A new form of interferometer

The Lummer and Gehrke interferometer consists of a thin plate of glass with its two surfaces worked optically true and paralled to one another. High accuracy is required in the plane parallelism, and the flatness of surface and the difficulty of working such long plates for high resolving powers, and the cost- involved, sets a limit to our achievement in this direction. Such plates, even when finished, furnish results which have given rise to a considerable amount of discussion and raise doubt as to whether the appearances are trustworthy at such high resolving powers and whether they are not partly due to ghosts formed by the defects in the glass. Hence the present attempt to devise a parallel plate that would meet these objections and serve as an instrument of resolving power and precision sufficient to extend our knowledge of the constitution of spectrum lines, and a variety of other phenomena requiring the aid of very high resolving power. The late Lord Rayleigh used the free surface of a sheet of water as a true optical flat for testing plane surfaces, and describes the results of observations of interference with large differences of path between two interfering beams in a thick slab of water floating on mercury. The optical flatness of the surface of a liquid at rest suggests that, by floating a layer of water or other transparent liquid over another, like mercury, with which it does not mix, we can get an optically perfect plane parallel plate, of any length and thickness, which may be used as a parallel-plate interferometer. The following is a brief account of the results of a series of experiments, conducted under great experimental difficulties during 1918-1919, to test the practicability of this idea.