Chronotropic actions of morphine sulfate were studied by selective perfusion of the sinus node in 111 dogs. Choice of anesthetic did not influence the results. The following results were obtained: The minimal effective concentration of morphine was 1,000 μg/ml, which caused sinus slowing (25 beats/min, lasting 21 min) in control dogs but had no effect in dogs pretreated with reserpine. Tachyphylaxis was distinct and prevented the use of any dog as its own control. Bradycardia was not altered by atropine or vagotomy; in fact, intranodal administration of morphine produced a small but consistent vagal blocking effect. Intranodal administration of morphine had no effect on the acceleration produced by stellate stimulation or by selective perfusion of the sinus node with norepinephrine, and propranolol did not prevent the bradycardia from morphine. Even with high concentrations of morphine, which cause bradycardia, there is little influence on local autonomic control. However, morphine induced bradycardia does require the presence of some slowing substance existing in the normal canine sinus node, but not in the sinus node of reserpinized dogs.