Yield-independent variation in grain nitrogen and phosphorus concentration among Ethiopian wheats

New semidwarf wheat (Triticum aestivum L.) cultivars and new land management practices for Vertisols are being introduced in Ethiopia, Our objectives were to (i) determine the variation of N and P content and concentration in the grain and whether these are related to gain yield, (ii) test cultivar response to different fertility levels, and (iii) assess component traits of N and P yield, Five bread wheat cultivars and three durum wheat (Triticum durum Desf.) cultivars were sown in Exp. 1 at three locations in Ethiopia on two dates. In Exp. 2, seven of these cultivars were grown on a P-deficient soil at four N levels (0, 20.5, 41, 61.5 kg N ha(-1)) and four P levels (0, 10, 20, 30 kg P ha(-1)); in Exp. 3, two cultivars were grown all possible combinations of the same four N and P levels. Grain yields did not differ among cultivars, but significant variations were found for total shoot N and P, grain N and P yield, and grain N and P concentration. Cultivar differences in these traits were fairly consistent across the treatments and were corroborated by Exp. 3. The N and P concentrations in the grain were not related to grain yield (r = 0.36 NS for N; r = 0.28 NS for P). There was a positive association between grain N and P concentrations in Exp. 1 (r = 0.66; P = 0.07) and in Exp. 2 (r = 0.92; P < 0.01), The N and P grain yields were related to total shoot N and P at maturity (r = 0.99; P < 0.001), However, postanthesis accumulation of N was more closely related to postanthesis dry matter accumulation (r = 0.84; P < 0.05) than to the postanthesis accumulation of P (r = 0.56 NS). Total shoot P varied by as much as 50%. Thus, cultivar choice is an important factor determining removal of P from the soil.