Allopatric cryptic diversity in the alpine species complex Phtheochroa frigidana s. lat. (Lepidoptera: Tortricidae)

Allopatric alpine populations of Phtheochroa frigidana s. lat. (Lepidoptera, Tortricidae) are reviewed. In addition to traditional diagnostic characters of external morphology, the genitalia structures of everted vesicae in male genitalia and DNA barcodes are analysed. This new approach supports the existence of fi ve rather than two species in Europe: Phtheochroa schawerdae (Rebel, 1908) comb. nov. (Dinaric Mts, Rila Mts, Pirin Mts) = P. drenowskyi (Rebel, 1916) syn. nov.; P. alpinana sp. nov. (SW Alps); P. apenninana sp. nov. (Apennines); P. frigidana (Guenée, 1845) stat. rev. (Pyrenees) = P. fl avidana (Guenée, 1845) = P. sulphurana (Guenée, 1845) = P. andorrana (Millière, 1865); P. cantabriana sp. nov. (Cantabrian Mts). In order to stabilize the nomenclature, a neotype for Eupoecilia frigidana is designated.

Widespread though disjunct species such as Phtheochroa frigidana (Guenée, 1845) stat.rev.are subject to isolation in their alpine environment, resulting in interrupted gene-fl ow.They therefore deserve particular attention as model groups of potential speciation processes.Literature sources suggest that P. frigidana is distributed at high altitudes in the Apennines, the south-western Alps, the Pyrenees and the Dinaric Mts.The closely related sister species P. drenowskyi (Rebel, 1916) has been reported from the Apennines, the Dinaric Mts, the Carpathian Mts and the Rila Mts (Razowski 2009;Trematerra 2003;Kovács & Kovács 2004), thus in putative sympatry with P. frigidana.However, initial DNA barcoding of specimens from the Cantabrian Mts and from the Pyrenees to the Balkan Peninsula indicated quite a different taxonomy.
We therefore revise the group, introducing for the fi rst time new morphological techniques for species delimitation, particularly eversion of the vesica in the male genitalia and analysis of the COI standard barcode region of the mitochondrial DNA.

Material and methods
We examined 38 specimens of the P. frigidana species group.The moths collected by the authors were pinned and partially spread or prepared following standard techniques.The genitalia were processed following the methodology of Robinson (1976) and Zlatkov (2011).The phalli with everted vesicae were submerged in Euparal essence on a cavity slide, immobilized with a curved minute needle inserted through the phallic opening, and drawn under a compound microscope with camera lucida attached.The phalli were observed and drawn from two aspects: dorsal and left.Only completely everted and maximally infl ated vesicae were used.The angles of the diverticula were measured from drawings and photographs with a protractor.The description of cornuti follows the terminology of Anzaldo et al. 2014; wing pattern elements are those defi ned by Razowski (2008).
DNA barcode sequences of the mitochondrial COI gene (cytochrome c oxidase 1) were obtained from 11 specimens.DNA samples from dried legs were prepared according to prescribed standards using the high-throughput protocol of deWaard et al. (2008).Samples were processed at the Canadian Centre for DNA Barcoding (CCDB, Biodiversity Institute of Ontario, University of Guelph) to obtain DNA barcodes (Ratnasingham & Hebert 2007).DNA sequencing resulted in a full barcode fragment of 658 bp for eight specimens and a further three sequences with 632, 628 and 612 bp.Details of successfully sequenced voucher specimens, including complete voucher data and images, can be accessed in the Barcode of Life Data Systems (Ratnasingham & Hebert 2007).Degrees of intra-and interspecifi c variation in the DNA barcode fragments were calculated under the Kimura 2 parameter (K2P) model of nucleotide substitution using analytical tools in BOLD systems v. 3.0 (http://www.boldsystems.org).A maximum likelihood (ML) tree of the DNA barcode data was constructed using MEGA 6 (Tamura et al. 2013)

Diagnosis
Externally, this species resembles P. apenninana sp.nov., with a deep yellow forewing upperside and a dark grey hindwing.The most conspicuous difference can be seen in the vesica: the axes of the right diverticulum and the sclerotized phallus form an angle of 100-120° in P. schawerdae and a maximum of 90° in P. apenninana sp.nov.; the right part of the vesica is larger in P. apenninana sp.nov.Other characters of less importance are the shape of the valva (wider in P. schawerdae, more slender in P. apenninana sp.nov.) and transtilla (short, with wide base in P. apenninana sp.nov.but variable in P. schawerdae).The ventral phallic process is strongly curved to the right, in contrast to all other species of the group.The upperside wing colour easily distinguishes P. schawerdae from P. alpinana sp.nov., P. frigidana and P. cantabriana sp.nov.
ABDOMEN.Grey with scattered white scales.GENITALIA (Fig. 2A-B).Valva with weak dorsal curvature, round apically, with parallel costa and sacculus, both moderately sclerotized.Valva may appear wider or narrower depending on preparation.Uncus relatively well sclerotized, tapered apically.Socii rounded, convergent and connected medially with membrane.Transtilla (Fig. 3A-B) dorsally spinulous, variably shaped in specimens from same population: apically rounded or relatively angled, wider or narrower.Phallus (Fig. 5) slightly curved ventrally, medioventrally with large, slender, sharp-tipped process (Fig. 4A-C), with apex curved to the right at nearly right angle.Vesica with large main part protruded at right from sclerotized phallus and two large diverticula bearing long, robust cornuti.Small conical, undulating structure (diverticulum) emerges ventrally from main part.Gonopore located dorsally, surrounded by semicylindrical sclerotized plate with minute lentiform tubercles.Main part medially constricted by circular furrow in which gonopore is sunk.Diverticula emerge close to each other from right side, right one straight and directed  Female (Fig. 1C) Generally resembles male, but forewing length 7.3-9.2mm (n = 2), underside with larger creamy area, hindwing underside creamy or whitish.Female genitalia (Fig. 8A) with tergum 8 medially membranous, sterigma with two lateral protuberances, antrum nearly rectangular, with round anterior angles.Apophyses anteriores longer than apophyses posteriores.Ductus bursae ventrally membranous, with conical diverticulum of thick cuticle curved to the left and triangular sclerite dorsally.Corpus bursae with large sclerite on ventral side forming several (5-6) folds at the right, then expanding medially and dorsally on left side, reaching dorsal area of ductus bursae.Right side of corpus bursae with extensively folded membranous area.Ductus of accessory bursa emerges medioventrally, folded cuticle and small spines present to the left of its emerging area.Ductus seminalis inserted on ventral side.1, Fig. 9) BIN URI: BOLD:ABA3262.The maximum intraspecifi c divergence of the barcode region is considerable with 1.59% (n = 4), supporting two geographically separated clusters in the Dinaric Mts. and Rila Mts.The minimum distance to the nearest neighbour, P. apenninana sp.nov. is 3.69%.

Ecology
The moths fl y in July-August at daytime, in the morning and late afternoon until twilight, in sunny, calm weather.Preimaginal stages and larval host plant are unknown.Habitat: humid subalpine meadows at altitudes from 2000 to 2600 m a.s.l.Distribution (Fig. 10) Known from the highest Balkan mountain systems: Dinaric Mts, Rila Mts, Pirin Mts, and most probably Carpathian Mts (material not checked by us).

Diagnosis
The forewing upperside is paler than in P. schawerdae and P. apenninana sp.nov.and somewhat deeper yellow than in P. frigidana and P. cantabriana sp.nov.The hindwings are pale grey as in P. frigidana and P. cantabriana sp.nov.The vesica easily distinguishes this species from all other members of the group: with large dorsal and ventral diverticula; P. frigidana has only one diverticulum (dorsal); the diverticula on the other three species are oriented more or less laterally.The shape of the valva resembles those of P. frigidana.The phallic process is slightly curved to the right and resembles that of P. cantabriana sp.nov.The female genitalia are generally similar to these of P. schawerdae, but the sclerotized area of the corpus bursae is larger, with three long folds on the right ventral side; the folds in P. schawerdae are short and more numerous (5-6).

Etymology
The specifi c name is a feminine adjective derived from the name of the Alps.rust dorsal blotch, underside dark grey with scattered white scales in apical area, cilia pale yellow.Hindwings with upperside grey, underside grey with white costa and white longitudinal lines, cilia white with pale grey basal part.
GENITALIA (Fig. 2C).Curvature of valva prominent.Rest of valva, uncus and socii resembles those in P. schawerdae.Transtilla (Fig. 3C) wide, rectangular apically, with shallow median incision.Phallus (Fig. 6A-B) curved ventrally, with long, slender medioventral process (Fig. 4D-E), with apex slightly curved to the right.Vesica with subspherical proximal part and large part protruded from the right, bearing two large diverticula ending with cornuti.A furrow divides the two parts.Gonopore sunk in dorsal portion of furrow and surrounded by large sclerotized plate with uneven surface (distal part of ductus ejaculatorius).Right part of vesica protrudes in wider, conical, almost straight dorsal and narrower, cylindrical, curved ventral diverticulum.Cornuti aciculate and capitate, very robust, large, ventral one larger and slightly curved near apex.Proximal part of vesica bears a large conical ventral protrusion (diverticulum) with small processes.Acanthae present on posterior side of proximal part of vesica and at end of diverticula.
Female (Fig. 1F) Considerably smaller than male.Head and thorax resemble those of male, but with forewing length 7.2-9.2mm (x = 7.9, n = 5), upperside with rust brown dorsal blotch and subterminal fascia.Hindwing upperside grey, underside white with scattered grey scales.Abdomen grey.Female genitalia (Fig. 8B) with tergum 8 medially membranous, sterigma with two lateral protuberances, antrum nearly rectangular, with round anterior angles.Apophyses anteriores longer than apophyses posteriores.Ductus bursae ventrally membranous, with conical diverticulum of thick cuticle slightly curved to the left.Corpus bursae with large sclerite that starts on right side of ductus bursae with folds, then expands anteroventrally forming large ventral sclerotization and laterally to the left, then passes on dorsal side where it forms several folds on ductus bursae.Anterior medial part of corpus bursae with folded membranous area extending to the right.Ductus of accessory bursa emerges medioventrally, folded cuticle and small spines present to the left of its emerging area.Ductus seminalis inserted on ventral side.
Molecular data (Table 1, Fig. 9) BIN URI: BOLD:AAL5381.The intraspecifi c divergence of the barcode region is unknown (n = 1).The minimum distance to the nearest neighbour, P. schawerdae from the Dinaric Mts is 4.01%.

Ecology
Preimaginal stages and larval host plant are unknown.The type material was collected in July at altitudes of about 2200 m a.s.l.According to Bassi & Scaramozzino (1989) the moths fl y in two generations from April to July.However, material collected by these authors in the Italian Alps dates from mid-to late May and was collected at relatively low altitudes from 1200 to 1350 m a.s.l.From available data it thus seems more likely that the species is univoltine with a fl ight period from May to July, depending on climatic conditions and elevation.Habitat: the type material was collected in alpine grassland above the tree-line.
Distribution (Fig. 10) So far known only from the south-western Alps (France, Italy).A record from the Slovenian Alps (as P. frigidana) (Lesar et al. 2009) is doubtful and may refer to P. schawerdae.

Etymology
The specifi c name is a feminine adjective derived from the name of the Apennines.
THORAX.Dorsally, including tegulae, with yellow and brown scales, ventrally whitish, fore-and middle legs brown, hindlegs white.Forewing length 7.2-7.4mm (n = 2), upperside deep yellow, with remnants of rust dorsal blotch, cilia concolorous with upperside, underside dark grey, with white scales in costal area.Hindwing upper-and underside dark grey, with white costal area and white longitudinal line between veins M 1 and M 2 starting from discal cell and ending in cilia. ABDOMEN.Grey.
GENITALIA (Fig. 2D).Valva with weak curvature, appears slender but this may be due to preparation.Rest of valva, uncus and socii very similar to those of P. schawerdae.Transtilla (Fig. 3D) large, round.Phallus (Fig. 6C-D) slightly curved ventrally, with wide and blunt medioventral process (Fig. 4F-G), with apex curved to the right.Vesica in general resembles that of P. schawerdae, with large main part protruded at right from sclerotized phallus and two large diverticula bearing long, robust cornuti.Small, conical undulating structure (diverticulum) emerges ventrally from main part.Gonopore located dorsally, surrounded by semicylindrical sclerotized plate with minute, lentiform tubercles.Main part medially constricted by circular furrow in which gonopore is sunk.Portion located distally of furrow very large, heart-shaped in dorsal view.Main part of vesica looks ovoid in lateral left view.Diverticula emerge from right portion in opposing directions, right one straight, emerging from ventral half of right part and directed laterally, left one curved and also pointed laterally.Length and orientation of diverticula relatively constant.Axes of right diverticulum and sclerotized phallus form an angle of maximum 90°.Cornuti aciculate, capitate, of equal length, left one slightly bent.Barely discernible acanthae present on left portion of main part of vesica and at end of diverticula.

Female
Unknown.

Ecology
Preimaginal stages and larval host plant are unknown.The moths fl y in July.Habitat: the type material was collected in alpine grassland above the treeline.
THORAX.Dorsally covered with white scales, ventrally with whitish scales.Tegula whitish, laterally with ochreous scales.Forewing length 8.8-10.2mm (x = 9.6, n = 4).Forewing long, narrow, with pointed apex, upperside ground colour creamy, more intense yellowish at base and along proximal half of costa; no pattern recognizable in studied specimens, but a specimen fi gured in Millière ( 1865) has rust dorsal blotch and subterminal fascia.Underside dark grey, with scattered white scales, especially in apical area.Cilia whitish, with darker basal line.Hindwing upperside pale grey, underside white, with scattered grey scales.Cilia white. ABDOMEN.Grey.
GENITALIA (Fig. 2E).Curvature of valva prominent.Rest of valva, uncus and socii resembles those of P. schawerdae.Transtilla of neotype (Fig. 3E) short and wide at base, narrow and round apically, but in other specimens with angular apical part.Phallus (Fig. 7A-B) almost straight, with very large, straight, slender, blunt medioventral process (Fig. 4H).Vesica with ovoid part protruded to the right bearing anteroventrally directed cornutus.Long, slender, bent at basal third, anterodorsally directed diverticulum with cornutus emerges from the left dorsal part of vesica.Gonopore located dorsally, at base of diverticulum, with sclerotized plate and collar-like fold expanding laterally.Cornuti unequal, ventral one shorter, thicker and curved, dorsal one long, slender, sharp-tipped and only slightly curved.
Acanthae present in small area on dorsal part, near gonopore, and scarcely around attachment areas of cornuti.

Female
A drawing by Razowski (1970: pl. 112, fi g. 3) may represent the female genitalia of a specimen of this species from the Pyrenees.

Ecology
Preimaginal stages and larval host plant are unknown.According to the label data, the moths fl y from the middle of July to the beginning of August.Guenée (1846) reported May and June.The habitat of the neotype was alpine meadows above the tree-line, but based on supplementary material this species also occurs at lower altitudes.

Remarks
The erroneous type locality of "Dalecarlia" in the original description of P. frigidana is a major hindrance in the interpretation of this taxon.Therefore, in order to maintain the stability of the nomenclature, we have designated a neotype for Eupoecilia frigidana in accordance with past usage of this name for populations from the Pyrenees (Razowski 1970), with Cochylis fl avidana and C. andorrana as junior synonyms (Razowski 2002).
Phtheochroa cantabriana sp.nov.urn:lsid:zoobank.org:act:AB5396E9-E6BF-487D-8CC3-D5E7F2B1674EFigs 1I-J, 2F, 3F, 7C-D, 9, 10; Table 1 Diagnosis Externally, this species is practically indistinguishable from P. frigidana; the same is valid for most of the genital characters, but the vesica is strikingly different.The general shape of the vesica resembles that of P. schawerdae and P. apenninana sp.nov.but in lateral left view its main part looks angular (not ovoid), and the angle between the axes of the phallus and right diverticulum (in dorsal view) is usually larger (more than 130°).The wing upperside colouration of P. cantabriana sp.nov. is much paler than that of P. schawerdae and P. apenninana sp.nov.

Etymology
The specifi c name is a feminine adjective derived from the name of the Cantabrian Mountains.
THORAX.Dorsally with white and grey scales, tegula the same but laterally rust, ventrally whitish, foreand middle legs brownish, hind legs whitish.Forewing long and narrow with pointed apex, length 8.8-10.4mm (x = 9.6 mm, n = 4), upperside pale yellow with rust base, sometimes with rust dorsal blotch and subterminal fascia, cilia white, underside pale grey, with white scales in costal and anal areas.
Hindwing upperside pale grey, cilia white with pale grey basal part, underside white with pale grey scales along veins R and M. ABDOMEN.Grey.

Female
Unknown.

Ecology
Preimaginal stages and larval host plant are unknown.The moths fl y in June-July.Habitat: the type material was collected in montane meadows.

Discussion
Eupoecilia frigidana was the fi rst taxon described from this group (Guenée 1845a).The origin of the type material is uncertain; according to the original description the type locality is "Dalecarl.[ia]",which is a region in Sweden, but this taxon has never been reported outside southern Europe.Most probably, the type material originates from the Pyrenees (Razowski 1970) and we follow this interpretation in our selection of a neotype (see above).Guenée (1845b) described two further species from the Pyrenees, belonging to this group, Cochylis fl avidana and Aphelia sulphurana, which were subsequently synonymized with P. frigidana (Ragonot 1894).Conchylis andorrana Millière, 1865, another synonym of P. frigidana, also originates from the Pyrenees (Ariège Mts).The type specimens of all these taxa are most likely lost (Gilligan et al. 2014), leaving only speculation as to their real taxonomical assignment.
The other available names from this species group originate from material collected on the Balkan Peninsula.Conchylis schawerdae Rebel, 1908 was described from the Dinaric Mts in Bosnia and later synonymized with P. frigidana (Razowski 1970), an action based, presumably, on the original description, since the type specimens of C. schawerdae are lost.Euxanthis drenowskyi Rebel, 1916 was described after material from the Rila Mts in Bulgaria and is considered as valid species in recent literature.It is the only species from the group with preserved type specimens.
In summary, two valid names exist in the present-day literature for taxa of the Phtheochroa frigidana species group: P. frigidana and P. drenowskyi.However, our molecular and morphological comparison of specimens from all main southern European mountain ranges revealed much greater diversity than previously recognized.Five geographically separated putative species can be recognized from the Cantabrian Mts, Pyrenees, SW Alps, Apennines and the Balkans (Fig. 10).These allopatric species are supported by subtle morphological characters of diagnostic value and considerable divergence of barcode sequences.It should be emphasized that the main morphological differences concern the threedimensional structure of the vesica, a character never studied before in this group.Some differences can be detected in the phallic process (Fig. 4), but in general, the rest of the male genitalia do not provide reliable diagnostic characters in this group (Figs 2-3).The shape of the uncus, socii, transtilla and valvae are commonly used characters in the taxonomy of Phtheochroa, but there is considerable intraspecifi c variation and these features are not of great signifi cance in this particular group.Overreliance on these structures has been misleading taxonomists for a long time, producing unnecessary synonymy or omitting well-defi ned taxa.  . unionana (Kennel, 1900), than to those of the other two species in the frigidana group, P. frigidana and P. alpinana sp.nov.(Zlatkov & Huemer 2016).These facts support the assumption that the male genitalia, especially the "internal" (phallus and vesica) ones, in many cases diverge very rapidly and cannot provide a reliable phylogenetic signal.An intriguing structure presented in all the above-cited species of Phtheochroa, with the exception of P. frigidana, is a small conical ventral diverticulum of the vesica devoid of cornuti and acanthae.It is intensively stained by Chlorazol Black, has an uneven surface, sometimes small processes and folds, and is variably shaped among members of the same species, in contrast to the other large diverticula which have a relatively constant shape.Its function is entirely unknown, but one can speculate that it is an internal, non-eversible (in the living moth) structure, probably an apodeme for intrinsic phallic muscles rather than a real diverticulum.
The female genitalia of only two species of the group are known to us: P. schawerdae and P. alpinana sp.nov.(Fig. 8); those of probable P. frigidana were fi gured by Razowski (1970).The illustrations available from the literature do not correctly represent all characters.An important character not presented in the existing drawings is the presence of a ventral diverticulum of thick, colourless cuticle of the ductus bursae with its tip curved to the left, most probably serving to accommodate the ventral phallic process of the male.In general, the female genitalia of both studied species demonstrate considerable similarity, although some differences can be noticed in the shape of sclerotizations of the ductus and corpus bursae.Comparison of the vesicae demonstrates that those of the Pyrenean specimens (Fig. 7A-B) are strikingly different from all other studied specimens.We consider the Pyrenean population as the true P. frigidana, because the original description was likely based on specimens from the Pyrenees (Razowski 1970).Also in this regard, the earlier synonymy of P. fl avidana, P. andorrana and P. sulphurana with P. frigidana seems justifi ed.The male genitalia of this species were probably illustrated by Razowski (1970: pl. 29, fi g. 3).The structure of the vesica is not visible from these drawings, but the shapes of the transtilla and phallus do fi t specimens from the Pyrenees studied by us.The vesica of the Alpine specimens is considerably different from those of the Pyrenean specimens and also differs from that of specimens collected in the other mountain massifs; therefore, we consider the southern Alpine population as a separate, undescribed taxon, a decision which is supported by considerable divergence of the barcode.Additionally, the genitalia of a female from the same locality in the Alps was compared with the drawing of "frigidana" genitalia of a specimen from the Pyrenees (Razowski 1970); differences in several characters can easily be noticed, which supports the contention that the Pyrenean and Alpine populations are separate species.The Cantabrian, Apenninian and Balkan populations demonstrate a similar structure of the vesicae, but can easily be separated by some details in the position of the diverticula and shape of the phallic process.The Cantabrian and Apenninian populations appear to be undescribed taxa.Two nominal taxa are known from the Balkans: P. schawerdae and P. drenowskyi.
A specimen originating from the Dinaric system (Korab Mts, Macedonia) and presumably representing P. schawerdae (described fromVucija Bara, Bosnia) was examined.However, a morphological comparison with specimens of P. drenowskyi from the Rila Mts (Bulgaria) did not provide any reliable differences between these two taxa; thus, they appear conspecifi c, despite a considerable intraspecifi c barcode divergence.In the absence of specifi c morphological characters, P. drenowskyi should be treated as a junior synonym of P. schawerdae.The distribution range of P. schawerdae probably covers all higher Balkan mountains (above 2000 m); it is reported and illustrated also from the Carpathian Mountains in Romania (Kovács & Kovács 2004).The fi gures of male and female genitalia in this paper fi t specimens from the Rila Mts, but the comparison is not conclusive, because the vesica is not everted.
These diagnostic morphological characters are supported by distinct barcode gaps for all putative taxa and are concordant to unique BINs (Barcode Index Number) in BOLD (Ratnasingham & Hebert 2013).
The intraspecifi c divergence is low, as far as is known from the relatively few samples, with the single exception being the allopatric populations of P. schawerdae in the Dinaric and Rila Mts.Conversely, the interspecifi c divergence varies from 3.69% to 6.11% (Table 1, Fig. 9).Similar or even lower values have been supportive for the delimitation of cryptic species of Lepidoptera in various groups (e.g., Pazhenkova et al. 2015).
Species delimitation of allopatric taxa is a critical task, as the biological species concept applying reproductive incompatibility cannot be proved under natural conditions.In such species the phylogenetic concept (Cracraft 1989) may be adopted and species concepts are rather based on diagnosability.To avoid artifi cial oversplitting, Mutanen et al. (2012) suggested species status for allopatric populations with a divergence in two or more independent characters.We follow this compromise and reserve species status for fi ve allopatric populations, supported by both morphological characters and DNA barcode divergence.
The evolution of several alpine species is likely to be explained by climatic changes during the Pleistocene (Varga & Schmitt 2008).Cyclic changes in temperature caused up-and down-range shifts and the disruption of formerly connected populations.The timing of alleged speciation processes is unknown, though molecular clock estimations of the COI gene suggest average substitution rates of 1.0 to 2.3% sequence divergence per million years in Lepidoptera (Kandul et al. 2004).However, the range of interspecifi c barcode variation in P. frigidana s. lat.with a minimum 3.69%, at least indicates that gene fl ow was already interrupted by the early pleistocenic glaciation periods, commencing about 2.58 Mya.
) curved ventrally, medioventral process slender, pointed, with apex slightly curved to the right.Vesica in general resembles that of P. schawerdae, with large main part protruded at right from sclerotized phallus and two large diverticula bearing long, robust cornuti.Small, conical undulating structure (diverticulum) emerges ventrally from main part.Gonopore located dorsally, surrounded by semicylindrical sclerotized plate with minute tubercles.Main part medially constricted by circular furrow, in which gonopore is sunk.Main part of vesica looks angular in lateral left view.Diverticula emerge close to each other from right side, right one straight, emerging from dorsal half of right part and pointed posterolaterally, left one curved and oriented laterally.In dorsal view, axes of right diverticulum and sclerotized phallus usually form angle of 120-140°.Cornuti aciculate, capitate, of equal length, left one slightly bent.Barely discernible acanthae present on left portion of main part of vesica and at end of diverticula.

Fig. 9 .
Fig. 9. Maximum likelihood tree (built with MEGA6) of cytochrome c oxidase subunit I (COI) barcode fragments.Values at the nodes are bootstrap support values based on 500 replicates.
DiagnosisExternally, this species is practically indistinguishable from P. cantabriana sp.nov.The characters of the phallus and vesica easily separate this species from all other members of the group: the phallic process is straight, in contrast to the other species with a curved process; the vesica has a single long dorsal diverticulum, the other species have two diverticula.The paler fore-and hindwing upperside colour distinguishes P. frigidana from P. schawerdae and P. apenninana sp.nov., both with deep yellow forewings and dark grey hindwings.08349"[greenlabel]// Gen. prep.♂1/16.7.2012 [genitalia slide number] // NEOTYPE Eupoecilia frigidanaGuenée, 1845 Zlatkov & Huemer, 2017des.[red label] // (TLMF).
The shapes of the vesicae and phalli in P. schawerdae, P. apenninana sp.nov.andP.cantabriana sp.nov.demonstrateconsiderablesimilarity; the most obvious difference between them is the angle formed by the axes of the phallus and right diverticulum (Figs5-7).This is due not only to the mutual orientation of these structures, but also to the different development of other parts, which can be recognized only after careful morphological examination.It is interesting to note that the most isolated taxa, P. schawerdae and P. cantabriana sp.nov., differ in relatively few characters.At the same time, the differences in vesicae in P. apenninana sp.nov.and P. schawerdae are more prominent despite their smaller genetic and geographic distances.The phallus and vesica of P. frigidana are strikingly different from those of all other species in the group.Genetically, the nearest species to P. frigidana is P. cantabriana sp.nov., which in no way is supported by the genital morphology.It is interesting to note that the vesicae of P. schawerdae, P. apenninana sp.nov.and P. cantabriana sp.nov.are more similar to those of species belonging to a relatively distant group of Phtheochroa, P. procerana (Lederer, 1863) and P