Experiments were conducted in 1995, 1996, and 1999 to evaluate clomazone preemergence in flue-cured tobacco seedbeds. The seedbeds were fumigated with methyl bromide to kill all the weeds and other pests, and then the subplots designated as grassy were sown with a mixture of equal mass of crowfootgrass, goosegrass, and rhodesgrass. Clomazone rates were 0, 0.14, 0.28, 0.42, and 0.56 kg ai/ha for 1995 and 0, 0.42, 0.47, 0.52, and 0.94 kg ai/ha for 1996 and 1999. The rates of clomazone in 1996 and 1999 were based on the 1995 results where 0.42 kg/ha decreased the dry mass of grasses by 98% compared with the nontreated control, while clomazone at 0.28 kg/ha decreased the dry mass of grasses by 75%, and this was not considered adequate. In 1995, 0.56 kg clomazone/ha decreased the tobacco seedling number but had no effect on the growth of tobacco plants. In 1996 and 1999 clomazone had no effect on the number or dry mass of tobacco seedlings despite causing slight discoloration 4 wk after sowing at all rates in 1996. This discoloration had disappeared by transplanting time. The effect of grass competition on the emergence and growth of tobacco was severe in the nontreated subplots. More grasses emerged in 1999 than in 1996, possibly due to higher temperatures; however, few grasses survived until transplanting time in clomazone-treated areas. All the clomazone rates tested in 1996 and 1999 were satisfactory as there was no crop injury or detrimental effect of the grasses on tobacco in the clomazone-treated grassy subplots. Therefore, the 0.42-kg clomazone/ha rate is considered optimal.