The minimal inhibitory concentrations (MICs) of 110 anaerobic bacteria (84% fresh clinical isolates) to nine antimicrobial agents were determined simultaneously by the tentative reference agar-dilution method of the National Committee for Clinical Laboratory Standards and a prototype commercial microbroth panel (Microscan). MICs (determined visually for either system) differed by greater than 1 log2 dilution in 17% of all tests. With the exceptions of the MICs of tetracycline and moxalactam, however, all compared antimicrobial MICs were within 1 log2 dilution for at least 82% of all tests. Twenty-three anaerobic bacterial strains (21%) failed to grow in the microdilution panel (Wilkins-Chalgren broth), and seven strains (6%) failed to grow on the reference Wilkins-Chalgren agar. However, for those bacterial strains that could be tested, the microbroth panel appeared to be less cumbersome than the agar-dilution procedure for susceptibility testing of anaerobic microorganisms.