JEther and Electrons. 695 



field be established throughout the region. If we suppose 

 such a field to be constituted by a uniform flow of aether 

 along the lines of force, the effect upon our material system 

 will be simple. Precisely as in the case examined in § 2, the 

 system will participate in the purely translational motion of 

 the aether, without experiencing any of the effects which 

 should be produced by an electric field. 



If, on the other hand, with a like theory as to the constitu- 

 tion of atomic matter, we adopt the view that a magnetic 

 field (in free aether) is constituted by a bodily flow of aether 

 along the lines of magnetic force, we are led by precisely 

 similar reasoning to a similar contradiction. 



Thus it would seem that a fairly strong case may be made 

 out for the proposition 



That neither the electric nor the magnetic vector is to be 

 identified with translational velocity of the aether . . (A) 



5. In view of this result, the question arises : Do the 

 volume-elements of aether concerned in electromagnetic pro- 

 cesses suffer any bodily displacement ? For example, in a 

 region through which radiation is being propagated, we have 

 to consider elements which, though large enough to be 

 treated as homogeneous portions of aether, are in every 

 dimension very small compared with any effective wave- 

 length. May we then conclude that the residtant translational 

 velocity of an aether-element never differs from zero ? Such a 

 conclusion, though not necessarily implied by the admission 

 of proposition (A) above, seems indicated as highly probable. 

 It appears indeed evident that, if translational aether-velocity 

 is an essential of an electric or of a magnetic field, it is only 

 absolute velocity in space which can be involved *, since the 

 choice of standards of position, relatively to which motion is 

 to be measured, in no way affects our judgment as to whether 

 or not an electric or magnetic field is present, or as to the 

 direction and intensity of such a field. Let us write, then, 

 tentatively, 



= u=v = w, universally, . . . , . (1) 



where u, v, w are the components of translational velocity of 

 a volume-element of aether having- its centroid at anv given 

 point. 



The motion through the aether of electrons and atomic 

 matter generally is not improbably a purely electromagnetic 



* This is not to be taken as implying a belief in absolute motion as a 

 valid physical conception ; the qualification is added only to avoid passing 

 by a conceivable alternative without mention. 



