Coronary vasomotion, which controls myocardial perfusion, is closely regulated by myocardial metabolic demand and has been assessed by changes of coronary vascular resistance. Only two decades ago evidence of neural and hormonal control overriding the metabolic control of coronary arterioles, which represent the "resistance vessels" was reported (fig 1). Numerous groups reported persistence of a coronary vasodilator reserve or progressive vasoconstriction despite ischaemia.1-11 Studies showed that a adrenergic receptor mediated coronary constriction may restrict metabolic dilatation to avoid wasting blood flow to the heart muscle. ' Springer Verlag, 1988:454-64.) culature for various mediators, and the physiological significance of such mediators remains obscure.Vasomotion of large conductance vessels does not normally contribute to regulation of coronary blood flow. However, in the presence of atherosclerotic lesions, coronary thrombosis, and spasm, large vessels limit myocardial perfusion and induce ischaemia, angina pectoris, and infarction. Abnormal vasomotion of large coronary vessels has been appreciated in the last decades, when coronary angiography became available during acute coronary syndromes. Atherosclerotic vessels seem to respond with vasoconstriction in situations where vasodilatation should be required. The endothelium and mediators released by the endothelium play a critical role in mediating vasomotion in normal and diseased large and small coronary vessels. However, multiple potential mediators confuse our understanding of coronary vasomotion.Previously, agonists and antagonists of various mediator systems have been used to identify a potential role for coronary vasomotion. Effects of agonists are dose dependent, and it remains unclear from most studies which concentrations of agonists may occur under physiological or pathophysiological conditions. The use of antagonists might increase our understanding, but the specificity of available inhibitors is limited. In addition, the effect of mediators is balanced by other mediators, and blockade of one may activate others. For the coronary circulation, the situation is even more complicated since coronary blood flow is under the influence of aortic pressure, compressive forces of the ventricle, and myocardial metabolic demand. Thus interventions in the coronary circulation with any agonist or antagonist result in complex changes which are difficult to analyse. Response to mediators may be different in various sections of coronary circulation,'9 and diseases which alter vasomotion may affect large and small vessels differently. Finally, acute effects of an agonist or antagonist may be different from chronic effects. Changes of coronary vascular muscle tone are responsible for acute coronary vasomotion, but coronary vasomotion may well be influenced by chronic morphological changes of the coronary vessel wall and perivascular tissue.