In 1951, the advocacy of treatment for recent vascular accidents by vasodilators or by stellate block in some clinics led us to examine closely the mechanism of "vasospasm," at that time commonly held to be the basis of transient hemiplegic episodes. The increasing performance of arteriography had by then confirmed the hitherto unsuspected frequency of occlusion of one carotid,1,2,10 and my co-workers and I, like others, were struck by the common association of such carotid occlusion with a history of recurrent episodes of transient hemiplegia, often over a period of months or years, to which attention was first drawn by Egas Moniz, Lima, and de Lacerda38 in 1937. In a series of our own cases 5 we cited the common precipitation of a transient attack by a situation that lowered the level of systolic blood pressure or the cardiac output. Loss of blood, gastrointestinal bleeding, syncope, and the use