Preliminary interpretive zone criteria were calculated for cefetamet (Ro 15-8074) and cefteram (Ro 19-5247) by using 10and 30-,ig disks and three possible MIC susceptibility breakpoints. Absolute interpretive agreement between MICs and zone size criteria ranged from 91.8 to 97.2%. Very major errors (false susceptibility) were s1.2% for both cephalosporin disk tests. Morganella morganii strains appeared to produce the highest rates of very major interpretive errors with cefetamet disks. Cefetamet (formerly Ro 15-8075) and cefteram (formerly Ro 19-5248 and T-2588) are orally administered cephalosporin pivoxyl esters which, as free acids, possess an antimicrobial activity most similar to that of céfixime (1, 4, 5, 9-11, 13). Each free acid (Ro 15-8074 and Ro 19-5247 or T-2525) has been linked to form the pivaloyloxymethylester that allows gastrointestinal tract absorption but produces a somewhat lower concentration in serum than do currently available cephalosporins (12; data on file at Hoffmann-La Roche, Inc., Nutley, N.J.). Even with these lower drug levels in blood, cefetamet and cefteram have sufficient potency to cover members of the family Enterobacteriaceae, Haemophilus influenzae, pathogenic neisseriae, Branhamella catarrhalis, and most streptococci (1, 4, 9-11, 13). In this brief report, we compare the antimicrobial activities of the free acids of these two new orally administered cephalosporins. We additionally present the preliminary determinations of disk diffusion test interpretive criteria with 10and 30-pLg cefetamet and cefteram disks. Cefetamet and cefteram were provided by Hoffmann-La Roche, Inc. (R. Cleeland). Both drugs were tested in cationsupplemented Mueller-Hinton broth by dilution methods described by the National Committee for Clinical Laboratory Standards (8). Additional medium supplementation for fastidious organisms was prepared according to the National Committee for Clinical Laboratory Standards M7-A recommendations (8). The disk diffusion tests were also performed by National Committee for Clinical Laboratory Standards procedures (7) by using 10-and 30-,ug disks prepared by the investigators to 110% of the stated potency. Regression analyses were calculated by the least-squares method as adapted to a computer by using the zone diameter as the independent variable. The error rates were calculated for three possible susceptibility MIC breakpoints to try to minimize the false-susceptibility error rate to s1.0% (6).