The responses in cervicogenic headache to four different agents have been studied. Nitroglycerin was given sublingually to 27 patients. Eighteen patients got more than 20% increase of their headache. Of those with any headache increase at all, 12 got bilateral and 12 unilateral pain. The typical late cluster headache response to nitroglycerin was not seen in cervicogenic headache. The provocative effect of nitroglycerin seemed less marked in cervicogenic than in cluster headache. Oxygen inhalation, a frequently used treatment for cluster headache, was given to 14 patients with cervicogenic headache. In general, the effect seemed uncertain and probably clearly inferior to the effect in cluster headache. Ergotamine treatment (given to 13 patients) also seemed to be of little avail in cervicogenic headache. Morphine injections given to 11 cervicogenic headache patients resulted in "marked" improvement in 4, but complete pain freedom was only seen in 2 cases. In our opinion, the present results add further evidence to the view that different etiologic and pathogenetic factors underlie cervicogenic headache and cluster headache.