Rating cultivars and trials in applied plant breeding

The number of means "m", from which a particular cultivar differs significantly, is used to arrange cultivars in order of descending superiority. For n cultivars under study, the maximum "m" value corresponding to the best cultivar is n-1, indicating that it exceeds significantly the n-1 ether cultivars, and the minimum zero. Because "m" represents an objective and reliable statistical measure of cultivar performance, it has been used to calculate two indexes: (1) performance index. P = 100 m (n-1), giving the percentage of means which a particular cultivar exceeds significantly, and (2) differentiation index. D = 200 Sigma m/n (n-1), giving the percentage of paired comparisons between cultivars that show significant differences. Rating cultivars according to P across years (Py), sites (Ps), and years plus sites (Pys), gives an overall picture of the relative superiority of the entries. Rating trials according to D allows comparisons between trials and offers the possibility when the same sets of entries are compared each year across sites, to identify sites that give maximum differentiation.