By means of the bromsulphalein method it has been shown in man that insulin hypoglycaemia is accompanied by an increase in liver blood flow (Bearn, Billing & Sherlock, 1952). These experiments, however, threw little direct light on the relative importance of hepatic artery and portal vein, nor could they really be regarded as conclusive in regard to the importance of nervous or endocrine factors.The first object of the present work was, therefore, to apply the method of 'internal calorimetry', a heated thermocouple technique which depends on the measurement of thermal conductivity (Grayson, 1952; Hensel, Ruef & Golenhofen, 1954), to the fuller investigation of this phenomenon in rats and baboons.The present work has had, too, a subsidiary object, namely, to determine changes in hepatic heat production after giving insulin.
METHODSWistar strain albino rats (200-250 g) of either sex were used. Many of the experiments were carried out on the conscious animal. In these experiments liver recorders were implanted under ether anaesthesia, usually on the previous day. The leads from the animals were soldered to the leads from the apparatus. No restraint other than the animal's usual cage was used, the leads from the apparatus hanging freely from above.In other experiments the rats were anaesthetized with pentobarbitone sodium (Nembutal, Abbott Laboratories; 60 mg/kg intraperitoneally). Polythene cannulae were inserted into the trachea, a carotid artery, a femoral artery and, usually, an external jugular vein, after heparinization (Liquemin, Roche Products, 100 ug/kg intraperitoneally). Liver blood-flow recorders were inserted through a mid-line abdominal incision. Baboons (Papio anubos) were anaesthetized by ethyl chloride induction followed by open ether and intraperitoneal pentobarbitone sodium (75 mg/kg).Liver blood flow. The technique of internal calorimetry fully described elsewhere (Birnie & Grayson, 1952;Grayson, 1952) was employed for assessment of liver blood flow. This technique depends on a thermo-electric measurement of liver thermal conductivity. It has been shown elsewhere (Grayson, 1952) that conductivity increment (i.e. thermal conductivity of living liver less the thermal conductivity of dead liver) is an approximately linear function of blood flow.