Investigation of the use of PEF in the measurement of airway variability over a single day or several weeks suggests that serial measures of PEF may prove useful in detecting short-term adverse responses to a variety of environmental agents. Several investigators have already shown how the visual inspection of temporal trends can prove useful in identifying individuals with occupational asthma. The use of statistical summaries of diurnal variation, i.e., the amplitude as a percent of the daily mean, may provide important added information about airway reactivity in response to occupational exposures in a variety of populations. Indeed, there are several groups of investigators presently examining more refined ways, including autoregressive time series models, of linking short-term temporal patterns in serial measurements of PEF with simultaneously measured short-term exposures. As our understanding of the sources and nature of the variability in PEF improves, together with refinements in the statistical techniques used to describe the temporal data, serial measures of PEF are likely to see a far broader application in the study of acute respiratory hazards. Does the utility of PEF as a pulmonary function measurement stem solely from ease of collection and the consequent increase in the amount of data one can gather, or is the PEF a distinctly different and inherently useful measure of pulmonary function? This question cannot yet be answered. The increasing availability of accurate reliable hand-held spirometers means that one can now plan studies using the powerful repeated measures designs previously feasible only with the peak flow meter (see article on Instrumentation, pp 397-408, in this issue). These data will enable side-by-side comparisons of PEF, FEV1, maximum mid-expiratory flow, and other parameters collected simultaneously. Will the FEV1 once again become the preferred measure of airflow? Time will tell.