For many decades, and until fairly recently, asbestos was commonly found in most sectors of South African industry. Consequently there is a large but indeterminate pool of formerly exposed workers, some of whom will present to medical practitioners for evaluation of possible asbestosis, the pneumoconiosis caused by the fibre. Fundamental to the diagnosis of asbestosis is a history of asbestos exposure sufficient to cause the disease. Attending practitioners need to be aware of the common asbestosis-inducing industries and jobs and the duration of exposure reported by patients if we are to obtain and interpret their exposures. This paper describes asbestos exposure in 141 cases of asbestosis. Cases were identified from patient records at the Occupational Medicine Clinic of the National Institute for Occupational Health (NIOH, formerly NCOH), for the years 1980-2000. Patients were only included in the series if they had no asbestos exposure in mining, if they had been certified with asbestosis by a compensation panel, and if on re-reading of the chest radiograph a radiologist reported irregular opacities (profusion 1/0 or greater on the International Labour Organisation (ILO) scale). Asbestos exposure was taken from the patient's records. Only one patient was exposed in an industry where asbestos was incidental to the enterprise's operation, while 54% of cases arose from exposure in primary asbestos industries, i.e. companies selling, distributing, refining, milling or using raw asbestos to manufacture products. The mean reported duration of exposure was 17.5 years. Surprisingly, 21 cases (15%) reported less than 5 years' exposure. Unexpectedly, 7 cases had a latency period from first exposure to diagnosis of less than 6 years. The data presented should assist practitioners in the purposeful exploration of asbestos exposure and in interpretation of its significance with regard to asbestosis.