From 1931 to 1977 a total of 144 cases of silicosis have been diagnosed in the iron-ore mine in Kiruna, Sweden. Drilling, loading and tapping caused all cases of silicosis. In 24% of the cases, the disease had progressed after the diagnosis. The progression of the disease after diagnosis showed no significant correlation to the length of the exposure but a dose-response relationship was present between the cumulative quartz exposure and the stage of silicosis 20 years after the diagnosis. Signs of tuberculous infection were found in 17%. There was a significant relationship between tuberculosis and progression of the silicosis. Mortality was increased in association with silicosis stage II-III and in stages with concomitant tuberculosis infection. Half of the cases of silicosis had been diagnosed before 1942, and after 1960 no more cases in stages II or III were diagnosed. The mean concentration of respiratory quartz was approximately 0.8 mg/m3 in the 1950s and early 1960s and decreased progressively to below 0.05 mg/m3 in the late 1970s. The cumulative incidence rate, with respect to the decade in which the exposure began, was 0.021% in 1951-1960. With an unaltered dose-response, less than one case of silicosis per 500 workers may be expected among those who start working underground in the 1980s. Evidently silicosis is no longer a major health risk among those who start working underground in this mine today, and it is recommended that the 40-year-old regulations for the medical prevention of silicosis be revised.