Questions about the adequacy of the existing ozone (O3) standard prompted an examination of relationships between concentration (C) and exposure time (T) and the impact of changes in the C x T product on toxic responses. Using protein concentration of bronchoalveolar lavage fluid (BALP) as an index of O3-induced lung damage, models were developed from a matrix of C (0.0, 0.1, 0.2, 0.4, and 0.8 ppm) and T (2, 4, and 8 h) values in rat and guinea pig. Equal C x T products with different levels of C and T were incorporated into the protocol. Polynomial and exponential least-squares models were developed and the lognormal linear model (Larsen et al., 1991) was evaluated for the rat and guinea pig data. For equal C x T products the results showed similar BALP responses at low C x T products. Calculations from the data and the models showed that (1) the models were consistent with reported experiments from the author's laboratory (Hatch et al., 1986), (2) exercising humans were more responsive to O3 exposure (without adjustments for ventilation rates) than were either rats or guinea pigs as measured by changes in BALP (Koren et al., 1989), and (3) the exponential model provided more generality than Haber's law by providing estimates of BALP levels for various C x T. (Copyright (c) 1992 by Hemisphere Publishing Corporation.)