The antimicrobial susceptibility profiles of a total of 318 aerobic and facultatively anaerobic bacteria (255 gram-negative bacilli and 63 gram-positive cocci) were determined, using a new commercially available breakpoint broth microdilution procedure (Sensititer Breakpoint System (SBS), Gibco Diagnostics, Inc., Madison, WI) that categorizes test results in the form of susceptibility categories: susceptible, intermediate, and resistant. Results obtained with the SBS were compared with those achieved with a standardized disk diffusion procedure. Among a total of 4,414 organism-antimicrobic comparisons, concordance between the results of the SBS and the disk diffusion procedure was observed in 3,888 cases (88.1%). Four hundred twenty-three (9.6%) minor discrepancies, 45 (1.0%) major discrepancies, and 58 (1.3%) very major discrepancies were noted. Arbitration of major and very major discrepancies with a full-range minimum inhibitory concentration (MIC) procedure confirmed the results of the SBS in 53.4% of cases. A single organism-antimicrobial combination, the nonenterococcal streptococci tested against the aminoglycosides, yielded a significant number of very major errors which were arbitrated in favor of the disk diffusion result. These errors were probably due to poor growth of the test organism in the broth medium used for performing the SBS test (i.e., cation-supplemented Mueller-Hinton broth). With this exception, the SBS was found to be at least as accurate as the standardized disk diffusion procedure.