CONICET | Buscador de Institutos y Recursos Humanos

Premise of research. The geographical distribution of ploidy levels provides insights into evolutionary pathways. For the subtropical tree line species Polylepis australis (Rosaceae), we tested the hypotheses that (1a) incidence of polyploidy is higher in the northern parts than in the southern parts of the species range due to the presence of related species that might favor hybridization (allopolyploidy), (1b) incidence of polyploidy is higher in the southern part of the range because the species here presumably reaches the limit of its environmental tolerance (autopolyploidy), and (2) ploidy levels increase with elevation, as polyploids are believed to perform better in stressful environments. Methodology. We used flow cytometry to assess the ploidy levels of 361 individuals from 27 populations across most of the distribution range of the species in two disjunct Argentinean high mountain regions. Pivotal results. The northern stands had lower ploidy levels (diploid) than the southern populations, in which we found mainly tetraploids intermixed with diploids, triploids, and a single hexaploid. We did not find any environmental correlates to the geographical distribution of ploidy levels. Conclusions. Polyploidy appears to have arisen in P. australis via autopolyploidy, either twice in different parts of the range or, more likely, once followed by long-distance dispersal. This is also supported by amplified fragment length polymorphism (AFLP) data from a previous study that confirmed higher numbers of AFLP fragments and private bands in the southern populations. The checkerboard distribution of ploidy levels in the southern Sierras de Co´ rdoba may represent a time capture of the spread of the polyploid condition. We propose that polyploidy represents a key factor to understanding the high morphological variation in P. australis and should be taken into account in ongoing reforestation activities.