408 Scientific Proceedings, Royal Duhlin Society. 



sented on the model.- Observe that half the indiarubber bands are 

 strained, and that in lines running round the bodies the tight side 

 of a band is always away from one body and next the other. 

 This represents the polarisation of the ether. Clerk-Maxwell de- 

 fines polarisation as a state in which the opposite sides of each 

 element are in opposite states. Now the opposite sides of each 

 band are in opposite states, one side loose, the other tight ; and so 

 it can very well represent the polarised state of the ether. The 

 displacement producing the polarisation is due to the different 

 rotation of the wheels carrying the band causing more of the band 

 to be at one side of the wheels than at the other : less at the tight 

 and more at the loose side of the pair of wheels, and this represents 

 the electric displacement producing the polarisation. The direction 

 of this displacement is at right angles to the line of the bands that 

 are strained, and is out from one body and in towards the other, 

 all round. Now, one of the first things one would expect a model 

 of the ether to represent would be the force of attraction between 

 these two electrified bodies, and which is not represented on my 

 model. This, however, is because this force depends entirely on 

 the connexions between the ether and matter; and as I have 

 already explained, this connexion is not represented on my model, 

 and in consequence of this the attraction depending on this con- 

 nexion is quite rightly not represented on my model. I may, how- 

 ever, be permitted to suggest a way in which my model might be 

 modified so as to represent this force, especially as it will emphasize 

 a point about this force that is frequently overlooked in describing 

 electrical phenomena. This point is that the force is proportional 

 to the square of the electric displacement, and consequently cannot 

 be directly and simply due to it, as in ordinary solids forces are 

 produced by displacement : for in them the force is proportional to 

 the displacement, and not to its square. The result is that the 

 sign of the force is independent of the sign of the displacement, 

 and is the same whether it be positive and negative, or vice versa. 

 This is a very important distinction, and is obvious when pointed 

 out, for it is evident that it is electromotive force, and not the 

 mechanical force, that depends on electric displacement in the same 

 way as mechanical forces depend on displacements of solids. 



Now, if the wheels over which the bands run were made of 

 some deformable, not merely incompressible, substance, such as 



