be reflected from the surface of the metal, not a ray is reflected from the surface of the glass that is in contact with it, consequently no interference can take place; which proves beyond a doubt that Newton’s rings result from the interference of the light reflected from both the surfaces apparently in contact (N. 199). Notwithstanding the successful adaptation of the undulatory system to phenomena, the dispersion of light for a long time offered a formidable objection to that theory, which has been removed by Professor Powell of Oxford. A sunbeam falling on a prism, instead of being refracted to a single point of white light, is separated into its component colours, which are dispersed or scattered unequally over a considerable space, of which the portion occupied by the red rays is the least, and that over which the violet rays are dispersed is the greatest. Thus the rays of the coloured spectrum, whose waves are of different lengths, have different degrees of refrangibility, and consequently move with different velocities, either in the medium which conveys the light from the sun, or in the refracting medium, or in both; whereas rays of all colours come from the sun to the earth with the same velocity. If, indeed, the velocities of the various rays were different in space, the aberration of the fixed stars, which is inversely as the velocity, would be different for different colours, and every star would appear as a spectrum whose length would be parallel to the direction of the earth’s motion, which is not found to agree with observation. Besides, there is no such difference in the velocities of the long and short waves of air in the analogous case of sound, since notes of the lowest and highest pitch are heard in the order in which they are struck. In fact, when the sunbeam passes from air into the prism, its velocity is diminished; and, as its refraction, and consequently its dispersion, depend solely upon the diminished velocity of the transmission of its waves, they ought to be the same for waves of all lengths, unless a connexion exists between the length of a wave and the velocity with which it is propagated. Now, this connexion between the length of a wave of any colour, and its velocity or refrangibility in a given medium, has been deduced by Professor Powell from M. Cauchy’s investigations of the properties of light on a peculiar modification of the undulatory hypothesis. Hence the refrangibility of the various coloured rays, computed from this relation for any given medium, when compared with their refrangibility in the same medium determined by actual observation, will show whether the dispersion of light comes under the laws of that theory. But, in order to accomplish this, it is clear that the length of the waves should be found independently of refraction, and a very beautiful discovery of M. Fraunhofer furnishes the means of doing so.