A review of the Cochylimorpha perfusana (Guenée, 1845) species group (Lepidoptera, Tortricidae) in Europe, with the description of a new species from the Southern Carpathians (Romania)

. European species of the Cochylimorpha perfusana (Guenée, 1845) species group are revised. Four species are recognized, all closely related to C. perfusana (Guenée, 1845) with which they have previously been confused. Based on morphological differences and DNA barcode data, C. callosana (Herrich-Schäffer, 1856), sp. rev. and C. dorsimaculana (Preissecker, 1908), sp. rev. , are re-instated to their original species rank. Inconsistencies in their taxonomy leading to their synonymy are shown and corrected. Cochylimorpha buce-giana Z. Kovács, S. Kovács & P. Buchner, sp. nov. is described from the subalpine meadows of the Bucegi and Făgăraș Mountains (Southern Carpathians, Romania), and the other three species are redescribed. Adults and genitalia of both sexes of all four, and the habitat of three taxa are figured. The geographic distribution of each species is re-evaluated based on the examined material and a critical analysis of the literature. Literature data for which no material could be found for examination are treated as doubtful and in need of confirmation. Cochylimorpha perfusana is widely distributed in mountain areas of Europe (Alps, Carpathians and Pirin Mountains), C. callosana is widespread only in the north-western Balkan Peninsula and north-eastern Italy, and local in eastern France, Corsica and Hungary, the other two seem to be endemics, C. dorsimaculana to Wachau and Retz in Lower Austria and C. bucegiana sp. nov. to the Southern Carpathians in Romania. Cochylimorpha perfusana is recorded for the first time from Bulgaria; the first confirmed records of C. callosana from Slovenia and Hungary are given; and in the north-eastern Italian fauna C. callosana replaces C. perfusana . Records of C. perfusana from Ukraine, the Czech Republic, Slovenia, and north-western Italy require confirmation.

The suspicion that C. perfusana might be more than a single species arose more than 20 years ago, when the first two authors revised the Romanian Cochylini.They observed that material from two populations differed.Owing to the similar male genitalia and the lack of a female from one of the populations, they treated them as the two forms of C. perfusana, the typical form and f. callosana, following the species concept accepted at the time (Kovács and Kovács 2005: 87-88).When comparing data from old literature sources (Herrich-Schäffer 1851, 1856;Kennel 1913) to those in the monograph of the Palaearctic Cochylini (Razowski 1970), they discovered inconsistencies in the taxonomy which, together with the similarity of the male genitalia, probably led to the synonymy of all known taxa in this species group (Razowski 1970: 38-39, 162).They also noticed that in the earlier and subsequent works of Razowski (1970Razowski ( , 1987Razowski ( , 2001Razowski ( , 2002Razowski ( , 2009) ) two different genitalia of both sexes were figured under the name C. perfusana.Later, when the female genitalia of both Romanian populations became available for study, they realised that those from the Eastern Carpathians are similar to the specimens originating from Austria and the genitalia figured by Razowski (1970), but the genitalia of the other population, from the Southern Carpathians, differ from all genitalia figured by Razowski (1970Razowski ( , 1987Razowski ( , 2001Razowski ( , 2002Razowski ( , 2009)).They concluded that in Europe there may be at least three different species.In the meantime, the last two authors of the present study observed that the material in their collections originating from different regions of Europe was heterogeneous, and also suspected the existence of more species.Type specimens of the available names were examined, and material collected from the type localities of C. perfusana and its synonyms and other regions of Europe were examined and barcoded.The differences in the external morphology and the genitalia of both sexes, combined with the results of the molecular studies, confirmed the hypothesis of a species group and revealed the existence of four related taxa in the European fauna.We describe a new species and compare it with the three previously named species.

Material and methods
A total of 157 specimens were examined from four major mountain ranges of Europe: the Austrian, Slovenian, Italian and French Alps; the Romanian Carpathians; the Pirin Mountains in Bulgaria; and the Velebit Mountains and its neighbouring low-mountain habitats in Croatia, Slovenia and Italy.The examined material was dried, pinned and partly set.The terminology of wing pattern and genitalia follows Razowski (1970Razowski ( , 2002)), except that of the cornuti, which follows Anzaldo et al. (2014), and the costal roll (Pérez Santa-Rita et al. 2022).Photos were taken with a Canon EOS 5D Mark III camera, adults with a Canon lens MP-E 65, using ring flash and a magnification of 2:1, specimen details with 4:1 magnification, and genitalia with a microscope (Wild Heerbrugg) using a 10× objective and a 2.5× ocular lens.Images were stacked using Helicon Focus 4.80 software.Genitalia preparations followed standard techniques (Robinson 1976).Male preparations were stained with mercurochrome and females with chlorazol black, which gives a better result than using the same stain for both sexes.Samples from all major collecting areas were studied to assess possible geographical variation.Additional specimens were also examined from the different localities to detect potential intraspecific variability.
DNA samples from 19 specimens of the C. perfusana species group were processed at the Canadian Centre for DNA Barcoding (CCDB, Biodiversity Institute of Ontario, University of Guelph) to obtain DNA barcodes (658 base-pair long segments of the 5' terminus of mitochondrial cytochrome c oxidase I gene) following the standard protocol (whole barcode region sequenced at once, 3 samples) (deWaard et al. 2008), degraded material protocol (barcode region split into 2 fragments which were sequenced separately, eight samples) (Wilson 2012) and next-generation sequencing protocol (barcode region split into six short fragments which were sequenced separately, 8 samples) (Prosser et al. 2016).These resulted in seven full length sequences (658 bp), and four shorter sequences (one of 638 bp, one of 550 bp, two of 145 bp); eight samples failed.Two full length sequences of Cochylimorpha jucundana (Treitschke, 1835) are used as the outgroup.The sequences were submitted to GenBank, where further details including complete voucher data and images can be accessed through the public dataset DS-DEEUR398 (http://www.boldsystems.org/index.php/Public_SearchTerms?query=DS-DEEUR398; https://dx.doi.org/10.5883/DS-DEEUR398) (BOLD; Ratnasingham and Hebert 2007).Degrees of intra-and interspecific variation of DNA barcode fragments were calculated under the Kimura 2 parameter model of nucleotide substitution using analytical tools of BOLD systems v. 4.0.(http://www.boldsystems.org).A neighbour-joining tree was constructed using MEGA6 (Tamura et al. 2013) under the K2P model for nucleotide substitutions.
Photographs of the habitats, immature stages and host-plants in nature were taken using Canon EOS 10D and 50D digital cameras and a Huawei Nova 5T mobilephone.

Results
The C. perfusana species group can be defined by a combination of the following characters.Adult (Figs 1-8) medium sized, the reticulate pattern of the forewing varying from conspicuous to barely discernible, with reduced sexual dimorphism, male forewing without costal fold and hindwing without costal roll, female hindwing usually with 3 bristles in the frenulum.The male genitalia (Figs 9-12) of all taxa are very similar, possessing a short, wide and tapering tegumen ending in a short terminal process similar to an uncus (pseudo-uncus); lobe-like socius covered with medium to long setae; short, wide and tapering valva, densely covered with setae on the inner surface and margins, costa varying from slightly concave to convex, sacculus short (about 1/3 length of valva) without a process, cucullus round; median process of transtilla long, wide and strongly sclerotized, apically with a group of small thorns; rod-like vinculum; membranous saccus; caulis with ventro-lateral folds; phallus about of the length of valva with tapering ventral phallic process and single non-deciduous, aciculate cornutus, half as long as phallus basally or latero-basally attached.The female genitalia (Figs 13-16) possess a sterigma with wide lateral sides and strengthened middle, with distinct anteostial sclerite, densely covered with microspines; wide ostium; short ductus bursae; and variable shaped corpus bursae with signum and accessory bursa.The species inhabit open meadows (Figs 37,39,46) from the lowlands up to 2100 m.There is a single generation per year; larvae seem to be confined to Centaurea (Asteraceae) species; adults fly during the day and are rarely attracted to light; and the species group has a European distribution.
Romania: 1 ♂, Carpații Orientali, Cheile Bicazului, 14-16.vi.1984;1 ♂, Ibidem, 4.vii.1987, genit. prep. Kovács 285;8 ♂, 3 ♀, Ibidem, 26-28.vi.1989, genit. prep Diagnosis.Externally, Cochylimorpha perfusana is characterized by a fine, smooth, regularly dispersed olive-green reticulate pattern that lacks an admixture of brown scales, and the roundish, almost uniformly small, yellowish white spots of ground colour.In the male genitalia the tegumen has convex margins and an elongated and pointed terminal process; the valva is slightly variable; the median process of the transtilla wide and long, has parallel margins and a tapering apex; the caulis has shallow ventro-lateral folds; and the phallus is straight with one long, straight, latero-basally attached cornutus.In the female genitalia the papillae anales are lanceolate and densely covered with strong short setae and a longitudinal row of very long setae; the apophyses are stout; the ductus bursae is very short; and the corpus bursae has the signum in the posterior 1/3, in the form of a very weakly sclerotized plate with distinct longitudinal folds.
Redescription.Adult, male (Fig. 1).Head.Frons and vertex covered with very pale olive-green scales.Labial palpus about 2.5 times length of diameter of eye, first segment short, second segment long and widened distally, third segment short, all covered with off-white scales medially, light olive-green dorsally, a mixture of olive-green and brown scales laterally, and long brown scales ventrally.Antenna filiform, reaching ½ length of the forewing, brown, dorsally covered with offwhite scales.
Thorax.Dorsally covered with light olive-green scales, long and erect in posterior 1/3, tegula pale olive-green laterally edged with a row of long off-white scales.Forewing length 7-9 mm, wingspan 15.5-19.5 mm.Forewing almost 3 times as long as wide, trapezoidal, only slightly widening from base to apex, costal margin without costal fold, slightly convex in basal 1/5, remainder straight, apex rounded, termen straight; almost evenly scattered small spots of off-white ground colour evenly distributed with fine and distinct pale olive-green reticulate pattern, lacking admixture of brown scales.Fringe pale olive-green.Hindwing without costal roll, dark grey, fringe off-white with light grey basal line.Underside of thorax and forewing brown, yellowish white distally on costal margin, along basal ½ of subcostal vein and fringe; hindwing light grey, with an indistinct light brown reticulate pattern in apical 1/3, light yellow along M 2 vein, fringe off-white.Legs brown, but fore-and mid-leg medially, hind-leg on both dorsal and ventral surfaces covered with off-white scales.
Abdomen.Dorsally uniformly covered with olive-green, ventrally a mixture of off-white and olive-green scales, each segment distally edged with a row of long, off-white scales, last segment off-white.
Variation.The spots of ground colour may vary from small and regularly scattered to fairly large and irregularly scattered in median and subterminal areas of forewing.Both the olive-green markings and the off-white ground colour may vary from greyish to yellowish.
Variation.Length of terminal process of tegumen slightly variable, costa of valva straight, ventral edge varying from slightly concave to straight or slightly convex.Cornutus sometimes with slightly curved distal apex.
Female (Figs 2,17).Forewing length 7-8.5 mm, wingspan 15-18.5 mm, generally slightly shorter and widening more from base to apex than that of male.Underside of forewing brown with yellowish white reticulate pattern on the apical 1/3.Hindwing with 3 bristles in the frenulum.
Variation.Hindwing grey with indistinct white spots along the external and dorsal margin, rarely with 2 bristles in the frenulum on one or both sides.
Female genitalia (Figs 13,32).Papilla analis about 1 ½ as long as segment VIII, narrow, elongated, lanceolate, densely covered with strong, short setae, with row of 10-13 strong, very long setae parallel to lateral edge.Posterior apophysis 1 ½ as long as segment VIII, with wide, weakly sclerotized posterior ½ and rod-like, stout, strongly sclerotized anterior ½.Segment VIII with group of strong and very long setae along and parallel to posterior margin.Anterior apophysis 1 ½ as long as posterior apophysis, rod-like, stout, strongly sclerotized, except its anterior end.Sterigma with wide lateral sides and strongly strengthened middle covered with tiny microspines.Habitat.Exposed mountain and high-mountain meadows on limestone substrate where its hostplants are abundant, at elevations from about 550 m to Dauphiné (France) up to 1700-1800 m in the Alps (France, Austria, Switzerland), between 1200-1300 m in the Eastern Carpathians (Romania) (Fig. 37) and at 2000-2100 m in the Pirin Mountains (Bulgaria).
Distribution (Fig. 48).Widespread in scattered localities in mountain areas of Europe.We examined voucher specimens from the Austrian Alps and France (type localities), Eastern Carpathians (Romania), Pirin Mountains (first record for Bulgaria) and only a historic specimen from Serbia; for details see above the examined material.
Further material was identified from figures in the literature or available on the world wide web: Ukraine (Crimea) (Razowski 1970: colour pl. 8 fig.79-1), however, in our opinion, this may be a misinterpretation of the label data ("Kr.Kennel (1913: 320) reports the Austrian Alps, Serbia and Switzerland.Caradja (1916: 54) mentions La Grave in France.In Razowski (1970: 163;2001: 37;2002: 43;2009: 37, all as perfusana) detailed distribution is given, most of which were confirmed by examined material: France, Switzerland and the Schneeberg in Austria refer to C. perfusana; records from north-eastern Italy and Croatia (Dalmatia) refer to C. callosana; and those from Wachau and Retz in Lower Austria refer to C. dorsimaculana.We were not able to confirm the following two: Hardegg may refer to the latter species, being a similar habitat also in Lower Austria; and Romania (Transylvania), specified as dorsimaculana, may refer to C. bucegiana sp.nov.Razowski (2001: 37) mentions C. perfusana from the Czech Republic, which is plausible, but we cannot confirm it.In both the catalogue of the Italian Tortricidae (Trematerra 2003: 51) and the on-line checklist of the fauna of Italy (Stoch 2003), the species is mentioned from northern Italy.In the former, Valle d'Aosta, Piemonte, Lombardia, Trentino-Alto Adige and Friuli-Venezia Giulia are given, the latter without details.Valle d'Aosta, Piemonte and Trentino-Alto Adige are high-mountain habitats which may refer to C. perfusana, but we cannot confirm this.All Italian specimens examined in this study, most of them from Friuli-Venezia Giulia and also from Trentino, proved to be C. callosana.Records from Lombardia, a low-mountain habitat, may also refer to the latter species, but require re-examination.Two historic specimens from Hungary deposited in ZSM were recorded from Budapest as C. perfusana by Fazekas (1994: 40), but these records refer to C. callosana (see also that species).Another record of C. perfusana from Hungary (Mátra Mountains, 1 specimen from Jászberény and 3 from Nagykáta, Cseh-domb, all F. Buschmann leg., coll.Mátra Múzeum) (Buschmann 2004: 224) was also recently disproved, as all specimens proved to be Cochylimorpha straminea ( Haworth, 1811) (Buschmann 2022: 159; Fazekas 2022: 118, 120; 2023: 24).The species was mentioned for the first time from Romania, in Transylvania, as Euxanthis dorsimaculana Preiss.by Galvagni and Preissecker (1913: 49), based on the written communication of J. Kennel, and which, in our opinion, may not refer to this species but to C. bucegiana sp.nov.This record was also cited by Razowski (1970: 163;2001: 37).Kovács and Kovács (2005: 88) recorded material from the Bicaz Gorge as forma perfusana which refers to this species, but the material from the Bucegi Mountains, mentioned as forma callosana, refers to C. bucegiana sp.nov.Székely and Cernea (2007: 141) recorded four specimens from the environs of Braşov (Codlea, Vlădeni) and Lacu Roşu, but we consider the record as doubtful, because two of these specimens, one from each of the latter two collecting localities, examined by us proved to belong to Aethes rubigana (Treitschke, 1830), and the other two specimens, not examined by us, may also be misidentified.The record by Lesar and Godevič (2010: 77) (Razowski 1987: 250, 313;1970: 163;2001: 37;2002: 43, pl. 5 fig. 91, pl. 47 fig. 91;2009: 37, pl. 9 fig. 97, pl. 40 fig. 97) and north-east Italy (Gradisca d'Isonzo, Monte Bondone) (Mann 1854: 576;Klimesch 1951: 24;Razowski 1970: 163;2001: 37;2002: 43;2009: 37; Peter Huemer pers.comm.)refer to C. callosana, for details see below.
As a consequence, C. perfusana is replaced by C. callosana in the north-eastern Italian and Croatian fauna, and the former is regarded as requiring confirmation from the Czech Republic, north-western Italy and Slovenia.We consider the record from Ukraine (Crimea) (Razowski 1970: colour pl. 8 fig.79-1) to be doubtful.
Taxonomic notes.Argyrolepia perfusana was described by Guenée (1845: 302) based on an unspecified number of specimens.Razowski (1970: 162), under Stenodes perfusana, states that the holotype, an undissected female labelled as "Type Guenée", is deposited in NHMUK (Fig. 17), giving page 301 for Guenée's description of perfusana, but later (Razowski 2002: 43;2009: 37) he gives the correct page number (i.e., 302); however, all publications mention only the Austrian Alps as the type locality ignoring Dauphiné in France (see also below in discussion).In accordance with the ICZN article 74.6 Razowski's (1970: 162) treatment of "Type Guenée" as the holotype must be considered a lectotype designation.(Preissecker, 1908), sp.rev.Figs 3,4,10,14,[21][22]29,33,38 Euxanthis dorsimaculana Preissecker, 1908, Verhandlungen der zoologisch-botanischen Gesellschaft in Wien 58: 70.Type locality: Wachau and Retz in Lower Austria.Synonymized with Stenodes perfusana (Guenée, 1845) by Razowski (1970: 162).Euxanthis dorsimaculana Kennel 1913: 340, pl. 15  Diagnosis.Cochylimorpha dorsimaculana is externally characterized by a slightly yellowish white ground colour that differs only slightly from the light brownish yellow reticulate pattern; only the large yellowish brown dorsal blotch with an admixture of brown scales is distinct.In the male genitalia the tegumen has straight margins and a pointed terminal process; the valva is wide; the median process of the transtilla is rather short, very wide, and with a round apex; the caulis is wide with shallow ventro-lateral folds; and the phallus is long and S-shaped with a ventrally and laterally curved and latero-basally attached cornutus ½ as long as the phallus.In the female genitalia the papillae anales are lanceolate and densely covered with strong short setae; the apophyses are stout; the ductus bursae is very short; and the corpus bursae is large with the signa posteriorly consisting of a small sclerotized plate with a few longitudinal folds and centrally a large group of thorns.

Cochylimorpha dorsimaculana
Redescription.Adult, male (Figs 3,38).Head.Frons and vertex covered with brownish yellow scales.Labial palpus about 2.5 times diameter of eye, first segment short, second segment long and wide, third segment short, all covered with brownish yellow scales on the medial surface and yellowish brown on the lateral surface.Antenna filiform, reaching ½ forewing length, brown, dorsally covered with brownish yellow scales.
Thorax.Dorsally covered with light brownish yellow scales similar to head, long and erect scales in posterior 1/3.Tegula light brownish yellow.Forewing length 7-8 mm, wingspan 16-18.5 mm.Forewing short and trapezoidal, gradually widening from base to apex, about 2 1/3 times as long as wide, without costal fold, costal margin and termen straight, apex rounded.Ground colour shiny yellowish white differing only slightly in colour from the light brownish yellow reticulate pattern forming an indistinct basal blotch, a median transverse fascia and a subterminal blotch, distinct dorsal part of median transverse fascia a large, yellowish brown dorsal blotch.Admixture of brown scales forming an indistinct narrow line along basal ½ of costal margin distally with small spots towards apex, distinct spots on edges of dorsal blotch and narrow oblique line on basal edge of subterminal blotch.Fringe yellowish white.Hindwing without costal roll, grey, darkest along margins, fringe slightly yellowish white with light grey basal line.Underside of thorax and forewing dark brownish grey, distal ½ of costal margin with a row of light yellow spots, yellowish white line along basal ½ of subcostal vein and along posterior margin, fringe yellowish white with an interrupted grey line; hindwing yellowish white densely covered with brownish grey scales in anterior ½, light yellow along M 2 vein, fringe yellowish white.Legs yellowish white with admixture of brownish grey scales on medial and lateral surfaces.
Abdomen covered with grey scales, light grey at distal edge of segments, and last segment with light brownish yellow scales.
Variation.Ground colour varying from almost pure white to yellowish white, markings and admixture of brown scales more or less pronounced.
Variation.Cornutus variably curved, its base of varying size and shape.Female (Fig. 4).Forewing length 7-8 mm, wingspan 16-18 mm.Forewing slightly wider than that of male.Hindwing with 3 bristles in the frenulum, nearly uniformly dark grey, fringe with distinct basal line.
Variation.Margins at end of hindwing veins with a row of dark grey dots.Female genitalia (Figs 14,33).Papilla analis 1 1/3 as long as segment VIII, narrow, elongated, lanceolate, densely covered with strong and short setae.Posterior apophysis almost 2 times as long as segment VIII, strongly sclerotized, posterior 1/3 wide, rod-like and stout in remainder.Segment VIII with long, strong setae dispersed on posterior ½.Anterior apophysis similar to posterior apophysis, but slightly longer.Ostium ½ as wide as segment VIII.Antrum trapezoidal, strongly sclerotized, 2/3 as long as wide.Sterigma with widened lateral sides and strengthened medially, densely covered with microspines.Anteostial sclerite weak, indistinct, flattened, with microspines.Ductus bursae membranous, as wide as antrum and slightly wider than long.Corpus bursae large, roundish, 1 ½ as long as wide, membranous; signum a posteriorly situated, strongly sclerotized plate with distinct longitudinal folds and a group of large thorns arranged in a sinuous row extending from posterior 1/3 into median 1/3 of bursa.Accessory bursa membranous, about 1/3 size of corpus bursae, with a short, wide postero-ventral join.
Molecular data (Fig. 36).BIN URI: BOLD:ADI4764.The intraspecific divergence of the barcode region is 0% (n = 2).The DNA barcoded specimens share the BIN with C. perfusana, but cluster separately as the sister group with a minimum distance of 1.07%.The minimum distance to C. callosana is 3.12%, and to C. bucegiana sp.nov. is 3.62%.Cochylimorpha dorsimaculana seems to be phylogenetically the most recently evolved member of the group.
Biology.In the original description the flight period was given as 27 May to 13 July, and Centaurea rhenana Bor.(= Centaurea stoebe L.) was identified as the presumed host-plant (Preissecker 1908: 72).Two ♂ and one ♀ from Oberloiben in the Wachau area were reared by P. Buchner in 2008.Larvae were found in the flower-heads of Centaurea triumfettii All.(Asteraceae) in mid-April, and pupated in the same place at the end of April .Adults emerged between 8-10 May, which is the earliest recorded date for the species.Given the restricted distribution of C. dorsimaculana and its confirmed host-plant, Centaurea triumfettii, it is doubtful that the widespread Centaurea stoebe is a host-plant.
Habitat (Fig. 39).Larvae have been collected in xerothermic open to semi-open habitats from Centaurea triumfettii, predominantly on south-east to south-west facing slopes at around 200-300 m elevation.These habitats are part of the Wachau, a 35 km long section of the river Danube, where it cuts through the southern tip of the Bohemian Massif.It therefore belongs to the southern edge of the Austrian natural area of granite and gneiss highlands.
Distribution (Fig. 48).Cochylimorpha dorsimaculana was described from the wine-growing area of Wachau and the neighbouring Retz in Lower Austria (Preissecker 1908: 72), and is only known from these two small areas of Lower Austria.All of the collecting sites are at low elevations between 200 and 300 m.The species is considered an endemic of Wachau and Retz in Lower Austria.
We examined material from Lower Austria: Wachau (Dürnstein, Gaisberg -type locality, Spitz an der Donau, Oberloiben) and Retz (Muzion); for details see the examined material above.Further material was identified from figures in the literature or available on the world wide web (all as C. perfusana): the same locations in the Wachau area and in Retz in Lower Austria (Razowski 1970: 162-163, pl. 8 fig. 79-2;2001: pl. 3 fig. 51a;2002: pl. 5 fig. 91a;2009: pl. 3 Razowski (1970: 163, colour pl. 8 fig. 79-2;2001: 37;2002: 43;2009: 37, all as C. perfusana) only records from Wachau and Retz in Lower Austria refer to this species.However, specimens from Hardegg, a similar habitat also in Lower Austria and also given by Razowski, may refer to this species, but we were not able to confirm this.

Remarks.
We were not able to locate the voucher material of Euxanthis dorsimaculana Preiss.recorded from Romania (Transylvania) based on the written communication of J. Kennel given by Galvagni and Preissecker (1913: 49).Razowski's (1970: 163) record from Romania (Transylvania) is based on the previous literature source.In our opinion these records may not refer to this species, but to C. bucegiana sp.nov.
The reasons why Cochylimorpha dorsimaculana (Preissecker, 1908), sp.rev. is elevated to species rank are: 1) the divergence in DNA barcodes: the DNA barcoded specimens of C. dorsimaculana and C. perfusana share the same BIN, but cluster in two sister groups with a minimum distance of 1.07% between them.2) the strikingly different external morphology of the two taxa: C. dorsimaculana has an indistinct light brownish yellow reticulate pattern and the dominant element of the marking is a large, brown dorsal blotch in contrast to a conspicuous light olive-green reticulate pattern uniformly dispersed on the entire forewing of C. perfusana.3) the differences in the genitalia of both sexes.The male C. dorsimaculana has the tegumen with straight margins, a short and pointed terminal process, a rather short, very wide and parallel-sided median process of the transtilla with a round apex and a gradually narrowing valva, whereas C. perfusana has the tegumen with convex margins and an elongated terminal process, a long and wide median process of the transtilla with a tapering apex and a narrower and variable shaped valva.The female C. dorsimaculana possesses signa consisting of both a sclerotized plate and a group of thorns, whereas C. perfusana has a signum which is only a very weakly sclerotized plate.4) the differences in habitats: C. dorsimaculana inhabits meadows at low elevations (200-300 m), in contrast to mountain and high-mountain meadows (between 550-2100 m) inhabited by C. perfusana.
Description.Adult, male (Figs 5,43,44).Head.Frons and vertex covered with yellowish white scales.Labial palpus about 2.5 times diameter of eye, first segment short, second segment long and wide, third segment short, all covered with yellowish white scales on dorsal and medial, yellowish brown on lateral and ventral surfaces.Antenna filiform, ½ forewing length, brown, covered with yellowish white scales dorsally.
Thorax.Dorsally covered with yellowish brown scales, long and erect in posterior 1/3, laterally and tegula covered with light brownish yellow scales.Forewing length of holotype 7.5 mm, paratypes 6-8.5 mm, wingspan 14-19 mm.Forewing short and trapezoidal, gradually widening from base to apex, about 2 2/3 times as long as wide, costal margin slightly convex, without costal fold, apex pointed, termen straight.Ground colour shining yellowish white.Light yellowish brown reticulate pattern forming an indistinct basal blotch and distinct median transverse fascia, a subapical blotch, a weakly delimited subterminal blotch, and slender terminal fascia.Median transverse fascia perpendicular to costal margin, bent and parallel to termen immediately below costa, expanding towards middle of dorsal margin.Dense admixture of brown scales along basal 1/3 of costal margin, on basal blotch, margins of median transverse fascia and subapical blotch.Fringe shining yellowish white.Hindwing without costal roll, brownish grey, fringe shining yellowish white with grey basal line.Underside of thorax and forewing dark brownish grey, indistinct yellowish white line along median 1/3 of subcostal vein, fringe yellowish white, hindwing grey, fringe yellowish white.Legs covered with brownish grey scales on medial and yellowish white on lateral surface.
Abdomen covered with dark grey scales, distal edge of segments and last segment with yellowish white scales.
Variation.Terminal process sometimes truncated; basal half of costa of valva varies from slightly to fairly convex.
Female (Figs 6,45).Forewing length 7.5-8 mm, wingspan 16.5-17 mm.Forewing markings slightly more extended and with a more pronounced admixture of brown scales than in male.Hindwing with 3 bristles in the frenulum.
Female genitalia (Figs 15,34).Papilla analis 2 times as long as segment VIII, wide, roundish, densely covered with medium-long and long setae.Posterior apophysis almost 2 times as long as segment VIII, rod-like, strongly sclerotized, slender.Posterior part of segment VIII covered with strong and long setae, with a row of small setae along anterior margin.Anterior apophysis as long as posterior, rod-like, strongly sclerotized and slender.Antrum as wide as segment VIII.Sterigma with wide and strongly sclerotized lateral sides and weakly sclerotized strengthened medially and densely covered with small microspines.Anteostial sclerite large, distinct and densely covered with microspines.Ostium ½ as wide as segment VIII.Ductus bursae ½ width of ostium, short, membranous.Corpus bursae narrow at posterior 1/4, slightly narrower than half of segment VIII, anterior 3/4 wide, roundish, slightly wider than segment VIII; signum a large sclerotized plate occupying posterior 4/5 of corpus bursae, posteriorly with strongly sclerotized ribbon-like folds, only a small anterior part of corpus bursae membranous.Accessory bursa membranous, longer than corpus bursae, more than 2 times as long as wide, with long and narrow postero-ventral join and about ½ of the size of corpus bursae.
Habitat (Fig. 46).The new species inhabits high-mountain to subalpine mesophilic meadows with straggling Larix decidua Mill.(Pinaceae) trees on south-east facing steep slopes on conglomerate substrate in the Bucegi Mountains and subalpine mesophilic meadows on silicate substrate in the Făgăraș Mountains, at 1500-2200 m altitude, where its presumed host-plant, Centaurea kotschyana Heuff.(Asteraceae) is abundant.
Biology.Poorly known, adults were collected from mid-July to the end of August.
Distribution.Only known from Romania, Southern Carpathians: Bucegi Mountains (Jepii valley -type locality; Piatra Arsă) and the Făgăraș Mountains (environs of the Bâlea Lake) at 1500-2200 m elevation (Fig. 48).Possibly a Southern Carpathian endemic.However, the range of its presumed host-plant, Centaurea kotschyana, is much wider, occurring scattered localities throughout the high-mountain and subalpine zone of the Carpathians, from Poland and Ukraine through the Romanian Eastern and Southern Carpathians, the Apuseni Mountains to the mountains of the Balkan Peninsula (Stara Planina) (Prodan 1930;A. Bartók pers. comm. 2024), suggesting a wider distribution of the moth as well.
Remarks.A record by Kovács and Kovács (2005: 61, 64 fig. 53, 70 fig. 72, 87, 88) from the Bucegi Mountains under the name Cochylimorpha perfusana f. callosana (Herrich-Schäffer, 1851) was based on the above mentioned examined material and refers to this species.We were not able to locate the voucher material of Euxanthis dorsimaculana Preiss.recorded from Romania (Transylvania) based on the written communication of J. Kennel by Galvagni and Preissecker (1913: 49).Razowski's (1970: 163) record from Romania (Transylvania) is based on the previous literature source.In our opinion, these records may not refer to C. dorsimaculana, but to this species.
Etymology.The specific name is a feminine adjective derived from the name of the type locality, the Bucegi Mountains.male genitalia the tegumen has convex margins and a long terminal process; the valva is wide; the median process of the transtilla is long and wide with concave margins and a truncated apex; the caulis has deep ventro-lateral folds; and the phallus is ventrally curved at about 30 degrees with a latero-basally attached and ventrally curved cornutus, about ½ as long as the phallus.In the female genitalia the papillae anales are lanceolate and densely covered with short setae; the apophyses are stout; the ductus bursae is short; and the corpus bursae is long with the signa posteriorly consisting of a large sclerotized plate and in the median part of the bursa with a large group of thorns.
Redescription.Adult, male (Figs 7,47).Head.Frons and vertex covered with straw-yellow scales.Labial palpus about 2.5 times diameter of eye, first segment short, second segment long and wide, third segment short, all segments with straw-yellow scales on medial surface, brownish yellow on lateral surface.Antenna filiform, reaching ½ forewing, brown, covered with straw-yellow scales dorsally.
Thorax.Dorsally covered with straw-yellow scales, long and erect in posterior 1/3.Tegula covered with straw-yellow scales.Forewing length 7-8.5 mm, wingspan 15.5-19 mm.Forewing short and trapezoidal, slightly widening from base to apex, about 2 1/4 times as long as wide, costal margin without costal fold, convex in basal 1/3, remainder straight, apex rounded, termen straight.Shiny yellowish white spots of ground colour larger and irregularly shaped basally, smaller and roundish apically.Reticulate pattern consisting of numerous wide, sinuous, straw-yellow crosslines, often forming a narrow, elongated and indistinct dorsal blotch.Scattered admixture of brown scales forming small indistinct spots along costal margin and middle of dorsal margin, dense and consisting of scattered spots on median and subterminal areas, and only dispersed brown scales along termen and apex.Fringe shiny yellowish white.Hindwing without costal roll, dark grey, fringe white with grey basal line.Underside of thorax and forewing brownish grey, yellow along distal ½ of the costal margin and along basal ½ of subcostal vein, fringe yellowish white, with grey basal line, hindwing yellowish white with admixture of grey scales, yellow along M 2 vein, fringe yellowish white with indistinct grey basal line.Fore-and mid-leg covered with brownish-grey scales on medial and yellowish white on lateral surface, hindleg ochreous white.Abdomen.Covered with grey scales, light grey at distal edge of segments, and last segment with yellowish white scales.
Variation.Colour of markings varying from straw-yellow to light brownish yellow; narrow and indistinct dorsal blotch often light brownish yellow.In fresh specimens straw-yellow reticulate pattern dominates.
Variation.Terminal process of tegumen variable, more or less pointed; median process of transtilla more or less strengthened medially, small thorns at apex arranged in two parallel rows.
Female (Fig. 8).Forewing length 6-7.5 mm, wingspan 14-16.5 mm.Forewing markings more conspicuous, admixture of brown scales less pronounced and ill-defined spots of ground colour smaller than in male.Hindwing with 3 bristles in the frenulum.
Female genitalia (Figs 16,35).Papilla analis 1 ½ as long as segment VIII, narrow, elongated, lanceolate, densely covered with strong, short setae.Posterior apophysis 1 ½ as long as segment VIII, posterior 1/3 wide and weakly sclerotized, anterior 2/3 rod-like, stout and strongly sclerotized.Segment VIII slightly longer than wide, dorsally with 2 rows of strong and long setae along posterior margin, ventrally with a single row along margins.Anterior apophysis slightly shorter than posterior apophysis, entirely rod-like, stout and strongly sclerotized.Sterigma with wide lateral sides and strengthened middle covered with weak microspines.Anteostial sclerite small, covered with weak microspines.Ostium almost as wide as segment VIII.Antrum ½ as long as wide, strongly sclerotized.Ductus bursae ½ as wide as ostium, shorter than wide, membranous.Corpus bursae about 2 times as long as wide, rounded anteriorly, membranous; posterior ½ occupied by signa consisting of a large sclerotized plate posteriorly with weak longitudinal folds; in middle of corpus bursae a large group of tiny thorns.Accessory bursa membranous, almost ½ of size of corpus bursae, with a short and wide postero-lateral join.
Habitat.Dry and stony open hilly meadows and semi-open mountain slopes on limestone substrate, from 350 to 1500 m.
Biology.Poorly known.Adults fly in the daytime (Mann 1854: 576) and evening (Mann 1857: 165).The most recently collected specimens were taken in the evening and early morning by net, usually around places where Centaurea (Asteraceae) species grow, but some specimens were also attracted to light (J.Junnilainen personal observation).Moths are on wing from mid-May to mid-June.
Distribution (Fig. 48).Widespread in the north-western part of the Balkan Peninsula (Croatia and Slovenia) and north-eastern Italy, localized in eastern France (Prémanon), and known only from historic records from Corsica (France) and Hungary.The type locality is Rijeka (formerly Two arguments support our presumption that the name C. callosana might be a commonly used manuscript name before the Herrich-Schäffer's formal description.In the Index of the fourth volume, Herrich-Schäffer (1856: 7) mentions "Mtzn."(= Metzner) as the author of the species name, and the second is that Mann (1854: 576;1855: 552) twice used the name callosana to record a well-defined moth from southern Carniolia and Corsica before the description of the species by Herrich-Schäffer (1856: 157).Cochylis callosana was defined by Herrich-Schäffer (1856: 157) as having a short forewing, straw-yellow coloured markings with only two black spots on the forewing, neither on costal margin nor on termen (remarks: the black spots are the scattered spots formed by the admixture of brown scales).All these characters fit the material examined in this study from Croatia, Slovenia, Italy, France and Hungary.It has also been treated as a valid species, Phalonia callosana HS., by Kennel (1913: 283-284, pl. 12 fig.85).Kennel's description also fits the original one, except for the figure which, as mentioned by Kennel himself, is too light yellow ("Taf.XII, fig.85 ♂ (zu hochgelb)") (Kennel 1913: 283); however, this feature characterizes worn specimens.This species has also been treated by Klimesch (1951: 24) as Euxanthis callosana HS.In the monograph on the Palaearctic Cochylini, Razowski (1970: 38, 162, colour pl. 8, pl. 50, pl. 127) mentions a lectotype deposited in MfN Berlin, but incorrectly gives Schneeberg in Lower Austria as the type locality.Neither the lectotype nor the paralectotype possesses locality data.Razowski confused and synonymized C. callosana with C. perfusana, and gave the reference as: "Cochylis callosana Herrich-Schäffer, 1851, Systematische Bearbeitung der Schmetterlinge von Europa 4: 183", however, this is wrong and in fact it refers to C. perfusana.The correct reference is: Cochylis callosana Herrich-Schäffer, 1856, Systematische Bearbeitung der Schmetterlinge von Europa 6: 157 (different year, volume and page), where Fiume (= Rijeka, Croatia) is specified as the type locality.Razowski figured two male adults, one of C. perfusana and another of C. dorsimaculana (as perfusana f. dorsimaculana), and the genitalia of both sexes of C. perfusana.In his subsequent works Razowski (2001: 13;2002: 43;2009: 37) continues using the incorrect data and taxonomic status, and figures the adult male of C. perfusana, the female of C. dorsimaculana, and the genitalia of both sexes of C. callosana, all under the name C. perfusana.
The reasons why Cochylimorpha callosana (Herrich-Schäffer, 1856), sp.rev. is reinstated to species rank are: 1) the divergence in the DNA barcodes: the DNA barcoded specimens cluster in a separate BIN (BOLD:ADI3050), at a minimum distance of 3.12% to the nearest neighbour C. dorsimaculana, at 3.13% to C. perfusana and 4.43% to C. bucegiana sp.nov.2) the different external morphology: straw-yellow reticulate pattern with scattered admixture of brown scales and grey hindwing characterizes C. callosana, in contrast to the olive-green reticulate pattern without an admixture of brown scales and the light grey hindwing of C. perfusana.3) the clear differences in the structure of the genitalia of both sexes: the male with slightly concave margins and truncated apex of the median process of the transtilla and ventrally curved phallus of C. callosana, in contrast to C. perfusana male with a parallel sided and tapering apex of the median process of the transtilla and straight phallus; and female with a long corpus bursae and signa consisting of a large sclerotized plate posteriorly with weak longitudinal folds and in the middle of bursa a large group of tiny thorns of C. callosana, in contrast to C. perfusana female with a short corpus bursae and the signum being a very weakly sclerotized plate in the posterior 1/3 of the bursa, with distinct longitudinal folds.4) the clear differences in the habitat and distributional pattern: C. callosana inhabits meadows at lower elevations (from 350-1500 m) and is widespread only in the north-western Balkan Peninsula and north-eastern Italy, and is local in eastern France, Corsica and Hungary, in contrast to the montane meadows (between 550-2100 m) of the mountain ranges inhabited by the even more widely distributed C. perfusana, from the Carpathians and the Pirin Mountains through the Austrian Alps and Switzerland to the Hautes Alpes in France.

Discussion
Small unintended erroneous statements can lead to subsequent longstanding taxonomic errors.Such taxonomic or nomenclatural problems spanning generations are not rare, indeed frequent for Cochylini.One problem regarding Longicornutia epilinana (Duponchel, 1842) has already been discussed by Kovács and Kovács (2020: 7).
The case of the Cochylimorpha perfusana species group is also problematic.Treating two related species in two different volumes, Cochylis perfusana in the fourth volume of his Systematische Bearbeitung der Schmetterlinge von Europa (Herrich-Schäffer 1851: 183) and C. callosana in the sixth volume (Herrich-Schäffer 1856: 157), but in the index giving for both the same cross-reference, that referring to perfusana, Herrich-Schäffer (1856: Index 7, 32) created a misleading situation.This confusing statement was probably the starting-point which led to the subsequent confusion of the two taxa, to the incorrect reference of the description of C. callosana, to its wrong type locality and also to the subsequent synonymy of three related taxa (Razowski 1970: 38-39, 162).The latter act, the consequence of the original mistake, has resulted in an incorrect species concept, which had been accepted for more than 50 years.
Guided by the data in Razowski (1970: 162) referring to the depository institution of the "holotype" of Argyrolepia perfusana Guenée, 1845, the specimen labelled as "Type Guenée" has been located in the NHMUK, and the photograph of the adult examined through the courtesy of D.C. Lees, R.J. Heckford and S.D. Beavan.The fact that the species was described based on an unspecified number of specimens, and a large distribution area given as "In summis montibus Austriae et in alpibus Delphinatus" (Guenée 1845: 302), suggests that the description was not based on a single specimen.As a consequence, the type material might consist of a number of syntypes.However, in the NHMUK there are no other historic specimens that support this hypothesis (D.C.Lees pers.comm.).Further studies may clarify this possibility and identify the putative syntypes.For the time being, in accordance with the ICZN article 74.6, Razowski's (1970: 162) treatment of the type specimen as the holotype must, instead, be considered a lectotype designation.Razowski (1970: 162) and in his subsequent works (Razowski 2002: 43;2009: 37) erroneously cites "HERRICH-SCHÄFFER, 1851, Syst.Bearb.Schmet.Eur.4: 183" for the description of callosana, also the type locality is incorrectly specified as "Niederösterreich: Schneeberg".These data in fact refer to C. perfusana.The correct year of Herrich-Schäffer's description of callosana is 1856, the volume 6, the page 157 and the type locality Fiume (= Rijeka) in Croatia.
DNA sequencing has an important role in the taxonomy of the 21 st century.DNA barcoding, in particular, has become the most important tool in uncovering unjustified synonymy and cryptic diversity.Several cases were previously remedied within Lepidoptera including a few in the European Cochylini (Mutanen et al. 2012;Zlatkov and Huemer 2017;Corley and Ferreira 2020;Kovács et al. 2020).The C. perfusana species group is now a further example.Molecular studies can also reveal the relationships among the taxa of such species groups, which in this case fully supports the close relationship of the four species that form the informal C. perfusana species group.At the time of their descriptions, the authors considered them closely related based only on the similar external morphology.The diversity within the group was questioned during the study of the genitalia which, being very similar especially in the males, led to the synonymy of the earlier known three taxa.Finally, owing to the molecular studies, we were able to reinstate all of the previously named taxa to their original specific rank and to discover and describe one more cryptic species in this species group.Furthermore, the availability now of DNA barcodes for all four species of the C. perfusana species group makes possible COI based identification.Cochylimorpha dorsimaculana shares a BIN with C. perfusana, but clusters separately as its sister group with a minimum distance of 1.07%, whilst the minimum distances to the other two species exceed 3%.However, the strikingly different external morphology and the differences in the genitalia structure of both sexes compared to all of the other three species necessitated its treatment as a separate species.The small minimum inter-specific distance indicates that phylogenetically it is the most recently evolved member of the group.
Similarity in the genitalia of closely related species seems to characterize Cochylini.This is already the fourth species group of related taxa earlier synonymized mainly following the study of the genitalia and recently taken out of synonymy after their DNA sequencing.In three of them, the species pairs Phalonidia manniana (Fischer von Röslerstamm, 1839)/P.udana (Guenée, 1845) (Mutanen et al. 2012) and Cochylimorpha discopunctana (Eversmann, 1844)/C.punctiferana (Ragonot, 1881) (Corley and Ferreira 2020), and the C. perfusana species group studied here, the careful re-examination of the external morphology and the genitalia allowed the clear differentiation of the taxa.However, in the fourth grouping, the Phtheochroa frigidana (Guenée, 1845) species group (Zlatkov and Huemer 2017;Kovács et al. 2020), the use of special techniques, i.e., in the male, the eversion and study of the three-dimensional structure of the vesica, was necessary to reveal the differences in the otherwise similar male genitalia, and in the female, the discovery of a previously ignored membranous structure of ductus bursae, the ventral diverticulum, together with other small differences, allowed the morphological differentiation of the taxa.

Figure 17 .
Figure 17.Argyrolepia perfusana Guenée, 1845, lectotype ♀ and its labels, photograph and courtesy of David C. Lees and with the kind permission of The Trustees of NHMUK.