sky by affirming that they have originally been fragments torn from the earth. But that such is indeed the case I shall endeavour to demonstrate. I have no doubt to some extent discussed this matter in previous writings. I have found, however, that the force of the argument is often imperfectly understood, and as the evidence that can be adduced seems to be strengthening in proportion as our knowledge of these bodies accumulates, I have thought it desirable to take this opportunity of treating the question fully with the help of additional information which recent investigations have placed at our disposal.
The notion that terrestrial volcanoes should have been the agent by which meteorites have been driven off into space seems at first to be refuted by the circumstance that we have no volcanoes which possess at present anything like the intensity of explosive energy which would be required. There is, I believe, sufficient testimony to prove that certain bombs projected from Krakatoa in 1883 fell to earth at a distance of fifty miles from their source. It is easy to demonstrate that a body which performed a flight so tremendous must have been expelled from the crater with an initial velocity considerably in excess of any speed that has been attained by a missile in artillery practice. This statement offers a remarkable illustration of volcanic energy. To appreciate it fully we should reflect on the dimensions of a volcanic crater, and on the free opportunity it affords for the escape of the explosive gases, in contrast with the conditions found in the discharge of a missile from a cannon, where all the force of the explosion can be concentrated on the projectile.
It must, however, be admitted that there is no observation, so far as I know, of any volcanic explosion in which
