Relevance of Specific versus General Combining Ability in Winter Wheat

Knowledge of quantitative genetic parameters is crucial in allocating resources for different steps of multistage selection programs. The main objective of the present study was to investigate the magnitude of variance of general combining ability (GCA) and specific combining ability (SCA) effects and their interaction with environments in wheat (Triticum aestivum L.). We used four data sets (Exp. 1, 2, 3, and 4) generated in commercial hybrid winter wheat breeding programs with a total of 940 hybrids evaluated in multienvironment trials in 2010 and 2011 in France. In at least three out of four experiments, general combining ability variance for females (sigma(2)(GCA-F)), specific combining ability variance (sigma(2)(SCA)), general combining ability of females x environment interaction variance (sigma(2)(GCA-FxE), and general combining ability of males x times environment interaction variance (sigma(2)(GCA-MxE)) were significantly larger than zero but general combining ability variance for males (sigma(2)(GCA-M)) was significant only in Exp. 4. For Exp. 3, we found no significant difference between the best performing hybrid and the best performing commercial variety. In contrast, 1.8% of hybrids in Exp. 2 and more than 21% of the hybrids in Exp. 1 and 4 significantly (P < 0.05) outperformed the best commercial variety included in the respective experiment. This superiority was even apparent when projecting our findings on line varieties of the same cycle of selection. In Exp. 2, 3, and 4, general combining ability variance (sigma(2)(GCA)) was more pronounced compared to sigma(2)(SCA). Moreover, correlation between GCA predicted and observed hybrid performance was medium to high (r(Exp.1) = 0.50, P < 0.01; r(Exp.2) = 0.90, P < 0.01; r(Exp.3) = 0.59, P < 0.01; and r(Exp.4) = 0.92, P < 0.01). Consequently, selection based on GCA effects is promising in hybrid wheat breeding program.