Usually, adaptive phenotypic differentiation is paralleled by genetic divergence between locally adapted populations. However, adaptation can also happen in a scenario of non-significant genetic divergence due to intense gene flow and/or recent differentiation. While this phenomenon is rarely published, findings on incipient ecologically-driven divergence or isolation by adaptation are relatively common, which could confound our understanding about the frequency at which they actually occur in nature. Here, we explore genome-wide traces of divergence between two populations of the lacertid lizard Psammodromus algirus separated by a 600 m elevational gradient. These populations seem to be differentially adapted to their environments despite showing low levels of genetic differentiation (according to previously studies of mtDNA and microsatellite data). We performed a search for outliers (i.e. loci subject to selection) trying to identify specific loci with FST statistics significantly higher than those expected on the basis of overall, genome-wide estimates of genetic divergence. We find that local phenotypic adaptation (in terms of a wide diversity of characters) was not accompanied by genome-wide differentiation, even when we maximized the chances of unveiling such differentiation at particular loci with FST-based outlier detection tests. Instead, our analyses confirmed the lack of differentiation on the basis of more than 70,000 SNPs, which is concordant with a scenario of local adaptation without any degree of isolation by environment. Our results add evidence to previous studies in which local adaptation does not lead to any kind of isolation (or early stages of ecological speciation), but maintains phenotypic divergence despite the lack of a differentiated genomic background.

During geographical expansion of a species individual colonizers have to confront different ecological challenges, and the capacity of the species to broaden its range may depend on the total amount of adaptive genetic variation supplied by evolution. We set out to test whether the distribution of loci under selection along a contrasting environmental gradient can be turned into a model that accurately predicts a species' range. If positive, this may shed light on the genetic source of adaptive limits that shape range boundaries. We sampled five populations of the western Mediterranean lizard Psammodromus algirus that inhabit a noticeable environmental gradient of temperature and precipitation. We used 21 SNPs putatively under selection to correlate the genotypes of 95 individuals with environmental variation among their populations, using 1x1 km2 grid cells as sampling units. By extrapolating the resulting model to all possible combinations of alleles, we inferred the locations that were theoretically suitable for the species. The inferred distribution range overlapped to a large extent with the realized range of the species, including an accurate prediction of internal gaps and range borders. Our results suggest an adaptability threshold determined by the amount of genetic variation available that would be required to warrant adaptation beyond a certain limit of environmental variation. These results support the idea that the expansion of a species' range may be ultimately linked to the arising of new variants under selection.

During geographical expansion of a species individual colonizers have to confront different ecological challenges, and the capacity of the species to broaden its range may depend on the total amount of adaptive genetic variation supplied by evolution. We set out to test whether the distribution of loci under selection along a contrasting environmental gradient can be turned into a model that accurately predicts a species' range. If positive, this may shed light on the genetic source of adaptive limits that shape range boundaries. We sampled five populations of the western Mediterranean lizard Psammodromus algirus that inhabit a noticeable environmental gradient of temperature and precipitation. We used 21 SNPs putatively under selection to correlate the genotypes of 95 individuals with environmental variation among their populations, using 1x1 km2 grid cells as sampling units. By extrapolating the resulting model to all possible combinations of alleles, we inferred the locations that were theoretically suitable for the species. The inferred distribution range overlapped to a large extent with the realized range of the species, including an accurate prediction of internal gaps and range borders. Our results suggest an adaptability threshold determined by the amount of genetic variation available that would be required to warrant adaptation beyond a certain limit of environmental variation. These results support the idea that the expansion of a species' range may be ultimately linked to the arising of new variants under selection.