This review has focused on the importance of cellular proliferation in neoplasia induced by nongenotoxic carcinogens and has not attempted to address the mechanisms involved in the progression from the hyperplastic to the neoplastic state. Major advances in this area are likely in the next decade and should explain how cells in an abnormally high proliferative state are rendered more vulnerable either to the action of endogenous or environmental mutagens or to defects in cell reproduction, thus providing the stimulus for progression to neoplasia (165). In this review evidence has been presented to support the hypothesis that sustained tissue damage induced by nongenotoxic compounds in rodents predisposes to tumor development. The evidence has been obtained by drawing from the published findings of experimental rodent studies conducted over the past 50 years. The experience gained at various sites in the rodent including the connective tissue, liver, bladder, and forestomach shows the existence of a threshold dose for various test materials/agents, below which neither sustained tissue damage nor tumor induction occurs but above which level both effects are manifest. Taken collectively, an overall picture emerges that sustained cell proliferation renders various sites in the rodent vulnerable to tumor development. The validity of this hypothesis is of importance to the evaluation of carcinogenic risk of chemicals to humans. For those nongenotoxic carcinogens known to cause characteristically defined and sustainable tissue damage as a precursor of tumor development in rodents, it should be possible to establish a threshold dose below which both effects disappear and upon which a safety margin in humans can be based. Some limited evidence supports an association between chronic tissue injury and neoplastic development in humans. Note, however, that certain types of induced tissue damage may be rodent-specific and therefore have no relevance for humans. We conclude that the appearance of persistent tissue damage predisposes to tumor development in rodents exposed to nongenotoxic carcinogens and that systematic studies in rodents should provide a rational basis for arriving at a safety margin in humans exposed to such agents.