A case-control study of cancer of the nose and paranasal sinuses was conducted in France to determine whether occupational exposure to formaldehyde was associated with an increased risk of sinonasal cancer. Exposures to 14 other substances or groups of substances were also studied (wood dust, leather dust, textile dust, flour dust, sugar dust, coal/coke dust, nickel compounds, chromium compounds, chromium VI, welding fumes, soldering fumes, cutting oils, paints and lacquers, glues and adhesives). Cases (n = 207) and controls (n = 409) were interviewed to obtain detailed information on job history and other potential risk factors for sinonasal cancer. In addition, a questionnaire specially designed for this study was used to help assess exposures to formaldehyde and other substances of interest. The questionnaires were translated into history of occupational exposure by an expert in industrial hygiene, without knowledge of case-control status. Several exposure variables (lifetime average level, duration, cumulative level) were used to describe the risk related to exposure to formaldehyde. Potential confounding factors (occupational and non-occupational) were examined and adjusted for when necessary. No significant association was found between exposure to formaldehyde and squamous-cell carcinomas of the sinonasal cavities. Because of the strong association between exposure to wood dust and nasal adenocarcinoma, it was not possible to assess an independent effect of formaldehyde on this type of cancer. However, among males exposed to medium or high levels of wood dust, the risk of adenocarcinoma associated with formaldehyde was significantly elevated for the highest exposure categories for average level (OR = 5.3, 95% confidence interval = 1.3-22.2), cumulative level (OR = 6.9, 95% CI = 1.7-28.2) and duration of exposure (OR = 6.9, 95% CI = 1.7-27.8). Although a residual confounding effect of wood dust could not be excluded, this study suggests that exposure to both formaldehyde and wood dust may increase the risk of nasal adenocarcinoma, by comparison with the risk due to wood dust alone. This study also indicated an increased risk among males who had been exposed to glues and adhesives, for all histologic types, which was not explained by a confounding effect of paints and lacquers, wood dust or formaldehyde. No other significant association was observed.