New insights on phylogeography and distribution of painted frogs (Discoglossus) in northern Africa and the Iberian Peninsula.

Painted frogs (Discoglossus) contain five to six species of Western Palearctic anurans that are mainly distributed in allopatry. We here provide the first comprehensive assessment of the phylogeography of the Moroccan species D. scovazzi and geographically characterize its contact zone with D. pictus in Eastern Morocco. Discoglossus scovazzi shows, in general, a weak phylogeographic structure across Morocco on the basis of mitochondrial DNA sequences of the cytochrome b gene, with only populations centered in the Atlas Mountains characterized by the presence of slightly divergent haplotypes. In eastern Morocco, all populations east of the Moulouya River were clearly assignable to D. pictus. This species was also found along the Mediterranean coast west of the Moulouya, in the cities of Nador and Melilla, suggesting that not the river itself but the wide arid valley extending along much of the river (except close to the estuary) acts as a possible distributional barrier to these frogs. No sympatry of D. scovazzi with D. pictus was observed, and all specimens were concordantly assigned to either species by DNA sequences of cytochrome b and of the nuclear marker RAG 1. Species distribution models of the two taxa show largely overlapping areas of suitable habitat, and the two species' niches are significantly more similar than would be expected given the underlying environmental differences between the regions in which they occur. Comparative data are also presented from the southern Iberian contact zone of D. galganoi galganoi and D. g. jeanneae. These taxa showed less clear-cut distributional borders, extensively shared RAG1 haplotypes, and had instances of sympatric occurrence on the basis of cytochrome b haplotypes, in agreement with the hypothesis of a yet incomplete speciation. In this wide contact zone area we found mitochondrial sequences containing double peaks in electropherograms, suggesting nuclear pseudogenes or (less likely) heteroplasmy, possibly related to the ongoing admixture among the lineages. [ABSTRACT FROM AUTHOR]