Pedigree Analysis of North American Oat Cultivars Released from 1951 to 1985

Pedigree analysis of large collections of cultivars is useful for evaluating effects of crop improvement and provides a basis for selection of cultivars for use as parents in breeding programs. Coefficients‐of‐parentage among 205 North American oat (Avena sativa L.) cultivars were estimated and used to: (i) measure relative changes in genetic diversity through time based on diversity of cultivars released for the two time periods: 1951–1960 and 1976–1985, (ii) measure contributions to the germplasm pool by 89 landraces and introduced cultivars (ancestral parents) of A. sativa and A. byzantina C. Koch, and (iii) identify major groupings of related cultivars through cluster analysis. The average coefficients‐of‐parentage (rp) among all cultivars released with pedigree information from 1951 to 1960 was 0.09 and for cultivars released from 1976 to 1985 rp was 0.08. The 10 most important ancestral parents in each time period, based on cumulative rp values for all surveyed cultivars, declined in their relative contribution to the germplasm pool from 79% of the parentage of cultivars in 1951 to 1960 to 54.2% for cultivars released from 1976 to 1985. The most important ancestral parent was the South American cultivar, ‘Victoria’, followed by the Swedish landrace, ‘Milton’, and the A. byzantina landrace, ‘Red Rustproof’. Cluster analysis by the unweighted pair group mean method produced seven groups of cultivars. Six of the cultivar groups corresponded to either regional germplasm pools such as the southern USA cultivars or cultivars with high degrees of relationship to specific crop introductions such as Victoria. The largest cluster contained 116 of the 205 cultivars, and most of the cultivars from the midseason growing region as well as 85% of all North American cultivars released from 1976 to 1985. Although the average relationship among cultivars has remained nearly constant, current cultivars are more broadly based due to recombination of germplasms from regional clusters with a high intracluster relationship and a low intercluster relationship to form a single large germplasm pool with a low intracluster relationship and a higher than average intercluster relationship.