The effects of selected agricultural land-management practices on water quality were assessed in a comparative study of four small basins in the Piedmont province of North Carolina. Agricultural practices, such as tillage and applications of fertilizer and pesticides, are major sources of sediment, nutrients, and pesticides in surface water, and of nutrients and pesticides in ground water. The four study basins included two adjacent row-crop fields, a mixed land-use basin, and a forested basin. One of the row-crop fields (7.4 acres) was farmed by using conservation land-management (CLM) practices, which included strip cropping, contour plowing, field borders, and grassed waterways. The other row-crop field (4.8 acres) was farmed by using standard land-management (SLM) practices, which included continuous cropping, straight-row plowing without regard to land topography, and poorly maintained waterways. The mixed land-use basin (665 acres) was monitored to compare water quality in surface water as SLM practices were converted to CLM practices during the project. The forested basin (44 acres) provided background surface-water hydrologic and chemical-quality conditions. Surface-water flow was reduced by 18 percent by CLM practices compared to surface-water flow from the SLM practices basin. The thickness of the unsaturated zone in the row-crop basins ranged from a few feet to 25 feet. Areas with thick unsaturated zones have a greater capacity to intercept and store nutrients and pesticides than do areas with thinner zones. Sediment concentrations and yields for the SLM practices basin were considerably higher than those for the other basins. The median sediment concentration in surface water for the SLM basin was 3.4 times that of the CLM basin, 8.2 times that of the mixed land-use basin, and 38.4 times that of the forested basin. The total sediment yield for the SLM basin was 2.3 times that observed for the CLM basin, 14.1 times that observed for the mixed land-use basin, and 19.5 times the yield observed for the forested basin. Nutrient concentrations in surface water from the row-crop and mixed land-use basins were higher than those measured in the forested basin and in precipitation collected near the row-crop basins. The SLM basin generally had the highest concentrations of total nitrogen, nitrite plus nitrate, total phosphorus (equivalent to the mixed land-use basin), and potassium. Nutrient concentrations in soil water and ground water were less than concentrations in surface water for the row-crop basins. Nutrient concentrations generally were at least slightly below the root zone (3-foot depth) and in ground water. Differences in nutrient yields among basins had patterns similar to those observed for nutrient concentrations. The total nitrogen yield for the SLM basin was 1.2 times the yield for the CLM basin, 1.9 times the yield for the mixed land-use basin, and 4.2 times the yield for the forested basin. The total phosphorus yield for the SLM basin was 1.7 times the yield for the CLM basin, 3.3 times the yield for the mixed land-use basin, and 7.8 times the yield for the forested basin. No significant differences in pesticide concentrations in surface water were identified between those measured in the SLM basin and those measured in the CLM basin. Significantly higher pesticide concentrations were observed at the row-crop basins compared with those observed at the mixed land-use basin probably because sampling sites for the row-crop basins were closer to the pesticide sources. No pesticides were detected in the forested basin. Comparisons of pesticide concentrations in soil from the two row-crop basins indicated some differences. Concentrations of the soil pesticides isopropalin and flumetralin were higher in the SLM basin than in the CLM basin. The surface-water quality of the mixed land-use basin generally was less affected by agricultural nonpoint sources than that of the smaller row-crop b