Field experiments were conducted in 1996, 1999, and 2000 to evaluate weed control and snap bean response to postemergence applications of fomesafen at registered and reduced rates. S-Metolachlor was applied preemergence to all plots to suppress annual grasses. Snap bean injury generally increased as fomesafen rate increased, but at rates up to 0.28 kg ai/ha, injury by fomesafen was similar to or less than that from bentazon. Fomesafen at rates as low as 0.07 kg/ha provided near-complete control of common ragweed, and rates of 0.14 kg/ha or more of fomesafen controlled ivyleaf and pitted morningglories and 5-cm or smaller common lambsquarters as effectively as did bentazon. Control of all weed species from fomesafen alone at 0.21 kg/ha did not improve with the addition of bentazon at 0.28 kg/ha. Although snap bean injury from fomesafen was as high as 43% 1 wk after treatment, snap bean yield and net returns were similar to those from S-metolachlor alone. In a rate and application timing study, fomesafen at 0.14 kg/ha applied to three-trifoliolate snap bean was the least injurious to the crop, whereas applications at 0.28 kg/ha to one- or two-trifoliolate snap bean provided the best weed control.