In the last few years, there has been considerable concern over the effects of long term exposure to low levels of toxic materials-especially chemical agents. Controlled data for such exposure paradigms are extremely limited; most studies have been for relatively short exposures to high levels. Where long-term data exist, chemical agents were often administered by routes inappropriate to the exposure scenario for which allowable levels are required, and there was no long-term follow-up after termination of exposure. Although it is known that some agents produce long-term. effects and others may, it is often not clear that long-term effects are produced in the absence of a symptomatic exposure. Extrapolation of such data to "safe" exposure levels is not advocated, but it is recognized that, too often, interim exposure limits must be derived from existing data. In doing the risk assessment for estimating allowable exposure levels (AELs), it is important to fully characterize the exposure scenario, and to understand the concentration-time relationship for the chemical(s) in question. Although the fundamental process of establishing exposure limits is independent of the type of available data, the quality of the data affects the confidence in the exposure limits so derived. Risk Management should be the final step in the process of implementing AELs and should be independent of the Risk Assessment.