For many years attempts to control nosocomial infection consisted of programs to measure and then reduce the number of microorganisms present in the hospital environment. Recently, however, investigation of documented episodes of infection in hospital patients or personnel has been emphasized, and the role of routine microbiologic monitoring has been minimized. Further improvements in infection control will require a better understanding of the relative importance of objects in the animate and inanimate environment as reservoirs and vectors for infection rather than development of newer and better ways to culture the environment.