To evaluate the general combining ability of male—sterile lines of sorghum for grain yield and some related traits in a genetically broad base population of sorghum [Sorghum bicolor (L) Moench], in 1990, nineteen randomly selected male-sterile lines (A-line) were top crossed to a broad genetic base tester (a mixture of equal amount of seed from 120 fertility restorer(R—lines) and 19 top crosses were produced. Genotypes, consisting of 19 top crosses, 19 male—fertile lines (B—lines) and tester were grown in 1991 at the Isfahan University of Technology Research Farm located in Najaf-Abad. Genotypes were evaluated in two favorable and unfavorable environments in a split—split plot design with complete randomized block arrangment in 3 replications. The main plots were two planting dates, 1st and 16 May. Two within rows plant spacings (10 and 20 cm) considered as the sub plots. Differences between means of top crosses and their respective B—lines, that indicates heterosis and superiority relative to the female parent, were statistically different from zero, for most of the traits studied. Heterosis led to an increase in grain yield, test weight, panicle lenght, plant height, panicle weight and earliness.
Grain yield and test weight had the smallest heritability estimates, based on variance components and regression of top crosses on female parents. The highest estimate of heritability based on variance component and regression belonged to plant height and panicle lenght. Due to high genetic variability and unadequate number of environments studied, the heritability estimates were high for most of the traits. The stimates of heritability for most of the traits in unfavorable environmental conditions were lower than those in favorable conditions. Howevere, the high estimates of heritabilites indicated the efficiency of selection for superior genotypes based on these traits. Therefore, it is possible to increase the mean of these traits by selection.
In favorable environment A22 and A23 had Significant general combining ability (GCA) effects for all of the traits,e.g. grain yield, and can be used as better parent in hybrid production. A12 had only a better performances for most of the traits in favorable environmental conditions. A22 showed a good performance in both environments, and its GCA effects for all of the traits, except plant height and test weight were positive and significant. Therefore, it can be used as a potential parent in hybrid seed production for a wide range of environments.
)