Soybean Genetics Newsletter


In this communication, the results obtained from the material given earlier (Gupta et al., 1981) on the basis of individual regression analysis by using Perkins and Jinks (1968) model, have been presented for different traits groupwise. In the individual regression analysis, genotypes having nonsignificant regression m.s. as well as remainder m.s. against error m.s. were described as exhibiting absence of genotype x environment (g x e) interaction, genotypes having regression m.s. significantly different from error m.s. were designated as showing predictable g x e (linear g x e) interaction, and genotypes having only significant remainder m.s. or significant regression m.s. not significantly different from significant remainder m.s. were categorized as genotypes showing unpredictable g x e (nonlinear g x e) interaction. On the basis of this all 40 genotypes were classified into three groups and their distribution with respect to different groups of characters studied is given in Table 1.