It was through this property that Michelson attempted to measure the earth's velocity through the ether. The apparatus was simple in principle. A circular table ABCD (see fig. 1) was ar ranged so as to be capable of slow rotation about its centre 0. Light sent along CO was divided up at 0 into two beams which were made to travel along perpendicular radii OA, OB. The arms OA, OB were made as equal as possible and mirrors were placed at A and B to reflect the beams of light back to 0. An extremely sensitive optical method made it possible to detect even a very slight difference in the times of the total paths of the two beams from 0 back to 0. There would in any case be a difference owing to the necessarily imperfect equalization of the lengths of the arms OA, OB, but if the earth is moving through the ether in some di rection OP, and if the table is made to rotate slowly about 0, then this difference ought itself to vary on account of the earth's mo tion through the ether. Michelson, and afterwards Michelson and Morley in collaboration, attempted to estimate the amount of this variation. No variation whatsoever could be detected, although their final apparatus was so sensitive that the variation produced by a velocity through the ether of even 'km. a second ought to have shown itself quite clearly.
Thus to the question "What is our velocity through the ether?" nature appeared to give the answer "None." It was never sug gested that this answer should be accepted as final ; this would have brought us back to a geocentric universe. Clearly either the question had been wrongly framed or the answer wrongly inter preted. It was pointed out in 1893 by Fitzgerald, and again in dependently, in 1895, by Lorentz, that the null result of the Michelson-Morley experiment could be explained if it could be supposed that motion through the ether altered the linear dimen sions of bodies. They both showed that the experiment would in variably and of necessity give a null result if every body moving through the ether with a velocity u was contracted in the direction of its motion in the ratio being the velocity of light.
The supposition that such a contraction occurred was not only permissible—it was almost demanded by electrical theory. Indeed, Lorentz had already shown that if matter were a purely electrical structure, the constituent parts would of necessity readjust their relative positions when set in motion through the ether and the final position of equilibrium would be one showing precisely the contraction just mentioned.
On this view, there was no prima facie necessity to abandon the attempt to measure the earth's velocity through the ether. The answer to the problem had merely been pushed one stage farther back, and it now became necessary only to measure the shrinkage of matter produced by motion. It was obvious from the first that no direct material measurement could disclose the amount of this shrinkage, since any measuring rod would shrink in exactly the same ratio as the length to be measured; but optical and electrical methods appeared to be available. Experiments to this end were
devised and performed by Rayleigh, Brace, Trouton and Noble, Trouton and Rankine and others. In every case a null result was obtained. It appeared then that if the earth moved through the ether this motion was concealed by a universal shrinkage of mat ter, and this shrinkage was in turn concealed by some other agency or agencies.
At this time the word "conspiracy" found its way into the tech nical language of science. There was supposed to be a conspiracy on the part of the various agencies of nature to prevent man from measuring his velocity of motion in space. If this motion produced a direct effect x on any phenomenon, the other agencies of nature seemed to be in league to produce a countervailing effect —x. A long train of experiments had not revealed, as was intended, our velocity through the ether; they had merely created a conviction that it was beyond the power of man to measure this velocity. The conspiracy, if such there was, appeared to have been perfectly organized.
A perfectly organized conspiracy of this kind differs only in name from a law of nature. The inventor who tries to devise a perpetual-motion machine may come to the conclusion that the forces of nature have joined in a conspiracy to prevent his ma chine from working, but wider knowledge shows that he is in con flict not with a conspiracy, but with a law of nature—the conser vation of energy. In 1905 Einstein, crystallizing an idea which must have been vaguely present in many minds, propounded the hypothesis that the apparent conspiracy might be in effect a law of nature. He suggested that there might be a true law to the effect that "it is of necessity impossible to determine absolute mo tion by any experiment whatever." This hypothetical law may again be put in the equivalent form : "The phenomena of nature will be the same to two observers who move with any uniform velocity whatever relative to one another." This may be called the hypothesis of relativity.
The hypothesis in itself was not of a sensational character. In deed, from the quotations which have already been given from Newton's works, it appears probable that Newton himself would have accepted the hypothesis without hesitation : he might even have regarded it as superfluous. The true significance of the hy pothesis can only be understood by a reference to the scientific history of the two centuries which had elapsed since Newton. The Newtonian view that absolute rest was to be found only "in the remote regions of the fixed stars, or perhaps far beyond them," had given place to a belief that absolute rest was to be found all around us in an ether which permeated all bodies. What was strik ing about the hypothesis was its implication—either that we could not measure the velocity relative to ourselves of a medium which surrounded us on all sides, or else that no such medium existed.