Diversity and Association Analysis of Physiological and Yield Indices in Rice Germplasm

Abstract

Rice is an important staple food crop, but in many countries, average rice yields are much lower than their yield potential. The objective of the present study was to evaluate the phenotypic performance of diverse rice genotypes (310) for yield traits, identify high-yielding early-duration genotypes with greater partitioning efficiency, and classify the best and worst genotypes based on their performance in the 2019–20 growing season under randomized complete block design (RCBD) with three replications. The analysis of variance showed significant differences for all the traits between genotypes (p ≤ 0.001). Correlation analysis revealed a significant correlation between grain yield plant^–1 and flag leaf area, panicle grain weight, panicle length, number of spikelets panicle^–1, spikelet fertility, number of grains panicle^–1, 1000 grain weight, grain length, net photosynthesis, and water use efficiency. Principal component analysis indicated genetic variation between all genotypes. The cumulative genetic variation in the first two principal components (PCs) was 69.18% (PC₁: 57.74% and PC₂: 11.44%). PC₁ added more toward yield and related traits to the separation of rice genotypes and contributed to the variability for 1000 grain weight (7.74%), spikelet fertility (7.56%), number of spikelets panicle^–1 (7.54%), flag leaf area (7.41%), and shoot dry weight (7.13%). Projection in biplot analysis confirmed that all the best genotypes were opposite to only the worst genotype G-19 and all others were positively associated with each other. Thus, the selection of these traits which are positively related to grain yield and the selection of best genotypes in rice would be useful for improving yield. Diversity and association of physiological and yield-related traits could be useful to improve the crop through the selection of useful traits to increase productivity and meet the demand of the growing population.