A wide variety of nutritional exposures have been proposed as possible risk factors for Parkinson's disease (PD) with plausible biological hypotheses. Many studies have explored these hypotheses, but as yet no comprehensive systematic review of the literature has been available. MEDLINE, EMBASE, and WEB OF SCIENCE databases were searched for existing systematic reviews or meta-analyses of nutrition and PD, and one meta-analysis of coffee drinking and one meta-analysis of antioxidants were identified. The databases were searched for primary research articles, and articles without robust methodology were excluded by specified criteria. Seven cohort studies and thirty-three case–control (CC) studies are included in the present systematic review. The majority of studies did not find significant associations between nutritional factors and PD. Coffee drinking and alcohol intake were the only exposures with a relatively large number of studies, and meta-analyses of each supported inverse associations with PD. Factors that were reported by at least one CC study to have significantly increased consumption among cases compared with controls were: vegetables, lutein, xanthophylls, xanthins, carbohydrates, monosaccharides, junk food, refined sugar, lactose, animal fat, total fat, nuts and seeds, tea, Fe, and total energy. Factors consumed significantly less often among cases were: fish, egg, potatoes, bread, alcohol, coffee, tea, niacin, pantothenic acid, folate and pyridoxine. In three cohort studies, two reported borderline decreased relative risks and one a significant increased risk with vitamin C intake. One cohort reported an inverse association between caffeine intake and PD. Three cohorts reported significant positive association in men between dairy products and PD.