Recent studies of the association between lung cancer and silicosis and silica dust have been inconclusive; some showing positive association and some showing none. The present study matched 231 cases of lung cancer with 318 controls by year of birth. Subjects were selected from the necropsy records of the National Centre for Occupational Health. Data on intensity and duration of exposure to silica dust were obtained from personnel records. Presence or absence of lung cancer and the presence and severity of silicosis of the parenchyma, pleura, and hilar glands were documented from necropsy reports. Smoking data were abstracted from records of routine examinations. No case-control differences were noted for any of the exposure indicators including cumulative dust exposure, total dusty shifts, weighted average intensity of exposure, total underground shifts, and shifts in high dust. Similarly, no association was found between lung cancer and the presence or severity of silicosis and any site. Stratified analyses showed neither significant nor suggestive trends when case-control comparisons for silicosis were examined by level of dust exposure or smoking. Reasons for disparity between these results and those of some other studies may include concomitant exposures to radon daughters, asbestos, diesel emissions, and cigarette smoking; idiosyncracies of the compensation process; and the possibility of a threshold in the relation(s).