The action of gravity, if propagated with finite velocity, should differ for bodies preserving an invariable distance from its source, and for bodies travelling towards or away from it. Their movement would modify the law of attraction.[62] Yet, up to the present, it has proved impossible to detect the slightest deviation from the plain rule of inverse squares. Again, the penetrative faculty of this strange force seems absolute and unlimited. We know by ordinary experience that we cannot diminish the weight of an object by interposing any kind of shield between it and the earth; and no refinement in experimentation avails to alter this result. That it is so, is a fortunate circumstance for the harmony of the world. We can dimly imagine the riot of confusion that would ensue if a transiting globe could intercept the attraction, as it does the light, of a central governing mass. But from the wonderfully adjusted universal order gravitational eclipses are excluded; nor does the densest body throw even the faintest gravitational shadow. The nature of a power so singularly conditioned is almost inconceivable; yet attempts have not been wanting to fathom the mystery that surrounds it. Professor Osborne Reynolds, in the Rede Lecture for 1902, claimed to have arrived at 'a complete, quantitative, purely mechanical explanation of the cause of gravitation,' based on the 'dilatancy' of a granular medium in close piling. But his working model of the universe will probably be remembered only as a lesson in the 'inversion of ideas,' showing that with skill and ingenuity a fairly concordant outcome of phenomena may be derived from antagonistic hypotheses. In this author's view matter is equivalent to a deficiency of mass, the spaces where his cosmic grains are relatively few, because their arrangement is out of gear, being driven towards one another by the pressure of the surrounding medium, in which they are compactly stowed, and therefore numerous. Thus, the acting forces in nature are made to depend upon the compression by the denser medium of interspatial tracts of rarer consistence, forming what we call matter. The theory is difficult, if not impossible of acceptance, not because it involves the overthrow of conceptions which may be rooted in habitual modes of thought, rather than in absolute truth, but because of its startling postulates and large vacuities. To be valid, it should be complete; and there are obvious chasms in the vast expanse of ground which it covers with surprising, though only partial success.