Toxicity/carcinogenicity studies in rodents have played a pivotal role in identifying chemicals that are potentially hazardous to humans. In fact, nearly all of the known human carcinogens are also carcinogenic in 1 or more rodent species. During the past 20 yr the quality and consistency of rodent studies has improved considerably, and much has been learned about mechanisms whereby chemicals initiate or promote the carcinogenic process in rats and mice. The process of identifying chemicals that cause toxicity or carcinogenicity in rodents is quite well established, but the procedures for extrapolating this data for risk management decisions in the protection of human health have lagged far behind. While many would accept the assumptions that genotoxic chemicals that cause cancer in animals pose a cancer risk to humans and that genotoxic chemicals causing cancer at high doses pose a risk at lower doses, there is much less certainty with respect to nongenotoxic chemicals. The confusion about risk extrapolation for nongenotoxic chemicals has often lead to criticism of the hazard identification process for chemicals in general. There is increasing awareness of the complexity of the carcinogenic process that has made species extrapolation and dose extrapolation from rodent studies to humans more complex. Although newer molecular biological techniques and cell kinetic measurements offer exciting possibilities for better risk assessment, it is the combination of well-designed rodent studies with appropriate mechanistic studies that offers the best hope for regulatory decisions based on sound scientific principles.