Most painstaking inquiry into the clinical history of patients suffering from clinical tuberculosis fails, in nearly all instances, to reveal with any degree of certainty the time when the infection occurred. Especially is this true in adults, for tuberculosis is not the disease that we have long thought it to be. The time of infection and the time when clinical symptoms manifest themselves is, as a rule, most probably separated by years. We must not expect an exposure to infection and an immediate manifestation of symptoms. It is just as important, in fact, more important, to know that the patient associated intimately with some one suffering from tuberculosis when a child, as to find an exposure immediately prior to the manifestation of symptoms. Some German authors,1,2,3 are considering tuberculosis as a disease of three stages, the same as syphilis; first, that of the infection of the regional lymph-nodes; second,