Bacterial air sampling in an animal care laboratory showed that dense aerosols are generated during cage changing and cage cleaning. Reyniers and Andersen sampling showed that the airborne bacteria numbered 50 to 200 colony-forming units (CFU)/ft3 of air. Of the viable particles collected by Andersen samplers, 78.5% were larger than 5.5 ,m. A low velocity laminar air flow system composed of high-efficiency particulate air (HEPA) filters and a ceiling distribution system maintained the number of airborne viable particles at low levels, generally less than 2 CFU/ft3. Vertical air flow of 15 ft/min significantly reduced the rate of airborne infection by a strain of Proteus mirabilis. Other factors shown to influence airborne infection included type of cage utilized, the use of bedding, the distance between cages, and the number of animals per cage. Accidental airborne infection in an animal care facility can influence the validity of experimental results and the economics of laboratory animal care (6). It is not an uncommon practice to maintain laboratory animals under observation for their normal life span, especially in toxicity