Combining ability among single grosses and predicting double cross performances in poultry12

Abstract

Thirteen traits were examined on progeny of 40 egg‐type double crosses sampled over 2 years in order to evaluate the importance of general and specific combining ability variances among single cross parents in double cross combination and the feasibility of predicting double cross performance from parental and non‐parental single crosses. General combining ability variance was significant for 8‐and 55‐week body weight, egg shape index, egg weight, percentage egg production, meat spots and adult mortality, while specific combining ability variance was not significantly different from zero for any traits. General combining ability variance was not significant for age at first egg, interior egg quality, specific gravity and blood spots. Multiple regression procedures using as independent variables ordered means of parental and non‐parental single crosses and inbreds yielded higher predictability of double cross performance than using the respective means of each set. Deletion of parental single crosses from the complete model did not result in a statistically significant reduction in the sum of squares due to regression for any trait, but deletion of non‐parental single crosses yielded a significant reduction in regression sum of squares for 11 traits. However, part of this result may be explained on purely statistical grounds. Predictability of double cross performance was significant for age at first egg, 8‐ and 55‐week body weight, egg weight, specific gravity, egg production, meat spots and adult mortality.