416 Scientific Proceedings, Royal Dublin Society. 



form as the heat energy in material hodies ; but I see no reason 

 for assuming that, because the ether has a structure, it must be 

 capable of being a vehicle for heat energy of exactly the same form 

 as that in material bodies ; in fact, unless there were another ether 

 bearing to the known ether the same relation as the known ether 

 bears to matter, it would be impossible that the energy of the 

 known ether should be in exactly the same form as the heat 

 energy in matter. If the medium have a structure we can ex- 

 plain electric polarisation much more satisfactorily as a change of 

 structure than as a displacement. Maxwell defines polarisation 

 (Elect, and Mag., vol. i. s. 60) generally thus: — "An elemen- 

 tary portion of a body may be said to be polarised when it acquires 

 equal and opposite properties on two opposite sides." He, how- 

 ever, immediately afterwards calls it electric displacement, and 

 there is danger from this of its being supposed to be merely a dis- 

 placement of the medium in the direction of the electric displace- 

 ment. It may be this, but I think is is very desirable that the 

 more general notion of polarisation as a change of structure of the 

 elements and not as a displacement of the elements should be 

 emphazised, and it is in the latter form that polarisation is ex- 

 hibited in the model. Another point to which attention is called 

 by the model is the difference between the mechanical and the 

 electrical stress in the medium. The model does not represent the 

 mechanical stress at all. This latter depends necessarily on matter, 

 for mechanical stress is a material phenomenon, and its connexion 

 with the electric polarisation depends on the connexion between 

 matter and ether. This is true of the mechanical forces, due to 

 both electrical and magnetic polarisation. That the mechanical 

 stress is not connected with the electric displacement in the same 

 way as material stresses are connected with their material strains 

 in solids is obvious from its being proportional to the square of the 

 displacement, and so being independent of its direction. All 

 theories of the ether that suppose it to be simply a jelly with 

 matter spread through it, like grapes in a jelly, hardly seem to 

 attribute sufficient importance to the difficulty of explaining upon 

 any such simple hypothesis such phenomena as electricity and 

 magnetism; and although the equations of motion of the jelly 

 may fairly well represent the equations of motion of the ether, as 

 regards its propagation of light, yet the properties of a jelly 



