The purpose of this survey was to evaluate the effects of smoking and occupational exposures on the decline in forced expiratory volume in one second (FEV1), and the presence of airflow limitation (FEV1 x100/forced vital capacity (FVC) being < 65) at follow-up. A random sample of 1,933 men aged 22-54 years in Bergen, Norway, were invited into the survey. Smoking habits and measurements of FEV1 were recorded at the initial survey in 1965-1970 (n = 1,591) and at follow-up in 1988-1990 (n = 951). Past or present self-reported occupational exposures to eleven airborne agents (dusts, gases, vapours and fumes) and measurements of FVC were recorded at follow-up only. The decline in FEV1 was associated (p < 0.001) with age, body height and smoking. Smoking cessation reduced the decline to the level of lifetime nonsmokers. Accelerated decline in FEV1 was observed in subjects exposed to sulphur dioxide gas and to metal fumes. The adjusted decline in FEV1 increased progressively in subjects exposed to increasing numbers of occupational agents (test for trend: p < 0.01). Airflow limitation was observed in 9.5% at follow-up, and increased with age and cigarette consumption. In this community follow-up survey in men, smoking and occupational exposures to sulphur dioxide gas, metal fumes and the numbers of specific agents were found to be important predictors for accelerated decline in FEV1.