Biosolids and composted municipal solid wastes were surface-applied (0 and 80 Mg ha(-1)) to a degraded soil in a semi-arid environment to determine their effects on the quantity and quality of run-off water. Three and 4 years after application, a simulated rainfall was performed (intensity=942.5 ml min(-1) and kinetic energy=3.92 J m(-2)) on 0.078 m(2) plots using a portable rainfall simulator. The run-off from the different treatment plots was collected and analysed. The type of treatment was highly related to infiltration, run-off and sediment production. The biosolid-treated plots showed the minimum value of total run-off, maximum time to the beginning of run-off and maximum run-off ratio (the relationship between total rainfall and run-off). The MSW-treated plots showed values intermediate between biosolid-treated plots and control plots. Soil losses were also closely related to treatment type. Control plots showed the maximum sediment yield, MSW-treated plots showed intermediate values, and biosolid plots the minimum values for washout. The concentrations of NH(4)-N and PO(4)-P in the run-off water were significantly higher in the treated plots than in control plots. The highest PO(4)-P value, 0.73 mg l(-1), was obtained in the soil treated with biosolids; NO(3)-N concentration also increased significantly with respect to the control and MSW treatments. NH(4)-N concentrations of 15.6 and 15.0 mg l(-1) were recorded in the soils treated with biosolids and MSW, respectively, values approximately five times higher than those obtained in run-off water from untreated soil. However, the concentrations of all these constituents were lower than threshold limits cited in water quality standards for agricultural use. With the exception of Cu, all trace metals analysed in the run-off water were below detection limits.