From January 1988 to May 1989, one hospital in South Carolina reported 12 isolates of Tsukamurella paurometabolum from 10 patients. There were no common risk factors among the patients. Case-control studies revealed that the positive specimens were significantly more likely to have been processed in the TB/fungal room, to have been tissue samples, and to have been handled by one technician. Typing on the basis of biochemical, antimicrobial resistance, Southern blot, and ribotype profiles showed that the isolates from the outbreak were essentially identical and that they were distinguishable from each of two isolates obtained after the outbreak and from two type strains. These findings support the hypothesis of a common-source outbreak of pseudoinfection. There are reasons to believe that T. paurometabolum is present both in the environment and as a culture contaminant more often than has been recognized and that it is very rarely the true cause of infection in humans. Typing results show differences between one type strain and all of the other isolates studied in terms of colonial morphology, biochemistry, antimicrobial susceptibility, and ribotyping; these differences suggest that the nomenclature of T. paurometabolum may require further clarification.