I. Tm~ CAPACITY 0]~ TUBERCLE BACILLI TO SURVIVE STERILIZATION IN MOUSE

An unusual form of microbial persistence displayed by tubercle bacilli of human origin was described in previous reports from this laboratory (3-5). Relatively large populations of bacilli (one to three million cells) could be made to "vanish" uniformly from the tissues of mice that had received two compounds; the nicotinamide derivative pyrazinamide, and isoniazid, in appropriate time-dose relationships. The criteria for "vanishing" were failure to detect any bacilli by; microscopy, culture, or guinea pig inoculation of the tissues of the infected mice. The "vanishing" did not mean that the bacilli were totally eliminated from the tissues. On the contrary, after a 90 day treatment-free interval, culturable tubercle bacilli could be demonstrated in approximately one-third of the animals. The question immediately arose whether all of the tubercle bacilli had been eradicated in the other two-thirds of the animals in each experiment, or whether the bacilli were still present in them but were in a iatent state. By latent is meant a phase during which the infection is hidden beyond all diagnostic reach. In such a phase, the bacilli conceivably could be present in some altered form or they could be unaltered, but so few in number t l~t they escape detection. In view of the demonstrated sensitivity of the culture techniques employed, it seemed clear that at the most only a very few ~naltered bacilli could be present and escape detection. I t was noted (4), however, that if a microbial alteration took the form of sterilization without death, fairly large populations, being nonculturable, might vanish from the tissues because of the insensitivity of microscopy as a detection technique. In the studies reported (3), acid-fast bacilli could not be found with any regularity on microscopy of tissue sections when the organ census was below I × 105 per ml and only