This investigation was conducted to clarify the epidemiology of Parkinson's disease in Finland. A community survey was made in a selected area in southwest Finland in addition to an analysis of Finnish mortality statistics for Parkinson's disease. The annual mortality rate from Parkinson's disease was found to average 2 per 100,000 population. Almost 70 per cent of the deaths occurred between 65-79 years of age. The total and age-specific mortality rates for males were greater than those for females whereas the proportionate mortality rates were almost identical. This was considered to indicate that suggestions of greater male prevalence, based on mortality statistics, do not appear justified. On prevalence day, Dec. 31st, 1971, 484 patients with Parkinson's disease (of which 444 were personally examined) lived in the area of investigation (population 402,988), the prevalence rate being 120.1 per 100,000 population. The highest annual incidence rate was 16.6 per 100,000 population. The age-specific prevalence rates showed a rapid increase after the 50th year of age. The greatest prevalence was shown by the age group 70-79 years of age in which almost 0.8 per cent of the population are affected. Age-specific incidence rates also displayed an increase after the 50th year of age. The greatest incidence was observed in the age group 70-79 years of age in which almost 1 per 1,000 of the population are annually affected by the disease. A difference between the sexes was demonstrable in the prevalence and incidence rates showing greater values for females, but in the age-specific frequencies the differences were reduced. This probably reflects the difference between the age structures of the male and female populations, suggesting that both sexes have a similar risk of being affected by the disease. The permanent age structure shown by idiopathic patients in comparison with previous investigations as well as the increased mean age and proportionate decrease of postencephalitic patients was found to be in disagreement with the cohort theory according to which all parkinsonian patients are previously victims of encephalitis lethargica.