Efficiency of three selection methods in two bread wheat populations

A three-seasons study (2021/22–2023/24) were carried out at the Experimental Farm, Faculty of Agriculture, Al-Azhar University, Assiut, Egypt to compare three selection methods i.e., pedigree (PSM), single-seed descent (SSD) and bulk (BM) from F2 to F4 generations in two bread wheat populations i.e., Sids14 x Giza 171 (population I) and Sids 14 × Gommiza 11 (population II). The studied characters were grain yield/plant, number of spikes/plant and number of grains/spike. The analysis of variance indicated highly significant differences among the F3 and F4 selected families for all the studied characters in both populations, confirming sufficient genetic variation for selection. Across all selection methods and both populations, the mean performance of F4 selected families consistently surpassed that of F3 selected families for all the studied characters. The results also revealed minimal differences between the phenotypic and genotypic coefficients of variability for the F3 and F4 selected families. This suggests that the environmental effects on these traits were relatively small. High values of broad-sense heritability were observed for all the studied traits in both populations and across all selection methods, indicating that the traits are highly heritable and selection would be effective. Furthermore, a significant and positive phenotypic correlation was found between grain yield/plant and both number of grains per spike and 100-grain weight. This suggests that selecting these component traits could indirectly improve grain yield. The substantial differences observed among families in both populations confirm that selection for grain yield/plant is a feasible and effective strategy for genetic improvement.