A spatial stochastic simulation model was developed to describe the dynamics of bovine tuberculosis in badger populations in southwest England, based on data from the literature and from unpublished sources. As there are no data on intra- and intergroup infection probabilities, estimates of these were obtained through repeated simulations based on field observations of the spread and prevalence of the disease.The model works on a grid-cell basis, with each grid cell potentially occupied by one badger social group; immigration to and emigration from the main grid are incorporated. Population regulation is assumed to occur at the group level through density-dependent fecundity and cub mortality, and the model can be run for various disease-free equilibrium group sizes (which are determined by the carrying capacity of the environment). The model works on a quarterly (three-monthly) basis and processes are stochastic at the individual level. Three classes of individual (adults, yearlings and cubs) and three classes of infection (susceptible, infected-but-not-infectious and infectious) are recognized. Bovine tuberculosis was shown to persist in badger populations for long periods of time, even in populations with a disease-free equilibrium group size of only four adults and yearlings. However, with standard rates of intergroup infection and movement, the disease only became endemic in populations with a disease-free equilibrium group size greater than six adults and yearlings. In the endemic situation, the prevalence of the disease ranged between 11-22% depending on the combination of inter- and intragroup infection probabilities used. Endemic infection within the homogeneous environment of the grid was characterized by a high degree of heterogeneity. Patches of infection were spatio-temporally unstable, but shifted in location relatively slowly.Spread of the disease from a point source of infection with standard rates of intergroup movement and infection only occurred to any marked extent in populations with disease-free equilibrium group sizes of eight or more adults and yearlings. Increasing the intergroup infection probability had a significant effect on increasing the probability and rate of spread, and considerably lowered the threshold group size for spread from a point source to around four adults and yearlings. However, increasing the rates of intergroup movement reduced the probability of spread of the disease except at the largest group sizes. When both intergroup infection and movements were increased, the effects of increased infection in enhancing spread were offset to some degree by the increased movements. Perturbation to the badger population, as may be caused by control operations, could therefore increase the probability of persistence or spread of an infection.