Taxonomic Investigation of the Genus Ablepharus (Sauria; Scincidae) with Molecular and Morphological Methods in Anatolian Populations

Home , Ablepharus, Ablepharus kitaibelii, Asymblepharus, Heremites, Lygosominae, Scincinae

Turkish Journal of Zoology Turk J Zool (2020) 44: 134-145 http://journals.tubitak.gov.tr/zoology/ © TÜBİTAK Research Article doi:10.3906/zoo-1911-14

Taxonomic investigation of the genus Ablepharus (Sauria; Scincidae) with molecular and morphological methods in Anatolian populations

1, 2 Emin BOZKURT *, Kurtuluş OLGUN  1 Department of Veterinary Sciences, Eldivan Vocational School of Health Services, Çankırı Karatekin University, Çankırı, Turkey 2 Department of Biology, Faculty of Science and Arts, Adnan Menderes University, Aydın, Turkey

Received: 11.11.2019 Accepted/Published Online: 28.01.2020 Final Version: 04.03.2020

Abstract: Ablepharus Lichtenstein, 1823, which includes ten species, is distributed in eastern Europe and Asia. Four species are recorded in Turkey: A. kitaibelii, A. chernovi, A. bivittatus, and A. budaki. After molecular and morphological studies in Anatolia, the phylogenetic relationship of the genus is still very complicated. Here, we investigate the taxonomic status of Ablepharus in Anatolia using morphological and molecular methods. The genetic structure ofAblepharus populations in Anatolia was analyzed using both the nuclear (CMOS) and mitochondrial (cyt b and COI) gene regions. The maximum likelihood and Bayesian inference methods were applied for phylogenetic tree reconstruction, together with calculation of p-distances and reconstruction of haplotype networks. The morphological analysis was evaluated with SPSS by using 12 metric and 17 meristic characteristics. In total, 41 specimens and 279 specimens were used for the molecular and morphological analysis, respectively. A. bivittatus has highly different genetic and morphological characteristics compared to the others, and it shares similar morphological characteristics with the genus Asymblepharus, which are elliptical tympanum, hidden upper eye opening under 3 or 4 large shields, light and dark longitudinal stripes on the back, adpressed hind-limbs reaching to knees, and well-developed limbs with five toes. For this reason,A. bivittatus was placed in the genus Asymblepharus as Asymblepharus bivittatus comb. nov. A. budaki anatolicus was elevated to species level as A. anatolicus based on morphological and genetic data. In this study, Anatolian A. kitaibelii comprised two sister subclades, A. kitaibelii kitaibelii and A. kitaibelii stepaneki, and this research is the first record of A. kitaibelii stepaneki in Turkey. A. chernovi was divided into three subclades: A. chernovi chernovi, A. chernovi eiselti, and A. chernovi isaurensis.

Key words: Skinks, Turkey, Asymblepharus, pholidosis, phylogeny

1. Introduction Asia and colonized the Balkan Peninsula from Anatolia Scincidae is the largest lizard family with 1598 species about 5.51 million years ago (Skourtanioti et al., 2016). representing more than 25% of the world’s lizard species Ablepharus kitaibelii (Bibron and Bory, 1833) was (Bauer, 1998; Hedges, 2014; Skourtanioti et al., 2016). This described from the Pylos ruins. In 1953, A. chernovi was family was divided into seven subfamilies by Hedges (2014): described by Darevsky in Armenia. Fuhn (1970) defined Acontinae, Egerniinae, Eugongylinae, Lygosominae, the subspecies of A. k. stepaneki and introduced five Mabuyinae, Sphenomorphinae, and Scincinae. Pyron subspecies: A. k. chernovi, A. k. fabichi, A. k. fitzingeri, A. et al. (2013) stated that Scincinae and Feylininae are k. kitaibelii, and A. k. stepaneki. Eremchenko and Sczerbak synonymous. Ten species belonging to Scincidae are (1986) considered A. chernovi as a separate species. distributed in Anatolia and these are Ablepharus bivittatus, Göçmen et al. (1996) described a new subspecies, A. A. budaki, A. chernovi, A. kitaibelii, Chalcides ocellatus, kitaibelii budaki, in Cyprus. Eumeces schneideri, Heremites auratus, H. septemtaeniatus, Schmidtler (1997) made a comprehensive revision H. vittatus, and Ophiomorus kardesi (Schmidtler, 1997; of the genus Ablepharus in Anatolia and the Middle East Ilgaz et al., 2007; Kumlutaş et al., 2007; Poulakakis et al., and (1) raised A. kitaibelii chernovi to the species level as 2008; Durmuş et al.; 2011; Baran et al., 2012; Karin et al., Ablepharus chernovi Darevsky, 1953 and recognized four 2016, Kornilios et al., 2018). subspecies (A. c. chernovi, A. c. eiselti, A. c. isaurensis, Ablepharus Lichtenstein, 1823 is distributed in eastern and A. c. ressli) in Anatolia; (2) raised A. k. budaki to the Europe and Asia and is currently represented by ten species level as Ablepharus budaki Göçmen et al. 1996 and species (Poulakakis et al., 2005). It originated in southeast described two subspecies (A. b. budaki and A. b. anatolicus) * Correspondence: [email protected] 134

This work is licensed under a Creative Commons Attribution 4.0 International License. BOZKURT and OLGUN / Turk J Zool in southern Anatolia; and (3) restricted the distribution The mitochondrialcyt b region was also amplified via of A. kitaibelii kitaibelii from the Balkan Peninsula to PCR using the following primers: PL 5’-AAC CAA GAC Central Anatolia. Skourtanioti et al. (2016) proposed that CTG TGA YAY GAA-3’ and PH 5’-GGC TTA CAA GAC the western Anatolian and Kastelorizo populations could CAR KGC TTT-3’. The PCR reaction was carried out in a represent a distinct species with the available name A. total volume of 25 µL, with the following conditions: an anatolicus. initial cycle of 94 °C for 3 min, followed by 35 cycles of 94 The aim of this study was to solve the taxonomic °C for 30 s, 52 °C for 40 s, and 72 °C for 90 s, and a final issues of the genus Ablepharus in Anatolian populations cycle of 72 °C for 10 min (Huang et al., 2013). In addition with morphological and molecular analyses. Here, to the previous gene region, the mitochondrial COI region two mitochondrial markers (cyt b and COI) and one was amplified via PCR using the RepCOI-F 5’-TNT TMT nuclear marker (CMOS) were used and a morphological CAA CNA ACC ACA AAG A-3’ and RepCOI-R 5’-ACT analysis was also performed using metric and meristic TCT GGR TGK CCA AAR AAT CA-3’ primers. The PCR characteristics. reaction was carried out in a total volume of 25 µL, with the following conditions: an initial cycle of 94 °C for 3 min, 2. Materials and methods followed by 40 cycles of 94 °C for 40 s, 49 °C for 30 s, and 2.1. Genetic methods 74 °C for 1 min, and a final cycle of 72 °C for 7 min (Nagy To analyze the molecular sequences, 41 samples (Figure et al., 2012). A commercial automated sequencer was used 1; Table 1) with total DNA extracted by using a standard to sequence the amplified products (GATC-Biotech AG, phenol/chloroform procedure (Hillis and Moritz, 1990) Konstanz, Germany). were used. The nuclearCMOS region was amplified via DNA sequences were aligned using the CLUSTAL W polymerase chain reaction (PCR) using the following algorithm as implemented in BioEdit version 7.2.5 (Hall, primers: G73.1 5’-GGC TRT AAA RCA RGT GAA GAA 1999) while manually adjusting the alignments. Two A-3’ and G74.1 5’-GAR CWT CCA AAG TCT CCA ATC- different phylogenetic approaches were used, maximum 3’. The PCR reaction was carried out in a total volume of likelihood (ML) and Bayesian inference (BI), for the 25 µL, with the following conditions: an initial cycle of 92 partitioned concatenated alignment. The best models of °C for 12 min, followed by 40 cycles of 94 °C for 1 min, 56 nucleotide substitution for each gene were calculated in Mr °C for 1 min, and 72 °C for 1 min, and a final cycle of 72 °C Modeltest v3.7 (Posada and Crandall, 1998) based on the for 5 min (Whiting et al., 2003). Akaike information criterion. The bootstrap method with

Figure 1. Distribution of the lizards used in the molecular and morphological analysis. For more details on samples, check Table 1. Colors indicate the lineages based on the analysis (orange: A. bivittatus, claret red: A. chernovi chernovi, pink: A. chernovi eiselti, yellow: A. chernovi isaurensis, red: A. budaki budaki, green: A. budaki anatolicus, blue: A. kitaibelii kitaibelii, and turquoise: A. kitaibelii stepaneki).

135 BOZKURT and OLGUN / Turk J Zool

Table 1. List of the materials used in the molecular and morphological analysis.

Number of samples used in Working morphological GenBank accession number, cyt b/ Taxon Locality number analyses COI/CMOS (♂♂/♀♀/ juveniles) 1 0/1/0 A. c. isaurensis Between Gülnar and Mut, 18 km/Mersin MN447752/*/MN418805 2 1/2/0 A. c. isaurensis Meke Saline/Konya MN447755/MN447788/MN418805 3 0/1/0 A. c. isaurensis Alahan/Mersin MN447758/*/MN418805 4 0/0/1 A. c. isaurensis Berendi village/Karaman MN447762/MN447793/MN418805 5 1/1/0 A. c. isaurensis Ortaören/Mersin MN447757/MN447790/MN418805 6 0/3/0 A. c. isaurensis Beyşehir/Konya MN447765/*/MN418805 7 7/9/0 A. c. isaurensis Kohu Mountain/Antalya MN447761/MN447792/MN418808 8 0/1/0 A. c. eiselti Mansurlu/Adana MN447749/MN447785/MN418808 9 3/1/0 A. c. eiselti Çamlıyayla/Mersin MN447759/MN447791/MN418808 10 1/1/0 A. c. chernovi Evcihüyük village/Kahramanmaraş MN447748/*/MN418808 11 1/2/0 A. c. chernovi Akoluk village/Kayseri MN447751/*/MN418808 12 0/1/0 A. c. chernovi Suceyin village/Malatya MN447750/*/MN418808 13 0/1/0 A. c. chernovi Yumaklıcerit village/Adıyaman MN447753/MN447786/MN418808 14 0/0/2 A. c. chernovi Sarısalkım village/Gaziantep MN447756/MN447789/MN418808 15 1/1/0 A. c. chernovi Between Kilis and Hassa, 7 km/Kilis MN447760/*/MN418808 16 0/1/0 A. c. chernovi Milkisağ village/Tunceli MN447754/MN447787/MN418808 17 2/1/1 A. c. chernovi Akdamar Island/Van MN447763/*/MN418808 18 1/0/0 A. k. kitaibelii Osmaniye village/Aydın MN447764/*/MN418804 19 1/1/0 A. k. kitaibelii Eber Lake/Konya MN447766/MN447794/MN418804 20 4/6/2 A. k. kitaibelii Between Eğirdir and Gelendost/Isparta MN447773/*/MN418804 21 3/4/1 A. k. kitaibelii Eğirdir/Isparta MN447772/MN447796/MN418806 22 1/1/0 A. k. kitaibelii Türkönü village/İzmir MN447774/*/MN418806 23 0/1/0 A. k.kitaibelii Toparlar/Muğla MN447770/*/MN418806 24 1/1/0 A. k. kitaibelii Bafa Lake/Muğla MN447768/*/MN418804 25 0/1/0 A. k. kitaibelii Salih Island/Muğla MN447769/*/MN418804 26 0/5/0 A. k. stepaneki Between Kalabak-Kalkım/Balıkesir MN447767/MN447795/MN418804 27 1/0/0 A. k. stepaneki Demirci village/İstanbul MN447771/*/MN418804 28 0/1/0 A. b. budaki Düziçi/Osmaniye MN447734/*/MN418807 29 1/0/0 A. b. budaki İskenderli village/Gaziantep MN447735/*/MN418807 30 2/3/2 A. b. budaki Hassa/Hatay MN447738/*/MN418807 31 0/1/0 A. b. budaki Kozan/Adana MN447737/MN447778/MN418807 32 5/1/0 A. b. budaki Müftüler village/Hatay MN447743/MN447783/MN418807 33 0/2/0 A. b. anatolicus Between Cevizli and Akseki, 27 km/Antalya MN447736/MN447777/MN418807 34 1/0/2 A. b. anatolicus Kapıkargın village/Muğla MN447739/MN447779/MN418807 35 0/1/0 A. b. anatolicus Ulupınar Village/Antalya MN447742/MN447782/MN418807 36 3/1/0 A. b. anatolicus Büyükçaltıcak/Antalya MN447744/MN447784/MN418807 37 1/1/0 A. b. anatolicus Cemilli Village/Mersin MN447740/MN447780/MN418807 38 1/0/0 A. b. anatolicus Between Silifke and Mut 6. km/Mersin MN447741/MN447781/MN418807

136 BOZKURT and OLGUN / Turk J Zool

Table 1. (Continued).

39 8/2/0 A. bivittatus Saray/Van MN447745/MN447775/MN418809 40 8/11/0 A. bivittaus Murataldı Village/Van MN447747/MN447776/MN418809 41 7/6/0 A. bivittatus Between Çaldıran and Doğubeyazıt/Van MN447746/*/MN418809 42 1/2/0 A. c. chernovi Turnadere Village/Tunceli 43 0/1/1 A. c. chernovi Adıyaman 44 1/0/0 A. c. chernovi Between Başçiftlik and Reşadiye/Tokat 45 1/5/0 A. c. chernovi Polateli/Kilis 46 3/7/2 A. b. budaki Yayladağ/Hatay 47 1/2/0 A. c. eiselti Saimbeyli/Adana 48 1/1/0 A. c. chernovi Hanyeri Village/Adana 49 3/6/1 A. c. isaurensis Çıralı Lake/Konya 50 2/5/0 A. k. kitaibelii Beyşehir/Konya 51 1/2/1 A. b. anatolicus Manavgat/Antalya 52 1/0/0 A. b. anatolicus Gebiz/Antalya 53 2/2/0 A. b. anatolicus Termessos Ruins/Antalya 54 1/1/0 A. b. anatolicus Altınyaka/Antalya 55 2/2/0 A. c. isaurensis Söğüt village/Antalya 56 4/1/0 A. b. anatolicus Rhodiapolis/Antalya 57 2/2/0 A. b. anatolicus Kaş/Antalya 58 6/7/0 A. b. anatolicus Kalkan/Antalya 59 1/0/0 A. b. anatolicus Karaağaç village/Muğla 60 1/0/0 A. b. anatolicus Kayaköy/Muğla 61 1/0/0 A. b. anatolicus Taşbaşı village/Muğla 62 0/1/0 A. b. anatolicus Gökbel village/Muğla 63 2/2/1 A. k. kitaibelii Köyceğiz/Muğla 64 5/1/0 A. k. kitaibelii Kahya Island-Marmaris/Muğla 65 4/8/0 A. k.kitaibelii Madran Mountain/Aydın 66 2/4/0 A. k. kitaibelii Aydın 67 1/0/0 A. k. kitaibelii Selçuk/İzmir 68 0/1/0 A. k. kitaibelii Buca/İzmir 69 2/3/1 A. k. kitaibelii Güzelçamlı National Park/Aydın

1000 replicates was used to assess the nodal support for network (Bandelt et al. 1999) was calculated with Network the ML tree with PAUP version 4.0b10 (Swofford, 2002), 4.5.1 (available from http://www.fluxusengineering.com). and BI was conducted in Mr.Bayes 3.2.1 (Huelsenbeck and 2.2. Morphological methods Ronquist, 2001). The analysis of BI started with a randomly This analysis was conducted on 279 Ablepharus specimens generated tree and was run for 5 × 107 generations, saving from field studies between 2012 and 2016, and deposited a tree after 1000 generations. The first 25% of these trees samples from the Zoology Museum of Aydın Adnan were discarded while generating a consensus tree from the Menderes University and Dokuz Eylül University (Figure post-burn-in trees, and Eumeces schneideri was used as an 1; Table 1). Specimens were anesthetized with ether and outgroup. fixed with 96% ethanol. The genetic distances between taxa were estimated in During morphological investigation, metric, meristic, MEGA version X (Kumar et al., 2018), as percentages of and index (proportional) characteristics were determined p-distance values, as in the work of Sindaco et al. (2014). For according to the literature (Göçmen et al., 1996; Schmidtler, the nuclear marker (CMOS), a median-joining haplotype 1997; Kumlutaş et al., 2005; Ilgaz et al., 2007). The meristic

137 BOZKURT and OLGUN / Turk J Zool characteristics were counted with an Olympus brand Within A. chernovi three subgroups were recognized: A. binocular microscope, and metric characteristics were c. chernovi (central and eastern part of Anatolia), A. c. measured with a 0.01-mm caliper. isaurensis (northern part of Taurus Mountain), and A. The following metric measurements were taken: SVL c. eiselti (Cilician part of Anatolia). The second major (snout-vent length), tip of snout to anal cleft; PW (pileus group is Anatolian A. kitaibelii and is divided into two width), at widest point between parietal plates; PL (pileus subgroups: A. k. kitaibelii (western part of Anatolia) length), tip of snout to the posterior margins of parietals; and A. k. stepaneki (northwestern part of Anatolia). The BW (body width), at the widest of the body posterior of third group is Anatolian A. budaki and splits into two the forelimb; FLL (left forelimb length), outstretched limb subgroups: A. b. budaki (southeastern part of Anatolia) from shoulder joint to tip of toe; HLL (left hind-limb and A. b. anatolicus (southwestern part of Anatolia). The length), outstretched limb from hip joint to tip of toe; fourth group is Anatolian A. bivittatus and it is represented FHLL, forelimb to the hind-limb; IL (interparietal length); by one group (Figure 2). IW (interparietal width); DE (diameter of ear opening), The eight major groups exhibited high genetic the largest diameter of the ear opening; 2SL (second divergence among themselves. The mean p-distance supraciliar plate length); and 2SW (second supraciliar values in Ablepharus varied in the range of 0%–19.0% in plate width). Furthermore, the following morphometric cyt b, 0%–19.4% in COI, and 0%–1.6% in CMOS. Higher indexes were calculated: PW/SVL, BW/SVL, 2SW/2SL, percentages of mean p-distance values in all gene regions IW/IL, and FLL/HLL. are seen between A. bivittatus and other groups. Table 2 Meristic (pholidolial) characteristics were considered shows the percentage of mean p-distance values between in the following counts: supraciliar plates (left–right, the species and subspecies of Ablepharus in Anatolia. SCPa–SCPb), supralabial plates (left-right, SRLa-SRLb CMOS returned a very low haplotype diversity. - number of labials both anterior and posterior to the Specifically, A. budaki budaki and A. budaki anatolicus center of eye), front supralabial plates (left–right, FSLa– have the same haplotype, while A. bivittatus is highly FSLb), sublabial plates (left–right, SLa–SLb), ventral plates divergent from others as shown in Figure 3. In addition (longitudinal, VP), subdigital lamellae in the 4th toe (left– to the network analysis, the DNA polymorphism values right, SDLa–SDLb), transversal series of dorsal scales at of CMOS were calculated as h = 6, Hd = 0.815, and π = the midtrunk (DS), transversal series of scales in a row at 0.00491. the neck (NS), the scales composing the rings around the eyes (left–right, ESa–ESb), and number of scales between 3.2. Meristic analysis eye and ear (left–right, SBEEa–SBEEb). According to the Kolmogorov–Smirnov test, only the Statistical analysis was performed with SPSS 22.0, and number of the ventral plates (VP) shows significant descriptive statistics were used for each taxon. To detect differences between males and females (P ≤ 0.05), as shown sexual dimorphism, the Mann–Whitney U Test was in Table 3, while the other meristic characteristics do not applied. Nondimorphic characteristics were analyzed with present sexual dimorphism. The number of the VP ofA. one-way ANOVA and the Tukey HSD test. Discriminant budaki budaki and A. budaki anatolicus is significantly function analysis was performed for multiple comparisons. lower than the other Ablepharus for both females and males (P ≤ 0.05). For this characteristic, the values for A. 3. Results bivittatus are significantly higher thanA. budaki budaki 3.1. Genetic analysis and A. budaki anatolicus, while they are significantly lower A total of 2185 bp of concatenated DNA sequences (cyt b than other Ablepharus taxa in both sexes. 1088 bp, COI 680 bp, and CMOS 417 bp) were used. The 41 The meristic characteristics, except VP, have no haplotypes of the ingroup sequences included 637 variable significant differences between sexes. Hence, the combined and 575 parsimony-informative sites [409 (37.6%) variable data were analyzed for male and female lizards, which is and 373 (34.3%) parsimony informative in cyt b; 217 summarized in Table 3. (31.9%) variable and 193 (28.4%) parsimony informative According to result of the Mann–Whitney U test and in COI; and 11 (2.6%) variable and 9 (2.2%) parsimony descriptive statistics, A. bivittatus is different from all other informative in CMOS]. taxa in terms of all characteristics. In addition, A. budaki The phylogenetic results based on the mitochondrial budaki and A. budaki anatolicus have significantly higher and nuclear DNA show four major groups in the genus numbers of SDLa and SDLb than the other taxa, except Ablepharus in Anatolia, with high support values for each A. bivittatus. The maximum number of mean SBEEa and group (ML bootstrap value ≥70, BI posterior probability SBEEb were obtained within the A. chernovi group while value ≥0.95) (Figure 2). As already known (Skourtanioti et the minimum number of mean SBEEa and SBEEb were al., 2016), A. kitaibelii and A. chernovi are monophyletic. observed in the A. kitaibelii group.

138 BOZKURT and OLGUN / Turk J Zool

Figure 2. Maximum likelihood (ML) and Bayesian tree presenting the phylogenetic relationship between the specimens of the genus Ablepharus in Turkey based on the analysis of the CMOS, cyt b, and COI gene regions. Numbers in terminal nodes refer to working codes. Numbers above branches are BI posterior probabilities and numbers below branches are ML bootstrap values. Asterisk indicates 1 BI posterior probability value and 100 bootstrap value.

Table 2. Mean genetic divergences (% p-distance values) among taxa for CMOS (above diagonal), cyt b, and COI (below diagonal). Bold values are genetic distance within taxa (CMOS/cyt b/COI).

A. c. isaurensis A. c. eiselti A. c. chernovi A. k. kitaibelii A. k. stepaneki A. b. budaki A. b. anatolicus A. bivittatus A. c. isaurensis 0.1/2.3/2.2 0.4 0.4 0.8 0.7 0.7 0.7 1.6 A. c. eiselti 5.7/5.3 0.0/2.3/2.7 0.0 0.4 0.2 0.2 0.2 1.2 A. c. chernovi 8.3/7.9 8.4/8.2 0.0/2.5/2.8 0.4 0.2 0.2 0.2 1.2 A. k. kitaibelii 13.2/13.0 13.0/13.9 12.8/12.9 0.2/1.3/2.4 0.1 0.6 0.6 1.6 A. k. stepaneki 13.4/11.8 6.8/12.1 12.4/11.9 13.7/6.5 0.0/3.6/- 0.5 0.5 1.4 A. b. budaki 16.0/16.0 16.2/15.3 14.9/15.8 15.2/15.8 14.8/14.7 0.0/4.6/5.3 0.0 1.0 A. b. anatolicus 14.6/15.1 13.7/14.9 13.6/14.6 14.5/15.8 14.5/14.6 14.9/15.1 0.0/6.1/5.2 1.0 A. bivittatus 17.5/18.7 17.3/18.0 17.8/17.2 17.5/19.4 18.0/18.6 18.1/18.3 19.0/16.6 0.0/1.5/0.2

The connection of the prefrontal plate was observed three large plates under the supraciliar plates, called with a higher percentage in A. bivittatus (97.60%), whereas preablepharid by Eremchenko and Sczerbak (1980), were the percentage of prefrontal plate connection was 8.00%, seen only in A. bivittatus. 6.40%, 7.20%, 0.00%, 10.30%, 13.20%, and 0.00% in A. For all taxa, differences were observed for SCPa (F budaki budaki, A. budaki anatolicus, A. chernovi chernovi, = 1140.945, P ≤ 0.001), SCPb (F = 1337.995, P ≤ 0.001), A. chernovi eiselti, A. chernovi isaurensis, A. kitaibelii SRLa (F = 265.276, P ≤ 0.001), SRLb (F = 1064.407, P ≤ kitaibelii, and A. kitaibelii stepaneki, respectively. However, 0.001), FSLa (F = 265.276, P ≤ 0.001), FSLb (F = 711.995,

139 BOZKURT and OLGUN / Turk J Zool

Figure 3. Median-joining network of CMOS. Colors of circles represent taxa as shown in Figure 2 and size of circles is proportional to the number of individuals per haplotypes. Numbers in the boxes show specimen numbers and correspond to Table 1.

P ≤ 0.001), SLa (F = 9.811, P ≤ 0.001), SLb (F = 9.442, chernovi chernovi and A. chernovi isaurensis, respectively. P ≤ 0.001), SDLa (F = 230.118, P ≤ 0.001), SDLb (F = All in all, the discriminant and descriptive analysis highly 248.702, P ≤ 0.001), DS (F = 150.728, P ≤ 0.001), SBEEa (F supports the separation of species and subspecies for the = 240.435, P ≤ 0.001), and SBEEb (F = 311.823, P ≤ 0.001). Ablepharus genus. According to the discriminant analysis, A. bivittatus and 3.3. Metric analysis A. budaki anatolicus tend to be different from the other The diameter of ear openings was measured in all taxa as shown in Figure 4. taxa except A. chernovi’s subspecies. To test for sexual As a result of the discriminant function analysis, a total dimorphism, the Mann–Whitney U test was used, and of 79.80% among-group variation was explained by the sexual dimorphism was seen for SVL, FHLL, IW/IL, and first canonical variate, which provides the discrimination 2SW/2SL. Females are significantly longer than males in between A. bivittatus and other taxa (Figure 4). The terms of SVL and FHLL in all taxa. According to one-way discriminant analysis classifiesA. bivittatus, A. budaki analysis (ANOVA) performed between taxa, significant budaki, A. budaki anatolicus, and A. kitaibelii kitaibelii in differences were found for DE (df = 4, 178; F = 49.228, P ≤ their correct taxa, while the discrimination of A. chernovi 0.001), 2SL (df = 7, 248; F = 12.017, P ≤ 0.001), 2SW (df = chernovi, A. chernovi eiselti, and A. kitaibelii stepaneki is 7, 248; F = 30.209, P ≤ 0.001), IL (df = 7, 248; F = 60.178, not clearly provided. The misclassified specimens were P ≤ 0.001), IW (df = 7, 248; F = 39.795, P ≤ 0.001), FLL represented by 13 of 23 A. chernovi chernovi as A. chernovi (df = 7, 248; F = 223.094, P ≤ 0.001), HLL (df = 7, 246; F isaurensis as well as 3 and 4 of the A. chernovi eiselti as A. = 217.153, P ≤ 0.001), PL (df = 7, 248; F = 109.378, P ≤

140 BOZKURT and OLGUN / Turk J Zool

Table 3. Descriptive statistics of the meristic and metric characteristics of taxa (N = number of individuals; SD = standard deviation; abbreviations are given in Section 2).

A. budaki A. b. A. chernovi A. kitaibelii A. k. A. bivittatus A. c. eiselti A. c.isaurensis budaki anatolicus chernovi kitaibelii stepaneki ♂♂+♀♀ ♂♂+♀♀ ♂♂+♀♀ ♂♂+♀♀ ♂♂+♀♀ ♂♂+♀♀ ♂♂+♀♀ ♂♂+♀♀ (N = 42) (N = 25) (N = 47) (N = 26) (N = 8) (N = 39) (N = 68) (N = 6) Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD SCPa 5.00 0.32 2.00 0.00 2.00 0.00 2.00 0.00 2.00 0.00 2.07 0.26 1.99 0.20 2.00 0.00 SCPb 5.05 0.38 1.97 0.19 2.02 0.14 2.00 0.00 2.00 0.00 2.02 0.16 1.99 0.12 2.00 0.00 SRLa 5.93 0.26 5.21 0.41 5.98 0.14 5.03 0.18 5.00 0.00 5.00 0.00 5.00 0.00 5.00 0.00 SRLb 5.95 0.22 5.00 0.00 5.98 0.14 5.00 0.00 5.00 0.00 5.00 0.00 5.00 0.00 5.00 0.00 FSLa 3.93 0.26 3.21 0.41 3.98 0.14 3.03 0.18 3.00 0.00 3.00 0.00 3.00 0.00 3.00 0.00 FSLb 3.93 0.26 3.00 0.00 3.98 0.14 3.00 0.00 3.00 0.00 3.00 0.00 3.00 0.00 3.00 0.00 SLa 6.64 0.53 5.21 0.41 6.12 0.48 6.10 0.31 6.13 0.35 5.93 0.41 6.10 0.34 6.00 0.00 SLb 6.55 0.50 5.00 0.00 6.04 0.20 6.07 0.25 6.13 0.35 5.98 0.42 6.08 0.32 6.17 0.41 SDLa 17.80 0.90 13.82 1.19 13.65 0.85 11.57 0.96 11.88 0.64 11.95 0.97 11.21 0.86 11.17 1.17 SDLb 18.00 0.85 13.90 1.15 13.69 0.99 11.56 0.97 11.50 0.54 12.06 0.94 11.17 0.85 11.00 0.63 VP 43.22 2.85 40.10 2.57 38.47 2.74 49.27 4.43 46.38 5.04 47.88 4.14 47.68 3.43 47.83 3.92 DS 22.69 0.92 19.69 0.76 19.46 0.89 18.00 0.00 17.75 0.71 18.03 0.16 19.24 0.91 20.00 0.00 NS 3.02 0.75 3.24 0.69 3.41 0.58 3.43 0.97 3.12 0.64 3.32 0.57 3.47 0.60 3.17 0.41 SBEEa 3.17 0.38 3.07 0.26 2.94 0.37 4.13 0.35 4.13 0.35 4.20 0.40 2.01 0.12 2.00 0.00 SBEEb 3.19 0.40 3.03 0.19 2.94 0.24 4.03 0.18 4.13 0.35 4.20 0.40 2.01 0.12 2.00 0.00 SVL 43.84 4.98 38.57 5.09 32.66 3.81 38.41 6.44 36.94 3.57 39.38 5.64 38.63 4.31 34.78 1.88 PW 4.46 0.32 3.50 0.48 3.05 0.30 3.12 0.34 3.20 0.24 3.34 0.44 3.21 0.26 3.01 0.15 PL 8.05 0.48 6.27 0.56 5.59 0.43 5.61 0.47 5.91 0.48 6.04 0.69 5.73 0.41 5.61 0.15 FLL 10.13 0.81 6.23 0.79 5.28 0.69 5.24 0.67 5.34 0.46 6.07 0.74 5.52 0.57 5.80 0.50 HLL 15.04 1.01 9.52 1.32 8.12 1.00 7.90 1.04 8.15 0.60 9.21 1.23 8.22 0.77 7.73 0.55 FHLL 24.69 3.97 22.79 3.86 18.99 3.08 24.51 5.22 22.18 3.61 24.54 4.35 24.65 3.91 20.48 2.26 IL 1.59 0.14 1.32 0.20 1.06 0.14 1.09 0.17 1.25 0.14 1.10 0.16 1.05 0.13 1.09 0.19 IW 1.33 0.14 1.06 0.14 0.98 0.11 0.92 0.10 1.03 0.10 0.96 0.13 1.00 0.12 1.02 0.20 BW 6.05 0.50 4.20 0.89 3.29 0.48 3.70 0.69 3.56 0.58 4.04 0.82 4.02 0.46 3.24 0.30 2SL 0.71 0.19 0.91 0.15 0.66 0.09 0.76 0.10 0.80 0.08 0.80 0.14 0.69 0.10 0.71 0.07 2SW 0.51 0.10 0.32 0.05 0.33 0.05 0.41 0.07 0.44 0.05 0.38 0.05 0.38 0.07 0.35 0.04 FLL/HLL 0.68 0.05 0.66 0.07 0.65 0.06 0.67 0.07 0.66 0.06 0.66 0.05 0.67 0.06 0.75 0.05 IW/IL 0.84 0.08 0.81 0.10 0.94 0.12 0.86 0.10 0.82 0.06 0.88 0.11 0.97 0.12 0.94 0.12 2SW/2SL 0.77 0.32 0.36 0.07 0.50 0.10 0.55 0.12 0.55 0.05 0.49 0.09 0.55 0.08 0.50 0.07 PW/SVL 0.10 0.01 0.09 0.01 0.09 0.01 0.08 0.01 0.09 0.01 0.09 0.01 0.08 0.01 0.09 0.01 BW/SVL 0.14 0.01 0.11 0.02 0.10 0.01 0.10 0.01 0.10 0.02 0.10 0.01 0.11 0.01 0.09 0.01

0.001), PW (df = 7, 248; F = 71.850, P ≤ 0.001), BW (df The results of the Kolmogorov–Smirnov test show = 7, 248; F = 75.797, P ≤ 0.001), and FLL/HLL (df = 7, normality for PW/SVL and BW/SVL. Also, these 246; F = 2.498, P ≤ 0.05). Based on the Mann–Whitney U characteristics do not have sexual dimorphism according test and descriptive statistics, A. bivittatus is significantly to the Mann–Whitney U test. Differences were found for bigger than the others in terms of IL, IW, FLL, HLL, PL, PW/SVL (F = 22.039, P ≤ 0.001) and BW/SVL (F = 42.120, PW, and BW (Table 3). P ≤ 0.001) in all taxa. According to discriminant analysis,

141 BOZKURT and OLGUN / Turk J Zool

Figure 4. Discriminant function analysis of thirteen meristic characteristics of the eight Ablepharus taxa in Anatolia.

A. bivittatus tends to be different in terms of PW/SVL and The genusAsymblepharus was identified by elliptical BW/SVL characteristics from the other taxa (Figure 5). tympanum, hidden upper eye opening under the 3 or A total of 49.00% of originally grouped cases were 4 large shields, 2–6 striped plates on the outside of the correctly classified. Similarly, 90.50% ofA. bivittatus forelimb, light and dark longitudinal strips on the back, and A. budaki budaki, 71.40% A. budaki anatolicus, and adpressed hind-limbs reaching to knees, well-developed 82.10% A. kitaibelii kitaibelii were found to be accurately limbs with five toes, and transversal series of dorsal scales classified within their taxa. As a result, the discriminant at the midtrunk 24–30 (Eremchenko and Sczerbak, 1980). function and descriptive analysis highly supports species A. bivittatus also shares these characteristics, except the and subspecies separation for the Ablepharus genus. transversal series of dorsal scales at the midtrunk 22–24. Hence, A. bivittatus must be classified asAsymblepharus 4. Discussion bivittatus comb. nov. based on the morphological and There is limited knowledge about Ablepharus bivittatus in molecular analysis, while it must also be studied with a terms of molecular and morphological data. Eremchenko molecular approach representing all distribution areas for and Sczerbak (1980) completed a comprehensive other Asymblepharus species. study about Scincella, Ablepharus, and Asymbelpharus Ablepharus budaki was first described as a subspecies genera with morphological aspects and reported some of A. kitaibelii from Cyprus by Göçmen et al. (1996). morphological differentiation of A. bivittatus from other Schmidtler (1997) raised A. kitaibelii budaki to the species Ablepharus species. There are two extensive studies level as A. budaki and described two subspecies: A. budaki about the morphology of the species by Eremchenko and budaki and A. budaki anatolicus. Poulakakis et al. (2005), Sczerbak (1986) and Ilgaz et al. (2007), which mostly using mitochondrial DNA markers, confirmed that A. focused on the comparison of the populations. Since then, budaki was a distinct species from A. kitaibelii and showed publications have mainly focused on new locality reports. that the specimens of A. budaki from Syria and Cyprus In our study, samples of A. bivittatus are completely were a monophyletic group. Skourtanioti et al. (2016), divergent from the other Ablepharus taxa in Anatolia in using mitochondrial and nuclear DNA markers, suggested terms of the network analysis, genetic distance, BI and that A. budaki was a paraphyletic group and its specimens ML topology, and all meristic and metric characteristics. from southwest Turkey were differentiated compared to

142 BOZKURT and OLGUN / Turk J Zool

Figure 5. Discriminant function analysis of PW/SVL and BW/SVL of the eight Ablepharus taxa in Anatolia. the rest of the A. budaki populations, and related to A. 2005; Skourtanioti et al., 2016). Since these studies did not kitaibelii from southwest Turkey and Kastelorizo Island. A. include samples from Armenia and the east of Turkey, they budaki was found to be monophyletic in our incomplete reported that the biogeographic pattern and taxonomic phylogeny that does not include several Ablepharus status of A. chernovi were still unclear. Anatolian subspecies and geographic regions, as in the morphological Ablepharus chernovi was found to be monophyletic based study conducted by Schmidtler (1997). A. budaki budaki on our study, and it is separated into three subclades, which and A. budaki anatolicus are recognized here as sister were morphologically identified by Schmidtler (1997). In clades, but it is known that they are in fact not sisters from this study, we also collected samples from Çamlıyayla, the other more complete studies (Skourtanioti et al., 2016). type locality for A. chernovi ressli, but we did not find any These taxa are easily distinguishable by front supralabial differences between A. chernovi eiselti and the Çamlıyayla plates. A. budaki anatolicus can be raised to species level population. For this reason, we consider A. chernovi eiselti as A. anatolicus based on the genetic data of Skourtanioti and A. chernovi ressli as synonymous. According to the et al. (2016) and morphological data of Schmidtler (1997) results of our genetic and morphological analysis, the (Table 3). Skourtanioti et al. (2016) also reported that A. separation of these three subclades is not at species level. anatolicus separated from the other Ablepharus taxa at The subspecies taxonomy ofAblepharus kitaibelii about 12.87 Mya, while A. budaki diverged from the others was revised several times by various authors, and it still at about 10.08 Mya. remains unclear. Before Schmidtler (1997), most of the Ablepharus chernovi was first described as a subspecies studies focused on the locality record and morphological of A. kitaibelii by Darevsky (1953). Schmidtler (1997) comparison in Turkey. Schmidtler (1997) made a revision raised A. kitaibelii chernovi to the species level as A. of the genus Ablepharus and reported that A. kitaibelii was chernovi and described four subspecies of A. chernovi: A. distributed in the Balkan Peninsula and western and central chernovi chernovi, A. chernovi eiselti, A. chernovi ressli, Anatolia, but there was no taxonomic differentiation and A. chernovi isaurensis. A. chernovi populations from in these regions. Poulakakis et al. (2005) reported that Syria and Turkey form a monophyletic group while A. A. kitaibelii was divided in two lineages: (1) Greece, chernovi and A. kitaibelii are sister groups, with the split Peloponnesus, Cyclades, and Kithira and (2) Crete, the East between them dated at about 11.24 Mya (Poulakakis et al., Aegean islands, and Turkey. A. kitaibelii and A. chernovi

143 BOZKURT and OLGUN / Turk J Zool were sister groups (Poulakakis et al., 2005). Skourtanioti et Thrace and Marmara regions. The exact distribution of al. (2016) split A. kitaibelii into two groups: (1) a western A. kitaibelii stepaneki in Turkey can be revealed by future group (A. kitaibelii kitaibelii from Greece, A. k. stepaneki studies with the addition of new samples from the Thrace from Romania, and A. k. fitzingerifrom Serbia) and (2) and Marmara regions. an eastern group (A. k. kitaibelii from Turkey and the Aegean islands and A. k. fabichi Crete and Karpathos), and Acknowledgments they also stated that the morphological subspecies of A. This study was based on the PhD study of Emin Bozkurt. k. kitaibelii was not monophyletic. Fuhn (1970) identified The thesis was supervised by Kurtuluş Olgun. This project a new subspecies, namely, A. kitaibelii stepaneki, and gave was supported by Aydın Adnan Menderes University the distribution range including Thrace. Schlüter (2005) Scientific Research Projects Unit with number FEF- also reported this subspecies along the border of Turkey 13012. We would also like to thank TÜBİTAK (Project and Bulgaria. In this study, the İstanbul and Balıkesir No. 113Z752) for additional materials and scholarship populations were evaluated as A. kitaibelii stepaneki based support. The authors would like to thank Aziz Avcı, Nazan on the molecular analysis. Our study is the first record of Üzüm, Can Yılmaz, Çetin Ilgaz, and Yusuf Kumlutaş for A. kitaibelii stepaneki in Turkey. As a result of this study, their valuable contributions. We also thank Süleyman A. kitaibelii kitaibelii is distributed in central and western Ceylan, Mehmet Tural, Yusuf Geroğlu, and Özkan Sercan Anatolia, while A. kitaibelii stepaneki is distributed in the for their help during field studies.

References

Bandelt HJ, Forster P, Röhl A (1999). Median-joining networks for Hillis DM, Moritz C (editors) (1990). Molecular Systematics. 1st ed. inferring intraspecific phylogenies. Molecular Biology and Sunderland, MA, USA: Sinauer. Evolution 16: 37-48. Huang WS, Lin SM, Dubey S, Pike DA (2013). Predation drives Baran İ, Ilgaz Ç, Avcı A, Kumlutaş Y, Olgun K (2012). Türkiye Amfibi interpopulation differences in parental care expression. Journal ve Sürüngenleri, Ankara, Turkey: TÜBİTAK Popüler Bilim of Animal Ecology 82: 429-437. Kitapları (in Turkish). Huelsenbeck JP, Ronquist F, Nielsen R, Bollback JP (2001). Bayesian Bauer AM (1998). Lizards. In: Halliday T, Adler K (editors). inference of phylogeny and its impact on evolutionary biology. Encyclopedia of Reptiles and Amphibians. 1st ed. San Diego, Science 294: 2310-2314. CA, USA: Academic Press, pp. 126-173. Ilgaz Ç, Kumlutaş Y, Olgun K, Baran İ (2007). The morphology Durmuş SH, Kumlutaş Y, Özdemir A, Avcı A, Ilgaz Ç (2011). and distribution of Ablepharus bivittatus (Menetries, 1832) Morphology, taxonomic status and distribution of Trachylepis (Reptilia: Sauria: Scincidae) in Turkey. Russian Journal of aurata (Linnaeus, 1758) in southeast Anatolia (Squamata: Herpetology 14: 91-97. Sauria: Scincidae). Herpetozoa 24: 61-71. Karin BR, Metallinou M, Weinell JL, Jackman TR, Bauer AM (2016). Eremchenko VL, Sczerbak NN (1980). Ablepharus lindbergi sp. n. Resolving the higher-order phylogenetic relationship of the (Reptilia, Sauria, Scincidae) from Afghanistan-A new species circumtropical Mabuya group (Squamata: Scincidae): an out- of snake-eyed lizard. Vestnik Zoologii 6: 84-86. of-Asia diversification. Molecular Phylogenetics and Evolution 102: 220-232. Eremchenko VL, Sczerbak NN (1986). Ablefaridnyye Yashcheritsy Fauny SSSR i Sopredel’nykh Stran. 1st ed. Frunze, Russia: Ylym Kornilios P, Kumlutaş Y, Lymberakis P, Ilgaz Ç (2018). Cryptic Publications (in Russian). diversity and systematics of the Aegean Ophiomorus skinks (Reptilia: Squamata), with the description of a new species. Fuhn IE (1970). Über die unterarten von Ablepharus kitaibelii Journal of Zoological Systematics and Evolutionary Research (Bibron & Bory de St. Vincent, 1833) (Sauria, Scincidae). Acta 56: 364-381. Societatis Zoolooicae Bohemoslovacea 1: 9-17 (in German). Kumar S, Stecher G, Li M, Knyaz C, Tamura K (2018). MEGA X: Göçmen B, Kumlutaş Y, Tosunoğlu M (1996). A new subspecies, Molecular Evolutionary Genetics Analysis across computing Ablepharus kitaibelii (Bibron & Borry, 1833) budaki n. ssp. platforms. Molecular Biology and Evolution 35: 1547 - 1549. (Sauria: Scincidae) from Turkish Republich of Northern Kumlutaş Y, Arıkan H, Ilgaz Ç, Kaska Y (2007). A new subspecies, Cyprus. Doğa Turkish Journal of Zoology 20: 397-405. Eumeces schneideri barani n. ssp (Reptilia: Sauria: Scincidae) Hall TA (1999). BioEdit: A user-friendly biological sequence from Turkey. Zootaxa 1387: 27-38. alignment editor and analysis program for Windows 95/98/ Kumlutaş Y, Öz M, Özdemir A, Tunç MR, Durmuş H et al. (2005). NT. Nucleic Acids Symposium Series 41: 95-98. On the populations of Abelpharus kitaibelii (Bibron and Bory, Hedges SB (2014). The high-level classification of skinks (Reptilia, 1833) (Sauria: Scincidae) from South-Western Anatolia. Squamata, Scincomorpha). Zootaxa 3765: 317-338. Pakistan Journal of Biological Sciences 8: 461-465.

144 BOZKURT and OLGUN / Turk J Zool

Nagy ZT, Sonet G, Glaw F, Vences M (2012). First large-scale DNA Schmidtler JF (1997). Die Ablepharus kitaibelii – Gruppe in Süd- barcoding assessment of reptiles in the biodiversity hotspot of Anatolien und benachbarten Gebieten (Squamata, Sauria, Madagascar, based on newly desinged COI primers. PLoS One Scincidae). Herpetozoa 10: 35-63 (in German). 7: 1-11. Sindaco R, Kornilios P, Sacchi R, Lymberakis P (2014). Taxonomic Posada D, Crandall KA (1998). Modeltest: Testing the model of DNA reassessment of Blanus strauchi (Bedriaga, 1884) (Squamata: substitution. Bioinformatics 14: 817-818. Amphisbaenia: Blanidae), with the description of a new species from south-east Anatolia (Turkey). Zootaxa 3795: 311-326. Poulakakis N, Lymberakis P, Tsigenopoulos CS, Magoulas A, Mylonas M (2005). Phylogenetic relationships and evolutionary history Skourtanioti E, Kapli P, Ilgaz Ç, Kumlutaş Y, Avcı A et al. (2016). of snake-eyed skink Ablepharus kitaibelii (Sauria: Scincidae). A reinvestigation of phylogeny and divergence times of the Molecular Phylogenetics and Evolution 34: 245-256. Ablepharus kitaibelii complex (Sauria, Scincidae) based on mtDNA and nuDNA genes. Molecular Phylogenetics and Poulakakis N, Pakaki V, Mylonas M, Lymberakis P (2008). Evolution 103: 199-214. Molecular phylogeny of the Greek legless skink Ophiomorus punctatissimus (Squamata: Scincidae): the impact of the Mid- Swofford DL (editor) (2002). PAUP: Phylogenetic Analysis Aegean trench in its phylogeography. Molecular Phylogenetics Using Parsimony (And Other Methods), Version 4. 1st ed. and Evolution 47: 396-402. Sunderland, MA, USA: Sinauer. Pyron RA, Burbrink FT, Wiens JJ (2013). A phylogeny and revised Whiting AS, Bauer AM, Sites JW Jr (2003). Phylogenetic relationship classification of Squamata, including 4161 species of lizards and limb loss in sub-Saharan Africa scincine lizards (Squamata: and snakes. BMC Evolutionary Biology 13: 93-146. Scincidae). Molecular Phylogenetics and Evolution 29: 582- 598. Schlüter U (2005). Die herpetofauna der bulgarischen schwarzmeerküste – Teil 2: Echsen. Elaphe 13: 52-58 (in German).

145