Revealing the identity of Mordellistena minima and M. pseudorhenana (Coleoptera: Mordellidae) based on re-examined type material and DNA barcodes, with new distributional records and comments on morphological variability

Abstract The current interpretation of two common European species, Mordellistena minima Costa, 1854 and M. pseudorhenana Ermisch, 1977, is based on misidentification. The confusion regarding the identity of the species is fixed based on the revised type material. Here, the species are redescribed, and diagnostic characters are provided. Mordellistena pseudorhenana is revalidated. Mordellistena emeryi Schilsky, 1895 is recognised as a new synonym of M. minima. Mordellistena sajoi Ermisch, 1977 is recognised as a new synonym of M. pseudorhenana. Lectotype and paralectotypes of M. emeryi are designated. Mordellistena pseudorhenana is reported for the first time from Bosnia and Herzegovina, Slovenia, and Switzerland. Two morphotypes of M. pseudorhenana differing in size and shape of the parameres are recognised. Morphological differences are quantified and displayed using principal component analysis. In addition, DNA barcodes have been used for the first time in family Mordellidae to examine the divergences between the species and to interpret the morphological variability observed in M. pseudorhenana. Low genetic divergences did not provide the evidence for considering the morphotypes as separate species. The discrepancy between the morphological and molecular evidence raises questions about the efficiency of the CO1 gene for Mordellidae identification and the stability of morphological traits conventionally used for species separation.


Introduction
The genus Mordellistena Costa, 1854 (Coleoptera: Mordellidae) is represented in Europe by approximately 170 species (Horák 2008;Odnosum 2009;Selnekovič and Kodada 2019;Selnekovič and Ruzzier 2019;Selnekovič and Improta 2020). Most of the common and widespread European species were described during the 19th century by Costa (1854), Mulsant (1856), Emery (1876), and Schilsky (1894Schilsky ( , 1895Schilsky ( , 1898Schilsky ( , 1899. Their work was later followed up by specialists and prolific authors such as K. Ermisch, M.E. Franciscolo, and R. Batten, who greatly contributed to the knowledge of the family with descriptions of dozens of new species. Unfortunately, during our recent studies, it became clear that the type material of some previously described taxa remained unstudied, leading to several cases of incorrect species interpretations and descriptions of taxa that already bore a name (Horák 1990(Horák , 1996Selnekovič and Kodada 2019;Selnekovič and Improta 2020). Mordellistena minima Costa, 1854 (Fig. 1A) and M. pseudorhenana Ermisch, 1977 (Fig. 1B) discussed in the present paper may serve as examples.
Mordellistena minima was described by Costa (1854) based on a specimen from the island of Ischia, Italy. Later, Emery (1876) considered the type specimen of M. minima "just a small specimen of M. micans (Germar, 1817), which varies greatly in size". His opinion was then followed by all subsequent authors until Ermisch (1954) treated M. minima as a valid species but did not provide any description or diagnostic characters to separate it from its allies. Batten (1977), without seeing the type specimen, characterised M. minima based on a unique combination of characters: short antennomeres, long and pointed galea, and expanded protibiae in males. Subsequently, Batten (1980) examined the holotype of M. pseudorhenana Ermisch, 1977 and considered it to be conspecific with M. minima.
The re-examination of the type specimen of M. minima surprisingly revealed a unique set of characters that differ significantly from the abovementioned and currently accepted interpretation of the species as presented by Batten (1977). The present paper aims to resolve the confusion regarding the identity of M. minima and M. pseudorhenana and to provide redescriptions of both species based on the examined type material. We integrated morphometric and DNA barcode analyses to interpret the observed morphological variability in specimens of M. pseudorhenana. Furthermore, we have been able to add DNA barcodes for the first time to five species of the Mordellistena confinis species group, with recently re-examined and documented type material (Horák 1996;Selnekovič and Kodada 2019;Selnekovič and Improta 2020). This allowed us to examine the interspecific genetic divergences at the species-group level and set the baseline for future studies with the use of DNA markers.

Materials and methods
The present study is based on examination of 242 adult specimens, including a lectotype of Mordellistena minima Costa, 1854, a lectotype and paralectotypes of M. emeryi Schilsky, 1895, two syntypes of M. micans (Germar, 1813), a holotype and paratypes of M. pseudorhenana Ermisch, 1977, and a holotype of M. sajoi Ermisch, 1977. Freshly collected specimens used for the morphological observations were killed using ethylacetate, dissected, and glued on a cardboard mounting card. Specimens used for the molecular analyses were killed and stored in 96% ethanol. Observations were made using a Leica MZ16 stereomicroscope (Leica Microsystems) with magnification up to 120×, illuminated with diffuse light (neon bulb, 6400 K; Philips, Amsterdam, The Netherlands). Dry specimens were soaked in water with a small amount of acetic acid. Dissected body parts used for drawings were treated with lactic acid for several days, then washed in water or dehydrated in ethanol and mounted on slides in Berlese's fluid (Swan 1936) or Euparal (Paradox Co., Cracow, Poland). Drawings were made using a Leica drawing tube attached to a Leica DM 1000 microscope (Leica Microsystems), then scanned and traced in Adobe Illustrator CC (Adobe, San Jose, California, United States of America). All dissected body parts were glued with 5,5-dimethyl hydantoin formaldehyde on the same card as the respective specimen or put in the microvials filled with glycerine and pinned under the specimen. Digital photographs were made using a Canon EOS 5D mark II camera (Canon, Tokyo, Japan) attached to Zeiss Axio Zoom.V16 stereoscope (Carl Zeiss AG, Oberkochen, Germany). Image stacks were produced manually, combined using the Zerene Stacker 1.4 software (Zerene Systems LLC, Richland, Washington, United States of America), and edited in Adobe Photoshop CC (Adobe). Measurements were taken using a calibrated eyepiece graticule. Morphometric parameters are provided as range and mean ± standard deviation. The following abbreviations are used for the measured characters: BLbody length from anterior margin of pronotum to elytral apices along midline; HLhead length from anterior margin of clypeus to occipital margin along midline; HWmaximum head width; PLpronotal length along midline; PWmaximum pronotal width; ELelytral length from apex of scutellar shield to apices of elytra along suture; EWmaximum elytral width combined; PyLmaximum length of pygidium; RPLmaximum length of right paramere; LPLmaximum length of left paramere. Terminology used in morphological descriptions follows Franciscolo (1957), Lu et al. (1997), and Lawrence and Ślipiński (2010). All nomenclatorial acts follow regulations of the International Code of Zoological Nomenclature (International Trust of Zoological Nomenclature 1999). The examined material is deposited in the following collections: Dávid Selnekovič collection, Bratislava, Slovakia (DSBS), Hungarian Natural History Museum, Budapest, Hungary (HNHM), the Museum für Naturkunde, Humboldt-Universität zu Berlin, Berlin, Germany (MNHU), the Museo Zoologico dell'Università Federico II, Naples, Italy (MZFN), and Senckenberg Deutsches Entomologisches Institut, Müncheberg, Germany (SDEI).
Principal component analysis was performed using PAST 3.12 software (Hammer et al. 2001), using log-transformed variables of three morphometric characters: elytral length, right paramere length, and left paramere length (Supplementary material, Table S1). The dataset consisted of 60 male specimens of M. pseudorhenana from Bulgaria, Cyprus, Hungary, Israel, Italy, Montenegro, Slovakia, and Turkey, including holotype and all male genetic vouchers. Plots were subsequently edited in Adobe Illustrator CC.
A total of 30 adults were used for the DNA analyses (Table 3). Genomic DNA was extracted from whole individuals using E.Z.N.A.® Tissue DNA kit (OMEGA Bio-tek Inc., Norcross, Georgia, United States of America) according to the manufacturer's protocol. Extracted and purified DNA is stored at -25°C at the Department of Zoology of Faculty of Natural Sciences, Comenius University, Bratislava, Slovakia. A 568-bp-long fragment of cytochrome oxidase subunit 1 (CO1) was amplified with primers LCO1490 and HCO2198 (Folmer et al. 1994). Standard polymerase chain reaction was performed using DreamTaq™ Green DNA Polymerase (Thermo Fisher Scientific Inc., Waltham, Massachusetts, United States of America) for a total volume of 25.0 μL, comprising 100-200 ng genomic DNA, 2.5 μL DreamTaq™ Buffer, 2.5 μL 25 mM MgCl 2 , 2.0 μL of dNTP (deoxynucleotide triphosphohydrolase) mix, 1.0 μL of 3.0 pmol/mL each primer, 0.4 μL (5 U/ μL) DreamTaq™ DNA polymerase and nuclease-free water to 25.0 μL. Polymerase chain reaction was carried out on an Eppendorf thermal cycler (Eppendorf, Hamburg, Germany), with initial denaturation at 94°C for 1 minute, followed by 35 cycles of 94°C for 30 seconds, 52°C for 40 seconds, and 72°C for 1 minute, and 10 minutes of final extension at 72°C. All polymerase chain reaction products were detected on 1% agarose gel stained with GoldView (SBS Genetech, Beijing, China). Purification and Sanger sequencing were done in the commercial laboratory of Macrogen Europe Inc. (Amsterdam, The Netherlands) using both amplification primers. Consensus sequences, alignment, and final matrix were produced in Geneious 6.1.8 software (Kearse et al. 2012). Mordella aculeata Linnaeus, 1758 and Mordellistena variegata (Fabricius, 1798) were used as outgroups. Estimates of evolutionary divergence between CO1 sequences were calculated using the Kimura two-parameter model (Kimura 1980). The dendrogram was based on the maximum likelihood method, and bootstrap support values were calculated in MEGA X software (Kumar et al. 2018). The best-fitted substitution model (GTR I G) was selected by jModelTest 2 (Darriba et al. 2012) using 1000 replicates. Voucher identifiers and GenBank and BOLD accession numbers are listed in Table 3.

Morphology and systematics
Examination of the male lectotype of Mordellistena minima Costa, 1854 deposited in the Museo Zoologico dell'Università Federico II revealed a unique set of characters separating the species from other congeners (see differential diagnosis). The presence of yellow metatibial spurs, in combination with an entirely black body and short antennomeres, observed in the lectotype of M. minima is a rather unique condition that appears only in two other taxa from the M. confinis group: M. emeryi Schilsky, 1895 andM. lindbergi Ermisch, 1963. Re-examination of the lectotype of M. emeryi and comparison of the male genitalia with the lectotype of M. minima (Fig. 5F,G) revealed that the specimens are conspecific, and therefore we propose M. emeryi as a new junior subjective synonym of the latter. The redescription of the species provided below is based on a lectotype of M. minima, the type series of M. emeryi, and a series of specimens recently collected in the type locality, Ischia, Italy.
The identity of M. minima is not consistent with a previously accepted interpretation of the species presented by Ermisch (1963) and Batten (1977) and followed by subsequent authors (Odnosum 1992(Odnosum , 1993(Odnosum , 2003(Odnosum , 2005(Odnosum , 2010Horák 2008;Ruzzier 2013). The definition of M. minima as a species, with long apically pointed galea, expanded protibiae in males with distinct clusters of extended setae, and short antennomeres 5-10, was based on a misidentification. The aforementioned species interpretation was found to correspond with the holotype of M. pseudorhenana Ermisch, 1977 previously synonymised with M. minima by Batten (1980). After a re-examination of the holotype, we consider M. pseudorhenana to be a valid species, which can be separated from other members of M. confinis species group by the presence of long and pointed galea (Fig. 6A) and the combination of characters listed in the differential diagnosis section. The holotype of M. sajoi Ermisch, 1977 shares the important diagnostic characters with the holotype of M. pseudorhenana, and we consider M. sajoi a new junior subjective synonym of the latter species.
Among the material examined for the present study, we were able to identify two morphotypes of M. pseudorhenana that differ in the size and shape of the parameres. Morphotype 1 is represented here by a holotype and 88 additional male specimens from several localities in Europe (Fig. 2), while morphotype 2 is represented by 27 male specimens from Cyprus, Israel, and Turkey. The two morphotypes differ in the size and shape of the parameres: morphotype 1's parameres are shorter and smaller in proportion to the elytral length than they are in morphotype   Table 1); basal portions of the parameres in morphotype 1 are shorter in proportion to the distal processes than they are in morphotype 2 (Fig. 6E,F); and the dorsal process of the left paramere in morphotype 1 is shorter and wider than it is in morphotype 2 (Fig. 6E, F). The differences in dimensions are also shown by the results of the principal component analysis (Fig. 3A). Despite the morphological differences, the genetic divergence in CO1 fragment between the representatives of the two morphotypes is very low (Table 5; discussed in the Molecular Analyses section).  Ermisch, 1977 along the first two components of the principal component analysis. The analysis is based on three morphometric characters: elytral length, right paramere length, and left paramere length. The black cluster represents the specimens from central and southern Europe, the red cluster represents specimens from Cyprus. Entire dataset used for the analysis is provided in Supplementary material, Table S1; B, differences in the length of parameres between the specimens of M. pseudorhenana from southern and central Europe versus the specimens from Cyprus.
For the principal component analysis, we focused on two morphotypes of M. pseudorhenana that can be distinguished based on the shape and size of the parameres. The first group representing morphotype 1 consisted of 39 male specimens from Bulgaria, Hungary, Italy, Montenegro, and Slovakia, including the holotype and voucher specimens used for the molecular analyses. The second group consisted of 21 male specimens from Cyprus, Israel, and Turkey, also including the genetic vouchers. The principal component analysis was based on a set of three characters (elytral length, right paramere length, and left paramere length) that best reflect the differences in morphology. The analysis revealed two separate clusters that represent the two morphotypes ( Fig. 3 A). The first principal component explained 68.5% of the variance and correlated strongly with the length of the right paramere ( Table 2). The second principal component explained 29.8% of the variance and correlated strongly with elytral length ( Table 2). Results of the principal component analysis are congruent with differences in the actual measurements ( Fig. 3B; Table 1; Supplementary material, Table S1).

Molecular analyses
The sequences of CO1 gene fragment were obtained from 30 out of 35 amplified samples representing five species of M. confinis group, plus two outgroup species (Table 3). The analysed CO1 fragment was 568 bp long, with no indels and stop codons. The maximum likelihood analysis revealed all five presumed ingroup species as distinctly separate clades, each with bootstrap value of 100 (Fig. 4). The Kimura two-parameter genetic divergences between species were high and ranged from 13.9%   Table S2). The mean interspecific distance between the five analysed species from the M. confinis species group was 18.5%. The mean intraspecific distances ranged from 0% in M. minima to 1.9% in M. hirtipes (Table 4). The M. pseudorhenana morphotype 1 was represented in the analyses by 13 specimens from Bulgaria, Italy, and Slovakia. Morphotype 2 was represented by five specimens from Cyprus. The analyses revealed four haplotypes in the morphotype 1 and two haplotypes in morphotype 2 (Tables 3 and 5). Based on the Kimura two-parameter distances, M. lindbergi was recovered as the closest neighbour of M. pseudorhenana, with the smallest interspecific distance (16.5%)   Table S2). The divergences between the two M. pseudorhenana morphotypes ranged from 0.2% between haplotypes from Cyprus and Bulgaria to 1.4% between haplotype 3 from Cyprus and haplotype 4 from Italy (Table 5). The highest intraspecific Kimura two-parameter distance between the two morphotypes is 11.8 times less than the smallest interspecific distance between M. pseudorhenana and M. lindbergi. The analysed CO1 fragment did not provide any evidence to consider the two morphotypes separate species.

Distribution
Distributional records for M. minima that were published by Ermisch (1963), Batten (1976), Odnosum (1993Odnosum ( , 2003Odnosum ( , 2010, and Horák (2008Horák ( , 2020 were identified to refer to M. pseudorhenana, based on the revised material and the illustrations of male genitalia presented in the publications. The large series of examined material revealed new distributional records for M. pseudorhenana from Bosnia and Herzegovina, Slovenia, and Switzerland. The range of M. pseudorhenana reaches from Spain in the west, across the whole Mediterranean basin to Turkey, from Israel and Jordan in the south, across the Pannonian basin to Hungary and Slovakia in the north, and along the Black and Caspian seas to Ukraine, Azerbaijan, and Kyrgyzstan in the east (Fig. 2). A new record from Rajecké Teplice, Slovakia marks the northernmost known extent of the species' distribution (Fig. 2).  Heyden et al. (1906: 456) [catalogue]; Schaufuss (1916: 766) [catalogue, first report from Germany]; Roubal (1934: 5) [first report from Morocco]; Franciscolo (1942: 7) [localities], Franciscolo (1956: 4) [localities].
Mordellistena minima may be assigned to the M. confinis species group as defined by Ermisch (1956). Within this group, the combination of yellowish metatibial spurs and completely black body, including legs and antennae, is shared by three other species: M. lindbergi Ermisch, 1963, M. eversi Ermisch, 1965, and M. canariensis Ermisch, 1965. The species most closely resembles M. lindbergi and can be distinguished by the following characters: (1) the pubescence on elytra in M. minima has a distinct purple sheen, whereas that of M. lindbergi has a distinct greenish sheen; (2) protibiae are, in males of M. minima, slightly expanded, sometimes with several extended setae, compared to those of M. lindbergi, which are not expanded and are without extended setae; (3) the metatibiae in M. minima usually possess 4-5 lateral ridges, whereas those of M. lindbergi usually possess three lateral ridges, the last of which is very short; (4) abdominal sternite VIII in males of M. minima is ca. 2.00 times longer than wide with a rounded apex (Fig. 5C), whereas that of M. lindbergi is approximately 1.60 times longer than wide, with lateral margins distinctly convergent and a slightly emarginated apex; (5) the parameres of M. minima are as illustrated in Figure 5F,G, and those of M. lindbergi are as shown in Horák (1996); (6) the species are separated by ca. 19% divergence in the barcoding fragment of the CO1 gene (Table 3). Both M. eversi and M. canariensis are known only from the Canary Islands and differ from M. minima by having distinctly longer antennae, with antennomeres 5-10 almost twice as long as wide.
Redescription. Body slender (Fig. 1A), wedge-shaped, widest before middle of elytra, dorsum moderately convex, venter strongly so. Basic metric characters are listed in Tables 1 and 6. Colour of almost entire integument black, with very fine bluish sheen; maxillary palpi and four basal antennomeres sometimes dark reddish-brown; metatibial spurs yellowish with black apices. Vestiture consisting of dense, decumbent, dorso-ventrally flattened setae; colour uniformly yellowish on head and sternal thoracic parts; in anterior portions of pronotum yellowish, somewhat darkened posteriorly; in anterior portions of elytra yellowish, gradually darkened posteriorly to completely black at apices; on first to abdominal ventrites yellowish, on following ventrites gradually darkened to completely black on ventrite 5 and pygidium; on legs yellowish, somewhat darkened towards apices; vestiture on pronotum and elytra with distinct purple sheen.
Pronotum moderately convex, slightly wider than long (Table 6), widest behind middle, finely microreticulate with dense rasp-like punctures; anterior margin slightly produced in middle, margination complete, anterior angles rounded; lateral carinae rounded in dorsal aspect and very slightly concave in lateral aspect, margination inapparent but complete; posterior angles obtuse and rounded in lateral aspect. Prosternal process obliterated. Scutellar shield triangular, with rasplike setiferous punctures. Mesoventral process truncate at apex, as wide as mesotibia. Metaventral discrimen distinct, reaching shortly before middle. Metanepisternite rather wide, with lateral margin concave and mesal margin straight.
Elytra moderately convex, about twice as long as combined width (Table 6), widest around end of anterior one-third, moderately narrowed, with lateral carinae convergent behind middle (Fig. 1A); apices separately rounded; surface with rasp-like, setiferous punctures, and with fine microreticulation consisting of transverse, undulate lines.
Protibiae straight, in males expanded basally and sometimes with few extended setae; mesotibiae slightly bent medially; metatibiae with short apical ridge and 3-5 short lateral ridges parallel to apical tibial margin, reaching ca. one-quarter of tibial width, subequal in length except the last one usually shorter, sometimes inapparent; metatibial spurs yellowish with black apices, outer one ca. 0.60 times as long as inner one. Protarsi slightly longer than protibiae, first protarsomere slightly longer than following two segments combined, penultimate protarsomere distinctly longer than wide, with anterior margin slightly concave, claws tridentate; mesotarsi ca. 1.30 times as long as mesotibiae; metatarsomere 1 with 3-5 ridges, metatarsomere 2 with two ridges, metatarsomere 3 without ridges.
Sexual dimorphism. Females generally larger than males, with elytra somewhat wider in proportion to length (Table 6). Protibiae in males slightly expanded basally, sometimes with several extended setae; in females not expanded and without longer setae. Terminal maxillary palpomere in males wider than in females (Fig. 5A,B). BL, body length from anterior margin of pronotum to elytral apices along midline; HL, head length from anterior margin of clypeus to occipital margin along midline; HW, maximum head width; PL, pronotal length along midline; PW, maximum pronotal width; EL, elytral length from apex of scutellar shield to apices of elytra along suture; EW, maximum elytral width combined; PyL, maximum length of pygidium; RPL, maximum length of right paramere; LPL, maximum length of left paramere.
Variability. The individual variability is, besides the dimensions (Table 6), most strongly pronounced in the colouration of the pubescence, which may be pale yellowish on most of the dorsal surfaces or darkened to various extent in posterior portions of pronotum and elytra or entirely brownish. The number of lateral ridges on metatibiae varies from three to five and those on metatarsomere 1 also from three to five.
DNA sequences. Two DNA sequences of 568-bp CO1-gene fragment are deposited in GenBank and BOLD databases with accession numbers listed in Table 3.
Natural history. The adults were collected on the xeric Mediterranean grasslands (Fig. 7A) and ruderal vegetation on the flowers of Apiaceae plants. The larva is not known.
Based on the morphology, the species can be assigned to the M. confinis species group as defined by Ermisch (1956). From other members of the group, it can be differentiated based on the long and pointed galea (Fig. 6A), in combination with short antennomeres 5-10, expanded protibiae with distinct clusters of extended setae in males, and completely black-coloured body. Such form of galea is also present in M. grisea Mulsant, 1856(sensu Batten 1977, but it can be differentiated from M. pseudorhenana by its protibiae not being expanded in males and by parameres of different shape. Redescription. Body slender, wedge-shaped, widest at proximal one-half of elytra, dorsum moderately convex, venter strongly so (Fig. 1B). Metric characters are provided in Tables 1 and 6. Colour of almost all surfaces uniformly black with fine bluish sheen; anterior margin of frontoclypeus, mandibles, lacinia, galea, and labium, including palpi, brownish. Vestiture consisting of dense, decumbent, dorso-ventrally flattened setae; colour uniformly light-yellowish on head and sternal thoracic parts; in anterior one-half of pronotum light-yellowish and slightly darkened postero-medially; in antero-lateral portions of elytra light-yellowish and gradually darkened postero-medially to completely black in apical portions; on femora and proximal portions of tibiae light-yellowish and gradually darkened distally; on abdominal ventrites 1-2 light-yellowish, gradually darkened on following ventrites to entirely black on ventrite 5 and pygidium; vestiture on elytra with strong purple sheen.
Elytra moderately convex, evenly and strongly narrowed posteriorly, widest at end of first one-quarter, EL/EW ratio in Table 6; lateral carinae rounded, strongly convergent behind first one-quarter (Fig. 1B); apices separately rounded; surface with fine microreticulation formed by transverse, undulate lines and with rasp-like setiferous punctures.
Protibiae straight, distinctly dilated basally and with fringe of long, medially oriented setae in males; mesotibiae slightly bent medially; metatibiae with one apical and three lateral ridges parallel to apical tibial margin; second lateral ridge is usually distinctly longer than first one, not reaching beyond one-half of tibial width; third lateral ridge often short and inapparent; metatibial spurs black, outer one ca. 0.70 times as long as inner one. Protarsi about as long as protibiae, protarsomere 1 as long as following two tarsomeres combined; protarsomere 4 distinctly longer than wide with anterior margin slightly concave; protarsal claws tridentate; mesotarsi ca. 1.30 times as long as mesotibiae; metatarsomere 1 with three ridges, metatarsomere 2 with two ridges, metatarsomere 3 without ridges.
Sexual dimorphism. Antennomeres 5-10 in males slightly longer, ca. 1.20-1.30 times longer than wide, whereas in females they are ca. 1.10 times longer than wide. Maxillary palpomeres 1-3 in males bearing very long setae on the ventral surface; palpomeres 2 and 4 in males wider than in females (Fig. 6 A,B). Protibiae in males basally expanded, with group of longer, medially oriented setae; in females, simple, without extended setae.
Variability. The colour of pubescence on the dorsal surfaces varies from pale yellowish to darkened to various extents in posterior portions of pronotum and elytra to completely brownish. Rather distinct differences can be found in the shape and dimensions of the parameres between the morphotype 1 represented by holotype plus all the examined male specimens from Europe and the morphotype 2 represented by male specimens from Cyprus, Israel, and Turkey (Figs. 2, 3, 6 E,F; Table 1; Supplementary material, Table S1). The specimens of morphotype 2 have their parameres longer (Table 1), with the basal part of the right paramere longer and the dorsal process of the left paramere longer and narrower than in the morphotype 1 (Fig. 6 E,F). The barcoding region of CO1 gene shows very small differences between the representatives of the two morphotypes (0.18-1.43%; Table 5). between M. pseudorhenana and M. lindbergi (16.8%). Furthermore, the intraspecific Kimura 2-parameter distances within M. pseudorhenana ranged from 0.2 to 1.4%, with no distinct gap. Such low genetic divergence between the two morphotypes does not provide evidence for establishing the morphotypes as separate species.
The discrepancy between the morphological and molecular evidence opens the discussion about the efficiency of using the conventional DNA barcoding marker (CO1) for testing the taxonomic boundaries and verifying the status of the species within the family Mordellidae. It also raises questions about the validity of species that have been defined by rather weak morphological differences. The broader datasets, a combination of multiple genes, and the use of more advanced tools in molecular taxonomy such as character-based DNA barcoding should yield more insights into the taxonomy of this problematic group.