Field studies were conducted in 1986 to determine the effects of labeled rates and simulated drift rates on cotton of propanil and thiobencarb, two commonly used rice herbicides in Arkansas. The influence of cotton growth stage at time of application and the presence or absence of an in-furrow cotton insecticide were also evaluated. Herbicide treatments were propanil + thiobencarb (3.36 + 3.36, 0.34 + 0.34, and 0.03 + 0.03 kg/ha), propanil alone (4.48, 0.45, and 0.05 kg/ba), and an untreated check. In-furrow insecticide treatments consisted of aldricarb (0.5 kg/ha), disulfoton (0.8 kg/ha), and no insecticide. Growth stages were preemergence, cotyledon cotton, and 4-leaf cotton.

The experiment was arranged as a split-plot with a randomized complete block design with five replications. Main plots were the in-furrow insecticide treatments. Sub-plots were the herbicide treatments which represented 1, 0.1, and 0.01 times the normal rate. The same experimental design was used to evaluate each cotton growth stage as a separate experiment. All treatments were applied at a carrier rate of 93.5 1/ha. No preemergence or incorporated herbicides were used in the tests to insure injury was due only to treatment effects. Weed control was obtained by cultivation, hand-weeding, and layby treatment of linuron (1.1 kg/ha).

In the preemergence and cotyledon experiments, there was generally greater injury from simulated herbicide drift where in furrow insecticide was used. Aldicarb tended to result in more injury and yield reduction than disulfoton. Labeled rates of propanil alone (4.46 kg/ha) or propanil + thiobenearb (3.36 + 3.36 kg/ha) resulted in significant cotton injury or yield reduction regardless of the insecticide treatment in each of the growth stage experiments. Drift rates (0.1 and 0.01 times the normal rate) caused significant yield reductions in the cotyledon and 4-leaf tests but not in the preemergence experiment. The addition of thiabencarb to propanil caused more cotton injury and yield reductions than propanil alone in the preemergence experiment, but not in the cotyledon or 4-leaf test.