| Summary: | The present experiment was undertaken to study the variability, trait association and principle component analysis
among eighteen characters in 53 rice genotypes including landraces and cultivars. Higher estimates of GCV and
PCV coupled with high heritability and GAM were observed for a number of filled grains per panicle followed by
a number of spikelets per panicle and single plant yield indicated that these traits could be utilised for selection in
rice improvement. Study of association of yield attributing traits revealed that the single plant yield had a significant
positive association with the total number of spikelets per plant, number of filled grains per plant and number of
primary branches per plant and had a significant negative association with the length of the primary branches. Hence,
direct selection of positively associated characters can improve the grain yield and increased primary branch length
reduces the number of primary and secondary branches that might lead to a reduction in yield. Path coefficient analysis
showed that the number of spikelets per panicle, primary branches per plant and length of secondary branches had
a positive direct effect on yield. Therefore, giving weightage during selection to the above the mentioned traits could
improve the yield and yield attributes. The principal component analysis showed that out of 18 traits studied, only
eight principal components (PCs) had Eigen values greater than 1.00 and accounted for approximately 80.18% of the
total cumulative variability. Single plant yield, number of fertile spikelets and total number of spikelets per plant and
number of tillers showed maximum vector length confirming that the maximum contribution to the total diversity and
the genotypes with the above characters could be utilized as donors to improve the yield and its attributing traits in
future breeding programme.
|
|---|