Rehmannia glutinosa was extremely tolerant to paraquat at the levels of leaf disk and chloroplast. The tolerance level of R. glutinosa was much greater at the level of leaf disk than at the level of chloroplast, indicating that the tolerance factor(s) are mainly located external to the chloroplast envelope. Possible tolerance mechanisms of R. glutinosa to paraquat were examined in comparison with paraquat-susceptible corn and soybean. Epicuticular wax and cuticle contents did not correspond to differential paraquat effects on the three plant species. Cuticular penetration of paraquat was almost the same in the three plant species. Autoradiography experiments with [C-14]paraquat indicated that the paraquat absorption and translocation in R. glutinosa leaves were not restricted. In addition, the time-and the concentration-dependent paraquat accumulation into the vacuole of R. glutinosa was found to be very similar to those of corn and soybean. However, paraquat binding to the cell wall was greater in R. glutinosa than in corn and soybean, presumably resulting in the reduction of paraquat movement to its site of action in R. glutinosa. The levels of lipid peroxidation in the thylakoid membrane caused by paraquat were found to be almost the same in the three plant species, suggesting that altered site of action is not a mechanism of paraquat tolerance in R. glutinosa. The activities of protective enzymes of superoxide dismutase, ascorbate peroxidase, dehydroascorbate reductase, and glutathione reductase in the stromal extracts were ail higher in R. glutinosa than in corn and soybean. However, the enhanced activities of the protective enzymes are not likely to account for the paraquat tolerance of R. glutinosa, since the activities did nor correlate with the differential paraquat effects on the three plant species. Thin-layer chromatographic analysis of leaf extracts revealed that paraquat metabolism occurred in intact leaves of R. glutinosa, but not in the chloroplast. Our results suggest that paraquat can be metabolized by unknown factor(s) in the cytosol, nor in the chloroplast, of R. glutinosa leaves. Collectively, it can be concluded that the tolerance of R. glutinosa to paraquat might mostly be due to the consequence of paraquat metabolism outside of the chloroplast, although paraquat binding to the cell wall is also likely to confer the tolerance to some extent. (C) 1997 Academic Press.