THE MOLECULAR PROCESS IN MAGNETIC INDUCTION. 



261 



9). A weak force produces no iiiore tliaii slight quasi-elastic deflections 5 

 a stronger force breaks up the old lines, and forms new ones more fa- 

 vorably inclined to the direction of the force (Fig. 10). A very strong 

 force brings alxtut saturation (Fig. 11). 



In an actual piece of iron there are multitudes of groups lying differ- 

 ently directed to begin with — perhaps also different as regards the 

 spacing of their mend)ers. Some enter the second stage while others 

 are still in the first, and so on. Hence, the curve of magnetization 

 does not consist of i^erfectly sharp steps, but has the rounded outlines 

 of Fig. 1. 



\ \ 



I 1 



I \ \ 



\ 



\ \ ^ 



' ', \ \ 



I I 1 ^ ^ ^ 



III'', 

 ill' 



1 1 '. I 

 111'' 



// 



// 



Fl(i. 10. Kiu. 11. 



Notice, again, liow the behavior of these assemblages of elementary 

 magnets agrees with what I have said about residual magnetism. If 

 we stop strengthening the field before the first stage is passed — before 

 any of the magnets have beconu^ unstable and have tund)led round 

 into new idaces — the small detlection simply disappears, and there is no 

 residual effect on the (configuration of the group. But if we carry the 

 l)rocess far enough to have unstable deflections, the etfects of these 

 l)ersist when the force is i-emoved, for the magnets then retain the new 

 gronping into wliicli they have fallen (lug. 10). And again, the quasi- 

 elastic dclh'ctions which go on dnring tlie third stagedo not add to the 

 residual magnetism. 



Notice, further, what hapjxMis to the gronp if after ai)plying a magnetic 

 forceinonedirection and rcmo\ ingit, 1 l)egin to apjjly force in thcojipo- 

 site direction. At first there is little leduction of the residual polarity, 



