EXPERIMENTS ON ETHER-DRIFT 37 



travel. Whatever the modification of the ether produced by the 

 presence of matter may be, it is a natural question to ask, 

 " When a piece of material moves, does it carry with it the 

 whole of the particular part of the ether which it contains ; or 

 does it merely remove the modification to the various portions 

 of the ether through which it passes ? " Of these two alterna- 

 tives, the former seems to be the less likely. It is somewhat 

 difficult to believe that an abrupt change in the velocity of the 

 ether occurs at the boundary of moving matter. Lodge's ex- 

 periment shows that, just outside pieces of material of ordinary 

 size, there is no sensible drag on the ether ; while the above 

 supposition leads to the conclusion that, just inside, the ether 

 has the full velocity of a moving body. Moreover, the second 

 supposition has considerable experimental support. As has 

 been shown by Fresnel, it is convenient to regard the modifica- 

 tion produced in the ether by the presence of matter as an 

 effective change in density ; or, the ether is conceived as being 

 loaded by the matter and, hence, incapable of conveying luminous 

 vibrations so rapidly as in vacuo. If we denote by p the density 

 of the unmodified ether, and by p the density inside a moving 

 body, the latter is continually changing from p to p the density 

 of the different portions of the ether which it from time to time 

 includes. The effect of this is equivalent to the carrying forward 

 of an additional ethereal density of magnitude p — p with the 

 full velocity with which the body is travelling; this is again 

 equivalent to saying that the whole of the ether within the body 

 is dragged forward by the motion of the latter with a velocity 



less than the full velocity in the ratio , . On Fresnel's 



p 



assumption that the density of the ether in any material is 



inversely proportional to the velocity of light in that material, 

 we obtain the following relation : 



*-G -?)«-(-£)- -*>* 



where u is the velocity of the body, 11' the effective velocity with 

 which the ether in the body is dragged forward, V and V the 

 velocities of light in vacuo and in the body respectively, and n 

 the refractive index of the material forming the body. It is 

 probable that this partial drag of the ether does not occur, and 

 that what really occurs is a successive modification of the 

 different parts of the ether ; but the effect on the motion of 



