American Research Journal of Agriculture

Grain yield and grain protein content (GPC) of bread wheat are crucial determinants of its profitability and product quality. In general, low levels of nitrogen fertilizer result in a lower protein content in wheat grains and lower grain yield. The main objective of this investigation was to get information on gene action, heritability and expected genetic advance from selection (GA) for GPC and grain yield/plant (GYPP) under contrasting soil- nitrogen conditions. Diallel crosses in F1 and F2 generations among six contrasting wheat parents in GPC and GYPP were evaluated in field experiments across two seasons in a split plot design under two N levels, i.e. low-N (0kg N/fed) and high-N (75kg N/fed). In both F1's and F2's, the magnitude of dominance (H1) was much greater than that of additive (D) component for GPC under both high N and low N and GYPP under high-N, while the opposite was true for GYPP in F1's under low-N, where the additive was greater than dominance variance and GYPP in F2's under low-N, where the additive and dominance components were approximately equal. The average degree of dominance (H1/D)1/2 in F1's was in the range of over-dominance, except for GYPP under low N, which showed partial dominance, but in F2's was in the range of partial dominance for GYPP under both high and low N and GPC under high N, while GPC under low N showed no dominance. Narrow-sense heritability (h2 n) in F1's and F2's was generally of low magnitude; it ranged in F1's from 6.18% (GPC under low-N) to 20.41% (GYPP under high-N); the exception was GYPP under low-N, where h2 n was of high magnitude (67.14%) and ranged in F2's from 10.86% for GPC under low N to 46.90% for GYPP under low-N. The h2 n was higher under low--N for GYPP and under high-N for GPC. In F1's, the values of GA were higher under low-N than under high N for GYPP, but were higher under high-N than low-N for GPC. In F2's, the values of GA for both traits were higher under low-N than under high-N, suggesting that it is better to practice selection in F2's for both traits under low-N conditions to obtain higher values of selection gain.