404 Prof. Thomson, On some experiments on the [May 10, 



same. The polarization in an electrolyte is due to the presence 

 of the products of decomposition on the electrodes ; if we had any- 

 mechanical appliance to wipe them off as fast as they were formed, 

 or if we could prevent them from settling on the electrodes, we 

 should get rid of polarization. Now this is just what is done 

 by the violent, explosion which takes place when the spark passes, 

 the explosion drives the products of the decomposition violently 

 about, and prevents them from settling on the electrodes, so that 

 in this case we should not expect any polarization. 



The only assumption we have made about the electric field is, 

 that in it there is polarisation of the motion of the ether. 

 Now the motion of the ether must depend to a certain ex- 

 tent upon the motion of the molecules moving about in it. 

 This is very clearly seen in the case of the vortex ring theory 

 of matter, and would seem to be a necessary consequence of 

 any conceivable theory of the relation of matter and the sur- 

 rounding ether. Thus whenever we have a polarization of the 

 molecules we should expect a polarization of this motion of 

 the ether. Now though it does not follow that the polarization 

 of the ether is always accompanied by electrostatic effects ; — the 

 existence of permanent magnets with presumably a polarized 

 arrangement of molecules, and therefore a polarized arrangement 

 of the ether surrounding them, without any corresponding electro- 

 static phenomenon, seems to shew that it is not — yet if we consider 

 the system of polarized molecules with which we are acquainted, 

 we shall find that in nearly every case they are accompanied by 

 electric phenomena. The property of exhibiting pyro-electric 

 phenomena which all crystals seem to possess, shews, according to 

 Sir WilliamThomson's view of the phenomenon, that in crystals the 

 polarized state of the molecules is accompanied by electric polari- 

 zation. 



Again, the Thomson effect in thermo-electricity i.e. the pro- 

 duction of electric effects by the polarized motion of the molecules 

 of a body consequent upon differences of temperature in the body, 

 shews that in this case the polarization of the molecules is accom- 

 panied by electrical phenomena. The production of currents in a 

 solution of varying strength, the phenomenon of galvanic polariza- 

 tion, the electrification of bodies by strain, the effects produced 

 in thermo-electric circuits by inequalities in the strain or state of 

 magnetisation, are only some among many of the instances when 

 a polarization of the molecules is accompanied by an electrical 

 effect. 



These considerations would seem to have an application to 

 the electric discharge at low pressures. Let us suppose that the 

 electrodes are plates, as in the experiments described above, and 

 suppose the pressure is so low that the discharge spreads to a 



