Fitzgerald — On a Model Illustrating Properties of Ether. 409 



indiarubber, it is evident that they would be deformed by the 

 straining of the bands. Now, if inextensible threads were 

 wound round them, and the ends of the threads connected respec- 

 tively with the bodies a and h, it is evident that as the deformation 

 of the circular wheels would enlarge their circumferences, these 

 threads would tend to draw the two bodies together. Now, the 

 shortening of the threads would be proportional to the square of 

 the linear deformation of the wheels, and so the resulting force 

 might represent the electrostatic attraction of the two bodies. If 

 we consider the force that would be thus produced between two 

 infinite planes, we see that it would be the same no matter what 

 their distance apart so long as the polarisation is kept the same ; 

 and that consequently it will represent a force varying inversely as 

 the square of the distance from a point. I mention this crude 

 mechanism in order to show that it is possible to represent by 

 mechanism the connexion between the ether and matter, and also 

 to illustrate how electrostatic attractions depend entirely on the 

 way the ether is connected with matter. 



Let us now return to consider what other properties of the 

 ether are represented by the model. Observe, in the first place, 

 that during the time polarisation is taking place the wheels are 

 rotating, and that the rate of rotation of the wheels is pro- 

 portional to the rate of increase of polarisation, and that the 

 direction of the axis of rotation is perpendicular to the direc- 

 tion of the displacement. We, of course, at once conclude that 

 the magnetic force is properly represented by the rate of rota- 

 tion of the wheels, and its direction by the axis of rotation. We 

 thus see that any change of displacement produces rotation of the 

 wheels, i. e. magnetic force ; and we will presently see that what 

 represents an electric current is similarly accompanied by rotation 

 of the wheels. The momentum of the wheels while rotating repre- 

 sents the kinetic energy of the currents producing the magnetic 

 force, and in the case of one current is its self-induction. Now, if 

 we suppose something to loosen the hold of the bands on the 

 wheels, this will represent the presence of some partially conducting 

 matter, for evidently perfectly conducting matter must be repre- 

 sented by the fact that no amount of difference of rotation of 

 neighbouring wheels, i. e. no amount of electric displacement can 

 produce any polarisation, and we have represented it by removing 



SCIEN. PROC. K.D.S. VOL. IV. FT. VII 2 M 



