Postprandial serum bile acid estimation was recently proposed as the most sensitive test of liver function. In our study, the fasting and postpranidal serum bile acid measurements were performed on 19 normal subjects, 20 patients with cirrhosis, 10 with acute hepatitis, 4 with resolving viral hepatitis, and 6 with chronic active hepatitis. A gas-chromatographic method was used. One healthy subject had postprandial serum bile acid levels above the normal range, while 7 patients with liver disease had postprandial levels within normal limits. Of the latter group, 2 had chronic active hepatitis in remission and 3 had resolving viral hepatitis. Significant correlations were seen between serum bile acid levels and most of the conventional "liver function" tests. Our data indicate that the postprandial serum bile acid determination is better than any of the other conventional tests taken separately, but no better than their combined use. No significant modification of the cholic acid/chenodeoxycholic acid ratio was observed between the fasting and the postprandial determinations.