The possibility of the airborne spread of foot-and-mouth disease during the 2001 epidemic in the UK has been investigated in three epidemiological case studies. On the basis of evidence from field investigations, and a simple meteorological analysis, it is concluded that the spread of disease was consistent with the airborne transport of virus. The distances ranged from less than 1 km to 16 km; six of the farms were over 6 km from the source and involved the passage of virus over the sea combined with meteorological conditions which strongly favoured airborne disease transmission. The results of detailed atmospheric modelling demonstrated that airborne virus could have challenged livestock on all the farms studied. However, with one exception the 24-hour average daily concentrations of the virus were significantly below the experimentally estimated threshold for infection. A detailed model intercomparison established that, under stable atmospheric conditions, peak concentrations of virus up to two orders of magnitude higher might have been experienced for short periods, owing to fluctuations within the plume of virus, and model limitations. This finding would significantly reduce the apparent discrepancy between the experimentally estimated threshold for infection and the modelling results.