Information on long-term fertilizer P recovery by grain crops from a single broadcast application in the northern Great Plains is limited. Uptake of P by grain and straw was measured in 10 or 11 crops (1967 or 1968 to 1983) grown on a Williams loam (fine-loamy, Typic Argiborolls) in a split-plot, randomized complete block experiment. Fertilizer N, as main plots, was applied each crop year, except two, at rates of 0, 45, and 90 kg N/ha. Concentrated superphosphate P fertilizer (subplots) was applied only once, at study initiation, at rates of 0, 22, 45, 90, and 180 kg P/ha. The first six wheat crops (Triticum aestivum) were grown in a crop-fallow system and the last four or five crops, including wheat, barley (Hordeum vulgare), and safflower (Carthamus tinctorius L.), in an annual cropping system. Average P uptake in the grain (21 crops) was 6.4, 7.2, 8.0, 9.1, and 10.0 kg P/ha per crop for the 0, 22, 45, 90, and 180 kg P/ha treatments, respectively, over all N levels. Average P uptake in straw (21 crops) was 0.6, 0.7, 0.9, 1.0, and 1.2 kg P/ha per crop for the same respective P rates. Nitrogen fertilization generally decreased P uptake at low P rates (0 and 22 kg P/ha) but increased P uptake at high P rates (90 and 180 kg P/ha). Average P uptake in grain (21 crops) was 7.6, 8.4, and 8.4 kg P/ha per crop for the 0, 45, and 90 kg N/ha treatments, respectively, over all P rates. Fertilizer P recovery in the grain for the 22, 45, 90, and 180 kg P/ha treatments averaged 32, 25, 23, and 13%, respectively, without N fertilization and 45, 38, 37, and 24% with 45 kg N/ha. The results of this study suggest that a one-time broadcast application of P fertilizer at rates as high as 90 kg P/ha is an efficient way to meet crop needs and utilize P fertilizer resources.