Magnetic Field about a Revolving Magnet. 101 



charged negatively below and positively above in the present 

 case (fig. 1) . It remains, then, to investigate the nature of 

 this charge in order to prove which hypothesis is true — i. e. 

 in order to show whether the magnetic field rotates with the 

 magnet or does not do so. 



For it will be perceived that under hypothesis (2), when 

 the magnetic field (per contra) is assumed to revolve with the 

 magnet, the field will cut through or intersect the disk p 

 (fig. 2) as the magnet revolves ; and so the disk will, accord- 

 ing to known principles, be charged through electrodynamic* 

 induction with electricity of opposite sign at its centre and 

 periphery respectively, as fig. 2 shows. 



But it will be observed that the decomposition of the elec- 

 tricity in the disk takes place under hypothesis (2) in a 

 direction at right angles to that which occurs under hypo- 

 thesis (1). So that it will be evident that we have a physical 

 test between the two opposite hypotheses here. Under hy- 

 pothesis (1), namely, the disk is polarized by electrostatic 

 induction in the direction of its thickness ; and under hypo- 

 thesis (2) the disk is polarized by electrodynamic induction 

 in a direction at right angles to the former, viz. transversely, 

 or along all its radii, as fig. 2 indicates. These diverse phy- 

 sical effects then, as it appears, are capable of constituting a 

 test to decide between the two rival hypotheses. 



But I am informed by Prof. H. Hertz, of Bonn, to whom 

 I had communicated the above by letter, that with a normally 

 sensitive electrometer a considerable velocity of rotation with 

 a magnet of large size would be calculably required in order 

 to produce a distinct deflexion under either hypothesis : and 

 facts of this kind discourage from trying the experiment. 



I would therefore venture to ask — in order to facilitate the 

 experiment — whether with the aid of Mr. Vernon Boys's 

 discoveries, or those of others, it is not possible to bring 

 the electrometer to a much higher degree of sensitiveness ? 

 Prof. Hertz lays stress on the great interest attaching to the 

 inquiry, and agrees that the conditions of test as above pro- 

 posed should! be capable of deciding this question, — presup- 



* This charge will react electrostatically on the magnet, to produce a 

 feeble (secondary) charge there, which need not be taken into account, 

 as it does not affect our experiment. 



t It may be, of course, that the magnetic field partly partakes of the 

 motion of the revolving magnet, or that something between the two 

 hypotheses is true. Then a modified distribution of charge on the disk 

 would obviously present itself, consistent with such conditions. In other 

 words, if neither vertical nor horizontal polarization of the disk were 

 found to be precise (but a superposition of the one upon the other), then 

 this would obviously indicate that the field partly shared the rotation of 

 the magnet, so that both magnet and disk were intersected by the field. 



