Using principle component analysis to compare genetic diversity across polyploidy levels within plant complexes: an example from British Restharrows (Ononis spinosa and Ononis repens).

The investigation of genetic diversity between related plant populations which differ in ploidy levels is problematic, with common statistical methods developed for diploids being inappropriate for polyploid species. Studies into gene flow in such complexes are critical and can shed light on the mechanisms that generate and maintain populations of different polyploidy levels. We have investigated the use of principle component (PCO) analysis as one approach to elucidate population structure within British Restharrows (Leguminsoae, Ononis spp). Restharrows were common agricultural weed species until the advent of mechanical ploughing and both diploid (2n=2x=30; O. spinosa and O. intermedia) and tetraploid (2n=4x=60; O. repens and O. maritima) taxa exist. Patterns of genetic diversity were investigated among British Restharrows using 10 microsatellite loci with 21 Restharrow populations analysed (411 individual plants) from Central and Eastern Britain. PCO analysis revealed clear genetic differentiation of the sampled plants into two groups, corresponding to O. spinosa/O. intermedia (diploid) and O. repens/O. maritima (tetraploid) plants. Evidence of genetic differentiation by distance was also revealed for O. repens/O. maritima, but not for O. spinosa/O. intermedia. The data suggest the presence of strong reproductive barriers between diploid and tetraploid Restharrows in Britain, but not within ploidy levels. This genetic isolation between ploidy levels is confirmed by a detailed analysis of a sympatric site (Harton Down Hill). These results demonstrate that PCA analysis is a suitable general tool for comparing related species of different ploidy levels.