274 MAGNETISM 



already employed the susceptibility for that change was nearly tin- 

 same as when the specimen was initially unmagnetised. 



Thus if GAP, Fig. 2 10, represents the magnetisation curve with 

 an increasing current, OA is the part under small forces with constant 

 JUL and K and no retentivity. If the magnetisation is carried up to 



N M H 



Fio. 210. 



P and then a small decrease MN is made in II, PQ parallel to OA 

 is the magnetisation curve. If H is again increased 1>\ VM \\r 

 arrive again at P. 



But when the initial magnetisation is much increased and the 

 region of saturation is approached. 1\) has a less slope, or (In- 

 susceptibility for the small change, which we may denote by TTT> 



decreases. 



With annealed iron there- is a lime lag in the magnetisation. 

 This effect has been examined by Ewing, and we refer tin- reader 

 to his work, p. 129, for an account of the experiments. 



Magnetisation in very strong fields. Ewing' s 

 isthmus method. In Fig. Jll NS an the pole pieces of a 

 very powerful electro-magnet, the tips being bored out cylindrically 

 about an axis through o perpendicular to the plane of the figure. 

 A bobbin B just fits in between the poles. It has a coil of out- 

 turn wrapped close about the central neck and with its ends 

 connected to a ballistic galvanometer. It can be rotated through 

 180 about the axis through o so that the induction through it 

 is reversed. The throw of the galvanometer on this rotation : 

 the value of the induction through the neck. Since there is no 

 free magnetism on the surface of the iron about the neck the field 



