum usually quite short, convexly rounded with spiracleelliptical; anterior transverse carina usually complete except centrally, other carinae often reduced, areasuperomedia usually not delineated; gaster inserted high up on propodeum, above level of hind coxae.Fore tibia with a short tooth on outer distal margin; fore tarsus unspecialized; hind coxa of female with ananterior carina continued ventrally as a process, the area behind this carina concave and closely punctate; 118 I. D. GAULD & G. A. HOLLOW AY tarsal claws simple. Fore wing with cu-a opposite or proximal to base of Rs&M; 3r-m usually present,areolet almost triangular, often petiolate above; 2m-cu sinuous, with two close bullae. Hind wing withdistal abscissa of Cu\ present or incomplete or absent; first abscissa of Cu\ shorter than cu-a; basal cellslender; Sc with one or two hamuli. Gaster quite long, tergite 1 from stout to quite slender, with spiraclesbefore centre; sternite 1 usually reaching nearly to level of spiracles; laterotergites 2-4 membranous,folded under; tergite 8 highly modified, projecting laterally as a pair of prominences at either side ofovipositor base, dorsally with a detached plate projecting through concave orifice in hind margin, tergite 7often mediodorsally incised. Ovipositor projecting beyond apex of gaster by 3-0-8-0 times length of hindtibia, its apex compressed, the upper valve with weak blunt serrations, the lower valve enclosing the upperand bearing fine file-like teeth. REMARKS. Certonotus is a large genus centred in Australia with a few species present in New Guinea, NewZealand and South America. Previously, the species now included in this taxon were divided betweenCertonotus and Asperellus (Townes, 1969; Gauld, 1984) but more detailed study has revealed that thelatter genus is almost certainly polyphyletic. In the present work 23 species are recognized as occurring inAustralia. Certotonus species are easily recognized by the possession of transverse rugae on the mesoscutum, afeature immediately distinguishing the genus from all other Australian labenines. Certonotus species mayinadvertently be confused with Rhyssini, especially Epirhyssa species. Unlike rhyssines, which have anundeveloped submetapleural carina, Certonotus has a broad expanded lobe present anteriorly. Further-more, Certonotus has fine file-like teeth on the ovipositor apex, not the coarse teeth found in Epirhyssa(Gauld, 1984). The relationships of the Australian species. Gauld (1984) suggested that Asperellus, as defined by Townes(1969), was merely a specialized species-group of Certonotus. Further study suggests that Asperellus is noteven a monophyletic group, but rather an assemblage of species of Certonotus that lack the distal abscissaof Cu\. The majority of species (farrugiai, pineus, mogimbensis, zebrus, toolangi, hinnuleus andleeuwinensis) do comprise a natural group, the leeuwinensis-group. All have rather similar propodealcarination, in that the posterior transverse carina is absent but a large smooth area is enclosed by theanterior carina and the lateral carinae; they possess a very small areolet, have one or more spine-likebristles present on the hind tibia and have a slightly convex face. Most have a rather short first sternite, aquite deeply divided tergite 7 and a pronounced lobe on tergite 8. Only in farrugiai is tergite 7 barelyindented posteriorly. The species ixion, paluma, celeus and Certonotus species A would all run toAsperellus in Townes' (1969) key but these do not appear to be closely related to the others (i.e. theleeuwinensis-group). The first three are closely related to Certonotus talus and belong to the humeralifer-group, which is defined by having a very deeply divided tergite 7 and possessing rather long narrowprocesses on tergite 8. The group includes seven Australian species - humeralifer, apicalis, talus, cestus,ixion, paluma and Certonotus species A. The New Zealand species C. fractinervis apparently also belongsto this group. The majority of other Australian species (annulatus, nitidulus, geniculatus, rufescens, andrewi, avitus,sisyphus and celeus) constitute a third group, having a weakly to moderately deeply divided tergite 7,rounded lobes on tergite 8 and a relatively long first sternite. Within the nitidulus-group one ratherdistinctive lineage can be recognized comprising rufescens, geniculatus, sisyphus and celeus. These speciesall have flat lower faces, and most have the posterior end of the lateral propodeal carina broadened to forma raised keel. The remaining Australian species, C. monticola, belongs to the/Zav/ce/w-group which is characterized bypossession of elongate glossae and a very short occipital carinal stub. Males of species in this group arespecialized in having a very elongate gaster (with tergite 7 about 1-5 times as long as broad or longer).Frequently the hind margins of the tergites are concave. This elongate form of the gaster is characteristic ofwood-boring species in which the male copulates with the female prior to her emergence from the burrow(see Nuttall, 1973) . The male of C. fractinervis resembles males of theflaviceps-group though the structureof the female suggests the species is best placed in the humeralifer-group. The relationships of the species-groups are difficult to determine as there is considerable conflict in thecharacters. The leeuwinensis-group, a holophyletic clade, may well be the sister-lineage of the humeralifer-group. Both have a similarly deeply divided tergite 7 and many often possess spine-like bristles on the hindtibia and show reduction in the distal abcissa of Cu\. This arrangement leads to difficulty in placing C.farrugiai which has tergite 7 barely divided. The humeralifer-group could be paraphyletic with respect tothe leeuwinensis-group. The nitidulus-group could well be a paraphyletic assemblage, the stem group fromwhich all others have arisen, although the rufescens-subgroup is clearly a holophyletic clade. Theflaviceps-group is also undoubtedly a holophyletic clade. The geographical distribution of the species- groups is as follows: AUSTRALIAN LABENINI & POECILOCRYPTINI 119 nitidulus-group: New Guinea, Australia;rufescens-subgroup: Australia;humeralifer-group: Australia, New Zealand;flaviceps-group: Moluccas, New Guinea, tropical Australia;leeuwinensis-group: New Guinea, Australia, New Hebrides, New Caldeonia. Key to species of Certonotus occurring in Australia 1 Hind wing with distal abscissa of Cui complete 2 - Hind wing with distal abscissa of Cui either incomplete, that is present in membrane but not joining Ci/i & cu-a, or entirely absent 13 2 Gaster more or less entirely yellowish or orange-brown , the tergites in dorsal view , unicolorous or slightly infurcate along posterior margin 3 - Gaster dark reddish brown or black, usually with conspicuous yellow spots or bands along margin of tergites , some marks with gaster unicolorous dark brown 7 3 Subalar prominence, in dorsal view, strongly raised, with a blunt, back-curved, thorn-like protuberance which is more sharply pointed in the male than the female (Fig. 22); malarspace long (Fig. 25), in female 0-9-1-0 times as long as basal mandibular width, in male0-8-0-9 times as long; submetapleural carinal flange with a ridge delimiting an anterolateraltriangular area rufescens Morley (p. 133) - Subalar prominence in dorsal view at most convex, not produced into a spine-like protuberance (Figs 23, 24); malar space fairly short (Fig. 26), in female 0-7-0-8 or male 0-4-0-6 times aslong as basal mandibular width; submetapleural carinal flange without a delimited triangulararea anterolaterally 4 4 Pronotum in dorsal view with part before upper corner strongly convex or pyramidal (Figs 23, 24); anterior transverse carina of propodeum present centrally (Fig. 27); 2r-m and 3r-m fusedanteriorly so areolet is petiolate (Fig. 30) 5 - Pronotum in dorsal view with part before upper corner flat to weakly convex; anterior transverse carina of propodeum absent centrally (Figs 28, 29); 2r-m and 3r-m joining Rsseparately so areolet not petiolate (Fig. 31) 6 5 Upper part of pronotum pyramidal in dorsal view (Fig. 24); female with ovipositor about 5 times as long as hind tibia; hind tibia with one spine-like bristle on posterior margin;flagellum entirely black humeralifer Krieger (p. 127) - Upper part of pronotum simply strongly convex in dorsal view (Fig. 23) ; female with ovipositor 7-5 or more times as long as hind tibia; hind tibia without a spine-like bristle or posteriormargin; flagellum with distal end white apicalis Morley (p. 122) 6 Posterior transverse carina of propodeum centrally strongly raised (Fig. 28); antenna with apical 10 or so segments whitish; spiracular area bounded posteriorly by carina, thus beingseparated from lateral area nnnulatus Morley (p. 121) - Posterior transverse carina of propodeum absent centrally (Fig. 29); antenna with apical segments black but with a subapical white band; spiracular area confluent with lateral area andrewi sp. n. (p. 121) 7 Males and females with tergites 2-5 with yellow band along hind margin 8 - Females with tergites 2-5 with paired yellow spots or the males either with paired yellow spots or without yellow marks at all 9 8 Metapleuron centrally with conspicuous longitudinal wrinkles; propodeum with a tubercle below spiracle (Fig. 32) ; flagellum distally uniformly black; anterolateral part of mesoscutumwithout yellow marks geniculatus Morley (p. 124) - Metapleuron centrally smooth; propodeum without a tubercle below spiracle (Fig. 33); flagellum distally white-marked; anterolateral part of mesoscutum yellow-marked nitidulus Morley (p. 131) 9 Subalar prominence bearing a long slender back-curved spine (Fig. 34); female with tergite 8 posteriorly extended into a narrow truncate projection (Fig. 35) talus sp. n. (in part) (p. 135) - Subalar prominence without a spine; female with tergite 8 bluntly rounded apically (Figs 36, 37) 10 10 Labium with glossae very long and slender, extending ventrally a distance of approximately theheight of the eye; occipital carina represented by a short vestige at its junction withhypostomal carina (Fig. 38); propodeum (at least part behind anterior transverse carina)yellow monticola Morley(p. 130) 120 I. D. GAULD & G. A. HOLLO WAY Labium with glossae short, barely projecting below head; occipital carina present ventrally, atleast as long as abscissa of hypostomal carina between mandible and junction with occipitalcarina (Fig. 39); propodeum not or only partially yellow-marked behind anterior transversecarina 11 1 1 Propodeum with anterior and posterior transverse carinae strong , complete and almost parallel to each other (Fig. 40); lateromedian carinae not present between transverse carina; hind tibia uniformly reddish; females with malar space brown sisyphus sp. n. (p. 134) Propodeum with anterior or posterior transverse carina weak or missing in part, the carinae notparallel and often with discernible traces of lateromedian longitudinal carinae between them(Fig . 41 ) ; hind tibia proximally pale-marked ; females with malar space whitish 12 12 Sternite 1 long, extending behind spiracles and being about length of hind coxa; submetapleu- ral carina anteriorly broadened to form a rectangular flange (Fig. 42); mesoscutum with pair of pale stripes extending back from anterior margin avitus sp. n. (p. 123) Sternite 1 short, reaching at most (in males) to level of spiracles, and being distinctly snorterthan length of hind coxa; submetapleural carina anteriorly broadened to form a roundedlobe (Fig. 43) ; mesoscutum with a median pale rectangular mark cestus sp. n. (p. 123) 13 Hind wing with distal abscissa of Cu\ absent; hind tibia with posterior margin bearing one or more spine-like bristles 14 Hind wing with distal abscissa of Cu\ present in membrane, not joining nervellus; hind tibiawith posterior margin devoid of spine-like bristles 22 14 Tergite 2 of gaster entirely white; alitrunk anteriorly reddish brown, propodeum infuscate, gaster in greater part black (Fig. 58); wings uniformly infumate; antenna of female black; female with posterior margin of tergite 7 only slightly indented farrugiai sp. n. (p. 126) Tergite 2 of gaster yellowish brown, or dark brown or pale spotted but never entirely white(Figs 59-62) ; alitrunk variously coloured, if brownish then concolorous with much of gaster ;wings hyaline or apically infumate; antenna of female with white bands; female withposterior margin deeply indented medially 15 15 First segment of gaster exceptionally long and slender, the Sternite far longer than length of hind coxa; propodeum with first and second lateral areae clearly defined and separated (Fig. 44); apex of fore wing of female narrowly infumate celeus sp. n. (p. 125) - First segment of gaster stouter , the Sternite not longer than hind coxa ; propodeum with first and second lateral areae confluent, often not defined laterally (Fig. 45); apex of fore wing notinfumate 16 16 Gaster more or less uniformly yellowish or orange; female with at least distal two flagellar segments entirely black so white ruptive mark is a subapical band 17 - Gaster predominantly black with white maculae; female with distal end of flagellum white except for tip of apical segment which is black 18 17 Pronotum convexly produced before upper hind corner, the convexity almost pyramidal in dorsal view (Fig. 46); female with tergite 9 in dorsal view transverse; female hind tarsus black; male with 3r-m present pineus sp. n. (p. 132) Pronotum weakly convex before upper hind corner (Fig. 47); female with tergite 9 in dorsalview elongate; female hind tarsus yellow; male with 3r-m absent. mogimbensis Cheesman (p. 129) 18 Tergite 2-5 with hind margin banded with yellow, occasionally with bands very faint so tergite is almost unicolorous 19 Tergite 2-5 with yellow spots on posterolateral corners, these marks not confluent centrally 21 19 Propodeum with area superomedia distinct, hexagonal, with only posterolateral sides rather weak (Fig. 48); tergite 8 produced into long slender processes (Fig. 50); tergites 1-2 not clearly whitish or yellow-marked species A (p. 137) Propodeum with area superomedia undefined laterally and posteriorly (Fig. 49); tergite 8produced into blunt or moderately long processes (Fig. 51); tergites 1-2 clearly whitish oryellow-banded posteriorly (Fig. 59) 20 20 Metapleuron closely punctate; lower face distinctly transverse (Fig. 52); hind coxa dorsally brown; propodeum in lateral aspect brownish leeuwinensis Turner (p. 128) Metapleuron virtually smooth; lower face longer than broad (Fig. 53); hind coxa entirely black;propodeum in lateral aspect predominantly whitish zebrus sp. n. (p. 136) 21 Pronotum convexly produced before upper hind corner; propodeum in lateral view extensively whitish; metapleuron virtually impunctate; hind tibia with an indistinct yellowish markproximally hinnuleus Krieger (p. 126) AUSTRALIAN LABENINI & POECILOCRYPTINI 121 Pronotum very weakly convex before upper hind corner; propodeum in lateral view brown(Fig. 60); metapleuron closely punctate; hind tibia with proximal 0-5 whitish toolangi sp. n.(p. 136) 22 Subalar prominence bearing a long slender back-curved spine (Fig. 34); female flagellum with a subapical white band, the distal 5 or so segments black talussp. n. (in part) (p. 135) Subalar prominence simple; flagellum with apical segments white, only extreme distal apex oflast segment blackish or entirely black 23 23 Tergites 4 and 5 of gaster with anterolateral corners broadly white (Fig. 62) ; hind coxa in profile rather short and stout, the ovipositor guide reaching to about the centre (Fig. 54); process on tergite 8 very slender; ovipositor about 4 times as long as hind tibia paluma sp. n. (p. 132) - Tergites 4 and 5 of gaster with only small white triangular marks in posterolateral corners (Fig.61); hind coxa in profile slender, the ovipositor guide reaching about 0-3 of its length (Fig.55); process on tergite 8 moderately slender; ovipositor about 5-5 times as long as hind tibia ixion sp. n.(p. 128) Certonotus andrewisp. n. (Fig. 29) Female. Medium to large-sized species, fore wing length 7-14 mm. Labium with glossae unspecialized.Lower face at narrowest point 1-0-1-1 times as broad as high; malar space 0-7-0-8 times as long as basalmandibular width. Occipital carina ventrally more than twice as long as abscissa of hypostomal carinabetween it and base of mandible. Upper part of pronotum, slightly before posterior corner, almost flat, indorsal view barely projecting beyond scutal margin; subalar prominence moderately convex. Scutellumtransversely striate, crest strong; metapleuron smooth and polished; submetapleural carina anteriorlyexpanded into a moderately broad, rather long lobe; metanotum with very strong tooth opposite anteriorend of lateral carina. Propodeum moderately short with anterior transverse carina incomplete centrally;posterior transverse carina absent centrally; lateromedian longitudinal carina present only before anteriorcarina; pleural carina complete, but weak posteriorly; area superomedia undefined; area spiracularisconfluent with first lateral area; first and second lateral area separated by a carina. Fore wing with 3r-mconverging towards 2r-m, joining Rs separately; 2m-cu joining M 0-25 to 0-50 from 3r-m towards 2r-m.Hind wing with distal abscissa of Cu^ distinct to wing margin. Hind tibia with posterior margin withoutspine-like bristles. Gaster with segment 1 short, the sternite 0-5-0-6 times as long as hind coxa, reaching tolevel of spiracle. Tergite 7 mediodorsally without any indentation; tergite 8 posteriorly with process shortand rounded; tergite 9 in dorsal view transverse. Ovipositor projecting beyond apex of gaster by 3-5 to 4-0times length of hind tibia. Coloration. Orange-yellow species with frons centrally, vertex, mesoscutal stripes, most of gaster andhind legs more brownish or orange; flagellum black with subapical white band. Pterostigma brown, wingshyaline. Male. Similar to female. i REMARKS. Very like C. annulatus which it resembles in colour, venation, possession of short malar spaceand an almost flat pronotum. C. andrewi appears to be a more southern species than annulatus as it hasbeen collected in southern Queensland, New South Wales and Victoria. MATERIAL EXAMINED Holotype $, New South Wales: Heathcote, near Sydney, x.1979 (Holloway) (AM). Paratypes. Queensland: 1 <j>, Brisbane, xii.1972 (SedlaceK) (TC); 4 <j>, Mt Glorious, near Brisbane,xii.1976 (Boucek) (BMNH); 1 $, 6 cf , Mt Glorious, xi (TC). New South Wales: 1 $, Cabbage Tree Ck,Clyde Mtn, ix.1979 (Naumann & Cardale) (ANIC); 1 $, Iluka, Clarence R., rain forest, xi.1970(McAlpine) (AM); 1 $, Warren, ix.1982 (Holloway) (AM). Victoria: 1 $, 'Victoria' (NMV). Certonotus annulatus Morley(Figs 28, 31) Certonotus annulatus Morley, 1913: 31; Turner, 1919: 551. LECTOTYPE $ , QUEENSLAND (BMNH), heredesignated [examined]. Female. Moderately large species, fore wing length 9-13 mm. Labium with glossae very slightly leng-thened. Lower face at narrowest point 0-9 times as broad as high; malar space 0-7-0-8 times as long as basalmandibular width. Occipital carina ventrally about as long as abscissa of hypostomal carina between it and 122 I. D. GAULD & G. A. HOLLOWAY mandibular base. Upper part of pronotum, slightly before posterior corner, very weakly convex, in dorsalview projecting only slightly beyond scutal margin; subalar prominence weakly convex. Scutellumpunctate, transverse crest distinct; metapleuron virtually smooth except for a few striae ventrally andposteriorly; submetapleural carina anteriorly expanded into a broad rounded lobe that usually bearsconcentric striae; metanotum with a blunt tooth opposite anterior end of lateral carina. Propodeummoderately long with anterior transverse carina centrally absent; posterior transverse carina centrallypresent; lateromedian longitudinal carina complete only anteriorly and with stub behind anterior trans-verse carina; pleural carina complete; area superomedia indicated, but incomplete anteriorly and laterally;area spiracularis complete; first and second lateral areae separated, distinctly delineated. Fore wing with3r-m converging towards 2r-m, joining Rs separately; 2m-cu joining M 0-2-0-3 from 3r-m towards 2r-m.Hind wing with distal abscissa of Cu\ distinct to wing margin. Hind tibia with posterior margin withoutspine-like bristles. Gaster with segment 1 short, the sternite 0-5-0-6 times as long as hind coxa, reachingbehind level of spiracle. Tergite 7 mediodorsally with a short, narrow slit on posterior margin; tergite 8posteriorly with process long and truncate; tergite 9 in dorsal view elongate and weakly roundedposteriorly. Ovipositor projecting beyond apex of gaster by 4-5 times length of hind tibia. Coloration. Orange species with face and orbits, notauli, subalar prominences, scutellar margins andhind margins of gastral tergites more yellowish; flagellum black with distal 10 or so segments white.Pterostigma blackish, wings very weakly infumate. Male. Similar to female but with fore wing length 8 mm; malar space 0-4 times basal mandibular width;3r-m present; gaster with segment 1 moderately slender, the sternite 0-5 times as long as hind coxa. Apex ofgonosquama flattened, bearing a tuft of long hairs. Coloration as for female. REMARKS. C. annulatus is rather similar to and probably the sister species of C. andrewi. The two speciesare fairly easily separated by the characters given in the key.C. annulatus seems to be a north Queensland species. MATERIAL EXAMINED Lectotype $ , Queensland: Kuranda near Cairns, xii.1901 (BMNH). Queensland: 2 $ (paralectotypes), Kuranda, xii.1901, iv.1902 (BMNH); 1 $, Kuranda, ii.1935 (Burns)(NMV); 1 , Mission Beach, near Tully, iv.1971 (Moulds} (AM); 1 $, 2 cf , Moses Ck, 4 km N. by E. MtFinnigan (15-47S 145-17E), at light, x.1980 (Cardale) (ANIC); 2 cf , Paluma (19-OOS 146-12E), 900 m,Malaise trap, x.1980 (Frith) (ANIC); 1 $, Shipton's Flat (15-47S 145-14E), at light, x.1980 (Cardale)(ANIC). Certonotus apicalis Morley(Fig. 23) Certonotus apicalis Morley, 1913: 31; Turner, 1919: 551. LECTOTYPE $, QUEENSLAND (BMNH), heredesignated [examined] Female. Large species, fore wing length 12-15 mm. Labium with glossae slightly lengthened. Lower face atnarrowest point 1-0-1-1 times as broad as high; malar space 0-7-0-8 times as long as basal mandibularwidth. Occipital carina ventrally sinuous, more than twice as long as abscissa of hypostomal carina betweenit and base of mandibles. Upper part of pronotum, slightly before posterior corner, convex, in dorsal viewprojecting as a rounded protuberance; subalar prominence quite strongly convex. Scutellum punctate,transverse crest weak; metapleuron with fine sparse punctures; submetapleural carina anteriorly expandedinto a broad rounded lobe; metanotum with a weak swelling in front of lateral carina. Propodeummoderately short with anterior transverse carina complete; posterior transverse carina present centrally;lateromedian longitudinal carina present before anterior carina only; pleural carina strong anteriorly,weak posteriorly; area superomedia not defined laterally; area spiracularis complete; first and secondlateral areae more or less completely separated, the latter usually undefined externally. Fore wing with3r-m fused anteriorly with 2r-ra, so areolet is petiolate; 2m-cu joining M 0-20-0-25 basad of 3r-m. Hindwing with distal abscissa of Cu\ distinct to wing margin. Hind tibia with posterior margin without spine-likebristles. Gaster with segment 1 short, the sternite 0-5 times as long as hind coxa, not reaching level ofspiracle. Tergite 7 mediodorsally divided about 0-5 times its length; tergite 8 posteriorly with long narrowprocess; tergite 9 in dorsal view transverse, rounded apically. Ovipositor projecting beyond apex of gasterby 7-5 or more times length of hind tibia. Coloration. Yellowish brown species, antenna black with distal segments white, the apical one or twosegments slightly infuscate. Pterostigma black, wings weakly infumate. Male. Similar to female but with fore wing length 9 mm; malar space 0-5 times basal mandibular width; AUSTRALIAN LABENINI & POECILOCRYPTINI 123 3r-m fused anteriorly with 2r-m; gaster with segment 1 short, the sternite 0-7 times as long as hind coxa,apex of gonosquama flattened, with a tuft of long close hairs. Colour similar to female but pterostigma darkbrown. REMARKS. The females of this species are easily distinguished by their very long ovipositors, the apices ofwhich are fairly bluntly pointed. The fore femur is also rather distinctive, being slender proximally, thenabruptly inflated proximocentrally. This is most similar to C. humeralifer and the two are probablysister-taxa. C. apicalis is a tropical species, only recorded from Queensland. MATERIAL EXAMINED Lectotype $ , Queensway: Kuranda near Cairns, iv.1902 (BMNH). Queensland: 3 $ (paralectotypes), same data as holotype (BMNH); 1 d", Kuranda, v-vi.1913 (Turner)(BMNH); 1 $, Paluma, mv lamp, i.1970 (Holloway) (AM). Certonotus avitus sp. n. (Figs 41, 42) Female. Medium-sized species, fore wing length 8-10 mm. Labium with glossae unspecialized. Lower faceat narrowest point 1-0 times as broad as high; malar space 0-4-0-6 times as long as basal mandibular width.Occipital carina ventrally sinuous, more than 3 times as long as abscissa of hypostomal carina between itand base of mandible. Upper part of pronotum, slightly before posterior corner, almost flat, in dorsal viewbarely projecting beyond scutal margin; subalar prominence weakly convex. Scutellum finely punctate,crest strong; metapleuron finely punctate, obsolescent striae posteriorly; submetapleural carina anteriorlyexpanded into a rectangular elongate flange; metanotum laterally produced into a stout tooth in front oflateral carina. Propodeum moderately long with anterior transverse carina incomplete centrally; posteriortransverse carina absent except laterally; lateromedian longitudinal carina present only before anteriorcarina; pleural carina incomplete posteriorly; area superomedia indistinct; area spiracularis complete; firstand second lateral areae separated, the second not defined laterally. Fore wing with 3r-m convergingtowards 2r-m, widely separated on Rs; 2m-cu joining M 0-3 from 3r-m towards 2r-m. Hind wing with distalabscissa of Cui distinct to wing margin. Hind tibia with posterior margin without spine-like bristles. Gasterwith segment 1 long, the sternite 1-2 times as long as hind coxa, reaching well behind level of spiracle.Tergite 7 mediodorsally with indentation to 0-25 of tergite length; tergite 8 posteriorly with short, roundedprocess; tergite 9 in dorsal view transverse. Ovipositor projecting beyond apex of gaster by 5 times lengthof hind tibia. Coloration. Flagellum black, distal apex white; head brown, orbits entirely and all of lower face whitish;alitrunk dark brown, notaular stripes, subalar prominence, tegula, scutellum, postscutellum, mesopleuralstripe and posterior part of propodeum/metapleuron yellow. Gaster reddish brown, tergites laterallyyellow and with posterolateral yellow spots. Tergite 1 anteriorly yellowish. Anterior two pairs of coxaewhitish, hind ones brown with white apices; all tibiae and tarsi infuscate; femora brown, distally pale.Pterostigma dark brown; wings hyaline. Male. Similar to female but with fore wing length 8 mm; malar space 0-3-0-4 basal mandibular width;3r-m present; gaster with segment 1 slender, the sternite 1-0 times as long as hind coxa. Gonosquama withapex slightly flattened, with a tuft of close moderately long hairs. Similar in colour to female. REMARKS. C. avitus is a distinctive species with a very characteristic submetapleural carina. For its size thisspecies has a rather large areolet and a fairly long first sternite.It is known from eastern Australia, from southern Queensland to Victoria. MATERIAL EXAMINED Holotype $ , New South Wales: Mt Tomah, Blue Mtns, viii.1979 (Rodd) (AM). Paratypes. Queensland: 1 $ , Eungella, xi (TC); 1 $ , Mt Glorious, i (TC); 1 cf , Mt Nebo, viii (TC); 1 $ ,Mt Tamborine, ix-x.1978 (Galloway) (BMNH). Victoria: 1 cT, Healesville, xi.1943 (NMV); 1 $, KingLake, x.1953 (Burns). (NMV); 1 C?, King Lake, x.1954 (Burns) (NMV); 4 $>, 1 cf, Toolangi, xi.1982(Farrugia) (BMNH). Certonotus cestus sp. n. (Figs 37, 43) Female. Medium-sized species, fore wing length 9 mm. Labium with glossae unspecialized. Lower face atnarrowest point 1-1 times as broad as high; malar space 1-0 times as long as basal mandibular width. 124 I. D. GAULD & G. A. HOLLO WAY Occipital carina ventrally longer than abscissa of hypostomal carina between it and lobe of mandible.Upper part of pronotum, slightly before posterior cornet, weakly convex, in dorsal view barely protrudingbeyond scutal margin; subalar prominence quite strongly convex. Scutellum coarsely punctate, creststrong; metapleuron small and polished; submetapleural carina anteriorly expanded into a rounded lobe,metanotum barely produced before lateral carina. Propodeum moderately long with anterior transversecarina obsolescent centrally; posterior transverse carina weak but more or less complete; lateromedianlongitudinal carina present anteriorly and weakly between transverse carina; pleural carina anteriorlystrong, posteriorly weaker or absent; area superomedia more or less distinct, not defined anteriorly; areaspiracularis complete; first and second lateral areae separated, the first very short, the second ill-definedexternally. Fore wing with 3r-m converging towards 2r-ra, joining Rs separately; 2m-cu joining M opposite3r-m. Hind wing with distal abscissa of Cu\ distinct to wing margin. Hind tibia with posterior margin withtwo spine-like bristles. Gaster with segment 1 short and broad posteriorly, the sternite 0-5 times as long ashind coxa, reaching almost to level of spiracle. Tergite 7 mediodorsally with broad indentation; tergite 8posteriorly short and rounded; tergite 9 in dorsal view very broad. Ovipositor projecting beyond apex ofgaster by 6 times length of hind tibia. Coloration. Very dark brown species, flagellum with a white subapical band; facial, frontal and genalorbits, upper and lower margin of pronotum, mesoscutum in a central quadrate spot and indistinctlylaterally, scutellum, postscutellum, subalar prominence, anterior and posterior mesopleural spots, most ofhind part of propodeum and metapleuron whitish. Gaster dark brown with large anterolateral spots.Anterior coxa whitish, hind coxa brown with whitish mark. Pterostigma blackish, wings hyaline. Male. Similar to female but with fore wing length 9 mm; malar space 0-7 times basal mandibular width;3r-m present; gaster with segment 1 quite stout, the sternite 0-5 times as long as hind coxa. Apex ofgonosquama with a small lobe that bears scattered long hairs. Colour similar to female but face entirelypale. REMARKS. This species is readily recognizable by its colour pattern. Structurally it seems to be related to theC. humeralifer-group as it has tergite 7 very deeply divided. Tergite 8 is less pronounced apically in thisspecies than others in the humeralifer-group.It is known from Queensland and New South Wales. HOST RECORD. Buprestidae: Diadoxussp. (ANIC). MATERIAL EXAMINED Holotype $ , New South Wales: State Forest 854, viii.1952 (Martin) (ANIC), parasite of Diadoxus.Paratype. Queensland: 1 cf , viii.1926 (Jarvis) (BMNH). Certonotus geniculatus Morley(Fig. 32) Certonotus geniculatus Morley, 1913: 28. LECTOTYPE $, VICTORIA (BMNH), here designated [ex-amined] . Female. Large species, fore wing length 12-17 mm. Labium with glossae slightly lengthened. Lower face atnarrowest point 1-1-1-2 times as broad as high; malar space 0-9-1-0 times as long as basal mandibularwidth. Occipital carina ventrally about as long as abscissa of hypostomal carina between it and base ofmandible. Upper part of pronotum, slightly before posterior corner, very weakly convex, in dorsal viewbarely projecting beyond scutal margin; subalar prominence weakly convex. Scutellum punctate withstrong transverse crest; metapleuron longitudinally striate; submetapleural carina anteriorly expandedinto a moderately broad lobe; metanotum produced into a tooth opposite anterior end of lateral carina.Propodeum quite short with anterior transverse carina incomplete centrally; posterior transverse carinavestigial; lateromedian longitudinal carina present only as vestige anteriorly; pleural carina strongly raisedinto tubercle below spiracle, posteriorly obsolescent; area superomedia undefined; area spiracularis notcompletely delineated posteriorly; first and second lateral areae separated, but not clearly delineatedinternally or laterally. Fore wing with 3r-m and 2r-m converging but not joining anteriorly. Hind wing withdistal abscissa of Cu\ distinct to hind margin. Hind tibia with posterior margin without spine-like bristles.Gaster with segment 1 long and narrow, the sternite 1-0-1-3 times as long as hind coxa, reaching behindlevel of spiracle. Tergite 7 mediodorsally with very small indentation on posterior margin; tergite 8posteriorly short and rounded. Tergite 9 in dorsal view broad. Ovipositor projecting beyond apex of gasterby 5-5 to 7-0 times length of hind tibia. Coloration. Antenna black; head black, lower face, frontal and genal orbits yellow. Alitrunk red-brownwith periphery of sclerites infuscate, only tegula, scutellum posteriorly, postscutellum and hind end of AUSTRALIAN LABENINI & POECILOCRYPTINI 125 propodeum yellow. Gaster reddish brown, tergite 1 anteriorly, laterally and posteriorly yellow, othertergites with lateral and posterior margins yellow. Fore leg yellow, proximal part of femur black; mid legyellow, coxa basally, femur proximally and tarsus black. Hind leg black, coxa distally, trochantersegments, distal apex of femur, base and apex of tibia yellow. Pterostigma dark brown, wings virtuallyhyaline. Male. Similar to female but with fore wing length 11 mm; malar space 0-7-0-9 times basal mandibularwidth; 3r-m present; gaster with segment 1 very slender, the sternite 1-3 times as long as hind coxa. Apex ofgonosquama flattened with a dense tuft of long hairs. Colour similar to female. REMARKS. C. geniculatus belongs to the rufescens-subgroup of the tasmaniensis-group. It is probably thesister-species of C. rufescens. Both have the prementum and glossae somewhat lengthened, have a very flatface, the anterior portion of the lateral carina strongly raised and the pleural carina raised into a tuberclebelow the propodeal spiracle. C. geniculatus is most easily distinguished by the possession of a longitudin-ally striate metapleuron.This is a southern species, only recorded from Victoria. MATERIAL EXAMINED Lectotype $, Victoria: Nulla Wurren, near Berwick (BMNH). Victoria: 1 $, 2 O" (paralectotypes), same data as lectotype (BMNH); 1 9> Buckland River, xi.1964(Neboiss) (NMV); 1 cf , Fernshaw (NMV); 1 $, 1 cf , Trafalgar (NMV). Certonotus celeussp. n. (Fig. 44) Female. Medium-sized species, fore wing length 8-9 mm. Labium with glossae slightly lengthened. Lowerface at narrowest point 0-9 times as broad as high; malar space 1-3-1-4 times as long as basal mandibularwidth. Occipital carina ventrally slightly longer than abscissa of hypostomal carina between it and base ofmandible. Upper part of pronotum, slightly before posterior corner, very weakly convex, in dorsal viewprojecting slightly beyond mesoscutal margin; subalar prominence moderately convex. Scutellum sparselypunctate with strong crest posteriorly; metapleuron smooth, very sparsely punctate; submetapleural carinaanteriorly strongly expanded into a quadrate lobe that is abruptly narrowed posteriorly; metanotum with aweak lateral tooth. Propodeum quite long with anterior transverse carina complete; posterior transversecarina complete; lateromedian longitudinal carina present only anterior to anterior carina; pleural carinacomplete; area superomedia not defined; area spiracularis clearly delineated; first and second lateral areaedistinct, separated. Fore wing with 3r-m converging towards 2r-m, joining Rs separately; 2m-cu joining M0-3 from 3r-m towards 2r-m. Hind wing with distal abscissa of Cu\ absent. Hind tibia with posterior marginwith one spine-like bristle. Gaster with segment 1 very long and narrow, the sternite 1-2-1-6 times as longas hind coxa, reaching well behind level of spiracle. Tergite 7 mediodorsally with shallow indentation;tergite 8 posteriorly short and rounded; tergite 9 in dorsal view short and pointed. Ovipositor projectingbeyond apex of gaster by 4-5 times length of hind tibia. Coloration. Red-brown species, flagellum black with subapical white band, lower face yellowish,anterior leg orange. Pterostigma dark brown; wings hyaline, apex of fore wing strongly infumate. Putative male. Similar to female but with fore wing length 8 mm; malar space 0-6 times basal mandibularwidth; 3r-m present; propodeal carinae weaker; gaster with segment 1 slender, the sternite 0-9 times aslong as hind coxa. Apex of gonosquama rounded, with scattered fine hairs. Pale orange, antenna black,head yellowish and wings uniformly hyaline. REMARKS. A distinctive species on account of the infumate tip to the fore wing of the female, the very longpetiole, long hind coxa and rather flat face. It is probably related to C. rufescens though it is the only speciesin this subgroup without the distal abscissa of Cu\. The male, here tentatively associated, has similarly longcoxae and a rather flat face but has no clear posterior transverse propodeal carina nor has the wing apicesinfumate.It is only known from Queensland. MATERIAL EXAMINED Holotype $, Queensland: Baldy Mtn Rd, via Atherton, Malaise trap, vi.1981 (Brown) (QM). Paratypes. Queensland: 1 $, same data as holotype (BMNH); 1 cf, Eungella Nat. Park, xi.1976(Boucek) (BMNH); 1 $, Windsor Tableland, iii.1981 (Storey) (BMNH). 126 I. D. GAULD & G. A. HOLLO WAY Certonotusfarrugiaisp. n. (Fig. 58) Female. Medium-sized species, fore wing length 6-10 mm. Labium with glossae unspecialized. Lower faceat narrowest point 1-1-1-2 times as broad as high; malar space 1-2-1-3 times as long as basal mandibularwidth. Occipital carina ventrally as long as abscissa of hypostomal carina between it and base of mandible.Upper part of pronotum, slightly before posterior corner, almost flat, in dorsal view barely projectingbeyond mesoscutal margin; subalar prominence quite strongly convex and medially raised in dorsal aspect.Scutellum smooth, virtually impunctate with weak transverse carina apically; metapleuron anteriorlysmooth, posteriorly with few coarse punctures; submetapleural carina anteriorly expanded into a broadrounded lobe; metanotum with a small lateral tooth before vestige of lateral carina. Propodeummoderately short with anterior transverse carina complete; posterior transverse carina absent; laterome-dian longitudinal carina absent; pleural carina present, posteriorly obsolescent; area superomediaundefined; area spiracularis defined except medially; first and second lateral areae confluent. Fore wingwith 3r-m converging towards 2r-m, joining latter at Rs or sometimes slightly separated; 2m-cu joining Mslightly basad of 3r-m. Hind wing with distal abscissa of Cu\ absent. Hind tibia with posterior margin withone spine-like bristle. Caster with segment 1 short and broad, the sternite 0-4-0-6 times as long as hindcoxa, reaching to level of spiracle. Tergite 7 mediodorsally with wide indentation; tergite 8 posteriorlyshort and broadly rounded; tergite 9 in dorsal view short and broad. Ovipositor projecting beyond apex ofgaster by 3-0-3-5 times length of hind tibia. Coloration. Head and anterior part of alitrunk reddish brown; antenna, much of metapleuron andpropodeum black; gaster black, tergite 2 entirely and posterior margins of tergites 3+ white; anteriorlyred-brown, partially infuscate; hind leg mainly black. Pterostigma black, wings strongly infumate. Male. Similar to female but with fore wing length 6 mm; malar space 0-8 times basal mandibular width;3r-m present; gaster with segment 1 stout, the sternite 0-6 times as long as hind coxa. Apex of gasterrounded, bearing long scattered hairs. Similar in colour to female but with gaster entirely red; mid legbrownish. REMARKS. C. farrugiai is probably the most distinct species in the genus on account of its striking colourpattern which resembles that of a number of other unrelated species of Hymenoptera occurring insouth-eastern Australia. Structurally it is also distinctive in having the posterior margin of tergite 7 onlyweakly indented.This species occurs from south-eastern Queensland to Victoria. MATERIAL EXAMINED Holotype $, Victoria: Toolangi, xii.1982 (Farrugia) (AM). Paratypes. Queensland: 4 $ , Mt Glorious, xi-iii (TC); 1 $ , Mt Nebo, 500 m, iii (TC); 1 $ , Tambourine,x.1977 (Galloway) (BMNH); 1 tf , Mt Tambourine, iv. 1935 (Turner) (BMNH); 1 $ , 2 cT, Mt Tambourine,x-xii (TC). Victoria: 2 $, same data as holotype (BMNH). Certonotus hinnuleus Krieger comb. rev. Certonotus hinnuleus Krieger, 1901: 123; Turner, 1919: 551. Holotype $, NEW SOUTH WALES (MNHU) [not examined] . Certonotus n. sp.; Fullaway, 1942: 244.Asperellus hinnuleus (Krieger) Townes et al. , 1961: 114. Female. Fairly small species, fore wing length 4-7 mm. Labium with glossae unspecialized. Lower face atnarrowest point 1-2 times as broad as high; malar space 1-2-1-4 times as long as basal mandibular width.Occipital carina ventrally almost twice as long as abscissa of hypostomal carina between it and mandiblebase. Upper part of pronotum, slightly before posterior corner, very convex, in dorsal view appearing as aconical projection; subalar prominence moderately convex. Scutellum sparsely punctate with strongtransverse keel; metapleuron smooth and polished; submetapleural carina anteriorly expanded into abroad triangular lobe; metanotum without a tooth. Propodeum short with anterior transverse carinacomplete, close to anterior margin; posterior transverse carina absent; lateromedian longitudinal carinapresent before anterior transverse carina; pleural carina present anteriorly, posteriorly absent; areasuperomedia not delineated; area spiracularis not delineated internally; first and second lateral areaeindistinctly delineated, confluent. Fore wing with 3r-m converging towards 2r-m, sometimes forming apetiolate areolet or joining at one point on Rs; 2m-cu joining M at 3r-m or slightly basad. Hind wing withdistal abscissa of Cui absent. Hind tibia with posterior margin with one spine-like bristle. Gaster withsegment 1 short and broad, the sternite 0-6 times as long as hind coxa, reaching to level of spiracle. Tergite 7 AUSTRALIAN LABENINI & POECILOCRYPTINI 127 mediodorsally with large indentation; tergite 8 posteriorly short, broad, truncate; tergite 9 in dorsal viewshort, rounded posteriorly. Ovipositor projecting beyond apex of gaster by 5-0-5-5 times length of hindtibia. Coloration. Antenna black, distal flagellar segments white; head white, vertex, interocellar area andfrons centrally whitish; mesoscutum brownish or blackish with a yellow central quadrate mark; most ofpronotum, tegula, subalar prominence, metapleuron and propodeum almost entirely yellowish; meta-pleuron brownish. Gaster dark brown with lateral margins and spots near posterolateral corners yellow;legs brownish; tibia proximally and femur distally somewhat paler; fore coxa whitish. Pterostigmablackish, wings very weakly infumate. Male. Similar to female but with fore wing length 3-6 mm; malar space 0-9-1-0 times basal mandibularwidth; 3r-m present; gaster with segment 1 quite stout, the sternite 0-5 times as long as hind coxa; apex ofgonosquama quite weakly sclerotized, slightly flattened, bearing long fine hairs. Colour similar to femalebut mid coxa yellow. REMARKS. Certonotus hinnuleus belongs to the leeuwinensis-group. It is the most distinctive taxon in thegroup on account of the swollen pronotal corner. Structurally it is most closely related to C. zebrus fromwhich it differs in having a longer ovipositor and shorter stouter petiole. In Australia it has only beenrecorded from New South Wales and Victoria. The female from Western Australia referred to by Morley(1913) is a distinct species, C. ixion. Townes et al. (1961) note that this species occurs in New Caledonia. MATERIAL EXAMINED New Caledonia: 2 $ , 7 km SE. La Foa, i. 1945 (Remington) (TC) (compared with type); 1 d", hills behindNoumea, x.1940 (Williams) (TC). Queensland: 1 C?, N. slope, Bluff Range, Biggenden, viii.1976 (Frauca)(ANIC); 1 C?, Broken R., near Eungella, xii.1961 (McAlpine & Lossiri) (AM); 1 $, 1 cf, Kuranda,v-vi.1913 (Turner) (BMNH); 1 C?, Mackay, 1909 (Turner) (BMNH); 3 C?, Mt Cootha, iv-v (TC); 1 C?, MtGlorious, i (TC); 2 d" , Mt Tambourine, xi.1977 (Galloway) (BMNH). Certonotus humeralifer Krieger(Figs 24, 26, 27, 30) Certonotus humeralifer Krieger, 1901: 121. Lectotype $, NEW SOUTH WALES (MNHU), designated byTownes et al., 1961: 113 [examined]. Female. Medium to large species, fore wing length 7-14 mm. Labium with glossae unspecialized. Lowerface at narrowest point 1-2-1-3 times as broad as high; malar space 0-7-0-8 times as long as basalmandibular width. Occipital carina ventrally sinuous, at least twice as long as abscissa of hypostomal carinabetween it and base of mandible. Upper part of pronotum, slightly before posterior corner, very stronglyswollen, in dorsal view pyramidal, projecting; subalar prominence strongly convex. Scutellum punctate,crest distinct, metapleuron smooth and polished; submetapleural carina anteriorly expanded into a toothbefore lateral carina. Propodeum moderately short with anterior transverse carina weak centrally butusually complete; posterior transverse carina absent; lateromedian longitudinal carina present anteriorly;pleural carina strong anteriorly, posteriorly vestigial; area superomedia not delineated; area spiracularisnot clearly delineated posteriorly; first and second lateral area confluent, weakly defined laterally. Forewing with 3r-m fused anteriorly with 2r-m, so areolet petiolate; 2m-cu joining M-basad by 0-2 from 3r-m.Hind wing with distal abscissa of Cui distinct to wing margin. Hind tibia with posterior margin with onespine-like bristle. Gaster with segment 1 short, the sternite 0-5 times as long as hind coxa, reaching to levelof spiracle. Tergite 7 mediodorsally with wide indentation narrowing to slit to 0-5 of tergite length; tergite 9in dorsal view short, rectangular. Ovipositor projecting beyond apex of gaster by 5 times length of hindtibia. Coloration. Bright yellow species; flagellum, interocellar area, mesoscutum in central spot andscutoscutellar groove black; hind tarsus infuscate. Pterostigma black, wings hyaline. Male. Similar to female but with fore wing length 7-8 mm; malar space 0-6 times basal mandibular width;3r-m present; gaster with segment 1 quite stout, the sternite 0-5 times as long as hind coxa; apex ofgonosquama strongly flattened, with periphery bearing only fine sparse hairs. Similar in colour to female. REMARKS. The bright yellow ground colour of this species and the conical pronotal process distinguish C.humeralifer from other Australian Certonotus species. C. humeralifer belongs to the humeralifer-group; itappears to be most closely related to C. apicalis. Both have a characteristically specialized fore femur.C. humeralifer has been collected in Queensland and New South Wales. 128 I. D. GAULD & G. A. HOLLO WAY MATERIAL EXAMINED Lectotype $, New South Wales (MNHU). Queensland: 1 cf , Brisbane, 1956 (Ken) (NMV); 2 $, Fraser Island, ix.1930 (BMNH); 1 $, 1 cf , MtGlorious, 650 m, x-xiii (TC); 1 $, Mt Nebo, 500 m, iii (TC); 5 $, 1 cf, Mt Tambourine, x-xi.1977(Galloway) (BMNH); 1 $, Mt Tambourine, xii.1911 (Hacker) (BMNH); 2 $, Mt Tambourine, xi-xii(TC); 1 cf , Toowoomba, iii (TC). New South Wales: 1 $, Terrigal, 1900 (Froggatt) (BMNH); 1 $, 'NewSouth Wales' (NMV). Certonotus ixion sp. n. (Figs 55, 61) Female. Medium-sized species, fore wing length 8 mm. Labium with glossae unspecialized. Lower face atnarrowest point 1-2 times as broad as high; malar space 0-9 times as long as basal mandibular width.Occipital carina ventrally slightly longer than abscissa of hypostomal carina between it and base ofmandible. Upper part of pronotum, slightly before posterior corner, weakly convex, in dorsal view parallelwith mesoscutal margin; subalar prominence moderately convex. Scutellum closely, finely punctate withstrong transverse crest; metapleuron with close shallow punctures; submetapleural carina anteriorlyabruptly expanded into a broad lobe. Propodeum moderately short with anterior transverse carinacomplete; posterior transverse carina almost complete; lateromedian longitudinal carina absent except forvestiges before anterior carina; pleural carina obsolescent posteriorly; area superomedia not delimitedlaterally; area spiracularis almost complete; first and second lateral areae barely delineated laterally,confluent. Fore wing with 3r-m strongly converging towards 2r-m, joining Rs separately; 2m-cu joining Mat 3r-m. Hind wing with distal abscissa of Cu\ present over distal 0-5 to wing margin, not joined to firstabscissa of Cu\. Hind tibia with posterior margin with a small spine-like bristle. Gaster with segment 1 longand narrow, the sternite 0-8 times as long as hind coxa, reaching behind level of spiracle. Tergite 7mediodorsally with narrow indentation to 0-5 of length; tergite 8 posteriorly short and truncate; tergite 9 indorsal view broad. Ovipositor projecting beyond apex of gaster by 5-5 times length of hind tibia. Coloration. Antenna blackish, apex (except extreme distal part of last segment) whitish; head reddishbrown, lower face, frontal orbits and most of genae whitish; alitrunk reddish, peripherally darker;scutellum posteriorly, postscutellum, tegula, subalar prominence and upper margin of pronotum yellow-ish. Gaster dark red-brown; tergites 1-6 with posterolateral triangular yellow marks, tergites 2-3 withanterolateral pale spots. Legs predominantly reddish brown, anterior two pairs of coxae and distal apicesof fore and mid femora whitish. Pterostigma red-brown, wings hyaline. Male. Quite similar to female but with fore wing length 5-7 mm; malar space 0-6 times basal mandibularwidth; 3r-m present; gaster with segment 1 slender, the sternite 1-0 times as long as hind coxa; apex ofgaster with a small lobe bearing long scattered hairs. Colour similar to female but with mesopleuronwhite-marked and pale maculae on gaster smaller, rather inconspicuous, flagellum entirely black. REMARKS. The species is one of the three Certonotus that have incomplete distal abscissa of Cu\, It is mostsimilar to C. paluma from which it differs strikingly in colour pattern, length of ovipositor and shape oftergite 8. The coxae are more elongate than in many other species and the sculpture of the alitrunk iscoarser than that of C. paluma. C. ixion has been collected in Queensland and Victoria. MATERIAL EXAMINED Holotype , Victoria: Ferntree Gully, x.1921 (Burns) (NMV). Paratypes. Queensland: 1 cf , Lamington N.P. , xi.1961 (Common & Upton) (ANIC); 1 cf , Mt Glorious,xii (TC); 2 cf , Mt Nebo, 500 m, iii (TC). Certonotus leeuwinensis Turner comb. rev.(Figs 49, 52) Certonotus leeuwinensis Turner, 1919: 551. Holotype cf , WESTERN AUSTRALIA (BMNH) [examined].Asperellus leeuwinensis (Turner) Townes et al. , 1961: 115. Female. Medium-sized species, fore wing length 6-7 mm. Labium with glossae unspecialized. Lower faceat narrowest point 1-2-1-3 times as broad as high; malar space 1-1-1-2 times as long as basal mandibularwidth. Occipital carina ventrally weak, only slightly longer than abscissa of hypostomal carina between itand mandibular base. Upper part of pronotum, slightly before posterior corner, weakly convex, in dorsalview slightly protruding from mesoscutal margin; subalar prominence weakly convex. Scutellum sparsely AUSTRALIAN LABENINI & POECILOCRYPTINI 129 punctate, with strong transverse keel; metapleuron closely, quite finely punctate; submetapleural carinaanteriorly expanded into a moderately broad, triangular lobe; metanotum with weak lateral tooth.Propodeum quite short with anterior transverse carina complete; posterior transverse carina absent;lateromedian longitudinal carina present before anterior transverse carina; pleural carina more or lesscomplete; area superomedia undefined; area spiracularis incomplete internally; first and second lateralareae confluent. Fore wing with 3r-m convergent towards 2r-m in some specimens; 2m-cu joining M justbasad of 2r-m. Hind wing with distal abscissa of Cui absent. Hind tibia with posterior margin with onespine-like bristle. Caster with segment 1 short, the sternite 0-5 times as long as hind coxa, reaching nearlyto level of spiracle. Tergite 7 mediodorsally with indentation to 0-5 its length; tergite 8 posteriorly short androunded; tergite 9 in dorsal view short and broad. Ovipositor projecting beyond apex of gaster by 5-5 timeslength of hind tibia. Coloration. Antenna black with apical segments white (except for distal apex of last segment which isblack); head black, orbits entirely whitish, lower face centrally brownish; alitrunk reddish brown, thesclerites often margined with black irregularly; pronotum dorsally as a stripe, ventrally, tegula, subalarprominence, scutellum and postscutellum pale whitish yellow; gaster brownish or even blackish, tergites1-7 laterally and posteriorly margined with yellow. Legs with fore and mid coxae yellow above, blackbelow; hind coxa brown above, blackish below; anterior two pairs of legs otherwise brownish except fordistal apices of femora which are yellow; hind femur brownish; hind tibia darker brown, proximallyindistinctly paler, tarsus infuscate. Pterostigma dark brown, wings hyaline. Male. Similar to female but with fore wing length 3-6 mm; malar space 0-9 times basal mandibular width;3r-m present; gaster with segment 1 moderately stout, the sternite 0-6 times as long as hind coxa; apex ofgonosquama flattened slightly and bearing long scattered hairs. Colour similar to female but lower face andanterior two pairs of coxae almost entirely yellow. REMARKS. Certonotus leeuwinensis belongs to the leeuwinensis-group. It is structurally most similar to C.toolangi in having a noticeably punctate metapleuron. The most obvious difference between the twospecies is in coloration. The gaster of leeuwinensis has the hind margins of tergites 2-5 banded with yellowwhereas those of toolangi are spotted with yellow. The hind tibia of toolangi is bicoloured, that ofleeuwinensis is almost unicolorous. C. leeuwinensis has a slightly longer ovipositor and malar space thantoolangi. MATERIAL EXAMINED Holotype cf , Western Australia: Yallingup near Cape Naturaliste ix-x.1913 (Turner) (BMNH). New South Wales: 1 $, Killara, xii.1935 (Day) (ANIC). Tasmania: 1 $, Coles Bay, ii-iii (TC); 1 cf,Georgetown, xi.1917 (Cole) (NMV); 1 $, Mt Barrow, 1200 m, xiii-i (TC). Western Australia: 1 $(paratype), same data as holotype (BMNH); 1 cf , 21 km SW. by S. Donnybrook (23-44S 115-41E), x.1981(Naumann & Cardale) (ANIC); 1 cf , Mt Chudalup, S. of Northcliffe, x.1970 (Colless) (ANIC); 1 cf ,Yallingup, near Cape Naturaliste, ix-x.1913 (Turner) (BMNH). Certonotus mogimbensis Cheesman comb. rev.(Fig. 47) Certonotus mogimbensis Cheesman, 1936: 180. Holotype $, NEW HEBRIDES (BMNH) [examined].Asperellus mogimbensis (Cheesman) Townesefa/., 1961: 115. Female. Small species, fore wing length 5 mm. Labium with glossae slightly elongate. Lower face atnarrowest point 1-0 times as broad as high; malar space 0-8-1-0 times as long as basal mandibular width.Occipital carina ventrally from slightly shorter to slightly longer than abscissa of hypostomal carinabetween it and base of mandible. Upper part of pronotum, slightly before posterior corner, weakly convex,in dorsal view parallel to mesoscutal margin; subalar prominence very weakly convex. Scutellum punctatewith strong transverse keel; metapleuron smooth with some fine punctures posteriorly; submetapleuralcarina anteriorly expanded into a rounded lobe; metanotum without a lateral tooth. Propodeum short withanterior transverse carina complete; posterior transverse carina absent; lateromedian longitudinal carinavestigial; pleural carina complete; area superomedia undefined; area spiracularis not defined internally;first and second lateral areae confluent. Fore wing with 3r-m present for half the distance or less from Rstowards M; 2m-cu joining M well distad of 2r-m. Hind wing with distal abscissa of Cui absent. Hind tibiawith posterior margin with one or two spine-like bristles. Gaster with segment 1 short and broad, thesternite 0-5 times as long as hind coxa, reaching to level of spiracle. Tergite 7 mediodorsally withindentation to 0-5 of length; tergite 8 posteriorly long and rounded; tergite 9 in dorsal view elongate.Ovipositor projecting beyond apex of gaster by 4-0-4-5 times length of hind tibia. 130 I. D. GAULD & G. A. HOLLOWAY Coloration. Antenna black with a subapical broad white band; head whitish; clypeus and mouth partsbrownish, frons centrally and interocellar area dark brown. Alitrunk, legs and gaster uniformly orange.Pterostigma dark brown, wings hyaline. Male. Similar to female but with fore wing length 3-4 mm; malar space 0-9-1-1 times basal mandibularwidth; 3r-m absent; gaster with segment 1 moderately slender; the sternite 0-6 times as long as hind coxa;apex of gonosquama with small lobe bearing fine scattered hairs. Similarly coloured to female but withhead uniformly orange, flagellum entirely black, hind tibia and tarsus weakly to strongly infuscate, sometergites of gaster infuscate and wings slightly infumate. REMARKS. C. mogimbensis most closely resembles C. pineus in colour and structure and the two may beclosely related. However, mogimbensis is distinctive, not only in having a flatter pronotum but in having3r-m incomplete in the male and lacking dark maculae on the mesoscutum. This species is known to occurin tropical Queensland and on Vanuatu (New Hebrides). MATERIAL EXAMINED Holotype $, New Hebrides: Malekula, Ounua, ii.1929 (Cheesman) (BMNH). Queensland: 1 $, 1 cf, Claudie R, near Mt Lamond, xii.1971 (McAlpine & Holloway) (AM); 1 cf,Shipton's Flat (15-47S 124-14E), x.1980 (Cardale) (ANIC). Certonotus monticola Morley (Fig. 38)Certonotus monticola Morley, 1913: 29. Holotype $, QUEENSLAND (BMNH) [examined]. Female. Medium-sized species, fore wing length 9-12 mm. Labium with glossae elongate, projectingbeyond clypeus by a distance equal to or greater than facial height. Lower face at narrowest point 0-9-1-0times as broad as high; malar space 0-5-0-6 times as long as basal mandibular width. Occipital carinaventrally obsolescent, represented by a stub that is shorter than abscissa of hypostomal carina between itand base of mandible. Upper part of pronotum, slightly before posterior corner, weakly convex, in dorsalview barely projecting beyond scutal margin; subalar prominence very weakly convex. Scutellum closelyand coarsely punctate, with transverse crest; metapleuron anteriorly punctate, posteriorly and dorsally;submetapleural carina anteriorly abruptly expanded into a broad rounded lobe; metanotum without atooth before lateral carina. Propodeum quite short with anterior transverse carina centrally incomplete;posterior transverse carina absent; lateromedian longitudinal carina present only before anterior carina;pleural carina complete; area superomedia not delineated; area spiracularis complete, short; first andsecond lateral areae confluent. Fore wing with 3r-m converging on 2r-m but joining Rs separately; 2m-cujoining M midway between 3r-m and 2r-m. Hind wing with distal abscissa of Cu^ distinct to wing margin.Hind tibia with posterior margin without spine-like bristles. Gaster with segment 1 short and broad, thesternite 0-7 times as long as hind coxa, reaching just beyond level of spiracle. Tergite 7 mediodorsally withsmall indentation; tergite 8 posteriorly elongate and narrow; tergite 9 in dorsal view long and truncate.Ovipositor projecting beyond apex of gaster by 4-5 times length of hind tibia. Coloration. A reddish brown species with lower face, frontal and genal orbits, central mesoscutal spot,scutellum, most of pronotum, much of mesopleuron and virtually all of metapleuron/propodeum behindspiracle, yellow. Gaster dark brown, tergites 2+ with anterolateral triangular marks. Fore leg brownishyellow, coxa infuscate; mid leg blackish brown, femur distally yellow marked, coxa darker; hind leg withcoxa black with dorsal yellow marks, femur and tarsus strongly infuscate, tibia less strongly infuscate.Pterostigma black, wings weakly infumate. Male. Slender gaster, otherwise similar to female but with fore wing length 7-11 mm; malar space0-5-0-6 times basal mandibular width; 3r-m present; gaster exceptionally elongate with segment 1 quitestout, the sternite 0-6 times as long as hind coxa; gonosquama long, apex a little flattened, with scatteredhairs of various length. Male similar in general colour pattern to female, although with fewer yellowmaculae. REMARKS. C. monticola belongs to the /7av/ce/w-group. It is the only taxon in this complex to occur inAustralia where it is easily recognizable by the elongate glossae and short occipital carina. C. monticola isknown to occur in both north Queensland and Papua New Guinea. MATERIAL EXAMINED Holotype 9> Queensland: Tambourine Mt (BMNH). Queensland: 1 $, Middle Claudie R., Iron Range, x.1974 (Daniels) (AM). Papua New Guinea: 1 $,Amuk, 165 m, i.1960 (Mao) (BPBM); 1 $, 2 cf , Bulolo, ix.1981, exMyristica sp. (Roberts) (BMNH); 7 $, AUSTRALIAN LABENINI & POECILOCRYPTINI 131 3 Cf , Kiunga, Fly River, viii-x. 1957 (Brandt) (BPBM) ; 2 , Normanby Is. , Wakaiuma, Sewa Bay, xii. 1956(Brandt) (BPBM); 1 $, 10 mi. W. Vudal, xi.1970 (TC); 1 $, Waris, S. of Hollandia, 450-500 m, viii.1959(Maa) (BPBM). Certonotus nitidulus Morley(Fig. 33) Certonotus nitidulus Morley, 1913: 29. Holotype cf , VICTORIA (BMNH) [examined].Certonotus tasmaniensis Turner, 1919: 550; Hocking, 1967: 57; Short, 1978: 41. Holotype cf , TASMANIA(BMNH) [examined]. Syn. n. Female. Medium to large species, fore wing length 7-14 mm. Labium with glossae unspecialized. Lowerface at narrowest point 1-0-1-1 times as broad as high; malar space 0-6-0-7 times as long as basalmandibular width. Occipital carina ventrally strong, slightly longer than abscissa of hypostomal carinabetween it and base of mandible . Upper part of pronotum , slightly before posterior corner , weakly convex ,in dorsal view barely projecting beyond scutal margin; subalar prominence weakly convex. Scutellumpunctate, transverse keel strong; metapleuron smooth with isolated punctures; submetapleural carinaanteriorly expanded into a broad rounded lobe; metanotum with a weak tooth opposite anterior end oflateral carina. Propodeum quite short with anterior transverse carina centrally incomplete; posteriortransverse carina absent; lateromedian longitudinal carina present only before anterior carina; pleuralcarina present, complete but rather weak; area superomedia undefined; area spiracularis complete,defined posteriorly by an arched carina that is very close to spiracle; first and second lateral areaeseparated, distally delineated. Fore wing with 3r-m converging towards 2r-m, joining Rs separately; 2m-cujoining M to 0-3 basad of 3r-m. Hind wing with distal abscissa of Cu\ distinct to wing margin. Hind tibia withposterior margin without spine-like bristles. Gaster with segment 1 short, the sternite 0-6 times as long ashind coxa, reaching behind level of spiracle. Tergite 7 mediodorsally with indentation almost to 0-5 lengthof tergite; tergite 8 posteriorly short and rounded; tergite 9 in dorsal view transverse. Ovipositor projectingbeyond apex of gaster by 5-0 times length of hind tibia. Coloration. Colour somewhat variable but generally with antenna black, distally, except for apex of lastsegment, white. Head red-brown, lower face laterally to entirely, genal and frontal orbits pale yellow.Alitrunk reddish brown, yellow marks on pronotum medially, anterior 'corners' and central spot onmesoscutum, scutellum laterally, tegula, subalar prominence, anterior and posterior mesopleural spots,postscutellum and hind part of propodeum. Gaster reddish with tergites posteriorly and laterally yellowishmargined, and also with spots in anterolateral corners. Fore leg yellow, femur red centrally, peripherallyblack with distal apex yellow. Mid leg similar with coxa black-marked and tibia and tarsus infuscate. Hindleg as mid leg but with coxa reddish brown, ventrally black. Pterostigma black, wings hyaline. Male. Similar to female but with fore wing length 7-10 mm; malar space 0-4-0-6 times basal mandibularwidth; 3r-m present; gaster with segment 1 moderately slender, 0-6 times as long as hind coxa; apex ofgonosquama flattened, bearing two widely separated dense tufts of long hairs. Similarly coloured tofemale. REMARKS. C. nitidulus is apparently one of the commonest species in the genus as it has adapted toparasitizing an introduced siricid in Pinus radiata plantations in Victoria and Tasmania (Hocking, 1967). Incoloration C. nitidulus most strongly resembles C. geniculatus (from which it can be separated by referenceto couplet 8 in the key) but structurally it is most similar to C. annulatus and C. andrewi. C. nitidulus differsfrom both in colour pattern and in the form of the propodeal carinae. This is a quite widespread species inthe south-east, extending from southern Queensland to Tasmania. HOST RECORDS. Siricidae: SirexnoctilioF. (Hocking, 1967). MATERIAL EXAMINED Holotype cf (Gertonotus nitidulus Morley), Victoria (BMNH). Holotype cf (Certonotus tasmaniensisTurner), Tasmania: Mt Wellington, i-ii.1913 (Turner) (BMNH). Queensland: 1 $ , Mt Glorious, xi (TC); 1 cf , Mt Norman area, Wallangarra, x.1972 (Monteith) (ANIC).New South Wales: 1 $ , Barrington Nat. Park, i (TC); 1 $ , Boyd River crossing, Kanangra-Boyd Nat. Park,xii. 1977 (Daniels) (AM); 1 cf , Leather Barrel Ck, Kosciusko, xi.1961 (Colless) (ANIC); 1 $, Lord HoweIs., ii-iii.1957 (Leipa) (BPBM); 1 cf , Monga, x.1957 (Riek) (ANIC); 1 $, Moonee, xi.1947 (NMV); 1 cf ,Moss Vale, xi.1919 (Duquef) (AM); 1 $, 2 cf, Mt Tomah, xi.1983 (Rodd) (AM); 1 cf, Pebbly Beach,xi.1960 (Common & Upton) (ANIC); 1 cf , Tubrabucca, xi.1953 (Burns) (NMV). Victoria: 1 $, DynamiteCk, Bonang Hwy, x.1961 (Colless) (ANIC); 1 $, Harrietville, i.1924 (Oke) (NMV); 3 cf, Healesville, 132 I. D. GAULD & G. A. HOLLOWAY xi.1943 (Oke) (NMV); 2 $ , 1 cf , South Melbourne, breeding cages, x.1969, ex Sirex sp. (Waugh) (NMV);3 $ , 3 Cf , Mirboo North, x.1967, ex Sirex noctilio (Elliott) (ANIC); 3 $ , Mt Dandenong, 300 m, ii (TC); 3Cf , Mt Buffalo, 1600 m, ii (TC); 1 $, 1 cf , Noorimbee, xi.1965 (Neboiss) (NMV); 5 cf , Toolangi, xii.1982(Farrugia) (BMNH) ; 5 $ , Toolangi , i-ii . 1983 (Farrugia & Gauld) (BMNH) ; 2 $ , Warburton , ii-iii (TC) ; 8$, 2 cf , Vic. Dept Agric., ii.1968, ex Finns logs (Irvine) (ANIC). Tasmania: 1 $, Harrison Ck, betweenCracroft and Blakes Opening, ii.1966 (Neboiss) (MNV); 1 $, E. Blakes Opening, ii.1966 (Neboiss)(NMV); 1 $, 1 Cf , Bruny Is., vii.1964, from Pinus radiata (Wilson & Hocking) (AM); 1 $, Catamaran, ii(TC); 1 $, Collinsville, 300 m, ii.1983 (Gauld) (BMNH); 2 $, Frenchman's Gap Trig at Franklin River,ii-iii (TC); 1 $, Geeveston, ii (TC); 1 $, 1 cf, Hartz Mtns, ii-iii (TC); 1 $, Hellyer Gorge, ii.1983(Naumann & Cardale) (ANIC); 1 $ , Hellyer Gorge, xii.1981 (Naumann & Cardale) (ANIC); 2 $ , HellyerGorge, ii.1967 (Riek) (ANIC); 1 (7, Hellyer Gorge, 300 m, i-ii (TC); 1 $ , Huon-Picton Junction, ii.1967(Riek) (ANIC); 1 cf , King William Ck, xii.1981 (Gauld) (BMNH); 2 $ , 1 cf , King William Range i (TC);28 $ , 18 cf , Launawanna, Bruny Is. , ix-x.1964 (Hocking) (ANIC, BMNH, TC); 1 $ , 1 cf , Mt Barrow, 300m, xii-i(TC). Certonotus paluma sp. n. (Figs 54, 62) Female. Medium-sized species, fore wing length 8-10 mm. Labium with glossae unspecialized. Lower faceat narrowest point 1-1-1-2 times as broad as high; malar space 0-8 times as long as basal mandibular width.Occipital carina ventrally as long as abscissa of hypostomal carina between it and base of mandible. Upperpart of pronotum, slightly before posterior corner, very weakly convex, in dorsal view almost parallel withmesoscutal margin; subalar prominence weakly convex. Scutellum sparsely punctate, with a posteriortransverse crest; metapleuron polished, smooth, virtually impunctate; submetapleural carina anteriorlyexpanded into a broad rounded lobe; metanotum barely produced opposite anterior margin of lateralcarina. Propodeum quite short with anterior transverse carina complete; posterior transverse carinapresent at least centrally; lateromedian longitudinal carina present only anterior to anterior transversecarina; pleural carina complete anteriorly, posteriorly obsolescent; area superomedia not boundedlaterally; area spiracularis completely delineated; first and second lateral areae separated by carina. Forewing with 3r-m converging towards 2r-ra, joining the latter at Rs; 2m-cu joining M at 3>r-m. Hind wing withdistal abscissa of Cu\ present over distal 0-5 to wing margin, not joined to first abscissa of Cu\. Hind tibiawith posterior margin without spine-like bristles. Gaster with segment 1 long and narrow, the sternite 0-7times as long as hind coxa, reaching to level of spiracle. Tergite 7 mediodorsally with narrow indentation,over 0-5 of its length; tergite 8 posteriorly long and narrow; tergite 9 in dorsal view short and pointed.Ovipositor projecting beyond apex of gaster by 4 times length of hind tibia. Coloration. Antenna black, distally white with extreme apex of last segment black; head black with faceand orbits white; ali trunk black or reddish black, with central mesoscutal stripe, scutellum, pronotum,tegula, subalar prominence, mesopleuron centrally, metapleuron and propodeum laterally white; gasterblack with very large lateral triangular areas white. Anterior two pairs of legs white with femora, tibiae andtarsi variously infuscate ; hind leg black , coxa dorsally with white spot . Pterostigma blackish , wings hyaline . Male. Unknown. REMARKS. C. paluma is one of the three quite closely related Certonotus species that have an incompletedistal abscissa of Cu\ in the hind wing. It is distinguishable from Certonotus sp. A by its unspecializedsubalar prominence. It is structurally similar to C. ixion, from which it may be separated by reference to thekey. MATERIAL EXAMINED Holotype 9, Queensland: Paluma, 900 m (19-OOS 146-12E), x.1980, ex Malaise trap (Frith) (ANIC).Paratype. Queensland: 1 $, Windsor Tableland, iii.1981 (Storey) (BMNH). Certonotus pineussp. n. (Figs 45, 46) Female. Medium-sized species, fore wing length 7-9 mm. Labium with glossae unspecialized. Lower faceat narrowest point 1-0-1-1 times as broad as high; malar space 0-9-1-0 times as long as basal mandibularwidth. Occipital carina ventrally almost 3 times as long as abscissa of hypostomal carina between it and baseof mandible. Upper part of pronotum, slightly before posterior corner, very convex, in dorsal viewsubpyramidal; subalar prominence moderately convex. Scutellum moderately, closely punctate, trans- AUSTRALIAN LABENINI & POECILOCRYPTINI 133 verse keel weak; metapleuron anteriorly smooth, posteriorly punctate; submetapleural carina anteriorlyexpanded into a broad almost quadrate lobe; metanotum with a very weak tooth before vestige of lateralcarina. Propodeum quite short with anterior transverse carina present, very close to anterior part ofpropodeum; posterior transverse carina absent; lateromedian longitudinal carina vestigial before anteriorcarina, otherwise absent; pleural carina more or less complete; area superomedia undefined; areaspiracularis more or less complete except internally; first and second lateral areae not separated. Fore wingwith 3r-m only slightly convergent towards 2r-m, joining Rs away from 2r-m, making areolet quadrate;2m-cu joining M at 3r-m. Hind wing with distal abscissa of Cu\ absent. Hind tibia with posterior marginwith one spine-like bristle. Gaster with segment 1 short and broad, the sternite 0-3 times as long as hindcoxa, just not reaching level of spiracle. Tergite 7 mediodorsally with indentation to 0-5 its length; tergite 8posteriorly long and rounded; tergite 9 in dorsal view short and broad. Ovipositor projecting beyond apexof gaster by 4-5 times length of hind tibia. Coloration. Antenna black, with subapical white band; head yellowish, only interocellar area and edgesof mandibles blackish. Alitrunk and gaster yellowish; mesoscutum with lateral marks, anterocentrallyinfuscate; tip of tergite 8 blackish. Legs yellow, hind tarsus and sometimes distal apex of tibia blackish.Pterostigma black, wings hyaline. Male. Similar to female but with fore wing length 4 mm; malar space 0-7 times basal mandibular width;3r-m present; gaster with segment 1 quite stout, the sternite 0-3 times as long as hind coxa; apex ofgonosquama rounded, bearing scattered hairs. Similar in colour to female but with yellowish areas a littlepaler, more strongly contrasted with infuscate areas; flagellar white mark very indistinct. REMARKS. A stocky species easily recognized by the almost pyramidal pronotal convexity. The virtuallyquadrate areolet and very short first sternite are also quite characteristic of this species. C. pineus is onlyknown to occur in Queensland. MATERIAL EXAMINED Holotype $, Queensland: Moses Ck, 45 km N. by E. Mt Finnigan (15-47S 145-17E), x.1980 (Cardale)(ANIC). Paratypes. Queensland: 1 $, Lake Barrine, ii.1935 (Burns) (NMV); 1 cf , Mt Webb Nat. Park (15-04S145-07E), iv.1981 (Naumann) (ANIC); 1 cf , Palm Is. nearTownsville, x (TC). Certonotus rufescens Morley(Figs 22, 25) Certonotus rufescens Morley, 1913: 30. LECTOTYPE $, QUEENSLAND (BMNH), here designated[examined] . Female. Moderately large to large species, fore wing length 8-16 mm. Labium with glossae slightlylengthened. Lower face at narrowest point 0-9-1 -0 times as broad as high; malar space 0-9-1-0 times as longas basal mandibular width. Occipital carina ventrally rather straight, more than twice as long as abscissa ofhypostomal carina between it and mandible base. Upper part of pronotum, slightly before posteriorcorner, weakly convex, in dorsal view rounded, just protruding beyond scutal margin; subalar prominencevery strongly raised, in dorsal view with a blunt, back-curved, thorn-like protuberance. Scutellumpunctate, with a moderately strong transverse crest; metapleuron smooth, virtually impunctate; submeta-pleural carina anteriorly expanded into a broad almost quadrate lobe which usually has ridge near anteriorcorner delimiting a narrow triangular area; metanotum with small tooth opposite anterior end of lateralcarina. Propodeum moderately long with anterior transverse carina complete; posterior transverse carinapresent as lateral vestige; lateromedian longitudinal carina present before anterior carina; pleural carinavery strong anteriorly, absent behind anterior carina; area superomedia not delineated laterally; areaspiracularis complete; first and second lateral areae confluent, not delineated laterally. Fore wing with3r-m converging towards 2r-m and joining latter at Rs; 2m-cu joining M at 3r-m to 0-3 towards 2r-m. Hindwing with distal abscissa of Cui distinct to wing margin. Hind tibia with posterior margin with one to threespine-like bristles. Gaster with segment 1 long and narrow, the sternite 0-9-1-0 times as long as hind coxa,reaching behind level of spiracle. Tergite 7 mediodorsally at most only slightly indented on posteriormargin; tergite 8 posteriorly with process short and rounded; tergite 9 in dorsal view elongated and bluntlypointed posteriorly. Ovipositor projecting beyond apex of gaster by 4-5 times length of hind tibia. Coloration. Brownish orange; flagellum black with a white subapical band which is usually welldeveloped but is entirely absent in specimens from the northern part of the range. Pterostigma dark brown,wings hyaline. Male. Similar to female but with fore wing length 5-7 mm; malar space 0-8-0-9 times basal mandibular 134 I. D. GAULD & G. A. HOLLO WAY width; 3r-m present; gaster with segment 1 long, slender, the sternite 0-8-1-0 times as long as hind coxa;apex of gonosquama simply truncate, not broadened, but with long sparse hairs. Similarly coloured tofemale, flagellum without a white band. REMARKS. C. rufescens is easily distinguished from other Australian species by the thorn-like subalarprominence. The face of this species is flatter than most other Certonotus and laterally it is abruptly, almostangularly rounded before the malar space. The characteristic submetapleural carina and possession of astrong, usually Y-shaped vestigial carina above the insertion of the hind coxa are useful confirmatorycharacters. A widespread species extending from northern Queensland south to near Melbourne, Victoria. MATERIAL EXAMINED Lectotype 9, Queensland: Mackay, ix.1901 (BMNH). Queensland: 2 $ (paralectotypes), same data as lectotype (BMNH); 2 cf , Mackay, 1909 (BMNH); 1 $ ,Montville, ix.1935 (Burns) (NMV); 1 cf , Moses Ck, 4 km N. by E. Mt Finnigan (15-47S 145-17E), x.1980,at light (Cardale) (ANIC); 1 $, Mt Glorious, xii.1979 (Galloway} (BMNH); 1 $, Mt Glorious nearBrisbane, xii.1976 (Boucek) (BMNH); 3 cf, Mt Glorious, i-iii (TC); 2 $, Mt Tambourine, ix-x.1978(Galloway) (BMNH); 2 $ , 4 cf , Mt Tambourine, x-xi. 1977 (Galloway) (BMNH). New South Wales: 1 $ ,19 km S. Coff s Harbour, i.1958 (Riek) (ANIC); 1 cf , Otford, xii.1962 (Colless) (BMNH); 1 $ , SassafrassGully, Springwood, ix.1972 (McAlpine) (AM). Victoria: 2 $, Toolangi, i-ii.1983 (Farrugia & Gauld)(BMNH); 1 Cf , Yellingbo, xi.1976 (Neboiss) (NMV). Certonotus sisyphussp. n. (Figs 36, 39, 40)[Certonotus hinnuleus Krieger; Morley 1913: 32. Misidentification.] Female. Medium to large-sized species, fore wing length 8-18 mm. Labium with glossae slightly leng-thened. Lower face at narrowest point 1-1-1-3 times as broad as high; malar space 0-9-1-0 times as long asbasal mandibular width. Occipital carina ventrally sinuous, much longer than abscissa of hypostomalcarina between it and mandibular base. Upper part of pronotum, slightly before posterior corner, weaklyconvex, in dorsal view projecting slightly beyond scutal margin, subalar prominence very strongly convex,in dorsal view slightly pyramidal. Scutellum sparsely punctate, transverse crest distinct; metapleuronsmooth with scattered punctures, on larger specimens with punctures closer; submetapleural carinaanteriorly expanded into a rectangular flange; metanotum with strong tooth opposite end of lateral carina.Propodeum moderately long with anterior transverse carina complete; posterior transverse carina strong,complete, almost parallel to anterior carina; lateromedian longitudinal carina present only as vestigesbefore anterior carina; pleural carina weak, posteriorly evanescent; area superomedia not defined; areaspiracularis complete; first and second lateral areae clearly separated though latter weakly definedexternally. Fore wing with 3r-m strongly converging towards an almost vertical 2r-m, joining Rs separately;2m-cu joining M 0-2 from 3r-m towards 2r-m. Hind wing with distal abscissa of Cui distinct to wing margin.Hind tibia with posterior margin with one to three spine-like bristles. Gaster with segment 1 long andnarrow, the sternite 1-0-1-2 times as long as hind coxa, reaching behind level of spiracle. Tergite 7mediodorsally with wide indentation to 0-5 length of tergite; tergite 8 posteriorly short and rounded; tergite9 in dorsal view as long as wide with blunt point posteriorly. Ovipositor projecting beyond apex of gaster by4-5 times length of hind tibia. Coloration. Reddish brown, flagellum darker with subapical white band; clypeus, facial, frontal andgenal orbits. Subalar prominence, pair of central stripes on mesoscutum, scutellum, mesopleural spot,spots in posterolateral corners of tergites 1+ and also on lateral margins of 3+ yellow. Legs yellowishbrown, anterior two pairs of coxae and distal apex of mid femur yellow. Pterostigma dark brown, wingshyaline. Male. Similar to female but with fore wing length 8 mm; malar space 0-7 times basal mandibular width;3r-m present; gaster with segment 1 slender, the sternite 1-1 times as long as hind coxa; apex ofgonosquama with a small angulate lobe that bears scattered long hairs. Similar in colour to female but withface entirely yellow, mid tibia yellow and yellow maculae on gaster virtually absent. REMARKS. The fairly elongate petiole, rather flat face and posteriorly flanged lateral propodeal carinasuggest this species is related to C. rufescens though the pleural carina is not tuberculate as it is in rufescensand geniculatus . Superficially C. sisyphus is similar to C. avitus from which it can be distinguished, not onlyby the characters in the key, but also in having a distinctly longer prementum and having the hypostomaland occipital carinae meeting further from the base of the mandible than the basal mandibular width. AUSTRALIAN LABENINI & POECILOCRYPTINI 135 Widely distributed in the south-east of Australia; recorded from New South Wales, Victoria andTasmania. MATERIAL EXAMINED Holotype $, Victoria: Toolangi, xi-xii.1982 (Farrugia & Gauld) (AM). Paratypes. Victoria: 8 $ , 1 C?, same data as holotype (BMNH). New South Wales: 2 $ , Acacia Plateau,3000 ft, x.1961 (Common & Upton) (ANIC). Tasmania: 1 $, Duck River, 6 km SE. Roger R., xii.1974(Neboiss) (NMV); 1 $, MeridithR., 20 km from Corinna, i. 1954 (Campbell) (ANIC); 1 $, Mt Field Nat.Park, 250 m, i-ii.1983 (Gauld) (BMNH); 1 $, Picton R. bridge, i.1983 (Gauld) (BMNH); 1 ?, no data(ANIC). Certonotus talus sp. n. (Figs 34, 35) Female. Medium-sized species, fore wing length 8-12 mm. Labium with glossae unspecialized. Lower faceat narrowest point 1-1 times as broad as high; malar space 0-7-0-8 times as long as basal mandibular width.Occipital carina ventrally more than twice as long as abscissa of hypostomal carina between it and base ofmandible. Upper part of pronotum, slightly before posterior corner, very weakly convex, in dorsal viewbarely projecting beyond scutal margin; subalar prominence convex, with a long sharp back-curved spine.Scutellum punctate with strong transverse keel; metapleuron very sparsely punctate; submetapleuralcarina anteriorly expanded into a broad triangular lobe; metanotum with a large tooth before anterior endof lateral carina. Propodeum moderately long with anterior transverse carina complete; posteriortransverse carina almost complete; lateromedian longitudinal carina vestigial, discernible only beforeanterior carina; pleural carina strong anteriorly, posteriorly poorly developed or obsolescent; areasuperomedia not delineated; area spiracularis complete, posteriorly very weakly delineated; first andsecond lateral areae present, separated by strong carina. Fore wing with 3r-m converging towards 2r-m,joining latter at Rs; 2m-cu joining M at 3r-m. Hind wing with distal abscissa of Cu\ distinct to wing margin.Hind tibia with posterior margin with or without one spine-like bristle. Gaster with segment 1 short, thesternite 0-8-1-0 times as long as hind coxa, reaching behind level of spiracle. Tergite 7 mediodorsally withwide indentation almost to 0-7 length of tergite; tergite 8 posteriorly with long, narrow process; tergite 9 indorsal view long and evenly rounded . Ovipositor projecting beyond apex of gaster by 5 times length of hindtibia. Coloration. Antenna black with subapical white band; head reddish brown, orbits and clypeus entirelypale; alitrunk reddish, mesoscutum with longitudinal stripes, tegula, subalar prominence, scutellumpartly, postscutellum, anterior mesopleural spot and paired spots on hind edge of propodeum yellow;gaster reddish, lateral margins of all tergites yellow and tergites also with paired large yellow spots inposterolateral corners. Fore and mid legs yellow, femur for proximal 0-7 and tibia externally reddish, distaltarsal segment blackish; hind leg virtually entirely reddish brown, tarsus infuscate. Pterostigma brown,wings hyaline. Male. Unknown. REMARKS. The possession of a long, backwardly curved, slender spine arising from the subalar prominenceimmediately distinguishes talus from all other Australian Certonotus. It appears to belong to thehumeralifer- group but differs in having a more slender petiole with a much longer sternite. One specimenhas asymmetric hind wings; the left wing has the distal abscissa of Cu\ incomplete, not joining Cu and cu-a,whilst in the right wing the vein is complete. This incomplete condition is found in C. ixion andpaluma andwe believe these are closely related species. All have similarly modified posterior gastral tergites, andrather small, oblique areolets. The petiole is more slender in these species than in other taxa in this groupand the sternite is longer. C. talus occurs in New South Wales and Victoria and has been collected in subtropical and temperate wetforest. MATERIAL EXAMINED Holotype $, New South Wales: Dorrigo Nat. Park, E. end of Blackbutt Track, 710 m, ii-iii.1980, insubtropical rainforest (Newtown & Thayer) (ANIC). Paratypes. Victoria: 2 <J>, Toolangi, xi.1982 (Farrugia) (BMNH). New South Wales: 1 <j>, Tubrabucca,i. 1948 (Bums) (NMV). 136 I. D. GAULD & G. A. HOLLOW AY Certonotus toolangi sp. n. (Fig. 60) Female. Medium-sized species, fore wing length 6-7 mm. Labium with glossae unspecialized. Lower faceat narrowest point 1-1 times as broad as high; malar space 0-9-1-1 times as long as basal mandibular width.Occipital carina ventrally as long as basal mandibular width. Occipital carina ventrally as long as or slightlylonger than abscissa of hypostomal carina between it and base of mandible. Upper part of pronotum,slightly before posterior corner, almost flat, in dorsal view parallel to mesoscutal margin; subalarprominence moderately convex. Scutellum sparsely punctate, transverse keel weak; metapleuron withshallow coarse close punctures; submetapleural carina anteriorly expanded into a broad triangular lobe;metanotum with a weak tooth laterally. Propodeum quite short with anterior transverse carina complete;posterior transverse carina absent; lateromedian longitudinal carina vestigial, only present before anteriortransverse carina; pleural carina present anteriorly; area superomedia undefined; area spiracularisincompletely defined internally; first and second lateral areae confluent, undefined laterally. Fore wingwith 3r-m convergent towards 2r-m, joining Rs separately; 2m-cu joining M very near 3r-m. Hind wing withdistal abscissa of Cui absent. Hind tibia with posterior margin with one spine-like bristle. Gaster withsegment 1 short and broad, the sternite 0-6 times as long as hind coxa, just reaching level of spiracle.Tergite 7 mediodorsally with indentation to 0-5 of length; tergite 8 posteriorly short and rounded; tergite 9in dorsal view short and broad. Ovipositor projecting beyond apex of gaster by 4-0-4-5 times length of hindtibia. Coloration. Very similarly coloured to leeuwinensis but differing in having the anterior two pairs ofcoxae entirely pale and the tibiae with at least proximal 0-5 whitish and mesoscutum centrally with a pair ofshort yellowish stripes. Male. Unknown. REMARKS. C. toolangi belongs to the leeuwinensis-group. It is most readily distinguishable by its colourpattern. Structurally it is rather similar to C. leeuwinensis from which it may be distinguished by its shortermalar space and ovipositor. It is only known from the south-east of Australia, where it has been taken inwet sclerophyll forest. MATERIAL EXAMINED Holotype $ , New South Wales: New England Nat. Park, x.1962 (Colless) (ANIC). Paratypes. Victoria: 2 $, Toolangi, xi.1982 (Farrugia) (BMNH); 1 $, Toolangi, xii.1982 (Farmgia &Gauld) (BMNH). Certonotus zebrus sp. n. (Figs 51, 53, 59) Female. Medium-sized species, fore wing length 7-8 mm. Labium with glossae unspecialized. Lower faceat narrowest point 0-8-0-9 times as broad as high; malar space 0-8-0-9 times as long as basal mandibularwidth. Occipital carina ventrally weak, nearly twice as long as abscissa of hypostomal carina between it andmandibular base. Upper part of pronotum, slightly before posterior corner, very weakly convex, in dorsalview parallel with mesoscutal margin; subalar prominence strongly convex, centrally somewhat produced.Scutellum sparsely punctate, transverse crest quite weak; metapleuron anteriorly smooth, posteriorly withscattered punctures; submetapleural carina anteriorly expanded into a broad rounded lobe; metanotumwith a weak tooth opposite anterior end of lateral carina. Propodeum short with anterior transverse carinacomplete; posterior transverse carina absent; lateromedian longitudinal carina discernible as traces beforeanterior transverse carina; pleural carina incomplete posteriorly; area superomedia not delineated; areaspiracularis virtually complete; first and second lateral areae confluent, not defined laterally. Fore wingwith 3r-/n converging towards 2r-ra, well separated at Rs; 2m-cu joining M 0-3 from 3r-m towards 2r-m.Hind wing with distal abscissa of Cui absent. Hind tibia with posterior margin with one spine-like bristle.Gaster with segment 1 long, the sternite 0-6 times as long as hind coxa, just reaching level of spiracle.Tergite 7 mediodorsally with indentation to 0-5 its length; tergite 8 posteriorly short and roundedposteriorly. Ovipositor projecting beyond apex of gaster by 4 times length of hind tibia. Coloration. A black species that sometimes has pale yellow-white marks on distal flagellar segments(except extreme apex of distal one); lower face entirely, orbits, most of pronotum, tegula, scutellum,postscutellum, subalar prominence, mesopleural stripe, most of mesopleuron and propodeum laterallyand posteriorly, base of petiole, posterior and lateral margins of gastral tergites whitish. Anterior two pairsof legs whitish, femora and tibiae centrally, tarsi partly brownish marked. Hind leg black. Pterostigmablackish, wings hyaline. AUSTRALIAN LABENINI & POECILOCRYPTINI 137 Male. Similar to female but with fore wing length 3 mm; malar space 0-7 times basal mandibular width;3r-m present; gaster with segment 1 quite stout, the sternite 0-5 times as long as hind coxa; apex ofgonosquama flattened slightly with scattered hairs. Similarly coloured to female but with flagellum entirelyblack as is most of mesopleuron and metapleuron. REMARKS. The holotype differs from the female paratype in having entirely black antennae. C. zebrusbelongs to the leeuwinensis-group. It is the only species in the complex which is virtually entirely black andwhite. Structurally it is rather unremarkable but it has a more strongly raised subalar prominence than anyother species in this group.It is only known from tropical Queensland. MATERIAL EXAMINED Holotype $, Queensland: Baldy Mtn Road, via Atherton, vi.1981, ex Malaise trap (Brown) (QM). Paratypes. Queensland: 1 cT, 3 km N. by E. Mt Tip Tree (17-02S 145-37E), x.1980, at light (Cardale)(ANIC); 1 $, Windsor Tableland via Mt Carbine, xii.1980, ex Malaise trap (BMNH). Certonotussp. A(Figs 48, 50) Female. Medium-sized species, fore wing length 7 mm. Labium with glossae unspecialized. Lower face atnarrowest point 1-3 times as broad as high; malar space 1-0 times as long as basal mandibular width.Occipital carina ventrally long, about 3 times length of abscissa of hypostomal carina between it and base ofmandible. Upper part of pronotum, slightly before posterior corner, moderately convex, in dorsal viewprojecting beyond mesoscutal margin; subalar prominence weakly convex, rather sharp and ridge-like.Scutellum coarsely, closely punctate, transverse keel weak; metapleuron closely, coarsely punctate;submetapleural carina anteriorly expanded into a broad rounded flange, metanotum with a distinct toothnear anterior end of lateral carina. Propodeum moderately long with anterior transverse carina centrallystrong; lateromedian longitudinal carina present anteriorly and weak between transverse carinae; pleuralcarina strong anteriorly, posteriorly obsolescent; area superomedia clearly discernible, hexagonal; areaspiracularis complete; first and second lateral areae indistinctly delineated, confluent. Fore wing with 3r-mconverging towards 2r-m, joining Rs separately; 2m-cu joining M opposite 3r-m. Hind wing with distalabscissa of Cu\ absent. Hind tibia with posterior margin without spine-like bristles. Gaster with segment 1short, the sternite 0-2 times as long as hind coxa, not reaching level of spiracle. Tergite 7 mediodorsally withindentation on posterior margin; tergite 8 posteriorly very elongate and narrow; tergite 9 in dorsal viewquadrate, rounded posteriorly. Ovipositor projecting beyond apex of gaster by 4 times length of hind tibia. Coloration. Head blackish with lower face, genal and frontal orbits yellow; alitrunk reddish brown, thetergites peripherally black and only tegula and subalar prominence yellow. Gaster with tergites 3+ withposterior and lateral margins very narrowly yellow. Legs reddish brown, anterior two pairs of coxaeyellowish marked, hind coxa, trochanter, femur proximally and distally, and tarsus infuscate. Pterostigmablackish, wings hyaline. Male. Unknown. REMARKS. Certonotus sp. A is easily recognized by its colour pattern and possession of very slender tergalprocesses. The area superomedia is almost regularly hexagonal and strongly delineated except for the weakposterolateral sides. Despite lacking any trace of a distal abscissa of Ci in the hind wing this speciesappears to be quite closely related to C. ixion, suggesting it belongs in the humeralifer-group rather than inthe leeuwinensis-group.It is known only from Western Australia. MATERIAL EXAMINEDWestern Australia: 1 $ , Swan River (BMNH). Tribe POECILOCRYPTINI This relatively small tribe is characterized by the lack of an occipital carina dorsally and thepresence of a single bulla in vein 2m-cu in the fore wing. It is restricted to Australia where it isrepresented by three genera, Alaothyris, Urancyla and Poecilocryptus . Very little is known about the biology of poecilocryptines, but the available records suggestthe species all oviposit into nutritious plant tissue. It is possible that the larva is partiallyphytophagous, as it has rather massive mandibles (Short, 1978). 138 I. D. GAULD & G. A. HOLLOWAY Key to genera of Poecilocryptini 1 Fore wing with areolet and 2r-m obliterated by fusion of Rs and M (Fig. 63); gaster strongly laterally compressed; ovipositor very long, at least 6-0 times length of hind tibia ALAOTHYRIS Gauld(p. 138) - Fore wing with areolet distinct, bounded internally by 2r-m (Fig. 5); gaster cylindrical or depressed ; ovipositor less than 5 -0 times length of hind tibia 2 2 Hind wing with distal abscissa of Cu\ present; propodeal carinae vestigial dorsally; ovipositor strongly decurved (Fig. 64) URANCYLA Gauld(p. 141) - Hind wing with distal abscissa of Cu\ absent; propodeum with at least some carinae dorsally; ovipositor more or less straight POECILOCRYPTUS Cameron (p. 139) ALAOTHYRIS Gauld Genus A; Gauld, 1983: 169. Alaothyris Gauld, 1984: 93. Type-species: Alaothyris elongissimus Gauld, by original designation. Medium-sized species, fore wing length 6 mm; clypeus small, flat, truncate; labrum moderately large,exposed; mandible quite short, tapered, twisted about 20 times, almost evenly bidentate; malar spaceslightly less than basal mandibular width. Occipital carina absent on dorsal part of head, ventrally joininghypostomal carina above base of mandible. Antenna long, not tapered. Mesoscutum polished, almostsmooth; notaulus deep on anterior 0-2 of scutum, notaular crest occluding extreme anterior end; scutellumweakly convex, not laterally carinate; propodeum long, evenly rounded without carinae dorsally;propodeal spiracles circular; gaster inserted at end of short propodeal neck, above and far behind coxalinsertion. Fore tibia with a small tooth on outer side; mid and hind coxae very elongate; tarsal claws simple.Fore wing with cu-a proximal to base of Rs&M ; 3r-m absent; Rs and M fused to obliterate areolet and 2r-m;2m-cu with a single bulla. Hind wing with distal abscissa of Ct/i absent; basal cell slender; Sc bearing onehamulus. Gaster very long and slender, laterally compressed; tergite 1 slender, with spiracles a little behindcentre; sternite 1 reaching far behind level of spiracles; tergites 2-3 with laterotergites folded under.Ovipositor very long and slender, projecting beyond apex of gaster by more than 6-0 times length of hindtibia; apex cylindrical. REMARKS. A very distinctive genus easily recognized by its slender facies and characteristic venation. Thesystematic position of this genus is questionable. The mandible and elongate structure suggest arelationship with the Labenini, but the position of the petiolar spiracle, venation and shape of propodeumsuggest that it is perhaps more closely related to the Poecilocryptini, especially Poecilocryptus. Unlike theLabenini, Alaothyris does not have fine, file-like teeth on the ovipositor apex, nor apparently does it have alobe at the base of the ovipositor sheath. A single species is known. Alaothyris elongissimus Gauld (Fig. 63)Alaothyris elongissimus Gauld, 1984: 94. Holotype 9, QUEENSLAND (ANIC) [examined]. Female. Lower face slightly elongate, with a pronounced central tubercle; eye surface finely pubescent;ocelli arranged in an equilateral triangle; flagellum with about 28 segments. Mesoscutum polished,impunctate; mesopleuron and metapleuron similarly smooth, epicnemial carina dorsally obsolescent;submetapleural carina broad anteriorly. Gaster highly polished. Coloration. Predominantly orange-brown species with flagellum, hind leg and gaster darker brown.Pterostigma brown, wings hyaline. Male. Similar to female. REMARKS. The holotype and paratype emerged from the seeds ofAraucaria cunninghamii. What their hostwas is not known. Probably it will be found to be some seed-feeding beetle, but the possibility (given thesemi-phytophagous tendencies of some labenines) that this is partially a seed-feeding ichneumonid cannotbe ruled out. MATERIAL EXAMINED Holotype $, Queensland: Yarraman, vii.1969 (Heather) (ANIC).1 O" (paratype), same data as holotype. AUSTRALIAN LABENINI & POECILOCRYPTINI 139 POECILOCR YPTUS Cameron Poecilocryptus Cameron, 1901: 527. Type-species: Poecilocryptus nigromaculatus Cameron, by mono- typy.Poecilopimpla Morley, 1914: 35, 36. [Unnecessary replacement name for Poecilocryptus Cameron.] [Homonym of Poecilopimpla Cameron, 1903.] Medium-sized species, fore wing length 6-10 mm; clypeus rather small, apically very thin, truncate; labrumsmall, exposed; mandible short, slightly twisted, strongly narrowed, bidenjtate; malar space shorter thanbasal mandibular width. Occipital carina dorsally absent; eye with a weak indentation opposite antennalsocket. Antenna moderately long, clavate. Mesoscutum polished, virtually impunctate; notauli deep onanterior 0-2 of scutum, notaular crests strong. Propodeum abruptly rounded with spiracle oval; areasuperomedia large, quadrate, often confluent with area petiolaris; area externa not usually definedlaterally; gaster inserted well above level of hind coxae. Fore tibia simple, its apex not bearing a long spine;tarsal claws large, simple, or in some species those of the anterior two pairs of legs basally lobate. Fore wingwith cu-a slightly proximal to base of Rs&M\ 3r-m complete, enclosing a large, transverse pentagonalareolet; Im-cu with a single bulla, straight but inclivous. Hind wing with distal abscissa of Cui absent; basalcell not exceptionally broad; Sc bearing about two hamuli. Gaster long, quite slender; tergite 1 slender,evenly broadened posteriorly with spiracles at or slightly behind centre, sternite reaching to spiracles, thatof female bearing a pair of knob-like protuberances near anterior end. Ovipositor moderately long,projecting beyond apex of gaster by 2-8-3-3 times length of hind tibia, its apex cylindrical, with lower valvepartially enclosing the upper, with an indistinct matt area laterally, the upper valve with weak dorsal teeth. REMARKS. Poecilocryptus is an endemic Australian genus with species widely distributed throughout thecontinent. They seem to be associated with a variety of galls on trees of the genera Eucalyptus and Acacia.Parrott (1954) recorded three species from Australia. In this work four species are recognized, two ofwhich are described as new. One, P. galliphagus, has previously been incorrectly known as P. nigromacula-tus. In fact nigromaculatus is a senior synonym of P. stramineus. The relationships of the species. The four species may be placed in two species-groups, the nigromacula-tus-group (containing nigromaculatus and galliphagus) which is characterized by having acute lobes on theclaws of the anterior two pairs of legs, and the nigripectus-group (containing nigripectus and coloratus)which has the first lateral area very reduced in size. Key to species of Poecilocryptus 1 Tergites 1 and 2 of gaster white ; alitrunk predominantly black , only pronotum and anterior parts of mesoscutum and mesopleuron orange coloratus sp. n. (p. 139) - Tergites 1 and 2 of gaster bright yellow, sometimes with black marks; alitrunk predominantly bright yellow with profuse black spots 2 Propodeum with area superomedia delineated posteriorly by a strong carina (Fig. 56) ; flagellum with a subapical pale mark; sternite 1 of gaster with weak antero-ventral sublateral keels;tergite 2 entirely yellow nigripectus Turner & Waterston(p. 140) - Propodeum with area superomedia not delineated posteriorly (Fig. 57); flagellum entirely black; sternite 1 of gaster with strong antero-ventral sublateral tubercles; tergite 2 alwaysblack-marked 3 3 Distal apex of hind femur black; ovipositor projecting beyond apex of gaster by 2-0-2-3 times length of hind tibia galliphagus sp. n.(p. 140) - Distal apex of hind femur yellow; ovipositor projecting beyond apex of gaster by 2-8-3-3 times length of hind tibia nigromaculatus Cameron(p. 141) Poecilocryptus coloratus sp. n. Female. Unknown. Male. Small species, fore wing length 4-5 mm. Hypostomal carina, above mandibular base, weaklyraised. Metapleuron smooth and highly polished. Propodeal carinae quite weak; area superomedia notdelineated by a carina posteriorly; lateral longitudinal carina present above spiracle; first lateral area quitesmall, less than the area of area dentipara. Fore wing with Rs between 2r-m and 3r-m distinctly shorter thanlength of 2r-m. Fore and mid tarsal claws simple. Gaster with sternite 1 simple. Coloration. Face whitish, head, pronotum and anterior parts of mesothorax orange. Remainder ofalitrunk black. Gaster with tergites 1 and 2 white, 3 white with a central black mark, 4+ black withposterior and lateral margins broadly white. Flagellum black. Fore leg orange; mid leg with femur and 140 I. D. GAULD & G. A. HOLLOW AY proximal segments white, tibia and tarsus blackish; hind leg black, femur proximally and distally slightlyreddish orange. Pterostigma blackish, wings infumate. REMARKS. This small species is easily recognized by its atypical colour pattern. It is the only species in thegenus that is not predominantly yellow. The propodeal carination is also quite distinctive. The simple tarsalclaws and small first lateral area suggest P. color atus may be related to P. nigripectus . MATERIAL EXAMINED Holotype cf , Tasmania: Coles Bay, ii-iii.19-- (TC).Paratypes. Tasmania: 3 cf , same data as holotype (BMNH, TC); 1 cf , Mt Barrow, 700 m, ii.19-- (TC). Poecilocryptus galliphagussp. n. [Poecilocryptus nigromaculatus Cameron; Parrott, 1954: 240. Misidentification.] Female. Medium-sized species, fore wing length 7-10 mm. Hypostomal carina, above mandibular base,moderately raised. Metapleuron smooth and highly polished. Propodeal carinae strong; area superomedianot delineated by a carina posteriorly; lateral longitudinal carina not complete above spiracle; first lateralarea very large, more than twice the area of area dentipara. Fore wing with Rs between 2r-m and 3r-mabout twice as long as length of 2r-m. Fore and mid tarsal claws with a well-developed acute basal lobe.Gaster with sternite 1 bearing a pair of antero-ventral sublateral tubercles. Ovipositor projecting beyondapex of gaster by 2-0-2-3 times length of hind tibia. Coloration. A bright yellow species with black marks on interocellar area, vertex behind ocelli, threelongitudinal stripes on mesoscutum, anterior propodeal areae, anterior margin of all abdominal tergites,hind femur centrally and hind tibia distally. Flagellum and ovipositor sheath black. Distal tarsal segmentsslightly infuscate. Pterostigma dark brown, wings hyaline. Male. Similar to female though slightly more slender, often with sternal tubercles very weak. REMARKS. This species is rather similar to P. nigromaculatus, with which it is frequently confused incollections. The two species may easily be separated by the characters given in the key. HOST RECORDS. This species has been reared from galls on Eucalyptus delegatensis and E. pauciflora. MATERIAL EXAMINED Holotype $ , Victoria: Wiseleigh via Bruthen, ix.1962, ex eucalypt gall (Hobb) (NMV). Paratypes. Queensland: 1 $, no further data (Riek) (BMNH). New South Wales: 1 $, Dainer's Gap,xi.1972 (ANIC); 1 $, Deer Vale, i.1931 (Burns) (NMV); 1 cf , Mt Victoria, x.1930 (Burns) (NMV); 1 $,Sydney (Froggatt) (ANIC). Victoria: 2 $, Eildon area, ix.1959 (Irvine) (NMV); 2 $, Mt Pinniber, iv.1961(Taylor) (ANIC); 1 $, Toolangi, xi.1982 (Farrugia) (BMNH); 3 $, Warrandyte, viii-x.1928 (Hill)(ANIC); 2 Cf, no further locality data (French) (BMNH). Tasmania: 3 $, Collinsvale, Fairy Glen,i-ii.1983 (Williams & Gauld) (BMNH). Poecilocryptus nigripectus Turner & Waterston (Fig. 56)Poecilocryptus nigripectus Turner & Waterston, 1920: 24. Holotype $, TASMANIA (BMNH) [examined]. Female. Medium-sized species, fore wing length 7-11 mm. Hypostomal carina, above mandibular base,strongly raised. Metapleuron with distinct longitudinal wrinkles, moderately polished. Propodeal carinaestrong; area superomedia rectangular, delineated by a carina posteriorly; lateral longitudinal carinacomplete above spiracle; first lateral area not exceptionally large, of approximately the same area as areadentipara. Fore wing with Rs between 2r-m and 3r-m subequal to length of 2r-m. Fore and mid tarsal clawssimple. Gaster with sternite 1 bearing a pair of weak antero-ventral sublateral keels. Ovipositor projectingbeyond apex of gaster by 2-2-2-4 times length of hind tibia. Coloration. Bright yellow species with interocellar area, vertex behind eyes, posterior part of mesoscu-tum, anterior part of propodeum and metapleuron, anterior 0-7 of tergite 1, and most of tergites 3 and 6black. Legs yellow, hind femur broadly black centrally, distal hind tarsal segment infuscate. Flagellumblack, with a broad subapical yellowish white band. Pterostigma dark brown, wings hyaline. Male. Similar to female but slightly more slender and with gastral tergites more coarsely punctate. REMARKS. This is a particularly characteristically patterned species. Structurally it is more similar to P.coloratus than it is to P. nigromaculatus, a species that it superficially resembles in ground-colour. AUSTRALIAN LABENINI & POECILOCRYPTINI 141 HOST RECORDS. In the BMNH is a specimen reared from anthribid galls. MATERIAL EXAMINED Holotype $ , Tasmania: Mt Wellington (Turner) (BMNH). Queensland: 4 $ , Brisbane, ii.1969 (Campbell) (QM); 1 cf , Iron Range, v.1975 (Moulds) (AM); 1 $, nodata (NMV). New South Wales: 1 cf, Bateman's Bay, x.1969 (Riek) (ANIC); 1 $> Bendigo (Froggatt)(ANIC). Australian Capital Territory: 2 $, Canberra, x.1930 (Bruce) (ANIC). Poecilocryptus nigromaculatus Cameron Poedlocryptus nigromaculatus Cameron, 1901: 528. LECTOTYPE $, AUSTRALIA (BMNH), here desig-nated [examined] . Poeciloeryptus (sic) nigro-maculatus Cameron; Cameron, 1911: 335. Poecilopimpla nigromaculata (Cameron) Morley, 1914: 36. Poecilopimpla nigromaculata var. straminea Morley, 1914: 36. Holotype $, NEW SOUTH WALES (BMNH)[examined]. Syn. n. Poecilocryptus stramineus (Morley) Parrott, 1954: 241. Poecilocryptus nigromaculatus Cameron; Townes et al. , 1961: 117. Female. Medium-sized species, fore wing length 6-10 mm. Hypostomal carina, above mandibular base,weakly raised. Metapleuron smooth and highly polished. Propodeal carinae strong; area superomedia notdelineated by a carina posteriorly; lateral longitudinal carina absent above spiracle; first lateral area verylarge, more than twice the area of area dentipara. Fore wing with Rs between 2r-m and 3r-m about twice aslong as length of 2r-m. Fore and mid tarsal claws with a well-developed acute basal lobe. Gaster withsternite 1 bearing a pair of antero-ventral sublateral tubercles. Ovipositor projecting beyond apex of gasterby 2-8-3-3 times length of hind tibia. Coloration. A bright yellow species with black marks on interocellar area, vertex behind ocelli, threelongitudinal stripes on mesoscutum, anterior propodeal areae, anterior margin of all abdominal tergitesand hind femur centrally. Flagellum and ovipositor sheath black. Distal hind tarsal segments stronglyinfuscate. Pterostigma dark brown, wings hyaline. Male. Similar to female though slightly more slender, often with sternal tubercles very weak. REMARKS. In the BMNH are two specimens labelled as 'Cameron types' of nigromaculatus. They areconspecific and one has been labelled and is here designated as lectotype. Morley (1914) clearly referred tothe female that was reared by Froggatt as the 'typical' specimen, and this must be construed as a valid typerestriction (Art. 73(a)(i) of the Code). Parrott (1954) was incorrect to refer to the male in Froggatt'scollection as the holotype. The female holotype is a slightly undersized specimen, but clearly conspecificwith nigromaculatus. P. nigromaculatus appears to be the sister-species of P. galliphagus. Both species have a well-developedbasal lobe on the fore and mid tarsal claws and have the first lateral area of the propodeum greatlyenlarged. P. nigromaculatus may be recognized by its elongate ovipositor and entirely yellow hind tibia.We have examined the differences between this species and galliphagus tabulated by Parrott (1954) andhave found that only the ovipositor character holds up. We failed to find any difference in the ratio ofinterocellar to orbital-ocellar distance, and the range of numbers of flagellar segments for both species isvery similar. Small specimens of either species have fewer flagellar segments than large individuals. Theextent of the black banding on third to fifth gastral tergites is quite variable in both species. HOST RECORDS. P. nigromaculatus has been reared from anthribid and chalcid galls on Acacia longifolia,and eriococcid galls on Eucalyptus. MATERIAL EXAMINED Lectotype $ (nigromaculatus Cameron), 'Australia': no further data (BMNH). Paralectotype, 1 $,same data (BMNH). Holotype $ (nigromaculata var. straminea Morley), New South Wales: no furtherdata (Froggatt) (BMNH). 'Australia': 1 $ (paralectotype of nigromaculatus) (BMNH). 38 $,34 cf, Queensland, New SouthWales, Australian Capital Territory, Victoria, Tasmania (ANIC, BMNH, NMV). URANCYLA Gauld Genus U; Gauld, 1983: 169. Urancyla Gauld, 1984: 95. Type-species: Urancylafulva Gauld, by original designation. Medium-sized species; fore wing length 6 mm; clypeus flat, small, apically truncate with margin thin; 142 I. D. GAULD & G. A. HOLLO WAY mandible strongly tapered, twisted and with upper tooth slightly the longer; malar space shorter than basalmandibular width. Occipital carina absent dorsally , ventrally joining hypostomal carina well above base ofmandible. Antenna long, neither tapered nor clavate distally. Mesoscutum polished, punctate; notauluspresent near front margin, with a small crest occluding extreme end; scutellum flat, without lateral carinae;propodeum evenly rounded with vestiges of carinae though areae superomedia and petiolaris are notdefined; gaster inserted low on propodeum, near level of hind coxae. Fore tibia with a small tooth on outerside, femur with a weak longitudinal ventral furrow; tarsal claws of female with a basal lobe. Fore wing withcu-a subopposite Rs&M; 3r-m present, weakly pigmented, enclosing a small pentagonal areolet; 2m-cuwith a single bulla. Hind wing with distal abscissa of Cu\ present; first abscissa of Cu\ shorter than cu-a;basal cell moderately broad; Sc bearing one hamulus. Gaster moderately long, tergite 1 slender withspiracle slightly behind centre; tergites 2 and 3 with pendant laterotergites which are almost membranous.Ovipositor about as long as gaster, evenly decurved, apex simply acute with inconspicuous teeth andindistinct matt area laterally.Male. Unknown. REMARKS. In Townes' (1969) key to Labiinae this genus runs to the tribe Clasini but it does not appear to berelated to the genera in this group. Clasines have a nodus on the ovipositor apex, long, simple claws andtwo bullae in 2m-cu. They also have no trace of a notaular crest. The twisted mandible and single bulla in2m-cu are characters that Urancyla shares with Poecilocryptus and the two genera appear to be closelyrelated. Urancyla fulva Gauld (Fig. 64)Urancyla fulva Gauld, 1984: 95. Holotype $, QUEENSLAND (TC) [examined]. Female. Lower face elongate, regularly punctate; frons polished and finely punctate; ocelli arranged in anequilateral triangle. Flagellum with 30 segments. Mesoscutum polished, regularly and finely punctate;mesopleuron highly polished, smooth, almost impunctate; metapleuron polished with scattered finepunctures; submetapleural carina moderately wide, evenly tapered anteriorly. Gaster highly polished,finely punctate. Coloration. Predominantly orange-brown species; face, upper orbits, genae, propleuron, anteriormargin of pronotum, diagonal stripe across mesopleuron, fore coxa and trochanters and a stripe on midcoxa pale yellowish; flagellum, except centrally, scape and pedicel, frons centrally, vertex, interocellararea, occiput, mesoscutum and ovipositor sheath, pterostigma black. Male. Unknown. REMARKS. This species is known only from the holotype. MATERIAL EXAMINEDHolotype $, Queensland: Brisbane, xi.1972 (Sedlacek) (TC). Acknowledgements We thank the curators of the institutions that kindly loaned us material, and the individuals who ranMalaise traps to secure additional specimens. We are particularly grateful to Dennis Farrugia for collectinga considerable amount of material in southern Victoria. IDG would like to thank ABRS for providing agrant to study in Australia. GAH would like to thank the New South Wales Premier's Department forsupport whilst overseas, and Nestle (Australia) Ltd for funds. References Askew, R. R. 1971. Parasitic insects, xvii + 316 pp. London. Bosch, R. van den 1978. The pesticide conspiracy. 226 pp. New York. Brulle, M. A. 1846. In Lepeletier de Saint-Fargeau, A. L. M.,Histoire naturelle des insectes 4. Hymenop- teres. viii + 680 pp. Paris.Cameron, P. 1901. Descriptions of seventeen new genera of Ichneumonidae from India and one from Australia (concluded). Annals and Magazine of Natural History (7) 7: 523-531.1903. Descriptions of new genera and species of Hymenoptera taken by Mr Robert Shelford at Sarawak, Borneo. Journal of the Straits Branch of the Royal Asiatic Society 39: 89-181.1911. On a collection of parasitic Hymenoptera (chiefly bred) made by W. W. Froggatt, in New South AUSTRALIAN LABENINI & POECILOCRYPTINI 143 Wales, with descriptions of new genera and species. Part 1. Proceedings of the Linnean Society of New South Wales 36: 333-346.Cheesman, L. E. 1936. Hymenoptera of the New Hebrides and Bank Islands. Transactions of the Royal Entomological Society of London 85: 169-195.Cresson, E. T. 1864. Descriptions of North American Hymenoptera in the collection of the Entomological Society of Philadelphia. Proceedings of the Entomological Society of Philadelphia 3: 397-402.Fullaway, D. T. 1942. A checklist of parasitic Hymenoptera collected by J. X. Williams in New Caledonia. Proceedings of the Hawaiian Entomological Society 11: 243-247.Gauld, I. D. 1983. The classification, evolution and distribution of the Labeninae, an ancient southern group of Ichneumonidae (Hymenoptera). Systematic Entomology 8: 167-178. 1984. An introduction to the Ichneumonidae of Australia. 413 pp. London. Hardy, D. E. 1982. The role of taxonomy and systematics in integrated pest management programmes. News Bulletin of the Entomological Society of Queensland 10: 1924.Hocking, H. 1967. A native ichneumonid Certonotus tasmaniensis Turner, parasitizing Sirex noctilio (F.) (Siricidae). Journal of the Australian Entomological Society 6: 57-60.Huffaker, C. B. & Messenger, P. S. (Eds). 1976. Theory and practice of biological control. 788 pp. New York. Kriechbaumer, J. 1889. Nova genera et species Pimplidarum. Entomologische Nachrichten 15: 307-312.1890. Ichneumon-Studien. Neue Ichneumoniden des Wiener Museums. Annalen des Naturhistoris- chen Hofmuseums Wien 5: 479491.Krieger, R. 1901. Ueber die Ichneumoniden-Gattung Certonotus Kriechbaumer. Zeitschrift fur Systema- tische Hymenopterologie und Dipterologie 1: 113-126. Matsuda, R. 1976. Morphology and evolution of the insect abdomen with special reference to developmen-tal patterns and their bearing upon systematics. International Series of Monographs on Pure and Applied Biology (Zoology Division) 56: 1-532.Morley, C. 1913. A revision of the Ichneumonidae ... 2. Tribes Rhyssides, Echthromorphides, Anomalides and Paniscides. ix + 140 pp. London. 1914. A revision of the Ichneumonidae . . .3. Tribes Pimplides and Bassides. ix + 149 pp. London. Muldrew, J. A. 1967. Biology and initial dispersal of Olesicampe (Holocremnus) sp. nr. nematorum (Hymenoptera; Ichneumonidae), a parasite of the larch sawfly recently established in Manitoba. Canadian Entomologist 99: 312-321.Nuttall, M. J. 1973. Pre-emergence fertilization oiMegarhyssa nortoni (Hymenoptera: Ichneumonidae). New Zealand Entomologist 5: 112-117.Parrott, A. W. 1954. Australian species of Poecilocryptus Cameron (Ichneumonidae: Hymenoptera). Pacific Science 8: 239-242.1955. A new genus and species of the tribe Labenini from Australia (Pimplinae: Ichneumonidae). Proceedings of the Linnean Society of New South Wales 79: 230-232.Schulz, W. A. 1906. Spolia Hymenopterologica. 356 pp. Paderborn.Short, J. T. R. 1978. Larvae of Ichneumonidae. Memoirs of the American Entomological Institute 25: 1-508.Smith, E. L. 1970. Evolutionary morphology of the external insect genitalia. 2 Hymenoptera. Annals of the Entomological Society of America 63: 1-27.Taylor, K. L. 1978. Evaluation of the insect parasitoids of Sirex noctilio (Hymenoptera: Siricidae) in Tasmania. Oecologia 32: 1-10. Townes, H. 1969. Genera of Ichneumonidae 1. Memoirs of the American Entomological Institute 11: 1-300.Townes, H. & Townes, M. 1960. Ichneumon-flies of America north of Mexico: 2, subfamilies Ephialtinae, Xoridinae, Acaenitinae. Bulletin of the United States National Museum 216 (2): 1-676.Townes, H., Townes, M. & Gupta, V. K. 1961. A catalogue and reclassification of Indo- Australian Ichneumonidae. Memoirs of the American Entomological Institute 1: 1-522.Tryon, H. 1900. Caterpillar plague. Queensland Agricultural} 'ournal 6: 135-147.Turner, R. E. 1919. Notes on the Ichneumonidae in the British Museum 1 . Annals and Magazine of Natural History (9) 3: 550-558.Turner, R. E. & Waterston, J. 1920. A revision of the ichneumonid genera Labium Brulle and Poecilocryptus Cameron. Proceedings of the Zoological Society of London 1920: 1-26.Viereck, H. L. 1914. Type species of the genera of ichneumon-flies. Bulletin of the United States National Museum 31: 1-186. Walsh, B. D. 1866. Borers. Practical Entomologist 1: 25-31.Wilson, F. 1960. A review of the biological control of insects and weeds in Australia and Australian New Guinea. Technical Communication of the Commonwealth Institute of Biological Control 1: 1-102. I. D. GAULD & G. A. HOLLOWAY 14 Figs 1-14 1, 2, face of (1) Labium sp.; (2) Labena sp. 3, fore wing of Labena sp. 4, base of ovipositorsheath, Certonotus nitidulus. 5, 6, fore wings of (5) Urancyla fulva; (6) Adelphion sp. 7, fore leg ofLabena sp. 8-10, gonosquama of (8) Labena annulata; (9) L. keira; (10) L. pudenda. 11, 12, hind wingsof (11) L. annulata; (12) L. fce/ra. 13, 14, tergite 2 of (13) L. annulata; (14) L. chadwickii. AUSTRALIAN LABENINI & POECILOCRYPTINI Figs 15-33 15, 16, end of gaster of (15) Labena grandis; (16) L. malecasta. 17, 18, mid tibia of (17) L.jacunda; (18) L. keira. 19, 20, propodeum of (19) L. keira; (20) L. pudenda. 21, apex of ovipositor, L.pudenda. 22-24, pro- and mesoscutum of (22) Certonotus rufescens; (23) C. apicalis; (24) C. humeralifer.25, 26, face of (25) C. rufescens; (26) C. humeralifer. 27-29, propodeum of (27) C. humeralifer; (28) C.annulatus; (29) C. andrewi. 30, 31, areolet of fore wing of (30) C. humeralifer, (31) C. annulatus. 32, 33,propodeum of (32) C. geniculatus; (33) C. nitidulus. 146 I. D. GAULD & G. A. HOLLOWAY Figs 34-49 34, subalar prominence of Certonotus talus. 35-37, end of gaster of (35) C. talus; (36) C.sisyphus; (37) C. cestus. 38, 39, mouthparts and back of head of (38) C. monticola; (39) C. sisyphus. 40,41, dorsal view of propodeum of (40) C. sisyphus; (41) C. avitus. 42, 43, lateral view of propodeum of(42) C. avitus; (43) C. cestus. 44, 45, propodeum and tergite 1 of (44) C. celeus; (45) C. pineus. 46, 47,alitrunk of (46) C. pineus; (47) C. mogimbensis. 48, 49, propodeum of (48) Certonotus sp. A; (49) C.leeuwinensis. AUSTRALIAN LABENINI & POECILOCRYPTINI 147 Figs 50-64 50, 51, end of gaster of (50) Certonotus sp. A; (51) C. zebrus. 52, 53, head of (52) C.leeuwinensis; (53) C. zebrus. 54, 55, hind coxa of (54) C. paluma; (55) C. ixion. 56, 57, propodeum of(56) Poecilocryptus nigripectus; (57) P. nigromaculatus . 58-62, body in profile to show colour pattern of(58) Certonotus farrugiai; (59) C. zebrus; (60) C. toolangi; (61) C. ixion; (62) C. paluma. 63, fore wing ofAlaothyris elongissimus . 64, alitrunk and gaster of Urancylafulva. 148 I. D. GAULD & G. A. HOLLO WAY H0) CO 8 0) 00 4,5 10 7,8 Fig. 65 Cladogram showing putative phylogenetic relationship of Australian species of Labena. (Note.The holophyly of this grouping vis a vis the Neotropical species has not been established.) Theapomorphic features supporting this cladogram are: 1 , mesopleuron smooth; 2, ring of long hairs aroundapex of gonosquama; 3, apex of fore wing infumate; 4, apex of gonosquama indented; 5, apex ofovipositor with coarse teeth; 6, fore tibia slender; 7, vein Cu\ incomplete in hind wing; 8, flagellum withapical white band; 9, gaster closely punctate; 10, mesoscutum striate. Acacia longifolia 141Buprestidae 114, 124 Cerambycidae 113Diadoxus sp. 124 INDEX Index to hosts Host plants, as well as host insects, are included in this index. Ethon affine 114Eucalyptus 141Eucalyptus delegatensis 140Eucalyptus pauciflora 140 Pinusradiata 114 Sirex noctilio 131Sirexsp. 114,132Siricidae 131 Ur acanthus strigosus 113 149 Index to Ichneumonidae Invalid names are in italics; principal references are in bold. Alaothyris 109, 138andrewi 108, 119, 121annulata 109, 112, 113apicalis 108, 119, 122Asperellus 108, 117avitus 108, 120, 123 celeus 108, 120, 123Certonotus 108, 111,117cestus 108, 120, 123chadwickii 109, 112, 113coloratus 109, 139 elongissimus 109, 138 farrugiai 108, 120, 126fulva 109, 142 galliphagus 109, 139, 140geniculatus 108, 119, 124grandis!09, 112, 114 hinnuleus 108, 120, 126humeralifer 108, 119, 127 ixion 108, 121, 128jacunda 109, 112, 115keira 109, 112, 115 Labena 109, 111, 112Labeninae 108, 109, 110Labenini 108, 109, 110, 111leeuwinensis 108, 120, 128 malecasta 109, 112, 116mogimbensis 108, 120, 129monticola 108, 119, 130 nigripectus 109, 139, 140nigromaculatus 109, 139, 141 nitidulus 108, 119, 131 paluma 108, 121,132pineus 108, 120, 132Poecilocryptini 108, 109, 137, 138Poecilocryptus 109, 138, 139pudenda 109, 113, 117 rufescens 108, 119, 133 sisyphus 108, 120, 134species A 109, 120, 137straminea 109, 141 talus 108, 119, 121,135tasmanienis 108, 131toolangi 109, 121, 136 Uraneyla 109, 138, 141zebrus 109, 120, 136 /. D. Gauld In the important field of biological and integrated control of pests the parasitic Hymenopteraare of particular significance, and this work considers one of the largest families of Parasitica,the Ichneumonidae. The group has received little attention in Australia - though it has alreadybeen utilized successfully in curtailing the ravages caused by accidentally introduced pests. Forselective control programmes to be effective, however, a sound knowledge of the biology ofboth the pest and its parasites is essential - and a sound taxonomic base is vital for thedevelopment of such knowledge. Ironically, considering the group's economic importance, the parasitic Hymenoptera areamongst the least studied of any group of living organisms, and taxonomic difficulties havepresented major problems to many entomologists working with the Parasitica. AnIntroduction to the Ichneumonidae of Australia will go a long way towards rectifying thissituation, being a taxonomic treatment, by genus, of the Australian ichneumonids, acomprehensive illustrated identification guide, and a summary of all available information onthe group. It will also serve as an introduction to the biology and distribution of Australianichneumonids, and provide a check-list of the described species and an index to their knownhosts. It provides an important revision of ichneumonid nomenclature in order to bring thegroup into line with the generally accepted principles of zoological nomenclature. 1984, 413pp, 3 maps, 580 figs. Paperback. 565 008% X 40.00 Titles to be published in Volume 53 A review of the Miletini (Lepidoptera: Lycaenidae) By J.N.Eliot Australian ichneumonids of the tribes Labenini and Poecilocryptini By I. D. Gauld & G. A. Holloway The tribe Pseudophloeini (Hemiptera: Coreidae) in the Old World tropics with a discussion onthe distribution of the Pseudophloeinae By W. R. Dolling The songs of the western European grasshoppers of the genus Omocestusin relation to theirtaxonomy (Orthoptera: Acrididae) ByD.R. Ragge Photoset by Rowland Phototypesetting Ltd, Bury St Edmunds, SuffolkPrinted in Great Britain by Henry Ling Ltd, Dorchester :i98< r Bulletin of the British Museum (Natural History) The tribe Pseudophloeini (Hemiptera:Coreidae) in the Old World tropics with adiscussion on the distribution of thePseudophloeinae W. R. Dolling Entomology series Vol53 No 3 30 October 1986 The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in fourscientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology,and an Historical series. Papers in the Bulletin are primarily the results of research carried out on the unique andever-growing collections of the Museum, both by the scientific staff of the Museum and byspecialists from elsewhere who make use of the Museum's resources. Many of the papers areworks of reference that will remain indispensable for years to come. Parts are published at irregular intervals as they become ready, each is complete in itself,available separately, and individually priced. Volumes contain about 300 pages and severalvolumes may appear within a calendar year. Subscriptions may be placed for one or more ofthe series on either an Annual or Per Volume basis. Prices vary according to the contents ofthe individual parts. Orders and enquiries should be sent to: Publications Sales, British Museum (Natural History),Cromwell Road, London SW75BD,England. World List abbreviation: Bull. Br. Mus. nat. Hist. (Ent.) Trustees of the British Museum (Natural History), 1986 The Entomology series is produced under the general editorship of theKeeper of Entomology: Laurence A. MoundAssistant EiJi ISBN 565 06021 XISSN 0524-6431 British Museum (Natural History)Cromwell RoadLondon SW75BD Entomology seriesVol 53 No 3 pp 151-212 Issued 30 October 1986 ,AL nia The tribe Pseudophloeini (Hemiptera: Coreidae) inthe Old World tropics with a discussion on thedistribution of the Pseudophloeinae W. R. Dolling Department of Entomology, British Museum (Natural History), Cromwell Road, LondonSW7 5BD Contents Synopsis 151 Introduction 151 Diagnosis of Pseudophloeinae and its division into tribes 152 Rejected genera 154 Abbreviations of depositories 155 Terminology and measurements 155 Key to the genera of Pseudophloeini found in the Old World tropics 156 Risbecocoris Izzard 156 HoplolomiaS&l 161 Indolomia gen. n 163 Paramyla Linnavuori 165 Psilolomia Breddin 168 Pungragen.n 179 Pseudomyla gen. n 180 Neomevaniomorpha gen. n 183 Mevaniomorpha Reuter 185 Mevanidea Reuter 187 Arenocoris Hahn 191 A/v/flStal 193 Distribution of Pseudophloeinae 205 Acknowledgements 209 References 209 Index 212 Synopsis The major morphological features of the Pseudophloeini are outlined and their bearing on the classifica-tion of the tribe is discussed. Twelve genera and 43 species are recognized in the tropical regions of Africaand Asia; descriptions and keys for their separation are provided. Four genera and 16 species are describedas new. Three new synonymies are established at the genus-level and four at the species-level; 10 newcombinations are established; two 'forms' are raised to the status of species and nine lectotypes aredesignated. Introduction Plant-feeding bugs of the family Coreidae are characteristic inhabitants of the herb and shrublayers of tropical and, to a lesser extent, of temperate ecosystems. They are frequentlyencountered in surveys of crop pests, since most of the world's tropical crops are herbs or shrubs,and are usually represented in collections made during ecological studies in the tropics. Most ofthe literature available for the identification of Coreidae is out of date, fragmentary or lacksidentification keys. Pseudophloeinae may be recognized, with a little practice, by their general habitus and size,and by the absence of a dorsal sulcus on the tibiae; other groups of Coreoidea lacking tibial sulci Bull. Br. Mus. not. Hist. (Ent.) 53 (3): 151-212 30 October 1986 152 W. R. DOLLING - Alydidae, Rhopalidae, Stenocephalidae and the coreid subfamily Hydarinae - are unlikely tobe mistaken for Pseudophloeinae because they are all of characteristic appearance. Coreinaesimilar in size and build to Pseudophloeinae all have very distinct tibial sulci. Twenty-eight validgenera and 166 species of Pseudophloeinae have been described. Most of the species are 7-10mm in length; a species of typical appearance is illustrated in Fig. 1. The food plants of all species of the subfamily, where known, are herbaceous legumes(Fabaceae). Host plant records for the tropical Pseudophloeini are very scarce, as are otherdetails of their biology. There is only a single, New World, species of the subfamily in whichfacultative brachyptery is known; all other species are fully macropterous. Flight, therefore, isprobably important, yet there are very few records of them being caught in light traps,interception traps or yellow trays; presumably they fly only rarely and then not far. Most specieshave rather restricted ranges and only one is recorded from oceanic islands. Tropical-cropentomologists and collectors in Europe report them to move sluggishly even when disturbed andto be reluctant to fly. The typical habitats of Pseudophloeinae reflect those of their host plants:open woodlands and grasslands with scattered trees; a few species have penetrated driergrasslands on the one hand and forest clearings on the other, but deserts and dense forests, tojudge from the distribution patterns of the tropical species, are impenetrable barriers to most ofthem. Several species are restricted to high altitudes, suggesting that low temperatures are notbarriers to dispersal in the long term, a reasonable supposition in view of the richness of thePalaearctic pseudophloeine fauna. The subfamily is represented in almost all parts of the major land masses except fornon-tropical Australia. Recent revisions by Froeschner (1963), Dolling (1977) and Dolling &Yonke (1976) enable the Nearctic and Neotropical species to be identified. Most Palaearcticspecies are covered by the keys of Stichel (1960), which should be used in conjunction with thenotes of Chernova (1979). The largest Palaearctic genus, Coriomeris Westwood, was revised byChernova (1978). A new Palaearctic species of Microtelocerus Reuter was described from Sinaiby Dolling (19796) and a new genus with a single new species was added by Puchkov (1979). Thetribe Clavigrallini, which is confined to the Old World tropics and contains a number of pestspecies, has been monographed by Dolling (1978: 1979a). The present revision covers all of thegenera of Pseudophloeinae not treated in the above works. Diagnosis of Pseudophloeinae and its division into tribes Stal's original (1868: 535) diagnosis of Pseudophloeinae mentioned only the presence of anantevannal vein. Later (1873: 33-34), he mentioned also the broad cells at the base of thehemelytral membrane, the form of the metathoracic scent-gland auricle, the lack of a mediandorsal impression on the head, the prominent and gently declivent tylus and juga, the prominentposterior angles of the seventh abdominal segment in both sexes, and the non-sulcate tibiae.Dolling (1978: 282) gave a fuller diagnosis, adding mention of the outer apical processes of theantennifers and several features of the genitalia. The antevannal vein is not present in fivegenera: the Neotropical Vilga Stal; the Oriental Hoplolomia Stal and Indolomia gen. n., theAfrotropical Paramyla Linnavuori and the Afro- Asian Risbecocoris Izzard. All five of thesegenera accord well with the revised diagnosis of the subfamily and there seems no reason toexclude them from it because of the absence of this single character; in fact, Stal (1873) includedHoplolomia in Pseudophloeinae, presumably on the basis of other features and in ignorance ofthe venation of the hind wing. The tribe Clavigrallini was erected (as division Clavigrallaria) by Stal (1873: 81) in a key to theAfrican and Asian genera of the subfamily. It was characterized by him as having the scutellumconvex, the base of the posterior femur devoid of the small tubercle which is present in manygenera of Pseudophloeinae, the body compressed laterally, the male genital capsule notbiemarginate posteriorly, the propleuron emarginate on its posterior margin near the postero-dorsal angle, the second antennal segment equalling or exceeding the third in length and theprescutellar angles of the pronotum armed with a spine. The first couplet of Stal's key contrastedMevania Stal, My la Stal and Hoplolomia with the 'Clavigrallaria', uniting these three non- TRIBE PSEUDOPHLOEINI 153 Fig. 1 Psilolomia pundaloyae , dorsal view of male . 154 W. R. DOLLING clavigralline genera on the basis of their common possession of a flat or almost flat scutellum, atubercle at the base of the posterior femur, the posteriorly biemarginate genital capsule with theemargination filled by the apices of the parameres, the posterior tibia not or slightly shorter thanthe femur and the second antennal segment shorter than the third. Although Stal did notformally diagnose or name a 'division Pseudophloearia' to contrast with his 'Clavigrallaria', hisrecognition of the latter implies the existence of the nominate tribe. An additional feature, bothuniversal among Clavigrallini and restricted to them, is the presence of a pair of small tuberclesat the base of the mesoscutellum adjacent to the posterior margin of the pronotum. In a fewClavigrallini, the second antennal segment is shorter than the third, or the scutellum is flat, orthe genital capsule is biemarginate posteriorly, or the posterior tibiae are subequal in length tothe femora, or the body is depressed rather than compressed. The emargination of thepropleuron, the absence of a femoral tubercle and the presence of the basal tubercles of thescutellum are constant features of the tribe, as is the presence of an antevannal vein. Among thenon-clavigralline genera, the scutellum is occasionally convex and the propleuron emarginate(both conditions well developed in Vilga westwoodi (Kolenati)). The tubercle at the base of thefemur is absent in Risbecocoris and Vilga. No further tribes have been described, so the subfamily at present comprises the Clavigralliniand the nominate tribe, Pseudophloeini. The unique characters of the Clavigrallini are probablyapomorphies within Pseudophloeinae while those of Pseudophloeini are, in general, plesiomor-phies. Both the presence of the basal tubercle of the posterior femur and the presence of theantevannal vein of the metathoracic wing are probably apomorphies within the subfamily(though it is interesting to note that the genus Spathocera Stein, in the Coreinae, has anantevannal vein, and a group of genera including Riptortus Stal, in the Alydinae, have a femoraltubercle). Pseudophloeinae can, therefore, be divided into four unequal groups (Table 1): Vilgaand Risbecocoris, lacking both vein and tubercle; Paramyla, Hoplolomia and Indolomia,lacking the vein but possessing the tubercle; Clavigrallini, possessing the vein but lacking thetubercle; and all the remaining genera, possessing both vein and tubercle. Assuming that thelack of both vein and tubercle is primitive, a possible interpretation of the phylogeny of thePseudophloeinae is that Clavigrallini and most Pseudophloeini are united by descent from acommon ancestor that had no femoral tubercle but had acquired an antevannal vein; from thisstock developed two lines: on the one hand Clavigrallini and on the other all those genera withboth a femoral tubercle and an antevannal vein. On this interpretation, Paramyla, Hoplolomiaand Indolomia would have acquired their femoral tubercles independently of most Pseudo-phloeini; alternatively, these three genera may have secondarily lost their antevannal veins.Recently, Stys (1978) suggested that the antevannal vein was primitively present in the commonancestor of all Coreidae. This interpretation is rejected here because it would involve recogni-tion of a group comprising the genera Vilga, Risbecocoris, Paramyla, Hoplolomia and Indolo-mia sharing the apomorphy of the loss of this vein. This group, based on a loss character thatunites these five morphologically diverse and geographically widely dispersed genera, is unlikelyto have any phyletic validity. A classification that reflected the most probable phylogeny of the subfamily would require theerection of two or three additional tribes to accommodate Vilga, Risbecocoris and, possibly,Paramyla plus Hoplolomia plus Indolomia. An assumption of independent acquisition of theantevannal vein by Clavigrallini and the main group of Pseudophloeini would require separatetribes for Vilga and Risbecocoris but not the other three of these genera. In the present worksuch new tribes are not erected. The subfamily Pseudophloeinae and the tribe Clavigrallini areboth believed to be holophyletic in composition, leaving the nominate tribe paraphyletic. Rejected genera Four genera represented in the geographical area covered by the present revision were originallydescribed in Pseudophloeinae or associated with genera that belong in the subfamily. Brotheolus Bergroth (1908: 107) is a replacement name for the preoccupied Brotheus Distant TRIBE PSEUDOPHLOEINI 155 (19020: 248) which was placed by Distant (p. 246) in Pseudophloeinae. It was transferred toCoreinae: Gonocerini by Dolling (19796: 97). Trallianus Distant (19026: 404-405) was originally placed in Pseudophloeinae but wastransferred to Coreidae: Gonocerini by Dolling (19796: 97). Cristovallia Distant (1920: 149-150) was said by its author to have 'affinity with the generaClavigralla and Ceraleptus' (both Pseudophloeinae). Brown (1958: 514) synonymized itstype-species, C. typica Distant, with Amblypelta bilineata Stal in Coreinae: Dasynini. Austrocoris Hsiao (1965: 426) was described in Pseudophloeinae. Later, Hsiao (1977: 253)synonymized it with Chariesterus Laporte but placed the latter genus in Pseudophloeinae. Itscorrect position is in Coreinae: Chariesterini. Abbreviations of depositories Specimens mentioned in the text are held in a number of different depositories; the addresses of 20 of these have been abbreviated as follows. AMNH American Museum of Natural History , New York , U . S . A . BMNH British Museum (Natural History), London, U.K. BPBM Bernice P . Bishop Museum , Honolulu , Hawaii , U . S . A . CAS California Academy of Sciences , San Francisco , U . S . A . IAR Institute of Agricultural Research, Samaru, Nigeria IP Institut f iir Pflanzenschutzforschung , Eberswalde , D . D . R . IRSNB Institut Royal des Sciences Naturelles, Brussels, Belgium IZ Institute of Zoology , Academy of Sciences , Leningrad , U . S . S . R . IZPAN Institut Zoologiczny, Polska Akademia Nauk, Warsaw, Poland MNHN Museum National d'Histoire Naturelle, Paris, France MNHU Museum fur Naturkunde der Humboldt-Universitat , Berlin , D . D . R . MRAC Musee Royal de 1'Afrique Centrale, Tervuren, Belgium NMB National Museum, Bulawayo, Zimbabwe NMP Natal Museum, Pietermaritzburg, South Africa NMV Naturhistorisches Museum, Vienna, Austria NR Naturhistoriska Riksmuseet, Stockholm, Sweden TM Transvaal Museum, Pretoria, South Africa UG University of Ghana, Legon, Ghana UM University Museum, Oxford, U.K. ZMU Zoological Museum of the University, Helsinki , Finland Terminology and measurements Antennal and rostral segments are numbered I to IV starting with the segment attached to thebody. The term 'rostral' is used in preference to 'labial' because the base of the first segment ofthe labium is usually obscured; the first segment of the rostrum is arbitrarily taken to commenceat the base of the labrum, which is almost always visible. Measurement of antennal segmentsexcludes the narrow, unsculptured bases of segments I, II and III and the small ring-segmentbetween HI and IV. Ranges of measurements are given where this procedure seems useful (it isnot used in the case of ratios of lengths of rostral and antennal segments, where it would be toocumbersome and would be of minimal use in identifying species). Means were calculated duringthe preparation of this revision but were discarded: because of the unequal representation ofdifferent populations in the samples available it was felt that they would be unhelpful andpossibly misleading. Surprisingly often, means were found to fall almost exactly half-waybetween the extremes of the ranges cited. The reliability of the range of measurements of anyspecies as a guide to what might be encountered by the reader in examining his own material willdepend largely on the number of specimens examined by the author and the number of localitiesfrom which they were collected; this information is given under the heading 'Material examined'for each species. It should be borne in mind that the antennae of these insects are fragile and thatthe figures given for the ratios of the lengths of the antennal segments are in all likelihood basedon progressively fewer specimens as one proceeds towards the distal end of the antenna. The 156 W. R. DOLLING term 'posterior' as applied to angles and spines of the pronotum is avoided in favour of theunambiguous terms 'posterolateral' and 'prescutellar' ; the former usually project laterally oranterolaterally from the sides of the body while the latter are situated, if they are present at all,on the posterior margin of the pronotum close to the lateral angles of the scutellum and projectposteriorly. Detailed descriptions of the body sculpture, pubescence and colour are usuallygiven either under the description of the genus or under the description of one species of eachgenus that typifies the condition of these characters throughout the genus, and only deviationsfrom this pattern, if they occur, are given for the other species. In descriptions of sculpture, theterm 'granule' is applied to projections that are no higher than their width and 'tubercle' appliesto projections longer than this; on the femora, there is a continuous gradation from granulesthrough tubercles to small spines; all of these structures are probably derived from enlarged hairbases that, in the case of spines, have either lost the hair completely or have its insertiondisplaced from the apex of the setiferous tubercle. Key to the genera of Pseudophloeini found in the Old World tropics 1 Posterior femur without tubercle adjacent to base of trochanter. Appearance characteristic (Fig. 2) RISBECOCORIS(p. 156) - Posterior femur with tubercle adjacent to base of trochanter, rarely obsolete and insect then of general appearance of Fig. 1 2 Antennal segment II less than one-third as long as segment III ARENOCORIS(p. 191) - Antennal segment II more than half as long as segment III 3 3 Posterior margin of pronotum with a pair of spines projecting backwards over bases of clavi at rest (Figs 20, 25, 77) 4 - Posterior margin of pronotum smooth or at most with a few , low tubercles or granules 5 4 Head dorsally and pronotum laterally with long spines; antennal segment I strongly clavate, with many long spines and tubercles (Fig. 77) MEVANIDEA(p. 187) - Head, pronotum and antennae without spines PARAMYLA (p. 165) 5 Main pubescence of body and hemelytra of short , decumbent , scale-like hairs (Figs 97 , 98) 6 - Main pubescence of body and hemelytra of longer , erect or suberect , bristle-like hairs 8 6 Male with apical tooth of paramere upcurved, apex of paramere not filling posterior emargina- tion of genital capsule (Fig. 96). (Africa) MYLA(p. 193) - Male with parameres club-shaped, apical tooth short and not curved, apex of paramere filling posterior emargination of genital capsule (as in Fig. 21) . (Asia) 7 7 Pronotum with posterolateral angles strongly produced anterolaterally (Figs 63, 64). PSEUDOMYLA(p. 180) - Pronotum with posterolateral angles slightly prominent (Fig. 58) PUNGRA (p. 179) 8 Abdominal sternites III to VII with posterolateral angles right-angled or acute, not projecting as triangular teeth (maximum degree of serration as in Fig. 42) PSILOLOMIA (p. 168) - Abdominal sternites III to VII produced into triangular teeth, making outline of abdomen coarsely serrate (Figs 13,19) 9 Antevannal vein present in metathoracic wing. (Africa) 10 - Antevannal vein absent from metathoracic wing. (Asia) 11 10 Scutellum terminating apically in a small, elevated, white blob MEVANIOMORPHA (p. 185) - Scutellar apex pointed, neither elevated nor white NEOMEVANIOMORPHA (p. 183) 11 Pronotum (Fig. 12) coarsely granulate-tuberculate, posterolateral spines arising abruptly from posterolateral angles HOPLOLOMIA(p. 161) - Pronotum (Fig. 18) finely granulate-tuberculate, posterolateral angles tapering smoothly into posterolateral spines INDOLOMIA(p. 163) RISBECOCORIS Izzard Risbecocoris Izzard, 1949: 478-479. Type-species: Risbecocoris tomentosus Izzard, by original designation. Body rather elongate, about 2-5-3-0 times as long as broad, strongly depressed. Connexivum moderatelyexpanded. Head about as long as pronotum, strongly convex; eyes small, prominent; ocelli dorsally obscured bypads of tomentose pubescence; dorsum of head strongly granulate and, at level of base of antennifers, witha pair of prominent setiferous tubercles; each antennifer laterally with two or three similar tubercles and a TRIBE PSEUDOPHLOEINI 157 broad, weakly deflexed and ventrally incurved apical process. Antennal segment I varying in length fromabout four- fifths to almost equal to head width including eyes, with long, outstanding tubercles, cylindricalthrough most of its length but narrowed gradually towards base in proximal one-quarter; segments II andIII distinctly more slender than I, tuberculate, II shorter than I, III longest of all; IV shortest, slightlythicker than II or III, elongate fusiform, specialized sensory area occupying its apical two-thirds. Bucculaeoccupying about one-half of ventral midline of head. Rostrum at rest reaching to posterior margin ofmesosternum, segments I and II subequal, HI about half as long as I and IV about two-thirds as long as I.Posterior half of head with two rows of tubercles flanking rostrum, in line with bucculae. Pronotum shallowly declivent, granulate, posterior margin straight, prescutellar spines absent, postero-lateral angles weakly elevated, lateral margins almost straight, with five long, laterally directed spines.Scutellum about 1-2 times as long as its basal width, almost flat, apex pointed. Mesosternum deeply sulcatelongitudinally. Metasternum convex with fine, median, longitudinal groove. Metathoracic scent-glandperitreme with dorsal ridge modified into a short, spout-like structure, circular in outline and completelysurrounding orifice (Fig. 4). Corium with costal margin convex, apical margin weakly convex, all veins verystrongly prominent; membrane of hemelytron with venation reticulate, prominent. Metathoracic wing(Fig. 5) with subcosta free in apical two-thirds, antevannal vein absent. Femora and tibiae with rows ofprominent tubercles or granules; posterior femur without subapical spines but with one or two smalltubercles in this position and no apical series; base of femur adjacent to trochanter lacking a tubercle. Abdominal sternites III-VII with posterolateral angles produced into progressively longer spines,lateral margins of sternites and spines bearing prominent granules and tubercles. Abdominal spiraclessituated very close to lateral margins of sternites, prominent. Male genital capsule (Figs 9-11) broadlyemarginate posteriorly, emargination filled by apices of parameres. Phallotheca with a ventral sclerite,produced dorsolaterally, and two rather broad dorsal longitudinal sclerites. Conjunctive with mediandorsal and ventral lobes, distal and apical lobes various. Vesica short, not protected by any basal sclerites,wings of ejaculatory reservoir complex articulating with broad, distal dorsolateral sclerites. Spermathecalduct short, almost straight except for one sharp bend; bulb narrowly lunate. Anus of female directedventrally. Dense, off-white, tomentose pubescence present on head, pronotum, scutellum, clavus, corium,connexivum, underside, antennae up to and including base of segment IV and legs up to and includingbases of tibiae. Erect hairs of body and of first one or two antennal segments very long and curved,remaining parts of antennae and other appendages with shorter, straight, erect hairs. REMARKS. All species of this genus have a very distinctive appearance (Fig. 2) by virtue of their long, densepubescence and long, lateral pronotal spines. Reticulate venation of the forewing membrane is characteris-tic of ground-dwelling Coreidae. The distally free subcosta in the metathoracic wing is unique in the family.The features that give the insects of this genus their remarkable appearance have probably arisen inresponse to the demands of what is, for a coreid, an unusually arid habitat. DISTRIBUTION. All records come from a belt of semi-arid terrain stretching across Africa from Senegal toKenya and thence into the Indian Desert. Key to species 1 Larger, length of male 7-2 mm or more, of female 7-7 mm or more; antennal segment III about 1-3 times as long as segment I. (Sudan, Nigeria, Kenya, Chad) numidianus(p. 160) - Smaller, length of male 7-1 mm or less, of female 7-6 mm or less; antennal segment III between 1 1 and 1-2 times as long as segment 1 2 2 Posterior femur with low granules on dorsal surface. (Niger) airensis(p. 159) - Posterior femur with outstanding tubercles on dorsal surf ace 3 3 Spines of posterolateral angles of abdominal segments with weak sigmoid curvature, those of segment VII apically divergent. (Pakistan) quadrocephalus(p. 161) - Spines of posterolateral angles of abdominal segments with stronger sigmoid curvature , those of segment VII apically convergent 4 4 Larger,lengthofmale6-7-7-lmm,offemale7-l-7-6mm. (Senegal) tomentosus(p. 157) - Smaller, length of female 6-4-6-7 mm, male unknown. (India) delhiensis(p. 160) Risbecocoris tomentosus Izzard (Figs 5-11)Risbecocoris tomentosus Izzard, 1949: 479-480, pi. 7. Holotype cf , SENEGAL (BMNH) [examined]. 158 W. R. DOLLING Figs 2-11 Risbecocoris species. 2, 3, numidianus: (2) dorsal view; (3) apical view of conjunctiva andvesica. 4, quadrocephalus , left metapleuron and surrounding area, showing scent-gland peritreme. 5-11,tomentosus: (5) metathoracicwing; (6) dorsal view of conjunctiva and vesica; (7) apical view of same; (8)ventral view of same; (9) posterior view of male genital capsule with parameres; (10) dorsal view ofsame; (11) lateral view of same. TRIBE PSEUDOPHLOEINI 159 Length: <D", 6-7-7-1 mm; $,7-1-7-6 mm. Head quadrate, its dorsal spine-like tubercle shorter than diameter of eye; tubercles on lateral margin ofantennifer about 4 in number, conspicuous, 2-4 times as long as wide. Antenna with first segment about0-95 times as long as width of head including eyes; ratio of lengths of segments about 1-00:0-93: 1-11:0-56.Tubercles of segment I about twice as long as wide, those of segment II about 1-5 times as long as wide.Ratio of lengths of rostral segments about l-00:0-90:0-43:0-67. Apex of corium at rest reaching to basalquarter or basal third of laterotergite VI. Width across hemelytra at rest about 0-72 times abdominal width(excluding spines and tubercles) in male, 0-77 times in female; abdominal width about 0-38 times total bodylength. Posterolateral spines of abdominal sternites with strong sigmoid curvature, those of segment VIIapically slightly convergent. Tubercles on margins of abdominal sternites 1-50-2-25 times as long as wide,absent from segments I-IV, absent or single on segment V, one or two (rarely absent) on segment VI, oneor two on segment VII. All femora dorsally with two rows of short tubercles. Conjunctiva (Figs 6-8) with low, M-shaped dorsomedian lobe; large, membranous, paired distaldorsolateral lobes; globose, membranous apical dorsal lobes, small, membranous, digitiform apicalventral lobes, globose, membranous distal ventrolateral lobes and a small, median, apically bifidventromedian lobe. Distal ventrolateral lobes each supported by a slender sclerite along its distal wall, thissclerite projecting anteriad as a narrowly triangular appendage, probably representing the totallysclerotized mesal lobe of the bilobed distal ventrolateral lobe. Distal dorsolateral lobes each with a strong,heavily sclerotized appendage arising near its base, adjacent to the apex of the respective wing of theejaculatory reservoir complex, these sclerotized appendages not expanded apically. Overall coloration appearing pale grey-brown to the naked eye. Colour of integument mid-brown;corium between veins, tibiae except for basal and apical annuli and spines of head yellow; lateral spines ofpronotum yellow with apices black; apices of abdominal spines black. Forewing membrane slightly milkyhyaline, veins white with occasional brown markings. REMARKS. This species differs from R. numidianus in its smaller body size and less elongated antennalsegments III and IV, as well as in the form of the sclerotized distal appendages of the conjunctiva; the otherknown African species, R. airensis, has the tubercles of the femora reduced to granules only about as highas wide on the anterior two pairs of legs and obsolete on the posterior pair, whereas in tomentosus thesetubercles are distinctly longer than wide. DISTRIBUTION. Known only from the type-locality in Senegal. MATERIAL EXAMINED Senegal: 1 C? (holotype), Bambey, 26.ix.1942 (/. Risbec) (BMNH). Senegal: 5 C?, 7 $ (paratypes), Bambey, 26.ix.1942 [Izzard, 1949, gives dates as ix.1940, 26.xi.1943 and1946.] (BMNH). Risbecocoris airensis VilliersRisbecocoris airensis Villiers, 1950a: 323-324. Holotype $ , NIGER (MNHN) [examined]. Length: cf, unknown; 9 , 7-2 mm. Very similar to R. tomentosus. First antennal segment as long as head. Ratio of lengths of antennalsegments as 1-00:0-96: 1-18:0-68. Tubercles on lateral margin of antennifer short, hidden among thepubescence. Dorsal spine-like tubercle of head much shorter than diameter of eye. Lateral spines ofholotype all with apices broken (not short as stated in original description). Anterior and intermediatefemora with two dorsal rows of hispid granules about as high as wide; posterior femora with granulesobsolete. Apex of corium at rest just reaching base of laterotergite VI. Lateral margins of abdominalsternites I-V without long tubercles, VI with a tubercle on one side only, VII with a tubercle on both sides.Width of abdomen (excluding spines) 0-38 times body length; width across closed hemelytra 0-82 timesabdominal width. REMARKS. This species can be distinguished from others of the genus by the obsolete granulation of theposterior femora. The male is unknown; the holotype, stated to be of this sex in the original description, isin fact a female and it remains the only known specimen. DISTRIBUTION. Air Mountains. MATERIAL EXAMINEDNiger: 1 $ (holotype), Air Sud, Agadez, 525 m, l-5.viii.1947 (A. Villiers & L. Chopard) (MNHN). 160 W. R. DOLLING Risbecocorisnumidianussp. n. (Figs 2, 3) Length: cf, 7-2-8-0 mm; $,7-7-8-3 mm. Head quadrate, dorsal spine-like tubercles longer than diameter of eye; antennifers laterally with 4-6conspicuous tubercles up to 5 times as long as wide. Antennae with segment I 0-89-0-95 times as long aswidth of head including eyes; ratio of lengths of segments about 1-00:0-90: 1-34:0-72. Tubercles of antennalsegment I twice as long as wide or more, those of II up to twice as long as wide, those of III and base of IVabout as long as wide. Ratio of lengths of rostral segments about l-00:0-90:0-43:0-52. All femora dorsally with two rows of tubercles about 2-5 times as long as wide, tubercles of tibiae about1-5 times as long as wide. Apex of corium at rest reaching one-third or one-half of the way alonglaterotergite VI. Spines of posterolateral angles of abdominal segments with weak sigmoid curvature,apices of spines of segment VII weakly divergent. Numbers of tubercles present on lateral margins ofabdominal sternites (excluding those on the spines themselves) varying from none to three on segments IIIand IV, one to three on V and VI and three to five on VII. Conjunctiva (Fig. 3) differing from that of tomentosus in that apical dorsal and apical ventral lobes areabsent and sclerotized appendages of distal dorsolateral lobes are broadly expanded apically. Overall coloration dark grey-brown; integument generally dark brown; forewing membrane milkyhyaline, its veins white with some brown streaks; tibiae, except for basal and apical annuli, yellow-brown;corium and clavus between veins pale brown; tubercles of head, abdomen and appendages pale brown;lateral spines of pronotum and abdomen pale brown, black-tipped; antennal segment IV very darkred-brown. REMARKS. This is the largest species of the genus. It differs from all its relatives in the proportionatelylonger antennal segments III and IV and in the apically expanded distal sclerotized appendages of theconjunctiva; a distinctive species in an otherwise structurally rather uniform genus. DISTRIBUTION. Widespread in the eastern Sahel. MATERIAL EXAMINED Holotype cf , Sudan: Umm Berembeita [H51' N, 3040' E; spelt Umberumbeita on label], on ground,22.iii.1931 (F. G. S. Whitehead) (BMNH). Paratypes. Sudan: 1 $, Upper Nile Province, Steamer Falweil, swamps, 21.x. 1933 (A. D. Sherati)(BMNH); 2 $, U.N.P., Renk-Malakal, 2-^.i.l963 (R. Linnavuori) (BMNH and R. Linnavuori coll.); 1Cf , 1 $, U.N.P., nearMalakal, 5-20.U963 (R. Linnavuori); 1 cf , Blue Nile, IngessanaMts, 17-22. id. 1962(R. Linnavuori); 1 cf, Blue Nile, Umm Banein, 14. xi. 1962 (R. Linnavuori) (all R. Linnavuori coll.).Chad: 1 cf , Bas-Chari, near Fort-Lamy, Farcha, Forest, 8.viii.l963 (J. Pericart); 1 cf , Bas-Chari, nearDouggia, 13.viii.1963 (/. Pericart) (both R. Linnavuori coll.). Nigeria: 1 $, NW. State, Badeggi RRS,19.iii.1972 (/. T. Medler) (University of Ibadan). Kenya: 1 <j>, Northern Frontier District, Wajeir,24.U955, at light (/. Lansbury) (BMNH). Risbecocoris delhiensis (Bose) comb. n. Hoplolomia delhiensis Bose, 1946: 75-76. Holotype (sex unknown), INDIA: Delhi, ix.1937, at light (Kerr)(Institute of Agricultural Research, New Delhi) [not examined]. Length: cf , unknown; $ , 6-4-6-7 mm. Very similar to R. tomentosus. Dorsal spine of head much shorter than diameter of eye; antennifersbearing four tubercles up to 3 times as long as wide on their lateral margins. Head slightly wider than long.First segment of antennae distinctly shorter than head, about 0-85 times as long as head width includingeyes; ratio of lengths of antennal segments about 1-00:0-85:1-14:0-71. Tubercles of antennal segment I atmost twice as long as wide, of segment II about 1-5 times as long as wide, of segment III about as long aswide. Ratio of lengths of rostral segments about 1-00:0-87:0-54:0-71. Femora with two dorsal rows oftubercles about 1-5 times as long as wide, tibiae with granules about as high as wide. Abdominal sterniteswith posterolateral spines stout and with rather strong sigmoid curvature, those of segment VII slightlyconvergent. Abdominal margins between the spines (which themselves bear tubercles) with tubercles up to3 times as long as wide, borne singly on segments V-VII. Male unknown. REMARKS. In the original description, the body length is given as 7 mm. This species is the smallest of thegenus. The holotype, said to be in the Agricultural Research Institute at New Delhi, was not available forstudy; the paratype, allegedly deposited in the BMNH, was not found. TRIBE PSEUDOPHLOEINI 161 DISTRIBUTION. Eastern fringes of the Indian Desert. MATERIAL EXAMINEDIndia: 2 $, Rajasthan, Pilani, ix.1965 (5. C. Goel) (BMNH). Risbecocoris quadrocephalus Ahmad & Shadab(Fig. 4) Risbecocoris quadrocephalus Ahmad & Shadab, 1969: 151-155. Holotype $, PAKISTAN: Karachi, Malir,on grass near lucerne field, 7.12.1967 (Fareed Ahmed) (Karachi University Natural History Museum)[not examined]. Length: cT, 6-7 mm; $,7-2-7-3 mm. Similar to R. delhiensis but slightly larger and tubercles of body and appendages slightly longer. Lengthof antennal segment I divided by head width including eyes 0-88-0-92. Ratio of lengths of antennalsegments I:II:III about 1-00:0-89:1-17; length of segment IV divided by that of segment I 0-66 (type) or0-69 in female, 0-77 in male. Posterolateral spines of abdominal sternite VII slightly divergent apically. Afull description of the female is given in the original description. Conjunctiva of male similar to that of R.tomentosus in all respects except that the membranous lobe of the distal ventrolateral lobes is smaller andthe ventromedian lobe is somewhat larger. REMARKS. The differences between the Asian species of Risbecocoris are very slight. Substantially morematerial would be required for a definitive treatment of these forms. The orientation of the spines ofabdominal sternite VII can be seen in the figures given in the original descriptions of delhiensis andquadrocephalus and may be reliable characters for distinguishing these two species. The two describedAsian species are structurally very similar to tomentosus and the Arabian specimen mentioned below. DISTRIBUTION. Southern Pakistan. MATERIAL EXAMINED Pakistan: 1 d", 1 $, Sind, Mainiforest, grass, 7.xi.l975 (Ali Khan) [determined as quadrocephalus byAhmad and Khan] (BMNH). Risbecocoris sp. A single damaged specimen is mentioned here because of its biogeographical interest. It is a male from theisland of Kamaran in the People's Democratic Republic of the Yemen ('South Yemen'), collected by G. C.Champion and deposited in the BMNH. It is 6-5 mm long; the length of its first antennal segment is 0-90times the width of the head including the eyes and the second antennal segment is 0-87 times as long as thefirst; the remaining antennal segments are missing. Structurally, it is identical in almost every detail to R.tomentosus, including the detailed structure of the conjunctiva. The only differences apparent are the moreslender vesica, the slightly smaller size and the slightly paler colour. The presence of a tomentosus-like formon an island near the Arabian shore of the Red Sea suggests that similar forms were and perhaps still arepresent throughout the northern edge of the Afrotropical Region and it forms a link between theSenegalese and Indo-Pakistan species. HOPLOLOMIA StalHoplolomia Stal, 1873: 82, 84. Type-species: Hoplolomia scabricula Stal, by monotypy. Body oblong, slightly depressed, connexivum greatly expanded, insect strongly spinose-tuberculate andhispid. Head slightly longer than pronotum, tuberculate; eyes rather small. Antennifers strongly divergent,their outer apical processes directed obliquely forwards and downwards, slightly curved inwards apically.Antennal segment I clavate, tuberculate; segment III the longest, IV shortest, fusiform, specializedsensory area of IV occupying three-quarters of its length. Bucculae rather short, occupying somewhat lessthan one-third of ventral midline of head. Rostrum at rest reaching disc of metasternum; segments I and IIsubequal, IV shorter, III very short. Pronotum (Fig. 12) moderately or weakly declivent, tuberculate; posterolateral angles somewhatproduced, bearing spines; posterior margin almost straight, prescutellar spines absent. Scutellumequilateral, flat or weakly convex, apex slightly raised. Dorsal ridge of metathoracic peritreme reniform,prominent. Meso- and metasternum shallowly sulcate. Corium with costal margin shallowly concave, 162 W. R. DOLLING apical margin straight, apex not produced. Metathoracic wing (Fig. 17) without antevannal vein. Femoraand tibiae strongly granulate to tuberculate; posterior femur with small basal tubercle, distally (Fig. 14)with two major spines and an apical series of three or four. Abdominal sternites with lateral margins tuberculate, posterolateral angles spinously produced (Fig.13). Male genital capsule posteriorly emarginate, emargination filled by apices of parameres. Conjunctiva(Figs 15, 16) with ventral lobes denticulate. Spermatheca with bulb narrowly lunate, duct about twice aslong as bulb, simply looped. REMARKS. This small genus is characterized by the presence of a basal tubercle on the posterior femur inconjunction with the absence of an antevannal vein in the metathoracic wing and the strongly tuberculateand hispid body. Hoplolomia delhiensis Bose is transferred to Risbecocoris (p. 160). DISTRIBUTION. Oriental region. Key to species 1 Scutellum flat. (India) campbelli(p. 163) Scutellum distinctly convex. (South East Asia) scabricula(p. 162) Hoplolomia scabricula Stal (Figs 12-14)Hoplolomia scabricula Stal, 1873: 84. Holotype $, 'INDIA' (NR) [examined]. Length: cf, $,6-4 mm. Length of antennal segment I about 0-93 times as long as width of head including eyes ; ratio of lengths ofantennal segments about l-00:0-81:l-20:0-62 (female). Ratio of lengths of rostral segments aboutl-00:0-96:0-38:0-67. Pronotum (Fig. 12) moderately declivent, spines of posterolateral angles directed laterad and slightlyanteriad, width across apices of spines divided by width of head including eyes 2-17-2-54. Scutellum weaklybut distinctly convex. Conjunctiva with dorsomedian lobe deeply cleft, appearing almost as a pair of membranous dorsolaterallobes; distal dorsomedian lobe small, conical, membranous; distal dorsolateral lobes rather globular, eachwith a finger-like, membranous lobe arising at junction with dorsomedian lobe and a short, curved,sclerotized appendage arising opposite apex of wing of ejaculatory reservoir apparatus and curvingdownwards; apical ventral lobes membranous, globular; distal ventrolateral lobes each divided into aslender, finger-like, membranous lobe adjacent to apical ventral lobe, a large, spreading membranous lobeand two small, membranous but minutely denticulate ventral lobes. Sclerotized parts of conjunctivaconsisting of these denticles, appendages of distal dorsolateral lobes and cup-like sclerite protecting base ofvesica. Vesica moderately long, its apex obliquely truncate. Ejaculatory reservoir apparatus with wings butwithout straps. Head granulate, dorsally with some stout tubercles; antennae granulate, segment I also tuberculate.Pronotum punctate, granulate and tuberculate with a pair of large, somewhat irregular, whitish tuberculesnear midline between posterolateral angles. Scutellum granulate-punctate, margins tuberculate. Thoracicpleura granulate-punctate. Clavus and corium granulate and deeply punctate, costal margin of corium inbasal half tuberculate. Femora and tibiae granulate and tuberculate. Connexivum granulate-punctate.Abdominal sterna granulate, lateral margins tuberculate. Pubescence of body and appendages of moderate length, semidecumbent, with longer, suberect or erecthairs arising from larger tubercles of appendages and dorsal surface of body, hairs arising from thetubercles of head and pronotum particularly long and erect. Colour rufous brown. Anterior midline of pronotum, margins of clavi adjacent to scutellum, distal veinsof corium, abdominal spines, large areas of laterotergites IV and V and anterior and posterior margins ofVI and VII and apical two-thirds of posterior femora conspicuously darker. Bases of femora, first segmentof each tarsus, tibiae except for basal and apical annuli, most of clavus and basal half of corium and spots ondarker parts of femora stramineous. Hemelytral membrane fuscous hyaline, some groups of somewhatreticulate veins dirty white, veins of anal angle conspicuously white, the membrane surrounding them notinfuscate. REMARKS. The type-locality is cited only as 'India Orientalis'; it is assumed here that it was in South EastAsia rather than in the Indian subcontinent on the basis of the distribution of the other material seen. TRIBE PSEUDOPHLOEINI 163 DISTRIBUTION. South East Asia. Hsiao (1964: 252) records this species from Yunnan, China. MATERIAL EXAMINED South East Asia 'India orientalis': 1 $ (holotype) (Stevens) (NR). Burma: 1 cf , Bhamo, viii. 1885 (Fea, ex Distant coll.) (BMNH). Vietnam: 1 $ , Dalat, 6 km S. , 1400-1500m,9.vi.!961 (N. R. Spencer) (BPBM). Laos: 1 cf, 1 $, WapikhamthongProv.,KhongSedone, 18.ix.1965(native collector) and 3. viii. 1965 (Rondon) (BPBM). Hoplolomia campbellisp. n. (Figs 15-17) Length: cf, 5-7-5-8 mm; $,6-7 mm. Very similar to H. scabricula in size, form and colouring, including in the structure of the conjunctiva(Figs 15, 16) but body slightly more depressed and slightly narrower. Length of antennal segment I dividedby width of head including eyes 85-0 93 ; ratio of lengths of antennal segments about 1 -00 : 83 : 1 22 : 72 .Pronotum shallowly declivent, width across apices of posterolateral spines divided by width of headincluding eyes 2-03-2-11. Scutellum flat. REMARKS. Very similar to H. scabricula, this species is distinguished from it by the slightly moredorsoventrally flattened body, this difference being most apparent in the shallower slope of the pronotaldisc and the flat, not convex, scutellum. DISTRIBUTION. Southern India. MATERIAL EXAMINED Holotype cf , India: Chikaballapura, iii.1915 (T. V. Campbell) (BMNH).Paratypes. India: 1 cf, Chikaballapura (Campbell), ii.1918; 1 $, Chikaballapura ll.i.1918 (BMNH). i INDOLOMIA gen. n. Type-species: Indolomia conculata sp. n. Body robust, connexivum widely expanded. Head slightly longer than pronotum, with uniform, low granulation. Antennifers weakly divergent,outer apical processes triangular, apically curved inwards. Antennal segment I weakly clavate, slightly butdistinctly shorter than head; segment III slightly shorter than I, II slightly shorter than III, IV shortest.Bucculae short, occupying only one-quarter of length of ventral midline of head. Rostrum with segment Ireaching posteriorly almost to level of posterior margin of eye; segment II slightly shorter than I, III half aslong as II, IV two-thirds as long as II. Pronotum strongly declivent, its posterolateral angles strongly produced anterolaterally, posterolateralmargin straight, without prescutellar spines; mostly rather weakly and uniformly granulate, with a pair ofshort, oblique ridges near midline between posterolateral angles, anterior and anterolateral margins andanterior half of midline with some larger and more prominent granules. Scutellum equilateral, convex, itsapex slightly produced and swollen. Meso- and metasternum strongly sulcate longitudinally. Dorsal ridgeof metathoracic peritreme shortly reniform. Corium with costal margin weakly concave and apical marginweakly convex, apex not produced. Metathoracic wing without antevannal vein. Anterior and intermedi-ate femora each with a small subapical spine beneath; posterior femur with two or three major subapicalspines and an apical series of tubercles and with a prominent basal tubercle. Posterior coxae separated by aspace equal to about two-thirds of the width of a coxa. Abdominal sterna HI- VII with posterolateral angles triangularly produced, lateral margins with veryobsolete granulation. Male genital capsule posteriorly emarginate, emargination filled by apices ofparameres. Phallotheca with a broad ventral sclerite, produced laterally, and two thin, longitudinal dorsalsclerites. Phallotheca, conjunctiva and vesica subequal in length. Conjunctiva with sclerotized, ventrallydirected distal dorsolateral appendages and a pair of sclerites protecting coiled base of vesica, otherwisecompletely membranous; dorsomedian lobe low, distal dorsomedian lobe conical, apical ventral lobespaired, weakly developed, distal ventrolateral lobes trilobed. Ejaculatory reservoir complex with wingswell developed, straps absent. REMARKS. This genus resembles Hoplolomia in that it lacks an antevannal vein in the metathoracic wingand has a basal tubercle on the hind femur. Because of the shape of the pronotum, the short pubescenceand the absence of prominent tubercles from most of the body surface and appendages, it superficially 164 W. R. DOLLING 18 19 Figs 12-19 12-14, Hoplolomia scabricula: (12) dorsal view of pronotum; (13) ventral view of abdominalmargin; (14) dorsal view of apical half of posterior femur. 15-17, H. campbelli: (15) dorsal view ofconjunctiva and vesica; (16) ventral view of same; (17) metathoracic wing. 18, 19, Indolomia conculata:(18) dorsal view of pronotum; (19) ventral view of abdominal margin. resembles some My la species but the differences in the form of the male genitalia show that the two generaare not closely releated. DISTRIBUTION. A single species, in southern India. Indolomia conculata sp. n. (Figs 18, 19) Length: d", 7-8 mm; $ unknown. Head as long as its width including eyes. Length of antennal segment I divided by head width includingeyes 0-89. Ratio of lengths of antennal segments as l-00:0-89:0-94:0-83. Specialized sensory setae TRIBE PSEUDOPHLOEINI 165 occupying about nine-tenths of length of segment IV. Antennae without prominent granules or tubercles.Rostral segment 1 0-66 times as long as width of head including eyes; ratio of lengths of rostral segments asl-00:0-89:045:0-64. Pr.onotum (Fig. 18) with posterolateral angles terminating in short spines; width across apices of spines2-43 times width of head including eyes. Legs moderately long, posterior tibia 0-96 times as long asposterior femur. Left femur of holotype with three major spines, the second half as long as the third andtwice as long as the first, right femur with only two major spines, similar to second and third of left femur;an apical series of three spines present on both femora, one tubercle and two granules present between lastand penultimate spines, a few granules present on rest of apical half of ventral surface of both femora.Scutellum, clavus and corium strongly punctate and weakly granulate throughout. Abdominal sterna withposterolateral angles triangularly produced but not drawn out into spines (Fig. 19). Pubescence of body,antennal segment I and most of femora short, semidecumbent; that of rest of antennae, of tibiae and tarsiand parts of femora short or rather short , suberect ; front of head and some large granules of pronotum witha few longer, erect hairs. Colour various shades of brown. Head with midline and a U-shaped stripe adjacent to eyes andinterrupted by ocelli conspicuously paler. Pronotum with midline and posterior disc conspicuously darker,the two oblique ridges of posterior disc conspicuously paler. Femora piceous with stramineous spots.Tibiae stramineous with basal and apical annuli piceous. Tarsi piceous, first segment dorsally stramineous.Connexivum banded pale and dark brown, abdominal dorsum rufous brown, anteriorly and anterolateral-ly piceous. Thoracic sterna piceous. Abdominal sterna mottled pale brown and piceous, piceous markingstending to form five longitudinal stripes. Hemelytral membrane fuscous hyaline; veins darker brown,interrupted by milky spots; anal angle of membrane, including veins, milky. REMARKS. Only the holotype, which is complete but slightly crushed, is known. MATERIAL EXAMINEDHolotype cf , India: Anaimalai Hills, Cinchuna, 3500 ft [1067 m], v.1976 (T. Nathan) (BMNH). PARAM YLA Linnavuori stat. n. Myla (Paramyla) Linnavuori, 1971: 177-178. Type-species: Myla suspecta Schouteden, by originaldesignation. Body depressed, narrowly ovate, not conspicuously tuberculate or hispid, connexivum moderatelyexpanded. Head slightly shorter than pronotum, slightly longer than antennal segment I, dorsally granulate withtwo conspicuous, smooth, non-granulate lines arising between ocelli and diverging anteriorly, reachinghalf-way to antennifers, and a shorter, smooth line along outer margin of each ocellus. Antennifersdivergent, outer apical processes of antennifers short, porrect. Antennal segment I longest, weakly butrather abruptly clavate, II shortest, III next shortest; I-III densely granulate, granules of I larger; IV withspecialized sensory area occupying between two-thirds and four-fifths of its length. Bucculae occupyingabout one-quarter of ventral midline of head. Rostrum with segment I distinctly longest, II longer than IV,III shortest, apex of rostrum at rest reaching to disc of mesosternum. Pronotum shallowly declivent, posterolateral angles scarcely prominent, bearing a short, triangulartooth, lateral margins granulate, disc granulate to tuberculate, posterior margin weakly convex, prescutel-lar spines well developed. Scutellum flat, equilateral, apex slightly or strongly elevated. Mesosternumsulcate, metasternum convex with slight longitudinal groove in anterior one- third. Dorsal ridge ofmetathoracic peritreme reniform. Metathoracic wing without antevannal vein. All femora granulate,anterior two pairs with or without a subapical spine beneath, posterior femur with a moderatelywell-developed basal tubercle and with two or three major subapical spines with only granules betweenthem and an apical series of two to four tubercles. Posterior tibiae more than 0-90 times length of posteriorfemora. Abdominal sternites with their posterolateral angles not produced, their densely granulate lateralmargins forming a single, smooth curve. Genital capsule (Fig. 21) short, posteriorly emarginate, emargina-tion filled by apices of parameres (Fig. 22). REMARKS. Members of this genus are readily recognized among tropical Pseudophloeinae by their stronglydepressed and largely non-tuberculate body. They have more the appearance of some Palaearctic genera,but differ from these in the absence of an antevannal vein in the metathoracic wing. Linnavuori (1971)included Myla niokoensis Schouteden in his new subgenus Paramyla but the male genitalia of niokoensisare typical of the genus Myla and the resemblances between niokoensis and suspecta are confined to the W. R. DOLLING Figs 20-26 Paramyla species. 20, australis, dorsal view of body with hemelytra at rest and posterior femurand tibia. 21-26, suspecta: (21) posterior view of male genital capsule with parameres; (22) dorsomedialview of left paramere; (23) lateral view of phallotheca, conjunctiva and vesica; (24) ventral view of same;(25) dorsal view of head and pronotum; (26) dorsal view of posterior femur. superficial characters of a lack of abdominal spines, very short pronotal spines and very short pubescence.DISTRIBUTION. Mountainous regions of central Africa and South Africa. Key to species 1 Pronotum with prescutellar spines broad, triangular; hairs of pubescence longer than distance between adjacent hair insertions. (Mountains of central Africa) suspecta (p. 166) Pronotum with prescutellar spines slender; hairs of pubescence shorter than distance betweenadjacent insertions. (South Africa) australis(p. 167) Paramyla suspecta (Schouteden) comb. n.(Figs 21-26) Myla suspecta Schouteden, 1938: 294. Holotype cf , ZAIRE (MRAC) [examined].Myla (Paramyla) suspecta (Schouteden); Linnavuori, 1971: 177-178. Length: cf, 7-4-7-6 mm; $,7-7-8-0 mm. Antennal segment I with two different sizes of granulation, its length approximately equal (0-89-1-01times) to width of head including eyes; ratio of lengths of antennal segments about l-00:0-67:0-81:0-85; TRIBE PSEUDOPHLOEINI 167 specialized sensory area of segment IV occupying about four-fifths of its length. Ratio of lengths of rostralsegments about 1-00:0-80:0-45:0-70. Pronotum (Fig. 25) with disc granulate; lateral margins shallowly concave; posterolateral angles scarcelyelevated above general level of posterior disc, width across apices of blunt posterolateral spines 1-73-1-82times width of head including eyes; prescutellar angles with triangular spines. Scutellum with apex slightlyelevated. Anterior and intermediate femora without subapical spines or tubercles, posterior femur (Fig.26) with two major spines, the proximal one about half as long as the other, with only granules betweenthem, and an apical series of two or three tubercles. Corium with costal margin markedly convex. Male with phallotheca short, consisting of a single ventral sclerite extending laterally almost to meet tworather less well-sclerotized dorsal sclerites. Conjunctiva (Figs 23, 24) with dorsomedian lobe large, itslateral angles prominent, distal dorsomedian lobe low, scarcely developed, distal dorsolateral lobesfinger-like, wholly membranous and without sclerotized appendages, apical ventral lobes absent, distalventrolateral lobes each divided into three: a dorsal, wholly membranous part, a ventral, largelysclerotized part, denticulate along its anterior edge, and a narrow, denticulate lobe situated anterior andmedial to the two larger parts. Vesica stout, rather long, protected at base by a pair of sclerites. Ejaculatoryreservoir complex with wings long, L-shaped, straps absent. Head granulate throughout, granules larger along midline of tylus, gular region punctate. Antennalsegment I with small, densely packed granules and some larger, outstanding granules, segments II-III withvery small granules. Pronotal disc punctate, with scattered granules, anterior half with some largergranules or low tubercles, lateral margins with larger granules; scutellum weakly granulate-punctate;thoracic sterna and pleura punctate; femora and tibiae minutely granulate. Clavus and corium punctatethroughout, punctures of clavus larger, of apex of corium smaller than average size of those on disc ofcorium, veins of clavus and corium granulate. Abdominal sternites and laterotergites minutely granulate,laterotergites also minutely punctate. Insect clothed in short, semidecumbent pubescence, the hairs longer than the distances between theirinsertions; tubercles and larger granules of head and pronotum with longer, erect hairs. Colour largely mid-brown. Membrane of hemelytra infuscate between veins, veins brown, with short,paler interruptions. Antennal segment IV, femora, especially posterior pair, and mesosternum dark brownto black; tibiae, except for basal and apical annuli, paler brown. REMARKS. This species seems to be one of the tropical Pseudophloeinae most highly adapted for life on theground. The unusual form of the conjunctiva, lacking apical ventral lobes and sclerotized appendages tothe distal dorsolateral lobes and with very strongly developed distal ventrolateral lobes, is presumablycharacteristic for the genus, and sets it somewhat apart from other genera. DISTRIBUTION. Highlands of central Africa. MATERIAL EXAMINED Zaire: 1 cf (holotype), Ituri, Nioka, vii.1934 (/. V. Leroy) (MRAC). Zaire: 4 cf, 4 $, data as holotype except that one male and two females have fuller date: 7. vii.1934(probably all paratypes); 1 cf, Ituri, de Buba a Petro (Nizi), 23.iii.1929 (A. Collarf) (paratype); 1 $,Kibali-Ituri, Nioka, xii.1952 (J. Hecq) (MRAC). Malawi: 1 cf , Masuku Mts, 6000-7000 ft [1800-2100 m],vii.l896(A Whyte) (BMNH). Paramyla australis sp. n. (Fig. 20) Length: cf , unknown; $,7-5 mm. Antennae with segment I proportionately longer and segment IV much shorter than in P. suspecta, ratioof lengths of segments as l-00:0-71:0-87:0-65; granulation of segment I uniform, segment IV with junctionof specialized sensory area and area of normal pubescence oblique, specialized area occupying betweenthree-quarters and two-thirds of its length. Body form (Fig. 20) slightly narrower than that of suspecta;width of pronotum (Fig. 20) across apices of the blunt, triangular posterolateral spines 1-72 times width ofhead including eyes; prescutellar spines slender; posterolateral angles rather strongly elevated above levelof posterior disc; lateral margins strongly concave. Scutellum with apex strongly raised. Corium with costalmargin only weakly convex. Intermediate femur with a small subapical spine beneath; posterior femur(Fig. 20) with three major spines, of which the first is very small and the second about three-quarters thelength of the last, a single tubercle between these two biggest spines and an apical series of four tubercles. Sculpture of body, especially pronotum, rather more pronounced than in suspecta; pronotal disctuberculate. Pubescence of body and appendages very short, semidecumbent, white, the hairs distinctly 168 W. R. DOLLING shorter than the distances between their insertions; tarsi and apices of tibiae with longer, suberectpubescence but body without longer, suberect hairs. REMARKS. This species diverges less from the typical body form of the subfamily than does its congener; thebody is less flattened and there are more femoral spines. In contrast, the pubescence is abnormally short inthe new species. DISTRIBUTION. Known only from the type-locality in South Africa. MATERIAL EXAMINED Holotype , South Africa: Cape Province, Cape Peninsula, Red Hill, 30.U968 (/. A. & S. Slater, T.Schuh, M. H. Sweet) (J. A. Slater collection). PSILOLOMIA Breddin Psilolomia Breddin, 1909: 292. Type-species: Psilolomia brevitibialis Breddin, by monotypy.Neohoplolomia Distant, 1913: 284. Type-species: Neohoplolomia typica Distant, by monotypy. Syn. n.Psilocoris Hsiao, 1964: 251, 259. Type-species: Psilocoris clavipes Hsiao, by original designation. Syn. n. Body oblong, not depressed, connexivum rather narrowly to moderately expanded, body not conspicu-ously spinose, tuberculate or hairy. Head about as long as pronotum, dorsally granulate to rather weakly tuberculate. Antennifersdivergent, their outer apical processes short, broad, porrect. Antennae with segment I very weakly clavate,granulate, without conspicuously outstanding granules or tubercules; segments II and III never ofconspicuously unequal length, minutely granulate. Bucculae occupying about one-quarter of ventralmidline of head . Rostrum with segment I much the longest , II to IV subequal in length , apex of IV reachingto meso-metasternal suture. Pronotum moderately declivent, its posterolateral angles not or rather weakly elevated and not or ratherweakly produced laterally, posterolateral spines short, blunt or acute and arising abruptly from the angles,directed laterally; posterior margin straight, without humeral spines; disc and lateral margins with largegranules or tubercles. Scutellum flat or weakly convex, equilateral, anterior angles prominent, apexelevated as a minute, whitish blob. Mesosternum shallowly sulcate; metasternum almost flat to distinctlyconvex, anteriorly concavely emarginate. Metathoracic peritreme (Fig. 38) with dorsal ridge bilobed, theanterior lobe much the larger. Clavus and corium punctate throughout, corium less heavily so apically andwith apex not produced. Antevannal vein present in metathoracic wing (Fig. 39) but sometimes weak.Anterior and intermediate femora with zero to two small subapical spines beneath, posterior femur withthree major subapical spines, some smaller ones and an apical series; tubercle at base of posterior femursmall, sometimes obsolete. Posterior coxae not approximated. Abdominal sterna III- VII with posterolateral angles not or slightly produced (maximum development inP. brunneofusca, Fig. 42). Spermatheca with bulb lunate, duct short, not convoluted. Male genital capsuleposteriorly emarginate, emargination filled by apices of parameres. Conjunctiva with dorsomedian lobelarge, distal dorsomedian lobe smaller, both membranous; distal dorsolateral lobes slender, their posteriorfaces supported for much or, usually, all of their length by sclerites continuous with or articulating withwings of ejaculatory reservoir, which lacks straps; no sclerites protecting base of vesica; apical ventral lobeobsolete or transverse, sometimes weakly bilobed but never divided into two globular lobes; distalventrolateral lobes membranous; ventral wall of conjunctiva with a pair of posteriorly diverging, toothed,sclerotized strips, these weakly developed or absent in the smallest three species. Pronotum, scutellum and thoracic pleura granulate-punctate, veins of hemelytra with occasional smallgranules, femora, tibiae, abdominal sternites and laterotergites granulate. Pubescence colourless oramber, simple, mostly rather short, semidecumbent to suberect, some species with long, erect hairs onpronotum and dorsum of head; pubescence never scale-like or tomentose. REMARKS. This, one of the largest genera of the tribe, lacks any striking distinguishing features. Thetype-species of Psilolomia and Neohoplolomia are synonymized here. P. brevitibialis, the type-species ofPsilolomia, is unique in that its vesica is about twice as wide as those of the other species examined and thedistal dorsolateral lobes of the conjunctiva are sclerotized for only about half their length. It shares withonly P. parva both the absence of ventral, toothed, sclerotized strips from the conjunctiva and thepossession of an antennal segment II that is shorter than segment III; these are the only two species inwhich some of the dorsal granules of the head are developed into tubercles that are higher than wide. Theconjunctiva of P. amphrysia, the third small species, has very weakly developed ventral sclerotized stripsand thus forms a link between the other two and the large species of the genus. These differences do notseem of sufficient value to warrant the separation of P. brevitibialis or of P. brevitibialis plus P. parva from TRIBE PSEUDOPHLOEINI 169 Figs 27-35 Psilolomia species, outline dorsal view of right half of pronotum of 27, amphrysia; 28,nigeriensis; 29, lamottei; 30, brevitibialis; 31,parva; 32, vulgaris; 33, brunneofusca; 34, steeleae; 35, lata. the remaining species at the generic level. Linnavuori (1970) placed the two African species known to himin Mevaniomorpha, which is undoubtedly close to Psilolomia and might be regarded as a derivative of itwith more pronounced pronotal and abdominal spines. DISTRIBUTION. Afrotropical and Oriental regions; wholly continental, with no insular records. Key to species 1 Side of head above each antennifer with a large tubercle almost as large as an eye (Figs 36, 37) . (India) d/spar(p. 174) - Head without such tubercles 2 2 Posterolateral angles of pronotum bearing acute spines that are longer than their basal width (Figs 1,27, 30, 31, 33) ..... 3 - Posterolateral angles of pronotum with usually blunt spines no longer than their basal width (Figs 28, 29, 32, 34, 35) 7 170 W. R. DOLLING 3 Larger insects: length of male 7-6 mm or more; length of female 7-9 mm or more. Antennal segment I as long as or longer than width of head including eyes 4 Smaller insects: length of male 7-2 mm or less; length of female 7-6 mm or less. Antennalsegment I not as long as width of head including eyes 5 4 Coloration ochreous yellow. (India, Sri Lanka) pundaloyae (p. 177) Coloration ferruginous or fuscous brown. (S. China to Malaya) brunneofusca(p. Ill), clavipes(p. 178) 5 Antennal segment II longer than segment III. (Africa) amphrysia (p. 170) Antennal segment II shorter than segment III. (Asia) 6 6 Pronotum (Fig. 31) with lateral margins distinctly concave; abdominal sternites V and VI with posterolateral angles produced into short, acute spines. (Laos, Thailand) parva (p. 174) Pronotum (Fig. 30) with lateral margins almost straight; abdominal sternites V and VI withposterolateral angles right-angled. (India, Sri Lanka) brevitibialis (p. 172) 7 Pronotum with lateral margins (Figs 32, 34, 35) bearing coarser tubercles and granules and longer and straighter pubescence. (Asia) Pronotum with lateral margins (Figs 28, 29) bearing finer tubercles and shorter, more curvedpubescence. (Africa) 10 8 Pronotum with lateral margins (Fig. 35) strongly concave just anterior to posterolateral angles. (Thailand) Ma (p. 179) Pronotum with lateral margins (Figs 32, 34) weakly concave 9 9 Antennal segment I longer than width of head including eyes. General coloration ferruginous brown. (Burma) steeleae(p. 178) Antennal segment I shorter than width of head including eyes. General coloration ochreousyellow. (India, Sri Lanka) vulgaris(p. 175) 10 Antennal segment I not longer than width of head including eyes. Pronotum with lateral margins (Fig. 28) almost straight. (Nigeria, Central African Republic) nigeriensis (p. 171) Antennal segment I longer than width of head including eyes. Pronotum with lateral margins(Fig. 29) distinctly concave 11 11 Larger insects: length of male 9-9 mm. (Sudan) ferruginea(p. 172) Smaller insects: length of male 9-0 mm; length of female 9-7 mm. (Guinea, Nigeria) Iamottei(p. 171) Psilolomia amphrysia (Linnavuori) comb. n.(Figs 27, 49) Mevaniomorpha amphrysius Linnavuori, 1970: 43-45. Holotype C?, IVORY COAST: Lamto, Toumodi,1-5. vi. 1962 (Gillori) (R. Linnavuori coll.) [not examined]. Length: d", 6-6-7-2 mm; $,7-2-7-6 mm. Head dorsally granulate. Antennifers distinctly but not strongly divergent. Length of antennal segment Idivided by width of head including eyes 0-88-1-00. Ratio of lengths of antennal segments in male aboutl-00:0-78:0-73:0-82, in female about l-00:0-77:0-71:0-74. Ratio of lengths of rostral segments in maleabout l-00:0-65:0-50:0-73, in female about 1-00:0-58:0-52:0-61. Pronotum (Fig. 27) with lateral margins slightly concave, bearing small tubercles that are slightly longerthan wide; posterolateral spines acute, longer than their basal width; width across apices of spines dividedby width of head including eyes 1-80-2-07. Metasternum moderately convex. Anterior and intermediatefemora with or without one small, subapical spine beneath, posterior femur with three major spines, threeor four minor spines between the second and third and an apical series of four. Length of posterior tibiaabout 0-7 times length of posterior femur. Abdominal sternites III-VII with posterolateral angles acute, slightly prominent. Conjunctiva withdistal dorsolateral lobes (Fig. 49) short, sclerotized throughout their length; ventral sclerotized stripsweakly developed. Pubescence of most of body and of hemelytra short, curved, semidecumbent; that of antennae and legsstraighter, suberect; head also with some longer, suberect hairs arising from tubercles and larger granules;declivent part of pronotum also with pubescence of erect hairs about as long as diameter of an eye. Colour sordid yellow, apical half of posterior femur heavily marked piceous; scattered, small piceousmarkings of greater or less extent present elsewhere on body and appendages, especially abdominallaterotergites and sternites. Granules of appendages usually piceous. REMARKS. This is the smallest of the African species and the only one of them with long pronotalpubescence (about as long as diameter of an eye); its pronotal posterolateral spines are longer than those of TRIBE PSEUDOPHLOEINI 171 other African species. It differs from the two small Asian species in having the second antennal segmentlonger than the third, not shorter. DISTRIBUTION. West Africa and Zambia. MATERIAL EXAMINED Ivory Coast: 1 cf , Lamto, 3.iii.l967 (Gillon), 1 $, Lamto, 7-10.ii.1962 (Gillon) (R. Linnavuori coll;paratypes); 6 cf , Lamto, various dates 26.vi.1962 to 21.vii.1964 (Gillon) (BMNH). Nigeria: 1 cf , Samaru,Zaria, in dry plant detritus, 30.U955 (M.G.E.); 1 cf, 1 $, Ilorin, under dead leaf , 19.iii.1955 (A/.G.E.);3$ , Gombe, Matzoro Lakes, i.1929 (Lloyd) (BMNH). Zambia: 1 cf , Chisinga Plateau, Kalungwisi District,4500 ft (1350 m), 25.ix.1908 (Neave) (UM). Psilolomia nigeriensis sp. n. (Figs 28, 57) Length: cf , 7-9-9-3 mm; $,8-7-9-1 mm. Head dorsally granulate. Antennifers distinctly but not very abruptly divergent. Length of antennalsegment I divided by width of head including eyes 0-86-1-00. Ratio of lengths of antennal segments innorthern males about l-00:0-89:0-78:0-93; in southern males about l-00:0-93:0-80:0-82; in northernfemales about l-00:0-93:0-82:0-86; in southern females about l-00:0-93:0-82:0-82. Ratio of lengths ofrostral segments about 1.00:0-57:0-58:0-55. Pronotum (Fig. 28) with lateral margins very shallowly concave, almost straight, bearing small granules;posterolateral angles each with a short, triangular spine; width across apices of spines divided by width ofhead including eyes 1-94:2-02. Metasternum rather strongly convex. Anterior and intermediate femoraeach without or with one small, subapical spine beneath; posterior femur with three major spines, rarelywith a small spine preceding them, three or four minor spines between the penultimate and last majorspines and an apical series of four or five. Length of posterior tibia divided by length of posterior femur0-82-0-92. Abdominal sternites III-VII with posterolateral angles acute and slightly prominent. Conjunctiva withventral toothed sclerotized strips well-developed, distal dorsolateral lobes (Fig. 57) sclerotized throughouttheir length but less strongly so apically and ventrally, dorsomedian lobe bearing a narrow, membranous,apically bifid, median process; vesica longer than in any other species of the genus. Pubescence of most of body and of hemelytra short, curved, semidecumbent; of antennae and legssuberect, longer and more erect ventrally on femora; head with uniform, short, semidecumbent pubesc-ence; declivent part of pronotum with longer, less curved, suberect pubescence distinctly shorter thandiameter of eye. Colour sordid yellow- ochreous, usually with a slight ferruginous tinge and with piceous markings alongapical margin of corium, thoracic and abdominal sterna and laterotergites and apical half of posteriorfemur; all tibiae with basal and apical annuli fuscous. REMARKS. This species differs from the other African species in its almost straight lateral pronotal margins.It is very similar to the Asian P. vulgaris but lacks suberect pubescence along the midline of the head anddiffers in the shape of the sclerotized distal dorsolateral lobes of the conjunctiva. DISTRIBUTION. West Africa. MATERIAL EXAMINED Holotype cf , Nigeria: U.C. Ibadan, on leaf of Vigna unguiculata, 31.1.1955 (G. H. Caswelf) (BMNH). Paratypes. Nigeria: 1 cf , data as holotype (Ibadan University) ; 1 $, Enugu, 30.x. 1955 (Bechyne, Exped.Mus. G. Frey); 1 cf , 1 $ , Zaria, Samaru, at light, 12.vii.1966 (J. Deeming); 1 cf , 2 $ , Samaru, Zaria, in dryplant detritus, dry grass and plant bases, 30.i.l955 (M.G.E.) (BMNH). Central African Republic: 1 cf ,Bambari, u.v., 1964 (G. Pierrard) (MRAC). Psilolomia lamottei ( Villiers) comb. n. (Figs 29, 54)Acanthomia lamottei Villiers, 19506: 654-655. Holotype $, GUINEA (MNHN) [examined]. Length: cf,9-0mm; $,9-7 mm. Head dorsally granulate. Antennifers strongly divergent. Length of antennal segment I divided by widthof head including eyes about 1-1. Ratio of lengths of antennal segments about l-00:0-84:0-78:0-76. Ratio oflengths of rostral segments about 1 -00:0-58:0-61 :0-58. 172 W. R. DOLLING Pronotum (Fig. 29) with anterolateral margins concave, granulate; posterolateral angles prominent,bearing broadly triangular posterolateral spines; width across apices of spines divided by width of headincluding eyes 2-05 in male, 1-93 in female. Metasternum rather strongly convex. Anterior femur with one,minute, subapical spine beneath, intermediate femur with one small spine, posterior femur with threemajor spines (preceded by a small spine in holotype) with three minor spines between the penultimate andlast major spines and an apical series of four; tubercle at base of posterior femur obsolete. Length ofposterior tibia divided by length of posterior femur 0-84-0-85. Abdominal sternites V and VI with posterolateral angles slightly prominent, slightly more acute thanright-angled; those of VII angled at about 85, slightly prominent. Vesica long and slender; conjunctivawith ventral toothed strips well developed; distal dorsolateral lobes (Fig. 54) sclerotized throughout, notapically angled or expanded. Pubescence rather short, suberect to semidecumbent, of uniform length, no erect hairs present even ondorsal midline of head. Coloration of holotype from Guinea sordid yellow with slight rufous tinge, antennal segment IV,tubercles of all appendages, basal annulus of posterior tibia and, less conspicuously, of other tibiae, meso-and metasterna and extensive mottling on abdominal sterna, laterotergites and apical half of posteriorfemur piceous or black. Male from Nigeria generally ferruginous without black or piceous markings exceptfor those of abdomen and thoracic sterna. REMARKS. There is a close resemblance in structure, though not in colour, between the two specimensexamined. This species resembles P. ferruginea in having the first antennal segment longer than the widthof the head across the eyes and in the prominent posterolateral angles of the pronotum, but it is smaller andhas fewer minor spines on the posterior femur. The two species are evidently closely related. DISTRIBUTION. Uplands of West Africa. MATERIAL EXAMINED Guinea: 1 $ (holotype), Nimba, ii.vi.42 (M. Lamotte) (MNHN).Nigeria: 1 tf, Jos, 14.X.1955 (Bechyne, Exped. Mus. G. Frey) (BMNH). Psilolomia ferruginea (Linnavuori) comb. n. Mevaniomorpha ferruginea Linnavuori, 1970: 43 [in key]; 1978: 36-37 [full description]. Holotype cf,SUDAN (R. Linnavuori coll.) [examined]. Length: cf , 9-9 mm; $, unknown. Head dorsally granulate. Antennifers strongly divergent. Length of antennal segment I divided by widthof head including eyes about 1-2. Ratio of lengths of antennal segments as l-00:0-90:0-77:0-63. Ratio oflengths of rostral segments as 1 -00:0-71 :0-57:0-68. Lateral margins of pronotum very distinctly concave, bearing granules about as high as wide; postero-lateral angles prominent, bearing short, broad posterolateral spines. Metasternum rather strongly convex.Anterior and intermediate femora each with a strong subapical spine beneath; posterior femur with threemajor spines, four minor spines between the first pair, three or four between the middle and apical majorspines and an apical series of five. Abdominal sternites HI- VII with posterolateral angles more acute than right-angled, slightly promin-ent. Coloration generally ferruginous. REMARKS. This species is closely similar to P. lamottei. It is the largest species of the genus.DISTRIBUTION. Known only from the type-locality in the foothills of the Ethiopian massif. MATERIAL EXAMINEDSudan: 1 cf (holotype), Blue Nile, Ingessana Mts, 17-22. xi. 1962 (R. Linnavuori) (R. Linnavuori coll.). Psilolomia brevitibialis Breddin(Figs 30, 44, 50) Psilolomia brevitibialis Breddin, 1909: 293. LECTOTYPE $, SRI LANKA (IP), here designated [ex-amined] . Neohoplolomia typica Distant, 1913: 284-285. LECTOTYPE cf, INDIA (BMNH), here designated[examined]. Syn. n. TRIBE PSEUDOPHLOEINI Length: cf, 5-8-6-4 mm; $,6-4-7-3 mm. Head dorsally granulate, tuberculate throughout vertex and midline. Antennifers distinctly divergent.Length of antennal segment I divided by width of head including eyes 0-78-0-81. Ratio of lengths ofantennal segments about l-00:0-79:0-96:0-83. Ratio of lengths of rostral segments about 1-00:0-50:0-54:0-54. Head and pronotum together constituting less than 0-39 (male) or 0-35 (female) of total length. Pronotum (Fig. 30) with lateral margins straight, tuberculate; posterolateral angles not prominent;posterolateral spines short, acute; width across apices of spines divided by width of head including eyes1-82-1-92. Metasternum moderately convex. Anterior and intermediate femora each without or with onesmall subapical spine beneath; posterior femur with three major spines, the first about half as long as theothers, one to three minor spines between the two large ones and an apical series of four. Length ofposterior tibia divided by length of posterior femur about 0-74. Abdominal sternites III-VII with posterolateral angles right-angled, not prominent. Male conjunctiva(Fig. 44) with apical ventral lobes obsolete and ventral sclerotized strips absent; distal dorsolateral lobes(Fig. 50) short, supported by sclerotizations for only about half their length. Vesica unusually broad. Pubescence suberect, short; tubercles of head and pronotum bearing longer, erect hairs shorter thanthree-quarters of the diameter of an eye. Colour ochreous yellow, often sprinkled with minute red dots. Granules of antennae, femora andsometimes of tibiae piceous; antennal segment IV often piceous; anterior midline of pronotum, apical andbasal annuli of tibiae, apical halves of posterior femora and more or less extensive markings ofconnexivum, especially posterior halves of laterotergites IV- VI, of pleura, of thoracic and abdominalsterna and of corium, especially in two places on the apical margin, piceous. REMARKS. The short, broad vesica and only partially sclerotized distal dorsolateral conjunctival lobes ofthis species are unique for the genus. The lack of ventral sclerotized conjunctival strips, the small body sizeand the possession of an antennal segment II that is shorter than segment III are all shared with P. parva.The two species can be distinguished externally by the form of the posterolateral angles of the abdominalsternites, which are acute in parva but right-angled in brevitibialis . 36 37 Figs 36-43 Psilolomia species. 36, 37, dispar: (36) dorsal view of head; (37) lateral view of head. 38, 39,vulgaris: (38) left metathoracic scent-gland aperture; (39) metathoracic wing. 40-43, ventral view ofabdominal margin of (40) steeleae; (41) lata; (42) brunneofusca; (43) vulgaris. 174 W. R. DOLLING DISTRIBUTION. Southern India and Sri Lanka. MATERIAL EXAMINED Sri Lanka: 1 $ (lectotype of brevitibialis), Trincomalee (Horn) (IP). India: 1 cf (lectotype of typica),Chikkaballapura (T. V. Campbell) (BMNH). India: 26 cf , 31 $, data as holotype of N. typica (paralectotypes of typica) (BMNH); 1 $, Coimbatore,x.1953 (P. 5. Nathan) (IRSNB). Sri Lanka: 1 cf , Pundaloya (BMNH). Psilolomia parva sp. n. (Figs 31, 45, 51) Length: cf , 5-4-5-9 mm; $,5-9-6-5 mm. Head dorsally granulate, tuberculate on vertex and in midline. Antennifers distinctly but rather weaklydivergent. Length of antennal segment I divided by width of head including eyes about 0-86. Ratio oflengths of antennal segments about 1-00:0-73:0-89:0-81. Ratio of lengths of rostral segments aboutl-00:0-55:0-55:0-55. Head and pronotum together constituting more than 0-39 (male) or 0-35 (female) oftotal length. Pronotum with lateral margins (Fig. 31) weakly concave, posterolateral angles scarcely prominent;posterolateral spines acute, at least 1-5 times as long as their basal width; width of pronotum across apicesof spines divided by width of head including eyes 1-74-1-83. Metathoracic wing with antevannal vein weak,difficult to see. Metasternum rather strongly convex. Anterior and intermediate femora without or with asingle subapical spine beneath; posterior femur with three major spines, the first about half as long as theother two, one to three minor spines between the two large spines and an apical series of four. Length ofposterior tibia divided by length of posterior femur about 0-72. Abdominal sternites III- VII with posterolateral angles acute, slightly but distinctly prominent. Maleconjunctiva (Fig. 45) without ventral sclerotized strips, distal dorsolateral lobes (Fig. 51) short, supportedby sclerotization throughout their length; vesica slender. Pubescence short, semidecumbent on hemelytra and most of body, slightly longer and less decumbenton legs and antennae, tubercles of head and pronotum bearing longer, almost erect hairs about as long astwo-thirds diameter of an eye. Colour ochreous yellow; ground colour of clavus, corium and posterior half of pronotum pale cinereous,without yellow tinge. Grey or piceous markings of greater or lesser extent present on midline and posteriorborder of pronotum, on scutellum, on hemelytra, especially costal and apical areas of corium; on anteriorand posterior borders of laterotergites, including more than posterior halves of laterotergites IV and V; onapical halves of femora, especially posterior pair; and on thoracic pleura and abdominal sterna. Mesoster-num and metasternum entirely piceous. Tibiae with basal, apical and median annuli grey. REMARKS. This is the smallest species of the genus. It differs from the African P. amphrysia in having thesecond antennal segment shorter than the third and from the Asian P. brevitibialis in having theposterolateral angles of the abdominal sternites acute. The antevannal vein is very indistinct, probablybecause it is functionally obsolete in such a small wing. DISTRIBUTION. South-east Asia. MATERIAL EXAMINED Holotype cf , Laos: Wapikhamthong Prov., Khong Sedone, 15.viii.1965 (native collector) (BPBM). Paratypes. Laos: 50 cf , 49 $, Wapikhamthong Prov., Khong Sedone, various dates from 30. iv. 1965 to5.xi.l965 (native collector); 13 Cf, 17 $, Khong Sedone, 3.viii.l965 (as 3.8.65) (Rondon) (BPBM;BMNH). Thailand: 2 cf, Chiangmai Prov., Chiangmai (Zoo), 16.vi.1965 (P. D. Ashlock); 3 cf,l $,Chiangmai Prov., Chiangmai (Arboretum), 16.vi.1965 (P. D. Ashlock) (BPBM). Psilolomia dispar (Walker) comb. n.(Figs 36, 37, 48) Clavigralla dispar Walker, 1872: 6. Holotype cf (no locality cited) (BMNH) [examined].[Clavigralla aliena (Walker); Distant, 1901: 426. Misidentification.] Length: cf, 8-4-9-0 mm; $,8-9-9-9 mm. Head (Figs 36, 37) dorsally granulate, with antennifers very short, slightly divergent, outer apicalprocess obsolete; a large, granular tubercle, almost as large as the eye, present on side of head above eachantennifer. Length of antennal segment I divided by width of head including eyes 0-81-0-90. Ratio of TRIBE PSEUDOPHLOEINI 175 lengths of antennal segments in male about l-00:l-01:0-88:0-87, in female about l-00:0-99:0-84:0-92. Ratioof lengths of rostral segments about 1-00:0-63:0-55:0-61. Pronotum with lateral margins shallowly concave, granulate, posterolateral angles elevated but onlyweakly produced laterally, posterolateral spines acute, about 1-5 times as long as their basal width; width ofpronotum across apices of spines divided by width of head including eyes 1-96-2-04. Metasternum weaklyconvex. Anterior femur with one small subapical pine beneath, intermediate femur with one small spineand one very small one, posterior femur with three major spines, the first pair with no minor spinesbetween them, the second and last with three or four minor spines between them and with an apical seriesof four; holotype and some other specimens with a small spine preceding the three major spines. Abdominal sternites IH-VII with posterolateral angles slightly more acute than right-angled and slightlyprominent. Conjunctiva (Fig. 48) with ventral, toothed sclerotized strips well developed, distal dorso-lateral lobes sclerotized throughout, L-shaped. Coloration generally ferruginous; dorsal midline of head with distinct, narrow, yellow stripe; antennalsegments III-VII yellow with granulation piceous; segment IV ferruginous; pronotum with anteromedianmacula and posterolateral spines brown; legs yellow but extensively covered with brown granules, tibiaewith basal, apical and median annuli brown, femora extensively mottled brown or ferruginous, especiallyapically; thoracic sterna brown; abdominal sterna yellow with brown markings largely confined to twolongitudinal bands; laterotergites yellow with brown markings medially and posteriorly; clavus and coriumferruginous, the latter with a few obscure spots on veins brown, membrane fuscous, veins spotted darker,apex of corium and membrane throughout with large, cream spots. Pubescence rather short, semidecumbent to suberect; erect hairs about as long as two-thirds diameter ofeye present rather densely on declivent part of pronotum and on ventral surfaces of femora; mainpubescence of head consisting of short, almost decumbent hairs but larger granules, especially of dorsalmidline, bearing short, suberect to erect hairs. REMARKS. This species is immediately recognizable by the presence of the large tubercles above theantnnifers; these could be mistaken for the antennifers themselves in dorsal view. The first antennalsegment is unusually short for one of the larger species of the genus. Walker's original description refers tothree longitudinal brown bands on the pronotum; the lateral ones are, in fact, shadows in the shallowtroughs between the gently convex pronotal disc and the elevated posterolateral angles. Distant (1901:426) redescribed the type-specimen of Clavigralla dispar Walker under the impression that it was the typeof Cletus alienus Walker. The latter, which is now missing, was a female from New Guinea whereas the typeof dispar was specified by Walker as a male without locality data donated by Saunders. .DISTRIBUTION. India. MATERIAL EXAMINED 1 cf (holotype), no data (Saunders) (BMNH). India: 1 $ , Pondicherry State, Karikal, vii.1963 (P. S. Nathan) (BMNH); 2 cf , 1 $ 'E. Ind.' (Bacon); 1Cf , 'Bgl' (Haw.); 1 $, no locality (Haw.) (all UM); 1 cf , Pondicherry State, Karikal, l.i.1959 (Nathan); 1Cf, Karikal Terr., Kurumbaragum, viii.1953 (Nathan); 6 cf , 2 $ , Tranquebar, vii.1953 (Nathan) (J. A.Slater coll.). Psilolomia vulgaris sp. n. (Figs 32, 38, 39, 43, 55) Length: cf, 7-9-8-8 mm; $, 8-7-9-7 mm. Head dorsally granulate. Antennifers abruptly divergent. Length of antennal segment I divided by widthof head including eyes in male about 0-89, in female about 0-95. Ratio of lengths of antennal segments inmale about l-00:0-89:0-75:0-89, in female about l-00:0-96:0-81:0-93. Ratio of lengths of rostral segmentsabout l-00:0-56:0-56:0-56. Pronotum (Fig. 32) with lateral margins very shallowly concave, bearing large, conical tubercles about ashigh as wide; posterolateral spines short, blunt, not longer than wide; width across apices of spines dividedby width of head including eyes 1-98-2-11. Metasternum moderately convex. Anterior and intermediatefemora each without or with a single, small subapical spine beneath; posterior femur with three majorspines, rarely preceded by an additional small spine, the first pair of large spines without or with a singleminor spine between them, three or four minor spines between the middle and distal spines and an apicalseries of four. Length of posterior tibia divided by length of posterior femur 0-75-0-79. Metathoracic wing(Fig. 38) and peritreme (Fig. 39) typical of the genus. Abdominal sternites III-VII with posterolateral angles slightly more acute than right-angled, very 176 44 W. R. DOLLING 55 Figs 44-57 Psilolomia species. 44-46, dorsal view of conjunctiva and vesica of (44) brevitibialis; (45)parva; (46) brunneofusca. 47, 48, ventral view of conjunctiva and vesica of (47) brunneofusca; (48)dispar. 49-57, apical view of vesica, ejaculatory reservoir apparatus and distal dorsolateral lobes ofconjunctiva of (49) amphrysius; (50) brevitibialis; (51) parva; (52) brunneofusca; (53) /ate; (54) lamottei;(55) vulgaris; (56) pundaloyae; (57) nigeriensis. weakly prominent (Fig. 43). Male conjunctiva with ventral sclerotized strips rather weakly developed,distal dorsolateral lobes fully sclerotized, strongly angled before their spatulately expanded apices (Fig. Antennal segments I-III with suberect pubescence shorter than width of segment; rostrum with short,suberect and decumbent pubescence; head with short, semidecumbent hairs and dorsally with longer,suberect hairs not confined to midline; pronotum with short, semidecumbent pubescence posteriorly andsuberect pubescence on disc and anteriorly, interspersed except posteriorly with rather longer, erect hairs;legs with short to moderately long, suberect pubescence, femora with longer, erect hairs ventrally; clavusand corium with short, semidecumbent pubescence; thoracic pleura and sterna and abdominal sterna andlaterotergites with slightly longer, semidecumbent pubescence. Coloration generally pale yellowish brown; ocellar tubercles and granules of antennae, rostrum and legsbrown. Pronotum with anteromedian macula, lateral granules and posterolateral spines in part brown;scattered markings on pleura, scutellum and corium, especially apical areas of corium, brown; thoracicsterna piceous; femora sparsely mottled with brown, apical half of posterior femur, especially on posteriorface, reddish brown. Abdominal laterotergites and bands along abdominal sterna mottled dark brown.Membrane of hemelytra very palely infuscate, veins paler with interrupted brown streaks. TRIBE PSEUDOPHLOEINI 177 REMARKS. This species closely resembles P. pundaloyae and the African P. nigeriensis; it differs from theformer in the shorter pronotal posterolateral spines, from the latter in the longer pubescence of the headand from both in the form of the distal dorsolateral lobes of the conjunctiva. The sexual dimorphism in therelative length of antennal segment I is unusual; the shortness of this segment in the female results in thelength of segment IV divided by that of segment I being greater in the female than in the male, the reverseof the usual situation in the subfamily, where segment IV is relatively longer in the male. DISTRIBUTION. Southern India, Sri Lanka. MATERIAL EXAMINED Holotype cf , India: Madras, Karikal, Kurumbagarum, xi.1953 (P. S. Nathan) (J. A. Slater coll.). Paratypes. India: 8 Cf, 1 $, Karikal Territory, Kurumbagarum, ix.1954 (Nathan) (IRSNB); 1 cf,Karikal, v.1957 (Nathan); 1 cf, Tranquebar, viii.1956 (Nathan) (BPBM); 1 cf, 2 $, Pondicherry orKarikal, vii.1953 (Nathan) (BMNH); 9 specimens, Tranquebar, vii.1953 (Nathan); 36 specimens, Kurum-bagarum, various dates vii-xi.1953 and viii.1954 (Nathan) (J. A. Slater coll.). Psilolomiapundaloyaesp. n. (Figs 1,56) Length: cf , 8-2-8-8 mm; $,9-1 mm. Head dorsally granulate. Antennifers rather strongly divergent. Length of antennal segment I divided bywidth of head including eyes 1-00-1-05. Ratio of lengths of antennal segments about 1-00:0-92:0-75:0-80(IV missing in female). Ratio of lengths of rostral segments about l-00:0-58:0-49:0-56. Pronotum (Fig. 1) with lateral margins shallowly concave, bearing short tubercles slightly longer thantheir basal width; posterolateral spines longer than wide; width across apices of spines divided by width ofhead including eyes 2-00-2-12. Metasternum weakly convex, its anterior emargination extending post-eriorly as a shallow sulcus for over half its length. Anterior femur with a minute subapical spine beneath,intermediate femur with none, posterior femur with three major subapical spines, one minor spinebetween the first two, three between second and third and an apical series of four. Length of posterior tibiadivided by that of posterior femur about 0-82. Abdominal sternites (Fig. 1) with posterolateral angles distinctly acute and slightly prominent. Maleconjunctiva with ventral, toothed, sclerotized strips well developed; distal dorsolateral lobes (Fig. 56)wholly sclerotized, not angled before the spatulate apex. Pubescence rather short, suberect, tending to semidecumbent on ventral surface of body; longer, erecthairs throughout declivent part of pronotum and dorsal surface of head. Longest hairs of head andpronotum more than two-thirds as long as diameter of an eye. Coloration pale yellowish brown with darker markings, as in P. vulgaris. REMARKS. This species closely resembles P. vulgaris but differs in the longer and more slender postero-lateral pronotal spines and the longer pubescence. DISTRIBUTION. India, Sri Lanka. MATERIAL EXAMINED Holotype $, Sri Lanka: Pundaloya (Green) (BMNH). Paratypes. India: 1 cf, Chikkaballapura, i.1915 (Campbell); 1 cf, Kodai Kanal, v.1918 (Campbell)(BMNH). Psilolomia brunneofusca sp. n. (Figs 33, 42, 46, 47, 52) Length: cf , 7-6-8-8 mm; $,7-9-8-5 mm. Head dorsally granulate. Antennifers abruptly divergent. Length of antennal segment I divided by widthof head including eyes 1-15-1-20. Ratio of lengths of antennal segments in male about 1 -00:0-81 :0-68:0-74,in female about l-00:0-83:0-77:0-73. Ratio of lengths of rostral segments about 1-00:0-62:0-56:0-60. Lateral margins of pronotum (Fig. 33) concave, bearing granules about as high as wide; posterolateralspines acute, longer than their basal width and slightly curved posteriad; width across apices of spinesdivided by width of head including eyes 2-11-2-14. Metasternum weakly convex. Anterior femur with asmall, subapical spine beneath, sometimes with an incipient second one proximal to it; intermediate femurwith two spines; posterior femur with three major spines and usually a fourth, smaller one proximal to theothers, one to three minor spines between the first pair of large spines, three to four between the 178 W. R. DOLLING penultimate and last and an apical series of four or five. Length of posterior tibia divided by that ofposterior femur 0-78-0-80. Abdominal sternites III-VII (Fig. 42) with posterolateral angles acute and prominent. Male conjunctiva(Figs 46, 47) with ventral sclerotized strips well developed and distal dorsolateral lobes (Fig. 52) fullysclerotized, with slightly expanded apices. Pubescence short, semidecumbent; rather longer and suberect on legs, antennae, declivent part ofpronotum and dorsum of head ; length of hairs of suberect pubescence equalling half to two-thirds diameterof an eye. General colour dark ochreous brown to almost black. Antennae and legs yellow with all granulationdark brown, antennal segment IV strongly infuscate, apical halves of anterior and intermediate tibiaemottled with dark brown, apical half of posterior femur heavily mottled dark brown. Head, pronotum andscutellum ochreous with dark brown markings, the raised, white apex of the scutellum (universal in thegenus) strongly contrasting. Thoracic pleura and abdominal sterna and laterotergites reddish yellow withmore or less extensive dark brown markings; thoracic sterna piceous or black. Clavus and corium darkochreous or grey-brown, main veins of corium spotted both darker and paler; membrane darkly infuscate,veins brown; membrane and apex of corium with more or less conspicuous whitish spots, those ofmembrane often interrupting darker colour of veins. REMARKS. This is a widespread, dark brown species similar in build to the much paler, yellowish P. vulgaris.It may be synonymous with P. davipes, from southern China, but specimens of the latter were not availablefor study. It may be that the abdominal sternites of davipes lack the acutely produced posterolateral anglesthat are more developed in P. brunneofusca than in any other species of the genus. The Indian specimenexamined differed from the south-east Asian specimens only in having slightly broader posterolateralpronotal spines. The specimen from peninsular Malaya reported below represents the most south-easterlyrecord of any species of the tribe; only some Clavigrallini represent the subfamily in the Malay Archipelagoand northern Australia. DISTRIBUTION. South-eastern India, Laos, Malaya. MATERIAL EXAMINED Holotype cf , Laos: Pakse, iii.1965 (/. A. Rondori) (BPBM). Paratypes. Laos: 1 $, Wapikhamthong Prov., Khong Sedone, 18. ix. 1965 (native collector) (BMNH); 1Cf, Borikhame Prov., Pakkading, 100-200 m, 23. iv. 1965 (J. L. Gressitt); 1 cf, Vientiane Prov., ThaNgone, 3.i.l965 (native collector) (BPBM); 1 cf , Luang Prabang, Pak Leung, 5.iii.l920 (R. V. de Salvaza)(BMNH). India: 1 cf, Anamalai Hills, Cinchona, 3500 ft [1070 m], v.1957 (P. S. Nathan) (BMNH).Malaya: 1 cf , Perlis, Kaki Bukit, 27.V.74 (P. Roche) (BMNH). Psilolomia davipes (Hsiao) comb. n. Psilocoris davipes Hsiao, 1964: 251-252; 259. Holotype $ , CHINA: Yunnan, 1250 m, 17.vi.1956 (AcademiaSinica, Beijing) [not examined]. Length: cf , unknown; $,8-7 mm. Length of antennal segment I about 1 15 times width of head including eyes. Ratio of lengths of antennalsegments as 1-00:0-87:0-80:0-80. Ratio of lengths of rostral segments as 1-00:0-69:0-54:0-61. Pronotumwith posterolateral angles not produced; posterolateral spines longer than their basal width. General coloration 'dark ochraceus'. REMARKS. The partial description given here is based on that of Hsiao (1964: 251-252, 259). No materialwas available for study. A photograph of this species is reproduced by Hsiao (1977: pi. 46, fig. 603). It looksvery similar to P. brunneofusca and the two species may be synonymous. DISTRIBUTION. Southern China (Yunnan). Psilolomia steeleae sp. n. (Figs 34, 40) Length: cf , unknown; $ , 8-3 mm. Head dorsally granulate. Antennifers weakly divergent. Length of antennal segmentl divided by widthof head including eyes 1-03. Ratio of lengths of antennal segments as l-00:0-86:0-76:0-79. Ratio of lengthsof rostral segments as l-00:0-64:0-54:0-68. Pronotum (Fig. 34) with lateral margin distinctly concave, bearing tubercles slightly longer than their TRIBE PSEUDOPHLOEINI 179 basal width; posterolateral spines short, broadly triangular; width across apices of spines divided by widthof head including eyes 1 -91 . Metasternum almost flat. Anterior and intermediate femora each with a small,subapical spine beneath; posterior femur with three major spines, the first half as long as the other two, nominor spines between this and the middle spine, two minor spines between the two large spines and anapical series of three. Posterior tibiae rather long, 0-88 times as long as posterior femur. Abdominal sternites III-VII with posterolateral angles (Fig. 40) slightly more acute than right-angled,weakly prominent. Pubescence short, semidecumbent; rather longer and suberect on legs, antennae, declivent part ofpronotum and dorsum of head; length of hairs of suberect pubescence equalling half to two-thirds diameterof an eye. Coloration paler than in P. brunneofusca but not as pale as in P. vulgaris; darker markings distributed asin vulgaris; membrane of hemelytra fuscous with traces of pale spotting. REMARKS. This new species differs from P. vulgaris in the longer first antennal segment and darker colour,and from P. brunneofusca in its shorter pronotal and abdominal spines. DISTRIBUTION. Known only from the type-locality in Burma. MATERIAL EXAMINEDHolotype $, Burma: Mishmi Hills, Lohit River, 24.iii.1935 (M. Steele) (BMNH). Psilolomia lata sp. n. (Figs 35, 41, 53) Length: d", 7-8-8-5 mm; $,8-5-9-3 mm. Head dorsally granulate. Antennifers rather strongly and often abruptly divergent. Length of antennalsegment I divided by width of head including eyes 1-03-1 16. Ratio of lengths of antennal segments in maleabout l-00:0-90:0-76:0-79, in female about 1-00:0-90:0-77:0-70. Ratio of lengths of rostral segments about1-00:0-67:0-53:0-60. Pronotum (Fig. 35) with lateral margins deeply concave, bearing tubercles slightly longer than theirbasal width; posterolateral spines short, triangular; width across apices of spines divided by width of headincluding eyes 1-96-2-09. Metasternum flat, midline broadly and very shallowly concave throughout.Anterior and intermediate femora each bearing a single subapical spine, rarely a second spine present onintermediate femur; posterior femur with three, rarely four, major spines, penultimate spine preceded by asingle minor spine or not, succeeded by three or four minor spines; terminal series consisting of four, rarelyfive, spines. Posterior tibia rather long, its length divided by that of posterior femur 0-86-0-87. Abdominal sternites III-VII with posterolateral angles (Fig. 41) slightly more acute than right-angled,not or very slightly prominent. Conjunctiva with ventral sclerotized strips well developed; distal dorso-lateral lobes (Fig. 53) supported by sclerotization right up to their apices, which are not expanded. Pubescence short, semidecumbent; rather longer and suberect on legs, antennae, declivent part ofpronotum and dorsal midline of head; length of hairs of suberect pubescence equalling half to two-thirdsdiameter of an eye. General coloration ferruginous, dark markings distributed as in P. vulgaris; membrane of hemelytraslightly infuscate, veins brown, interrupted by paler stretches. REMARKS. This broad-bodied, ferruginous species may be distinguished from all others of the genus by thevery deeply concave pronotal lateral margins. DISTRIBUTION. Thailand. MATERIAL EXAMINED Holotype C?, Thailand: Chiangmai Prov., Doi Pui, 17.vi.1965 (P. D. Ashlock) (BPBM). Paratypes. Thailand: 4 cf, data as holotype; 2 d", 2 $, Chiangmai, Doi Suthep, 1300 m, 8.vi.l965(Ashlock) (BPBM; BMNH). PUNGRA gen. n. Type-species: Pungra angusta sp. n. Body form rather narrowly oblong, slightly depressed, connexivum slightly widened. Head longer than pronotum. Antennifers moderately divergent, outer apical processes porrect, broadlytriangular. Antennal segment I about as long as head width across antennifers, abruptly but weakly 180 W. R. DOLLING clavate, slightly longer than II, distinctly shorter than III, IV equal to I in male, shorter in female. Bucculaeoccupying about one-quarter of length of ventral midline of head. Rostrum at rest reaching to anterior discof metasternum, segment II subequal to I, IV much shorter, III shortest. Pronotum gently declivent, much broader than long, with a pair of small, raised calli on posterior discbetween posterolateral angles, posterolateral angles weakly produced, posterolateral spines small, lateralmargins with four stout tubercles each, posterior margin weakly convex, prescutellar spines absent.Scutellum almost flat, equilateral, its apex scarcely raised. Mesosternum and metasternum deeply sulcatethroughout. Dorsal ridge of metathoracic peritreme reniform. Metathoracic wing with antevannal vein.Anterior and intermediate femora without spines or tubercles, posterior femur with basal tubercle welldeveloped, with two major subapical spines preceded by granules and with three or four spine-liketubercles between them and an apical series of four tubercles. Abdominal sternites with posterolateral angles only slightly prominent, lateral margins not tuberculate.Male genital capsule emarginate, emargination filled by apices of parameres. REMARKS. This rather undistinguished genus is probably close to Pseudomyla, which it resembles in thepresence of stout tubercles on the lateral margins of the pronotum, but differs from it in the shape of theposterolateral pronotal angles. DISTRIBUTION. India. Pungra angusta sp. n. (Fig. 58) Length: cf, 7-3-7-7 mm; $,8-0-8-4 mm. Length of antennal segment I 0-76-0-89 times width of head across eyes; ratio of lengths of segmentsabout 1-00:0-96:1-28:1-00 (male) or l-00:0-98:l-32:0-85 (female); specialized sensory area of segment IVoccupying about five-sixths of its length. Ratio of lengths of rostral segments about l-00:0-92:0-30:0-56. Pronotum (Fig. 58) with lateral margins straight for much of their length, posterolateral angles abruptlybut shortly produced, posterolateral spines short, width across apices of spines 1-63-1-68 times width ofhead across eyes. Calli of posterior disc granular, close together. Length of posterior tibia 0-84-0-86 timeslength of posterior femur. Abdominal sternites with posterolateral angles acute, very slightly prominent. Aedeagus similar in allrespects, including the unpaired apical ventral lobe, to that of Pseudomyla spinicollis except that distaldorsomedian lobe is broadly conical and lacks finger-like processes at the lateral angles. Antennal segment I strongly, II and III weakly granulate. Head granulate, dorsally with a few largergranules and tubercles. Pronotum weakly granulate, densely punctate, with a few tubercles anteriorly inaddition to the four stout tubercles on the lateral margins. Scutellum, abdominal sternites including theirlateral margins and laterotergites weakly granulate, thoracic pleura moderately granulate-punctate,thoracic sterna weakly punctate. Clavus and corium strongly, seriately punctate except for impunctate areanear middle of apical margin of corium. Femora rather weakly granulate. Pubescence of antennae, head, thorax, femora and abdomen of short, crisped, semidecumbent, pale,rather flattened hairs. Femora, tibiae and tarsi with short, suberect, fine hairs. Tubercles of head andpronotum with short, erect, curved, flattened hairs. Colour yellowish stramineous, ocellar tubercles dark brown, antennal segment I, sides of head,pronotum, pleura, thoracic and abdominal sterna with diffuse brown markings; apices of all femora,especially the posterior pair, with brown mottling, all tibiae with proximal and distal incomplete brownannuli, anterior tibiae also with incomplete median brown annuli. Membrane of hemelytra colourless, itsveins white, veins of membrane and of corium with a few brown spots. Laterotergites with small brownpatches at base, at apex and in middle of lateral margins. DISTRIBUTION. India. MATERIAL EXAMINEDHolotype C?, India: Nagpur (BMNH).Paratypes. India: 1 $, Bombay; 1 CM $, Maharashtra, Jalna, 4.ix.l971 (/. C. Deeming) (BMNH). PSEUDOMYLA gen. n. Type-species: Merocoris spinicollis Spinola. Body rather narrowly oblong, not depressed, connexivum moderately widened.Head shorter than pronotum; antennifers moderately divergent, outer apical processes of antennifers TRIBE PSEUDOPHLOEINI 181 58 Figs 58-64 58, Pungra angusta, dorsal view of head and pronotum. 59-63, Pseudomyla spinicollis: (59)dorsal view of conjunctiva and vesica; (60) ventral view of same; (61) left lateral view of same (dorsal toleft of figure); (62) ventral view of abdominal margin; (63) dorsal view of pronotum. 64, P. cornuta,dorsal view of pronotum. about as long as broad, porrect. Antennal segment I abruptly but weakly clavate, as long as or slightlyshorter than head; segments I-IV subequal in length; specialized sensory area occupying about eight-ninths of length of antennal segment IV. Bucculae occupying about one-quarter of ventral midline of head.Rostrum at rest reaching to anterior margin or disc of metasternum, segment II almost as long as I, IVabout two-thirds as long and III about two-fifths as long as I. Pronotum (Figs 63, 64) with posterolateral angles strongly produced and directed forwards, stronglydeclivent, posterior margin straight or very shallowly emarginate, without prescutellar spines, lateralmargins each with three or four stout tubercles, disc between posterolateral angles with a pair of small,transverse, granular calli. Scutellum almost flat with apex and anterior angles slightly elevated. Mesoster-num and metasternum sulcate. Dorsal ridge of metathoracic peritreme reniform. Metathoracic wing withantevannal vein present but feebly developed. Anterior and intermediate femora with or without a singlesubapical spine beneath; posterior femur with a well-developed basal tubercle, three major subapicalspines of which the proximal one is about half as long as the others, two or three tubercles between the twolargest spines and an apical series of three or four tubercles. Posterior tibia 0-8-0 -9 times as long as femur. Abdominal sternites III-VII with posterolateral angles produced into small spines (Fig. 62). Malegenital capsule posteriorly emarginate, emargination filled by apices of parameres. Spermatheca with bulbgibbously lunate, bulb short. REMARKS. These reddish ochreous insects with strongly produced and anteriorly directed posterolateralpronotal angles have been placed in the African genus Myla on the basis of the wholly superficialsimilarities of body shape. The genitalia of the males are quite unlike the characteristic form encounteredin that genus and the nearest relative of Pseudomyla may be Pungra, which it resembles in the presence ofstout tubercles on the lateral margins of the pronotum and granular calli on its posterior disc; the form ofthe conjunctiva is very similar in the two genera. DISTRIBUTION. Oriental region. Key to species 1 Pronotum (Fig. 63) with posterolateral margins behind the posterolateral spines weakly concave, almost straight. (Central and southern India, Sri Lanka) spinicollis (p. 182) - Pronotum (Fig. 64) with posterolateral margins behind the posterolateral angles distinctly convex. (S. China, Laos) cornuta(p. 182) 182 W. R. DOLLING Pseudomyla spinicollis (Spinola) comb. n.(Figs 59-63) Merocoris spinicollis Spinola, 1837: 216-217. Syntype(s), INDIA: Bombay (Duponf) [lost].Clavigralla concolor Dohrn, 1860: 403. LECTOTYPE cf, SRI LANKA (IZPAN), here designated [ex-amined]. Syn. n.Myla concolor (Dohrn) Stal, 1873: 84. Length: cf , 6-75-7-1 mm; $ unknown. Length of antennal segment I equal to width of head including eyes; ratio of lengths of antennal segmentsabout 1 -00:0-92: 1 -00:0-96. Ratio of lengths of rostral segments about 1 -00:0-91 :0-41 :0-68. Pronotum (Fig. 63) with posterolateral margins (extending backwards from posterolateral spines) veryweakly concave, almost straight. Metasternum moderately convex, shallowly sulcate throughout. Anteriorand intermediate femora each with a single, small (Sri Lanka) or obsolete (Bombay) subapical spine. Conjunctiva (Figs 59-61) with M-shaped dorsomedian lobe with pronounced dorsal angles; low, broaddistal dorsomedian lobe with two anteriorly projecting finger-like appendages; distal dorsolateral lobesdescending, supported throughout their ventral faces by a sclerotized strip articulating with apices of wingsof ejaculatory reservoir complex; apical ventral lobe single, weakly sclerotized; distal ventrolateral lobesmembranous except for a pair of toothed strips in ventral wall, which project at their anterior ends on verysmall, free lobes. Vesica slender, protected at base by a single, asymmetrical sclerite; ejaculatory reservoirasymmetrical but with wings symmetrical, straps absent. Granulation of antennal segment I dense, uniform, that of segments II and III sparser and smaller; headgranulate, pronotum and all pleura strongly punctate and rather weakly granulate, scutellum moderatelygranulate-punctate, thoracic and abdominal sterna weakly granulate, laterotergites weakly granulate-,punctate, clavus and corium densely and strongly sedately punctate except for impunctate area nearmiddle of apical margin of corium, femora and tibiae sparsely and weakly granulate. Semidecumbent, short, white pubescence of flattened hairs present on antennal segments I-III, head,thorax, clavus, corium and femora; hairs of thoracic pleura, pronotum and scutellum confused, those ofother areas lying more uniformly parallel; femora, tibiae and tarsi with rather longer, fine, colourless,semi-erect pubescence. Entire insect ochreous red except for black meso- and metasterna, diffuse blackish markings on ventralsurfaces of head, pronotum and abdomen and colourless hemelytral membrane with veins marked whiteand brown; tibiae paler red than rest of insect. REMARKS. Spinola's type-material of Merocoris spinicollis, sent him by Dupont from Bombay, could not befound in Spinola's collection by Prof. Dott. Carlo Vidano, of Turin, who kindly searched for it on mybehalf, nor is it catalogued by Casale (1981). The genus is of such distinctive appearance that there can beno doubt that spinicollis belongs here. Spinola's description is clear and the length given (3 French lines:equal to 7 mm) is within the measured range of specimens available. The slight difference in the size of thespines of the anterior and intermediate femora of the specimens from Bombay and those from Sri Lankadoes not seem sufficient to warrant a division of the material into two species, particularly in view of thesmall number of specimens available. If it should prove that the Sri Lanka specimens are specificallydistinct from the Bombay ones, Dohrn's name concolor is available for the species. DISTRIBUTION. India and Sri Lanka. Hsiao's (1964: 252) record of Myla concolor from China refers to thematerial upon which he later (Hsiao, 1965: 427) founded the following species. MATERIAL EXAMINED Sri Lanka: 1 cf (lectotype of concolor), (Nietner) (IZPAN). Sri Lanka: 1 cf , Pundaloya (BMNH); 1 cf , Pundaloya (Atkinson) (UM). India: 2 cf , Bombay (Dixon),ICf, Bombay (BMNH). Pseudomyla cornuta (Hsiao) comb. n.(Fig. 64) Myla cornuta Hsiao, 1965: 427, 433. Holotype $, CHINA: Yunnan, Xishuang, Banna, 20.V.1958(Academia Sinica, Beijing) [not examined]. Length: cf , 7-0 mm (fide Hsiao, 1965); $ , 6-5-7-2 mm. Length of antennal segment I 0-94 times width of head including eyes; ratio of lengths of antennalsegments about l-00:0-90:l-06:0-88. Ratio of lengths of rostral segments about l-00:0-95:0-41:0-68. TRIBE PSEUDOPHLOEINI 183 Pronotal margins behind posterolateral spines gently but distinctly convex (Fig. 64). Metasternummoderately convex, rather deeply sulcate throughout. Anterior and intermediate femora unarmed (fideHsiao) or with a single, small subapical spine beneath. Posterior femur with three major spines, the mostproximal less than one-third as long as the others, between which are two to four tubercles and beyondwhich is an apical series of four tubercles. Coloration, especially of clavus and corium, more ochreous thanthat of P. spinicollis. REMARKS. Type-material was not available but the material examined agrees rather closely with theoriginal description, and the locality from which it was obtained is near to the type-locality. DISTRIBUTION. China: Yunnan (type-locality) and Laos. MATERIAL EXAMINEDLaos: 4 $, Sedone Province, Paksong, 18.V.1965 (P. D. Ashlock) (BPBM). NEOMEVANIOMORPHA gen. n. Type-species: Mevaniomorpha annulipes rodhaini Schouteden, 1938: 296. Body oblong, neither compressed nor depressed, connexivum broad. Head about as long as pronotum; eyes small, prominent. Antennifers divergent, outer apical processabout as long as broad, porrect. Antennal segments subequal in length, segment I weakly clavate,granulate to tuberculate, distinctly shorter than head, segments II and III slender, very weakly granulate,IV elongate fusiform with specialized sensory area occupying about seven-eighths of its length. Bucculaeoccupying about one-quarter of ventral midline of head. Rostrum at rest reaching disc of metasternum;segment III the shortest, II and IV subequal, slightly shorter than I. Pronotum (Fig. 65) rather strongly declivent, granulate-tuberculate, lateral margins bearing severalshort tubercles, posterolateral angles strongly produced anterolaterally, tapering to terminal spine;posterior margin weakly convex, with a few granules, prescutellar spines absent. Scutellum almost flat,equilateral, apex not elevated or swollen. Mesosternum and metasternum strongly sulcate throughout.Dorsal ridge of metathoracic peritreme bilobed, anterior lobe much the larger. Corium with apical marginstraight. Metathoracic wing with antevannal vein present. Anterior and intermediate femora withsubapical spines beneath, posterior femur with three major subapical spines and some minor ones and withbasal tubercle well developed. Posterior coxae separated by about half the width of a coxa. Abdominal sterna III- VII produced into broad, triangular spines, lateral margins almost smooth.Spermatheca (Fig. 72) with bulb lunate, duct long and tightly convoluted. Sclerites of wall of gynatriumtriradiate, with three arms of comparable length, one arm rising to meet upper ends of rami, onedescending almost to meet base of second valvula and one extending horizontally posteriorly. Ovipositorwith valvulae longer and more slender than in related genera. Male genital capsule (Fig. 69) short,posteriorly emarginate, emargination filled by apices of the short parameres (Figs 70, 71). Aedeagus (Figs66, 67, 68) showing phallotheca of form typical in the tribe; conjunctiva with dorsomedian lobe M-shaped,membranous, distal dorsomedian lobe membranous, small, rounded-conical; distal dorsolateral lobessmall, weakly sclerotized, apical ventral lobes paired, membranous, subglobular, distal ventrolateral lobeslarge, membranous, bilobed; ventral wall of conjunctiva with two longitudinal tracts of denticles thatextend anterolaterally up sides of conjunctiva; vesica protected at base by a dorsal, cup-like sclerite;ejaculatory reservoir with long wings that extend inside the distal dorsolateral lobes to their apices, withoutstraps. REMARKS. The anterolaterally directed pronotal angles are shared with the African genus My la and theOriental Pseudomyla, both of which have decumbent, scale-like pubescence in contrast to the suberect toerect hairs of Neomevaniomorpha. The raised, white tip of the scutellum that is characteristic ofPsilolomia, Mevaniomorpha and Mevanidea is not found in this genus; it is least well developed inPsilolomia but the form of the conjunctiva is dissimilar. Probably the closest relatives of the genusNeomevaniomorpha are Mevaniomorpha and Mevanidea. DISTRIBUTION. Central Afro tropical region. Neomevaniomorpha rodhaini (Schouteden) comb n., stat. n. (Figs 65-72) Mevaniomorpha annulipes rodhaini Schouteden, 1938: 296. LECTOTYPE cf, ZAIRE (MRAC), heredesignated [examined] . 184 65 W. R. DOLLING 68 72 Figs 65-72 Neomevaniomorpha rodhaini. 65, dorsal view of pronotum; 66, dorsal view of aedeagus; 67,lateral view of phallotheca, conjunctiva and vesica; 68, ventral view of conjunctiva and vesica; 69, lateralview of genital capsule; 70, ventrolateral view of left paramere; 71, dorsomedial view of left paramere;72, spermatheca. Length: cf , 7-4-7-7 mm; $,7-9-9-0 mm. Antennal segment I about equal to width of head across antennifers, bearing some granules about ashigh as wide or, in one male paralectotype, with a few tubercles slightly higher than wide among thegranules. Ratio of lengths of antennal segments about 1-00:0-96: 1-08:0-96. Ratio of lengths of rostralsegments about l-00:0-92:0-64:0-87. Shape of pronotum as in Fig. 65. Corium with apex reaching or almost reaching suture betweenlater otergites V and VI. Sulcus of mesosternum and metasternum very broad. Anterior and intermediatefemora each with a single subapical spine beneath; posterior femur with first major spine less than half aslong as the other two, followed by about three very small minor spines, space between the two larger spineswith two or three minor spines and an apical series of four minor spines present. Length of posterior tibiadivided by length of posterior femur 0-80-0-84. Abdominal sternites III- VII with their posterolateral angles drawn out into spines equal in length atmost to one-third of the width of their respective laterotergites. Spermatheca and male genitalia as in Figs66-72 (see description of genus for discussion of salient features). Head granulate; gular region punctate; vertex, frons, tylus and juga shortly tuberculate. Pronotumgranulate-punctate, declivent area and lateral margins tuberculate, pleura granulate-punctate, sternagranulate, scutellum granulate-punctate with the granules on its lateral margins prominent. Hemelytrawith clavus and corium punctate throughout, veins anteriorly, especially costal margin of corium, stronglygranulate. Legs, especially femora, abdominal sternites and laterotergites granulate. Pubescence of moderate to rather short, suberect hairs; dorsum of head and declivent part of pronotumwith long, erect hairs. Coloration generally mid-brown; ventral midline of thorax and of abdomen piceous. Membrane ofhemelytra fuscous, the veins darker with occasional small, white spots; posterior basal angle white,appearing as a conspicuous, white triangle just behind end of claval commissure when hemelytra are atrest. REMARKS. This is the only species of the genus. It seems strange that Schouteden could have described thisinsect as a form of Mevaniomorpha annulipes, from which it differs strikingly in the shape of the pronotumand in coloration. TRIBE PSEUDOPHLOEINI 185 DISTRIBUTION. Zaire and adjacent part of Zambia. MATERIAL EXAMINED Zaire: 1 cf (lectotype), Sankisia, 1911 (Dr Rodham) (MRAC). Zaire: 1 cf , Lulua, Kapanga, viii.1932 (F. G. Overlaet); 1 <f>, Lulua, Kapanga, ix.1932 (F. G. Overlaet)(both paralectotypes; 2 cf , Sankuru, M'Pemba Zeo (Gandajika), 28.ix.1958 (R. Marechat); 1 cf , Tshupa,Lukolela (De Guide); 1 $, Lulua, Kapanga, i.1933 (F. Overlaet); 1 <j>, Wombali, viii.1913 (P. Vanderijst)(labelled 'allotypus') (MRAC). Zambia: 1 $, Upper Kalungwisi Valley, 4200 ft [1260 m], 11. ix. 1908 (5. A.Neave) (UM). MEVANIOMORPHA Reuter Mevaniomorpha Reuter, 1883: 13. Type-species: Mevaniomorpha annulipes Reuter, by monotypy. Body oblong, slightly depressed; connexivum broad. Head about as long as pronotum, granulate-tuberculate. Antennifers strongly divergent, outer apicalprocess porrect. Antennal segment I weakly clavate, granulate or granulate-tuberculate, longer or shorterthan head; III longer than II, both slender, granulate, IV shortest, narrowly fusiform, specialized sensoryarea occupying four-fifths to five-sixths of its length. Bucculae occupying about one-quarter of ventralmidline of head. Rostrum at rest reaching to disc of metasternum, its first two segments subequal in length,fourth shorter, third shortest of all. Pronotum rather shallowly declivent, granulate-tuberculate and punctate, its posterolateral anglesmoderately produced laterally with posterolateral spines arising abruptly from them or greatly producedlaterally and tapering gradually into the laterally-directed spines; lateral margins with several outstandingtubercles; posterior margin straight, smooth in front of scutellum, granulate further towards sides of body,prescutellar spines absent but granules present in their usual position on each side. Scutellum equilateral,its apex elevated into a small, white swelling occupying about one-fifth of its length. Mesosternum andmetasternum longitudinally sulcate throughout. Metathoracic scent-gland peritreme with dorsal ridgebilobed, anterior lobe much the larger, constriction between lobes shallow, sometimes obsolete. Coriumwith costal and apical margins slightly convex. Metathoracic wing with antevannal vein. Anterior andintermediate femora with one or more small subapical spines beneath; posterior femur with three majorsubapical spines, some minor ones and a basal tubercle. Posterior tibia slightly more than four-fifths as longas femur. Posterior coxae separated by rather less than width of one coxa. Abdominal sternites III-VII with posterolateral angles projecting as broad, triangular teeth, lateralmargins almost smooth. Spermatheca with bulb lunate, duct tightly convoluted within concavity of bulb.Sclerites of wall of gynatrium L-shaped. Male genital capsule posteriorly emarginate, emargination filledby apices of parameres. Phallotheca with ventral sclerite short ventrally, strongly produced posterol-aterally, dorsal sclerites very weakly sclerotized. Conjunctiva with flat-topped dorsomedian lobe, conicaldistal dorsomedian lobe, short distal dorsolateral lobes, broad apical ventral lobes, large distal dor-soventral lobes, all of them membranous; two strips of denticles present in ventral wall of conjunctiva;coiled base of vesica protected by two sclerites of which the left sclerite is twice as long as the right one. REMARKS. This genus differs from Psilolomia in the greater projection of the posterolateral angles of theabdominal sternites and the presence of sclerites at the base of the vesica. The head and first antennalsegment are not nearly so spiny as in Mevanidea. DISTRIBUTION. Tropical Africa. Key to species 1 Antennal segment I shorter than width of head including eyes; posterolateral spines ofpronotum arising abruptly from the posterolateral angles; width across apices of spines lessthan 2-5 times width of head including eyes. (Widespread in tropical Africa) annulipes(p. 185) - Antennal segment I longer than width of head including eyes ; posterolateral angles of pronotumtapering gradually into spines; width across apices of spines more than 2-5 times width of headincluding eyes. (Forestsof Central and West Africa) picta(p. 187) Mevaniomorpha annulipes Reuter (Figs 73, 75, 76) [Clavigralla elevator (Fabricius) sensu Dallas, 1852: 511. Misidentification; not Coreus elevator Fabricius,1803: 194.] 186 W. R. DOLLING Mevaniomorpha annulipes Reuter, 1883: 13-14. Holotype $, GHANA (ZMU) [examined].[Mevaniomorpha annulipes subsp. picta Schouteden sensu Linnavuori, 1970: 43-45. Misidentification.] Length: cT, 7-2-8-2 mm; $, 7-8-9-9 mm. Antennal segment I about as long as width of head across antennifers. Ratio of lengths of antennalsegments in male about l-00:l-04:l-08:0-88, in female about l-00:l-04:l-12:0-84; total length of antennaless than 3-7 times width of head including eyes. Ratio of lengths of rostral segments aboutl-00:0-92:0-54:0-78. Posterolateral angles of pronotum (Fig. 73) prominent, posterolateral spines arising abruptly fromthem, directed laterally; width across apices of spines divided by width of head including eyes 2-2-2-4. Aedeagus as in Figs 75, 76 (see description under genus). Head dorsally granulate-tuberculate, laterally granulate, ventrally punctate. Pronotum granulatethroughout, tuberculate anteriorly, punctate posteriorly; scutellum and pleura punctate-granulate; thor-acic and abdominal sterna and laterotergites weakly granulate. Clavus and corium punctate throughout,veins sparsely and weakly granulate; femora granulate. Pubescence of rather short, suberect hairs, almost decumbent on head and pleura; longer, erect hairspresent on dorsum of head and pronotum. Colour pattern of medium brown and piceous markings on a paler, yellow ground. The most prominentdark markings are a stripe along anterior midline of pronotum, a pair of spines in front of anterolateralangles of pronotum, apical margin of corium, spot in apical part of disc of corium usually separate fromdark apical margin, posterior two-thirds of laterotergites III-V, posterior margins of laterotergites V andVI, anterior margins of laterotergites III- VII and extensive mottling on apices of posterior femora. REMARKS. Some specimens in the north-eastern part of the range have shorter posterolateral pronotal andabdominal spines than the main population; such specimens were referred to the variety picta Schoutedenby Linnavuori (1970: 43-^45); picta in fact has longer spines than true annulipes and appears to be a goodspecies. DISTRIBUTION. Widespread in tropical Africa. MATERIAL EXAMINED Ghana: 1 cf (holotype), Addah (Reitter) (ZMU). Guinea: 1 cf, Nimba, Yalanzou, 'ii.vi.42' (Lamotte), 1 cf, Nimba, Keoulenta, 'ii.vi.42' (Lamotte)(MNHN). Sierra Leone: 1 $ (leg. Morgan fide Dallas, 1852: 511) (labelled '425a' and 'Clavigrallaelevator,'); 1 cf , Mopeille, l.vii.1926 (Hargreaves) (BMNH). Ivory Coast: 1 $ , Tai, 12.xi.1979 (Couturier)(MNHN); 1 Cf , Bingerville, xii.1963 (Decelle); 1 $, Korea, au sud de Daloa, ix.1961 (Decelle) (MRAC).Nigeria: 1 cf, Calabar, 9.xi.l955 (Bechyne, Exped. Mus. G. Frey); 1 cf, Akpasha Udi, 27.X.1955(Bechyne, Exped. Mus. G. Frey); 1 cf , near Bida, between Dabba and Kutiwenji, 20. xi. 1970 (Deeming); 1$ , Gombe, Matzoro Lakes, i.1929 (Lloyd) (BMNH). Cameroun: 2 cf , Batouri District, 345'N, 1345'E,750 m, I.v-6.vi.l935 (Merfield) (BMNH); 1 $, Baigom, reg. Bamoun (MNHN). Zaire: 1 cf , Gamangui,ii.1910 (Lang & Chapin) (AMNH); 1 cf, Haut-Uele, Manda, 18.iii.1925 (Schouteden); 1 cf, Sankuru,M'Pemba Zeo (Gandajika), 1960 (Marechal); 1 cf, Kivu, Mulungu, 1939 (Hendrickx); 1 cf, Kivu,Mulungu-Tshibinda, xi.1951 (Lefevre); 1 cf, Kivu, Mwenga, Kitutu, rive de 1'Elila, 650 m, foretmarecageuse primaire, humus, iv.1958 (Leleup); 1 cf, Katanga, Luembe, viii-ix.1956 (de Caters); 1 $,Lulua, Kafakumba, xii.1932 (Overlaet); 1 $, Maulema, Wamaza, iii-iv.1957 (Cotonco); 1 $, Gandajika,iv.1959 (Decelle); 1 $, Bas-Congo, Thysville, 1959-1963 (Michaux); 1 $, Bas-Congo, Mayidi, 1942 (vanEyen); 1 $, Kivu, Mulungu, 1938 (Hendrickx); 1 $, Kivu, Mulungu, Tshibinda, xi.1951 (Lefevre); 1 $,Kivu, Ibanda, 1952 (Vandelannoite) (MRAC). Uganda: 1 cf, Mutunda, l.vi.1911 (Marshall); 1 $,Western Ankole, 4500-5000 ft (1350-1500 m), 10-14.X.1911 (Neave) (BMNH). Kenya: 1 cf, Embu,12.ix.1914 (Browne); 1 $, N. slopes of Mt Kenya, on Embu-Meru road, 4500-5000 ft (1350-1500 m),13-14.ii.1911 (Neave); 1 $, Ngong, v.1926 (Gedye); 1 cf , 1 $, Kaimosi, iii-iv.1932 (Turner); 1 , ChyuluHills, 5200 ft (1600 m), iv.1938 (Coryndon Mus. Exped.) (BMNH). Zambia: 1 cf , Lake Bangweulu,N'Sumba Island, 25.xi.1946, beaten from climbers (Steele); 1 , Lake Bangweulu, Kapola, N. of Kapata,27.X.1946 (Steele); 2 cf , 1 $, Lake Bangweulu district, N'Salushi Island, 13.xi.1946 (Steele); 4 cf, UpperLuanga River, 27.vii-13.viii. 1910 (Neave); 1 9, Luangwa to Petauke, 14-17.ix.1910 (Neave) (BMNH); 1Cf , Lower Kalungwisi valley, 3500 ft (1050 m), dense forest, 12-13. ix.1908 (Neave) (UM). Malawi: 1 $,Ruo valley, 1000-2000 ft (300-600 m), 21-25.iv.1910 (Neave) (BMNH). South Africa: 1 $, Transvaal,Louis Trichard, 20-30. xii. 1956 (Capener) (J. A. Slater coll.); 1 $ Natal, 'P. town' (?=Pinetown), Gramin(BMNH). TRIBE PSEUDOPHLOEINI 187 Mevaniomorpha picta Schouteden stat. n.(Fig. 74) Mevaniomorpha annulipes picta Schouteden, 1938: 296. LECTOTYPE cf, ZAIRE (MRAC), here desig-nated [examined]. Length: cf, 8-1-8-4 mm; $,8-6-9-3 mm. Antennal segment I longer than width of head including eyes; ratio of lengths of antennal segmentsabout 1 -00:0-91 :0-96:0-71; total length of antennae more than 4-1 times width of head including eyes. Ratioof lengths of rostral segments about l-00:0-93:0-55:0-79. Pronotum (Fig. 74) with posterolateral angles merging gradually into posterolateral spines, width acrossapices of spines 2-6-3-0 times width of head including eyes; tubercles of disc and margins long. Male genitalia similar to those of M. annulipes but ejaculatory reservoir complex with wings ratherlonger. Pubescence as in M. annulipes; granulation and tuberculation more pronounced. Colour pattern verysimilar to that of annulipes but all elements of pattern darker; antennae rather dark brown throughout. REMARKS. This species differs from M. annulipes in the longer appendages, more pronounced posterolat-eral spines of the pronotum and darker colour. Linnavuori (1970: 43-45), presumably basing hisidentification on coloration, referred dark specimens of true annulipes to this taxon; these specimens in facthad slightly shorter spines and appendages than typical annulipes. After the above description wascompleted, Dr V. van Zeijst showed me two males of picta from the Tai forest in the Ivory Coast; thesespecimens had very long, slender pronotal spines and were almost black; superficially, they closelyresembled the unrelated Clavigralla hystrix Dallas, which occurred in the same forest. DISTRIBUTION. Rain forests of Central and West Africa. MATERIAL EXAMINED Zaire: 1 cf (lectotype), Sankuru, Komi, iv.1930 (/. Ghesquiere) (MRAC). Zaire: 1 cf, Sankuru, Komi, 2.V.1930 (Ghesquiere); 1 cf, Stanleyville (= Kisangani), 20.xii.1929(Collart); 1 $, Uele, Dingila, ix.1933 (Bredo); 1 $, Lomani, Kambaye, vii.1930 (Quarre) (paralectotypes)(MRAC); 1 $, Yangambi, xi.1937 (Henrard) (MRAC). Cameroon: 1 $, Batouri (MNHN). MEVANIDEA Reuter Mevania Stal, 1866: 110. Type-species: Clavigralla spiniceps Signoret, by monotypy. [Homonym of Mevania Walker, 1854: 442-^43.] Mevanidea Reuter, 1883: 11-12. Type-species: Mevanidea granulifera Reuter, by monotypy.Mevaniella Bergroth, 1907: 146. [Replacement name for Mevania Stal.] Syn. n. Body not depressed, conexivum moderately expanded. Aspect, especially of head and pronotum, spinose(Fig. 77). Head slightly shorter than pronotum, dorsally with numerous, long, spine-like tubercles. Eyes small,prominent. Antennifers moderately divergent, external apical processes porrect and directed slightlydownwards. Antennal segment I strongly clavate, bearing numerous spine-like tubercles on its expandedapical part. Antennal segments I and IV subequal in length, IV fusiform; II subequal to or rather longerthan I, III much the longest; II-IV almost smooth, bearing hairs about as long as width of segments II andIII. Bucculae occupying about one-quarter of ventral midline of head. Rostrum at rest reaching to anteriormargin or disc of metasternum; segment II slightly and IV distinctly shorter than I, III obviously shortest. Pronotum strongly declivent, posterior margin straight or slightly convex, prescutellar spines welldeveloped, tubercles present between them and posterolateral angles, along lateral margins and on disc,including some on posterior part. Posterolateral angles of pronotum slightly produced, posterolateralspines arising abruptly from them, directed laterally and slightly anteriorly. Scutellum equilateral, slightlyconvex, its apical one-third conspicuously swollen, china- white. Mesosternum deeply sulcate throughout;metasternum sulcate for half or all of its length. Metathoracic peritreme with its dorsal ridge simple orunequally bilobed, the anterior lobe in the latter case much the larger. Metathoracic wing with antevannalvein. Anterior and intermediate femora with two rows of tubercles and granules beneath. Posterior femurwith basal tubercle and beneath with three major subapical spines, the first spine two-thirds as long as theothers; usually three minor spines between the two biggest ones and an apical series of four. Posterior tibiashort, two-thirds to three-quarters as long as femur and conspicuously curved near base. Abdominal sternites III-VI with posterolateral angles produced into broad, triangular spines, lateral 188 W. R. DOLLING 73 Figs 73-78 73, Mevaniomorpha annulipes, dorsal view of head, pronotum, scutellum and first antennalsegment. 74, M. picta, dorsal view of head, pronotum and first antennal segment. 75, 76, M. annulipes:(75) ventral view of conjunctiva and vesica; (76) dorsal view of phallotheca, conjunctiva and vesica. 77,Mevanidea hystrix, dorsal view omitting legs and last three antennal segments. 78, Arenocoris interme-dius, antenna. margins of sterna granulate. Spermatheca with bulb very narrow, duct three to four times as long as bulb,convoluted. Male conjunctiva with dorsomedian lobe M-shaped, membranous, distal dorsomedian lobe ofsimilar size, also membranous, distal dorsolateral lobes obsolete, distal ventrolateral lobes large, bifid,weakly sclerotized on their posterior faces, apical ventral lobes membranous, paired, ventral wall ofconjunctiva with or without a pair of toothed, sclerotized strips; basal, coiled part of vesica protected bytwo subequal sclerites. TRIBE PSEUDOPHLOEINI 189 REMARKS. The appearance of the two species of this genus is very characteristic (Fig. 77), the spiny headand first antennal segment being particularly notable. The genus is probably derived from a Mevaniomor-p/io-like form. DISTRIBUTION. Africa and Madagascar. Key to species 1 Membrane of fore wing with an irregular, opaque, piceous or dark brown spot occupying about half of its area. (Africa) hystrix(p. 189) - Membrane of forewing translucent, with some darker markings along veins and sometimes withdark amber shading of area between veins, but never with an opaque spot. (Madagascar) spiniceps(p. 190) Mevanidea hystrix (Gerstaecker)(Fig. 77) Mevania hystrix Gerstaecker, 1873: 408-409. Holotype cf , KENYA (MNHU) [examined]. Mevanidea granulifera Reuter, 1883: 12. Holotype $, GHANA (ZMU) [examined]. [Synonymized by Linnavuori, 1978: 36?] Mevanidea kilimana Schouteden, 1910: 157. [Nomen nudum.]Mevaniella kilimana Schouteden, 1912: 55. LECTOTYPE cf , KENYA (NR), here designated [examined]. [Synonymized by Schouteden, 1938: 295.]Mevanidea hystrix (Gerstaecker) Schouteden, 1938: 295. Length: cf , 5-7-6-8 mm; $,6-2-7-6 mm. Ratio of lengths of antennal segments in male about 1-00: 1-02: 1-30: 1-02, in female about1-00:1-14:1-41:1-00. Ratio of lengths of rostral segments in male about l-00:0-97:0-55:0-79, in femaleabout l-00:0-97:0-50:0-75. Head dorsally with numerous granules and tubercles of various lengths, mostnoticeably a pair of spine-like tubercles close together between eyes with a third immediately behind them,a long, spine-like tubercle on each side of vertex just posterior to inner basal angles of antennifers, astaggered row of long tubercles all along tylus and others, rather shorter, on juga; sides and ventral surfaceof head punctate-granulate. Pronotum with lateral margins almost straight, bearing three or four long tubercles, lateral anglessuddenly prominent; prescutellar spines long, slender, sinuous; centre of disc with a widely spaced pair ofconical tubercles, two groups of partially fused, blunt granules behind them but less widely spaced and arow of three similar clusters of granules between these and posterior margin in addition to more generallydistributed granules and tubercles. Scutellum and thoracic pleura and sterna punctate-granulate. Hemely-tra with clavus and corium, except for smooth area in middle of its apical part, punctate, veins anteriorlyand basally granulate. Femora strongly granulate, shortly tuberculate dorsally and ventrally, with one ortwo subapical tubercles of the anterior and intermediate femora enlarged and occupying the usual positionsof subapical spines; posterior femur with three major subapical spines beneath, one to four minor spines inthe space between the first two and three or four in the space between the middle and last, with an apicalseries of three or four. Posterior tibia about two-thirds as long as femur. Ventral wall of conjunctiva of aedeagus with a pair of apically divergent, toothed, sclerotized strips;vesica short, about as long as either of the wings of the ejaculatory reservoir complex. Pubescence throughout of moderate to short, suberect hairs with some long, erect hairs on pronotum. Colour pattern as in Fig. 77; the pale areas being stramineous and the dark areas varying from darkbrown to dark red in different specimens. The most striking features of the colour pattern are thechina- white, swollen apex of the scutellum, an irregular, opaque, dark brown blotch in the middle of themembrane of the hemelytra and a dark brown (never red) transverse spot near middle of corium edgedproximally with white; laterotergites III to VI largely dark with a pale band anteriorly, VII largely pale. REMARKS. I can trace no earlier synonymy of granulifera with hystrix than that of Linnavuori (1978: 36);however, the synonymy is not indicated as new in that publication. The holotype of M. hystrix Gerstaeckeris a small (6-0 mm) and very red male from Mombasa (Kenya) and that of M. granulifera Reuter a large (7-1mm) blackish brown female from 'Addah' (i.e. Ada, Ghana). The great distance (4500 km) between thetype-localities and the striking difference between specimens at the extremes of the range of colourvariation have led earlier workers to recognize two species on the African continent. The reddishcoloration is in general more pronounced in specimens from eastern and southern Africa and the brownish 190 W. R. DOLLING coloration more pronounced in those from western Africa. The dark brown blotch in the middle of themembrane of the hemelytra is variable in extent but not in colour. The egg of this species is illustrated byCobben (1968: 97, fig. 84). DISTRIBUTION. Widespread in subsaharan Africa. MATERIAL EXAMINED Kenya: 1 cf (holotype of hystrix), Mombasa (van der Deckeri) (MNHU). Ghana: 1 9 (holotype ofgranulifera), Addah (Reitter) (ZMU). Kenya: 1 $ (lectotype ofkilimana), Kilimanjaro (Sjoestedt) (NR). Sierra Leone: 1 $, Port Lokko, 30.viii.1925 (Hargreaves) (BMNH). Ivory Coast: 1 $, Adiopodoume,iv-v.1964 (Cobben) (BMNH); 1 $, Bouake, ii.1963 (Schmitz) (MRAC). Ghana: 1 cf, Kintampo,7.xii.l965 (Leston) (BMNH). Nigeria: 1 cf, Kumba, 15.xi.1955 (Bechyne) (BMNH). Sudan: 1 cf,Ingessana Mts, Blue Nile, 18-22.xi.1982 (Linnavuori) (BMNH). Zaire: 1 cf , Yangambi, xii.1959 (De-celle); 1 $, Muranan 'Deke, v.1953 (De Francqueri) (MRAC); 1 cf, Mpese, ll.vi.1937 (Cooremari)(IRSNB); 7 Cf, 5 $, Faradje, 2940'E, 340'N, i.1913 (Lang & Chapiri) (AMNH). Uganda: 1 cf , BussuBusoga, v.1909 (Bayon); 1 $, Chugwe, Mabira Forest, 3500-3800 ft (ca 1100 m), 16-25.vii.1911 (Neave)(BMNH). Kenya: 1 cf, Rabai, vii.1937 (van Someren); 1 cf, Mombasa, iii.1921 (van Someren); 1 <j>,Chyula Hills, 5200 ft (1600 m), iv.1938 (Coryndon Museum Expedition) (BMNH). Tanzania: 1 cf , Uvira,28-29.viii.1931 (Ogilvie); 1 cf, Zanzibar, near Mazi Moja, 20.viii-ll.ix.1924 (Snelt) (BMNH); 1 $,Zanzibar (Ashby) (AMNH). Malawi: 1 cf , between Fort Mangoche and Chikala Boma, about 4000 ft(1200 m), 20-25.iv.1910 (Neave); 1 $, Mlanje, 2.xii.l912 (Neave) (BMNH). Zambia: 2 cf, 2 $, UpperLuanga River, 27.vii-13.viii.1910 (Neave); 1 $, Lake Bangweulu district, ix-xii.1946 (Steele); 1 $, LakeBangweulu, Chiluwi Island, beaten from lemon (Citrus limon), 29. xi. 1946 (Steele) (BMNH). South Africa:1 Cf , 1 $ , Transvaal, Kruger National Park, 3 miles E. of Satara Camp, Nwandzi River, 29. iv. 1968 (Schuh,Slater & Sweet) (J. A. Slater collection); 1 cf , Transvaal, Johannesburg, 6000 ft (1800 m), ii.1895 (Cregoe);1 $, Transvaal, Johannesburg, 6000 ft (1800 m), iv.1899 (Cregoe); 1 cf, Natal, Estcourt, i.1897; 2 cf,Natal, Tugela River, near Weenen, 14.iii.1897; 1 $, Natal, Durban, 1874 (Bell-Marley) (BMNH). Mevanidea spiniceps (Signoret) comb. n. Clavigralla spiniceps Signoret, 1861: 944. LECTOTYPE $, MADAGASCAR (NMV), here designated [examined]. Mevania spiniceps (Signoret) Stal, 1866: 110-111.Mevaniella spiniceps (Signoret) Bergroth, 1907: 146. Length: cf, 6-8-7-8 mm; $,7-3-8-2 mm. Ratio of lengths of antennal segments in male about 1-00:1-30:1-55:1-15, in female about1-00:1-14:1-48:1-04. Ratio of lengths of rostral segments in male about l-00:0-92:0-54:0-87, in femaleabout l-00:0-92:0-55:0-78. Head strongly spiny-tuberculate but with fewer tubercles and granules than M.hysterix; the pair of long tubercles on vertex between eyes without a tubercle behind them. Posterior tibiaabout three-quarters as long as femur. Aedeagus with no ventral sclerotized, toothed strips on conjunctiva and with vesica about twice as longas either wing of ejaculatory reservoir. Colour pattern of dark to pale, often somewhat rufous (but never strongly reddish) brown areas andstramineous areas; membrane of hemelytra without opaque, brown blotch though with some brownmarkings along veins and occasionally with some dark amber shading of membrane between veins.Laterotergites III to V largely dark, each with a pale band anteriorly, VI and VII largely pale. REMARKS. This species resembles M. hystrix in general appearance. The main differences are in theaedeagus, where the longer vesica of spiniceps is presumably plesiomorphic and the absence of ventralsclerotized strips in the conjunctival wall apomorphic, and the absence of the dark brown blotch on thehemelytral membrane. The presence of only two spines, rather than three, on the vertex between the eyesis constant in all the specimens examined. M. spiniceps and a few species of Clavigralla (tribe Clavigrallini)are the only Pseudophloeinae known from Madagascar. DISTRIBUTION. Madagascar. MATERIAL EXAMINED Madagascar: 1 $ (lectotype) (NMV). Madagascar: 1 $, Tulear Province, Tongobory, 200 m, 27.iii.1968 (K.M.G. & P.D.) (BMNH); 4 cf , 2$ , [no precise locality], 1930 (Sicard); 2 cf , Baie de Baly, Plateau de Soalala, 1930 (Joly); 2 cf , no data; 1Cf, Region de Sud-est, Fort-Dauphin, i.1901 (Alluaud); 1 cf, Nossi-Be, 5.vii.l900 (Alluaud); 1 cf, TRIBE PSEUDOPHLOEINI 191 Ivondro, vii.1900 (Alluaud); 1 cf [no precise locality], (Fairmaire); 1 $, Region de Sud-est, Vallee duFanjahira, xii.1901 (Alluaud); 1 $, Vallee du Fanjahira, Isaka, xii.1901 (Alluaud); 1 $, Nossi-Be, ForetLoukoube, 1897 (Alluaud); 1 $, Sahambava, Fianarantsoa (MNHN). AtfEJVOCORISHahn Arenocoris Hahn, 1834: 109-110. Type-species: Coreus falleni Schilling, by subsequent designation of Blanchard, 1842: 312. [Atractus Laporte sensu Curtis, 1834: legend to pi. 500. Misidentification.]Pseudophloeus Burmeister, 1835: 301, 302, 308. Type-species: Coreus falleni Schilling, by subsequent designation of Brulle, 1835: 359. Ammocoris Agassiz, 1848: 48, 94. [Unjustified emendation of Arenocoris Hahn.]Psammocoris Marshall, 1868: 281. [Unjustified emendation of Arenocoris Hahn.]Boudicca Kirkaldy, 1909: 30. [Unnecessary replacement name for Pseudophloeus Burmeister.] Body strongly depressed, ovate; connexivum considerably expanded in middle. Body and appendagesconspicuously granulate-tuberculate . Head distinctly longer than pronotum. Antennifers divergent with outer apical process either porrect ordeflexed. Antennae with segment I much shorter than width of head, strongly incrassate except for basalone-eighth; segments II and III slender, II much shorter than I, III about 4 times as long as II, IV short,fusiform, intermediate in thickness between I and II. Bucculae occupying about two-fifths of ventral lengthof head. Rostrum at rest reaching to posterior margin of mesosternum. Pronotum weakly declivent, abruptly widened posteriorly; posterolateral angles truncate, sometimesauriculate, not pointed and not bearing posterolateral spines; posterior margin almost straight, weaklydeveloped prescutellar lobes (not spines) present. Scutellum equilateral, lateral margins elevated anteri-orly, apex elevated and bilobed. Mesosternum throughout and metasternum anteriorly deeply sulcate inmidline. Metathoracic scent-gland peritreme with dorsal lobe entire, its dorsal margin evenly rounded,descending anteriorly and posteriorly to enclose orifice in at least its dorsal half. Membrane of hemelytronwith venation reticulate. Metathoracic wing with antevannal vein well developed. Anterior and intermedi-ate femora without subapical spines beneath, sometimes with a very slightly enlarged tubercle in thisposition; posterior femur with a single, large tubercle or blunt spine subapically beneath, very rarely withan additional, slightly enlarged tubercle distad of this; base of posterior femur with a prominent tubercleadjacent to trochanter. Abdominal margin evenly rounded, posterolateral angles of sternites obtuse or right-angled, notprominent. Male genital capsule short with tongue triangular, lip obtusely angled in middle; parameresbroad, flat, their apices filling posterior emargination of capsule. Phallotheca comprising two slender,dorsal sclerites and a slightly shorter, broad, ventral sclerite. Conjunctiva with dorsomedian lobe broad,membranous; distal dorsolateral lobes very long, their posteroventral sides supported by sclerotizedextensions of wings of ejaculatory reservoir apparatus; vesica not protected basally by sclerites; apicalventral lobes small, membranous; distal ventrolateral lobes large, bilobed, the lower lobe sclerotized on itsposterior face; ventral wall of conjunctiva with two slender, parallel sclerites; ejaculatory reservoircomplex with wings long, straps absent. Ovipositor with second valvulae emarginate at apex. Sclerites ofdorsal wall of gynatrium short, each comprising a narrow, longitudinally elongate, vertical, mesal platesupporting a thin, sclerotized ring and a narrow, horizontal, transverse, anterior plate articulating laterallywith second valvifer and its ramus. Spermatheca with bulb lunate or half-moon shaped, duct about 1-5times as long as bulb, slightly undulate but not convoluted. INCLUDED SPECIES. Stichel (1962: 194-195) lists six species: egenus (Horvath, 1917), falleni (Schilling,1829), gestroi (Bergevin, 1930), intermedius (Yakovlev, 1883), latissimus Seidenstiicker, 1960 and waltli(Herrich-Schaeffer, 1834). Chernova (1979: 579) established the synonymy of egenus with falleni, and thesynonymy of gestroi with intermedius is established below. REMARKS. This is essentially a Palaearctic genus, with a single eremic species intrusive into the north-eastern part of the Afrotropical region. Within the tropical fauna it is readily distinguished from othergenera of Pseudophloeinae by the very short antennal segments I and II. DISTRIBUTION. Europe, North Africa, Canary Islands, western Asia; mostly eremic but two species extendinto cooler parts of Europe and another into Ethiopia and Chad. 192 W. R. DOLLING Key to species 1 Outer apical process of antennifer deflexed - Outer apical process of antennifer porrect 3 2 Antennal segment I longer, length of incrassate part 0-46 or more times width of head including eyes. (Spain, Portugal, western Asia and northern Africa southwards to Chad and Ethiopia) intermedius(p. 192) - Antennal segment I shorter, length of incrassate part less than 0-46 times width of head including eyes. (Europe, western Asia, North Africa from Morocco to Libya) fallen! (Schilling) 3 Head shorter, its length equal to its width including eyes. (Europe, Canary Islands, coastal countries of North Africa, western Asia, Central Asia) waM(Herrich-Schaeffer) - Head longer, its length about 1-2 times its width including eyes. (Turkey and adjacent areas of U.S.S.R.) /atfssimusSeidenstucker Arenocoris intermedius ( Yakovlev)(Fig. 78) Pseudophloeus intermedius Yakovlev, 1883: 101-103. Holotype (sex unknown), U.S.S.R.: Krasnovodsk (Bekker) (IZ) [not examined].Pseudophloeus angustus Reuter, 1891: 139. Holotype $, EGYPT (ZMU) [examined]. [Synonymized by Kirichenko, 1952: 165.]Pseudophloeus gestroi Bergevin, 1930: 32. Holotype cf , LIBYA: Giarabub, 1926-1927 (Confalonieri) (Museo Civico di Storia Naturale, Geneva) [not examined]. Syn. n.Arenocoris intermedius (Yakovlev) Kirichenko, 1952: 165. Length: cf , 6-0-7-2 mm, $,5-6-7-4 mm. Head granulate-tuberculate. Length of head equal to its width inclusive of eyes; antennifers stronglydivergent, their outer apical processes deflexed. Antennae (Fig. 78) with segment I abruptly incrassatefrom a slender base, incrassate part tuberculate; segments II and III slender, weakly and densely granulate;IV shortly ovate-fusiform, nearly as thick as I and with apical sensory area occupying about one-half ofits length. Length of antennal segment I (excluding slender basal part) divided by width of head includingeyes 0-47-0-61. Ratio of lengths of segments (again excluding slender base of I) in male aboutl-00:0-58:2-34:0-80, in female about l-00:0-57:2-25:0-83. Ratio of lengths of rostral segments aboutl-00:0-98:0-52:0-72. Pronotum granulate-tuberculate, lateral margins each bearing three or four prominent tubercles, discwith two longitudinal rows of semiglobular granules, most of these in contact with the neighbouring ones,not standing separately; anterolateral margins distinctly but rather shallowly concave. Scutellum, pleura,femora and veins of corium granulate. Abdomen with laterotergites granulate. Colour pale yellowish brown, usually with darker brown mottling, to piceous with paler markings.Antennal segments I to III usually pale yellowish brown throughout, infuscate in darkest specimens, IIInever darkened at apex, intercalary segment and segment IV always black. Membrane of hemelytracolourless, veins white with short, piceous streaks. REMARKS. Pseudophloeus gestroi was described from a single male taken at the oasis of Jarabub (Giarabub)in Libya by Confalonieri. In his original description, Bergevin (1930: 32-33) compared it with falleni andwaltli but not with intermedius, which suggests that he was not familiar with this last species. His descriptionagrees well with intermedius, especially in the 'almost smooth' antennal segment III and in the shape of thepronotum, which is usually less abruptly widened posteriorly in this species than in most specimens offalleni. In falleni, antennal segments II and III are distinctly more strongly granulate. Bergevin did notmention the length of antennal segment I in relation to the width of the head, which is greater inintermedius than in falleni. A drawing of the holotype of P. gestroi, kindly supplied by Dr R. Poggi,confirms that this species is synonymous with the former rather than the latter. I have seen specimens offalleni from several localities each in Britain, Spain and Greece and from single localities in Austria,Gibraltar and Tunisia (Sbeitla). The specimen of intermedius listed below from Ethiopia has shorterappendages than the others examined. Its antennal segment I was found to be almost as short as in falleni. Itappears to have suffered some damage to one posterior leg and to the abdominal apex during development,which suggests that it may not be typical of the species in Ethiopia. Dr I. M. Kerzhner (pers. comm.)reports two specimens from Addis Ababa in IZ. DISTRIBUTION. Eremic. Spain and Portugal but no other countries in Europe to the north of the TRIBE PSEUDOPHLOEINI 193 Mediterranean Sea; North Africa south to Chad and Ethiopia; Sinai; Iraq; Iran; Turkey; Turkmenia (inpart fide Stichel, 1962: 194). MATERIAL EXAMINED Egypt: 1 $, (holotype of angustus), Sinai, Ai'oun Mousa, ii.1899 (Autran) (data fide Reuter, 1891)(ZMU). Chad: 2 cf, Kanem District, N'Gouri, viii.1958, x-xi.1958 (Renoud) (MRAC). Sudan: 1 cf, UmEnderaba, 2.x. 1927, at light (Johnston) (BMNH). Ethiopia: 1 cf , Simien, Ras Degien, below the pass,over 4300 m, swept from Senecio farinaceus and Helichrysum citrispinum, ll.xii.1952 (Scott) (BMNH). Extralimital material. Spain: 4 $ , Cangas (Champion). Algeria: 3 cf , 2 9 , Biskra (Champion). Tunisia:1 $, Tozeur (Champion); 2 $>, Gafsa (Champion); 2 $, Sfax (de Vauloger). Libya: 1 cf, Cyrenaica,xi.1942. Palestine: 1 cf , 1 $, Gaza district, Deir El-Belah, lO.v.1917, 14.V.1917 (Austen). Iraq: 1 cf , 2 $,Baghdad, xii.1918, 27.xi.1918 (Harwood). Iran: 2 $, Shiraz, Bushire, 1-2.V.1927 (Siyazov). U.S.S.R.: 1Cf, 2 $, Uzbekistan, Termez, 24.V.1912, 5.V.1912, 25.V.1912 (Kirichenko); 2 cf, 1 $, Tadzhikistan,Molotovabad (= Pyandzh), 5-7.iii.1944 (Kirichenko). (All BMNH.) MFLAStalMyla Stal, 1866: 111. Type-species: Myla nigrispina Stal, by monotypy. Body form parallel-sided, length about 3 times breadth, sometimes broader (M. lata), sometimessomewhat depressed (M. niokensis), connexivum narrow to broad. Head about as long as broad, weakly granulate. Antennifers weakly divergent, their outer apicalprocesses porrect. Antennae with segment I varying in length from about 0-50-1-25 times width of headincluding eyes, almost isodiametric throughout its length, weakly granulate; segment III about 1-25 timesas long as II, both segments slender and weakly granulate; segment IV elongate, cylindrical, variable inlength between the species so that in some it is the longest and in others the shortest segment, its specializedsensory area occupying nine-tenths of its length or more. Bucculae occupying about one-third of ventrallength of head. Rostrum at rest reaching posteriorly to base of metasternum. Pronotum strongly declivent anteriorly, disc weakly or very weakly granulate; lateral margins with manysmall granules or a few large granules or tubercles; posterolateral angles scarcely to very strongly producedanterolaterally or laterally, bearing obsolete to long, slender and acute posterolateral spines; posteriormargin weakly convex, smoothly rounded, sometimes appearing slightly trilobed but without distinctemarginations or projections. Scutellum equilateral or slightly elongate, its apex slightly elevated aboveclavi at rest, usually white; lateral margins, especially anteriorly, slightly elevated; disc flat, weaklygranulate-punctate. Mesosternum sulcate throughout. Metasternum strongly convex, sulcate only anter-iorly. Metathoracic scent-gland peritreme with dorsal ridge reniform. Corium with distal margin weakly orvery weakly concave, its apex reaching to level of suture between laterotergites V and VI when at rest.Metathoracic wing with antevannal vein present. Anterior and intermediate femora usually with a single,prominent, subapical granule or tubercle beneath; posterior femur with well-developed basal tubercle andsubapically beneath with three major spines, of which the first (proximal) one is very small, two to foursmall granules or tubercles between the two larger spines and an apical series of four tubercles. Posteriorcoxae separated by a distance equal to approximately half the width of a coxa. Abdominal sterna III to VII with posterolateral angles acute, weakly to strongly produced, lateralmargins of these sternites finely or obsoletely granulate. Male genital capsule posteriorly emarginate, theemargination with a central cusp; parameres with blades flat, their apices not occluding emargination ofcapsule (Fig. 96), proximal tooth of paramere simple, apical tooth long, upcurved. Phallotheca (Figs111-113) comprising a single dorsal and a single ventral sclerite, both broad, each with a narrow, median,posterior projection, ventral sclerite proximally produced dorsad on each side to meet dorsal sclerite.Conjunctiva (Figs 111-114) with dorsomedian lobe membranous, transverse, its lateral angles prominent;distal dorsolateral lobes membranous, arising close together over ejaculatory reservoir, finger- like,reflexed anteriad; apical ventral lobes large, membranous, paired; distal ventrolateral lobes deeply bifid,wings of ejaculatory reservoir complex extending into the dorsal, descending arm of each lobe, ventral,anteriorly directed arm of each lobe membranous and without sclerotized support; ventral lobes lightlysclerotized, wide apart, narrow, anteriorly directed and slightly upcurved; base of vesica withoutprotective sclerites. Sclerites of wall of gynatrium in female each comprising a curved plate parallel to ramibearing a longer, ventral, mesal and a shorter, dorsal, lateral process, both directed posteriorly.Spermatheca (Figs 115, 116) with bulb lunate, duct adjacent to bulb tightly convoluted, connected withgynatrium by an S-shaped portion of the duct. Main pubescence of body, hemelytra, femora and first antennal segment consisting of short, semidecum- 194 W. R. DOLLING bent to adpressed, white, flattened, scale-like hairs. Coloration, probably of all species, variable, basicallygrey-brown with more or less of a yellowish or reddish suffusion. REMARKS. This genus differs from all others in the tribe in the form of the parameres, of which the distaltooth is long and sharply upcurved at right angles to the blade and the apices do not fill the posterioremargination of the genital capsule. The form of the phallotheca is also unique within the tribe. Thescale-like pubescence is sufficient to distinguish Myla species from all other African Pseudophloeinaeexcept Paramyla australis. Myla species are sometimes mistaken for members of the tribe Clavigralliniwhich, however, lack the tubercle that is present in Myla at the base of the posterior femur. Eleven species,all African, are accepted here as members of the genus. For M. concolor Dohrn and M. cornuta Hsiao, seePseudomyla; for M. suspecta Schouteden, see Paramyla. M. schnelli Villiers was transferred to Clavigrallaby Dolling (19790: 34). DISTRIBUTION. Africa south of the Sahara, including Ethiopia. Not so far recorded from any islands or fromArabia. Key to species 1 Length of antennal segment I divided by width of head including eyes 0-84 or less; in doubtful cases width of pronotum across apices of posterolateral spines divided by width of head 2-22 or less 2 - Length of antennal segment I divided by width of head including eyes 85 or more ; in doubtful cases width of pronotum across apices of posterolateral spines divided by width of head morethan2-30 4 2 Length of antennal segment I divided by width of head less than 0-54. (East Africa, rare) d/spar(p. 198) - Length of antennal segment I divided by width of head more than 0-54 3 3 Pronotum (Fig. 85) with posterolateral spines very short. (Mountains of East Africa)ni'olcensis(p. 201) - Pronotum (Fig. 80) with posterolateral spines longer. (Cameroun) onceroma (p. 203) 4 Pronotum (Fig. 89) with posterolateral spines short, laterally directed; tibiae (Fig. 95) with alternate dark and light annuli. (East Africa, rare) /afa (p. 200) - Pronotum (Figs 79, 81-84, 86-88) with posterolateral spines longer, directed anterolaterally; tibiae stramineous, unicolorous 5 5 Pubescence of body (Fig. 98) long, individual scale-like hairs of thorax longer than distance between their insertions. (West and Central Africa) gracilis (p. 204) - Pubescence of body (Fig. 97) short, individual scale-like hairs of thorax shorter than distance between their insertions 6 6 Ocelli (Fig. 81) very large, distance between ocellus and eye less than half diameter of ocellus. (East Africa) somalica (p. 200) - Ocelli (Figs 79, 82) smaller, distance between ocellus and eye equal to or greater than diameter of ocellus 7 7 Head (Fig. 82) elongate, apically acute, its midline fuscous; eyes small. (Widespread in Africa) microphthalma(p. 202) - Head (Fig. 79) short , apically obtuse , its midline not fuscous ; eyes larger 8 8 Lateral margins of pronotum (Fig. 86) bearing large, prominent granules. (Southern Africa, rare) granula (p. 198) Lateral margins of pronotum (Figs 79, 84, 87, 88) with smaller granules 9 9 Posterolateral angles of pronotum and abdominal sternites (Fig. 92) less prominent. (Ethiopia) abyssinica(p. 197)Posterolateral angles of pronotum (Figs 79, 84, 87, 88) and abdominal sternites (Fig. 91) more prominent 10 10 Paramere (Figs 99, 100) with apical tooth projecting beyond apex of shaft; second valvula (Fig. 119) apically rounded. (West and Central Africa) hoploxys(p. 194) Paramere (Fig. 104) with apical tooth not projecting beyond apex of shaft; second valvula (Fig. 120) obliquely truncate, with apex acute. (Central, eastern and southern Africa) . . . calida(p. 197) Myla hoploxys (Dallas)(Figs 79, 84, 91, 96, 99, 100, 119)Clavigralla! hoploxys Dallas, 1852: 515. Holotype cf , GAMBIA (BMNH) [examined]. TRIBE PSEUDOPHLOEINI 195 79 Figs 79-94 My la species. 79-82, dorsal view of head and pronotum of (79) hoploxys; (80) onceroma; (81)somalica; (82) microphthalma. 83-89, dorsal view of pronotum of (83) gracilis , lectotype; (84) hoploxys;(85) niokensis; (86) granula; (87) calida from Zaire: Mpese; (88) calida, holotype from Uganda; (89)lata, holotype. 90-94, ventral view of abdominal margin of (90) niokensis; (91) hoploxys; (92) abyssinica;(93) lata, holotype; (94) gracilis . 196 W. R. DOLLING Myla nigrispina Stal, 1866: 111. LECTOTYPE $, 'GUINEA' (NMV), here designated [examined]. Syn. n. Myla hoploxys (Dallas) Stal, 1873: 84. Myla hoploxys nigrispina Stal; Linnavuori, 1971: 176. Length: cf, 9-8-11-7 mm; $,9-7-12-0 mm. Head (Fig. 79) slightly longer than wide, broadly rounded anteriorly. Length of antennal segment Idivided by width of head including eyes in male 88-1 16 , in female 89-1 09 . Ratio of lengths of antennalsegments in male about l-00:0-80:l-03:0-96, in female about l-00:0-84:l-04:0-85. Length of rostralsegment I divided by width of head including eyes in both sexes 0-69-0-81, ratio of lengths of rostralsegments about 1 -00:0-84:0-43 :0-64. Pronotum (Figs 79, 84) with posterolateral angles produced anterolaterally and terminating in shortspines, lateral margins very weakly granulate. Width across apices of posterolateral spines divided by widthof head including eyes in male 2-14-2-84, in female 2-20-2-92. Length of posterior tibia divided by that ofposterior femur 0-89-1-06. Abdomen rather narrow, its lateral margin (Fig. 91) weakly convex, posterolateral angles of sternitesIII- VII projecting as fairly short, narrow, posteriorly directed spines; abdominal laterotergites narrow,laterotergite V usually at least twice as long as wide, rarely slightly less. Paramere (Figs 96, 99, 100) withapical tooth long, upcurved, its posterior margin projecting distinctly beyond apex of shaft. Aedeagus verysimilar in all respects to that of M. calida, q.v. Female with second valvula of ovipositor (Fig. 119)terminating in an apically rounded lobe. Pubescence of head, all exposed parts of thorax, abdominal laterotergites, antennal segment I, femora,clavus and corium composed of short, flattened, scale-like, decumbent or adpressed, white hairs, each hairmuch shorter than the distance between its own insertion and that of its nearest neighbour, each insertionborne on a large granule on the head and a small granule elsewhere, granules of thoracic pleurites,posterior lobe of pronotum, scutellum and hemelytra each positioned on the anterior (or proximal onhemelytra) border of a large puncture. Pubescence of disc of abdominal venter suberect, not scale-like,grading laterally into white, scale-like type. Pubescence of antennal segments II to IV, tibiae and tarsisuberect, rather short, not scale-like. General coloration of body and appendages greyish yellow, more or less heavily suffused red. Antennalsegments II and III, tibiae and tarsi stramineous. Apex of rostral segment IV, posterolateral angles andspines of pronotum and often a line along lateral margins of pronotum piceous. Membrane of hemelytrawhitish hyaline, its veins brown with frequent whitish hyaline interruptions. Elevated apex of scutellumand dorsal ridge of metathoracic peritreme white, the latter often margined piceous. REMARKS. This and the next four species form a group of closely related taxa which are probably all rathervariable in pronotum shape, size and coloration. Examination of the genitalia is the only completelyreliable way of distinguishing between this species and M. calida, with which it is partly sympatric.Although the genitalia of the female lectotype of M. nigrispina were not dissected out, the specimencorresponds in every other way with the West African species recognized here as M. hoploxys. No otherspecies of this complex was detected in the area of the type-locality of M. nigrispina ('Guinea', = WestAfrica). There is no justification for Linnavuori's (1971: 175, 176) recognition of two, largely sympatricsubspecies distinguished by size, colour and some slight, probably allometric structural features. Large,red individuals with long posterolateral pronotal spines predominate in the western part of the range of thespecies and are rare in the eastern part. This species includes cowpea (Vigna) in its diet and, according toProfessor A. E. Akingbohungbe (pers. comm.), it is approaching the pest status of the Clavigralla specieson this crop. DISTRIBUTION. Proven distribution covers West Africa from Senegal and Gambia eastwards to Zaire andAngola. Linnavuori (1978: 36) records a single specimen from Sudan. MATERIAL EXAMINED Gambia: 1 cf (holotypeof hoploxys), (no other data) (BMNH). 'Guinea': 1 $ (lectotype of nigrispina)(coll. Signoret), (NMV). Gambia: 1 cf (no further locality data; pin and locality label identical with those of holotype of C.hoploxys but also with label: 'Saunders Coll.'); 1 $ , 3.iii.l911 (Simpson) (BMNH). Senegal: 1 cf , Bambey,6.V.1943 (Risbec) (BMNH); 1 Cf, 2 $, M'Bambey, 28.x-6.xi. 1939 (Risbec) (MRAC). Ivory Coast: 4 cf ,Lamto, 7.V.1963, 14.V.1963, 28.V.1963, 18.vi.1963; 3 $, Lamto (Toumodi), 18. ii. 1964, 24.vii.1964; 1 $>,Bouake, 21.vii.1962 (probably all leg. Gillon) (BMNH); 2 cf, Bouake, ii.1968 (Schmitz) (MRAC).Ghana: 1 $, Tafo, 8.xi.l965 (Leston); 1 cf, Labadi, 12.ix.1965 (Lestori) (BMNH); 1 cf, Namgua,30.ii.1964, oncowpeas; 1 cf , 1 $, Sokode, 7.xi.l967 (Leston); 1 $, (locality illegible), 8.xii.l965 (Leston);1 $ , Legon, 22.ix.1968, in u.v. trap (Kumar); 1 $, Abouadi, 7.xi.l976 (Leston) (UG). Mali: 1 cf , 40 miles TRIBE PSEUDOPHLOEINI 197 (64 km) SW. of San, 300 m, 26.viii.1966 (Ross, Lorenzen) (CAS). Nigeria: 1 cf, Ibadan, 30.xi.1976(Deeming) (IAR); 1 cf, He-Ife, 5.xii.l970 (Medler); 1 $, Ile-Ife, 14.ii.1974 (Akingbohungbe); 1 $,Badeggi, 19.iii.1972 (Medler) (He University); 1 $, Azare, 1928-1929 (Lloyd); 1 cf , 1 $, Zaria, Samaru,9.viii.l966, 21. ix. 1967 (Deeming); 1 $, Zaria, Samaru, 14. ix. 1966, on cowpea (Vigna) (Deeming); 1 9,Samaru, ll.iii.1955, at grass roots (Emsley); 4 cf , 1 $, Samaru, 7-14. vii. 1970, 28.viii.1970, 8-15.ix.1970(Ward); 1 $, Ilorin, 19.iii.1955, in dry humus (Emsley); 1 $, Jos, 9.X.1955 (Bechyne); 1 cf , Ibadan, DFRnursery, yellow tray trap, 30. ix. 1963 (White); 1 cf , Enugu, 22.x. 1955 (Bechyne); 1 $, Lagos 'hinterland'(Rowland) (BMNH); 1 cf , 1 $, Jos province, 7-26.iv.1963 (Meussen, Bouquiaux) (MRAC). Zaire: 1 $,Bambesa, 22.U938 (Vrydagh); 1 $, Katanga, Nyonga (? = Nyanga), v.1925 (de Witte); 1 $, Wombali,vii.1919 (Vanderijst); 1 cf , 2 $, Lulua, Kapanga, iii.1932, iii.1933 (Overlaet) (MRAC). Angola: 1 cf , Joaode Almeida, 29.iii.1972 (BMNH Expedition) (BMNH). 'Tropical Africa': 2 cf (UM). Myla abyssinica Linnavuori(Figs 92, 97, 102) Myla abyssinica Linnavuori, 1971: 175. Lectotype cf , ETHIOPIA: near Nazareth, 20-21.vi.1963 (Linna-vuori) (R. Linnavuori coll.), designated by Linnavuori, 1978: 36 [not examined]. Length: cf , 9-9-10-1 mm; $ unknown. Very similar to M. hoploxys. Length of antennal segment I divided by width of head including eyes0-88-0-93. Ratio of lengths of antennal segments about 1-00:0-80: 1-05: 1-04. Length of rostral segment Idivided by width of head including eyes 0-69-0-75. Ratio of lengths of rostral segments about1 -00:0-84:0-45 :0-64. Pronotum with posterolateral spines as short as in the shortest-spined examples of M.hoploxys; width across apices of spines divided by width of head including eyes 2-16-2-21. Connexivumnarrow, laterotergite V about 2-5 times as long as wide. Posterolateral spines of sternites III to VII short(Fig. 92). Length of posterior tibia divided by length of posterior femur 0-96. Paramere (Fig. 102) withapical tooth relatively shorter than that of M. hoploxys. REMARKS. This species was described by Linnavuori (1971: 175) in a dichotomous key. Subsequently, thesame author (Linnavuori, 1978: 21, 36, 103, figs 16f, 16g, 69d) provided a fuller description, with figures,suggesting at the same time that it might be only a race of M. hoploxys. In his 1978 publication, hedesignated his unique specimen as holotype. As he already validly described the species in 1971, withoutdesignating a type-specimen, his 1978 action must be regarded as a lectotype designation. In view of thesmall differences separating species in this genus it seems appropriate to accord full specific status to thistaxon, especially as the two specimens examined here are very similar to the one described by Linnavuori.The strong reddish suffusion noted by Linnavuori (1978) in the type-specimen is not apparent in the twospecimens I have seen. DISTRIBUTION. Endemic to Ethiopia on the available evidence. MATERIAL EXAMINED Ethiopia: 1 cf , Djoudjou Abayi, Didessa R., Goma, 2000 ft (600 m), lO.v.1905 (Zaphiro) (BMNH); 1Cf, Gemu-Gofa Prov., 45 km N. of Demika, near Jinka road, 1460 m, 30. iv. 1974 (de Rougemont)(MRAC). Myla calida sp. n. (Figs 87, 88, 101, 111-114, 120) Length: cf , 9-0-10-6 mm; $ , 9-0-10-3 mm. Very similar to M. hoploxys. Length of antennal segment I divided by width of head including eyes inmale and female 0-92-1-05. Ratio of lengths of antennal segments in male about l-00:0-75:0-95:0-94, infemale about l-00:0-80:0-95:0-82. Length of rostral segment I divided by width of head including eyes inmale 0-66-0-75, in female 0-69-0-79. Ratio of lengths of rostral segments in male aboutl-00:0-85:0-43:0-64, in female about 1-00:0-85:0-42:0-61. Pronotal shape variable (Figs 87, 88), posterolateral spines on average more strongly divergent thanthose of M . hoploxys; width across apices of posterolateral spines divided by width of head including eyesin male 2-30-2-85, in female 2-31-2-62. Granulation of lateral margins of pronotum slightly morepronounced than in M. hoploxys. Length of posterior tibia divided by length of posterior femur 0-89-0-95. Connexivum and abdominal spines as in M. hoploxys. Paramere (Fig. 101) with apical tooth shorter thanthat of M. hoploxys, its posterior margin not or scarcely projecting posteriad of apex of shaft. Aedeagus 198 W. R. DOLLING (Figs 111-114) of the form usual in the genus. Ovipositor with second valvula (Fig. 120) obliquely truncate,apex acute and slightly upturned. Coloration on average more yellowish than greyish, never as deeply suffused red as the most extremeexamples of M. hoploxys. REMARKS. The second valvula of this species is very distinctive. The shape of the paramere is rather variablebut it is never so long as that of M. hoploxys, nor is its apical tooth so long or so prominent. The shape of thepronotum is also variable (compare Figs 87, 88), as is the colour. In many individuals there is a conspicuousblack line along the lateral margins of the pronotum. DISTRIBUTION. Eastern, central and southern Africa. Absent from West Africa. Northern limit unknown. MATERIAL EXAMINED Holotype cf , Uganda: W. shores of Vic. Nyanza, Buddu, 3700 ft (1100 m), 19-25.ix.1911 (5. A. Neave)(BMNH). Paratypes. Zaire: 1 cf, 1 $, Ngowa, 5.vi.l939, 16.vi.1939 (Mertens); 1 cf, Kwango, Ngowa, 3.X.1937(Mertens); 1 C? , 1 ?, Ngowa-Kwango, no date and 17. ii. 1938 (Mertens); 1 cf , Kibangula, 1957 (Henry); 1Cf, 2 $, Mpese, 7-26.vi.1937 (Cooreman) (IRSNB); 1 tf, Sankuru, M'Pemba Zeo (Gandajika),17.vii.1958 (Marechaf); 1 cf , Kwango, Popokabaka, v.1952 (Pierquin); 1 cT, Kasai, Ilebo (Thery); 1 cf ,Bokala, 20.V.1915 (Mayne); 1 cf , Tolo, early xii.1913 (Maes); 1 cf , Kivu, Kavimvira (Uvira), vi.1955, atlight (Marlier); 1 cf, Kinshasa (Tynant); 1 cf, Lita, 25.V.1912 (Christy); 1 cf, 'Dans le ebenal. PostTelegraphiq au dessous de Block river, 22 Mai 1909. Herbes et arbustes a 3-4 au dessus du fleuve, endroitnon boise, 1 a 17 heures' (Voyage de S.A.R. le Prince Albert); 1 $ , Kisantu, 1925 (Varderijst); 2 $ , Lulua,Kapanga, i.1933, v.1933 (Overlaet); 1 $, Sankuru, M'Pemba Zeo (Gandajika), 28.iv.1960 (Marechaf); 1$, Mayidi, 1942 (van Eyen); 1 ex. without abdomen, Bas-congo, Lemfu, vi.1945 (de Beir) (MRAC); 6 cf ,13 $, Faradje, 2940'E, 340'N (Lang, Chapin) (AMNH). Zambia: 1 cf, 1 $, High Plateau, L.Tanganyika, 4500 ft (1350 m), 1 8-2 l.viii. 1908 (Neave) (UM). Angola: 1 cf (other data illegible) (BMNH).Zimbabwe: 1 cf , Salisbury, v.1963 (NMB); 1 $ , Umtali (BMNH). South Africa: 1 cf , Durban, 1902 (Muir)(BMNH). /V/v/a d/sparsp. n. (Fig. 104) Length: cf , 9-6 mm; $ , unknown. Resembles M. hoploxys in most respects, including the shape of the paramere (Fig. 104), whichdistinguishes it from M. calida. Lateral margins of pronotum rather more strongly granulate than those ofM. calida and posterolateral spines of pronotum short. Length of antennal segment I divided by width ofhead including eyes 0-53. Ratio of lengths of antennal segments as 1-00:0-82:1-06:1-18. Width of pronotumacross apices of posterolateral spines divided by width of head including eyes 2-22. Length of posterior tibiadivided by that of posterior femur 0-92. REMARKS. Perhaps this is an outlying representative of M. hoploxys but it differs from that species and fromM. calida in the relative lengths of the antennal segments, of which I and II are very short and III and IVlonger with respect to I. DISTRIBUTION. Zambia. MATERIAL EXAMINEDHolotype cf , Zambia: L. Bangweolo, Chirui Island, 3800 ft (1140 m), 5-7. vii. 1908 (Neave) (UM). Myla granula sp. n. (Figs 86, 121) Length: cf , unknown; $, 10-2-10-4 mm. Resembles M. hoploxys and M. calida in general form. Length of antennal segment I divided by width ofhead including eyes 0-85-0-93. Ratio of lengths of antennal segments in holotype as 1-00:0- 84: 1-00:0-95, inparatype as 1-00: 0-82: 0-92:0-91. Length of rostral segment I divided by width of head including eyes0-66-^0-69. Ratio of lengths of rostral segments in holotype as 1-00:0-87:0-40:0-60, in paratype as1-00:0-94:0-44: 0-67. Pronotum (Fig. 86) with posterolateral angles produced laterally almost perpendicu-lar to axis of body, posterolateral spines very short, width across apices of spines divided by width of headincluding eyes 2-34-2-37; lateral margins bearing very large granules. Length of posterior tibia divided by TRIBE PSEUDOPHLOEINI 199 95 97 98 99 101 102 103 104 106 107 108 109 Figs 95-110 My la species. 95, lata, posterior tibia. 96, hoploxys, posterior view of male genital capsulewith parameres. 97, 98, detail of pubescence of left metapleuron of (97) abyssinica; (98) gracilis. 99,hoploxys, dorsomedial view of left paramere. 100, hoploxys, lateral view of same. 101, 102, dorsomedialview of left paramere of (101) calida; (102) abyssinica. 103, somalica, dorsomedial view of rightparamere, 104-110, dorsomedial view of left paramere of (104) dispar; (105) microphthalma; (106)niokensis; (107) lata; (108) gracilis from Zaire: Ngowa; (109, 110) gracilis, two specimens from Zaire:Faradje. 200 W. R. DOLLING length of posterior femur 0-95-0-99. Ovipositor with second valvula (Fig. 121) symmetrically taperingtowards narrowly rounded apex. REMARKS. This species differs from M. hoploxys and M. calida mainly in the shape and granulosity ofpronotum. The second valvulae of the ovipositor are characteristic; males are unknown. DISTRIBUTION. South Africa. MATERIAL EXAMINED Holotype $, South Africa: Natal, Malvern, ll.vi.1897 (BMNH). Paratype. South Africa: 1 $ , Natal, Lewombo Mission, 3140'E, 2835'S, 15.x. 1977, 365 m, under brickpile(/?eave//)(NMP). Myla somalica Linnavuori(Figs 81, 103) Myla somalica Linnavuori, 1982: 14-16. Holotype cf , SOMALI REPUBLIC: Giohar, 18-20. iv. 1968 (Spedi-zione Biologica in Somali del Centra di Studio per la Faunistica ed Ecologia Tropicali del ConsiglioNazionale delle Ricerche) (Museo Zoologico delPUniversita di Firenze) [not examined]. Length: cf, 10-5-11-5 mm; $ unknown. An elongate species related to M. calida. Ocelli very large (Fig. 81). Length of antennal segment Idivided by width of head including eyes 0-85-0-99. Ratio of lengths of antennal segments about1-00:0-82:0-97: 1-08. Length of rostral segment I divided by width of head including eyes 0-70-0-71. Ratioof lengths of rostral segments about 1 -00:0-87:0-42:0-66. Pronotum (Fig. 81) with posterolateral angles shortly produced laterally, with slight anteriad inclination;width across apices of posterolateral spines divided by width of head including eyes 2-31-2-44. Posteriortibia long, its length divided by that of posterior femur 0-96-1-03. Paramere (Fig. 103) with shaft long, as in M. hoploxys, teeth of blade short, as in M. calida. REMARKS. This species is readily distinguished from all others of the genus by its enormous ocelli. Theparatype was received after the above measurements were taken. It is 9-8 mm long, its first antennalsegment is 0-90 times the width of the head and the antennal ratio is 1-00:0-79: 1-07: 1-08. The paramere isindistinguishable from that of the inland specimens with which it seems to be conspecific despite thedifference in habitat. DISTRIBUTION. Highlands of eastern central Africa to the Somali coast. MATERIAL EXAMINED Somali Republic: 1 cf (paratype). Giohar, 18-20.iv.1968 (L. Simonettd) (R. Linnavuori coll.). Kenya: 1Cf , Nakuru, viii.1912 (van Someren) (BMNH). Uganda: 1 ex. without abdomen (probably cf), Chisinga,viii.1911 (Marshall) (BMNH). Zaire: 3 cf, Pare National Albert, Camp Ruindi (or Rwindi), 1000 m,20-28.xi.1934 (de Witte); 2 cf , P.N.A., Cp Ruindi, 13.ix.1932 (Burgeon) (MRAC). Myla lata sp. n. (Figs 89, 93, 95, 107) Length: cf , 9-8-10-8 mm; $ unknown. A very broad-bodied species. Length of antennal segment I divided by width of head including eyes inholotype 0-88. Remainder of antennae in holotype and whole of antennae in paratype missing. Length ofrostral segment I divided by width of head including eyes in paratype 0-70; ratio of lengths of its rostralsegments as l-00:0-80:0-45:0-63; rostrum not visible in holotype mount. Pronotum (Fig. 89) very broad, lateral margins strongly granulate, posterolateral angles weaklyproduced, terminating in small, laterally directed spines; width across apices of posterolateral spinesdivided by width of head including eyes 2-52-2-56. Abdomen broad, its lateral margins (Fig. 93) strongly convex, posterolateral angles of sternites III to VIIproduced into broad, triangular spines. Connexivum broad, laterotergite V about five-sixths as broad aslong. Paramere (Fig. 107) with both teeth of blade on a common stem. Coloration ferruginous; all tibiae stramineous with basal, median and apical annuli piceous (Fig. 95). REMARKS. This species is very distinctive by reason of its great breadth. It is the only species of the genus inwhich the tibiae are conspicuously bicolorous. TRIBE PSEUDOPHLOEINI 201 DISTRIBUTION. Central Africa. MATERIAL EXAMINED Holotype cf , Zambia: N. Lake Bangweolo, 4200 ft [1260 m], Luwingu, 21.vii.1908 (Neave) (UM).Paratype. Zaire: 1 cf , Katanga, Lufira River, 3500 ft [1050 m], 10.ix.1907 (Neave) (BMNH). Myla niokensis Schouteden(Figs 85, 90, 106) Myla niokensis Schouteden, 1938: 294. LECTOTYPE cf , ZAIRE (MRAC), here designated [examined].Myla (Paramyla) niokensis Schouteden; Linnavuori, 1971: 178. Length: cf, 8-2-9-3 mm; $,8-7-9-7 mm. Head similar in shape to that of M. hoploxys. Antennae, especially segment I, short. Length of antennalsegment I divided by width of head including eyes in male 0-56-0-74, in female 0-55-0-69. Ratio of lengthsof antennal segments in male about l-00:0-95:l-14:l-48, in female about l-00:l-04:l-09:l-33. Length ofrostral segment I divided by width of head including eyes in both sexes 0-64-0-75 . Ratio of lengths of rostralsegments about l-00:0-85:0-44:0-64. Pronotum (Fig. 85) with posterolateral angles produced forward, their apices directed laterally but veryweakly produced in that direction, terminating in very short spines, margins behind the produced anglesconcave, diverging posteriorly so that in most cases greatest width of pronotum is not across apices ofposterolateral spines but across posterior lobe; width across apices of posterolateral spines divided bywidth of head including eyes in male 1-88-2-16, in female 2-00-2-22. Length of posterior tibia divided bylength of posterior femur in both sexes 0-94-1-06. Abdomen with lateral margin (Fig. 90) moderately convex; posterolateral angles of sternites III to VIweakly prominent. Connexivum moderately broad, laterotergite V about 1-7 times as long as wide.Paramere (Fig. 106) very similar to that of M. hoploxys. Colour yellow-brown with punctures of pronotum and hemelytra and often parts of underside andposterior femora piceous or black; body and appendages often with a reddish tinge. The darkest species ofthe genus. Pubescence as in M. hoploxys. Granules of lateral margins of pronotum about as prominent asthose of M. hoploxys. REMARKS. This species may easily be recognized by the form of the pronotum, except that it might beconfused with M. abyssinica from which it differs in that its antennal segment I is much shorter and segmentIV is very long compared with I. Specimens from Zaire have the apical tooth of the paramere slightlyshorter than it is in those from Ethiopia and Kenya. Ethiopian specimens differ from the others in having aslightly longer rostral segment IV. Antennal segment I is shortest, both in proportion to head width and inproportion to the other segments, in Kenyan specimens and longest in those from Zaire. The numbers ofspecimens available are too low for it to be certain that these distinctions are not artefacts of sampling. DISTRIBUTION. Records suggest that the species occurs at high altitudes in three separate areas: Ethiopia;near Nairobi in Kenya; and on the common borders of Zaire, Uganda, Tanzania and Rwanda. MATERIAL EXAMINED Zaire: 1 cf (lectotype), Ituri, Nioka, 7.vii.l934 (Leroy) (labelled: 'holotypus') (MRAC). Ethiopia: 1 cf , Mt Chilalo, forest, ca 8500 ft (2550 m), 24.xi.1926 (Scott); 1 cf , Gojjam, Fasilo Bahr Dar,scrub, 2.viii.l965 (Kitching); 1 cf, Addis Ababa, Filoa, vii.1941 (Meneghetti); 1 $, Simien, AtghebaGhiyorghis, ca 10,900 ft (3270 m), 4.xii.l952, arable land (Scott); 1 ? , Urgessa R., Jimu, 2300 ft (690 m),3.V.1905 (Zaphiro); 1 $, Mt Zuqala, in crater, ca 9000 ft (2700 m), 26.X.1926 (Omer Cooper) (BMNH).Zaire: 3 $ (paralectotypes), Ituri, Nioka, vii.1934 (Leroy); 1 $ (paralectotype), Ituri, Nioka, 7.vii.l934(Leroy); 1 cf , Nioka, vii.1934 (Leroy); 1 cf , Tshamgussa, 8-15. viii. 1934 (de Witte); 1 $, Pare Nat. Albert,Lac Gando, 2600 m, Mt Tamira, ll.iii.1935 (de Witte); 1 cf , Kivu, Mulungu-Tshibinda, xi.1951 (Lefevre)(MRAC); 1 $, Ituri Forest, 40 miles (64 km) NNE. of Beni, 3000 ft (900 m), 10-12.ix.1959, indigenousforest with cultivated patches (Cambridge Expedition); 1 $ , Butembo-Beni road, 2930'E, 025'N, 4000 ft(1200 m), ll.ix.1959, agricultural district (Cambridge Expedition) (BMNH). Uganda: 1 cf, KasingaChannel, Katunguru, 1931 (Worthingtori) (BMNH). Kenya: 1 cf , Limuru, 10.iii.1911 (Anderson); 1 cf ,Kinangop, iii.1930 (Turner); 2 cf , W. Aberdares, above 9000 ft (2700 m), iii-iv.1934 (Turner); 1 $, SE.slopes of Mt Kenya, 6000-7000 ft (1800-2100 m), 3-12.ii.1911 (Neave) (BMNH). Rwanda: 1 cf,Rubengera, 1900 m, terr. Kibuye, 12.ii.1953 (Basilewsky) (MRAC). Tanzania: 1 $, Mt Mem, Olkokola,facing NW. , 2800 m, 24.vi-l.viu.1957 (Basilewsky, Leleup) (MRAC). No data: 1 cf (ex coll. Schouteden)(MRAC). 202 W. R. DOLLING Myla microphthalma Linnavuori (Figs 82, 105, 115, 117) Myla microphthalma Linnavuori, 1971: 175, 176-177. Holotype cf , IVORY COAST: Lamto, 3.iii-21.iv.l967(Gillon) (R. Linnavuori coll.) [not examined]. Length: cf, 8-3-9-7 mm; $,8-6-9-9 mm. Head (Fig. 82) more elongated and more acutely pointed anteriorly than is usual in the genus, eyes small.Length of antennal segment I divided by width of head including eyes in males from the type-locality (IvoryCoast) 0-91-0-97, in females from type-locality 0-90-0-97, in males and females from elsewhere respective-ly 0-76-0-93 and 0-76-0-91. Ratio of lengths of antennal segments in males and females from Ivory Coastrespectively about l-00:0-86:l-05:0-98 and l-00:0-88:l-04:0-85; in males and females from elsewhererespectively about 1-00:0-88:1-05:1-17 and 1-00:0-90:1-05:1-03. Length of rostral segment I divided bywidth of head including eyes in both sexes 0-72-0-84; ratio of lengths of rostral segments aboutl-00:0-85:0-44:0-67. Pronotum (Fig. 82) with posterolateral angles rather strongly produced anterolaterally and lateralmargins finely granulate with a few larger granules; width of pronotum across apices of posterolateralspines divided by width of head including eyes 2-15-2-65, not differing significantly between the sexes.Length of posterior tibia divided by length of posterior femur 0-89-0-98. Abdomen with posterolateral angles of sternites III to VII produced as short, acute spines. Paramere(Fig. 105) with blade broad, distal margin of apical tooth evenly curved, apical tooth rather longer thanbasal tooth. Spermatheca (Fig. 115) with bulb short, duct tightly convoluted. Second valvula of ovipositor(Fig. 117) apically narrow, elongate, parallel-sided. Coloration reddish or greyish brown, dorsal midline of head always with median fuscous stripe;pronotum with anterolateral margins, posterolateral spines and anterior half of midline fuscous. Colourpattern otherwise and pubescence as in M. hoploxys. REMARKS. The fuscous midline of the head distinguishes this species from all others in the genus. The seriesfrom the type-locality (Ivory Coast: Lamto, Toumodi) was found to differ from the rest of the materialexamined in several biometric ratios, as indicated above. A particularly striking difference was found in theratio of the lengths of antennal segments I and IV. The length of segment IV divided by that of segment I infive males and two females from the Ivory Coast was, respectively, 0-91-1-00 and 0-84-0-86; in 12 malesand six females from the rest of Africa this value was, respectively, 1-02-1-30 and 1-00-1-10. Although theIvory Coast sample size is small, the measurements of this ratio fall completely outside the rangedetermined for specimens from the rest of Africa. It may be that the Ivory Coast population has beenisolated from the remainder for a considerable period. Perhaps an ancestral population of this species wasisolated in mountain refugia in east Central Africa and Guinea and has re-invaded the lower-lying areasfrom these two centres. One male is omitted from the description above because of its even more elongatehead and its aberrant measurements. It is from Zaire: Mongende; its length is 9-0 mm; the width of itspronotum, length of antennal segment I and length of rostral segment I, all divided by the width of thehead, are respectively 2-82, 1-03 and 0-84. The lengths of the four antennal segments are in the ratio of1-00:0-84:0-97:1-00. DISTRIBUTION. The species is known from two widely separated areas: (1) Ivory Coast and (2) southernAfrica, avoiding desert and dense forest and extending northwards to 4S in the west and 2N in theequatorial mountains of the east. MATERIAL EXAMINED Ivory Coast: 6 cf, 3 $, Lamto (Toumodi), 5.vi.l962, 12.vi.1962, 10.vii.1962, 4.vi.l963, ll.vi.1963,27.viii.1963, 21.vii.1964 (Gillon); 1 $, Bouake, 21.vii.1962 (Gillon) (BMNH). Kenya: 1 cf, Kaimosi,iii-iv.1932 (Turner) (BMNH). Rwanda: 1 cf (paratype), Kibungu, 1500 m, 2.ii.l953 (Basilewsky)(MRAC). Urundi: 1 cf (?paratype - locality but not date cited in original description), Kitega, xi.1935(Lefevre); 1 $ (paratype), Kitega, vii.1935 (Lefevre); 2 $ (paratypes), Usumbura, 1934 (Lefevre)(MRAC). Zaire: 1 $, Uele, Wango, 29.vii.1931 (Bredo); 1 $, Ituri, Nioka, vii.1934 (Leroy); 1 cf , Kivu,Mulungu, 1938 (Hendrickx); 1 cf, Mayidi, 1942 (van Eyen); 1 cf, Kisantu, 1919 (Vanderijst) (allparatypes); 1 cf , Tshambi, xii.1934 (de Witte); 1 $, Reg. Thysville (Bas-Congo), 1959-1963 (Michaux); 1$, Lulua, Kapanga, i.1933 (Overlaet); 1 cf, Katanga, Kipopo, 29.X.1961 (Marechal) (MRAC); 2 $>,Elisabethville (= Lubumbashi), 25.iv.1939, 15.xi.1939 (Bredo) (IRSNB). Zambia: 1 cf, KamankundjuRiver, Ikelenge, N. Mwinilunga, 5.V.1972 (Pinhey, de Moor) (NMB). Angola: 1 cf , 12 miles (19 km) SW.of Luimbale, ca 1680 m, 20-2 l.iii. 1972, general sweeping (BMNH Southern African Expedition)(BMNH). Malawi: 1 cf , SE. shore Lake Nyasa, between Fort Maguire and Fort Johnston, 6-17. iii. 1910 TRIBE PSEUDOPHLOEINI 203 111 112 114 118 115 Figs 111-121 Myla species. 111-114, calida: (111) dorsal view of phallotheca, conjunctiva and vesica;(112) lateral view of same, dorsal to left; (113) ventral view of same; (114) apical view of same. 115, 116,spermatheca of (115) microphthalma; (116) onceroma. 117-121, left lateral view of apex of left secondvalvula of ovipositor of (117) microphthalma; (118) onceroma; (119) hoploxys; (120) calida; (121)granula. (Neave) (BMNH). South Africa: 1 cf , Transvaal, Pretoria, 26.xi.1973 (Jacobs); 1 cf , Utrecht, 4.xi.l967(Kroon); 1 $, Camperdown, 17.iv.1908 (Leigh) (TM); 1 cf , Transvaal, 10 miles (16 km) NE. of Cullinan,27.ii.1962 (Capener); 1 <j>, no precise data (Capener) (J. A. Slater coll.); 1 cf, Umbilo (Bevis); 2 $,Transvaal, Argent, 29.iii.1942 (Capener); 1 $, Pondoland, Port St John, 15-31.V.1923 (Turner); 1 cf,Natal, Tugela River, nrWeenen,iii. 1897; 1 cf, Natal, Karkloof, 10.ii.l897(?or2.x.l897); 1 cf, 1 $, Natal,Howick; 1 cf , Natal, Estcourt; 3 cf , 2 $, Estcourt, xi.1896; 4 cf , Estcourt (Marshall) (BMNH). Possiblybelonging to this species: Zaire: 1 cf , Mongende, 17. iv. 1921 (Schouteden) (falsely labelled as a paratype ofMyla gracilis Schouteden but not from a locality mentioned in the original description) (MR AC). Myla onceroma sp. n. (Figs 80, 116, 118) Length: cf , 8-2-9-2 mm; $,8-6 mm. Head (Fig. 80) not elongate or acute anteriorly. Eyes small. Length of antenhal segment I divided bywidth of head including eyes in male 73-0 84 , in female (one only seen) 78 . Ratio of lengths of antennalsegments in male about 1-00:0-83:1-02:1-19, in female as 1-00:0-81:1-04:1-04. Length of rostral segment Idivided by width of head including eyes in male 0-70-0-75, in female 0-78. Ratio of lengths of rostralsegments in male about l-00:0-82:0-43:0-65, in female as l-00:0-77:0-42:0-63. Pronotum (Fig. 80) with posterolateral angles shortly produced anteriorly, bearing short spines; widthacross apices of spines divided by width of head including eyes in male 1-99-2-20, in female 2-03. Length ofposterior tibia divided by length of posterior femur 0-94-0-97. Abdominal sternites III to VII with posterolateral angles produced into short, acute spines. Malegenitalia similar to those of M. microphthalma. Female with bulb of spermatheca (Fig. 116) longer thanthat of M. microphthalma and with apical projection of second valvula (Fig. 118) very short. General coloration dark red-brown or brown. Midline of head and pronotum concolorous, withoutdarker median stripe. REMARKS. This small species is probably most closely related to M. microphthalma but may be disting-uished from it by the much shorter antennae, shorter head and much less strongly produced pronotalangles, which approach the form of those of M. niokensis but bear longer apical spines. It is a montane 204 W. R. DOLLING form, perhaps descended from the ancestral population that gave rise to M. microphthalma but is nowgreatly specialized, particularly in the form of the second valvulae of the ovipositor, after long isolation. DISTRIBUTION. Known only from three localities at high altitudes in Cameroun and south-eastern Nigeria. MATERIAL EXAMINED Holotype cf , Cameroun: Bamenda, 17.xii.1955 (Bechyne, Exped. Mus. G. Frey} (BMNH). Paratypes. Cameroun: 4 cf, as holotype but dates 7.xii.l955, 8.xii.l955, 13.xii.1955 and 17.xii.1955(BMNH); 2 Cf , 1 $ , Mt Bamboutos (MRAC). Non-paratypic material (received after description was drawn up). Nigeria: 3 d", 4 9 , SE. State, ObuduC.R., 21 and 23.iii.1971 (Medler) (University of Ibadan). Myla gracilis Schouteden(Figs 83, 94, 98, 108-110) Myla gracilis Schouteden, 1938: 293. LECTOTYPE cf , ZAIRE (MRAC), here designated [examined].Myla graciloides Linnavuori, 1971: 175. Lectotype cf, SUDAN: Equatoria prov., Yei-Maridi, 13- 15. iv. 1963, swampy meadow (Linnavuori) (R. Linnavuori coll.), designated by Linnavuori, 1978: 36 [not examined]. Syn. n. Length: cf, 8-4-9-8 mm; $, 9-00-10-0 mm. Head distinctly longer than wide, narrowly rounded anteriorly. Length of antennal segment I divided bywidth of head including eyes in male 1-06-1-27, in female 1-06-1-25. Ratio of lengths of antennal segmentsin male about l-00:0-87:0-94:0-86, in female about 1-00:0-87:0-93:0-80. Length of rostral segment I dividedby width of head including eyes in both sexes 0-71-0-82. Ratio of lengths of rostral segments aboutl-00:0-85:0-46:0-66. Pronotum (Fig. 83) with posterolateral angles strongly produced anterolaterally, terminating in slenderspines, anterolateral margins with one to three prominent, piceous or black tubercles; width across apicesof posterolateral spines divided by width of head including eyes in male 2-57-3-18, in female 2-74-3-09. Abdomen with posterolateral angles of sternites III to VII produced as fine spines (Fig. 94). Connex-ivum rather narrow; laterotergite V twice as long as wide. Paramere (Figs 108-110) slender, with apicaltooth long and both teeth on a common stem that arises terminally or subterminally from the blade. Femalewith second valvula tapering to narrowly rounded apex. Pubescence of head, pronotum, scutellum, thoracic pleura, abdominal venter laterally and basally,laterotergites, clavus, corium, antennal segment I, coxae, trochanters, femora and rostrum comprisingfairly short, flattened, scale-like, decumbent (except on appendages), white hairs; hairs on thoracic pleura(Fig. 98) and declivent area of pronotum very distinctly longer than distance between adjacent hairinsertions and tending to form three or four transverse bands on deolivent area of pronotum. Areas ofabdominal venter and appendages that are devoid of scale-like pubescence with short, suberect, colourlesshairs. Coloration generally as in M. hoploxys but never heavily tinged red. Scutellum with midline and small,apical knob usually creamy white. Large tubercles of anterolateral margins of pronotum and apices ofposterolateral spines piceous to black. Veins of hemelytral membrane brown, uninterrupted. REMARKS. This species differs from all others of the genus in the greater length of the scale-like pubescence.The posterolateral pronotal angles are more strongly produced in this species than in any of its congeners.The combination of these two features gives M. gracilis a superficial resemblance to Clavigralla schnelli(Villiers). Indeed, the type-series of M. gracilis contains three individuals of C. schnelli, one of which bearsa label with the legend, 'holotypus'. This type designation has never been published and Schouteden's verydetailed description of the antennifers, rostrum, antennae, femoral spines and male genital capsule leavesno doubt that he was describing a species of Myla. I have, therefore, ignored the unpublished 'holotype'designation and here select one of the 'paratypes' as lectotype. Linnavuori (1971: 175; 1978: 35-36) usedthe name M. gracilis for what is evidently C. schnelli and established the unnecessary new nominal speciesMyla graciloides for M. gracilis proper. In his 1971 paper, Linnavuori cites the localities Sudan, Congo andIvory Coast but in 1978 he cites only the 'type' from Sudan; this citation is effectively a lectotypedesignation. There is some variation in the pronotal posterolateral spines which are usually laterallydirected but sometimes anterolaterally directed. DISTRIBUTION. West and Central Africa. TRIBE PSEUDOPHLOEINI 205 MATERIAL EXAMINED Zaire: 1 cf (lectotype of gracilis), Yangambi, xii.1925 (Ghesquiere) (labelled as paratype on pinkrectangle but lectotype on purple-bordered disc) (MRAC) Ivory Coast: 14 cf, 20 $, Lamto, variously 'Lamto (Toumodi)', various dates from 4.xii.l962 to14.viii.1964, all months except i, ii, xi, greatest numbers in vi; 5 cf , 1 $ , Bouake, 22.V.1962, 21.vii.1962; 1CT, Dabov, S.v.1964 (BMNH). Ghana: 1 cf, 'Gold Coast' (CottereU); 1 $, Yapi (Simpson}; 1 $, Tafo,9.V.1957, in yellow trays (Eastop) (BMNH); 2 cT, 1 ? , Tafo, 14.X.1966, 20.ii.1967, pyrethrum knockdownfrom cocoa (Kumar, Leston); 1 $ , Legon, 2.x. 1968 (Kumar); 1 $ , Ho, 7.xi.l967 (Leston); 1 $ , Shai Hill,29.vii.1967 (Leston) (UG). Togo: 1 cf, Missahou<, 650 m, vi.1963 (Schach) (MRAC). Nigeria: 1 $,Akpashe District, Udi, 27.X.1955 (Bechyne) (BMNH). Zaire: 73 cf, 61 $, Faradje, 2940'E, 340'N,i.1913, 3 $, Faradje, i.1912; 1 cf , 1 $, Faradje, i.1917; 1 cf , Medje, 2715'E, 225'N, 10.iv.1910 (all Lang,Chapin) (AMNH); 1 cf, Lomani, xii.1931 (Collart); 1 cf, Ngowa, vi.1938 (Mertens); 1 $, Ngowa,7-27.iv.1939 (Mertens) (IRSNB); 2 $, Congo da Lemba, 1911 (Mayne); 1 $, Haut-Uele, Moto, 1920(Burgeon); 1 $ , Elisabethville (R. Lubumbashi), 7.xi.l920 (Bequaert); 1 $, Katanga, Kando, l-4.iv.1931(de Witte); 1 $, Kisanfu, 1919 (Vanderijst); 1 ex. without abdomen, Magabi-Niarembe, v.1935 (Scops)(these seven exx all paralectotypes of gracilis); 1 cf , Kivu, Bwito, ca 1700 m, 26. vi. 1934 (Marlier); 1 cf,Ituri, Bunia, vi.1938 (Lefevre); 1 cf, Bas-Congo, Mayidi, 1942 (van Eyen); 1 cf, Rutshuru, Fuku,30.V.1936 (Lippens); 1 cf , Lulua, Kapanga, Hi. 1933 (Overlaet) (MRAC). Paralectotypes of My la gracilis Schouteden not belonging to this species. 3 exx determined as Clavigrallaschnelli (Villiers) by Dolling, 1979: 36, q.v. Distribution of Pseudophloeinae Preamble Without a complete phylogenetic analysis of the subfamily, incorporating the Palaearcticgenera, a comprehensive account of its zoogeography is not possible. In undertaking the presentrevision I have considered every species of the subfamily, including the Palaearctic ones, inrelation to the diagnostic features of every genus, and I am reasonably certain that all of thegenera currently recognized are monophyletic in origin and holophyletic in composition.However, I have not fully investigated the relationships among all the genera of the subfamily.Two presumably holophyletic groups of genera are recognized: the tribe Clavigrallini (seeDolling, 1978; 1979a) and a large group of genera including all of those confined to the NorthTemperate Zone and all of the tropical genera except for Vilga, Risbecocoris, Paramyla,Hoplolomia and Indolomia. The first two of these genera are probably not descended from themost recent common ancestor of the two big groups while the last three may be related to thenon-clavigralline group in some way (see Table 1). The distribution of the various genera ofPseudophloeinae in the major zoogeographical regions is summarized in Table 2. The Afro-tropical region in this table excludes the Mascarene area (Madagascar, Mauritius, Reunion andRodriguez), which is shown separately. The Oriental region is taken to include New Guinea andthe Solomon Islands but Australia is excluded and its pseudophloeine fauna is tabulated in aseparate column. Table 1 Distribution of two characters in the subfamily Pseudophloeinae. Antevannal veinabsent Antevannal veinpresent Posterior femur without basal tuberclePosterior femur with basal tubercle Vilga,Risbecocoris Hoplolomia, Indolomia, Paramyla Clavigrallini Remaininggenera Neotropical The only genus of Pseudophloeinae represented in the Neotropical region is Vilga, which isendemic to the New World. The 14 species were divided among six subgenera by Dolling (1977), 206 W. R. DOLLING who commented that the differences among these subgenera were of the same degree as thoseseparating some Palaearctic genera. One subgenus, whose single species is facultativelybrachypterous, occurs from the extreme southern U.S.A. to El Salvador; all the othersubgenera are confined to the Neotropical region and are widely sympatric there, with theexception of one subgenus whose three species are confined to the temperate and largelymountainous regions of Argentina, Chile and Peru. This latter subgenus, through a generalreduction of spininess of the body and legs and a flattening and broadening of the body,superficially resembles the tropical montane and temperate South African genus Paramyla andthe Palaearctic Arenocoris and Bathysolen Fieber. Nearctic The presence of a single species of the otherwise Neotropical genus Vilga in Mexico and adjacentareas has been mentioned above. In addition, two genera of Pseudophloeini are shared betweenthe Palaearctic and Nearctic regions. Coriomeris Westwood has 16 species in the Palaearctic and another four in the Nearctic. TheNearctic species (see Dolling & Yonke, 1976) are mutually very similar and have strictlyparapatric distributions, suggesting recent entry into and radiation within the region. Species ofCoriomeris are found within the Arctic Circle in Alaska, Yakutia and Karelia, so there wouldseem to be no climatic barrier to migration across the Bering Straits even today. Ceraleptus Costa has five Palaearctic and four Nearctic species, none of them shared betweenthe two regions. Its nearest relatives appear to be Microtelocerus Reuter and UrartucorisPuchkov, both genera with very restricted distributions in the mountains of the central southernPalaearctic, so a Palaearctic origin for Ceraleptus seems probable. Its American species arerather diverse and fall geographically into two pairs, one pair sympatric along the Pacific coast ofNorth America and the other pair sympatric in the south-eastern U.S.A. (Froeschner, 1963).Members of the genus do not occur at such high latitudes as the extremes penetrated byCoriomeris and this fact, coupled with the greater diversity and more complex distributionpattern shown by the species of Ceraleptus, suggests that the genus has been present in theNearctic region longer than has Coriomeris, presumably arriving in one or more immigrationepisodes during an era of warmer climate than that of the present day. Palaearctic Twelve genera and about 39 species of Pseudophloeini occur in the Palaearctic region; only oneof these species is shared with another region: Arenocoris intermedius , whose range includeswestern temperate Asia, Iberia and the northern coastal countries of Africa and extendssouthwards into Chad and Ethiopia. As indicated above, the Palaearctic genera Coriomeris andCeraleptus have both entered North America and have endemic species there. The Palaearctic intrusions of two species of the tribe Clavigrallini, Clavigralla scutellaris (intoAfghanistan) and Gralliclava horrens (into Japan) are mentioned below; no other genus of thesubfamily is represented in both the Palaearctic and Afrotropical or the Palaearctic and Orientalregions. Afrotropical and Oriental Clavigrallini The well-defined tribe Clavigrallini comprises four genera whose interrelationships are unclear.With the exception of two minor intrusions into the Palaearctic region, mentioned below, thetribe is confined to the tropical regions of the Old World. Clavigralla Spinola is the largest genus of Pseudophloeinae, with 46 described species. It hasclearly undergone most of its radiation on the African mainland, where 38 of its species arefound. Only one species, C. elongata Signoret, which is widespread on the African mainland, ispresent on the smaller offshore Afrotropical islands. It has reached Sao Tome and the CapeVerde Islands in the Atlantic (Dolling, 1979a: 28 erroneously reported it from the CanaryIslands on the basis of a mislabelled specimen from the Wollaston collection; the card bears a TRIBE PSEUDOPHLOEINI 207 Table 2 Number of species of each genus of Pseudophloeinae in the regions indicated. Species occurringin more than one region are scored in parentheses in regions other than those of their putative origin asare genera with no endemic species. 1 Anoplocerus, Bathysolen, Bothrostethus, Loxocnemis, Microtelocerus, Nemocoris, Strobilotoma, Ulmicola, Urar-tucoris. colour-coded mark that indicates its true provenance as the Cape Verde Islands), and Madagas-car, Mauritius, Reunion, Rodriguez and the Seychelles in the Indian Ocean. It has also beenreported from Arabia. In 1979 1 had seen no specimens of the species from the mainland of WestAfrica, where it seems to be largely replaced by C. shadabi Dolling and C. breviceps Dolling. Ihave now seen specimens of it from two localities in Senegal. C. elongata is the onlyrepresentative of the subfamily in these Atlantic and Indian Ocean islands with the exception ofsix endemic species in Madagascar. One of these, C. tuberculicollis (Reuter), is a taxonomicallyisolated species with several plesiomorphic features; it has probably been present in Madagascarmuch longer than any other species of the tribe. Three Malagasy species are closely similar to C.elongata and may be either local derivatives of it or descendants of a single, elongata-likeimmigrant. The other two endemics are members of the C. tomentosicollis Stal group but belongto different subgroups and clearly derive from two separate immigrations. Thus, Madagascarseems to have been colonized on at least four and possibly five separate occasions byClavigrallini, but there is evidence for local radiation stemming from only one of these invasions.Clavigralla is evidently a recent arrival in the Oriental region. Its three Oriental species allbelong to the same subgroup of the tomentosicollis-group. One of them, C. scutellaris (West-wood), which is probably the sister-species of C. tomentosicollis, is also found in north-easternAfrica and Arabia; in Asia it extends as far northwards as Afghanistan. The other two arevicariant sister-species, one confined to Sri Lanka and southern India and the other distributedfrom northern India to southern China and Vietnam; the Indian population of this secondspecies is recognizably distinct from the populations further east. 208 W. R. DOLLING Oncaspidia Stal has a single, widespread species endemic to Africa. Clavigralloides Dolling has the most easterly centre of distribution of any genus of Pseudo-phloeinae. Three of its five species occur in New Guinea; one of these three is, so far as is known,endemic while the other two extend southwards into northern Australia and one of these alsospreads northwards to southern China. Closely related to these two is a species restricted to theLesser Sunda Islands. The fifth member of the genus is rather isolated taxonomically and isdivided into northern Indian and a southern Chinese subspecies. Gralliclava Dolling species are distributed from India to Australia. They fall into two groups,one with a plesiomorphic condition of the male genital capsule and a variety of unusualparamere shapes, and the other with an apomorphic condition of the capsule and rather uniformparameres. The first group comprises six known species with restricted distributions; three ofthese are Indian; one is found in the Himalayas and at high altitudes in southern China and Laos,one is known only from its high altitude type-locality in Java and one, surprisingly, is knownfrom a single specimen taken on a small Australian island in the Torres Strait. The four species ofthe second group between them occupy an almost continuous range from Sri Lanka and India toJapan, the Solomon Islands and tropical Australia. One is known from two specimens, one fromTimor and the other from Flores. The most widespread species, G. horrens Dohrn, is alsoknown from Timor, from Wetter and the Sulawesi group, and thence northwards to Japan(Japanese records, all from Kyushu, are summarized by Kawazawa, 1978) and westwards toIndia and Sri Lanka. A third species is common in New Guinea and is also recorded fromAmbon and the Admiralty, Bismarck and Solomon groups. The fourth species has beencollected from many localities in Queensland and the Northern Territory of Australia but it hasnever been found elsewhere, nor has any other species of the genus been taken in Australia. Thealmost complete allopatry of these four species and the fact that the Torres Strait separates twoof them suggests a very recent radiation of this group. Afrotropical and Oriental Pseudophloeini Apart from the intrusive Arenocoris intermedius , the present revision recognizes seven generawith 25 species in the Afrotropical region and six genera with 17 species in the Oriental. Two ofthese genera and none of the species are shared between the two regions. The largest genera arethe wholly Afrotropical My la, with 11 species, and the shared Psilolomia, with 13. Risbecocoris , with five described species, occupies a probably almost continuous belt ofgrassland and semi-desert from Senegal via Arabia to Delhi, taking in the Air mountains andnorthern Kenya as well. This type of habitat has probably been available in the area for a verylong time and dispersal across it between Africa and Asia would present no problems for anorganism as well adapted to it as is Risbecocoris. It makes little sense to attribute part of this aridzone to the Afrotropical region and part to the Oriental in the context of the biogeography ofsuch organisms. Psilolomia displays its greatest diversity in the Oriental region. A group of dark-colouredspecies, whose characteristic habitat is probably forest clearings, extends no further westwardsthan eastern India while a group of yellowish species, probably characteristic of dry, openwoodland, is distributed from West Africa to South East Asia. Only four species of this groupoccur in Africa; they have had little or no success south of the Equator. One of them, Psilolomianigeriensis, shows a tendency to split into northern and southern races in Nigeria, perhaps inresponse to differing climatic conditions. A single migration in the fairly recent past from Indiainto Africa via a more humid Arabia seems the most probable explanation for its occurrence inboth continents. Large areas of apparently suitable terrain remain unoccupied by this genus ineastern and southern Africa. Mevanidea is the only genus of Pseudophloeinae, apart from Clavigralla, to have been foundin Madagascar. The two species of the genus show classic allopatry, one in mainland Africa andthe other endemic to Madagascar. TRIBE PSEUDOPHLOEINI 209 Discussion The faunistic picture that emerges is one of greatest diversity in the tropics of the Old World withrather less diverse but long established faunas in the Palaearctic and Neotropical regions. TheNeotropical and Palaearctic faunas seem to have developed in almost complete isolation fromeach other and from the Palaeotropical faunas. The Nearctic region, Australia and Madagascarhave no endemic genera and have played only a peripheral part in the evolutionary history of thesubfamily. Apart from Risbecocoris, whose range is in a transitional zone, only one genus,Psilolomia, is shared between the Oriental and Afrotropical regions. It is the only genus that canbe shown to have successfully invaded and begun to diversify in a region where a diversepseudophloeine fauna was already present. The genus Vilga has evidently been present in the Neotropics for a very long time, perhapssince before the opening of the Atlantic Ocean; this would place the early radiation of thesubfamily before the end of the Cretaceous period. Among tropical Coreoidea it is frequentlythe case that the distribution of the subfamilies is pantropical while the tribes are restricted toeither the Old World or the New. Dense forest and water seem to have been very effective barriers to dispersal of Pseudo-phloeini but less effective in the case of Clavigrallini. The more elongate condition of thehemelytra of the latter, with a tendency for the corium to extend apically along the costal marginof the hemelytron, and the deeper thorax, presumably housing larger flight-muscles, aresuggestive of a more sustained power of flight which would enable its possessor to overcome theobstacles that inhibit the dispersal of Pseudophloeini. The distribution patterns of two groups ofrecently differentiated species illustrate this difference in dispersal ability, and hence in rates ofgenetic interchange. The genus Coriomeris in North America has split into four recognizablydistinct though closely related species without any intervening barriers of forest or water,whereas the species of the horrens-group of the clavigralline genus Gralliclava are, with one veryminor exception, separated by sea barriers. Parapatric speciation may also be in progress in thecase of the northern and southern forms of Psilolomia nigeriensis. Acknowledgements I am sincerely grateful to Dr Roberto Poggi, Museo Civico di Storia Naturale, Genoa, for sending me adrawing of the holotype of Pseudophloeus gestroi; to Prof. Dott. Carlo Vidano, Turin, for searchingSpinola's collection for the type-material of Merocoris spinicollis; to Dr H. Brailovsky, UNAM, Mexico,for donating the unique type-specimen of Indolomia conculata; and to the following colleagues for theirkindness in arranging loans of material in the depositories mentioned: Dr A. E. Akingbohungbe, IfeUniversity, Ile-Ife, Nigeria; Dr P. H. Arnaud, CAS, San Francisco; Mr J. C. Deeming, formerly IAR,Samaru; Dr U. Gollner-Scheiding, MNHU, Berlin; Mrs C. N. Higa and the late Dr J. L. Gressitt, BPBM,Honolulu; Dr A. Kaltenbach, NMV, Vienna; Prof. R. Kumar, UG, Legon; Drs I. Lansbury and M. W. R.de V. Graham, UM, Oxford; Dr R. Linnavuori, Raisio, Finland (private collection to be deposited inZMU, Helsinki); Dr J. G. H. Londt, NMP, Pietermaritzburg; Dr J. T. Medler, University of Ibadan,Nigeria; Drs M. Meinander and A. Jansson, ZMU, Helsinki; Dr J. Nast, IZPAN, Warsaw; DrT. Nyholm,NR, Stockholm; Dr G. Peterson, IP, Eberswalde; Dr E. C. G. Pinhey, NMB, Bulawayo; Mr P. Reavell,Empangeni, Zululand (private collection, type-material to be deposited in NMP, Pietermaritzburg); Dr G.Schmitz, MRAC, Tervuren and IRSNB, Brussels; Drs M. U. Shadab and P. Wygodzinsky, AMNH, NewYork; Prof. J. A. Slater, University of Connecticut, U.S.A. (private collection); Dr M. Taou and the lateDr A. Villiers, MNHN, Paris; Mr J. A. van Reenen, TM, Pretoria. References Agassiz, L. 1848. Nomendator zoologici sive index universalis x + 1135 pp. Solothurn. Ahmad, I. & Shadab, M. U. 1969. A new species of a monotypic Ethiopian genus Risbecocoris Izzard 1949 (Heteroptera, Coreidae, Pseudophloeinae) from Pakistan. Pakistan Journal of Zoology 1: 151-155, 2 figs.Bergevin, E. de 1930. Risultati zoologici della missione inviata dalla R. Societa Geografica Italiana per 1'esplorazione dell'Oasi di Giarabub (1926-1927). Liste des HSmpiteres. Annali del Musei Civico di Storia Naturale (di Genova) Giacomo Doria 55: 29-39. 210 W. R. DOLLING Bergroth, E. 1907. Hemipteren von Madagaskar und angrenzenden Inseln gesammelt von Herrn T. Tschitscherin. Russkoe Entomologicheskoe Obozrenie 6: 145-149. 1908. Neue Hemiptera aus Sud-Abyssinien. Russkoe Entomologicheskoe Obozrenie 7: 106-110.Blanchard, E. 1842. Entry: 'Atractus', p. 312. In D'Orbigny, Ch., Dictionnaire universel d'histoire naturelle 2 (Aph-By). Paris. Bose, M. 1946. A new Coreid from Delhi. Indian Journal of Entomology 7: 75-76, 1 fig.Breddin, G. 1909. Rhynchoten von Ceylon, gesammelt von Dr. Walter Horn. Annales de la Societe Entomologique de Belgique 53: 250-309; 27 figs. Brown, E. S. 1958. Revision of the genus Amblypelta Stal (Hemiptera, Coreidae). Bulletin of Entomolo-gical Research 49: 509-541, pi. 18.Brulle, A. 1835. Tome X. Orthopteres et Hemipteres. In Audouin, V. Histoire naturelle des Insectes 415 pp., 11 pis. Paris.Burmeister, H. 1835. Handbuch der Entomologie 2. Besondere Entomologie. Erste Abtheilung. Schnabelkerfe. Rhynchota. iv + 400 pp., 2 pis. Berlin.Casale, A. 1981. Museu Regionale di Scienze Naturali. Cataloghi2. Collezione emitterologica di Massimil- liano Spinola. 120 pp., illus. Torino.Chernova, G. P. 1978. Palaearctic species of the genus Coriomeris Westw. (Heteroptera, Coreidae). [In Russian.] Entomologicheskoe Obozrenie 57: 551-566, 46 figs.1979. New data on the taxonomy of some Palaearctic Coreidae (Heteroptera). [In Russian.] Entomologicheskoe Obozrenie 58: 578-581.Cobben, R. H. 1968. Evolutionary trends in Heteroptera. Part 1. Eggs, architecture of the shell, gross embryology and eclosion 475 pp., 316 figs. Wageningen. Curtis, J. 1834. British entomology. Part 11: 4 pp. (unnumbered), col. pis 482-529 with legends. London.Dallas, W. S. 1852. List of the specimens of Hemipterous Insects in the collection of the British Museum. Part 2: 369-590, pis 12-15. London.Distant, W. L. 1901. Rhynchotal notes. - IX. Heteroptera: Fam. Coreidae. Annals and Magazine of Natural History (7) 7: 416-432. 19020. Rhynchotal miscellanea. Annals of the South African Museum 2: 237-254, pi. 15. 1902ft. The Fauna of British India, including Ceylon and Burma. Rhynchota 1 (Heteroptera). xxxviii + 438 pp. London. 1913. Contributions to a knowledge of Oriental Rhynchota. Annals and Magazine of Natural History (8) 12: 283-287. 1920. Rhynchota from New Caledonia. Annals and Magazine of Natural History (9) 6: 143-164. Dohrn, A. 1860. Zur Heteropteren-Fauna Ceylon's. Stettiner Entomologische Zeitung 21: 399-409.Dolling, W. R. 1977. A revision of the neotropical genus Vilga Stal. Systematic Entomology 2: 27-44, 45 figs.1978. A revision of the Oriental pod bugs of the tribe Clavigrallini (Hemiptera: Coreidae). Bulletin of the British Museum (Natural History) (Entomology) 36: 281-321, 108 figs.1979a. A revision of the African pod bugs of the tribe Clavigrallini (Hemiptera: Coreidae) with a checklist of the world species. Bulletin of the British Museum (Natural History) (Entomology) 39: 1-84, 247 figs.1979ft. Systematic changes in the Coreidae (Hemiptera). Entomologist's Monthly Magazine 114: 97-98. 1979c. A second species of Microtelocerus Reuter (Hem. , Coreidae) with a redescription of the genus. Entomologist's Monthly Magazine 114: 99-101, 1 fig.Dolling, W. R. & Yonke, T. R. 1976. The genus Coriomeris in North America. Annals of the Entomological Society of America 69: 1147-1153, 6 figs. Fabricius, J. C. 1803. Systema Rhyngotorum x + 21 + 314 pp. Leipzig & Hamburg.Froeschner, R. C. 1963. The genus Ceraleptus Costa in the western hemisphere (Hemiptera: Coreidae). Journal of the Kansas Entomological Society 36: 109-113, 4 figs.Gerstaecker, A. 1873. Gliederthiere (Insekten, Arachniden, Myriopoden und Isopoden). xvi + 542 pp. , 18 col. pis. In von der Decken, C. C. , Reisen in Ost-Afrika 3 (2). Hahn, C. W. 1834. Die Wanzenartigen Insecten 2 (5): 101-120, col. pis 67-72. Nurnberg.Hsiao, T. Y. 1964. New Coreidae from China (Hemiptera, Heteroptera) III. [In Chinese: English summary.] Acta zoologica sinica 16: 251-262, 8 figs.- 1965. New Coreids from China (Hemiptera, Heteroptera) IV. [In Chinese; English summary.] Acta Zoologica Sinica 17: 421-434, 10 figs. 1977. A handbook for the determination of the Chinese Hemiptera- Heteroptera [In Chinese.] 1: iii + 330 pp., 52 pis. Beijing. TRIBE PSEUDOPHLOEINI 211 Izzard, R. J. 1949. A new genus of the subfamily Arenocorinae (Pseudophloeinae) from Senegal (Heteroptera, Coreidae). Annals and Magazine of Natural History (12) 2: 478-480, pi. 7.Kawazawa, T. 1978. Collection of Momobutohariheri Kamemushi from the lower slopes of Mount Kaimon. [In Japanese.] Genseitt: 14.Kirichenko, A. N. 1952. New and little known bugs (Hemiptera-Heteroptera) of Tadzhikistan. Trudy Zoologicheskogo Instituta Akademlya Nauk SSSR 10: 140-198. Kirkaldy, G. W. 1909. Hemiptera: new and old. No. 1. Canadian Entomologist 41: 30-32.Linnavuori, R. 1970. On the genus Mevaniomorpha Rt. (Het., Coreidae). Suomen Hydnteistieteellinen Aikakauskirja 36 (1): 43-45, 1 fig.1971. On the family Coreidae (Het.). Suomen Hydnteistieteellinen Aikakauskirja 37 (3): 163-178, 24 figs. 1978. Hemiptera of the Sudan, with remarks on some species of the adjacent countries. 6. Aradidae, Meziridae, Aneuridae, Pyrrhocoridae, Stenocephalidae, Coreidae, Alydidae, Rhopalidae, Lygaeidae.Acta Zoologica Fennica 153: 1-108, 74 figs. 1982. Some hemipterous fauna from Somalia and Ethiopia. Monitore Zoologico Italiano (N.S.) Suppl. 16(1): 1-33, 9 figs. Marshall, T. A. 1868. A few more words on bad spelling. Entomologist's Monthly Magazine 4: 280-282.Puchkov, P. V. 1979. A new genus and species of the subfamily Pseudophloeinae (Heteroptera, Coreidae) from Transcaucasia. [In Russian.] Trudy Vsesoyuznogo Entomologicheskogo Obshchestva Akademiya Nauk SSSR 61: 62-64, 1 fig.Reuter, O. M. 1883. Ad cognitionem Heteropterorum Africae occidentalis. Ofversigt af Finska Vetens- kaps-Societens Forhandlingar 25: 1-43. 1891. Heteropteres de Suez. Revue d' Entomologie 10: 137-142. Schouteden, H. 1910. Hemiptera. 9. Reduviidae, Pyrrhocoridae, und Coreidae. In Sjoestedt, Y. (Ed.), Wissenschaftliche Ergebnisse der schwedischen zoologischen Expedition nach dem Kilimandjaro, dem Meru und den umgebenden Massaisteppen Deutsch-Ostafrikas 1905-1906 12: 145-160.1912. Coreides nouveaux de la region du Kilimandjaro et du Meru. Annales de la Societe entomolo- gique de Belgique 56: 52-59. 1938. Catalogues raisonnes de la faune entomologique du Congo Beige. Hemipteres, Coreidae. Annales du Musee du Congo Beige Tervuren (Belgique) C. Zoologie. Serie 3 (2) 1 (4): 221-308. Signoret, V. A. 1861. Faune des Hempiteres de Madagascar. (Suite et fin.) 2e partie. Heteropteres.Annales de la Societe Entomologique de France (3) 8: 917-972, col. pis. 13, 14. Spinola, M. 1837. Essai sur les genres d'insectes appurtenants a I'ordre des Hempiteres, Lin. ou Rhyngotes,Fab. eta la Section des Heteropteres, Dufour 383 pp, 15 tables. Genes. Stal, C. 1866. Hemiptera Africana 2: 181 pp. Stockholm. 1868. Bidrag till Hemipternas Systematik. Oefversigt af Kungliga Vetenskaps-Akademiens Forhand-lingar 24(7): 491-560. 1873. Enumeratio Hemipterorum 3. Kungliga Svenska Vetenskapsakademiens Handlingar (N.F.) 11 (2): 1-163.Stichel, W. 1960. Illustrierte Bestimmungstabellen der Wanzen. II. Europa (Hemiptera-Heteroptera) 4 (13): 385^416, illus. Berlin-Hermsdorf.1962. Verzeichnis der palaarktischen Hemiptera-Heteroptera. 4 Pentatomomorpha (Coreoidea- Pentatomoidea) pp. 187-362. Berlin-Hermsdorf.Stys, P. 1978. Hind wing venation in Coreidae (Heteroptera): a history of misinterpretation. Acta Entomologica Musei Nationalis Prague 39: 263-269, 6 figs.Villiers, A. 1950a. Contribution a 1'etude de 1'Air (Mission L. Chopard et A. Villiers) Hemipteres Pyrrhocoridae, Coreidae, Pentatomidae et Coptosomatidae. Memoires de I'lnstitut Frangais d'Afrique Noire 10: 322-328, 4 figs.19506. Note d'entomologie Ouest-africaine. I. Descriptions de nouveaux Hemipteres Heteropteres. Bulletin de 1'Institutfranc.ais d'Afrique noire (A) 12: 634-659, 17 figs.Walker, F. 1984. List of the specimens of lepidopterous insects in the collection of the British Museum Part 2: (279)-581. London.Yakovlev, V. 1883. Materials on the Heteropterous fauna of Russia and the neighbouring area. X. New species in the family Coreidae. [In Russian.] Bulletin de la Societe Imperiale des Naturalistes de Moscou 57 f3): 98-112. 212 INDEX Index Invalid names are in italics; references to descriptions are in bold. abyssinica 194, 197, 201 airensis 157, 159 aliena 174 alienus 175 Amblypelta 155 Ammocoris 191 amphrysia 168, 170, 174 angusta 179, 180 an gust us 192 annulipes 183, 184, 185, 187 Anoplocerus 207 Arenocoris 156, 191, 206, 207 Atractus 191 australis 166, 167, 194 Austrocoris 155 Bathysolen 206, 207 bilineata 155 Bothrostethus 207 Boudicca 191 breviceps 207 brevitibialis 168, 170, 172, 174 Brotheolus 154 Brotheus 154 brunneofusca 168, 170, 177, 178, 179 calida 194, 196, 197, 198, 200 campbelli 162, 163 Ceraleptus 155, 206, 207 Chariesterus 155 Clavigralla 155, 190, 194, 206, 207, 208 Clavigralloides 207, 208clavipes 168, 170, 178concolor 182, 194conculata 163, 164Coriomeris 152, 206, 207, 209cornuta!81,182, 194Cristovallia 155 delhiensis 157, 160, 161,162dispar (Myla) 194, 198dispar (Psilolomia) 169, 174 egenus 191elevator 185elongata 206, 207 fallen! 191, 192 ferruginea 170, 172 gestroi!91,192gracilis 194, 204graciloides 204Gralliclava 207, 208, 209granula 194, 198granulifera 187, 189 Hoplolomia 152, 154, 156, 161, 163, 205,207hoploxys 194, 197, 198, 200, 201, 202, 204 horrens 206, 208, 209hystrix (Clavigralla) 187hystrix (Mevanidea) 189, 190 Indolomia 152, 154, 156, 163, 205, 207intermedius 191, 192, 206, 208 kilimana 189 lamottei 170, 171, 172lata (Myla) 193, 194, 200lata (Psilolomia) 170, 179Iatissimusl91,192Loxocnemis 207 Mevania 152, 187 Mevanidea 156, 183, 185, 187, 207, 208 Mevaniella 187Mevaniomorpha 156, 183, 185, 189, 207 microphthalma 194, 202, 203, 204Microtelocerus 152, 206, 207Myla 152, 156, 165, 181, 193, 207, 208 Nemocoris 207Neohoplolomia 168Neomevaniomorpha 156, 183, 207nigeriensis 170, 171, 177, 208, 209nigrispina 193, 196niokensis 165, 193, 194, 201, 203numidianus 157, 159, 160 Oncaspidia 207, 208 onceroma 194, 203 Paramyla 152, 154, 156, 165, 194, 205,206,207parva 168, 170, 173, 174picta 185, 186, 187picta 186, 187Psammocoris 191 Pseudomyla 156, 180, 183, 194, 207Pseudophloeus 191Psilocoris 168Psilolomia 156, 168, 183, 185, 207, 208, 209 pundaloyae 170, 177Pungral56, 179, 181,207 quadrocephalus 157, 161 Riptortus 154 Risbecocoris 152, 154, 156, 162, 205, 207, 208, 209rodhaini 183 scabricula 161, 162, 163schnelli 194, 204, 205scutellaris 206, 207shadabi 207somalica 194, 200Spathocera 154spiniceps 187, 189, 190spinicollis 180, 181, 182, 183steeleae 170, 178Strobilotoma 207suspecta 165, 166, 194 tomentosicollis 207tomentosus 156, 157, 159, 161Trallianus 155tuberculicollis 207typica (Cristovallia) 155typica (Neohoplolomia) 168, 172 Ulmicola 207Urartucoris 206, 207 Vilga 152, 154, 205, 207, 209vulgaris 170, 171, 175, 177, 178, 179 waltli!91,192westwoodi 154 Occasional Papers on Systematic Entomology New Series The economic importance of insects, and the enormous number of species, have resulted in avast literature written in many languages; that which is of direct economic importance andrecently published can, to an increasing extent, be searched by using computerized data bases,but a great amount of more general information is unlikely to be available so readily in thenear future. The objective of this new occasional series is to make available in hard copy someof the basic data that is essential to the preparation of comprehensive accounts of the worldinsect fauna. The papers have been fully researched bibliographically and consist of checklistsof nominal taxa, and faunal lists with information on host plants and localities, based mainlyon the collections and libraries of the British Museum (Natural History). No. 1. A checklist of Neotropical arctiine and pericopine tiger moths. A. Watson & D. T. Goodger 72pp. inc. 4 colour plates 27 February 1986 No. 2. An annotated checklist of the Carabidae (including Cicindelinae, Rhysodinae and Paussinae) recorded from Borneo. N. E. Stork 26pp. , 1 map 29 May 1986 Catalogues of BM (NH) Bulletins and Books free on request to Publication Sales. Titles to be published in Volume 53 A review of the Miletini (Lepidoptera: Lycaenidae) ByJ. N.Eliot Australian ichneumonids of the tribes Labenini and Poecilocryptini By I. D. Gauld & G. A. Holloway The tribe Pseudophloeini (Hemiptera: Coreidae) in the Old World tropics with a discussion onthe distribution of the Pseudophloeinae By W. R. Dolling The songs of the western European grasshoppers of the genus Omocestusin relation to theirtaxonomy (Orthoptera: Acrididae) By D. R. Ragge Photoset by Rowland Phototypesetting Ltd, Bury St Edmunds, SuffolkPrinted in Great Britain by Henry Ling Ltd, Dorchester pRr Bulletin of the British Museum (Natural History) The songs of the western Europeangrasshoppers of the genus Omocestus inrelation to their taxonomy (Orthoptera:Acrididae) D. R. Ragge Entomology series Vol53 No 4 27 November 1986 The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in fourscientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology,and an Historical series. Papers in the Bulletin are primarily the results of research carried out on the unique andever-growing collections of the Museum, both by the scientific staff of the Museum and byspecialists from elsewhere who make use of the Museum's resources. Many of the papers areworks of reference that will remain indispensable for years to come. Parts are published at irregular intervals as they become ready, each is complete in itself,available separately, and individually priced. Volumes contain about 300 pages and severalvolumes may appear within a calendar year. Subscriptions may be placed for one or more ofthe series on either an Annual or Per Volume basis. Prices vary according to the contents ofthe individual parts. Orders and enquiries should be sent to: Publications Sales, British Museum (Natural History),Cromwell Road, London SW75BD,England. World List abbreviation: Bull. Br. Mus. nat. Hist. (Ent.) "rustees of the British Museum (Natural History), 1986 The Entomology series is produced under the general editorship of the Keeper of Entomology: Laurence A. Mound Assistant Editor: W. Gerald Tremewan ISBN 565 06022 8ISSN 0524-6431 British Museum (Natural History)Cromwell RoadLondon SW7 5BD Entomology seriesVol53No4pp213-249 Issued 27 November 1986 The songs of the western European grasshoppers ofthe genus Omocestus in relation to their taxonomy(Orthoptera: Acrididae) D. R. Ragge Department of Entomology, British Museum (Natural History), Cromwell Road, LondonSW75BD Contents Synopsis 213 Introduction 213 Acknowledgements 214 Abbreviations of depositories 214 Methods 215 Recording and analysing the songs 215 Song terminology 215 Presentation 215 The song as a taxonomic character 219 New lectotype designations 220 Omocestus Bolivar 221 Keys to the principal western European species of Omocestus 222 Key based primarily on morphological characters 222 Key based on song characters 224 The songs of the western European Omocestus 226 References .- 246 Index 249 Synopsis The male calling songs of 11 western European and one Canary Island species of the Gomphocerine genusOmocestus are described and fully illustrated with oscillograms at three or four different speeds. Thecourtship song is also described for the six species in which such a song is known. The songs of six species(panteli, bolivari, uhagonii, minutissimus, llorenteae and simony i} are described for the first time. For eachspecies a full list is provided of references to any past published work on the songs, classified according tothe kind of information given. Notes on recognition, using both morphology and song, are given for eachspecies, and two identification keys are provided, one based mainly on morphology and the other on song.The value of the song as a taxonomic character is discussed. Introduction The Gomphocerinae, of which Omocestus is one of the better known Old World genera, includemost of the common European grasshoppers of open grassland and moorland. They produce themost highly developed communicative sounds of all grasshoppers and are widely used inbio-acoustic, ecological and cytogenetic research. The subfamily includes about 150 Europeanspecies, of which over 20 are pests in parts of their range, seven of them being regarded as majorones (Tsyplenkov, 1970). Omocestus is a widespread Palaearctic genus, including some 40 species living in habitatsranging from lush lowland meadow to steppe, open woodland and high mountains. Twentyspecies have been recorded from western Europe, of which three, rufipes, petraeus andparticularly haemorrhoidalis , become pests when they occur in sufficient numbers, causingdamage to cereal grasses, hayfields, alfalfa and other cultivated plants in western and,especially, eastern Europe (Bei-Bienko & Mishchenko, 1951; Tsyplenkov, 1970). The conspicuous songs of Gomphocerine grasshoppers have attracted attention for centuries, Bull. Br. Mus. not. Hist. (Ent.) 53 (4): 213-249 Issued 27 November 1986 214 D. R. RAGGE but it is only since the pioneering work of Faber (1928) that they have been considered seriouslyas an aid to identification, and only during the past 30 years that their importance in taxonomicresearch, and particularly in resolving complexes of sibling species, has been fully recognized.This importance stems from the current belief that they provide the main mate recognitionsystem in these insects and are thus able to maintain reproductive isolation in groups ofsympatric species that would otherwise be able to interbreed freely. Perdeck (1957) establishedexperimentally that this is true of the well-known sibling pair Chorthippus brunneus (Thunberg)and C. biguttulus (Linnaeus), and, although comparable studies on other groups of Gom-phocerinae are still greatly needed, it seems likely that such studies would confirm that thisfunction of the song is widespread in the subfamily. Grasshopper songs can be quite easily recorded on tape and their rhythmic patterns, whichprovide the most useful taxonomic information, lend themselves well to oscillographic analysis.It is thus rather surprising that there has so far been no comprehensive account of these songs,presented group by group and illustrated with oscillograms. The main aim of the present paper,and of similar papers in preparation on other groups of Orthoptera, is to provide such anaccount, arranged systematically by group and fully illustrated with oscillograms at severaldifferent speeds. It is hoped that this form of presentation will make the information most usefulin identification and taxonomy, as well as providing a basic reference work on the songs of theseinsects. No information has previously been published on the songs of six of the 12 species included inthis study (panteli, bolivari, uhagonii, minutissimus, llorenteae and simonyi) and no oscillogramshave previously been published for a further two (petraeus and raymondi). For the species inwhich published information on the song is already available, references to the sources aregiven, classified according to the kind of information published. O. simonyi, known only fromthe Canary Islands, is not strictly within the scope of this paper but is included for convenience. The song is still unknown in the following nine species of Omocestus recorded from westernEurope: antigai, burri, corsicus, femoralis, kaestneri, knipperi, lopadusae, navasi and uvarovi.These species, which are all very local and often of doubtful status, are therefore excluded fromthis study. Acknowledgements I am much indebted to the following, who have kindly lent me type-specimens or other material from theirrespective institutions: Drs R. Danielsson, M. Donskoff, P. Grootaert, K. K. Giinther, A. Kaltenbach, V.Llorente, F. Pascual and J. J. Presa. My thanks are also due to Dr N. D. Jago, who kindly provided me with tape recordings of the song of O.viridulus made by him in the French Alps; to Mr W. G. Tremewan, who was kind enough to collect livemales of O. panteli for me in Teruel Province, Spain; and to Mr B. Alexander, who kindly brought me a livemale of O. simonyi from Lanzarote, Canary Islands. I am particularly grateful to my colleague Mr W. J. Reynolds, who made the studio recordings used inthis study, helped me in reviewing the past literature on the songs, tested the identification key based onmorphological characters and provided valuable comments at various stages in the preparation of thepaper. Finally I wish to thank my wife, who has given me invaluable help in all my field-work. Abbreviations of depositories BMNH British Museum (Natural History), London IRSNB Institut Royal des Sciences Naturelles de Belgique, Brussels MNCN Museo Nacional de Ciencias Naturales, Madrid MNHN Museum National d'Histoire Naturelle, Paris MNHU Museum fur Naturkunde der Humboldt-Universitat, Berlin ZI Zoologiska Institution, Lunds Universitet, Lund SONGS OF THE WESTERN EUROPEAN OMOCESTUS 215 Methods Recording and analysing the songs All the field recordings of the songs used in this study were made while the insects were in fullsunshine using a Uher 4000, 4200 or 4200IC tape recorder and AKG D202 or (recording 222/3only) Shibaden microphone; the tape speed was always 19 cm/s. The subjects were approachedstealthily so that the microphone could be held (or rested on the ground) about 10 cm from themwithout causing any disturbance. The studio recordings were made in the BMNH Acoustic Laboratory using a Kudelski NagraIV tape recorder and Sennheiser MKH405 microphone; the tape speed was 19 or 38 cm/s. Inevery case a bench lamp was used to provide light and radiant heat. Further data are given in Tables 1 and 2 for the recordings used for the oscillogramsreproduced in Figs 12-128. However, the song descriptions given for each species are in mostcases based on many more recordings of numerous songs, all of which were analysed oscillo-graphically. The oscillograms reproduced in the figures were chosen as being typical of thespecies concerned and, in some cases, showing the extent of intraspecific variation. Although Ihave given in Tables 1 and 2 the ambient air temperatures at the time of recording, it is really thebody temperature of the singing insect that is important, and this is more dependent on radiantheat than on the air temperature; as the insects were always receiving radiant heat from the sunor a bench lamp, the conditions under which the recordings were made were more uniform thanthe different air temperatures might suggest. All the oscillograms were made from recorded songs with a Mingograf 34T ink- jet recorder. Song terminology The bio-acoustic terms used are defined as follows. Calling song. The song produced by an isolated male. Courtship song. The special song produced by a male when close to a female. Syllable. The sound produced by one complete up and down movement of the hind legs (Fig- 1). Echeme. A first-order assemblage of syllables (Fig. 1).Echeme-sequence. A first-order assemblage of echemes (Fig. 1). Momentary breaks in the sound (of at least 1-25 ms) during the course of a syllable arereferred to as 'gaps'. Presentation As the main purpose of this paper is to give information on songs, I have not included either a fullsynonymy or a formal morphological diagnosis in the account of each species. I have, however,listed references to all past descriptive accounts of the songs (of any significance), classifiedaccording to whether they include oscillograms (including sound-level tracings), diagrams (i.e.hand-drawn representations of the songs), frequency information, musical notation, or verbaldescription without any of these additions; any commercially available disc or cassette record-ings of the songs are also listed. These references are not intended to be exhaustive - there aremany brief statements about the songs, especially in the earlier literature, that do not warrantinclusion; my aim has been to list all sources that the reader might find useful to refer to foradditional or confirmatory information on the songs. The only reference I have included to awork published before the present century is to Yersin's (1854) short but admirable account ofthe songs of 38 European species of Orthoptera (including two of Omocestus), in which heattempted, with some success, to represent them in terms of musical notation; in this largelyforgotten work he demonstrated clearly for the first time the striking differences between thesongs of the three closely similar species Chorthippus brunneus (Thunberg), C. biguttulus(Linnaeus) and C. mollis (Charpentier).Notes are given for each species on recognition by both morphology and song, and these are 216 D. R. RAGGE Table 1 Data for the field recordings of male songs of Omocestus used for oscillograms reproduced in thisstudy. All these recordings were made from different males. Recordists' names are abbreviated asfollows: NDJ = N. D. Jago; DRR = D. R. Ragge; WJR = W. J. Reynolds. Species Locality Date recorded Shade air Recording and collected Recordist temperature No. viridulus rufipes haemorrhoidalis petraeus FRANCE: Alpes-Maritimes, near StDalmasdeTende,Vallon de la Miniere,1830m AUSTRIA: Tyrol, nearKufstein, KaisertalFRANCE: Hautes-Pyrenees, nearVielle-Aure, Nestede Couplan FRANCE: Lozere, nearMende, Col deMontmiratFRANCE: Alpes-Maritimes, near StMartin Vesubie, Col StMartin, 1400 m FRANCE: Haute- Vienne,near St Yrieix-la-PercheFRANCE: Lozere,Causse Mejean, near 26.viii.1977 NDJ13.viii.1973 DRR 24C 20.ix.1976 WJR 14C 10.viii.1977 DRR 21C 26.viii.1977 DRR 22C 30.viii.1978 WJR 25C 222/398/3 183/2 232/3 241/11268/3 SONGS OF THE WESTERN EUROPEAN OMOCESTUS 217 Table 1 - cont. Species Locality Date recorded Shade air Recording and collected Recordist temperature No. uhagonii mmutissimus SPAIN: Madrid,Sierra de Guadarrama,La Bola del Mundo,2200mSame localitySame locality SPAIN: Madrid,Sierra de Guadarrama,Puerto de los Leones,1500m 6.viii.l9836.viii.l9836.viii.l983 DRRDRRDRR 21C21C 24C 5.viii.l983 DRR 23C 508/6509/1509/5 508/3 Table 2 Data for the studio recordings of male songs of Omocestus used for oscillograms reproduced inthis study. All these recordings were made from different males except for 310 and 312, which were madeon different days from the same male ofpanteli. The recordist was W. J. Reynolds in every case. haemorrhoidalis FRANCE: Pyrenees-Orientales, nearSaillagouse, Vald'Eyne 22.ix.1976 29.ix.1976 26C 177 218 D. R. RAGGE ECHEME HI I ' 'I 1 1 it H M |l**|i|M SYLLABLE GAPS Omocesfus raymonc/i ECHEME-SEQUENCE I 0-05 s +-* HMH-H > I ECHEME 0-05 s Omocesfus pefraeus Fig. 1 Oscillograms of the male calling songs of two species of Omocestus, showing the terminology usedin this paper. SONGS OF THE WESTERN EUROPEAN OMOCESTUS 219 followed by a more detailed descriptive account of the calling song and, when known, courtshipsong. I have not attempted to describe any further kinds of sound produced by the males (e.g.'rivalry song', 'copulation song') as these are not stereotypically patterned and seldom ofdiagnostic value. Neither have I considered it appropriate to describe the songs produced byfemales when in a sexually receptive state; these songs are seldom heard in the field and are inany case broadly similar in pattern to the calling songs of conspecific males. The songs of each species are illustrated by oscillograms at three, or sometimes four, differentspeeds. As the scale lines indicate, the three speeds given for all species are (after reduction forprinting) 10 mm/s, 80 mm/s and 320 mm/s, and the fourth speed included for the songs of threespecies (see below) is 1280 mm/s. In addition, oscillograms at 10 mm/s of typical male callingsongs of all 12 species are shown together in Figs 12-23 so that the more obvious differencesbetween them can be seen at a glance. Grasshopper songs are normally produced by both hind legs being rubbed simultaneouslyagainst the fore wings, and oscillograms of such 'two-legged' songs show the sounds produced bythe movement of one hind leg superimposed on those produced by the other. The two hind legsare not always well synchronized and in some species follow different patterns of movement, sothat there is sometimes a degree of 'blurring' of the oscillogram. Moreover, in 'two-legged'oscillograms that are fast enough to show the impacts of individual stridulatory pegs, it isimpossible to determine which peg-impacts are produced by one leg since these are mixed withthose produced by the other. Although these points are not particularly relevant to thediagnostic value of the songs, I have thought it worth including for three species (viridulus,rufipes and panteli} oscillograms taken from the calling songs of males with only one hind leg.The difference between the 'one-legged' and 'two-legged' oscillograms is hardly apparent at theslower speeds, but becomes more obvious at 320 mm/s and clearer still at 1280 mm/s. The highestspeed 'one-legged' oscillogram of panteli (Fig. 89) shows the individual peg-impacts withparticular clarity. The song as a taxonomic character Recognition of the diagnostic value of the songs of Orthoptera may be said to have begun in themid-nineteenth century, when Fischer (1849, 1850, 1853) published a number of song descrip-tions in primarily faunal or systematic works, and Yersin (1852, 1853 and especially 1854)produced a series of papers specifically on the songs and their value in identification. Yersin'swork of 1854, in which he described the songs of 38 European species of Orthoptera, wasparticularly noteworthy in that it included the first attempt to illustrate the songs graphically, inthe form of musical notation. Little further progress was made in exploiting the songs in taxonomy until Faber's work of1928, in which he gave, for the first time, an identification key to the German Orthoptera basedon their songs. Many German studies on the songs of the European Orthoptera followed,notably by Faber (especially 1929, 1932 and 1936) and Jacobs (especially 1950), culminating intwo major works published almost simultaneously (Faber, 1953; Jacobs, 1953). Most of thiswork was concerned with the detailed description of the songs and associated behaviour, andthere was still no serious attempt to use the songs in taxonomic research. Jacobs (1950, 1953) used simple diagrams against a time scale to illustrate the rhythmicpatterns of the songs. Ragge (1965) introduced a more sophisticated kind of diagram, amountingto a simplified oscillogram, and diagrams of this kind have since been used by Hoist (1970),Luquet (1978), Wallin (1979) and, in a slightly different form, by Duijm & Kruseman (1983) andBellmann (1985). Identification keys based on the songs were provided, for restricted parts ofEurope, by Luquet (the Mont Ventoux area of the French Alps, Acrididae only), Wallin(Sweden), Duijm & Kruseman (Benelux) and Bellmann (West Germany). All these studies areconcerned with regional faunas and contain no taxonomic research, although Luquet gives muchemphasis to the importance of the songs in taxonomy. The first serious application of the songs of Orthoptera to their taxonomy was in work on theNorth American crickets (Gryllidae). Following the pioneering work of Fulton (1931, 1952), 220 D. R. RAGGE Alexander (1957) and Alexander & Thomas (1959) reviewed the taxonomy of North Americanspecies of Acheta and Nemobius, and Walker (1962, 1963) carried out similar studies onOecanthinae; in all these studies the songs played a crucial part in resolving the taxonomy ofmorphologically similar species. Similar work on other groups followed, some of it reviewed byAlexander (1962) and Walker (1964). In 1972 Alexander, Pace & Otte produced an account ofthe 'singing insects' of Michigan in which they gave keys based mainly on the songs for all theEnsifera. More recent taxonomic studies on Orthoptera in which the songs have played a significant partinclude those of Bailey (1975, 1980) on African and Australian Copiphorine Tettigoniidae,Walker & Greenfield (1983) on Caribbean Neoconocephalus (Tettigoniidae), Otte & Alexander (1983) on Australian Gryllidae, Otte & Cade (1983 and later) on African Gryllidae, Heller (1984) on Poecilimon (Tettigoniidae) and Ragge & Reynolds (1984) on western EuropeanEuchorthippus (Acrididae). As can be seen at a glance from Figs 12-23, the male calling songs of the western Europeanspecies of Omocestus provide most useful taxonomic characters. Recorded songs, withappropriate oscillographic analyses, can be used to identify the species even if no specimens areavailable for morphological study (see the key on p. 224). The calling songs of the 12 species embraced by the present study fall into two clearly differentkinds: the rapid sequence of echemes produced by petraeus, minutissimus and llorenteae, andthe single or widely separated echemes produced by the remaining species. As would beexpected, the rapidly repeated echemes of petraeus, minutissimus and llorenteae are composedof very rapidly repeated syllables, almost always more than 80/s, whereas the syllable repetitionrate of the remaining species is almost always less than 40/s. Within this trio of species petraeusstands a little apart from the other two in having even shorter and more rapidly repeatedechemes, but the songs of minutissimus and llorenteae are broadly similar to each other inpattern, differing mainly in the duration of the echeme-sequence and the number of echemes ofwhich it is composed. Among the remaining species, viridulus and rufipes have much longer echemes than all theothers; they are very similar to each other in both calling and courtship songs, differing mainly inthe duration of the calling song echeme. The songs of all the other species consist of relativelyshort echemes distinguished from one another by duration, syllable repetition rate, and numberand distribution of gaps. The song differences shown by the four brachypterous species bolivari,uhagonii, minutissimus and llorenteae are particularly useful for field identification; two or threeof these species sometimes occur together in the same locality and are sufficiently similar inappearance to be quite easily confused with one another. Extreme caution has to be used in drawing phylogenetic conclusions from resemblances ordifferences in grasshopper songs, since it is well known from studies on the Chorthippusbiguttulus group that apparently closely related species, capable of producing vigorous andfertile hybrids, can have strikingly different songs. From the known songs of Omocestus I aminclined to make such inferences in only two cases of song resemblance. The first is thewell-known pair viridulus and rufipes, in which the strong resemblance in both calling andcourtship songs suggests that they are sister species. The second is the pair of brachypterous,montane species minutissimus and llorenteae, in which the close resemblance in calling song (aswell as morphology) also suggests a comparatively recent common ancestry. New lectotype designations In the course of examining type-material I have taken the opportunity of designating lectotypesfor the following species included in this study. Gryllus rufipes Zetterstedt, 1821: 90. Following Ander (1943: 10) I am regarding only two adult males from Zetterstedt's original material (in the ZI, Lund) as being eligible as syntypes, and I have selected and labelled one of these, bearing Zetterstedt's SONGS OF THE WESTERN EUROPEAN OMOCESTUS 221 ochre-yellow label (indicating Ostergotland), as lectotype. I have labelled the other eligible adult male,which bears no colour-coded label, as a paralectotype. Acridium petraeum Brisout, 1855: cxiv. There is in the IRSNB, Brussels a long series of both sexes of this species, all labelled 'Lardy', thetype-locality. These specimens are also labelled as being from the collection of de Selys Longchamps, whohad earlier acquired Brisout's collection, and it seems almost certain that they include at least part ofBrisout's type-series. I have selected and labelled a male bearing the handwritten label 'Lardy pres deParis' as lectotype. As there are many other specimens of this and various other species of Omocestus (andother genera) in the de Selys Longchamps Collection labelled 'Lardy', I have not attempted to determinewhich of these are also likely to belong to Brisout's type-series and so have labelled no specimens asparalectotypes. Stenobothrus Brolemanni Azam, 1906: 128. Dr M. Donskoff of the MNHN, Paris has kindly sent me the male from Azam's type-series that seems most eligible as a lectotype and I have so-labelled it. Omocestus BolivariChopard, 1939: 172. The type-series of this species (in the MNHN, Paris) consists of one male and one female, and I haveselected and labelled the male as lectotype. The altitude given in the original description, '2000 m.environ', clashes with the '3000 m' given on the locality label, which I think is more likely to be accurate. Ihave labelled the female as a paralectotype. Gomphocerus (Stenobothrus) Uhagonii Bolivar, 1876: 324. Dr J. J. Presa of the Universidad de Murcia very kindly delivered to me by hand a male syntype of thisspecies (from the MNCN, Madrid) labelled as being from Navarredonda, one of the type-localities, and Ihave labelled it as lectotype. OMOCESTUS Bolivar Omocestus Bolivar, 1878: 427 [as subgenus of Gomphocerus Thunberg]; Burr, 1904: 320 [raised to genus]. Type-species: Gryllus Locusta viridulus Linnaeus, by subsequent designation (Kirby, 1910: 172).Dirshius Harz, 1975: 710 [as subgenus of Omocestus Bolivar]. Type-species: Gryllus haemorrhoidalis Charpentier, by original designation. Syn. n. DIAGNOSIS, cf $ Head of typical Gomphocerine shape, foveolae well developed. Antennae not clubbed.Pronotal lateral carinae varying from sharply incurved to almost straight. Brachypterous to macropterous.Fore wings without bulge on anterior margin of precostal area; medial area not conspicuously widened(except in O. uvarovi Zanon , whose generic assignment is uncertain) . Hind wings varying from transparentto strongly smoky. Ovipositor varying in length, normally without lateral teeth but sometimes showingtendency towards their development (e.g. O. broelemanni). Tympanal aperture slit-like. DISCUSSION. Omocestus can be defined only on negative characters: it lacks the precostal bulge ofChorthippus Fieber, the toothed ovipositor of Stenobothrus Fischer, and the clubbed antennae ofGomphocerus Thunberg, Gomphocerippus Roberts and Myrmeleotettix Bolivar. Like several other generain the large and rather intractable subfamily Gomphocerinae, it is no more than a group of convenience andis very unlikely to be holophyletic. Harz (1975) attempted to split it into two subgenera on the basis of thedegree of curvature of the pronotal lateral carinae, but the species show every gradation between sharplyincurved and almost straight carinae, and it is thus impossible to draw a clear dividing line between twosubgroups. Jago (1971) treated Omocestus as a subgenus of Stenobothrus, thus reverting to the status accorded to itby Bolivar (1897) and Yakobson & Bianki (1902). Jago's study was based entirely on males, however, andso did not take into account the toothed ovipositor as the main diagnostic character of Stenobothrus s. str. Iprefer to follow Harz (1975) and other recent workers on the European fauna in regarding this character assufficient for a generic distinction. Jago also regarded Kirby 's designation of viridulus as the type-species ofOmocestus as invalid on the grounds that, at the time of the original description of the genus, this specieswas listed by Bolivar (1878: 460) under Chorthippus rather than Omocestus. However, on p. 427 ofBolivar's work, where Omocestus is first established and diagnosed, viridulus is clearly included in it; theheading 'Chorthippus' in the list on p. 460 has clearly been inadvertently misplaced, appearing aboverufipes instead of in its correct position above pullus. I am therefore accepting Kirby 's type-speciesdesignation as valid. The male callings songs show a wide range of patterns in Omocestus, varying from the prolonged single 222 D. R. RAGGE echemes of viridulus and rufipes to the rapid sequence of short echemes oipetraeus. This is not in itself anindication of polyphyly, however, as some complexes of sibling grasshopper species that are capable ofinterbreeding (e.g. the Chorthippus biguttulus group) show equally striking differences in male callingsong. DISTRIBUTION. All Europe, the larger Mediterranean islands, North Africa and temperate Asia as far asChina. INCLUDED SPECIES. O. alluaudi Uvarov, O. antigai (Bolivar), O. aymonissabaudiae Salfi, O. bolivariChopard, O. broelemanni (Azam), O. burri Uvarov, O. caucasicus Tarbinskii, O. corsicus Chopard, O.cuonaensis Yin, O. demokidovi Ramme, O. enitor Uvarov, O. femoralis Bolivar, O. haemorrhoidalis(Charpentier), O. heymonsi (Ramme), O. hingstoni Uvarov, O.. kaestneri (Harz), O. knipperi Harz, O.lecerfi Chopard, O. lepineyi Chopard, O. llorenteae Pascual, O. lopadusae La Greca, O. lucasii (Brisout),O. minutissimus (Bolivar), O. minutus (Brulle), O. megaoculus Yin, O. motuoensis Yin, O. nanus Uvarov,O. navasi Bolivar, O. nyalamus Xia, O. panteli (Bolivar), O. petraeus (Brisout), O. raymondi (Yersin), O.rufipes (Zetterstedt), O. simony i (Krauss), O. tibetanus Uvarov, O. tzendsureni Giinther, O. uhagonii(Bolivar), O. uvarovi Zanon, O. viridulus (Linnaeus), O. znojkoi Mishchenko. Keys to the principal western European species of Omocestus The only western European species not included in these keys are very local forms, usually ofdoubtful status, in which the song is so far unknown. Two keys are provided here, one based on morphological characters but with notes on songdifferences added where appropriate, and one based entirely on the male calling songs. Some ofthe species are very difficult to separate morphologically and, in the absence of song informa-tion, some specimens may not be correctly identified using the morphological key. Referenceshould in any case always be made to the notes given on recognition in the accounts of eachspecies. Key based primarily on morphological characters This key is for sexually mature adults, either freshly collected or showing good colour preservation.O. simony i, which is known only from the Canary Islands, is not included. 1 Rather brachypterous , fore wings less than 3 times length of pronotum - Fully winged, fore wings more than 3 times length of pronotum 10 2 Larger: length of hind femur more than 9-2 mm in cf, more than 11-0 mm in $. (Eastern Pyrenees and Catalonia) broelemanni (p. 236) Smaller: length of hind femur less than 9-2 mm in cf, less than 11 -Omm in $. (Not known fromPyrenees) 3 3 Fore wings not or hardly projecting beyond hind wings (when flexed); medial area with large dark spots 4 Fore wings pro j ecting well beyond hind wings (when flexed) ; medial area without spots 5 4 Fore wings more than 2-5 times length of pronotum in cf , more than 2-1 times in 9 llorenteae (p. 243)Fore wings less than 2-5 times length of pronotum in cf , less than 2-1 times in $ minutissimus (p. 243) 5 Male 6 Female 6 Cerci laterally compressed towards tip uhagonii (p. 237) Cerci simply conical, not laterally compressed towards tip 7 Hind wings reaching less than halfway along fore wings (when flexed). Length of pronotum usually more than 2-5 mm. (Song a single echeme, as in Fig. 19) bolivari (p. 237) - Hind wings reaching more than halfway along fore wings (when flexed). Length of pronotum usually less than 2-5 mm. (Song a sequence of echemes, as in Fig. 21) minutissimus (p. 243) 8 Fore wings less than 1-3 times length of pronotum bolivari (p. 237) Fore wings more than 1 -4 times length of pronotum 9 SONGS OF THE WESTERN EUROPEAN OMOCESTUS 223 ruf uho 5 mm roy 10 pan11 Figs 2-11 Ovipositors and pronota of species of Omocestus. 2-5. Lateral view of the ovipositor of (2) O.viridulus, (3) O. rufipes, (4) O. uhagonii, (5) O. minutissimus. 6-11. Dorsal view of the pronotum of (6)O. viridulus, (7) O. rufipes, (8) O. haemorrhoidalis , (9) O. petraeus, (10) O. raymondi, (11) O. panteli. 9 Ovipositor shaped as in Fig. 5, lower valves with strongly sigmoid ventral profile minutissimus (p. 243) - Ovipositor shaped as in Fig. 4, lower valves with weakly sigmoid ventral profile uhagonii (p. 237) 10 Pronotal lateral carinae straight or almost so in prozona (Fig. 11). (Iberian Peninsula) panteli (p. 236) - Pronotal lateral carinae distinctly incurved or angled in prozona (not as in Fig. 11) 11 11 Fore wings falling well short of hind knees, their length less than 7-4 mm in cf , less than 8-0 mm in $ . (Mountains in S. Spain) Ilorenteae (p. 243) - Fore wings usually reaching at least to hind knees, their length more than 7-4 mm in d" , more than8-0mmin$ 12 12 Hind wings completely transparent (except sometimes for part of the subcostal area) . Pronotal lateral carinae as in Fig. 9. (Song a rapid sequence of echemes, as in Fig. 15) petraeus (p. 233) 224 D. R. RAGGE Hind wings at least slightly smoky towards tip. Pronotal lateral carinae as in Figs 6, 7, 8 or 10.(Song a single echeme , repeated after an interval of at least several seconds) 13 13 Male 14 Female 17 14 Abdomen with no orange or red colouring. (Song echeme lasting more than 12 s). . viridulus(p. 226)Abdomen with at least some orange or red colouring. (Song echeme lasting less than 12s) 15 15 Abdomen with at least the distal 5 sternites coloured red or orange-red. Pronotal lateral carinae gently incurved in prozona, as in Fig. 7. (Song echeme lasting more than 4 s) rufipes(p. 227) Abdomen without reddish colouring on sternites or with such colouring restricted to the distal 3or 4 sternites. Pronotal lateral carinae more strongly incurved in prozona, as in Figs 8 or 10.(Song echeme lasting less than 4s) 16 16 Fore wings reaching well beyond hind knees, more than 4 times length of pronotum. (Song echeme lasting less than 2s) raymondi(p. 233) Fore wings not or hardly reaching hind knees, less than 4 times length of pronotum. (Songecheme lasting more than 2 s) haemorrhoidalis(p. 227) 17 Ovipositor long, as in Fig. 2. Sides of body often green (but sometimes brown or reddish) viridulus(p. 226) - Ovipositor shorter, similar in length to Fig. 3. Sides of body never green 18 18 Abdomen with at least the more distal sternites coloured red or orange-red. Pronotal lateral carinae gently incurved in prozona, as in Fig. 7 rufipes(p. 227) - Abdominal sternites with no reddish colouring. Pronotal lateral carinae more strongly in- curved in prozona, as in Figs 8 or 10 19 19 Fore wings reaching beyond hind knees, more than 4 times length of pronotum . . . raymondi (p. 233) - Fore wings not or hardly reaching beyond hind knees, less than 4 times length of pronotum haemorrhoidatis(p. 227) Key based on song characters (see especially Figs 12-23) This key is based on the calling songs produced by isolated males in warm, sunny conditions. As far ascouplet 5 the song differences can be easily detected by the human ear assisted by the second hand (ordigital count) of a watch. From couplet 6 onwards some of the differences require oscillographic analysis,or at least the ability to play back the recorded song at a slower tape speed.See p. 00 for definitions of the terms 'echeme', 'syllable' and 'gap'. 1 Song a series of echemes in rapid succession (at least 1/s) 2 - Song a single echeme or a series of echemes separated by intervals of at least 3s 4 2 Echemes repeated at the rate of at least 3/s (Figs 55-63) petraeus(p. 233) Echemes repeated at the rate of about 2/s 3 3 Song consisting of about 1 1-15 echemes and lasting about 6-8 s (Figs 123-128) .... llorenteae (p. 243) - Song consisting of about 6-10 echemes and lasting about 3-5 s (Figs 117-1 19) minutissimus(p. 243) 4 Song consisting of a single echeme lasting more than 4s 5 Song consisting of one or more echemes, each lasting less than 4s 6 5 Echeme lasting more than 12 s (Figs 24-33) viridulus (p. 226) Echeme lasting less than 12s (Figs 34-45) rufipes(p. 227) 6 Syllable repetition rate less than 10/s Syllable repetition rate more than 10/s 8 7 Each syllable with more than 7 gaps (Figs 108-1 16). (Iberian Peninsula) uhagonii(p.231) - Each syllable with fewer than 6 gaps (Figs 120-122). (Canary Islands) simonyi (p. 246) 8 Syllable repetition rate less than 20/s 9 - Syllable repetition rate more than 20/s 11 9 Echeme lasting more than 2 s (Figs 90-98) broelemanni(p. 236) - Echeme lasting less than 2s (Figs 64-72 or 99-107) 10 SONGS OF THE WESTERN EUROPEAN OMOCESTUS "IO 00 c (N CN O s K 00 cs o O) o OCN 226 D. R. RAGGE 10 Gaps becoming obscured towards the end of the echeme (Figs 64-72). (Widespread in Iberian Peninsula, S. France, NW. Italy and North Africa) raymondi(p. 233) Gaps persisting to the end of the echeme (Figs 99-107). (Known only from the higher parts ofthe Sierra Nevada in S. Spain) bolivari(p. 237) 1 1 Echeme lasting more than 2 s (Figs 46-54) haemorrhoidalis(p. 227) - Echeme lasting less than 2 s (Figs 73-89) panteli(p. 236) The songs of the western European Omocestus Omocestus viridulus (Linnaeus) (Figs 2, 6, 24-33) Gryllus Locusta viridulus Linnaeus, 1758: 433. Lectotype cf, SWEDEN: Gotland, Burgsvik (LinneanSociety of London), designated by Marshall (1983: 394) [examined]. REFERENCES TO SONG. Oscillogram: Eisner, 1974, 1975; Grein, 1984; Haskell, 1957, 1958; Kutsch, 1976;Kutsch & Schiolten, 1979; Loher & Broughton, 1955; Zhantiev, 1981. Diagram: Bellmann, 1985a; Duijm& Kruseman, 1983; Haskell, 1957; Hoist, 1970; Jacobs, 1950, 1953; Ragge, 1965; Wallin, 1979. Frequencyinformation: Haskell, 1957, 1958. Musical notation: Yersin, 1854. Verbal description only: Beier, 1956;Faber, 1928, 1953; Haskell, 1955; Harz, 1957; Skovmand & Pedersen, 1978, 1983; Weber, 1984; Weih,1951. Disc recording: Grein, 1984; Ragge, Burton & Wade, 1965. Cassette recording: Bellmann, 19856;Wallin, 1979. RECOGNITION. Males of this common species may be easily distinguished from the rather similar speciesrufipes and haemorrhoidalis by the lack of any red colouring on the abdomen, and females by theconspicuously longer ovipositor (Fig. 2). Both sexes also lack the dark markings in the medial area of thefore wings that are normally present in the other two species, and the pronotal lateral carinae are usuallymuch less incurved in the prozona (Fig. 6). In the field the conspicuously loud calling song of the male provides an easy means of identifying thisspecies. Although similar in the quality of sound to the songs of rufipes and haemorrhoidalis, the song ofviridulus is more prolonged, usually lasting for more than 12 s. SONG (Figs 24-33). The calling song is an echeme usually lasting about 12-25 s and consisting ofsyllables repeated at the rate of about 15-20/s. The echeme beings quietly (the first few legmovements producing no audible sound) and gradually increases in loudness until the maximumintensity is reached after a few seconds; the echeme then continues at a constant intensity untilreaching an abrupt end. The syllable repetition rate is highest at the beginning of the echeme(usually 18-20/s), gradually lessening towards the end (when it is usually 15-18/s). The song is aconspicuous summer sound in much of the European countryside, louder than the songs of mostother common grasshoppers. In the presence of a female the male produces longer echemes, usually lasting more than 30 sand occasionally more than a minute; one of the hind legs is moved through a noticeably widerangle than the other and produces most of the sound. After a series of these echemes with shortpauses (about 10-15 s) between them, there is a quite different and much quieter echeme lastingabout 3-5 s and composed of syllables repeated at the rate of about 12-16/s (Fig. 26). This isnormally followed by a series of loud syllables and an attempt to copulate with the female. If thisattempt is unsuccessful the male will usually produce a series of sharp 'ticks' (Fig. 27) by kickingbackwards with the hind tibiae (as in the calling song of Stethophyma grossum (Linnaeus))before beginning another sequence of courtship echemes. Usually the two hind tibiae are kickedsimultaneously, but sometimes one at a time, changing haphazardly (rarely regularly) from oneside to the other. The number of ticks is very variable but is usually between 5 and 15, and therepetition rate is rather irregular, generally about 1-2/s. DISTRIBUTION. This common species occurs in fairly moist habitats throughout Europe, except for theextreme north and the southern parts of the Iberian, Italian and Balkan peninsulas; its range extendseastwards to Siberia and Mongolia. SONGS OF THE WESTERN EUROPEAN OMOCESTUS 227 Omocestus rufipes (Zetterstedt)(Figs 3, 7, 34-45) Gryllus rufipes Zetterstedt, 1821: 90. LECTOTYPE cf , SWEDEN: Ostergotland, Larketorp (ZI, Lund),here designated (see p. 220) [examined]. [Validity established by von Borck (1848: 124) under Article 24of the International Code of Zoological Nomenclature.} Gryllus ventralis Zetterstedt, 1821: 89. Holotype $, SWEDEN: Skane, near Tranas, Asperod ['Esperod'](ZI, Lund) [examined]. REFERENCES TO SONG. Oscillogram: Broughton, 1955; Grein, 1984; Loher & Broughton, 1955; Schmidt &Baumgarten, 1977. Diagram: Bellmann, 1985a; Duijm & Kruseman, 1983; Hoist, 1970; Jacobs, 1950,1953; Luquet, 1978; Ragge, 1965; Wallin, 1979. Frequency information: Busnel, 1955; Loher & Brought-on, 1955. Musical notation: Yersin, 1854. Verbal description only: Beier, 1956; Chopard, 1922; Faber,1953; Harz, 1957; Weih, 1951. Disc recording: Andrieu & Dumortier, 1963; Grein, 1984; Ragge, Burton &Wade, 1965. Cassette recording: Bellmann, 19856; Wallin, 1979. RECOGNITION. The red colouring on the underside of the abdomen and the conspicuously pale-tipped palpsenable both sexes of this species to be distinguished from its relatives. The distal part of the hind wings ismuch more strongly smoky than in haemorrhoidalis , in which the male abdomen is occasionally reddish onthe underside. In the field the calling song of the males enables them to be recognized easily. Although quite similar inbasic pattern to those ofviriduku and haemorrhoidalis, the song of rufipes differs in duration, lasting abouthalf as long as that of viridulus and about three times as long as that of haemorrhoidalis. SONG (Figs 34-45) . The calling song is an echeme usually lasting 5-10 s and consisting of syllablesrepeated at the rate of about 13-23/s. The echeme begins quietly and gradually increases inloudness until reaching an abrupt end, thus resembling the first half of the calling song ofviridulus. As in viridulus the syllable repetition rate is highest at the beginning of the echeme(usually 17-23/s), gradually lessening towards the end (when it is usually 13-17/s). In the presence of a female the male produces a courtship song quite similar to that ofviridulus. There is first a series of echemes similar to those of the calling song but usually ratherlonger; this is followed by a quite different and quieter echeme lasting about 5-10 s andcomposed of syllables repeated at the rate of about 10-15/s (Fig. 37) . There are then several loudsyllables followed by an attempt to copulate with the female. The 'ticks' produced by backwardkicks of the hind tibiae during the courtship song of viridulus seem never to occur in that ofrufipes. DISTRIBUTION. This species can tolerate drier conditions than viridulus and often occurs in more shadedhabitats. Its distribution is very similar, but extends further south in the Iberian, Italian and Balkanpeninsulas. Omocestus haemorrhoidalis (Charpentier) (Figs 8, 46-54) Gryllus haemorrhoidalis Charpentier, 1825: 165. Syntypes of both sexes, POLAND/CZECHOSLOVAKIA:'Silesia'. There are a male (in bad condition) and two females in the MNHU, Berlin that have beenregarded as syntypes of this species, but none is labelled 'Silesia' and there is some doubt that they belongto the type-series. REFERENCES TO SONG. Oscillogram: Grein, 1984; Schmidt & Schach, 1978. Diagram: Bellmann, 1985a;Duijm & Kruseman, 1983; Jacobs, 1950, 1953; Wallin, 1979. Verbal description only: Faber, 1953; Harz,1957; Weber, 1984. Disc recording: Grein, 1984. Cassette recording: Bellmann, 1985ft; Wallin, 1979. RECOGNITION. For the distinction between this species and petraeus see the remarks under that species.Both sexes may be distinguished from viridulus, rufipes and raymondi by the almost transparent hind wings(strongly smoky in the distal part in those three species). In the field the males can be recognized quiteeasily by their calling song, in which the echemes are shorter than those of viridulus and rufipes, but muchlonger than those of raymondi. SONG (Figs 46-54). The calling song is an echeme lasting about 2-4 s and consisting of syllables 228 D. R. RAGGE CS CN CS CN CN ^Q U t/5 5 8 M 2r *- r* ' o <A .5 rN y B , m 0> OO -CT3 J3 CM *- a o . o o |.8J! -^ O S* 8 2 . P * 5 B K " <*- 2 CM " O 'S T3 C 3wa 2 C ^ D O fe , < ^^ > I a|H5^^!^ ir r^ -> ~ C O o in o p t; 12 ** 9 Q ifl 2 Si r- <- 3 _ i illl SONGS OF THE WESTERN EUROPEAN OMOCESTUS 229 Ioo o 60 U. I If II s f3 O*O 'j* - o g 3 6 I OCO CO CNCO COCO 230 D. R. RAGGE CO 10CO >OCO I S I CO 00CO co C/5 -Jj ISf If SONGS OF THE WESTERN EUROPEAN OMOCESTUS 231 I 00 00 _ c ! c - o-o V3 C t; rt CN 10 232 D. R. RAGGE f -3II * U H c 00 2 u S T3 O 2 IISB <U CO S o -3 o iz Od. u </) D C <U <u si oo -- CO ^ (N UO C -T e g5 g c ts > ^ O _, r r- cs CO o SONGS OF THE WESTERN EUROPEAN OMOCESTUS 233 repeated at the rate of about 25-40/s (about double the rate of viridulus and rufipes). Theecheme begins quietly but soon reaches maximum intensity. The syllable repetition rategradually lessens during the course of the echeme. DISTRIBUTION. Widely distributed in continental Europe, but not known from the Scandinavian mainlandand limited to mountains in the southern peninsulas. The range extends eastwards across central Asia toMongolia, Manchuria and Korea. Omocestus petraeus (Brisout)(Figs 9, 55-63) Acridium petraeum Brisout, 1855: cxiv. LECTOTYPE cf , FRANCE: near Paris, Lardy (Brisout) (IRSNB,Brussels), here designated (see p. 221) [examined]. REFERENCES TO SONG. Diagram: Luquet, 1978. Verbal description only: Faber, 1953; Harz, 1957. RECOGNITION. This species lacks the smoky coloration of the distal part of the hind wings shown byviridulus, rufipes and raymondi, and is also noticeably smaller. Distinguishing it from haemorrhoidalis ismore difficult, especially if reliably identified specimens of both species are not available for comparison.O. petraeus is again noticeably smaller, and the head is larger in comparison to the pronotum and moreconvex above with shorter foveolae. The males lack the red colouring shown by the distal part of theabdomen in haemorrhoidalis, showing a yellowish colouring in this region instead. The highly distinctive calling song, consisting of about 10-20 rapidly repeated echemes, enables males ofthis species to be recognized easily in the field. SONG (Figs 55-63). The calling song consists of a sequence of about 10-20 echemes lasting about2-5 s. The sequence begins quietly, reaching maximum intensity after about 6 echemes; even atits maximum intensity the song is rather quiet. The echeme repetition rate is usually about 4/s,and the syllable repetition rate within each echeme about 80-1 10/s. Each of the louder echemeslasts for about 150 ms and contains about 10-15 syllables. In the presence of a female the male produces echemes of a quite different kind (Figs 57, 60,63). Each echeme begins with a relatively long syllable, lasting about 20-50 ms, and this isimmediately followed by a series of very short sounds, each lasting less than 5 ms, repeated at therate of about 50/s. These echemes are produced either singly or, more often, in groups of about2-5. DISTRIBUTION. Western, central and eastern Europe, Asia Minor and southern Siberia. Largely absentfrom the Iberian Peninsula, but quite widespread in Italy and the Balkan Peninsula. Omocestus raymondi ( Yersin)(Figs 10, 64-72) Stenobothrus raymondi Yersin, 1863: 289. Syntypes of both sexes, FRANCE: Var, near Hyeres (Raymond)(lost); the two 'neotypes' designated by Harz (1975: 713) are clearly invalid. REFERENCES TO SONG. Diagram: Luquet, 1978. Verbal description only: Chopard, 1922, 1952. RECOGNITION. This all-brown species is most likely to be confused with haemorrhoidalis, but can bedistinguished from it in both sexes by the strongly smoky distal part of the hind wings (almost transparent inhaemorrhoidalis). In the field males may be distinguished from those of haemorrhoidalis by the very shortechemes (lasting about 1 s) and slower syllable repetition rate (less than 20/s) of the calling song. SONG (Figs 64-72). The calling song is an echeme lasting about 1-0-1-5 s and consisting of about18-25 syllables repeated at the rate of 15-20/s. Each echeme usually begins quietly, rapidlyincreasing in intensity. Oscillograms show the syllables to have a characteristic pattern of gaps,which often becomes obscured towards the end of the echeme (Figs 67-69). The echemes areoften produced singly and repeated at irregular intervals (10-15 s is typical), but sometimes theyare in groups of 2-4 with much shorter intervals between them (often 2-5 s) (Figs 65, 66). DISTRIBUTION. Known only from southern France, the Iberian Peninsula, north-western Italy and NorthAfrica. 234 D. R. RAGGE O10 i o = 00 o oo CNO t^ S m o OO f> c oo f M 00^ c o M " U (u jj 360 ^j "Jj CN 3 =^ 5 ' rH n r ^ O QJ Hg O c 111 _ _ 3 *2c |]l o .cM . .S c u og 13 -aS S a 111 U o CS ts <t^ u r^ *t il E 9 N OO ^ l-i ^f ^ __ U V3 t/3 *- O OS "5 3 O COO SONGS OF THE WESTERN EUROPEAN OMOCESTUS 235 II 5 2 II ..; C ^1 |o i! o " j* G si 0) <U 11 S "3 o (4 1) O J3 ItI? <G o=6 8 a Io g 00O 236 D. R. RAGGE Omocestus panteli (Bolivar)(Figs 11, 73-89) Stenobothrus panteli Bolivar, 1887: 95. Syntypes of both sexes, SPAIN: Cuenca, near Ucles (7MNCN,Madrid). REFERENCES TO SONG. No published work known to me. RECOGNITION. In the Iberian Peninsula, to which it is confined, panteli is most likely to be confused withviridulus; it is, however, much smaller than that species and can be further distinguished from it by the redtip of the male abdomen and the relatively short ovipositor. There is also a superficial resemblance toStenobothrus stigmaticus (Rambur), which is similar in size and also widespread in the Iberian Peninsula,but the pronotal lateral carinae are straighter in panteli, the male cerci are conical throughout (laterallycompressed at the tip in stigmaticus) and the female lacks the teeth on the ovipositor shown by all species ofStenobothrus. There is a large form (var. meridionalis Bolivar) found in the south of the peninsula,especially in mountains; this form is even more like viridulus, but confusion is unlikely as that species doesnot occur in southern Spain. In the field the song may be recognized by its short duration (less than 2 s) and high syllable repetitionrate (usually more than 20/s). SONG (Figs 73-89). The calling song is an echeme lasting 1-2 s and consisting of about 30-55syllables repeated at the rate of about 18-30/s. The echeme begins quietly (sometimes after a fewlouder syllables) but soon reaches maximum intensity. The syllable repetition rate graduallylessens during the course of the echeme. In the presence of a female the male first produces a different kind of echeme lasting about 2-3s, and this is immediately followed by an echeme similar in duration and syllable repetition rateto that of the calling song but with an increase in intensity continuing through the whole echeme(Figs 77, 82, 87). DISTRIBUTION. This species is confined to the Iberian Peninsula, where it is widespread and very common. Omocestus broelemanni ( Azam)(Figs 90-98) Stenobothrus Brolemanni Azam, 1906: 128. LECTOTYPE cf , FRANCE: Pyrenees-Orientales, Val d'Eyne,1700-2200 m, 14.viii.1905 (Brolemann) (MNHN, Paris), here designated (see p. 221) [examined]. REFERENCES TO SONG. Oscillogram: Reynolds, 1986. RECOGNITION. This Pyrenean species is brachypterous, the fore wings not usually reaching the hind knees inthe male and reduced to short lobes (about 1-5 times the length of the pronotum) in the female; in bothsexes the hind wings fall short of the fore wings by quite a large gap. There is thus no risk of confusion withviridulus, rufipes, haemorrhoidalis and the other fully winged species of Omocestus. Confusion is alsounlikely with the brachypterous species occurring further south in Spain, such as uhagonii and minutissi-mus, which are much smaller and not known from the vicinity of the Pyrenees. The relationship of O.broelemanni with O. antigai (Bolivar) and O. navasi Bolivar has been recently discussed by Reynolds(1986). In the field the calling song of the male is quite distinctive, consisting of echemes lasting about as long asthose of haemorrhoidalis but with about half the syllable repetition rate of that species. SONG (Figs 90-98). The calling song, recently described for the first time by Reynolds (1986), is aseries of echemes, each lasting about 1-5-3-0 s and consisting of about 25-45 syllables repeatedat the rate of about 15-16/s. Each echeme begins quietly, reaching maximum intensity afterabout 1 s. Although the echemes are sometimes produced singly, they are more often repeatedfairly regularly, about every 5-10 s, in a series of indefinite duration. During the courtship song, which has also been described by Reynolds (1986), the male oftenproduces a series of rather longer echemes (each lasting up to 5 s) and this is followed by avariable number of short echemes of small numbers of syllables (or even single syllables) SONGS OF THE WESTERN EUROPEAN OMOCESTUS 237 produced by one leg only (Figs 92, 95, 98). There is then a series of single syllables produced byboth legs, followed by an attempt at copulation. DISTRIBUTION. Known only from the vicinity of the Pyrenees, particularly the eastern end of the range. Omocestus bolivari Chopard(Figs 99-107) Omocestus Bolivari Chopard, 1939: 172. LECTOTYPE cf , SPAIN: Granada, Sierra Nevada, slopes ofMulhacen, 3000 m ['2000 m. environ' according to Chopard, loc. cit.], 10.vii.1934 (Balachowsky)(MNHN, Paris), here designated (see p. 221) [examined]. REFERENCES TO SONG. No published work known to me. RECOGNITION. This species, known only from the higher parts of the Sierra Nevada in southern Spain, isvery similar to uhagonii, llorenteae and minutissimus . Males may be distinguished from all three of thesespecies by their short hind wings, which are less than half the length of the fore wings, and the females bytheir short fore wings, which are usually less than 1-3 times longer than the pronotum. In the field the isolated echemes of the calling song enable males to be easily distinguished fromllorenteae and minutissimus; from uhagonii they may be distinguished by the much faster syllable repetitionrate. SONG (Figs 99-107) . The calling song is an echeme lasting 0-5-2-0 s and consisting of about 10-30syllables repeated at the rate of 14 16/s. Each echeme begins quietly, soon reaching maximumintensity. Oscillographic analysis shows that there are gaps in each syllable (mainly in the laterpart of the syllable) and that they occur throughout the echeme; there are, however, only 2-4gaps per syllable (Figs 102-107), many fewer than in uhagonii. As in uhagonii the echemes arerepeated at irregular intervals, varying from a few seconds to over a minute. DISTRIBUTION. Known only from the higher parts of the Sierra Nevada in southern Spain, usually ataltitudes above 1500 m. Omocestus uhagonii (Bolivar)(Figs 4, 108-116) Gomphocerus (Stenobothrus) Uhagonii Bolivar, 1876: 324. LECTOTYPE cf , SPAIN: Madrid, Navarre-donda (Avila) (MNCN, Madrid), here designated (see p. 221) [examined]. REFERENCES TO SONG. No published work known to me. RECOGNITION. Males of this brachypterous montane species can be distinguished from the rather similarspecies bolivari, llorenteae and minutissimus by the cerci, which are laterally compressed towards the tip(simply conical in the other three species). Females can be separated from bolivari and llorenteae by thelength of the fore wings, which are 1-4-1-8 times longer than the pronotum (the corresponding ratios forbolivari and llorenteae are 0-9-1-3 and 2-1-2-5, respectively), and from minutissimus by the much lessstrongly sigmoid ventral profile of the lower valves of the ovipositor (cf. Figs 4,5). In the field the calling song of the males enables this species to be easily distinguished from llorenteae andminutissimus, which produce echeme-sequences of a quite different kind (cf. Figs 20-22); from bolivari itmay be distinguished by the much slower syllable repetition rate (cf. Figs 19, 20). SONG (Figs 108-116). The calling song is an echeme lasting 1-2 s and consisting of about 10-15syllables repeated at the rate of 6-7/s. Each echeme begins quietly, rapidly increasing inintensity. Oscillographic analysis shows that each syllable has a large number of gaps (commonlyas many as 8) and that this pattern of gaps is maintained until the end of the echeme (Figs 111,112). The echemes are repeated at irregular intervals, varying from a few seconds to over aminute. In the presence of a female the male produces a series of slightly longer echemes (lasting 2-3 s)at much more regular intervals (usually 6-8 in 40 s) (Fig. 110). The echemes consist of about15-20 syllables and at the end of some of them (usually about half) there is a group of 3-4syllables of a different kind (Figs 113, 116), repeated more rapidly (at the rate of about 16/s). In 238 D. R. RAGGE o 00 00 cs 00 II ijj o H a |< .W>"O |.s 2 a ja 8 | g " J2 O ^ 4) (SO ^ * t~~ o ^"^ ** r^ 1^1 R ob S O *O So So " .S j> ^H "O I c- <D O C t3 3 ^2 C 3 r^ '^ g o N ooU i- x SONGS OF THE WESTERN EUROPEAN OMOCESTUS 239 CO00 00 00 o 00 KCO CdON00 TJ CO E (N I o 8 o I 60 OJ~, u CO c. Q S o. CO 4) & o ^-^ = 00 o>^> "co JC 240 D. R. RAGGE OCS ~ r CO a. t/2 C GO o t/2 V3C'.S S T3 ^^ UN u 8t-^ t- ON 1) ON ^ ON X) r- o es ON a ;> G vO 2 )d5 * exS w ^^ S2H^ gB 2 g g-o I Ceo s "8 D 2 C si* =3 S c a o 15 ops t- o '8 o^ . o oo o ON o ri-2)ON ooex SONGS OF THE WESTERN EUROPEAN OMOCESTUS 241 oo a - CNO COO oo KO 242 ooo exo D. R. RAGGEI CM CO OB <L>C J= - E - oo (N C s! O 3O 00 SP O0> i-H^ ^ (N o oo d c/3 e ^^ D . ,_, Us C C PSO T3 ^ 13 li-a <*- <u 2 8. 00 * z ^ o o _c " S 5 5- S ^ c ?: o 2 <u ^ SO O ^ 1=1 C o o 5 n OX) SONGS OF THE WESTERN EUROPEAN OMOCESTUS 243 the intervals between the echemes there are quieter 'ticking' sounds repeated fairly regularly atthe rate of about 3-4/s. The number of echemes in the series is variable but can be more than 20.The series of echemes is followed by a variable period (often more than a minute) in which thequiet ticking continues and which ends in a number of sequences of a different kind of sound, stillquiet, and then an attempt at copulation. If this is unsuccessful the cycle begins again withanother series of echemes. DISTRIBUTION. This species has been previously known only from the mountains of central and northernSpain, but my wife and I collected a series of adults of both sexes at El Chorillo (2700 m), south of theMulhacen peak in the Sierra Nevada, on 31 July 1983. Omocestus minutissimus (Bolivar)(Figs 5, 117-119) Gomphocerus (Omocestus) minutissimus Bolivar, 1878: 424. Syntypes of both sexes, SPAIN: Navarra,Cascante (Perez Areas); Madrid, Escorial (MNCN, Madrid; BMNH, London; probably other deposi-tories) [$ syntype from Escorial in BMNH examined]. REFERENCES TO SONG. No published work known to me. RECOGNITION . Males of this small upland species may be distinguished from uhagonii by the cerci , which aresimply conical (laterally compressed towards the tip in uhagonii}. They can be separated from bolivari andllorenteae by the length of the hind wings, which are more than half the length of the fore wings but do notreach the fore wing-tips. Females can be distinguished from bolivari and llorenteae by the length of the forewings, which are usually 1-4-2-1 times the length of the pronotum; the corresponding ratios for bolivari andllorenteae are 0.9-1-3 and 2-1-2-5, respectively (these two species are in any case known only f r om theSierra Nevada or Sierra Espuna, where minutissimus does not apparently occur). Females of minutissimusmay be separated from those of uhagonii by the strongly sigmoid ventral profile of the lower valves of theovipositor (cf. Figs 4, 5). In the field the calling song of the males, consisting of a rapid sequence of echemes, enables this speciesto be easily distinguished from uhagonii and bolivari; from llorenteae it may be distinguished by beingshorter (usually lasting less than 6 s) and having fewer echemes in each sequence (usually fewer than 10). SONG (Figs 117-119). The rather quiet calling song is a sequence of about 6-10 echemes lastingabout 3-5 s. The sequence begins very quietly, reaching maximum intensity after about 3-4echemes. The echemes are separated by intervals of about 100-200 ms, often becoming morewidely spaced towards the end of the sequence. Each echeme begins and ends very quietly,giving a spindle-shaped oscillogram (Fig. 118). Each of the later echemes in the sequence lastsabout 300-400 ms and contains about 20-30 syllables repeated at the rate of about 80-90/s. Thecalling song is thus very similar to that of llorenteae, but is quieter and shorter, and has fewerechemes composed of fewer syllables. DISTRIBUTION. North-eastern, central and south-eastern Spain. Omocestus llorenteae Pascual(Figs 123-128) Omocestus llorenteae Pascual, 1978: 159. Holotype O" , SPAIN: Granada, Sierra Nevada, Dornajo, 2000 m,19. ix. 1975 (Pascual) (Universidad de Granada) [examined]. REFERENCES TO SONG. No published work known to me. RECOGNITION. This species, known only from the Sierra Nevada and Sierra Espuna in southern Spain, canbe distinguished from the similar montane species uhagonii and bolivari by the dark spots in the medialarea of the fore wings and by the longer hind wings, which reach the tips of the fore wings; the longer hindwings also enable this species to be distinguished from minutissimus, which is in any case not known tooccur in the Sierra Nevada or Sierra Espuna. Females of llorenteae also differ from those of uhagonii andbolivari in that the fore wings are more than twice the length of the pronotum. In the field the calling song of the males, consisting of a rapid sequence of echemes, provides an easymeans of distinguishing this species from most other species of Omocestus occurring in southern Spain. 244 D. R. RAGGE 106 CD CN CNCN SONGS OF THE WESTERN EUROPEAN OMOCESTUS 245 COCM CS CN CS CS 00CN 246 D. R. RAGGE From minutissimus , which has a similar song-pattern, it may be distinguished by being longer (usuallylasting more than 6 s) and having more echemes in each sequence (usually more than 11). SONG (Figs 123-128). The calling song is a sequence of about 11-15 echemes lasting about 6-8 s.The sequence begins quietly, reaching maximum intensity after about 5-6 echemes. Theechemes are usually separated by intervals of about 100-200 ms, often becoming more widelyspaced towards the end of the sequence. Each echeme begins and ends quietly, giving aspindle-shaped oscillogram (Figs 125, 126). By the middle of the sequence each echeme lastsabout 350-600 ms and contains about 30-50 syllables repeated at the rate of about 85-95/s. Thecalling song is thus remarkably similar to that of Myrmeleotettix maculatus (Thunberg), thoughusually of shorter duration and composed of fewer echemes. DISTRIBUTION. Known only from the Sierra Nevada and Sierra Espuna in southern Spain. Omocestus simony i (Krauss)(Figs 120-122) Stenobothrus simonyi Krauss, 1892: 166. Syntypes of both sexes, CANARY ISLANDS: Lanzarote (Simony)(lost). REFERENCES TO SONG. No published work known to me. RECOGNITION. This is the only species of Omocestus known from the Canary Islands and so in practice thereis no problem in recognizing it. It is similar in appearance to raymondi but is smaller (total length less than14 mm in the male, less than 19 mm in the female). The calling song of the male is superficially very similarto that of uhagonii, oscillographic analysis being required to show that it has far fewer gaps in each syllable;it differs from that of raymondi in having a much slower syllable repetition rate. SONG (Figs 120-122). The calling song is an echeme lasting about 1-0-2-5 s and consisting ofabout 10-20 syllables repeated at the rate of 5-10/s. Each echeme begins quietly, reachingmaximum intensity about midway through its duration. Oscillographic analysis shows that thereare gaps in each syllable, similar in number to those in the calling song of raymondi (far fewerthan in uhagonii) but not arranged in the same pattern and maintained until the end of theecheme (Figs 121, 122). The echemes are repeated at irregular intervals. DISTRIBUTION. Known only from the Canary Islands, where it occurs on Lanzarote and Fuerteventura. References Alexander, R. D. 1957. The taxonomy of the field crickets of the eastern United States (Orthoptera: Gryllidae: Acheta). Annals of the Entomological Society of America 50: 584-602. 1962. The role of behavioral study in cricket classification. Systematic Zoology 11: 53-72. Alexander, R. D., Pace, A. E. & Otte, D. 1972. The singing insects of Michigan. Great Lakes Entomologist 5: 33-69.Alexander, R. D. & Thomas, E. S. 1959. Systematic and behavioral studies on the crickets of the Nemobius fasciatus group (Orthoptera: Gryllidae: Nemobiinae). Annals of the Entomological Society of America 52: 591-605.Ander, K. 1943. Revision der Orthopterensammlungen Zetterstedts. Lunds Universitets Arsskrift (N.F., Avd.2)38(7),23pp. Andrieu, A.-J. & Dumortier, B. 1963. Chantes d'Insectes. 30 cm disc, 33 rpm. Neuilly (Pacific).Azam, J. 1906. Description d'un Stenobothrus nouveau des Pyrenees [Orth.]. Bulletin de la Societe Entomologique de France 1906: 128-131.Bailey, W. J. 1975. A review of the African species of Ruspolia Schulthess [Orthoptera Tettigonioidea]. Bulletin de I'lnstitut Fondamental d'Afrique Noire (A) 37: 171-226.1980. A review of Australian Copiphorini (Orthoptera: Tettigoniidae: Conocephalinae). Australian Journal of Zoology 27: 1015-1049.Bei-Bienko, G. Ya. & Mishchenko, L. L. 1951. Locusts and grasshoppers of the U.S.S.R. and adjacent countries. Part II. Opredeliteli po Faune SSR, Izdavaemye Zoologicheskim Muzeem Akademii Nauk, Leningrad 40: 381-667. [In Russian. English translation published by Israel Program for Scientific Translations, Jerusalem, 1964, iv + 291 + xxi pp.] SONGS OF THE WESTERN EUROPEAN OMOCESTUS 247 Beier, M. 1956. Feldheuschrecken. Neue Brehm-Biicherei 179, 48 pp. Bellmann, H. 1985a. Heuschrecken. Beobachten, Bestimmen. 210 pp. Melsungen. 19856. Die Stimmen der heimischen Heuschrecken. Tape cassette. Melsungen (Neumann- Neudamm).Bolivar, I. 1876. Sinopsis de los ortopteros de Esparia y Portugal. Segunda parte. Anales de la Sociedad Espanola de Historia Natural 5: 259-372. 1878. Analecta orthopterologica. Anales de la Sociedad Espanola de Historia Natural 7: 423-470. 1887. Especies neuvas 6 criticas de ortopteros. Anales de la Sociedad Espanola de Historia Natural 16: 90-114. 1897. Catalogo sinoptico de los ortopteros de la fauna iberica [second part]. Annaes de Sciencias Natvraes 4; 2X8-232. Borck, J. B. von 1848. Skandinaviens ratvingade insekters natural-historia. xvi + 146 pp. Lund.Brisout de Barneville, L. 1855. Seance du 26 decembre 1855. Bulletin de la Societe Entomologique de France (3) 3: cxiv-cxvi.Broughton, W. B. 1955. L'analyse de 1'emission acoustique des Orthopteres a partir d'un enregistrement sur disque reproduit a des vitesses ralenties [pp. 82-88]. In Busnel, R.-G. [Ed.], Colloque sur 1'acoustique des Orthopteres. Annales des Epiphyties (fascicule special de 1954), 448 pp.Busnel, R.-G. 1955. Sur certains rapports entre le moyen d'information acoustique et le comportement acoustique des Orthopteres [pp. 281-306]. In Busnel, R.-G. [Ed.], Colloque sur 1'acoustique des Orthopteres. Annales des Epiphyties (fascicule special de 1954), 448 pp. Charpentier, T. de 1825. Horae entomologicae, adjectis tabulis novem coloratis. xvi + 262 pp. Bratislava.Chopard, L. 1922. Orthopteres et Dermapteres. Faune de France 3, vi + 212 pp.1939. Description d'une espece nouvelle du genre Omocestus [Orth. Acrididae]. Bulletin de la Societe Entomologique de France 44: 172-173. 1952. Orthopteroides. Faune de France 56 (1951), 359 pp. Duijm, M. & Kruseman, G. 1983. De krekels en sprinkhanen in de Benelux. 186 pp. Amsterdam. Eisner, N. 1974. Neuroethology of sound production in gomphocerine grasshoppers (Orthoptera: Acridi-dae). I. Song patterns and stridulatory movements. Journal of Comparative Physiology 88: 67-102. 1975. Neuroethology of sound production in gomphocerine grasshoppers (Orthoptera: Acrididae). II. Neuromuscular activity underlying stridulation. Journal of Comparative Physiology 97: 291-322. Faber, A. 1928. Die Bestimmung der deutschen Geradfliigler (Orthopteren) nach ihren Lautausserungen.Zeitschrift fur Wissenschaftliche Insektenbiologie 23: 209-234. 1929. Die Lautausserungen der Orthopteren. (Lauterzeugung, Lautabwandlung und deren biologis- che Bedeutung sowie Tonapparat der Geradfliigler.) Vergleichende Untersuchungen I. Zeitschrift furMorphologic und Okologie der Tiere 13: 745-803. 1932. Die Lautausserungen der Orthopteren II. (Untersuchungen uber die biozonotischen, tierp- sychologischen und vergleichend-physiologischen Probleme der Orthopterenstridulation. Methodik derBearbeitung und Auswertung von Stridulationsbeobachtungen. Einzeldarstellungen.) Zeitschrift furMorphologic und Okologie der Tiere 26: 1-93. 1936. Die Laut- und Bewegungsausserungen der Oedipodinen. Biologisch, tierpsychologisch und vergleichend-funktionell beschreiben; mit einem Erstnachweis von tonhaften Lauten bei Acridiidenund mehrfachen Lautformen bei Weibchen. Allgemeines zur Biologie de Paarungseinleitung undPaarung. Zeitschrift fur Wissenschaftliche Zoologie 149: 1-85. 1953. Laut- und Gebardensprache bei Insekten. Orthoptera (Geradfliigler). Teil 1. 198 pp. Stuttgart. Fischer, H. 1849. Beitrage zur Insekten-Fauna um Freiburg im Breisgau. Orthoptera. Jahresbericht des Mannheimer Vereines fur Naturkunde 15: 23-51.1850. Beitrage zur Insekten-Fauna um Freiburg im Breisgau. (Erste Fortsetzung.) Orthoptera. Jahresbericht des Mannheimer Vereinesfilr Naturkunde 16: 25-40.1853. Orthoptera Europaea. xx + 454 pp. Leipzig. Fulton, B. B. 1931. A study of the genus Nemobius (Orthoptera: Gryllidae). Annals of the Entomological Society of America 24: 205-237. 1952. Speciation in the field cricket. Evolution 6: 283-295. Grein, G. 1984. Gesange der heimischen Heuschrecken. Akustisch-optische Bestimmungshilfe. 30 cm disc, 33 rpm [with oscillograms] . Hannover (Niedersachsisches Landesverwaltungsamt).Harz, K. 1957. Die Geradfliigler Mitteleuropas. xxiv + 494 pp. Jena. 1975. Die Orthopteren Europas II. Series Entomologica 11, [viii +] 162 pp. Haskell, P. T. 1955. Intensite sonore des stridulations de quelques Orthopteres britanniques [pp. 154-167]. In Busnel, R.-G. [Ed.], Colloque sur 1'acoustique des Orthopteres. Annales des Epiphyties (fascicule special de 1954), 448 pp. 248 D. R. RAGGE 1957. Stridulation and associated behaviour in certain Orthoptera. 1. Analysis of the stridulation of, and behaviour between, males. British Journal of Animal Behaviour 5: 139-148. 1958. Stridulation and associated behaviour in certain Orthoptera. 2. Stridulation of females and their behaviour with males. Animal Behaviour 6: 27-42.Heller, K.-G. 1984. Zur Bioakustik und Phylogenie der Gattung Poecilimon (Orthoptera, Tettigoniidae, Phaneropterinae). Zoologisches Jahrbucher (Systematik) 111: 69-117. Hoist, K. T. 1970. Kakerlakker, Graeshopper og 0rentviste XXVII. Danmarks Fauna 79, 221 pp.Jacobs, W. 1950. Vergleichende Verhaltensstudien an Feldheuschrecken. Zeitschrift fur Tierpsychologie 7: 169-216.1953. Verhaltensbiologische Studien an Feldheuschrecken. Zeitschrift fur Tierpsychologie, Beiheft 1, vii + 228 pp.Jago, N. D. 1971. A review of the Gomphocerinae of the world with a key to the genera (Orthoptera, Acrididae). Proceedings of the Academy of Natural Sciences of Philadelphia 123: 205-343.Kirby, W. F. 1910. A synonymic catalogue of Orthoptera. Vol. III. Orthoptera Saltatoria. Part II. (Locustidce velAcridiidce). x + 674pp. London.Krauss, H. 1892. Systematisches Verzeichnis der canarischen Dermapteren und Orthopteren mit Di- agnosen der neuen Gattungen und Arten. Zoologischer Anzeiger 15: 163-171.Kutsch, W. 1976. Post-larval development of two rhythmical behavioural patterns: flight and song in the grasshopper, Omocestus viridulus. Physiological Entomology 1: 255-263.Kutsch, W. & Schiolten, P. 1979. Analysis of impulse rate and impulse frequency spectrum in the grasshopper, Omocestus viridulus, during adult development. Physiological Entomology 4: 47-53.Linnaeus, C. 1758. Systema naturae, 10th edition. 1, [iv +] 824 pp. Stockholm.Loher, W. & Broughton, W. B. 1955. Etudes sur le comportement acoustique de Chorthippus bicolor (Charp.) avec quelques notes comparatives sur des especes voisines (Acrididae) [pp. 248-277]. In Busnel, R.-G. [Ed.], Colloque sur 1'acoustique des Orthopteres. Annales des Epiphyties (fascicule special de 1954), 448 pp.Luquet, G. C. 1978. La systematique des Acridiens Gomphocerinae du Mont Ventoux (Vaucluse) abordee par le biais du comportement acoustique [Orthoptera, Acrididae]. Annales de la Societe Entomologique de France (N.S.) 14: 415-450.Marshall, J. A. 1983. The orthopteroid insects described by Linnaeus, with notes on the Linnaean collection. Zoo logical J ournal of the Linnean Society 78: 375-396.Otte, D. & Alexander, R. D. 1983. The Australian crickets (Orthoptera: Gryllidae). Academy of Natural Sciences of Philadelphia, Monograph 22, [vi +] 477 pp.Otte, D. & Cade, W. 1983. African crickets (Gryllidae). 1. Teleogryllus of eastern and southern Africa. Proceedings of the Academy of Natural Sciences of Philadelphia 135: 102-127.Pascual, F. 1978. Description de una nueva especie de Omocestus Bolivar, 1878, de Sierra Nevada, Esparia (Orth., Acrididae, Gomphocerinae). Eos 52 (1976): 159-165.Perdeck, A. C. 1957. The isolating value of specific song patterns in two sibling species of grasshoppers (Chorthippus brunneus Thunb. and C. biguttulus L.) [viii +] 75 pp. Leiden.Ragge, D. R. 1965. Grasshoppers, crickets and cockroaches of the British Isles, xii + 299 pp. London & New York.Ragge, D. R., Burton, J. F. & Wade, G. F. 1965. Songs of the British grasshoppers and crickets. 17 cm disc, 33 rpm. London (Warne).Ragge, D. R. & Reynolds, W. J. 1984. The taxonomy of the western European grasshoppers of the genus Euchorthippus , with special reference to their songs (Orthoptera: Acrididae). Bulletin of the British Museum (Natural History) (Entomology) 49: 103-151.Reynolds, W. J. 1986. A description of the song of Omocestus broelemanni (Orthoptera: Acrididae) with notes on its taxonomic position. Journal of Natural History 20: 111-116.Schmidt, G. H. & Baumgarten, M. 1977. Untersuchungen zur raumlichen Verteilung, Eiablage und Stridulation der Saltatorien am Sperbersee im Naturpark Steigerwald. Abhandlungen des Naturwissens- chaftlichen Vereins Wurzburg 15 (1974): 33-83.Schmidt, G. H. & Schach, G. 1978. Biotopmassige Verteilung, Vergesellschaftung und Stridulation der Saltatorien in der Umgebung des Neusiedlersees. Zoologische Beitrage (N.F.) 24: 201-308.Skovmand, O. & Pedersen, S. B. 1978. Tooth impact rate in the song of a shorthorned grasshopper: a parameter carrying specific behavioral information. Journal of Comparative Physiology 124: 27-36.Skovmand, O. & Pedersen, S. B. 1983. Song recognition and song pattern in a shorthorned grasshopper. Journal of Comparative Physiology 153: 393-401.Tsyplenkov, E. P. 1970. Harmful Acridoidea of the U.S.S.R. 272 pp. Leningrad. [In Russian. English translation published by Amerind Publishing Co., New Delhi, 1978, viii + 208 pp.] SONGS OF THE WESTERN EUROPEAN OMOCESTUS 249 Walker, T. J. 1962. The taxonomy and calling songs of United States tree crickets (Orthoptera: Gryllidae: Oecanthinae). I. The genus Neoxabea and the niveus and varicornis groups of the genus Oecanthus. Annals of the Entomological Society of America 55: 303-322.1963. The taxonomy and calling songs of United States tree crickets (Orthoptera: Gryllidae: Oecanthinae). II. The nigricornis group of the genus Oecanthus. Annals of the Entomological Society of America 56: 772-789. 1964. Cryptic species among sound-producing ensiferan Orthoptera (Gryllidae and Tettigoniidae). Quarterly Review of Biology 39: 345-355.Walker, T. J. & Greenfield, M. D. 1983. Songs and systematics of Caribbean Neoconocephalus (Orthop-tera: Tettigoniidae). Transactions of the American Entomological Society 109: 357-389.Wallin, L. 1979. Svenska grdshoppors och vartbitares sangldten. 9 pp. Uppsala. [With tape cassette of recorded songs.]Weber, H. E. 1984. Bestimmungsschlussel fur Heuschrecken und Grillen in Westfalen nach akustischen Merkmalen. Natur und Heimat 44: 1-19.Weih, A. S. 1951. Untersuchungen iiber das Wechselsingen (Anaphonie) und iiber das angeborene Lautschema einiger Feldheuschrecken. Zeitschrift filr Tierpsychologie 8: 1-41.Yakobson, G. G. & Bianki, V. L. 1902-1905. Orthoptera and Pseudoneuroptera of the Russian Empire and neighbouring countries, x + 952 pp. Leningrad. [In Russian.]Yersin, [A.] 1852. Seance du 18 fevrier 1852. Bulletins des Seances de la Societe Vaudoise des Sciences Naturelles 3: 100-104.1853. Seance annuelle et publique du 29 juin 1853. Bulletins des Seances de la Societe Vaudoise des Sciences Naturelles 3: 239-242.1854. Memoire sur quelques fails relatifs a la stridulation des Orthopteres et leur distribution geographique en Europe. Bulletins des Seances de la Societe Vaudoise des Sciences Naturelles 4: 108-128.1863. Description de deux Orthopteres nouveaux d'Europe. Annales de la Societe Entomologique de France (4) 3: 285-292.Zetterstedt, J. W. 1821. Orthoptera Svecice, disposita et descripta. 152 pp. Lund.Zhantiev, R. D. 1981. Bioacoustics of insects. 256 pp. Moscow. [In Russian.] Invalid names are in italics;principal page references inbold. Acheta 220alluaudi 222antigai 214, 222, 236aymonissabaudiae 222 biguttulus 214, 215, 220bolivari 214, 220, 221, 222, 226, 237, 243 broelemanni 221, 222, 224, 236brunneus 214, 215burri 214, 222 caucasicus 222 Chorthippus 214, 215, 220, 221corsicus 214, 222cuonaensis 222 demokidovi 222Dirshius 221 enitor 222Euchorthippus 220 femoralis 214, 222Gomphocerippus 221 INDEX Gomphocerus 221grossum 226 haemorrhoidalis 213, 222, 224, 226, 227, 233, 236heymonsi 222hingstoni 222 kaestneri 214, 222knipperi 214, 222 lecerfi 222 lepineyi 222 llorenteae 214, 220, 222, 223, 224, 237, 243 lopadusae 214, 222lucasii 222 maculatus 246megaoculus 222meridionalis 236minutissimus 214, 220, 222, 223, 224, 236, 237, 243, 246minutus 222mollis 215motuoensis 222Myrmeleotettix 221, 246 nanus 222 navasi 214, 222, 236 Nemobius 220 Neoconocephalus 220nyalamus 222 panteli 214, 219, 222, 223, 226, 236petraeus 213, 214, 220, 221, 222, 223,224,227,233Poecilimon 220pullus221 raymondi 214, 222, 224, 226, 227, 233, 246rufipes 213, 219, 220, 222, 224, 226, 227, 233, 236 simonyi 214, 222, 224, 246Stenobothrus221,236Stethophyma 226stigmaticus 236 tibetanus 222tzendsureni 222 uhagonii 214, 220, 221, 222, 223, 224, 236, 237, 243, 246uvarovi214,221,222 ventralis 227 viridulus 214, 219, 220, 221, 222,224, 226, 227, 233, 236 znojkoi 222 Occasional Papers on Systematic Entomology New Series The economic importance of insects, and the enormous number of species, have resulted in avast literature written in many languages; that which is of direct economic importance andrecently published can, to an increasing extent, be searched by using computerized data bases,but a great amount of more general information is unlikely to be available so readily in thenear future. The objective of this new occasional series is to make available in hard copy someof the basic data that is essential to the preparation of comprehensive accounts of the worldinsect fauna. The papers have been fully researched bibliographically and consist of checklistsof nominal taxa, and faunal lists with information on host plants and localities, based mainlyon the collections and libraries of the British Museum (Natural History). No. 1. A checklist of Neotropical arctiine and pericopine tiger moths. A. Watson & D. T. Goodger 72pp. inc. 4 colour plates 27 February 1986 No. 2. An annotated checklist of the Carabidae (including Cicindelinae, Rhysodinae and Paussinae) recorded from Borneo. N. E. Stork 26pp., 1 map 29 May 1986 Catalogues of BM (NH) Bulletins and Books free on request to Publication Sales. Titles to be published in Volume 53 A review of the Miletini (Lepidoptera: Lycaenidae) By J.N.Eliot Australian ichneumonids of the tribes Labenini and Poecilocryptini By I. D. Gauld & G. A. Holloway The tribe Pseudophloeini (Hemiptera: Coreidae) in the Old World tropics with a discussion onthe distribution of the Pseudophloeinae By W. R. Dolling The songs of the western European grasshoppers of the genus Omocestus in relation to theirtaxonomy (Orthoptera: Acrididae) ByD.R. Ragge Photoset by Rowland Phototypesetting Ltd, Bury St Edmunds, SuffolkPrinted in Great Britain by Henry Ling Ltd, Dorchester 19 PRE Bulletin of the British Museum (Natural History) The structure and affinities of theHedyloidea: a new concept of thebutterflies M. J. Scoble Entomology series Vol53 No 5 18 December 1986 The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in fourscientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology,and an Historical series. Papers in the Bulletin are primarily the results of research carried out on the unique andever-growing collections of the Museum, both by the scientific staff of the Museum and byspecialists from elsewhere who make use of the Museum's resources. Many of the papers areworks of reference that will remain indispensable for years to come. Parts are published at irregular intervals as they become ready, each is complete in itself,available separately, and individually priced. Volumes contain about 300 pages and severalvolumes may appear within a calendar year. Subscriptions may be placed for one or more ofthe series on either an Annual or Per Volume basis. Prices vary according to the contents ofthe individual parts. Orders and enquiries should be sent to: Publications Sales, British Museum (Natural History),Cromwell Road, London SW75BD,England. World List abbreviation: Bull. Br. Mus. nat. Hist. (Ent.) Trustees of the British Museum (Natural History), 1986 The Entomology series is produced under the general editorship of the Keeper of Entomology: Laurence A. Mound Assistant Editor: W. Gerald Tremewan ISBN 565 06023 6 ISSN 0524-6431 Entomology series Vol53No5pp251-286British Museum (Natural History)Cromwell RoadLondon SW7 5BD Issued 18 December 1986 The structure and affinities of the Hedyloidea: a new concept of the butterflies .. M. J. Scoble Department of Entomology, British Museum (Natural History), Cromwell Road, London SW75BD Contents Synopsis .................................................................................................... 25 1 Introduction .............................................................................................. 25 1 Material ................................................................................................. 253 Hedylidae Guenee ...................................................................................... 253 Generic synonymy .................................................................................... 254 Macrosoma Hiibner ..................................................................................... 254 General appearance ..................................................................................... 254 Description ................................................................................................ 254 Comments on selected structures .................................................................... 260 Adult .................................................................................................... 260 Juvenile stages ......................................................................................... 280 Discussion ................................................................................................. 281 Epilogue : a suggested phylogeny ..................................................................... 284 Acknowledgements ..................................................................................... 284 References ................................................................................................ 284 Synopsis The affinities of the Hedylidae, a group of about 40 Central and South American Lepidoptera, arereassessed. Currently, the taxon is treated as a tribe ('Hedylicae') of the Oenochrominae (Geometridae),but evidence is presented here to show that hedylids are butterflies although they look moth-like in manyrespects. The five generic names in the family are synonymized in this work with the oldest, MacrosomaHiibner. The taxonomic history of the Hedylidae is discussed, and the reasons for the proposed assignment of thefamily to the Rhopalocera are given. A description of the family is followed by comments on charactersselected for their biological or systematic interest. In this section comparisons are made between theHedylidae and other Lepidoptera, particularly the butterflies, and their phylogenetic relationships arediscussed. Hedylidae + Hesperioidea -I- Papilionoidea (s.str.) probably form a monophyletic taxon,although doubts have been expressed about the relationship between skippers and true butterflies. It istentatively suggested that within the Rhopalocera, the Hedylidae are more closely related to thePapilionoidea than are the Hesperiidae to the Papilionoidea; that is, the butterflies without the Hedylidaeare paraphyletic. The evidence of the butterfly affinities of hedylids comes from adults, larvae, and pupae; the structure ofthe thorax and the base of the abdomen of the adult was found to be particularly revealing. Key words: Lepidoptera, Rhopalocera, Hedyloidea stat. n., Hedylidae, Geometridae, morphology,phylogeny. Introduction Guenee (1857) described the Hedylidae as one of the 26 component families of the Phalenites.The Phalenites are, with modification, the equivalent of what are now called the Geometridae.The Hedylidae were treated by Prout (1910; 1931) as a tribe ('Hedylicae') of the Oenochromi-nae (Geometridae), and no reassessment of the relationships of the group has been attemptedsubsequently. In fact the Hedylidae are misplaced in the Geometridae, the most obvious reasonbeing the absence of tympanal organs from the base of the abdomen in the former - structuresthat are characteristic of the latter. Bull. Br. Mus. nat. Hist. (Ent.) 53 (5): 251-286 Issued 18 December 1986 252 M. J. SCOBLE The Hedylidae consist of some 40 species, from South or Central America. Below, I arguethat not only is this group misplaced in the Geometridae but also that its phylogenetic affinitiesare with the butterflies (Rhopalocera: Hesperioidea or skippers, and Papilionoidea or truebutterflies) despite the moth-like external appearance of the adult insects. Both Guenee (1857) and Prout (1910; 1931) noted certain similarities between hedylids andbutterflies. Guenee even went so far as to name one species napiaria because it resembled thepierid butterfly Pieris napi (L.). He named another species heliconiaria since it reminded him ofcertain species of the papilionoid genus Heliconius Kluk. Prout (1910; 1931) noted thebutterfly-like shape of the adults, and the remarkable girdled pupa in the two species for whichhe had information. However, neither of these workers suggested that these features meant thathedylids and butterflies were related, and they presumably regarded the similarities as indepen-dent derivations. Prout (1931) was clearly uncomfortable about the geometrid affinities of thegroup since he suggested that the 'Hedylicae' might deserve even a separate family status.However, neither he nor Guenee apparently considered that the butterfly-like features theyobserved might actually be homologous in the taxa concerned and thus indicative of true affinity.Why? There are two reasons, one philosophical and one practical. We tend to develop a conceptionof a given taxon. We think of butterflies as a package, and develop what might be termed abutterfly Gestalt - colourful, diurnal lepidopterans with clubbed antennae and amplexiformrather than frenate wing-coupling. This is not to say that we do not recognize exceptions. Malesof the skipper Euschemon Doubleday have a frenulum and retinaculum; satyrines generally arenot brightly coloured, many butterflies do not have strongly clubbed antennae, and a few arenocturnal. By definition, exceptions do not destroy a Gestalt. But hedylids do not conform to thetraditional butterfly package. They do not have clubbed antennae; in some species thesestructures are even bipectinate. In males of most species the frenulum and retinaculum are verywell developed, and the wing pattern in most species bears little resemblance to that ofbutterflies (Figs 1-18). Most species seem to be nocturnal. Most taxa are diagnosed by a set of attributes rather than by a single character or very fewcharacters. Even so, such diagnoses often need qualification to incorporate those taxa that donot fit - for example, primitive species that have not yet acquired all the attributes of themajority, or those specialized species that have lost them secondarily. Crowson (1970), notingthe essentialist nature of such diagnoses, called the groups Aristotelian to distinguish them fromtaxa delimited phylogenetically. Crowson also coined the phrase 'the non-congruence principle'to suggest that it is most unlikely that a taxon can be defined on more than one or two characters,since when a taxon originates many of the characters found in later members of the group havenot yet evolved. Characters we regard as typical of a given taxon do not evolve at once.Therefore, in the present case, we should not expect the ground plan of the butterflies as a wholeor even the true butterflies to exhibit the characteristics of the more advanced members of thegroup. It should not, then, be unexpected to find a group strongly affiliated to the butterflies thatdoes not fit our conception of them . However , if it is our stated aim to produce , as far as possible ,a phylogenetic classification, then merely one derived character shared between two taxaprovides evidence of monophyly. In fact, the Hedylidae possess several such characters. Having rejected the idea that hedylids might be related to butterflies, neither Guenee norProut looked further. This may largely have been because morphological studies were lessextensive in Guenee's time than they are today. Much the same can be said of Prout, who had aremarkable ability to observe structures and their significance on dried specimens, but did littledissection. (Only late in his career did he study genitalia.) In fact many of the characters thatsupport the suggestion that hedylids are butterflies are to be observed only after detailedexamination of the sclerites of the thorax and the base of the abdomen. The present study evolved from a project to re-diagnose the Oenochrominae (Geometridae),and to exclude those groups that currently make the subfamily unnatural (non-monophyletic).The extraordinary combination of attributes of the Hedylidae demonstrates that their affinitieslie neither with Oenochroma Guenee (the type genus of the Oenochrominae) and its relatives,nor with the Geometridae as a whole. In particular, abdominal tympanal organs, which are NEW CONCEPT OF BUTTERFLIES 253 typical of geometrids, are absent from hedylids. Also, prolegs are present on abdominalsegments 3 to 6 and 10 of the larva, unlike most geometrids where their number is reduced. Thefirst tergum of the abdomen is strongly 'pouched' (a condition described for papilionoids byEhrlich, 1958&), and pre- and postspiracular bars are present. The abdomen is curved andflattened laterally, particularly in the male. The remarkable pupa is attached to a leaf by a silkengirdle: there is no cocoon. The apical tibial spurs of the hindleg are absent in all species exceptthe male of one. The precoxal (or paracoxal) sulcus of the mesothorax joins the 'marginopleural'sulcus. In the metathorax the furcal apophysis is weakly sagittate in dorsal view. The pretarsus islost from the forelegs of the male although minute claws are present, and the tarsomeres arereduced to two by fusion. The head of the larva is extended into a pair of extremely long horns,and there is an anal comb on abdominal segment 10. These characters are all representedin various butterflies. In contrast, the moth-like features include the absence of clubs onthe antennae, and a well-developed frenulum and retinaculum in the males of mostspecies, although they are reduced and functionless in a few. The wing venation is peculiar:Rsi and Rs 4 share a common stem, but Rs 2 is separate from the cell and is not stalked with Rs 3 .Rsi and Rs^ are distinctly sinuate. Special modifications of the wings, including a complex ofsmall chambers at the base of the forewing and a prominent flap extending from Sc, aredescribed below. The body of this work aims to provide the evidence that suggests that hedylids are butterflies.Are the apparent butterfly attributes genuinely shared-derived characters of hedylids andbutterflies or are they convergent or parallel developments? In short, are the Hedylidae moreclosely related to the butterflies than to any other lepidopterans? A further question is: are the Hedylidae more closely related to the true butterflies(Papilionoidea) than to the skippers (Hesperioidea), or does the family represent the sistergroup of the Rhopalocera as a whole? A general description of the characters found in this family is provided, together with asummary of its biology and distribution. This is followed by observations on selected structureswith comments on their systematic importance. Finally, the evidence for the affinity of theHedylidae with the Hesperioidea and the Papilionoidea is assessed. Material The observations were based on specimens housed in the British Museum (Natural History).Juvenile stages of one species were lent by Mr R. O. Kendall, San Antonio, Texas. Examination of the structure of the thorax was necessarily limited to avoid breaking upnumerous specimens. From the study of more easily observed characters the family Hedylidaeappears to be a homogeneous group, and thoracic structures probably do not differ significantlybetween species. Thoracic structure was examined in species that showed most structuralvariation in the genitalia. HEDYLIDAE Guenee Hedylidae Guenee, 1857: 521. Type genus: Hedyle Guenee, 1857: 521. Guenee (1857) described the family for three genera, Hedyle Guenee, Venodes Guenee, andPhellinodes Guenee. With the exception of Prout (1910; 1931), most authors have been contentto describe new species or genera without discussing the taxonomic relationships of the group asa whole. Prout, in both these works, dealt with all the genera and species. He treated the taxonas a tribe - the 'Hedylicae' - of the subfamily Oenochrominae (Geometridae). Authors whopreceded Prout assigned the genera to various infrafamilial categories of the Geometridae,some of which are not currently in use, with no justification for their action. Kendall (1976), inhis notes on the life-history of one species, used the name Hedylidae, therefore effectivelyreviving Guenee's original status. The use of the name Hedylidae is fully justified by Guenee's priority, Kendall's subsequentusage, and the absence of any competing name. 254 M. J. SCOBLE Generic synonymy The species of the family show no fundamental structural difference from each other, and form ahighly compact group, particularly in the male genitalia, wing venation, wing shape, legs andantennae. Below, I synonymize the five genera that are currently accepted. Three of these aremonotypic. Macrosoma tipulata Hiibner and Lasiopates hyacinthina Warren are species that exhibit thegreatest differences from the majority of the family. The valvae of the male genitalia ofhyacinthina (Fig. 79) are of a more complex shape, the two components of the gnathos are notfused medially, and the juxta is characteristic. One reason for the proliferation of genera in the Lepidoptera is that since the study of themale genitalia became universal, there has been a strong tendency to treat as genera thosegroups where there is a morphological 'gap' in these structures. The consequence has been thatalthough a family may contain a few large genera there are usually several anomalous species. Ifthese are then excluded from one of the main genera each has to be assigned to a new genus. In afamily with several large genera it may be impossible to decide to which genus such 'outlier'species should be assigned. The description of a new genus therefore becomes inevitable. In afamily with one large genus there is often a good case to be made for including all species withinthat genus and using lower categorical rank, e.g. subgenus, or simply species-group, todistinguish outliers. In the present case, despite the synonymy, I have not so divided Macrosomain this way. This is best left until such time as a revision of all the species is undertaken. Thesynonymized names can then be used as subgenera, if required. MACROSOMA Hiibner Epirrita Hiibner, 1808. [Name unavailable, see Fletcher, 1979.] Macrosoma Hiibner, 1818: 10. Type species: Macrosoma tipulata Hiibner, 1818: 10 (by monotypy). Hedyle Guenee, 1857: 521. Type species: Hedyle heliconiaria Guenee, 1857 (by monotypy). [Fourdescribed species.] Syn. n. Venodes Guenee, 1857: 522. Type species: Venodes napiaria Guenee, 1857: 522 (by monotypy). [Mono-typic.] Syn. n. Phellinodes Guenee, 1857: 523. Type species: Phellinodes satellitiata Guenee, 1857: 523 (by monotypy).[Thirty-three described species.] Syn. n. Macrophila Walker, 1862: 1463, 1465. Type species: Macrosoma tipulata Hiibner, 1818 (by monotypy).[Junior objective synonym of Macrosoma, see Fletcher, 1979.] Hyphedyle Warren, 1894: 375. Type species: Hedyle rubedinaria Walker, 1862: 1464 (by originaldesignation). [Synonymized with Hedyle Guenee by Prout, 1910.] Lasiopates Warren, 1905: 310. Type species: Lasiopates hyacinthina Warren, 1905: 311 (by originaldesignation). [Monotypic.] Syn. n. Hedylidae remains the valid name of the taxon since family names are not rejected when, after 1960,their type genera are found to be junior synonyms. General appearance Hedylids are fairly delicate insects with curved abdomens, particularly in the male (Fig. 63). Figs1-18 show the main variation in wing patterns: many of the species bear semi-transparentpatches on the wings. The fore wings are weakly or strongly emarginate at the apex. The larva(Figs 19, 100, 101) bears a pair of very long 'horns' on the head capsule, prolegs on abdominalsegments 3 to 6 and 10, a pair of furcae on the anal segment, and an anal comb. The pupa (Figs20, 21) lacks a cocoon, and is attached to its substrate by a silken girdle. Description Adult Head (Figs 22-29). Frons narrow and protuberant (Figs 22, 23). Compound eyes large. Ocelli absent.Chaetosemata present, one chaetosema behind each eye (Fig. 22). Antenna (Figs 25-28): bases close NEW CONCEPT OF BUTTERFLIES 255 1 Figs 1-6 Adults of Hedylidae. 1, Macrosoma heliconiaria (Guenee), x 1-5; 2, M. hyacinthina (Warren),x 1-1; 3, M. tipulata Hiibner, x 1-1; 4, M. rubedinaria (Walker), x 1-5; 5, M. ustrinaria (Herrich-Schaffer), x 1-5; 6, M. leucophasiata (Thierry Mieg), x 1-5. together; filiform (Figs 26, 27) or bipectinate (Fig. 25); scaled dorsally. Pilifers present. Maxillae: palpiminute, one-segmented; galeae form a well-developed proboscis, not scaled at base. Labial palpiascending (Fig. 24), three-segmented with a deep sensory invagination on distal segment (Fig. 29).Tentorium: medial swelling on each tentorial arm (Fig. 23), but dorsal arms not developed. Thorax (Figs 30, 31). Prothorax with membranous patagia and parapatagia; patagia protuberant.Mesothorax: anepisternum well developed, not reduced (Fig. 30); precoxal sulcus (paracoxal sulcus ofScott, 1985) joins 'marginopleural' suture (Fig. 31). Metathorax: furcal apophyses weakly sagittate indorsal view; dorsal lamellae of secondary furcal arms fused mesally in their extension anteriorly from thearms to the furcal apophyses. Wing venation (Fig. 32). (Following the recommendations of Wootton (1979), in this work I label the 256 M. J. SCOBLE 8 10 11 12 Figs 7-12 Adults of Hedylidae. 7, Macrosoma subornata (Warren), x 1-5; 8, M. paularia (Schaus), X1-5; 9, M. coscoja (Dognin), x 1-5; 10, M. satellitiata satellitiata (Guenee), x 1 1 , hyaline patch arrowed;11, M. albifascia albifascia (Warren) x 1-1; 12, M. hedylaria (Warren), x 1-1. branches of the radial sector as Rsi, Rs 2 , etc., to distinguish them from the radius (/?i).) Fore wing:accessory cells absent. Rs\ and Rs 2 sinuous after branching from cell; Rs 4 stalked with /?s 3 , Rs 2 not stalkedwith Rs 3 (although it shares a common stem with Rs 3 and Rs 4 before the three branches arise); Rs 4 remotefrom MI; CuP often weakly indicated; M 2 about equal distance from M l as from M 2 ; anals forked at base.Hind wing: Rs separate from Sc+R] from near base of wing; M 2 about equidistant from MI and M 3 . Wing coupling (Figs 33-37). Male usually with strong frenulum and a very long, well-developedretinaculum; frenulum occasionally reduced and non-functional (Fig. 35). Female: frenulum composed ofa few bristles (Fig. 36); retinaculum absent. Wings .-pattern and specializations. Figs 1-18 illustrate the variety of shape and pattern. Vestiture: ratherweakly scaled; prominent piliform scales present particularly near base of wings (Fig. 38), and more NEW CONCEPT OF BUTTERFLIES 257 13 14 15 16 18 Figs 13-18 Adults of Hedylidae. 13, Macrosoma lucivittata (Walker), cf, x 1-1; 14, M. lucivittata(Walker), $, x 1-1; 15, M. nigrimacula( Warren), x 1-1; 16, M. leptosiata (Felder) , x 1-1, hyaline patcharrowed; 17, M. napiaria (Guenee), x 1-1; 18, M. hedylaria (Warren), ventral surface with wings foldedvertically, x 1-5. numerous on hind wing than fore wing. Upperside and underside with the same pattern. Areas ofsemi-transparent scales often present. Fore wing long, not broad, weakly falcate often with apex weaklyemarginate; posterior edge expanded into a small lobe at base of wing in males of some species (Fig. 43),sometimes merely a very weak expansion. Holding area (Haftfeld) not present on ventral surface (nor onmetascutum). Two small, but prominent chambers at base of wing (Figs 37, 42); one, predominantly in the 258 M. J. SCOBLE 19 20 21 Figs 19-21 Larva and pupa of Hedylidae. 19, larva of Macrosoma heliconiaria (Guenee), x 3-5; 20, 21.pupal exuviae of (20) M. lucivittata (Walker), x 1-5; (21) M. nigrimacula (Warren), x 1-5. ventral plane, is a distension of base of Sc, the other, predominantly in the dorsal plane, a distension of Cu.Prominent ventral fold expands from Sc on underside (Figs 39, 40). Scale-enclosed pocket present at verybase of underside of wing (Figs 40, 41). Hindwing rounded: in males often with folded area (of glassyappearance) present (Figs 10, 16, 44, 47); basally usually with associated sclerotized protuberances (Figs44-47, 50) and specialized scales (Figs 48, 49); base of costa weakly or, when frenulum is reduced, stronglyexpanded into a 'shoulder' (Fig. 35), composed of extended frenulum-plate and costal region. Base of fore wing (Figs 37, 51). Second median plate with substantial part hidden under base of analveins. Legs (Figs 52-62). Microtrichia present on ventral surface at base of tarsi of mid leg and hind leg of maleand all legs of female. Fore leg. Epiphysis present (Figs 52-54); tibial spurs absent. Male: tarsus (Figs 52, 55, 57, 58)two-segmented, number reduced by fusion but tarsus not shortened; pretarsus reduced to pair of minuteclaws (Fig. 58), empodium and pulvilli absent; tarsi not spined, but hairy; occasionally with a scent brush.Female: tarsomeres not fused, tarsi five-segmented (Fig. 54); pretarsus not reduced, bears a pair ofstrongly curved claws (Fig. 59) that are weakly and asymmetrically forked; pulvilli present; tarsus spinedventrally, not hairy. Mid leg: with one pair of tibial spurs; tarsus five-segmented, tarsomeres not fused; pretarsus notreduced, tarsal claws weakly forked; tarsus spined on ventral surface. Hind leg: with one pair (Fig. 61) or occasionally two pairs (Fig. 62) of tibial spurs; tibia somewhatswollen; otherwise as mid leg. Abdomen. Weakly to moderately curved (Fig. 63), flattened laterally not dorso-ventrally, and narrow-ing slightly towards base. Tympanal organs absent. Pregenital abdomen (Figs 64, 65, 68). Segment I with narrow prespiracular bar'and broad postspiracularbar (tergopleural). Tergum I with large membranous pouch and strong tergal braces. Sternum II with pairof mediolateral sclerotizations anteriorly; sternal apodemes reduced. NEW CONCEPT OF BUTTERFLIES 259 Male postabdomen (Figs 66, 67). Tergum VIII longer than sternum VIII forming a short hood. Female postabdomen . Segment VII in form of a short hood over the genitalia ; tergum VII large ; sternumVII shorter, usually bearing a series of prominent setae on fold just anterior to the ostium bursae (Fig. 90). Male genitalia (Figs 69-80). Genital capsule is 'deep' dorso-ventrally owing to length of lateral arms ofvinculum. Segment IX comprises a completely sclerotized ring, i.e. tergum and sternum fused. Tegumennarrow, sometimes prominently bilobed. Uncus large, usually triangular in ventral view since it narrowsapically; alternatively apex is truncated or otherwise modified (Fig. 83). Vinculum: ventral plate narrowand produced into a prominent narrow saccus; rarely, saccus is rounded. Gnathos usually large and heavilysclerotized; occasionally lateral arms do not meet medially. Valvae (Figs 77-82) simple, undivided, usuallyapproximately triangular. Diaphragm unsclerotized, or occasionally sclerotized as a juxta into a smallsimple plate or, rarely, a prominent U-shaped structure (Fig. 84). Aedeagus (Figs 85-89) a simple tube;vesica unmarked, or with weak sclerotizations. Female genitalia (Figs 90-99). Tergum VIII a well-sclerotized band. Anal papillae usually in form of apair of ear-like lobes. Anterior apophyses sometimes reduced (Figs 93, 96). Posterior apophysesdeveloped normally. Bursa copulatrix: a narrow ductus leads into a globose or sub-globose corpus; signumabsent or present; if present then of a characteristic shape (Figs 98, 99). Egg An egg, found while dissecting the female genitalia of a specimen, was oval. Larva Based on M. heliconiaria (comb, n.) (Fig. 19, and see Kendall, 1976), with further information on M.nigrimacula (Warren) (comb, n.) derived from larval exuviae housed in the British Museum (NaturalHistory). When this study was at an advanced stage, Mr T. Fox (London Butterfly House) provided acolour transparency of a larva (subsequently reared through) of M. cascaria (Schaus) (comb, n.), which hecollected in Monte Verde, Costa Rica. Head. Head-capsule bears a pair of long horn-like processes, flattened and broad in heliconiaria (Fig. 100),not broad in nigrimacula (Fig. 101). (Tips of processes out of focus in transparency of cascaria.}Hypognathous. Mandibles toothed. Six ocelli present. Secondary setae present. Thorax and abdomen. Not pigmented. Secondary setae on dorsum capitate in heliconiaria, shorter andtrifurcate in nigrimacula; row of knobbed setae run along midline in heliconiaria, absent from nigrimacula.Scoli absent. Primary setae not on pinaculi. True legs well developed, each bearing a single claw. Prolegspresent on abdominal segments III to VI, and on X (Fig. 19). Plantae of ventral prolegs with crochetsbiordinal and uniserial; arranged in two transverse bands in heliconiaria and in a penellipse in nigrimacula.Abdominal segment X: anal plate extended into a pair of furcae; anal comb present. Pupa (Figs 20, 21) Known for five species (heliconiaria, notes and illustrations by Kendall (1976), and personal examination;tipulata, exuvial remains; M. lucivittata (Walker) (comb, n.), nigrimacula, and cascaria (well-preservedpupal exuviae). The pupa is rather flattened, although raised protuberances arise from the thorax of heliconiaria,lucivittata, and nigrimacula, and from the abdomen of lucivittata, nigrimacula, and cascaria. The pupa of all five species is attached to the substrate by a fine silk girdle around the thorax, and by awell-developed cremaster. There is no cocoon. The temporal cleavage line appears to be absent, althoughthere is a very slight indication of a line running transversely across the top of the head. In each case (exceptfor heliconiaria where a specimen was stored in alcohol) the pupa was attached to a leaf. However, as far asI know, pupation occurred in captivity in all specimens available so whether or not hedylid pupae arealways attached to leaves in the wild requires confirmation. Host-plant Known only for heliconiaria: Buettneria aculiata Jacq., (Sterculiaceae). Recorded by Kendall (1976). Habits The following comments, on heliconiaria, are based on Kendall (1976), who has provided what appears tobe the only published information. The eggs are laid singly on top of leaves of Buettneria aculiata. The larvaeats a series of small holes in the leaf. At rest it lies along the midrib. Kendall notes that the adults ofheliconiaria are diurnal, as opposed to nocturnal, which may be exceptional for hedylids. Similarobservations were made for the larva of cascaria collected by T. Fox from Costa Rica (pers. comm.). Thespecimen was found lying along the midrib on the upper surface of the leaf. The leaf of the food-plant(unidentified) to which the pupal exuviae is attached is largely consumed. From what remains it is clear thatthe larva had eaten several holes in the leaf. 260 M. J. SCOBLE Parasites Mirax sp. and Apanteles sp. (Braconidae), identified by P. M. Marsh, USDA, Washington D.C., havebeen reared from heliconiaria larvae (R. O. Kendall, pers. comm.). Distribution Found in Central and South America: recorded from Paraguay, Brazil, Bolivia, Peru, Ecuador, Colombia,Venezuela, Guyana, Surinam, French Guiana, Panama, Costa Rica, Honduras, Guatemala, BritishHonduras, Mexico, Cuba, and Trinidad. Specimens have been collected mostly from montane rain forest, occurring in many localities in theAndes. Material has been collected at localities up to 7000 ft. Comments on selected structures Adult In this section further detail is provided about structures selected for their biological or systematic interest.Emphasis has been placed on comparing the Hedylidae with other macrolepidopterans, and in particularthe Rhopalocera. I include as macrolepidopterans the following taxa: Castnioidea, Hesperioidea, Papi-lionoidea, Geometroidea (sensu Common, 1970), Calliduloidea, Bombycoidea (including Sphingidae andSaturniidae), and Noctuoidea. (Geometroidea is used as a term of convenience in this work. Minet (1983)argued that the superfamily is not monophyletic.) Head. The compound eyes are large and constitute a relatively large portion of the total area of the head(Figs 22-24). There is some variation in size between species. The inner, dorsal angle of each eye is weaklyemarginate where it meets the base of the antenna (Fig. 22). Eye-size is of little value in assessingphylogenetic relationships of higher taxa since it is related to habits. Large eyes tend to be associated withlow light intensity. However, the large eyes of most true butterflies, a group that is typically diurnal, 'arespecialized for high visual acuity at the expense of relatively poor absolute sensitivity'. The reverse isapparently true of moths (Bernard, quoted in Ferguson, 1985). Most hedylids are apparently nocturnal; onindependent occasions, specimens have been collected at light by A. Watson, M. J. Matthews, and V. O.Becker (pers. comm.), and by the late C. L. Collenette (label data). However, heliconiaria is said to bediurnal by Kendall (1976). 22 23 Figs 22, 23 Head of Macrosoma tipulata Hiibner. 22, anterior aspect; 23, lateral aspect. The narrow frons bulges moderately (Fig. 23), but not to the degree found for example in theHeliothinae (Noctuidae). The narrowness of the frons means that the bases of the antennae are closetogether (Fig. 22). The antennae are usually filiform (Figs 26, 27), but in some species they are bipectinate(Fig. 25). In those with filiform antennae, the flagellar segments are rectangular, sometimes almost square,and each bears various sensilla. Usually the sensilla are long and arranged approximately in the shape of a'U' on each side of the segments. They are shorter in females. In tipulata the segments are uniformlycovered with shorter sensilla (Fig. 27). The more distal flagellar segments bear two or three longer setae,and also a sensillum basiconicum. NEW CONCEPT OF BUTTERFLIES 261 25 Figs 24-29 Head and associated structures of Hedylidae. 24, head, lateral aspect, of Macrosomahyadnthlna (Warren), x 7-4; 25, antenna of M. semlermis (Prout), d"; 26-28, antennal segments of (26)M. hyacinthina( Warren); (27) M. tipulataH\ibner;(28)M. venodes(Guene), $, arrow indicates reducedpectination; (29) labial palpus of M. tipulata Hiibner. Bipectinate antennae are found in both sexes of Macrosoma (= Hedyle) heliconiaria, and its closerelatives M. semiermis (Prout) (comb, n.), M. inermis (Prout) (comb, n.), and M. albipannosa (Prout)(comb, n.), and in males of Macrosoma (= Venodes) napiaria (comb. n.). In the females of napiaria eachpectination is reduced to a small knob on each side of every flagellar segment (Fig. 28). In those specieswith bipectinate antennae each pectination bears sensilla. The sensilla in males are slightly longer than infemales. 262 M. J. SCOBLE The antennae extend about one-half to just under two-thirds of the length of the fore wing. Whetherfiliform or bipectinate they bear lamellar scales dorsally. Chaetosemata are present, one chaetosema behind each eye (Fig. 22). They are fairly small, but can beseen on dried specimens. Ocelli are absent. The presence of chaetosemata and the absence of ocelli areconditions found in Hesperioidea, Papilionoidea, and in many Geometroidea amongst macrolepidopter-ans. Chaetosemata are absent from the front of the head. According to Jordan (1923), most Hesperiidaehave a pair of chaetosemata additional to those on the vertex, sited between the crest of scales on the fronsand the antennae. This additional pair is absent from papilionoids. Mouthparts. The maxillary palpi are minute (Figs 22, 23), a condition found generally in largerlepidopterans. The proboscis is well developed. Its base is not scaled, which contrasts with the typicallyscaled condition found in Pyralidae. The labial palpus (Fig. 29) is three-segmented. The palpi are generallymoderately ascending (Fig. 24) as a result of curvature of the basal segment, but are not appressed to thehead as they are in papilionids and pierids. The invaginated sensory pit of the terminal segment is acharacter of the lepidopteran ground plan (Kristensen, 1984ft). In those hedylids examined for thischaracter, the pit is deep, or moderately deep, and narrow. Vestiture. Long narrow scales form the main covering of the frons and vertex (Fig. 24). These scales areneither closely appressed to the head, nor do they form a strongly erect tuft. The labial palpi are coveredwith scales; on the lower surfaces, particularly of the two proximal segments, they are long and narrow,forming a fringe (Fig. 24). Thorax. To judge from the distribution of the sclerotized and membranous conditions of the patagia in truebutterflies (Ehrlich, 1958ft), the membranous condition must have arisen independently on severaloccasions. The sclerotized state is regarded as primitive. The development of membranous patagia hasprobably occurred independently in other macrolepidopterans, although the general condition (includingthat of hesperiids) is of sclerotized patagia. The loss of sclerotized patagia is given as a butterfly 'trend' byEhrlich (1958ft). The size of the anepisternum of the mesothorax (Fig. 30) is not reduced, i.e., there is no pronounceddorsal movement of the anapleural cleft. A reduction is a prominent and specialized feature of Papi-lionoidea (Brock, 1971; figures of Ehrlich, 1958ft). In Hesperiidae the reduction is not so pronounced.Scott (1985) states that it is very large in moths and skippers; however, it is not large in all skippers. In the mesothorax, the precoxal sulcus of hedylids appears to be fused with the sulcus (Fig. 31) called'marginopleural' by Shepard (1930) and Brock (1971). Usually these sulci are separate in the Lepidoptera.The precoxal sulcus does not curve towards the midline as it does in moths and skippers, and thebasisternum is 'open' - not 'closed' by the sulcus. Brock (1971) drew attention to the divergent fusedcondition in papilionoids and noted that the primitive arrangement was retained in some Hesperiidae, butalmost reached the specialized condition in advanced members of the group. Scott (1985) observed that inhesperiids, although the precoxal sulcus (his paracoxal sulcus) might meet the 'marginopleural', it does notfuse with it as in papilionoids. These structures are difficult to interpret and need a broader survey to testtheir soundness. There are two characters of particular importance in the metathoracic furca. In his study of theintegumental anatomy of the Monarch butterfly, Danaus plexippus (L.), Ehrlich (1958a) noted that vieweddorsally the apophyses of the furca were shaped like an arrowhead. That this sagittate shape might be thegeneral condition of both Hesperioidea and Papilionoidea was suggested by Brock (1971). (Brock treatedthe Hesperiidae as a family of the Papilionoidea.) In the Hedylidae the arrowhead is blunt, butnevertheless it is similar to that of other butterflies. The sagittate shape was not seen in the variousmacrolepidopterans examined by Brock. The shape is quite different in Oenochroma vinaria Guenee andO. polyspila Lower (Geometridae), andJosiafornax Druce (Dioptidae), which were examined during thecourse of this work. The character of the metathoracic furca considered by Brock (1971) to be the most prominent anddivergent for the butterflies was that which he referred to as 'the peculiar mesal fusion of the dorsal laminaeof the secondary arms from the point of their association with the apophyses to a point level with thethoraco-abdominal conjunctival insertion'. This character was also included as one of the shared derived(apomorphic) features of the Hesperioidea and Papilionoidea by Kristensen (1976). If I understandBrock's description correctly, this mesal fusion is present in the Hedylidae. It can best be seen in dorsalview. Despite the weight given to this character by Brock (1971), the structure seems to be present inOenochroma vinaria (Geometridae: Oenochrominae) and also in Archiearis Hiibner, a member of theprimitive geometrid subfamily Archiearinae. Wing venation (Fig. 32a). The venation does not seem to vary between species either in the number ofveins present, or their branching pattern and disposition. In the fore wing the arrangement of the veins ofthe radial sector (Rs) is characteristic of the family. The sinuous course run by Rs l and Rs 2 is such that they NEW CONCEPT OF BUTTERFLIES 263 a Figs 30-32 Hedylidae. 30, 31, mesothorax of Macrosoma rubedinaria (Walker): (30) ventrolateralaspect, (31) lateral aspect; 32, M. nigrimacula (Warren) (a) fore wing, (b) hind wing; anepisternum(am), katepisternum (km), basisternum (bm), precoxal sulcus (pcs), 'marginopleural' sulcus (mps). nearly meet at one point. The only stalking that occurs is that formed by the common stem of Rs 3 and Rs 4 .Veins R l , Rsi, and Rs 2 are all separate (at least from the cell). These veins fail to coalesce at any point so,unlike many macrolepidopterans, no secondary cells are formed. This pattern is generalized, andapproximates to that exhibited by the more primitive of the macrolepidopteran elements. Unlike severallineages of higher Lepidoptera, there is no inward migration ('splitting back') of veins. The arrangement of 264 M. J. SCOBLE 33 Figs 33-36 Wing-coupling apparatus in Hedylidae. 33, frenulum of Macrosoma hyacinthina (Warren);34, retinaculum of M. bahiata (Felder), view from ventral surface; 35, 36, frenular bristles of (35) M.napiaria (Guenee), cf, (36) M. satellitiata satellitiata (Guenee), $. Rs therefore provides few clues to relationships, although it further supports the exclusion of the groupfrom advanced geometroids, calliduloids, bombycoids, and noctuoids. Vein MI is remote from Rs 4 , a condition widely encountered, unlike the arrangement found inEpiplemidae and Uraniidae where these branches are associated. The usual geometroid condition, andindeed that of most higher Ditrysia, is that /?s 3 and Rs 2 share a common stem, and Rs 4 arises from this jointstem nearer to the cell (i.e. proximally). A notable exception is the Papilionoidea where Rsj and Rs 4 areoften associated as in hedylids. In the Hesperioidea there is no 'stalking' of the veins whatsoever; all runindependently from the cell to the apex of the termen. Vein M is very weakly, but definitely, visible in thecell. CuP is often present, but weak. Hind wing venation (Fig. 32b). Vein Sc+/?j is remote from Rs from the base of the cell; the conditioncontrasts with that found in geometrids and drepanids, but is not dissimilar from Epiplemidae andUraniidae. The presence of two anal veins in the hind wing is more frequently encountered in thebutterflies (including Hesperiidae), bombycoids and noctuoids amongst the higher Ditrysia, than ingeometroids (s.l.). The rather generalized occurrence of both these attributes therefore is of little value inassessing the relationships of Macrosoma. Nevertheless both features indicate yet further the remotenessof the genus from the Geometridae. M 2 does not arise nearer MI than M 3 . Wing coupling (Figs 33-37). In males, a single-spined frenulum is usually well developed on the hind NEW CONCEPT OF BUTTERFLIES 265 37 Fig. 37 Macrosoma hyacinthina (Warren), base of fore wing, ventral aspect; anterior chamber (ac),posterior chamber (pc), subcostal fold (fd). wing (Fig. 33). It engages a long retinaculum on the fore wing (Figs 34, 37, 39). The length of theretinaculum is related to the need for it to extend over the prominent subcostal fold so as to engage thefrenulum (Figs 37, 39). The frenulo-retinacular system is reduced and probably non-functional in the malesof three species: M. subornata (Warren) (comb, n.), M. leucoplethes (Prout) (comb, n.), and napiaria. Inthe first two species the frenulum remains as a single spine, but it is much weaker and shorter. In napiaria itis reduced to a few short bristles (Fig. 35). The retinaculum of napiaria is lost, while in subornata andleucoplethes it is reduced to a very short, and presumably functionless structure. Although the frenulumand retinaculum are reduced in subornata and leucoplethes, the system is well developed in the closelyrelated M. desueta (Prout) (comb. n.). In females there are sometimes a few weak costal bristles on the hindwing (Fig. 36). There is no sign of aretinaculum, and functionally the frenulo-retinacular system is lost in this sex. In Hesperioidea and Papilionoidea the frenulum and retinaculum are lost but remarkably, thesestructures are present in males of the hesperiid Euschemon. Scott (1985) notes that Euschemon isotherwise a typical pyrgine hesperiid and suggests an independent development of the frenulum andretinaculum through perhaps a reverse mutation. The loss of the frenulum and retinaculum or even of theretinaculum alone has occurred many times within the Lepidoptera besides the butterflies, e.g. inbombycoids, uraniids and some drepanids. Wing surface. There is no holding area (Haftfeld) on the underside of the base of the fore wing and themetascutum of the thorax. These patches of velcro-like microtrichia (aculei) are absent from Hesper-ioidea, Papilionoidea, Geometroidea, and most Bombycoidea, a feature considered to be related to theresting position of the wings (Common, 1969; Kuijten, 1974). In general, those Lepidoptera with wingsfolded over the body have holding areas on the fore wings and thorax, whereas those with wings heldvertically at rest or at right angles to the body do not. Vestiture. Two kinds of scales predominate on the wing-surface (Fig. 38): (a) broad, 'typical',overlapping scales, and (b) hair-like scales. The wings appear to be hairy in parts. The broad scales do notprovide a dense cover to the wings; hedylids are relatively weakly scaled. In several species there are smallhyaline patches devoid of scales, or larger areas with semi-transparent scales. The small hyaline patchesmay be seen with the naked eye on some specimens (Figs 10, 16). The hair-like scales are more dense on thebasal section of the upperside, and somewhat more dense on the underside. Figs 1-18 illustrates the main kinds of wing patterns in the Hedylidae. Of particular note is that themarkings on both dorsal and ventral surfaces are, with very minor exceptions, the same, compare Fig. 12with Fig. 18. At least one hedylid may hold its wings vertically at rest so exposing the underside (R. O.Kendall, pers. comm.) although this requires confirmation. Such a resting position may be related to thestrong patterning on the underside . In M. subornata the posterior half of the underside is pale and contrasts 266 M. J. SCOBLE sharply with both the rest of the underside and with that of the upper surface. The habit of holding thewings vertically is common among butterflies, and is also to be seen in several geometrids. (I am grateful toDr K. Sattler for suggesting the resting position of hedylids.) There are several relatively distinct wing-patterns in hedylids, as well as other patterns that are lessdistinct. Frequently, elements of one subgroup are found in another. For example, large areas covered bysemi-transparent scales are present in many species as are small, but distinctive, white triangular marks onthe costa of the fore wing near the apex. In some species the apex of the fore wing bears a large patch ofeither chestnut or dark brown. The illustrations show the distinctive emargination of the apex of theforewing, a condition found in many species, and the weakly falcate shape of them all. The comments thatfollow are an adjunct to the black and white photographs to provide a guide to colour. M. heliconiaria (Guenee) (Fig. 1). Ground colour darkish grey-brown with translucent patches on foreand hind wings. The small triangular mark on the costa of the fore wing is white. M. hyacinthina (Warren) (comb, n.) (Fig. 2). Ground colour dark grey-brown with a purple iridescence.More brown than grey at apex of fore wing. The pale patches on both wings are covered withsemi-transparent scales except for the subtriangular mark on the costa of the fore wing, which is white. M. tipulata Hiibner (Fig. 3). A pale species. The wings bear large areas of semi-transparent scales. Thedarker areas are pale brown, and the small irregular patches on the fore wing are white, edged with darkerbrown. M. rubedinaria (Walker) (comb, n.) (Fig. 4). Ground colour brown. The small dark streaks are darkbrown, edged with white. M. ustrinaria (Herrich-Schaffer) (comb, n.) (Fig. 5). A pale species. The darker scales scatteredunevenly over the wing surfaces are pale brown. M. leucophasiata (Thierry-Mieg) (comb, n.) (Fig. 6). The pale areas are white and the dark parts dark tovery dark grey-brown. M. subornata (Warren) (Fig. 7). Ground colour brown, dark brown near apices of wings. The patch onthe costa is cream, and the patch below it and the tiny spot to one side are white. On the underside theposterior half of the fore wing is pale, contrasting markedly with the equivalent section of the uppersurface. M. paularia (Schaus) (comb, n.) (Fig. 8). Ground colour of fore wing pale grey-brown flecked andstreaked with dark brown. The large spot at the apex of the fore wing is a warm pale brown. The large palearea on the hind wing is composed of semi-transparent scales. M. coscoja (Dognin) (comb, n.) (Fig. 9). Ground colour pale brown. Apex of fore wing dark brown. Thesmall patch at the base of the strong apical patch is composed of semi-transparent scales. M. satellitiata (Guenee) (comb, n.) (Fig. 10). Ground colour a darker brown than that of coscoja.Otherwise similar to that species. The small 'white' spot at the base of the hind wing represents the glassymodification in males discussed below. M. albifascia albifascia (Warren) (comb, n.) (Fig. 11). Ground colour grey-brown. The pale area on thefore wing is semi-transparent. The apex of the fore wing is a warm pale brown thickly edged with darkbrown. The very dark streak and patches on the fore wing are also dark brown. M. hedylaria (Warren) (comb, n.) (Figs 12, 18). The colour pattern is extremely similar to heliconiaria,although the apex of the fore wing is more brown than grey-brown. The name hedylaria is obviously acomposite of Hedyle (the genus in which heliconiaria was described but which is treated as a juniorsynonym of Macrosoma in the present work) and heliconiaria. M. lucivittata (Walker) c? (Fig. 13). Ground colour brown. The pale areas are covered with semi-transparent scales and many brown piliform scales. 9 (Fig. 14). Similar to the male but with a distinctsemi-transparent spot near the apex of the fore wing. M. nigrimacula (Warren) (Fig. 15). Ground colour brown, flecked and spotted with dark brown. Thesemi-transparent areas are poorly defined on the upper surface in this species. On the underside of the forewing the posterior half bears a large pale grey area that contrasts with the rest of the wing. M. leptosiata (Felder) (comb, n.) (Fig. 16). Ground colour brown. Both fore and hind wing edged withdarker brown distally, particularly on the fore wing. M. napiaria (Guenee) (Fig. 17). Ground colour off-white, formed by a covering of semi-transparentscales. The veins of the fore wing are a more strongly pigmented shade of cream. Modification of the fore wings (Figs 37, 39-43). The base of the fore wing bears some apparently uniquemodifications. These take the form of a ventral expansion of vein Sc into a fold, and the presence of twosmall, rounded chambers developed within the bases of certain veins. A pocket is also present and isenclosed by a dense fringe of scales. These modifications occur within both males and females. In addition,a distinct lobe is formed near the anal edge of the fore wing in males of some species. NEW CONCEPT OF BUTTERFLIES 267 Figs 38-43 Fore wing structure (ventral aspects) of Hedylidae. 38, vestiture of fore wing of Macrosomatipulata Hiibner; 39, base of fore wing of M. hyacinthina (Warren) to show subcostal fold (fd); 40, 41,scanning electron micrographs of base of fore wing of M. tipulata Hiibner; (40) x 16, large arrowindicates pocket, small arrow shows base of retinaculum, (41) detail of pocket x 40,? tympanum (tm);42, base of fore wing of M. hyacinthina (Warren), arrow indicates position of membrane (see text); 43,lobe at base of fore wing of M. hyacinthina (Warren). 268 M. J. SCOBLE The expansion of Sc on the underside of the fore wing is prominent (Figs 37, 39, 40). There is somevariation in its extent between species, but the form is basically similar. This fold is a hollow outgrowth ofthe vein (although both membranes are close together and not 'ballooned'). The fold extends a short wayalong the wing. At the extreme base of the fore wing, Sc is expanded into a hollow chamber (Figs 37, 42) themain part of which lies ventral to the plane of the wing. This chamber is divided from the fold by amembrane that runs at right angles to the plane of the wing. The position of this membrane is arrowed inFig. 42; it is visible as a dark line in this view. The chamber is invaginated caudally effectively providinganother pocket, which is enclosed by a thick fringe of hair-scales (Figs 40, 41). The posterior wall of thechamber is invaginated. From the ventral exterior lip of this invagination arises a thick fringe of hair-scales(Fig. 40 and particularly Fig. 41). Below this fringe, and presumably protected by it, is an extremelydelicate, transparent membrane. This is supported by a cuticular frame at the base of Cu, also lying in theventral plane of the wing. Stretched over the other (posterior) side of this frame is a whitish membrane,which closely resembles the tympanum found in the auditory organs of certain Lepidoptera (Fig. 41). Lying mainly within the dorsal plane of the fore wing is another small , closed chamber (Figs 37 , 42) . Thischamber is an expansion of the base of the cubitus, 'Cu^ in the terminology of Sharplin (1963). 47 .->ja ?? Figs 44-47 Scanning electron micrographs of specialized structures on hind wing (ventral aspects) ofHedylidae. 44-46, Macrosoma tipulata Hiibner: (44) general view of area x 16, (45) detail of squatprotuberance, x 140, (46) detail of finger-like protuberance, x 160; 47, M. zikani conifera (Warren)(comb, n.), base of hind wing x 16. NEW CONCEPT OF BUTTERFLIES 269 50 51 Figs 48-51 Wing structures of Hedylidae. 48, 49, scale from specialized area of hind wing (ventralsurface) ofMacrosoma tipulata Hiibner, (48) whole scale, x 2600, (49) detail to show serrations x 6500;50, specialized structure near base of hind wing (ventral surface) of M. nigrimacula (Warren); 51, base offore wing of M. hyacinthina (Warren) to show position of second median plate partly hidden under \A(arrowed). Although the function of these chambers is unknown, there are several similarities with the tympanalorgans found at the base of the fore wing in many nymphalid butterflies. The structure of these alartympanal organs is summarized by Bourgogne (1951). In both hedylids and those nymphalids with theseorgans, the tympanum is situated at the base of vein Cu. Unlike thoracic or abdominal hearing organs,there is no tympanic cavity; instead, the tympanum is superficial. In nymphalids, the membrane isprotected by scales, but in hedylids it is more exposed. In most of those nymphalids with alar hearingorgans, veins Sc and Cu are swollen basally, and the tracheal air sacs within the swellings communicate withthe tympanic sac (Bourgogne, 1951). It is uncertain whether there is a functional relationship between thesacs, but it should be noted that not all nymphalids with hearing organs have swollen veins. In theHedylidae the veins are not swollen. It is well known that some butterflies respond to sound. Swihart (1967) demonstrated the sensitivity to 270 M. J. SCOBLE sound of a small chamber at the base of each hind wing of Heliconius erato (L.). He also found a similar, butsmaller, chamber at the base of each fore wing. No information is available on the question of whether hedylids respond to sound, but on circumstantialevidence it seems likely that the modifications at the base of the fore wing function as an auditory organ. The possibility that the chamber enclosed by the fringe of hairs is a scent pouch cannot be discounted , butthe pocket is present in both males and females. The presence of 'scent pockets' tends to be sexuallydimorphic in lepidopterans, but is not always so. A small lobe extends from the anal edge of the fore wing near its base in the males of some species. It isprominent in hyacinthina (Fig. 43). A very weakly developed expansion of this area is found in manyspecies. Modifications of the hind wing (males). These are present in many species. Viewed with the naked eyethey appear as small, oval, glassy areas at the base of the wing on both the upper and the under sides (Figs10, 16). The modifications take the form of a folding of the wing membrane, making the surface appearrather crinkled or distorted when viewed under a dissecting microscope. Their appearance under thescanning electron microscope is shown in Figs 44 and 47. From this area one or more protuberances arise onthe under side of the wing (Figs 44-47, 50). The area is covered with specialized scales. In hyacinthina the modified area may be roughly divided into an inner, approximately oval section,which bears short piliform scales on both surfaces, and an outer section with both piliform scales andsemi-transparent laminar scales. A prominent knob-like structure is present on the underside. The modifications of the hind wing are well developed in many other species, and less pronounced inseveral more. Modifications present in one species may be absent from closely related ones. In M. bahiata(Felder) (comb, n.) for example they are present, while in coscoja, a close relative, they are not. In thesame group of species the structures are less strongly pronounced in M. albistria (Prout) (comb, n.) and M.uniformis (Warren). In tipulata two protuberances are present (Fig. 44) of which one is relatively squat(Fig. 45), and the other is a long digitate structure, which is slightly swollen apically (Fig. 46). The functionof these protuberances is uncertain, but since they are confined to the males they are probably scent organsof some kind. Figs 48 and 49 illustrate the specialized scales found on the surface of the structure in tipulata.The longitudinal ridges are strongly serrated (Fig. 49). 54 Figs 52-54 Fore legs of male Hedylidae. 52, Macrosoma semiermis (Prout), tibia and tarsus; 53, 54,epiphysis of (53) M. tipulata Hiibner, (54) M. napiaria (Guenee). NEW CONCEPT OF BUTTERFLIES 271 Wing base. According to Dr R. de Jong (pers. comm. and in prep.), to whom I am indebted for sketchesand information, the second median plate of the fore wing is always partly hidden under the base of vein 1Ain both Hesperioidea and Papilionoidea. The general ditrysian condition seems to be that the plate is notobscured by 1A. In the Hedylidae the second median plate is small and partly obscured by 1A (Fig. 51). Legs. An epiphysis is present on the fore leg of both males and females (Figs 52-56). The presence of thisstructure is widespread among macrolepidopterans including the Hesperiidae (with a few exceptions) andPapilionidae, but it is lost in many Pieridae, Nymphalidae, Libytheidae, and Lycaenidae. In males the fore tarsi are modified (Figs 55, 58). Although they are not reduced in length (in fact they arelong or very long) the segments are mostly fused so that only a long proximal segment and a short distalsegment are present. In tipulata the proximal segment is extremely long (Fig. 55). The pretarsus of the fore leg is reduced to a pair of small claws in males (Figs 52, 58). In some speciesthese claws are minute, and in all species they are only very weakly curved. There is no arolium, orempodium, nor are there pulvilli on the fore tarsi of males. The fore tarsi bear a prominent fringe ofpiliform scales giving them a feathery appearance. There are no spines. In females the fore tarsi are not modified (e.g. tipulata, Fig. 56). There are five tarsomeres and thepretarsus bears claws that are strongly curved and asymmetrically forked, pulvilli, and a large arolium. Thetarsomeres are covered with laminar, not piliform, scales, and a series of small spines is present. The mid and hind tarsi in both males and females are five-segmented, spined, and with a normallydeveloped pretarsus with claws, pulvilli and an arolium. The mid leg is generally longer than the hind leg,but shorter than the fore leg. Microtrichia or aculei ('tiny spines' of Scott, 1985) are not present on the dorsal aspect of the tarsi, unlikethe Papilionidae, Pieridae, some Nymphalidae and some Lycaenidae. According to Scott they are absentfrom hesperiids and from other macrolepidopterans. Microtrichia are, however, present on the ventralaspect of the tarsi of Hedylidae although their distribution is patchy, being best developed on the pulvilli.They are absent from the fore tarsi of males. Ventral microtrichia are probably widespread in Lepidoptera. A single pair of tibial spurs is present on the midleg of both sexes (Fig. 61). The hind leg of the male oftipulata bears two pairs of tibial spurs (Fig. 62), but usually hedylids (including the female of tipulata) haveone pair distally on the hind tibia. The proximal pair of spurs is found in many Lepidoptera, e.g. someThyrididae, some Hesperiidae, some Drepanidae, many Bombycoidea (including Sphingidae), and manyNoctuoidea (including Notodontidae). This loss is a classic example of a character that is prone toindependent reduction in many groups. Therefore the loss of spurs in most hedylids and in papilionoidsdoes not provide strong evidence for their monophyly. In one female specimen of hyacinthina there isapparently no pretarsus at all on the hind leg: the tarsus ends in a flask-shaped tarsomere (Fig. 60). There isno sign of claws arising from this last tarsomere. Of the two other females in the collection of the BritishMuseum (Natural History), one has lost both hind legs while in the other the tarsi, including the pretarsus,are normal. In many species the hind tibia is slightly swollen. A scent pencil occurs on the fore tibia of males of five species (hyacinthina, lucivittata, M. klagesi (Prout)(comb, n.), and nigrimacula, and probably (only one specimen available) on M. latiplex (Dognin) (comb.n.)). The pencil comprises a bunch of long hairs arising from the inner surface of the tibia. It is particularlywell developed in hedylaria. The modifications of the legs of the Hedylidae are all found within other butterflies. Information forHesperioidea and Papilionoidea is summarized in a useful table by Scott (1985: table 1). Pulvilli and thearolium are lost from the legs of Papilionidae, some Pieridae, and some Nymphalidae. In the Papilionidaeand the Pieridae the tarsomeres are not fused. In Nymphalidae the male fore legs are small although thetarsomeres are not lost or fused. In Lycaenidae the male fore legs are reduced to a single tarsomeretogether with reduction or loss of claws. The proximal pair of tibial spurs on the hind leg is absent from alltrue butterflies (Ehrlich, 1958ft; Scott, 1985) and hedylids (with the exception of the male of tipulata), butthese spurs are present in skippers. Scent brushes, when present, tend to be found on the hind legs ofbutterflies; they occur on the fore legs of Hedylidae. Abdomen. The abdomen is distinctly curved, particularly in males (Fig. 63), and is laterally flattened. Notonly are these features butterfly-like, but from my initial observations they resemble true butterflies(Papilionoidea) rather than skippers (Hesperioidea). This interpretation, however, remains tentative. Tergum I is 'pouched' (Figs 64, 68), that is, there is a pocket-like outgrowth of tergum I posteriorly thatoverlaps tergum II. Ehrlich (19586) used the word 'pouched' to describe the typical papilionoid condition.The pouching in hesperioids is not strong, and in some papilionids the pouch is reduced (Ehrlich, 1958ft). On segment I there are pre- and postspiracular bars. The prespiracular bar (Figs 64, 68) is a narrowsternopleural structure, which extends anteriorly from the antero-lateral margin of sternum II, and curvesaround the spiracle. The presence of a prespiracular bar is the general condition in macrolepidopterans,including hesperioids and papilionoids, although it is lost in pierids (Ehrlich, 1958ft). The presence of this 272 M. J. SCOBLE N *. 55 56 57 58 Figs 55-58 Fore legs of Hedylidae. 55, 56, Macrosoma tipulata Hiibner: (55) cf , (56) $ ; 57, M. hedylaria(Warren), O"; 58, M. semiermis (Prout), tarsal claws of O", arrowed. NEW CONCEPT OF BUTTERFLIES 273 59 60 61 Figs 59-62 Legs of Hedylidae. 59, pretarsus of Macrosoma tipulata Hiibner, 9, showing weak fork ofclaw; 60, last (5th) tarsal segment of hind leg of M. hyadnthina (Warren), $ ; 61, 62, hind tibia of (61) M.hedylaria (Warren), c? , (62) M. tipulata Hiibner, cf . structure in the Hedylidae is therefore of no particular phylogenetic significance. Brock (1971) considersthat both pre- and postspiracular bars are secondarily developed within the Papilionoidea and Hesper-ioidea (his Papilionoidea) and analogous to similar structures in certain Pyraloidea, Geometroidea andNoctuoidea. It is not entirely clear why Brock does not consider them homologous. The postspiracular bar is well-developed in the Hedylidae. It is a tergopleural structure (Figs 64, 68),which just fails to meet the sternum. According to Brock (1971) the postspiracular bar is derived from tergaI and II. Ehrlich (19586) considered the presence of a postspiracular bar to be specialized within thePapilionoidea, where he presumably considers it to have evolved independently within the group. Scott(1985) records the bar as absent in Papilionidae (Ehrlich said it was reduced or absent), present or absent inNymphalidae, present and large in Libytheidae, and absent in Lycaenidae. He notes its presence in orabsence from Hesperioidea and its absence from other macrolepidopterans. Both Ehrlich and Scottconsider the structure as derived. The occurrence of a similar structure in some other macrolepidopterans 274 M. J. SCOBLE 63 64 66 tg Vlll 67 65 Figs 63-67 Abdomen of Hedylidae. 63, 64, Macrosoma bahiata (Felder), lateral aspect: (63) wholeabdomen, (64) base of abdomen, small arrow - pre spiracular sclerotization, large arrow - postspiracular sclerotization; 65, sternum II (st II) of M. paularia (Schaus); 66, segment VIII of M. semiermis(Prout), d", lateral aspect; 67, tergum VIII (tg VIII) of M. tipulata Hubner, cf . (e.g. Thyrididae) noted by Brock is probably too isolated to suggest that the structure in the butterflies isnot derived. The tergal groove is well developed and leads into a strong tergal brace - the internal ridge of the groove(Fig. 64). There are no abdominal tympanal organs in the Hedylidae. Male genitalia (Figs 69-89). The lateral arms of the vinculum are long, so the genital capsule appears'deep', that is, expanded dorso-ventrally. The natural way to mount these structures for microscopic slidepreparation is laterally, since if they are mounted ventrally they tend to topple over on the slide. Most NEW CONCEPT OF BUTTERFLIES 275 68 69 Figs 68-71 Abdomen and male genitalia of Hedylidae. 68, segments I and II of Macrosoma bahiata(Felder), lateral aspect, abbreviations as for Fig. 64; 69-71, male genitalia, lateral view, of (69) M.semiermis (Prout), (70) M. hyacmthina (Warren), (71) M. paularia (Schaus). butterfly systematists mount male genitalia laterally for the same reason. The genitalia of the Hedylidae aresimilar to those of many butterflies in the shape of the genital capsule. They might be regarded as fairlygeneralized butterfly genitalia, but this does not provide strong evidence for the monophyly of hedylids andother butterflies since the arrangement might be primitive. Certainly the Castniidae also have genitalia thatare 'deep'. The lateral arms are fused with the tegumen so that segment IX of the abdomen forms a closedring (e.g. Figs 73, 74), a feature noted as a tentative ground plan character of the Amphiesmenoptera(Kristensen & Nielsen, 1979: 126; Kristensen, 1984&: 150). The condition frequently encountered in theLepidoptera is that of a distinct separation of the tergum and sternum of segment IX. The valva is approximately triangular, and unmodified. This condition is found in many species ofmacrolepidopterans including many butterflies. There are no particularly striking modifications of the main components of the genitalia. The greatest 276 M. J. SCOBLE Figs 72-76 Male genital capsules of Hedylidae. 72-74, lateral aspects of (72) Macrosoma tipulataHiibner, (73) M. lamellifera (Prout), (74) M. napiaria (Guenee); 75, 76, ventral aspects of (75) M.napiaria (Guenee); 76; M. nigrimacula (Warren) (a) valvae and juxta removed, (b) valvae and juxta. modifications are the shapes of the uncus in heliconiaria (Fig. 83), the valva (Fig. 79) and juxta (Fig. 84) ofhyacinthina, and the simplification of the genitalia of tipulata (Fig. 72). Female postabdomen and genitalia (Figs 90-99). Tergum VII is longer than tergum VI and forms a weakhood over the genitalia, when they are not extended. Tergum VIII is well-sclerotized and narrow (Fig. 90),and gives rise to a pair of thin, anterior apophyses. In most species examined these are sclerotized, but insome they are membranous (staining in Chlorazol black E), and short (Figs 93, 96). Presumably, in thosespecies with reduced apophyses the muscles are functionless during oviposition. In hyacinthina and NEW CONCEPT OF BUTTERFLIES 277 78 Figs 77-84 Male genitalia of Hedylidae. 77-82, valvae of (77) Macrosoma heliconiaria (Guenee), (78)M. semiermis (Prout), (79) M. hyacinthina (Warren), (80) M. lamellifera (Prout) (comb, n.), (81) M.nigrimacula (Warren), (82) M. napiaria (Guenee); 83, uncus of M. heliconiaria (Guenee), ventral view;84, juxta of M. hyacinthina (Warren), valvae at different plane of focus. 278 M. J. SCOBLE Figs 85-89 Male genitalia, aedeagus, of Hedylidae. 85, Macrosoma heliconiaria (Guenee); 86, M.hyacinthina (Warren); 89, M. napiaria (Guenee). hedylaria the reduced anterior apophyses are usually bent inwards (medially) at their ends at about rightangles. Reduced anterior apophyses are found in many butterflies. Terga IX and X form a sclerotized band more narrow than tergum VIII. The posterior apophyses, whichare derived from this, are sclerotized rods. Ventrally, sternum VII is folded before the ostium bursae thus forming a lip to this aperture. The lip isnearly always fringed with strong setae (Fig. 90). The sclerotizations around the ostium bursae includeanterior, posterior, and lateral components. The antevaginal and postvaginal sclerites are small, while thelateral sclerites (derived from sternum VIII) are large and extend laterally to unite with the anteriorapophyses in those species where they are sclerotized. The extent of these ostial sclerotizations variesbetween species. In heliconiaria the lamella postvaginalis is absent as are the setae on the fold of sternumVII. In tipulata the sclerotizations are virtually absent (Fig. 97). NEW CONCEPT OF BUTTERFLIES 279 91 Figs 90-91 Female postabdomen and genitalia of Hedylidae. 90, Macrosoma hedylaria (Warren),abdominal sternum VII, ventral aspect, antrum (an); 91, M. lucivittata (Walker), lateral aspect. Figs 92-95 Female genitalia of Hedylidae. 92-94, ventral aspect of (92) Macrosoma rubedinaria(Walker), (93) M. hyacinthina (Warren), anterior apophysis arrowed, (94) M. rubedinaria (Walker) toshow well-sclerotized anterior apophyses; 95, M. lucivittata (Walker), lateral aspect. 280 M. J. SCOBLE 98 Figs 96-99 Female genitalia of Hedylidae. 96, 97, lateral aspect of (96) Macrosoma hedylaria (Warren),anterior apophysis arrowed, (97) M. tipulata Hiibner; 98, 99, signum of (98) M. lucivittata (Walker), (99)M. rubedinaria (Walker). The ostium bursae leads into the ductus bursae, the first part of which is funnel-shaped with sclerotizedwalls, the antrum (Fig. 94). Beyond this the ductus bursae is narrow, membranous, and usually long. Itexpands suddenly into a globose or slightly elongate-globose membranous corpus bursae (Fig. 91). Thecorpus bursae may bear a single signum of a characteristic shape (Figs 98, 99) , or the signum may be absent. The weakly telescoped ovipositor ends in a pair of large, soft, setose, ear-like lobes (Figs 92, 93). Juvenile stages The shape of the egg varies within the Lepidoptera, but it is essentially flat or upright (Chapman, 1896).Many butterflies have upright eggs. In the Hesperiidae, however, they are flat. The oval egg dissected fromthe macerated abdomen of a female hedylid appears to be of the flat variety, but its shape did not resemblethat of the hesperiid egg. Since both flat and upright eggs are found within single families, or even singlegenera, (Hinton, 1981), the gross shape of the egg is of limited phylogenetic significance. The larva (Fig. 19) exhibits features of various butterfly families rather than any one of them. Itshorn-like processes (Figs 19, 100, 101) strongly resemble those of many Nymphalidae. The secondary setaeon the body are like those of some pierids. The crochets on the ventral prolegs of the final instar , which arearranged in a penellipse or as transverse bands, are neither like those of hesperiids (in which they form acircle) nor like papilionoids (where, in post first instars, they are usually arranged in a mesoseries) . An analcomb (found in Hedylidae) is usually present in Hesperiidae and at least some Pieridae among themacrolepidopterans. The pupa (Figs 20, 21) resembles that found in Papilionidae (see Igarashi, 1984) or Pieridae, amongst thebutterflies, in that it is attached to the substrate by both a girdle and a cremaster, and that there is nococoon. A girdle is present in some sterrhine geometrids, e.g. Anisodes Guenee (see Common, 1986) andCyclophora Hiibner. However, these genera lack the other butterfly characters discussed in the present NEW CONCEPT OF BUTTERFLIES 281 100 Figs 100, 101 Half of head capsule of larval exuviae of Hedylidae. 100, Macrosoma heliconiaria(Guenee); 101, M. nigrimacula (Warren). work. Furthermore, whereas the girdles of butterflies (including hedylids) are thoracic, those of sterrhinegeometrids are spun around the abdomen (Holloway, pers. comm. - information derived from unpub-lished manuscripts housed in the British Museum (Natural History); and see Fig. 3 of Common, 1986 for anAnisodes girdle). The presence of a thoracic girdle is probably a ground plan character of the Rhopalocera,although weak cocoons are usually present in hesperiids. Loss of the cocoon is a typically papilionoidcharacter, although it is present in Parnassius Latreille (Papilionidae) and weak and web-like in somesatyrines. The girdle is present in Pieridae and Papilionidae, but is lost in Nymphalidae and manyLycaenidae. A termporal cleavage line is said by Scott (1985) to be present in Hesperiidae, but absent fromPapilionoidea. Mosher (1916) referred to the line as the epicranial suture. The apparent absence of the linein the Hedylidae is therefore like that of true butterflies rather than that of skippers. A temporal cleavageline is found in many moths. Mosher notes its presence in Lycaenidae, although Scott (1985: table 1)records it as absent from that family. Discussion The Hedylidae exhibit a high degree of structural uniformity. Several characters seem to beunique to the family, which provides evidence for its suspected monophyly. The most strikingare the small chambers at the base of the fore wing, the protuberances on the underside of thehind wing, the sinuous course of veins Rsi and Rs 2 of the radial sector in the fore wing, and thelong fore tarsi of the male composed of only two tarsomeres. Two important questions about the phylogeny of the Hedylidae need to be discussed. First,are hedylids members of a taxon composed of Hesperioidea plus Papilionoidea plus themselves,i.e., are they butterflies in the broadest sense? Second, if so, what are the relationships ofhedylids, hesperiids, and papilionoids to each other: do hedylids represent the sister group of thetaxon Hesperioidea plus Papilionoidea (i.e. the Rhopalocera), or are they more closely relatedto the Hesperioidea or to the Papilionoidea than are either of these two related to each other? Ishall argue that the Hedylidae are indeed 'butterflies' in the broadest sense. That they mayrepresent the sister group of the Papilionoidea remains an intriguing possibility. This possibilityis not a conclusion of this work: it would be premature and requires further comparison of thethree taxa involved. 282 M. J. SCOBLE A separate, but related, problem is how to treat the nomenclature. Papilionoidea andHesperioidea are widely used names, although Hesperioidea, consisting as it does of a singlefamily, is redundant (Farris, 1976; Wiley, 1979), assuming that the Megathyminae are notviewed as an independent family. How should the Hedylidae be treated? Although the butterflies are one of the best studied groups of organisms, the publishedevidence for their monophyly is certainly not overwhelming. Ehrlich (19586) regarded hesper-iids as papilionoids that retain a great many primitive characters. Both Ehrlich and Kristensen(1976) retained the two superfamilies and did not synonymize them, Ehrlich because heconsidered the phenetic distance between them sufficiently great, and Kristensen because heconsidered the sister-group relationship a reasonable working hypothesis but one that requiredconfirmation. Although the consensus is that hesperioids and papilionoids do form amonophyletic group, the characters on which this argument is based are not entirely convincing. Kristensen (1976) records six possible specialized characters (apomorphies) shared by the twotaxa, two of which he considers doubtful. However, none of them have been examined in a greatnumber of species of butterflies and moths. Until they are examined they must remain somewhatin doubt. A particular problem with these characters is that they are not observable on driedspecimens. Some of the characters can only be seen on alcohol-preserved, or fixed material (e.g.the twist in the oblique lateral dorsal muscle of the mesothorax, and the structure of the aorta inthe mesothorax). Those that can be observed on dried specimens require maceration of the heador thorax, and have been examined, inevitably, in a limited number of Lepidoptera. A detailedstudy of the relationships of the Hesperioidea and the Papilionoidea is being undertaken by DrR. de Jong, who has found several specialized characters considered to support the sister-grouprelationship of the two taxa. If hedylids are the sister group of papilionoids then several of deJong's characters will have to be treated as homoplasious. It would be premature to treat theHedylidae as the sister group of the Papilionoidea while de Jong's detailed work is in progress.For this reason it has not been possible to make unequivocal statements about precisely whichtaxa some of the characters 'define'. Recently Scott (1985) has discussed the phylogeny of the butterflies. From his table 1, thecharacters that appear to support the monophyly of the papilionoids and hesperioids are: thepresence of a postspiracular bar, found in skippers but not in Papilionidae, some Nymphalidae,and Lycaenidae, and the presence of a secondary sternopleural sulcus, variable in size, butpresent in all butterflies. No other specialized features are tabulated by Scott as unique to the adults of all butterflies,with the possible exception of the absence of spurs on the tibia of the mid leg. However,although the spurs are absent from skippers, they are lost only in Papilionidae within the truebutterflies so the condition is likely to be a parallel development. Of the attributes of the Hedylidae described and discussed above, 13 are of particularsignificance in the discussion of the question of the relationship of the Hedylidae to the skippersand true butterflies. Of these several resemble the condition in true butterflies rather thanskippers. (1). Apophyses of metathoracic furca sagittate. The sagittate condition appears to be aspecialized character of the three taxa. It certainly deserves examination in many more speciesto establish that it is definitely absent from moths and therefore diagnostic of butterflies. (2). Pupa girdled. Ehrlich (19586) and Scott (1985) regard a girdled pupa in butterflies asprimitive and its absence as specialized. Scott considers this condition to have been present inthe ancestor of hesperioids and papilionoids. There is no general survey of the occurrence ofgirdled pupae in the Lepidoptera, so to suggest that the condition is diagnostic of hedylids,hesperioids, and papilionoids needs to be treated with caution. This attribute derives itstaxonomic strength in conjunction with others. (3). Second median plate of fore wing lies partly under the base of vein L4. Although thischaracter requires examination in a wide range of Lepidoptera, it appears to support the view ofthe monophyly of the Hesperioidea plus the Papilionoidea (de Jong, pers. comm.) plus theHedylidae. (4). Presence of an anal comb in the larva. This structure, which is present in Hesperioidea NEW CONCEPT OF BUTTERFLIES 283 and Hedylidae, seems to be confined to the Pieridae in the Papilionoidea. It may reasonably beassumed to be a ground plan character of the Papilionoidea. The presence of an anal comb in theTortricidae is probably an independent development. However, if the Macrolepidoptera are notmonophyletic then the closest relatives of the Rhopalocera may be found among the microlepi-dopterans. The tortricoid/cossoid assemblage might be a serious contender. (5). Presence of a postspiracular bar on the first abdominal segment. Postspiracular scler-otizations are present on the first abdominal segment of non-ditrysian moths (see for exampleKyrki, 1983; Kristensen, 1984a - 'lateral lobe of tergum F lo in his figure 2). That the presence ofa secondary sclerotization (postspiracular bar) is a character of the ground plan of hedylids,hesperioids, and papilionoids is a reasonable supposition. The presence of a postspiracular barin certain Pyraloidea (those without tympanal organs), certain Geometroidea (Drepanidae,Thyatiridae, and most Uraniidae), and Noctuoidea (many families) (Brock, 1971) must castsome doubt on the value of the structure as an indicator of the monophyly of hedylids,hesperioids, and papilionoids. The postspiracular bar is recorded as absent from macrolepidop-terans by Scott (1985: table 1), although he made no comment on the postspiracular sclerotiza-tions in the macrolepidopteran families discussed by Brock (1971). (6). Reduction of anterior apophyses in female genitalia. The weak, reduced anteriorapophyses in some Hedylidae is a condition found also in many butterflies (de Jong, pers.comm.). Whether this is a ground plan character of hedylids, hesperioids, and papilionoids, orwhether it is one that is subject to extensive parallelism is unknown. (7). Abdomen curved, particularly in males. As noted above, the shape of the abdomenresembles that of the true butterflies rather than that found in skippers. This character is onlydoubtfully a unique derivation of Hedylidae + Papilionoidea. A detailed study, preferablymeasuring the degree of curvature, is needed to ascertain whether the condition in Hedylidaeand Papilionoidea is greater than that found in Hesperiidae. (8). Abdominal tergum I strongly 'pouched'. (9). Precoxal (paracoxal) sulcus joins 'marginopleural' sulcus. This character appears tobe a specialization of the Hedylidae and the Papilionoidea. The main objections to this vieware that the sulci are difficult to observe, and that they have been examined in relatively fewspecies. (10). Pupal cocoon lost. Unlike the condition in hesperioids, the cocoon of hedylids and mostpapilionoids is lost. Loss of a cocoon occurs elsewhere in the Lepidoptera, so this character is ofdoubtful value and derives its strength in consideration with others. (11). Loss of temporal cleavage line in pupa. The loss of this cleavage line in both hedylidsand papilionoids is further, although rather weak, evidence for their close relationship. (12). Crochets of ventral prolegs of larva not forming a complete circle. This character isbased on an examination of only two species. The penellipse ofnigrimacula, and the transversebands of heliconiaria are neither the typical circle of hesperiids nor the usual mesoseries ofpapilionoids. It is possible, considering other characters of the family, that the hedylid conditionis part of a transformation series circle > penellipse > mesoseries, but this remains a tentativesuggestion. (13). Loss of pretarsus in the fore leg of males. Reduction or loss of the pretarsus iswidespread in butterflies. A reduction of the pulvilli and arolium is the condition found in all legsof the most primitive papilionoids. The initial reduction, confined as it is to the fore legs of themale, appears to be the first stage of a reduction fully established in the papilionoids. Thereforeeither the character is a shared specialization of the Hedylidae plus Papilionoidea or it hasdeveloped in parallel in the two groups. The possibility remains that the pretarsus is redeveloped(a reversal) in hesperiids. Nomenclature The Papilionoidea are generally regarded as a separate superfamily from the Hesperioidea, althoughBrock (1971) combined the two in Papilionoidea. In the present study I follow convention and retain bothtaxa as superfamilies since the relationships between the Hesperioidea and the Papilionoidea is beingstudied in detail by R. de Jong. Also, Kristensen (1976) noted some uncertainties about the sister group 284 M. J. SCOBLE relationship of the two groups, and recently Stallingseffl/. (1985, quoting C. L. Remington) made a similarpoint. It would be premature to formally asign the Hedylidae to the Papilionoidea at this stage, so to retainequivalence of rank they are treated as a superfamily (Hedyloidea stat. n.). Consequently, theRhopalocera now include three superfamilies Hesperioidea, Hedyloidea, and Papilionoidea, all ofinterchangeable position (sedis mutabilis, Wiley, 1979). Epilogue: a suggested phytogeny Although this work concluded with the conservative suggestion that hedylids are rhopaloceransof which the precise relationships remain uncertain, the following dendrogram (Fig. 102) is afully resolved three taxon statement presented to stimulate critical assessment. 7-13 1-6 Fig. 102 Dendrogram to suggest possible phylogenetic relationships of Hesperioidea, Hedyloidea andPapilionoidea. Numbers represent the characters in the Discussion. Relative strengths and weaknessesof the characters are not indicated, but are considered in the text. Acknowledgements I am indebted to my colleagues for their interest in this work, and for their critical comments andsuggestions throughout its course. In particular I thank P. R. Ackery, D. J. Carter, J. D. Holloway, I. J.Kitching, K. Sattler, J. A. Schoorl, R. I. Vane-Wright, A. Watson and P. E. S. Whalley. I am grateful toM. J. Matthews, who helped to record distribution data. Useful information was also received from V. O.Becker. R. O. Kendall supplied Juvenile stages of Macrosoma heliconiaria, which added an extradimension to this study. I am extremely grateful to him. R. de Jong and N. P. Kristensen provided veryvaluable comments on the manuscript. I thank both, and further express my gratitude to R. de Jong whomade available unpublished sketches and other information from his work on the phylogenetic rela-tionships within the butterflies. I thank T. Fox and D. Lees for generously supplying collection details,photographs of a larva and adult, and pupal exuviae of M. cascaria. I am grateful to members of thePhotographic Unit of the British Museum (Natural History) for many of the illustrations: Ms S. H. Barnes,scanning electron micrographs; P. V. York, photo micrographs; H. Taylor, photo macrographs. References Bourgogne, J. 1951. Ordre des Lepidopteres, pp. 174-448. In Grasse, P. P., Tralte de Zoologie 10. Paris.Brock, J. P. 1971. A contribution towards an understanding of the morphology and phylogeny of theDitrysian Lepidoptera. Journal of Natural History 5: 29-102. NEW CONCEPT OF BUTTERFLIES 285 Chapman, T. A. 1896. On the phylogeny and evolution of the Lepidoptera from a pupal and ovalstandpoint. Transactions of the Entomological Society of London 1896: 567-587. Common, I. F. B. 1969. A wing-locking or stridulatory device in Lepidoptera. Journal of the AustralianEntomological Society 8: 121-125. - 1970. Lepidoptera. In Mackerras, I. M. (Ed.), The Insects of Australia. 1029 pp. Melbourne. 1986. Exposed pupae in some Australian moths. Entomological Society of Queensland News Bulletin13: 120-123. Crowson, R. A. 1970. Classification and Biology. 350 pp. London. Ehrlich, P. R. 1958a. The integumental anatomy of the Monarch butterfly Danaus plexippus L. (Lepidop-tera: Danaiidae). University of Kansas Science Bulletin 38: 1315-1349. - 19586. The comparative morphology, phylogeny and higher classification of the butterflies. Uni-versity of Kansas Science Bulletin 39: 305-370. Farris, J. 1976. Phylogenetic classification of fossils with recent species. Systematic Zoology 19: 83-92.Ferguson, D. C. 1985. Contributions towards reclassification of the world genera of the tribe Arctiini, Part 1 - Introduction and a revision of the Neoarctia - Grammia group (Lepidoptera: Arctiidae; Arctiinae). Entomography 3: 181-275.Fletcher, D. S. 1979. Geometroidea. In Nye, I. W. B., The Generic Names of Moths of the World!: 243 pp. London.Guenee, A. 1857. Species general des Lepidopteres. In Boisduval, J. A. & Guenee, A., Histoire Naturelle des Insectes 10: 584 pp. Paris. Hinton, H. E. 1981. Biology of Insect Eggs 2: 475-778. Oxford.Hiibner, J. 1808. Erste Zutrage zur Sammlung exotischer Schmetterlinge 4. Augsburg. 1818. Zutrage zur Sammlung exotischer Schmetterlinge 1: 1-32, 18 pis. Augsburg. Igarashi, S. 1984. The classification of the Papilionidae mainly based on the morphology of their immature stages. Tyo to Ga 34: 41-96.Jordan, K. 1923. On a sensory organ found on the head of many Lepidoptera. Novitates Zoologicae 30: 155-158, 1 pi.Kendall, R. O. 1976. Larval foodplants and life history notes for eight months from Texas and Mexico. Journal of the Lepidopterists' Society 30: 264-271.Kristensen, N. P. 1976. Remarks on the family-level phylogeny of butterflies (Insecta, Lepidoptera, Rhopalocera). Zeitschrift fur Zoologische Systematik und Evolutionsforschung 14: 25-33. 1984a. The pregenital abdomen of the Zeugloptera (Lepidoptera). Steenstrupia 10: 113-136. - 19846. Studies on the morphology and systematics of primitive Lepidoptera (Insecta). Steenstrupia10: 141-191. Kristensen, N. P. & Nielsen, E. S. 1979. A new subfamily of micropterigid moths from South America. A contribution to the morphology and phylogeny of the Micropterigidae, with a generic catalogue of the family (Lepidoptera: Zeugloptera). Steenstrupia 5: 69-147.Kuijten, P. J. 1974. On the occurrence of a hitherto unknown wing-thorax coupling mechanism in Lepidoptera. Netherlands Journal of Zoology 24: 317-322.Kyrki, J. 1983. Adult abdominal sternum II in ditrysian tineoid superfamilies - morphology and phylogenetic significance (Lepidoptera). Annales Entomologici Fennici 49: 89-94.Minet, J. 1983. Etude morphologique et phylogenetique des organs tympaniques des Pyraloidea. 1 - Generalites et homologies (Lep. Glossata). Annales Societe Entomologique de France 19: 175-207.Mosher, E. 1916. A classification of the Lepidoptera based on characters of the pupa. Bulletin of the Illinois State Laboratory of Natural History no. 212: 15-159, 9 pis.Prout, L. B. 1910. Lepidoptera Heterocera, Fam. Geometridae, Subfam. Oenochrominae. In Wytsman, P., Genera Insectorum 104: 1-119, 2 pis. 1931. The American Geometridae. In Seitz, A. , The Macrolepidoptera of the WorldS: 1-144, 15 pis.Scott, J. A. 1985. The phylogeny of butterflies (Papilionoidea and Hesperioidea). Journal of Research on the Lepidoptera 23 (1984): 241-281.Sharplin, J. 1963. Wing base structure in Lepidoptera: I Fore wing base. Canadian Entomologist 95: 1024-1050.Shepard, H. H. 1930. The pleural and sternal sclerites of the lepidopterous thorax. Annals of the Entomological Society of America 23: 237-260.Stallings, D. B., Stallings, V. N. T., Turner, J. R. & Turner, B. R. 1985. Courtship and oviposition patterns of two Agathymus (Megathymidae). Journal of the Lepidopterists' Society 39: 171-176.Swihart, S. L. 1967. Hearing in butterflies (Nymphalidae: Heliconius, Ageronia}. Journal of Insect Physiology 13: 469-476. 286 M. J. SCOBLE Walker, F. 1862. List of the Specimens of lepidopterous Insects in the Collection of the British Museum 26: 1479-1796.Warren, W. 1894. New genera and species of Geometridae. Novitates Zoologicae 1: 366-466. 1905. New American Thyrididae, Uraniidae, and Geometridae. Novitates Zoologicae 12: 307-379.Wiley, E. O. 1979. An annotated Linnaean hierarchy, with comments on natural taxa and competing systems. Systematic Zoology 28: 308-337.Wootton, R. J. 1979. Function, homology and terminology in insect wings. Systematic Entomology 4: 81-93. British Museum (Natural History) Milkweed butterflies: their cladistics and biology P. R. Ackery & R. I. Vane- Wright The Danainae, a subfamily of the Nymphalidae, contains only some 150 species, yet aspects oftheir biology have stimulated far more attention than can be justified by species numbersalone. In recent years, an expansive literature has grown, considering aspects of theircourtship and pre-courtship behaviour, migration, larval hostplan associations, mimicry andgenetics. The popularity of danaines among biologists can certainly be attributed to thiscombination, within one small group, of so many of the factors that make butterflies such aninteresting group to study. The obvious need to place this wealth of biological data within anacceptable systematic framework provided the impetus for this volume. Started eight years ago within the conventions of evolution by natural selection andHennig's phylogenetic systematics, the book is now largely about natural history (what theanimals have and do, where they live and how they develop) and natural groups - as revealedby a form of analysis approaching that practised by the new school of 'transformed cladistics' .The authors have prepared a handbook that will appeal to a wide range of biologists, frommuseum taxonomists to field ecologists. 425 pp, 12 pp colour, 73 b/vv plates, line and graphic illustrations, maps, extensive bibliography.ISBN 565 00893 5. 1984. Price 50. Titles to be published in Volume 53 A review of the Miletini (Lepidoptera: Lycaenidae) By J. N. Eliot Australian ichneumonids of the tribes Labenini and Poecilocryptini By I. D. Gauld & G. A. Hollo way The tribe Pseudophloeini (Hemiptera: Coreidae) in the Old World tropics with a discussion onthe distribution of the Pseudophloeinae By W. R. Dolling The songs of the western European grasshoppers of the genus Omocestus in relation to theirtaxonomy (Orthoptera: Acrididae) ByD. R. Ragge The structure and affinities of the Hedyloidea: a new concept of the butterflies ByM. J. Scoble Photoset by Rowland Phototypesetting Ltd, Bury St Edmunds, SuffolkPrinted in Great Britain by Henry Ling Ltd, Dorchester f 5