Dr. Stoney, like Mr. Crookes, considers that the chemical elements are subject to decay. That they are not only generated but destroyed—that they are subject not only to evolution but dissolution. He believes that the generative process probably takes place only at, or beyond, the confines of the universe, and the destructive process at the centres of overgrown stars, which is the position of lowest potential. Dr. Stoney thinks that this extinction of the chemical elements in the centre of a star is a cause which limits its size and prevents its overgrowth. THE IMPACT THEORY IN RELATION TO THE FOREGOING THEORIES OF THE PRE-NEBULAR CONDITION OF MATTER. In all these theories, as has already been observed, the primitive condition of the universe was that of matter in a state of extreme tenuity, while by aggregation the materials became successively larger and larger until they assumed the magnitude of suns and planets. For example, according to the meteoric theory, meteorites are formed out of “cosmical dust,” “fire-mist,” or condensed vapour, and then suns and planets are formed by aggregation from these meteorites. Facts seem, however, to point to the very reverse as being the true course of events. Meteorites are undoubtedly the fragments of larger masses. It looks more likely that they are, as has already been stated, fragments of stellar masses which have been shattered to pieces by collision, and that this “cosmical dust,” from which the meteorites are alleged to have been formed, are simply the dust arising out of the destruction of the masses. After the two bodies had collided and been shattered to pieces, some of the fragments would undoubtedly be projected with a velocity that would carry them beyond the attractive power of the general mass, and thus they would escape being volatilised. These fragments would continue their wanderings through space as meteorites. I cannot but think that the number, as well as the importance, of these wanderers has been greatly over-estimated. Mr. Lockyer states that Dr. Schmidt, of Athens, found that the mean hourly number of luminous meteors visible on a clear moonless night by one observer was fourteen. Certainly no such quantity is visible in this country. In Scotland, at least, one may often watch night after night under the most favourable conditions without having the good fortune to see a single meteor.