A 4-h bioluminescence method for methicillin susceptibility determination was compared with reference methods. Of the Staphylococcus aureus strains tested, 80 were methicillin resistant, 180 were methicillin susceptible, and 10 were borderline susceptible. There was 100% correlation between bioluminescence and reference methods for methicillin-susceptible and methicillin-resistant strains. All borderline-susceptible strains were identified as methicillin resistant by bioluminescence. Rapid, accurate detection of methicillin-resistant Staphylococcus aureus (MRSA) is important therapeutically, epidemiologically, and economically for health care institutions. Current methods include the use of screen plates containing 4% NaCi and 6 ,ul of oxacillin per ml (9), broth dilution using 2% NaCI in cation-supplemented Mueller-Hinton broth (7), and standard disk diffusion (6). Automated, commercial MIC methods are also being used, but their accuracy is questionable (1, 3, 4, 8). In addition, most of these methods require 18 to 24 h of incubation. This study evaluated a 4-h bioluminescence bacterial ATP assay for the detection of MRSA and methicillin-susceptible S. aureus (MSSA). Of 270 S. aureus isolates included in this study, 180 were MSSA, 80 were MRSA, and 10 were borderline-susceptible S. aureus (BSSA) (5). The MRSA strains were provided by the Centers for Disease Control