The persistence of [14C]2,4‐D at a rate equivalent to 1 kg/ha was studied in the laboratory on a heavy clay and a sandy loam at 85%of field capacity and 20°C both alone and in the presence of 1 kg/ha dicamba, dichlorprop, difenzoquat, TCA, and 2,4,5‐T. The persistence of 2,4,5‐T was also monitored in both soils under the same conditions in the presence and absence of [14C]2,4‐D. All soils were extracted at weekly intervals using aqueous acidic acetonitrile and analysed for [14C]2,4‐D remainining radiochemical techniques. The extracts containing 2,4.5‐T were additionally analysed gas chromatographically for that herbicide. In each soil type the half‐life of the 2,4‐D was similar regardless of whether applied singly or in combination with the five herbicides tested. Similarly, [14C]2,4‐D did not affect the breakdown of 2,4,5‐T in either soil type. The persistence of tri‐allate (1·5 kg/ha) and trifluralin (0·75 kg/ha) both singly and in combination were compared using small field plots at two locations in Saskatchewan. Applications were made during May of 1977 and 1978 and the plots were sampled and analysed for herbicide(s) remaining after 10 and 20 weeks, respectively. The results indicate that within experimental error the loss of both tri‐allate and trifluralin from the plots treated with the mixture was the same as from plots treated with the individual compounds. Copyright © 1979, Wiley Blackwell. All rights reserved