Soybeans [Glycine max (L.) Merr.] are often the replacement crop when cotton (Gossypium hirsutum L.) is abandoned because of stand failure in the southern United States. Injury from cotton herbicides may be reduced if the soybean planting is delayed more than 4 weeks after cotton planting or if the original herbicide-treated area is fully tilled and a new seedbed formed. Planting delay intervals were compared with five cotton preemergence herbicides on Calloway silt loam at one location. Seedbed-preparation methods were included in a similar experiment at two locations on Sharkey silty clay. Herbicides compared were fluometuron [1,1-dimethyl-3-(α,α,α-trifluoro-m-tolyl)urea], norflurazon [4-chloro-5-(methylamino)-2-(α,α,α-trifluoro-m-tolyl)-3(2H)-pyridazinone], cyanazine {2-[[4-chloro-6-(ethylamino)-s-triazin-2-yl] amino]-2-methylpropionitrile}, perfluidone {1,1,1-trifluoro-N-[2-methyl-4-phenylsulfonyl)phenyl] methanesulfonamide}, fluridone {1-methyl-3-phenyl-5-[3-(trifluoromethyl)phenyl]-4(1H)-pyridinone}, and diuron [3-(3,4-dichlorophenyl)-1,1-dimethylurea]. Seedbeds compared were no-till and conventional (fully tilled). Soybean planting delays after cotton planting were 20, 29, and 56 days (first year), and 15 and 29 or 15 and 30 days (second year). We found in the 2-yr studies that fluridone severely damaged soybeans both years. Fluometuron and diuron also caused damage the second year when we experienced wet, cool conditions in the spring. Most injury occurred on the clay, with yield reductions occurring even after the 30-day delay. Norflurazon was most injurious at this location. The no-tillage planting method resulted in the least herbicide injury on the silt loam and the conventional tillage method was better on the clay. Detailed studies with fluometuron under incubation conditions showed that degradation was inhibited most by low temperatures and high soil moisture.