L. Page — Relativity and the Ether. 171 



Lorentz-Einstein transformation. Hence the divergence of E 

 will be zero in ether devoid of matter, and every charged particle 

 will be a continuously emitting source of strains. One other 

 assumption, which is suggested by symmetry, is necessary. 

 To an observer relative to whom the charged particle is, for the 

 instant, at rest, the strains must be emitted uniformly in all 

 directions. 



The most convenient way to represent these strains geomet- 

 rically will be by means of tubes of strain spreading out from 

 each charged particle. While we may consider these tubes, 

 which are analogous to "tubes of force," to have a cross- 

 section which increases as one proceeds outwards from the 

 charged particle in such a manner that they fill all space, yet we 

 must endow them with some property that will enable us to 

 determine the direction of the strain (which in general will be 

 different from its direction of motion) at each point in space. 

 E will have the direction of the tube at the point and instant 

 considered, and will be proportional in magnitude to the num- 

 ber of tubes per unit cross-section. We will choose our units 

 so as to make this factor of proportionality, as well as that 

 between the divergence of E and the density of charge, equal to 

 unity. This amounts to the adoption of Lorentz's units of charge 

 and electric force. Any point on a tube of strain will move 

 with the velocity of light in a straight line of length I, drawn 



from the position of the particle at a time — earlier, to the 



point in question. To an observer relative to whom the parti- 

 cle is at rest at the instant considered, the tubes of strain 

 in its immediate vicinity will appear to diverge from the 

 particle in such a way that equal solid angles will contain 

 equal numbers of tubes of strain. If a number of particles 

 are producing strains in the ether, the resultant strain at any 

 point will be the vector sum of the strains due to the individ- 

 ual particles. 



Before proceeding to a consideration of the strains due to a 

 moving charge, let us consider the simple case of a charge 

 which is permanently at rest relative to the observer. Accord- 

 ing to former theories, the charge would be surrounded by a 

 uniformly diverging field of stationary strains, whereas, accord- 

 ing to the theory developed here, the charge would be a 

 continuously emitting source of strains which move outward 

 through the ether with the velocity of light. Since, however, 

 the direction of motion would be parallel to the strains them- 

 selves, the number of tubes of strain per unit cross-section at any 

 point in the field would be constant, and the same as in former 

 theories of the ether. A little consideration will show that 



