Ray and Wolff in a landmark study of human patients under local anesthesia, concluded that the brain was not sensitive to pain; however, at the time of their study, the anatomy and physiology of pain transmission and modulation were largely unknown and their stimulating electrodes were not implanted in the brainstem or thalamic cells or projections now known to be important to pain perception. We now report 15 patients, previously headache‐free, who underwent electrode implantation in the periaqueductal gray between 1977 and 1982 who immediately at implantation or in the few days subsequent to implantation reported severe continuous head pain usually with florid “migrainous” feature that persisted for 2 months to 10 years. Ten of these patients were treated with reserpine and all were dramatically responsive to it, but 8 patients rapidly became tolerant. Seven patients who were treated with dihydroergotamine rapidly became headache‐free; 2 of the 7 became tolerant quickly. One patient developed the “cough headache” syndrome after implantation, was responsive to indomethacin, the syndrome abating in 6 months. These data suggest that perturbation of brain may generate head pain.