Continuously measuring the myocardial tissue blood flow by heat clearance method using the cross-thermocouple probes whose character has been described elsewhere, the roles, in the regulation of the flow, of the sympathetic nerve, vagal nerve, carotid sinus and the sympathetic afferent fibers with the receptor sites in the thoracic cavity were examined in the anesthetized dogs with the results that their role varied considerably by the layers of the left ventricular wall. In this paper, the differences between the nervous control of the coronary blood flow and that of the tissue blood flow and those between the nervous control of the blood flow in the inner layer and that in the outer layer were represented and the possible mechanisms for the differences were discussed. Method s Experiments were carried out on the mongrel dogs, weighing from 18 to 25 kg, anesthetized with pentobarbital sodium (35-45 mg/kg, I.V.). The arterial Po2 was maintained to normal level by a Bird's respirator. The thoracic cavity was opened by resecting the ribs from the first to 8th on the left side. After pericardiectomy, the anterior descending branch of the left coronary artery proximal to its origin was carefully separated from the surrounding connective tissues not to damage the nerves which run close to it. Thereafter, a probe of electromagnetic flow-meter was set on it. In order to measure the myocardial tissue blood flow, a pair of the thermoprobes were implanted into the interior of the left ventricular wall. The two thermojunctions of the probes were settled nearly in the same distance from either the epicardium or endocardium to avoid the influence on the thermoelectromotive force of the temperature gradients, different heat production and different heat or temperature conductivity in the wall. Thereafter, the strain-gauge bridges for tension measurements were sewn onto the wall adjacent to the probes. In addition, a monopolar electrode for e.k.g. was attached to the wall close to the probes. The left stellate ganglion was freed from the surroundings. After transsecting the branches of the central side, the ganglion was placed across a pair of platinum-ilidium stimulating electrodes. The right cervical vagus was prepared in the same way for electrical stimulation. The nerves were stimulated with the electric pulses of 0.5 msec duration, square wave, 12 V, and of 10 cps. The stimulus frequency was altered when necessary from 0.5 up to 10 cps.