This analysis of the relationship between long-term exposure to carcinogens and the risk of cancer relies upon a model which depicts exposure, burden and genetic damage as time series. It is shown that the dose of the carcinogen and the resulting damage to susceptible tissues can be related to cumulative exposure provided that linear kinetics are maintained and the exposure series remains stationary. Since saturable processes can lead to nonlinear behavior, the role of metabolism is investigated. It is argued that by maintaining the mean burden below 1/8 of the value of KM the contribution of nonlinear kinetics to the dose should be minimal, at least in occupational settings where saturable metabolism is most likely. Under this conjecture an expression is derived for the maximum mean air concentration, designated Xmax, which should maintain linear kinetics. By comparing values of Xmax, estimated for five genotoxic and/or carcinogenic substances (benzene, styrene, tetrachloroethylene, trichloroethylene, and vinyl chloride) with the corresponding U.S. occupational exposure limits, it is shown that saturable metabolism is unlikely to occur in some situations but is likely in others. This suggests that biological monitoring can play an important role in defining dose-response relationships for some, but not all, carcinogenic substances.