The identification of tuberculosis vaccines and vaccination strategies, which in small animal models appear to be more effective than BCG, offer some exciting possibilities for control of human tuberculosis in the future. However, some major problems remain including selecting which vaccines should go into human trials and the length of time it will take for testing these vaccines in humans. The cattle model is well suited for the secondary screening of tuberculosis vaccines as there is a strong similarity between the disease in cattle and humans and outbred animals are used. Moreover, there are many similarities in the results from field trials of BCG in both cattle and humans, with BCG often failing to protect when the trials are extended over a number of years. In addition, calves, like human infants, are immunocompetent at birth. Recent studies in calves have shown that BCG vaccination of calves within hours of birth is highly effective in protecting animals against bovine tuberculosis, but BCG revaccination at 6 weeks of age is contraindicated. Prime boost vaccination strategies using BCG and DNA vaccines have provided evidence that these combinations may give better protection in calves than either vaccine alone. Based on antigens whose genes are absent from the BCG genome, advances have also been made to develop diagnostic reagents distinguishing infected and vaccinated animals (differential diagnosis). The cattle model has been particularly useful in prioritizing such antigens for testing in humans. Finally, there is an urgent need to identify an immunological correlate of protection against tuberculosis. The cattle model can be particularly helpful in this area as it is relatively easy to collect large volumes of blood from calves at intervals following vaccination and challenge, and a large number of immunological reagents are now available for cattle.