INVESTIGATORS have recently emphasized the very obvious fact that ultraviolet stimulation of the skin or tissues of a patient, or other room occupant, occurs only when the light rays actually reach the individual; and that, as might be expected, such stimulation varies directly with the amount of light received, depending upon the individual's nearness to the window and whether the window contains ordinary window glass or glass which is to ultraviolet light. The same limitations exist with regard to bacterial destruction by light, of course; but since air circulates throughout a room, any agent which kills the bacteria in any part of it must affect the bacterial content of the room as a whole. From this point of view, there would, therefore, be a real advantage in replacing ordinary window glass with glass which is more to ultraviolet rays; and two different types of experiments were undertaken to determine the relative bacterial destruction which may be obtained with (1) unobstructed or open sunlight, (2) sunlight through plate glass, and (3) sunlight through one of the more permeable glasses now on the market.' In both types of experiments we used cubical wooden boxes about twenty inches in diameter. Each box had a removable top, which could be taken out to allow the sun's rays to enter the box unimpeded, or which could be replaced by plate glass or by the more glass. The boxes