12 



Proceedings of the Royal Irish Academy. 



It is obvious that there is something unsound in the assumptions 

 which lead to such results. The explanation that the ions are not 

 describing circuits or are not influenced by the magnetic field in the 

 manner conceived would by no means be a satisfactory solution of the 

 question. It is more feasible to regard the discrepancy as arising in the 

 application of the theory to the experimental facts. If the ions were 



describing circuits of molecular size, then would have 1/10000 of 



TjCZ 



the values given above, and consequently such a system could not 

 produce more than 1/1000 of the observed rotation. Another way of 

 stating the case is that the magnetic field produced by the rotating 

 ions would require to be about 10,000 times greater than it is to 

 produce the actual rotation. This suggests that, if the magnetic 

 susceptibility were calculated by means of the data obtained above 

 concerning the ionic circuits, we should obtain also impossible values. 

 This additional form of the argument is given here, as it utilises 



directly the calculated values of and thus avoids the question of 



TXZ 



the periodic time. 



If we make the usual hypothesis that the ions in diamagnetic 

 substances on the application of magnetic force are set rotating in a 

 plane normal to the field in such a direction that the force due to the 

 ion, whether positive or negative, is opposed to the field, all terms in 

 the expression for the magnetic susceptibility can then be taken 

 positive. We have the following expression for h from (6) 



It simplifies the argument if we refer these quantities to the 

 molecule and the molecular constants, writing 



= number of ions of class {p) per molecule, 

 JV = number of molecules per molecular weight in grammes, 

 V = volume of the molecular weight in grammes, 



Magnetic Rotation and Magnetic Susceptibility. 



k = 2 



where 



n = number of ions per unit volume, 

 e = charge on the ions. 



