One drawback to using perfluorochemical emulsions as blood substitutes is that perfluorochemical particles are cleared from the blood by the reticuloendothelial system, primarily liver and spleen. We measured the impact of two perfluorochemical emulsions on clearance of colloidal carbon (< 1 μm) and 51Cr-sheep red blood cells (about 8 μm) by the reticuloendothelial system in vivo and in the isolated perfused liver. Male rats were injected with 2 ml/100 gm body wt of Fluosol-DA or Oxypherol-ET for 4 consecutive days. Carbon (1 ml/100 gm body wt) or sheep red blood cells (0.05 ml of 5% vol/vol/100 gm body wt) were then injected intravenously (in vivo) or added to perfusate. Samples were taken at several time points for 1 hr. In the isolated perfused liver, carbon clearance was depressed by 25% 1 day after treatment. Rates returned to control levels by 12 days in Fluosol-DA—treated rats but remained depressed by 67% in Oxypherol-ET—treated rats. Sheep red blood cell (8 μm) clearance was two to five times slower than carbon clearance and depressed by 40% in livers from Fluosol-DA rats 1 day and 12 days after treatment. Added serum did not improve phagocytosis. In vivo carbon clearance remained normal in Fluosol-DA—treated rats but decreased by 74% in Oxypherol-ET—treated rats 1 day after treatment, returning to normal by 12 days. Clearance rates were similar in control rats in vivo and in the perfused liver. We conclude that the isolated perfused liver is a good model to measure liver clearance function. Although low doses of perfluorochemical emulsions may depress Kupffer cell phagocytosis, general reticuloendothelial system function is not significantly compromised. (HEPATOLOGY 1991;14:169–174.)