Information on P release potential in relation to labile P and P fractions in sandy soils is limited. In this study, P release potential was determined by leaching, and labile P, soil P fractionation, and P adsorption capacity were measured in the laboratory using 96 Florida sandy soil samples to evaluate the relationship between P release in water and soil P status. The sandy soils had a very low P adsorption capacity. The adsorption maximum, as calculated from the Langmuir equation, averaged 40.4 mg P kg(-1). More than 10% of the soil P was water soluble, indicating a high risk of P leaching from soil to water. Successive leaching using deionized water released, on average, 7.7% of total P (144.5 mg kg(-1)) in different soils, whereas labile P recovered by successive water extraction accounted for 39.2% of the total P. Variation in release potential among the different soils could be explained more by the difference in amounts of extractable P than the adsorption capacity. Total amounts of P released by successive leaching were significantly correlated with all labile P indices measured by different methods and all soil P fractions except for residual P. The correlation coefficients (r) were 0.97** for water-soluble P, 0.96** for 0.01 M CaCl2-P, 0.94** for Olsen P, 0.86** for Mehlich 1-P, 0.77*** for Mehlich 3-P, and 0.64*** for Bray 1-P. There were no obvious turning points in the relationships between Olsen-P, water-soluble P, or CaCl2-P and the amounts of P released from the sandy soils. The release of P from the sandy soils appeared to be controlled by a precipitation-dissolution reaction rather than a P sorption-desorption process. Furthermore, the sequential extraction of soils using deionized water indicated that P released was not limited to the labile P (H2O-P, NaHCO3-IP) and potentially labile P (NaOH-P) pools, but also from the HCl-P, indicating that all of P fractions except for residual P in the sandy soils can contribute to P release.