Magnetism on a permanent Electric Current. 325 



the action we have been considering, placing at our command, 

 as it does, a new point of view from which to study the inte- ^ 

 rior workings of the substances examined, is destined to teach 

 us a good deal in regard to the molecular structure of bodies, 

 while helping us toward an understanding of the physical 

 nature of electricity and magnetism. 



We return now to the remarkable anomaly presented by 

 the direction of the action in iron. That the direction in this 

 metal, a magnetic substance, should be different from that in 

 gold, a diamagnetic substance, is remarkable, but not perhaps 

 surprising. ^Ye find, however, that nickel and platinum, 

 both magnetic substances, resemble in the particular above 

 mentioned, not iron, but gold and the other diamagnetic sub- 

 stances. This fact has to be taken into account in endeavour- 

 ing to apply the newly discovered action to explain the mag- 

 netic rotation of the plane of polarization in accordance with 

 the principles of Maxwell's electromagnetic theory of light. 

 Professor Rowland, therefore, in view of this difference of 

 behaviour of iron and nickel with respect to electricity, was 

 very desirous to know whether these two metals would mani- 

 fest a similar disagreement in their action upon light. I have 

 therefore, at his suggestion, repeated Kerr's experiment on the 

 rotation of the plane of polarization of Mght by reflection from 

 the pole of a magnet, using nickel for the latter instead of 

 iron. The reflecting surface used was the nickel plating on 

 one of the disks of Professor Rowland's absolute electrometer. 

 This disk, for the purpose of the experiment, was placed 

 between the poles of the electromagnet. The action upon the 

 plane of polarization, though apparently much weaker than in 

 iron, has, in the plate used, unmistakably the same direction. 

 This nickel plating, however, was executed in Germany; and 

 Professor Rowland thinks that, as the nickel of that country is 

 very impure, this specimen may possibly contain iron enough 

 to mask the true action of the nickel. 



I have already spoken of the fact that, when a strongly mag- 

 netic substance is experimented upon, complications are in- 

 troduced by the influence of the induced magnetism, which 

 affects the condition of the magnetic field through which the 

 current flows, making the value of M different from that de- 

 termined by means of the test-coil. It does not seem probable 

 that in this fact can be found an explanation of the anomalous 

 behaviour of iron ; but there is no doubt that an interesting 

 research is here suggested. For instance, it might be profitable 

 to subject to experiment a thin plate of hard steel, and deter- 

 mine to what extent the permanent magnetization induced 

 therein by the electromagnet would be accompanied by a per- 



