In developed countries, Mycobacterium bovis infection in cattle is now mostly confined to the respiratory system, which reflects transmission and establishment of infection mainly by this route. A single bacillus transported within a droplet nucleus is probably sufficient to establish infection within the bovine lung. Infected cattle should always be considered as potential sources of infection, since studies have demonstrated that a significant proportion of tuberculous cattle excrete M. bovis. In general, the dynamics of M. bovis transmission are poorly understood and the conditions under which a tuberculous animal becomes an effective disseminator of infection are currently not defined although environmental contamination appears to be a less effective method of disease transmission. Field studies indicate a wide spectrum of transmission rates but generally the spread of M. bovis infection is still considered to be a relatively slow process. Slaughter of diseased cattle detected by tuberculin testing and at meat plant inspection has been shown to be an effective policy for tuberculosis eradication, provided there are no other reservoirs of infection and all involved in the cattle industry are committed to a policy of eradication. Epidemiological approaches, particularly case-control studies, seem to provide the best method for quantifying the relative importance of the various sources of M. bovis transmission to cattle and modelling techniques can be used to assist in the design of cost-effective control measures that may lead to tuberculosis eradication.