Improving wheat's tolerance to environmental stress is of utmost importance in the current era due to climatic changes. This study was carried out to determine the relative merits of pedigree selection for grain yield per plant on old and new land. To quantify the response of selection, two cycles of pedigree selection for grain yield per plant were applied to a segregating population of bread wheat crosses (Misr 3 Line #1) in F3 and F4 generations under new land stress conditions. The F5-selected families were evaluated in both old and new land habitats after the second cycle. Under both circumstances, the genotypic variance was much less than the phenotypic variance, and it generally decreased from the F3-generation to the F5-generation. Furthermore, compared to the old land environment, broad-sense heritability estimates for grain yield plant-1 were lower in the new land environment. After Cycle 1 and Cycle 2, the realized heritability in old land was 52.03 and 84.52 percent, respectively, compared to 34.08 and 62.80 percent in new land. In both cases, the instant reaction to selection that was found on ancient soil was examined, and the results showed a significant increase in grain production from both the bulk and the best parent of 5.40 and 5.02 percent and 17.28 and 7.03 percent, respectively. Selected families for grain yield under new land that were studied under both conditions revealed negligible increases of 12.29 and 2.80 percent from the bulk and considerable increases of 28.24 and 22.34 percent from the better parent, under new and old land, respectively. The results indicate these genotypes could be used as sources of tolerance or factors contributing to general adaptation. Furthermore, selection for grain yield/plant under new land stress was superior to selection under old land stress, regardless of whether selection entries were evaluated under stress or non-stress.