Accurate fertilizer N recommendations for irrigated continuous corn (Zea mays L.) production in the Great Plains are important for maximizing productivity and profit, while minimizing environmental impact of fertilizer use. The objectives of this study were to determine the effects of long-term N and P fertilization on grain yield and fertilizer recovery by irrigated continuous corn and to compare predicted N recommendations with actual N requirements. Six rates of N (0, 40, 80, 120, 160, and 200 lb/acre) and two rates of P (0 and 18 lb P/acre) in a factorial amangement were applied annually from 1961 to 1991 on a Ulysses silt loam (fine silty, mixed, mesic Aridic Haplustoll) near Tribune, KS. Grain yields were greater with a combination of N and P than with either applied alone. The optimal N rate was about 160 lb N/acre and was relatively constant over time. Averaged over the last 10 yr, application of 160 lb N/acre increased grain yields 46 % without P and 103 % with P. The yield benefit from applied P has increased over time. With adequate N, P increased grain yields by 29 % averaged over 31 yr and 48 % over the last 10 yr. Phosphorus fertilization reduced grain moisture from 28 to 23 %. Apparent fertilizer N recovery in the grain at 160 lb N/acre was twice as high with P as without P. A N recommendation model was used to compare estimated N requirements with actual N rates required to meet yield goals. Yields obtained with 160 lb N/acre were estimated to require 220 to 260 lb N/acre, or an overestimation of 30 to 60 %. These data suggest that the crop factor used in the model may need to be reduced to improve accuracy of N recommendations