The influence of temperature on gastric blood flow was estimated in anesthetized dogs. The temperature of the gastric wall was controlled by circulating a fluid at a given temperature through an intragastric balloon. Gastric venous blood flow was measured in 2 preparations; one in which only a portion of the blood returned through the splenic vein and the other in which all gastric venous flow returned through the splenic vein. A "Y" cannula was inserted in the splenic vein. With appropriate clamping of one or other of its arms, returning blood would either flow into the portal vein or could be collected in a graduated cylinder. Where the gastric venous isolation was incomplete, reduction of the stomach wall temperature to 15[degree] C resulted in a mean fall in blood flow in 6 animals of 67%. Following complete venous isolation of the stomach, flows were significantly higher; however, the percent fall in blood flow at 15[degree] C was the same. During hyperthermia (48[degree] C) of the stomach wall the mean blood flow in 6 animals increased 29%. The volume in the intragastric balloon was the same in all phases of each experiment; inflation of the balloon had little effect in itself on the blood flow.