PROFESSOR J. A. EWING ON THE MAGNETIC QUALITIES OP NICKEL. 337 



been applied is sensibly nil. It was only when the value of Jp was raised to about 

 2 c.g.s. units that any trace of residual magnetism could be detected with certainty, 

 although by that time the induced magnetism had become great enough to allow 

 a tenth of it, or less, to have been determined without difficulty. As to the induced 

 magnetism, it is to be noticed that the presence of stress has much less effect on 

 the initial value of the permeability than it has when a later stage in the process of 

 magnetisation is reached. 



The experiments were completed by examining the magnetisation of nickel in very 

 intense fields, by means of the " isthmus " method.* The same nickel bar was fitted 

 with conical expanding end pieces of soft wrought iron, into which the ends of the 

 bar were sunk, the whole forming a built-up bobbin with a short narrow neck of 

 nickel, and the diameter of the neck was turned down to 0'399 cm. On the neck 

 two induction coils were wound, one close to the metal and the other a little way out, 

 so as to enclose an annular space, in which the field was measured by observing the 

 difference in the inductive effects on the two coils. The bobbin was placed between 

 the pole-pieces of the large electro-magnet of the Edinburgh University Laboratory, 

 and its magnetism was measured by suddenly withdrawing it from between the poles, 

 while the magnet was more or less strongly excited. Afterwards readings of the 

 residual magnetism (which of course did not show itself when the bobbin was with- 

 drawn from the field) were taken by removing one of the conical end pieces and 

 slipping off the induction coil. 



Measurements were made in fields ranging from 3450 c.g.s. to 13,000 c.g.s. 

 Allowing for residual magnetism these produced values of the induction 9), which 

 ranged from 9850 to 19,800. Treating the magnetising field, which was measured 

 outside the metal, as equal to the magnetising field within the metal itself (an 

 assumption not far from true), we may calculate the permeability p. and the intensity 

 of magnetisation 3- The permeability ranged from 2'9 with a field of 3450 down 

 to 1'5 with the highest field that was applied. The values of 3 fluctuated irregularly 

 between about 480 and 540, but showed no distinct progressive change either in the 

 way of increase or decrease as the field was strengthened. The mean of 3 in six 

 determinations was 515, and this may be taken as fairly representing the limiting or 

 saturation value of the intensity of magnetism for the particular specimen of annealed 

 nickel dealt with. Saturation was practically reached at the lowest field, viz., 3450. 

 Of the whole quantity, 515 for 3. the residual part was 160, and this also was 

 sensibly constant throughout the range of these experiments. 



Roy. Soc. Proc,' vol. 42, p. 200. 



MDCCCLXXXVIII. A. 2 X 



