1. Hepatic alcohol dehydrogenase activity and leucocyte ascorbic acid content were measured in thirty-five patients with liver disease and in ten control subjects with duodenal ulcer. The patients with liver disease were divided into three groups consisting of non-drinkers, moderate drinkers and alcoholic/heavy drinkers.
2. There was no significant difference in hepatic alcohol dehydrogenase activity between the groups with liver disease, but all patients had less than half the hepatic alcohol dehydrogenase activity of the control subjects (P < 0·001).
3. The ascorbic acid in leucocytes was significantly lower in the alcoholic/heavy drinker group than that in the control subjects (P < 0·02) when the Student's t-test was applied, but no significant difference was found when the Mann—Whitney U-test was used.
4. A correlation coefficient of r = 0·77 (P < 0·001) was observed among the thirty-five patients with liver disease when hepatic alcohol dehydrogenase activity was compared with leucocyte ascorbic acid content. An insignificant correlation (r = 0·332) was found in the control subjects with no liver disease.
5. This comparison was also significant among non-drinkers with liver disease (r = 0·873; P < 0001), moderate drinkers (r = 0·739; P < 0·02) and alcoholic/heavy drinkers (r = 0·702; P < 0·005).
6. The addition of ascorbic acid in vitro (0·5–10 mmol/l) had no effect on the activity of alcohol dehydrogenase.
7. The relation between hepatic alcohol dehydrogenase activity and leucocyte ascorbic acid content is probably a consequence of liver disease, as opposed to any specific effect of ascorbic acid deficiency or alcohol consumption on alcohol dehydrogenase activity.