Yield and Trait Associations with Long-Duration Boro Rice Varieties in Northern Bangladesh using Correlation and PCA Approach

Abstract

For modern breeding program, the role of genetic variation is most important in the desirable genotype selection. This study was conducted to assess the adaptability and selection potential of newly released rice varieties in terms of their yield and analyze the relationships among yield and yield component traits by partitioning correlation coefficients and principal component analysis (PCA). Four long-duration boro rice varieties were assessed in a Randomized Complete Block Design. Yield-boosting factors included panicle number (PN), panicle length (PL), filled grains (FG), and 1000-grain weight (TGW), while spikelet sterility (%) hindered yield. Positive correlations with grain yield were observed for biological yield (BY), harvest index (HI), straw yield (SY), PN, plant height (PH), and PL. Traits like PN, PL, FG, TGW, HI, total tiller numbers (TN), BY, SY, PH, days to 50% maturity (DF), and days to 80% maturity (DM) positively impacted yield. Conversely, spikelet sterility and unfilled grains had negative correlations. PCA demonstrated that the initial three PCs explained 100% of the variance. PC1 highlighted the significance of DF, DM, PH, TN, PN, FG, TGW, SY, BY, and HI. PC2 emphasized the importance of unfilled grains, while PC3 underscored spikelet sterility. These findings imply that grain yield is predominantly influenced by FG, PH, DM, DF, PN, TN, PL, SY, BY, and HI. These insights can guide future breeding for enhanced yield potential as well. BRRI dhan89 and BRRI dhan92 might be recommended as superior yield producing rice varieties for northern Bangladesh considering productivity.

SAARC J. Agric., 23(2): 43-61 (2025)