We tested the antibiotic susceptibilities of 100 Ureaplasma urealyticum strains from 99 patients using a broth-disk method and two types of inocula: urine sediments and overnight broth subcultures of the sediments. Of the 100 ureaplasma-positive urine sediments tested, nine (9%) of the ureaplasmas were found resistant to all four tetracyclines. When overnight broth cultures were used as the inoculum, 54 (54%) were found to be resistant to all four tetracyclines, an increase in resistance of 45 (45%). Thirty-seven susceptible strains remained susceptible upon subculture. The nine resistant strains remained resistant. Loss of susceptibility was not related to the pH or titer of ureaplasmas in the urine sediment inoculum but was related to the pH and titers when subcultures were used as the inoculum. Results of cultures following treatment, available for 53 patients, showed that treatment successes and treatment failures were significantly related to antibiotic susceptibility tests done with urine sediments but not to those done with broth subcultures as the inocula. Because reliable susceptibility testing is essential for appropriate therapy for U. urealyticum infections, all factors influencing this test need to be recognized and defined.