A steep hill country area (1.66 hal of gorse and pasture was sprayed with 2,4,S-T butyl ester by air. Soil and vegetation samples were collected at regular intervals and a stream at the bottom of the trial area was monitored continuously over 6 months. Only 0.6070 of the applied herbicide entered the stream during the trial, with 0.4070 being washed out in the first storm. Measurable traces of the original ester formulation, 2,4,S-T (acid), and the principal breakdown product, 2,4,S-trichlorophenol (2,4,S-TCP), with a combined average of about 0.4 mg/kg, were found in pasture vegetation 6 months after initial application. The persistence of these residues may be a cause of dietary intake by grazing livestock. Soil residues (expressed as butyl ester) persisted throughout the study, principally as 2,4,S-TCP (12-100 ng/g after 18S days), and were also detected from a treatment 1 year before the trial began (20 -90 ng/ g). An estimate of the masses of residues and losses 6 days after spraying accounted for about 20% of the applied mass. A significant part of the balance (80%) is thought to have been lost in spray drift.