Estimation of genetic parameters in a recurrent selection program in Apple

Abstract

Combining ability was estimated for 8 tree and 11 agronomic traits in a multi-location apple genetics population to select individuals for the next cycle of selections. Families more than three standard errors away from the general mean were identified for each trait and extreme individuals (with regard to that trait) from those families were selected. On the whole, 148 individuals were chosen based upon individual traits and, in a few cases a combination of traits, in a way that maintains the original diversity within the population. Narrow-sense heritability estimated separately for each site for fruit ribbing, fruit russet, fruit overcolour amount and fruit weight ranged from 0.0–0.13, 0.05–0.58, 0.34–0.40 and 0.27–0.90 respectively. Phenotypic correlation and additive genetic correlation were small and from small to moderately high respectively, between pairs of traits. The implication of these genetic parameter estimates in apple breeding is discussed.