138 ANNUAL OF SCIENTIFIC DISCOVERY. 



ble to execute the experiment for want of the proper instruments. 

 That difficulty has been at last vanquished, and the result is the com- 

 plete defeat of the Newtonian corpuscular theory. The experiment 

 was made by M. Foucault, by means of an ingenious instrument found- 

 ed upon the revolving mirror of Wheatstone, and is thus described by 

 Arago : " Let us cause to start from two points near each other, in 

 the same vertical line, two rays of light parallel to the horizon, and 

 let their direction conduct them to two points on the median line of 

 a mirror turning about this line. The direction in which the two 

 ravs will be reflected will depend upon the moment at which they 

 reach the turning mirror. If they arrive simultaneously, they will be 

 r-'iiected simultaneously by the mirror, and exactly as if the mirror 

 itself were still. The rays will remain parallel, as they w r ere before 

 reflection. But if one of the rays, accelerated in its movement, reaches 

 the mirror sooner than the other, then the two reflections will not take 

 place at the same instant ; the mirror will have turned in the interval, 

 and the two rays of light, striking it at different angles, will be re- 

 flected in diverging directions. According to the corpuscular theory, 

 light moves more quickly in water than in air. According to the un- 

 dulatory theory, the reverse is true. Let us cause, then, one of the 

 rays, the superior, for instance, to traverse a tube filled with water 

 before reaching the mirror. If the theory of emission, or the corpus- 

 cular theory concerning the nature and propagation of light be true, 

 the movement of this ray will be quickened, and it will reach the mir- 

 ror first, and be reflected before the inferior ray, forming with it a 

 certain angle ; and the deviation will be such that the inferior ray will 

 appear more advanced than the other, and will seem to have been 

 drawn on quicker by the turning mirror. Now, suppose the undula- 

 tory theory to be true ; the tube of water will retard the movement of 

 the superior ray. It will reach the turning mirror later than the in- 

 ferior ray, and, of course, will be reflected later. The position of the 

 reflecting surface will be no more the same as when an instant before 

 the inferior ray was reflected. The two rays will form the same 

 angle as in the first hypothesis, but with this remarkable difference : 

 the deviation will be such, just the reverse of w r hat appeared in the 

 first hypothesis, that the superior ray w r ill now be the more advanced, 

 and always in the direction of the rotation of the mirror." 



M. Foucault thus states the result : " Light traverses water more 

 slowly than air ; and the difference is detected by the deviation of the 

 r;iv, which is reflected at a given moment from a mirror turning with 

 great velocity. Cccteris pnribus, the deviations appear sensibly propor- 

 tional to the indices of refraction of air and of water. No doubt is 

 possible as to the reality of these results. They have been obtained in 

 two different manners. The two deviations were, in the first place, 

 observed successively, and found to be unequal, the velocity of the 

 rotation of the turning mirror being the same. The deviations were 

 afterwards observed simultaneously, which makes the experiment 

 more satisfactory." Similar experiments made by Fizeau and Bre- 

 quet conduct to the same conclusion. 



