OBJECTIVE The aim of the study was to test the hypothesis that platelet serotonin 5HT2 receptors are important in the genesis of thrombosis in stenosed coronary arteries. METHODS The specific serotonin 5HT2 receptor antagonist, ritanserin was used as a pharmacological tool to examine the effect of removal of the participation of the 5HT2 receptors on thrombus growth, in a paired statistical design. METHODS The study involved 10 open chest anaesthetised dogs, with constrictors of critical diameter applied to the left circumflex coronary artery. RESULTS Blood flow was monitored in the left circumflex coronary arteries, distal to the critical stenosis. Flow reductions occurred that have previously been shown to be caused by the accumulation of platelet thrombi. By embolising the thrombi, the process could be monitored cyclically (cyclic flow reductions). The specific serotonin 5HT2 receptor antagonist, ritanserin, abolished cyclic flow reductions at a dose of 0.5 mg.kg-1. There was no effect on blood pressure or heart rate on administration of ritanserin at any dose. The serotonin blockade by ritanserin also prevented the reestablishment of cyclic flow reductions by adrenaline infusion (0.4 micrograms.kg-1.min-1), but required ritanserin doses up to 1.5 mg.kg-1. Ex vivo aggregation of platelets was reduced in blood taken from the dogs after ritanserin administration. CONCLUSIONS These results constitute further evidence of the possible importance of serotonin as a mediator of platelet thrombosis in stenosed coronary arteries.