Tests for adaptive RAPD variation in population genetic structure of wild barley, Hordeum spontaneum Koch

We examined the adaptive importance of RAPD variation in the population genetic structure of wild barley, Hordeum spontaneum. The test involved (1) a nested sampling design with four population groups representing four distinct environments; and (2) a comparison of observed variation with that expected as a result of natural selection. Analyses of selection on fitness-related traits by reciprocal introductions served as guidelines for the expected pattern of RAPD variation. We found no concordance between the observed pattern of population genetic structure and that expected under the null hypothesis of environment-specific natural selection. There was no relationship between genetic distance and environmental similarity; none of 54 putative loci exhibited an allele distribution in accordance with that expected and no favoured epistatic allele combinations were detected across the four environments. The fact that environmentally induced adaptation, detected by fitness-related traits, was not reflected in inter-population RAPD structure (1) strongly enhances the neutralist viewpoint and (2) casts doubt on the notion that significant correlations between some environmental parameters and allele frequencies in one or more loci are evidence of selection on the latter.