Phosphorus fertilization has increased soil test P (STP) values in many fields in recent years, and has created a need for information concerning how these soils should be fertilized in the future. This study was conducted to evaluate changes in STP values, yields of corn (Zea mays L.) and soybean [Glycine max (L.) Merr.], and economic returns to P fertilization resulting f rom 14 yr of annual applications of P fertilizer at various rates to plots having various initial STP values. The experimental site was located in north central Iowa on Webster (fine-loamy, mixed, mesic, Typic Haplaquoll) and Canisteo (fine-loamy, mixed, calcareous, mesic, Typic Haplaquoll) soils that tested medium in STP (18 ppm, Bray-P1 method). Corn and soybean were grown in rotation each year from 1976 to 1989 by alternating crops between halves of the site. Initial P fertilization treatments of 0, 130, and 260 lb P/acre were applied to large plots in the spring of 1975. Four annual P treatments of 0, 10, 20, and 30 lb P/acre were superimposed on the initial P treatments beginning in the fall of 1975. Annual P additions required to maintain STP values increased with increases in the initial STP value and ranged from 15 to more than 30 lb P/acre. For both crops, economic returns to investments on annual P applications were positive when STP values were less than 16 to 20 ppm (medium range) and negative when STP values were above this range. The results suggest that crop producers with high-testing soils could increase profits by not applying P fertilizers until STP values decrease to levels that are profitable to maintain.