188 Dr. E. H. Kennard on 



The author yields to none in appreciation of Barnett's work 

 on unipolar induction, and in respect for his long experience 

 with electromagnetic theory ; but on this point it seems clear 

 that all of the writers mentioned have, through an oversight, 

 arrived at a wrong conclusion. 



The matter has been fully discussed elsewhere*, but perhaps 

 sl summary of the argument will not be amiss here. Barnett 

 urges that an adherent of the moving-line theory must 

 suppose the lines (i. e. the rotating magnetic system) to set 

 up in the cether the same electromotive intensity that is caused 

 to act upon a material dielectric, and shows from this 

 assumption that the charges developed upon conductors 

 would be the same whether the lines moved or not. The 

 •objection which seems to the author conclusive is that this 



electromotive intensity, of magnitude -[Vx B], does not in 



general satisfy Laplace's equation. (The proof is simple.) 

 In spite of this fact Barnett employs the two additional 

 assumptions : 



/=i[VxB]rf«+E, (1) 



where /= total electromotive intensity, e — electric force 

 due to changes in the magnetic field and vanishes in the 

 present instance, E = electric force due to electrostatic causers 

 alone ; and the equation 



div(K/) = 47r / 3, (2) 



where p=. electric density. 



Now in free aether we know from electrostatics that 



■divE = 0. Hence in free aether div/= - div [Vx B], and 



may not vanish, which contradicts (2). We are thus forced 

 to the conclusion that if the moving-line theory is correct, 

 then the lines do not act on the aether, but only on material 

 bodies, and instead of (2) we must write (with Lorentz) 

 div (KE) =47170. 



The real root of the matter seems to be that the aether 

 cannot be treated as only a particular species of the genus 

 " dielectric," but must be regarded as complementary to all 

 material dielectrics. The displacement or polarization in 

 matter does not need to be solenoidal, because its sources can 

 be cancelled by sources of opposite sign in the displacement 

 in the aether, and the total displacement may then be 



* E. H, Kennard, Phys. Kev. May 1918, p. 355. 



