Investigating the relationship between grain yield and yield components in spring rapeseed cultivars using multivariate analysis

Rapeseed with the scientific name (Brassica napus L.) are one of the most important oilseed crops in the world. Considering the country's increasing need to import oil and the importance of rapeseed among oilseeds plants, increasing the yield and percentage of oil is very important. Therefore, identification the traits that increase seed yield, play an important role in the success of breeding programs. Yield is a complex trait that is affected by many factors. direct selection of a variety for yield is not often very effective, so investigating the relationships between grain yield and other traits for indirect selection may be effective. In such a situation, correlations may not clarify the relationships well, therefore, multivariate statistical methods and path analysis can help in understanding the nature of the relationship between traits for direct or indirect selection and improve the efficiency of selection in breeding programs plants become useful.

In this study, eight genotypes of spring rapeseed (SPN-202, SPN-204, SPN-206, SPN-207, SPN-217, SPN-225, SPN-227, SPN-182) along with 56 rapeseed hybrids (F1) were planted in a randomized complete block design with three replications in 2019-2020 crop years in the research farm of Gorgan Agricultural Research Station. During the experiment, phenological, morphological traits, yield and yield components were recorded. The normality of the data was evaluated based on the Kolmogorov-Smirnov method. In order to understand the relationships between traits and to identify the traits affecting on grain yield, at first, correlation coefficients were estimated, then stepwise regression were done. Then, path analysis based on correlation coefficients was used to determine the direct and indirect effects of traits. Finally, all traits were grouped based on principal component analysis.

The results of analysis of variance indicate a significant difference for all studied traits at the probability level of one percent. The coefficient of variation (C.V) for the traits varied from 1.37 (physiological maturity) to 10.31 (number of sub-branches), which indicates that there is sufficient accuracy in conducting the research. The results of correlation evaluation showed that number of pod per lateral branches (0.864), number pods per Plant (0.865), number of lateral branches (0.466), physiological maturing (0.329) and number of grain per pod (0.358) had a positive and significant correlation with grain yield (P<0.01). Stepwise regression analysis showed that the traits of number of pods per plant, number of grain per pod and 1000-grain weight has a decisive role on grain yield. Based on the results of path analysis, number of pods per plant had the most direct effect (0.881) and number of grain per pod had the most indirect effect (0.053) through the number of pods per plant on grain yield. In order the results of principal component analysis, four components were able to explain 75.91% of the changes in the measured data of 64 rapeseed varieties.

Results of the path analysis showed that number of pods per plant, number of grain per pod and 1000-grain weight had a direct positive effect on grain yield. Also, the stepwise regression analysis showed that the stated traits have the highest regression coefficients. Therefore, selection for these traits can be very effective in achieving high performance