1889.] On Time-lag in the Magnetisation of Iron. 



271 



through a Bunsen flame. After it was put in place the method of 

 reversals was applied to extract a small amount of magnetism which 

 it had acquired in being handled. In the experiments which I shall 

 first describe the effects of very feeble magnetic force were examined 

 by making and breaking the circuit of the magnetising solenoid while 

 the current was adjusted to produce a force of less than 0*1 c.g.s. It 

 was found that the immediate effect of each make and break could be 

 balanced very exactly by adjusting the position of the compensating 

 coil, and so long as the magnetising force was considerably less than 

 01 c.g.s. the distance at which the coil had to be set to give this 

 balance was as nearly as possible independent of the value of the force, 

 and was the same for " break " as for " make." The position of the coil 

 was adjusted so that at the instant when the magnetising current was 

 set up by pressing down a contact key, there was no sudden deflection 

 of the magnetometer mirror to either side. When the compensation 

 was right the spot of light simply began to drift slowly towards the 

 side corresponding to increase of magnetism ; when there was over- 

 compensation, the spot of light gave a quiver to the opposite side 

 before beginning to drift, and the position of the coil was adjusted 

 by drawing it back little by little until the quiver on pressing down 

 the key disappeared. The amount of magnetism that was balanced 

 was afterwards measured by removing the iron, but leaving the mag- 

 netising solenoid and the compensating coil in place, and observing 

 the deflection of the magnetometer when the same current was passed 

 through the empty solenoid and the compensating coil. This deter- 

 mined the immediate magnetic effect of the magnetising current on 

 the iron, and the subsequent creeping up of the magnetism was of 

 course determined by observing the drifting of the magnetometer 

 needle which had ensued after applying the current while the iron 

 was in its place. 



In the following experiment a current of 21 on the arbitrary gal- 

 vanometer scale (equivalent in this case to a magnetising force of 

 O044 c.g.s.) was made, after the wire had been completely demagne- 

 tised, and after the compensating coil had been adjusted to balance 

 the immediate effect. Magnetometer readings were taken 5 seconds 

 and 60 seconds after "make;" and at 60 seconds the current was 

 broken, and magnetometer readings were taken 5 seconds and 

 60 seconds after " break." The immediate effect (balanced by the 

 compensating coil) was equivalent to twenty-five divisions of the 

 magnetometer scale. 



Time after 



"make." Magnetometer. 





 5" 8 



60" 



VOL. XLVI. 



Time after 



" break." Magnetometer. 



13 



5" 5 

 60" 



