Salinity stress poses a significant challenge to rice production, particularly in regions where rice is a staple food. Breeding programs aimed at developing salt-tolerant rice varieties are crucial for enhancing food security and agricultural sustainability. In this experiment, 11 rice genotypes were evaluated as split-plot based on a randomized complete block design with three replications in summer 2012, at the Yasouj University. The main plots included 4 levels (0, 44, 88, and 132 mM) of salinity, and 11 rice genotypes considered as the subplots. According to the analysis of variance, sufficient genetic diversity was observed among genotypes for all studied traits. Breeding programs may therefore take into account the association of various component traits with yield and among themselves. According analysis of correlation coefficients under salt stress conditions revealed that traits such as flag leaf length (0.53), flag leaf area (0.61), hundred-grain weight (0.64), sterility percentage (0.67), and number of grains per panicle (0.80) had a moderate to high impact on the number of filled grains per panicle. The first two principal components (PC1, PC2) explained 64% of the data variation under non-stress conditions (PC1 = 33.50%, PC2 = 30.50%) and 64.6% under salt stress (PC1 = 37.20%, PC2 = 27.40%). This revealed that genotypes 2 and 6 were located near vectors associated with traits such as the number of grains per panicle, number of filled grains per panicle, and hundred-grain weight, indicating their superior yield under salt stress conditions. Using Ward’s method, genotypes were grouped into three clusters under both non-stress and salinity stress conditions. Under non-stress conditions, Group 2 (genotypes 2, 6, 7, 8) showed the highest filled grains and grain weight, while Group 1 (9, 10, 11) had the lowest. Under salinity stress, Group 1 (2, 4, 6) exhibited superior performance, suggesting genetic potential for salt tolerance, making them valuable for breeding programs aimed at improving yield under stress. Genotypes with high variability and desirable traits can be used in further breeding programs.
