Genetic Architecture of Yield and Components of Yield in Wheat¹

A 6 X 6 diallel set of crosses in bread wheat (Triticum aestivum L.) was studied in the parental, F₁, F₂ and F₃ generations during 1965–66 and 1966–67 and subjected to biometrical analysis with a view to determining the genetic architecture of grain yield per plant and its principal components, viz., 1,000-grain weight (kernel weight), the number of grains per ear, weight of grains per ear and the number of ears per plant.

Heterotic expression was maximum for grain yield and the character showed high specific combining ability variances, high H₁ component, low heritability and preponderance of duplicate epistasis. Among component characters, 1,000-grain weight was found to be the most highly heritable and to have high general combining ability variances, a large magnitude of D component and preponderance of complementary epistasis.

General combining ability X year interaction was more than specific combining ability X year for all the characters. Sonora 64, was the most genetically divergent parent in the material studied. Sonora 64, N.P. 876 and H. 26 were found to be good general combiners for most of the characters studied. On the basis of combining ability estimates and on the nature of gene action involved in the expression of yield and its component characters, Sonora 64 X H. 26, Sonora 64 X N.P. 876 and N.P. 876 X H.S. 990–483 appeared to be most suitable for exploitation in practical plant breeding programmes. Use of different methods of analysis for a better understanding of the genetic architecture and the implications of such information in future wheat breeding programmes are discussed.