Several reports of investigations in which peffused rat livers were used to study plasma protein synthesis (1, 2) as well as other basic hepatic functions (3-5) suggested that these preparations offered many possibilities for the investigation of the factors involved in protein metabolism under well defined experimental conditions. Consequently, a peffusion apparatus was constructed and preliminary experiments were conducted to test the disappearance rate of Pa~-labeUed chromic phosphate in colloidal form from the perfusate. The results of a typical experiment, together with the bile flow prevailing at the same time are presented.Attention was then turned to plasma protein synthesis using L-methionine-S 85 as the tracer amino acid. The results showed that the livers were capable of rapidly incorporating this amino acid into protein which appeared in the perfusate. Regardless of the initial experimental conditions employed, however, all the specific activity curves for the protein in the perfusate approached a plateau at the end of 4 hours. Similar results have been obtained by others (1). This phenomenon could be due to either the loss of synthetic ability by the liver, or to the depletion of substrate, or to a decrease in the specific activity of the labelled amino acid. The first possibility appeared unlikely, since it was found in confirmation of other work (1) that when the incorporation was proceeding at a reduced rate the addition of more labelled amino acid resulted in an immediate increase. Consequently, studies were designed to ascertain whether the cause of this plateau was due to the rapid dilution *