Although a considerable proportion of cancer research is devoted to the identification of possible human carcinogens by the use of experimental systems, past and present experience has shown that once such evidence has been acquired the data are not necessarily used for the implementation of primary prevention of cancer. In fact, a recent survey of legislation related to protection against occupational carcinogens in 14 industrial countries has revealed a large number of inadequacies and inconsistencies, even with regard to chemicals that are indisputably carcinogenic in humans. The authors believe that, even in the absence of epidemiological data, a clearer distinction between those chemicals for which there is and those for which there is not adequate evidence of a possible carcinogenic effect in humans can be made by applying the revised criteria for evaluating the carcinogenicity of chemicals, and by using further carcinogenicity and related data on chemicals, together with new knowledge about the mechanism(s) of carcinogenesis as it becomes available. At present, there are no acceptable methods for measuring possible errors in an approximate quantitative evaluation of human risk for a given level of exposure to a chemical made on the basis of data that provide 'sufficient evidence' of carcinogenicity in experimental animals. The limited data available suggest, however, that such a relationship might exist, at least for certain classes of carcinogenic chemicals, provided that one takes into account the nature of the chemical concerned and the possible physiological, pharmacological, and toxicological differences between the test animals and humans. The experience of the past has also shown that for several environmental chemicals, experimental evidence of carcinogenicity preceded evidence in humans and would have predicted similar effects in humans. It is therefore hoped that the provision of adequate, critically analyzed experimental data will serve the purpose of better assisting national and international authorities in formulating decisions concerning preventive measures, at least with regard to occupational and iatrogenic exposures, in which intervention can be the most direct and effective.