Antispila oinophylla new species (Lepidoptera, Heliozelidae), a new North American grapevine leafminer invading Italian vineyards: taxonomy, DNA barcodes and life cycle Author Nieukerken, Erik J. van Author Wagner, David L. Author Baldessari, Mario Author Mazzon, Luca Author Angeli, Gino Author Girolami, Vicenzo Author Duso, Carlo Author Doorenweerd, Camiel text ZooKeys 2012 170 29 77 http://dx.doi.org/10.3897/zookeys.170.2617 journal article http://dx.doi.org/10.3897/zookeys.170.2617 1313-2970-170-29 Antispila oinophylla Van Nieukerken & Wagner sp. n. Figs 1 -69- 296263 Antispila sp.; Baldessari et al. 2009 : 68 [first record for Italy]; Duso et al. in press [pest status]. Antispila ampelopsifoliella ; Needham et al. 1928 : 289 [partim]; Davis 1983 : 4 [partim]; van Nieukerken 2011 : Fauna Europaea database; Lastuvka 2009 : S57; van Nieukerken et al. 2011a : 51. Misidentifications. Antispila ampelopsiella ; Dyar et al. 1903 : 539 [partim]; Barnes and McDunnough 1917 : 181 [partim]; Forbes 1923 : 226 [partim]; McDunnough 1939 : 91 [partim]; Brower 1984 : 29 [partim]. Misidentifications. Type material. Holotype ♂, USA: Georgia, Murray Co., Chattahoochee Nat. Forest, E of Chatsworth, GA rd 52, 523 m, 34.74066N , 84.71852W , hardwood forest along highway, leafmines on Vitis aestivalis var. aestivalis, 14.x.2010, EvN2010266, emerged 14. iv- 4.v.2011, E.J. van Nieukerken & C. Doorenweerd, Genitalia slide EJvN 4204, RMNH.INS.24204 (RMNH). Paratypes. 32♂, 31♀. Italy: 1♂, 3♀ (all dissected), Trento, Borgo Valsusana, leafmines 2007, on Vitis vinifera , emerged 1. iii- 26.iv.2008, M. Baldessari; 3♀, same locality, 13.viii.2008; 10♂, 1♀ (1♂ RMNH.INS.23920 dissected & DNA barcode), same locality, 18.viii.2008; 17♂, 18♀ (1♂ RMNH.INS.24038, 1♀ RMNH.INS.24039 dissected & DNA barcode), same locality, 29.vi.2009, leafmines on Vitis vinifera , EvN no 2009903, emerged in Leiden, 14. vii- 6.viii.2009, M. Baldessari (all RMNH). Canada: 1♂, Ontario, Ottawa, mines on Vitis , rearing 57-112, emerged 31.iii.1958, Freeman & Lewis (CNC); 1♀, Quebec, Hull, mines on Vitis , rearing 55-228, emerged 26.vi.1956, T.N. Freeman (CNC). USA: 1♂, Connecticut , Tolland Co., Mansfield, 22.viii.1989, leafmines on Vitis, DLW89H37 breeding, emerged 4.v.1990, D.L. Wagner (DLW); 1♀ (dissected), Connecticut, Windham Co., Hampton, 916 Pudding Hill Rd., leafmines on Vitis 1-5.ix.1988, DLW 88J7, emerged 20.vii.1989, D.L. Wagner (DLW); 1♂ 1 ♀, Georgia, same data as holotype; 1♀, Georgia, Murray Co., Chattahoochee Nat. Forest, Cohutta Overlook, 730 m, 34.785356N , 84.627323W , shrub in forest clearing, leafmines on Vitis aestivalis var. bicolor, 14.x.2010, EvN2010270, emerged 19.iv.2011, E.J. van Nieukerken & C. Doorenweerd (RMNH); 1♀ (dissected, EvN 4211), Kentucky, [Covington], bred, [19th century], Chambers, "pseudotype," MCZ Type 1367 (MCZ); 1♂, 1♀ (♂ dissected), Vermont, Chittenden Co., South Burlington, leafmines on Vitis 11.viii.1988, DLW 88H23, emerged 30. iii- 15.v.1989, D.L. Wagner (DLW). Non-type material (all in RMNH).Italy: leafmines & larvae, Borgo Valsusana, 29.vi.2009, on Vitis vinifera , EvN no 2009903, M. Baldessari. USA: 1 larva, Connecticut, Tolland Co., Storrs campus, on Vitis labrusca , 185 m, 8.ix.2011, EvN2011168, B. Gagliardi; leafmines and larvae (being reared), Connecticut, New London Co., Connecticut College Arboretum, 34 m, 41.37929N , 72.11121W , on Vitis labrusca , 10.ix.2011, EvN2011193, E.J. van Nieukerken; leafmines and larvae (being reared), Connecticut, New Haven Co., West Rock Ridge SP, 125 m, 41.33353N , 72.96423W , on Vitis aestivalis var aestivalis, 10.ix.2011, EvN2011198, E.J. van Nieukerken; leafmines & larvae (DNA barcode RMNH.INS.18394), Georgia, same data as holotype; leafmines & larvae (DNA barcode RMNH.INS.18392), Georgia, Murray Co., Chattahoochee Nat. Forest, Cohutta Overlook, 730 m, 34.78535N , 84.62732W , shrub in forest clearing, leafmines on Vitis aestivalis var. bicolor, 14.x.2010, EvN2010270, E.J. van Nieukerken & C. Doorenweerd (RMNH); leafmines & 2 larvae (DNA barcode RMNH.INS.18533), Massachusetts, Berkshire Co., Beartown State forest, SW margin, 480 m, 42.19814N , 73.28928W , on Vitis riparia , 12.ix.2011, EvN2011208, E.J. van Nieukerken; leafmines & larvae (DNA barcode RMNH.INS.18558), New York, Essex Co., Hwy 9N, 3.5 km WSW Keeseville, 142 m, 44.49233N , 73.52042W , on Vitis riparia , 14.ix.2011, EvN2011237, E.J. van Nieukerken; leafmines & larvae (DNA barcode RMNH.INS.18555), New York, Essex Co., Wilsboro, Noblewood Park, 62 m, 44.35216N , 73.36435W , on Vitis riparia , 14.ix.2011, EvN2011244, E.J. van Nieukerken; leafmines & larvae (DNA barcodes RMNH.INS.18298, 18300), Tennessee, Blount Co., NP Great Smoky Mts, Rich Mountain Gap, 619 m, 35.64557N , 83.80537W , rich forest on limestone ridge, leafmines on Vitis vulpina , 2.x.2010, EvN2010119, E.J. van Nieukerken & C. Doorenweerd (RMNH); mine and larva, (DNA barcode LGSME035-06), Tennessee, Cocke Co., Cosby, ATBI house, 35.77771N , 83.21359W , on Vitis sp. 12.viii.2006, DLW 2006H55, D.L. Wagner (DLW); leafmines & larvae (being reared and DNA barcode RMNH.INS.18669), Vermont , Addison Co., Button Bay SP, Lake Champlain borders, 44 m, 44.18154N , 73.36892W , on Vitis riparia , 16.ix.2011, EvN2011253, E.J. van Nieukerken. Differential diagnosis. In North America, at least four other species have an apical silver spot (together forming the ampelopsifoliella group): Antispila ampelopsifoliella , Antispila voraginella , which has a darker head, an unnamed species from Vitis (here Antispila "vitis2" ) and Antispila hydrangaeella Chambers, 1874. The latter, which is closely similar in appearance, can be separated by the greater number of white flagellomeres at the antennal tip (six segments) and feeds on Hydrangea arborescens L. ( Hydrangeaceae ). Dissection of genitalia is needed to distinguish Antispila oinophylla from other members of the ampelopsifoliella group. Male genitalia are characterised by the long carinal spine at the phallotrema and several other details; female genitalia differ by the number of cusps on the ovipositor from at least Antispila ampelopsifoliella . In Europe, Antispila oinophylla differs from all other Heliozelidae with a similar forewing colour pattern (species of Antispila , Antispilina and Holocacista ) by the presence of a small silvery spot in the apical part of forewing and the distinctly white head. Some Elachistidae are superficially similar, but differ in long-pointed and upcurved palpi, longer antennae and more elongate habitus. The leafmine of Antispila oinophylla differs from that of Holocacista rivillei by its short initial gallery, which is later usually completely incorporated into the blotch, whereas the initial gallery of Holocacista rivillei mines is usually as long as or longer than the blotch, and remains intact. In Eastern North America other Vitis -feeding Antispila do not show the concentric arrangement of frass that is typical for Antispila oinophylla - particularly in thinner leaves -and the mines are often larger. Mines of Antispila cf isabella and related species are much larger, and also have much larger cut-outs, 5 mm or longer. Since not all Vitis miners have been comprehensively studied, mine identification cannot yet be relied on. Description. Adult (Figs 1-5). Head face and vertex covered with appressed, strongly metallic, silvery-white scales, more prominently raised in male. Palpi porrect, white; base of proboscis covered with white scales. Antenna fuscous, apical 1 or 2 flagellomeres white. Labial palp silvery white, slightly upturned. Thorax lead-coloured, shiny, contrasting with forewings. Legs grey, tarsi mostly yellowish white, especially on undersides. Forewing dark fuscous with silver-golden patterning; an outwardly oblique fascia from 1/8 of posterior margin to 1/4 of costa, narrowing towards costa; triangular (dorsal) spot at middle of posterior margin, reaching to middle of wing, smaller triangular costal spot just beyond middle, sometimes touching dorsal spot; small, silvery subapical spot in middle of wing at 3/4; fringe line distinct. Terminal fringe paler. Hindwing pale grey. Abdomen lead-coloured, including vestiture on external genitalia. Figures 1-5. Antispila oinophylla , adult habitus. 1 Male holotype, RMNH.INS.24204 2 Female paratype, RMNH.INS.24039, Italy, Borgo Valsusana. 3-5 Alive male,Georgia, paratype, emerged 29.iv.2011. Measurements: male: forewing length 2.5-2.8 mm (2.6 +/- 0.10, n=11), wingspan 5.5-6.2 mm, 25-31 antennal segments (29.1 +/- 1.9, n=11); female: forewing length 2.3-2.8 mm (2.5 +/- 0.16, n=10), wingspan 4.8-5.6 mm, 25-29 antennal segments (27.2 +/- 1.4, n=8). Venation (Fig. 6). Forewing with Sc barely visible. R1 a separate vein, connected by persistent trachea to Rs+M stem. Rs+M terminating in five branches, interpreted as Rs2 (possibly with 1) to costa, Rs3+4 to costa just before apex, M1 to dorsum just beyond apex, M2+3 to dorsum and a weakly developed CuA. A1+2 a strong separate vein. Hindwing with Sc barely or not visible, Rs+M a strong vein, bifurcate from ca. 1/4th, upper vein ending in two branches: Rs and M1, lower vein single (M3); Cu and A1+2 separate veins. Compared to the complicate venation of many other Antispila species, including the type species Antispila metalella , (example in Fig. 7, Antispila treitschkiella ) venation reduced with loss of forewing cell, separate M stem and connection between R1 and Rs, loss of Rs1 and in hindwing loss of M2. The venation more closely resembles that of Holocacista rivillei (Fig. 8), which is even more reduced and also lacks Cu in the forewing. Figures 6-7. Antispila , venation. 6 Antispila oinophylla , male, Italy, RMNH.INS.24257 7 Antispila treitschkiella , male, Netherlands, Leiden, RMNH.INS.24258. Figure 8. Holocacista rivillei , venation.Female, Italy, RMNH.INS.24259. Male genitalia (Figs 9-16). Uncus bar-shaped, with two large setae dorsally. Vinculum very long, anteriorly rounded, posteriorly shallowly bilobed. Valva more or less triangular, pecten on pedicel, with 10-13 comb teeth (Fig. 15); inner margin of valva with setose lobe anterior to pecten pedicel; basally with a triangular protuberance, almost touching that of other valva; transtilla with trapezoid medial plate, sublateral processes relatively short. Juxta anteriorly spade-shaped, about half as long as phallus. Phallus long, anteriorly much widened, at phallotrema with a comb of about 10-12 strong teeth and at left side a very long curved process (Figs 10-12, 16). Figures 9-16. Antispila oinophylla , male genitalia.Paratype, Italy,RMNH.INS.23920 (9, 15, 16), Paratype, Italy, RMNH.INS.15247 (12),Holotype, RMNH.INS.24204 (10, 11, 13-14). 9 Complete genitalia with separate phallus in ventral view 10-12 Phallus and juxta in ventro-lateral view 15-16 Complex of tegumen, uncus, valvae and transtilla 15 Detail of valval tips and pectinifers 16 Detail of spines near phallotrema. Female genitalia (Figs 17-20). Ovipositor with 4-5 cusps at either side (Fig. 19). S8 medially indented, with many papillate setal sockets. Vestibulum with broad, indistinct sclerotization and no spines (Fig. 18). Figures 17-20. Antispila oinophylla , female genitalia. 17 Terminal segmentsandapophyses, ventral view, paratype, EJvN4211,USA, Kentucky (pseudotype ampelopsifoliella) 18 Internal genitalia, lateral view, showing sclerotisation in vestibulum,paratype, EJvN4206,USA, Connecticut 19 Ovipositor tip, dorsal view, EJvN4206 20 Detail of S8, ventral view,paratype, Italy,RMNH.INS.15244. Biology. Host plants.In North America reared from or found as larva on summer grape Vitis aestivalis Michx., both var. aestivalis and var. bicolor Deam, fox grape Vitis labrusca L., riverbank grape Vitis riparia Michx. and frost grape Vitis vulpina L. Literature records of Antispila "ampelopsifoliella" from Vitis or grape likely refer to this species ( Chambers 1874a , b ; Forbes 1923 ; Needham et al. 1928 ). We did not find any reports of this species occurring in vineyards in North America. In Italy mines produced by Antispila oinophylla were detected on various Vitis vinifera cultivars, hybrids (e.g. Vitis riparia x rupestris) and French-American grapes (e.g. Clinton). Infestation levels on the latter were comparable with those observed on commercial vineyards. A preference for some grape cultivars (e.g. Cabernet Sauvignon, Chardonnay, Muscat) is suggested from observations carried out in mixed cultivar vineyards. It is interesting that we also found active mines on Virginia creeper Parthenocissus quinquefolia in Italy (Levico and Caldonazzo, Trento province) (identification of larvae confirmed by DNA barcodes, no rearing attempted), whereas we have as yet no records of Antispila oinophylla from this host in North America. Leafmines (Figs 21-28). The egg is inserted on the underside of a leaf, usually within 1-2 mm from a vein. The mine starts as a rather straight or slightly contorted gallery towards the vein, usually forms a right angle and often follows the vein for a short distance, then again turns away from the vein where it expands into a blotch. The gallery portion, of variable length, is usually later incorporated into the blotch mine. The frass is linear, usually occupies the complete mine width, but occasionally is deposited in a thin line (Fig. 27). In the blotch much of the blackish-brown frass is deposited close to the origin in semicircular concentric frass lines. This characteristic pattern is best seen in thin shade leaves (e.g., Figs 25, 26); in sun-exposed leaves the frass pattern is often obscured. The whole mine occupies as a rule an area of less than 10 x 10 mm; only in thin leaves are mines appreciably larger. The larva cuts out an elliptic case of about 3.2-4.0 mm long. Figures 21-28. Antispila oinophylla , life history: leafmines on several species of Vitis and different localities. 21, 23, 24 Italy, Borgo Valsusana, Vitis vinifera , 25.vi.2009 22 USA: Vermont, Button Bay SP, Vitis riparia 16.ix.2011 25 USA:Tennessee, NP Great Smoky Mts, Vitis vulpina , 2.x.2010, mine in shade leaf 26, 28 USA: Georgia, type locality, Vitis aestivalis var. aestivalis, 14.x.2010 27 USA: Vermont, Button Bay SP, Vitis riparia , 16.ix.2011, DNA barcode,RMNH.INS.18589. Distribution (Fig. 29, 62). In North America, Antispila oinophylla is known with certainty (material cited) from Canada: Ontario, Quebec; USA: Connecticut, Georgia, Kentucky, New York, Tennessee, and Vermont. Records under Antispila ampelopsifoliella from Maine, Missouri, and Ohio ( Brower 1984 , Forbes 1923 ) may partly refer to this species. In Europe introduced into northern Italy, see below. In our experience in the southern Appalachians and New England, at least in the fall, Antispila oinophylla is often the most abundant Antispila species occurring on Vitis . Figure 29. Antispila oinophylla , distribution in North America. Etymology. The epithet oinophylla, a noun in apposition, is from the Greek οινος (oinos = wine) and φυλλον , plural φυλλα (phyllon, phylla = leaf), "wine leaves," because the larva lives in the leaves of the grapevine from which wine is made. Justification for status as new species. Four species feeding on Vitaceae have been named previously from North America. No name-bearing types are available for three species, only for Antispila voraginella is a holotype extant. The latter is clearly different from Antispila oinophylla , and restricted to western North America. For the eastern species Antispila isabella , Antispila viticordifoliella and Antispila ampelopsifoliella , we have only the original descriptions and subsequent interpretations to establish identities. The fact that our preliminary sampling of DNA barcodes for grape-feeding Antispila show great diversity, complicates matters further. Below, we will discuss these three species in the chronological order of their descriptions. Antispila isabella was described from mines on "Isabella grape" (a cultivar of Vitis labrusca ) and adults ( Clemens 1860 ). The description unequivocally describes a relatively large species without a silvery apical spot. Clemens characterizes the case (shield) as large and almost roundish - both features exclude our species. We have tentatively named one larger barcode cluster as Antispila cf. isabella, because mines and adults conform to this description. Antispila viticordifoliella was also described by Clemens in 1860, from mines on "wild grapes" only, differing by a smaller case (shield) and a larva "without dots." Although the foodplant was not explicitly mentioned by Clemens, from the species name it is evident that the host must have been Vitis cordifolia Michx. (a synonym of Vitis vulpina ). In fact his very brief description could fit the mines of Antispila oinophylla , but subsequently the name has always (e.g. Forbes 1923 ) been used in the sense of Chambers (1874a) , who first described the moth (as "viticordifoliella N. sp.?"), without an apical spot and with several, white, distal flagellomeres. He reared that moth from the same hostplant ( Vitis cordifolia )as Clemens did, and was not able to find the mine on any other Vitis ( Chambers 1874a : 169). One of the species that we studied from Parthenocissus has similar externals, and is named here Antispila cf viticordifoliella (Fig. 37). Because we haven't been able to find or rear any similar adults from Vitis we are at the moment unable to establish if the Parthenocissus miner is indeed the same as Antispila viticordifoliella , but clearly it is not our species (because it lacks an apical spot). In a future revision a neotype will need to be selected to firmly anchor the identity of this species, material from the Chambers' collection (two extant "syntypes" , see Miller and Hodges 1990 ) probably is most suitable for that goal. In collections and websites (e.g., http://mothphotographersgroup.msstate.edu/) the name Antispila viticordifoliella is often misinterpreted as the species that we call Antispila cf isabella or a closely related one. Antispila ampelopsifoliella :Chambers(1874a: 168) only briefly described the mine and larva from " Ampelopsis quinquefolia " [= Parthenocissus quinquefolia ] (and stated that he "never succeeded in breeding it."). Only a month later he described the moth under the name " Antispila ampelopsisella " [sic, considered as a subsequent incorrect spelling], writing: "Since that paper was placed in the hands of the Editor, many months ago, I have succeeded in rearing it from the mine [from Parthenocissus ]" ( Chambers 1874b ). Theconfusionof the new specieswith Antispila ampelopsifoliella dates from Chambers' original description, because he also described a moth that he reared from Vitis and shows the external characters of both species: "Last summer I found its leaves [referring to a Vitis species] mined by a larva closely resembling that of Antispila ampelopsifoliella , supra, and which I suspect to be the same. ..... From it I bred the species described below, which I do not now name, as it may prove to be identical with Antispila ampelopsifoliella ." ( Chambers 1874a ). One month later he wrote: "but I believe it to be the same" ( Chambers 1874b ). Ever since these two publications, the species has been considered to feed both on Parthenocissus and Vitis (e.g., Forbes 1923 ; Brower 1984 ). However, our rearing and barcode data show that two or three species of Antispila are feeding on Parthenocissus , which show large barcode distances to Antispila oinophylla or other Vitis miners (Fig. 30), and thus are not identical. Figure 30. Neighbor-joining tree for heliozelid COI barcodes, based on uncorrected pairwise distances. Numbers on branches are bootstrap values, 10,000 replicates. Vitaceae -feeding clusters are coloured differently, others in black. Labels include species name or informal name, codes for country and state (in North America) and sample numbers (Genbank numbers for sequences taken from Genbank). In Chambers' collection at MCZ there are three specimens under the name Antispila ampelopsifoliella that probably served as the basis for the adult description. These specimens were termed pseudotypes ( Miller and Hodges 1990 ), since they were not available at the time of the original description, because then Chambers only had mines and larvae available. Of the three specimens, one is completely missing from the pin. The one labelled as from Parthenocissus unfortunately is heavily damaged, only a forewing and hindwing being present. A third specimen, a female, is complete and was dissected (Fig. 17). This specimen, however, appears to be Antispila oinophylla . This is no surprise, since Chambers (1874a , 1874b ) considered the Vitis miner to be the same as the Parthenocissus miner, and thus he would have placed specimens reared from both hosts under the same name. There is no indication of the hostplant or the collecting year on this particular specimen, so it is useless for confirmation of the identity of Antispila ampelopsifoliella . We restrict here the usage of the name Antispila ampelopsifoliella to the species feeding on Parthenocissus , with an apical spot (The generic name for Parthenocissus quinquefolia was Ampelopsis at the time Chambers described the species.) Although we have not obtained a DNA barcode form such an adult, the fact that an adult from the other cluster on this host (see below) does not have such a spot and is tentatively identified as Antispila cf viticordifoliella, we can associate Antispila ampelopsifoliella adults with one of the larval types. When adults are available for all barcode clusters, we suggest that a neotype be selected from material reared from Parthenocissus from the vicinity of Covington, Kentucky, to fix the identity of Chambers' name.