The relationships between the inhaled dose of foot and mouth disease virus and the outcomes of infection and disease were examined by fitting dose-response models to experimental data. The parameters for both the exponential and beta-poisson models were estimated using maximum likelihood and Bayesian methods. The median probability of infection given a single inhaled TCID50 was estimated to be 0·031 with 95% Bayesian credibility intervals (CI) of 0·018–0·052 for cattle, and 0·045 (CI = 0·024–0·080) for sheep. These estimates were used to construct dose-response curves and uncertainty distributions for use in quantitative risk assessments.