A case-control study on risk factors for cerebral tumors was conducted on an adult Italian population by the four Neurosurgical Departments of the Veneto Region, i.e. Padua, Treviso, Verona and Vicenza. The study recruited 195 cases of histologically-confirmed cerebral glioma. One hospital control was selected for each case. Cases and controls were matched for age, sex, data of hospitalization and residence. Information on both cases and controls was obtained from a relative. Uninvolved interviewers administered a structured questionnaire including items on the subject's education, occupation, lifestyle, medical history, exposure to radiation for diagnosis or therapy, head trauma and blood group and the medical history of family members. The series of cerebral tumors was first considered as an indistinct set: none of the risk factors examined showed a statistically significant association. A positive association was found with blood group A (OR = 6) when low-grade astrocytomas (n = 41) were considered separately. As for the malignant astrocytomas (n = 132), there was a suggestive but not statistically significant association with the presence of CNS tumors among first- and second-degree relatives (OR = 7.0). On the whole, this study yielded no clear and meaningful association for the various risk factors analyzed.