A broad variety of bacterial strains, including 79 Pseudomonas aeruginosa , were studied in an in vitro evaluation of carbenicillin disk susceptibility testing. Regression analysis with both 50-μg and 100-μg carbenicillin disks was carried out. Organisms having minimal inhibitory concentration values of 100 to 200 μg/ml demonstrated zones of less than 11 mm with the 50-μg disk, resulting in very little opportunity for appropriate discrimination of results. The line of regression for the 50-μg disk intersected the ordinate at a point just above the minimal inhibitory concentration value considered to be the limit of intermediate susceptibility for Pseudomonas . These considerations, together with evidence of greater disk content variation in the 50-μg than in the 100-μg disks assayed, considerable manufacturer-to-manufacturer variability with the 50-μg disk, and the more appropriate performance of the 100-μg disk, lead us to conclude that the 100-μg disk better serves the clinical test requirements for this agent than does the 50-μg disk, which is currently the only disk available for laboratory testing.