The eye is sensitive only to light of wave-length 7600 to 4000 A.U. With the ordinary dry photographic plate and glass lenses we can get an effect down to wave-length 3300 A.U.; with the same plate and quartz lenses we can get an effect from a wave-length as short as 2000 A.U. The gelatine of the plate absorbs the wave-lengths immediately above this limit, and their effect is very faint. A camera fitted with a quartz lens can take a picture in which all the wave-lengths down to 2000 A.U. produce their share, and hence can extend the range of the eye another octave, but the disadvantage of this picture is that it integrates all the different colours. We cannot, for example, tell whether a mark is due to light of wave-length 3500 A.U. or 2500 A.U. Such a picture makes the ultraviolet appear the same as the visible would appear to a man with monochromatic vision. All the detail due to variety and wealth of colour is lost. Now if we could photograph the same objects in succession with monochromatic light of wave-length say 3500 A.U., 3000 A.U., 2500 A.U., and 2000 A.U., it is possible that a great amount of new detail might be obtained of the utmost value to science. It was with the purpose of obtaining monochromatic photographs in the ultraviolet that the present research was undertaken.