The procedure currently recommended by the University of Arizona for predicting the nitrogen (N) requirements of durum wheat has proven to be quite accurate at sites where grain yields exceeded 5,400 lbs/acre. However, the method slightly overestimated N needs when the yield possibility was below that level. Additional information on the relationships between N rates, stem NO₃⁻N levels and grain yields are needed for the wide range of agronomic conditions found in Arizona. Three N fertility trials were conducted at the Maricopa Agricultural Center to: 1) document the accuracy of the currently recommended soil + stem NO₃⁻N testing procedure in predicting the N needs of durum wheat on soils of varying residual N content and grain yield potentials; and 2) to evaluate the use of the current stem testing procedure on two durum varieties 'Aldura' and 'Westbred -881. The University of Arizona procedure was found to accurately predict the minimum amount of N required for optimum production of durum wheat on two sites where yield potentials were 5,400 and 4300 lbs. grain /A, but it slightly overpredicted N rates on two sites with maximum yield levels of 5,400 lbs /a. 'Aldura' consistently out yielded 'Westbred -881' by about 12 % but 'Aldura' also averaged 0.78 %lower in grain protein content. Little statistical or practical differences were observed in the quantities of NO₃⁻N contained in the stem tissue of these two varieties, which should simplify the interpretation of stem NO₃⁻N values for various wheat cultivars. The currently recommended procedure for predicting optimum N rates in durum wheat production has proven to be accurate when yield levels exceed 5,400 lbs. grain /A. A slight modification of the procedure may be needed to more closely predict N requirements on lower yielding sites.

The relative efficiencies of residual soil NO₃⁻N and fertilizer Nin basin - irrigated wheat production are not well defined. A two-year field study was conducted at the Safford Agricultural Center to investigate what these N efficiencies are under optimum yielding conditions. 'Aldura' durum wheat was grown on the same field site two years in succession. In 1987 a wide range of fertilizer N (0 to 419 lbs /A) applications resulted in residual NO₃⁻N accumulations of 36 to 140 lbs /A in the surface four feet of soil. Residual N plots were split in 1988 with one subplot receiving no additional N while the other was treated with 145 lbs of fertilizer N /A. Grain yield response curves for the –N and +N subplots were used to estimate the marginal rate of substitution (MRS) of soil NO₃⁻N for fertilizer N. The marginal efficiency of residual NO₃⁻N was a constant 16.7 lbs. grain produced /lb. of N across the range of profile N values in this study, while the marginal efficiency of fertilizer N varied from over 17 to below 6 lbs. grain /lb. N. When basin- irrigated wheat is supplied with adequate, but not excessive N, the MRS of soil vs. fertilizer N is about 1:1 although absolute N efficiencies under basin irrigation are considerably lower than those achieved in other grain production systems.

Collecting additional data to calibrate and refine current guidelines for interpreting soil and plant test values for P is an ongoing need in Arizona. An experiment was conducted at the Safford Agricultural Center during the 1987 -88 crop year to evaluate the response of 'Aldura' durum wheat to P fertilizer applications on a clay loam soil testing low in available P. Maximum grain yields of more than 5,500 lbs. /A were obtained by banding 40 lbs. P₂O₅/A as triple super phosphate with the seed at planting. A December 9 planting date subjected plants to cold soil temperatures early in the season (up to 80 days after planting), rendering soil P less available for plant utilization. Preplant soil P analyses predicted the yield increase observed with P fertilization (11.2 %);however, economic returns were not sufficient to offset the cost of P fertilizer. The recommended preplant soil test for P proved accurate in predicting P status and stem PO₄⁻P tissue analyses seemed reliable in monitoring P nutrition of durum wheat. A critical nutrient range of 1200 - 2000 ppm PO₄⁻P is proposed for basal stem tissue sampled prior to the joint growth stage, and 1500 - 1706 ppm PO₄⁻P is suggested for flag leaf tissue sampled at the boot stage.