Cryptic diversity within the Anatololacerta species complex (Squamata: Lacertidae) in the Anatolian Peninsula: Evidence from a multi-locus approach

摘要

The rapid development of innovative molecular tools for characterizing biodiversity is leading to an extensive and sometimes unexpected renovation of taxonomic classifications. Particularly, for species having allopatric or parapatric distributions or resulting from recent speciation processes, the absence of clear phenotypic differentiation may hinder the recognition of closely related taxa, while intraspecific polymorphism may be confused with the presence of more than one single species. In the present work, we apply different phylogenetic methods in order to infer relationships within the genus Anatololacerta, and to assess the taxonomy of this morphologically diversified group of lizards endemic to western and southern Anatolia and some neighboring Aegean islands. According to morphology, three species have been recognized (Anatololacerta anatolica, A. oertzeni and A. danfordi) as well as several subspecies, but small variation at immunological markers led some authors to join all the populations into one single taxon, A. danfordi. By selecting both mitochondrial and nuclear informative markers, we tested the effectiveness of classical "gene tree" (i.e. Bayesian Inference) vs. innovative (i.e. coalescent-based) "species tree" methods in resolving the Anatololacerta taxonomic enigma, as a case in point for similar studies on species complexes resulting from non-obvious and cryptic diversification patterns. According to our results, the gene tree method failed in resolving phylogenetic relationships among clades, whereas the multi-locus species tree approach, coupled with species delimitation methods, allowed the identification of four well distinct species. These species probably diversified in different allopatric refugia located in southern and western Anatolia, where isolated populations may have persisted during Pleistocene glacial cycles. (C) 2014 Elsevier Inc. All rights reserved.