650 Dr. 0. Y. Burton on the 



conclusion ^hich may be admitted to have a prima facie- 

 reasonableness. For an observer who knows that his 

 velocity relatively to the aether must at some time of the year 

 amount to 28 kilometres per second at the least, might well 

 hesitate to believe that between an intense magnetic field and 

 a space free from magnetic force the only difference was a 

 minute difference of aetherial velocity: a velocity of (perhaps) 

 a fraction of a millimetre per second in the former case, as 

 against zero velocity in the latter. 



The electric vector in free cether is not constituted by 

 translational cetlierial velocity. 



17. By suitably modifying the two trains of plane-polarized 

 plane waves referred to in § 11, a uniform electrostatic field 

 free from magnetic induction is equally readily obtained, and 

 the proof of the proposition just enunciated follows on lines 

 precisely similar to the proof given in the magnetic case. 

 Such a demonstration, however, is hardly called for, in view 

 of von Helmholtz's objection that if a continuous flow of 

 aether were taking place along the lines of electric force, 

 every charged body would behave as an unlimited source of 

 aether or as a corresponding sink. 



The Poynting vector in free cether is not identifiable 

 with translational cetherial velocity. 



18. In any purely progressive electromagnetic wave-train, 

 the vector product of the electric and magnetic vectors 

 (Prof. Poynting's vector) is persistently in the direction of 

 propagation, and never in the opposite direction. For 

 example, if we take the train represented by (5), (6), putting 

 for (J>'(z — ~Vt) the more definite expression A cos s(z — Yt) r 

 the Poynting vector is 



{0, 0, A 2 Vcos 2 ^-V0}, 

 that is 



{0, 0, 1A 2 V + J,A 2 V cos 2s(z~Vt) } : . . (10) 



and if we suppose the aether to be flowing at each point in 

 the direction of this vector, the motion corresponding to (10) 

 will be a wave-train of compressional-rarefactional type (like 

 sound-waves in air) superposed upon a general drift of aether 

 in the direction of wave-propagation. 



19. We have thus to face the conclusion that the aether is 

 compressible ; its compressibility being not merely of that 

 minute residual order which has been assumed as a means of 

 accounting for the phenomena of gravitation, but of such 



