108 



that it was made in the circuit of metal round the cylinder, 

 and that my object was to find the relative motion of the 

 cylinder and the electricity. Altogether I think it may be 

 taken as exjDerimental proof of the fact previously stated 

 that if a steel top were spinning under the inductive 

 influence of a body charged with negative electricity the 

 effect would be that of a current round the top such as 

 would render it magnetic. 



The cause of terrestrial magnetism has not been the 

 subject of so much speculation as many much less important 

 phenomena. It seems to have been regarded as part of the 

 original nature of things like gravity, and the heat of the 

 sun, as a cause from which other phenomena might result, 

 but not as itself the result of other causes. 



Yet, when we come to think of it, it has none of the 

 characteristics of a fundamental fact ; it appears intimately 

 connected with other things, and when two phenomena 

 have a relation to each other, there is good reason for 

 believing them to be connected, either as parent and child, 

 or else as brother and sister, the one to be derived from the 

 other, or else them both to spring from the same cause. 



Now the direction of the earth's magnetism bears a 

 marked relation to the eaith's figure, and yet it can have 

 had no hand in giving the earth its shape, which is fully 

 explained as the result of other causes ; therefore, we must 

 assume that the figure of the earth has something to do 

 with its magnetism, or what is more likely, that the rota- 

 tion which causes the earth to keep its shape, also causes it 

 to be magnetic. 



If this is the case then there must be some influence at 

 work with which we are as yet unacquainted — some cause 

 which coupled with the rotation of the earth, results in 

 magnetism. From the influence which the sun exerts on 

 this magnetism we are at once led to associate it with the 

 cause. Yet the cause itself cannot be the result of either 



