Present models for predicting human pulmonary toxicity of ozone from the toxic effects observed in animals rely on dosimetric measurements of ozone mass balance, and species comparisons of tissue protective mechanisms against ozone. The goal of the present study was to identify a method to directly compare ozone dose and effect in animals and humans using bronchoalveolar lavage (BAL) fluid markers. Feasibility of estimating ozone dose to lung alveoli was demonstrated through assay of reaction products of oxygen-18 labeled ozone in lung surfactant and macrophage pellets of rabbits. Feasibility of using BAL protein measurements to quantify the ozone toxic response in humans was demonstrated by the finding of significantly increased BAL protein in 10 subjects exposed to 0.4 ppm ozone for 2 hr with intermittant periods of heavy exercise.