THE OBJECTIVE of this study was to assess correlation, path and regression analysis in 39 F6 bread wheat (Triticum aestivum L) genotypes during two seasons, i.e., 2015/2016 (F7) and 2016/2017 (F8) under irrigation and drought stress conditions at Fac. Agric. Edu. Farm, Minia University, Egypt. A positive correlation was found between grain yield and each of the number of spikes/plant and number of grains/spike under the two conditions. Path analysis revealed high positive direct effects on grain yield/plant via the number of grains/spike (0.929) under irrigation and number of spikes/plant (0.973) under drought. The direct effect of 100-grain weight on grain yield/plant was positive under irrigation (0.649) and drought (0.260). The indirect effects of these traits were negative under the two conditions. Stepwise regression analysis revealed that three models no. 8, 9 and 10 were fitted for each environment. The model no. 8 included one trait; the number of grains/spike under irrigation and number of spikes/plant under drought. Its relative contributions in grain yield were 0.180 and 0.693 under irrigation and drought; respectively. The model no. 9 in the two environments included two traits; the number of grains/spike and number of spikes/plant, its relative contributions in grain yield/plant were 0.544 and 0.836 under normal and drought conditions, respectively. The model no. 10 included three traits number of grains/spike, number of spikes/plant and 100-grain weight, its relative contribution in grain yield/plant were 0.0.931 and 0.978 under normal and drought conditions, respectively. This model no. 10 is fit and superior to use in selection for grain yield/plant.
