270 MAGNKTISM 



maximum value of K is raised from 185 to '-MH5 and /<. is in- 

 creased practically in the same proportion. 



The magnetometer method is not the best for accurate absolute 

 measurements. The polarity is not all concentrated at or close 

 to the ends of the wire and its distribution cannot be calculated. 

 The value obtained for I is therefore inexact. The correction to 

 H, too, is only an approximation to the true correction. M on- 

 exact results are obtained by the ballistic- method described below. 

 But the magnetometer method is specially ncll adapted for in- 

 vestigations on the effects of >trc^c> on magnetisation, .since the 

 wire tested can be subjected to end pulls or to Uists while within 

 the solenoid. It serves well, too, to give a general idea of the 

 results obtained on subjecting a .specimen to a magnetising force 

 and on carrying the magnetising force through a cycle. ina>- 

 much as it is easy to work and the theory i- simple and L 

 fairly good results when the .specimen trxted i> UTV thin compared 

 with its length. By it \\<- ma\ obtain the curve alnad\ Ini 

 described in Chapter \I\. p. 1S:J, and represented in I _ W7, 



Fi<;. 207. 



It is interesting to note that in the ti st represented b\ 

 the residual magnetisation when the current is reduced to < 

 still 1162, or 92 per cent, of the value it attained \\ith the 

 increasing current. AVhen the current is reversed the magnetisa- 

 tion does not disappear till the field H= 2*75 gausses, and this 

 is the value of the "coercive force" for the specimen. It is to l>e 

 remembered that though the external eflict of the magnetisation 

 disappears under this coercive force, the wire is not in the de- 

 magnetised condition from which it started. It is still affected by 



