The objective of the study is to compare the use of several indices of exposure in describing the relationship between O3 and reduction in agricultural crop yield. No attempt has been made to determine which exposure-response models best fit the data sets examined. Hourly mean O3 concentration data, based on two-three measurements per hour, were used to develop indices of exposure from soybean and winter wheat experiments conducted in open-top chambers at the Boyce Thompson Institute, Ithaca, NY NCLAN field site. The comparative efficacy of cumulative indices (i.e. number of occurrences equal to or above specific hourly mean concentrations, sum of all hourly mean concentrations equal to or above a selected level, and the weighted sum of all hourly mean concentrations) and means calculated over an experimental period to describe the relationship between exposure to O3 and reductions in the yield of agricultural crops was evaluated. None of the exposure indices consistently provided a best fit with the Weibull and linear models tested. The selection of the model appears to be important in determining the indices that best describe the relationship between exposure and response.