We have measured the activity of S-adenosyl-L-methionine synthetase, the ratio between the high- and low-molecular-weight forms of this enzyme and the concentration of S-adenosyl-L-methionine in liver biopsies from a group of controls (n = 6) and in six cirrhotics (five posthepatitic and one alcoholic). The total activity of S-adenosyl-L-methionine synthetase was markedly reduced in cirrhosis (37.5% of that found in the control group). This was due to a specific reduction in the high-molecular-weight S-adenosyl-L-methionine synthetase in the group of cirrhotics (73.9 pmoles per min per mg protein) when compared with that observed in controls (460.3 pmoles per min per mg protein). Despite this reduction in the rate of synthesis of S-adenosyl-L-methionine (the high-molecular-weight form of the enzyme is 15 times more active than the low-molecular-weight form at physiological concentration of substrates), the concentration of this metabolite was the same in the control group (17.3 ± 2.6 μM) and in the group of cirrhotics (17.8 ± 3.1 μM). To explain these findings, it is postulated that in human liver, where the concentration of S-adenosyl-L-methionine is lower than the Km values of a variety of enzymes that use this metabolite (around 50 to 100 μM), a reduction in the synthesis of S-adenosyl-L-methionine is compensated by a reduction in the rate of utilization of this molecule without affecting the intrahepatic concentration of S-adenosyl-L-methionine. This postulate was tested by decreasing the activity of the high-molecular-weight S-adenosyl-L-methionine synthetase in rat liver by treatment with galac-tosamine. In the rat liver, the concentration of S-adenosyl-L-methionine is about 100 μM. Galactosamine treatment reduced the high-molecular-weight S-adenosyl-L-methionine synthetase from 66.5 pmoles per min per mg protein in the control group (n = 9) to 31.4 pmoles per min per mg protein (n = 8). This reduction of the activity of the high-molecular weight S-adenosyl-L-methionine synthetase was accompanied by a reduction of the concentration of S-adenosyl-L-methionine from 111.3 ± 5.7 μM in the control group to 56.9 ± 3.2 μM in the group of animals treated with galactosamine. Our results support the hypothesis that in human cirrhotic liver there is a marked reduction in the rate of synthesis and utilization of S-adenosyl-L-methionine.