A case-control study was performed to investigate the significance of arteriosclerosis, heredity and some infections in the etiology of Parkinson's disease. The study group consisted of all traceable patients with Parkinson's disease living in a defined area, a total of 444 patients, and of control subjects for each patient, matched in sex and age, chosen from among the general population residing in the same area. No significant differences were found between the patients and the controls concerning the occurrence of cardiac insufficiency, coronary heart disease, or stroke. The Parkinsonian patients, however, had a significantly lower incidence of clinical arterial hypertension when compared with the controls. In addition, the patients more often had low systolic blood pressures and more rarely high pressures than the controls. Even the mean systolic blood pressure was significantly lower in the patients than in the controls. The low blood pressure seems to be an effect of Parkinson's disease itself with a minor contribution of levodopa therapy. The observations above are considered to indicate that arteriosclerosis and Parkinson's disease are probably only concurrent disorders and not in etiological relationship with each other. There was no statistically significant difference in the proportion of the patients and the controls with relatives with Parkinson's disease or essential tremor, which suggests that genetic factors do not have a significant role in Parkinson's disease and on the other hand that essential tremor and Parkinson's disease are two separate disease entities. No other encephalitis than a lethargic one was found to precede Parkinson's disease and the occurrence of meningitis was rare both among the patients and the controls. The history of Spanish influenza was found to be as frequent in the patients as in the controls, thus not supporting the idea that influenza has etiological importance in Parkinson's disease.