genetic analysis of some physiological, phenological and morphological traits in rapeseed (brassica napus l.) genotypes using diallel method

production of high rapeseed yield is important in rapeseed breeding programs. with regards to the importance of estimation of general combining ability (gca), specific combining ability (sca), additive and dominance genetic variances, gene action, heterosis and heritability, an experiment was carried out in a 9×9 diallel design. nine parents with 36 crosses were studied in a randomized complete block design (rcbd) with three replications at the dryland agricultural research institute, sararood station during 2006-2007. for the analysis of data griffing method 2 and hayman’s graphical analysis method were used. diallel analysis showed both additive and dominant type of gene action in the inheritance of the studied traits. the licord line and kiristina × pastill hybrid revealed maximum rate of general combining ability and specific combining ability for grain yield, respectively. the involvement of additive gene action in the inheritance of cellular membrane stability exhibited the effectiveness of selection for the improvement of this trait. since chlorophyll content is controlled by dominant gene action, methods based on hybridization are effective for the improvement of this trait. finally, additive and dominant effects were involved in the genetics of seed yield, thousand kernel weight, seed per pod, pod per plant, pod length, days to maturity, days to flowering, plant height, sub-branches and chlorophyll content, therefore, methods based on hybridization and selection in segregating population are suggested for their assessment. graphical analysis of the characters, displayed over dominance for seed yield, thousand kernel weight, plant height, sub-branches and chlorophyll content, while partial dominance was observed for other traits. calculation of the average degree of dominance for all traits except days to maturity, days to flowering and pod length also indicated dominant effect for control of these traits, thereby to increase and improve these traits by the phenomenon of heterosis. distribution of parents around the regression line indicated that the parents such as parade, pastill, kvintell and modena for most traits had the dominant genes more than other varieties, while option-500 had the recessive genes for most of these traits. parents with recessive alleles should not cross with individuals with much lower traits.