عنوان مقاله [English]
Background and objective: According to importance of food rice, and also increase of population, the development of new rice lines, and following that selection for improvement of grain yield is essential. Identification of direct and indirect effects of traits affecting grain yield facilitate successful selection. This research was done according to the importance of hidden factors in the formation of traits and their effects on grain yield, the need to determine the paths of the interaction between yield components on grain yield to improve in breeding programs, understanding the inter-relationships between traits and determining the most important traits related to grain yield for selecting the advanced mutant lines of Tarom Mahalli.
Material and methods: Twelve advanced mutant lines derived from Tarom mahalli together with Tarom Hashemi, Neda and Tarom mahalli were evaluated in a randomized complete block design with three replications at two locations of Sari and Tonekabon in 2016. The measured traits include: days to 50% flowering (DF), days to full maturity (DM), plant height (PH), no. of fertile tiller (FT), panicle length (PL), no. of filled and unfilled seeds per panicle (FUS), 1000 grain weight (1000 GW), grain length and width (GL&W), grain length / width ratio (GL/W), flag leaf length and weight (FLW), flag leaf length / width ratio (FL/W) and grain yield. Through stepwise regression, independent variables that had little effect on the function variable were eliminated, and fit the best model. Path analysis was calculated to determine the direct and indirect effects of traits on grain yield and also factor analysis was used in order to better justify and interpret the inter-relationships between traits and better understanding of hidden factors.
Results: Results showed that grain yield had positive significant correlation with 1000 grain weight (0.354* and 0.304* at Sari and Tonekabon location respectively) and fertile tiller (0.627** and 0.442**), and negative significant correlation with plant height (-0.300* and -0.501**). It expressed that the shorter cultivars having more fertile tillers and heavier 1000 grain weight illustrating more performance. Based on results of path analysis revealed that the most direct effect related to no. of fertile tiller (0.613) and days to maturity (0.242) respectively, so later maturity genotypes with more no. of fertile tiller produced more grain yield. At Tonekabon location, plant height (-0.452) in the opposite direction and days to full maturity (-0.431) in the positive direction had the greatest impact on yield and shorter and late maturity genotypes exhibited higher performance. According to factor analysis at Sari location, five factors were selected so that totally more than 77% of yield variance was identified by the first factor was called as morpho-phenology. The second, third, fourth and fifth factors were called as panicle and its components, grain production, seed size and flag leaf size, respectively. At Tonekabon location four factors were known which are able to identify more than 70% of yield variance. The first factor was called as grain characteristic and phenology, the second together with the fourth factor, were defined as morphology and grain production component and the third factor was defined as grain production and late maturity.
Conclusion: Path analysis showed that days to full maturity, increase of number of fertile tiller and decrease of plant height have greater efficiency and can use as a selection index in breeding programs. Based on results obtained of the factor analysis found that selection for increased of grain yield, increased of the period of vegetative growth and increased of number of fertile tiller in investigated mutant lines is possible simultaneously.