Trichloroacetic acid (TCA) has been implicated as an airborne pollutant responsible for adverse effects on forest health. There is considerable debate as to whether TCA observed in trees and forest soils is derived from atmospheric deposition or from in situ production. This experiment reports the results from treating 4-year-old Sitka spruce (Picea sitchensis (Bong.) Carr) plants in a greenhouse over a growing season with TCA supplied either to the soil or to the foliage at concentrations of 10 and 100 ng mL(-1). Similar uptake of TCA by needles was observed for both modes of treatment, with significant accumulation of TCA (300 ng g(-1) dry wt) at the higher concentration. Larger concentrations in stem tissue were seen for the foliar-applied TCA (280 ng g(-1)) than for the soil-applied TCA (70 ng g(-1)), suggesting that direct stem uptake may be important. Six months after treatments stopped, TCA concentrations in the needles of plants exposed to 100 ng mL(-1) TCA were still enhanced, showing that biological degradation of TCA in needles was slow over the winter. By contrast, no significant enhancement of TCA in soil could be detected in the directly treated soils even during the experiment. The protein content of needles treated with the higher concentration of TCA by either route was significantly smaller than for the controls, but there was no effect of TCA on the conjugation of 1-chloro-2,4-dinitrobenzene in roots nor on the conjugation of 1,2-dichloro-4-nitrobenzene in needles.