Historical climate changes and orogenesis are two important factors that have shaped intraspecific biodiversity patterns worldwide. Although southern South America has experienced such complex events, there is a paucity of studies examining the effects on intraspecific diversification in this part of the world. Liolaemus pictus is the southernmost distributed lizard in the Chilean temperate forest, whose genetic structure has likely been influenced by Pleistocene glaciations. We conducted a phylogeographic study of L. pictus in Chile and Argentina based on one mitochondrial and two nuclear genes recovering two strongly divergent groups, Northern and Southern clades. The first group is distributed from the northernmost limit of the species to the Araucanía region while the second group is distributed throughout the Andes and the Chiloé archipelago in Southern Chile. Our results suggest that L. pictus originated 751 Kya, with divergence between the two clades occurring in the late Pleistocene. Demographic reconstructions for the Northern and Southern clades indicate a decrease in effective population sizes likely associated with Pleistocene glaciations. Surprisingly, patterns of genetic variation, clades age and historical gene flow in populations distributed within the limits of the Last Glacial Maximum (LGM) are not explained by recent colonization. We propose an “intra-Andean multiple refuge” hypothesis, along with the classical refuge hypothesis previously proposed for the biota of the Chilean Coastal range and Eastern Andean Cordillera. Our hypothesis is supported by niche modelling analysis suggesting the persistence of fragments of suitable habitat for the species within the limits of the LGM ice shield. This type of refuge hypothesis is proposed for the first time for an ectothermic species.

The present distribution of Patagonian species is the result of a complex history involving Quaternary refugial populations, Holocene range expansions and demographic changes occurring during the Anthropocene. Invasive salmonids were introduced in Patagonia during the last century, occupying most rivers and lakes, preying on and competing with native species, including the fishGalaxias platei. Here, we usedG. plateias a case study to understand how long-term (i.e. population differentiation during the Holocene) and short-term historical processes (salmonid introductions) affect genetic diversity. Using a suite of microsatellite markers, we found that the number of alleles is negatively correlated with the presence of salmonids (short-term processes), withG. plateipopulations from lakes with salmonids exhibiting significantly lower genetic diversity than populations from lakes without salmonids. Simulations (100 years backwards) showed that this difference in genetic diversity can be explained by a 99% reduction in population size. Allelic richness and observed heterozygosities were also negatively correlated with the presence of salmonids, but also positively correlated with long-term processes linked to Quaternary glaciations. Our results show how different genetic parameters can help identify processes taking place at different scales and their importance in terms of conservation.