We studied the inactivation of herbicides applied to the soil by activated carbon in the greenhouse and field. Degree of inactivation depended upon herbicide type, dosage, plant species, and quantity of activated carbon applied. In greenhouse experiments with Toa silty clay, 2-chloro-4,6-bis(isopropylamino)-s-triazine (propazine) was inactivated at activated carbon: herbicide ratios of 66:1, whereas 4-amino-3,5,6-trichloropicolinic acid (picloram) required ratios of 3600:1. Beans (Phaseolus vulgaris L., var. Black Valentine) were used as a bioassay. In the field, activated carbon rates up to 600 lb/A did not completely prevent injury to oats (Avena sativa L., var. Markton), cucumbers (Cucumis sativus L., var. Ashly), and Black Valentine beans by propazine at 2.5 lb/A in Toa silty clay 0, 37, 72, and 93 days after herbicide application. Activated carbon protected oats from picloram at 0.5 lb/A but did not completely protect cucumbers or beans at rates up to 600 lb/A. Activated carbon in untreated herbicide soils was not harmful to growth of beans, oats, and cucumbers in Toa silty clay.