Evaluation of Diversity and Relationship among Yield and Yield Components of Rapeseed Genotypes (Brassica napus L.)

In order to study of variation and correlation between yields and yield components in oilseed rape, an experiment was designed using line×tester method. Fifteen hybrid derived from crosses of three spring type and high yielding testers (SPN34, RGS003, SPN1) with five spring lines with moderate yield (SPN3, SPN9, SPN36, SPN30, DH4) were tested using a randomized complete blocks design with two replications in Seed and Plant Improvement Institute during 2014-15. Totally 21 traits were studied. Mean squares for all 21 traits except flowering time were significant at 5% to 1% levels. Grain yield positively correlated with, oil yield (0.999), biological yield (0.868), first pod height from ground level (0.627), the number of seeds per pod (0.546) and the pod length (0.523). According to the results of path analysis, oil yield (1.016) and oil percent (0.379) had the most direct and indirect effects on grain yield, respectively. Cluster analysis based on the square Euclidean distance and Ward's method, classified crosses in three clusters and the accuracy of the results of clustering was confirmed by discriminant analysis. Six independently principal components were identified in the analysis of principal components, and this component could explain 89.20% of the total variation. In general, the most important affecting traits was grain yield, which can be used to for direct selection of grain yield in segregating generations.