Selected paternal occupations as well as specific occupational exposures to chemicals such as organic solvents have been suggested as possible risk factors for neural tube defects (NTD). We investigated data from a population-based, case-control study of fetuses and liveborn infants with NTDs among 1989-91 California births and fetal deaths. Interviews were conducted with mothers of 538 NTD cases and 539 non- malformed controls. Mothers were asked to report the occupations that the fathers had in the period 3 months before and 3 months after conception. Each job title and industry reported was coded in accordance with the 1990 US Census. Considering those fathers who worked in managerial and professional occupations as the reference group, elevated odds ratios (OR) for NTDs were observed for the categories: technical, sales and administrative, OR = 1.5 [95% confidence interval 1.0, 2.4]; service, OR = 2.0 [1.2, 3.1]; farming, forestry and fishing, OR = 2.1 [1.3, 3.3]; operators, fabricators and labourers, OR = 1.8 [1.2, 2.7]; and military, OR = 1.9 [0.7, 5.0]. Stratification by NTD phenotype revealed that these elevated ORs were primarily observed for spina bifida phenotypes. Analyses adjusted for maternal body mass index, maternal periconceptional use of multivitamins containing folic acid, paternal race/ethnicity, maternal race/ethnicity and maternal education revealed attenuated risk estimates for most of the occupational groups. Analyses of 182 more specific occupational groups defined by aggregating fathers on similar job titles, when compared with fathers who worked in managerial and professional occupations, revealed that four job title groups were associated with fairly precise effect estimates of >or=1.5. These groups were: cooks; janitors and cleaners; farm workers; and groundsmen/gardeners. Using occupational titles as defined in previous investigations, we did not observe an elevated OR associated with paternal occupational solvent exposures, OR = 0.8 [0.5, 1.3]. These analyses generated potential clues regarding paternal occupational exposures as NTD risk factors. Risk variation observed by spina bifida phenotype is interesting and will need to be investigated further.