Increasing emphasis on phosphorus (P)-based nutrient management underscores the need to understand P behavior in soils amended with biosolids and manures. Laboratory and greenhouse column studies characterized P forms and leachability of eight biosolids products, chicken manure (CM), and commercial fertilizer (triple superphosphate, TSP). Bahiagrass (Paspalum notatum Flugge) was grown for 4 mo on two acid, P-deficient Florida sands, representing both moderate (Candler series: hyperthermic, uncoated Typic Quartzipsamments) and very low (Immokalee series: sandy, siliceous, hyperthermic Arenic Alaquods) P-sorbing capacities. Amendments were applied at 56 and 224 kg P(T) ha(-1), simulating P-based and N-based nutrient loadings, respectively. Column leachate P was dominantly inorganic and lower for biosolids P sources than TSP. For Candler soil, only TSP at the high P rate exhibited P leaching statistically greater (alpha = 0.05) than control (soil-only) columns. For the high P rate and low P-sorbing Immokalee soil, TSP and CM leached 21 and 3.0% of applied P, respectively. Leachate P for six biosolids was <1.0% of applied P and not statistically different from controls. Largo biosolids, generated from a biological P removal process, exhibited significantly greater leachate P in both cake and pelletized forms (11 and 2.5% of applied P, respectively) than other biosolids. Biosolids P leaching was correlated to the phosphorus saturation index (PSI = [Pox]/[Al(ox) + Fe(ox)]) based on oxalate extraction of the pre-applied biosolids. For hiosolids with PSI < or = approximately 1.1, no appreciable leaching occurred. Only Largo cake (PSI = 1.4) and pellets (PSI = 1.3) exhibited P leaching losses statistically greater than controls. The biosolids PSI appears useful for identifying biosolids with potential to enrich drainage P when applied to low P-sorbing soils.