SummaryPresent methods for in vitro sensitivity testing of non-sporing anaerobic bacteria to antibiotics have been found to be unreliable and inconsistent. Studies have been carried out, therefore, to identify the variable factors involved. The results showed that in addition to the pH of the medium, variations in a number of other factors relating both to the methodology and to the reagents used could result in major changes in disc sensitivity testing and MIC determinations for erythromycin. Procedures for standardization of method and reagents are briefly discussed.