Objective: To investigate the associations between Parkinson's disease and other degenerative parkinsonian syndromes and environmental factors in five European countries. Methods: A case-control study of 959 prevalent cases of parkinsonism (767 with Parkinson's disease) and 1989 controls in Scotland, Italy, Sweden, Romania and Malta was carried out. Cases were defined using the United Kingdom Parkinson's Disease Society Brain Bank criteria, and those with drug-induced or vascular parkinsonism or dementia were excluded. Subjects completed an interviewer-administered questionnaire about lifetime occupational and hobby exposure to solvents, pesticides, iron, copper and manganese. Lifetime and average annual exposures were estimated blind to disease status using a job-exposure matrix modified by subjective exposure modelling. Results were analysed using multiple logistic regression, adjusting for age, sex, country, tobacco use, ever knocked unconscious and family history of Parkinson's disease. Results: Adjusted logistic regression analyses showed significantly increased odds ratios for Parkinson's disease/parkinsonism with an exposure-response relationship for pesticides (low vs no exposure, odds ratio (OR) =1.13, 95% Cl 0.82 to 1.57, high vs no exposure, OR =1.41, 95% Cl 1.06 to 1.88) and ever knocked unconscious (once vs never, OR= 1.35, 95% Cl 1.09 to 1.68, more than once vs never, OR = 2.53, 95% Cl 1.78 to 3.59). Hypnotic, anxiolytic or antidepressant drug use for more than 1 year and a family history of Parkinson's disease showed significantly increased odds ratios. Tobacco use was protective (OR = 0.50, 95% Cl 0.42 to 0.60). Analyses confined to subjects with Parkinson's disease gave similar results. Conclusions: The association of pesticide exposure with Parkinson's disease suggests a causative role. Repeated traumatic loss of consciousness is associated with increased risk.