In the course of routine tuberculin testing of preschool children in a clinic it became apparent that a number of positive reactions to intermediate-strength tuberculin could not be attributed to exposure to persons infected with active tuberculosis. Furthermore, it appeared that many of these unexplained tuberculin conversions represented low-grade hypersensitivity and showed a propensity to revert to negativity on retesting 6 to 12 months later.1Since most of these children have been recommended for prophylactic antituberculous therapy by a Committee of the American Academy of Pediatrics2based on the U.S. Public Health Service tuberculosis prophylaxis trial3findings, it became important to determine whether these reactions indicated the presence of tuberculosis. Edwards, Edwards, and Palmer4have noted that in some geographic areas many people had small reactions to intermediate-strength 5 TU (tuberculin units) tuberculin. They postulated that the unusual number of small reactors could be explained by