The effect of intracoronary arterial injections of normal saline, contrast agent, hyperosmotic glucose, and mannitol solutions on the sinus rate was studied in 18 patients. Sinus slowing was not produced by isosmotic saline (0.3 osmol/kg of water) when injected into either coronary artery, whereas 76%, Urografin (1.9 osmol/kg of water), 20% glucose (1.3 osmol/kg of water), and mannitol (1.2 osmol/kg of water) decreased the sinus rate significantly (P < 0.001). Among these solutions, Urografin with the highest osmolality produced the most marked sinus slowing, whereas no significant difference was detected between change in the sinus rate with hyperosmotic glucose and mannitol solutions (P > 0.5). There was no significant difference between the degree of sinus slowing produced by intracoronary injections into the coronary artery of the side which the sinus node artery originated from and the slowing produced by contralateral injections with hyperosmotic solutions (P > 0.2). It seems unlikely that this sinus slowing would be a result of a direct action on the sinus node including hypoxia and elevated perfusion pressure. Reflex-mediated parasympathetic stimulation may be operative.