A series of tests were made to determine the cause of some infections that develop in patients postoperatively. A number of Petri plates were exposed to people in clean operating rooms, these persons wearing a variety of types of face protection. One half of the plates was exposed to people talking and the other half was exposed to people not talking. A series of twenty Petri plates was exposed to a man with an upper respiratory infection, both talking and not talking. As a result of the colony count of organisms from these Petri plates, it was possible to determine that many more bacteria were spread by talking people than by non-talking people, although in the two groups wearing the greatest amount of face covering the colony counts were about the same for talking and non-talking. It was concluded that, at the very least, anyone in the operating room should wear a mask which covers the nose and that a person with an upper respiratory infection should not participate in an operation unless absolutely necessary, and then he should wear a double mask; preferably with a plastic sheet between the two masks. It was shown that the talking person with an upper respiratory infection produced as many colonies, or more, on ten Petri plates as were found on 25 plates exposed to people without upper respiratory infections, non-talking and with a variety of face coverings, including none. Staphylococcus albus, coagulase negative, and hemolytic Staphylococcus, coagulase negative, made up the greatest majority of the organisms found.