From terrestrial physics let us take the case of sound propagated to a great distance. Whenever we are directly conscious of the causation of sound (namely, when we produce it ourselves), its invariable antecedent is force. The immediate sequence of this force we know to be motion—first, of our own organs, and then of the body which we set vibrating. The vibrations so generated we can discern both through the fingers and through the ears; and that the sensations received by the ears are the equivalents of mechanical force communicated to the air, and by it impressed on surrounding objects, we have clear proof when objects are fractured: as windows by the report of a cannon; or a glass vessel by a powerful voice. On what, then, rests the reasoning when, as occasionally happens under favourable circumstances, men on board a vessel a hundred miles from shore, hear the ringing of church-bells on placing their ears in the focus of the main sail; and when it is inferred that atmospheric undulations have traversed this immense distance? Manifestly, the assertion that the motion of the clapper, transformed into the vibrations of the bell, and communicated to the surrounding air, has propagated itself thus far on all sides, diminishing in intensity as the mass of air moved became greater, is based solely upon a certain change produced in consciousness through the ears. The listeners are not conscious of motion; they are conscious of an impression produced on them—an impression which implies a force as its necessary correlative. With force they begin, and with force they end: the intermediate motion being simply inferred. Again, where, as in celestial physics, the continuity of motion is quantitatively proved, the proof is not direct but inferential; and forces furnish the data for the inference. A particular planet can be identified only by its constant power to affect our visual organs in a special way—to impress upon the retina a group of forces standing in a particular correlation. Further, such planet has not been _seen_ to move by the astronomical observer; but its motion is _inferred_ from a comparison of its present position with the position it before occupied. If rigorously examined, this comparison proves to be a comparison between the different impressions produced on him by the different adjustments of the observing instruments. Going a step further back, it turns out that this difference is meaningless until shown to correspond with a certain calculated position which the planet must occupy, supposing that no motion has been lost. And if, finally, we examine the implied calculation, we find that it makes allowances for those accelerations and retardations which ellipticity of the orbit involves, as well as those variations of velocity caused by adjacent planets—we find, that is, that the motion is concluded to be indestructible not from the uniform velocity of the planet, but from the constant quantity of motion