Density of the JEtlur. 491 



Wherefore, in the case when the field is excited by a single 

 spherical moving charge, e, 



_ esinfl // ft \ 

 r 2 VWp/ 



At the equator of the sphere, where r=a and = 90°, w is 

 a maximum, say iv Q , such that 



u a 2 V \47rp/' 



47T^ y 



and so iv a 2 . a 



Thus the magnetic circulation round an electron, though 

 certainly intense in its immediate neighbourhood, decreases 

 as the square of the distance ; and at any perceptible micro- 

 scopic interval of space is utterly insignificant. 



9. Certain imperfect cog-wheel, or rack-and-pinion, ana- 

 logies, elsewhere developed — with a screw motion on the axes 

 of the pinions by reason of the gyrostatic structure of the 

 medium — suggest that w and u are velocities of the same 

 order of magnitude, — suggest even that they may be equal, or 

 connected by the ratio it. 



.Now although this is uncertain and hypothetical, yet it is 

 difficult to suppose w Q greater than u, since it is excited by it ; 

 moreover u is already often so enormous, sometimes nearly 

 equal to the velocity of light. While if w be regarded as 

 decidedly less than w, the density of the aether, shortly to be 

 estimated on this basis, will be still further increased, in 

 proportion to the square of the ratio u : w , to an amount 

 wdiich becomes even more immense. As a working hypo- 

 thesis, therefore, I propose to assume that ic and u are of the 

 same order of magnitude and practically equal. 



10. An immediate consequence of this is that, in every 

 system of units, 



and as there are reasons (indicated in ' Modern Views') for 

 supposing that the quantity under the square root is unity, the 

 electronic charge comes out as equal, both in magnitude and 

 dimensions, to the superficies of its supposed spherical bulk. 

 The ordinary electrostatic unit of charge, moreover, happens 

 to equal the superficies of the conventional atom : and the 

 dimensions of all electrostatic and magnetic quantities are as 

 given in Appendix p of ' Modern Views of Electricity ' (1889). 



11. But this is a minor matter; let us proceed to estimate 



= 4 ™V(- 



