While one can not assert that the conclusions which experiments appear to offer must of necessity correspond to the conditions which arise clinically in man, nevertheless these results, if definite, must tend to show the direction in which the truth lies. The conclusions which may be drawn from this research are as follows: 1. Direct infection of the medulla of a long bone is unlikely to lead to the development of a tuberculous osteomyelitis. 2. Inoculation of the interior of a joint with tubercle bacilli readily produces tubercle of the synovial membrane. 3. From such an infected joint the epiphysis or metaphysis of the bone becomes diseased. 4. Infection of the arterial heart blood does not result in the local development of tubercle of the bones or joints. 5. Infection of the main artery supplying a limb leads to the development of tuberculous disease of certain of the joints of that limb. 6. Direct infection of the nutrient artery does not result in tuberculous osteomyelitis of the bone.