01 APRIL 1993 No. 2 61 2* 101 , 2 ril 1993 Journal of the New York Entomological Society (ISSN 0028-7 199) Devoted to Entomology in General JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Editor: James M. Carpenter, Department of Entomology, American Mu- seum of Natural History, Central Park West at 79th Street, New York, New York 10024 Book Review Editor: James S. Miller, Department of Entomology, Amer- ican Museum of Natural History, Central Park West at 79th Street, New York, New York 10024 Publications Committee: James K. Liebherr, Cornell University, Ithaca, Chairman; David L. Wagner, University of Connecticut, Storrs; Alfred G. Wheeler, Jr., Pennsylvania State Department of Agriculture, Har- risburg. The New York Entomological Society Incorporating The Brooklyn Entomological Society President: Randall T. Schuh, Department of Entomology, American Mu- seum of Natural History, New York, New York 10024 Vice President: Sule Oygur, Department of Entomology, American Mu- seum of Natural History, New York, New York 10024 Secretary: Rajeev Vaidyanathan, Medical Entomology Laboratory, New York Medical College, Valhalla, New York 10595 Treasurer: Louis Sorkin, Department of Entomology, American Museum of Natural History, New York, New York 10024 Trustees: Class of 1991 — Christine Falco, White Plains, New York; Theo- dore Weisse, Northport, New York. Class of 1992— Durland Fish, Valhalla, New York; Richard Falco, White Plains, New York. Annual dues are $23.00 for established professionals with journal, $10.00 without journal. $15.00 for students with journal, $5.00 without journal. Sustaining memberships are $53.00 per year, institutional memberships are $125.00 per year, and life memberships are $300.00. Subscriptions are $45.00 per year domestic, $50.00 Mexico and Canada. $55.00 all other countries. All payments should be made to the Treasurer. Back issues of the Journal of the New York Entomological Society, the Bulletin of the Brooklyn Entomological Society, Ento- mologica Americana can be purchased from Lubrecht and Cramer. RD 1 , Box 244, Forestburgh, New York 12777. The Torre-Bueno Glossary of Entomology can be purchased directly from the society at $45.00 per copy, postage paid. Meetings of the Society are held on the third Tuesday of each month (except June through September) at 7 p.m. in the American Museum of Natural History, Central Park West at 79th Street, New York, New York. Mailed June 2, 1993 The Journal of the New York Entomological Society (ISSN 0028-7199) is published 4 times per year (January. April. July. October) for the Society by Allen Press. Inc., 1041 New Hampshire. Lawrence. Kansas 66044 Second class postage paid at New York. New York and at additional mailing office. Postmaster: Send address changes to the New York Entomological Society. % American Museum of Natural History . Central Park West at 79th Street. New York. New York 10024-5192. Known office of publication: American Museum of Natural History. New York. New York 10024. Journal of the New York Entomological Society, total copies printed 700, paid circulation 602. mail subscription 602, free distribution by mail 19. total distribution 621. 79 copies left over each quarter. THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER. J. New York Entomol. Soc. 101(2): 143-299, 1993 REVISION AND CLADISTIC ANALYSIS OF THE WORLD GENERA OF THE FAMILY HEMEROBIIDAE (INSECTA: NEUROPTERA) John D. Oswald1 Department of Entomology, Cornell University, Ithaca, New York 14853 Abstract. — The extant world genera of the family Hemerobiidae are revised; and the literature on fossil hemerobiids is reviewed. A key is provided to the 25 extant genera recognized. An intergeneric cladistic analysis utilizing 1 07 characters, 24 ingroup genera, and 3 outgroup species provides a basis for (1) justifying the holophyly of the family, (2) recognizing holophyletic subgroups within the family, (3) proposing a new nine-subfamily classification of the family, and (4) identifying putative synapomorphies for nearly all recognized named taxa. Traits used in the cladistic analysis include characters from the head, forewing, and male and female terminalia. The biogeographic patterns exhibited within the family are discussed in light of the cladistic classification. For each genus, a synonymical listing, a differential diagnosis, a list of proposed synapo- morphies, and notes on distribution and included species are given. To the extent possible from the material available for study, representative forewings, hind wings, and several aspects of the male and female terminalia, are illustrated for each genus. Phylogenetically important character complexes are analyzed in the comparative framework of the familial classification. The hemerobiid genitalic structure previously termed the “para- meres” is here considered a unique evolutionary novelty of the family, and renamed the par- abaculum. This structure is not a homologue of the “parameres” (=9th gonocoxites sensu Adams) of other neuropterous families. A new terminology is proposed for the putatively homologous regions of the gonarcus, facilitating comparative analysis of this complex sclerite. CONTENTS Introduction 144 Materials and Methods 145 Adult Morphology 147 Cladistic Analysis 162 Methods and Data 162 Results 194 Discussion 195 Taxonomic History and Early Classifications 203 Family Hemerobiidae Latreille 205 Key to World Hemerobiid Genera 207 Subfamily Carobiinae Oswald 211 Genus Carobius Banks 211 1 Current address: Department of Entomology, National Museum of Natural History, Wash- ington, DC 20560 144 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Subfamily Hemerobiinae Latreille 213 Genus Hemerobius Linnaeus 213 Genus Nesobiella Kimmins 217 Genus Wesmaelius Kruger 219 Genus Hemerobiella Kimmins 221 Subfamily Sympherobiinae Comstock 222 Genus Nomerobius Navas 222 Genus Neosympherobius Kimmins 224 Genus Sympherobius Banks 225 Subfamily Psychobiellinae Oswald 227 Genus Psychobiella Banks 227 Subfamily Notiobiellinae Nakahara 228 Genus Notiobiella Banks 229 Genus Psectra Hagen 230 Genus Anapsectra Tjeder 232 Genus Zachobiella Banks 232 Subfamily Drepanacrinae Oswald 234 Genus Conchoplerella Handschin 235 Genus Austromegalomus Esben-Petersen 235 Genus Drepanacra Tillyard 237 Subfamily Megalominae Kruger 239 Genus Megalomus Rambur 239 Subfamily Drepanepteryginae Kruger 242 Genus Neuronema McLachlan 242 Genus Gayomyia Banks 244 Genus Drepanepteryx Leach in Brewster 246 Subfamily Microminae Kruger 248 Genus Noius Navas 248 Genus Nusalala Navas 249 Genus Megalomina Banks 251 Genus Micromus Rambur 253 Unplaced Genera 257 Genus Notherobius New 257 Nomina Dubia 258 Genus Subboriomyia Steinmann 258 Neuropterida Incertae Sedis 258 Genus Ormiscocerus Blanchard in Gay 258 Fossil Hemerobiidae 259 Biogeography 265 Acknowledgments 274 Literature Cited 274 Appendices 1-5 288 The family Hemerobiidae, brown lacewings, contains approximately 550 species which are collectively distributed on all continents except Antarctica. Several genera, e.g., Hemerobius , Micromus, and Sympherobius, are, or are nearly, themselves cos- 1993 WORLD GENERA OF HEMEROBIIDAE 145 mopolitan. All known hemerobiid larvae are strictly predaceous. Adults are generally believed to be principally predaceous, but at least one recent work (Stelzel, 1990) documents unexpectedly high levels of honeydew feeding. All species apparently inhabit, both as larvae and adults, the surfaces of emergent terrestrial vegetation, upon which their foodstuffs are found. The fragmented taxonomic literature on the Hemerobiidae consists largely of de- scriptive treatments, and accounts of local and regional faunas. Very few works of both broad taxonomic and broad geographic scope have been published. This situ- ation has hindered the recognition of phylogenetically meaningful (i.e., holophyletic sensu Ashlock, 1971) groups on a global scale, and has discouraged the pursuit of studies such as comparative biology and biogeography, which are dependent upon the recognition of such groups. This paper is a comprehensive revision of the family Hemerobiidae at, and above, the rank of genus. Its single most important objective has been to reorganize the family into a set of nested holophyletic taxa. To accomplish this, an extensive mor- phological survey was undertaken to identify characters of potential phylogenetic significance. The cladogram resulting from the cladistic analysis of the characters identified during this survey represents the first detailed hypothesis of hemerobiid intergeneric relationships. Although the cladistic relationships proposed here are subject to future corroboration, as are all hypotheses, the putative synapomorphies expressed on the cladogram support and identify specific clades which are capable of being confirmed or refuted by future work. In addition, the cladogram provides a much needed initial framework of sister-group relationships within the family. These relationships, in conjunction with the autapomorphies identified for most genera, provide a logical basis from which to begin comprehensive revisionary studies of individual genera. In addition to its phylogenetic component, this work contains: (1) the first com- prehensive key to world hemerobiid genera, (2) a new subfamily classification of the family, (3) a preliminary analysis of hemerobiid biogeography, and (4) a summary of available information on hemerobiid fossils. For each genus, diagnostic and dis- tributional data are presented, and citations to published information on immature stages are given. When this project was begun in 1986, the family Hemerobiidae contained ap- proximately 45 valid genera and subgenera. Several of these have since been either formally removed from the family (Nyrrna [to the Berothidae], by U. Aspock, 1989), or synonymized with other hemerobiid genera (e.g., Anotiobiella [with Hemerobius ], by Gonzalez-Olazo, 1987; Oxybiella and Drepanomina [with Megalomina], and Kimminsiella [with Psectra], by New, 1988; Sympheromima [with Sympherobius ], by Oswald, 1988b; and Idiomicromus [with Micromus ], by Monserrat, 1990). The changes proposed herein further restrict the number of valid genera to 25, and eliminate all previously recognized subgenera. This substantial reduction in the num- ber of genus-group taxa results from the recognition of only demonstrably holophy- letic taxa. MATERIALS AND METHODS Material. This study is based on the examination of specimens of approximately 185 hemerobiid species borrowed from the following collections: American Museum of Natural History, New York, NY (AMNH); Bernice P. Bishop Museum, Honolulu, 146 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) HI (BPBM); British Museum (Natural History), London, England (BMNH); Cali- fornia Academy of Sciences, San Francisco. CA (CAS); Cornell University Insect Collection, Ithaca. NY (CUIC): Florida State Collection of Arthropods. Gainesville, FL (FSCA); James B. Johnson, private collection (JOHNSON); Museum National d'Histoire Naturelle, Paris, France (MNHP); Ellis G. MacLeod, private collection (MacLeod); Museum of Comparative Zoology, Harvard University, Cambridge. MA (MCZ); National Collection of Insects, Pretoria, South Africa (NCIP); National Mu- seum of Natural History. Washington. DC (NMNH); Norman D. Penny, private collection (PENNY). Specimen Preparation and Examination. Head capsules and terminalic structures were cleared in a cold 10 percent solution of KOH, generally overnight, then rinsed in 70 percent EtOH or a dilute solution of acetic acid. Most terminalic preparations were, in addition, stained for ca. 1-2 minutes in a saturated solution of 70 percent EtOH and Chlorazol Black, then rinsed in 70 percent EtOH. Structures were trans- ferred directly from 70 percent EtOH to glycerine for examination under dissecting and/or compound light microscopes. This procedure yielded generally satisfactory results. However, care was taken not to overstain multilayered structures, as overstained preparations may not be suffi- ciently translucent to allow examination with transmitted light. Such examination w as found to be critical for accurately distinguishing the attachments of membranes to male genitalic structures (particularly to the gonarcus and parabaculum) and for obtaining a detailed understanding of the morphological configurations of complex genitalic structures. Illustrations. Drawings were prepared with the aid of camera lucidas attached to Wild M5 dissecting and Olympus CH compound microscopes. Wings were detached and held flat between a pair of glass slides for illustration. To simplify wing figures, marginal trichosores have been omitted throughout. All hemerobiids possess prom- inent trichosores on the wing margins, although in a few taxa they are poorly de- veloped or absent proximally. Wing flexion and fold lines are indicated by dashed lines. Terminalic structures were mounted in glycerine jelly on depression slides to pre- vent rotation during illustration. Lateral views of male and female terminal abdom- inal segments are illustrated with little or no sclerite overlap. In actual preparations, telescopic foreshortening of these segments is commonplace. The trichobothria-bear- ing callus of the ectoproct is outlined only; individual trichobothria are not shown. Surface micro-ornamentation and setae are shown only where particularly diagnostic. The gonarcus and parabaculum of one species of each genus are illustrated in dorsal and lateral views. These figures are oriented with anterior to the right, and dorsal (lateral view figures) or left side (dorsal view figures) up. An attempt has been made in many of these figures to show where the membranes associated with these sclerites are attached. This feature should aid in interpreting these structures, particularly with respect to the new terminologies proposed herein. To minimize hidden lines and maximize views of sclerotized surfaces, the free membrane margins in these figures are shown in artificially cropped and stretched positions. Most terminalic figures have been drawn from material stained with Chlorazol Black. Some details shown in the illustrations may not be apparent in unstained material. This is particularly true of minute structural details of the gonarcus and 1993 WORLD GENERA OF HEMEROB1IDAE 147 parabaculum. It should also be noted that the edges of internal terminalic apodemes (e.g., those of the intragonarcus, the parabacular apophysis, and the anteroventral angles of some male 9th tergites) show evidence of adult growth. The absolute shapes of these structural components may vary as a function of post-ecdysial age. Text references to figures found in this work are cited with a upper case “F” (“Fig.”), those in other works with a lower case “f” (“fig.”). Ordinal Nomenclature. The following system of “neuropteroid” ordinal names is used here (after Kristensen, 1981): Neuropterida (adjective, neuropterid) = Neurop- tera + Megaloptera + Raphidioptera; Neuroptera (adjective, neuropterous) = the lacewings and their allies (Planipennia); Megaloptera = Sialidae + Corydalidae; Raphidioptera = Inocelliidae + Raphidiidae. Terminology. The general entomological terminology employed here follows The Torre-Bueno Glossary of Entomology (Nichols, 1 989). Most terms used for particular character systems are explained below under the heading Adult Morphology. Abbreviations and Annotations. Collection acronyms are recorded above under Material. The following annotations are used in the synonymical listings: A, adult description or characterization; Bdy, body; Bio, biology; Dst, distribution; FT, female terminalia; ITSD, incorrect type species designation; Key, key or keyed; Lst, list or listed; Msc, miscellaneous notes; MT, male terminalia; Nom, nomenclature; OD, original description; RD, redescription; Syn, synonymy, synonymized; Tax, taxon- omy (synonymy, homonymy, type data, etc.); W, wing. An asterisk (*) following an annotation indicates a figure (e.g., FW*, forewing figure). ADULT MORPHOLOGY This section presents pertinent information on general hemerobiid morphology. It is intended primarily to supplement the characters used in the cladistic analysis (see below), not as a comprehensive morphological analysis or survey. The treatment by Killington (1936: 1 3-44) is still the best general review of hemerobiid morphology, although some of its terminology is now out-of-date. Discussions of character po- larities in this section are based on the results of the cladistic analysis presented in the next section. Head Cranium. Well sclerotized; posterior surface with complete or incomplete post- ocular carinae running ventrally from the dorsolateral aspects of the occipital foramen to the posterior articulations of the mandibles; a variety of internal costae may be present (Fig. 1 ), these may or may not be evidenced by external sulci; frons unmodified externally, except in Neosympherobius which bears a pair of lateral cavities connected by a transverse depression, and a pair of low prominences above the anterior tentorial pits (Fig. 78); clypeus planate, bearing several pairs of homologizable primary setae (Fig. 2) in addition to more irregular setae; eyes prominent, pulvinate (Fig. 78) to hemispherical (Fig. 1); ocelli always absent; antennae moniliform, scape large, its shape variable but generally somewhat flattened (with a medially directed process in one Zachobiella species), pedicel somewhat larger than first flagellomere; tentorium strongly sclerotized, anterior arms with prominent mesal lobes (Fig. 4, mtl), devel- 148 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 1-16. Head terminology. 1 , Cranium, anterior (diagrammatic, showing principal cos- tae). 2, Lower cranium, anterior (diagrammatic, showing insertions of principal clypeal setae). 3, Epipharyngeal surface of labrum ( Megalomus tortricoides) [Scale bar E], 4, Tentorium, an- terodorsal ( Notiobiella multifurcata) [A]. 5. Lower cranium, lateral (Drepanacra binocula) [B]. 6. Maxillary palpus ( Hemerobius humulimis) [B]. 7. Labial palpus (Hemerobius humulinus) [B], 8. Penicilhform sensillum of galea (Megalomus tortricoides) [C]. 9. Maxillary palpus (Megalomus tortricoides ) [D], 10, Labial palpus (Megalomus tortricoides) [D]. 1 1, Ventral part of occipital foramen, posterior (Micromus montanus) [B], 12-16. Right mandibles, anterior. 12, Carobius pulchellus [B], 13, H. humulinus [B]. 14, Sympherobius amiculus [B], 15, Psectra diptera [B], 16. Psychobiella sordida [B], Abbreviations: ant, antennal costa; atp, anterior tentorial pit; clg, clypeogenal costa; clp, clypeopleurostomal costa; cor, coronal costa; cvm. cervical membrane; dls. dorsolateral seta; dms, dorsomedial seta; dtc, distal convexity; ftc, frontoclypeal costa; ftg. frontogenal costa; gpl, genopleurostomal costa; gpt. genal punctulae; lbm, labrum; mdf. mid- frontal costa; mtl, mesal tentorial lobe; ocv, orad cavity; pis, pleurostomal costa; pma, palpi- 1993 WORLD GENERA OF HEMEROB1IDAE 149 opment of dorsal arms highly variable, absent to very long, posterior tentorial bridge stout, generally with a short anteromedian process (Fig. 4, stp). Mouthparts. Labrum approximately rectangular, aboral surface prominently setose, oral surface bearing sensory organs and, in Megalomus, two columns of transverse strigae (Fig. 3, str); mandibles moderately asymmetrical, left mandible modified to receive and overlap right mandible when opposed, both mandibles rarely with well- developed molar ridges (some Megalomus)-, mesal surface of left mandible with dorsal and ventral blades enclosing a groove which receives opposed right mandible, dorsal blade always bearing a prominent basal tooth; mesal surface of right mandible with a well-developed dorsal blade bearing one or two prominent cusps or a rounded lobe (Figs. 12-16), ventral blade vestigial (Fig. 12) or absent; maxillary palpus (Figs. 6, 9) 5-segmented, basal palpomere longitudinally divided, 5th palpomere simple (Fig. 9) or bisubsegmented (Fig. 6); galea 2-segmented, distal segment bearing several types of sensilla and an apical sensillar knob, the latter often minute and concealed by adjacent trichoid sensilla; laciniae attenuate and prominently setose; labial palpus (Figs. 7, 10) 3-segmented, 3rd palpomere simple (Fig. 10) or bisubsegmented (Fig. 7), 3rd palpomere always bearing a palpimacula (Fig. 10, pma) on its outer surface, the palpimacula is composed of a small group of recumbent setae overlying a shallow depression; glossae and paraglossae fused into a rounded ligula of variable shape. Thorax Prothorax. Pronotum generally transverse, i.e., wider than long, occasionally more produced anteriorly; usually with a pair of distinguishable parasagittal calli. Pterothorax. Very homogeneous throughout the family, see Killington (1936) for a general morphological account. Legs. Procoxa elongate and subcylindrical, meso- and metacoxae short and conical; pro- and mesotibiae subequal or somewhat longer that pro- and mesofemora, protibia generally with an antennal comb composed of elongate, aligned, setae located distally on its anteromesal face; metatibia much longer that metafemur; all tarsi with five tarsomeres, all but the ultimate generally bearing a row of stout spines ventrodistally; pretarsus always with a pair of simple, lateral, arched claws and a median empodium. Wings General. Trichosores always present, but sometimes vestigial or absent on proximal wing margins. Wing membranes always microtrichose, never macrotrichose (except very rarely along some wing veins). Nygmata never present. Pterostigma never sharp- ly defined, but a ’pterostigmal area’, marked by altered subcostal veinlets, is almost always identifiable in the distal portion of the costal space. Marginal end-twigging of longitudinal veins generally pronounced. macula; ptf, postfrontal costa; ptm, postmentum; pxc, proximal convexity; sbo, subocular costa; ss, sagittal seta; stp, sagittal tentorial process; str, strigae; tmp, temporal costa; trf, transfrontal costa; trt, transtorular costa; vis, ventrolateral seta; vms, ventromedial seta. Scale bars (mm); A = 0.2; B = 0.2; C = 0.02; D = 0.2; E = 0.1. 150 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) SCV CCV Figs. 17-18. Forewing terminology. 17, Venation (composite). 18, Wing spaces, flexion lines, and miscellaneous (composite). Abbreviations: Al, A2, A3, anal veins; ahs, antehumeral space: als, anal space; atr, anterior; cafl, cubitoanal flexion line; cas, cubitoanal space; ccv, costal crossvein; cfl. costal flexion line; els, costal space; CuA (cua), anterior cubitus; CuP (cup), posterior cubitus; cus, cubital space; dst, distal: hp, humeral plate; hvt, humeral veinlet; i, intra; ials. intra-Al space: iafl, intraanal flexion line; ics, intracubital space; ims, intramedial space; jfl. jugal fold line: jls, jugal space; jv. jugal vein; M (m), media; mefl, mediocubital flexion line; mcs. mediocubital space; ms, medial space; orb#, oblique radial branch of anterior radial trace = “radial sector”; pht, proximal humeral trace; prx, proximal; pst, posterior; pts, pterostigmal space: R (r). radius; rls, R1 space; rds, radial space; rmfl, radiomedial flexion line; rms, radiome- dial space: Sc (sc), subcosta; scs, subcostal space; sev, subcostal veinlet. 1993 WORLD GENERA OF HEMEROBIIDAE 151 Venation. Three categories of wing veins are differentiated here: (1) longitudinal veins— longitudinal or oblique, setose, veins, (2) veinlets — longitudinal veins of the costal space, and (3) crossveins— asetose veins joining adjacent longitudinal veins. All veins are composed of one or more simple internodes bounded by nodes where they fork or intersect other veins. Crossveins always consist of a single intemode. The vein expanse between two adjacent forks of a longitudinal vein is called an interramus. An interramus will consist of more than one intemode if intersected by one or more crossveins. Longitudinal Veins. The terminology adopted for the principle longitudinal veins of the forewing is given in Fig. 17. No “MA” arising from the apparent R is assumed present (see discussion under Characters 29-33 below). Traces. Any continuous sequence of vein intemodes may be referred to as a trace. This new venational concept allows great flexibility, precision, and conciseness in describing venational features, particularly in defining axes of vein pectinations and wing space boundaries. The major traces are the anterior and posterior traces of the principle longitudinal veins. These traces may be defined, respectively, as the most anterior and posterior sequences of intemodes of each principle longitudinal vein. Other traces, e.g., the proximal humeral trace (Fig. 17, pht [=“recurrent vein” of authors]), may be defined in a similar manner. Gradate Series r Crossveins are frequently arranged in more or less aligned tracts called gradate series. Up to five gradate series are commonly found in hemerobiid forewings (Fig. 17), these are the costal gradate series, and up to four series posterior to the anterior radial trace. In this work, the number and position of the latter series are based on the inferred plesiomorphic number (4) and positions of the m-cu cross- veins. These series are numerically designated 1 to 4 starting at the base of the wing. Crossveins belonging to each series are prefixed with the same numerical designator. The serial assignment of individual crossveins is sometimes problematic; and extra, short, interpolated, series are occasionally developed in some large-winged taxa (e.g.. Fig. 166). The proper assignment of crossveins lying behind the anterior cubital trace to the 3rd or 4th gradate series is often especially difficult. The numerical designation of gradate series is preferred here over the common prior practice of using relative positional designators (i.e., “inner”, “middle” and “outer”). The numerical system proposed here is intended to imply specific crossvein (for some crossveins) and series homologies. The relative positional designator system has generally been used as a system of convenience, without specific regard for homologies. Specific hypotheses about principle crossvein homologies are required here to facilitate critical assessment of gradate series variations, particularly, to infer gains or losses of entire gradate series and gains or loses of individual principle crossveins. Crossveins. The principle crossveins joining different principle longitudinal veins are designated by a numerical prefix and a pair of vein abbreviations in lowercase letters, e.g., the four principle mediocubital crossveins are designated lm-cu, 2m-cu, 3m-cu, and 4m-cu (Fig. 17). The number indicates the gradate series association of the crossvein, and the letters the longitudinal veins it intersects. Crossveins joining different branches of the same longitudinal vein are designated by a numerical prefix, the letter “i” [’intra-’], and one lowercase vein abbreviation, e.g., the two principle intramedial crossveins are designated 3im and 4im (Fig. 1 7). 152 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Membranes. Two categories of wing membrane divisions are differentiated: (1) cells, unbroken expanses of membrane bounded by veins and/or the wing margin, and (2) spaces, aggregates of cells (and vein segments) bounded by veins and the wing margin. Cells. Wing cell designations are formed from the designation of the crossvein bounding the cell distally and the prefix “c” [’cell’], e.g., the intramedial cell bounded distally by crossvein 3im is designated c3im (Fig. 17). Spaces. Ten principle, and five secondary, wing spaces are differentiated here. The principle spaces are bounded by the anterior and posterior traces of most of the principle longitudinal veins. Secondary spaces are subdivisions of principle spaces. All spaces used herein are illustrated in Fig. 18. Flexion and Fold Lines. Flexion and fold lines are designated after the wing space in which they occur. Up to five flexion lines are present (Fig. 18): (1) costal (cfl), (2), radiomedial (rmfl), (3) mediocubital (mcfl), (4) cubitoanal (cafl), and (5) intraanal (iafl). A single fold line, the jugal (jfl), is always present. Forewing. Always macropterous. Membrane frequently hyaline, but more often darkly mottled with browns or blacks on a more or less hyaline ground, rarely secondarily coriaceous. Veins whitish, yellowish, reddish, blackish, brownish, or green (rare), frequently with alternating light and dark regions. Forewing outlines extremely variable, continuously convex shapes vary from narrowly elliptical (Fig. 115) to nearly circular (Fig. 132), shapes with one or more marginal concavities (especially on the posterior wing margin) have been independently derived in several lineages. Costa prominent. Costal space of variable width, attenuating distally; humeral area extremely broad (Fig. 1 32) to very narrow (Fig. 1 24); humeral veinlet (most proximal subcostal veinlet) simple to many branched; proximal humeral trace (=“recurrent vein” of authors) strongly recurrent to simply transverse; posthumeral subcostal veinlets simple or branched, but only rarely all simple. Subcostal space generally with 2 or 3 crossveins, 1 substigmal and 1 (Fig. 17, lsc-r) or 2 (Fig. 17, lsc-r and 2sc-r) proximal, rarely with proliferated supernumeraries (Fig. 1 74); proximal width wider than adjacent subcosta, except in Notiobiella. Radius highly modified, with one or more branches (the ORB’s [oblique radial branches] or “radial sectors”) of ancestral Rs origin, secondarily originating from the anterior radial trace (see discussion under Characters 29-33 below). CuA and M3 + 4 fused in Nusalala and many Micromus. Jugal lobe well formed, one jugal vein present. Hind Wing. Shorter than forewing, usually macropterous, rarely brachypterous, micropterous, or absent [lost]. Membrane generally pale, sometimes weakly mottled, never coriaceous. Humeral plate generally forming an oblique strut at base of wing (e.g.. Fig. 34), and often with a short distal lobe bearing several stout, frenulum-like, bristles. Humeral veinlet (most proximal subcostal veinlet), and most or all posthu- meral subcostal veinlets, simple. Subcosta stout. Crossveins absent proximally in subcostal space. Radius not modified as in forewing, but occasionally with Rs stem closely associated with or briefly fused to Rl. R1 space frequently with a prominent proximal irl crossvein which mimics a second “radial sector” (Fig. 99). Remnant of inferred MA always present as a well-developed, generally sigmoid, vein which terminates on the posterior radial trace. CuA stout. CuP well developed to absent. 1993 WORLD GENERA OF HEMEROBIIDAE 153 possibly size correlated, with greater development in larger wings. A1 usually deeply forked. Jugal lobe vestigial, jugal vein present. Additional Terminological Considerations. The phrase “radial sectors” is used here in its traditional sense, i.e., to collectively refer to the two or more prestigmal oblique branches of the forewing anterior radial trace. Although technically incorrect, this phrase is used in several places in this work (e.g., in the generic key) because it is concise and widely understood in the intended sense. The “radial sectors” are called “oblique radial branches” (ORB’s), when a designation relatively free of implications of homology is necessary or convenient. ORB's are counted consecutively from the base of the wing. Abdomen General. 10-segmented; segments 1-8 with paired lateral spiraculae; 10th tergite always bipartite, each half ( ectoproct ) bearing an oval aggregation of trichobothria set in rosettiform alveoli and borne on a low callosity; segments 8-10 always sexually dimorphic, more proximal segments only occasionally sexually dimorphic. Basal Segments (1-6). Tergites 1-6 and sternites 2-6 free and hemiannular; each stemite generally with a transverse costa internally, costae gradually decreasing in prominence on more posterior segments, stemite 1 reduced to a transverse strip; male tergites 3-4 (sometimes other tergites also) frequently notched posteromedially; male tergites 4-6 rarely (some Psectra ) with marginal posteromedian processes. Female Terminalia (Abdominal Segments 7-10) 7th Segment. Tergite free and hemiannular; sternite free and normally hemian- nular, rarely divided ( Anapsectra ). 8th Segment. Lateral tergite margins produced ventrally below, and enclosing, spiraculae of 8th segment (Fig. 26), contralateral margins often lying closely adjacent ventrally or, rarely, fused; 8th sternite ( subgenitale ) variously developed, large and prominent to absent, when prominent, its apex usually enclosed at rest by a pair of small concave or rod-like sclerites, the gonapophyses posteriores , which are associated with the anteroventral angles of the 9th gonocoxites. 9th Segment. Lateral portions of tergite greatly expanded ventrally, and extended posteriorly to subtend ectoprocts, contralateral margins generally closely adjacent on ventral midline; tergite sometimes (e.g., most Microminae) sagittally divided; tergite often narrowed, sometimes divided (e.g., some Psectra , Fig. 1 10), laterally; stemite absent; paired gonocoxites always prominent and usually free, elongated (e.g., some Wesmaelius, Fig. 52), rarely partially fused to each other (e.g., Anapsectra , some Hemerobius) or to adjacent ipsilateral margins of 9th tergite (e.g., Zachobielta , Fig. 126), stylus present (Fig. 26) or absent [lost] (Fig. 35). 10th Segment. Ectoprocts rarely strongly lobed or ornamented (in contrast to males); stemite absent. Internal Reproductive Tract. Vulva opening between paired 9th gonocoxites, lead- ing into a broad, membranous (rarely partially sclerotized), genital chamber (bursa), which is presumed to receive male’s spermatophore, and into which open the col- leterial gland, bursal glands, and the proximal end of the insemination-fertilization 154 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Table 1. General terminology for gonarcus and associated membranes. Gonarcus1 Paleogonarcus Neogonarcus Intragonarcus Extragonarcus Neogonarcus Gonopons Intragonopons Extragonopons Neogonopons Hemigonarcus Intrahemigonarcus Extrahemigonarcus Neohemigonarcus Associated Paragonosaccal Paragonosaccal Paragonosaccal membranes & Gonosaccal [& Gonosaccal] ' See also Figures 19, 20. Unfortunately, proper orthography dictates that the singular and pleural forms of words based on the Latin work “arcus,” arch, bear the same ending, -us. canal. The latter canal opens at a small rounded pore (not a slit) at the anterior end of the bursa, its distal end presumably emptying into the common oviduct. The shape of the canal is highly variable, generally a narrow, more or less irregularly convoluted tubule, rarely helical (e.g., some Noius), often with an expanded area somewhere along its length. The term “spermatheca”, which has been used in prior literature to indicate either any well-sclerotized portion of the insemination-fertilization canal, or a prominently expanded region along its length, requires clarification and stan- dardization within neuropterous taxa. Consequently, this term is not used here. Male Terminalia (Abdominal Segments 7-10) Tergites and Stemites 7th Segment. Tergite and stemite nearly always free and hemiannular, tergite occasionally with processes (some Megalomus, Psectra, Zachobiella). 8th Segment. Tergite and stemite normally free and hemiannular; tergite rarely with processes (some Psectra, Zachobiella), or linked to 8th stemite (some Zacho- biella, Fig. 127) or 7th tergite; stemite rarely (some Zachobiella , Fig. 127) with a posteromedian projection (=?fused 9th sternite); spiraculae rarely opening through prolonged lateral margins of tergite (some Notiobiella and Zachobiella) or stemite (some Zachobiella, Fig. 127). 9th Segment. Tergite generally free, and approximately hemiannular overall, but with numerous modifications in different taxa, including: (1) anteroventral angles produced internally as apodemes (e.g., Nesobiella, Fig. 47; Anapsectra, Fig. 1 17), (2) posteroventral angles produced as distinct processes (e.g., Sympherobiinae, Fig. 69; Psectra, Fig. 108; Zachobiella, Fig. 1 27), (3) posterior margin produced as an elongate process (some Psectra), (4) a sagittal dorsal division (most Microminae), (5) a trans- verse dorsal division, separating anterior antecostal portion of tergite from posterior portion (some Psectra), or (6) partial fusion with ectoproct (most Microminae); ster- nite extremely variable in shape, from small, flat or hemiannular plates to prominently produced tubular structures (e.g.. Fig. 47), occasionally (1) with distinctive orna- mentation (Fig. 194), or (2) apparently lost or fused with 8th stemite (some Zacho- biella, Fig. 127); 9th gonocoxites prominent in Carobius, absent in all other taxa. 10th Segment. Ectoprocts articulating with (Fig. 27), or fused to (Fig. 222), posterior margins of 9th tergite, and very diversely ornamented within in the family; oma- 1993 WORLD GENERA OF HEMEROBIIDAE 155 Figs. 19-20. Male gonarcus terminology. 1 9, Gonarcus, dorsal (diagrammatic). 20, Gonar- cus. lateral (diagrammatic). Abbreviations: a, b, c, d, reference points; egps, extragonopons; egs. extragonarcus; ehgs. extrahemigonarcus; gsm, gonosaccal membrane; hgc, hemigonarcal con- junction; hgs. hemigonarcus; igps, intragonopons; igs, intragonarcus; ihgs, intrahemigonarcus; med. mediuncus; ngs. neogonarcus; pgs, paleogonarcus; pgsm. paragonosaccal membrane; x-x. reference line. mentation types include ( 1 ) broad, hollow, cuticular lobes, (2) solid projecting spines, and (3) numerous modifications of setae and their bases. Gonarcus The hemerobiid gonarcus is a highly variable and frequently complex sclerite. It’s morphology has been extensively studied for this work. To provide a foundation for accurately assessing the homologies of its parts, and to clear up earlier misinterpre- tations, its structural morphology is treated in detail below. Some of the terminology employed was proposed earlier by Oswald ( 1 988b), but much is new. The terms used below for the principal divisions of the gonarcus are summarized in Table 1; and most are identified in Figs. 19 — 20. Commissures , Intragonarcus and Extragonarcus. The hemerobiid gonarcus is a complex sclerite composed of surface (external) and apodemal (internal) components, which are freely suspended in the membranous posterior body wall of the abdomen. Gonarcal components are membranously attached to the body wall along lines, called commissures, which constitute the principal landmarks for delimiting putatively homologous gonarcus regions. There are three primary commissures: (1) the antex- tragonarcal, (2) the postextragonarcal, and (3) the neogonarcal. The antextragonarcal commissure ( Figs. 19, hgc-b-a-b-hgc; 20, hgc-b-a) marks the line of attachment of the paragonosaccal membrane (Fig. 20, pgsm) to the paleo- gonarcus (Fig. 1 9. pgs). This commissure divides the posterior (external) components of the paleogonarcus (collectively the extragonarcus [Fig. 19, egs]), from its anterior 156 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) (apodemal) components (collectively the intragonarcus [Fig. 19, igs]). The antextra- gonarcal commissure extends dorsad from its pair of termini at the hemigonarcal conjunctions (Figs. 1 9-20, hgc), which are the two ventrolateral points on the gonarcus where the free margin of the intragonarcus meets the ventral margin of the extra- gonarcus. The hemigonarcal conjunctions are important gonarcal landmarks and are generally easily located in stained preparations. The path of the antextragonarcal commissure may be obscured by secondary sclerotization of the paragonosaccal membrane (see Neogonarcus and Neogonarcal Commissure below). The postextragonarcal commissure (Figs. 19, hgc-hgc; 20, hgc-c) marks the line of attachment of the gonosaccal membrane (Fig. 20, gsm) to the ex tragonarcus (Fig. 19, egs). Like the antextragonarcal commissure, the postextragonarcal commissure ter- minates ventrally at the hemigonarcal conjunctions. The path of the postextragonarcal commissure is generally distinct, but may be discontinuous (when the mediuncus is membranously separated from the gonarcus arch), highly convoluted, or obscure (when the posterior margin of the extragonarcus is poorly defined). Gonopons and Hemigonarcus. The gonarcus arch has historically been partitioned into three parts, a pair of lateral “wings”, the hemigonarcus (Fig. 20, hgs), and a dorsomedial connecting “bridge”, the gonopons (Fig. 19, igps + egps). These regions are approximately delimited in Figs. 19-20 by lines marked “x-x”. Although the positions of these boundaries are sometimes imprecise, they are generally clear enough to be of use descriptively, and this tripartite division is retained here. By combining the two preceding gonarcal division schemes (intragonarcus/extra- gonarcus, and gonopons/hemigonarcus), the paleogonarcus can be partitioned into six basic regions (see Table 1): (1) an anterodorsal, apodemal intragonopons (Fig. 19, igps), (2) a posterodorsal, external extragonopons (Fig. 19, egps), from which the mediuncus (Fig. 19, med) extends posteriorly, (3) a pair of anterolateral, apodemal intrahemigonarcus (Fig. 20, ihgs), and (4) a pair of posterolateral, external extrahemi- gonarcus (Fig. 20, ehgs). Neogonarcus and Neogonarcal Commissure. In addition to sclerites of paleogonar- cal origin (i.e., the intragonarcus + extragonarcus), many hemerobiid gonarcus con- tain sclerotized structural elements derived from the secondary sclerotization of the plesiomorphically membranous paragonosaccal membrane. The term neogonarcus (Fig. 20, ngs) is used here to collectively refer to these elements. It is sometimes convenient to partition the neogonarcus into a dorsomedian neogonopons and a pair of lateral neohemigonarcus. named for the analogous portions of the paleogonarcus. The neogonarcal commissure (Figs. 19, b-b; 20, b-d) marks the line of attachment of the membranous (distal) portion paragonosaccal membrane (Fig. 20, pgsm) to the margin of its sclerotized (proximal) part, the neogonarcus (Fig. 20, ngs). In almost all cases, the ends of the neogonarcal commissure are confluent with some part of the antextragonarcal commissure. The neogonarcal commissure is discontinuous when lateral neohemigonarcal plates are developed without a dorsally connecting neogonopons (e.g.. Fig. 72). Poor definition of the posterior margin of the neogonarcus sometimes prevents precise definition of the neogonarcal commissure. Intragonarcus. The intragonarcus (Fig. 19, igs) is the internal, apodemal, portion of the paleogonarcus. It is bounded posteriorly by the antextragonarcal commissure. It is generally divisible into a narrowly transverse, dorsomedial bar, the intragonopons (Fig. 19, igps), and a pair of broad lateral lobes, the intrahemigonarcus (Fig. 19, ihgs); 1993 WORLD GENERA OF HEMEROBIIDAE 157 although, the boundary between these regions is not always clear, particularly when the intrahemigonarcus are poorly developed. Surface irregularities of the intragonar- cus include localized depressions and occasional carinae. Although well sclerotized, the intragonarcus is usually poorly pigmented; consequently, the details of its structure may not be fully apparent in unstained preparations. The intragonarcus generally exhibits evidence of growth along its free margin in the form of sequentially arranged sclerotized arcs. Since the margin of the intrahemigonarcus appears to grow at a greater rate than that of the intragonopons, the shape of the intragonarcus may vary as a function of adult age. Extragonarcus. The extragonarcus (Fig. 1 9, egs) is the externally exposed portion of the paleogonarcus. It is bounded anteriorly by the antextragonarcal commissure and posteriorly by the postextragonarcal commissure. The conformation of the ex- tragonarcus is extremely variable within the Hemerobiidae. It is generally composed of a sclerotized, anterior (proximal), arch which bears one or more posteriorly directed processes. The proximal arch is generally divisible into a dorsomedial area, the extragonopons (Fig. 19, egps), and a pair of lateral areas, the extrahemigonarcus (Fig. 19, ehgs). The frequently prominent dorsomedian process of the extragonarcus is called the mediuncus (Fig. 19, 20, med). The approximate division of the mediuncus from the extragonopons is shown in Figs. 19 and 20 by a heavy vertical line. Neogonarcus. The term neogonarcus (e.g.. Fig. 20, ngs [a-b-d-a]) is used here to collectively refer to any secondary sclerotization(s) of the paragonosaccal membrane which is(are) attached to the gonarcus along the antextragonarcal commissure (Fig. 20, a-b-hgc). Non-homologous neogonarcal elements of varying configuration and complexity have evolved independently in different hemerobiid lineages. In some cases these elements considerably alter the overall appearance of the gonarcus. In some genera, the neogonarcus is present as a prominent arched plate, which extends roof-like over the base of the clearly identifiable mediuncus (as is shown in Fig. 20, also in some Megalomina, Fig. 211, ngs; Neuronema, Fig. 170, ngs; Psy- chobiella, Fig. 97, ngs). Occasionally, prominent neogonarcus bear attenuate, pos- teriorly directed, processes (e.g., Micromus , Figs. 219, ngs, ngsp; 220). The extent of lateral development of the neogonarcus is highly variable, but frequently the neo- hemigonarcus enclose much of the extrahemigonarcus (Figs. 96-97). In some cases the structural role of the extrahemigonarcus are taken over by the neohemigonarcus, and the extrahemigonarcus are reduced. In a few extreme cases (e.g., Megalomus flint i and nigratus, and some Neuronema) the ventral portions of the extrahemigonar- cus are entirely lost and the gonosaccal membrane is attached directly to the internal surface of the neohemigonarcus (not the posterior margin of the extragonarcus). In these cases, it is imperative to make reference to the gonarcal commissures to establish the correct homology of gonarcal elements. In Sympherobius (Fig. 84), the extragonopons and mediuncus are absent [lost]; and the gonosaccal membrane terminates transversely on the ventral surface of the gonopons. The structure which occupies the position of the mediuncus in this genus is actually a narrow neogonarcal element, the pseudomediuncus (pmd). This sclerite articulates with the posterior margin of the gonopons and, at rest, generally lies partially suspended below the gonopons. Less-developed pseudomediunci are present in the genera Nomerobius (Fig. 72) and Neosympherobius (Fig. 76). 158 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) die Ic ape dor ape Figs. 21-22. Male parabaculum terminology. 21. Parabaculum. dorsal (diagrammatic). 22. Parabaculum. lateral (diagrammatic). Abbreviations: al. apophyseal lamella: ape. apophyseal cavity; as. apophyseal shaft: air. anterior; die. dorsolamellar cusp: dor. dorsal: lc. lamellar conjunction: pa. parabacular apophysis: pbm. parabacular membrane: pst. posterior; sap. su- praapophyseal plate; tc. terminal cleft: tl. terminal lobe: ven. ventral. Parabaculum The hemerobiid genitalic structure previously called the "parameres” is here re- named the parabaculum (name from the Greek para, near, and the Latin baculum, rod. in reference to the inferred plesiomorphic shape of its apodeme; plural, para- baculi). A new name is required for this structure because, as argued below under Homolog}’, it is not a homologue of the 9th gonocoxites (sensu Adams. 1969), which have been frequently called “parameres” in other Neuroptera. The fine structure of the parabaculum is highly variable and frequently of taxonomic value. Below. I discuss the general morphology and structural variability of the parabaculum. General Morphology (Figs. 21-22). The parabaculum is composed of two funda- mental sclerotized regions: an anterior parabacular apophysis (pa), and a pair of posterior terminal lobes (tl). The terminal lobes are separated medially by an ante- riorly convergent space, the terminal cleft (tc). The parabaculum is freely suspended in the membranous body wall at the posterior end of the abdomen. The terminal lobes lie parasagittally. below the gonarcus-mediuncus complex, and above the 9th stemite and hypandrium internum. The membrane adjacent and attached to the parabaculum is called the parabacular membrane (pbm). The parabacular apophysis is generally elongate, and usually extends medially for a considerable distance between the hemigonarcus. Although freshly preserved material was not examined, it seems likely that the muscles which control the motion of the parabacular apophysis insert principally on the inner surfaces of the hemigonarcus. Orientational Terminology'. The parabaculum is composed of external surface sclerites and sclerotized apodemal components. Its suspension in the membranous abdominal body wall allows it great range of motion and orientation. These factors make it difficult to conveniently apply self-evident descriptors of relative position (e.g.. proximal, posterior) to the parabaculum. because the terms would lack a clearly evident orientational basis. The orientational basis arbitrarily imposed upon the parabaculum for the descriptive purposes of this work is shown associated with Fig. 21. Parabacular Apophysis. The plesiomorphic condition of the parabacular apophysis is presumed to be that of a simple elongate apodeme composed of two subregions: a posterior, baculiform, apophyseal shaft (as), and an anterior, sagittally oriented 1993 WORLD GENERA OF HEMEROBIIDAE 159 plate, the apophyseal lamella (al). The apophyseal lamella is an apodemal process which extends freely into the body cavity. The generally narrower apophyseal shaft connects the apophyseal lamella to the terminal lobes. The point at which the free dorsal margin of the apophyseal lamella meets the parabacular membrane is called the lamellar conjunction (lc). The parabacular mem- brane is plesiomorphically attached to the parabacular apophysis along two contig- uous lines which begin at the lamellar conjunction and extend parasagittally along the dorsal margin of the apophyseal shaft. The lamellar conjunction is frequently located at the apex of an acute process of the posterodorsal margin of the apophyseal lamella, the dorsolamellar cusp (die). The dorsolamellar cusp frequently extends posteriorly over a portion of the dorsal surface of the apophyseal shaft. When this occurs, a compressed space, the apophyseal cavity (ape), becomes dorsally enclosed by the dorsolamellar cusp, ventrally enclosed by the apophyseal shaft, and laterally enclosed by the parabacular membrane. Apophyseal cavity configurations exhibit varying degrees of complexity and de- velopment within the family. A compressed, shallowly concave space with mem- branous lateral walls is the putative plesiomorphic condition. In a slightly more derived condition, the cavity remains shallow and compressed, but the lateral walls become partly or entirely sclerotized. In a still more derived condition, the lateral walls are well sclerotized with the dorsolamellar cusp reaching, or nearly reaching, the terminal lobes. This results in a condition in which an elongate and compressed tube is enclosed within a baculiform apophysis (Fig. 70, 71). The parabacular apophysis often bears prominent sclerotized structures of sec- ondary origin. Collectively, these structures are referred to as supraapophyseal plates (Fig. 21, sap), but they are almost certainly not homologous among the genera which exhibit them. These plates represent secondarily sclerotized portions of the para- bacular membrane associated with the dorsal surface of the parabacular apophysis. They are generally more or less bilaterally symmetrical and may be articulated or rigidly fixed to the parabacular apophysis. Supraapophyseal plates may be erect and parasagittally adpressed (e.g., Drepan- acra. Figs. 155-156; some Micromus), or laterally splayed (e.g., Megalomus, Figs. 164-165; Nusalala, Figs. 205-206, lbc; some Wesmaelius and Neuronema). Close inspection of parasagittally adpressed supraapophyseal plates usually reveals some of evidence of an apophyseal cavity, or its remnant, between them. Splayed supraa- pophyseal plates may or may not be associated with a compressed apophyseal cavity, and may additionally bear prominent processes or spines (e.g., some Nusalala, Neu- ronema). Several genera (e.g., Nusalala, Gayomyia, most Neuronema) possess supraapo- physeal sclerites which articulate with the parabacular apophysis. In Nusalala and Gayomyia, these appear to represent novel secondary sclerotizations of the para- bacular membrane which have developed in close functional association with the parabacular apophysis, and have maintained a plesiomorphic articulation. The struc- tures in these genera have been given separate names, laterobacula in Nusalala (Fig. 205, lbc) and penniform sclerites in Gayomyia (Fig. 179, pen), to indicate their presumed independent origin. Modifications of the Apophyseal Shaft. The plesiomorphic, slender, apophyseal shaft may exhibit one or more of several types of modifications. The following are 160 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) exemplary, not exhaustive. The apophyseal shaft may be: (1) deeply (e.g., Drepa- nepteryx, Fig. 189; Nesobiella , Fig. 48) or entirely ( Hemerobius , Fig. 37) divided sagittally, (2) superficially unrecognizable due to secondary sclerotization of the ad- jacent parabacular membrane, which may obscure the dorsal surface of the apoph- yseal shaft (e.g., some Micromus and Megalomus), (3) modified as a broad horizontal plate with the parabacular membrane attached along its lateral margins (e.g., some Micromus ; Zachobiella , Fig. 128), or (4) prominently grooved medioventrally (e.g., some Neuronema and Micromus). Modifications of the Apophyseal Lamella. The apophyseal lamella is, with very few exceptions, always a distinct structure. In the few taxa where the lamella is vestigial or lost (e.g. some Megalomus ), its muscle insertions have apparently shifted to ad- jacent apophyseal plates, which are always well developed when the lamella is absent. In Hemerobius , where the parabaculum is composed of a pair of independent, bi- laterally symmetrical, halves, each half bears a small anterior lamella (Figs. 37-38). In other species, the ventral groove of the apophyseal shaft may extend anteriorly onto the apophyseal lamella resulting in a tricuspidate cross-section. Terminal Lobes. The configurations of the terminal lobes of the parabaculum are extremely diverse within the family, and are of considerable taxonomic value. In their presumed plesiomorphic condition they are a small bilaterally symmetrical pair of lobes with limited dorsal sclerotization. More derived conditions include such diverse modifications as spinose or multiply-lobed structures or, in the one Zacho- biella male examined, with the lobes apparently fused into a single median process (Fig. 128). The terminal lobes are external sclerites and often bear surface micro- ornamentation. Parabaculum Homology /. Adams (1969:8-10) discussed the homology and origin of the principal components of the male terminalia in the Neuropterida. He ho- mologized neuropterid male “parameres” with the 9th gonocoxites of other insects, and traced the evolution of neuropterid 9th gonocoxites from their putative ancestral association with the 9th tergite (retained in Raphidioptera, most Megaloptera, and a few Neuroptera) to their derived association with the gonarcus (found in most Neuroptera and a few Megaloptera). According to Adams’ interpretation, which is adopted here, the structures previously called “parameres” in most neuropterous families are 9th gonocoxite homologues. Based on Adams’ discussion (pp. 9-10), three general recognition criteria can be stated for neuropterous 9th gonocoxites: (1) proximal association with the posterior margin of the extragonarcus, (2) prominent setation, and (3) clear evidence of paired ancestral structure. If the hemerobiid parabaculum is also a 9th gonocoxite homolog (sensu Adams), one could reasonably expect to recognize it on the basis of one or more of these criteria; the hemerobiid parabaculum, however, fails all three. Criterion 1 : Proximal Association with Extragonarcus. The hemerobiid paraba- culum is never fused to or articulated with the extragonarcus. In only a few clearly derived cases are parts of the parabaculum even closely associated with the posterior margin of the extragonarcus. For example: (1) in a few Micromus the posterior margins of the hemigonarcus are strongly bent medially, and encroach upon the parabaculum, and (2) in some Megalomus (e.g., Figs. 164-165) the supraapophyseal plates are well developed and laterally splayed, and their margins may approach the extragonarcus laterally. 1993 WORLD GENERA OF HEMEROBIIDAE 161 Criterion 2: Setation. Hemerobiid parabaculi frequently bear minute microtrichia or microspines, but they are apparently never setose. Criterion 3: Paired Ancestral Structure. All previous authors appear to have re- garded the hemerobiid parabaculum as a structure derived from a pair of independent ancestral sclerites. This assumption is implicit in frequent statements in the heme- robiid literature regarding “fused parameres”. This idea appears to have drawn support from two primary observations: (1) the nearly universal presence of paired terminal lobes on hemerobiid parabaculi, and (2) the presence of a completely bi- partite parabaculum in the genus Hemerobius. This apparent support is examined below. Although paired terminal lobes of the parabaculum are apparently present in all hemerobiid genera except Zachobiella, this condition does not necessarily imply derivation of this structure from a pair of incompletely fused ancestral sclerites (where the terminal lobes would represent a remnant of paired structure). If the fusion hypothesis was true, evidence of a suture marking the line of sagittal fusion between the ancestral sclerites could reasonably be expected. I have been unable to find any such suture in the parabaculum. The apophyseal lamella is always a simple lamellate apodeme; it never exhibits a sagittal suture. This structure is a plate-like ingrowth of the integument of the outer body surface (as indicated by the concentric growth rings generally visible on its lateral surfaces), not a remnant of fused sclerites. Like- wise, no suture is evident sagittally along the apophyseal shaft. In some taxa, the supraapophyseal plates are adpressed sagittally and may appear to form a suture. However, this is a secondary condition involving secondarily sclerotized portions of the parabacular membrane, not the true apophyseal shaft. The absence of a distin- guishable sagittal suture in the parabaculum seems strong evidence against any hy- pothesis suggesting its origin from a pair of ancestral sclerites. The fact that Hemerobius— the oldest and one of the best known, most speciose, and most widespread of hemerobiid genera— possesses a completely bipartite para- baculum (Fig. 37) has clearly biased the interpretation of the homology of the par- abaculum. As the type genus of the Hemerobiidae, Hemerobius was the natural focal point of comparative studies involving hemerobiid taxa. This comparative bias led to two intertwined misconceptions. First, that the paired “parameres” (parabaculum) in Hemerobius were homologous with the paired “parameres” (9th gonocoxites) of other neuropterous families; and second, that the unpaired “parameres” (paraba- culum) in non -Hemerobius hemerobiids were “fused” (note the twice implied po- larity, “parameres” [pleural, implying ancestral pairing] and “fused” [implying an- cestral separation]). The first conclusion misinterpreted the homology of the parabaculum within the order Neuroptera, whereas the second misinterpreted the polarity of the parabaculum states “fused” and “not fused” (or, more clearly, “not divided” and “divided”). In the genus Carobius, both the 9th gonocoxites and parabaculum are present (Figs. 29-32). The gonocoxites fit criteria 1 and 3, but are not setose (criterion 2). This clearly demonstrates that these structures are not homologs. Furthermore, the cla- distic analysis (see below) strongly supports the polarity “not divided” [plesiomor- phic] and “divided” [apomorphic] for the parabacular states. To summarize, the principal evidence supporting recognition of the parabaculum as a hemerobiid novelty not homologous with the 9th gonocoxites of other Neuroptera 162 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) are: (1) the absence of a sagittal parabacular suture. (2) the rarity of close associations between the parabaculum and the posterior margin of the extragonarcus. and (3) the joint presence of paired 9th gonocoxites and a parabaculum in the genus Carobius. Hypandrium Internum Always present, membranously associated with the ventral surface of the para- baculum near the anterior end of the terminal lobes: in dorsal view an approximately triangular sclerite with a depressed median keel and more or less revolute lateral margins (e.g.. Tjeder 1961. figs. 480-481. 578 — 579. 685-686). The configuration of this sclerite varies only slightly within the family; and it is of little taxonomic value. CLADISTIC ANALYSIS Methods and Data Overview Hemerobhd intergeneric relationships were investigated cladistically. Twenty-four ingroup genera and three outgroup species were evaluated for 107 characters. The resulting data matrix was analyzed using the cladistic computer program HENNIG86 (Farris. 1988). Cladistic Methodology Cladistic methods seek to identify (i.e.. estimate) the cladogenetic component of phylogeny by analysis of the distributions of putatively synapomorphic (shared de- rived) characters exhibited by collections of studied taxa (Hennig. 1966: Sneath and Sokal. 1973; Wiley, 1981). Numerical formulations of cladistic methods— generally intended for implementation on computers— are designed to construct trees con- taining intemodes justified by putative synapomorphies from Taxon x Character matrices of numerically coded character state data. On the basis of a parsimony criterion, the tree which minimizes the number of character state changes required to explain the observ ed state distributions is regarded as the best estimate of the relative genealogical relationships among the studied taxa. In cases where multiple equally parsimonious cladograms are found to exist, selection among these, if desir- able. must employ one or more secondary criteria, for example, biogeography or posthoc character weighting. Cladogram rooting and character state polarization are generally accomplished by some form of outgroup comparison (Watrous and Wheeler. 1981: Maddison et al., 1984). The outgroup may consist of one or more real taxa. or of a hypothetical taxon whose character state array is composed of individually justified plesiomorphic states. Data For the present analysis, approximately 185 hemerobiid species were examined (ca. one third of the currently valid world species). An exact species tally is not possible due to the presence of unidentifiable, questionably identified, and unde- scribed species among the material examined. Because of the impracticability of scoring all species for all 107 characters (see Characters below), one or more repre- 1993 WORLD GENERA OF HEMEROBIIDAE 163 sentative species of each genus were selected for comprehensive examination. These species are indicate by “+” symbols in Appendix 2. To the extent allowed by the available material, three criteria were adopted to limit geographic and/or taxonomic bias in the selection of representative species: (1) samples were selected to include species from geographically distant and/or disjunct portions of generic ranges, (2) samples were selected to include species from presently or formerly differentiated taxonomic subgroups (i.e., species from different subgenera, or species from subjec- tively synonymous genera), (3) larger samples were generally taken from larger genera. Species not selected for comprehensive scoring were subsequently examined for the presence of the synapomorphic characters identified by the cladistic analysis of the representative species; these additional species provided a means of more broadly assessing the distribution of putative synapomorphies. After scoring each representative species, the character states of congeners were evaluated, and a single state adopted for each character/genus. These character states were used in the cladistic analysis, and are given in matrix form in Appendix 5. Selection of a single state was trivial in cases where all congeners possess the same state. However, where intrageneric variation was present among congeners the choice of a representative state was problematic. In each such case the state interpreted as plesiomorphic within the genus was adopted as representative; however, no general methodology— short of independently analyzing each polymorphic genus— can en- sure identification of the correct plesiomorphic state. The following approach was used here to address this problem: ( 1 ) the state initially considered plesiomorphic was that which was most widespread of among other hemerobiid genera; then, (2) a preliminary cladistic analysis was run using this state; then, (3) the initially selected state of the polymorphic genus was compared to the state exhibited by its sister group; finally, (4) if the outgroup state matched one of the states exhibited by the polymorphic genus, the latter was (if necessary) reset to match the outgroup state, and the analysis was rerun. Under this methodology, the initial selection of character states may have some biasing effect on the topology of the tree finally adopted; however, since the number of intragenerically polymorphic characters in this analysis is low (64/2889 = 2.2%), it is believed that any such bias is minimal. Since the present analysis addresses the resolution of intergeneric relationships, detailed analyses of intrageneric variation in the characters examined are not pre- sented. However, intragenerically variable characters are identified in the character treatments, and are indicated by bold values in Appendix 5. Future intrageneric analyses incorporating these characters will provide further information on their polarities, and allow for reassessment of their effects on inferred intergeneric rela- tionships. Computational Methods The coded character state data given in Appendix 5 was analyzed with the micro- computer program HENNIG86 (Version 1.5; Farris, 1988). Minimum length clado- grams were computed by sequentially applying options “mhennig*” (a tree calculation command which constructs several trees, each by a single pass through the data matrix, followed by branch-swapping on each tree) and “bb*” (which applies extended branch-swapping to all trees in the current tree file). As a means of identifying 164 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) common elements among equally parsimonious cladograms. Nelson consensus trees were calculated using the “Nelsen” option. Output from the “xsteps” option was used in character and cladogram analysis. Cladograms were rooted using the out- groups defined below. Outgroup Taxa Three outgroup species, Nothochrysa calif ornica (Chrysopidae: Nothochrysinae), Psvchopsis birmana (Psychopsidae), and Polystoechotes punctatus (Polystoechotidae), were included as terminal taxa in the overall cladistic analysis as a check on the holophyly of the Hemerobiidae. and as a means of polarizing character states within the family. These outgroup species were selected as available representatives of the three families most often included with the Hemerobiidae in the neuropterous su- perfamily Hemerobioidea (Withycombe, [1925]; Riek. 1970; Henry, 1982; Schliiter, 1986). Characters The 107 characters used in the cladistic analysis include 11 multistate (3 or 4 states) characters (10, 12, 23, 24, 26, 53, 56, 78. 86, 96, and 106); all others are binary. Characters 10, 12 and 106 were analyzed as nonadditive (unordered); all others were treated as additive (ordered). Binary characters have the plesiomorphic state coded “0”, and the derived state coded “1”. Transformation series for additive multistate characters are presented under the appropriate character treatments. Any descriptive or analytical information concerning characters, their states, or polarities, determined prior to the analysis are given under the subheading Comments. Post- analysis notes on the polarities of character states, and the fits and lengths of characters to the preferred cladogram (see Cladistic Analysis: Results: Cladograms, below, and Figi 23) are given under the subheading Results. Character lengths are reported in the format “L = #”, where # is the numerical length of each character on the preferred cladogram. These values are taken from the HENNIG86 output of the preferred cladogram and represent the number of times each character exhibits a character state change on the cladogram. Nonterminal lineages are identified by the reference numbers given on the preferred cladogram. HEAD (Characters 1-22) Cranial Costae (Characters l^J) The relative positions of most cranial costae are shown in Fig. 1. Internal costae are not always marked by external sulci, and vice versa. Character 1 : Temporal costae. (0) absent or poorly developed; (1) well developed (Fig. 1, tmp). Comments: The genera Hemerobius (1). Psectra (1), and Zachobiella (1) are poly- morphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. 1993 WORLD GENERA OF HEMEROBIIDAE 165 Results: L = 4. State (1) is uniquely derived in lineage 2, but is independently reversed (1—0) in lineages 19, 22, and Conchopterella. Character 2: Transtorular costa. (0) absent or poorly developed; (1) well developed (Fig. 1, trt). Comments: Tjeder (1961, e.g., fig. 467) noted the presence of a “postfrontal suture” in several hemerobiid taxa. This “suture” is certainly not homologous with the orthopteroid postfrontal sutures of Snodgrass (1935, fig. 57B), which diverge laterally from the coronal suture and pass behind the ocelli and antennal toruli. The heme- robiid “suture” is never continuous with the coronal suture and always extends between, not behind, the toruli. As Tjeder’s “postfrontal suture” appears to be only a secondary internal thickening of the cranium between the toruli, it is treated here under the name transtorular costa. The genus Micromus is polymorphic for this character; state (1) was used in the cladistic analysis. Results: L = 2. State (1) is uniquely derived in lineage 21, but is reversed (1—0) in Megalomina. Character 3: Midfront a l costa. (0) absent or weakly developed; (1) well developed (Fig. 1, mdf). Comments: The midfrontal costa is a simple internal costa of the midline of the frons. It is frequently expressed as a ventral extension of the median cusp of the transtorular costa. The genus Micromus is polymorphic for this character; state (1) was used in the cladistic analysis. Results: L = 2. State (1) is uniquely derived in lineage 21, but is reversed (1 -0) in Megalomina. Character 4; Frontoclypeal costa. (0) entirely absent, or present only as a pair of spurs arising from the clypeogenal costae near the anterior tentorial pits, spurs not united medially (Fig. 1, ftc); (1) present, a strong, transversely continuous, costa across the frontoclypeus below the anterior tentorial pits. Comments: The genus Micromus is polymorphic for this character; state (0) was used in the cladistic analysis. Results: L = 1. State (1) is a unique autapomorphy of Noius. Mouthparts (Characters 5-12) Character 5: Columns of transverse strigae on epipharyngeal surface of labrum. (0) absent; (1) two columns present (Fig. 3, str). Results: L = 1. State (1) is a unique autapomorphy of Megalomus. The function of this structure, noted here for the first time, is unknown. 166 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Character 6: Penicilliform sensilla of galea. (0) absent; (1) present (Fig. 8). Comments: The apices of hemerobiid galeae possess a distinctive type of penicil- liform sensillum. These sensilla consist of an elongate proximal rod terminated by a small distal cone (Fig. 8). Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 2. This character, first reported here, is considered a tentative synapomorphy of the Hem- erobiidae. Similar sensilla are lacking in the outgroups examined. Character 7; 5th palpomere of maxillary palpus. (0) not subsegmented (Fig. 9); (1) composed of two subsegments (Fig. 6). Comments: Prior works have characterized the hemerobiid maxillary palpus as 5- or 6-segmented. Similarly, the labial palpus (Character 8) has been characterized as 3- or 4-segmented. The “subsegmentation” terminology adopted here is preferred because it explicitly states the inferred homology of the added palpal unit. There is no doubt that the extra palpal units, the “6th” maxillary and “4th” labial, have been derived respectively from the 5th maxillary and 3rd labial palpomeres. This is best illustrated by the labial palpus, where the palpimacula (Fig. 10, pma) always occurs on the ultimate palpomere in taxa with “3-segmented” palpi (Fig. 10), but always on the penultimate “segment” in taxa with “4-segmented” palpi. Tjeder (1961:340) reported that the ultimate palpomeres of South African Sym- pherobius species lack subsegmentation. I have examined material from this area and have found subsegmentation to be present. Results: L = 3. State (1) is uniquely derived in lineage 3, but is independently reversed (1—0) in lineages 12 and 17. State changes of characters 7 and 8 occur identically on the cladogram. Although these characters are logically independent, their identical state-change patterns and similar natures suggest that they may share a common functional or developmental origin. Character 8: 3rd palpomere of labial palpus. (0) not subsegmented (Fig. 10); (1) composed of two subsegments (Fig. 7). Comments: See Comments under Character 7. Results: L = 3. State (1) is uniquely derived in lineage 3, but is independently reversed (1 —0) in lineages 12 and 17. See also Results for Character 7. Character 9: Dorsal (cervical) margin of postmentum. (0) transverse or irregularly parabolic and poorly defined; (1) regularly parabolic and well defined (Fig. 1 1). Comments: The genus Nusalala is polymorphic for this character; state (1) was used in the cladistic analysis. 1993 WORLD GENERA OF HEMEROBIIDAE 167 Results: L = 2. State (1) is uniquely derived in lineage 21, but is reversed (1—0) in Megalomina. Character 10: Orad cavity of right mandible. (0) absent [lost]; (1) present, inconspicuous (Fig. 12, ocv); (2) present, prominent. Comments: The states of this character were analyzed as nonadditive. Results: L = 3. State (0) is independently derived (1—0) in lineage 3 and the outgroup Psychopsis birmana\ state (2) is uniquely derived (1—2) in the outgroup Polystoechotes punctatus. Carobius pulchellus was the only hemerobiid examined which possessed an orad cavity on the right mandible. This cavity is enclosed by the prominent dorsal mandibular blade and a smaller ventral shelf. In this analysis, this character is interpreted as plesiomorphic within the Hemerobiidae. In all other hem- erobiids this cavity has apparently been lost through the atrophy of the ventral shelf. Additional study of this character in other Carobius species and other neuropterous families is warranted. Character 1 1: Proximal convexity of orad margin of right mandible. (0) prominently convex, rounded (Fig. 16, pxc); (1) prominently convex, strongly angulate (Figs. 13, 15). Results: L = 3. State (1) is independently derived in lineages 4, 12, and Noius. Character 12: Distal convexity of orad margin of right mandible. (0) absent (Fig. 13) or poorly developed; (1) well developed, rounded (Fig. 14); (2) well developed, strongly angulate (Fig. 15; 16, dtc). Comments: The states of this character were analyzed as nonadditive. Results: L = 3. State (1) is uniquely derived (0—1) in lineage 7; state (2) is inde- pendently derived (0 — 2) in lineage 12 and Psychobiella. Clypeal Setae (Characters 13-17) The relative positions and terminology applied to the principal long setae of the clypeus are illustrated in Figure 2. These setae may be identified by their size, and their positions relative to each other, the anterior tentorial pits, and the clypeogenal and clypeopleurostomal costae. Additional setae are generally present on the clypeus, but are too variable in size and/or position to be of taxonomic value. The four main pairs of principal clypeal setae— the dorsolaterals, dorsomedials, ventrolaterals, and ventromedials— are broadly homologous throughout the Hemerobiidae. Character 13: Paired dorsolateral setae of clypeus. (0) absent; ( 1 ) present (Fig. 2, dls). 168 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 2. Character 14: Sagittal seta of clypeus. (0) absent; (1) present (Fig. 2, ss). Results: L = 1. State (1) is an autapomorphy of Hemerobius. Character 15: Paired mesolateral setae of clypeus. (0) absent or short; (1) present and long. Comments: Species of the genus Notiobiella possess an extra pair of prominent clypeal setae just dorsad of the ventrolaterals, herein called the mesolaterals. The genus Neuronema is polymorphic for this character; state (0) was used in the cladistic analysis. Results: L = 1. State (1) is an autapomorphy of Notiobiella. Character 16: Paired ventrolateral setae of clypeus. (0) absent; (1) present (Fig. 2, vis). Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 2. Character 17: Paired ventromedial setae of clypeus. (0) absent; (1) present (Fig. 2, vms). Results: L = 2. State (1) is uniquely derived in lineage 2, but is reversed (1 —0) in Psychobiella. Tentorium (Characters 18-19) Character 18: Sagittal process of tentorial bridge. (0) present (Fig. 4, stp); (1) absent [lost]. Results: L = 3. State (1) is independently derived in lineages 1, 20, and Noius. Character 1 9: Mesal lobes of anterior tentorial arms. (0) widely separated; (1) very closely adjacent or overlapping (Fig. 4, mtl). Results: L = 1 . State ( 1 ) is an autapomorphy of Notiobiella. Head Miscellaneous (Characters 20-22) Character 20: Setal foveae of cranium. 1993 WORLD GENERA OF HEMEROBIIDAE 169 (0) absent or few (< 10) and inconspicuous; (1) numerous (> 10) and prominent. Comments: In Gayomyia falcata most of the cranium is prominently punctate. Prominent punctures are rare or absent in other hemerobiids. Results: L = 1. State (1) is an autapomorphy of Gayomyia. Character 2 1 : Field of punctulae on posterior margin of gena. (0) absent; (1) present (Fig. 5, gpt). Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 16. Character 22: Bilobed cavity of male frons. (0) absent; (1) present (Fig. 78). Results: L = 1. State (1) is an autapomorphy of Neosympherobius. FOREWING (Characters 23-49) General forewing structure and terminology are given in Figures 17 and 18. Costal Space (Characters 23-26) Character 23: Trichosores of anterior wing margin. (0) present proximally and distally; (1) lost proximally, present distally; (2) lost proximally and distally. Comments: Pre-analysis transformation hypothesis: 0- 1 —2. This transformation suggests that trichosores are lost sequentially from the wing margin, beginning at the base of the wing and proceeding distally. This idea is supported by trichosore dis- tributions in other Neuroptera (e.g., Mantispidae, see Lambkin, 1986), where prox- imal trichosores appear to be lost before those located more distally. Results: L = 4. State (1) is independently derived (0— 1) in lineage 21, Zachobiella , and the outgroup Nothochrysa californica, and further uniquely derived (1-2) in Nothochrysa. This pattern corroborated the pre-analysis transformation hypothesis of this character. Character 24: Humeral veinlet. (0) branched, proximal humeral trace recurrent and with 3 or more rami (Fig. 18, pht); (1) branched, proximal humeral trace recurrent but with only 1 or 2 rami (Fig. 1 15); (2) simple, branching absent [lost], straight or slightly bent. Comments : Pre-analysis transformation hypothesis: 0-1-2. This transformation assumes sequential reduction in the prominence of the humeral veinlet, particularly in association with reduction in the width of the humeral area. The genera Micromus 170 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) (1), Noius (1), Psectra (0), Sympherobius (0), and Zachobiella (1) are polymorphic for this character. Tne states used in the cladistic analysis for these genera are given in parentheses. Results: L = 5. State (1) is independently derived (0—1) in lineages 13, 21, and the outgroup Nothochry’sa californica , but is reversed (1—0) in Megalomina and further derived (1—2) in Nothochrysa. The pre-analysis transformation hypothesis of this character was corroborated, except for the anomalous reversal in Megalomina. Character 25: Humeral marginal silhouette. (0) convex (Fig. 1 7); (1) linear or shallowly concave (Fig. 191). Results: L = 3. State (1) is independently derived in lineages 21, Zachobiella , and the outgroup Nothochrysa californica. Character 26: Crossveins [not veinlets] of costal space. (0) numerous (>4/wing) and regularly positioned, forming a distinct gradate series in costal space (Fig. 182); (1) absent (Fig. 24) or few in number (rarely > 4/wing); (2) numerous (> 4/wing) but irregular in position, forming, with costal veinlets, a reticulate network in costal space at base of wing (Fig. 174). Comments: Pre-analysis transformation hypothesis: 1—0; 1—2. This transforma- tion suggests the independent acquisition of numerous costal crossveins in Drepa- nepteryx and Gayomyia based on their different arrangements. The genera Con- chopterella (1), and Micromus (1) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 2. State (0) is uniquely derived in Drepanepteryx\ state (2) is uniquely derived in Gayomyia. The pre-analysis transformation hypothesis was corroborated. Subcostal Space (Characters 27-28) Character 27: Prestigmal width of subcostal space. (0) narrow to broad, but > width of adjacent Sc throughout (Fig. 17); (1) very narrow, < width of adjacent Sc throughout (Fig. 98) (sometimes slightly wider immediately adjacent to crossvein lsc-r). Results: L = 1. State (1) is an autapomorphy of Notiobiella. Character 28: Subcostal crossvein 2sc-r. (0) absent (Fig. 24); (1) present (Figs. 17,90). Comments: The genera Notiobiella (1), and Zachobiella (1) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. 1993 WORLD GENERA OF HEMEROBIIDAE 171 Table 2. State transformations for Characters 29-33. Character Transformation' 29 0-1 30 0-2 31 1-3 32 1-4 33 3-5 1 Unit length transformations. Results: L = 2. State (1) is uniquely derived in lineage 9, but is reversed (1-0) in lineage 22. In Noius , the sister-group to lineage 22, crossvein 2sc-r is frequently poorly developed or vestigial. This state may be transitional between a fully developed 2sc-r crossvein and its loss in lineage 22. Radial Space (Characters 29-37) Characters 29-33: Radial vein configuration. (0) 1 ORB, the true radial sector. (1) 2 ORB’s, ORB1 deeply forked (Fig. 66). (2) 2 ORB’s, ORB1 shallowly forked (Fig. 24). (3) 3 ORB’s, ORB 1 deeply forked (Fig. 90). (4) 3 ORB’s, ORB1 shallowly forked (Fig. 33). (5) 4+ ORB’s (e.g.. Fig. 207). All genera characterized by 4 or more “radial sectors” are placed here. Comments. Interpret “deeply forked” as: most proximal anterior branch of ORB1 arising from proximal half of posterior ORB1 trace, “shallowly forked” as: most proximal anterior branch of ORB1 arising from distal half of posterior ORB1 trace. Characters 29-33 involve transformations among the six states listed above. The following pre-analysis transformation series were hypothesized: (Series 1) 0 — 2; (Se- ries 2) 0— 1—3 — 5; (Series 3) 1 —4. The transformation assigned to each character is given in Table 2. Data matrix values for characters 29-33 (columns 29-33 in Ap- pendix 5) contain additive binary coding sequences encoding the preceding trans- formation series. The binary coding sequence of each state is given in Table 3. The homologies and polarities of radial vein states are discussed below. Discussion. The homologies of the oblique radial branches (ORB’s) of the forewing have been difficult to assess. Three principle areas of ambiguity are involved. Phrased in the form of questions, these are: ( 1 ) Do the ORB’s contain veins homologous with the MA?, (2) Do the ORB’s contain veins homologous with the Rl?, and (3) What is the correspondence between individual ORB’s and portions of the ancestrally pectinate neuropterous Rs. These questions are addressed below. In the discussion, the following definitions are employed: (1) the apparent R is taken to be the longi- tudinal vein posterior to the Sc and anterior to the apparent M, (2) the apparent M is taken to be the longitudinal vein immediately anterior to the Cu, which is always clearly identifiable, and (3) the apparent Rs is taken to be the most proximal ORB. 172 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Table 3. Additive binary coding values for characters 29-33. Slate Character 29 30 31 32 33 (0)° 0 0 0 0 0 (i)' 1 0 0 0 0 (2)2 0 1 0 0 0 (3)3 1 0 1 0 0 (4)4 1 0 0 1 0 (5 y 1 0 1 0 1 0 Nothochrysa, Polystoechotes, Psychopsis (outgroups). 1 All sympherobiine and notiobielline genera. 2 Carobius. - Psychobiella. 4 All hemerobiine genera. 5 All drepanacrine, megalomine, drepanepterygine, and micromine genera. Do the ORB’s Contain Veins Homologous with the MA? Many, if not most, present neuropterists implicitly or explicitly accept the hy- pothesis that an ancestral anterior branch of the media, the MA, has fused anteriorly with the R, and that its remnant is represented in the fore- and hind-wings of most extant Neuroptera by the most proximal branch of the apparent R or apparent Rs. This acceptance is reflected in the terminological conventions commonly adopted for forewing venation. Martynov (1928) first introduced the idea that the apparent neuropterous R is a hybrid R+MA, an idea subsequently adopted by Carpenter (e.g., 1936, 1940), and his students (e.g., Adams, [I960]; Adams, 1967; MacLeod and Adams, [1968]; Adams, 1970). Unfortunately, a comprehensive review of this con- cept is beyond the scope of this work, and its applicability to the Hemerobiidae is judged here solely on the basis of evidence found in hemerobiid taxa. At least two general lines of supportive evidence have been advanced to suggest the existence of a hybrid forewing R + MA in the Hemerobiidae. These are; (1) the presence of a sigmoid vein (Fig. 167, sv) connecting the apparent M and R in the hind wings of all hemerobiids, and (2) the presence of a “basal” or “b” crossvein in some hemerobiids (Kevan and Klimaszewski, 1987; fig. 2, “b”). The presence of a sigmoid vein connecting the apparent M and R in the hind wing is very widespread within the Neuropterida (see, for example, the wing illustrations in Aspock et al., 1980). Tillyard (1916, text-fig. 4) interpreted this vein as the basal portion of the ancestral Rs which had become secondarily fused to the base of the M. Martynov, however, interpreted it as a basal branch of the M, which he designated the MA. A substantial body of evidence seems to support Martynov’s interpretation. For example: (1) the invariable origin of this vein from the base of the apparent M, (2) its frequently long uncrossvein-like appearance, and (3) its broad distribution within the Megaloptera (Van der Weele, 1910b) and Raphidioptera (Aspock et al., 1980), where the identity of the true Rs appears to be clear, as well as in the Neu- roptera. If this hind wing vein is the MA (which seems likely), and if one is willing to assume parallel venational development of the forewing and hind wing, the fusion of MA in the hind wing could be used as evidence of a similar fusion in the forewing. However, because differential development of other forewing and hind wing venation 1993 WORLD GENERA OF HEMEROBIIDAE 173 is common within the Neuroptera, there is little a priori reason to suspect that the second premise of this argument is true. Carpenter (1940) reported that a “basal” or “b” crossvein (Kevan and Klimaszew- ski, 1987: fig. 2, “b”) was present in most Hemerobius species, and occasionally present in other North American hemerobiids. 1 have found the “b” crossvein only in species of Hemerobius. Occasionally, in other genera, forewing cell c2r-m is some- what darkly pigmented or sclerotized proximally in the cleft between the apparent R and apparent M. This may falsely give the impression of a crossvein at that position. This manner of modification could be misinterpreted as a “b” crossvein. Carpenter interpreted the “b” crossvein as the vestigial remnant of the basal piece of the forewing MA. I interpret it as forewing crossvein 2r-m, which is lost, or proximally displaced, in Hemerobius. The latter interpretation seems better sup- ported: (1) the “b” crossvein in Hemerobius possesses all of the characteristics nor- mally associated with true crossveins, i.e., it is transverse, weakly sclerotized, non- tracheate, and asetose; (2) the “b” crossvein is present only in Hemerobius ; and (3) Hemerobius is the only hemerobiid genus apparently lacking a forewing 2r-m cross- vein. The coincidence of the last two points appears to be particularly compelling evidence supporting the homology of the “b” crossvein with crossvein 2r-m of other hemerobiids. In conclusion, I find no clear evidence to support the hypothesis that the most proximal branch of either the apparent R or apparent Rs in the hemerobiid forewing has been derived from the M. Consequently, in this work, I adopt the terminological convention of calling the apparent M the media (M) and the apparent R the radius (R). This terminology implies that if an anterior branch of the M did fuse anteriorly with the R, its terminal branchings have subsequently been suppressed, and its stem entirely lost. Do the ORB’s Contain Veins Homologous with the Rl? Previous workers have not discussed the possibility that at least some ORB’s may represent branches of the Rl that have been displaced proximally by the basad “migration” of distal R 1 forks. Comstock (1918) regarded a similar fork “migration” mechanism, operating on the “R2”, to have produced the pectinate Rs found in many Neuroptera. The principal evidence opposing the evolution of such a mechanism on the hem- erobiid anterior radial trace is the presence in most genera of a long substigmal interramus, i.e., the portion of the anterior radial trace located between the most distal prestigmal ORB and the most proximal terminal fork of the R 1 . The ubiquitous presence of this long interramus is difficult to reconcile with branching patterns expected from the operation of an Rl fork migration mechanism. Such a mechanism would be expected to produce veins separated by interrami of subequal length along the entire distal portion of the anterior radial trace. But, veins nearly meeting this expectation are found in only a few genera, e.g.,Austromegalomus and Drepanepteryx, which contain species characterized by especially numerous ORB’s. Therefore, I conclude that Rl fork “migration” has not been an important mechanism in pro- ducing hemerobiid ORB’s. Although the distal ORB’s in some Austromegalomus and Drepanepteryx species may originate from this source, taxa possessing 2 to 4 ORB’s (the ORB states of most concern here) are regarded as not containing ORB’s of Rl origin. 174 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) How do the ORB’s Correspond to the Branches of the Ancestral Rs? Having outlined above my reasons for dismissing possible claims of partial ho- mology of the ORB's with the R1 and/or M. it seems clear that the ORB's must be collectively homologous with the Rs, since the R 1 and M bound the Rs anteriorly and posteriorly. This interpretation supports the homologies of Comstock (1918) over those proposed by Tillyard (1916) or Carpenter (1940). Radial State Polarities Tillyard (1916:290) proposed that forewing states with few (2-3) “radial sectors” (ORB’s) were derived through reduction from ancestral states with numerous “radial sectors”. This opinion was quoted and accepted by Nakahara (1960b: 1-2). Although I agree that a morphocline in “radial sector” number is clear, I disagree with Tillyard’s assessment of its polarity. I suggest that evolution in “radial sector” number has generally (although not necessarily exclusively) increased, not decreased, within the Hemerobiidae, and that convergence toward higher “radial sector” numbers (4+) is more prevalent than convergence toward lower numbers. At least two lines of evi- dence suggest the latter polarity. First, since a single ORB. the Rs, is clearly plesiomorphic within the Neuroptera. an increase in ORB number must, at some level, be derived within the family. Logically, a sequential increase in ORB number, 1 ->2 — 3 — 4 + , would appear to be reasonable. Second, independent increases in radial sector number are evident within some hemerobiid genera. For example: (1) most Wesmaelius are characterized by 3 ORB’s (apparently the plesiomorphic number within the Hemerobiinae), but several species normally possess 4 (e.g., quadrifasciatus , concinnus), and (2) Oswald (1988b:400) suggested, based on a cladistic analysis of Nearctic Sympherobius species, that the 3-ORB state found in the four species of the European fuscescens species group is derived, relative to the 2-ORB state predominant within that genus and apparently plesiomorphic within the Sympherobiinae. Confirmation of a general trend toward increase in ORB number cannot, however, completely establish ORB homologies, since portions of the ancestrally pectinate neuropterous Rs may have become secondarily associated with the anterior radial trace at different times and/or at different points. Starting with a plesiomorphic forewing radial configuration consisting of a pectinately branched Rs, a simple Rl, and one or more crossveins joining the two, at least three general classes of mech- anisms can be imagined which could account for the shifting of the origins of ancestral Rs branches to the anterior radial trace. The first class of mechanisms postulates the partial basal fusion of the Rs with the Rl. In simple formulations, these mechanisms would appear to result in a distal “migration” of the origin of the Rs along the anterior radial trace. However, it is possible to escape this difficulty by suggesting concurrent or subsequent basal “mi- gration” of individual branches following the partial anastomosis of Rl and the anterior sectoral trace. The second class of mechanisms invokes a concept which might be called “cross- vein capture”. These mechanisms require the existence of one or more crossveins 1993 WORLD GENERA OF HEMEROBIIDAE 175 joining the ancestral R1 and Rs. These crossveins may “capture” a portion of the Rs by rotating from a transverse to an oblique orientation and coming to serve as the functional base of a portion of the Rs, or by decreasing in length and thus “pulling” a segment of the anterior sectoral trace toward the anterior radial trace. In the latter case, eventual elimination of the crossvein after a period of shortening could result in the attachment an Rs fragment directly to the anterior radial trace. The third mechanism, adventitious fragmentation of the Rs, with subsequent reas- sociation of the fragments along the anterior radial trace, is not theoretically or mechanistically satisfying, but may account for much of the intraspecific variability in ORB numbers and origination positions observed in taxa which generally possess 5 or more ORB’s. This mechanism cannot be conclusively differentiated from a mechanism which supposed the adventitious origin of entirely new branches from the Rl. All three types of mechanisms have been discussed by several previous authors (e.g.. Carpenter, 1940; Comstock, 1918; Tillyard, 1916). Each general mechanism appears plausible; but, none appear to leave behind an unambiguous “signature”. Consequently, the evolution of radial configurations in the Hemerobiidae is likely to remain ambiguous. Three principal criteria were used to define and polarize the radial states recognized here. First, the observed number of ORB’s [up to 4] was used to divide radial states into four groups. As a working hypothesis, the state adopted for each taxon (species or genus) was the one most frequently observed in the taxon. However, since taxa characterized by four or more ORB’s generally exhibit high levels of intraspecific and intrageneric variation in ORB number, these taxa were grouped under a single state. Second, the observed location of the most proximal fork of the most proximal ORB was used to divide the 2-ORB and 3-ORB groups into two states each, two in which the most proximal ORB was deeply forked (Figs. 66, 90), and two in which this vein was only shallowly forked (Figs. 24, 33). Third, and finally, the six states were polarized by assuming a sequential increase in ORB number and a progression from deeply forked to shallowly forked states. Results: Character 29: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 3. Character 30: L = 1. State (1) is an autapomorphy of Carobius. Character 31: L = 2. State (1) is independently derived in lineages 14 and Psy- chobiella. Character 32: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 4. Character 33: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 14. These results generally corroborate the pre-analysis polarities. In particular, the assumption that ORB numbers generally increase within the family is borne out. The subfamilies Carobiinae, Hemerobiinae, Sympherobiinae, and Notiobiellinae, which are characterized by low (2-3) ORB numbers, clearly cluster at the base of the cladogram, while those possessing higher ORB numbers (Drepanacrinae, Megalom- inae, Drepanepteryginae, Microminae) are grouped further from the base. The par- allelism of character 31 in lineages 14 and Psychobiella , and the occurrence of both characters 31 and 33 on lineage 14, are the only two apparently anomalous features of this character with regard to the pre-analysis polarities. 176 JOURNAL OF THE NEW YORK. ENTOMOLOGICAL SOCIETY Vol. 101(2) Character 34: Most proximal fork of posterior sectoral trace. (0) located in proximal one-third of posterior sectoral trace (Fig. 66); (1) located in distal two-thirds of posterior sectoral trace (Fig. 24). Comments: The genera Megalomus (0). Neuronema (0), and Sympherobius (0) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 5. State (1) is independently derived in lineages, 4, 21, Carobius , Xotiobiella. and Gayomyia. Post-analysis consideration of this character suggests that the derived state (1) has evolved by several independent mechanisms. In No- tiobiella. the basally simple posterior sectoral trace appears to have resulted from the lengthening of the Rs stem due to the distal migration of its most proximal fork (Fig. 98). In all other taxa. a simple posterior sectoral trace appears to have resulted from a shift in the origin of its anterior branches to the anterior radial trace; however, this association is inferred to have occurred at different times in different lineages. Thus, these states are probably not homologous, and this character should be dropped or revised in subsequent analyses. This character may also not be completely inde- pendent from characters 29-33. Character 35: Pre-3irl intraradial crossveins. (0) present (Figs. 17, 166); (1) absent (Fig. 50). Results: L = 3. State (1) is uniquely derived in lineage 2. but is independently reversed (1 —0) in Neuronema and Gayomyia. Character 36: Intraradial crossvein 4irl. (0) present (Figs. 17, 33); (1) absent [lost] (Fig. 74). Comments: The genera Conchopterella (0), Micromus (0), and Psectra ( 1 ) are poly- morphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 2. State (1) is independently derived in lineage 12 and Neosympher- obius. Character 37: Posterodistal marginal silhouette. (0) convex (Fig. 24); (1) concave (Figs. 149, 174). Comments: The genera Conchopterella (0), Drepanepteryx (0). Hemerobius (0), Megalomina (0), Micromus (0), and Sympherobius (0) are polymorphic for this char- acter. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 2. State (1) is independently derived in Drepanacra and Gayomyia. The derived state is also independently derived in some species of other genera. 1993 WORLD GENERA OF HEMEROBIIDAE 177 Radiomedial Space (Characters 38-40) Character 38: Radiomedial crossvein 2r-m. (0) present (Figs. 17, 24); (1) absent [lost] (Fig. 33) or displaced proximad of crossvein 2m-cu. Comments: In Hemerobius crossvein 2r-m is absent or displaced toward the base of the wing. When present, it always lies proximal to the position of crossvein 2m- cu, and generally lies proximal to fork Ml +2/M3 + 4. The genus Notiobiella is poly- morphic for this character; state (0) was used in the cladistic analysis. Results: L = 2. State (1) is independently derived in Hemerobius and the outgroup Psychopsis birmana. Character 39: Radiomedial crossvein 3r-m. (0) present (Figs. 17, 33); (1) absent [lost] (Fig. 24). Comments: The genus Micromus is polymorphic for this character; state (0) was used in the cladistic analysis. Results: L = 2. State (1) is independently derived in Carobius and the outgroup Nothochrysa californica. Character 40: Radiomedial crossvein 4r-m. (0) present (Figs. 17,41); (1) absent [lost] (Fig. 66). Comments: The genus Notiobiella is polymorphic for this character; state (1) was used in the cladistic analysis. Results: L = 3. State ( 1 ) is independently derived in lineages 7,11, and the outgroup Nothochrysa californica. Media Posterior Space (Characters 41-42) Character 41: Intramedial crossvein 2im. (0) absent (Fig. 24); (1) present (Fig. 182). Comments: The genera Conchopterella (0), Megalomina (0), and Micromus (0) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 2. State (1) is independently derived in lineage 19 and the outgroup Nothochrysa californica. Character 42: Intramedial crossvein 4im. (0) present (Figs. 17, 41); (1) absent [lost] (Fig. 98). 178 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Results: L = 3. State (1) is independently derived in lineage 11, Sympherobius , and the outgroup Nothochrysa californica. Mediocubital Space (Characters 43-46) Character 43: Mediocubital crossvein 2m-cu. (0) present (Figs. 17, 41); (1) absent [lost] (Figs. 199. 207). Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 22. Character 44: Mediocubital crossvein 3m-cu. (0) present (Figs. 17, 41); (1) absent [lost] (Fig. 24). Comments: The genera Hemerobius (0), Megalomus (0), and Micromus (0) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 4. State (1) is independently derived in lineage 6. Carobius , Nusalala , and the outgroup Nothochrysa californica. Character 45: Mediocubital crossvein 4m-cu. (0) present (Figs. 17,41); (1) absent [lost] (Fig. 82). Comments: The genus Micromus is polymorphic for this character; state (0) was used in the cladistic analysis. Results: L = 3. State (1) is independently derived in lineage 11, Sympherobius , and the outgroup Nothochrysa californica. Character 46: Mediocubital flexion line. (0) present (Figs. 18. 33); (1) absent [lost] (Fig. 174). Results: L = 2. State (1) is independently derived in Gavomyia and the outgroup Nothochrysa californica. Loss of the mediocubital flexion line in Gavomyia is as- sociated with the proliferation of finely reticulate veins at the base of the forewing in this genus. Cubital Space (Characters 47 — 48) Character 47: Intercubital crossvein lcua-cup. (0) absent (Figs. 17, 50); (1) present (Fig. 199). Comments: The genus Neuronema is polymorphic for this character; state (1) was used in the cladistic analysis. Results: L = 2. State (1) is independently derived in lineages 19 and 22. 1993 WORLD GENERA OF HEMEROBIIDAE 179 Character 48: Depth of most proximal fork of forewing CuP. (0) fork absent or shallow, length of CuP stem before fork > length of posterior CuP trace beyond fork (Fig. 41); (1) deep, length of CuP stem before fork < length of posterior CuP trace beyond fork (Figs. 17, 98). Comments: The genus Conchopterella is polymorphic for this character; state (1) was used in the cladistic analysis. Results: L = 4. State (1) is independently derived in lineage 14, Notiobiella, and the outgroup Polystoechotes punctatus, but is reversed (1 — 0) in lineage 22. Anal Space (Character 49) Character 49: Depth of most proximal fork of forewing A] . (0) absent or shallow, length of A1 stem before fork > length of posterior A1 trace beyond fork (Fig. 24); (1) fork deep, length of A1 stem before fork < length of posterior A1 trace beyond fork (Figs. 1 7, 149). Comments: The genera Conchopterella ( 1 ), Megalomus (0), and Neuronema (0) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 3. State (1) is independently derived in lineages 15, 20, and the outgroup Psychopsis birmana. MALE TERMINALIA (Characters 50-95) 9th Tergite (Characters 50-55) Character 50: 9th tergite. (0) not completely divided sagittally (anterior and/or posterior margin may be sag- ittally emarginate); (1) completely divided sagittally. Results: L = 2. State (1) is independently derived in lineage 22 and Gayomyia. Character 51: Antecosta of 9th tergite (lateral view). (0) not obliquely crossing tergite from anterior to posterior margin (Fig. 36); (1) obliquely crossing tergite from anterior to posterior margin (Fig. 148). Results: L = 2. State (1) is independently derived in lineage 16 and Psychobiella. Character 52: Cusps of posterolateral margins of male 9th tergite. (0) absent (Fig. 47); (1) present (Fig. 69, cus). Results: L = 1 . State ( 1 ) is an autapomorphy of Nomerobius. 180 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Character 53: Posteroventral angle of 9th tergite. (0) produced as a narrow membrane-margined lobe (Fig. 69); (1) not prominently produced (Fig. 47) or. if somewhat produced, expanded as a broad membrane-margined lobe; (2) produced as an elongate subectoproctal lobe (usually) with a free distal process (Fig. 108. 117). Comments: Pre-analysis polarity hypothesis: 1—0; 1—2. Results: L = 2. State (0) is uniquely derived in lineage 7; state (2) is uniquely derived in lineage 12. The pre-analysis hypotheses of polarity were corroborated. Character 54: Horizontal costae of inner lateral surfaces of 9th tergite. (0) absent; (1) present (Fig. 65. cos). Results: L = 1. State (1) is an autapomorphy of Hemerobiella. Character 55: 9th and 10th tergites. (0) not fused (Fig. 27); (1) fused (Fig. 222). Results: L = 2. State (1) is independently derived in lineage 22 and the outgroup Nothochrysa californica. 9th Stemite (Characters 56-63) Character 56: 9th stemite. (0) absent [lost] (Fig. 127); (1) present, a free sclerite (Fig. 27); (2) present, fused with 8th stemite. Comment: Pre-analysis polarity hypothesis: 1—0; 1—2. Results: L = 2. State (0) is uniquely derived in Zachobiella ; state (2) is uniquely derived in the outgroup Xothochmsa californica. The pre-analysis polarity hypotheses were corroborated. Character 57: Sagittal costa of dorsal surface of 9th stemite. (0) absent; (1) present (Fig. 144. cos). Comments: The midline of the male 9th stemite Austromegalomus insulanus is marked in its anterior lh to 'h by a low sagittal costa. This costa appears to be restricted to Austromegalomus. Results: L = 1. State (1) is an autapomorphy of Austromegalomus. 1993 WORLD GENERA OF HEMEROB11DAE 181 Character 58: Apex of 9th sternite (A). (0) not produced as a long, straight, attenuate tube; (1) produced as a long, straight, attenuate tube (Fig. 47). Results : L = 1. State (1) is an autapomorphy of Nesobiella. Character 59: Apex of 9th sternite (B). (0) not tubularly produced and abruptly upturned distally; (1) tubularly produced and abruptly upturned distally (Fig. 69). Results: L = 2. State (1) is independently derived in Nomerobius and the outgroup Polystoechotes punctatus. Character 60: Digitiform posterior process of 9th sternite. (0) absent; (1) present (Fig. 194). Comments: Each of the three species of Noius examined possess a narrow scler- otized process issuing medially from the posteroventral margin of the male 9th sternite. This derived state is unique to Noius. It differs from the drawn-out 9th sternite processes of Nomerobius and Nesobiella by lacking sclerotization proximo- dorsally and/or laterally, and by differences in setation. The last trait is particularly diagnostic. In Nomerobius the apical portion of the 9th sternite has lost its setae or bears only modified spine-like or peg-like setae (Oswald, 1990). In Nesobiella , the attenuate apex bears normal, unmodified macrotrichia along its entire length. In Noius oceanicus and noumeanus the posteroventral process bears a small field of unmodified setae apically and a dense tuft of setae posterobasally. An undescribed species of Noius from Samoa has the apical setae thickened and arranged in a fan- like array, and the basal setal tuft absent (?lost). The protruded apices of the male 9th sternite in Noius , Nomerobius , and Nesobiella, are considered here to be parallel, analogous developments, a hypothesis supported by their distant cladistic relation- ships (Fig. 23). Results: L = 1 . State ( 1 ) is an autapomorphy of Noius. Character 6 1 : Posterior margin of 9th sternite. (0) without a pair of acute cusps; (1) with a pair of acute cusps (Fig. 53). Comments: The posterior margin of the male 9th sternite in 1 Vesmaelius bears a pair of cusps, which are frequently drawn out as slender processes. Less prominent cusps are found in some Hemerobius species. These are considered parallel devel- opments; and absence of such cusps (state 0) is presumed plesiomorphic in Heme- robius. Results: L = 1 . State ( 1 ) is an autapomorphy of Wesmaelius. 182 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Character 62: Posterolateral lobes of 9th sternite. (0) absent; (1) present (Fig. 194). Results: L = 1 . State ( 1 ) is an autapomorphy of Noius. Character 63: Paired elongate setae of 9th sternite. (0) absent: ( 1 ) present (Fig. 69). Comments: In addition to normal unmodified macrotrichia, all known species of Nomerobius possess a pair of elongate, chalazate macrotrichia posterolaterally on the male 9th sternite. Results: L = 1. State (1) is an autapomorphy of Nomerobius. 10th Tergite [Ectoproct] (Characters 64-68) Character 64: Mesal surfaces of ectoprocts,. (0) largely membranous (excluding narrowly protruding lobes); (1) largely sclerotized (narrowly protruding lobes absent) (Fig. 136). Comments: The male ectoprocts of Conchopterella are large, planar, setose, struc- tures which are broadly sclerotized on their mesal surfaces. In all other genera, the mesal surfaces of the ectoprocts are largely membranous, with mesal sclerotization restricted to the inner surfaces of narrowly protruding lobes and/or the revolute margins of outer surfaces. Results: L = 1. State (1) is an autapomorphy of Conchopterella. Character 65: Fusiform setae of ectoproct. (0) absent; ( 1 ) present (Fig. 46). Results: L = 1. State (1) is an autapomorphy of Nesobiella. Character 66: Pecten of ectoproct. (0) absent; (1) present (Fig. 59, pec). Comments: The male ectoprocts of Wesmaelius each possess a row of contiguous cuticular pegs. These pegs are apparently not derivatives of modified setae and are borne on a mesal process of varying length (sometimes very long). Results: L = 1. State (1) is an autapomorphy of Wesmaelius. Character 67: Articulated posteroventral lobe of ectoproct. (0) absent; (1) present (Fig. 108, 109). 1993 WORLD GENERA OF HEMEROBIIDAE 183 Comments: In Psectra, each male ectoproct possesses a small hinged or articulated posteroventral lobe which carries near its apex a row of variously modified setae. Results: L = 1. State (1) is an autapomorphy of Psectra. Character 68: Modified subapical seta of ectoproct. (0) absent; (1) present (Fig. 69). Comments: The ectoprocts of the four known species of Nomerobius each possess a single distinctive subapical seta, which is thickened, elongate and borne on a short pedicel. Results: L = 1. State (1) is an autapomorphy of Nomerobius. Gonarcus and Associated Structures (Characters 69-83) Character 69: Sagittal emargination of posterior margin of extragonopons [not med- iuncus]. (0) absent; (1) present (Fig. 195). Comments: All three species of Noius examined possess a peculiarly modified gonopons in which a pair of small parasagittal plates protrude posteriorly, resulting in the formation of a sagittal emargination between them. Results: L = 1. State (1) is an autapomorphy of Noius. Character 70: Ventral margins of extrahemigonarcus. (0) not turned outward; (1) distinctly turned outward (Fig. 77). Comments: In Neosympherobius cinereus the ventral portions of the reduced ex- trahemigonarcus are strongly flared laterally. Results: L = 1. State (1) is an autapomorphy of Neosympherobius. Character 7 1 : Horizontal costae of inner surfaces of intrahemigonarcus. (0) absent; (1) present (Fig. 177, cos). Results: L = 1. State (1) is an autapomorphy of Gayomyia. Character 72: Horizontal costae of outer surfaces of intrahemigonarcus. (0) absent; (1) present (Fig. 76, cos). Results: L = 1. State (1) is an autapomorphy of Neosympherobius. Character 73: Lateral neogonarcal carinae of gonarcus. (0) absent; (1) present (Fig. 72, car). 184 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Comments: Each of the four species of the genus Nomerobius possesses a pair of small neogonarcal carinae or plates. These are oriented perpendicular to their re- spective extrahemigonarcus and are located ventrally along the antextragonarcal commissure. Results: L = 1. State (1) is an autapomorphy of Nomerobius. Character 74: Gonofenestra. (0) absent (Fig. 28); (1) present (Fig. 153, gfa). Results: L = 4. State (1) is uniquely derived on lineage 10. but is reversed (1^0) on lineages 20. 23. and Zachobiella. Character 75: Gonofenestral plate. (0) absent (or gonofenestra absent); (1) present (Fig. 134. gfp). Comments: In Conchopterella the anteromedian portion of the gonofenestral mem- brane has become secondarily sclerotized and forms a broad gonofenestral plate, which is separated from the mediuncus by a constricted gonofenestral membrane. Results: L = 1. State (1) is an autapomorphy of Conchopterella. Character 76: Secondary gonopons of gonarcus. (0) absent; (1) present (Fig. 73. sgp). Comments: Nomerobius is unique among hemerobiid genera in possessing a sec- ondarily derived sclerotized bridge, the secondary gonopons. joining the postero- ventral margins of the extrahemigonarcus. All four known species of Nomerobius possess this structure (Oswald. 1990). The transverse bar joining the hemigonarcus of some Psectra species is interpreted here as a derivative of the mediuncus. Results: L = 1. State (1) is an autapomorphy of Nomerobius. Character 77; Transverse subbasal costa of ventral surface of mediuncus. (0) absent; (1) present (Fig. 96, cos). Comments: In Psychobiella the mediuncus is narrowed and bears a transverse costa on its ventral surface near the base. Results: L = 1. State (1) is an autapomorphy of Psychobiella. Character 78: Apex of mediuncus. (0) mediuncus absent (Fig. 76); (1) apex pointed or rounded (not emarginate) (Figs. 19. med; 176); (2) emarginate (Fig. 153); (3) terminating in a pair of long, slender, pliable processes (Fig. 162. med). 1993 WORLD GENERA OF HEMEROBIIDAE 185 Comments: Pre-analysis polarity hypothesis: 1—0; 1-2-3. Results: L = 7. State (1) is most plesiomorphic. State (2) is independently derived (1-2) on lineage 10 and in the outgroup Psychopsis birmana. State (0) is uniquely derived (1 —0) on lineage 8. State (3) is uniquely derived (2 — 3) in Megalomus. State (2) is independently reversed (2—1) on lineages 18 and Anapsectra , and rereversed (1—2) in Megalomina. Although the pre-analysis transformations were generally upheld, the configuration of the apex of the mediuncus seems particularly plastic in the Hemerobiidae, and convergence among some of these states may be expected. Character 79: Pseudomediuncus. (0) absent; (1) present (Fig. 84, pmd). Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 7. Character 80: Base of pseudomediuncus. (0) narrow (or pseudomediuncus absent) (Fig. 73); (1) laterally expanded (Fig. 85). Results: L = 1. State (1) is an autapomorphy of Sympherobius. Character 8 1 : Pseudomediuncus divisions. (0) absent, pseudomediuncus absent or undivided (Fig. 72, pmd); (1) present, divided into two sclerites (Fig. 84). Results: L = 1. State (1) is an autapomorphy of Sympherobius. Character 82: Gonosaccal membrane below mediuncus. (0) without a scabriculous region (Fig. 143) (or mediuncus absent); (1) with a scabriculous region (Fig. 135, sbr). Results: L = 3. State ( 1 ) is independently derived on lineage 1 5 and in Psychobiella (poorly developed), but is reversed (1 -0) in Austromegalomus . Character 83: Phatlolingua. (0) absent; (1) present. Comments: Nakahara ( 1 960b) coined the term phallolingua for a tubular, eversible, structure found in species of the genus Notiobiella. At rest the phallolingua is tele- scopically withdrawn below the mediuncus and above the parabaculum; and it ap- pears to be a modification of part of the gonosaccal membrane. Its external surface (when everted) is often extensively ornamented with small spicules. Results: L = 1. State (1) is an autapomorphy of Notiobiella. 186 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Parabaculum and Associated Structures (Characters 84-92) Character 84: Parabaculum (A). (0) absent; (1) present (Fig. 21). Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 2. Character 85: Parabaculum (B). (0) not deeply divided, or if deeply divided not as described for state (1) below (or parabaculum absent); ( 1 ) deeply (entirely or nearly entirely) divided, composed of a pair of adjacent, narrow sclerotized straps, which are enlarged posteriorly as small sclerotized surfaces surrounded by membrane, sagittal apophyseal lamella absent (Figs. 37, 48). Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 5. Character 86: Sagittal division of parabaculum. (0) terminal cleft complete, parabaculum comprised of a pair of laterally adjacent sclerites (Fig. 37); (1) terminal cleft incomplete (Fig. 22, tc) (or parabaculum absent); (2) terminal cleft absent, parabaculum with a single undivided terminal lobe (Fig. 128). Comments: Pre-analysis polarity hypothesis: 1—0; 1—2. Results: L = 2. State (0) is uniquely derived in Hemerobius\ state (2) is uniquely derived in Zachobiella. The pre-analysis hypotheses of polarity were corroborated. Character 87: Dorsal subapical spinose processes of parabaculum. (0) absent; (1) present, superposed over small terminal lobes (Figs. 95, 156). Comments: Spiniform processes derived from the terminal lobes of the paraba- culum are sporadically distributed throughout the Hemerobiidae. Such processes may be apical or subapical. Well-sclerotized distal processes found in genera such as Drepanepteryx, Wesmaelius , and Micromus, appear to represent simple, convergent, modifications of the terminal lobes (by elongation, attenuation, and increased scle- rotization). Distinctly spiniform subapical processes occur in Drepanacra , Psycho- biella , and some Megalomus. Only those of found in Drepanacra and Psychobiella were assumed homologues for this character. In these genera the processes arise from the dorsal margin of the terminal lobes and project posteriorly over small, terminal lobe remnants. In Megalomus , analogous processes arise ventrally or laterally from the terminal lobes. Megalomus is polymorphic for such processes and their absence is presumed plesiomorphic. Results: L = 2. State (1) is independently derived in Psychobiella and Drepanacra. This result suggests that the processes in Drepanacra and Psychobiella are not ho- mologous, as coded above. 1993 WORLD GENERA OF HEMEROBIIDAE 187 Character 88: Ventrodistal region of parabaculum. (0) not expanded as a pair of rounded bulbous lobes (or parabaculum absent); (1) expanded as a pair of rounded bulbous lobes (Fig. 1 12). Results: L = 1. State (1) is an autapomorphy of Psectra. Character 89; Bow in apophyseal shaft of parabaculum immediately anterior to ter- minal lobes. (0) absent; (1) present (Fig. 139). Results: L = 1. State (1) is an autapomorphy of Conchoptere/la. Character 90: Penniform sclerites. (0) absent; (1) present (Fig. 179, pen). Comments: Gayomyia falcata possesses a pair of penniform (narrow, arched, and toothed) sclerites attached dorsally to the parabaculum at the base of the terminal cleft. Results: L = 1. State (1) is an autapomorphy of Gayomyia. Character 9 1 : Supra-penniform sclerite. (0) absent; (1) present (Fig. 178, spn). Comments: The supra-penniform sclerite is a small triangular sclerite located dorsal to the penniform sclerites in Gayomyia. Results: L = 1. State (1) is an autapomorphy of Gayomyia. Character 92: Laterobacula. (0) absent; (1) present (Fig. 205, lbc). Comments: The laterobacula are a pair of small sclerites which flank the dorsal surface of the parabaculum in Nusalala. Results: L = 1. State (1) is an autapomorphy of Nusalala. Male Terminalia Miscellaneous (Characters 93-95) Character 93: Gonosaccal accessory sclerite. (0) absent; (1) present (Figs. 190, acs; 186). Comments: This sclerite is a small bilobed structure lying in the gonosaccal mem- brane above the parabaculum in Drepanepteryx. Results: L = 1. State (1) is an autapomorphy of Drepanepteryx. 188 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Character 94: Supragonopontal setal group. (0) absent; (1) present. Comments: This feature is a small group of generally well-developed setae which lie medially in the paragonosaccal membrane adjacent to the anterior margin of the extragonopons. The genus I Vesmaelius is polymorphic for this character; state (1) was used in the cladistic analysis. Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 4. Character 95: 9th gonocoxites. (0) present, articulating with or fused proximally to gonarcus (Fig. 29, 9gcx; 30); (1) absent [lost]; Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 3. For a discussion of this character see above under the heading Adult Morphology >: Ab- domen: Male Terminalia : Parabaculum. FEMALE TERMINALIA (Characters 96-107) Tergites and Stemites (Characters 96-98) Character 96: 7th Sternite. (0) posterior margin sagittally emarginate; (1) not divided and posterior margin not emarginate; (2) sternite completely divided sagittally (Fig. 121). Comments: Pre-analysis polarity hypothesis: 1—0; 1—2. The genus Psectra is poly- morphic for this character: state (1) was used in the cladistic analysis. Results: L = 2. State (0) is uniquely derived in the outgroup Psychopsis birmana: state (2) is uniquely derived in Anapsectra. The pre-analysis hypotheses of polarity were corroborated. Character 97: Ventral margins of 9th tergite. (0) without chalazate setae; (1) with chalazate setae (Fig. 118). Comments: The genus Micromus is polymorphic for this character; state (0) was used in the cladistic analysis. Results: L = 1. State (1) is an autapomorphy of Anapsectra. Character 98: Anteroventral lobes of 9th tergite. (0) absent (Fig. 52); (1) present (Fig. 64, avl). Comments: In Hemerobiella the anteroventral angle of the 9th tergite is promi- nently produced anteriorly as a broad, nonapodemal, lobe. Results: L = 1. State (1) is an autapomorphy of Hemerobiella. 1993 WORLD GENERA OF HEMEROBIIDAE 189 9th Gonocoxites (Characters 99-102) Character 99: 9th gonocoxites. (0) membranously separated from ipsilateral 9th tergite margins (Fig. 26, 9gcx); (1) fused with ipsilateral 9th tergite margins (Fig. 126, 9t + 9gcx). Comments: All hemerobiid females except those of Zachobiella possess free 9th gonocoxites, i.e., the gonocoxites are membranously separated from the adjacent posteroventral margins of the 9th tergite. New (1988, figs. 229, 237, 247) published the first illustrations of Zachobiella female terminalia and showed free gonocoxites to be present in the Australian species Zachobiella lobata, pallida and submarginata. However, in the four Zachobiella females examined for this analysis (including spec- imens of pallida and submarginata) the 9th gonocoxites were found to be fused to the adjacent ipsilateral margins of the 9th tergite. This results in a condition in which the gonocoxites appear to be lacking. This configuration should be verified in other species of Zachobiella. Results: L = 1. State (1) is an autapomorphy of Zachobiella. Character 100: Styli of 9th gonocoxites. (0) present (Fig. 26, sty); (1) absent [lost] (Fig. 35). Results: L = 5. State (1) is independently derived in lineages 4, 12, 21, Drepanep- teryx, and the outgroup Nothochrysa californica. The polarity of this character seems well established by outgroup comparison with other neuropterid taxa, particularly the Megaloptera and Raphidioptera, where styli are nearly universally present. How- ever, recognizable styli are present in only a few neuropterous families, e.g., Psy- chopsidae, Polystoechotidae, Osmylidae, Ithonidae, and Hemerobiidae, having ap- parently been lost in most other families. Within the Hemerobiidae, 13 of 25 genera possess styli, and styli have been lost independently in four lineages: ( 1 ) the subfamily Hemerobiinae, (2) the clade Psectra + Anapsectra + Zachobiella , (3) the genus Drepanepteryx, and (4) the subfamily Microminae. Character 101: Sclerotized sulcus dividing 9th gonocoxites. (0) absent; (1) present (Fig. 1 18, sul). Comments: The genera Neuronema (0), Noius (0), Notiobiella (0), and Psectra (0) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 1 . State ( 1 ) is an autapomorphy of Anapsectra. In females of Anapsectra medleri the 9th gonocoxites are partially joined by secondary sclerotization of their dividing membrane, the channel between them taking the form of a deep sclerotized sulcus. Character 102: Ventral lobes of 9th gonocoxites. (0) absent; (1) present (Fig. 118, vlb). 190 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Comments: The genera Hemerobius (0), Megalomus (0), and Micromus (0) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 1. State (1) is an autapomorphy of Anapsectra. Female Terminalia Miscellaneous (Characters 103-107) Character 103: Gonapophyses posteriores. (0) present; (1) absent [lost]. Comments. The term gonapophyses posteriores is applied here to the pair of small, concave, sclerites which border or enclose the apex of the female subgenitale. These structures tend to be poorly developed or absent when the subgenitale is reduced or absent. The genera Megalomus (0). Micromus (0), Notiobiella (0). and Sympherobius (0) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 5. State (1) is independently derived on lineages 5, 12. Carobius , Nomerobius , and Noius. Character 104: Insemination-fertilization canal. (0) slit-entry' type; ( 1 ) pore-entry type. Comments: All hemerobiid females possess an elongate, generally well sclerotized, duct joining the bursa to the median oviduct. This duct is here called the insemi- nation-fertilization canal. During insemination, sperm are presumed to pass from the male’s spermatophore (implanted in the female’s bursa during copulation) into the canal, where they are temporarily stored (frequently in an enlarged “spermathe- cal” region) before passing to the oviduct, where fertilization occurs. The presence, or likely presence, of similar canals of equivalent function have been reported in most other families of Neuroptera (e.g., Chrysopidae [Philippe, 1972: fig. 5. “canal de la spermatheque” + “spermatheque” + the short duct between the bursa and spermatheca]; Osmylidae [Adams, 1969: fig. 6. “fertilization canal”]; Mantispidae [Lambkin, 1986: figs. 30-31, “spermatheca” + “fertilization canal"]) and Raphi- dioptera (Raphidiidae [Achtelig, 1978: fig. 6, "ductus receptaculi seminis” + “re- ceptaculum seminis” + “ductus seminalis”). The hemerobiid insemination-fertilization canal differs from those found in the outgroups examined in the configuration of the entrance to the canal. In all heme- robiids the canal entrance is a small circular opening located at or near the anterior end of the bursa. In the outgroups, the entrance to the canal is by means of an elongate dorsal slit. It seems likely that the “slit-entry” canal type arose from a plesiomorph- ically membranous bursa in which a narrow midventral strip became secondarily sclerotized; subsequent revolution and opposition of the lateral margins of the scler- otized strip could create an elongate hollow cleft possessing a narrow dorsal slit ringed by bursal membrane. From this “slit-entry ” configuration, the “pore-entry” config- uration found in hemerobiids could be easily derived through progressive closure of one end of the slit. Such closure would simultaneously account for the small size of 1993 WORLD GENERA OF HEMEROBI1DAE 191 the opening into, and the elongate form of, hemerobiid insemination-fertilization canals. The comparative morphology of the neuropterous insemination-fertilization canal has not been well studied. A detailed survey of females of representative neuropterous families is needed to clarify homologies and standardize terminological conventions for this structure. Particularly problematic is the term “spermatheca”, which has been (1) loosely applied to any sclerotized portion of the insemination-fertilization canal (e.g., many Neuroptera), (2) restricted to some prominently enlarged portion of the canal (if one is present) (e.g., Chrysopidae, some Hemerobiidae), or (3) applied to a pair of bulbs each joined to the canal by an elongate duct (Osmylidae). All these structures are not homologous and should not bear the same name. Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 2. Character 105: Subgenitale. (0) present; (1) absent [lost]. Comments: When present, the female subgenitale is generally a small, apically bilobed, sclerite arising from the medioventral body wall between the ventrally ex- panded lobes of the 9th tergite. The subgenitale is probably a remnant of the female 8th sternite. The genera Megalomus (0), and Micromus (0) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 2. State (1) is independently derived in lineages 5 and 12. Character 106: Sclerotizations of female bursa. (0) bursa membranous, without sclerotized areas; (1) bursa broadly but weakly sclerotized dorsally; (2) bursa with lateral walls prominently sclerotized; (3) bursa with a sclerotized dorsal arch transversely connecting lateral margins of 8th sternite. Comments: The states of this character were analyzed as nonadditive. Secondary sclerotizations of the female bursa are rare in hemerobiids. The genera Micromus (0), Neuronema (0), and Notiobiella (0) are polymorphic for this character. The states used in the cladistic analysis for these genera are given in parentheses. Results: L = 3. State ( 1 ) is uniquely derived (0 - 1 ) in Gayomyia ; state (2) is uniquely derived (0 — 2) in Nusalala ; state (3) is uniquely derived (0 — 3) in the outgroup Po- lystoechotes punctatus. Character 107: Distal diverticulum of insemination-fertilization canal. (0) absent; (1) present. Comments: In many Microminae the distal end of the insemination-fertilization canal possesses a distinct constriction, setting off a small diverticular lobe. Results: L = 1. State (1) is a unique, unreversed, synapomorphy of lineage 23. 192 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) II “ _ O'! ^ „ O eg so = E O' , o t 5/3 CJ u > CJ CJ .2 « 11 So oo c3 2 ■o J S .5 E o 5 -5 f-s£ O •— o 00 > — 3 y (« o C > r-* o II t/3 c c— w c u ^ 2 > ; c5 2 # cj c 3 C/3 W *o c c CJ CS 00 - o £ > C3 o g> » ■§ « u T3 .2 £ “ C3 c3 “afeScS^sa C/3 c-» 2 '-J „ c/3 W CJ te 3 £ U - c *— 2 5 & e -5 ii a 2 -o 3 Sb £ 6 g O C3 §■ ob -g .2 o = - -o £ •- U c OC J= v ■- So O £ g ^ O 5 rt ^ - O ^ CJ I '5 w i E | 0 £ 15 r; •— CJ > -n £ O e i — C Ti n y C H.- s „-a r ~0 - c o 3 ~ o X) £ % 3 X3 C/3 a ■- a x: x: £ H .2 C CJ w *3 S2 ^ ? > 1 •o S £ u £ &■ g “ Z m — C! •— ob E s _r h: o e ’o o -u 4> X! 2^ lc J5 00 g c - E -c E § 1 00 CJ a CJ 00 a o *3 o= o cj i2 2 C ■£ ° O C ^ CJ > o CJ c £ - a — i a § 00 CJ o sz E -3 C/3 T 3 -2 O £ cj a > — C O cj 00 00 — he *3 ^ O O t z £ CJ O a r"- o ^ ■ 00 U- so ■C £ 2 E c > C3 > •tr - £i . o a ~o * ~0 73 x C e a 22 u ra <3 » u o r „ o 3 -O E -2 n u »- "O oO - XJ CJ O CJ C3 ■- O CJ C3 O CJ C/3 tr o i: 3 o 1993 WORLD GENERA OF HEMEROBIIDAE 193 V) o> ■ □ 74,0-1 3U 78.1-2 2 3(cont.) 194 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Results Cladograms Two minimal length tree topologies were identified by the HENNIG86 analysis (L = 196, Cl = .61, RI = .70). The Nelson consensus tree of these topologies was identical to one of the two original trees. This tree is illustrated in Fig. 23, and is adopted as the preferred cladogram of hemerobiid intergeneric relationships. The lineage numbers cited in the following sections refer to the lineage reference numbers given in Fig. 23. The second original tree differs from the preferred tree only by its dichotomous resolution of the trichotomy formed by lineages 4, 7, and 9 on the preferred tree, though this resolution is very tenuous. The two tree topologies differ only due to two different equally parsimonious optimizations for character 34 (see Appendix 3). This character has a length of five steps, with three steps the same for both optimizations: 34(0 ^ 1 ) on lineages 21. Gayomyia , and Notiobiella. The other two steps differ in the two optimizations. In optimization 1 the transformation 34(0— 1) occurs also on lineages 4 and Carobius; in optimization 2 the transformation 34(0—1) would occur on lineage 2 and be reversed [34(1—0)] on a lineage uniting lineages 7 and 9. but not 4. The tree specified by optimization 1 is preferred here. This choice appears conservative in preferring a trichotomous resolution of clades 4, 7, and 9, over a fully dichotomous resolution supported only by a single trans- formation of character 34. which post-analysis evaluation suggests consists of several nonhomologous character states (see Results under this character above). Characters The fit of each character to the preferred cladogram is given under the heading Results in the appropriate character treatment above. These fits are summarized by- character in Appendix 3. and by lineage in Appendix 4. Classification A new generic and subfamily classification of the Hemerobiidae is proposed here, based on the holophyletic groups of genera identified by the cladistic analysis. This classification is given in full in Appendix 1 . Its subfamily taxa are compared to earlier hemerobiid classifications in Table 4. This new classification was constructed using the conventions recommended by Wiley (1981) for annotated Linnaean classifica- tions. The most important of these conventions is the adoption of a phyletic se- quencing scheme. Phyletic sequencing is a method for converting cladograms into classifications. Using this method, one or more pectinately branched cladogram lineages are selected as "axes" for the classification. The sequential branches originating along these “axes” are named and given equal categorical rank. A sequential listing of taxa (a classifi- cation) formulated in this manner has the following desirable properties: (1) each named taxon is holophyletic; (2) each named taxon which is linked to the “axis" at a dichotomous node is the sister-group of the collection of taxa of equal rank below it. The phrase sedis mutabilis , sequence interchangeable, is used to identify taxa stemming from polytomous nodes. Adjacent taxa in the classification which bear 1993 WORLD GENERA OF HEMEROBIIDAE 195 Table 4. Family-group classifications of the Hemerobiidae proposed since 1916. Comstock (1918) Kruger (1922) Navas ([1925], 1933) Nakahara (1960) Present classification — — — — Carobiinae Hemerobiidae Hemerobiinae Hemerobiini Hemerobiinae Hemerobiinae, s.m.' Sympherobiidae Sympherobiinae Sympherobiini - Sympherobiinae, s.m. — — — — Psychobiellinae — — — Notiobiellinae Notiobiellinae — — — — Drepanacrinae — Megalominae Megalomini - Megalominae - Drepanepter- Drepanopter- - Drepanepteryginae yginae ygini — Microminae Micromini — Microminae - - Nyrmini - - 1 s.m.: sedis mulabilis, sequence interchangeable. this annotation are of interchangeable position in the classification, since the sister- group relationships among such taxa are unresolved. The “axis” selected as the basis of the subfamily classification proposed here was chosen to balance the conflicting desiderata of: (1) maximizing the number of sub- family taxa recognized, thereby increasing the specificity of the classification, and (2) minimizing the number of new family-group names to be proposed. The chosen lineage— composed of preferred cladogram lineages 2, 3, 9, 10, 14, 17, and 18 — bears 9 pectinate branches. Six of these possess available family-group names (Hemero- biinae, Sympherobiinae, Notiobiellinae, Megalominae, Drepanepteryginae, and Mi- crominae); new names are proposed here for the other three (Carobiinae, Psycho- biellinae, and Drepanacrinae). Formal names are not proposed at this time for the two clades within the Hemerobiinae (Hemerobius + Nesobiella , and Wesmaelius + Hemerobiella)\ but these groups may be informally called, respectively, the Heme- robius and Wesmaelius genus-groups. Discussion Numbers enclosed in brackets in the following discussion are character numbers keyed to the character treatments above. Holophyly of the Hemerobiidae Tillyard (1916) appears to be the only previous author to have attempted to differentiate critically the Hemerobiidae from other neuropterous families on the basis of explicitly defined characters. Other authors have described or characterized the family, but they have not fully differentiated it from other families. The 10 characters forming the basis of Tillyard’s concept of the Hemerobiidae are reviewed below, followed by an assessment of the putative synapomorphies of the family proposed here. 196 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Character 1: “The presence of more than one radial sector in the forewing”. In the terminology used here, this statement is equivalent to “anterior radial trace with two or more ORB's”. This character is clearly apomorphic within the Hemerobiidae; however, it is not synapomorphic at the level of the entire family. Two different forewing radial vein configurations characterized by two ORB’s appear to have evolved independently in the Hemerobiidae, one in Carobius, and one in its sister-group, lineage 3. Character 2: "The absence of unspecialised cross-veins”. Tillyard observed that hemerobiids possess fewer forewing crossveins than most other families of Neurop- tera. e.g.. Ithonidae, Dilaridae. Osmylidae, and the myrmeleontoid families. The latter families possess crossveins which are both more numerous, and of more in- determinate wing position. Although hemerobiids possess fewer, and by implication more “specialised”, forewing crossveins than many neuropterous families, the phy- logenetic implications, if any, of this trait are difficult to assess, particularly since Tillyard proposed no specific crossvein homologies which could be subjected to testing. Other neuropterous families possessing relatively few forewing crossveins include the Coniopterygidae, Sisyridae, and Berothidae. Character 3: “The presence of at least one false or secondary origin for the radial sector in the hindwing”. Tillyard (1916: text-fig. 4) misinterpreted the sigmoid vein (Fig. 175. sv) which joins the hind wing M to the Rs as the basal segment of the Rs. This vein is now generally interpreted as the MA. Under the latter interpretation, adopted here, the true Rs is (as expected) the most proximal branch of the R, and does not originate on the M. Thus. Tillyard's first false origin of the Rs is actually its true origin. Since this Rs configuration is plesiomorphic within the pterygote Insecta. it cannot justify the holophyly of the Hemerobiidae. Character 4: “The presence. . .of a coupling apparatus at the base of the wings”. Tillyard believed that the jugal lobe of the forewing and the “jugal process” of the anteroproximal aspect of the hind wing (see Tjeder, 1961. figs. 460-466. “frenulum”) functioned as a coupling mechanism in flight. To the best of my knowledge, such a function for these structures remains unconfirmed. Furthermore, as noted by Tillyard. similar putative coupling mechanisms are present in other neuropterous families, and thus would not appear to constitute a synapomorphy at the level of the Hem- erobiidae. Character 5: “The absence of any distal fusion between Sc and R”. Fusion of the Sc and R is clearly derived within the Neuroptera. Lack of fusion, which is char- acteristic of the Hemerobiidae. is plesiomorphic, and therefore not evidence for hemerobiid holophyly. Character 6: “The archaic, unspecialised form of Rs”. Tillyard misinterpreted the collective homology of the Rs. He assumed that only the most distal prestigmal branch of the anterior radial trace (which is itself typically pectinately branched) represented the homologue of the Rs of other neuropterous families (Tillyard, 1916: 276). Under this interpretation, hemerobiids, and all other neuropterous families, have an “unspecialized" pectinately branched Rs. This interpretation of the forewing Rs is considerably at odds with the interpretation adopted here (see Characters 29- 33 above). All ORB's in aggregate are treated here as veins of ancestral Rs origin, and the retention of a pectinately branched most distal prestigmal ORB is interpreted as a plesiomorphic remnant of a once more extensively branched Rs. This character, therefore, cannot justify the holophyly of the Hemerobiidae. 1993 WORLD GENERA OF HEMEROBIIDAE 197 Character 7: “The retention of the archaic branchings of the veins as they approach the wing-border”. This character refers to the ’marginal twigging' of hemerobiid longitudinal veins. Since similar states are present in most other families of Neu- roptera, as was well known to Tillvard, ’marginal twigging’ is certainly plesiomorphic at the level of the Hemerobiidae, and is not evidence for the holophyly of the family. Character 8: “The fusion of M with R basally in the forewing, and the fusion of M in the hindwing with the weakly formed, original, basal portion of Rs”. As noted above (see Characters 29-33), Tillyard’s interpretations of the forewing R and hind wing Rs are here regarded as erroneous. Consequently, this character is not relevant to the question of hemerobiid holophyly. Character 9: “The unspecialised form of the antennae”. The momliform character of hemerobiid antennae was considered by Tillyard to be “unspecialized” (i.e., ple- siomorphic), relative to antennal configurations in the families Psychopsidae, Chry- sopidae, and Nymphidae. There is currently no reason to doubt this polarity; however, as a plesiomorphic trait, it cannot be used to support the holophyly of the family. Character 10: “Position of [the wings at] rest”. Tillyard proposed this character to distinguish hemerobiids, in which the wings are held at rest in a nearly vertical roof- like orientation over the thorax and abdomen, from the Psychopsidae and Ithonidae, in which the wings are oriented at rest in a more horizontal plane. The orientation found in Hemerobiidae is widespread within the Neuroptera and is certainly ple- siomorphic at the level of the Hemerobiidae. In summary, I conclude that none of the 10 traits used by Tillyard to characterize the Hemerobiidae is synapomorphic at the level of this family. In view of this, it seems remarkable that Tillyard's concept of the family is essentially identical to that adopted in this work. The present cladistic analysis strongly supports the holophyly of the family on the basis of eight entirely different characters (Fig. 23). The most significant of these are the presence of: penicilliform sensilla on the galea [6]; prom- inent paired setae on the clypeus [13, 16, 17]; a parabaculum in the male genital complex [84]; and a pore-entry type of female insemination-fertilization canal [104], Notes on Other Clades Carobiinae: This subfamily contains only the Australian genus Carotins. This genus is the sister-group to the rest of the Hemerobiidae, and occupies an isolated position within the family. The holophyly of Carobius is justified on the basis of five characters [30, 34, 39, 44, 103], three of which [34, 44, 103] lend only weak support to this clade, since they each occur independently four or five times within the Hemerobiidae. The most convincing synapomorphy of the genus is the configuration of the forewing radius [30], which is unique to Carobius (Fig. 24). Species of Carobius are also the only hemerobiids which retain, plesiomorphically, the 9th gonocoxites as clearly recognizable elements of the male genitalia. Lineage 3: This clade is well supported by five characters [7, 8, 10, 29, 95], The subsegmentation of the distal segment of both the labial and maxillary palpi [7, 8], the altered configuration of the radius [29], and the loss of 9th gonocoxites in the male [95] are the most significant characters. Hemerobiinae (Lineage 4): This grouping is justified on the basis of five characters [11, 32, 34, 94, 100], three of which [11, 34, 100] occur independently three to five times within the Hemerobiidae. The most convincing synapomorphies of the group 198 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) are the presence of a distinctive group of macrotrichia, the supragonopontal setal group, located medially on the paragonosaccal membrane above the gonopons [94]; and the configuration of the typically three forewing “radial sectors” [32], Heme- robiine species with four or five “radial sectors” are known, but are uncommon and clearly derived within the subfamily. Sympherobiinae (Lineage 7); The Sympherobiinae is justified on the basis of four characters [12, 40, 53, 79]. The presence of a pseudomediuncus associated with the male gonarcus [79], and the elongated postero ventral angle of the male 9th tergite [53], are particularly good synapomorphies of this clade. Lineage 9: The single putative synapomorphy of this clade is the presence of forewing crossvein 2sc-r [28]. Although this character involves the presence/absence of a wing crossvein, a character type which has historically been accorded low weight in Neuropteran taxonomy, this crossvein exhibits minimal intraspecific and intra- generic variation, and appears to be a phylogenetically significant trait. Within the clade, this crossvein is prominent in all taxa except (1) Noius, where it is present but vestigial, (2) lineage 22 ( Nusalala + Megalomina + Micromus ) and some Zacho- biella , where it is secondarily lost, and (3) Notiobiella, where it is secondarily short- ened or lost due to the extreme narrowing of the subcostal space. As one of the relatively few venational traits synapomorphic on the principal “classificatory axis” of the family, this character could prove important in assessing the phylogenetic positions of fossil hemerobiids. which are known almost entirely from preserved wings. Psychobiellinae: This subfamily contains only the Australian genus Psychobiella, whose holophyly is justified on the basis of seven characters [12, 17, 31, 51, 77, 82, 87], Of these, only the proximal transverse costa of the ventral surface of the male mediuncus [77] is unique to Psychobiella , although five of the six remaining characters occur only twice within the Hemerobiidae. It is interesting to note that four of the latter characters [31, 51, 82, 87] co-occur in the Drepanacrinae or on the lineages between the Psychobiellinae and Drepanacrinae. This suggests the possibility of a closer relationship between these two subfamilies than is indicated in the current analysis. Lineage 10: Of the two characters supporting this clade [74, 78], the opening of a membranous gonofenestra at the base of the mediuncus [74] appears most significant, although this structure has apparently been independently and secondarily lost several times within this clade. Notiobiellinae (Lineage 1 1): This clade is supported by three homoplasious char- acters [40, 42, 45], The observations (1) that all three of these characters involve the loss of individual forewing crossveins, and (2) that these losses co-occur in the apparently distantly related genus Sympherobius , cast some doubt upon the robust- ness of the grouping of Notiobiella with the clade Psectra + Anapsectra + Zachobiella. The holophyly of the latter clade is, however, particularly well supported. Lineage 1 4: This clade is supported by three characters [31,33, 48], of which, those involving alteration in the configuration of the radius [31,33] appear most significant. Drepanacrinae: This grouping is defined on the basis of two homoplasious char- acters [49, 82], One of these, a scabriculous membrane surface on the venter of the mediuncus [82] is also characteristic of the Psychobiellinae, and is part of a complex of characters suggesting a closer relationship between these two subfamilies than is indicated by the present analysis. The second character, a deeply forked forewing A 1 1993 WORLD GENERA OF HEMEROBIIDAE 199 [49], occurs twice within the Hemerobiidae at the level of this generic analysis, but also occurs sporadically in other genera. This character may be partially dependent on wing size, with larger wings showing more deeply forked A1 veins. Lineages 17 and 18: These two clades are relatively weakly supported. Lineage 17 isjustified solely on the basis of the presumed secondary fusion of the two subsegments of the ultimate labial and maxillary palpomeres [7, 8], The separation of the Me- galominae from the Drepanepteryginae + Microminae, Lineage 18, is even more tenuous, being supported only by a change in the distal shape of the mediuncus [18]. Based on the length of this character [L = 7] on the preferred cladogram, this feature appears to be especially plastic within the Hemerobiidae; and it may not provide convincing phylogenetic evidence at the present level of analysis. Megalominae: This subfamily contains only the genus Megalomus, whose holo- phyly isjustified by two unique synapomorphies [5, 78]. The male genitalia oi Me- galomus, particularly the convoluted mediuncus, are distinctive (Figs. 1 59, 162). The holophyly of this genus seems well established. Drepanepteryginae: This clade isjustified on the basis of three characters [1, 41, 47], two of which [1, 47] are homoplasious within the Hemerobiidae. The members of this group are all large hemerobiids, which may account for the presence of forewing 2im [41] and lcua-cup [47] crossveins in these taxa. This grouping should be reex- amined after a more comprehensive survey of the very poorly supported genus Neuronema, the putative sister-group to other drepanepterygines, has been under- taken. Microminae: This grouping is supported by eight characters [2, 3, 9, 23, 24, 25, 34, 100]. On the simple basis of the number of supporting characters, the Microminae is the best supported subfamily in this analysis; however, all eight characters are either paralleled elsewhere in the Hemerobiidae, or are reversed within the Microm- inae. The subordinate grouping Nusalala + Megalomina + Micromus is almost certainly holophyletic; however, the relationships among these three genera require further analysis. I was not able to identify any synapomorphies for the large, cos- mopolitan, genus Micromus, and I regard as very tentative the cladistic relationships shown in Fig. 23 between it and the genera Nusalala and Megalomina. In particular, the possibility that Nusalala may be a derived clade falling within the broad Micromus concept accepted here should be investigated. The sister-group to these three genera, Noius, appears distantly related both on the basis of venational and terminalic char- acters. Posterior Processes of the Male Extragonarcus The establishment of homologies among the posterior processes of the hemerobiid male extragonarcus has been an especially difficult problem; one which is only par- tially resolved here. Previous authors have applied a variety of terms to these pro- cesses (e.g., aedeagus, arcessus, entoprocesses, epimeres, mediuncus). But, uncer- tainties about homologies have often prevented critical application of these terms. Due to this uncertainty, extragonarcal process characters, with the exception of the mediuncus, were excluded from the present cladistic analysis. The discussion below is a retrospective look at the question of the homologies of these processes, based on the intergeneric relationships reflected in the preferred cladogram. Carobiinae: In Carobius, the sister-group to all other hemerobiids, three extragonar- 200 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) cal processes are typically present: ( 1 ) an unpaired dorsomedian process, the med- iuncus (Fig. 28. med), and (2) a pair of ventrolateral processes called here the par- amediunci (Fig. 28, pmed) [Etymology: para-, from Greek para , near, + mediuncus; pleural, paramediunci]. The male terminalic figures given by New (1988) show that this generalized arrangement is found in eight of the nine known species of Carobius. One species. C. elongatus , apparently lacks a mediuncus. This tripartite extragonarcal configuration is assumed here to be plesiomorphic within the Hemerobiidae. Hemerobiinae: Both the mediuncus and paramediunci are retained in Nesobiella and Wesmaelius. In Wesmaelius a prominent mediuncus (Fig. 56, med) is present, flanked by a pair of paramediunci (Fig. 56, pmed) which have retained their ple- siomorphic ventrolateral positions. In Nesobiella the paramediunci (Fig. 43, pmed) have converged medially to become partially fused sagittally, but the mediuncal remnant is distinctly visible as an acute cusp dividing the bases of the paramediunci. In Hemerobius the paramediunci (Fig. 39) have assumed the form of a pair of ventrally directed appendages, which are generally only narrowly attached to the posterior margin of the extragonarcus. In many Hemerobius species the paramediunci are medially convergent (not uncommonly partially fused sagittally), and the mediuncus is generally vestigial or lost. In Hemerobiella (Fig. 62), a simple but prominent mediuncus has been retained, but the paramediunci are entirely lost. Sympherobiinae: In Nomerobius (Figs. 72-73) the paramediunci are present as a pair of small posteriorly directed lobes which are flexibly associated with the lateral margins of a small, triangular mediuncus. In Neosympherobius (Fig. 76) and Sym- pherobius (Fig. 84) both the mediuncus and paramediunci are entirely lost. In all three of these genera, but particularly in Sympherobius, a pseudomediuncus (a neo- gonarcal. not extragonarcal, structure) is present projecting from the gonopons. In the remaining six subfamilies (Psychobiellinae, Notiobiellinae, Drepanacrinae, Megalominae, Drepanepteryginae, and Microminae), only the homology of the med- iuncus is here regarded as well established. Detailed assessments of the homologies of other extrahemigonarcal processes present in these subfamilies must await more comprehensive analyses of individual genera and subfamilies. In recent years the term entoprocess (-es) has been used extensively as a term of convenience to designate any paired posterior process of unknown homology on the neuropterous gonarcus. Until homologies (and appropriate new names) for paired extragonarcal processes can be established in these six subfamilies, I recommend that this term, or the phrase “extrahemigonarcal process(es)”, be applied to such structures. Use of the term “entoprocess” should be accompanied by a statement indicating that no homologies are to be inferred among the structures so designated. I also recommend that the term arcessus be abandoned in the Hemerobiidae. Most structures given this designation in prior literature are mediuncus homologs. Tjeder (1931:4) proposed the term mediuncus for the unpaired, posteriorly directed, dor- somedian, process of the extragonarcus in several species of Wesmaelius (as Borio- myia). This term has subsequently been very widely applied to processes of similar location throughout the Neuroptera. Tjeder (1936:7) proposed the term arcessus for the small sclerotized structure located below the “mediuncus” in some Neuronema species. Unfortunately, the membranous attachments to the gonarcus in Neuronema clearly reveal that Tjeder’s “mediuncus” and “arcessus” in Neuronema are in fact the neogonarcus and mediuncus respectively. Thus. Tjeder’s term “arcessus”, in its 1993 WORLD GENERA OF HEMEROBI1DAE 201 original sense, is a synonym of “mediuncus”. Similar misinterpretations have also been made for some Micromus species which possess well-developed neogonarcus (e.g., perelegans: Tjeder, 1936: pi. 10, figs, f, h; africanus : Tjeder, 1961:322, figs. 519- 520). Although the homologies of extragonarcus processes remain problematic in the remaining six subfamilies (Psychobiellinae, Notiobiellinae, Drepanacrinae, Mega- lominae, Drepanepteryginae, and Microminae), three other gonarcal developments present in some of these taxa deserve mention here. First, the mediuncus in these subfamilies is often modified as a broad, elongate, dorsal plate, which supports an extensive gonosaccal membrane. This contrasts with the generally small mediunci of the three preceding subfamilies. Usually associated with such mediuncal enlarge- ments are structural modifications which facilitate dorsoventral motion of the distal end of the mediuncus. Second, neogonarcal sclerites are extensively developed in some genera of these subfamilies (e.g., some Megalomus , Neuronema, and Micro- mus). The addition of a prominent neogonarcus in some of these taxa has resulted in considerable alteration in the overall appearance of the gonarcus. Third, in nearly all genera of these subfamilies, a small membranous “window” is present dorso- medially at the base of the mediuncus. This structure, named here the gonofenestra (Fig. 1 53, gfa) [Etymology: Greek gono. seed, + Latin fenestra , window], is interpreted as a secondarily desclerotized area at the base of the mediuncus, developed, perhaps, as one mechanism to facilitate movement of the mediuncus. Psychobiellinae: The gonarcal configuration in Psychobiella is highly modified. In P. sordida (Figs. 96-97), much of the extragonarcus is enclosed within a helmet-like neogonarcus. The mediuncus is prominently produced; and the paramediunci are apparently lost. The extrahemigonarcus are deeply cleft laterally along a pair of lines which begin near the hemigonarcal conjunctions and follow the trend of the antex- tragonarcal commissure. These clefts result in the production of a pair (one above each cleft) of flat extragonarcal lobes which flank the mediuncus. No gonofenestra is present. Notiobiellinae: Gonofenestra present or absent. In Notiobiella (Figs. 100-101) the mediuncus is prominent and the posterolateral margins of the extragonarcus fre- quently bear one or more processes of uncertain homology. The posteroventral angles of the extrahemigonarcus (=?paramediunci) are frequently produced to support the eversible phallolingua. In Psectra the mediuncus may be distinct (Figs. 1 13-1 14) or vestigial; lateral extragonarcal processes are absent. In Anapsectra medleri (Figs. 1 19- 1 20) the mediuncus is distinctly tricuspate; lateral extragonarcal processes are absent. In Zachobiella (Figs. 1 30-13 1 ) the mediuncus is distinct and divided apically; lateral extragonarcal processes are sometimes present. Drepanacrinae: Gonofenestra always prominent. In all three genera, Conchopterella (Figs. 1 34-1 35), Austromegalomus (Figs. 142-143), and Drepanacra (Figs. 153-154), the mediuncus forms a large, dorsal piate which is apically emarginate. The pos- teroventral margins of the extragonarcus are variously lobed. Megalominae: Gonofenestra present. In Megalomus the mediuncus (Fig. 1 62, med) is distinctively modified. It’s proximal portion is directed sharply downward at the gonopons, and its distal portion is bent along a transverse flexion line to project posteriorly. The distal portion is composed of a pair of filiform processes divided by a deep sagittal emargination. From the proximolateral angles of the base of the 202 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) mediuncus arise a pair of long narrow entoprocesses (Fig. 162, ent), the “epimeres” of Carpenter (1940). The homology of these processes remains uncertain, but they may represent a pair of ancestrally lateral extragonarcal processes, that have become secondarily associated with the base of the mediuncus. The posteroventral corners of the extragonarcus are sometimes produced. Neogonarcal sclerotization is often well developed in Megalomus. particularly in the South American species flint i and nigratus , a factor which has undoubtedly facilitated the observed general reduction in the prominence of the extragonarcus. Drepanepteryginae: In Neuronema the structure of the paleogonarcus (Fig. 170, pgs) is highly variable. The mediuncus (Fig. 170, med) may be well developed as a large, posteriorly directed plate or reduced as a small, ventrally directed lobe, but it is generally only membranously associated with the remainder of the gonarcus. The extragonarcus is poorly developed or absent ventrally, but is usually developed as a pair of fixed lobes (frequently large) dorsally. A neogonopons (Fig. 170, ngs) is often present over the base of the mediuncus. A gonofenestra is present. In Gayomyia (Figs. 176-177) the mediuncus is prominent and laterally continuous with the ex- trahemigonarcus. Situated laterally near the base of the mediuncus are a pair of elongate, movable processes of uncertain homology which articulate with the pos- terior margins of the extragonarcus; the neogonarcus and gonofenestra are absent. In Drepanepteryx (Figs. 184-185) the mediuncus is prominent and flanked by a pair of stout extragonarcal processes; again, the neogonarcus and gonofenestra are absent. Microminae: In Noius (Figs. 195-196) a large gonofenestra is present and the extrahemigonarcus are reduced to a pair narrow sclerotized straps which join the mediuncus to the gonarcus arch. The extragonopons is represented by a pair of plates divided by a sagittal incision. In Nusalala (Figs. 201—202) the mediuncus is flanked by two pairs of processes, an inner pair associated with the proximolateral angles of the mediuncus, and an outer pair extending as posteriorly directed outgrowths of the dorsal portion of the extrahemigonarcus. A gonofenestra is present. The extrahemigo- narcus are generally well developed, but are frequently partially overlain by extensive but inconspicuous neogonarcal plates. In Megalomina (Figs. 211-212) the mediuncus is prominent, generally emarginate distally, and laterally continuous with the extra- hemigonarcus. The posterolateral margins of the extrahemigonarcus are produced as inwardly revolute plates. The gonofenestra is absent. A neogonopons (Fig. 211, ngs) may be developed over the base of the mediuncus. In Micromus (Figs. 219-220) the mediuncus is always prominent, but the relative development of the extrahemigo- narcus and neogonarcus are extremely variable. In some species the mediuncus is broadly connected at its base to well-developed lateral extrahemigonarcus (presum- ably the plesiomorphic condition). In other taxa the extrahemigonarcus are largely or entirely lost, their structural functions apparently having been taken over by neogonarcal sclerites. Posterior processes of the gonarcus are particularly diverse in Micromus. However, many of these arise from the posterior margin of the neogo- narcus (e.g.. the dorsomedian processes in variegatus [Fig. 220, ngsp] and mullipunc- tatus , and the paired lateral processes in dissimilis, numerosus , and timidus). not the extragonarcus. Thus, these processes are not homologous with other extragonarcal processes in Micromus or other genera. In Micromus , true extrahemigonarcal pro- cesses generally take the form of broad posterior lobes rather that acute prominences. Well-developed shelf-like neogonopontes are present in many Micromus species. 1993 WORLD GENERA OF HEMEROBIIDAE 203 TAXONOMIC HISTORY AND EARLY CLASSIFICATIONS The modern era of hemerobiid taxonomy began with Tillyard’s 1916 revision of the Australian hemerobiid fauna, in which he formulated the first modern concept of the family. Pre-Tillyardian concepts, from Latreille (1802) to Handlirsch (1906- 1908), and the early formulations of Banks (1892, 1907, 1913), were considerably broader in scope and are not discussed here. Since Tillyard’s seminal work, four subsequent authors have advanced proposals to subdivide the Hemerobiidae. Each of these is briefly reviewed below. Table 4 summarizes these classifications, and contrasts them with the subfamily classification used here. The latter classification is presented in full in Appendix 1, and is substantiated above under the heading Cladistic Analysis: Results: Classification. Based on his extensive studies of wing venation, Comstock (1918) proposed a division of the Hemerobiidae into two families: (1) the Sympherobiidae, to include Sympherobius. Psectra, Notiobiella and other (unnamed) genera possessing only two “radial sectors”, and (2) the Hemerobiidae, to include all genera possessing three or more “radial sectors”. Although a similar division was proposed at lower rank in the subsequent classifications of Kruger and Navas, very few later workers (e.g., Brues and Melander, 1932; Kuwayama, 1962) followed Comstock in recognizing the Sym- pherobiidae at family rank. Comstock’s concept of the Sympherobiidae is rejected here because it is not a holophyletic grouping. The preferred cladogram (Fig. 23) shows that a suprageneric hemerobiid taxon based solely on the dichotomy 2 vs. 3 + “radial sectors” (i.e., a grouping of Carobiinae + Sympherobiinae + Notiobiellinae vs. the remaining genera) would be clearly polyphyletic. Kruger (1922) proposed a five subfamily classification based on another detailed venational study. Several pieces of evidence suggest that Kruger was unaware of Comstock’s prior work. First, Comstock’s work is not cited in Kruger’s paper; second, the genera Annandalia. Psectra , and Notiobiella , which were noted by Comstock are not mentioned by Kruger; and, third, Kruger cites himself (pp. 170-171) as the author of the names Hemerobiinae and Sympherobiinae. This evidence suggests that Kruger’s proposal of family-group names based on the type genera Hemerobius and Sympherobius was independent of Comstock’s proposal. Several factors have con- tributed to the rejection of Kruger’s classification by subsequent workers, principal among these are: (1) the eccentric format of Kruger’s paper, which has made the identification and critical assessment of any diagnostic subfamilial characters ex- tremely difficult, (2) the fact that an important intended companion paper was never published, and (3) the small number of genera upon which the classification was based. Furthermore, a comparison of Kruger’s subfamily concepts (as revealed by the taxonomic listings given by Kruger, 1922:170-172) against the preferred clado- gram (Fig. 23), suggests that at least his subfamilies Drepanepteryginae, Megalominae, and Hemerobiinae do not appear to be holophyletic. Navas ([ 1 925b]) divided the Hemerobiidae of the Iberian Peninsula into four tribes: Sympherobiini, Hemerobiini, Megalomini, and Micromini. It is not clear whether Navas was aware of Kruger’s earlier subfamily proposals, or whether he indepen- dently proposed his nearly identical tribal groupings. In 1933, however, Navas ex- plicitly recognized Kruger’s Drepanepteryginae at tribal rank (as Drepanopterygini [sic]), and also proposed a sixth tribe, Nyrmini, for the enigmatic genus Nyrma. The 204 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) latter tribe was recently removed to the family Berothidae as the subfamily Nyrminae (U. Aspock, 1989). Navas’s remaining subfamilies exhibit problems similar to those of Kruger’s classification, e.g., foundation on few— mostly European— genera, and implicit recognition of some groupings which are not holophyletic. Nakahara (1960b) proposed a division of the Hemerobiidae into two subfamilies: ( 1 ) the Notiobiellinae, for the genus Notiobiella (based on the presence of a distinctive derived genitalic feature, the phallolingua) and (2) the Hemerobiinae, for all other genera. Nakahara’s work is significant for its shift in emphasis from venational to male terminalic characters as the primary basis for hemerobiid family-group clas- sification, and as the first attempt to extensively survey hemerobiid diversity at the generic level beyond the confines of a regional or continental boundary. However, its classification is unsatisfactory for at least two reasons: (1) although the genus Notiobiella is certainly holophyletic, its elevation to subfamily status clearly renders Nakahara’s Hemerobiinae paraphyletic; (2) the recognition of only two subfamilies, one of which is monogeneric, leaves intergeneric relationships within the bulk of the family entirely unresolved. Shortly following the publication of Nakahara’s work. Tjeder (1961) criticized and rejected all previously proposed classifications of the Hemerobiidae. No critical de- fenses of earlier classifications have subsequently been advanced; and few works published following Tjeder’s influential paper have formally recognized family-group subdivisions within the Hemerobiidae. The new classification proposed here has been formulated with the deficiencies of earlier arrangements in mind. It’s principle advantages over older classifications are: (1) foundation on a most parsimonious cladistic hypothesis of intergeneric relation- ships, (2) recognition of only phylogenetically meaningful — demonstrably holophy- letic—taxa based on putative synapomorphies, (3) incorporation of the phyletic se- quencing convention, thereby allowing the estimated phylogeny upon which it is based to be retrieved from the classification, (4) expansion of the character base upon which the classification is based, and (5) inclusion of as many extant hemerobiid genera as practicable. Taxonomic Treatments Format. The following taxonomic treatments are divided into subheadings as follows (some treatments do not contain all subheadings): (1) Synonymical Listing. A listing of all supraspecific names which are considered synonyms of the treated taxon name, with relevant nomenclatural data and references. References cited as authoritative sources of subjective synonyms are generally recent revisionary works, not necessarily the first reference in which the synonymy was proposed. (2) Differential Diagnosis. A synoptic statement giving a character combination diagnostic for the treated taxon. (3) Proposed Synapomorphies. A listing of synapomorphic characters of the treated taxon (and unnamed subordinated clades, if any), as inferred from the cladistic analysis. Clades are identified by name or the lineage reference number given on the preferred cladogram (Fig. 23). Characters are identified by their numbers, in brackets [], and listed in numerical order. 1993 WORLD GENERA OF HEMEROBIIDAE 205 (4) Included Genera or higher laxa. A listing of family- and/or genus-group taxa included within each treated higher taxon. (5) Species. For genera containing 15 or fewer species, an alphabetical listing is given of all species and their known distributions. For larger genera, the ap- proximate number of valid species currently placed in the taxon is given; for more detailed species level data on these genera, the references cited under the heading Principal Revisions and Regional Faunas should be consulted. See also Appendix 2. (6) Distribution. A description of the known distribution of the treated taxon. (7) Principal [Revisions and] Regional Faunas. A reference list of revisional and faunal works which include data on species in the treated taxon. The geographic scope of each work is indicated in brackets. Generally, only the most recent faunal treatment for any area is cited. Some older works may be taxonomically out-of-date. Citations to faunal studies of individual European countries may be found in Aspock et al. (1980), and are not repeated here. (8) Preimaginal Stages. A reference list of literature pertaining to the preimaginal stages of the treated genus grouped by species. (9) Synonymical Notes. Justifications for newly proposed synonyms. (10) General Notes. Comments and discussion not conveniently treated under any other heading. Family Hemerobiidae Latreille, 1802 Hemerobini Latreille, 1802:288 (Type genus; Hemerobius Linnaeus). Hemerobida Leach in Brewster, 1815:138. Hemerobides Leach in Brewster, 1815:138. Hemerobiides Billberg, 1820:95. Hemerobidae Stephens, 1829:310. Hemerobioides Burmeister, 1829:22. Hemerobites Brulle, 1832:275. Hemerobiina Newman, 1835:380. Hemerobiites Newman, 1835:380. Hemerobii Ehrenberger, 1836:17. Hemerobiidae Westwood, [1838]:48. Hemerobiinae Swainson and Shuckard, 1840:356. Hemerobioidae Agassiz, [1842] 1842-1846:176. Hemerobina Schneider, 1851:35. Hemerobiidea Costa, 1855:1. Hemerobinae Handlirsch, [1906]:42. Hemerobiini Nakahara, 1915:13. Differential Diagnosis. Very small to small, generally brown or brownish, Neu- roptera. Forewing length 3-18 mm. Distinguished from other Neuroptera by the forewing character combination: (1) anterior radial trace bearing 2 or more (up to ca. 12) “radial sectors” and (2) nygmata absent. Proposed Synapomorphies. Hemerobiidae (Lineage 2): [1] Temporal costae well developed (sometimes secondarily reduced); [6] galeae bearing penicilliform sensilla; [13, 16, 17] clypeus bearing several pairs of homologizable primary setae; [35] pre- 206 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) 3irl intraradial crossveins lost (occasionally secondarily regained); [84] parabaculum present in male: [104] female insemination-fertilization canal opening of pore-entry type. Lineage 3: [7, 8] 5th maxillary and 3rd labial palpomeres bisubsegmented; [10] orad cavity of right mandible absent [lost]; [29] forewing radial sector segments Rs(stem) and Rs(b) branching from anterior radial trace; [95] male 9th gonocoxites absent [lost]. Lineage 9; [28] Forewing subcostal crossvein 2sc-r present. Lineage 10: [74] Male gonarcus with gonofenestra; [78] apex of mediuncus emar- ginate. Lineage 14: [31, 33] Forewing radial sector segments Rs(stem), Rs(a), Rs(b), and Rs(c) branching from anterior radial trace; [48] forewing CuP deeply forked, most proximal fork located in proximal half of posterior CuP trace. Lineage 17: [7, 8] 5th maxillary and 3rd labial palpomeres simple, bisubsegmen- tation absent [lost]. Lineage 18: [78] Apex of male mediuncus simple, emargination absent [lost]. Included Higher Taxa. Nine subfamilies containing 24 valid genera are recognized here, see Appendix 1. One additional genus, Notherobius New (1988), is also con- sidered valid, but was not examined and is not placed to subfamily. Species (ca. 550). Species richness for all recognized family- and genus-group taxa is summarized in Appendix 2. Distribution. Cosmopolitan, absent only from very high latitudes, nonvegetated deserts, and some remote oceanic islands. Principle Regional Faunas. Only complete, or nearly complete, faunal studies are listed below. Taxonomically limited works are listed under the individual genera treated. For faunas of individual European countries, and some adjacent areas see Aspdck et al. (1980). NEARCTIC REGION: Carpenter, 1940 [Nearctic (keys)]. NEOTROPICAL REGION: Alayo, 1968 [Cuba (keys)]; Penny and Monserrat, [1985] [Amazon Basin (keys)]; Stange, 1967 [Argentina and Uruguay (list)]. PALEARCTIC REGION: Aspock et al., 1980 [Europe (keys)]; Kuwayama, 1962 [Japan (keys)]; Makarkin, 1985, 1986 [USSR (keys)]. ETHIOPIAN REGION: Fraser, 1951 [Madagascar (keys)]; Tjeder, 1961 [southern Africa (keys)]. ORIENTAL REGION: Ghosh and Sen. 1977 [India (list)]. AUSTRALIAN REGION: New, 1988 [Australia (keys)]; New, 1988 [New Guinea (keys)]. ISLAND FAUNAS: See Table 10. General Notes. Comments on hemerobiid subgenera. Many recent revisionary and regional faunistic works which treat hemerobiids have recognized subgenera within the genera Hemerobius [( Hemerobius ), ( Monorobius ), (Brauerobius)], Megalomus [{Megalom us). (Pirionus)]. Micromus [(Micromus). ( Nesomicromus)\ , Sympherobius [( Sympherobius ), ( Niremberge )], and W'esmaelius [( Wesmaelius ), (Kimmmsia)]. The use of these subgenera clearly results from the desire to nomenclaturally distinguish intrageneric groupings within these large genera. Unfortunately, justification for these subgenera is available only in the form of restricted regional studies. Since all of the 1993 WORLD GENERA OF HEMEROBIIDAE 207 genera involved are quite broadly distributed, formal recognition of these subgenera is, in my view, premature. The subdivision of large genera on the basis of restricted regional studies is apt to result in the recognition of subgenera which are small paraphyletic splinter groups, large paraphyletic stem-groups, or both. If our classifications are to reflect our best hypotheses (estimates) of phylogeny (a position taken here), then all named taxa should represent demonstrably holophyletic groups. A corollary of this position is the necessity of removing taxa, including subgenera, which are not demonstrably holophyletic. Detailed analyses of intrageneric relationships within the five genera mentioned above were beyond the scope of this work. However, I have examined a sufficient number of species in each of these genera, encompassing distant portions of their ranges, to conclude that no good evidence exists to support the supposition that their pairs (or triplets) of subgenera represent pairs of holophyletic taxa. Although some of these subgenera represent small, apparently closely related, possibly holophyletic groups of species (e.g., ( Brauerobius ), ( Pirionus ), (Niremberge), ( Wesmaelius)), formal nomenclatural recognition of each of these subgenera would require the coordinate recognition of a larger subgenus which appears to be paraphyletic. Lacking compre- hensive phylogenetic studies upon which to base demonstrably holophyletic subdi- visions of these genera, the removal of their paraphyletic subgenera seems best achieved for the present by treating all of their existing subgeneric names as synonyms. For these reasons, no subgenera are formally recognized in this work. Rather than continue to use subgenera of unsubstantiated holophyly, I suggest that neuropterists expand their use of the informal “species-group” convention as a means of expressing preliminary intrageneric groupings within regional faunas. Formal sub- generic taxa should be proposed only for demonstrably holophyletic groups in con- nection with an explicit phylogenetic analysis, preferably accompanying a compre- hensive revision of the larger genus involved. Key to World Hemerobiid Genera (Adults) Notes and Instructions. The following key to the 25 genera recognized in this work is the first attempt to facilitate generic identifications for the entire family Heme- robiidae. Although it should allow the identification of most specimens, some alter- ations may be necessary after intrageneric variability has been more completely analyzed. The key is intentionally artificial (i.e., holophyletic groups are not always keyed as units), and utilizes forewing characters almost exclusively for maximum ease of use. Because hemerobiid venation is notoriously variable, the choice and sequence of couplets have been carefully considered to minimize errors from this source. However, some specimens with exceptionally abnormal venation may not key correctly. Distributional information is given in brackets. The genus Notherobius (not seen) is included based on information published by New (1988). 1. Both forewings with 2 or fewer prestigmal “radial sectors” (Fig. 82) [wide- spread] 27 1'. One or both forewings with 3 or more prestigmal “radial sectors” (Fig. 33) [widespread] 2 208 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) 2(1'). 2'. 3(2'). 3'. 4(3'). 4'. 5(4). 5'. 6(4'). 6'. 7(3. 6'). 7'. 8(7'). 8’. 9(8'. 18'). 9'. 10(9'). 10'. 11(10). 1 1'. Both forewings with 0 [rare] or 1 prestigmal sc-r crossveins (Fig. 33) [wide- spread] 16 One or both forewings with 2 or more prestigmal sc-r crossveins (Fig. 157) [widespread] 3 Both forewings with 2 or fewer prestigmal sc-r crossveins [widespread] 7 One or both forewings with 3 or more prestigmal sc-r crossveins [widespread] 4 Both forewings with 10 or more prestigmal sc-r crossveins; forewings partly or entirely finely reticulate [southern South America and Pacific islands] 5 One or both forewings with 9 or fewer prestigmal sc-r crossveins; forewings without fine reticulations (discal area sometimes coarsely reticulate) [wide- spread] 6 Forewings falcate and finely reticulate proximally (Fig. 174); subcostal space with one row of cells; hind wing macropterous [southern South America and adjacent islands] Gayomyia Forewings falcate or not, but entirely finely reticulate; subcostal space, at least proximally, with 2 or more irregularly parallel rows of cells; hind wing mi- cropterous or absent [Hawaiian Islands] Micromus (part) Forewings short and very broadly oval, nearly circular; “membrane” coria- ceous; humeral area extremely broad (Fig. 132) [Juan Fernandez Islands] . . Conchoplerella (part) Forewings elongate, not nearly circular; “membrane” usually membranous; humeral area broad or narrow [widespread] 7 Forewing costa continued proximally toward humeral plate as a simple vein segment beyond its intersection with proximal humeral trace (Fig. 199); fore- wing humeral area narrow, humeral veinlet simple or branched [widespread] Forewing costa terminating proximally at its point of intersection (sometimes incomplete) with distal end of proximal humeral trace, this point lying closely adjacent to humeral plate (Fig. 17); forewing humeral area broad, humeral veinlet always branched [widespread] 8 Both forewings with 3 or fewer m-cu crossveins (count any closely adjacent “doubled" crossveins as one) AND Forewing CuP simple or shallowly forked (i.e., most proximal fork of CuP lying in distal half of posterior CuP trace) [widespread] 19 One or both forewings with 4 or more m-cu crossveins OR Forewing CuP deeply forked (i.e.. most proximal fork of CuP lying in proximal half of pos- terior CuP trace) [widespread] 9 Forewing crossvein lcua-cup present (Fig. 182) AND Forewing costal space with a gradate series containing at least 5 (usually more) crossveins (not vein- lets) [Palearctic] Drepanepteryx Forewing crossvein lcua-cup absent (Fig. 33) OR Forewing costal space with 4 or fewer crossveins (not veinlets) [widespread] 10 Forewing distinctly falcate (Fig. 149) [South America. Australia, and eastern Asia] 1 1 Forewing not falcate, posterior margin convex throughout (Fig. 166) [wide- spread] 12 Antecubital portion of forewing 3rd gradate series strongly sigmoid [South America] Conchoplerella (part) Antecubital portion of forewing 3rd gradate series linear or only weakly sig- moid (Fig. 149) [Australia, eastern Asia] Drepanacra 1993 WORLD GENERA OF HEMEROBIIDAE 209 12(10'). 12'. 13(12). 13'. 14(13). 14'. 15(13', 14). 15'. 16(2). 16'. 17(16'). 17'. 18(17). 18'. 19(8, 18). 19'. 20(19'). 20'. Forewing with 2 or fewer irl crossveins (Fig. 90) ( 1 orfewerifRl fused distally with most distal “radial sector”; count closely adjacent “doubled” crossveins as one) OR R 1 area ill-defined because of proliferation of “radial sectors” (Fig. 140) [widespread] 13 Forewing with 3 or more irl crossveins (Fig. 166) (occasionally 2 if R 1 fused distally with most distal “radial sector”; count closely adjacent “doubled” crossveins as one) [eastern Asia] Neuronema Both forewings with 4 or fewer prestigmal “radial sectors” [widespread] 14 One or both forewings with 5 or more prestigmal “radial sectors” [widespread] 1 5 Both hind wings with 2 or fewer irl crossveins [widespread, but not Australia] 1 5 One or both hind wings with 3 irl crossveins (Fig. 91) [Australia] . . . Psychobiella Posterior margins of genae each with a field of minute punctulae (Fig. 5) [Tahiti Is., Rapa Is.] Austromegalomus Posterior margins of genae without fields of minute punctulae [widespread, but not Tahiti or Rapa] Megalomus Forewing costa continued proximally toward humeral plate as a simple vein segment beyond its intersection with proximal humeral trace (Fig. 199); fore- wing humeral area narrow, humeral veinlet simple or branched; forewing crossvein 2m-cu present or absent [widespread] 23 Forewing costa terminating proximally at its point of intersection (sometimes incomplete) with distal end of proximal humeral trace, this point lying closely adjacent to humeral plate (Fig. 17); forewing humeral area narrow or broad; forewing crossvein 2m-cu present (Fig. 157) [widespread] 17 Hind wing 4th (outer) gradate series with 3 or more (rarely 2) intraradial crossveins (Fig. 34) [widespread] 18 Hind wing 4th (outer) gradate series with 1 or fewer (rarely 2) intraradial crossveins (Fig. 42) [widespread] 21 Both forewings with 3 or fewer m-cu crossveins ( 1 eastern Asian species with 4, but then with a long sagittal clypeal seta; count closely adjacent “doubled” crossveins as one) AND Forewing CuP simple or shallowly forked (i.e., most proximal fork of CuP lying in distal half of posterior CuP trace) [widespread] 19 One or both forewings with 4 or more m-cu crossveins (long sagittal clypeal seta never present) OR Forewing CuP deeply forked (i.e., most proximal fork of CuP lying in proximal half of posterior CuP trace) [widespread] 9 Anterior border of forewing cell c3r-m strongly bowed anteriorly (Fig. 60); forewing radial veins between 3rd gradate series and wing margin distinctly sinuate (Fig. 60) [South America] Hemerobiella Anterior border of forewing cell c3r-m straight or only very weakly bowed anteriorly (Fig. 50); forewing radial veins between 3rd gradate series and wing margin not sinuate (Fig. 33) [widespread, including South America] 20 Clypeus with a long sagittal seta (Fig. 2); crossvein 2r-m absent, or if present, positioned proximad of crossvein 2m-cu; intersection of crossvein 2m-cu with M3 + 4 located more than the crossvein’s length distal to fork Ml +2/ M3 + 4, resulting in cell c2m-cu narrow distally (Fig. 33); male parabaculum completely divided (Fig. 37); male ectoproct without a pecten (Fig. 36) [wide- spread] Hemerobius Clypeus without a long sagittal seta; crossvein 2r-m present and positioned distad of crossvein 2m-cu (Fig. 50); intersection of crossvein 2m-cu with M not more than the crossvein’s length distal to fork Ml + 2/M3 + 4 (sometimes anterior to this fork), resulting in cell c2m-cu broad distally (Fig. 50); male 210 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) 21(17’). 21'. 22(2 L). 23(7, 16). 23'. 24(23 ). 24'. 25(24 ). 25'. 26(24). 26'. 27(1). 27'. 28(27’). 28'. 29(28). parabaculum not completely divided (Fig. 57); male ectoproct with a pecten (Fig. 59) [widespread] Wesmaelius Forewing with 3 m-cu crossveins [widespread, including South America and the Hawaiian Islands] Sympherobius (part) Forewing with 4 m-cu crossveins [South America and the Hawaiian Islands] 22 Posterior sectoral trace giving rise to 1 side-branch along its proximal half (Fig. 74): male frons with a deep bilobed cavity (Fig. 78) [South America] Neosympherobius (part) Posterior sectoral trace simple along its proximal half (Fig. 41); male frons without a deep bilobed cavity [Hawaiian Islands] Nesobiella Forewing crossvein 2m-cu present, separated from crossvein lm-cu by a distance approximately equal to the combined lengths of both crossveins (Fig. 191) [southwestern Pacific islands] Noius Forewing crossvein 2m-cu absent, if apparently present then distance sepa- rating crossvein 1 m-cu from next apparent m-cu crossvein much greater than their combined lengths (Fig. 207) [widespread, including Pacific islands] ... 24 Both forewings with M3 -4 and CuA free (not fused), assume no fusion if a transverse 3m-cu crossvein is present (Fig. 207) (it may be very short) [wide- spread] 26 One or both forewings with M3 + 4 and CuA fused, assume fusion if: (1) fusion is clear, or (2) no 3m-cu crossvein is apparent. (3) apparent 2m-cu or 3m-cu crossvein is strongly inclined or setose (Fig. 199). or (4) continuation of apparent CuA beyond its intersection with apparent 2m-cu or 3m-cu crossvein is unbranched except for small marginal forks (Fig. 215) [widespread] 25 Hind wing M3 — 4 not fused to CuA (Fig. 200) AND Forewing CuA simple (not forked) proximal to its fusion with M3-4 (proximal segment of M3->-4 appearing as a long, oblique, setose "crossvein") (Fig. 1 99); male parabaculum with a pair of articulated sclerites developed in the parabacular membrane adjacent to the apophyseal shaft (Fig. 205) [South and Central America. West Indies] Nusalala Hind wing M3 ^4 fused to CuA (Fig. 216) OR Forewing Cu\ forked proximal to its fusion with M3 -4 (proximal portion of anterior branch of CuA usually appearing to be crossvein 2m-cu or 3m-cu) (Fig. 215): male parabaculum without sclerites developed in the parabacular membrane adjacent to the apophyseal shaft (Fig. 217) [widespread] Wicromus (part) Most proximal hind wing irl crossvein intersecting anterior sectoral trace (1) in interramus 4 AND (2) proximal to pterostigmal region (Fig. 208) (interrami are vein segments located between sequential longitudinal vein branching points) [Australia and New Guinea] Megalomina Most proximal hind wing irl crossvein not intersecting anterior sectoral trace in interramus 4 (Fig. 216) OR not proximal to pterostigmal region (usually intersecting in interramus 2 or 3. or near distal end of pterostigma) [wide- spread. including Australia and New Guinea] Micromus (part) Forewing CuP forked proximal to crossvein 2cua-cup (Fig. 98) [widespread] 34 Forewing CuP not forked proximal to crossvein 2cua-cup (Fig. 82) [wide- spread] 28 Posterior sectoral trace giving rise to 1 side-branch along its proximal half (Fig. 82) [widespread] 29 Posterior sectoral trace simple along its proximal half (Fig. 24) [Australia] . Carobius Forewing costal space narrow or broad: some posthumeral costal veinlets 1993 WORLD GENERA OF HEMEROBI1DAE 211 29'. 30(29). 30'. 31(30). 31'. 32(30'). 32'. 33(32). 33'. 34(27). 34'. branched (or, if none branched, then trichosores evident in humeral area) AND proximal humeral trace recurrent and with at least 1 side-branch [wide- spread] 30 Forewing costal space very narrow; posthumeral costal veinlets unbranched; proximal humeral trace unbranched (or, if with at least 1 side-branch, then trichosores not evident in humeral area) [Australia, southeast Asia] Zachobiella Forewing 4th (outer) gradate series with 3 or more (usually 4) intraradial crossveins (Fig. 82) [widespread] 31 Forewing 4th (outer) gradate series with no more than 2 (usually 0) intraradial crossveins (Fig. 106) [widespread] 32 Forewing crossveins 4m-cu and 4im absent (Fig. 82); intraradial crossveins of forewing 4th (outer) gradate series frequently stepped alternately [wide- spread, including South America] Sympherobius (part) Forewing crossveins 4m-cu and 4im present (Fig. 66) (rarely one absent); intraradial crossveins of forewing 4th (outer) gradate series always stepped in one direction (Fig. 66) [South America] Nomerobius Forewing crossvein 2sc-r present (Fig. 1 06); male frons without a deep bilobed cavity [widespread, but not South America] 33 Forewing crossvein 2sc-r absent (Fig. 74); male frons with a deep bilobed cavity (Fig. 78) [South America] Neosytnpherobius (part) Male ectoproct with a small, articulated, posterior lobe, bearing a distal row of modified setae or setal pegs (Figs. 108-109); antecosta of male 9th tergite entire; female 9th gonocoxites small reniform plates, without prominent mar- ginal lobes or spinose setal bases (Fig. 1 10) [widespread, including Africa] Pseclra Male ectoproct with a lanceolate posterior process, not bearing a distal row of modified setae (Fig. 1 1 7); antecosta of male 9th tergite divided by an internal sagittal sulcus; female 9th gonocoxites greatly enlarged, each orna- mented ventrally with a prominent ventral lobe and spines derived from modified setal bases (Fig. 1 18) [Africa] Anapsectra All or part of prestigmal length of forewing subcostal space much wider than width of adjacent subcosta (New, 1988, fig. 192) [Australia] Notherobius Entire prestigmal length of forewing subcostal space no wider than width of adjacent subcosta (sometimes slightly wider immediately adjacent to cross- vein lsc-r) (Fig. 98) [primarily pantropical, including Australia] . . Notiobiella Subfamily Carobiinae Oswald, new subfamily Carobiinae Oswald, new subfamily (Type genus: Carobius Banks). Included Genera. Carobius. Genus Carobius Banks (Figs. 12, 24-32) Carobius Banks, 1909:78 (Type species: Carobius pulchellus Banks, 1909:79, by original designation. Etymology: Unexplained, probably Caro- [from Latin cams , dear or precious] + -bius [from Greek bios (masculine), life]. Gender: Masculine.): Tillyard, 1916 (RD, A, Dst, Key, W*, Tax); New, 1988 (RD, A, Dst, Key, FT*, MT*, W*, Tax). 212 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 24-32. Carotins pulchellus (except as noted). 24, Forewing [Scale bar A], 25, Hind wing [A], 26. Female terminalia ( Carobius sp.). lateral [B], 27. Male terminalia. lateral [B], 28. Gonarcus, dorsal [C]. 29, Gonarcus and 9th gonocoxite, lateral [C]. 30, 9th gonocoxite, dorsal [C], 31. Parabaculum. dorsal [C], 32. Parabaculum. lateral [C]. Abbreviations: 7t, 8t, 9t. tergites; 7s, 7th stemite: 9gcx. 9th gonocoxite; ect. ectoproct; med. mediuncus; pmed, paramediuncus; sty, stylus. Scale bars (mm): A = 2.0; B = 0.5; C = 0.1. Differential Diagnosis. Recognized by the forewing character combination (Fig. 24): (1) anterior radial trace bearing 2 prestigmal “radial sectors” and (2) posterior sectoral trace simple for at least the proximal two-thirds of its length. Proposed Synapomorphies. [30] Forewing with 2 ORB's, ORB 1 shallowly forked: [34] most proximal fork of most proximal “radial sector” located in distal two-thirds of posterior sectoral trace; [39] forewing radiomedial crossvein 3r-m absent; [44] forewing mediocubital crossvein 3m-cu absent; [103] female gonapophyses poster- iores absent. Species (9). angustus Banks: Queensland, New South Wales; curvatus New: South Australia; elongatus New: New South Wales; lateproctus New: Northern Territory; pectinatus New: Northern Territory; pedicellatus New-: South Australia and south- western Australia; pulchellus Banks: Queensland, New South Wales; spinosus New: southwestern Australia; trifurcatus Kimmins: Queensland. Distribution. Australia. Preimaginal Stages. Unknown. 1993 WORLD GENERA OF HEMEROBIIDAE 213 Subfamily Hemerobiinae Latreille (revised) Hemerobini Latreille, 1802:288 (Type genus: Hemerobius Linnaeus). See also the Synonymical Listing for the family Hemerobiidae. Differential Diagnosis. Recognized by the forewing character combination (Fig. 33): (1) crossveins 2sc-r and lcua-cup absent, (2) anterior radial trace bearing 3 to 5 prestigmal “radial sectors”, and (3) crossvein 4r-m present. Proposed Synapomorphies. Hemerobiinae (Lineage 4): [11] Proximal convexity of orad margin of right mandible strongly angulate; [32] forewing with 3 ORB’s, ORB1 shallowly forked; [34] most proximal fork of most proximal “radial sector” located in distal two-thirds of posterior sectoral trace; [94] male supragonopontal setal group present; [100] styli of female 9th gonocoxite lost. Hemerobius genus-group (Lineage 5): [85] Parabaculum deeply divided; [103] fe- male gonapophyses posteriores lost; [105] female 8th stemite (subgenitale) lost. Wesmaelius genus-group (Lineage 6): [44] Forewing mediocubital crossvein 3m- cu lost. Included Genera. Hemerobius, Nesobiella, Wesmaelius, and Hemerobiella. Distribution. All continents except Antarctica, but poorly represented in Australia (1 Hemerobius sp.). Present on numerous islands in the Atlantic, Indian and Pacific oceans. Genus Hemerobius Linnaeus (Figs. 6-7, 13, 33-40) Hemerobius Linnaeus, 1758:549 (Type species: Hemerobius humulinus Linnaeus, 1758:550, by subsequent designation by the International Commission on Zoo- logical Nomenclature, Opinion 21 1 (ICZN, 1954):3. Etymology: See Navas, [1924b] 1923:199; Hemero- [from Greek hemera, day] + -bius [from Greek bios (mas- culine), life]. Gender: Masculine.): Latreille, 1810 (ITSD); Curtis, 1828 (ITSD). Egnyonyx Wesmael, 1836:167 (Type species: Hemerobius humulinus Linnaeus, 1758: 549 [as humuli (sic)], by subsequent designation by Kimmins, 1963:202. Etymol- ogy: Unexplained, probably Egny- [origin unknown] + -onyx [from Greek onyx (masculine), claw]. Gender: Masculine.): Kimmins, 1963 (Syn). Mucropalpus Rambur, 1842:420 (Type species: Hemerobius lutescens Fabricius, 1793: 84, by subsequent designation by Aspock et al., 1980:208. Etymology: Unex- plained, probably Mucro- [from Latin mucro, sharp point] + -palpus [from Latin palpus (masculine), feeler]. Gender: Masculine.): McLachlan, 1863 (ITSD); Banks, 1905 (ITSD); Leraut, [1981] 1980 (ITSD); Klimaszewski and Kevan, 1985 (Syn). Stenolomus Navas, 1906:701 (Type species: Stenolomus cabrerai Navas, 1906:701 [=Hemerobius eatoni Morton, 1906:147], by subsequent designation by Aspock et al., 1980:208. Etymology: See Navas, 1906:701; Steno- [from Greek stenos, narrow] + -lomus [from (Mega)lomus]. Gender: Masculine.): Klimaszewski and Kevan, 1985 (Syn). Hemerodomia Navas, 1 909b:2 1 5 (Type species: Hemerodomia buyssoni Navas, 1 909b: 217 [= Hemerobius stigma Stephens, 1836:1 12], by monotypy. Etymology: See Navas, 1909b:216; Hemero- [from Hemero(bius)] + -domia [from Puy-de-Dome, a region in France]. Gender: Feminine.): Klimaszewski and Kevan, 1985 (Syn). Brauerobius Kruger, 1922:171 (Type species: Hemerobius marginatus Stephens, 1836: 214 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 33-40. Hemerobius humulinus. 33. Forewing [Scale bar A]. 34, Hind wing [A]. 35, Female terminalia. lateral [C], 36, Male terminalia. lateral [C], 37, Parabaculum. dorsal [B], 38. Parabaculum. lateral [B], 39, Gonarcus. dorsal [D], 40. Gonarcus. lateral [D], Scale bars (mm): A = 2.0; B = 0.1; C = 0.5; D = 0.1. 109, by original designation. Etymology: Unexplained, probably Brauer- [from the surname of Austrian entomologist Friedrich Moritz Brauer (1832-1904)] + -obius [from Greek bios (masculine), life]. Gender: Masculine.): Klimaszewski and Kevan, 1985 (Tax). NEW STATUS Hagenobius Kruger, 1922:171 (Type species: Hemerobius citrinus Hagen. 1861:204 [=Hemerobius conjunctus Fitch, [1855] 1854:798], by original designation. Ety- mology: Unexplained, probably Hagen- [from the surname of German entomol- ogist Hermann August Hagen ( 1 8 1 7-1 893)] + -obius [from Greek bios (masculine), life]. Gender: Masculine.): Klimaszewski and Kevan. 1985 (Syn). Reuterobius Kruger, 1922:171 (Type species: Hemerobius pini Stephens, 1836:1 1 1, by original designation. Etymology: Unexplained, probably Reuter- [from the sur- name of Finnish entomologist Odo Morannal Reuter (1850-1913)] + -obius [from Greek bios (masculine), life]. Gender: Masculine.): Klimaszewski and Kevan, 1985 (Syn). Schneiderobius Kruger, 1922:171 (Type species: Hemerobius nitidulus Fabricius, 1777:244. by original designation. Etymology: Unexplained, probably Schneider- 1993 WORLD GENERA OF HEMEROBIIDAE 215 [from the surname of German entomologist Wilhelm Gottlieb Schneider (181 4— 1889)] + -obius [from Greek bios (masculine), life]. Gender: Masculine.): Kli- maszewski and Kevan, 1985 (Syn). Anotiobiella Kimmins, 1928:364 (Type species: Anotiobiel/a withycombei Kimmins, 1928:365, by monotypy. Etymology: Unexplained, probably A- [from Greek a-, not] + Notiobiella. Gender: Feminine.): Gonzalez-Olazo, 1987 (Syn). Allemerobius Banks, 1940:183 (Type species: Allemerobius Jlaveolus Banks, 1940: 184, by monotypy. Etymology: Unexplained, probably All- [from Greek alios, other] + -emerobius [from (H)emerobius]. Gender: Masculine.): Yang, 1981 (RD, MP). NEW SYNONYM Dyshemerobius Tjeder, 1961:352. Type species: Dyshemerobius productus Tjeder, 1961:354, by original designation. Etymology: Unexplained, probably Dys- [from Greek dys , bad or with difficulty] + Hemerobius. Gender: Masculine.): Tjeder, 1961 (OD, Dst, Fig, Key, Not); Tjeder, 1963a (Tax, Dst); Monserrat, 1990 (Dst, Fig, MT, Tax). NEW SYNONYM Mesohemerobius Nakahara, 1966:202 [not Mesohemerobius Ping, 1928:42 (a fossil hemerobiid)] (Type species: Mesohemerobius subacutus Nakahara, 1966:202, by original designation. Etymology: Unexplained, probably Meso- [from Greek mesos, middle] + Hemerobius. Gender: Masculine.). NEW SYNONYM Semohemerobius Yang, 1983:128 [an objective replacement name for Mesoheme- robius Nakahara, 1966; not Mesohemerobius Ping, 1928] (Type species: Mesohem- erobius subacutus Nakahara, 1966:202, by original designation [for Mesoheme- robius Nakahara, 1966], Etymology: See Yang, 1983:128; An anagram of Mesohemerobius. Gender: Masculine.). NEW SYNONYM Monorobius Makarkin, 1985:167 (Type species: Hemerobius lutescens Fabricius, 1793:84, by original designation. Etymology: Unexplained, probably Mono- [from Greek monos, one] -I- -robius [from (Heme)robius]. Gender: Masculine.). NEW SYNONYM Differential Diagnosis. Recognized by the presence of a prominent sagittal seta on the clypeus (Fig. 2), and the completely divided male parabaculum (Fig. 37). Dif- ferentiated venationally from other Hemerobiinae by the absence of crossvein 2r-m (Fig. 33), or its displacement proximad of crossvein 2m-cu. Proposed Synapomorphies. [14] Prominent sagittal seta of clypeus present; [38] forewing radiomedial crossvein 2r-m absent or lying proximad of crossvein 2m-cu; [86] male parabaculum longitudinally divided into a pair of parasagittal sclerites. Species (ca. 125). No recent comprehensive listing is available. Distribution. Cosmopolitan. Widespread on all continents except Australia (1 sp.; New Guinea, 3 spp.) and Antarctica (absent), but apparently most diverse in northern temperate areas. Present also on numerous islands in the Atlantic, Indian and Pacific oceans. Principle Revisions and Regional Faunas. Alayo, 1968 [Cuba]; Aspock et al., 1980 [Europe]; Klimaszewski and Kevan, 1985, Kevan and Klimaszewski, 1987 [Canada and Alaska]; Kuwayama, 1962 [Japan]; MacLeod, 1964 [Hawaiian Islands]; Mak- arkin, 1985 [USSR]; New, 1988 [Australia]; New, 1988 [New Guinea]; Penny and Monserrat, [1985] 1983 [Amazon Basin]; Stange, 1967 [Argentina and Uruguay]; Tjeder, 1961 [southern Africa]. 216 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Preimaginal Stages. (1) atrifrons— Killington, 1937; (2) humulinus— Smith, 1923 (as humuli): Withycombe, 1923; Killington, 1936, 1 937; Nakahara, 1954 (as obtususf, Agekjan, 1973; (3) japonicus— Nakahara, 1954; (4) lutescens — Withycombe, 1923; Killington, 1937; (5) marginatus— -Killington, 1937; (6) micans— Withycombe, 1923; Killington, 1936, 1937; (7) nitidulus— Withycombe, 1923; Killington, 1936, 1937; Bansch, 1 964; (8) pacificus — Quayle, 1912; Moznette, 1915a, 1915b; (9 ) perelegans— Killington, 1934, 1936, 1937; (10) pini— Killington, 1936, 1937; (1 1) simulans — Killington, 1936, 1937; (12) stigma— Withycombe, 1922, 1923; Smith, 1925 (as stigmaterus ); Killington, 1936, 1937; Miller and Lambdin, 1982, 1984. Synonymical Notes. Allemerobius. Banks (1940) diagnosed Allemerobius on the basis of the slightly falcate forewing and several other minor venational features. These characters are variable within the revised concept of Hemerobius adopted here. I have examined two specimens (both females, MCZ) of the type species, Allemerobius flaveolus. These specimens exhibit the proposed synapomorphies of Hemerobius found in females, i.e., characters [14] and [38], The male terminalia of flaveolus were figured by Yang (1981: figs. 62-65). They are in all respects characteristic of Hem- erobius. For example, the ectoprocts bear acute processes distally, the gonarcus pos- sesses a pair of median paramediunci, and the parabaculum is completely divided sagittally. The latter characteristic is a unique synapomorphy of Hemerobius [char- acter 86], Dyshemerobius. Tjeder (1961) originally diagnosed Dyshemerobius on the basis of several unusual characteristics of the female terminalia (particularly the partially fused and ventrally produced 9th gonocoxites) of its type species, D. productus. Tjeder (1963a) described a second species ,falciger, and added the male terminalic character “gonarcus split in the middle-line” to the generic diagnosis of Dyshemerobius. Mon- serrat (1990) described the male of productus and showed that the split gonarcus of falciger was not characteristic of productus. In fact, all three of Tjeder’ s diagnostic features appear to be derived relative to the homologous states found in other Hem- erobius species. Although falciger and productus may yet prove to form a distinctive species group within Hemerobius , its exclusion from Hemerobius s.l. would render the latter paraphyletic. I have examined the male holotype of falciger. It possesses all three of the synapomorphies proposed here for Hemerobius [characters 14, 38, and 86], and the two genera are synonymized on this basis. Mesohemerobius Nakahara [nec Ping] and Semohemerobius. Nakahara (1966) di- agnosed Mesohemerobius on the basis of the fused paramediunci (fused “processes of aedeagus”) of the male gonarcus. Within Hemerobius , this condition is clearly derived relative to the presence of a pair of independent paramediunci. Excluding a group of species from Hemerobius on the basis of this character alone would leave the taxon comprising the remaining Hemerobius species paraphyletic. I have ex- amined a male of Mesohemerobius subacutus, the type species of Mesohemerobius, and found it to possess the three Hemerobius synapomorphies proposed here, and the two genera are synonymized on this basis. Semohemerobius is an objective re- placement name for Mesohemerobius Nakahara, 1966 (not Mesohemerobius Ping, 1928). Brauerobius. Several recent authors have followed Aspock et al. (1980) in recog- nizing Brauerobius as a subgenus distinct from Hemerobius ( Hemerobius ). In my view, recognition of Brauerobius as a formal subgenus is premature. Although the males of the marginatus species group (i.e., costalis, marginatus, and tristriatus) 1993 WORLD GENERA OF HEMEROBIIDAE 217 possess a distinctive synapomorphy, i.e., a field of setal bases modified as recurved denticles on the mesodistal surfaces of the male ectoprocts, removal of these species from Hemerobius likely leaves the remaining group of Hemerobius species paraphy- letic. If the marginatus group forms the sister group to the rest of Hemerobius (the presence of acute prominences on the male ectoprocts, absent in the marginatus group, is a possible although unconfirmed synapomorphy of this grouping) there may be justification for recognizing Brauerobius at subgeneric rank. However, the analyses needed to demonstrate this possibility have yet to be undertaken. Monorobius. Makarkin (1985: 101 [of English translation]) diagnosed the subgenus Hemerobius ( Monorobius ) on the basis of two characters: ( 1 ) “ectoproct with single tip (kataprocessus), anaprocessus in form of spine or lacking” and (2) “paired parts of arcessus [paramediunci] fused at base”. These characters do not appear to divide Hemerobius into a pair of holophyletic subgenera, consequently, the subgenus Hem- erobius (Monorobius) is not recognized here. There are currently insufficient data to establish the homologies of the male ec- toproct processes of Hemerobius. This is especially true when processes of highly divergent ectoprocts, such as those characteristic of Makarkin’s subgenera, are com- pared. Even if Makarkin’s homologies are accepted, it is difficult to accept that this character constitutes a synapomorphy for ( Monorobius ), since both processes (“an- oprocessus” and “katoprocessus”) are distinctly present in most species of both Hemerobius ( Monorobius ) and H. ( Hemerobius ) sensu Makarkin (see Makarkin (1985: figs. 10-27). Makarkin’s second character, proximal fusion of the paramediunci, is the same as that proposed by Nakahara for his genus Mesohemerobius. For a discussion of this character see under the heading Mesohemerobius and Semohemerobius above. It should be noted that a tendency for the paramediunci to migrate medially along the posterior margin of the extragonarcus may be a common convergent trait in the Hemerobiinae and Sympherobiinae. Paramediunci which lie closely adjacent, or are fused, medially along the posterior margin of the extragonarcus occur in Hemerobius , Nesobiella, and Nomerobius. The possibility that such a configuration has evolved convergently in independent lineages of Hemerobius cannot be presently ruled out. This tends to argue against the diagnostic use of this character for a subgeneric taxon. General Notes. Among hemerobiid genera, Hemerobius is second only to Micromus in its number of generic synonyms. This reflects a long history of attempts to subdivide this large and broadly distributed genus. Adequate faunistic treatments of Hemerobius are currently available for most regions of the world (South and Central America and southeast Asia being notable exceptions). Future advances in our understanding of this genus will only come from concerted efforts to expand phylogenetic and revisionary studies across regional and continental boundaries. Such studies are the only satisfactory basis upon which stable, holophyletic, intrageneric groupings can be based. Genus Nesobiella Kimmins (Figs. 41-49) Nesobiella Kimmins, 1935:618 (Type species: Megalomus hospes Perkins, 1899:36, by original designation. Etymology: Unexplained, probably Neso- [from Greek nesos, island] + -bi- [from Greek bios, life] + -ella [from Latin diminutive suffix 218 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 4 1-49. Nesobiella hospes. 4 1 . Forewing [Scale bar A]. 42, Hind wing [A], 43, Gonarcus. dorsal [C], 44. Gonarcus, lateral [C], 45, Female terminalia, lateral [B], 46, Apex of male ectoproct, mesal [C], 47, Male terminalia. lateral [B], 48, Parabaculum, dorsal [D], 49, Para- baculum. lateral [D], Abbreviations: pmed. paramediuncus. Scale bars (mm): A = 2.0; B = 0.5; C = 0.2; D = 0.1. -ellus]. Gender: Feminine.): Perkins, 1913 (Bio); Esben-Petersen, 1937 (Dst, Lst); Zimmerman. 1940 (Dst. Key, Lst); Zimmerman, 1957 (Dst. Key); MacLeod, 1964 (Mis). Differential Diagnosis. Differentiated from other Hawaiian hemerobiids by the character combination: (1) forewing anterior radial trace bearing 3 or 4 prestigmal “radial sectors” (Fig. 41), (2) forewing proximal humeral trace strongly recurrent (Fig. 41), and (3) 2 or fewer (usually 0 or 1) intraradial crossveins in hind wing 4th (outer) gradate series. Proposed Synapomorphies. [58] Apex of male 9th stemite produced as a long, straight, attenuate tube; [65] inner distal surface of male ectoproct bearing a field of shortened, fusiform setae. Species (1). hospes (Perkins): Hawaiian Islands. Distribution. Hawaiian Islands. Preimaginal Stages. Unknown. 1993 WORLD GENERA OF HEMEROBIIDAE 219 Genus Wesmaelius Kruger (Figs. 50-59) Wesmaelius Kruger, 1 922: 1 70 (Type species: Hemerobius concinnus Stephens, 1 836: 106, by original designation. Etymology: Unexplained, probably from the surname of Belgian entomologist Constantin Wesmael (1798-1872). Gender: Masculine.). Kimminsia Killington, 1937:254 (Type species: Hemerobius betulinus Strom, 1788: 387, by original designation. Note: The identity of H. betulinus is uncertain. Aspock et al., 1980:409, list this name as a nomen dubium, and treat usages of "betulinus” by subsequent authors under the junior name nervosus Fabricius (1793:85); see also Tjeder (194 1 :28— 29) and Esben- Petersen (1925:54) for discussions of the iden- tity of betulinus. Etymology: See Killington, 1937:254; From the surname of British entomologist Douglas Eric Kimmins (1905-1985). Gender: Feminine.): Tjeder, 1941 (ITSD); Klimaszewski and Kevan, 1987, 1987 (Tax). NEW STATUS Differential Diagnosis. Recognized by the character combination: in the forewing (Fig. 50), (1) crossvein 2sc-r absent, (2) anterior radial trace bearing 3 or more (up to ca. 6) prestigmal “radial sectors”, (3) proximal humeral trace strongly recurrent, (4) crossvein 2r-m present and positioned adjacent or distal to crossvein 2m-cu, (5) anterior border of cell c3r-m not strongly bent, and, in the hind wing (Fig. 51), (6) 4th (outer) gradate series with 3 or more intraradial crossveins. Distinguished from the similar and largely sympatric genus Hemerobius by the presence of forewing crossvein 2r-m in its normal position adjacent or distal to crossvein 2m-cu, and, in the male genitalia, by the presence of a pecten (Fig. 59, pec) on the mesal process of the male ectoproct, and the male parabaculum never completely divided. See also key couplet 20. Proposed Synapomorphies. [61] Posterior margin of male 9th stemite bicuspate; [66] pecten of male ectoproct present. Species (ca. 70). No recent comprehensive listing is available. Distribution. Widely distributed in North America, Africa, Europe and Asia. Most diverse in the temperate regions of North America and Eurasia. Many early records are recorded under the generic names “ Kimminsia ” and “ Boriomyia" (sensu Banks, 1905). Principle Revisions and Regional Faunas. Aspock et al., 1 980 [Europe]; Klimaszew- ski and Kevan, 1987, 1987 [Canada and Alaska]; Kuwayama, 1962 [Japan]; Mak- arkin, 1986 [USSR]; Tjeder, 1961 [southern Africa], Preimaginal Stages. (1) bihamita— Yang, 1980a; (2) concinnus— Withycombe, 1923 (in Boriomyia ); Killington, 1937; (3) navasi — Monserrat, 1983; (4) nervosus— Withy- combe, 1923 (in Boriomyia ); Miles, 1924 (in Boriomyia)', Killington, 1934, 1936, 1937 (as Boriomyia betulina): (5) quadrifasciat us — -'N'why combe, 1923 (in Borio- myia): Killington, 1934, 1936, 1937; (6) rava — Killington, 1937 (in Boriomyia ); (7) subnebulosus— Withycombe, 1923 (in Boriomyia): Killington, 1936, 1937 (in Bor- iomyia): Laffranque and Canard, 1975; (8) spp. — Gurney, 1947. Synonymical Notes. Kimminsia. Recent works treating Nearctic and Palearctic Wesmaelius have generally followed Aspock et al. (1980) in recognizing two sub- genera, (Wesmaelius) and ( Kimminsia ). Klimaszewski and Kevan (1987:163, 263) diagnosed the smaller subgenus (Wesmaelius) on four principal characters: (1) fore- 220 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 50-59. Wesmaelius concinnus. 50. Forewing [Scale bar B]. 51. Hind wing [B]. 52. Female terminalia. lateral [D]. 53. Male 9th stemite. posterior [D], 54. Male terminalia. lateral [D], 55. Gonarcus. dorsal [C]. 56. Gonarcus. lateral [C]. 57. Parabaculum. dorsal [A]. 58. Parabaculum. lateral [A]. 59. Ventrodistal portion of male ectoproct. mesal [E]. Abbreviations: 2ir. 2nd series intraradial crossvein; med. mediuncus: pec. pecten: pmed. paramediuncus. Scale bars (mm): A = 0.5: B = 2.0: C = 0.5: D = 0.5: E = 0.2. wing broadly oval. (2) a forewing 2ir crossvein present (Fig. 50: “r+m*” in their terminology). (3) male ectoprocts triangular, and (4) female 9th gonocoxites (“gon- apophyses laterales") elongate. The subgenus ( Kimminsia ) was characterized as pos- sessing the corresponding characters: ( 1 ) forewing narrowly oval to oval, (2) forewing 2ir crossvein absent. (3) male ectoproct rectangular, and (4) female 9th gonocoxites short. I regard the polarities of characters ( 1 ) and (3). forewing and male ectoproct shapes, as unresolved, and therefore not currently adequate to demonstrate the holophyly of 1993 WORLD GENERA OF HEMEROBIIDAE 221 Figs. 60-65. Hemerobiella sinuata (except as noted). 60, Forewing [Scale bar A], 61, Hind wing [A]. 62, Gonarcus, lateral [B], 63, Parabaculum, lateral [B]. 64, Female terminalia (Hemero- biella sp.), lateral [B]. 65, Male terminalia, lateral [B]. Abbreviations: avl, anteroventral lobe of 9th tergite; cos, costa. Scale bars (mm): A = 2.0; B = 0.5. either group. By outgroup comparison to other hemerobiine genera, the states of characters (2) and (4) in ( Wesmaelius ) are clearly derived; and, the plesiomorphic states in ( Kimminsia ) cannot justify its holophyly. Although (Wesmaelius) appears to be holophyletic, its removal from Wesmaelius s.l. appears to render ( Kimminsia ) paraphyletic; for this reason neither subgenus is recognized here. Genus Hemerobiella Kimmins (Figs. 60-65) Hemerobiella Kimmins, 1940:232 (Type species: Hemerobiella sinuata Kimmins, 1940:233, by original designation. Etymology: Unexplained, probably Hemero- [from Greek hemera, day] + -bi- [from Greek bios, life] + -ella [from the Latin diminutive suffix -ellus). Gender: Feminine.). Differential Diagnosis. Differentiated from other New World hemerobiids by the character combination: in the forewing (Fig. 60), (1) crossvein 2sc-r absent, (2) an- terior radial trace bearing 3 or more prestigmal “radial sectors”, (3) proximal humeral trace strongly recurrent, (4) anterior border of cell c3r-m strongly bent, and, in the hind wing (Fig. 61), (5) 4th (outer) gradate series with 3 or more intraradial crossveins. 222 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) The last character, and the distinctly sinuate distal portions of the forewing radial veins (Fig. 60), will distinguish this genus from other Hemerobiinae, including the similar genus Wesmaelius. The male ectoproct lacks a pecten. Proposed Synapomorphies. [54] Horizontal costae of inner lateral surfaces of male 9th tergite present; [98] anteroventral angles of female 9th tergite prominently pro- duced anteriorly. Species (1). sinuata Kimmins; Ecuador. A female specimen from Venezuela ex- amined for this study (NMNH) is presumed to belong to this genus, but may not be conspecific with sinuata. Distribution. Northwestern South America. Preimaginal Stages. Unknown. Subfamily Sympherobiinae Comstock (revised) Sympherobiidae Comstock, 1918:179 (Type genus: Sympherobius Banks). Sympherobiinae Kruger, 1922:171. Sympherobini Navas, [1925b]:91. Differential Diagnosis. Recognized by the forewing character combination: (1) crossvein 2sc-r absent, (2) anterior radial trace bearing 2 prestigmal “radial sectors” (3 in a few Sympherobius ), and (3) proximal third of posterior sectoral trace bearing one side branch, and (4) trichosores evident on humeral margin. Proposed Synapomorphies. Sympherobiinae (Lineage 7): [12] Distal convexity of orad margin of right mandible rounded; [40] forewing radiomedial crossvein 4r-m absent; [53] posteroventral angle of male 9th tergite produced as a narrow, membrane- margined lobe; [79] male pseudomediuncus present. Lineage 8 ( Neosympherobius + Sympherobius ): [78] Mediuncus lost. Included Genera. Nomerobius. Neosympherobius. and Sympherobius. Distribution. North, South and Central America, Europe, Africa, and Asia. Ap- parently absent from Australia, India and most or all of tropical southeast Asia. Genus Nomerobius Navas (Figs. 66-73) Nomerobius Navas, 1915a: 130 (Type species: Megalomus psychodoides Blanchard in Gay, 1 85 1 : 1 27, by monotypy. Etymology: Unknown. Gender: Masculine.): Stange, 1967 (Lst); Penny and Monserrat, [1985] 1983 (Tax); Oswald, 1990 (Tax). Differential Diagnosis. Distinguished from other South American hemerobiids by the forewing character combination (Fig. 66): (1) anterior radial trace bearing 2 prestigmal “radial sectors”, (2) intramedial space with 2 (rarely 1) crossveins, and (3) mediocubital space with 4 (rarely 3) crossveins. The posterolateral cusps of the male 9th tergite (Fig. 69) and the secondary medioventral connection between the extrahemigonarcus of the male gonarcus (Figs. 72-73) are also diagnostic. Proposed Synapomorphies. [52] Posterolateral margins of male 9th tergite cuspate; [59] apex of male 9th stemite tubularly produced and abruptly upturned; [63] male 9th stemite bearing a pair of elongate lateral setae; [68] male ectoproct with an elongate subapical seta borne on a short pedicel; [73] male gonarcus with lateral 1993 WORLD GENERA OF HEMEROB1IDAE 223 Figs. 66-73. Nomerobius psychodoides. 66, Forewing [Scale bar A], 67, Flind wing [A], 68, Female terminalia, lateral [B], 69, Male terminalia, lateral [B], 70, Parabaculum, dorsal [C], 7 1 , Parabaculum, lateral [C], 72, Gonarcus, lateral [C], 73, Gonarcus, dorsal [C], Abbreviations: car, carina; csp, cusp of 9th tergite; pmd, pseudomcdiuncus; sgp, secondary gonopons. Scale bars (mm): A = 2.0; B = 0.5; C = 0.1. neogonarcal carinae; [76] secondary gonopons of male gonarcus present; [103] female gonapophyses posteriores lost. Species (4). signatus (Hagen): Argentina, Chile, Peru; cuspidatus Oswald: Argentina, Chile, Bolivia; psychodoides (Blanchard in Gay): Argentina, Chile, Peru; spinosus Oswald: Chile. Distribution. Central and southern South America. Preimaginal Stages. Unknown. 224 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 74-8 1 . Neosympherobius cinereus. 74. Forewing [Scale bar A]. 75. Hind wing [A]. 76, Gonarcus. lateral [D]. 77, Gonarcus, dorsal [D]. 78, Cranium, anterior [C], 79. Male terminally, lateral [B], 80, Parabaculum. dorsal [D], 81, Parabaculum, lateral [D]. Abbreviations: cos, costa. Scale bars (mm): A = 2.0; B = 0.5; C = 0.5; D = 0. 1. Genus Neosympherobius Kimmins (Figs. 74-81) Neosympherobius Kimmins, 1929:187 (Type species: Neosympherobius cinereus Kimmins, 1929:187, by original designation. Etymology: Unexplained, probably Neo- [from Greek neos. new] + Sympherobius. Gender: Masculine.): Stange. 1967 (Lst). Differential Diagnosis. Distinguished from other South American hemerobiids by the forewing character combination (Fig. 74): (1) anterior radial trace bearing 2 1993 WORLD GENERA OF HEMEROBIIDAE 225 (occasionally 3) prestigmal “radial sectors”, (2) CuP not forked proximal to crossvein 2cua-cup, and (3) 4th (outer) gradate series lacking crossveins in intraradial space. The paired frontal cavities of the male (Fig. 78) are unique within the family. The highly derived configuration of the male gonarcus is also diagnostic (Fig. 76). Proposed Synapomorphies. [22] Male with a bilobed frontal cavity; [36] forewing intraradial crossvein 4irl lost; [70] ventral margins of extrahemigonarcus promi- nently turned outward; [72] each intrahemigonarcus with a more or less horizontal costa or keel ventrally. Species (1). cinereus Kimmins: Argentina. Distribution. Southern South America. Preimaginal Stages. Unknown. Genus Sympherobius Banks (Figs. 14, 82-89) Sympherobius Banks, 1904:209 (Type species: Hemerobius amiculus Fitch, [1855] 1854:799, by monotypy. Etymology: Unexplained, probably Symphero- [from Greek sympheron , useful] + -bius [from Greek bios (masculine), life]. Gender: Masculine.): Sharp, [1908] 1906 (ITSD); Oswald, 1988b (Rev). Spadobius Needham, 1905:16 (Type species: Hemerobius occidentalis Fitch, [1855] 1854:799, by original designation. Note: Carpenter (1940:227) indicates that the H. occidentalis of Needham, 1905, is a misidentification (of H. occidentalis Fitch) conspecific with H. amiculus Fitch, [1855] 1854, the type species of Sympherobius Banks. This is a case of a misidentified type species which may at a later date require submission to the Commission for type designation under ICZN Article 70b. Etymology: Unexplained, probably Spado- [from Latin spadix , frond or palm- tree branch] + -bius [from Greek bios (masculine), life]. Gender: Masculine.): Oswald, 1988b (Syn). Palmobius Needham, 1905:17 (Type species: Hemerobius amiculus Fitch, [1855] 1 854:799, by original designation. Etymology: Unexplained, possibly Palmo- [from Latin palma , hand or palm-tree] + -bius [from Greek bios (masculine), life]. Gen- der: Masculine.): Oswald, 1988b (Syn). Niremberge Navas, 1909a:377 (Type species: Niremberge limpida Navas, 1909a:377 [= Hemerobius fuscescens Wallengren, 1863:22], by monotypy. Etymology: See Navas, 1909a:377; An anagram of the surname of the early Spanish naturalist V. P. Juan Eusebio Nieremberg, S. J. (1590-1658). Gender: Feminine (implied from originally included species).): Navas, [1924b] 1923 (ITSD); Oswald, 1988b (Tax, Syn). NEW STATUS Coloma Navas, 19 15a: 129 (Type species: Megalomus marmoratipennis Blanchard in Gay, 1 85 1 : 1 27, by original designation. Etymology: See Navas, 1 9 1 5a: 1 29; From the surname of R. P. Luis Coloma, S. J. Gender: Feminine.): Oswald, 1988b (Syn). Nefasitus Navas, 1 9 1 5a: 1 3 1 (Type species: Sympherobius amicus Navas, 1 9 1 5b:332 [= Sympherobius fallax Navas, 1908:408], by original designation. Etymology: See Navas, [1924b] 1923:196; From Nefasit, a locality in Eritrea, Ethiopia. Gender: Masculine.): Oswald, 1988b (Syn). Eurobius Kruger, 1922:171 (Type species: Hemerobius elegans Stephens, 1836:1 13, by original designation. Etymology: Unexplained, probably Euro- [from Europe 226 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 82-89. Sympherobius amiculus. 82, Forewing [Scale bar A]. 83, Hind wing [A], 84, Gonarcus, lateral [C]. 85, Gonarcus, dorsal [C]. 86, Male terminalia, lateral [B], 87, Female terminalia, lateral [B], 88, Parabaculum, dorsal [C], 89, Parabaculum. lateral [C], Abbreviations: pmd, pseudomediuncus. Scale bars (mm): A = 2.0; B = 0.5; C = 0.1. or Europa] + -bius [from Greek bios (masculine), life]. Gender: Masculine.): Os- wald, 1988b (Syn). Lachlanius Kruger, 1922:171 (Type species: Hemerobius inconspicuus McLachlan, 1868:177 [^Hemerobius fuscescens Wallengren, 1863:22], by original designation. Etymology: Unexplained, probably from the surname of British entomologist Rob- ert McLachlan (1837-1904). Gender: Masculine.): Oswald, 1988b (Syn). Sympheromima Kimmins, 1928:363 (Type species: Sympheromima marginata Kim- mins, 1928:363, by original designation. Etymology: Unexplained, probably Sym- phero- [from Greek sympheron , useful] + -mima [from Greek mimos, imitator or actor]. Gender: Feminine.): Oswald, 1988b (Syn). 1993 WORLD GENERA OF HEMEROBIIDAE 227 Differential Diagnosis. Small hemerobiids recognized by the forewing character combination (Fig. 82): (1) crossvein 2sc-r absent, (2) anterior radial trace bearing 2 prestigmal “radial sectors” (3 in a few Eurasian spp.), (3) CuP not forked proximal to crossvein 2cua-cup, (4) proximal half of posterior sectoral trace bearing one side branch (simple in a few Eurasian spp.), (5) trichosores evident on humeral margin, and proximal humeral trace prominently recurrent, (6) 4th (outer) gradate series with 3 or more intraradial crossveins, and (7) crossveins 4m-cu and 4im absent. In males, the distinctive bipartite pseudomediuncus (Fig. 84) and processes of the ectoproct (Fig. 86) are diagnostic. Proposed Synapomorphies. [42] Forewing intramedial crossvein 4im lost; [45] forewing mediocubital crossvein 4m-cu lost; [80] base of pseudomediuncus laterally expanded; [81] pseudomediuncus distinctly bipartite. Species (ca. 55). For a recent world listing see Oswald (1988b). Distribution. Widely distributed in the temperate and tropical areas of North and South America, and the temperate areas of Europe, Africa, and Asia. Apparently absent from Australia, India and most or all of tropical southeast Asia. Principle Revisions and Regional Faunas. Alayo, 1968 [Cuba]; Aspock et al„ 1980 [Europe]; Kuwayama, 1962 [Japan]; Makarkin, 1986 [USSR]; Oswald, 1988b [Ne- arctic]; Penny and Monserrat, [1985] 1983 [Amazon Basin]; Stange, 1967 [Argentina and Uruguay]; Tjeder, 1961 [southern Africa]; Zimmerman, 1957 [Hawaiian Islands]. Preimaginal Stages. (1) amiculus — Smith, 1923; (2) californicus—Essig , 1910 (as angustus)\ (3) domesticus— Nakahara, 1954; (4) elegans— Withycombe, 1923; (5) fallax— Bodenheimer, 1930 (as amicus ); (6) fuscescens— Withycombe, 1923 (as in- conspicuus)-, Killington, 1931, 1937; (7) masucocciphagus — Yang, 1980b; (8) pellu- cidus— New, 1967a; (9) pygtnaeus— Withycombe, 1923; Killington, 1937; New, 1967b. Subfamily Psychobiellinae Oswald, new subfamily Psychobiellinae Oswald, new subfamily (Type genus: Psychobiella Banks). Included Genera. Psychobiella. Genus Psychobiella Banks (Figs. 16, 90-97) Psychobiella Banks, 1909:79 (Type species: Psychobiella sordida Banks, 1909:79, by monotypy. Etymology: Unexplained, probably Psycho- [from Greek psyche , breath, life or butterfly] + -bi- [from Greek bios , life] + -ella [from the Latin diminutive suffix -ellus]. Gender: Feminine.): Tillyard, 1916 (RD, A, Key, W*, Tax); New, 1988 (RD, A, Dst, Key, FT*, MT*, W*, Tax). Differential Diagnosis. Distinguished from other Australian hemerobiids by the forewing character combination (Fig. 90): (1) crossvein 2sc-r present, (2) anterior radial trace bearing 3 (occasionally 4) prestigmal “radial sectors”, (3) proximal hu- meral trace prominently recurrent and trichosores evident on humeral margin, and (4) forewing not falcate. Proposed Synapomorphies. [ 1 2] Distal convexity of orad margin of right mandible strongly angulate; [17] ventromedial setae of clypeus lost; [31] forewing with 3 ORB’s, 228 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 90-97. Psychobiella sordida. 90. Forewing [Scale bar A], 91, Hind wing [A], 92. Male terminalia. lateral [B], 93. Female terminalia. lateral [B], 94. Parabaculum. dorsal [C], 95. Parabaculum. lateral [C]. 96. Gonarcus. dorsal [C], 97. Gonarcus. lateral [C]. Abbreviations: cos. costa: ngs. neogonarcus. Scale bars (mm): A = 2.0; B = 0.5: C = 0.2. ORB1 deeply forked: [51] antecosta of male 9th tergite obliquely crossing tergite from anterior to posterior margin; [77] male mediuncus with a transverse ventral costa proximally; [82] male gonosaccal membrane with a scabriculous area below' mediuncus: [87] terminal lobes of parabaculum each composed of a subapical spinose process superposed over a small ventral lobe. Species (2). occidentals New': South Australia: sordida Banks: Tasmania and eastern Australia. Distribution. Australia and Tasmania. Preimaginal Stages. Unknown. Subfamily Notiobiellinae Nakahara (revised) Notiobiellinae Nakahara, 1960b:5 (Type genus: Notiobiella Banks). Differential Diagnosis. Recognized by the forewing character combination: (1) anterior radial trace bearing 2 prestigmal "radial sectors” and (2) crossvein 2sc-r present. OR. if absent (secondarily lost), then trichosores lacking on humeral margin (some Zachobiella) or prestigmal subcostal space no wider than adjacent subcostal vein (many Notiobiella). 1993 229 WORLD GENERA OF HEMEROBIIDAE Proposed Synapomorphies. Notiobiellinae (Lineage 1 1): [40] Forewing radiomedial crossvein 4r-m lost; [42] forewing intramedial crossvein 4im lost; [45] forewing mediocubital crossvein 4m-cu lost. Lineage 12 (Psectra + Anapsectra + Zachobiella ): [7, 8] 5th maxillary and 3rd labial palpomeres simple, bisubsegmentation lost; [11] proximal convexity of orad margin of right mandible strongly angulate; [12] distal convexity of orad margin of right mandible strongly angulate; [36] forewing intraradial crossvein 4irl absent; [53] posteroventral angle of male 9th tergite produced as an elongate subectoproctal lobe, generally with a free distal process; [100] styli of female 9th gonocoxite lost; [103] female gonapophyses posteriores lost; [105] female 8th stemite (subgenitale) lost. Lineage 13 ( Anapsectra + Zachobiella)'. [24] Humeral veinlet recurrent and with only 1 or 2 rami (rami sometimes lost). Included Genera. Notiobiella, Psectra , Anapsectra, and Zachobiella. Distribution. One species is present in temperate North America (possibly intro- duced), Europe, and northern Asia. The remainder range through Central and South America, Africa, southeast Asia, India, Australia, and several islands of the south- western Pacific. Genus Notiobiella Banks (Figs. 4, 98-105) Notiobiella Banks, 1909:80 (Type species: Notiobiella unita Banks, 1909:80, by orig- inal designation. Etymology: Unexplained, probably Notio- [from Greek notios, southern] + -bi- [from Greek bios, life] + -ella [from the Latin diminutive suffix -el/us]. Gender: Feminine.). Vaja Navas, 1925a: 192 (Type species: Vaja tumida Navas, 1925a: 192, by monotypy. Etymology: See Navas, 1925a: 192; An anagram of Java. Gender: Feminine.): Banks, 1932 (Syn). Buxtonia Esben-Petersen, 1928:93 [not Buxtonia Thomas, 1914] (Type species: Bux- tonia fulva Esben-Petersen, 1928:94, by original designation. Etymology: See Es- ben-Petersen, 1928:94; From the surname of British entomologist Patrick Alfred Buxton (1892-1955). Gender: Feminine.): Banks, 1932 (Syn). Ganchetus Navas, 1929a:21 (Type species: Ganchetus africanus Navas, 1929a:22, by monotypy. Etymology: Unknown. Gender: Masculine.): Banks, 1932 (Syn). Differential Diagnosis. Recognized by the forewing character combination (Fig. 98): (1) anterior radial trace bearing 2 prestigmal “radial sectors”, (2) CuP forked proximal to crossvein 2cua-cup, and (3) prestigmal subcostal space no wider than adjacent subcostal vein (sometimes slightly wider adjacent to crossvein lsc-r). In the male, the eversible phallolingua of the gonarcus is diagnostic. Proposed Synapomorphies. [15] Mesolateral setae of clypeus present; [19] mesal lobes of anterior tentorial arms closely adjacent or overlapping; [27] width of forewing subcostal space less than or equal to width of adjacent Sc; [34] most proximal fork of most proximal “radial sector” located in distal two-thirds of posterior sectoral trace; [48] forewing CuP deeply forked, most proximal fork located in proximal half of posterior CuP trace; [83] phallolingua present in male. Species (ca. 35). No recent comprehensive listing is available. Distribution. Widely distributed in Central and South America, Africa, southeast Asia, Australia and some southwestern Pacific islands. 230 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 98-105. Notiobiella multifurcata. ex. New Caledonia (except as noted). 98, Forewing (sp. near multifurcata. ex. Queensland. Australia) [Scale bar A]. 99. Hind wing (sp. near multi- furcata) [A]. 100. Gonarcus (phallolingua retracted, not shown), dorsal [C]. 101, Gonarcus (phallolingua retracted, not shown), lateral [C]. 102. Parabaculum. dorsal [C]. 103, Parabaculum, lateral [C]. 104. Male terminalia. lateral [B]. 105, Female terminalia (tumeri, ex. South Africa), lateral [B], Scale bars (mm): A = 2.0: B = 0.5; C = 0.2. Principle Revisions and Regional Faunas. Alayo, 1968 [Cuba]; Kuwayama, 1962 [Japan]; Monserrat, 1980, 1984a [Africa]: Monserrat and Penny, 1983, Monserrat, 1984b [Neotropical]; New. 1988 [Australia]; New. 1988 [New Guinea]; Penny and Monserrat, [1985] 1983 [Amazon Basin]; Tjeder. 1961 [southern Africa], Preimaginal Stages. Unknown. Genus Psectra Hagen (Figs. 15, 106-1 14) Psectra Hagen, 1866:376, 458 (Type species: Hetnerobius diptera Burmeister, 1839: 973. by monotvpy. Etymology: Unexplained, probably from Greek psektra (fem- inine), scraper or currycomb. Gender: Feminine.). Annandalia Needham, 1909:208 (Type species: Annandalia curia Needham, 1909: 1993 WORLD GENERA OF HEMEROBIIDAE 231 Figs. 106-1 14. Psectra diptera. 106, Forewing [Scale bar A], 107, Hind wing [A], 108, Male terminalia, lateral [B]. 109, Male ectoproct lobe, posterolateral [B], 110, Female terminalia, lateral [C]. Ill, Parabaculum, dorsal [D], 1 12, Parabaculum, lateral [D], I 13, Gonarcus, dorsal [D]. 1 14, Gonarcus, lateral [D], Scale bars (mm): A = 2.0; B = 0.5; C = 0.5; D = 0.1. 208 [=Hemerobius iniquus Hagen, 1859:208], by monotypy. Etymology: Unex- plained, probably from the surname of British zoologist Thomas Nelson Annandale (1876— 1924)]. Gender: Feminine.): Nakahara, 1960b (ITSD); Tjeder, 1961 (Syn). Eucarobius Esben-Petersen, 1928:95 (Type species: Eucarobius fasciatus Esben-Pe- tersen, 1928:95, by original designation. Etymology: Unexplained, probably Eu- [from Greek eu, good or true] + Carobius. Gender: Masculine.): Tjeder, 1961 (Syn). Kimminsiella Nakahara, 1960b: 14. Type species: Annandalia tillyardi Kimmins, 1940:228, by original designation. Etymology: Unexplained, probably Kimmins- [from the surname of British entomologist Douglas Eric Kimmins (1905-1985)] + -iella [from the Latin diminutive suffix -ellus]. Gender: Feminine.): New, 1988 (Syn). Differential Diagnosis. Not distinguishable from Anapsectra on the basis of ve- national traits. Together, Psectra and Anapsectra may be recognized by the forewing character combination (Fig. 106): (1) crossvein 2sc-r present, (2) anterior radial trace bearing 2 prestigmal “radial sectors”, (3) CuP not forked proximal to crossvein 2cua- cup, and (4) some posthumeral costal veinlets branched OR trichosores evident in humeral area. The male ectoprocts of Psectra each possess a small articulated lobe 232 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) bearing a row of terminal modified setae (Fig. 109). The maleectoproctsofHmt/?5e,c7ra each possess a lanceolate process (Fig. 1 1 7). The parabaculi of the two genera are also distinctive. In Psectra the parabacular apex possesses a pair (sometimes fused medially as a single structure) of broad, downtumed, weakly sclerotized plates (Figs. 111-112), in Anapsectra the parabacular apex possesses a pair of erect lanceolate terminal lobes (Fig. 123). Proposed Synapomorphies. [67] Male ectoproct with an articulating posteroventral lobe; [88] ventrodistal region of male parabaculum expanded as a pair of bulbously rounded lobes. Species (ca. 25). No recent comprehensive listing is available. Distribution. One species ( diptera ) is present in temperate eastern North America (possibly introduced) and broadly distributed across Europe and temperate Asia. The remaining species are distributed throughout sub-Saharan Africa, India, the primarily tropical areas of southeast Asia and eastern Australia, and several islands in the southwestern Pacific. Principle Revisions and Regional Faunas. Aspock et al., 1980 [Europe]; Tjeder, 1961 [southern Africa]; Makarkin, 1986 [USSR]; Kuwayama, 1962 [Japan]; New, 1988 [Australia]; New, 1988 [New Guinea]; Carpenter, 1961 [Micronesia]; Monser- rat, 1980 [central Africa]. Preimaginal Stages. (1) diptera — Killington, 1946; New, 1966. Genus Anapsectra Tjeder (Figs. 115-123) Anapsectra Tjeder, 1975:1 15 (Type species: Anapsectra medleri Tjeder, 1975:117, by original designation. Etymology: Unexplained, probably Ana- [from Greek ana , up, back or again] + Psectra. Gender; Feminine.): Monserrat (in press). Differential Diagnosis. See Differential Diagnosis for Psectra. Proposed Synapomorphies. [78] Emargination of apex of male mediuncus lost; [96] female 7th stemite completely divided sagittally; [97] ventral margins of female 9th tergite with chalazate setae; [101] female 9th gonocoxites separated ventrally by a broad sclerotized sulcus; [102] female 9th gonocoxites each with a prominent ventral lobe. Species (1). medleri Tjeder: Nigeria. The description of a second species by V. Monserrat is in press. Distribution. West central Africa. Preimaginal Stages. Unknown. Genus Zachobiella Banks (Figs. 124-131) Zachobiella Banks, 1920:335 (Type species: Zachobiella punctata Banks, 1920:335, by original designation. Etymology: Unexplained, probably Zacho- [origin un- known] + -bi- [from Greek bios, life] + -ella [from the Latin diminutive suffix - ellus ]. Gender: Feminine.). Differential Diagnosis. Recognized by the forewing character combination (Fig. 1993 WORLD GENERA OF HEMEROBI1DAE 233 Figs. 1 15-123. Anapsectra medleri. 1 15, Forewing [Scale bar A], 1 16, Hind wing [A], 1 17, Male terminalia, lateral [B], 1 18, Female terminalia, lateral [B], 1 19, Gonarcus, lateral [C], 120, Gonarcus, dorsal [C], 121, Female 7th stemite, ventral [B]. 122, Parabaculum, dorsal [C], 123, Parabaculum, lateral [C]. Abbreviations: 9gcx, 9th gonocoxite; sul, sulcus; vlb, ventral lobe of 9th gonocoxite. Scale bars (mm): A = 2.0; B = 0.5; C = 0.2. 1 24): ( 1 ) anterior radial trace bearing 2 prestigmal “radial sectors”, (2) all posthumeral costal veinlets simple, and (3) trichosores not evident in humeral area. Proposed Synapomorphies. [23] Trichosores lost proximally on anterior margin of forewing; [25] humeral silhouette not convex; [56] male 9th stemite lost (? fused to 8th stemite); [74] male gonofenestra lost; [86] terminal lobes of parabaculum fused into a single median lobe; [99] female 9th gonocoxites fused to adjacent margins of 9th tergite. Species (8). hainanensis Banks: China; jacobsoni Esben- Petersen: Sumatra; lobata New: Australia; marmorata Navas: Sumatra; pallida Banks: Australia, New Guinea; punctata Banks: Philippines; striata Nakahara: Taiwan; submarginata Esben-Peter- sen: Australia. Distribution. Southeast Asia and Australia. Principle Revisions and Regional Faunas. New, 1988 [Australia]; New, 1988 [New Guinea], Preimaginal Stages. Unknown. General Notes. The only published information available on the male terminalia of Zachobiella is New’s (1988) data on the Australian species pallida , submarginata 234 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 124-131. Zachobiella ? jacobsoni. ex. Cameron Highlands. Malaysia [MCZ] (except as noted). 124, Forewing [Scale bar A]. 125. Hind wing [A]. 126. Female terminalia ( submar - girtata ), lateral [B]. 127, Male terminalia. lateral [B]. 128, Parabaculum. dorsal [C]. 129. Para- baculum. lateral [C]. 130, Gonarcus. dorsal [C], 131. Gonarcus. lateral [C]. Abbreviations: 8s, 8th stemite; 8t, 8th tergite; 9t+9gcx, fused partial 9th tergite and 9th gonocoxite. Scale bars (mm): A = 2.0; B = 0.5: C = 0.1. and Iobata. The male terminalia of these species are abundantly apomorphic and provide good characters for species discrimination. The single male specimen used in this study (from peninsular Malaysia) has been tentatively identified as jacobsoni (described from Sumatra) based on wing characters. It is clearly not conspecific with any of the Australian species figured by New'. Subfamily Drepanacrtnae Oswald, new subfamily Drepanacrinae Oswald, new subfamily (Type genus: Drepanacra Tillyard). Differential Diagnosis. Not easily distinguished as a group on the basis of venational characters. Recognized by either the presence of a prominent gonofenestral plate of the gonarcus ( Conchopterella . Fig. 1 34, gfp), or each gena bearing an elongate field of minute punctulae ( Austromegalomus and Drepanacra , Fig. 5, gpt). Proposed Synapomorphies. Drepanacrinae (Lineage 15): [49] Forewing A1 deeply forked, most proximal fork located in proximal half of posterior A1 trace; [82] male gonosaccal membrane with a scabriculous area below' mediuncus. Lineage 16 ( Austromegalomus + Drepanacra)-. [21] Genae with fields of punctulae; 1993 WORLD GENERA OF HEMEROBI1DAE 235 [51] antecosta of male 9th tergite obliquely crossing tergite from anterior to posterior margin. Included Genera. Conchopterella, Austromegalomus, and Drepanacra. Distribution. Southern South America, southeastern Asia, Australia, New Zealand, and islands of the South Pacific. Genus Conchopterella Handschin (Figs. 132-139) Conchopterella Handschin, 1955:9 (Type species: Conchopterella kuscheli Handschin, 1955:10, by original designation. Etymology: Unexplained, probably Concho- [from Greek konchos, shell] + -pter- [from Greek pteryx, wing] + -ella [from the Latin diminutive suffix - ellus ]. Gender: Feminine.): Oswald, 1988a (Msc). Differential Diagnosis. Recognized by the male terminalic character combination: (1) ectoprocts large and flap-like, with well-sclerotized medial surfaces (Fig. 136), (2) gonarcus with a large gonofenestral plate (Fig. 1 34, gfp), and (3) parabaculum narrow and distinctly arched just before terminal lobes (Fig. 139). Proposed Synapomorphies. [ 1 ] Temporal costa lost or poorly developed; [64] mesal surfaces of male ectoprocts largely sclerotized; [75] gonarcus with a prominent gon- ofenestral plate; [89] apophyseal shaft of parabaculum prominently bowed before terminal lobes. Species (3+). kuscheli Handschin: Juan Fernandez Islands; maculata Handschin: Juan Fernandez Islands; stangei Gonzalez-Olazo (New Combination): Chile and Argentina. One additional undescribed species from southern Chili (NMNH) has been examined. Distribution. Southern South America and the adjacent Juan Fernandez Is. Preimaginal Stages. Unknown. General Notes. The concept of Conchopterella is broadened here. Conchopterella was originally created for two Juan Fernandez Islands species in which the forewings are coriaceous, nearly circular, and somewhat irregularly veined. I have examined two species from mainland South America [stangei (Gonzalez-Olazo), and an un- described species] which are clearly closely related to kuscheli and maculata on the basis of male terminalic characters (ectoprocts, gonarcus, and parabaculum). In the mainland species the forewings are falcate, membranous, and regularly veined. Since the venational characteristics of the Juan Fernandez Islands species are clearly de- rived, having probably evolved since their insular isolation, I regard forewing traits as variable within this genus. Based on the male terminalic characters noted in the cladistic analysis, the concept of Conchopterella is hereby extended to include the two mainland species. Genus Austromegalomus Esben-Petersen (Figs. 140-148) Austromegalomus Esben-Petersen, 1 935b: 1 39 (Type species: Austromegalomus brun- neus Esben-Petersen, 1935b: 140, by original designation. Etymology: Unexplained, probably Austro- [from Latin austrinus, southern] + Megalomus. Gender: Mas- culine.): Esben-Petersen, 1937 (Lst); Handschin, 1955 (Msc); Oswald, 1988a (RD, A, Dst, Key, FT*, MT*, W*, Tax). 236 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 132-139. Conchopterella kuscheli. 132, Forewing [Scale bar A], 133. Hind wing [A]. 134. Gonarcus. dorsal [B]. 135, Gonarcus, lateral [B], 136. Male terminalia. lateral [C]. 137, Female terminalia, lateral [C], 138. Parabaculum. dorsal [B], 139, Parabaculum, lateral [B]. Abbreviations: gfp. gonofenestral plate: sbr. scabriculous region. Scale bars (mm): A = 2.0; B = 0.5; C = 0.5. Differential Diagnosis. Recognized by the character combination: (1) each gena with a field of minute punctulae posteriorly (Fig. 5, gpt) and (2) gonosaccal membrane below mediuncus lacking a scabriculous region (Fig. 143). Proposed Synapomorphies. [57] Male 9th stemite with a small dorsosagittal costa; [82] scabriculous area of male gonosaccal membrane lost. Species (2). brunneus Esben-Petersen: Tahiti Is.; insulanus Oswald: Rapa Is. Distribution. French Polynesia (Tahiti and Rapa Islands). Preimaginal Stages. Unknown. General Notes. The presence of a sagittal costa on the dorsal surface of the male 9th stemite, proposed as a synapomorphy of Austromegalomus. requires confirmation in the type species brunneus. 1993 WORLD GENERA OF HEMEROBIIDAE 237 Figs. 140-148. Austromegalomus insulanus. 140, Forewing [Scale bar A], 141, Hind wing [A] , 142, Gonarcus, dorsal [C], 143, Gonarcus, lateral [C]. 144, Male 9th stemite, dorsal [B], 145, Parabaculum, dorsal [C], 146, Parabaculum. lateral [C], 147, Female terminalia, lateral [B] , 148, Male terminalia, lateral [B], Abbreviations: cos, costa. Scale bars (mm): A = 2.0; B = 0.5; C = 0.2. Genus Drepanacra Tillyard (Figs. 5, 149-156) Drepanacra Tillyard, 1916:293 (Type species: Drepanepteryx humilis McLachlan, 1863:1 16 [= Drepanepteryx binoculus Newman, 1 838:400], by original designation. Etymology: Unexplained, probably Drepan- [from Greek drepane or drepanon , sickle] + -acra [from Greek akra (feminine), highest or farthest point]. Gender: Feminine.): Esben-Petersen, 1 937 (Lst); Nakahara, 1 960b (RD, A, MT*, W*); New, 1988 (RD, A, Dst, FT*, MT*, W* Tax); Oswald, 1988a (Msc). Menopteryx Kruger, 1922:170 (Type species: Megalornus lanceolatus Gerstaecker, [1885] 1884: 1 10 [= Drepanepteryx binoculus Newman, 1838:400], by original des- ignation. Etymology: Unexplained, probably Meno- [from Greek menos, force. 238 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 149-156. Drepanacra binocula. 149, Forewing (Scale bar A], 150, Hind wing [A], 151, Female terminalia, lateral [B], 1 52, Male terminalia, lateral [B], 153, Gonarcus, dorsal [C]. 1 54, Gonarcus, lateral [C]. 155, Parabaculum, dorsal [C], 156, Parabaculum, lateral [C], Abbrevi- ations: gfa, gonofenestra. Scale bars (mm): A = 2.0; B = 0.5; C = 0.2. courage or strength] + -pteryx [from Greek pteryx (feminine), wing]. Gender: Feminine.): Nakahara, 1960b (Syn). Monopteryx [sic] Kruger, 1922:144 (an incorrect original spelling of Menopteryx\ correct original spelling fixed by Oswald and Penny (1991), acting as first revisers). Drepanacrella Kimmins, 1940:222 (Type species: Drepanacrella khasiana Kimmins, 1940:223, by original designation. Etymology: Unexplained, probably Drepan- [from Greek drepane or drepanon, sickle] + -acr- [from Greek akra, highest or farthest point] + -ella [from the Latin diminutive suffix -ellus]. Gender: Feminine.): Ghosh & Sen, 1977 (Lst). NEW SYNONYM Differential Diagnosis. Recognized by the character combination: (1) each gena with a field of minute punctulae posteriorly (Fig. 5, gpt) and (2) forewing falcate (Fig. 149). Proposed Synapomorphies. [37] Forewing falcate; [87] terminal lobes of paraba- culum each composed of a subapical spinose process superposed over a small ventral lobe. 1993 WORLD GENERA OF HEMEROBJIDAE 239 Species (4). binocula (Newman): Australia, Chatham Is., Kermadek Is., Lord Howe Is., New Caledonia (new record, BPBM), New Zealand, Norfolk Is., Tasmania; khas- iana (Kimmins): China (new record, MCZ), India; plaga Banks: Formosa; yunnanica Yang: China. Distribution. Southeastern Asia, Australia, New Zealand, and several smaller is- lands of the southwestern Pacific. Preimaginal Stages. (1) binocula — New, 1975. Synonymical Notes. Drepanacrella. Kimmins (1940:222 — 223) distinguished Dre- panacre/la from Drepanacra on the basis of (1) the configuration of its forewing cubital veins, (2) the highly autapomorphic male terminalia of Drepanacrella khas- iana, and (3) the apparently disjunct ranges of the two genera. The first of these characters is based on a venational misinterpretation (see below); the second character is, in my view, not sufficient to justify its generic separation from Drepanacra (par- ticularly since is removal may leave Drepanacra paraphyletic); and, the third char- acter is discounted by new records of khasiana from mainland China, and the sub- sequent description of other Drepanacra species from Taiwan and mainland China. Kimmins’ misinterpretation of the cubital venation of Drepanacrella stems from his interpretation of “the basal cross- vein between Cul [CuA] and Cu2 [CuP]”. In Drepanacrella, this apparent crossvein is actually the proximal segment of the anterior branch of the most proximal fork of the CuP (an interpretation supported by the presence of setae on this vein and the presence in most wings of a short, true 2cua- cup crossvein linking the anterior branch of the CuP to the CuA). However, Kim- mins’s interpretation is understandable, since the 2cua-cup crossvein in khasiana is often very short or even absent. In the latter case, the anterior branch of the CuP is fused directly to the CuA, and the proximal segment of the anterior branch of the CuA is extremely crossvein-like in appearance. Subfamily Megalominae Kruger (revised) Megalominae Kruger, 1922:170 (Type genus: Megalomus Rambur). Megalomini Navas, [1925b]: 106. Included Genera. Megalomus. Genus Megalomus Rambur (Figs. 3, 8-10, 157-165) Megalomus Rambur, 1 842:4 1 8 (Type species: Megalomus tortricoides Rambur, 1 842: 419, by subsequent designation by Banks, 1905:43. Etymology: See Navas, [1924b] 1923:217; Megal- [from Greek megale, large] + -omus [from Greek omos (mas- culine), shoulder]. Gender: Masculine.): Kruger, 1922 (ITSD). Boriomyia Banks, 1904:209 (Type species: Hemerobius fidelis Banks, 1897:27, by subsequent designation by Killington, 1937:256. Note: Killington’s type designa- tion was upheld by the International Commission on Zoological Nomenclature in Opinion 752 (1965:224). Etymology: Unexplained, probably Borio- [from Greek boreas, north] + -myia [from Greek myia (feminine), fly]. Gender: Feminine.): Banks, 1905 (ITSD). NEW SYNONYM Pleomegalomus Kruger, 1922:170 (Type species: Megalomus pictus Hagen, 1861: 198, by original designation. Etymology: Unexplained, probably Pleo- [from Greek pleo, swim or sail] + Megalomus. Gender: Masculine.): Carpenter, 1940 (Syn). 240 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 157-165. Megalomus tortricoides. 157, Forewing [Scale bar A], 158. Hind wing [A]. 159, Gonarcus, lateral (without mediuncus) [C], 160. Male terminalia, lateral [B]. 161, Female terminalia, lateral [B]. 162. Mediuncus and entoprocessus, lateral [C], 163, Gonarcus. dorsal (mediuncus not shown) [C]. 164. Parabaculum. dorsal [C]. 165. Parabaculum, lateral [C]. Ab- breviations: ent. entoprocessus: med. mediuncus. Scale bars (mm): A = 2.0; B = 0.5: C = 0.2. Pirionus Navas, [1929c] 1928:113 (Type species: Pirionus nigratus Navas, [1929c] 1928:1 14, by monotypy. Etymology: See Navas, [1929c] 1928:1 13; From the sur- name of R. P. Anastasio Pirion, S. J. Gender: Masculine.): Gonzalez-Olazo, 1981 (Tax). NEW STATUS Allotomvia Banks. 1930:224 (Type species: Hemerobius fidelis Banks. 1897:27, by subsequent designation by Oswald and Penny (1991). Etymology: Unexplained, probably Alloto- [from Greek allotrios, of another kind] + -myia [from Greek myia (feminine), fly]. Gender: Feminine.). NEW SYNONYM Spinomegalomus Nakahara, 1965:1 18 (Type species: Spinomegalomus flinti Naka- hara, 1965: 1 19, by original designation. Etymology: Unexplained, probably Spino- [from Latin spina, thorn] + Megalomus. Gender: Masculine.): Gonzalez-Olazo, 1981 (Syn). 1993 WORLD GENERA OF HEMEROBIIDAE 241 Differential Diagnosis. Recognized by either: (1) epipharyngeal surface of labrum bearing two columns of transverse strigae (Fig. 3, str), or (2) male mediuncus ter- minating in a pair of long attenuate processes (Fig. 162). Proposed Synapomorphies. [5] Epipharyngeal surface of labrum with two columns of transverse strigae; [78] mediuncus terminating in a pair slender, attenuate pro- cesses. Species (ca. 40). No recent comprehensive listing is available. Distribution. Widely distributed in North and South America, Europe, extreme northern Africa and much of Asia. Particularly speciose in Central and South Amer- ica. Apparently absent from Australia and sub-Saharan Africa. This poorly studied genus may contain species belonging to other genera, particularly in southeastern Asia. Principle Revisions and Regional Faunas. Alayo, 1968 [Cuba]; Aspock et al., 1980 [Europe]; Carpenter, 1940 [Nearctic]; Gonzalez-Olazo, 1981 [southern South Amer- ica]; Kimmins, 1935 [miscellaneous European and Neotropical species]; Makarkin, 1986 [USSR]; Penny and Monserrat, [1985] 1983 [Amazon Basin]. Preimaginal Stages. (\)fidelis— MacLeod, 1960; (2) hirtus— Killington, 1934, 1936, 1937. Synonymical Notes. Boriomyia and Allotomyia. Most of the early confusion sur- rounding the interpretation of the genus Boriomyia stemmed from controversy over its type species. This nomenclatural question has been resolved by the International Commission on Zoological Nomenclature (ICZN, 1965). There is no question that Boriomyia fidelis and speciosa belong within the genus Megalomus. Such an asso- ciation was suggested by Carpenter (1940) and MacLeod ([1961]) on the basis of both larval and adult characters, but was never formalized in synonymy. The present work fully corroborates this association. Both fidelis and speciosa share the two synapomorphies of Megalomus proposed here [characters 5 and 78], Furthermore, both species possess transverse ridges on the molar surfaces of the mandibles, a character elsewhere observed only in some (mostly New World) Megalomus species. On the basis of this last character, the removal of fidelis and speciosa from Megalomus to Boriomyia would appear to render Megalomus paraphyletic. The holophyly of fidelis + speciosa as a species group within Megalomus is supported by the apparently unique presence in these two species of a long, posteriorly directed, unpaired median lobe of the parabaculum. Allotomyia is a junior objective synonym of Boriomyia. Pirionus. Gonzalez-Olazo (1981) revised the Megalomus species of southern South America and recognized two subgenera, Megalomus ( Megalomus ) and M. (Pirionus). Megalomus ( Pirionus ) was diagnosed principally on the basis of three characters: (1) posterior margin of male 7th tergite prominently produced, (2) male ectoprocts strongly narrowed, and (3) configuration of male gonarcus highly modified. While all three of these characters strongly support M. nigratus + flinti as a holophyletic species group within Megalomus, the corresponding characters in the subgenus (Megalo- mus)— (1) male 7th tergite not prolonged, (2) male ectoprocts not strongly narrowed, and (3) male gonarcus not highly modified— are plesiomorphic, and cannot justify the holophyly of Megalomus (Megalomus). Consequently, neither subgenus is rec- ognized here. 242 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Subfamily Drepanepteryginae Kruger (revised) Drepanepteryginae Kruger, 1 922: 1 70 (Type genus: Drepanepteryx Leach in Brewster). Drepanopterygini Navas, 1933:224. Differential Diagnosis. Large hemerobiids. See Proposed Synapomorphies below. The following character combination is characteristic of the Drepanepteryginae and Megalominae: (1) crossvein 2sc-r present (not uniquely identifiable in Gayomyia), (2) anterior radial trace bearing 4 or more (up to ca. 12) “radial sectors”, (3) humeral area broad and with a well-branched humeral veinlet, and (4) 5th maxillary and 3rd labial palpomeres not subsegmented. Proposed Synapomorphies. Drepanepteryginae (Lineage 19): [1] Temporal costae lost or poorly developed; [41] forewing intramedial crossvein 2im present; [47] fore- wing intercubital crossvein lcua-cup present. Lineage 20 ( Gayomyia + Drepanepteryx ): [18] Anteromesal process of tentorial bridge lost; [49] forewing A 1 deeply forked, most proximal fork located in proximal half of posterior A1 trace; [74] gonofenestra of gonarcus lost. Included Genera. Neuronema , Gayomyia , and Drepanepteryx. Distribution. Southern South America and the Juan Fernandez Is., Europe, and Asia. Genus Neuronema McLachlan (Figs. 166-173) Neuronema McLachlan, 1869:27 (Type species: Hemerobius decisus Walker, 1860: 1 85, by monotypy. Etymology: Unexplained, probably Neuro- [from Greek neuron , nerve] + -nema [from Greek nema (neuter), thread]. Gender: Neuter.). Ninguta Navas, 1912b:420 [not Ninguta Moore, 1892] (Type species: Megalomus deltoides Navas, [1910] 1909:396 [=Hemerobius albostigmata Matsumura, 1907: 171], by original designation. Etymology: Unknown. Gender: Feminine (implied from originally included species).): Kuwayama, 1962 (Syn). Ninga Navas, 1 9 1 3c: 1 22 [a replacement name for Ninguta Navas, 1 9 1 2b; not Ninguta Moore, 1892] (Type species: Megalomus deltoides Navas, [1910] 1909:396 [=Hem- erobius albostigmata Matsumura, 1907:171], by original designation [for Ninguta Navas, 1 9 1 2b]. Etymology: See Navas, 1 9 1 3c: 1 22; A contraction of Ninguta. Gen- der: Feminine.): Kuwayama, 1962 (Syn). Kulinga Navas, 1936:49 (Type species: Kulinga pielina Navas, 1936:50, by mono- typy. Etymology: See Navas, 1936:50; From Kuling, a locality in China. Gender: Feminine.): Yang, 1964 (Syn). Sineuronema Yang, 1964:276, 280 (Type species: Neuronema sinensis Tjeder, 1936: 6, by original designation. Etymology: Unexplained, probably Si- [from Latin Sina, China] + Neuronema. Gender: Neuter.). NEW SYNONYM Differential Diagnosis. A poorly characterized genus of large species restricted to eastern Asia. Distinguished from other east Asian hemerobiids by the presence of one or more forewing pre-3irl crossveins (Fig. 166), or the forewing character com- bination (Fig. 166): (1) crossvein 2sc-r present, (2) anterior radial trace bearing 4 to 6 prestigmal “radial sectors” (rarely more or fewer), (3) costal space with 4 or fewer 1993 WORLD GENERA OF HEMEROBIIDAE 243 Figs. 166-173. Neuronema decisum (except as noted). 166, Forewing [Scale bar C). 167, Hind wing [C], 168, Male terminalia, lateral [D], 169, Female terminalia ( kuwayamai ), lateral [D]. 170, Gonarcus, lateral [A]. 171, Gonarcus, dorsal [A]. 172, Parabaculum, dorsal [B]. 173, Parabaculum, lateral [B]. Abbreviations: med, mediuncus; ngs, neogonarcus; pgs, paleogonarcus. Scale bars (mm): A = 0.5; B = 0.5; C = 2.0; D = 0.5. crossveins (not veinlets), (4) 2nd gradate series very well developed, and (5) forewing not falcate. Proposed Synapomorphies. [35] Forewing pre-3irl intraradial crossveins present. Species (ca. 30). No recent comprehensive listing is available. Distribution. China (mainland and Taiwan) and the adjacent areas of the USSR to the north, to Japan in the east, and India and Nepal to the southwest. Principle Revisions and Regional Faunas. Kimmins, 1943 [miscellaneous species]; Kuwayama, 1962 [Japan]; Makarkin, 1986 [USSR]; Yang, 1964, 1981 [China]. Preimaginal Stages. Unknown. Synonymical Notes. Sineuronema. Yang (1964) diagnosed Sineuronema princi- pally on the basis of four characters: ( 1 ) forewing M with only 2 primary branches, (2) intracubital reach of forewing 3rd (4th?) gradate series terminating on or near 244 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) anterior cubital trace proximal to crossvein 3m-cu, (3) forewing 2nd gradate series not extending posteriorly into anal space, and (4) parabacular cleft at least three- eights of total length of parabaculum. The corresponding characters in Neuronema are: (1) forewing M with at least 3 (usu. 4 or 5) primary branches, (2) intracubital reach of forewing 3rd (4th?) gradate series generally terminating on or near anterior cubital trace proximal between crossveins 3m-cu and 4m-cu, (3) forewing 2nd gradate series extending posteriorly into anal space, and (4) parabacular cleft less than three- eights of total length of parabaculum. In the material of Neuronema and Sineuronema which I have examined, there is considerable variation in these venational characters; and I can find no strong support for maintaining these taxa as separate genera. Because the parabacular apophysis may increase in length with age, the ratio of the depth of the parabacular cleft to total parabacular length would seem to be a particularly unreliable character. The Neuronema/ Sineuronema complex is poorly supported by synapomorphic characters in the present cladistic analysis. However, one of the more striking ter- minalic characters within this complex— the presence of a pair of prominent supra- parabacular processes— is almost certain to be found to be synapomorphic at some level within Neuronema s. lat. The apparent presence of these structures in all Si- neuronema species, and some but not all Neuronema species, suggests that the re- moval of Sineuronema from Neuronema s. lat. would leave Neuronema s. str. para- phyletic. Additional study of this complex of species is needed. General Notes. A convenient diagnostic character for the differentiation of Neu- ronema species from other large eastern Asian hemerobiids (particularly Drepanep- tery’x and Megalomus) is the particularly long R 1 space (Fig. 1 66), which results from the clustering of ORB’s at the proximal end of the anterior radial trace. Genus Gayomyia Banks (Figs. 174-181) Gayomyia Banks, 1913:216 (Type species: Megalomus falcatus Blanchard in Gay, 1851:1 25, by original designation. Etymology: Unexplained, probably Gayo- [from the surname of Chilean natural historian Claudio Gay ( 1 800- 1873)] + -myia [from Greek myia (feminine), fly]. Gender: Feminine.): Stange, 1967 (Lst). Gayomia [sic] Banks, 1913:217 (an incorrect original spelling of Gayomyia ; correct original spelling fixed by Oswald and Penny (1991), acting as first revisers). Porter Navas, 1 924a: 1 5 (Type species: Porter discolor Navas. 1 924a: 1 6 [= Megalomus falcatus Blanchard in Gay, 1851: 125], by monotypy. Etymology: See Navas, 1924a: 1 6; From the surname of Chilean entomologist Carlos Emilio Porter (1868-1 942). Gender: Masculine.): Handschin, 1955 (Syn). Differential Diagnosis. Recognized by the finely reticulate forewing base and the forewing subcostal space containing 10 or more sc-r crossveins which divide the space basally into a single row of cells (Fig. 174). Proposed Synapomorphies. [20] Cranium with numerous setal foveae; [26] costal space finely reticulate proximally; [34] most proximal fork of most proximal “radial sector” located in distal two-thirds of posterior sectoral trace; [35] forewing pre-3irl intraradial crossveins present; [37] forewing falcate; [46] forewing mediocubital flex- 1993 WORLD GENERA OF HEMEROBIIDAE 245 Figs. 174-181. Gayomyia falcata. 174, Forewing [Scale bar A], 175, Hind wing [A], 176, Gonarcus, dorsal [D], 177, Gonarcus, lateral [D], 178, Parabaculum, dorsal [B], 179, Paraba- culum, lateral [B], 180, Male terminally, lateral [C], 181, Female terminally, lateral [C]. Ab- breviations: cos, costa; pen, penniform sclerite; spn, suprapenniform sclerite; sv, sigmoid vein. Scale bars (mm): A = 2.0; B = 0.5; C = 0.5; D = 0.2. 246 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) ion line lost; [50] male 9th tergite sagittally divided; [71] inner surface of each intrahemigonarcus with a prominent horizontal costa; [90] parabaculum bearing a pair of penniform accessory sclerites; [91] gonosaccal membrane immediately dorsal to penniform sclerites containing a triangular, weakly sclerotized, area; [106] female bursa broadly, but weakly, sclerotized dorsally. Species (3). cinerea Kruger: Chile (?); falcata (Blanchard in Gay); Argentina, Chile, Juan Fernandez Is., Peru; stictica (Blanchard in Gay); Chile. The taxonomic place- ment and validity of cinerea and stictica require confirmation. Distribution. Southern South America and the Juan Fernandez Is. Preimaginal Stages. Unknown. Genus Drepanepteryx Leach in Brewster (Figs. 182-190) Drepanepteryx Leach in Brewster, 1815:138 (Type species: Hemerobius phalaenoides Linnaeus, 1758:550, by monotypy. Etymology: See Navas, [1924b] 1923:221; Drepane- [from Greek drepane or drepanon , sickle] + -pteryx [from Greek pteryx (feminine), wing]. Gender: Feminine.): Tjeder, 1963b (RD. A. Dst, MT*, FT*. Tax); Kuwayama, 1962 (Dst, Tax); Aspock et al., 1980 (Dst, MT*. FT*, Tax); Makarkin, 1986 (Dst, Tax). Drepanopteryx [sic]; Burmeister. 1839:975 (an incorrect subsequent spelling of Dre- panepteryx). Note: Burmeister’ s use of Drepanopteryx does not appear to be “de- monstrably intentional” (ICZN. 1985, Art. 33b(i)). Under a strict interpretation of the rules of the Code, it should be considered an unavailable incorrect subsequent spelling, not an available emendation, of Drepanepteryx. See also Drepanopteryx Agassiz below. Drepanopteryx Agassiz, [1847] 1842-1846:130 (an unjustified emendation of Dre- panepteryx). Note: Agassiz was apparently the first author to satisfy the “demon- strably intentional” criterion (ICZN, 1985, Art. 33b(i)), and therefore make the name Drepanopteryx available as an emendation. See also the Drepanopteryx of Burmeister above. Canisius Navas, 191 3a: 512 (Type species: Hemerobius algidus Erichson in Midden- dorff, 1851:69, by original designation. Etymology: See Navas, 191 3a: 513; From the surname of B. P. Pedro Canisio, S. J. Gender: Masculine.): Tjeder, 1963b (Syn). Oedobius Nakahara, 1915:44 (Type species: Oedobius infalcatus Nakahara, 1915:44 [=Megalomus punctatus Okamoto, 1 905: 1 14], by original designation. Etymology: Unexplained, probably Oedo- [from Greek oidos, swelling or tumor] + -bius [from Greek bios (masculine), life]. Gender: Masculine.): Kuwayama, 1962 (ITSD); Tjed- er, 1963b (Syn). Phlebonema Kruger, 1922:170 (Type species: Hemerobius algidus Erichson in Mid- dendorff. 1851:69, by original designation. Etymology: Unexplained, probably Phlebo- [from Greek phlebos, vein] + -nema [from Greek nema (neuter), thread]. Gender: Neuter.): Aspock et al., 1980 (Syn). Bestreta Navas, 1924c:222 (Type species: Bestreta iaponica Navas, 1924c:222 [ =Me - galomus punctatus Okamoto, 1905:1 14], by monotypy. Etymology: Unexplained, possibly from Catalonian bestreta (feminine), anticipation or advance. Gender: Feminine.): Kuwayama, 1962 (Syn). 1993 WORLD GENERA OF HEMEROBIIDAE 247 Figs. 182-190. Drepanepteryx phalaenoides. 182, Forewing [Scale bar C], 183, Hind wing [C], 184, Gonarcus, dorsal [A], 185, Gonarcus, lateral [A], 186, Accessory sclerite, posterior [F], 187, Male terminalia, lateral [E] . 188, Female terminalia, lateral [B], 189, Parabaculum, dorsal [D], 190, Parabaculum and accessory sclerite, lateral [D]. Abbreviations: acs, accessory sclerite. Scale bars (mm): A = 0.5; B = 1 .0; C = 2.0; D = 0.5; E = 0.5; F = 0.2. Differential Diagnosis. Recognized by the forewing character combination (Fig. 182); (1) crossvein 2sc-r present, (2) anterior radial trace bearing 8 or more (up to ca. 13) “radial sectors”, and (3) costal space with a gradate series of 5 or more crossveins. Proposed Synapomorphies. [26] Forewing costal space containing a gradate series with 5 or more crossveins; [93] gonosaccal accessory sclerite present; [100] styli of female 9th gonocoxite lost. Species (6). algida (Walker): Europe, USSR; falculoides Walker: Hindostan;/w.sr<2/<2 Nakahara: Japan; phalaenoides (Linnaeus): Canary Islands, China, Europe, Japan, USSR; pleshanovi Makarkin: eastern USSR; punctata (Okamoto): Japan, eastern USSR, Kuril Is. Distribution. Widely distributed in Eurasia from western Europe east to eastern USSR and Japan. Preimaginal Stages. (1) phalaenoides— Von Gleichen, 1770; Brauer, 1867; Morton, 1910; Killington, 1937; Fulmek, 1941. 248 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Subfamily Microminae Kruger (revised) Microminae Kruger, 1922:171 (Type genus: Micromus Rambur). Micromini Navas, [1925b]: 104. Micromidae Banks, 1939:466. Differential Diagnosis. Recognized by the forewing character combination: (1) anterior radial trace bearing 4 or more (up to ca. 10, very rarely 3) prestigmal "radial sectors”, (2) humeral margin lacking trichosores, its shape linear or shallowly concave (not distinctly convex). The majority of species (all except Noius) share the additional characters: (1) forewing crossvein 2m-cu absent, (2) forewing crossvein lcua-cup present, and (3) 9th tergite dorsosagittally divided, each half more or less fused to adjacent margin of ipsilateral ectoproct. Proposed Synapomorphies. Microminae (Lineage 21): [2] Transtorular costa well developed: [3] midfrontal costa well developed: [9] cervical margin of postmentum well defined and regularly parabolic; [23] trichosores lost proximally on anterior margin of forewing; [24] proximal humeral trace recurrent but with only 1 or 2 rami (rami and recurrency frequently secondarily lost); [25] humeral silhouette not convex; [34] most proximal fork of most proximal “radial sector” located in distal two-thirds of posterior sectoral trace; [100] styli of female 9th gonocoxite lost. Lineage 22 (Arusalala + Megalomina -I- Micromus ): [1] temporal costae lost or poorly developed; [28] forewing subcostal crossvein 2sc-r lost; [43] forewing mediocu- bital crossvein 2m-cu lost; [47] forewing intercubital crossvein lcua-cup present; [48] forewing CuP simple or shallowly forked, most proximal fork located in distal half of posterior CuP trace; [50] male 9th tergite sagittally divided; [55] male 9th and 1 0th (ectoprocts) tergites fused. Lineage 23 ( Megalomina + Micromus ): [74] gonofenestra of gonarcus lost; [107] female insemination-fertilization canal with a small distal constriction and lobe. Included Genera. Noius , Nusalala. Megalomina, and Micromus. Distribution. Cosmopolitan. Widespread on all continents, except Antarctica. Well represented in the island faunas of the Atlantic, Indian, and Pacific oceans. Genus Noius Navas (Figs. 191-198) Noius Navas. 1 929e: 5 1 (Type species: Noius oceanicus Navas, 1929e:51, by mono- typy. Etymology: Unexplained, possibly from Catalonian noi , (masculine or fem- inine), child or baby. Gender: Masculine.): Kimmins, 1958 (OD). Differential Diagnosis. Recognized by the forewing character combination (Fig. 191): (1) crossvein 2sc-r present, but often weak, (2) anterior radial trace bearing 4 to 8 (rarely 9) prestigmal "radial sectors”, and (3) crossveins 1 m-cu and 2m-cu closely adjacent, separated only by approximately their combined lengths. Proposed Synapomorphies. [4] Frontoclypeal costa strongly developed, transversely continuous across frontoclypeus; [11] proximal convexity of orad margin of right mandible strongly angulate; [18] anteromesal process of tentorial bridge lost; [60] apex of male 9th stemite produced as a digitiform process; [62] male 9th stemite bearing a pair of posterolateral lobes; [69] extragonopons sagittally emarginate; [103] female gonapophyses posteriores lost. 1993 WORLD GENERA OF HEMEROBIIDAE 249 Figs. 191-198. Noius oceanicus. 191, Forewing [Scale bar A], 192, Hind wing [A]. 193, Female terminalia, lateral [B], 194, Male terminalia, lateral [B], 195, Gonarcus, dorsal [C], 196, Gonarcus, lateral [C], 197, Parabaculum, dorsal [C], 198, Parabaculum, lateral [C], Scale bars (mm): A = 2.0; B = 0.5; C = 0.2. Species {2 noumeanus Kimmins: NewCaledonia, New Hebrides Is. (new record, S & 9, BPBM); oceanicus Navas: Fiji Is. (verified, <3 & 9, BPBM). I have also examined an undescribed species from Upolu Is., Western Samoa (BPBM). Distribution. Islands of the southwestern Pacific, including: Fiji Is., New Caledonia, New Guinea (new record, 2 only, BPBM), New Hebrides Is., Solomon Is. (new record, 2 only, BPBM), Western Samoa (new record, 6 & 2, BPBM), and American Samoa (new record, 2 only, BPBM). Preimaginal Stages. Unknown. Genus Nusalala Navas (Figs. 199-206) Nusalala Navas, 1913b:74 (Type species: Nusalala erecta Navas, 191 3b:75, by mono- typy. Etymology: Unknown. Gender: Feminine (implied from originally included species).). 250 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 199-206. Nusalala dispar (except as noted). 199, Forewing [Scale bar B], 200, Hind wing [B], 201. Gonarcus, lateral [C], 202, Gonarcus, dorsal [C], 203, Male terminalia, lateral [A], 204, Female terminalia ( championi ), lateral [A]. 205, Parabaculum, dorsal [C], 206, Para- baculum, lateral [C], Abbreviations: lbc. laterobaculum. Scale bars (mm): A = 0.5; B = 2.0; C = 0.2. Haarupiella Esben-Petersen, 1914:263 (Type species: Haarupiella neotropica Esben- Petersen, 1 9 1 4:264, by original designation. Etymology: See Esben-Petersen, 1914: 263; Haarup- [from the surname of Danish entomologist Anders Christian Jensen- Haarup (1863-1934)] + -iella [from the Latin diminutive suffix -ellus]. Gender: Feminine.): Kimmins, 1936a (Syn). Palaeomicromus Kruger, 1922:171 (Type species: Palaeomicromus schmidti Kruger, 1 922: 1 7 1 , by original designation. Etymology: Unexplained, probably Palaeo- [from Greek palaios, ancient or old] + Micromus. Gender: Masculine.): Nakahara, 1965 (Syn). Nesbe Navas, 1929d:320 (Type species: Nesbe irrebita Navas, 1929d:321, by mono- typy. Note: Navas (1935:58) emended irrebita to irretita. However, since there is no clear evidence within the original publication itself of an unintentional error (the etymology of the specific name was not originally stated), under a strict in- terpretation of the Code (ICZN, 1985, Arts. 32 and 33), the latter spelling should be considered an unjustified emendation, regardless of Navas’s (1935) statement 1993 WORLD GENERA OF HEMEROBIIDAE 251 that irretita was the intended spelling. Etymology: See Navas, 1929d:320; An anagram of “Esben”, from the surname of Danish entomologist Peter Esben- Petersen (1869— 1942). Gender: Feminine.): Navas, 1935:58 (Nom). NEW SYN- ONYM Differential Diagnosis. Distinguished from other South and Central American hem- erobiids by the forewing character combination (Fig. 1 99): (1 ) base of M3 + 4 forming a short, oblique, vein proximal to its fusion with CuA and (2) CuA simple proximal to its fusion with M3 + 4. In sympatric Micromus species, vein CuA is either forked proximal to its fusion with M3 + 4, or not fused to M3 +4. The presence of a pair of laterobacula associated with the parabaculum (Fig. 205, lbc) is also diagnostic. Proposed Synapomorphies. [44] Forewing mediocubital crossvein 3m-cu lost; [92] male laterobaculum present; [106] female bursa with prominently sclerotized lateral walls. Species (ca. 20). For a recent world listing see Penny, [1978]. Distribution. Central and South America and the Antilles. Principle Revisions and Regional Faunas. Alayo, 1968 [Cuba]; Kimmins, 1936a [miscellaneous Neotropical species]; Penny and Monserrat, [1985] 1983 [Amazon Basin]; Stange, 1967 [Argentina and Uruguay]. Preimaginal Stages. Unknown. Synonymical Notes. Nesbe. Navas apparently described Nesbe and its type species, irrebita , from a single specimen formerly in the Hamburg Museum (now presumed destroyed; see Weidner, 1972). The original descriptions of these taxa clearly support their placement in the Microminae, almost certainly in the genus Nusalala. The following passages concerning forewing characters support placement in the Microm- inae [free translations in brackets]: (1) “Ala anterior area costali basi angusta” [fore- wing costal space narrow basally]; (2) “sine ramo recurrente” [no recurrent vein, i.e., proximal humeral trace not recurrent]; (3) “radio fere 5 sectoribus” [anterior radial trace with five “radial sectors”]; (4) “area cubitali elongata, 2 venulis divisa” [inter- cubital space elongate, with 2 intercubital crossveins]. Three additional forewing characters support the synonymy of Nesbe with Nusalala'. (1) “venulis gradatis in tres series” [three post-subcostal gradate series], (2) “procubito. . .ramo posteriore brevi, cum cubito fuso” [posterior branch of media (i.e., M3 + 4) short and fused to cubitus], and (3) “venulis plerisque ad costam furcatis aut ramosis; medio biareolata, venulis pluribus serie venularum gradatarum connexis” [subcostal veinlets numerous and forked or branched, intervening spaces medially biareolate due to interconnecting crossveins of a costal gradate series]. The type locality of irrebita (San Jose, Costa Rica) is also consistent with its placement in Nusalala ; and irrebita may be synon- ymous with Nusalala kruegeri Nakahara (1965; Type Locality: Cordoba, Mexico), which also possesses a costal gradate series. Genus Megalomina Banks (Figs. 207-214) Megalomina Banks, 1909:78 (Type species: Megalomina acuminata Banks, 1909: 78, by original designation. Etymology: Unexplained, probably Megal- [from Greek megale, large] + -om- [from Greek omos, shoulder] + -ina [from the Latin di- minutive suffix -ina]. Gender: Feminine.): New, 1988; New, 1988. Drepanomina Tillyard, 1916:302 (Type species: Drepanomina gibbosa Tillyard, 1916: 252 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Figs. 207-214. Megalomina berothoides (except as noted). 207, Forewing ( bridwelli ) [Scale bar A]. 208, Hind wing ( bridwelli ) [A], 209, Parabaculum, dorsal [C], 210, Parabaculum, lateral [C], 211, Gonarcus. dorsal [C]. 212, Gonarcus, lateral [C]. 213, Male terminalia, lateral [B], 214, Female terminalia (bridwelli), lateral [B], Abbreviations: ngs, neogonarcus. Scale bars (mm): A = 2.0; B = 0.5; C = 0.2. 303 [=Drepanepteryx berothoides McLachlan, 1869:22], by original designation. Etymology: Unexplained, probably Drepan- [from Greek drepane or drepanon, sickle] + -om- [from Greek omos, shoulder] + -ina [from the Latin diminutive suffix -ina]. Gender: Feminine.): New, 1988 (Syn). Oxybiella Tillyard, 1916:305 (Type species: Oxybiella bridwelli Tillyard, 1916:305, by original designation. Etymology: Unexplained, probably Oxy- [from Greek oxys, sharp] + -bi- [from Greek bios , life] + -ella [from the Latin diminutive suffix -el/us]. Gender: Feminine.): New, 1988 (Syn). Differential Diagnosis. Distinguished from other Australian and New Guinean hemerobiids by the forewing character combination (Fig. 207): (1) crossveins 2sc-r and 2m-cu absent, (2) crossveins lcua-cup and 2cua-cup present, and (3) proximal humeral trace prominently recurrent and multiply branched. In sympatric Micromus species the humeral veinlet is simple, or at most possesses a short inconspicuous recurrent branch. Proposed Synapomorphies. [2] Transtorular costa lost or poorly developed; [3] midfrontal costa lost or poorly developed; [9] cervical margin of postmentum poorly defined and not regularly parabolic; [24] proximal humeral trace recurrent and usually with 3 or more rami; [78] apex of male mediuncus emarginate. 1993 WORLD GENERA OF HEMEROBIIDAE 253 Figs. 2 1 5-222. Micromus variegatus. 215, Forewing [Scale bar A]. 2 1 6, Hind wing [A], 2 1 7, Parabaculum, dorsal [C]. 218, Parabaculum, lateral [C]. 219, Gonarcus, dorsal [C]. 220, Gonar- cus, lateral [C]. 221, Female terminalia, lateral [B], 222, Male terminalia, lateral [B]. Abbre- viations: ngs, neogonarcus; ngsp, neogonarcal process. Scale bars (mm): A = 2.0; B = 0.5; C = 0.1. Species (3). acuminata Banks: eastern Australia, New Guinea; berothoides (Mc- Lachlan): southeastern Australia; bridwelli (Tillyard): Australia, New Guinea (new record, 2, BPBM). Distribution. Australia and New Guinea. Preimaginal Stages. Unknown. Genus Micromus Rambur (Figs. 1 1, 215-222) Micromus Rambur, 1842:416 (Type species: Hemerobius variegatus Fabricius, 1793: 85, by subsequent designation by Banks, 1905:44. Etymology: See Navas, [1924b] 1923:212; Micro- [from Greek mikros, small] + -mus [from Greek omos (mas- culine), shoulder]. Gender: Masculine.). Micromerus [sic] Hagen, 1858:483 (an incorrect subsequent spelling of Micromus). Nesomicromus Perkins, 1899:37 (Type species: Nesomicromus vagus Perkins, 1899: 37, by subsequent designation by Zimmerman, 1940:505. Etymology: Unex- plained, probably Neso- [from Greek nesos , island] + Micromus. Gender: Mas- culine.): Tjeder, 1961 (Tax); Klimaszewski and Kevan, 1988 (Tax, Syn). NEW STATUS 254 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Pseudopsectra Perkins, 1899:46 (Type species: Pseudopsectra lobipennis Perkins, 1899: 46. by monotypy. Etymology: Unexplained, probably Pseudo- [from Greek pseu- dos, lie] + Psectra. Gender: Feminine.): Tjeder, 1961 (Syn). Nesothauma Perkins, 1899:46 (Type species: Nesothauma haleakalae Perkins, 1899: 47 [currently a secondary homonym of Nesomicrotnus haleakalae Perkins, 1899: 42], by monotypy. Etymology: Unexplained, probably Neso- [from Greek nesos, island] -I- -thauma [from Greek thauma , wonder or marvel]. Gender: Feminine.): Tjeder, 1961 (Syn). Nenus Navas, 1 9 1 2a: 1 99 (Type species: Nenus longulus Navas. 1 9 1 2a:200 [=Micro- mus multipunctatus Matsumura. 1907: 1 7 1], by monotypy. Etymology: See Navas, 1 9 1 2a: 1 99; From Catalonian nen (masculine or feminine), infant or baby. Gender: Masculine.): Nakahara. 1960b (Syn); Monserrat (in press) (Tax). Nemis [sic] Banks, 1913:217 (an incorrect subsequent spelling of Nenus). Eumicronius Nakahara, 1915:36 (Type species: Micromus numerosus Navas, [1910] 1909:396, by original designation. Etymology: Unexplained, probably Eu- [from Greek eu. good or true] + Micromus. Gender: Masculine.): Carpenter, 1940 (Syn). Eumicronus [sic] Nakahara. 1915:37, 39 (an incorrect original spelling of Eumicro- mus ; correct original spelling fixed by Oswald and Penny (1991), acting as first revisers). Paramicromus Nakahara. 1919:137 [not Paramicromus Kruger, 1 922] (Type species: Eumicronius dissimilis Nakahara, 1915:43, by original designation. Etymology: Unexplained, probably Para- [from Greek para, beside or near] + Micromus. Gender: Masculine.): Tjeder, 1961 (Syn). Archaeomicromus Kruger, 1922:171. Type species: Micromus timidus Hagen, 1853: 481, by original designation. Etymology: Unexplained, probably Archaeo- [from Greek archaios, old or primitive] + Micromus. Gender: Masculine.): Kimmins, 1936b (Syn). Indomicromus Kruger. 1922:171 (Type species: Micromus australis Hagen, 1858: 483 [in Micromerus (sic)], by original designation. Etymology: Unexplained, prob- ably Indo- [from Latin Indicus, of India] + Micromus. Gender: Masculine.): Tjeder, 1961 (Syn). Stenomicromus Kruger, 1922:171 (Type species: Hemewbius paganus Linnaeus, 1767:912, by original designation. Etymology: Unexplained, probably Steno- [from Greek stenos, narrow] + Micromus. Gender: Masculine.): Tjeder, 1961 (Syn). Heteromicromus Kruger, 1922:171 (Type species: Heteromicromus audax Kruger, 1922:171, by original designation. Etymology: Unexplained, probably Hetero- [from Greek heteros, different] + Micromus. Gender: Masculine.): Tjeder, 1961 (Syn). Neomicromus Kruger, 1922:154 (Type species: Micromus tessellatus Gerstaecker, [1888] 1 887:129, by original designation. Etymology: Unexplained, probably Neo- [from Greek neos , new] + Micromus. Gender: Masculine.): Tjeder, 1961 (Syn [as Meomicromus (sic)]). Meomicromus [sic] Kruger, 1922: 1 72 (an incorrect original spelling of Neomicromus', correct original spelling fixed by Oswald and Penny (1991), acting as first revisers). Pseudomicromus Kruger, 1922:172 (Type species: Hemerobius angulatus Stephens, 1836:106, by original designation. Etymology: Unexplained, probably Pseudo- [from Greek pseudos, lie] + Micromus. Gender: Masculine.): Imperial Bureau of Entomology. [1924] 1922 (ITSD); Tjeder, 1961 (Syn). 1993 WORLD GENERA OF HEMEROBI1DAE 255 Paramicromus Kruger, 1 922: 172 [not Paramicromus Nakahara, 1919] (Type species: Micromus insipidus Hagen, 1 86 1 : 1 99 [=Micromus posticus Walker, 1853:283], by original designation. Etymology: Unexplained, probably Para- [from Greek para, beside or near] + Micromus. Gender: Masculine.): Carpenter, 1940 (Syn). Stenomus Navas, 1 922:55 [not Stenomus Rafinesque, 1815] (Type species: Stenomus nesaeus Navas, 1922:55 [=Micromus africanus Van der Weele, 19 10c: 17, NEW SYNONYM (holotype female examined, MNHP), by monotypy. Etymology: Unexplained, probably Steno- [from Greek stenos, narrow] + -mus [from Greek omos (masculine), shoulder]. Gender: Masculine.). NEW SYNONYM Phlebiomus Navas, [1923] 1922:24 (Type species: Phlebiomus yunnanus Navas, [1923] 1922:25, by monotypy. Etymology: See Navas, [1923] 1922:24; Phlebio- [from Greek ph/ebion, veinlet] + -mus [from Greek omos (masculine), shoulder]. Gender: Masculine.): Nakahara, 1960b (Syn). Tanca Navas, 1929b:373 (Type species: Tanca loriana Navas, 1929b:374, by mono- typy. Etymology: Unexplained, probably from Catalonian tanca (feminine), fence. Gender: Feminine.): Tjeder, 1961 (Syn). Menutus Navas, 1932:35 (Type species: Micromus haitiensis Smith, 1931:800, by original designation. Etymology: Unexplained, probably from the Catalonian ad- jective menut, little or small. Gender: Masculine.): Alayo, 1968:22 (Msc). NEW SYNONYM Idiomicromus Nakahara, 1955:8 (Type species: Idiomicromus kanoi Nakahara, 1955: 8, by original designation. Etymology: Unexplained, probably Idio- [from Greek idios, one’s own or personal] + Micromus. Gender: Masculine.): Nakahara, 1960b: 37 (Tax); Yang, 1981:303, 316 (NS, Tax); Monserrat (in press) (Syn). Spilomicromus Nakahara, 1960b:26 (Type species: Eumicromus maculatipes Na- kahara, 1915:39 [in Eumicronus (sic)], by original designation. Etymology: Unex- plained, probably Spilo- [from Greek spilos, spot] + Micromus. Gender: Mascu- line.): Tjeder, 1961 (Syn). A nomicromus Nakahara, 1960b: 30. (Type species: Nesomicromus paradoxus Perkins, 1899:39, by original designation. Etymology: Unexplained, probably Ano- [from Greek ano-, up] + Micromus. Gender: Masculine.): Tjeder, 1961 (Syn). Ameromicromus Nakahara, 1960b:33 [an objective replacement name for Parami- cromus Kruger, 1922; not Paramicromus Nakahara, 1919] (Type species: Micro- mus insipidus Hagen, 1861:199 [=Micromus posticus Walker, 1853:283], by orig- inal designation (for Paramicromus Kruger, 1 922. Note: The type species originally named by Nakahara, Micromus posticus, is nomenclaturally incorrect. A replace- ment genus-group name automatically assumes the type species of the name it replaces (ICZN, Article 67h). Etymology: Unexplained, probably Amero- [from America] + Micromus. Gender: Masculine.): Carpenter, 1940 (Syn). Afromicromus Nakahara, 1960b:34 (Type species: Micromus capensis Esben-Peter- sen, 1920:508, by original designation. Etymology: Unexplained, probably Afro- [from Africa] + Micromus. Gender: Masculine.): Tjeder, 1961 (Syn). Austromicromus Nakahara, 1960b:35. (Type species: Hemerobius tasmaniae Walker, 1860:186, by original designation. Etymology: Unexplained, probably Austro- [from Latin austrinus, southern] + Micromus. Gender: Masculine.): Tjeder, 1961 (Syn). Mixomicromus Ghosh, 1977:235 (Type species: Mixomicromus lampus Ghosh, 1977: 235, by original designation. Etymology: Unexplained, probably Mixo- [from Greek 256 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) mixis or Latin mixtus, to mix or mingle] + Micromus. Gender: Masculine.). NEW SYNONYM Differential Diagnosis. The forewing characters (1) crossveins 2sc-r and 2m-cu absent. (2) crossveins lcua-cup and 2cua-cup present, and (3) humeral area narrow, its margin lacking trichosores. will distinguish .\ficromus from all genera except Xusalala and Megalomina. The diagnoses of these genera should be consulted for characters to differentiate them from species placed in Micromus. See also General Xotes below. Proposed Synapomorphies. None. Species (ca. 95). No recent comprehensive listing is available. Distribution. Nearly cosmopolitan. Widespread in North and Central America. Europe. Africa. Asia, and Australia. Present also on numerous islands in the Atlantic, Indian and Pacific oceans. A notably diverse endemic fauna exists in the Hawaiian Islands. The taxonomic placement of the few species reported from South America requires confirmation; they may belong to Xusalala. Principle Revisions and Regional Faunas. Alayo. 1968 [Cuba]: Aspock et al., 1980 [Europe]; Klimaszewski and Kevan. 1988 [Nearctic]: Kuwayama. 1962 [Japan]; Mak- arkin. 1985 [USSR]: New. 1988 [Australia]: New, 1988 [New Guinea]: Tjeder. 1961 [southern Africa]: Zimmerman. 1957 [Hawaiian Islands]. Preimaginal Stages. (1) angulatus— Aubvook. 1935 (as aphidivorus ); Killington. 1936: Miermont and Canard. 1975; (2) numerosus— Nakahara. 1954: Kawashima. 1958: (3) paganus— -Withycombe. 1923: Killington. 1936 (in Eumicromus): (4) pos- ticus— Cutright. 1923: Smith. 1923: Miller and Cave. 1987; (5) tasmaniae— New and Boros, 1983: (6) vagus — Terry, 1908 (in .Xesomicromus); (7) variegatus— Brauer. 1871; Withycombe, 1924; Killington. 1936; Kimmins. 1939: Dunn. 1954. Synonymical .Xotes. Stenomus. Navas’ (1922) description of the genus Stenomus contains no characters which would exclude it from the broad concept of the genus Micromus adopted here. I have examined the female holotype of 5. nesaeus in the Paris Museum, and I propose it here as a new junior subjective synonym of Micromus africanus Van der Weele (1910c). Menutus. Smith's (1931: fig. 10) illustration of the forewing of haitiensis clearly shows it to be a small micromine species. Alayo (1968) reported the common North American species Micromus subanticus from the Greater Antilles [Cuba], and cited “ Micromus haitiensis ” as a possible synonym. I have seen the holotype and single paratype of haitiensis in the MCZ and they clearly fall within the broad Micromus concept adopted here. However, since the holotype of haitiensis is a female, and the paratype is missing its abdomen, additional males from Haiti are needed before the synonymical status of subanticus and haitiensis can be determined. Mixomicromus. In distinguishing Mixomicromus from Micromus Ghosh (1977: 236) states: the "Rs with 4 branches: two regularly arranged [series of] gradate cross- veins. absence of recurrent humeral veinlets in the forewings, elongated anal plate with long ventral process, very elongated, tubular 9th stemite produced beyond the apex of the anal plate and the genitalia of the male are some unique features to distinguish it [ Mixomicromus ] as a distinct genus”. Having examined more than 30 distinct species of Micromus from all regions of the globe (including specimens of 1993 WORLD GENERA OF HEMEROBIIDAE 257 lampus from India), the only character cited by Ghosh which is unusual within Micromus is the narrowly produced 9th stemite. In all other respects, lampus falls well within the broad concept of Micromus adopted here, and given the terminalic variability within Micromus , the modified 9th stemite is not sufficient to justify its ranking as a separate genus. General Notes. Characterization and subdivision of the large and widespread genus Micromus has long been a problem (see discussions in Tjeder, 1961, and Klimaszew- ski and Kevan, 1988), and is reflected in its extensive list of generic synonyms. The fact that no synapomorphies are known for the genus further attests to its intracta- bility. A worldwide revision of this genus will be necessary before any stable intra- generic groupings can be established. Such a revision should also include the closely related genera Nusalala and Megalomina , which represent potential outgroups (or, possibly, ingroups). Until such a revision has been completed, it is strongly recom- mended that interim intrageneric groupings be expressed in the literature as “species- groups”, not as formal genera or subgenera. Unplaced Genera Genus Notherobius New Notherobius New, 1988:358 (Type species: Notherobius nothofagi New, 1988:359, by original designation. Etymology: Unexplained, probably Noth- [from Greek nothos, spurious] + -erobius [from (Hem)erobius], Gender: Masculine.). Status. Apparently valid. Species (3). hastatus New: southeastern Australia; nebulosus New: southeastern Australia; nothofagi New: southeastern Australia, Tasmania. Distribution. Australia. General Notes. New (1988) proposed this genus for three newly described species from Australia. Unfortunately, none of these have been available for examination during the present study. Notherobius species are clearly distinct from other Austra- lian hemerobiids on the basis of forewing and male terminalic characters, and the proposal of a new genus for them appears justified. The placement of this genus in the Hemerobiidae is confirmed by the presence of a small, but distinct, parabaculum in the male, and the presence of two ORB’s branching from the anterior radial trace. A preliminary assessment of the cladistic position of Notherobius suggests that it would join the preferred cladogram as a fourth lineage arising from the trichotomy formed by lineages 4, 7, and 9 (see Fig. 23). This position is inferred from the absence in Notherobius of those synapomorphies of lineages 4, 7, 9, and 10 interpretable from New’s treatment (9 of 12 applicable characters), and the presence of two synapo- morphies of lineage 3 ([29] forewing anterior radial trace bearing two prestigmal ORB’s, the most proximal deeply forked, and [95] male 9th gonocoxites lost). On the basis of the two characters: (1) forewing anterior radial trace with two prestigmal ORB’s, and (2) presence of styli on the female 9th gonocoxites, Notherobius superficially resembles sympherobiine genera more than hemerobiine genera. How- ever, these characters are plesiomorphic “above” lineage 3, and thus cannot justify a sister-group relationship between Notherobius and the Sympherobiinae. 258 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Nomina Dubia Genus Subboriomyia Steinmann Subboriomyia Steinmann, 1 967: 1 59 (Type species: Subboriomyia fusca Steinmann. 1967:160. by original designation. Etymology: Unexplained, probably Sub- [from Latin sub , under or from] + Boriomyia. Gender: Feminine.): Aspock et al., 1980: 409 (Tax). Status. Treated here as a nomen dubium , following Aspock et al. (1980). Species ( 1 ). fusca Steinmann: Hungary. Distribution. Reported from Hungary. General Notes. Horst Aspock (pers. comm.), who has personally questioned Dr. Steinmann regarding the identity of Subboriomyia. has informed me that Subbor- iomyia and its type species, S. fusca. were described by Dr. Steinmann from memory, and that no type material exists for fusca. Aspock et al. (1980:41 1) were unable to identify any species in the European Neuroptera fauna possessing the character com- bination attributed to fusca by Steinmann. Having been described from memory, it seems probable these “taxa" represent fanciful combinations of characters. Neuropterida Incertae Sedis Genus Ormiscocerus Blanchard in Gay Ormiscocerus Blanchard in Gay. 1851:128 (Type species: Ormiscocerus nitidipennis Blanchard in Gay, 1851:129. by monotypy. Note: Blanchard’s work contains two specific names for the type species of Ormiscocerus— nitidipennis. in the text, and sticticopterus. in the legend to plate 2. Brauer (1866:989), acting as first reviser (ICZN. Article 24), fixed the correct original spelling as nitidipennis. Etymology: Unexplained, probably Ormisco- [from Greek ormiskos, a small necklace] + -cerus [from Greek keras or keros, horn or antenna]. Gender: Masculine.): Brauer, 1866: 986. 989 (Tax); Banks. 1913:21 1 (Key). Ormismocerus [sic]: Hagen, 1866:375, 454 (Lst). Status. Treated here as Neuropterida incertae sedis. The ordinal affinities of this taxon are here considered to be unknown. Species (1). nitidipennis Blanchard in Gay: Chile. Distribution. Reported from Chile. General Notes. Blanchard based the description of O. nitidipennis on an unrecorded number of specimens then contained in Claudio Gay’s collection of Chilean insects. Although Gay’s collection is known to reside in the Paris Museum (Papavero, 1971), the whereabouts of the type of nitidipennis is unknown. A search of the Paris Museum Neuroptera collection (personal visit, 1989) failed to turn up its type, though other specimens described by Blanchard from Gay’s collection were found. The possibility that the type is present in another part of the Paris Museum collection cannot be ruled out. The unavailability of type material together with Blanchard’s enigmatic figure render the identity and phylogenetic position of Ormiscocerus highly uncertain. The subsequent literature citations to this taxon also note uncertainty about its placement. Hagen (1866) listed nitidipennis under the generic name Ormismocerus [sic] and 1993 WORLD GENERA OF HEMEROBIIDAE 259 tentatively retained it in his restricted concept of the family “Hemerobidae”. Brauer ( 1 866), after noting similarities between Blanchard’s figure of Ormiscocerus and taxa now placed in the orders Raphidioptera, Megaloptera and Neuroptera, tentatively suggested that it might be most closely related to Raphidia. Banks (1913) keyed Ormiscocerus to his tribe Polystoechotini; however, some characters used in his key (e.g., “wings plainly falcate at tips”) clearly contradict the original figure and de- scription. None of the preceding references appear to have been based on a reex- amination of Blanchard’s type material or new specimens. Based on the little available evidence, Ormiscocerus is regarded here as Neuropterida incertae sedis. The following characters (based on Blanchard, 1 85 1 , pi. 2, fig. 1 1), justify removal of Ormiscocerus from the modem concept of the family Hemerobiidae: ( 1 ) profemora expanded [not cylindrical], (2) prothorax elongate [not transverse], (3) head apparently prognathous [not hypognathous], and (4) antennae inserted well before [not between] eyes. FOSSIL HEMEROBIIDAE General. Fossils confidently assignable to the Hemerobiidae are known from the Paleogene Tertiary of British Columbia and western Europe, and the Upper Jurassic of Kazakhstan, USSR; a questionable record from the Lower Cretaceous of China also exists (see Catalog below). Several Tertiary hemerobiid fossils have been referred to extant genera on the basis of venational similarities. However, until corroborating evidence from terminalic structures becomes available, such placements must be regarded as speculative. The scarcity of hemerobiid fossils and their uncertain phy- logenetic affinities limit their usefulness for drawing evolutionary and biogeographic conclusions about the family. Identification. Apart from specimens embedded in fossil resins (e.g., amber), fossil hemerobiids, like most other neuropterous fossils, are principally diagnosed and identified on the basis of venational traits. The single most important diagnostic character for fossil hemerobiids is the presence of two or more ORB’s diverging from the forewing anterior radial trace. The following additional forewing traits are ubiq- uitous or widespread, but apparently plesiomorphic, within the family; although these traits cannot definitively substantiate the placement of a specimen in the Hemero- biidae, the presence of alternative states may suggest the exclusion of a specimen from the family: (1) some trichosores always present, (2) most longitudinal veins prominently twigged near wing margin, (3) proximal humeral trace usually distinctly recurrent [occasionally simple], (4) subcostal veinlets numerous, frequently branched, particularly proximally, (5) pterostigma absent or very poorly developed, (6) Sc and anterior radial trace connected distally by a crossvein [rarely fused], and (7) one or two prestigmal sc-r crossveins present [rarely more]. Catalog. Existing literature records of fossil ’hemerobiids’ can be divided into three categories; (1) Unsubstantiated Records: Records of fossil ’hemerobiid’ taxa which cannot be substantiated because they lack adequate descriptions or figures, or references to them. These records are excluded from the catalog below. Most such records date from the 19th century (see listings in Handlirsch, 1906—1908 and Scudder, 1891), when the prevailing concepts of the family Hemerobiidae and the genus “ Hemero - 260 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Table 5. Synopsis of stratigraphic and geographic distributions of fossil hemerobiids. Era Period or epoch Local Taxon Citation Cenozoic Eoc./Olig. England Hemerobius tinctus Jarzembowski. 1980 Cenozoic Eoc. Olig. England Neuronema spp. Jarzembowski. 1980 Cenozoic Eoc./Olig. England Hemerobiidae spp. Jarzembowski, 1980 Cenozoic Eocene Denmark Hemerobiidae sp. Larsson. 1975 Cenozoic Eocene Denmark Megalomus densistriatus Henriksen. 1922 Cenozoic Eocene Denmark Megalomus sp. Henriksen. 1922 Cenozoic Pal./Eoc. Germany Megalomus sp. lilies. 1941 Cenozoic Tertiary British Co- lumbia Bothromicromus lachlani Scudder. [1878] Cenozoic O Baltic amber Prolachlanius resinatus (Hagen, 1856) Cenozoic 9 Baltic amber Prophlebonema resinatum Kruger. 1922 Cenozoic 9 Baltic amber Prospadobius moestus (Hagen. 1856) Creta- ceous Lower China Mesohemerobius jeholensis Ping. 1928 Jurassic Upper Kazakhstan. USSR Promegalomus anomalus Panfilov. 1980 bins” were much broader than at present. Many of these records almost certainly pertain to taxa properly assigned to other neuropterous families, or other related orders. (2) Invalid Records: Records of fossil taxa originally described in. or once placed in. the Hemerobiidae. but currently placed elsewhere. These taxa are excluded from the catalog below. The taxonomic histories of most such taxa are recorded by Han- dlirsch (1906-1908). (3) Apparently Valid Records: Records of fossil taxa which appear to fall within the modem concept of the Hemerobiidae. New taxa diagnosed in these records may- be unnamed, undescribed, or unfigured, but not all three. The catalog below gives all hemerobiid taxa of which I am aware which fit these requirements. The strati- graphic and geographic distributions of these taxa are summarized in Table 5. Re- marks contained under the Comments headings in the catalog are given based on information available in published figures and descriptions. Because the fossils them- selves have not been examined, and because few relevant characters are diagnosable from published descriptions and illustrations. I have refrained from assigning fossil taxa to extant subfamilies. The only taxonomic change formally suggested here is treatment of the Panfilov’s family Promegalomidae as a junior synonym of the Hemerobiidae. Family Hemerobiidae Latreille (Extant. Tertiary. Cretaceous, Jurassic) Hemerobiidae Latreille. 1802:288 (as Hemerobini). Type genus: Hemerobius Lin- naeus. 1993 WORLD GENERA OF HEMEROB1IDAE 261 Promegalomidae Panfilov in Dolin et al., 1980:87. Type genus: Promegalomus Pan- filov in Dolin et al. NEW SYNONYM (see comments under Promegalomus). Genus Hemerobius Linnaeus (Extant, Tertiary) Hemerobius Linnaeus, 1758:549. Type species: Hemerobius humulinus Linnaeus, 1758:550. See treatment of extant genus above for complete synonymy. tinctus Jarzembowski Hemerobius tinctus Jarzembowski, 1980:257 (OD, FW*). Etymology: Unexplained, probably from the Latin noun or adjective tinctus. Provenance: Late Eocene or early Oligocene, Bembridge Marls of the Isle of Wight, England. Comments: The forewing venation of tinctus (see Jarzembowski, 1980:254, fig. 31) resembles that of the genus Psychobiella (currently restricted to Australia, see Fig. 90) more than that of Hemerobius. The following similarities between tinctus and Psychobiella are particularly striking: (1) the anterior radial trace bears 3 ORB’s, the most proximal of which is forked in the basal third of the posterior sectoral trace [in extant Hemerobiinae the most proximal ORB is never this deeply forked (de- rived)], (2) crossvein 2sc-r is present (derived) [absent in extant Hemerobiinae], (3) four m-cu crossveins are present [three in almost all Hemerobius (derived)]. This combination of traits strongly suggests that tinctus is not a Hemerobius. Its possible relationship with Psychobiella should be investigated further. Hemerobius tinctus appears distinguishable from both genera by its deeply forked CuP. Genus Neuronema McLachlan (Recent, Tertiary) Neuronema McLachlan, 1869:27. Type species: Hemerobius decisus Walker, 1860: 185. See treatment of extant genus above for complete synonymy. ? Species A “? Neuronema Sp. A” Jarzembowski, 1 980:259 (OD, FW*). Provenance: Late Eocene or early Oligocene, Bembridge Marls of the Isle of Wight, England. Comments: The wing fragment illustrated by Jarzembowski (1980:258, fig. 39) appears consistent with this generic determination. ? Species B “? Neuronema Sp. B” Jarzembowski, 1980:259 (OD, FW*). Provenance: Late Eocene or early Oligocene, Bembridge Marls of the Isle of Wight, England. Comments: The wing fragments illustrated by Jarzembowski (1980:258, fig. 40) appear consistent with this generic determination, although costal crossveins are not as numerous in recent Neuronema species. Genus Megalomus Rambur (Extant, Tertiary) Megalomus Rambur, 1 842:4 1 8. Type species: Megalomus tortricoides Rambur, 1 842: 419. See treatment of extant genus above for complete synonymy. 262 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) densistriatus Henriksen Megalomus densistriatus Henriksen, 1 922: 1 4 (OD, FW*). Etymology: After the dense venation. Provenance: Lower Eocene, Mo-clay of Denmark. Comments: Placement of densistriatus in Megalomus is speculative since terminalic characters are lacking. The vein interpreted by Henriksen as the “M” is herein regarded as the most proximal branch of the R. Henriksen’s second specimen (“ Me- galomus ? densistriatus ” Henriksen, 1922: fig. 5), is certainly a forewing, not a hind wing. Based on the different configuration of its distal ORB’s, it is probably not conspecific with densistriatus. See also Comments below under Promegalomus an- omalus. Species A Megalomus sp. Henriksen, 1922:17 (Bdy*, W*). Provenance: Lower Eocene, Mo- clay of Denmark. Comments: Judging from Henriksen’s figure, this specimen is a poorly preserved body and wing with the general habitus of a hemerobiid. Species B Megalomus sp. lilies, 1941:21, fig. 4 (FW*). Provenance: Upper Paleocene or Lower Eocene of Havighorst (near Hamburg) Germany. Comments: Forewing figure only. Venation similar to wings figured by Henriksen (1922), Larsson (1975), and Panfilov in Dolin et al. (1980). See lilies ([1943]) for a treatment of the stratigraphy of the collection area. Genus Bothromicromus Scudder (Tertiary) Bothromicromus Scudder, [1 878]:462. Type species: Bothromicromus lachlani Scud- der, [1878], by monotypy. Etymology: Unexplained, probably Bothro- [from Greek bothros , hole or trench] + Micromus. Gender: Masculine. Bothriomicromus [sic]: Zoological Record, Literature year 1877:202 (an incorrect subsequent spelling of Bothromicromus). lachlani Scudder Bothromicromus lachlani Scudder, [1878]:462 (OD). Etymology: From the surname of British entomologist Robert McLachlan (1837-1904). Provenance: Tertiary of Quesnel, British Columbia, Canada.: Scudder, 1890:163-165 (RD, Bdy*, W*); Handlirsch, [1907]:909 (Tax). Comments: From Scudder’s description, this species appears to be closely related to the extant genus Drepanepteryx, particularly in the broad costal area, presence of a costal gradate series, and the large number of ORB’s. However, the specimen apparently lacks a recurrent proximal humeral trace (a preservational artifact?). 1993 WORLD GENERA OF HEMEROBIIDAE 263 Genus Prolachlanius Kruger (Tertiary) Prolachlanius Kruger, 1923:88. Type species: Hemerobius resinatus Hagen in Pictet and Hagen, 1856, by original designation. Etymology: Unexplained, probably Pro- [from Latin pro , before] + Lachlanius. Gender: Masculine. resinatus (Hagen in Pictet and Hagen) Hemerobius resinatus Hagen in Pictet and Hagen, 1856:88 (OD, FW*). Etymology: From the Latin adjective resinatus , resined. Provenance: Baltic Amber.: Hagen, 1854:228 (nom. nud., no description or indication. Art. 12a). Mucropalpus elegans Berendt in Pictet and Hagen, 1856:88 (nom. nud., invalid when first proposed and not subsequently made available. Art. 1 Id & e): Handlirsch, [1 907]:909 (Tax). Prolachlanius resinatus : Kruger, 1923:88 (RD). Genus Proph/ebonema Kruger (Tertiary) Prophlebonema Kruger, 1 923:85. Type species: Prophlebonema resinata [sic] Kruger, 1923, by original designation. Etymology: Unexplained, probably Pro- [from Latin pro, before] + Phlebonema. Gender: Neuter. resinatum Kruger Phlebonema resinata [sic] Kruger, 1922 (OD). Etymology: From the Latin adjective resinatus, resined. Provenance: Baltic Amber. Prophlebonema resinata [sic]: Kruger, 1923:85 (RD). Genus Prospadobius Kruger (Tertiary) Prospadobius Kruger, 1923:90. Type species: Hemerobius moestus Hagen in Pictet and Hagen, 1856, by original designation. Etymology: Unexplained, probably Pro- [from Latin pro, before] + Spadobius. Gender: Masculine. moestus (Hagen in Pictet and Hagen) Hemerobius moestus Hagen in Pictet and Hagen, 1856:88 (OD, FW*). Etymology: From the Latin adjective moestus [var. of maestus], sad or melancholy. Provenance: Baltic Amber.: Hagen, 1854:228 (nom. nud., no description or indication. Art. 12a); Handlirsch, [1907]:908 (Tax). Prospadobius moestus: Kruger, 1923:90 (RD). Genus Mesohemerobius Ping (Lower Cretaceous) Mesohemerobius Ping, 1928:42. Type species: Mesohemerobius jeholensis Ping, 1928: 42, by original designation. Etymology: Unexplained, probably Meso- [from Meso(zoic)] + Hemerobius. Gender: Masculine. 264 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) jeholensis Ping Mesohemerobius jeholensis Ping. 1928:42 (OD. FW*). Etymology: After Jehol. a region in China. Provenance: Lower Cretaceous. Upper Volcanic series of Pei-piao [=Beipiao]. Jehol. China. Comments: Mesohemerobius jeholensis is known only from the distal two-thirds of one forewing. If Ping's (1928:42, fig. 21) illustration and venational interpretation are accurate, this specimen possesses multiple ORB's along the anterior radial trace and would appear to be a hemerobiid. However, several aspects of Ping's figure are in potential conflict with this interpretation. First, absolutely no crossveins are shown (a drawing or fossil artifact?). No extant (or fossil?) neuropteran entirely lacks forewing crossveins. Second, the subcostal veinlets in the figure lack any proximal associations with a longitudinal vein. This calls into question Ping's venational homologies. If the subcosta of this specimen was inadvertently removed during its preparation (a possibility w hich would explain the lack of subcostal veinlet connections). Ping's veins “Sc" (simple) and “R” (pectinately branched) would actually represent the R1 and Rs respectively. If this is the case, the sole basis for considering this fossil a hemerobiid (the presence of a pectinately branched forewing anterior radial trace) would be eliminated. A critical reexamination of the type of jeholensis is needed to eliminate this possibility. Genus Promegalomus Panfilov in Dolin et al. (Upper Jurassic) Promegalomus Panfilov in Dolin et al.. 1980:87. Type species: Promegalomus an- omalus Panfilov in Dolin et al., 1980. by original designation. Etymology: Unex- plained. probably Pro- [from Latin pro , before] + Megalomus. Gender: Masculine. anomalus Panfilov in Dolin et al. Promegalomus anomalus Panfilov in Dolin et al.. 1980. 87 (OD. FW*). Etymology: From the Greek (or Latin) adjective anomalos (-us), anomalous. Provenance: Upper Jurassic of Kazakhstan. USSR. Comments: This species is known only from the complete forewing holotype. for w hich Panfilov created the new family Promegalomidae. However, as justified below, I consider Promegalomus a hemerobiid. and the Promegalomidae a junior synonym of the Hemerobiidae. Of the eight putative synapomorphies of the Hemerobiidae identified in the cla- distic analysis of recent genera, only Character 35 is a forewing character. Panfilov's figure (1980. fig. 91) of anomalus shows the derived state of this character, i.e., absence of pre-3irl crossveins. In addition, the figure shows the anterior radial trace of anomalus with two oblique radial branches, a further derived characteristic found in all extant hemerobiids. Finally, the prominently and pectinately branched posterior sectoral trace is very similar to the configuration found in some extant (e.g., some Drepanacrinae and Drepanepteryginae) and fossil hemerobiids (e.g.. the Lower Eo- cene specimens illustrated by Henriksen. 1922: Larsson. 1975:203: and lilies. 1941). 1993 WORLD GENERA OF HEMEROBIIDAE 265 Genus Unknown Hemerobiidae “Species A” Jarzembowski, 1980:260 (OD, W*). Provenance: Late Eocene or early Oligocene, Bembridge Marls of the Isle of Wight, England. Comments: Based on two small wing fragments. Genus undeterminable. Hemerobiidae “Species B“ Jarzembowski, 1980:260 (OD, Bdy*, W*). Provenance: Late Eocene or early Oligocene, Bembridge Marls of the Isle of Wight, England. Comments: Based on a body impression and several small wing fragments. Genus undeterminable. Hemerobiidae “Species C” Jarzembowski, 1980:260 (OD, W*). Provenance: Late Eocene or early Oligocene, Bembridge Marls of the Isle of Wight, England. Comments: Based on one small wing fragment. Genus undeterminable. Hemerobiidae sp. Larsson, 1975:203, fig. 6. Provenance: Lower Eocene, Mo-clay of Jutland, Denmark. Comments: The specimen illustrated by Larsson may be related to Promegalomus. Larsson mentions that hemerobiids are rather common in the Mo-clay of Denmark, but that the material collected to date remains largely undescribed. BIOGEOGRAPHY Age of the Clade Hemerobiidae Only limited data are available relevant to estimation of the absolute age of the Hemerobiidae. A minimum age may be inferred from the age of the oldest known hemerobiid fossil. If assignment of the genus Promegalomus to this family is correct, see Fossil Hemerobiidae above, the clade Hemerobiidae dates at least from the late Jurassic. The only other putative hemerobiid reported from Mesozoic strata is Me- sohemerobius jeholensis Ping, from the Lower Cretaceous of China; however, the assignment of this fossil to the Hemerobiidae requires confirmation, see Fossil Hem- erobiidae above. The existence of several genera with species occupying land areas in both the New and Old Worlds (i.e., Hemerobius, Wesmaelius, Sympherobius , Notiobiella, Mega- lomus, and Micromus), particularly those represented in the faunas of the southern continents of both hemispheres ( Hemerobius , Sympherobius , Notiobiella , and Mi- cromus), offers additional potential support for a Mesozoic origin for the family. If the broad distributions of these genera have resulted principally from the broad distributions of their respective common ancestors, species or groups of species within these genera found to be endemic to individual continents might be attributed to late Mesozoic vicariance events associated with the fragmentation of Gondwanaland, and support a somewhat earlier Mesozoic origin of the family. All of the preceding genera contain species restricted to single continents; however, estimates of phylo- genetic relationships among species within these genera, which will be necessary to critically interpret their biogeographic histories, are not currently available. 266 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Table 6. Distribution of hemerobiid genera by faunistic realm. Distribution* Subfamily genus New World Old World No. spp. NEA NEO PAL ETH ORI AUS Carobiinae Carobius 9 - - - - - X* Hemerobiinae Hemerobius 125 X X X X X X Nesobiella P — — — — — — Wesmaelius 70 X X X X X 2 Hemerobiella 1 - X* - - - - Sympherobiinae Nomerobius 4 X* Neosympherobius 1 — X* — — — — Sympherobius 55 X X X X 7 — Psychobiellinae Psychobiella 2 - - - - - X* Notiobiellinae Notiobiella 35 — X X X X X Psectra 25 2 — X X X X Anapsectra 1 — - - X* — - Zachobiella 8 — — — — X X Drepanacrinae Conchopterella 4 — X* — — - - A ustromegalom us 2J — — — - — — Drepanacra 4 — — — — X X Megalominae Megalomus 40 X X X - X — Drepanepteryginae Neuronema 30 — — X — X - Gayomyia 1 - X* - - — — Drepanepteryx 6 — — X* — — — Microminae Noius 3 — — — — - X* Nusalala 20 — X* — — — — Megalomina 3 — — — - — X* Aficromus 95 X ? X X X X Unplaced genera Nolherobius 3 - - — — — X* Endemic genera5 — 0 6 1 1 0 5 Total genera5 - 5 1 1? 9 7 9? 1 1 % genera endemic - 0 54 1 1 14 0 45 1993 WORLD GENERA OF HEMEROBIIDAE 267 Overview of Generic Distributions The distributions of hemerobiid genera are summarized by biogeographic realm in Table 6, and by continent in Table 7. No hemerobiids are known from Antarctica. The two most speciose hemerobiid genera, Hemerobius and Micromus, are nearly cosmopolitan in distribution. Hemerobius is broadly distributed in all temperate and tropical land areas except Australia (1 sp. in Queensland; 3 spp. in New Guinea), where it may be a relatively recent immigrant (? a member of the “Younger Northern Element” of Mackerras, 1970). Micromus is also broadly distributed throughout the temperate and tropical areas of all continents, with the possible exception of South America. Several Micromus species earlier reported from South America have proved to belong to Nusalala, and the remaining South American species currently assigned to Micromus require reexamination. Five other genera possess broad intercontinental distributions: Sympherobius (North and South America, Eurasia, and Africa), Megalomus (North and South America, Eurasia, and northern Africa), Wesmaelius (North America, Eurasia, and Africa), Psectra (North America [1 sp.], Europe [1 sp.], Africa, India, SE Asia, and Australia), and Notiobiella (South and Central America, Africa, SE Asia, and Australia). The remaining genera are restricted to ( 1 ) a single continent: Carobius, Psychobie/Ia, Notherobius, Noius, and Megalomina [Australia]; Conchopterella, Hemerobiella, Nomerobius, Neosympherobius, and Gayomyia [South America]; Anapsectra [Africa]; Neuronema and Drepanepteryx [Eurasia], (2) the neighboring portions of two adjacent continents: Zachobiella and Drepanacra [SE Asia and Australia], and Nusalala [South and Central America], or (3) isolated oceanic islands: Nesobiella [Hawaiian Islands] and Austromegalomus [Society Islands]. Overview of Subfamily Distributions Of the nine hemerobiid subfamilies recognized herein, three are monogeneric (Ca- robiinae, Psychobiellinae, and Megalominae), and, consequently, show no interge- neric distribution patterns. Five of the six remaining subfamilies (Hemerobiinae, Sympherobiinae, Notiobiellinae, Drepanacrinae, and Microminae) consist of one or two genera with broad intercontinental distributions, and one or two genera endemic to a single continent. The only subfamily containing multiple genera, each of which is endemic to a single continent, is the Drepanepteryginae, which contains two genera in Eurasia ( Drepanepteryx and Neuronema) and one in southern South America {Gayomyia). The extreme biogeographic disjunction exhibited by this subfamily * Genus endemic to a single faunistic realm. ? Presence uncertain. 1 NEA, Nearctic; NEO, Neotropical; PAL, Palearctic; ETH, Ethiopian; OR1, Oriental; AUS, Australian. 2 One or more species present, but presumed introduced. 3 Hawaiian Islands 4 Society Islands (Tahiti and Rapa). 5 Presumed introductions omitted. 268 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Table 7. Distribution of hemerobiid genera by continent. Distribution' Subfamily genus New World Old World No. spp. NA SA EUR AFR AUS Carobiinae Carobius 9 - - - - X* Hemerobiinae Hemerobius 125 X X X X X Nesobiella U — — — — — Wesmaetius 70 X — X X 2 Hernerobiel/a 1 - X* - - - Sympherobiinae Nomerobius 4 — X* — — — Neosympherobius 1 — X* — - — Sympherobius 55 X X X X — Psychobiellinae Psychobiella 2 - - - - X* Notiobiellinae Notiobiella 35 X X X X X Psectra 25 2 — X X X Anapsectra 1 — — — X* - Zachobiella 8 - - X — X Drepanacrinae Conchopterella 4 — X* — - — Austromegalomus 2J — — — — — Drepanacra 4 — — X — X Megalominae Megalomus 40 X X X X - Drepanepteryginae Neuronema 30 — — X* — — Gayomyia 1 — X* - - - Drepanepteryx 6 - — X* — — Microminae Noius 3 — — — — X* Nusalala 20 X X — — — Megalomina 3 — — - - X* Microtnus 95 X 9 X X X Unplaced genera Notherobius 3 - - - - X* Endemic genera5 — 0 5 2 1 5 Total genera5 - 7 10? 1 1 8 11 % genera endemic - 0 50 18 12 45 1993 WORLD GENERA OF HEMEROBIIDAE 269 seems best interpreted as relictual. No clear, repeated, patterns of intergeneric dis- tributions are evident among the multigeneric subfamilies. The sister-group relationship between the hemerobiine genera Hemerobius and Nesobiella is biogeographically intriguing. Nesobiella is monobasic and restricted to the Hawaiian Islands, while Hemerobius is speciose, nearly cosmopolitan, and absent from these islands except for the apparently recent introduction of H. pacificus from western North America. Zimmerman (1957) speculated that Nesobiella might have been recently introduced into the Hawaiian Islands from an unknown source area. To date however, Nesobiella has never been reported from outside the Hawaiian Islands, nor have any congeners been described from elsewhere in the world. Given the volcanic origin of the Hawaiian Islands, the ancestors of Nesobiella must have reached them by long-distance dispersal; and, given the absence of known close relatives, the present distribution of Nesobiella appears relictual. Faunal Similarities and Differences between Adjacent Continental Landmasses North America and Eurasia share seven genera: Hemerobius , Wesmaelius, Sym- pherobius, Notiobiella, Psectra, Megalomus, and Micromus. Of these, Hemerobius , Wesmaelius , Sympherobius , Megalomus, and Micromus are broadly distributed on both continents. Notiobiella (pantropical) is primarily restricted to the tropical regions of both continents. Psectra is most diverse in southeastern Asia, but a single species, P. diptera, is distributed across temperate Eurasia from Japan to Europe, and is present (probably introduced) in eastern North America. Four Eurasian genera are lacking in North America: Neuronema, an east Asian endemic; Drepanepteryx, a Palearctic endemic; and Zachobiella and Drepanacra, a pair of southeast Asian taxa shared with Australia. The genera comprising the North American and European faunas are identical, except for the additional presence of Drepanepteryx in Europe. North and South America share five genera: Hemerobius, Sympherobius, Notiobiel- la, Megalomus, and Nusalala. Of these, Hemerobius, Sympherobius, and Megalomus are broadly distributed on both continents. The genera Notiobiella (pantropical) and Nusalala (Neotropical), reach North America only in the tropical portions of Central America and the Antilles. The principal faunal differences between these regions are the apparent absence of Micromus species from South America (present in the An- tilles) and the presence in the South American fauna of five additional, small, endemic genera: Hemerobiella, Nomerobius, Neosympherobius, Conchopterella, and Gayo- myia. North America has no endemic genera. * Genus endemic to a single continent. ? Presence uncertain. 1 NA, North America; SA, South America; EUR, Eurasia; AFR, Africa; AUS, Australia (including New Guinea and New Zealand). 2 One or more species present, but presumed introduced. 3 Hawaiian Islands. 4 Society Islands (Tahiti and Rapa). s Presumed introductions omitted. 270 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Table 8. Hemerobiid genera endemic to northern and southern continents. Continents No. genera5 Northern' Southern-' Endemic 2 n Total 12 21 Percent endemic 17 52 ' North America, Europe. Asia. ; South America, Africa, Australia. 1 Presumed introductions omitted. Europe and Africa share six genera: Hetnerobius , W'esmaelius, Svmpherobius, Psec- tra. Megalomus. and Micromus. The genera Hetnerobius, Wesmaelius, Symphero- bius. Psectra and Micromus are broadly distributed in both regions. Megalomus has been reported from extreme northwestern Africa (? a recent immigrant from Europe), but is otherwise absent. The only generic-level faunal differences between these con- tinents are the additional presence of Drepanepteryx (a Palearctic endemic) in Europe, and of Anapsectra (an Ethiopian endemic) and Notiobiella (pantropical) in Africa. Australia/New Guinea and southeast Asia share six genera: Hemerobius, Notiobiel- la, Psectra. Zachobiella. Drepanacra. and Micromus. Psectra. Drepanacra and Mi- cromus are broadly distributed within both regions. In Australia, Hemerobius, No- tiobiella. and Zachobiella appear largely confined to the north, principally Queensland, but are more widely distributed in southeast Asia. The Australian fauna is especially distinctive because of five endemic genera: Carobius , Psychobiella, Noius, Megalom- ina, and Notherobius. There are no genera endemic to the adjacent areas of south- eastern Asia. Continental Generic Endemism Southern landmasses possess a percentage of endemic genera triple that of northern land areas (Table 8). The percentage of generic endemics is particularly high for South America and Australia. This is apparent both when genera are tabulated by biogeo- graphic realms and by continental boundaries (Tables 6 and 7). This pattern is consistent with the markedly different tectonic histories of the northern and southern continents. Since the division of Pangea into Laurasia and Gondwanaland in the late Mesozoic, east-west routes between North America and Eurasia (i.e., Beringia and northern trans-Atlantic routes [Matthews, 1980]) contin- ued. at least intermittently, to facilitate faunal exchange between the northern con- tinents. Faunal exchange via these routes may account for, in a general way, the low percentage of generic endemism ( 1 7%) and high percentage of common genera (33%) of the northern continents, relative to the same percentages (52% and 9% respectively, see Tables 8 and 9) for the southern continents, which were isolated by oceanic barriers subsequent to the fragmentation of Gondwanaland. The subfamily Drepanacrinae, which contains three genera ( Conchopterella , Aus- tromegalomus , and Drepanacra ) distributed entirely or principally in the southern hemisphere, exhibits the hemerobiid subfamily distribution which most closely re- sembles a classic southern Gondwanian distribution. 1993 WORLD GENERA OF HEMEROBI1DAE 271 Table 9. Hemerobiid genera common to northern and southern continents. Continents No. genera’ Northern1 Southern-’ In common 4 2 Total 12 21 Percent endemic 33 9 1 North America, Europe, Asia. 2 South America, Africa, Australia. ' Presumed introductions omitted. Island Endemism Hemerobiids are present on numerous oceanic islands and island groups (Table 10), many of which are not of continental origin. The widespread presence of hem- erobiids on non-continental islands supports the contention that hemerobiids, al- though relatively weak fliers, are good long-distance aerial dispersers. Most near- shore insular species, however, are conspecific with or appear to be closely related to species found on adjacent continental areas. Only three genera, Austromegalomus, Nesobiella , and Noius , are completely endemic to oceanic islands. Although intentional introductions of hemerobiids into island faunas have been documented for a few species (e.g., Micromus timidus [as M. vinaceus], introduced into the Hawaiian Islands, see Williams, 1931), the extent to which man has aug- mented insular distributions through inadvertent introductions has not been well studied. Certainly, some such introductions have been made, e.g., the introduction of the western North American species Hemerobius pacificus into Hawaii (MacLeod, 1964) and the introduction of the Holarctic species Wesmaelius subnebulosus into New Zealand (Wise, 1973), and some other records are highly suspect, e.g., the presence of the Palearctic species variegatus on Galiano Is., British Columbia, Can- ada. A few insular species exhibit remarkable secondary morphological modifications. These alterations are particularly apparent in the shape and extent of development of the wings. The most striking of these modifications are found in the complex of 20+ Micromus species endemic to the Hawaiian Islands (Zimmerman, 1957), and in the two species of Conchopterella endemic to the Juan Fernandez Islands (Hand- schin, 1955). Discussion As indicated above, no clear repeated patterns of intergeneric distributions are evident among the subfamilies of the Hemerobiidae. Thus, there is presently no good evidence to support the idea that a common vicariant history has significantly shaped its subfamily distributions. However, given the certain occurrence of vicariant events, the list of factors which can be invoked to explain the absence of a repeated, vicariantly induced, pattern in a taxon the age of the Hemerobiidae is lengthy. Patterns may be obscured by: (1) dispersal, particularly in association with changing environments, or extinction subsequent to a vicariant event, (2) the presence of ancestral taxa which were regionally, not catholically, distributed, (3) comparison of taxa [even if de- 272 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Table 10. Summary of principal literature on insular hemerobiid faunas (for additional titles see works cited in the papers given below). Ocean area Island or group: Reference Atlantic: Eastern Azores: Aspock et al., 1980: 378-379; Canaries: Aspock et al., 1980: 378-379; Cape Verde: Ohm and Holzel, 1982; General: Aspock et al., 1980; Iceland: Fristrup. 1942; Tjeder, 1964; Madeira: Aspock et al., 1980: 378-379; Mediterranean Islands: Aspock et al., 1980; St. Helena: Tjeder, 1976; Atlantic: Western Bermuda: Hilbum. 1990; Cuba: Alayo. 1968: Greenland: Nielsen, 1909; Richard and Harmston, 1972; Hispanola: Banks, 1941; Navas, 1932; Smith. 1931; Puerto Rico: W'olcott, [1950]; Indian: Eastern General: Van der Weele, 1909; Handschin, 1935; Java: Van der Weele, 1910a; Sri Lanka (Ceylon): Hagen, 1858: Hagen, 1859; Indian: Western Madagascar: Fraser, 1951; Reunion: Fraser, 1957; Seychelles: Kimmins. 1933: Enderlein, 1910; Pacific: Eastern Galapagos: Klimaszewski et al.. [1988]; Isla del Coco: Oswald, unpublished data; Juan Fernandez: Handschin, 1955; Revillagigedo: Oswald. 1988b; Pacific: Western and Central Borneo: Banks, 1931; Banks. 1934; General: Carpenter, 1961: Esben-Petersen, 1937: Van der Weele, 1909; Hawaiian: MacLeod. 1964; Perkins, 1899: Zimmerman. 1957; Japan: Kuwayama, 1962; Kuril: Kuwayama. 1956; Marquesas: Esben-Petersen, [1935a]: New Caledonia: Kimmins, 1958: Nakahara. 1960a; New Guinea: New, [1989]; New Hebrides: Kimmins, 1958; New Zealand (and adjacent islands): Wise, 1977; Wise, 1983; Philippines: Banks. 1937; Rapa: Oswald, 1988a; Ryukyu: Kuwayama. 1964; Nakahara, 1966; Samoa: Esben-Petersen. 1928; Tahiti: Esben-Peterson, [1935b]; Oswald. 1988a; Taiwan (Formosa): Nakahara, 1966 1993 WORLD GENERA OF HEMEROBIIDAE 273 monstrably holophyletic and of equal rank] of incongruent age, or (4) comparison of inappropriate [i.e., nonholophyletic] taxa. A further alternative is the true absence of a repetitive, vicariantly induced, pattern. This implies that factors such as dispersal, extinction, lack of speciation at vicariant events, and ancestral regionalism, may reduce the ability of sequential vicariant events to produce detectable, repeated, patterns of distribution in some kinds of organisms. This does not deny the existence of vicariant events, only their ability to universally produce repeated patterns of distribution. Clearly, our ability to detect common biogeographic patterns attribut- able to vicariant events depends upon the relative magnitude and frequency of both the pattern-producing vicariant events themselves and the diverse kinds of events which tend to obscure such patterns. In the case of the Hemerobiidae, little unambiguous information is currently avail- able regarding dispersal, extinction, and the ancestral distributions and absolute (or, at least, minimum) ages of its genera and subfamilies. Consequently, there are in- sufficient data from which to infer the principal causal agents which have resulted in the apparent lack of common distribution patterns among the genera comprising hemerobiid subfamilies. This said, a few general comments and preliminary spec- ulations regarding hemerobiid biogeography are given below. The high levels of generic endemism in South America and Australia are almost certainly attributable to their long isolation from other continental landmasses. The fact that these endemic genera did not aggregate in the cladistic analysis into sub- families exhibiting common area-cladogram elements, may be attributable to sub- sequent extinction of related taxa in other areas or restricted distributions of their ancestors. Distributions associated with the most widespread genera are probably due to a combination of vicariant and dispersive events. The balance between these two general processes, however, cannot be understood without more detailed investiga- tions of the phylogenies of these genera. The existence of several widespread Holarctic species in the genera Hemerobius, Wesmaelius , and Micromus, and the presence of few widespread species common to adjacent northern and southern continents, sug- gests that dispersal between North America and Eurasia may be less difficult than movement between the northern and southern continents. Such a phenomenon may be primarily due to ecological preferences of the taxa in question. The lack of concordance of intergeneric distribution patterns among subfamilies may be due to an attempt to compare taxa of substantially different ages. Although the relative (and implied absolute) ages of hemerobiid subfamilies increase toward the base of the preferred cladogram (Fig. 23), a basic cladistic inference, this does not necessarily imply that the immediate common ancestors of the collected extant genera of each subfamily are of increasing or even similar age, since older subfamily sister-groups may have become extinct and not be represented in the analysis. Since clades of differing ages experience at least partially different vicariant histories, there is no a priori reason to expect fully concordant distribution patterns among such clades. Thus, it may be generally inappropriate to limit the search for congruent distribution patterns strictly to holophyletic groups of identical rank. In the case of the Hemerobiidae, congruent patterns may yet be identified by comparisons of, for example, subfamilies and widely distributed genera. This underscores the need for additional phylogenetic and biogeographic studies of individual widespread genera to facilitate such comparisons. 274 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) ACKNOWLEDGMENTS This work represents research undertaken as a Ph.D. dissertation project in the Department of Entomology at Cornell University, Ithaca, NY. My sincere thanks are offered to the members of my graduate committee, Drs. James K. Liebherr (chairman), Amy R. McCune. and Quentin D. Wheeler for their valuable assistance and advice. I am especially grateful for the constant friendship and professional guidance offered by my committee chairman during my years at Cornell. I am happy to acknowledge the following curators, collection managers, and private indi- viduals who made hemerobiid specimens available for study from the collections under their care: Stephen J. Brooks (BMNH), Oliver S. Flint, jr. (NMNH), Dave Furth (MCZ), James B. Johnson (private collection). J. Legrand (MNHP), James K. Liebherr (CUIC), Ellis MacLeod (private collection), Mervyn Mansell (NCIP), Gordon M. Nishida (BPBM), Norman D. Penny (CAS and private collection), Randall T. Schuh (AMNH), Lionel A. Stange (FSCA); it is certainly a truism, but worth repeating, that this project could not have been completed without their support. My thanks also to James Carpenter and Stephen J. Brooks who offered much appre- ciated hospitality during my visits to the MCZ and BMNH respectively. While at Cornell, my studies were supported in part by teaching and research assistantships administered primarily through the Department of Entomology. Two of these in particular— a Liberty Hyde Bailey Research Assistantship and a James B. Palmer Distinguished Assis- tantship— allowed me to pursue my systematics research objectives with full and unrestricted vigor. Collection visitation grants were received from the American Museum of Natural History and the National Museum of Natural History. All of these sources of support are gratefully acknowledged. This paper is based upon research supported by the National Science Foundation under Grant No. BSR-8815124 and was published with support from NSF Grant No. DEB-9203962. The Foundation provides awards for research in the sciences and engineering. The awardee is wholly responsible for the conduct of such research and preparation of the results for publication. The Foundation, therefore, does not assume responsibility for such findings or their interpretation. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author and do not necessarily reflect the views of the National Science Foundation. Finally, I acknowledge my debt to my wife Diane who has been constantly encouraging and unflagging in her support during my undertaking of this project. LITERATURE CITED Achtelig, M. 1978. Entwicklung und Morphologie der inneren und ausseren weiblichen Gen- italorgane der Kamelhalsfliegen (Neuropteroidea: Raphidioptera). Ent. Germ. 4:140- 163. Adams, P. A. [1960] 1958 [29 July]. The relationship of the Protoperlaria and the Entopter- ygota. Psyche 65:1 15-127. [Dating: From reverse of wrapper for vol. 63, no. 3.] Adams. P. A. 1967. A review of the Mesochrysinae and Nothochrysinae (Neuroptera: Chry- sopidae). Bull. Mus. Comp. Zool. Harv. 135:215-238. Adams, P. A. 1969. A new genus and species of Osmylidae (Neuroptera) from Chile and Argentina, with a discussion of Planipennian genitalic homologies. Postilla 141:1-11. Adams, P. A. 1970. A review of the New World Dilaridae. Postilla 148:1-30. Agassiz, J. L. R. 1842-1847. NomenclatorZoologicus, continens nomina systematica generum animalium tarn viventium quam fossilium, . . .Soloduri. [Issued between 1842 and 1847 in 12 independently paginated fascicles. Fasciculus 5, containing the Order Neuroptera, is dated 1844. Fasciculus 12, the Index Universalis, is dated 1847. See D. R. Bowley, and H. M. Smith, 1968, Journal of the Society for the Bibliography of Natural History 1993 WORLD GENERA OF HEMEROBIIDAE 275 5:35-36, and McE. Kevan, D. K., 1970, Journal of the Society for the Bibliography of Natural History 5:286.] Agekyan, N. G. 1 973. Neuroptera feeding on bamboo aphids in Adzharia and their parasites [in Russian], Ent. Obozr. 52:549-564. [English translation: Ent. Rev. Wash. 52:362- 371.] Alayo, D. 1968. Los Neuropteros de Cuba. Poeyana Inst. Biol. (B)2: 1—1 27. Ashlock, P. D. 1971. Monophyly and associated terms. Syst. Zool. 20:63-69. Aspock, H.; Aspock, U.; Holzel, H. 1980. Die Neuropteren Europas. 2 Vols. Goecke and Evers, Krefeld, West Germany. 495 pp. Aspock, U. 1989. Nyrma kervillea Navas— eine Berothide (Neuropteroidea: Planipennia). Z. ArbGem. Ost. Ent. 41:19-24. Aubrook, E. W. 1935. Notes on the biology of Micromus aphidivorus Schrank (Planipennia, Hemerobiidae). NWest. Nat. 10:34-36. Banks, N. 1 892. A synopsis, catalogue, and bibliography of the neuropteroid insects of tem- perate North America. Trans. Am. ent. Soc. 19:327-373. Banks, N. 1897. New North American neuropteroid insects. Trans. Am. ent. Soc. 24:21-31. Banks, N. 1904 [12 November], A list of neuropteroid insects, exclusive of Odonata, from the vicinity of Washington, D. C. Proc. ent. Soc. Wash. 6:201-217. [Dating: From footnote on p. 6:245 ] Banks, N. 1905 [December]. A revision of the Nearctic Hemerobiidae. Trans. Am. ent. Soc. 32:21-51. [Dating: From footline of paper.] Banks, N. 1907. Catalogue of the neuropteroid insects (except Odonata) of the United States. American Entomological Society, Philadelphia. 53 pp. Banks, N. 1909. Hemerobiidae from Queensland, Australia. [Neuroptera, Hemerobiidae.] Proc. ent. Soc. Wash. 1 1:76-81. Banks, N. 1913 [26 August], Synopses and descriptions of exotic Neuroptera. Trans. Am. ent. Soc. 39:201-242. [Dating: From wrapper of separate.] Banks, N. 1920 [October], New neuropteroid insects. Bull. Mus. Comp. Zool. Harv. 64:[297]- 362. [Dating: From wrapper of separate.] Banks, N. 1930. New neuropteroid insects from the United States. Psyche 37:223-233. Banks, N. 1931. Some neuropteroid insects from North Borneo, particularly from Mt. Kin- abalu, 13,455 ft. J. fed. Malay St. Mus. 16:41 1^129. Banks, N. 1932. Concerning the genus Notiobiella (Neuropt. -Hemerobiidae). Psyche 39:103- 106. Banks, N. 1934. Supplementary neuropteroid insects from the Malay Peninsula, and from Mt. Kinabalu, Borneo. J. fed. Malay St. Mus. 17:567-578. Banks, N. 1937 [June]. Philippine neuropteroid insects. Philipp. J. Sci. 63:125-174. [Dating: From wrapper of separate.] Banks, N. 1939 [September], New genera and species of neuropteroid insects. Bull. Mus. Comp. Zool. Harv. 85:[439]-504. [Dating: From wrapper and title page of separate.] Banks, N. 1940. Report on certain groups of neuropteroid insects from Szechwan, China. Proc. U.S. Natn Mus. 88:173-220. Banks, N. 1941. New neuropteroid insects from the Antilles. Mems Soc. cub. Hist. nat. 15: 385-402. Bansch, R. 1964. Das Beutefangverhalten der aphidivoren Hemerobiidenlarven. Zool. Anz. 173:278-281. Billberg, G. J. 1820. Ordo Neuroptera. Pp. 94-97 in Enumeratio Insectorum in Museo Gust. Joh. Billberg. Typis Gadelianis, [Holmiae], 138 pp. Blanchard, C. E. 1851. Mirmeleonianos. Pp. 1 19-129 and Rafidianos. Pp. 1 29-1 35 in Historia Fisica y Politica de Chile, C. Gay, ed. Zoologia, Vol. 6. Paris. Bodenheimer, F. S. 1930. Sympherobius (Nefasitus) amicus Nav. Pp. 137-140 in Die Schad- 276 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) lingsfauna Palastinas. Monographien zur angewandten Entomologie. No. 10. Beih. Z. angew. Ent., Vol. 16. Brauer, F. 1866. Zusatze und Berichtigungen zu Hagen’s Hemerobidarum Synopsis Synon- ymica und Beschreibung einer neuen Nymphiden-gattung: Myiodactylus osmyloides aus Australien. Verh. zool-bot. Ges. Wien 16:983-992. Brauer. F. 1867. Larve von Hypochrysa nobilis Heyd. Verh. zool-bot. Ges. 17:27-30. Brauer. F. 1871. Beitrage zur Kenntniss der Lebensweise und Verwandlung der Neuropteren (Micromus variegatus Fabr., Panorpa communis L.. Bittacus italicus Klg. und Bittacus Hagenii Brau.). Verh. zool-bot. Ges. 21:107-1 16. Braes. C. T.: Melander. A. L. 1932. Order Neuroptera. Pp. 1 8 1-1 89 in Classification of Insects. Bull. Mus. Comp. Zool. Harv. 73:1-672. Bralle, G. A. 1832. Insectes. Nevropteres. Pp. 275-278 in Expedition scientifique de Moree. Tome 3. Part 1: Zoologie. Section 2: Des animaux articules. Levrault, Paris. Burmeister, H. C. C. 1829. De insectorum systemate naturali. Dissertatio inauguralis quam consensu facultatis medicae halensis. Halis Saxonum. [43] pp. Burmeister, H. C. C. 1839. Handbuch der Entomologie. Band 2, Abt. 2 (pp. xii + [397]- 1050 + [4]). Enslin. Berlin. Carpenter. F. M. 1936. Revision of the Nearctic Raphidiodea (recent and fossil). Proc. Am. Acad. Arts Sci. 71:89-157. Carpenter. F. M. 1940. A revision of the Nearctic Hemerobiidae. Berothidae, Sisyridae, Polystoechotidae and Dilaridae (Neuroptera). Proc. Am. Acad. Arts Sci. 4:193-280. Carpenter. F. M. 1961 [31 March], Neuroptera: Hemerobiidae. Insects of Micronesia 8(3): 35-43. [Dating: From back of title page of separate.] Comstock. J. H. 1918. The wings of the Neuroptera. Pp. 145-213 in The wings of insects. Comstock Publishing Co., Ithaca. New York. 430 pp. Costa. A. [1855-1871] 1860-1870. Fauna del Regno di Napoli ossia enumerazione di tutti gli animali. . .Nevrotteri. Napoli. [The original dates of publication of the parts are as follows: (1) Title page, dated ”1860-70”, (2) Prefazione, [2] pp.. December 1870, (3) Famiglia degli Ascalafidei, pp. [ 1 ]— 1 2. 3 November 1855, (4) Famiglia de’Formicaleonidei, pp. [l]-20, 3/4 December 1855, (5) Famiglia degli Emerobiidei. pp. [ 1 ]— 22, 15 December 1855 [as "1355”, sic], (5) Aggiunte alle precedenti famiglie, pp. [ 1 ]— 8 . 7 January 1871, (6) Famiglia de'Mantispidei, pp. [l]-2, 15 December 1855. (7) Famiglia de’Rafidiidei, pp. [ 1 ]— [8], 30 December 1855. (8) Famiglia de’Friganeidei. pp. [ 1 ]— 6, 1 February 1 87 1 , (9) Indice. [2] pp.. 1 February 1871, (10) 7 plates (numbered VII— XIII). Dates of pub- lication are given by Sherbom. C. D., 1937. Journal of the Society for the Bibliography of Natural History 1:35-47.] Curtis. J. 1824-1839. British entomology: being illustrations and descriptions of the genera of insects found in Great Britain and Ireland. . .London. [For a discussion of the editions of this work see Blackwelder. R. E.. 1947, Smithsonian Miscellaneous Collections 107(5): [ 1 ]— [3 1 ]. The four neuropterid plates are, and are dated, as follows: pi. 37 (vol. 1), Raphidia ophiopsis. 1824; pi. 202 (vol. 5), Hemerobius fimbriatus, 1828; pi. 520 (vol. 11), Chrysopa abbreviata. and pi. 528 (vol. 11), Coniopteryx psociformis, 1834. The date of publication is given at the bottom of each plate.] Cutright, C. R. 1923. Life history' of Micromus posticus Walker. J. Econ. Ent. 16:448^156. Dunn. J. A. 1 954. Micromus variegatus Fabricius (Neuroptera) as a predator of the Pea Aphid. Proc. R. ent. Soc. Lond. (A)29:76-80. Ehrenberger. F. J. 1836. Dissertatio inauguralis zoologica de Neuropteroram anatomia et phisio-biologia. . . .J. H. Pospjsil. Pragae. 21 pp. Enderlein. G. 1910 [November], Embiidina und Neuroptera (Coniopterygidae und Heme- robiidae). Trans. Linn. Soc. Lond., Zool. (2)14:55-58. [Dating: From wrapper of ser. 2, vol. 14. pt. L] 1993 WORLD GENERA OF HEMEROBIIDAE 277 Erichson, W. F. 1851. Nevroptera. Pp. 68-69 in Die Insecten, by E. Menetries in Reise in den aussersten norden und osten Sibiriens wahrend der jahre 1843 und 1844. . .auf veranstaltung der Kaiserlichen Akademie der Wissenschaften zu St. Petersburg ausgefuhrt und. . .herausgegeben von. . .A. Th. Middendorf. A. T. von Middendorf. ed. Bd. 2. Theil 1. St. Petersburg. Esben-Petersen, P. 1914. Descriptions of a new genus and some new or interesting species of Planipennia. Notes Leyden Mus. 36:263-270. Esben-Petersen, P. 1920. South African Neuroptera. I. Ann. S. Afr. Mus. 17:507-521. Esben-Petersen, P. [1925] 1924 [23 July]. Notes concerning Boriomyia betulina Strom. Norsk ent. Tidsskr. 2:54. [Dating: From footnote on p. 2:56.] Esben-Petersen, P. 1926. Fauna sumatrensis. Neuroptera II. Ent. Mitt. 15:404-407. Esben-Petersen, P. 1928. Neuroptera. Pp. 89-108 in Insects of Samoa and other Samoan terrestrial Arthropoda. Fasc. 3, Part VII. British Museum of Natural History. London. Esben-Petersen, P. 1929. Australian Neuroptera. [Part] VI. Qd Nat. 7:31-35. Esben-Petersen, P. [1935a] 1939 [2 January], Neuroptera from the Marquesas in Marquesan Insects III. Bull. Bernice P. Bishop Mus. 142:13-18. [Dating: From footnote on p. 142: 13.] Esben-Petersen, P. [1935b] 1939 [8 June]. Neuroptera from the Society Islands in Marquesan Insects III. Bull. Bernice P. Bishop Mus. 142:137-142. [Dating: From footnote on p. 142:137.] Esben-Petersen, P. 1937. Check list of Neuroptera Planipennia of Oceania. Occ. Pap. Bernice P. Bishop Mus. 13:49-60. Essig, E. O. 1910. The natural enemies of the citrus mealybug. I. Pomona Coll. J. Ent. 2: 1 43- 146. Fabricius, J. C. 1 777. Genera insectorvm eorvmqve characteres natvrales secvndvm nvmervm, figvram, sitvm et proportionem omnivm partivm oris adiecta mantissa speciervm nvper detectarvm. Chilonii. 310 pp. Fabricius, J. C. 1793. Entomologia systematica emendata et aucta secundum classes, ordines, genera, species adjectis synonimis, locis observationibus, descriptionibus. Hafniae. Tome 2. 519 pp. Farris, J. S. 1988. Hennig86 Reference. Version 1.5. Fitch, A. [1855] 1854. [Neuroptera.] Pp. 774-802 in Report [upon the noxious and other insects of the state of New-York]. Trans. N.Y. St. agric. Soc. 14:705-880. [Dating: From the title page of vol. 14, issued in 1855 with transactions for the year 1854. The identical text of pages 705-880 was reissued in 1856 as pages 1-176 (i.e., the First Report) of Fitch’s “First and second report on the noxious, beneficial and other insects, of the state of New-York.”] Fraser, F. C. 1951. A revision of the Madagascar Neuroptera with a key to their identifications and descriptions of new species. I. Osmylidae, Hemerobiidae and Chrysopidae. Natur- aliste malgache 3: 1 5-3 1 . Fraser, F. C. 1957. Odonata and Neuroptera of Reunion. Mem. Inst. Scient. Madagascar (E)8: 15-28. Fristrup, B. 1942 [20 April], Neuroptera and Trichoptera. Zoology of Iceland 3(43-44): 1-23. [Dating: From note on p. 3:23.] Fulmek, L. 1941. Uber die Aufzucht von Drepanepteryx phalaenoides L. ex ovo. (Neuroptera: Planipennia, Hemerobiidae). Arb. morph, taxon. Ent. Berl. 8:127-130. Gerstaecker, A. [1885] 1884. Zwei fernere decaden Australischer Neuroptera Megaloptera. Mitt, naturw. Ver. Neu-Vorpomm. 16:84-1 16. [Dating: From the wrapper and title page of vol. 16.] Gerstaecker, A. [1888] 1887. Weitere Beitrage zurartenkenntniss der Neuroptera Megaloptera. Mitt, naturw. Ver. Neu-Vorpomm. 19:89-130. [Dating: From the wrapper and title page of vol. 19.] 278 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Ghosh, S. K. 1977. A new genus and a new species of Neuroptera (Fam.: Hamerobiidae [sic]) from India. Proc. Indian Acad. Sci. (B)86:235— 237. Ghosh. S. K.; Sen. S. 1977. Check-list of Indian Planipennia (Order Neuroptera). Rec. Zool. Sun-. India 73:277-326. von Gleichen genannt Russworm. 1770. Versuch einer Geschichte der Blatlause und Bla- tlausfresser des Ulmenbaums. Niimberg. Russbiegel. [Not seen, cited in Horn & Shen- kling. Index Litteraturae Entomlogicae 1:432.] Gonzalez-Olazo. E. V. 1981. El genero Megalomus Rambur (Neurop.-Hemerobiidae) en Argentina y Chile. Acta zool. lilloana 36(2):97— 1 13. Gonzalez-Olazo. E. V. 1987. Notas sinonimicas sobre Anotiobiella withycombei Kimmins, 1929 (Neuroptera, Hemerobiidae). Acta zool. lill. 39:43^15. Gurney, A. B. 1 947. Notes on Dilaridae and Berothidae, with special reference to the immature stages of the Nearctic genera (Neuroptera). Psyche 54:145-169. Hagen. H. A. 1853. Hr. Peters berichtete iiber die von ihm gesammelten und von Hm. Dr. Hermann Hagen bearbeiteten Neuropteren aus Mossambique. Bericht fiber die zur Be- kanntmachung Geeigneten Verhandlungen der Konigl. Preuss. Akademie der Wissen- schaften zu Berlin 1 853:[479] — 482. Hagen. H. A. 1854. Ueberdie Neuropteren der Bemsteinfauna. Verh. zool-bot. Ges. 4:[22 1] — 232. Hagen, H. A. 1858. Synopsis der Neuroptera Ceylons [Pars I], Verh. zool-bot. Ges. 8:[47 1 ]— 488. Hagen. H. A. 1859. Synopsis der Neuroptera Ceylons. (Pars II.) Verh. zool-bot. Ges. 9:199- 212. Hagen. H. A. 1861 [July]. Synopsis of the Neuroptera of North America, with a list of the South American species. Smithson, misc. Colins 4(l):xx + 1-347. [Dating: From title page.] Hagen. H. A. 1866. Hemerobidarum Synopsis synonymica. Stettin, ent. Ztg 27:369^-62. Handlirsch. A. 1906-1908. Die fossilen Insekten und die Phvlogenie der rezenten Formen. W. Engelmann. Leipzig. 1430 pp. [Dating: From p. ix: 1906. pp. 1-640, pis. 1-36; 1907, pp. 641-1 120, pis. 37-51; 1908. pp. 1 121-1425.] Handschin. E. 1935. Indo-australische Neuropteren und Mecopteren. Revue suisse Zool. 42: 683-714. Handschin. E. 1955. Los insectos de las Islas Juan Fernandez. 15. Neuroptera. Revta chil. Ent. 4:3-20. Hennig. W. 1966. Phylogenetic Systematics. University of Illinois Press. Urbana. 263 pp. Henriksen. K. L. 1922. Eocene Insects from Denmark. Danm. geol. Unders. 2(37):[ 1 ]— 36. Henry, C. S. 1982. Neuroptera. Pp. 470-482 in Synopsis and Classification of Living Organ- isms, S. Parker, chief ed. Vol. 2. McGraw-Hill Book Co., New York. 1232 pp. Hilbum. D. J. 1 990. New records of Neuroptera. Mallophaga. and Strepsiptera from Bermuda. Fla Ent. 73:186-187. lilies. H. 1941. Das Insekten ftihrende Alttertiar von Havighorst. Z. Geschiebeforsch. Fla- chldgeol. 17:16-24. lilies. H. [1943] 1942 [30 April]. Uber das Eozan von Havighorst bei Hamburg und seine stratigraphische Stellung. Z. Geschiebeforsch. Flachldgeol. 18:42-49. [Dating: From title page of vol. 18.] Imperial Bureau of Entomology. [1924] 1922. Neuroptera, Mecoptera. Pp. 357-360 in The Zoological Record (Insecta). W. L. Schlater. superintendent. Vol. 59 (for literature year 1922). London. [Dating: From title page of vol. 59.] International Commission on Zoological Nomenclature. 1954. Opinion 211: Designation under the plenary powers of a type species for the genera “Hemerobius” Linnaeus, 1758, and “Chrysopa" Leach, 1815 (Class Insecta, Order Neuroptera) in harmony with ac- customed usage. Opin. Deck int. Comm. zool. Nom. 4:1-14. 1993 WORLD GENERA OF HEMEROBIIDAE 279 International Commission on Zoological Nomenclature. 1965. Opinion 752: Boriomyia Banks, 1904 (Insecta, Neuroptera): added to the official list of generic names. Bull. zool. Nom. 22:224-225. Jarzembowski, E. A. 1980 [27 March], Fossil insects from the Bembridge Marls, Palaeogene of the Isle of Wight, southern England. Bull. Br. Mus. nat. Hist., Geology 33:237-293. [Dating: From footnote on p. 33:237.] Kawashima, K. 1958. Bionomics and earlier stages of some Japanese Neuroptera (II) Eu- micromus numerosus (Navas) (Hemerobiidae). Mushi 32:43^16. Kevan, D. K. McE.; Klimaszewski, J. 1987. The Hemerobiidae of Canada and Alaska. Genus Hemerobius L. G. ital. Ent. 16:305-369. Kjllington, F. J. 1931. Notes on the life-history of Sympherobius fuscescens Wall. (=incon- spicuus McL.) [Neuroptera]. Entomologist 64:217-223. Killington, F. J. 1 934. On the life-histories of some British Hemerobiidae (Neur.). Trans. Soc. Br. Ent. 1:1 19-134. Kjllington, F. J. 1936. A monograph of the British Neuroptera. Vol. 1. Ray Society, London. xi + 269 pp. Kjllington, F. J. 1937. A monograph of the British Neuroptera. Vol. 2. Ray Society, London. xii + 306 pp. Kjllington, F. J. 1946. On Psectra diptera (Burnt.) (Neur., Hemerobiidae), including an account of its life-history. Entomologist’s mon. Mag. 82:161-176. Kimmins, D. E. 1928. New and little known Neuroptera of Central America. EOS, Madr. 4: 363-370. Kimmins, D. E. 1929. Some new and little known Argentine Neuroptera. Revta Soc. ent. argent. 2:187-192. Kimmins, D. E. 1933. New species of Notiobiella, from Africa and Seychelles (Neuroptera, Hemerobiidae). Entomologist 66:130-133. Kimmins, D. E. 1935. Notes on the genera Megalomus Rambur and Nesobiella, gen. nov. (Neuroptera), with descriptions of new species. Ann. Mag. nat. Hist. (10)16:602-619. Kimmins, D. E. 1936a. New species of Nusalala (Neuroptera, Hemerobiidae). Ann. Mag. nat. Hist. (10)17:568-576. Kimmins, D. E. 1936b. Odonata, Ephemeroptera, and Neuroptera of the New Hebrides and Banks Island. Ann. Mag. nat. Hist. (10)18:68-88. Kimmins, D. E. 1939. Supplementary notes on the life-history of Micromus variegatus (Fab.) (Neur.). J. Soc. Br. Ent. 1:240-241. Kimmins, D. E. 1940. New genera and species of Hemerobiidae (Neuroptera). Ann. Mag. nat. Hist. (1 1)6:222-236. Kimmins, D. E. 1943. New species of the genus Neuronema McL. (Neuroptera, Hemerobi- idae). Ann. Mag. nat. Hist. (1 1)10:40-53. Kimmins, D. E. 1958. Miss L. E. Cheesman’s expedition to New Hebrides, 1955. Orders Odonata, Neuroptera and Trichoptera. Bull. Br. Mus. nat. Hist., Entomology 6:239-250. Kimmins, D. E. 1963. Egnyonyx Wesmael, 1 836, a nomen oblitum in the family Hemerobiidae (Neuroptera). Entomologist’s mon. Mag. 98:202. Klimaszewski, J.; Kevan, D. K. McE. 1985. The brown lacewing flies of Canada and Alaska (Neuroptera: Hemerobiidae). Part I. The Genus Hemerobius Linnaeus: systematics, bionomics and distribution. McGill University, MacDonald College, Lyman Entomo- logical Museum and Research Laboratory, Memoire No. 15:iv + 1-1 19. Klimaszewski, J.; Kevan, D. K. McE. 1987. The brown lacewing flies of Canada and Alaska (Neuroptera; Hemerobiidae). Part II. The genus Wesmaelius Kruger. Neuropt. Int. 4: 153-204, 245-274. Klimaszewski, J.; Kevan, D. K. McE. 1988 [31 August]. The brown lacewing flies of Canada and Alaska (Neuroptera: Hemerobiidae). Part III. The genus Micromus Rambur. G. ital. Ent. 19:31-76. [Dating: From wrapper of separate.] 280 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Klimaszewski, J.; Kevan, D. K. McE.; Peck, S. B. [1988] 1987. A review of the Neuroptera of the Galapagos Islands with a new record for Sympherobius barberi (Banks) (Heme- robiidae). Can. J. Zool. 65:3032-3040. [Dating: From note in vol. 66, no. 3, p. iii.] Kristensen, N. P. 1981. Phylogeny of insect orders. A. Rev. Ent. 26:135-157. Kruger, L. 1922. Hemerobiidae. Beitrage zu einer Monographic der Neuropteren-Familie der Hemerobiiden. Stettin, ent. Ztg 83:138-172. Kruger, L. 1923. Neuroptera succinica baltica. Die im baltischen Bernstein eingeschlossenen Neuroptera des Westpreussischen Provinzial-Museums (heute Museum fur Naturkunde und Vorgeschichte) in Danzig. Stettin, ent. Ztg 84:68-92. Kuwayama, S. 1956. Further studies on the Neuroptera-Planipennia of the Kurile Islands. Insecta matsum. 20:77-82. Kuwayama, S. 1962. A revisional synopsis of the Neuroptera in Japan. Pacif. Insects 4:325- 412. Kuwayama, S. 1964. On the Neuroptera of the Ryukyus. Insecta matsum. 27:38-48. LafTranque, J. P.; Canard, M. 1975. Biologie du predateur aphidiphage Boriomyia subnebulosa (Stephens) (Neuroptera, Hemerobiidae): Etudes au laboratoire et dans les conditions hivemales du Sud-Ouest de la France. Annls Zool-Ecol. Amm. 7:331-343. Lambkin, K. J. 1986. A revision of the Australian Mantispidae (Insecta: Neuroptera) with a contribution to the classification of the family. I. General and Drepanicinae. Aust. J. Zool., Supplry Ser. 116:1-142. Larsson, S. G. 1975. Palaeobiology and mode of burial of the insects of the Lower Eocene Mo-clay of Denmark. Bull. geol. Soc. Denmark 24:193-209. Latreille, P. A. 1802. Histoire naturelle, generate et particuliere de Crustaces et des Insectes. Vol. 3. Dufart, Paris. 467 pp. Latreille, P. A. 1810. Considerations generates sur l’ordre naturel des animaux composant les classes des Crustaces, des Arachnides, et des Insectes; avec un tableau methodique de leurs genres, disposes en families. Schoell, Paris. 444 pp. [Facsimile reprint: 1970, Gregg International Pubis. Ltd., Westmead, England.] Leach, W. E. 1815 [April], Entomology. Pp. 57-172 in Edinburgh Encyclopaedia, D. Brewster, ed. Vol. 9, pt. 1, Edinburgh. [Dating: From Sherbom, C. D., 1937, J. Soc. Biblphy nat. Hist. 1:112.] Leraut, P. [1981] 1980 [30 April], Liste des Planipennes de France [Neuroptera], Bull. Soc. ent. Fr. 85:237-253. [Dating: From note on p. 85:308.] Linnaeus, C. 1758. Systema natura per regna tria naturae secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. 10th Ed. Vol. 1. Salvii, Hol- miae. 824 pp. Linnaeus, C. 1767. Systema natura per regna tria naturae secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. 12th Ed. Vol. 1, pt. 2 (pp. 533— 1327). Salvii, Holmiae. Mackerras, I. M. 1970. Chapter 9. Composition and distribution of the fauna [of Australia]. Pp. 187-203 in The insects of Australia. Melbourne University Press, Melbourne. 1029 pp. MacLeod, E. G. [1961] 1960 [16 February], The immature stages of Boriomyia fidelis (Banks) with taxonomic notes on the affinities of the genus Boriomyia (Neuroptera: Hemero- biidae). Psyche 67:26-40. [Dating: From title page of vol. 67.] MacLeod, E. G. 1964. The presence of the genus Hemerobius in Hawaii with a note on the wing venation of Nesobiella hospes (Perkins) (Neuroptera: Hemerobiidae). Pacif. Insects 6:1-4. MacLeod, E. G.; Adams, P. A. [1968] 1967 [20 February], A review of the taxonomy and morphology of the Berothidae, with the description of a new subfamily from Chile (Neuroptera). Psyche 74:237-265. [Dating: From inside of cover of vol. 74, no. 4.] Maddison, W. P.; Donoghue, M. J.; Maddison, D. R. 1 984. Outgroup analysis and parsimony. Syst. Zool. 33:83-103. 1993 WORLD GENERA OF HEMEROBIIDAE 281 Makarkin, V. N. 1985. A Review of Hemerobiidae (Neuroptera) of the fauna of the USSR. I. The genera Hemerobius L., Micromus Ramb.. and Paramicromus Nakah. [in Russian]. Ent. Obozr. 64:158-170. [English translation: 1985, Ent. Rev. 64(2):99— 1 13.] Makarkin, V. N. 1986. A Review of Hemerobiidae (Neuroptera) of the fauna of the USSR. II. The genera Wesmaelius Kruger, Sympherobius Banks, Psectra Hagen, Megalomus Ramb., Neuronema MacLach. and Drepanepteryx Leach [in Russian], Ent. Obozr. 65: 604-617. [English translation: 1987, Ent. Rev. 66(2):99— 1 13.] Martynov, A. V. 1928. Permian fossil insects of north-east Europe. Trav. Mus. geol. Acad, sci. URSS, Leningrad 4:1-1 18. Matsumura, S. 1907. Systematic Entomology [in Japanese] (transliterated title: Konchu Bun- ruigaku). Vol. 1. Keiseisha Co., Tokyo. 336 pp. [Not seen, cited in Kuwayama, 1962.] Matthews, J. V., jr. 1980. Tertiary land bridges and their climate: backdrop for development of the present Canadian insect fauna. Can. Ent. 1 12:1089-1 103. McLachlan, R. 1863. On some new species of neuropterous insects from Australia and New Zealand, belonging to the family Hemerobiidae. J. Ent., Lond. 2:1 1 1-1 16. McLachlan, R. 1868. A monograph of the British Neuroptera-Planipennia. Trans, ent. Soc. Lond. 1868:145-224. McLachlan, R. 1869. New species, &c., of Hemerobiina; with synonymic notes (first series). Entomologist’s mon. Mag. 6:21-27. Miermont, Y.; Canard, M. 1975. Biologie du predateur aphidiphage Eumicromus angulatus (Neur.: Hemerobiidae): Etudes au laboratoire et observations dans le sud-ouest de la France. Entomophaga 20:179-191. Miles, H. W. 1 924. On the life history of Boriomyia (Hemerobius) nervosa. Fab. (Planipennia, Hemerobiidae). Bull. ent. Res. 14:249-250. Miller, G. L.; Cave, R. D. 1987. Bionomics of Micromus posticus (Walker) (Neuroptera: Hemerobiidae) with descriptions of the immature stages. Proc. ent. Soc. Wash. 89:776- 789. Miller, G. L.; Lambdin, P. L. 1 982. Hemerobius stigma Stephens (Neuroptera: Hemerobiidae): external morphology of the egg. Proc. ent. Soc. Wash. 84:204-207. Miller, G. L., Lambdin, P. L. 1984. Redescriptions of the larval stages of Hemerobius stigma Stephens (Neuroptera: Hemerobiidae). Fla Ent. 67:377-382. Monserrat, V. J. 1980. Sobre las especies africanas del genero Notiobiella Banks, 1909 (Neu- roptera, Planipennia, Hemerobiidae). I. Neuropt. Int. 1:26-33. Monserrat, V. J. 1983 [June]. Sobre los Neuropteros de las Islas Canarias, IV: Wesmaelius (Kimminsia) navasi (Andreu, 1911) (Neur., Plan., Hemerobiidae). Boln Asoc. esp. Ent. 6:209-224. [Dating: From header on p. 209 of separate.] Monserrat, V. J. 1984a. Sobre las especies Africanas del genero Notiobiella Banks, 1909, II. (Neuroptera, Planipennia, Hemerobiidae). Pp. 99-124 in Progress in World’s Neurop- terology. Proceedings of the 1st International Symposium on Neuropterology, J. Gepp, H. Aspock, and H. Holzel, eds. Graz, Austria. 265 pp. [Symposium held in Graz, 1980.] Monserrat, V. J. 1984b. Sobre las especies Americanas del genero Notiobiella Banks, 1909, I. (Neuroptera, Planipennia, Hemerobiidae). Pp. 85-97 in Progress in World’s Neurop- terology. Proceedings of the 1st International Symposium on Neuropterology, J. Gepp, H. Aspock, and H. Holzel, eds. Graz, Austria. 265 pp. [Symposium held in Graz, 1980.] Monserrat, V. J. 1990. Some brown lacewings from Equatorial Guinea (Neuropteroidea, Planipennia: Hemerobiidae). J. Afr. Zool. 104:17-22. Monserrat, V. J.; Penny, N. D. 1983. Sobre las especies Americanas del genero Notiobiella Banks, 1909, II (Neuroptera, Planipennia, Hemerobiidae). Nouv. Revue Ent. 13:125- 132. Moore, F. [1892] 1890-1892. Lepidoptera Indica. Vol. 1. L. Reeve and Co., London. 317 pp. Morton, K. J. 1906. Notes on certain Palaearctic species of the genus Hemerobius: the Ma- 282 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) deira-Canarian species allied to H. humuli, and other species from the same islands. Entomologist’s mon. Mag. 42:146-148. Morton, K. J. 1910. Life-history of Drepanepteryx phalaenoides, Linn. Entomologist’s mon. Mag. 46:54-62. Moznette. G. F. 1915a. Notes on the brown lace-wing (Hemerobius pacificus, Bks.). J. econ. Ent. 8:350-354. Moznette. G. F. 1915b. The brown lacewing Hemerobius pacificus Banks. Pp. 181-183 in Second Biennial Crop Pest and Horticultural Report, 1913-1914. Oregon Agricultural College Experiment Station. Corvallis, Oregon. Nakahara. W. 1915. On the Hemerobiinae of Japan. Annotnes zool. jap. 9:1 1 — 48. Nakahara. W. 1919. Revisional notes on the Japanese Hemerobiidae [in Japanese], Insect Wld [=Konchu Sekai] 23:135-137. [Note: English translation of Paramicromus descrip- tion in Nakahara. W., 1956, Kontyu 24:182-191.] Nakahara. W. 1954. Early stages of some Japanese Hemerobiidae including two new species. Kontyu 21:41-46. Nakahara. W. 1955. Formosan Neuroptera collected by the late Dr. T. Kano. Kontyu 23: 6-12. Nakahara. W. 1960a [March]. The Hemerobiidae from New Caledonia. Bull. Osaka Mus. nat. Hist. 12:39^41. [Dating: From wrapper of No. 12.] Nakahara. W. 1960b. Systematic studies on the Hemerobiidae (Neuroptera). Mushi 34:1-69. Nakahara. W. 1965. Neotropical Hemerobiidae in the United States National Museum. Proc. U.S. Natn Mus. 1 17:107-122. Nakahara. W. 1 966. Hemerobiidae. Sisyridae and Osmylidae of Formosa and Ryukyu Islands (Neuroptera). Kontyu 34:193-207. Navas. L. 1 906. Catalogo descriptivo de los insectos Neuropteros de las Islas Canarias. Revta R. Acad. Cienc. exact, fis. nat. Madr. 4:687-706. Navas, L. 1908. Neuropteros nuevos. Mems R. Acad. Cienc. Artes Barcelona (3)6:40 1 — 423 [Separate: (3)6(25):[1 ]— 25]. Navas, L. 1909a. Neuropteros de los alrededores de Madrid. Suplemento I. Revta R. Acad. Cienc. exact, fis. nat. Madr. 8:370-380. Navas. L. 1909b. Sur deux Hemerobides (insectes nevropteres) nouveaux. Annls Soc. scient. Brux. 33(pt. 1):2 15-220. Navas, L. [1910] 1909 [14 March]. Hemerobides nouveaux du Japon (Neuroptera). Rev. Russ. Ent. 9:395-398. [Dating: From wrapper of vol. 9, no. 4.] Navas, L. 1912a [June], Insectos neuropteros nuevos o poco conocidos. Mems R. Acad. Cienc. Artes Barcelona (3)10:135-202 [Separate: (3)10(9):[1 ]— 70]. [Dating: From wrapper and title page of ser. 3, vol. 10, no. 9.] Navas, L. 1912b. Quelques Nevropteres de la Siberie meridionale-orientale. Rev. Russ. Ent. 12:414-422. Navas. L. 1913a. Mis excursiones por el extranjero en el verano de 1912 (25 julio-16 sep- tiembre). Mems R. Acad. Cienc. Artes Barcelona (3)10:479-514 [Separate: (3)10(24): [11-38], Navas. L. 1913b. Nevropteres. Pp. 69-77 in Mission du Service Geographique de l’Armee pour la mesure l'un arc de meridien equatorial en Amerique du Sud sous le controle scientifique de l'academie des sciences 1899-1906. Vol. 10 (Entomologie-Botanique), Fascicule 1 (Hymenopteres. Orthopteres, Nevropteres, Araignees). Paris. Navas, L. 1913c. [Ninga nom. nov. for Ninguta Navas, 1912 and other remarks.] Boln Soc. aragon. Cienc. nat. 12:122-123. Navas, L. 1915a [December], Neuropteros nuevos o poco conocidos (Sexta [VI] serie). Mems R. Acad. Cienc. Artes Barcelona (3)12: 1 19-136 [Separate: (3)12(7):[1 ]— 20]. [Dating: From wrapper of ser. 3, vol. 12, no. 7.] 1993 WORLD GENERA OF HEMEROBIIDAE 283 Navas, L. [1915b] 1914-1915 [20 February], Ordo Neuroptera. P. 263 and Sympherobius amicus Navas sp. n. Pp. 332-333 in F. Silvestri, Contribute alia conoscenza degli insetti dell’oliva dell’Eritrea e dell’Africa meridionale. Boll. Lab. Zool. gen. agr. Portici 9:240- 334. [Dating: From date given to Silvestri work in index to vol. 9.] Navas, L. 1922. Insectos exoticos. Broteria (Zoologica) 20:49-63. Navas, L. [1923] 1922 [16 November], Algunos insectos del Museo de Paris. [1 .a serie.] Revta R. Acad. Cienc. exact, fis.-quim. nat. Zaragoza 7:15-51. [Dating: From note on p. 7: 266.] Navas, L. 1924a. Algunos insectos de Chile. 2. a serie. Revta chil. Hist. nat. 28:12-16. Navas, L. [1924b] 1923 [5 February], Fauna de Catalunya. Entomologia de Catalunya. Neu- ropters. Fascicle I. Neuropters Propis. Publicacions de l’lnstitut d’Estudis Catalans, seccio de Ciencies, Barcelona. 270 pp. [Dating: From footline on p. [271].] Navas, L. 1924c. Insecta orientalia. Ill Series. Mem. Accad. pont. Nuovi Lincei (2)7:217- 228. Navas, L. 1925a [June], Insectos exoticos nuevos o poco conocidos. Segunda [II] serie. Mems R. Acad. Cienc. Artes Barcelona (3) 19: 18 1-200 [Separate: (3)19(5):[1 ]— 22], [Dating: From wrapper of ser. 3, vol. 19, no. 5.] Navas, L. [1925b] 1924 [1 1 April], Sinopsis de los Neuropteros (Ins.) de la peninsula iberica. Mems Soc. iber. Cienc. nat. 4:[ 1 ]— [ 1 50]. [Dating: From footnote on p. 4:[ 1 50].] Navas, L. 1926. Insecta orientalia. IV series. Mem. Accad. pont. Nuovi Lincei (2)9:1 1 1-120. Navas, L. 1929a. Insecta nova. Series XIII. Mem. Accad. pont. Nuovi Lincei (2)12:15-23. Navas, L. 1 929b. Insectos exoticos Neuropteros y afines del Museo Civico de Genova. Annali Mus. civ. Stor. nat. Giacomo Doria 53:354-389. Navas, L. [1929c] 1928 [February]. Insectos neotropicos. 4. a serie. Revta chil. Hist. nat. 32: 106-128. [Dating: From wrapper of vol. 32.] Navas, L. 1929d. Insectos Neuropteros del Museo de Hamburgo. Mems R. Soc. esp. Hist, nat. 15:315-322. Navas, L. 1929e. [Insecta orientalia.] VII Series. Mem. Accad. pont. Nuovi Lincei (2)12:43- 56. Navas, L. 1932. Neuropteros de Haiti. Boln Soc. ent. Esp. 15:33-37. Navas, L. 1933. De las cazas del Sr. Gadeau de Kerville en el Asia Menor. Pp. 221-225 in Travaux, Ve Congres International d’Entomologie (held in Paris, 1932). Vol. 2. Navas, L. 1935 [April]. Neuropteros exoticos. 2. a serie. Mems R. Acad. Cienc. Artes Barcelona (3)25:37-59 [Separate: (3)25(3):[1 ]— 25]. [Dating: From title page of ser. 3, vol. 25, no. 3.] Navas, L. 1936. Nevropteres et insectes voisins. Chine et pays environnants. Neuvieme [IX] serie. Notes Ent. chin. 3:37-62, 1 17-132. Needham, J. G. 1905. The summer food of the Bullfrog (Rana catesbiana) at Saranac Inn. Pp. 9-17 in May Flies and Midges of New York, Third Report on Aquatic Insects, J. G. Needham, K. J. Morton, and O. A. Johannsen. N.Y. St. Mus., Bull. 86 (Entomology 23). 352 pp. Needham, J. G. 1 909. Notes on the Neuroptera in the collection of the Indian Museum. Rec. Indian Mus. 3:185-210. New, T. R. 1966. Some notes on the biology of Psectra diptera (Burm.) (Neuroptera, Hem- erobiidae). Entomologist’s Gaz. 17:79-82. New, T. R. 1967a. Biological notes on Sympherobius pellucidus (Walk.), with a description of the larva (Neuroptera: Hemerobiidae). Entomologist’s Gaz. 18:50-52. New, T. R. 1967b. Sympherobius pygmaeus (Ramb.) (Neur., Hemerobiidae) in Berkshire, with a description of the cocoon and pupa. Entomologist’s mon. Mag. 103:171-172. New, T. R. 1975. The immature stages of Drepanacra binocula (Neuroptera: Hemerobiidae), with notes on the relationships of the genus. J. Aust. ent. Soc. 14:247-250. New, T. R. 1988. A revision of the Australian Hemerobiidae (Insecta: Neuroptera). Invert. Taxon. 2:339-41 1. 284 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) New. T. R. [1989] 1988. Hemerobiidae (Insecta: Neuroptera) from New Guinea. Invert. Taxon. 2:605-632. [Dating: From inside cover of vol. 2, no. 5.] New, T. R.; Boros. C. 1983. The early stages of Micromus tasmaniae (Neuroptera: Heme- robiidae). Neuropt. Int. 2:213-217. Newman. E. 1835. Attempted division of British insects into natural orders. Ent. Mag. 2: 379-431. Newman. E. 1838. Entomological Notes. Ent. Mag. 5:168-181, 372^102, 483-500. Nichols, S. W. 1989. The Torre-Bueno glossary of entomology. Revised Edition, compiled by Stephen W. Nichols. New York Entomological Society, New York. 840 pp. Nielsen. J. C. 1909. The insects of East-Greenland. Medd. Gronland 29:[363H*09. [Dating: From wrapper of vol. 29, no. 2.] Ohm. P.: Holzel. H. 1982. Tiergeographische und Okologische Aspekte der Neuropterenfauna der Kapverden. Courier Forschungsinstitut Senckenberg 52:159-165. Okamoto, H. 1905. Neuropterous insects of Hokkaido [in Japanese], Trans. Sapporo nat. Hist. Soc. 1:111-117. Oswald. J. D. 1988a. A review of the South Pacific genus Austromegalomus Esben- Petersen (Neuroptera: Hemerobiidae) with a description of a new species from Rapa. Proc. ent. Soc. Wash. 90:55-61. Oswald. J. D. 1988b. A revision of the genus Sympherobius Banks (Neuroptera: Hemero- biidae) of America north of Mexico with a synonymical list of the world species. J. N.Y. ent. Soc. 96:390—451. Oswald. J. D. 1990. Revision of the Neotropical brown lacewing genus Nomerobius (Neu- roptera: Hemerobiidae). Ann. ent. Soc. Am. 83:18-29. Oswald. J. D.; Penny. N. D. 1991. Genus-group names of the Neuroptera, Megaloptera and Raphidioptera of the World. Occ. Pap. Calif. Acad. Sci. 147:1-94. Panfilov. D. V. 1980. New representatives of lacewings (Neuroptera) from the Jurassic of Karatau [in Russian]. Pp. 82-1 1 1 in Mesozoic Fossil Insects, V. Dolin, D. Panfilov, A. Ponomarenko and L. Pritikina, eds. Naukova Dumka. Kiev. Papavero. N. 1971. Essays on the history of Neotropical Dipterology, with special reference to collectors (1750-1905). Vol. 1. Museu de Zoologia. Universidade de Sao Paulo, Sao Paulo, vii + 1-216. Penny, N. D. [1978] 1977. Lista de Megaloptera, Neuroptera e Raphidioptera do Mexico, America Central, ilhas Caraibas e America do Sul. Acta Amazonica 7(4)(Suplemento): 1-61. [Dating: Iowa State University Library date of receipt: 15 September 1978.] Penny, N. D.; Monserrat, V. J. [1985] 1983. Neuroptera of the Amazon Basin. Part 10. Hemerobiidae. Acta Amazonica 13:879-909. [Dating: Cornell University Library date of receipt: 29 October 1985.] Perkins, R. C. L. 1899 [25 September], Neuroptera. Pp. 31-89 in Fauna Hawaiiensis being the land-fauna of the Hawaiian Islands, D. Sharp, ed. Vol. 2. Cambridge University Press, London. [Dating: From cover page of vol. 2. pt. 2.] Perkins, R. C. L. 1913. Review of Neuroptera. Pp. clxx-clxxx in Fauna Hawaiiensis being the land-fauna of the Hawaiian Islands, D. Sharp, ed. Vol. 1 (Introduction). Cambridge University Press. London. Philippe. R. 1972. Les appareils genitaux male et femelle de Chrysopa perla [Neuroptera] etude anatomique histologique et fonctionnelle. Annls Soc. ent. Fr. (N.S.)8:693-705. Pictet-Baraban, F. J.; Hagen, H. A. 1856. Die im Bernstein befindlichen Neuropteren der Vorvelt. Bd. 2, Abt. 2 in Berendt, G. C., Die im Bernstein befindlichen organischen reste der Vorwelt. Berlin. Ping. C. 1928 [30 December], Cretaceous fossil insects of China. Palaeont. sin. (B) 1 3(1): 1— 56. [Dating: From note on p. [3].] Quayle. H. J. 1912. Red spiders and mites of citrus trees. Univ. Calif, agric. Exp. Sta. Bull. 234:483-530. 1993 WORLD GENERA OF HEMEROBIIDAE 285 Rafinesque, C. S. 1815. Analyse de la nature ou tableau de l’univers et des corps organises. Palerme. 224 pp. Rambur, M. P. 1842 [31 December], Hemerobides. Pp. 414-430 in Histoire Naturelle des Insectes, Nevropteres. Fain et Thunot. Paris. 534 pp. [Dating: See Esben-Petersen, 1922, Ann. Mag. Nat. Hist. (9)10:621.] Richard, P. W.; Harmston, F. C. 1972. Diptera and other arthropods of the Sukkertoppen Tasersiaq Area, southwest Greenland. Arctic 25:107-1 14. Riek, E. F. 1970. Neuroptera (Lacewings). Pp. 472—494 in The insects of Australia. Melbourne University Press, Melbourne. 1029 pp. Schluter, T. 1986. The fossil Planipennia— a review. Pp. 103-111 in Recent Research in Neuropterology. Proceedings of the 2nd International Symposium on Neuropterology, J. Gepp, H. Aspock, and H. Holzel, eds., Graz, Austria. 176 pp. [Symposium held in Hamburg, 1984], Schneider, W. G. 1851. Symbolae ad monographiam generis Chrysopae, Leach. Hirt, Vra- tislaviae. 178 pp. Scudder, S. H. [1878] 1876-1877. Additions to the insect-fauna of the Tertiary beds at Quesnel, British Columbia. Geological Survey of Canada. Report of Progress 1 876—1 877:457— 464. [Dating: From wrapper and title page of vol. for 1876-1877.] Scudder, S. H. 1890. Suborder Planipennia Burmeister. Pp. 146-172 in The Tertiary Insects of North America. Report of the United States Geological and Geographical Survey of the Territories, Vol. 13. Scudder, S. H. 1891. Index to the known fossil insects of the world including myriapods and arachnids. Bull. U.S. geol. Surv. 71:1-744. Sharp, D. [1908] 1906. Neuroptera. Pp. 371-373 in The Zoological Record (Insecta), D. Sharp, superintendent. Vol. 43 (for literature year 1906). London. [Dating: From title page of vol. 43.] Smith, R. C. 1923. The life histories and stages of some hemerobiids and allied species (Neuroptera). Ann. ent. Soc. Am. 16:129-151. Smith, R. C. 1931. The Neuroptera of Haiti, West Indies. Ann. ent. Soc. Am. 24:798-823. Sneath, P. H. A.; Sokal, R. R. 1973. Numerical Taxonomy. W. H. Freeman and Co., San Francisco. 573 pp. Snodgrass, R. E. 1935. Principles of Insect Morphology. McGraw-Hill Book Co., New York. 667 pp. Stelzl, M. 1990. Nahrungsanalytische Untersuchungen an Hemerobiiden-imagines (Insecta, Planipennia). Mitt, dtsch. Ges. allg. angew. Ent. 7:670-676. Stange, L. A. 1967. Catalogo de Neuroptera de Argentina y Uruguay. Acta zool. lilloana 22: 5-87. Steinmann, H. 1967. Raphidioptera, Megaloptera, Neuroptera es Mecoptera [in Hungarian], Fauna Hung. 82:1-203. Stephens, J. F. 1829. Neuroptera. Pp. 304-316 in A systematic catalogue of British insects: being an attempt to arrange all the hitherto discovered indigenous insects in accordance with their natural affinities. Vol. 1. London. Stephens, J. F. 1836. Illustrations of British entomology; . . .Mandibulata. Baldwin and Cra- dock, London. Vol. 6. 240 pp. Strom, H. 1788. Nogle insect larver med deres forvandlinger. Nye Samling af det Kongelige Norske Videnskabers Selskabs Skrifter, Kobenhaven 2:375— 400c. Swainson, W.; Shuckard, W. E. 1840. On the Neuroptera. Pp. 335-358 in On the history and natural arrangement of insects. London. 406 pp. Terry, F. W. 1908. Notes on the life-history of an endemic hemerobiid (Nesomicromus vagus Perk.). Proc. Hawaii, ent. Soc. 1:174-175. Thomas, I. 1914. The British Carboniferous Producti. I. Genera Pustula and Overtonia. Mem. geol. Surv. U.K. 1:197-366. 286 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Tillyard, R. J. 1916. Studies in Australian Neuroptera. No. iv. The families Ithonidae, Hem- erobiidae, Sisyridae. Berothidae, and the new family Trichomatidae; with a discussion of their characters and relationships, and descriptions of new and little-known genera and species. Proc. Linn. Soc. N.S.W. 41:269-332. Tjeder, B. 1931. Boriomyia persica Mort., rava With., and Baltica n. sp. Ent. Tidsskr. 52: 1-12. Tjeder. B. 1936 [31 December], Schwedisch-chinesische wissenschaftliche expedition nach den nordwestlichen provinzen Chinas, unter leitung von Dr. Sven Hedin und Prof. Sii Ping-chang. Insekten gesammelt vom schwedischen arzt der expedition Dr. David Hum- mel 1927-1930. 62. Neuroptera. Ark. Zool. 29A(8):l-36. [Dating: From footnote on p. 36.] Tjeder, B. 1941. Some remarks on “The generic names of the British Neuroptera.” Ent. Tidsskr. 62:24-31. Tjeder. B. 1961. Neuroptera-Planiepnnia. The Lace-wings of Southern Africa. 4. Family Hemerobiidae. Pp. 296-408 in South African Animal Life, B. Hanstrom, P. Brinck, and G. Rudebec, eds. Vol. 8. Swedish Natural Science Research Council, Stockholm. Tjeder, B. 1963a. A new Dyshemerobius from northern Rhodesia (Neur. Hemerobiidae). Opusc. Ent. 28:181-184. Tjeder. B. 1963b. Comments on Hemerobius algidus Erichson, with notes on the genus Drepanopteryx [sic] (Neur. Hemerobiidae). Opusc. Ent. 28:171-180. Tjeder, B. 1964. Neuroptera, Trichoptera and Diptera-Tipulidae from Iceland with a rede- scription of Rhabdomastix parva Siebke. Opusc. Ent. 29:143-151. Tjeder, B. 1975. A new genus and species of the Hemerobiidae from Nigeria (Insecta: Neu- roptera). Ent. Scand. 6:1 15-1 18. Tjeder, B. 1976 [December], La faune terrestre de l’ele de Sainte-Helene. 11. Neuroptera. Musee Royal de l'Afrique Centrale, Annales, Serie in-8°. Sciences Zoologiques No. 215: 2-17. [Dating: From reverse of title page of No. 215.] Van der Week, H. W. 1909 [25 January], Mecoptera and Planipennia of insulinde. Notes Leyden Mus. 31:1-100. [Dating: From note on p. 31:vi.] Van der Weele. H. W. 1910a [30 December], On Panorpata and Planipennia collected in western Java. Notes Leyden Mus. 32:199-202. [Dating: From wrapper of vol. 32, no. 4.] Van der Weele, H. W. 1910b. Megaloptera. Collections Zoologiques du Baron Edm. de Selys Longchamps. Catalogue Systematique et Descriptif. Fascicle 5:1-93. Van der Weele, H. W. 1910c. Neuroptera. 2. Planipennia et Panorpata. Pp. 11-23 in Wis- senschaftliche ergebnisse der Schwedischen Zoologischen expedition nach dem Kili- mandjaro. dem Meru und den Umgebenden massaisteppen Deutsch-Ostafrikas 1905- 1906, Y. Sjostedt, ed. Vol. 2, No. 13. Walker, F. 1853. List [Catalogue] of the specimens of neuropterous insects in the collection of the British Museum, Part II (pp. 193-476). (Sialides-Nemopterides). British Museum [Natural History], London. Walker, F. 1860. Characters of undescribed Neuroptera in the collection of W. W. Saunders. Trans, ent. Soc. Lond. (N.S.)5: 176-1 99. Wallengren. H. D. J. 1863. Bidrag till kannedomen af Sveriges Neuroptera. Ofvers. K. VetenskAkad. Forh. 20:15-26. Watrous, L. E.; Wheeler. Q. D. 1981. The out-group comparison method of character analysis. Syst. Zool. 30: 1-1 1 . Weidner. H. 1972. Die Entomologischen Sammlungen des Zoologischen Instituts und Zool- ogischen Museums der Universitat Hamburg. VIII Teil. Insecta V. Mitt. hamb. Zool. Mus. Inst. 68:107-134. Wesmael. C. 1836. Description d'un nouveau genre de Nevropteres, famille des Planipennes, tribu des Hemerobins. Bull. Acad. R. Sci. Belles-Lett. Brux. 3:166-168. 1993 WORLD GENERA OF HEMEROBI1DAE 287 Westwood, J. O. [1838] 1838-1840 [November], Order V. Neuroptera Linn. Pp. 46 — 48 in Synopsis of the genera of British insects. London. 154 pp. [Dating: From International Commission on Zoological Nomenclature, Direction 63.] Wiley, E. O. 1981. Phylogenetics, the theory and practice of phylogenetic systematics. John Wiley & Sons, New York. 439 pp. Williams, F. X. 1931. Order Neuroptera. Pp. 128-132 in Handbook of the insects and other invertebrates of Hawaiian Sugar Cane Fields. Experiment Station of the Hawaiian Sugar Planters’ Association, Honolulu. Wise. K. A. J. 1973. New records in the New Zealand Neuroptera: Hemerobiidae. N. Z. Ent. 5:181-185. Wise, K. A. J. 1977 [21 December], A synonymic checklist of the Hexapoda of the New Zealand sub-region. Bull. Auckland Inst. Mus. 1 l:iii+ 1-176. [Dating: From reverse of title page of No. 1 1 .] Wise, K. A. J. 1983 [23 December]. Lacewings and aquatic insects of New Zealand. 2. Fauna of the northern offshore islands. Records of the Auckland Institute and Museum 20: 259-271. [Dating: From wrapper of vol. 20. and footnote on p. 259.] Withycombe, C. L. 1922. The life-history of Hemerobius stigma, Steph. Entomologist 55: 97-99. Withycombe, C. L. [1923] 1922 [17 February]. Notes on the biology of some British Neu- roptera (Planipennia). Trans, ent. Soc. Lond. 1922:501-594. [Dating: From reverse of title page of vol. for 1922.] Withycombe, C. L. 1924. Further notes on the biology of some British Neuroptera. Ento- mologist 57:145-152. Withycombe, C. L. [1925] 1924-1925 [20 February], Some aspects of the biology and mor- phology of the Neuroptera. With special reference to the immature stages and their possible phylogenetic significance. Trans, ent. Soc. Lond. 1 924:303—4 1 1. [Dating: From reverse of title page of vol. for 1924.] Wolcott, G. N. [1950] 1948 [1 September], [Neuroptera s. str.] Pp. 88-92 in The insects of Puerto Rico. J. Agric. Univ. P. Rico 32:1-975. [Dating: From note on last page of vol. 32. no. 1 .] Yang, C.-k. 1 964. Notes on the genus Neuronema of China (Neuroptera; Hemerobiidae). Acta zootax. Sin. [=Tung Wu Fen Lei Hsueh Pao, Peichins] 1:261-282. Yang, C.-k. 1980a. Some new species of the genera Wesmaelius and Kimminsia (Neuroptera: Hemerobiidae). Acta ent. Sin. [=K’un Ch’ung Hsueh Pao] 23:54-65. Yang, C.-k. 1980b. Three new species of Sympherobius from China (Neuroptera: Hemero- biidae) [in Chinese]. Acta agric. Univ. Pek. 6:87-92. Yang, C.-k. 1981. Neuroptera: Hemerobiidae. Pp. 301-321 in Zhongguo Kexueyuan Qingzang Gaoyuan Zonghe Kexue Kaocha Dui [=The Series of the Comprehensive Scientific Expedition to the Qinghai-Xizang Plateau]. Xizang kunchong [=Insects of Xizang (Ti- bet)]. Vol. 1. Science Press, Peking, xi + 600 pp. Yang, C.-k. 1983. Semohemerobius nom. nov. for Mesohemerobius Nakahara 1966 (nee Ping 1928) [in Chinese]. Entomotaxonomia 5:128. Zimmerman, E. C. [1940] 1 939 [August]. Studies of Hawaiian Neuroptera. Proc. Hawaii, ent. Soc. 10:487-510. [Dating: From wrapper of vol. 10. no. 3.] Zimmerman, E. C. 1957. Order Neuroptera. Pp. 19-169 in Insects of Hawaii. Vol. 6. Uni- versity of Hawaii Press, Honolulu, Hawaii. 209 pp. Zoological Record. Literature year 1877. Neuroptera [s. lat.]. Pp. 198-206 in The Zoological Record. Vol. 14, Insecta. John Van Voorst, London. 288 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Appendix 1. Sequenced classification of world Hemerobiid genera. This classification in- corporates the conventions recommended by Wiley (1981: 205-213) for annotated Linnaean classifications. See also the discussion under the heading Phylogenetic Analysis: Results: Clas- sification. Classification of world Hemerobiidae Family Hemerobiidae Latreille Subfamily Carobiinae Oswald, new subfamily Genus Carobius Banks Subfamily Hemerobiinae Latreille, sedis mutabilis Hemerobius Genus-group Genus Hemerobius Linnaeus Genus Nesobiella Kimmins Wesmaelius Genus-group Genus Wesmaelius Kruger Genus Hemerobiella Kimmins Subfamily Sympherobiinae Comstock, sedis mutabilis Genus Nomerobius Navas Genus Neosympherobius Kimmins Genus Sympherobius Banks Subfamily Psychobiellinae Oswald, new subfamily Genus Psychobiella Banks Subfamily Notiobiellinae Nakahara Genus Notiobiella Banks Genus Psectra Hagen Genus Anapsectra Tjeder Genus Zachobiella Banks Subfamily Drepanacrinae Oswald, new subfamily Genus Conchopterella Handschin Genus Austromegalomus Esben-Petersen Genus Drepanacra Tillyard Subfamily Megalominae Kruger Genus Megalomus Rambur Subfamily Drepanepteryginae Kruger Genus Neuronema McLachlan Genus Gayomyia Banks Genus Drepanepteryx Leach in Brewster Subfamily Microminae Kruger Genus Noius Navas Genus Nusalala Navas Genus Megalomina Banks Genus Micromus Rambur Unplaced Genera Genus Notherobius New Nomina Dubia Genus Subboriomyia Steinmann Neuropterida Incertae Sedis Genus Ormiscocerus Blanchard in Gay 1993 WORLD GENERA OF HEMEROBIIDAE 289 Appendix 2. Hemerobiid species richness and species examined. This appendix summarizes the known species richness of all higher hemerobiid taxa, and lists the species examined for this work. Some examined species could not be identified to species level; such species are collectively recorded under the appropriate genus as “unknown spp.” Symbols: *. Includes one or more unidentified or undescribed species; +, Species selected for comprehensive examination (see discussion under the heading Phylogenetic Analysis: Methods and Data: Data)\ — , Includes species-group names of particularly doubtful validity; ?, Uncertain identification; M, Male; F, Female. Family-group taxon (no. spp., no. spp. examined) Genus Total no. spp. No. spp. examined Species examined (sex[s] examined) Family Hemerobiidae (ca. 550, ca. 185) Subfamily Carobiinae (9, 2*) Carobius 9 2* 4-pulcheIlus (MF), unknown sp. [1] (F). Subfamily Hemerobiinae (ca. 200, 56) Hemerobius ca. 125 37 adelgivorus (M), alpestris (M), Taper (M), + australis (M), chilensis (M), conjunclus (M), 4 costalis (MI7), domingensis (M), dorsatus (M), Tfalciger (MF), Jlaveolus (M), hageni (MF), 4- harmandinus (MF), 4-humulinus (MF), jamaciensis (M), 4-japonicus (MF), koka- neeanus (M), 4-lutescens (MF), 4-marginatus (MF), micans (M), montanus (M), 4-nairobi- cus (MF), neadelphus (M), 4-nigrans (M), nigrocornis (M), nitidulus (M), ovalis (M), 4-pacif- icus (M), 4-pini (MF7), pinidumus (M), reconditus (M), simulans (M), -l-stigma (M), 4-subacutus (MF7), tolimensis (M), 4-tristriatus (MF7), 4- withycombei (MF7). Nesobiella 1 1 4-hospes (MF7). F Vesmaelius ca. 70 17* 4-coloradensis (Ml7), 4-concinnus (MF7), fassnidgei (F7), Tfurcatus (M), involutus (MF), lon- gifrons (MF), longipennis (MY7), fassnidgei (F7), 4-navasi (MF), -mervosus (MF7). 4-nubilus (MF7), pretiosus (MF7), quadrifasciatus (MF7), reisseri (F7), Tsubnebulosus (MF), unknown spp. [2] (M). Hemerobiella 1 2* 4-sinuata (M), unknown sp. [1] (F7). Subfamily Sympherobiinae (ca. 60, 17) Nomerobius 4 4 Tcuspidatus (MF), 4-psychodoides (MF7), signatus (MF7), spinosus (MF7). Neosympherobius 1 1 4-cinereus (MF7). Sympherobius ca. 55 12 Tamiculus (MF), angustus (M), barberi (M), 4- bifasciatus (M), 4-domesticus (MF), Tfallax (MF7), fuscescens (F7), maculipennis (M), 4 marginatus (M), nigricornis (MF7), pygmaeus (M), tessellatus (MF). Subfamily Psychobiellinae (2, 1) Psychobiella 2 1 4-sordida (M), unknown sp. [1, sordidal ] (F). 290 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Appendix 2. Continued. Family-group taxon (no. spp., no. spp. examined) Genus Total no. spp. No. spp. examined Species examined (sex[s] examined) Subfamily Notiobiellinae (ca. 70, 24) Notiobiella ca. 35 10* costalis (MF), multifurcata (M), rosea (MF), +turneri (MF), valida (MF), viridis (F), un- known spp. [4] (M). Psectra ca. 25 8* Tdiptera (MF ),franzeni (M), iniqua (F), Aoblongus (M), pretiosa (F), Ttillyardi (MF), un- known spp. [2] (M). Anapsectra 1 1 Tmedleri (MF). Zachobiella 8 4* + ?jacobsoni (M), -t- pallida (F), submarginala (F), unknown sp. [1] (F). Subfamily Drepanacrinae (ca. 10, 7) Conchopterella 4* 4* + kuscheli (MF), + maculata (MF), stangei (MF), unknown sp. [1] (MF). Austromegalomus 2 1 Tinsulanus (MF). Drepanacra 4 2 Tbinocula (MF), + khasiana (MF). Subfamily Megalominae (ca. 40, 18) Megalomus ca. 40 18* angidatus (M ), fidelis (MF), hirtus (M), Tmoestus (MF), Tparvulus (MF), + nigratus (MF), pyraloides (F), +setosulus (MI7), speciosus (M), tineoides (F), + tortricoides (M), unknown spp. [7] (M). Subfamily Drepanepteryginae (ca. 40, 16) Neuronema ca. 30 12* Talbostigam (MF), decision (M), nr. gyironganum (M), + kuwayamai (MF), kwanshiensis (M), +laminatum (MF), + omeishanum (MF), shensiensis (M), zhamanumn (M), unknown spp. [3] (M). Gayomyia 3 — 1 -t-faleata (MF). Drepanepteryx 6 3 algida (F ), falculoides (MF), Tphalaenoides (MF). Subfamily Microminae (ca. 120, 45) Noius 3* 3* Tnoumeanus (M), -foeeanicus (MF), unknown sp. [1] (M). Nusalala ca. 20 9* + dispar (M), -t-krugeri (M), +rhegmatica (M), unknown spp. [6] (M). 1993 WORLD GENERA OF HEMEROBIIDAE 291 Appendix 2. Continued. Family-group taxon (no. spp,, no. spp. examined) Genus Total no. spp. No. spp. examined Species examined (sex[s] examined) Megalomina 3 3 + berothoides (MF), + acuminata (F), + bridwelli (F). Micromus ca. 95 30* + africanus (MF), +angulatus (MF), bifasciatus (F), brandti (M), dissimilis (MF), drepa- noides (F), + gradatus (MF), +lampus (MF), 4-maculatipes (MF), + marquesanus (MF), montanus (M), +multipunctatus (MF), -t-numerosus (M), Foblongus (MF), + paganus (MF), +posticus (MF), rubrinervis (F), +sjostedti (MF), +subanticus (MF), +tasmaniae (MF), +timidus (MF), + vagus (MF), +variegatus (MF), +variolosus (MF), unknown spp. [6] (M). Unplaced genera Notherobius 3 0 292 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Appendix 3. Character fits to the preferred cladogram (aggregated by character). This ap- pendix summarizes the “fit” of each character used in the cladistic analysis to the preferred cladogram (Fig. 23). Three “fit” characteristics are given: (1) character length, (2) all equally most parsimonious optimizations [if more than 1 exists, the optimization plotted on the pre- ferred cladogram is given in boldface type], and (3) the cladogram placement of each character state transformation. Terminal lineages are identified by generic names; nonterminal lineages are identifed by the reference numbers given on the preferred cladogram. Char. no. Length Opt. no. Lineage (state transformation) i 4 i 2(0— 1), 18(1 —0), Conchopterella (1—0), Noius (0—1). 2‘ 2(0-1), 19(1-0), 22(1-0), Conchopterella (1-0). 2 2 1 21(0—1), Xlegalomina (1—0). 3 2 1 21(0-1), Megalomina (1-0). 4 1 1 Noius (0- 1). 5 1 1 Megalomus (0—1). 6 1 1 2(0-1). 7 3 1 3(0-1), 12(1-0), 17(1-0). 8 3 1 3(0-1), 12(1-0). 17(1-0). 9 2 1 2 1 (0 — 1 ), Megalomina (1—0). 10 3 l2 3(1-0), Psychopsis (1-0), Polystoechotes (1-2). 2 Nothochrysa (0-1). Polystoechotes (0—2), Carobius (0-1). 3 1(0-1), Polystoechotes (1-2), Carobius (0-1). 4 1(0-2), Polystoechotes (2—1), Carobius (0-1). 1 1 3 1 4(0- 1), 12(0- 1), Noius (0-1). 12 3 1 7(0-1), 12(0-2 [+1]), Psychobiella (0-2 [+1]). 13 1 1 2(0-1). 14 1 1 Hemerobius (0—1). 15 1 1 Notiobiella (0—1). 16 1 1 2(0-1). 17 2 1 2(0— 1), Psychobiella (1—0). 18 3 1 1(0-1), 20(0- 1), Noius (0-1). 19 1 1 Notiobiella (0-1). 20 1 1 Gayomyia (0 - 1 ). 21 1 1 16(0-1). 22 1 1 Neosympherobius (0-1). 23 4 1 21(0-1), Nothochrysa (0-2 [ + 2]), Zachobiella (0—1). 24 5 1 13(0—1), 21(0—1), Nothochrysa (0-2 [+2]). Megalomina (1-0). 25 3 1 21(0—1), Nothochrysa (0—1), Zachobiella (0—1). 26 2 1 Gayomyia (1-2), Drepanepieryx (1-0). 27 1 1 Notiobiella (0—1). 28 2 1 9(0-1), 22(1-0). 29 1 1 3(0-1). 30 1 1 Carobius (0—1). 31 2 1 9(0-1), 11(1-0). 23 14(0-1). Psychobiella (0-1). 32 1 1 4(0-1). 33 1 1 14(0-1). 34 5 I4 4(0 — 1), 21(0-1), Carobius (0 — 1), Notiobiella (0-1), Gayomyia (0-1). 2 2(0- 1), 7 + 9(1 -0), 21(0-1), Notiobiella (0-1), Gayomyia (0-1). 35 3 Is 2(0-1), Neuronema (1-0), Gayomyia (1—0). 2 2(0—1), 19(1—0), Drepanepteryx (0— 1). 1993 WORLD GENERA OF HEMEROBI1DAE 293 Appendix 3 Continued. Char. no. Length Opt. no. Lineage (state transformation) 36 2 i 1 2(0 — 1 ), Neosympherobius (0 - 1 ). 37 2 i Drepanacra (0 — 1 ), Gayomyia (0—1). 38 2 l Psychopsis (0-1), Hemerobius (0-1). 39 2 i Nothochrysa (0-1), Carobius (0-1). 40 3 i 7(0- 1), 1 1(0 — 1), Nothochrysa (0—1). 41 2 i 1 9(0 - 1 ), Nothochrysa (0 — 1 ). 42 3 i 1 1(0- 1), Nothochrysa (0-1), Sympherobius (0—1). 43 1 i 22(0-1). 44 4 l 6(0— 1), Nothochrysa (0-1), Carobius (0—1), Nusalala (0—1). 45 3 i 1 1(0- 1), Nothochrysa (0—1), Sympherobius (0-1). 46 2 i Nothochrysa (0-1), Gayomyia (0-1). 47 2 i6 19(0-1), 22(0-1). 2 18(0-1), Noius (1—0). 48 4 1 10(0-1), 12(1-0), 22(1-0), Polystoechotes (0—1). 27 14(0-1), 22(1—0), Polystoechotes (0-1), Notiobiella (0-1). 49 3 1 1 5(0- 1), 20(0- 1), Psychopsis (0-1). 50 2 1 22(0—1), Gayomyia (0-1). 51 2 1 1 6(0 — 1 ), Psychobiella (0 - 1 ). 52 1 1 Nomerobius (0-1). 53 2 1 7(1-0), 12(1-2). 54 1 1 Hemerobiella (0—1). 55 2 1 22(0- 1), Nothochrysa (0—1). 56 2 1 Nothochrysa (1-2), Zachobiella (1-0). 57 1 1 Austromegalomus (0- 1 ). 58 1 1 Nesobiella (0—1). 59 2 1 Polystoechotes (0-1), Nomerobius (0-1). 60 1 1 Noius (0-1). 61 1 1 Wesmaelius (0 - 1 ). 62 1 1 Noius (0-1). 63 1 1 Nomerobius (0- 1 ). 64 1 1 Conchopterella (0-1). 65 1 1 Nesobiella (0-1). 66 1 1 Wesmaelius (0—1). 67 1 1 Psectra (0—1). 68 1 1 Nomerobius (0-1). 69 1 1 Noius (0-1). 70 1 1 Neosympherobius (0-1). 71 1 1 Gayomyia (0 - 1 ). 72 1 1 Neosympherobius (0 - 1 ). 73 1 1 Nomerobius (0—1). 74 4 1 10(0-1), 20(1-0), 23(1-0), Zachobiella ( 1-0). 75 1 1 Conchopterella (0—1). 76 1 1 Nomerobius (0-1). 77 1 1 Psychobiella (0-21). 78 7 1 8(1-0), 10(1-2), 18(2- 1), Psychopsis (1 -2), Anapsectra (2- 1), Megalomus (2 — 3), Megalomina (1 -2). 79 1 1 7(0-1) 80 1 1 Sympherobius (0-1). 81 1 1 Sympherobius (0-1). 294 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Appendix 3. Continued. Char. no. Length Opt. no. Lineage (state transformation) 82 3 1 Psychobiella (0—1), Conchopterel/a (0-1), Drepanacra (0-1). 21 2 3 4 5 6 7 8 15(0-1), Psychobiella (0-1), Austromegalomus (1-0). 83 1 1 Notiobiella (0—1). 84 1 1 2(0-1). 85 1 1 5(0-1). 86 2 1 Hemerobius (1-0), Zachobiella (1-2). 87 2 1 Psychobiella (0—1), Drepanacra (0—1). 88 1 1 Psectra (0— 1). 89 1 1 Conchopterella (0 - 1 ). 90 1 1 Gayomyia (0 - 1 ). 91 1 1 Gayomyia (0 - 1 ). 92 1 1 Nusalala ( 0-1). 93 1 1 Drepa nepteryx (0 — 1 ) . 94 1 1 4(0-1). 95 1 1 3(0-1). 96 2 1 Psychopsis ( 1 - 0), A napsectra (1-2). 97 i 1 A napsectra (0 - 1 ). 98 i 1 Hemerobiella (0—1). 99 i 1 Zachobiella ( 0—1). 100 5 1 4(0—1), 12(0— 1), 21 (0— 1), Nolhochrysa (0 — 1 ), Drepanepteryx (0-1). 101 1 1 A napsectra (0- 1). 102 1 1 Anapsectra (0— 1). 103 5 1 5(0 - 1 ), 1 2(0 - 1 ), Carobius (0 - 1 ), Nomerobius (0 - 1 ), Noius (0-1). 104 1 1 2(0-1) 105 2 T o rT t O in 106 3 1 Polystoechotes (0-3 [+1]), Gayomyia (0-1), Nusalala (0-2) [+1]). 107 1 196 1 23(0-1). 1 Optimization 2 is preferred. Optimization 1 would require a rereversal transformation series 0— 1 — 0 — 1 between lineages 2 and Noius. 2 Optimization 1 is arbitrarily plotted on the preferred cladogram. The three states of character 10 were analyzed as nonadditive. Since the two states (0 and 1), found basally within the Hemerobiidae are also found in the outgroups, the polarity of this character is ambiguous at the base of the cladogram. 3 Parallel increase in the number of “radial sectors” along the anterior radial trace (optimi- zation 2) is preferred over reversal in this character (optimization 1). 4 Optimization 1 is preferred, see discussion in text under the heading Phylogenetic Analysis, Results, Cladograms. 5 The occurrence of forewing pre-3irl crossveins in Gayomyia and Neuronema is interpreted as convergent (optimization 1). In Gayomyia these crossveins are associated with a general increase in the number of forewing crossveins, in Neuronema they are not. 6 The occurrence of a lcua-cup crossvein in the forewings of most Drepanepteryginae, and in the clade Nusalala + Megalomina + Micromus is interpreted as a convergence (optimiza- tion 1). 7 Optimization 2, which maximizes parallelism in this character, is preferred. 8 Optimization 2, which interprets the well-developed submediuncal scabriculous patches in Conchopterel/a and Drepanacra as homologs, is preferred. 1993 WORLD GENERA OF HEMEROBIIDAE 295 Appendix 4. Character fits to the preferred cladogram (aggregated by lineage). This appendix summarizes the “fit” of each character used in the cladistic analysis to the preferred cladogram (Fig. 23). This appendix is similar to Appendix 3, except that character transformations are aggregated by lineage, rather than by character. For characters 1, 10, 31, 34, 35, 47, 48, and 82, the character transformations specified by the preferred optimizations identified in Appendix 3 are used. Terminal lineages are identified by generic names; nonterminal lineages are identified by the reference numbers given on the preferred cladogram. Lineage Length Character transformations Outgroups i l 18(0-1). Psychopsis 5 10(1-0), 38(0-1), 49(0-1), 78(1-2), 96(1-0). Nothochrysa 15 23(0-2 [ + 2]), 24(0-2 [ + 2]), 25(0-1), 39(0-1), 40(0-1), 41(0- 1), 42(0- 1), 44(0- 1), 45(0- 1), 46(0-1), 55(0-1), 56(1-2), 100(0-1). Polystoechotes 4 10(1-2), 48(0-1), 59(0-1), 106(0-3 [+1]). Ingroups 2 8 1(0-1), 6(0-1), 13(0-1), 16(0-1), 17(0-1), 35(0-1), 84(0-1), 104(0-1). 3 5 7(0-1), 8(0-1), 10(1-0), 29(0-1), 95(0-1). 4 5 11(0-1), 32(0-1), 34(0-1), 94(0-1), 100(0-1). 5 3 85(0-1), 103(0-1), 105(0-1). 6 1 44(0-1). 7 4 12(0-1), 40(0-1), 53(1-0), 79(0-1). 8 1 78(1-0). 9 1 28(0-1). 10 2 74(0-1), 78(1-2). 11 3 40(0-1), 42(0-1), 45(0-1). 12 9 7(1-0), 8(1-0), 1 1(0-1), 12(0-2 [+1]), 36(0-1), 53(1-2), 100(0-1), 103(0-1), 105(0-1). 13 1 24(0-1). 14 3 31(0-1), 33(0-1), 48(0-1). 15 2 49(0-1), 82(0-1). 16 2 21(0-1), 51(0-1). 17 2 7(1-0), 8(1-0). 18 1 78(2-1). 19 3 1(1-0), 41(0-1), 47(0-1). 20 3 18(0-1), 49(0-1), 74(1-0). 21 8 2(0-1), 3(0-1), 9(0-1), 23(0-1), 24(0-1), 25(0-1), 34(0-1), 100(0-1). 22 7 1(1-0), 28(1-0), 43(0-1), 47(0-1), 48(1-0), 50(0-1), 55(0-1). 23 2 74(1-0), 107(0-1). Carobius 5 30(0-1), 34(0-1), 39(0-1), 44(0-2), 103(0-1). Hemerobius 3 14(0-1), 38(0-1), 86(1-0). Nesobiella 2 58(0-1), 65(0-1). Wesmaelius 2 61(0-1), 66(0-1). Hemerobiella 2 54(0-1), 98(0-1). Nomerobius 7 52(0-1), 59(0-1), 63(0-1), 68(0-1), 73(0-1), 76(0-1), 103(0-1). Neosympherobius 4 22(0-1), 36(0-1), 70(0-1), 72(0-1). 296 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Appendix 4. Continued. Lineage Length Character transformations Svmpherobius 4 42(0-1), 45(0-1), 80(0-1), 81(0-1). Psvchobiella 7 12(0-2 [+1]), 17(1-0), 31(0-1), 51(0-1), 77(0-1), 82(0-1), 87(0-1). Notiobiella 6 15(0-1), 19(0-1), 27(0-1), 34(0-1), 48(0-1), 83(0-1). Psectra 2 67(0-1), 88(0-1). Anapsectra 5 78(2-1), 96(1-2), 97(0-1), 101(0-1), 102(0-1). Zachobiella 6 23(0-1), 25(0-1), 56(1-0), 74(1-0), 86(1-2). 99(0-1). Conchopterella 4 1(1-0), 64(0-1), 75(0-1), 89(0-1). A ustromegalomus 2 57(0-1), 82(1-0). Drepanacra 2 37(0-1), 87(0-1). Megalomus 2 5(0-1), 78(2-3) Neuronema 1 35(1-0). Gayomyia 11 20(0- 1), 26(1-2), 34(0- 1), 35(1-0), 37(0-1), 46(0- 1), 50(0-1), 71(0-1), 90(0-1), 91(0-1). 106(0-1). Drepanepteryx 3 26(1-0), 93(0-1), 100(0-1). Noius 7 4(0-1), 11(0-1), 18(0-1), 60(0-1), 62(0-1), 69(0-1), 103(0-1). Nusalala 3 44(0-1), 92(0-1), 106(0-2 [+1]). Afegalomina Micromus 5 0 196 2(1-0), 3(1-0), 9(1-0), 24(1-0), 78(1-2). 298 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 101(2) Appendix 5. Cladistically coded data. This appendix presents the data matrix of 27 taxa (24 ingroup. 3 outgroup) and 107 characters used in the cladistic analysis. Bold values in the matrix indicate the presence of intrageneric variation for the possible characer states at these positions (see discussion in text under the heading Phylogenetic Analysis: Methods and Data: Data). Question marks (?) indicate unobtainable or uncertain character states. Character Number 00000000000000000000000000000000000000000 0000000001 11111111 12222222222333333333344 Taxon 12345678901234567890123456789012345678901 Anapsectra Austromegalomus Carobius Conchopterella Drepanacra Drepanepteryx Gayomyia Hemerobiella Hemerobius Megalomina Megalomus Micromus Neosympherobi us Nesobiella Neuronema Noius Nomerobius Notiobiella Nusalala Psectra Psychobiella Sympherobius Wesmaelius Zachobiella 1000010000121001 100000010101 1000001 100010 100001 1 100001001 100010000101 1010101000000 1 0000 100010010011 00000000 1 000 100011 000 1 00 000001 1 100001001 100000000101 1010101000000 100001 1 100001001 100010000101 1010101010000 00000 1 000000 100111 000000000 11010101 00000 1 00000 1 000000 10011101 0000020? 1010110010001 100001 1 100101001 100000000100100101 1000000 100001 1 100101 101 100000000100100101 1001000 00000 1 000000 10011 00000 1011001010111 000000 100011 000000 10011 00000000 1011010101 000000 0110010010001001 10000011110010101 1 1000000 1 0000 111000110011 0000 1 000 1 00 1 00000 1100010 100001 1 100101001 100000000100100101 1000000 00000 1 000000 10011 00000000 10110101 0000000 1 1 1 1 1010010101001 1 10000111 101101011 1000000 1 0000 111000110011 00000000 1 00 1 00000 1 0000 1 0 100001 1 10000101 1 101000000111100001 1000010 01 10010010001001 1000001 1 1 100101011 1000000 1 0000 100001210011 00000000 1011 00000 1100010 1 0000 1 1 1 0002 1 00 1 000000000 1011010001 000000 1 0000 111000110011 00000000 1 00 1 00000 1 0000 1 0 100001 1 100101001 100000000100100101 1000000 1000010000121001 100000111101 1000001 100010 Nothochrysa 000000000 1 0000000 1 000022 1 1 0000000?0?00 1 1 1 Polystoechotes 00000000020000000 1 0000000 1 000000000?00000 Psychopsis 0000000000000000000000000 1 0?0000000?0 1 000 1993 WORLD GENERA OF HEMEROBIIDAE 299 Appendix 5. Extended. Character Number 0000000000000000000000000000000000000000000000000000000000 11111111 444444445555555555666666666677777777778888888888999999999900000000 234567890123456789012345678901234567890123456789012345678901234567 10010000000200100000000000000000100010000010100000000121001 1111100 000000 1101010011 0000000000000000 1 000200000 1 0 1 00000000 1 1 0000000 1 000 00 1 00000000 1 00 1 000000000000000000000 1 00000 1 0 1 000000000 1 000000 1 1 000 00000011000 1 00 1 0000000 1 000000000 1 1 002000 10101001 00000 1 1 0000000 1 000 000000 110101001 00000000000000000 1 0002000 101011 0000000 1 1 0000000 1 000 00000 1110001001 000000000000000000000 1 00000 1 0 1 000000 101100010001 000 0000 17111001001 00000000000000 1 000000 1 00000 1010001100011 0000000 1010 00 1 00000000 1101 000000000000000000000 1 00000 1 0 1 0000000 11101010001000 00000000000 1 00 1 000000000000000000000 1 00000 1 1 00000000 1 1 1 000 1 00 111 00 010001001001011 000000000000000000000200000 1 0 1 00000000 110001 000 1 00 1 000000 1 0000 1 00 1 00000000000000000 1 000300000 1 0 1 00000000 1 1 0000000 1 000 010001001001011 000000000000000000000 1 00000 1 0 1 00000000 1100010001001 00000000000000 1 0000000000000 1 0 1 000000 1 0000 1 0 1 00000000 1 1 0000000 1 000 0000000000010010100000010000000000001000001 1 100000001 1100010011100 000001 1 0000 1 00 1 00000000000000000 1 000 1 00000 1 0 1 00000000 1 1 0000000 1 000 000000 1 0000 1001000101 000000 1 0000 1 000 1 00000 1 0 1 00000000 1100010011 000 0000000000 1 000 1 00 1 000 1 0000 1 0000 1001011 0000 1 0 1 00000000 1 1 000000 1 1 000 1001001 0000 1 00 1 00000000000000000 1 00020000 1101 00000000 1 1 0000000 1 000 011001001001011 00000000000000000 1 000 1 00000 1 0 1 00000 100110001 000 1 020 1 00 1 0000000200 1 0000000000 1 000000 1 000200000 10101 000000 1100010011100 000000000 101001 00000000000000000000 11000101011 0000000 1 1 0000000 1 000 1 00 1 0000000000 1 0000000000000000000000 11100101 00000000 1 1 0000000 1 000 00 1 00000000 1 00 1 0000 1 0000 1 00000000000 1 00000 1 0 1 00000001 110001 000 1 000 100100000002000000000000000000000000200000102000000001 100110011100 10111 700000 1 0 1 2000000000000000000000 1 0000000 1 000000000 1 000 1 0000000 000000 1 0000 1001001 000000000000000000 1 0000000 1 000000000 1 00000000030 0000000 1 000 1 00 1 00000000000000000000020000000 1 000000000000000000000 INSTRUCTIONS TO AUTHORS The Journal of the New York Entomological Society publishes original research resulting from the study of insects and related taxa. Research that contributes information on taxonomy, classification, phylogeny, biogeography, behavior, natural history, or related fields will be con- sidered for publication. The costs of publishing the Journal are paid by subscriptions, mem- bership dues, page charges, and the proceeds from an endowment established with bequests from the late C. P. Alexander and Patricia Vaurie. Manuscripts should be submitted in triplicate to: Dr. James M. Carpenter, Editor, Journal of the New York Entomological Society, Department of Entomology, American Museum of Natural Elistory, Central Park West at 79th Street, New York, New York 10024. Contributions must be written in English, double-spaced (including references, tables, captions, etc.) and prepared in the format of a recent issue of the Journal. Manuscripts of any length will be considered for publication. Longer papers will be printed as articles, shorter ones as “scientific notes.” Book reviews will be solicited by the Book Review Editor. Longer manuscripts intended for submission as articles should be accompanied by a brief abstract. Footnotes should be avoided. Tables should be prepared as separate pages; they should be kept to a minimum because of the high cost of typesetting, but may be submitted as photographically reproducible material (see below). The list of references is headed “Literature Cited" and should follow the format indicated in the CBE Style Manual or as found in a recent issue of the Journal. Manuscripts should be printed in roman, bold roman, or underlined roman or bold roman to conform to journal typesetting style. Manuscripts should not contain any italic fonts. Manuscripts containing cladistic analyses should be accompanied by a copy of the data matrix on diskette for verification of the calculations. Illustrations (originals whenever possible) should be submitted flat, never rolled or folded, with the manuscript. Size of illustrations should be kept manageable and suitable for shipment through the US mail without being damaged. Proportions of illustrations should conform to single column format, actual page size 1 1.5 x 17.5 cm. Please assemble illustrations in a com- pact fashion in order to use journal space effectively. Mount line drawings and halftones on separate plates. Figures should be numbered consecutively. Figure captions should be double- spaced, grouped together and placed at the end of the manuscript. If tables are submitted for photographic reproduction, they should be neat and clean and conform to journal format proportions. Authors will receive page proofs, with the original manuscripts and illustrations, directly from the printer. Corrected proofs should be returned promptly to the Editor to avoid publi- cation delays. Revisions in proof should be kept to the absolute minimum. Alterations in proof will be charged to the author at the rate of $ 3.25 per revision line. Society members will be charged a fee of $23.00 per printed page and $8.50 per plate of figures. Non-members will be charged $65.00 per printed page and $ 1 5.00 per plate of figures. Member authors who are students, or who do not have institutional affiliation may petition to the Society for waiver of page charges for no more than eight pages on a once a year basis. Because of limited funds, all such requests will be handled on a first-come first-serve basis. Authors will receive a reprint order blank with the proofs. Reprints are ordered directly from the printer with no benefit accruing to the Society. Journal of the New York Journal Entomological Society _T . . . „ AM. MUS. NAT. HIST. LIBRARY New York EntOHH Received on: 06-10-93 . 59.57.06(74.7) VOLUME 101 APRIL 1 CONTENTS Revision and cladistic analysis of the world genera of the family Henterobiidae (Insecta: Neuroptera) John D. Oswald 143-299 00234041