Tomato (Lycopersicon esculentum Mill.) was grown in southeastern Florida on sandy soils that tested very high in Mehlich-1 P to evaluate the yield response to P fertilization. One location was used in 1995–96, another in 1996–97. Prefertilization soil samples contained 290 (location 1) and 63 (location 2) mg·kg–1 Mehlich-1 P. Both soil test results were interpreted as very high in P, and P fertilizer was not recommended for the crop. Fertilizer treatments at both sites were 0, 25, 50, 100, 150, and 200 kg·ha–1 P. Neither total marketable yield nor yield in any fruit size category was affected by P fertilization in either season. Amounts of cull (undersized or misshapened) fruits increased quadratically with P fertilization in the second season. Whole-leaf P concentrations increased linearly or quadratically with P application, depending on sample periods, and were always above sufficiency values. Although many tomato growers apply P fertilizer irrespective of soil test recommendations, our results showed that added P was not needed on soils testing very high in P. Furthermore, withholding P applications to soils with high P concentrations will minimize potential P pollution of surface water and groundwater.