Evaluating Taxonomic Inflation Towards Evidence-Based Species

Home , Barred grass snake, Daboia, Hungarian meadow viper, Macrovipera schweizeri, Montivipera xanthina, Vipera aspis

Evaluating taxonomic inflation: towards evidence-based species

ANGOR UNIVERSITY delimitation in Eurasian vipers (Serpentes: Viperinae) Freitas, Inês; Ursenbacher, Sylvain; Mebert, Konrad; Zinenko, Oleksandr; Schweiger, Silke; Wüster, Wolfgang; Brito, José C.; Crnobrnja-Isailovi, Jelka; Halpern, Bálint; Fahd, Soumia; Santos, Xavier; Pleguezuelos, Juan M.; Joger, Ulrich; Orlov, Nikolay; Mizsei, Edvárd; Lourdais, Olivier; Zuffi, Marco A. L.; Strugariu, Alexandru; Zamfirescu, tefan Remus; Martínez-Solano, Íñigo; Velo- Antón, Guillermo; Kaliontzopoulou, Antigoni; Martínez-Freiría, Fernando Amphibia-Reptilia

PRIFYSGOL BANGOR / B DOI: 10.1163/15685381-bja10007

Published: 30/06/2020

Publisher's PDF, also known as Version of record

Cyswllt i'r cyhoeddiad / Link to publication

Dyfyniad o'r fersiwn a gyhoeddwyd / Citation for published version (APA): Freitas, I., Ursenbacher, S., Mebert, K., Zinenko, O., Schweiger, S., Wüster, W., Brito, J. C., Crnobrnja-Isailovi, J., Halpern, B., Fahd, S., Santos, X., Pleguezuelos, J. M., Joger, U., Orlov, N., Mizsei, E., Lourdais, O., Zuffi, M. A. L., Strugariu, A., Zamfirescu, . R., ... Martínez-Freiría, F. (2020). Evaluating taxonomic inflation: towards evidence-based species delimitation in Eurasian vipers (Serpentes: Viperinae). Amphibia-Reptilia, 41(3), 285-311. https://doi.org/10.1163/15685381-bja10007

Hawliau Cyffredinol / General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ?

Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

04. Oct. 2021 Amphibia-Reptilia (2020) DOI:10.1163/15685381-bja10007 brill.com/amre

Review

Evaluating taxonomic inﬂation: towards evidence-based species delimitation in Eurasian vipers (Serpentes: Viperinae)

Inês Freitas1,∗, Sylvain Ursenbacher2,3, Konrad Mebert4,5, Oleksandr Zinenko6, Silke Schweiger7, Wolfgang Wüster8,JoséC.Brito1, Jelka Crnobrnja-Isailovic´9,10, Bálint Halpern11, Soumia Fahd12, Xavier Santos1, Juan M. Pleguezuelos13, Ulrich Joger14, Nikolay Orlov15, Edvárd Mizsei16,17, Olivier Lourdais18, Marco A. L. Zufﬁ19, Alexandru Strugariu20,¸Stefan Remus Zamﬁrescu21, Íñigo Martínez-Solano22, Guillermo Velo-Antón1, Antigoni Kaliontzopoulou1, Fernando Martínez-Freiría1,*

Abstract. The designation of taxonomic units has important implications for the understanding and conservation of biodiversity. Eurasian vipers are a monophyletic group of viperid snakes (Serpentes, Viperinae), currently comprising four genera (Daboia, Macrovipera, Montivipera and Vipera) and up to 40 species. Taxonomic units have been described using a wide variety of methods and criteria, and consequently, considerable controversy still surrounds the validity of some currently listed species. In order to promote a consensus- and evidence-based taxonomy of Eurasian vipers, we analysed published mitochondrial and nuclear DNA sequences for this group to reconstruct phylogenetic relationships among currently recognized viper species. We also compiled information on external morphology to assess their morphological distinctiveness. Phylogenetic inference based on mtDNA sequences shows contrasting levels of divergence across genera and species and identiﬁes several instances of non-monophyly in described species. Nuclear DNA sequences show extremely low levels of genetic variation, with a widespread pattern of allele sharing among distant species, and even among genera. Revision of morphological data shows that most species designations rely on scalation traits that overlap extensively among species of the same genus. Based on our combined assessment, we recognize 15 taxa as valid species, three taxa which likely represent species complexes, 17 taxa of doubtful validity as species, and ﬁve taxa for which species status is maintained but further research is highly recommended to assess taxonomic arrangements. We stress the need to implement integrative taxonomic approaches for the recognition of evidence-based taxonomic units in Eurasian vipers.

Keywords: integrative taxonomy, morphology, mt-DNA, nuclear DNA, phylogeny, Viperidae.

1 - CIBIO/InBIO – Research Center in Biodiversity and Vienna, Burgring 7, 1010 Vienna, Austria Genetic Resources of the University of Porto, Vairão, 8 - Molecular Ecology and Fisheries Genetics Laboratory, Portugal School of Natural Sciences, Bangor University, Bangor 2 - Department of Environmental Science, Section of Con- LL57 2UW, Wales, UK servation Biology, University of Basel, Basel, Switzer- 9 - Faculty of Sciences and Mathematics, University of land Niš, Niš, Serbia 3 - info fauna-karch, Centre de coordination pour la pro- 10 - Institute for Biological Research “S. Stankovic”,´ Uni- tection des amphibiens et des reptiles de Suisse, versity of Belgrade, Belgrade, Serbia Neuchâtel, Switzerland 11 - MME BirdLife Hungary, Budapest, Hungary 4 - Global Biology, Waldmattstr, Birr, Switzerland 12 - Faculté des Sciences de Tétouan, Université Abdel- 5 - IDECC, Institute of Development, Ecology, Conserva- malek Essaâdi, Tétouan, Morocco tion and Cooperation, Rome, Italy 13 - Departamento de Zoología, Facultad de Ciencias, Uni- 6 - Museum of Nature, V. N. Karazin Kharkiv national versidad de Granada, Granada, Spain University, Kharkiv, Ukraine 14 - State Natural History Museum, Braunschweig, Ger- 7 - Herpetological Collection, Natural History Museum many

Downloaded from Brill.com04/29/2020 02:27:11PM © Freitas et al., 2020. DOI:10.1163/15685381-bja10007via free access This is an open access article distributed under the terms of the CC-BY 4.0 License. 2 I. Freitas et al.

Introduction species are therefore intended to identify bio- logically cohesive populations with recent com- The designation of taxonomic units has impor- mon ancestry rather than to recognize unusual tant implications for the way we study, describe patterns of distribution or morphology (Kaiser and understand biodiversity, as well as for how et al., 2013). we mobilize efforts and allocate resources to de- Eurasian vipers are a monophyletic group velop conservation strategies. Over the years, within the subfamily Viperinae (Serpentes, different criteria and tools have been used to de- Viperidae), whose members are distributed pri- fine species, leading to a succession of species marily in the Palaearctic region, i.e., non- concepts that resulted in extended controversy tropical Eurasia and North Africa (Phelps, within the research community (Mayden, 1997; 2010). This group is phylogenetically sister to de Queiroz, 2007). Nowadays, a species is of- a clade of Middle Eastern vipers, constituted ten defined as a separately evolving metapop- by the genera Eristicophis and Pseudocerastes ulation lineage that possesses relevant charac- (see Phelps, 2010; Zheng and Wiens, 2016), teristics that allow assessing its distinctiveness which are not considered in this work. At the from others (i.e., the unified species concept; time of writing, the most recent and comprehen- de Queiroz, 2007, and its precursor, the evolu- sive list of reptiles (i.e., The Reptile Database; tionary species concept; Simpson, 1961; Wiley, Uetz, Freed and Hošek, 2019) lists four gene- 1978; Frost and Hillis, 1990). This definition is ra and 40 species within Eurasian vipers (ta- linked to the integrative taxonomy framework, ble 1): Daboia, with 4 species; Macrovipera, which is based on the combination of different with 3 species; Montivipera, with 8 species; lines of evidence (e.g., genetic, morphological, and Vipera, with 25 species. However, Eurasian ecological) and methodologies (e.g., phyloge- vipers have a long taxonomic history, and dif- netic inference, ordination methods, ecological ferent authors have used a wide variety of meth- modelling) to objectively identify taxa (Dayrat, ods and criteria to define taxonomic units, as re- 2005) that – in an ideal case – would repre- flected in previous species lists (e.g., Mallow, sent independently evolving species. Names of Ludwig and Nilson, 2003; Phelps, 2010). At the genus level, the history of Eurasian

15 - Zoological Institute, Russian Academy of Sciences, St. vipers is relatively simple. Through most of the Petersburg, Russia 20th century, all species considered here were 16 - Department of Tisza River Research, Centre for Eco- included in the single genus Vipera (Boulenger, logical Research, Hungarian Academy of Sciences, De- 1896, 1913; Schwarz, 1936; Klemmer, 1963; brecen, Hungary Minton, Dowling and Russel, 1968). The mav- 17 - Department of Ecology, University of Debrecen, De- brecen, Hungary erick German herpetologist Albert Franz 18 - Centre d’Etudes Biologiques de Chizé, CNRS, UMR Theodor Reuss described numerous genera 7372, Villiers en Bois, France within the Eurasian vipers (reviewed by Krec- 19 - University of Pisa – Museum Natural History, Calci sák, 2007), but these gained little traction with (Pisa), Italy subsequent authors, except where the names 20 - Research Department – Faculty of Biology, “Alexandru Ioan Cuza” University of Ia¸si, Ia¸si, Romania had priority for subsequently validated clades. 21 - Department of Biology – Faculty of Biology, “Alexan- Obst (1983) was the first author to challenge the dru Ioan Cuza” University of Ia¸si, Ia¸si, Romania monogeneric classification of Eurasian vipers, 22 - Museo Nacional de Ciencias Naturales – CSIC, by separating the larger species into the genus Madrid, Spain ∗ Daboia, together with Pseudocerastes.This Corresponding author; e-mail: ifi[email protected]; split however, was not adopted by most subse- [email protected] quent researchers. Herrmann, Joger and Nilson Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 3 on, NT) Least D. = deserti EN EN EN 5% or more. LC should = M. M. M. mmended status, in some cases han one species, but delimitations should be M. lebetina M. M. bulgardaghica rom the closest sister species or clade are ld delimitation, single-species recognition 3or4%;High = , extent of gene flow among M. lebetina -group (2-3%), incl. M. bulgardaghica (see above), and M. wagneri divergence and extent of gene flow be examined extent of gene flow with relationship to clearly addressed, low genetic difference suggests a subspecific status for schweizeri evaluate extent of gene flow with wagneri low divergence within bornmuelleri albizona and these units should be evaluated 2% or less; Moderate = M. ,incl. (9%) subspecies should be examined for genetic M. xanthina M. (9%) complexes M. razii M. lebetina divergence categories: Low (2%), subspecific status was b (9%) M. raddei from the Turkish Mediterranean Critically Endangered. = Macrovipera Cyprus (2%) and shared hablotypelebetina with coast (Stümpel and Joger, 2009, Stümpel, 2012), despite 600 km distancerange between in its Milos to nearestpopulation mainland (Manavgat, Turkey) bulgardaghica suggested (Stümpel et al., 2016) (5-6%) and Endangered; CR = Vulnerable; EN morphological species high genetic divergence (9%) LC morphological species high genetic divergence from the remaining = + + (196 bp; supplementary table S3). “DVAS” means Doubtful Validity As Species, may represent geographic variation, a subspecies or diverged populati b morphological species high genetic divergence (15%) NT ( morphologicalmorphologicalgenetic species species high genetic divergence (15%) high genetic divergence (9%) LC morphological species high divergence from genetic morphological DVAS low divergence from morphological DVAS low genetic divergence from morphological species high genetic divergence from Near Threatened; VU = Currently listed species of Eurasian vipers according to Uetz, Freed and Hošek (2019), depicting the criteria used for species designation, our reco Gray, 1849 (Shaw and Nodder, 1797) (Smith, 1917) (Werner, 1938) (Linnaeus, 1758) Oraie et al., 2018 schweizeri (Werner, 1935) (Nilson, Andrén and Flärdh, 1990) bornmuelleri (Werner, 1898) provisional, including justification, suggestions forgiven further for work a small and fragment IUCN of red cyt list category (also for included taxa). Percentages of divergence f hence, currently we recommend toare decline not its clear species yet), hence, status; currentlyis “LSC” maintain currently means species maintained, Likely status; but Species and further Complex “pending” researchConcern; (i.e., indicates is NT that, group required despite to of some assess closely incongruences taxonomic related with status. taxa, the Cyt possibly divergence thresho more t Table 1. SpeciesDaboia mauritanica Criteria Status Reason Further work IUCN Daboia russelii Daboia siamensis Daboia palaestinae Macrovipera lebetina Macrovipera razii Macrovipera Montivipera albizona Montivipera Downloaded from Brill.com04/29/2020 02:27:11PM via free access 4 I. Freitas et al. M. LC LC CR EN NT ( albicornuta VU) CR LC M. M. might be M. latifii and lineages M. bornmuelleri and lineages evaluate extent of gene flow with albizona M. kuhrangica included, extent of gene flow withunits other should be evaluated evaluate extent of gene flow with other bornmuelleri evaluate extent of gene flow among Taurus populations recently described as a new subspecies based on morphology of one live and three road-killed specimensinsufficient is (Afsar et al., 2019); extentgeographic of variation, limits, and gene flow between Taurus, Greek, Aegean and Lycian populations must be evaluated M. V. and ;low and (6%) and M. V. ursinii M. xanthina ; low genetic ; low genetic , low genetic M. bornmuelleri should be included as V. transcaucasiana (9%) M. raddei M. raddei V. renardi M. bornmuelleri -group (2-3%) (2%) and M. albizona-M. bulgardahica (6-8%) (11-5%); includes several divergent (9%) M. albizona M. albizona (4%); ssp. (Stümpel et al., 2016) walser divergence (2%) divergence (2%) M. bornmuelleri to moderate divergence within bornmuelleri al., 2016), with paraphyletic positionlow and branch support; high genetic divergence to Vipera lineages (4-6%), two of them corresponding to raddei divergence (1%) morphological species high genetic divergence to morphological DVAS included within morphological DVAS low to moderate genetic divergence from morphological DVAS included within morphological species high genetic divergence to morphological DVAS sister to morphological LSC including 4 divergent lineages (Stümpel et morphologicalmorphological DVAS included within LSC high genetic divergence to the remaining (Continued.) Eiselt and Baran, 1970 kuhrangica Rajabizadeh, Nilson and Kami, 2011 bulgardaghica (Nilson and Andrén, 1985) (Mertens et al., 1967) (Boettger, 1890) (Nilson and Andrén, 1984) (Gray, 1849) Tuniyev, Nilson and Andrén, 2010 (Linnaeus, 1758) Vipera anatolica Montivipera Table 1. SpeciesMontivipera Criteria Status Reason Further work IUCN Montivipera latifii Montivipera raddei Montivipera wagneri Montivipera xanthina Vipera altaica Vipera ammodytes Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 5 VU) V. LC NT EN LC CR ebneri VU V. ursinii V. kaznakovi V. berus or V. kaznakovi must be clarified using must be included, as well as other V. berus nikolskii V. sakoi evaluate extent of gene flow among lineages evaluate extent of gene flow between western and eastern lineages search for new populations over larger areas and test for gene flow with V. barani subspecies and clades already recognized such as bosniensis relationships and gene flow to and integrative methods test for gene flow with nuclear inferences should be increased VU ( V. V. (6%); ,which from (4-5%) V. ebneri V. sakoi V. V. walser 1%) V. eriwanensis < V. seoanei V. kaznakovi , showing low genetic from Georgia, with from Russia, showing (11%); several species and V. berus V. ursinii V. kaznakovi V. kaznakovi (2-4%); it includes , renardi latastei-monticola recognized by Zuffi (2002) were subsequently rejected by genetic studies (Ursenbacher et al., 2006; Barbanera2009). et al., West and East lineages show moderate to high genetic divergence (4-5%) divergence (2%), although apparent geographic isolation; subspecific status should be more appropriate Russia (3-5%); private haplotypes in one nuDNA gene and apparent geographic isolation in relation to closelyspecies related geographic isolation to sister to moderate divergence (4%), and moderate to high genetic divergence to low genetic divergence ( renardi was previously synonymized and shows low genetic divergence to (1%); geographic isolation doubtful (Rajabizadeh et al., 2011, Tuniyev et2018); al., private haplotypes in some nuDNA genes; a subspecific status mustappropriate. be DVAS nested within + morphological pending moderate to high genetic divergence to + morphological species high genetic divergence to morphological geographic isolation morphological species high genetic divergence (5%) and morphological pending high genetic divergence to morphological DVAS sister to morphological DVAS low to moderate genetic divergence to genetic Böhme (Continued.) (Linnaeus, 1758) and Joger, 1983 (Linnaeus, 1758) Vedmederja et al., 1986 Nikolsky, 1913 (Reuss, 1933) (Nilson and Andrén, 1988) Table 1. SpeciesVipera aspis Criteria Status Reason Further work IUCN Vipera barani Vipera berus Vipera darevskii Vipera dinniki Vipera eriwanensis Vipera graeca Downloaded from Brill.com04/29/2020 02:27:11PM via free access 6 I. Freitas et al. EN EN VU NT EN CR NT , clade; V. latastei V. darevskii must be clarified, V. dinniki V. kaznakovi relationships between both mitochondrial lineages of also to understand the relationshipGeorgian of populations to Russian populations to populations from Iberia and Northmust Africa be recognized as distinct;gene extent of flow among further levels ofwithin structure each main clade should be evaluated extent of gene flow among lineagesthis within clade should be evaluated part of the North African and V. V. V. V. x x clade V. kaznakovi V. renardi . RAD-seq data V. kaznakovi , with low genetic V. monticola V. latastei V. kaznakovi V. kaznakovi (Baran et al., 2001; V. dinniki 1%), already suggested as a V. darevskii < )and ; another in Russia sister to ; Zinenko et al., 2016) ; Zinenko et al., 2016) (admixed population of V. renardi V. transcaucasiana Oleksandr Zinenko, unpublished; Mebert et al., 2015a) units, with high genetic divergence (10%): one in Georgia (type locality), sisterdarevskii to x indicates low differentiation among these mtDNA lineages (Oleksandr Zinenko, unpublished). orlovi included), and further divergent lineages within each clade geographically isolated, one in Iberiaanother and in North Africa ( divergence ( with low genetic divergence (1-2%) renardi within North African subspecies by Tuniyev et al. (2018) renardi hybrid between sympatric morphological pending it includes two very divergent paraphyletic morphological LSC it includes two divergent clades (7-8%), morphologicalmorphological DVASmorphological DVAS paraphyletic, included within admixed population ( pending highly divergent lineage (6%) included morphological DVAS admixed population ( morphological DVAS only two specimens known, represents a morphological DVAS nested within Nilson Tuniyev (Continued.) Nikolsky, 1909 Boscá, 1878 et al., 1995 Tuniyev and Ostrovskikh, 2001 Saint Girons, 1953 and Ostrovskikh, 2001 Billing et al., 1990 (Tuniyev et al., 2012) Table 1. SpeciesVipera kaznakovi Criteria Status Reason Further work IUCN Vipera latastei Vipera lotievi Vipera magnifica Vipera monticola Vipera orlovi Vipera pontica Vipera olguni Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 7 LC NT VU V. V. and V. lotievi , must be included V. ebneri , eriwanensis search for new populations over larger areas and test for gene flowdarevskii with V. altaica extent of gene flow must bebridge evaluated large to geographic gaps , V. V. is (less than V. renardi V. ammodytes V. darevskii V. darevskii from Russia V. eriwanensis from Russia (4-6%) V. berus V. kaznakovi (4-5%); however, only one and V. kaznakovi uncertain (4-6%) darevskii independent sample was tested and geographic isolation from ursinii isolation from 1%) with moderate to high genetic(3-5%); divergence recommended with subspecific status in Ursenbacher et al. (2008) and (6%), geographically isolated, private haplotypes in two nuDNA genes (Ghielmi et al., 2016) genetic DVAS low divergence to + morphological pending moderate to high genetic divergence to + genetic morphological species high divergence (5%) and geographic morphological species moderate to high genetic divergence to morphological morphological DVAS two lineages included within morphological species moderate to high divergence to genetic species high genetic divergence to (Continued.) (Tuniyev et al., 2018) Lataste, 1879 (Tuniyev et al., 2013) (Christoph, 1861) transcaucasiana Boulenger, 1913 (Bonaparte, 1835) Ghielmi et al., 2016 Vipera sakoi Vipera seoanei Vipera shemakhensis Table 1. SpeciesVipera renardi Criteria Status Reason Further work IUCN Vipera Vipera ursinii Vipera walser Downloaded from Brill.com04/29/2020 02:27:11PM via free access 8 I. Freitas et al.

(1992), using immunological distances, recog- were described using integrative taxonomy ap- nised Macrovipera for the lebetina group (in- proaches, first addressing phylogenetic relation- cluding the taxa mauritanica and deserti) and ships and later characterizing phenotypic vari- restricted Daboia to the species russelii.More ability (i.e., Vipera walser Ghielmi et al., 2016; recently, Nilson et al. (1999) described Mon- Vipera sakoi Tuniyev et al., 2018; Macrovipera tivipera as a new subgenus for the xanthina razii Oraie et al., 2018). However, the descriptions for the other six species were done in group. This was subsequently raised to full a traditional way, i.e., by solely recording or genus level by Joger (2005). Lenk et al. (2001), analysing morphological traits, without the sup- using mitochondrial DNA sequences, assigned port of molecular data or phylogenetic evi- the species mauritanica, deserti and palaesti- dence (Vipera magnifica Tuniyev and Ostro- nae to Daboia, leading to the current generic ar- vskikh, 2001; Vipera orlovi Tuniyev and Os- rangement of the group. The recent use of sub- trovskikh, 2001; Vipera altaica Tuniyev, Nilson genus Pelias (Merrem, 1820) as a full genus for and Andrén, 2010; Montivipera kuhrangica Ra- the berus and ursinii groups (e.g., Avcı et al., jabizadeh, Nilson and Kami, 2011; Vipera ol- 2010; Tuniyev et al., 2012, 2013, 2018a, b), on guni Tuniyev et al., 2012; Vipera shemakhensis the other hand, has remained a minority opinion Tuniyev et al., 2013). in the literature. Certain changes proposed out- Recent phylogenetic and phylogeographic side the peer-reviewed scientific literature are studies have transformed the taxonomic not considered here for reasons given in Kaiser panorama in Eurasian vipers considerably, vali- et al. (2013). dating some taxa as species (e.g., Vipera graeca, Mizsei et al., 2017), rejecting or synonymis- One third of the currently recognized species ing others (e.g., Vipera altaica with V. renardi, were described in the 18th and 19th centuries Zinenko et al., 2015, Montivipera albicornuta by recognized taxonomists and zoologists of with M. raddei and M. albizona with M. bul- the time (e.g., Carl Linnaeus described Colu- gardaghica, Stümpel et al., 2016; Daboia de- ber lebetinus, C. berus, C. aspis and C. am- serti with D. mauritanica, Martínez-Freiría et modytes (now in Macrovipera and Vipera)in al., 2017a; Vipera magnifica and V. orlovi were 1758; John Edward Gray described Clotho mau- identified as admixed populations, Zinenko et ritanica and Daboia xanthina (now in Daboia al., 2016), or modifying previously designated and Montivipera, respectively) in 1849; Eduard taxonomic units (e.g., assigning species to four Boscá described Vipera latastei in 1878; table genera; Garrigues et al., 2005). However, de- 1). During the 20th century, eighteen of the cur- spite this multitude of studies, there is still con- rently recognized species were described; ten of siderable uncertainty regarding the validity of them were described before or during the 1970s, some species, which is hampering the develop- and eight after that period. All species descrip- ment of optimized conservation strategy for the whole group. tions were based on morphological traits (i.e., Here, we apply an integrative approach to re- scale counts, biometric measures, colour pat- view the taxonomy of the Eurasian vipers, by terns), with the application of statistical analy- bringing together and analysing existing molec- ses of these morphological and other phenotypic ular and morphological data. We compiled and traits gradually becoming incorporated during analysed published and new mitochondrial and more recent times (e.g., Herrmann, Joger and nuclear DNA sequences for this group to re- Nilson, 1992; Nilson and Andrén, 2001). construct phylogenetic relationships among cur- Nine species were named since 2001 (Uetz, rently recognized species. We also compiled in- Freed and Hošek, 2019; table 1). Three of them formation on external morphology, as well as Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 9 on the criteria used for species delimitation, in and phylogeographic studies (e.g., Ursenbacher et al., 2006, order to assess the morphological distinctive- 2008a, b; Velo-Antón et al., 2012; Zinenko et al., 2015; Stümpel et al., 2016; Mizsei et al., 2017; Freitas et al., 2018; ness of currently recognized species. Our ob- Martínez-Freiría et al., 2020) to ensure that the selected se- jectives are: 1) to provide an updated taxon- quences represent the genetic structure and units reported in omy of Eurasian vipers; 2) to evaluate the valid- those studies. Mitochondrial DNA sequences were available ity of some of the newest species designations for a total of nine gene fragments, as well as the whole mitochondrial DNA genome for D. siamensis (mislabelled as under the unified species concept and recom- D. russelii in GenBank). For phylogenetic analysis, we se- mend an appropriate status reflecting our cur- lected DNA fragments from a subset of seven mtDNA mark- rent data base; 3) to identify remaining knowl- ers with a higher representation across species: CR (control region), COI (Cytochrome c oxidase subunit I), cyt b (cy- edge gaps and the research required to achieve tochrome b), ND2 (NADH dehydrogenase subunit 2), ND4 a robust, stable and evidence-based taxonomic (NADH dehydrogenase subunit 4), ND5 (NADH dehydro- framework for this group of vipers. genase subunit 5) and 16S (mitochondrial gene coding for 16S rRNA). Additionally, we also provide unpublished DNA sequences generated for other studies to complement available Material and methods data from GenBank. Sequences were concatenated when they originate from the same lineage or geographic local- Taxonomic inference ity, using SequenceMatrix software (Vaidya, Lohman and Meier, 2011). Since Eurasian viper species generally con- Our taxonomic evaluation is built upon an integrative and tain geographically cohesive, parapatric mitochondrial lin- evolutionary framework, based on the unified species con- eages, it is unlikely that they generate chimeras when concept (de Queiroz, 2007), under which species are defined catenating sequences from multiple individuals from the as separately evolving lineages and their biological proper- same geographic locality. The final dataset included 97 units ties (e.g., monophyly, reproductive isolation, differentiated (representing several lineages for 39 of the 40 recognized ecological niches, morphological distinctiveness) as “oper- species) with sequences ranging from 654 to 4621 base ational criteria” that ultimately provide evidence for their pairs (bp). Details of sequences used in mtDNA analyses are separation through a variety of methods. We employ the in- available in supplementary table S1. Sequences for nuDNA tegrative approach of Padial et al. (2010) by using phylo- were available for 17 gene fragments, from which we se- genetic analysis of mitochondrial (mtDNA) sequences, the lected six protein-coding genes, BDNF (Brain Derived Neu- most widely available and standardised marker, to identify rotrophic Factor), CMOS (Oocyte maturation factor Mos), divergent lineages, that are then tested, to the extent that MC1R (Melanocortin 1 Receptor), NT3 (Neurotrophin-3), the available data allow, with additional data, in particular PRLR (Prolactin Receptor), RAG1 (Recombination Acti- single-copy nuclear DNA (nuDNA) markers and morpholo- vating protein 1) and one intron, B-fib (Beta-fibrinogen in- gical data. We set a sequence divergence percentage to pro- tron 7), as they allowed the most comprehensive taxonomic pose a threshold for provisional taxonomic categorization coverage (24 species, 61.5% of the currently recognized to- (see phylogenetic inference section for more details). Above tal). This set includes the most relevant nuclear genes used this threshold, evolutionary lineages may simply confirm es- to support the designation of some of the most recently rec- tablished species, or if not described as such, they may be ognized species (e.g., RAG1, Ghielmi et al., 2016; NT3, considered as candidate species and should be targeted in fu- Mizsei et al., 2017). Details on nuDNA sequences used are ture studies to evaluate their taxonomic status. If currently provided in supplementary table S2. Sequences were man- recognized species are composed of several divergent lin- ually aligned and edited using Geneious v 4.8.5 (Kearse et eages, we classify them as a Likely Species Complex (LSC). al., 2012). For the nuclear genes, haplotype phases were pro- Below the cyt b threshold, currently recognized species are duced by a coalescent-based Bayesian reconstruction imple- categorized as Doubtful Valid as Species (i.e., DVAS), un- mented in PHASE (Stephens, Smith and Donnelly, 2001) til there is enough evidence to indicate species integrity. In available in DNAsp (Librado and Rozas, 2009). addition, we used the “pending” category to indicate that Phylogenetic relationships and time of divergence be- single-species recognition is currently maintained, despite tween species were inferred using a Bayesian Inference (BI) incongruences with the divergence threshold delimitation, method implemented in BEAST v 1.7.5 (Drummond et al., but further research is recommended to assess taxonomic 2012) on the concatenated mtDNA dataset. An exhaustive arrangements. search with PartitionFinder 1.1.1 (Lanfear et al., 2012) was conducted to select appropriate partitioning schemes and Phylogenetic inferences evolutionary models based on the Bayesian Information Cri- terion (BIC). The GTR + G + I model applied to all mtDNA We searched on GenBank for mtDNA and nuDNA gene fragments combined in a single partition was determined as sequences representing all the relevant lineages within the best-fit model and partitioning scheme. Eurasian vipers (see supplementary table S1 and S2). Se- Substitution rates were estimated under a strict molecular lection of sequences was based on published phylogenetic clock (Drummond et al., 2006) that assumes uniform rates Downloaded from Brill.com04/29/2020 02:27:11PM via free access 10 I. Freitas et al. across branches. A Yule model, most suitable for species- Morphological characterization level phylogenies, was implemented as tree prior. Since the fossil record of Eurasian vipers is fairly incomplete From the published literature, we compiled a list of the crite- and does not provide reliable and verified calibration dates ria used to identify each of the currently recognized species. (see Stümpel et al., 2016), our molecular dating strategy In addition, we gathered information on the variability of relied on secondary calibrations, including the splits of 14 external morphological traits in each species, includ- Vipera-Daboia and Macrovipera-Montivipera, dated at 26 ing one biometric, 11 pholidotic and two dorsal colour pat- Mya (Zheng and Wiens, 2016 – but see Šmíd and Tolley, tern characters. Information from the literature was collated 2019). We used a lognormal prior with a mean of 26.3 Mya with data from specimens measured by the authors in the and a standard deviation of 0.07 to constrain node ages. field or in museum collections. This allowed us to estab- Three independent runs of 100 million generations were lish the range of variation of different morphological traits, performed, sampling trees and parameter estimates every in some cases for each sex separately (see Results). In ad- 10 000 generations with 10% of the trees discarded as burn- dition, whenever possible, and in order to more accurately in. Convergence was verified by looking at the effective represent morphological variation in some pholidotic traits, sample sizes of all parameters (ESS > 300) using Tracer modal or mean values were retrieved for specific groups or v1.7 (Rambaut et al., 2018). Trees obtained from multiple subspecies within each species. Furthermore, we included independent runs were then combined using LogCombiner verbal descriptions of colouration, which provide an idea of v 1.7.5. (Drummond et al., 2012) and summary trees were variation in visually striking qualitative traits. generated with TreeAnnotator v1.7.1 (Drummond et al., 2012). Haplotype networks for the cyt b gene (the most widely used marker across studies) were reconstructed with TCS Results v 1.21 (Clement, Posada and Crandall, 2000), with a 90% parsimony connection limit; and the graphical output was Phylogenetic inferences from mtDNA visualized in TCSBU (dos Santos et al., 2015). The initial alignment of 1141 bp was trimmed down to 196 bp with We obtained a well-supported (most major no missing data across 96 units. The short length of the single sequence available for V. shemakhensis precluded its nodes with posterior probability > 0.95) mtDNA inclusion in the haplotype network. However, it was directly phylogenetic reconstruction for the 97 taxa compared to the remaining sequences (180 bp of overlap). (fig. 1, supplementary fig. S1). Three main Uncorrected p-distances between taxa were estimated based on the same cyt b fragment using MEGA ver. 5 (Tamura et clades are recognized, corresponding to gene- al., 2011). We propose the use of this short standardized cyt ra Daboia, Vipera and Macrovipera + Mon- b fragment as a candidate DNA barcode for the delimitation tivipera. Diversification times varied widely, of evolutionary units in Eurasian vipers (see Hebert and Grégory, 2005). Furthermore, we recommend a value of starting with Daboia (mean time to most re- 5% cyt b sequence divergence as a provisional threshold, cent common ancestor, TMRCA = 20.97 Mya), below which an untested species-level designation appears followed by Vipera (TMRCA = 16.65 Mya), inappropriate or premature, unless other lines of evidence would validate species classification. The cyt b-threshold and later the split between Macrovipera and results from two facts: 1) a 5% threshold provides a good Montivipera (TMRCA = 12.86 Mya). Diver- delimitation of currently recognized species within Eurasian sification within Macrovipera and Montivipera vipers, also recognizing deep evolutionary lineages within them; 2) cyt b divergence levels are consistently higher started later, around 8 Mya. than 5% between closely related species that co-exist with Vipera is the most diverse genus and includes restricted or no hybridization, e.g. sympatric species of three subclades: (1) Pelias, comprising one lin- the genus Vipera differ by more than 10% (Tarroso et al., 2014; Mebert et al., 2015b), sympatric watersnakes Nerodia eage with V. berus (including V. barani nested fasciata and N. sipedon differ by 9% (Mebert, 2008), Natrix in it) and V. seoanei, and another lineage with helvetica and N. natrix by 6.9% (Kindler et al., 2017) and V. renardi, V. ursinii, V. kaznakovi, V. graeca, 9% difference between Montivipera wagneri and M. raddei with no signs of mixing along a sharp contact line (Mebert V. sakoi, V. darevskii, V. walser and V. ana- et al., 2016; Stümpel et al., 2016). tolica (as well as other species nested inside Haplotype networks were drawn for each of the seven V. renardi, V. kaznakovi and V. darevskii); (2) nuclear genes following the same procedure. Sequences with high proportion of missing data (>30% of the to- Vipera 1, including V. aspis and the V. latastei- tal length) were excluded from the dataset. For B-fib and monticola complex; and (3) Vipera 2, com- RAG1, datasets were divided in two sets each based on prising the Vipera ammodytes-transcaucasiana sequence length (B-fib-1, B-fib-2 and RAG1-1, RAG1-2) and analysed independently to avoid excluding shorter se- complex. Diversification within each subclade quences from the haplotype networks. is <10 Mya, with Vipera 1 being the oldest Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 11

Figure 1. Dated Bayesian phylogenetic tree, obtained from the concatenated dataset of seven mitochondrial gene fragments, showing relationships for Eurasian vipers. The four genera (in uppercase) and major subclades (in italics) are highlighted. Lineages are grouped and distinctively coloured considering a divergence equal or higher than 5% for a small fragment (196 bp) of cyt b (supplementary table S3). Black and light-grey dots represent posterior probabilities higher than 0.95 and between 0.9-0.95, respectively. Names of taxa and lineages are given accordingly to publications from where sequences were retrieved. See supplementary table S1 for details.

Downloaded from Brill.com04/29/2020 02:27:11PM via free access 12 I. Freitas et al.

(TMRCA = 9.11 Mya), followed by Pelias Phylogenetic inferences from nuDNA (TMRCA = 6.6 Mya) and Vipera 2 (TMRCA = 5.27 Mya). Haplotype networks constructed for the seven The second most diverse clade is Mon- nuclear genes show a pattern of wide haplo- tivipera, composed of two subclades: (1) rad- type sharing and few mutational steps among dei, which diversified ca. 1.68 Mya, and (2) species; however, some species present dis- xanthina, which diversified in three lineages ca. tinct, well differentiated haplotypes (fig. 2; sup- 4.78 Mya. Macrovipera is the least diverse clade plementary table S4). Haplotype networks for and includes two subclades, M. razii and M. BDNF, CMOS, MC1R and RAG1 (both sets) lebetina + M. schweizeri; the latter diverged show very low levels of variability. Unique hap- about 2.62 Mya. Daboia includes three sub- lotypes were found for some species such as V. clades, corresponding to D. russelii + D. sia- berus and V. eriwanensis in BDNF, M. lebetina, mensis, D. palaestinae and D. mauritanica.The M. raddei and D. siamensis in CMOS, V. as- latter two are sister species (TMRCA = 16.17 pis in MC1R, or V. ammodytes, V. berus and Mya). Divergence of D. russelii and D. siamen- V. walser in RAG1. However, haplotype sharing V. ursinii V. renardi V. sis is estimated about 9.3 Mya. was found between , and graeca in BDNF, between V.berus, V.eriwanen- The range of divergence times between sis and V. aspis in CMOS, and between Mon- species and species complexes (i.e., group of tivipera species in MC1R and RAG1. Similarly, closely related species) is highly variable (fig. some phylogenetically distant species (accord- 1), e.g., 9.1 Mya (V. aspis from V. latastei- ing to mtDNA and previous multilocus phylo- monticola), 5.54 Mya (V. anatolica from the genies), such as M. lebetina and M. raddei or D. remaining kaznakovi-ursinii), 4.74 Mya (V. siamensis and V. ursinii have very similar hap- seoanei from V. berus-barani), 3.79 Mya (V. lotypes (one or two mutational steps different) ursinii from V. renardi complex), 2.02 (V. eriwa- in the BDNF network; the same occurs for M. nensis from V. renardi)or1.38Mya(V. barani lebetina and Montivipera spp. in the set 2 of from V. berus). The TMRCA between species RAG1. is found to be very recent for the pairs V. altaica Haplotype networks for B-fib, PRLR and V. renardi V. orlovi V. kaznakovi with –E, with – NT3 present higher variability than the other se- Russia, and V. olguni with V. darevskii kumlu- lected markers (fig. 2). The B-fib shows unique, tasi (fig. 1). Several species were recovered as well-separated haplotypes (more than three mu- paraphyletic (e.g., M. xanthina and V. lotievi)or tational steps) for D. mauritanica, D. siamen- polyphyletic (e.g., V. kaznakovi). sis, M. lebetina and V. aspis. However, B-fib- Haplotype networks based on the 196 bp 1 haplotypes are very similar in V. seoanei and cyt b fragment recover similar genetic relation- V. aspis, and identical in North African and ships to the mtDNA phylogenetic tree (supple- Iberian populations of V. latastei,aswellas mentary fig. S2). The 82 identified haplotypes in V. monticola. The PRLR dataset displays and matrices of uncorrected genetic distances unique, well-differentiated haplotypes in most based on these cyt b fragments are provided in taxa: V. berus, V. aspis, V. renardi, V. seoanei, supplementary material (supplementary tables D. mauritanica, D. siamensis, M. lebetina,some S1 and S3, respectively). The single sequence samples of V. ursinii and some samples of V. available for V. shemakhensis differed in one po- latastei from Iberia. However, it shows haplo- sition from the sequences of V. ebneri and V. eri- type sharing between V. ursinii and V. graeca, wanensis. between V. latastei from Iberia and North Africa Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 13

Figure 2. Haplotype networks for seven nDNA genes. B-ﬁb and RAG1 genes were divided in two sets each (see text and supplementary tables S2-S3 for details). as well as V. monticola, and between M. xan- variability of species, 13 assessing phylogenetic thina and M. albizona. The NT3 dataset in- relationships among species, and six using phy- cludes samples of several Vipera species, M. le- logenetic inferences and phenotypic characteri- betina, M. raddei, M. xanthina, D. mauritan- zation to describe taxa (see supplementary ref- ica and D. siamensis. It shows haplotype shar- erences S1). One book (Phelps, 2010) was used ing among species of the same genus (e.g., V. to extract maximum body size for some species aspis – V. latastei-monticola – V. seoanei, V. for which other published data were lacking. In- ursinii – V. renardi), and even of distinct ge- formation retrieved from publications, comple- nera (Macrovipera lebetina – Montivipera rad- mented with data collected by the authors from dei). Some species such as D. mauritanica, museum specimens, is shown in supplementary M. xanthina, V. berus, V. renardi, V. graeca table S5. or D. siamensis have distinct haplotypes, the The most commonly used criteria to diagnose two latter species being well separated from the species relied on head (24 species) and body rest. (26 species) pholidosis, and dorsal colouration (27 species). In contrast, phylogenetic analyses Morphological characterization based on molecular data were initially consid- We obtained data from 39 studies providing ered for species description in only ﬁve cases morphological descriptions or addressing the (table 1, supplementary table S5). Downloaded from Brill.com04/29/2020 02:27:11PM via free access 14 I. Freitas et al.

Data compilation showed that species within criteria used to identify them. By assessing phy- Macrovipera and Daboia are the largest in body logenetic and morphological variability, we pro- size, followed by Montivipera and with Vipera vide recommendations and future research di- being the smallest. Sexual dimorphism in body rections for robust species delimitation, which size is reported in most species (supplementary will aid the advancement towards a more in- table S5). formed and coherent taxonomy for this group Regarding pholidosis, some traits of head of vipers. scalation (e.g., canthal, supralabial or infral- abial scales) exhibit low variation, particularly Phylogenetic inference from mtDNA within each genus (supplementary table S5). Phylogenetic analyses based on seven mtDNA Modal or mean values of other head traits, how- fragments produced a mostly resolved topology ever, show important variation within genera that strongly supports the monophyly of the four (e.g., apical and intercanthal + intrasupraocu- genera (fig. 1). Phylogenetic relationships and lar scales within Vipera) and among them (e.g., divergence dates estimated in this study mostly loreal scales; fig. 3). Modal or mean values of agree with those reported in previous works body scalation exhibit variation among species (e.g., Alencar et al., 2016; Stümpel et al., 2016; (e.g., ventral and subcaudal scales; fig. 5) or ge- Zheng and Wiens, 2016), with the exception of nera (e.g., number of dorsal rows; fig. 4). Nev- divergence dates recently reported using alter- ertheless, the ranges of variation for most pholi- native time calibration procedures (Šmíd and dotic traits overlap extensively among species Tolley, 2019). of the same genus and even among distinct ge- Overall, our inferences suggest distinct di- nera (figs 3, 4). Sexual dimorphism in subcau- versification dates and divergence levels for dal scale counts is also mirrored in ventral scale each genus. Diversification within Daboia is variation in many species (fig. 4). estimated in the early Miocene, followed by With respect to dorsal colouration, there is the diversification within Vipera in the middle Macrovipera Montivipera high variability at both the inter- and intraspe- Miocene; and appear as the most recent genera, diverging from each cific levels, particularly in Vipera species, which other in the late Miocene (fig. 1). Despite its display more dorsal marks and a higher number old origin, Daboia shows low diversity levels, of distinct dorsal pattern types than species of with only four species recognized: two tropical the other genera (supplementary table S5). Asian species, D. russelii and D. siamensis, and two Mediterranean species, D. mauritanica and D. palaestinae. These species, however, are ex- Discussion tremely divergent, with genetic distances based on a small (196 bp), variable fragment of cyt b Eurasian vipers are a taxonomically challeng- ranging from 9%, between the Asian relatives ing group due to the long-standing use of di- D. russelii – D. siamensis, to 21%, between D. verse species criteria and the prevalence of siamensis and the Mediterranean D. palaesti- morphology-based classifications and species nae (supplementary table S3). The old diver- delimitations. Here, we inferred the molecular sification and wide distributional range of this phylogeny of Eurasian vipers based on mito- genus, together with the spatial gaps separating chondrial data, encompassing for the first time species ranges, suggest that this group likely ex- almost all the described species and evolution- perienced major extinctions along its evolutionary lineages within this group. We also gathered ary history, with only some of the representative information on the morphological variability of taxa currently persisting. This suggestion is sup- currently recognized species and reviewed the ported by the occurrence of Daboia-like vipers Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 15

Figure 3. Variation across species and genera in three pholidotic head traits (number of apical, intercanthal + intrasupraocular, and loreal scales). For each species, variation range (minimum-maximum) is represented as a vertical line. Modal, in apical and loreal scales, and mean values, in intercanthal + intrasupraocular scales, are represented as grey squares. Both values were retrieved for speciﬁc groups or subspecies (supplementary table S5). Species are displayed and grouped according to mtDNA phylogenetic relationships in ﬁg. 1.

Downloaded from Brill.com04/29/2020 02:27:11PM via free access 16 I. Freitas et al.

Figure 4. Variation across species and genera in three pholidotic body traits (number of ventral and subcaudal scales, and dorsal rows). For each species, variation range (minimum-maximum) is represented as a vertical line. Mean values are represented as circles (females) and rhomboids (males) in ventral and subcaudal scales, while modal values are represented as a grey square in dorsal rows. Both values were retrieved for speciﬁc groups or subspecies (supplementary table S5). Species are grouped according to mtDNA phylogenetic relationships in ﬁg. 1.

Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 17 in the fossil record from the Miocene in areas (ca. 1.68 Mya, 2% genetic divergence within such as Western Europe, where this group is the group) and an older divergence with higher currently absent (Paleobiology Database, 2019). levels of genetic diversity within the xanthina- With a more recent origin, Macrovipera also complex, including the bornmuelleri-group (ca. shows low diversity levels, comprising three de- 4.78 Mya, 6% genetic divergence within the en- scribed species (fig. 1): M. schweizeri, from tire complex). Major branches within the latter the western Cyclades, M. lebetina, widely dis- group are supported by high posterior proba- tributed from central Asia to the Middle East, bilities, except for the monophyly of M. xan- and M. razii from southern and central Iran. thina, for which low branch support was already High levels of polymorphism, especially in shown in Stümpel et al. (2016). colouration, led to the description of distinct Among all analysed genera, Vipera is the subspecies within M. lebetina (Nilson and An- most diverse and well-studied group. Exten- drén, 1988; Stümpel and Joger, 2009). Molec- sive phylogeographic and phylogenetic work ular studies have suggested the validity of four has been conducted within this genus (e.g., Gar- subspecies, i.e., lebetina, obtusa, turanica and cernovi (Stümpel and Joger, 2009), but rejected rigues et al., 2005; Ursenbacher et al., 2006a, the species status of M. schweizeri, which is b, 2008; Velo-Antón et al., 2012; Zinenko et suggested to be a subspecies of M. lebetina al., 2015, 2016; Martínez-Freiría et al., 2020). with a wider distribution than previously con- However, no previous study has included all sidered (Lenk et al., 2001; Stümpel and Joger, known taxa (e.g., 12 taxa in Zheng and Wiens, 2009). Our phylogenetic analyses recover M. 2016; 19 taxa in Šmíd and Tolley, 2019). Al- schweizeri and M. lebetina as a single unit (2% though traditionally grouped in two subgen- of cyt b genetic distance; supplementary ta- era (Garrigues et al., 2005), Vipera forms three ble S3). Shared mitochondrial haplotypes be- well-supported monophyletic groups (fig. 1): tween Macrovipera schweizeri and some south- Pelias, Vipera 1 and Vipera 2 (see Nilson and ern Turkish populations of M. lebetina further Andrén, 1997 for a similar designation). Both corroborate conspecificity (Stümpel and Joger, Vipera 1 and Vipera 2 present deep phyloge- 2009). Oraie et al. (2018) found cytochrome netic structure and high divergence between b divergences between moderate and high (up and within taxa. Vipera 1 includes the west- to 4.4%) among Iranian populations of M. le- ern and Mediterranean V. aspis and V. latastei- betina obtusa and M.l. cernovi and between monticola, and Vipera 2 is represented by V. these and specimens from Uzbekistan and Azer- ammodytes-transcaucasiana from the Balkans, baijan, suggesting that further diversity may Turkey and Asia Minor. V. latastei-monticola exist within this species. Moreover, increased and V. ammodytes-transcaucasiana are species geographical sampling may uncover additional complexes that comprise highly divergent units phylogenetic diversity, as has been the case with (9% and 6% of cyt b genetic distance within the description of M. razii from Iran, previously allocated to M. lebetina (Oraie et al., 2018). each complex, respectively; supplementary ta- The genus Montivipera consists of two well- ble S3). Phylogeographic studies on these taxa supported species complexes, the xanthina and show high levels of geographically structured raddei clades. This genus was recently sub- genetic diversity and the oldest divergences jected to a comprehensive phylogenetic study among Eurasian vipers, with main diversifica- based on a multilocus mitochondrial and nu- tion events likely occurring during late Miocene clear dataset for all its constituent taxa (Stümpel and early Pliocene (Ursenbacher et al., 2008; et al., 2016). Here we confirm previous findings Velo-Antón et al., 2012; Freitas et al., 2018; of a more recent origin of the raddei-complex Martínez-Freiría et al., 2020). Downloaded from Brill.com04/29/2020 02:27:11PM via free access 18 I. Freitas et al.

On the other hand, Pelias is the most diver- are still unclear but a historical occurrence of in- sified group in the phylogeny, with multiple de- trogressive hybridization with asymmetric mi- scribed species. In general, genetic divergence tochondrial DNA capture could explain this pat- within this group is shallower than that observed tern (e.g., Barbanera et al., 2009). The poly- within the other Vipera subclades, with many phyletic status of V. lotievi, on the other hand, taxa likely resulting from geographic splits dur- is thought to be the result of possible confusion ing Pleistocene climatic oscillations (see Zi- in the identification of species due to morpho- nenko et al., 2015). Our phylogeny recovered logical convergence or hybridization and intro- two highly divergent subclades (TMRCA = 7 gression leading to admixture of traits (Zinenko Mya, fig. 1; 9% genetic distance, supplemen- et al., 2015, 2016). tary table S3). One of the subclades includes Mito-nuclear discordance and low resolution V. berus (with V. barani nested within it) and of nuDNA V. seoanei. The other clade is highly diversified and includes three major groups: (1) V. renardi The nuclear data do not follow the same pattern (with V. shemakhensis, V. lotievi and V. altaica observed for mtDNA. The haplotype networks nested within it) and V. eriwanensis (with V. constructed for each nuDNA marker show a ebneri) as sister group, together with V. ursinii pattern of widespread haplotype sharing among and V. kaznakovi from Russia (V. orlovi,isan distant species, and even among genera, with admixed population of V. kaznakovi and V. re- extremely low levels of genetic variation (fig. nardi, Zinenko et al., 2016; and V. dinniki is 2), suggesting incomplete lineage sorting of an- nested within V. kaznakovi from Russia), and V. cestral polymorphism (Wan et al., 2004) or ex- graeca;(2)V. sakoi, V. kaznakovi from Georgia, tremely low levels of sequence evolution. Nu- V. darevskii (with V. olguni nested within) and V. clear genes were already known to provide in- walser; and (3) V. anatolica, which appears as a sufficient resolution to infer the phylogenetic separate lineage at the root of this clade. variability within Eurasian viper taxa (e.g., V. latastei-monticola and V. aspis Our results support most taxa and recognized , Velo-Antón et al., 2012; Freitas et al., 2018; Martínez- complexes (e.g., V. latastei-monticola) as mono- Freiría et al., 2020; D. mauritanica, Martínez- phyletic lineages, with the exception of V. kaz- Freíria et al., 2017a; Montivipera taxa, Stüm- nakovi and V. lotievi. As already shown by Zi- pel et al., 2016). Yet, relying on mtDNA as the nenko et al. (2015) and Ghielmi et al. (2016), V. only source for phylogenetic inference could be kaznakovi from Russia (Greater Caucasus) and problematic (see Ballard and Whitlock, 2004), Georgia (Lesser Caucasus) appear in the phy- and thus, the phylogenetic patterns retrieved logeny as two polyphyletic lineages separated here should be interpreted with some caution. by a considerable genetic distance (5%), and V. The existence of local hybridization between lotievi is also polyphyletic, with several lineages highly differentiated species (e.g., Tarroso et al., included within V. renardi (fig. 1, supplemen- 2014; Guiller, Lourdais and Ursenbacher, 2017) tary fig. S2). Multilocus RAD-sequencing data and the growing evidence for extensive gene recovered low differentiation between both mi- flow in more recently differentiated taxa (Zi- tochondrial lineages within V. kaznakovi (Olek- nenko et al., 2015, 2016) highlight the impor- sandr Zinenko, unpublished data), a pattern that tance of nuDNA analyses. However, despite be- is further supported by low morphological dif- ing commonly used in species delimitation stud- ferentiation and the existence of a continuous ies, the sequenced nuDNA markers were too area of suitable habitats connecting these popu- conserved to consistently differentiate between lations (Orlov and Tuniyev, 1990). Reasons for otherwise well supported species of Eurasian the discordance between mtDNA and nuDNA vipers. This applies even to introns and other Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 19 fast-evolving single copy nuclear genes (e.g., not representative region and/or particular diag- NT3, PRLR, Townsend et al., 2008). Conse- nostic traits (e.g., M. albizona Nilson, Andrén quently, other molecular approaches such as and Flärdh, 1990; M. kuhrangica Rajabizadeh, even faster evolving markers (e.g., Kindler and Nilson and Kami, 2011), robust studies using Fritz, 2018; Pöschel et al., 2018), phylogenomic multivariate comparative analyses of morpho- methods (e.g., Blair et al., 2019; Heinicke et al., logical traits have shown morphological dis- 2018), and increased sampling to evaluate cur- tinctiveness of some species (e.g., V. graeca, rent or past gene flow between most-proximate Nilson and Andrén, 2001; M. razii, Moradi, populations or contact zones of two or more Rastegar-Pouyani and Rastegar-Pouyani, 2014), closely related species (Mebert, 2008, 2015a, b, reinforcing the view that the combination of 2020; Hillis, 2019) should be favoured to infer evolutionary histories and resolve species limits traits can in fact identify taxonomic units in among these taxa. some cases. On the other hand, the occurrence of morphologically-cryptic species, that have Taxonomic relevance of morphological traits only been identified after molecular phylogenetic analyses (e.g., V. walser Ghielmi et al., The value of morphological traits to define 2016), suggests that other factors may be af- species in many groups has become ques- fecting the external morphological variability of tionable after the emergence of DNA-based Eurasian vipers. methods in taxonomy. Morphological varia- Both pholidotic and colouration traits fre- tion across populations often reflects local quently display geographic variation associ- adaptation processes or phenotypic plastic- ated with environmental gradients, reflecting ity, rather than historical relationships (e.g., Kaliontzopoulou et al., 2011; Kaliontzopoulou, adaptive processes (e.g., Shine, 2000; Sanders, Carretero and Llorente, 2012; Alhajeri, Hunt Malhotra and Thorpe, 2004; Martínez-Freiría and Steppan, 2015). Additionally, lack of mor- et al., 2009; Tomovic,´ Crnobrnja-Isailovic´ and phological differentiation does not necessarily Brito, 2010; Martínez-Freiría and Brito, 2013). imply shared evolutionary histories, as is the The role of local adaptation in shaping in- case in cryptic species (e.g., Bickford et al., traspecific morphological differentiation has 2007; Kaliontzopoulou et al., 2011; Ghielmi et been highlighted in multiple studies on reptile al., 2016) or taxa displaying convergent evolu- species (e.g., Thorpe and Baez, 1993; Malho- tion (e.g,. Harmon et al., 2005). tra and Thorpe, 1997; Kaliontzopoulou, Pinho The taxonomy of Eurasian vipers has been and Martínez-Freiría, 2018). In particular, traits traditionally based on differences in pholidotic related to fitness frequently present variation (head and body scalation) and colouration (dor- across different environmental and ecological sal pattern and colour) traits. Our data compila- conditions in order to meet the species-specific tion shows that the ranges of variation of most needs and enhance performance and fitness pholidotic traits overlap extensively among (Arnold, 1983; Kingsolver and Huey, 2003). In species of the same genus (figs 4, 5), while vipers, for instance, differences in dorsal pat- colouration traits exhibit high variability at both the inter- and intraspecific levels (supplemen- tern colouration can be an adaptive response to tary table S5). This apparent low prevalence of temperature gradients, enhancing thermoregula- diagnostic traits for taxonomic purposes must tion capabilities, or to predation pressures, lead- be taken with caution given the limitations of ing to aposematic signals or increased substrate- our data compilation. While some species de- crypsis (Wüster et al., 2004; Valkonen et al., scriptions relied on few specimens from a small, 2011; Santos et al., 2014; Dubey et al., 2015; Downloaded from Brill.com04/29/2020 02:27:11PM via free access 20 I. Freitas et al.

Martínez-Freiría et al., 2017). Increasing or de- recent species descriptions maintained the tra- creasing scale numbers within species can in- ditional methods of species delimitation, dis- fluence water loss along environmental gradi- regarding known limitations (e.g., Vipera al- ents (Malhotra and Thorpe, 1997; Sanders, Mal- taica Tuniyev, Nilson and Andrén, 2010; Mon- hotra and Thorpe, 2004) or enhance locomo- tivipera kuhrangica Rajabizadeh, Nilson and tion over distinct substrates (Kelley, Arnold and Kami, 2011; Vipera olguni Tuniyev et al., 2012; Gladstone, 1997). These traits can also be plas- Vipera shemakhensis Tuniyev et al., 2013). In tic and depend on the thermal conditions experi- agreement with previous molecular studies, our enced by the embryos during gestation. For ex- phylogenetic reconstruction shows a clear mis- ample, both field and experimental studies indi- match between relevant evolutionary units and recognized species. This is particularly evi- cate that thermal conditions at early embryonic dent within Vipera, in which species complexes stages influence the number of ventral scales, such as V. latastei-monticola or V. ammodytes- scale abnormalities, as well as dorsal coloura- transcaucasiana include highly divergent lin- tion in vipers (Lourdais et al., 2004; Lorioux et eages from the Miocene, while other species al., 2013). Additionally, reproductive programs are much younger (from late Pleistocene), poly- in low-effective-size populations have shown phyletic (e.g., V. lotievi) or are nested within that inbreeding can lead to subsequent decrease others (e.g., V. altaica, V. shemakhensis). This of scale counts in viper offspring (Üveges et puzzling scenario mainly results from the indis- al., 2012). Altogether, these studies suggest that criminate use of morphological traits and geo- morphological traits might exhibit high inter- graphic isolation as exclusive criteria for species population variability and little taxon-specific delimitation (see table 1). Polymorphism as a variation and thus, should no longer be used as result of local adaptation and plasticity may the only source of data to inform taxonomic de- lead to taxonomic inflation, whereas low mor- cisions in Eurasian vipers. phological variability can hamper the identification of cryptic species, such as in the case Taxonomic inflation and need for an integrative of the V. latastei-monticola and V. ammodytes- transcaucasiana complexes. One example that taxonomy illustrates well the problem of high morphological variation within species is the case of V. The taxonomy of Eurasian vipers has long been lotievi, for which morphological convergence under intense debate, especially as it provides across similar environments and confusion over crucial underpinnings to the formulation of con- species identification are highlighted as a pos- servation management strategies and the allo- sible explanation for its polyphyly (Zinenko et cation of economic resources for this purpose. al., 2015). Similarly, the occurrence of eco- Species have been described based on diverse types can lead to the designation of taxonomic criteria, and mostly using morphology and/or units which are not concordant with evolution- geographic isolation as the only source of in- ary history (e.g., V. aspis atra, Ursenbacher et ference. Not surprisingly, recent phylogenetic al., 2006b; V. aspis montecristi, Barbanera et al., studies have often revealed major inconsisten- 2009; Luiselli et al., 2015; V. monticola, Velo- cies in relation to current taxonomic units (e.g., Antón et al., 2012). Additionally, genetic intro- Ferchaud et al., 2012; Velo-Antón et al., 2012; gression was also shown to be a confounding Zinenko et al., 2015, 2016; Stümpel et al., factor on species classification within the Pelias 2016; Martínez-Freiría et al., 2017), also un- subgenus, leading to intermediate phenotypes in veiling the existence of morphologically cryp- admixed populations (e.g., the description of V. tic taxa (e.g., Ghielmi et al., 2016). Yet, many magnifica and V. orlovi; Zinenko et al., 2016). Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 21

Due to these limitations, an increasing num- are sometimes constrained by low financial sup- ber of studies have applied molecular methods port and by the scarcity or remoteness of popu- to inform taxonomic decisions. The amount of lations (e.g., Göçmen et al., 2014a, b, 2017; Fre- sequence divergence among groups has been itas et al., 2018), the systematic situation of the extensively used as a criterion to delimit taxa group will likely remain unresolved in the im- (e.g., in small mammals, Bradley and Baker, mediate future. 2001; in rat snakes, Hofmann et al., 2018). Al- Species classification is not the sole contro- though the use of a standard percentage of se- versial issue in Eurasian viper taxonomy; al- quence divergence to separate species is debat- location below (i.e., subspecies) or above (i.e., able, in this work, we draw an arbitrary but genera) species level is even more problematic conservative (in terms of recognizing most de- due to the lack of objective, operational con- scribed species) threshold for the delimitation cepts defining these ranks. While our phylogenetic inferences are directed towards assessing of evolutionary units in Eurasian vipers cor- the validity of species as independently evolv- responding to an uncorrected genetic distance ing lineages, some provisional consensus can equal or higher to 5% for a 196 bp cyt b be reached at higher hierarchical levels. For in- fragment (fig. 1; supplementary table S3). This stance, if Pelias is used to refer to Euro-Siberian strategy highlights the distinct levels of diver- Vipera species (as proposed in several studies, gence for the currently recognized species, al- see Tuniyev et al., 2009, 2012, 2013, 2018a, lowing the formulation of recommendations for b; Avcı et al., 2010), the designation of an ad- specific status of already described species (ta- ditional genus within the current Vipera and ble 1). Furthermore, our barcoding approach, the separation of Daboia into two or three ge- using a short cyt b fragment that fully recovers nera would be required to reflect equivalent all phylogenetic units (supplementary fig. S2), phylogenetic distances (see fig. 1; supplemen- may be a useful tool to assess the distinctiveness tary table S3). However, in order to avoid ill- of newly discovered populations that are sus- founded splitting procedures and even further pected to represent new taxa (Hebert and Gre- confusion, we advise against taking such steps gory, 2005). (see also Vences et al., 2013), and argue that the Ultimately, the taxonomic status of candidate use of subgenera may be a better way of pro- species should be best addressed in an integra- viding names for clades without disrupting the tive fashion, that is, by searching for concordant binomial nomenclature (Wallach, Wüster and differences in genetic, morphological and eco- Broadley, 2009). logical traits (Padial et al., 2010). Since mtDNA could be particularly misleading in vipers, both by inflating the number of taxa or by missing Concluding remarks lineages that have lost their mtDNA due to in- In this work, we integrate currently available in- trogression (e.g., Zinenko et al., 2016), we en- formation on the phylogenetic and morpholo- courage the use of multilocus genetic data (e.g., gical variability of Eurasian vipers to advance UCE loci, Blair et al., 2019). Until now, among into a more coherent and objective taxonomy all Eurasian vipers, only three species have been for this group. Based on our integrative assess- described using integrative approaches (i.e., by ment, we provide recommendations on the spe- addressing phylogenetic divergence and char- cific status of 40 described species and propose acterizing phenotypic variability; Vipera walser some guidelines to clarify the taxonomic status Ghielmi et al., 2016; Macrovipera razii Oraie et of some of them (table 1). Species complexes al., 2018; Vipera sakoi Tuniyev et al., 2018). In such as V. latastei-monticola or V. ammodytes- the absence of more integrative studies, which transcaucasiana require further analyses on the Downloaded from Brill.com04/29/2020 02:27:11PM via free access 22 I. Freitas et al. extent of gene flow among distinct lineages PORTUGAL 2020 Partnership Agreement, through the Eu- to delineate taxonomic units, while some of ropean Regional Development Fund (ERDF) within the scope of the project AGRIGEN-NORTE-01-0145-FEDER- the currently recognized species, described us- 000007. KM received financial support by funds from the ing morphological data only (e.g., V. altaica, Mohamed bin Zayed Species Conservation Fund, project V. magnifica, V. shemakhensis), must be explic- nos. 13057971 (2014), 150510677 (2015), 160513040 itly regarded as non-valid species due to low (2016), 170516395 (2017/18), 190520941 (2019), but also by the JCE private funding and the German Herpetologi- genetic differentiation in relation to other pre- cal Society DGHT (Deutsche Gesellschaft für Herpetolo- viously recognized species. Again, other taxa, gie und Terrarienkunde) via the Wilhelm Peters Fond 2013 such as eastern M. xanthina and V. sakoi re- and the branch in Zürich, Switzerland. OZ research was supported by the Volkswagen foundation (Project I/83 987) quire more extensive geographic sampling to ar- and the Ministry of Education and Science of Ukraine rive at robust conclusions. As discussed above, (grant 0117U004836). NO is partly supported by grant integrative taxonomic approaches bringing to- RFBR 19-04-00119 and ZISP AAAA-A19-119020590095- gether independent evidence and using different 9. JCI was funded by MESTD Republic of Serbia (grant ref. 173025). IF is financed by FCT through a grant methodological approaches, particularly incor- (SFRH/BD/148514/2019), and JCB, GV-A, AK and FM-F porating genomic data instead of relying solely are financed by FCT through contracts (refs. FCT-DL57, on mitochondrial data, will allow the robust de- IF/01425/2014, IF/00641/2014/CP1256/CT0008 and DL57/ 2016/CP1440/CT0010, respectively). lineation of coherent evolutionary units in a unified taxonomic framework.

Previous attempts to propose priorities for the Supplementary material. Supplementary material is avail- conservation of vipers (i.e., Maritz et al., 2016) able online at: suffered from taxonomic inflation and lack of https://doi.org/10.6084/m9.figshare.12044412 a geographic comprehensive sampling scheme (see table 1). It is striking that more than half of the Eurasian viper species listed as globally en- References dangered in our assessment (nine of 16 species Afasr, M., Yakin, B.Y., Çiçek, K., Dinçer, A. (2019): A with categories CR, EN and VU; table 1) were new subspecies of Ottoman viper, Montivipera xanthina classed as of doubtful validity as species. Ad- (Gray, 1849), (Squamata: Viperidae) from Geyik Moun- tains, Mediterranean Turkey. Ecologica Montenegrina vancing in a robust, evidence-based designation 22: 214-225. of taxonomic units is therefore essential for the Alencar, L.R., Quental, T.B., Grazziotin, F.G., Alfaro, M.L., future development of conservation strategies Martins, M., Venzon, M., Zaher, H. (2016): Diversifica- aimed to anticipate threats related to anthro- tion in vipers: phylogenetic relationships, time of divergence and shifts in speciation rates. Mol. Phylogenetics pogenic factors, while slowing down or even Evol. 105: 50-62. stopping the rapid decline of many populations Alhajeri, B.H., Hunt, O.J., Steppan, S.J. (2015): Molecular of Eurasian vipers. systematics of gerbils and deomyines (Rodentia: Gerbil- linae, Deomyinae) and a test of desert adaptation in the tympanic bulla. J. Zool. Syst. 53: 312-330. Arnold, S.J. (1983): Morphology, performance and fitness. Acknowledgements. Authors acknowledge curators from Am. Zool. 23: 347-361. the following museum collections: Museo Nacional de Avcı, A., Ilgaz, Ç., Ba¸skaya,S., ¸ Baran, I., Kumluta¸s, Y. Ciencias Naturales – CSIC (Madrid, Spain), Natural His- (2010): Contribution to the distribution and morphology tory Museum (London, UK), National Museum of Prague of Pelias darevskii (Vedmederja, Orlov and Tuniyev (Check Republic), Muséum national d’Histoire naturelle 1986) (Reptilia: Squamata: Viperidae) in northeastern (Paris, France) and Zoological Research Museum Alexan- Anatolia. Russ. J. Herpetol. 17:1-7. der Koenig (Bonn, Germany). This work was supported by Ballard, J.W., Whitlock, M.C. (2004): The incomplete nat- SYNTHESYS (projects ref. ES-TAF-5874, CZ-TAF-6627, ural history of mitochondria. Mol. Ecol. 13: 729-744. FR-TAF-6626, GB-TAF-5886), FEDER (COMPETE) and Baran, I., Joger, U., Kutrup, B., Türkozan, O. (2001): On Portuguese Foundation for Science and Technology (FCT) new specimens of Vipera barani Böhme and Joger, 1983, funds within the scope of project PTDC/BIA-EVL/28090/ from northeastern Anatolia, and implications for the 2017-POCI-01-0145-FEDER-028090, and Norte Portugal validity of Vipera pontica Billing, Nilson and Sattler, Regional Operational Programme (NORTE2020), under the 1990 (Reptilia, Viperidae). Zool. Middle East. 23: 47-53. Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 23

Barbanera, F., Zufﬁ, M.A., Guerrini, M., Gentilli, A., To- Eiselt, J., Baran, I. (1970): Ergebnisse zoologischer Sam- fanelli, S., Fasola, M., Dini, F. (2009): Molecular phy- melreisen in der Türkei: Viperidae. Ann Nat Hist Mus logeography of the asp viper Vipera aspis (Linnaeus, Wien 74: 357-369. 1758) in Italy: evidence for introgressive hybridization Ferchaud, A.-L., Ursenbacher, S., Luiselli, L., Jelic,´ D., and mitochondrial DNA capture. Mol. Phylogenetics Halpern, B., Major, A., Kotenko, T., Crnobrnja-Isailovic, Evol. 52: 103-114. J., Tomovic,´ L., Ghira, I., Ioannidis, Y., Arnal, V., Mont- Bickford, D., Lohman, D.J., Sodhi, N.S., Rahayu, D.L., gelard, C. (2012): Phylogeography of the Vipera ursinii Meier, R., Winker, K., Das, I. (2007): Cryptic species complex (Viperidae): mitochondrial markers revealed an as a window on diversity and conservation. Trends Ecol. east-west disjunction in the palearctic region. J. Bio- Evol. 22: 148-155. geogr. 39: 1836-1847. Billing, H., Nilson, G., Satler, U. (1990): Vipera pontica sp. Freitas, I., Fahd, S., Velo-Antón, G., Martínez-Freiría, F. n., a new viper species in the kaznakovi group (Reptilia, (2018): Chasing the phantom: biogeography and con- Viperidae) from northeastern Turkey and adjacent Tran- servation of Vipera latastei-monticola in the Maghreb scaucasia. Zoologica Scripta. 19: 227-231. (North Africa). Amphibia-Reptilia. 39: 145-161. Blair, C., Bryson Jr, R.W., Linkem, C.W., Lazcano, D., Frost, D.R., Hillis, D.M. (1990): Species in concept and Klicka, J., McCormack, J.E. (2019): Cryptic diversity practice: herpetological applications. Herpetologica. 46: in the Mexican highlands: thousands of UCE loci help 87-104. illuminate phylogenetic relationships, species limits and Garrigues, T., Dauga, C., Ferquel, E., Choumet, V., Failloux, divergence times of montane rattlesnakes (Viperidae: A.B. (2005): Molecular phylogeny of Vipera Laurenti, Crotalus). Mol Ecol. Resour. 19: 349-365. 1768 and the related genera Macrovipera (Reuss, 1927) Boettger, O. (1890): Eine neue Viper aus Armenien. Zool. and Daboia (Gray, 1842), with comments about neuro- Anz. 13: 62-64. toxic Vipera aspis aspis populations. Mol. Phylogenetics Böhme, W., Joger, U. (1984): Eine neue Art des Vipera Evol. 35: 35-47. berus-Komplexes aus der Turkei. Amphibia-Reptilia. 4: Ghielmi, S., Menegon, M., Marsden, S.J., Laddaga, L., 265-271. Ursenbacher, S. (2016): A new vertebrate for Europe: Bonaparte, L. (1835): Iconographia della Fauna Italica per the discovery of a range-restricted relict viper in the le quattro classi degli animali vertebrati. Tome. Rome.: western Italian Alps. J. Zool. Syst. 54: 161-173. 1832-1841. Göçmen, B., Mebert, K., I˙gci,˘ N., Akman, B., Yıldız, M.Z., Boscá, E. (1878): Note sur une forme vouvelle ou peu Oguz,˘ M.A., Altın, Ç. (2014a): New locality records connue de vipère. Bull. Zool. Soc. France 3: 116-121. for four rare species of vipers (Reptilia: Viperidae) in Boulenger, G.A. (1896): Catalogue of the Snakes in the Turkey. Zool. Middle East. 60: 306-313. British Museum Natural History, vol. III. British Mu- Göçmen, B., Mebert, K., Karı¸s, M., Oguz,˘ M.A., Ursen- seum (Natural History), London. bacher, S. (2017): A new population and subspecies Boulenger, G.A. (1913): On the geographical races of of the critically endangered Anatolian meadow viper Vipera ammodytes. Ann. Mag. Nat. Hist. 11: 283-287. Vipera anatolica Eiselt and Baran, 1970 in eastern An- Bradley, R.D., Baker, R.J. (2001): A test of the genetic talya province. Amphibia-Reptilia. 38: 289-305. species concept: cytochrome-b sequences and mammals. Göçmen, B., Mulder, J., Karı¸s, M., Oguz,˘ M.A. (2014b): J. Mammal. 82: 960-973. The poorly known Anatolian Meadow Viper, Vipera Christoph, H. (1861): Pelias renardi mihi. Bull. Soc. Nat anatolica: new morphological and ecological data. Her- Moscow 34: 599-606. petologica Romanica. 8: 1-10. Clement, M., Posada, D., Crandall, K.A. (2000): TCS: a Gray, J.E. (1849): Catalogue of the specimens of snakes in computer program to estimate gene genealogies. Mol. the collection of the British Museum. Edward Newman, Ecol. 9: 1657-1659. London 15: 1-125. Dayrat, B. (2005): Towards integrative taxonomy. Biol J Guiller, G., Lourdais, O., Ursenbacher, S. (2017): Hy- Linnean Soc. 85: 407-417. bridization between a euro-Siberian (Vipera berus)anda De Queiroz, K. (2007): Species concepts and species delim- para-Mediterranean viper (V. aspis) at their contact zone itation. Syst. Biol. 56: 879-886. in western France. J Zool. 302: 138-147. dos Santos, A.M., Cabezas, M.P., Tavares, A.I., Xavier, R., Harmon, L.J., Kolbe, J.J., Cheverud, J.M., Losos, J.B. Branco, M. (2015): tcsBU: a tool to extend TCS network (2005): Convergence and the multidimensional niche. layout and visualization. Bioinformatics 32: 627-628. Evolution. 59: 409-421. Drummond, A.J., Ho, S.Y., Phillips, M.J., Rambaut, A. Hebert, P.D., Gregory, T.R. (2005): The promise of DNA (2006): Relaxed phylogenetics and dating with conﬁ- barcoding for taxonomy. Syst. Biol. 54: 852-859. dence. PLoS Biol. 4: 699. Heinicke, M.P., Lemmon, A.R., Lemmon, E.M., McGrath, Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A. K., Hedges, S.B. (2018): Phylogenomic support for evo- (2012): Bayesian phylogenetics with BEAUti and the lutionary relationships of new world direct-developing BEAST 1.7. Mol. Biol. Evol. 29: 1969-1973. frogs (Anura: Terraranae). Mol. Phylogenetics Evol. Dubey, S., Zwahlen, V., Mebert, K., Monney, J.-C., Golay, 118: 145-155. P., Ott, T., Durand, T., Thiery, G., Kaiser, L., Geser, Herrmann, H.W., Joger, U., Nilson, G. (1992): Phylogeny S., Ursenbacher, S. (2015): Diversifying selection and and systematics of viperine snakes. III: resurrection of color-biased dispersal in the asp viper. BMC Evol Biol. the genus Macrovipera (Reuss, 1927) as suggested by 15: 99. biochemical evidence. Amphibia-Reptilia. 13: 375-392. Downloaded from Brill.com04/29/2020 02:27:11PM via free access 24 I. Freitas et al.

Hillis, D.M. (2019): Species delimitation in herpetology. Lanfear, R., Calcott, B., Ho, S.Y., Guindon, S. (2012): Parti- J. Herpetol. 53: 3-12. tionFinder: combined selection of partitioning schemes Hofmann, S., Mebert, K., Schulz, K.D., Helfenberger, N., and substitution models for phylogenetic analyses. Mol. Göçmen, B., Böhme, W. (2018): A new subspecies of Biol. Evol. 29: 1695-1701. Zamenis hohenackeri (Strauch, 1873) (Serpentes: Col- Lataste, F. (1879): Diagnose dúne vipère nouvelle d’És- ubridae) based on morphological and molecular data. pagne. Bull. Soc. zool. Fr. 4: 132. Zootaxa. 4471: 137-153. Lenk, P., Kalyabina, S., Wink, M., Joger, U. (2001): Evo- Joger, U. (2005): Montivipera Nilson, Tuniyev, Andrén, lutionary relationships among the true vipers (Reptilia: Orlov, Joger und Herrmann, 1999. In: Handbuch der Viperidae) inferred from mitochondrial DNA sequences. Reptilien und Amphibien Europas, Schlangen (Ser- Mol. Phylogenetics Evol. 19: 94-104. pentes), p. 61-62. Joger, U., Stümpel, N., Eds, Aula- Librado, P., Rozas, J. (2009): DnaSP v5: asoftware for Verlag, Wiesbaden. comprehensive analysis of DNA polymorphism data. Kaiser, H., Crother, B.I., Kelly, C.M., Luiselli, L., O’Shea, M., Ota, H., Passos, P., Schleip, W.D., Wüster, W. Bioinformatics. 25: 1451-1452. (2013): Best practices: in the 21st century, taxonomic Linnaeus, C. (1758): Systema naturæ per regna tria naturæ, decisions in herpetology are acceptable only when sup- secundum classes, ordines, genera, species, cum char- ported by a body of evidence and published via peer- acteribus, differentiis, synonymis, locis. Tomus I. Editio review. Herpetol. Rev. 44: 8-23. decima, reformata. Laurentii Salvii, Holmiæ. 10th Edi- Kaliontzopoulou, A., Carretero, M.A., Llorente, G.A. tion: 824. (2012): Morphology of the Podarcis wall lizards (Squa- Lorioux, S., Vaugoyeau, M., DeNardo, D.F., Clobert, J., mata: Lacertidae) from the Iberian Peninsula and north Guillon, M., Lourdais, O. (2013): Stage dependence of Africa: patterns of variation in a putative cryptic species phenotypical and phenological maternal effects: insight complex. Zool. J. Linn. Soc. 164: 173-193. into squamate reptile reproductive strategies. Am. Nat. Kaliontzopoulou, A., Pinho, C., Harris, D.J., Carretero, 182: 223-233. M.A. (2011): When cryptic diversity blurs the picture: a Lourdais, O., Shine, R., Bonnet, X., Guillon, M., Naulleau, cautionary tale from Iberian and north African Podarcis G. (2004): Climate affects embryonic development in a wall lizards. Biol. J. Linn. Soc. 103: 779-800. viviparous snake, Vipera aspis.Oikos.104: 551-560. Kaliontzopoulou, A., Pinho, C., Martínez-Freiría, F. (2018): Luiselli, L., Petrozzi, F., Mebert, K., Zuffi, M.A.L., Where does diversity come from? Linking geographical Amori, G. (2015): Resource partitioning and dwarfism patterns of morphological, genetic, and environmental patterns between sympatric snakes in a micro-insular variation in wall lizards. BMC Evol. Biol. 18: 124. Mediterranean environment. Ecol. Res. 30: 527-535. Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, Malhotra, A., Thorpe, R.S. (1997): Microgeographic varia- M., Sturrock, S., Buxton, S., Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P., Drum- tion in scalation of Anolis oculatus (Dominica, West In- mond, A. (2012): Geneious basic: an integrated and ex- dies): a multivariate analysis. Herpetologica. 53: 49-62. tendable desktop software platform for the organization Mallow, D., Ludwig, D., Nilson, G. (2003): True Vipers: and analysis of sequence data. Bioinformatics. 28: 1647- Natural History and Toxinology of Old World Vipers. 1649. Krieger Publishing Company. Kelley, K.C., Arnold, S.J., Gladstone, J. (1997): The effects Maritz, B., Penner, J., Martins, M., Crnobrnja-Isailovic,´ J., of substrate and vertebral number on locomotion in the Spear, S., Alencar, L.R., Sigala-Rodriguez, J., Messen- garter snake Thamnophis elegans. Funct. Ecol. 11: 189- ger, K., Clark, R.W., Soorae, P., Luiselli, L. (2016): Iden- 198. tifying global priorities for the conservation of vipers. Kindler, C., Fritz, U. (2018): Phylogeography and taxonomy Biol Cons. 204: 94-102. of the barred grass snake (Natrix helvetica), with a dis- Martínez-Freiría, F., Brito, J.C. (2013): Integrating classical cussion of the subspecies category in zoology. Vertebr. and spatial multivariate analyses for assessing morpho- Zool. 68: 269-281. logical variability in the endemic Iberian viper Vipera Kindler, C., Chèvre, M., Ursenbacher, S., Böhme, W., Hille, seoanei. J. Zool. Syst. Evol. 51: 122-131. A., Jablonski, D., Vamberger, M., Fritz, U. (2017): Hy- Martínez-Freiría, F., Crochet, P.A., Fahd, S., Geniez, P., bridization patterns in two contact zones of grass snakes Brito, J.C., Velo-Antón, G. (2017a): Integrative phylo- reveal a new central European snake species. Sci. Rep. geographical and ecological analysis reveals multiple 7: 1-12. Pleistocene refugia for Mediterranean Daboia vipers in Kingsolver, J.G., Huey, R.B. (2003): Introduction: the evo- north-west Africa. Biol. J. Linn. Soc. 122: 366-384. lution of morphology, performance and fitness. Integr. Comp. Biol. 43: 361-366. Martínez-Freiría, F., Freitas, I., Zuffi, M., Golay, P., Ursen- Klemmer, K. (1963): Liste der rezenten Giftschlangen. In: bacher, S., Velo-Antón, G. (2020): Climatic refugia Die Giftschlangen der Erde, Behringwerke, Marburg, boosted allopatric diversification in western Mediter- p. 255-464. Elwert, N.G., Ed. ranean vipers. J. Biogeogr. DOI:10.1111/jbi.13861. Krecsák, L. (2007): An account of the generic and specific Martínez-Freiría, F., i de Lanuza, G.P., Pimenta, A.A., Pinto, names, and type specimens of viperid taxa described T., Santos, X. (2017b): Aposematism and crypsis are not by Albert Franz Theodor Reuss (Squamata: Viperidae). enough to explain dorsal polymorphism in the Iberian Zootaxa. 1514: 1-36. adder. Acta Oecol. 85: 165-173. Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 25

Martínez-Freiría, F., Santos, X., Pleguezuelos, J.M., Lizana, Nilson, G., Andren, C. (1985): Systematics of the Vipera M., Brito, J.C. (2009): Geographical patterns of morpho- xanthina complex (Reptilia: Viperidae). Taxonomic sta- logical variation and environmental correlates in contact tus of the Bulgar Dagh viper in south Turkey. J. Herpetol. zones: a multi-scale approach using two Mediterranean 19: 276-283. vipers (Serpentes). J. Zool. Syst. Evol. 47: 357-367. Nilson, G., Andrén, C. (1988): Vipera lebetina trans- Mayden, R.L. (1997): A hierarchy of species concepts: mediterranea, a new subspecies of viper from north the denouement in the saga of the species problem. In: Africa, with remarks on the taxonomy of Vipera lebetina Species: the Units of Diversity, p. 381-423. Claridge, and Vipera mauritanica. Bonn. zool. Beitr. 39: 371-379. M.F., Dawah, H.A., Wilson, M.R., Eds, Chapman & Nilson, G., Andrén, C. (1997): Evolution, systematics Hall. and biogeography of Palaearctic vipers. In: Venomous Mebert, K. (2008): Good species despite massive hybridiza- Snakes: Ecology, Evolution and Snakebite, p. 31-42. tion: genetic research on the contact zone between the Thorpe, R.S., Wüster, W., Malhorta, A., Eds, Oxford watersnakes Nerodia sipedon and N. fasciata in the Car- University Press. olinas, USA. Mol. Ecol. 17: 1918-1929. Nilson, G., Andrén, C. (2001): The meadow and steppe Mebert, K., Göçmen, B., Igci, N., Oguz, M.A., Karis, vipers of Europe and Asia – the Vipera (Acridophaga) M., Ursenbacher, S. (2015a): New records and search ursinii complex. Acta zool. Acad. Sci. Hung. 47: 87-267. for contact zones among parapatric vipers in the Nilson, G., Andrén, C., Flärdh, B. (1990): Vipera albizona, genus Vipera (barani, kaznakovi, darevskii, eriwanen- a new mountain viper from central Turkey, with com- sis), Montivipera (wagneri, raddei), and Macrovipera ments on isolating effects of the Anatolian “Diagonal”. (lebetina) in northeastern Anatolia. Herpetol Bull. 133: Amphibia-Reptilia. 11: 285-294. 13-22. Nilson, G., Tuniyev, B., Andrén, C., Orlov, N., Joger, U., Mebert, K., Göçmen, B., Igci, N., Karis, M., Oguz, M.A., Herrmann, H.-W. (1999): Taxonomic position of the Yildiz, M.Z., Teynié, A., Stümpel, M., Ursenbacher, Vipera xanthina complex. Kaupia 8: 99-102. S. (2020): Mountain vipers in central-eastern Turkey: Nilson, G., Tuniyev, B.S., Orlov, N., Hoggren, M., Andren, huge range extensions for four taxa reshape decades of C. (1995): Systematics of the vipers of the Caucasus: misleading perspectives. Herpetol Conserv Bio. 15. polymorphism or sibling species?. Asiat. Herpetol. Res. Mebert, K., Jagar, T., Grželj, R., Cafuta, V., Luiselli, L., 6: 1-16. Ostanek, E., Golay, P., Dubey, S., Golay, J., Ursenbacher, Obst, F.J. (1983): Zur Kentniss der Schlangengattung S. (2015): The dynamics of coexistence: habitat sharing Vipera. Zool. Abh. Staatl. Mus. Tierk. Dresden 38: 229- vs. segregation patterns among three sympatric montane 335. vipers. Biol. J. Linn. Soc. 116: 364-376. DOI:10.1111/ Oraie, H., Rastegar-Pouyani, E., Khosravani, A., Moradi, bij.12582. N., Akbari, A., Sehhatisabet, M.E., Shafiei, S., Stümpel, Merrem, B. (1820): Versuch eines Systems der Amphibien N., Joger, U. (2018): Molecular and morphological anal- I (Tentamen Systematis Amphibiorum). J. C. Kriegeri, yses have revealed a new species of blunt-nosed viper Marburg. of the genus Macrovipera in Iran. Salamandra. 54: 233- Mertens, R., Darevsky, I.S., Klemmer, K. (1967): Vipera 248. latifii, eine neue Giftschlange aus dem Iran. Senckenb Orlov, N.L., Tuniyev, B.S. (1990): Three species in the Biol. 48: 161-168. Vipera kaznakovi complex (Eurosiberian Group) in the Minton, S.A., Jr., Dowling, H.G., Russell, F.E. (1968): Caucasus: their present distribution, possible genesis Poisonous Snakes of the World. A Manual for Use by and phylogeny. Asiat. Herpetol. Res. 3: 1-36. U.S. Amphibious Forces, 2nd Edition. Department of Padial, J.M., Miralles, A., De la Riva, I., Vences, M. (2010): the Navy and Bureau of Medical Surgery, Washington. The integrative future of taxonomy. Front. Zool. 78: 16. Mizsei, E., Jablonski, D., Roussos, S.A., Dimaki, M., Ioan- Paleobiology Database (2019): The Paleo Biology Database nidis, Y., Nilson, G., Nagy, Z.T. (2017): Nuclear markers Navigator. Search on March 2019. Available at: support the mitochondrial phylogeny of Vipera ursinii- https://paleobiodb.org/navigator/. renardi complex (Squamata: Viperidae) and species sta- Phelps, T. (2010): Old world vipers: a natural history of the tus for the Greek meadow viper. Zootaxa. 4227: 75-88. Azemiopinae and Viperinae. Ed. Chimaira. Moradi, N., Rastegar-Pouyani, N., Rastegar-Pouyani, E. Pöschel, J., Heltai, B., Graciá, E., Quintana, M.F., Velo- (2014): Geographic variation in the morphology Antón, G., Arribas, O., Valdeón, A., Wink, M., Fritz, U., of Macrovipera lebetina (Linnaeus, 1758) (Ophidia: Vamberger, M. (2018): Complex hybridization patterns Viperidae) in Iran. Acta Herpetol. 9: 187-202. in European pond turtles (Emys orbicularis)inthePyre- Nikolsky, A.M. (1909): Eine neue Vipern-Art aus dem nean region. Sci. Rep. 8: 15925. Kaukasus: Vipera kaznakovi sp. nov. Tiflis Mitt. Kaukas. Rajabizadeh, M., Nilson, G., Kami, H.G. (2011): A new Mus. 4: 174. species of mountain viper (Ophidia: Viperidae) from the Nikolsky, A.M. (1913): Herpetologia caucasica. Tiflis. 272. central Zagros Mountains, Iran. Russ. J. Herpetol. 18: Nilson, G., Andren, C. (1984): Systematics of the Vipera 235-240. xanthina complex (Reptilia: Viperidae). An overlooked Rambaut, A., Drummond, A.J., Xie, D., Baele, G., Suchard, viper within the xanthina species-group in Iran. Bonn. M.A. (2018): Posterior summarization in Bayesian phy- zool. Beitr. 35: 175-184. logenetics using Tracer 1.7. Syst. Biol. 67: 901-904. Downloaded from Brill.com04/29/2020 02:27:11PM via free access 26 I. Freitas et al.

Reuss, T. (1933): Fortsetzung und Schluss der Original- Townsend, T.M., Alegre, R.E., Kelley, S.T., Wiens, J.J., berichte. Nachr. Bl. Aquar. Terrar.-Ver.: 372-373. Reeder, T.W. (2008): Rapid development of multiple Saint Girons, H. (1953): Une vipère naine: Vipera latastei nuclear loci for phylogenetic analysis using genomic montana. Bull. Soc. zool. Fr. 78: 24-28. resources: an example from squamate reptiles. Mol. Sanders, K.L., Malhotra, A., Thorpe, R.S. (2004): Ecolog- Phylogenetics Evol. 47: 129-142. ical diversification in a group of Indomalayan pitvipers Tuniyev, B.S., Ostrovskikh, S.V. (2001): Two new species (Trimeresurus): convergence in taxonomically important of vipers of “kaznakovi” complex (Ophidia, Viperinae) traits has implication for species identification. J. Evol. from the Western Caucasus. Russ. J. Herpetol. 8: 117- 17 Biol. : 721-731. 126. Santos, X., Vidal-García, M., Brito, J.C., Fahd, S., Llorente, Tuniyev, B.S., Avcı, A., Ilgaz, Ç., Olgun, K., Petrova, G.A., Martínez-Freiría, F., Parellada, X., Pleguezuelos, T.V., Bodrov, S.Y., Geniez, P., Teynié, A. (2018a): On J.M., Sillero, N. (2014): Phylogeographic and environ- taxonomic status of shield-head vipers from Turkish mental correlates support the cryptic function of the zigzag pattern in a European viper. Evol. Ecol. 28: 611- Lesser Caucasus and East Anatolia. Proceedings of the 626. Zoological Institute RAS. 322: 3-44. Schwarz, E. (1936): Untersuchungen über Systematik und Tuniyev, B.S., Iremashvili, G.N., Petrova, T.V., Verbreitung der europäischen und mediterranen Ottern. Kravchenko, M.V. (2018b): Rediscovery of the In: Die europäischen und mediterranen Ottern und ihre steppe viper in Georgia. Proceedings of the Zoological Gifte, p. 159-362. Behringwerke-Mitteilungen. Institute RAS. 322: 87-107. Shaw, G., Nodder, F.P., Eds (1797): The naturalist’s miscel- Tuniyev, B.S., Nilson, G., Andrén, C. (2010): A new species lany. London, Nodder & Co. 8: 255-300. of viper (Reptilia, Viperidae) from the Altay and Saur Shine, R. (2000): Vertebral numbers in male and female Mountains, Kazakhstan. Russ. J. Herpetol. 17: 110-120. snakes: the roles of natural, sexual and fecundity selec- Tuniyev, S.B., Avcı, A., Tuniyev, B.S., Agasian, A.L., tion. J. Evol. Biol. 13: 455-465. Agasian, L.A. (2012): Description of a new species of Simpson, G.G. (1961): Principles of Animal Taxonomy. shield-head vipers – Pelias olguni sp. nov. from Basin of Columbia University Press, USA. Upper Flow of the Kura River in Turkey. Russ. J. Her- Šmíd, J., Tolley, K.A. (2019): Calibrating the tree of vipers petol 19: 314-332. under the fossilized birth-death model. Sci. Rep. 9: 5510. Tuniyev, S.B., Orlov, N.L., Tuniyev, B.S., Kidov, F.F. Stephens, M., Smith, N.J., Donnelly, P. (2001): A new sta- (2013): On the taxonomical status of steppe viper from tistical method for haplotype reconstruction from popu- foothills of the south macroslope of the east Caucasus. lation data. Am. J. Hum. Genet. 68: 978-989. Russ. J. Herpetol. 20: 129-146. Stümpel, N. (2012): Phylogenie und Phylogeographie eura- Uetz, P., Freed, P., Hošek, J., Eds (2019): The Reptile sischer Viperinae unter besonderer Berücksichtigung der orientalischen Vipern der Gattungen Montivipera Database, http://www.reptile-database.org, accessed on und Macrovipera. Unpubl. PhD Thesis, TU Carolo- March 2019. Wilhelmina zu Braunschweig, Germany. Ursenbacher, S., Carlsson, M., Helfer, V., Tegelström, H., Stümpel, N., Joger, U. (2009): Recent advances in phy- Fumagalli, L. (2006a): Phylogeography and Pleistocene logeny and taxonomy of near and middle eastern refugia of the adder (Vipera berus) as inferred from vipers – an update. ZooKeys. 31: 179-191. mitochondrial DNA sequence data. Mol. Ecol. 15: 3425- Stümpel, N., Rajabizadeh, M., Avcı, A., Wüster, W., Joger, 3437. U. (2016): Phylogeny and diversification of mountain Ursenbacher, S., Conelli, A., Golay, P., Monney, J.-C., Zuffi, vipers (Montivipera, Nilson et al., 2001) triggered by M.A.L., Thiery, G., Durand, T., Fumagalli, L. (2006b): multiple Plio-Pleistocene refugia and high-mountain to- Phylogeography of the asp viper (Vipera aspis) inferred pography in the Near and Middle East. Mol. Phyloge- from mitochondrial DNA sequence data: evidence for netics Evol. 101: 336-351. multiple Mediterranean refugial areas. Mol. Phylogenet- Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., ics Evol. 38: 546-552. Kumar, S. (2011): MEGA5: molecular evolutionary ge- Ursenbacher, S., Schweiger, S., Tomovic,´ L., Crnobrnja- netics analysis using maximum likelihood, evolutionary Isailovic,´ J., Fumagalli, L., Mayer, W. (2008): Molec- distance, and maximum parsimony methods. Mol. Biol. ular phylogeography of the nose-horned viper (Vipera Evol. 28: 2731-2739. ammodytes, Linnaeus (1758)): evidence for high genetic Tarroso, P., Pereira, R.J., Martínez-Freiría, F., Godinho, diversity and multiple refugia in the Balkan Peninsula. R., Brito, J.C. (2014): Hybridization at an ecotone: Mol. Phylogenetics Evol. 46: 1116-1128. ecological and genetic barriers between three Iberian vipers. Mol. Ecol. 23: 1108-1123. Üveges, B., Halpern, B., Péchy, T., Posta, J., Komlósi, I. Thorpe, R.S., Baez, M. (1993): Geographic variation in (2012): Characteristics and heritability analysis of head scalation of the lizard Gallotia stehlini within the island scales of the Hungarian meadow viper (Vipera ursinii of Gran Canaria. Biol. J. Linn. Soc. 48: 75-87. rakosiensis, Méhely 1893). Amphibia-Reptilia. 33: 393- Tomovic,´ L., Crnobrnja-Isailovic,´ J., Brito, J.C. (2010): The 400. use of geostatistics and GIS for evolutionary history Vaidya, G., Lohman, D.J., Meier, R. (2011): SequenceMa- studies: the case of the nose-horned viper (Vipera am- trix: concatenation software for the fast assembly of modytes) in the Balkan Peninsula. Biol. J. Linn. Soc. multi-gene datasets with character set and codon infor- 101: 651-666. mation. Cladistics. 27: 171-180. Downloaded from Brill.com04/29/2020 02:27:11PM via free access Species delimitation in Eurasian vipers 27

Valkonen, J., Niskanen, M., Björklund, M., Mappes, J. J., Moore, A.G., Niskanen, M., Pollard, C.P. (2004): (2011): Disruption or aposematism? Significance of dor- Do aposematism and Batesian mimicry require bright sal zigzag pattern of European vipers. Evol. Ecol. 25: colours? A test, using European viper markings. Proc. 1047-1063. Royal Soc. B. 271: 2495-2499. Vedmederja, V.J., Orlov, N.L., Tuniyev, B.S. (1986): On Zheng, Y., Wiens, J.J. (2016): Combining phylogenomic taxonomy of three viper species of the Vipera kaznakowi complex. Dokl. Akad. Nauk SSSR. 157: 55-61. and supermatrix approaches, and a time-calibrated phy- Velo-Antón, G., Godinho, R., Harris, D.J., Santos, X., logeny for squamate reptiles (lizards and snakes) based Martínez-Freiria, F., Fahd, S., Pleguezuelos, J.M., Brito, on 52 genes and 4162 species. Mol. Phylogenetics Evol. J.C. (2012): Deep evolutionary lineages in a western 94: 537-547. Mediterranean snake (Vipera latastei/monticola group) Zinenko, O., Sovic, M., Joger, U., Gibbs, H.L. (2016): and high genetic structuring in southern Iberian popula- Hybrid origin of European Vipers (Vipera magnifica tions. Mol. Phylogenetics Evol. 65: 965-973. and Vipera orlovi) from the Caucasus determined using Vences, M., Guayasamin, J.M., Miralles, A., de la Riva, genomic scale DNA markers. BMC Evol Biol. 16: 76. I. (2013): To name or not to name: criteria to promote Zinenko, O., Stümpel, N., Mazanaeva, L., Bakiev, A., economy of change in Linnaean classification schemes. Zootaxa. 3636: 201-244. Shiryaev, K., Pavlov, A., Kotenko, T., Kukushkin, O., Wallach, V., Wüster, W., Broadley, D.G. (2009): In praise of Chikin, Y., Duisebayeva, T., Nilson, G. (2015): Mito- subgenera: taxonomic status of cobras of the genus Naja chondrial phylogeny shows multiple independent eco- Laurenti (Serpentes: Elapidae). Zootaxa. 2236: 26-36. logical transitions and northern dispersion despite of Wan, Q.H., Wu, H., Fujihara, T., Fang, S.G. (2004): Which Pleistocene glaciations in meadow and steppe vipers genetic marker for which conservation genetics issue?. (Vipera ursinii and Vipera renardi). Mol. Phylogenetics Electrophoresis 25: 2165-2176. Evol. 84: 85-100. Werner, F. (1898): Über einige neue Reptilien und einen Zuffi, M. (2002): A critique of the systematic position neuen Frosch aus dem cilicischen Taurus. Zool. Anz. 21: 217-223. of the asp viper subspecies Vipera aspis aspis (Lin- Werner, F. (1935): Reptilien der Ägäischen Inseln. Sitzungs- naeus, 1758), Vipera aspis atra Meisner, 1820, Vipera ber. Akad. Wiss. Wien. 144: 81-117. aspis francisciredi Laurenti, 1768, Vipera aspis hugyi Werner, F. (1938): Eine verkannte Viper (Vipera palaestinae Schinz, 1833 and Vipera aspis zinnikeri Kramer, 1958. n. sp.). Zool. Anz. 122: 313-318. Amphibia-Reptilia. 23: 191-213. Wiley, E.O. (1978): The evolutionary species concept recon- sidered. Systematic Zoology 21: 17-26. Submitted: August 13, 2019. Final revision received: Wüster, W., Allum, C.S.E., Bjargardóttir, I.B., Bailey, March 25, 2020. Accepted: March 26, 2020. K.L., Dawson, K.J., Guenioui, J., Lewis, J., Mcgurk, Associate Editor: Uwe Fritz.

Downloaded from Brill.com04/29/2020 02:27:11PM via free access