Experiment alone has proved, and had the right to prove, that it is impossible to measure the velocity of an object relatively to the ether. At all events, this is now settled. After all, since it is evidently in the very nature of things that we cannot detect an absolute movement, is it not because the velocity of the earth in relation to the ether is an absolute velocity that we have been unable to detect it? Possibly; but it cannot be proved. If it is so—which is not at all certain—it is in the last resort _experience_, the one source of truth, which thus tends to prove, indirectly, that the ether is really identical with space. In that case, however, a space devoid of ether, or one containing spheres of ether, would no longer be conceivable, and there can be nothing but a single mass of ether with stars floating in it. In a word, the negative result of Michelson’s experiment could not be deduced _a priori_ from the problematical identity of absolute space and the ether; but this negative result does not justify us in denying the identity _a posteriori_. Let us return to our proper subject, the Fitzgerald-Lorentz hypothesis which explains the issue of the Michelson experiment, and which was in a sense the spring-board for Einstein’s leap. The hypothesis is as follows. The result of the experiment is that, whereas when the path of a ray of light between two mirrors is transverse to the earth’s motion through ether, and it is then made parallel to the earth’s motion, the path ought to be longer, we actually find no such lengthening. According to Fitzgerald and Lorentz, _this is because the two mirrors approached each other in the second part of the experiment_. To put it differently, _the frame in which the mirrors were fixed contracted in the direction of the earth’s motion, and the contraction was such in magnitude as to compensate exactly for the lengthening of the path of the ray of light which we ought to have detected_. When we repeat the experiment with all kinds of different apparatus, we find that the result is always the same (no displacement of the fringes). It follows that the character of the material of which the instrument is made—metal, glass, stone, wood, etc.—has nothing to do with the result. Therefore all bodies undergo an equal and similar contraction in the direction of their velocity relatively to the ether. This contraction is such that it exactly compensates for the lengthening of the path of the rays of light between two points of the apparatus. In other words, the contraction is greater in proportion as the velocity of bodies relatively to the ether becomes greater. That is the explanation proposed by Fitzgerald. At first it seemed to be very strange and arbitrary, yet there was, apparently, no other way of explaining the result of Michelson’s experiment.