The threat of P transport from land applied manure has resulted in water quality concerns. Research was conducted to evaluate gypsum as a soil amendment applied to grass buffer strips for reducing soluble P in surface runoff. A simulated concentrated flow was created in an established tall fescue (Festuca arundinacea Schreb.) pasture. Poultry litter (PL) was applied at a rate of 250 kg N ha(-1) to the upper 3.05 m of each plot, while gypsum was applied at rates of 0, 1, 3.2, and 5.6 Mg ha-1to the lower 1.52 m of the plot functioning as a grass buffer strip. Two 30-min runoff events ( approximately 4 L min(-1)) were conducted, immediately after PL application and 4 wk later to determined soluble P concentration in the surface water samples. The greatest concentration of soluble P was in the runoff event occurring immediately after the PL application. Gypsum applied to grass buffer strips was effective in reducing soluble P concentrations (32-40%) in surface runoff, while the untreated buffer strip was somewhat effective in reducing soluble P (18%). No significant differences were observed between gypsum rates, suggesting that land managers would achieve the greatest benefit from the lowest application rate (1Mgha(-1)). In the second runoff event, although concentrations of soluble P in the surface water runoff were greatly reduced, the effect of gypsum had disappeared. Thus, these results show that gypsum is most effective in reducing the initial P losses from PL application when applied to grass buffer strips. The information obtained from this study may be useful in aiding land managers in developing management practices that reduce soluble P loss at the edge of a field.