In a series of experiments aimed at evaluating copper oxide as a supplement, grazing sheep were dosed with varying quantities of copper oxide particles up to 64 g, and the toxicity, the rate of particle excretion, and copper storage in the liver and other tissues were recorded. The toxicities (LD50) of copper oxide particles were 1.17 and 1.80 g/kg liveweight for two groups of grazing adult fine-wool Merino sheep. Death usually occurred 88-96 days after the oral administration of the particles; mean hepatic copper concentrations of sheep dying from copper toxicity were 4122-4308 mg/kg DM. The rate of faecal copper excretion of copper-supplemented sheep, expressed as a percentage of the dose, was less when 50 g of particles were given than when the dose was 5 or 10 g, but excretion patterns were variable. The quantity of hepatic copper stored per g of copper oxide given declined as the dose increased from 0 to 32 g, but increased again at higher doses. Hepatic copper concentration reached a maximum 2-3 months after dosing and the rate of decline was positively related to dose rate; thus, large doses of copper are unlikely to extend the period of elevated copper status. Large doses also increased whole blood copper concentrations and elevated plasma aspartate transaminase activities; this was taken to indicate copper poisoning. Tissue copper concentrations from sheep given up to 64 g particles are reported; liver was the most sensitive to copper treatment. Copper contents of the lung and kidney also responded to copper therapy, but carcass components such as leg, shoulder and muscle did not. Weaned lambs given 2 g of particles (c. 0.13 g/kg liveweight) grew significantly faster than unsupplemented lambs. This dose rate was approximately one-seventh of that predicted to cause 5% mortality, and it is concluded that, at appropriate dose rates, this method of supplementation did not increase mortality or cause excessive increases in tissue copper concentrations, and could increase growth rate.