METHODS

The problem was investigated by means of pancreatic-femoral anastomoses. Seventy dogs were used. All animals were fasted for 16 hours preceding the experiment and were anesthetized with sodium amytal(70 mg/kg. intramuscularly, plus a single 50 to roe-mg. dose intravenously, if needed). This barbiturate was selected because it is reported to have little or no effect on blood sugar concentration(10, II). The experiments were performed as follows. A donor dug A was prepared by cannulating the pancreaticoduodenal and one of the femoral veins. The other femoral vein was exposed for withdrawal of blood samples. A recipient dug B, smaller than dog A, was prepared by cannulating the femoral vein and artery on one side and exposing the femoral vein on the other. The blood pressure of the dogs was measured with mercury manometers connected with the carotid arteries. Having thus prepared the animals, a control sample of blood was taken, the animals were heparinized, 2 the pancreaticoduodenal vein of dog A was connected with the femoral vein of dog B and the femoral artery of dug B with the femoral vein of dog A. The anastomoses were made with transparent polyethylene tubing and the free flow of blood ascertained by injecting a minute bubble of air into the tubing before every blood sampling. The pancreatic-femoral anastomosis was kept open continuously. The return flow from the artery of B to the vein of A was opened as needed to maintain arterial blood pressures of both dogs approximately equal to their initial values. Four samples of blood were taken at Is-minute intervals and, after this control period, 5 cc. of a Izo-per cent solution(I gm.) of glucose per kilo body weight or the same volume of iso-osmotic (4%) saline were rapidly injected into the exposed femoral vein of dog A. Blood samples were taken at Is-minute intervals for 75 minutes following the injection, the anastomoses were then disconnected and the sampling continued 15 minutes later and then every 30 minutes for 2 more hours. Blood sugar was determined in duplicate according to the method of Folin and Malmrose(12). In some control experiments a branch of the mesenteric vein of dog A, about equal in size to its