Since Wolff's original proposal regarding the vascular etiology of cyclic head pain, evidence has accumulated that the prodromes of migraine are due to cerebral vasoconstriction and headaches of both cluster and migraine are due to painful dilatation. Theories regarding their pathogenesis include cyclic release of vasoactive substances from platelets and/or other sources (such as serotonin, catecholamines, histamine, acetyl choline, prostaglandins, substance P, endogenous opiates). These substances influence vasomotor receptors bringing about abnormal constriction and/or dilatation. Drugs which modify receptors (such as methysergide, alpha and beta blockers, antihistaminics, anticholinergics, steroids and non-steriodal anti-inflammatory agents) have had some therapeutic success in migraine but provide little benefit for cluster patients. Ca2+ entry blockers (including nimodipine, nifedipine, verapamil) theoretically should diminish cephalic vasoconstriction and -dilatation no matter what their cause. To test this, 35 headache patients with classic (N = 13), common (N = 14) migraine or cluster (N=8) were evaluated by double-blind, cross-over, randomized assignment to high or low dose nimodipine therapy. Within 10 days migraine prodromes became infrequent and after 2–4 weeks headache frequency was significantly reduced for migraine and within 4–6 weeks for cluster. CBF measurements during oxygen inhalation showed reduced cerebral vasoconstrictive responses after high dose nimodipine. Associated muscular contraction headaches were not altered. Nifedipine and verapamil provided equivalent relief for cluster but produced more side effects, and were less effective, than nimodipine in control of migraine.