A B S T R A C T Extraction of insulin by the liver after administration of glucose in the duodenum has been studied in fourteen anesthetized dogs. Plasma insulin and glucose were measured in the portal vein hepatic vein and hepatic artery. During the control period 40 ±3% of the approximately 11 mU of insulin presented to the liver/min was removed during a single transhepatic passage. Within 5 min after glucose administration, the amount of insulin reaching the liver increased significantly. In some animals this increase preceded any significant increase in the glucose concentration of the femoral artery. After glucose administration, hepatic extraction of insulin remained unchanged in five animals and rose significantly in nine. In five of the latter animals, the increase may have been more apparent than real due to nonrepresentative sampling of hepatic venous blood. However, for the whole group of animals, comparison of arterial insulin levels with the amount of insulin delivered to the liver suggested a transient increase in insulin extraction between 5 and 50 min after glucose administration. In no animal was there a decrease in the proportion of insulin extracted by the liver after glucose administration. The results indicate that the extraction process is not saturable at physiological insulin levels. Prior to glucose administration, net hepatic glucose output averaged between 30 and 40 mg/min. After glucose administration, the liver began to take up glucose and there was a significant correlation between hepatic glucose uptake and the amount of insulin reaching the liver. However, since the amount of glucose presented to the liver also increased, it is not established that the insulin was responsible for the change in hepatic carbohydrate metabolism.The data demonstrate an increase in the absolute amount of insulin extracted by the liver after glucose administration and an important role for the liver in regulating peripheral insulin concentrations.