Combined agar diffusion and replica plating techniques in the study of antibacterial substances.

Much interest has recently been aroused in the field of antibiotics by two distinct but related problems. The first concerns the possibility that in certain types of infection, notably subacute bacterial endocarditis, the bactericidal action of a drug is of greater therapeutic significance than the inhibitory effect (Hunter 1950; Robbins and Tompsett, 1951 ; Cates, Christie, and Garrod, 1951). The second problem is the action of antibiotics used in combination. In this respect the findings of many workers indicate that different mixtures may show augmented or diminished bactericidal effect, and the result may be influenced by the species or strain of organism tested (Jawetz, Gunnison, Bruff, and Coleman 1952; Bliss, Warth, and Long, 1952). Any investigation of these two problems requires a method applicable to routine use for the assessment of the bactericidal action of antibiotics. None of the conventional methods used for these purposes is suitable for large-scale application, as they depend on some form of viable counting and are very laborious. Even the more convenient turbidimetric technique described by Burnell and Kirby (1951) is too elaborate for general use. Moreover, a disadvantage common to all these methods is that an appreciable amount of drug is inevitably transferred with any sample to the new medium used to measure the bactericidal action. If the proportion of surviving bacteria is low it may be necessary either to limit the range of concentration of the drug under study, or to destroy it in some way which does not affect the bacteria. This is not always possible or convenient. A method of transfer which carries over little more than the amount of drug adhering to individual organisms is therefore needed. The method described here, which is an application of the technique of " replica plating" described by Lederberg and Lederberg (1952) and used by them for studying problems of bacterial genetics, is believed to fulfil these requirements. It involves no significant transfer of drug from one medium to another, is simple enough for routine use, and may be applied as a continuation of a form of " sensitivity test " in common use. It consists essentially of the transfer of bacteria from one nutrient agar surface to another without disturbing their spatial relationships. The transferring agent is the pile of a piece of sterile velvet covering a flat surface. The success of the technique depends on the fact that moisture is absorbed by the pile and no lateral smearing of the impression occurs.