We have examined the mechanism of herbicide resistance in a biotype of wild oat (Avena fatua L.) that is resistant to diclofop-methyl and many other acetyl-coenzyme A carboxylase (ACCase) inhibitors. Resistance to diclofop-methyl and fenoxaprop-ethyl was not based on reduced uptake nor on enhanced metabolism of the herbicides to inactive products. In in vitro assays of crude or partially purified preparations, ACCase from the resistant (UM-1) and susceptible (UM-5) biotypes was equally sensitive to diclofop, with I50 values of 6.1 and 7.3 μM for UM-1 and UM-5, respectively. Corresponding values for fenoxaprop were 2.5 and 1.0 μM. These results suggest that the high level of resistance observed toward these herbicides is not based on an altered target enzyme. Root tissue from both UM-1 and UM-5 acidified an unbuffered bathing solution. Addition of 100 μM diclofop or fenoxaprop prevented acidification of the bathing medium by UM-1, but alkalinization occurred rapidly with UM-5. When diclofop was removed from the treatment solution, UM-1 resumed acidification of the solution, whereas the pH of the UM-5 bathing solution continued to increase. Diclofop (50 μM) rapidly depolarized the cell membrane in peeled coleoptile sections, with no difference between UM-1 and UM-5. However, when diclofop was removed from the treatment solution, the electrogenic membrane potential was quickly reestablished in UM-1, but remained collapsed in UM-5. These results provide support for the hypothesis that the effect of diclofop on the plasma membrane potential is an important component of its herbicidal activity. The reversibility of the effect of diclofop and fenoxaprop on transmembrane proton flux in UM-1 appears to be associated with resistance to these herbicides. © 1993 Academic Press.