Identification of Factors Affecting Trait Stability via Factorial Regression and Environmental Variance

Abstract

The stability basis in bread wheat still remains to be understood. Our objective was to determine factors affecting the stability of end-use quality in wheat, using a method developed in a companion paper that combined environmental variance and factorial regression. Kernel protein percentage (% protein), strength (W), tenacity (P) and extensibility (L) of dough were studied. Twenty-five environmental Nathalie Robert is affiliated with Institut Polytechnique LaSalle Beauvais, 19 rue Pierre Waquet, BP 30313, F-60026 Beauvais Cedex, France. Pierre Bérard is affiliated with INRA, UMR Amélioration et santé des plantes, 234 avenue du Brézet, 63039 Clermont-Ferrand cedex, France. Maryse Brancourt-Hulmel is affiliated with UMR INRA/USTL, 1281 SADV (Stress abiotiques et différenciation des végétaux cultivés), Estrées-Mons BP 50136, ). The authors thank P. Brabant and N. Galic, J. Le Gouis and M. Niarquin, M. Trottet and J.-Y. Morlais for running trials at Le Moulon, Mons and Rennes in 1999. The authors also thank Simon Hawkins for editorial assistance.covariates characterized grain formation and nitrogen yields, dry matter and nitrogen quantities at anthesis, grain-filling, and post-anthesis temperatures. Earliness was a genotypic covariate. For each trait, the factorial regression, introducing the two environmental covariates that best discriminated for stability and best explained the interaction, was selected. The best factorial regressions explained 31.5%, 53.3%, 49%, and 46.3% of the interaction sum of squares for the % protein, W, P, and L, respectively. The selected covariates indicated that nitrogen availability at anthesis, nitrogen grain-filling rate, and duration were involved in the quality stability. The stable varieties appeared to use poorly environments favorable to nitrogen accumulation, and to behave well in unfavorable ones. Such information will be helpful in future breeding programs.