1. When weanling rats were given a vitamin E-deficient diet and 0.15% (w/v) silver acetate in their drinking water, there was a high incidence of liver necrosis and death after 2–4 weeks. This was prevented by 120 ppm D-α-tocopheryl acetate in the diet, was partially prevented by selenium at 1 ppm but not at 0.05 ppm and was only marginally prevented by 0.15% DL-methionhe.2. All these effects were observed when the diet was free of fat and when it contained either methyl oleate or polyunsaturated methyl esters. The polyunsaturated lipid slightly increased the velocity of the terminal illness.3. In spite of the ‘anti-vitamin E’ effect of Ag and the clear demonstration that α-tocopherol played a major part in preventing toxicity, experiments with tracer amounts of [5-Me-14C]-D-α-tocopherol showed that rather than there being any destruction of tocopherol in the critical period preceding the onset of disease, there was an increase in tocopherol in the liver due to Ag administation. The metabolism of tocopherol in the remainder of the animal was unaffected during the pre-necrotic phase.4. The results are inconsistent with the view that the stress induced by Ag is caused by a pro-oxidant effect, either in the gastro-intestinal tract or in the tissues of the rat. They show that lipid peroxidation is not a causal factor in the aetiology of Ag-induced liver necrosis and suggest that stress in the vitamin E-deficient animal probably raises the requirement for tocopherol.