An isolated perfused rat heart preparation has been used in an attempt to define the mechanisms underlying the coronary vasoconstriction associated with the intraarterial infusion of cardioplegic solutions containing contaminant particles. Coronary infusion of commercially available intravenous solutions, compounded to create the St. Thomas' cardioplegic solution, resulted in a 40.3 +/- 1.9% decline in coronary flow rate at the end of a 20-minute period of infusion. Filtration of the solution through a 0.8 micron filter reduced this decline in flow rate to 23.4 +/- 2.3% of its control value. Inclusion of the 5HT2 receptor antagonist ketanserin (1.0 nanomoles/l), in the unfiltered cardioplegic solution, afforded the same protection as filtration. Inclusion of 5HT (0.1 micromole/l) in a filtered cardioplegic solution resulted in a reduction of coronary flow (49.7 +/- 1.9%) greater than that seen with the unfiltered solution. Studies with antagonists of the histamine receptor (mepyramine and cimetidine) and agonists and antagonists of adrenergic receptors suggested that, although involved in the control of vascular tone, these receptor systems are unlikely to be involved in the process of particle-induced vasoconstriction. Arguments are presented that 5 HT and its receptors may play a critical role and the hypothesis is advanced that particle-induced focal endothelial injury may be the primary lesion in a complex sequence of events leading to induction of transient vasoconstriction.