There has been a recent increase in the number of studies dealing with migraine-stroke. I describe five patients in whom migraine-stroke was the clinical diagnosis but in whom the subsequent clinical events or autopsy showed a different mechanism for the cerebral infarction. Three patients had arterial dissection (one proven at autopsy), one had marantic endocarditis that had been missed on two echocardiograms (proven at autopsy), and one had generalized atherosclerosis and diabetes. These patients demonstrate that important and different etiologies may produce what seems to be the migraine-stroke syndrome. The result may be failure to recognize specific therapeutic measures that could have vitally important benefit to the patient {Stroke 1991^22:1068-1074)