Abstract
Background: Cadmium (Cd) can reduce soybean growth and productivity, but responses may differ among cultivars. This study evaluated cultivar specific agronomic responses to increasing Cd stress using univariate and multivariate approaches. Methods: A greenhouse pot experiment was conducted using a 3 × 4 factorial completely randomized design with three replications. Atlas 3616, Arisoy and Superb were exposed to 0, 2.5, 5.0 and 7.5 mg Cd kg-1. Seven growth and yield traits were evaluated together with a composite tolerance index (CTI), principal component analysis (PCA) and Pearson correlations. Result: Significant cultivar × Cd addition interactions occurred for all seven traits. At 7.5 mg kg-1, seed yield declined by 24.32% in Atlas 3616, 36.65% in Arisoy and 43.79% in Superb. CTI ranked Atlas 3616 highest (83.67), followed by Arisoy (82.27) and Superb (80.40). The first three principal components explained 81.77% of total variation. Seed yield was positively associated with pod number (r= 0.659) and seed number (r = 0.613). Atlas 3616 showed the greatest overall agronomic trait retention under the highest Cd addition. However, because soil available and plant tissue Cd were not measured, these results cannot indicate Cd exclusion or food safety suitability.