The problem of drug resistance has become an important limiting factor in the therapeutic efficiency of streptomycin (Buggs et al., 1946; Finland et al., 1946; Bondi et al., 1946). We have previously shown in the case of streptomycin that of 13 strains tested all had the ability to throw off, spontaneously, variants resistant to streptomycin (Klein and Kimmelman, 1946; Klein, 1947). The destruction by streptomycin of the mass of susceptible bacteria and the multiplication of the fe.w highly resistant variants was indicated to be a mechanism for the development of streptomycin resistance. Alexander and Leidy (1947), working with Hemophilus influertae, have recently obtained similar results. Clinically, the inhibition of the rapid development of streptomycin resistance may then require the destruction of a relatively small number of resistant bacteria, which might be effected by the addition of a low concentration of another drug. In the present work we have therefore studied the combined action of streptomycin, penicillin, and sulfadiazine in vitro and determined the relationship between the synergistic action of the compounds and the inhibition of the development of streptomycin resistance. MATERIALS AND METHODS Staphylococcus aureus, susceptible to streptomycin, penicillin, and sulfadiazine, was used as the test organism. A casein hydrolyzate medium (Strauss, Dingle, and Finland, 1941) containing 0.5 per cent glucose provided a clear mediumwhich was convenient in the determination of growth rates turbidimetrically in the Klett-Summerson photoelectric colorimeter. The presence of the glucose resulted in a drop in pH after 24 hours that did reduce the streptomycin activity (Geiger, Green, and Waksman, 1946). However, this did not interfere with the interpretation of the results on the combined drug action. The tests for drug activity were performed as follows: Six ml of the casein hydrolyzate medium, containing the various drugs singly or in combination, were added to the Klett-Summerson tubes, and a standaxd inoculum of 0.1 ml of a 20-to 24-hour culture, diluted to give a reading of 50 on the Klett-Summmerson colorimeter (approximately 15,000,000 bacteria), was seeded into each of the tubes. This large inoculum provided a rapid initial growth, which permitted the taking of turbidity readings a.t 6 hours, in addition to the 12-, 24-, and 48hour readings. In preliminary assays it was found that the 24-hour growth This investigation has been aided by a grant from the Josiah Macy, Jr., Foundation.
