The epidemiology of cluster headache is virtually unknown. Using the unique resources of the Rochester Epidemiology Project for population-based studies, we identified 26 Olmsted County, Minnesota, residents who fulfilled the International Headache Society criteria for newly diagnosed cluster headache between 1979 and 1981. As part of a large study of migraine and cluster headache, we screened more than 6,400 patient records from several diagnostic rubrics to obtain this cohort, accomplished case ascertainment exclusively through medical record review, and assigned diagnoses by the consensus of two neurologists. The age-adjusted incidence was 15.6 per 100,000 person-years (p-y) (95% CI, 8.9 to 22.3) for males and 4.0 per 100,000 p-y for females (95% CI, 0.4 to 7.6). The overall age- and sex-adjusted incidence was 9.8 per 100,000 p-y (95% CI, 6.0 to 13.6) or approximately 1/25 that of migraine. The peak incidence was among men aged 40 to 49 years and women 60 to 69 years. There was a higher than expected prevalence of history of smoking among males with cluster headaches (p < 0.05), supporting the possibility that smoking predisposes to the development of cluster headaches in men.