In Spectro-Photometry one of the chief difficulties in making accurate measurements lies in the fact that the eye has to judge of the relative intensities of two spectral bands, which, instead of having their adjacent edges in contact, have them separated by a greater or less dark space. Accordingly, to facilitate a piece of spectrophotometrical work in which I am engaged, a Hüfner's rhomb was tried, which is a special attachment designed to overcome this difficulty. On trial, however, I found that by this means the fault is not entirely removed, although the separating dark space is by this means reduced to a very narrow line. The only alternatives to the Hiifner's rhomb which I know are the devices of Brace, and of Lummer and Brodhun. The first of these depends on the employment of a particular kind of prism in the spectroscope. The other device, due to Lummer and Brodhun, places before the ordinary spectroscope prism a special photometer cube. Both these devices, however, suffer from the drawback that they are unsymmetrical as regards the two beams of light, whose relative intensities it is our object to measure.