The loss of N and P by leaching is an important issue, especially on agricultural fields with subsurface tile drainage. The objective of this study was to evaluate how gypsum amendment and soil exchangeable Ca2+ and Mg2+ could affect the movement of P, NH4-N, and NO3-N in infiltrated water and soil. A column experiment was performed using a Miami silt loam soil, and the treatments were (i) control, (ii) gypsum applied at the surface, (iii) gypsum mixed into the 2.5-cm depth, and (iv) alteration of five different target exchangeable Ca/Mg ratios. A clear Plexiglas cylinder was filled with a 15-cm layer of soil; N, P, and K were applied in solution at the surface after the soil had been wetted and drained. Deionized water at a flow of 0.5 mL min(-1) was applied and eight leachate fractions, totaling about five pore volumes, were collected. Gypsum (5000 kg ha(-1)) applied at the surface and mixed into the 2.5-cm depth significantly decreased P and increased NH4-N concentration but had no significant effect on NO3-N leaching. Exchangeable Ca/Mg ratio treatments did not affect soluble nutrient losses; however, leaching of particulate P was significantly less in the Ca-treated soil than the Mg-treated soil. The overall practical conclusion of this study is that to control P transport, it is necessary to add gypsum even with a high soil exchangeable Ca/Mg ratio. The application of gypsum to the soil could be recommended as a best management practice to avoid water pollution by P; on the other hand, this could cause environmental problems by increasing the NH4 soil mobility.