An oat (Avena sativa L. ‘Dal’) primary root bioassay procedure was used to study the persistence of eight dinitroaniline herbicides in soil. Sensitivity of the bioassay to the herbicides depended on the length of time roots were exposed to the herbicide treatments. Root inhibition increased with time of exposure. Herbicide persistence was studied under growth chamber, greenhouse, and field conditions. The herbicides remained biologically active the longest under dry and cool soil conditions. At 5 C, little change in herbicide activity was observed over an 8-month period. Under greenhouse conditions, the herbicides in wet soil dissipated rapidly, as compared with those in dry soil. Soil samples taken from depths of 0 to 7.5 cm and from 7.5 to 15 cm in the field showed that the herbicides were retained primarily in the surface 7.5 cm. Small amounts were found from 7.5 to 15 cm only shortly after the herbicides were applied. Dinitramine (N4,N4-diethyl-α,α,α-trifluoro-3,5-dinitrotoluene-2,4-diamine) was the only herbicide tested of which soil residues from spring field applications were not detected the following fall.