Evidence for an involvement of aberrant homocysteine metabolism in the aetiology of schizophrenia is limited and controversial. A case-control study was performed to quantify the risk of schizophrenia in the presence of elevated homocysteine concentrations or homozygosity for the 677C → T polymorphism (677TT) in the methylenetetrahydrofolate reductase (MTHFR) gene in subjects of Dutch ancestry. We determined the 677C → T MTHFR genotype distribution in 254 well-defined patients and 414 healthy controls. Plasma homocysteine concentrations were measured in 62 patients with schizophrenia and 432 control subjects. When homocysteine concentrations were stratified into quartiles of the control distribution, we calculated an increased risk for schizophrenia in the fourth and third quartile versus the lowest quartile [odds ratio (OR) = 3.3; 95% confidence interval (CI): 1.2–9.2, and OR = 3.1; 95% CI: 1.2–8.0, respectively]. A significant dose-response relation of increasing homocysteine levels and increasing risk for schizophrenia was observed (P = 0.036). The 677TT genotype was associated with an OR of 1.6 [95% CI: 0.96–2.8] of having schizophrenia. Heterozygosity for the T allele compared to 677CC subjects accounted for an OR of 1.3 [95% CI: 0.91–1.8]. Elevated homocysteine levels and the MTHFR 677TT genotype are associated with an increased risk for schizophrenia. These observations support a causal relation between disturbed homocysteine metabolism and schizophrenia. © 2005 Wiley-Liss, Inc.