MOTION THROUGH ETHER 245 



stitutional ultra-microscopic whirl as may be required to 

 explain its elastic properties. The burden of proof rests on 

 those who deny the simple postulate of absence of gross loco- 

 motion in the ether. In so far as the elastic stress concerned 

 in potential or static energy has to be explained on kinetic 

 principles, it is perfectly legitimate to postulate a circulatory 

 or vortex structure for the ether, such as can account for its 

 optical elasticity and its electric and magnetic relations. But 

 such sponge-like vorticity is of the kind known as stationary 

 motion, and does not appeal to us as motion at all, only as 

 structure. 



Motion relative to the ether of space is, therefore, the 

 nearest approach to the idea of absolute motion that we are 

 able to form ; and it is already possible to show that we know 

 something, though not very much, about this motion. 



The fact that the effective or apparent mass of a body is 

 a function of its velocity squared, so that an expression for 

 it contains the factor (1 — z> 2 /c 2 )-*, and therefore has an 

 increased value as the velocity of light is approached, is the 

 fact we have to argue from. 



On the strength of that I assert that the solar system is 

 not moving through the ether at any excessive speed near 

 that of light ; at least that there is no component of such 

 great speed in the plane of the ecliptic. For if there were, 

 the earth's mass would vary according to its position in the 

 orbit : and this ought to be perceptible as a definite perturba- 

 tion, causing the orbit to revolve in a hypocycloidal manner. 



That an effect would be perceptible can be shown in ele- 

 mentary fashion, thus : Suppose the sun's proper motion in 

 the plane of the ecliptic to be within 1 per cent, of the 

 velocity of light, so that (1 — v^/c 2 )' 1 would equal 7, then the 

 whole solar system's apparent mass would be 7 times what 

 it would be if stationary. 1 It may be said that there would 

 be no means of detecting a uniform effect of that sort. 



1 The simple factor for inertia-increase m/m = (i - v 2 jc 2 )-i is a fairly close 

 approximation to a more complicated expression, such as is quoted in my book on 

 "Electrons," pp. 133 and 225, 



— = q-7 2 — ^(( 2 - cos 2& ) ~ ( 2 cos 2d - ~^-,) 



m 8(1 - cos 20) V ' x ' sin 28/ 



where sin 8 — v\c. This is supposed to be more correct, but the simple expression 



m — m sec 8 will do for the illustration in the text. The completer expression 



gives 6'4 instead of 7, when sin 8 = -99, and it always works out a trifle less than 



