Large spatial variability in soil properties is observed at the field scale in eroded landscapes typical of the northern Great Plains. We conducted herbicide dissipation studies to evaluate the effects of landscape position and soil properties on the rate of metolachlor dissipation in the northern Corn Belt. In this eroded landscape, the surface soil in upper slope positions is comprised of exposed subsoil low in organic carbon. Lower slope positions are characterized by deep accumulation (>40 cm) of high-organic-matter soil. Replicate plots were established that extended from the summit to the toeslope of the test area. Separate plots were treated with s-metolachlor according to label instructions in the fall and in the spring (preemergence) at a rate of 1.95 L per ha. Potassium bromide, a non-sorbed, non-degraded tracer, was applied to each plot at 60 kg per ha. Soil cores were collected after application and throughout the growing season to a depth of 1 m and sectioned into 0-10, 10-20, 20-40, 40-60, and 60-100 cm increments. Bromide and herbicide concentrations were measured in replicate samples at each depth at each sampling time. Metolachlor concentrations in the top 10 cm of soil in all landscape positions showed significant decreases in the first 14 days following fall application. Residual metolachlor concentrations will be determined in the spring. These results will be compared with the rate of metolachlor dissipation following spring application. Weed control efficacy will be determined for both the fall and spring application as a function of landscape position. These results will provide information for the development of best management practices for this herbicide.