A statistically significant association between cigarette smoking and radiologic readings of irregular opacities was observed in a case-control study of 430 South African asbestos miners (OR = 3.03; 95% confidence interval, 1.50; 6.11). In another case-control study of 487 dead South African asbestos miners, no positive association between smoking and autopsy findings for asbestosis was observed (OR = 0.50; 95% confidence interval, 0.20; 1.29). The association between greater than slight asbestosis (moderate and marked) and smoking (yes/no) was not significant (Fisher's exact test, p = 0.44), and the adjusted odds ratio was 1.25 with 95% confidence interval (0.20; 7.78). In addition, the certification in life for asbestosis, based on chest roentgenogram readings for irregular opacities, was also found to have a tendency to be related to smoking. Thus, the findings of these studies do not support the hypothesis that smoking is associated with the presence of asbestosis at autopsy or that smoking and asbestos dust act synergistically in producing asbestosis at autopsy. But the findings of this and other studies suggest that smoking does produce some change in the lung structure that can be visible on the roentgenogram and can be wrongly interpreted as irregular opacities caused by asbestosis.