One hundred and six consecutive patients with liver disease were selected on the basis of elevated serum transaminase levels. The patients were randomly allocated into a group treated with silymarin (treated) and a group receiving placebo (controls). Ninety-seven patients complete the 4-week trial-47 treated and 50 controls. In general, the series represented a relatively slight acute and subacute liver disease, mostly induced by alcohol abuse. There was a statistically highly significantly greater decrease of S-SGPT (S-ALAT) and S-SGOT (S-ASAT) in the treated group than in controls. Serum total and conjugated bilirubin decreased more in the treated than in controls, but the differences were not statistically significant. BSP retention returned to normal significantly more often in the treated group. The mean percentage decrease of BSP was also markedly higher in the treated. Normalization of histological changes occurred significantly more often in the treated than in controls.