عنوان مقاله [English]
In order to study of combining ability and heterosis in cotton (Gossypium hirsutum L.), six hopeful genotypes were crossed in a half-diallel scheme at Gonbad Research Station. Fifteen F1 hybrids along with six parents were grown in randomized complete block design with 3 replications, in 2004. Combining ability estimates were calculated according to method 2, model 1 of Griffing (1956a,b). Relative importance of general combining ability (gca) and specific combining ability (sca) was estimated by the Baker (1978) ratio. Analysis of variance showed that differences due to genotypes were significant for yield, boll weight, monopodial branches number, middle sympodial branches length and node distance from main stem (P≤0.01). The mean of squares due to gca effect were significant (P≤0.01) for all studied traits, and variances due to sca effect were significant for yield, boll weight and monopodial branches number. The results suggested that variation for number of monopodial branches, sympodial branches length and the distance between the first node on sympodial from the main stem, could be attributed to additive gene effects, whereas non additive gene effect accounts for seed cotton yield and boll weight is controlled by both additive and non additive gene effects. Among parents, tabladila and shirpan-603 had the highest GCA for seed cotton yield (P≤0.05). The highest heterosis for yield and boll weight was 23.6 and 15.8 percent, respectively. The results indicated the importance of heterosis breeding for effective utilization of non-additive genetic variance. The combinations with higher heterosis are sufficient for hybrid production in cotton breeding program.