Reports in the literature on 21 cohorts of workers exposed to asbestos were reviewed and analyzed to determine whether there is a causal association between asbestos and colorectal cancer. The latest report up to 1988 was used for each cohort. The end point was the standardized morbidity or mortality ratio as the measure of relative risk. Two additional cohorts using mortality as the end point were excluded because the authors failed to use comparable diagnostic methods for the asbestos-exposed populations and controls. The summary standardized morbidity or mortality ratio for all 21 cohorts was 0.97 (P greater than 0.05), and there was no dose-response relationship in the two studies with such data. The evidence does not meet the established criteria for making a judgment that there is a causal relationship between asbestos and colorectal cancer.