POPULATION-GENETIC STRUCTURE AND OPTIMAL SAMPLING OF LAND RACES OF BARLEY FROM IRAN

Allozyme variation at 25 genetic loci was assayed in 12 indigenous cultivars (land races) of barley from Iran. For these loci the average probability that 2 gametes drawn at random from 1 population would differ at a locus was 0.082. In the collection as a whole, this measure of genetic diversity was 0.161, so that about half the total diversity was distributed between populations. Of the total of 31 allozyme variants, about half were common (frequency > 0.10) in only 1 or 2 regions. The results were compared with diversity estimates based on spike morphological polymorphisms; with allozyme polymorphism in 2 composite crosses (CC21 and CC34) of cultivated barley; and with previous results for allozyme polymorphism in Israel populations of wild barley. In the land race samples, the diversity of spike types was a poor index of allozyme diversity. The total allozyme diversity in this collection of land races was intermediate between the moderate levels in composite crosses and the high levels in Hordeum spontaneum from Israel. These results emphasize the role of land races as valuable genetic resources for plant breeding. They support sampling strategies which, by taking samples of moderate size from many sites, emphasize the collection of locally common alleles, as against strategies framed to capture the rare, conspicuous morphological variant by intensive or biased sampling, or strategies based on collecting the extreme ecotypes from a cline. The chromosomal location of 20 of the allozyme loci as deduced from wheat-barley addition lines, is listed in the appendix.