Estimates of the atmospheric deposition of sulphur compounds to forests can be made from measurements of throughfall and stemflow below the forest canopy if internal cycling of sulphur can be neglected. The use of (SO4)-S-35 as a tracer isotope to study movement of sulphate from the soil to the tree and subsequent leaching by rain is described, and illustrated by reference to a field experiment in a Scots pine forest. Application of 1.85 GBq of (SO4)-S-35 to an area of 35 m2 of the forest floor resulted in measurable activity in both foliage and throughfall. Activity in foliage continued to increase through the winter after application in June. Activity in throughfall was initially high (20 Bq S mg-1), then fell to a steady value (<2 Bq S mg-1) from August onwards. There was good spatial correlation between activity in the canopy and in throughfall below canopy. The temporal changes showed that complete and rapid equilibration of the isotope with the pool of sulphur in the canopy could not be safely assumed. Laboratory experiments with excised shoots demonstrated a much smaller specific activity (Bq mg-1 S) in leached sulphate than in the whole needles. Nevertheless, leaching contributed only a small proportion (<15%) of the sulphate in net throughfall (throughfall + stemflow - rain), except during the period of needle expansion, which coincided with application of the isotope. The results suggest that the errors involved in neglecting internal cycling are of the same order as the uncertainties with which sulphur deposition in throughfall and stemflow can be measured. Such measurements may therefore be used to estimate sulphur deposition to forests with uncertainties comparable to those of current methods based on micrometeorological methods.