6 Dr. J. G. Gray on the Magnetic Pro]>erties of a 



30 minutes before measurements were made. The specimen 

 was now subjected to a "process of reversals" and tested. 

 Finally, with a view to ascertaining whether the cooling to 

 — 190° C. had transformed the material, the specimen was 

 allowed to warm to room-temperature, and then again sub- 

 jected to the action of a " process of reversals 5 ' and tested. 

 This procedure was gone through for each specimen in i ach 

 of the three conditions specified above. 



It is well known that before a magnetic test is commi need 

 all traces of previous magnetic operations musl be < liminated 

 from the specimen. This is accomplished by subjecting the 

 specimen to the action of a " process of reversals." I'm it 

 is far from well known that any thermal treatment applied 

 to a specimen between its neutralization by the "proc< 

 reversals" and the magnetic tests which follow is fatal to 

 accurate work. A specimen rendered neutral at one tempe- 

 rature ceases to be neutral if the temperature is changed. 

 Before carrying out tests at a definite tern] erature following 

 upon temperature change it is essential that the specin en 

 should be subjected to a "process of reversals" at tl e new 

 temperature. This important fact has been clearly shown 

 by a number of experiments carried out by Dr. A. 1). Ross* 

 and the writer. 



In PI. I. fig. 3 are shown the magnetization curves 

 yielded by a specimen of pure nickel when tested, at room- 

 temperature and at — 190 c C, in the annealed and quenched 

 conditions. It will be seen that for both condition- (^ the 

 metal the magnetization curve corresponding to the liquid- 

 air temperature lies everywhere below that corresponding to 

 room-temperature. The curves show that hail the mag- 

 netizing force been pushed to a sufficiently high value 

 crossing of the curves for the two temperatures would have 

 taken place ; but lor \)i\w annealed of quenched nickel the 

 susceptibility at — 1 90° ( '. is less than that at room- tern] erature 

 Tor all fields lying between and 200 C.G.S. unit-. 



The results obtained on testing the nickel-manganese 

 specimen containing 5 per cent. Mn are shown in Table I, 

 and in PI. I. fig. 4. The addition of the manganese has 



brOUghl about very marked changes in the magnetic pro- 

 perties of the material. The susceptibility of the alloj in 

 the annealed condition is inferior to that o\' pure annealed 

 nickel lor all \ allies of the magnetizing force. For this 

 condition of the alloy crossing of the magnetization curves 

 eon. -ponding to 15° C. and —190 0. occurs for a value of 

 the magnetizing force ol 1~> g.g.s. units. For H — i'oo ( .G.8. 

 * See Phil Ma* [6]auri. p. J (1911), 



