We applied gypsum (CaSO 4 ×2 H 2 O) amendments to 100 m 2 plots within two clay-textured fields, one under shallow cultivation to 10 cm depth and the other ploughed to 20 cm depth. Unamended plots and plots subjected to a CaCO 3 (finely ground limestone) application served as controls. Separate soil monoliths (30 cm in diameter, 40 cm in depth) were collected for laboratory rainfall simulations from all plots 7, 19 and 31 months after the initial application of the amendments. Water passed through the monoliths during these simulations was analysed for turbidity, dissolved and particulate phosphorus (DRP and PP), nitrogen species, dissolved organic carbon (DOC), as well as dissolved Ca 2+ , Mg
2+, K + and S, pH, and electrical conductivity (EC). Over the three-year monitoring period, gypsum amended soils exhibited substantial decreases in turbidity (45%), PP (70%), DRP (50%) and DOC (35%) relative to control samples. The effects gradually decreased with time, and after 31 months gypsum effects on P species were detectible, but no longer statistically significant. We consider gypsum amendments as a potential tool for slowing P loss from agricultural areas with high P loss potential.