Resumen de Genetic Diversity and Spatial Genetic Structure of Chamaedaphne calyculata (Ericaceae) at the Western Periphery in Relation to its Main Continuous Range in Eurasia
A previous phylogeography and genetic diversity study of Chamaedaphne calyculata (Ericaceae) showed that populations over its geographic range were strongly separated into two groups: a Eurasian/NW North American group and a NE North American one corresponding with the disjunct distribution of Sphagnum-dominated peatlands in north-western and central-eastern North America. Here, I have extended the survey and focused on the species� detailed postglacial origin and the effect of isolation on genetic diversity patterns, particularly within island-like populations at the western periphery of its range in Europe. Using AFLP markers, estimates of genetic diversity within 16 C. calyculata populations in the Eurasian group were low (percentage of polymorphic loci P PL=14.9�24.8 %, Nei�s gene diversity H=0.060�0.119). Genetic diversity patterns within this species did not support the hypothesis that genetic diversity decreases towards the periphery of the range. Bayesian clustering analysis showed that population-level admixture was present in almost all studied 16 populations, suggesting multi-directional gene flow. On the other hand, the majority of assigned individuals (ca. 98 % of individuals) were offspring of the original residents, confirming that C. calyculata populations in the present day acted as discrete genetic units both in its continuous range and at its western periphery, and that gene flow was historic rather than contemporary in Eurasia. There was no correlation between genetic and geographic distance in the Eurasian group (r=0.02, P>0.05, Mantel test) nor at the western periphery (r=0.15, P>0.05, Mantel test). The isolation-by-distance (IBD) scatterplot matched Hutchinson and Templeton�s interpretation (case III), and geographic distance between populations was not a reliable predictor of the degree of genetic differentiation between populations. It is suggested that the lack of IBD might be a result of random genetic drift in rather disconnected populations that have become increasingly fragmented relatively recently. Positive and significant relationships between genetic and geographic distance on a small population scale was the result of biparental inbreeding of C. calyculata and restricted seed rain. Despite sporadic generative reproduction and limited dispersal, the fine-scale genetic structure within populations has been maintained, even though population sizes have been reduced to small fragments in recent years.
