There is a need for more accurate, rapid and reliable means for diagnosis of tuberculosis. This paper describes the development and adaptqtion of the slide culture method for this purpose.

The adaptation of a modified slide culture technique to the routine diagnosis of pulmonary t~~berculosis is described in detail. Results of culture of 2248 specimens of sputum by this method and by conventional techniques are compared. The results demonstrate the superiority of the slide culture technique; 92.6% of positive-smear specimens yielded positive cultures, as against 70% by routine methods. Results were almost identical with both methods in the smear-negative group and an esplanation of this apparent discrepancy is advanced. 100% of the (positive) slide cultures were positive by the end of 14 days, whereas controls reached 96.5% only a t the end of eight weeks.Since Robert Icoch discovered the tubercle bacillus in 1882(4) a variety of media and methods have been introduced for the primary isolation of the organism from pathologic material. One of the problems has been to devise a procedure which would permit an earlier bacteriological diagnosis of the disease and which would be of greater assistance in clinical management during its course. I t is frustrating to the clinician to have to wait for three to eight weelis, or longer, for a bacteriological report on his patients, as is the case where the conventional cultural methods are used. On many occasions the first evidence of relapse is the reappearance of tubercle bacilli in the sputum, in too small numbers for microscopic detection, so that rapid cultural detection becomes of paramount importance. Implied in the over-all problem, of course, is not only speed but sensitivity to very small numbers of organisms as well. T h e ideal would be the cultural detection of one viable tubercle bacillus, from mixed culture, on a simple, selective directplating medium within 24 to 48 hr. This happy state has not yet been attained for no single method fulfills all our requirements. Indeed, it is fortunate that the tubercle bacillus is such a relatively hardy organism for otherwise, considering the punishment t o which we subject it, the number of positive cultures would be considerably less than a t present.In 1953 Reed (6) reported the results of experiments with a modified slide culture technique for the cultivation of tubercle bacilli from untreated sputum specimens. Early in the course of these experiments it became evident that the method held much promise and it was decided t o carry out full scale trials under actual diagnostic laboratory conditions. The present paper reports the results to date of these trials.