THE MICRO-LEPIDOPTERAN GENUS ECTOEDEMIA BUSCK (NEPTICULIDAE) IN NORTH AMERICA by CHRISTOPHER WILKINSON Department of Animal Systematics and Zoogeography, Vrije Universiteit, Amsterdam and P. J. NEWTON 20 Mellor Brow, Mellor, Blackburn, England With 71 text-figures Abstract This paper, together with the monograph by Wilkinson & Scoble (1979) forms a comprehensive taxo- nomie account of the genus Ectoedemia Busck in North America. Species are here fully diagnosed and illustrated with drawings of the external features and genitalia. Notes on their biology are also included. Contents Abstract P- 27 Introduction p. 27 Methods P- 28 Abbreviations P- 28 Acknowledgements P- 29 Taxonomie history P- 29 Taxonomie considerations P- 31 Generic description P- 32 Generic differential diagnosis p. 34 Check-list to genus, species and subspecies P- 36 Key to species P- 37 Table for the identification of Ectoedemia species from host plants and damage P-40 The populella group P- 41 The platanella group P- 51 The rubifoliella group P- 61 The castaneae group p. 72 References P- 86 An index to the names of taxa dealt with in the present paper can be found at the end of the fol- lowing paper. Introduction In this revision twenty-two species are considered. Five are new combinations, two are new species and nine are diagnosed and figured which have previously 27 28 Tijdschrift voor Entomologie, deel 124, afl. 2, 1981 been described by Wilkinson & Scoble (1979). Those species attacking bark are described at the end and form a clear group separable on the basis of genitalia from the leaf and petiole miners. In the same way other groups amongst the latter are recognisable. However, those attacking petioles, e.g. E. populella and canutus, are included in one of the groups of leaf-miners. This action is supported by the presence of an intermediate habit. In argyropeza the larva first bores into the pet- iole and later continues to mine the leaf. Thus in Ectoedemia it is now definitely established that three sites of attack exist: leaf, petiole and bark. There is also a reference to attack in the cortex. Both galls and mines are found in these locations. There has been much discussion in the past regarding the boundaries of the genus. This revision, with the advantage of additional material, sheds light on this debate. Methods The methods used are similar to those given in Wilkinson & Scoble (1979) as are also the terms used, although in the five years that have elapsed since the former work was completed we have modified some of our terminology. We have come to realise that a thorough comparative morphological study of the various nepticulid genera is long overdue, not only to ascertain homologies within the group but also with other families. At present one is not always certain that parts of genitalia given the same names as those in the Ditrysia are indeed homologous. For ex- ample the interpretation of the term "saccus" by Beirne (1945) is unclear and has not been adopted by other workers to mean a part of the integument as he seems to interpret it. Here "saccus" refers to the anterior extension of the vinculum in front of the ventral plate and is the area often extended into a bilobed protrusion. Whether this is homologous with the saccus of Ditrysia is uncertain and therefore the continued use of the term is open to question. "Pseuduncus" is a posterior extension of the tegumen and a true uncus is not present in Ectoedemia. However, some species have a thickened area in the region where an uncus might be expected to arise. Scoble (verbal communication) sug- gests that at least in the African species he has examined, the thickening rep- resents a forward fold of the pseuduncus and therefore the uncus is still absent. Expendable material of trinotata would be particularly useful for the investigation of this point. " Anellar lobes" refers to the large spines located at the end of the aedeagus. It is not always certain if they are homologous. It is assumed they provide the normal supporting role for the aedeagus. Sometimes, however, they might be more re- ferable to the juxta or aedeagus proper. Scale. — On the genitalia figures the scale line represents one tenth millimeter (1/10 mm) unless otherwise indicated. The magnification of the external figures can be ascertained from the al. ex. measurements (taken from centre of meso- thorax to wing tip and doubled) given for each species. Abbreviations Institutes from which material was borrowed: Wilkinson & Newton: Ectoedemia in North America 29 USNM — United States National Museum of Natural History, Smithsonian In- stitution, U.S.A. CNC — Canadian National Collection, Biosystematics Research Institute, Ottawa, Canada FIS — Forest Insect Survey, British Columbia, Canada SOO — Forestry Service, Environment Canada, Sault Ste. Marie, Canada ANS — Academy of Natural Sciences, Philadelphia, U.S.A. BM(NH) — British Museum (Natural History), London, U.K. MCZ — Museum of Comparative Zoology, Cambridge, Massachusetts, U.S.A. DFF — Department of Fisheries and Forestry, Forestry Service, (now Envi- ronment Canada), Ste. Foy, Quebec, Canada. Acknowledgements Our thanks are due to Mej. R. Kloos and B. J. van Cronenburg who assisted with the final production of the drawings and to Dr. Georgina Bryan who read the final manuscript and made helpful suggestions. P. J. N. wishes to acknowledge the support of Portsmouth Polytechnic during his tenure as research assistant to C. W. Taxonomic history The question of the significance of larval feeding habit and the pattern of forewing venation has led to some confusion in the past and is now the focus of discussions on the composition of this genus and its phylogeny with respect to the other members of the family. The genus was originally described by Busck (1907) as monotypic, because the species populella which, despite affinities with Stigmella Schrank could not be included within its confines. The genus and author were cited incorrectly by Busck as Nepticula Zeiler. His action was right but his reason was wrong. It was that the gall-forming habit of populella excluded it from the leaf- mining species of Stigmella. Busck also gives major morphological differences be- tween the types of Stigmella and Ectoedemia, including the closed cell of the forewing venation formed by the medial cross-vein, which is actually also present in other genera. For some time Braun had been aware of the two basic forms of forewing venation, those with a medial cross-vein and those without. This feature, however, was not given taxonomic status by her for a number of reasons. Firstly, she regarded otherwise closely related species and sometimes male and female in- dividuals of the same species as showing both conditions of the pattern of venation (Braun, 1917: 157, and figs. 1 and 2). This conclusion was the result of misidentifi- cation of a single species (nyssaefoliella), which she thought had both forms of ve- nation and, more generally, is due to the absence of any study of the genital morphology at that time. Again Braun and other of her contemporary workers re- garded the mining habit of the larvae as the most significant indication of phy- logeny. It was not until later examination of the genital morphology that evidence in favour of uniting leaf-mining species and gall-forming species within a single genus, became available. 30 Tijdschrift voor Entomologie, deel 124, afl. 2, 1981 Beirne (1945) divided the family into two basic groups on the form of the male genitalia, those with a ring-shaped vinculum and those with an incomplete vin- culum. This led him to erect the genus Dechtiria for those leaf-mining species with distinct differences in genitalia from the leaf-mining Stigmella (and Nepticula sensu Beirne) species. A study of the group by Svensson (1966: 200) with a comparison of genital morphology and venation of North American types, led him to synonymise the European genus Dechtiria with the North American Ectoedemia Busck. This decision has been accepted by Wilkinson & Scoble (1979), who included seven leaf-mining species, which were earlier assigned to Stigmella, in Ectoedemia. In the present study a further five species are combined with Ectoedemia, all of which are leaf-miners. Beirne (1945: 204) noted that in the genus Dechtiria, with the ex- ception of two species only, all species formed leaf-mines which terminated in blotches. In the present revision only one included species, virgulae, forms a linear leaf-mine. The remainder form either true blotches or linear mines terminating in blotches. Immediately after the publication by Zimmermann (1940) of work on the bark- miner, Ectoedemia liebwerdella Zimmermann, Hering (1940) erected the genus Zimmermannia, with this species as the type. Hering proposed this division on the basis of a difference in larval feeding habit, comparing the North American type populella Busck, which forms petiole galls in poplar, and the European species lieb- werdella, which burrows in the bark of beechwood. Hering noted that the two species were indistinguishable in the pattern of wing venation but that they dif- fered in that liebwerdella possesses a row of sharp bristles on the inner side of the hind-tibiae. Borkowski's study (1972: 693) points out that this character is present in all the European Ectoedemia species he examined, whether gall forming or bark burrowing. This view is also held here for the American species but there is some variation in the prominence of the bristles. Schönherr (1957: 127; 1958: 6) subsequently synonymises Zimmermannia with Ectoedemia and gives it subgeneric rank. This is confirmed by Borkowski (1972), who points out that the morphological differences given by Hering are not suffi- cient to warrant generic status being attached to Zimmermannia. He also presents a summary of the treatment of the three genera at the time and supports the di- vision of Ectoedemia sensu lato into three subgenera corresponding to larval feeding habit: the petiole-miners, Ectoedemia Busck; the leaf-miners, Dechtiria Beirne; and the bark-miners, Zimmermannia Hering. The group is treated similarly by Emmet (1976) in his revision of the British fauna. Although Borkowski is in some doubt that sufficient work was done by Svensson in his synonymy of Dechtiria with Ectoedemia, it is felt that the work by Wilkinson & Scoble (1979) on the Canadian fauna and the present study confirms the syn- onymy. On the basis of genitalia and wing venation the authors also support the synonymy of Zimmermannia with Ectoedemia. However, from this study it is pos- sible to identify morphological divisions, mainly based on genitalia structures, within the genus, which correspond with the sites of larval attack. Johansson (1971: 241) is of the opinion that the family Nepticulidae may be split into two genera, Nepticula and Trifurcula. He argues that Trifurcula has been subdi- vided into several genera including Ectoedemia and Dechtiria, but that these should Wilkinson & Newton: Ectoedemia in North America 31 be included as subgenera of Trifurcula on the basis of male genitalia. This scheme is followed more recently by Karsholt & Schmidt Nielsen (1976) in the catalogue of Danish Lepidoptera. Although there is merit in the proposal, this classification is not adopted in this revision for reasons given by Wilkinson & Scoble (1979). It was pointed out earlier that there are no significant structural differences between the bark-miners and bark gall-formers. For example, heinrichi and castaneae cannot be reliably separated without knowledge of the host plant and damage. However, examination of the genitalia figures shows that species which attack bark (whether galls or mines) have characteristic genitalia and leads one to guess at the habits of those whose life history is unknown. The three species recorded here as attacking petioles can be placed in the same morphological group on the basis of genital structures, but the group also contains leaf-miners. Taxonomic considerations The similarity in morphological structures, even genitalia, between castaneae and heinrichi, highlights problems over the species concept in this group of Lepi- doptera. Knowledge of the life history leads us to separate two species which otherwise might have been regarded as one. Speciation is much more extensive than has hitherto been recognised and many species are only known through the excellent rearing work of Braun and others. In this genus alone, many species are shown to have more than one form in the externals, particularly wing markings, which is not matched by differences in the genitalia. At first sight a species appears to be di- or polymorphic, but without knowledge of the life cycles, we cannot be sure that these variants are not distinct species. For example, three wing pattern forms occur in what has been called E. similella. Although they have identical gen- italia, one form occurs in Ohio, New Jersey and Virginia, another form in Florida. A single specimen with different markings has been collected from Arkansas. Host plants are only known for the Ohio form {Quercus palustris and Q. rubra). A specimen (nr. 7) probably represents a new species close to rubifoliella. It differs in ground colour and certain aspects of the genitalia. It was reared by Braun from a collection of similar mines on Quercus platanoides but the rest turned out to be Stigmella flavipedella (Braun) comb. n. This is one of a number of ex- amples which shows that each larva should be reared separately, as far as practi- cable, and also have a separate rearing number. The fact that the host plants of ru- bifoliella and specimen 7 are so different, supports the view that specimen 7 should be treated as a new species but this has not been done in case there is anything spurious about its breeding record. Otherwise its similarity to rubifoliella is quite striking. The male genitalia figures are drawn to accentuate the differences which may not be so easy to diagnose as they at first seem. Of course, the problem must also be approached from the other direction. It is quite possible that a species could attack more than one part of a plant. It is known that E. argyropeza downesi mines both the petiole and the lamina. It is also possible that the same species may create mines or galls depending upon the site of oviposition. Although many species are monophagous not all are, and we should therefore be prepared for dimorphism or different reactions from different host 32 Tijdschrift voor Entomologie, deel 124, afl. 2, 1981 plants to the same species. Some authorities think this unlikely but there is consid- erable danger in the philosophy — two species of host plant, therefore ipso facto two species of miner! However, hopefully these problems will encourage experi- mental work designed to discover the flexibility in miner and host plant relationships within the Nepticulidae. Meanwhile these cases underline the point that no one character, or series of characters, can be taken as necessarily species- diagnostic throughout the Nepticulidae or even for Ectoedemia. There has been debate regarding the division of the genus into sub-genera and also whether Ectoedemia comprises, in reality, more than one genus. Much more information on life cycles is needed to answer this, but this paper shows that the criteria previously used for separations are inadequate. The genitalia structures allow us to divide Ectoedemia into broad groups and it seems that those species at- tacking bark (both mining and gall-forming) are particularly diagnostic. However, we do not yet know whether all those species assigned to the group do, in fact, attack bark. In other groups it can be argued that one or two species {trinotata, nyssaefoliella) have males belonging to a different group from the females, thus suggesting that the groups are not fundamental. Our present level of knowledge recommends us to regard these species as respresenting one genus and our studies of Ectoedemia on other continents will shed further light on these conclusions. It is interesting to note how some genera of Nepticulidae are worldwide in their distri- bution whilst others are comparatively localised. Generic description External features: $ 9 . Head: palps extending well beyond labrum, pale grey or white; antennae approximately half length of forewings, pale or fuscous; tuft on front of head usually ochreous, sometimes white or brown, vertex concolourous; eye-caps always whitish; collar white or ochreous, sometimes fuscous. Thorax dark brown to black, sometimes white or buff and irrorate, usually iridescent. Ab- domen grey or brown, iridescing metallic grey beneath. Venation as in figs. 1 or 2. Forewings: Media coalescing with cubitus at base and passing obliquely to radius at or beyond R 2+3 and anastomosing beyond middle of wing; cubitus usually reaching margin; R 4 and R 5 separate; anal vein sometimes meeting cubitus dis- tally. Hindwings: Media single-branched. Forewings: elongate and ovate in shape, ground colour of dorsal surface generally fuscous with each scale darker at the tip, rarely pale and irrorate with fuscous; fringe pale grey or brown and iridescent, usually marked apically with a band of wing-scales or variously irrorate; markings usually in the form of up to four streaks or patches, never with more than one complete fascia. Hindwings: narrow and lanceolate, sometimes with costa emarginated; pale grey or brown and sometimes with white or ochreous patches of specialised scales. Legs: grey or brown, sometimes with scattered paler areas; proximal pair of spurs on hind-tibiae at or below the middle (fig. 3); hind-tibiae with many sharp bristles inwardly directed and variable in prominence. Male genitalia: Vinculum always ring-shaped; tegumen fused with vinculum dor- sally and produced into broad tapering, pointed or bluntly rounded pseuduncus. Uncus often not readily visible, only present as membranous lobe at base of pseu- Wilkinson & Newton: Ectoedemia in North America Sc 33 I Ectoedemia obrutella ó* Sc 2. Ectoedemia similella o Fig. 1. Ectoedemia obrutella (Zeiler) Busck, tf, wing venation. Fig. 2. E. similella (Braun), 9, wing ve- nation. duncus. Gnathos with medial arms fused to form central boss or tapering process, lateral arms rarely extended or bifurcate. Saccus bilobed to various degrees. Valves usually tapering and inwardly curved distally, sometimes weakly bifurcate and in some cases adorned with digitate setae. Transtillae usually U-shaped; trans- verse bars continuous, fused medially. Juxta absent. Aedeagus: generally regular in shape and longer than the capsule; vesica usually adorned with small denticles and cornuti; also with a plate of minute papillae. Anellus with heavily sclerotised spines, sometimes elaborate, rarely absent. Female genitalia: Apophyses usually shorter than ductus but posteriores very long in the castaneae group. Ductus at colliculum with double sclerotised ring and associated denticulate sac, or with single plate, or without sclerotisation. Bursa 34 Tijdschrift voor Entomologie, deel 124, afl. 2, 1981 Fig. 3. Ectoedemia species. Legs showing position of tibial spurs. copulatrix: large and variously pectinate; signum double, comprising ovate patches of reticulate cells. Larvae: may be divided into three groups: leaf-miners usually forming linear tracts terminating in blotches; petiole miners and bark miners both sometimes producing galls. Generic differential diagnosis Characters which differentiate North American genera of the family Nepticu- lidae are given. Ectoedemia Busck, 1907. Venation: Media of forewing coalescing with Cubitus at base, passing obliquely to Radius at or beyond R 2+3 and anastomosing to a point beyond middle of wing; R 4 and R 5 separate; Cubitus usually approaching margin; Media of hindwing single. Proximal pair of spurs on hind tibiae sometimes in the middle. Male geni- talia with gnathos W- or V-shaped, may vary according to method of mounting; vinculum ring-shaped and without associated lateral bars; tegumen extended into tapering or lobed pseuduncus; uncus absent or weakly membranous; valves in- wardly curved distally sometimes with digitate setae; juxta absent; aedeagus regular in shape with elaborate cornuti and usually anellar spines. Female genitalia with or without complex sclerotisation of the ductus and spiculate accessory lobe; apophyses shorter than ductus; signa comprising patches of reticulate cells. Larvae may mine in, or form galls on leaves, petioles, bark or cortex. Wilkinson & Newton: Ectoedemia in North America 35 Stigmella Schrank, 1802. Venation: Media of forewing coalescing with Radius at base and anastomosing to a point beyond the middle of the wing; R 4 coincident with R 5 ; Cubitus arising separately, approaching middle of the wing; Media of hindwing single. Forewings usually uniform and dark in colour, with one or two complete fasciae or patches; fringe with diffuse margin. Proximal pair of spurs on hind tibiae above the middle. Male genitalia usually with U-shaped vinculum; tegumen strap-like, articulating with vinculum dorsally; uncus bilobed; juxta, if present, membranous; aedeagus usually flask-shaped, vesica usually with many denticulate cornuti orientated in a ridge and rarely with plate-like cornuti at the anellus. Female genitalia with simple ductus and accessory sac; bursa copulatrix usually without signum, but if present often single and weakly sclerotised. Larvae mining leaves of trees and shrubs and sometimes herbs. M icrocalyptris Braun, 1925. Venation: Reduced; Media of forewing coalescing with Radius from base and anastomosing to a point beyond the middle of the wing; R 4 coincident with R s ; Cubitus vestigial; Media of hindwing single, unbranched. Ground colour of dorsal surface of forewing usually pale and variously irrorate. Proximal pair of spurs on hind tibiae below middle. Male genitalia, with membranous pseuduncus and strongly sclerotised bridge-like uncus; sclerotised gnathos with complex anterior and posterior projections; lateral arms of vinculum usually with associated sclero- tisations. Female genitalia with complex sclerotisations of the ductus; posterior apophyses very long, longer than the ductus; signa usually comprising linear row of spinose cells or plates. Larvae mining leaves. Fomoria B eirne, 1945. Venation: Media coalescing with Cubitus from base, both passing obliquely to Radius at R 2+3 and anastomosing to beyond middle of wing; Cubitus becoming ob- solete; R 4 and R 5 separate; Media of hindwing single, unbranched. Male genitalia with membranous pseuduncus and uncus as a spatulate sclerotisation; Y- or V- shaped gnathos; saccus weakly bilobed; valves sometimes with dorsal spine; ae- deagus regular in shape and usually with complex anellar spines and cornuti. Female genitalia, colliculum with sclerotised funicular antrum or complex plates; simple ductus; signa comprising linear patches of reticulate cells. Larvae often re- corded pupating within the leaf-mine. Obrussa Braun, 1915. Venation: Media of forewing coalescing with Cubitus at base, both passing obliquely to Radius at R 2+3 and anastomosing to beyond middle of wing; Media and Cubitus separate terminally; R 4 and R 5 separate; Media of hindwing single. Ventral surface of forewing and dorsal surface of hindwing in males with patch of brightly coloured specialised scales. Proximal pair of tibial spurs below middle of hind tibiae. Male genitalia with ring-shaped vinculum; tegumen extended into ta- pering pseuduncus; convex saccus; valves blunted distally and each with large dorsal arm projecting transversely to reach opposite side of capsule; vesica with 36 Tijdschrift voor Entomologie, deel 124, afl. 2, 1981 transverse plate expanded laterally. Female genitalia with plate-like sclerotisation at colliculum; signa comprising ovate reticulate patches. Larvae only known to mine fruits of Acer sp. Glaucolepis Braun, 1917. Venation: Media of forewing coalescing with Cubitus at base, both passing obliquely to Radius at R 2+3 and anastomosing to beyond middle of wing; Media and Cubitus separate terminally; R 4 and R 5 separate; Media of hindwing bifurcate. Hindwing of male with patch of brightly coloured specialised scales. Proximal pair of spurs on hind tibiae in the middle. Male genitalia with tegumen extending into tapering pseuduncus, gnathos with large transverse arms and medial dorso-lateral arms fusing terminally; valves markedly bifurcate distally; aedeagus with lateral cornuti extending full length of vesica and digitate distally. Female genitalia with simple ductus; signa comprising linear patches or rows of pectinations. Larvae mining leaves. Manoneura Davis, 1979 (Replacement for homonymie name Oligoneura Davis, 1978). Venation: greatly reduced; only two branches of Radius present; Media un- branched and arising from stem of R 4+J ; Cubitus absent; hindwing extremely slender and Media unbranched. Forewing dark fuscous with a single, narrow, pale golden yellow fascia at distal third. Proximal pair of spurs on hind tibiae near apex. Male genitalia with uncus vestigial; gnathos well developed, complex, consisting of two, largely separate, transverse sclerites of a highly irregular, but symmetrical outline; vinculum broad, quadrate; aedeagus moderately short and stout, with a relatively complex apex and no cornuti. Female genitalia unknown. Presumably a leaf mining genus. The type-species mines Coccoloba uvifera (L.). Artaversala Davis, 1978. Venation: greatly reduced; Radius unbranched, terminating well short of apex; Media unbranched, extending almost to apex; Cubitus shortened, indistinctly present; hindwing extremely slender and Media unbranched. Forewings with a pale yellow to whitish apex and a single, broad, median fascia. Proximal pair of spurs on hind tibiae near apex. Male genitalia with tegumen reduced to an ex- tremely slender dorsal ring; uncus absent; vinculum well developed and V-shaped valves deeply divided and aedeagus relatively simple, without cornuti. Female gen- italia with slender and elongate ductus; bursa copulatrix membranous; signa absent. Larvae mining leaves. Check-list to genus, species and subspecies Ectoedemia Busck, 1907. Type-species by original designation and monotypy: Ectoedemia populella Busck. Zimmermannia Hering, 1940. Type-species by original designation: Ectoedemia liebwerdella Zim- mermann, 1940: 264. Dechtiria Beirne, 1945. Type-species by original designation: Tinea subbimaculella Haworth, 1828: 583. Wilkinson & Newton: Ectoedemia in North America 37 populella Busck, 1907 argyropeza downesi Wilkinson & Scoble, 1979 1 ) canutus Wilkinson & Scoble, 1979 trinotata (Braun, 1914) marmaropa (Braun, 1925) platanella (Clemens, 1861) maximella (Chambers, 1873) clemensella (Chambers, 1873) similelia (Braun, 1917) virgulae (Braun, 1927) lindquisti (Freeman, 1962) rubifoliella (Clemens, 1860) «/me/fa (Braun, 1912) nyssaefoliella (Chambers, 1880) quadrinotela (Braun, 1917) obrutella (Zeller, 1873) bosquella (Chambers w Hayden, 1878a) acanthella sp.n. piper ella sp.n. heinrichi Busck, 1914a castaneae Busck, 1913 phleophaga Busck, 1914b chlorantis Meyrick, 1928a mesoloba Davis, 1978 Key to the species of North American Ectoedemia 1. Male genitalia with aedeagus totally lacking anellar projections. Female geni- talia with excessively broad anterior apophyses together with unequal signa and no colliculum 2 — Male genitalia with anellar spines on aedeagus. Female genitalia without above combination of characters 3 2. Male genitalia with valves relatively short; as in fig. 28. Female genitalia as in fig. 29. Externals as in figs. 58, tf, 59, 9. Larva forms ophistigmatonome on Nyssa leaves nyssaefoliella (p. 67) — Male genitalia with valves relatively long; as in fig. 30. Externals as in fig. 60. Female not known specimen 8 (p. 70) 3. Male genitalia with long tapering pseuduncus extending well beyond valves. Female genitalia with strongly sclerotised colliculum in the form of loops. Leaf and petiole miners and petiole gall formers 4 — Male genitalia with large rounded pseuduncus not usually projecting beyond valves and never far beyond. Female genitalia without colliculum or if weakly sclerotised, never in form of loops. Mostly species attacking bark 13 4. Male genitalia with palmate setae on the dorsal surface of valves, (not well de- veloped in one species). Female genitalia with enlarged spinose sac in asso- ciation with sclerotised colliculum • • • 5 — Male genitalia without palmate setae on valves. Female genitalia without spinose lobe at colliculum 8 ') The nominate subspecies of E. argyropeza (Zeiler, 1839) is European and therefore not included here. 38 Tijdschrift voor Entomologie, deel 124, afl. 2, 1981 5. Male genitalia with small three-pronged setae on dorsal surface of valves as in fig. 15. Hindwing of male without lance-shaped chitinous plate on costal margin. Female genitalia with equal signa, thin colliculum and minute spines on ductus as in fig. 16. Larva makes ophistigmatonome on leaves of Platanus occidentalis with long linear tract and large blotch . . . clemensella (p. 55) — Male genitalia with strongly palmate setae on dorsal surface of valves. Hindwing of male with lance-shaped chitinous plate on costal margin. Female genitalia not with above combination of characters 6 6. Male genitalia with strongly bilobed saccus; sharply pointed valves bearing palmate setae with shaft approximately equal to or shorter than digitate pro- cesses as in fig. 17c. Female genitalia with small signa, unequal in length and very small spicules on lobe associated with colliculum as in fig. 18. Externals occur in two forms as in figs. 54 and 55. Larva makes ophistigmatonome in leaves of Quercus sp similella(p. 56) — Male genitalia without strongly bilobed saccus; valves not sharply pointed and bearing palmate setae with shaft much longer than digitate processes. Female genitalia with signa approximately equal in length and large spicules on lobe associated with colliculum 7 7. Male genitalia variable, but as in fig. 19; some palmate setae on valves with shaft up to three times length of digitate processes. Female genitalia with signa short and narrow about half length of bursa copulatrix; short rows of spines on lobe associated with colliculum, as in fig. 20c. Externals in two forms as in figs. 56 and 57. Larva makes ophionome in Corylus sp. leaves virgulae (p. 59) — Male genitalia as in fig. 13; palmate setae on valves with shaft approximately twice length of digitate processes. Female genitalia with long, broad signa ex- tending almost full length and covering most of bursa copulatrix; very large single spines on the lobe associated with colliculum as in fig. 14c. Externals as in fig. 53. Larva makes ophistigmatonome in leaves of Platanus platanella(p. 51) 8. Male genitalia with long triangular pseuduncus tapering to point; aedeagus as in fig. 4 or 21. Female genitalia never with spiculate lobe associated with colli- culum 9 — Male genitalia with pseuduncus not triangular and rounded at least terminally. Female genitalia with spiculate lobe associated with colliculum 10 9. Male genitalia with dorsal arms of transtillae extending beyond saccus and comparatively small anellar projections; as in fig. 21. Female genitalia with thickened ring of colliculum and relatively long signa; as in fig. 22. Externals as in fig. 43. Larva forms stigmatonome on Betula sp. . . lindquisti (p. 61) — Male genitalia with dorsal arms of transtillae broad and short, not extending beyond saccus; large anellar projections; as in fig. 4. Female genitalia with thin ring of colliculum and short signa; as in fig. 5. A petiole gall former on Populus sp popullella(p. 41) 10. Male genitalia with small valVes only about half length of capsule; anellus with bifurcate spines totalling more than six, as in fig. 11. Female genitalia with large single or double spines on spiculate lobe; signa broad and short; as in fig. 12. Externals as in fig. 52. Larva makes ophistigmatonome on Rosa sp Wilkinson & Newton: Ectoedemia in North America 39 marmaropa (p. 49) — Male genitalia usually with valves larger than half capsule length; anellus with four spines. Female genitalia with spiculate lobe not as above; signa long 1 1 11. Male genitalia with large broad pseuduncus and anellus with four large spines as in fig. 7b. Female genitalia with broad signa tapering distally as in fig. 8. Ex- ternals as in fig. 44. Larva is probably a petiole gall-former on Populus sp canutus (p. 45) — Genitalia not as above 12 12. Male genitalia with highly characteristic pseuduncus and shape of anellar spines and valves; as in fig. 9. Female genitalia with long narrow signa and large spines on spiculate lobe of colliculum; as in fig. 10. Externals as in fig. 51. Larva makes ophistigmatonome on Carya sp trinotata (p. 46) — Males non-existent. Parthenogenetic females with genitalia as in fig. 6. Larva mines first in petiole and then makes stigmatonome in lamina of Populus sp argyropeza downesi(p. 45) 13. Male genitalia with not more than four large, simple spines near phallotreme; valves markedly incised along inner margin. Female genitalia without any thickening or sclerotisation at colliculum; apophyses usually not markedly dif- ferent in length 14 — Male genitalia with more than four spines near phallotreme; valves not markedly incised. Female genitalia with colliculum thickened; posterior apophyses usually markedly longer than anteriores 16 14. Male genitalia with saccus strongly concave; anterior arms of valves long; ae- deagus with thickened circular band of cornuti; as in fig. 35. Female not known acanthella (p. 75) — Male genitalia not with above combination of characters. Female genitalia as in fig. 24 or 27 15 15. Forewings mottled, indistinct fascia, dull. Male genitalia with valves extending as far as, or further than, extremity of pseuduncus; dorsal arms of transtillae short as in fig. 26. Female genitalia as in fig. 27. Larva makes ophistigma- tonome in leaves of Ulmus sp ulmella (p. 65) — Forewing with distinct fasciae, shining silver as in fig. 49. Male genitalia with valves not extending beyond pseuduncus; dorsal arms of transtillae reaching extremity of capsule as in fig. 23. Female genitalia as in fig. 24. Larva makes ophistigmatonome in Rubus sp. leaves rubifoliella (p. 63) 16. Forewings marked with four small silver patches as in fig. 48. Male genitalia with valves extending well beyond pseuduncus; crown of thorn-like spines present near phallotreme as well as large anellar spines; as in fig. 31. Female genitalia with simple circular thickening at colliculum. Anterior apophyses broad and unusually long reaching as far as posteriores as in fig. 32. Larva makes ophionome with blotch later, in leaves of Ostrya, Carpinus, Corylus and Betula sp quadrinotata (p. 70) — Male genitalia with valves not extending far beyond pseuduncus; anellar spines and cornuti characteristic, similar to figs. 33 and 37. Female genitalia with complex thickened folds at colliculum. Anterior apophyses always much shorter than posteriores 17 40 Tijdschrift voor Entomologie, deel 124, afl. 2, 1981 17. Male genitalia with valves just reaching end of pseuduncus; as in fig. 37. Female genitalia as in fig. 38 18 — Genitalia not as above 19 18. Larva mines the bark of Quercus sp. Externals as in fig. 61 " heinrichi (p. 78) — Larva makes galls on twigs of Castanea. Externals as in fig. 62 castaneae (p. 80) 19. Male genitalia without swellings or characteristic knobs on inner surface of valves; as in fig. 36. Externals as in fig. 55. Female not known piperella (p. 77) — Male genitalia with characteristic knobs on inner surface of valves. Female genitalia with spicules on colliculum 20 TABLE FOR THE IDENTIFICATION OF ECTOEDEMIA SPECIES FROM HOST PLANTS AND DAMAGE Ectoedemia host host site of larval form of larval probable Type species genus species damage damage No. of locality generations For Amelanchier see nyssaefoliella description. Wilkinson & Newton: Ectoedemia in North America 41 20. Male genitalia with simple rounded knob on inner side of valves; valves not reaching beyond pseuduncus; as in fig. 42. Externals as in fig. 50. Female not known mesoloba (p. 85) — Male genitalia different from above and valves extending beyond pseuduncus. Only females known in one species 21 21. Genitalia as in figs. 33 and 34. Male genitalia with three small knobs on inner side of valves. Anterior arms of valves long extending beyond capsule. Ex- ternals as in fig. 47 obrutella (p. 72) — Genitalia and externals not as above 22 22. Genitalia as in figs. 39 and 40. Male genitalia with characteristic knobs on inner side of valves. Dorsal arms of transtillae short. Female genitalia with comparatively simple ductus bursae. Externals in fig. 63. Larva makes a bark ophionome in Castanea sp phleophaga(p. 81) — Female genitalia as in fig. 41. Complex thickening in ductus bursae. Externals as in fig. 46. Male not known chlorantis (p. 84) The populella group The North American Ectoedemia can be divided into four species groups. The populella group contains the first five species. All males have a pointed pseuduncus except argyropeza downesi where males are not known to occur. They lack the multi-branched setae on the inner sides of the valves found in the platanella group. All females have a sclerotised ring-shaped colliculum and associated denticulate patch. E. lindquisti could belong in this group but it does not have the same charac- teristic anellar processes and has perhaps more affinity with the rubifoliella group. This group includes all the species which attack petioles: populella, argyropeza and probably canutus, as well as the leaf miners trinotata and marmar opa. Ectoedemia populella Busck (figs. 4, 5) Ectoedemia populella Busck, 1907: 98. Ectoedemia populella Busck; Braun, 1917: 197. Ectoedemia populella Busck; Braun in Forbes, 1923: 83. Ectoedemia populella Busck; McDunnough, 1939: 107 (no. 9783). Ectoedemia populella Busck; Borkowski, 1972: 697. Ectoedemia populella Busck; Wilkinson & Scoble, 1979: 74. Description. — Fully described from Canada by Wilkinson & Scoble (1979). Diagnosis. — Usually larger than the other members of the genus studied here and characterised by the cupreous reflections of forewings. Differs from canutus in the valves, which are shorter and broader, and the gnathos, which is not bifid lat- erally, in populella. The gall forming habit differentiates the larvae from those of argyropeza which produce more irregular callus tissue. Distribution. — USA: — New York, New Hampshire, Massachusets, Ohio (Braun). Canada: — Ontario, Manitoba. Material examined. — In USNM —USA: New York, Monroe Co.; on Populus 42 Tijdschrift voor Entomologie, deel 124, afl. 2, 1981 m )ik -3-mm 10 Fig. 4. Ectoedemia populella Busck, $ (a) and 9 (b) genitalia. sp.; 8 tf, 2 9, 2 ex., 1 — 16.V.1948 (Kimball). "Awene, Man."; 1 Ç, 24.V.1921 (Criddle). In CNC — Canada: Ontario, Bells Corner; 2 Ç, 30.V.1965 (Sattler); Lake Erie, Cultus; on Populus tremuloides; 1 tf, l.iii.1962; La Passe; on Populus sp.; 1