36 



Mr. G-. Barlow. Effects of Magnetisation [June 18, 



curve appears to touch the axis. The magnetic hysteresis loops were 

 also obtained for exactly the same cycle, and a comparison of the two 

 sets of curves indicates that the change of resistance is very closely 

 related to the magnetisation. Thus the " minimum points," described 

 above, appear to be identical with the points where the magnetisation 

 loops cat the axis j or in other words, when the field is such as to 

 reduce the magnetisation to zero, the change of resistance also vanishes. 

 Thus this critical value of the field is approximately equal to the 

 " coercive force'*' of the specimen. In all cases the curves represented 

 the cyclic state that is obtained after a large number of reversals of 

 the field. 



Q-r-1 I | | j | ; I I j ~r— [~7~| 



£00 



It may be remarked that the corresponding A®, I curves also possess 

 loops, and hence it appears impossible to express A9 as a function of I 

 simply. Unfortunately the magnetisation loops were not accurate 

 enough to allow of any attempt to establish a relation between the 

 quantities A©, H, and I for the general case of hysteresis. 



For nickel it was observed that if at any point of the cyclic process 

 the field be suddenly reduced to zero, the residual resistance will 

 depend on the particular point of the cycle which has been reached. 

 In the neighbourhood of the minimum points reducing the field to 

 zero may cause an increase of resistance. The minimum residual effect 

 was only about half that corresponding to the fields ±165 c.g.s. 

 In the other two metals a similar effect was noticed, but the investi- 

 gation was not carried further, owing to the difficulty of observing 

 such very small changes of resistance. These effects cannot be ex- 

 plained as resulting from the action of the earth's field. 



