Alectoris

Abstract: The nominal subspecies of rock partridge (Alectoris graeca graeca) is widely distributed in Greece, where populations are declining due to over-hunting and habitat changes. Captive-reared chukars (A. chukar) have been massively released throughout the country, raising fear that introgressive hybridisation might have disrupted local adaptations leading to further population declines. In this study we used mtDNA control-region sequences and Bayesian admixture analyses of multilocus genotypes determined at eight microsatellite loci, to assess the extent of introgressive hybridisation in 319 wild rock partridges collected in Greece. A neighbour-joining tree split the mtDNA haplotypes into three strongly supported clades, corresponding to rock, red-legged (A.␣rufa) and chukar partridges. We did not detect any case of maternal introgression. In contrast, admixture analyses of microsatellite genotypes identified from four to 28 putative hybrids (according to different assignment criteria), corresponding to 1.2–8.8% of the samples, which were widespread throughout all the country. Power and limits of admixture analyses were assessed using simulated hybrid genotypes, which revealed that a small number of markers can detect all first and second generation hybrids (F1 and F2), and up to 90% of the first generation backrossess. Thus, the true proportion of recently introgressed rock partridges in Greece might be ca. 20%. These findings indicate that introgressive hybridisation is widespread, suggesting that released captive-bred partridges have reproduced and hybridised in nature polluting the gene pool of wild rock partridge populations in Greece.

Abstract: We used mitochondrial DNA control-region and microsatellite data to infer the evolutionary history and past demographic changes in 332 rock partridges (Alectoris graeca) sampled from throughout the species' distribution range, with the exception of the central Balkans region. Maternal and biparental DNA markers indicated concordantly that rock partridge populations are structured geographically (mtDNA phiST = 0.86, microsatellite FST = 0.35; RST = 0.31; P < 0.001). Phylogenetic analyses of 22 mtDNA haplotypes identified two major phylogroups (supported by bootstrap values = 93%), splitting partridges from Sicily vs. all the other sampled populations at an average Tamura-Nei genetic distance of 0.035, which corresponds to 65% of the average distance between closely related species of Alectoris. Coalescent estimates of divergence times suggested that rock partridges in Sicily were isolated for more than 200000 years. This deep subdivision was confirmed by multivariate, Bayesian clustering and population assignment analyses of microsatellite genotypes, which supported also a subdivision of partridges from the Alps vs. populations in the Apennines, Albania and Greece. Partridges in the Apennines and Albania-Greece were probably connected by gene flow since recently through a late Pleistocene Adriatic landbridge. Deglaciated Alps were probably colonized by distinct and, perhaps, not yet sampled source populations. Bottleneck and mismatch analyses indicate that rock partridges have lost variability through past population declines, and did not expand recently. Deglaciated areas could have been recolonized without any strong demographic expansion. Genetic data partially supported subspecies subdivisions, and allowed delimiting distinct conservation units. Rock partridges in Sicily, formally recognized as A. g. whitakeri, met the criteria for a distinct evolutionary significant unit.