C. A. Perkins — Magnetic Permeability of Nickel. 223 



Froname* has shown that by repeatedly magnetizing in the 

 same direction, the permanent magnetism of an iron bar is 

 increased, while the total magnetism is not much altered, 

 according with Maxwell's idea of a force which causes a 

 molecule to return to the position from which it has been dis- 

 placed by the magnetizing force, unless the force exceeds a 

 certain amount. 



Tn 1S80 Wassmuth,f in a research on the effect of heat on 

 the magnetization coefficient of iron, found that for weak 

 forces it was increased by heating from 20° to 146 , but it 

 reached a maximum sooner and the maximum magnetization 

 was less by about three per cent. 



This result corresponds to that found by Eowland and shows 

 that the difference between this metal and nickel is only one of 

 degree. 



Bauer % also, using both a ring and a bar of iron and meas- 

 uring the induced current produced by reversing the mag- 

 netism, found that the same law continued at much higher 

 temperatures. For weak magnetizing forces the coefficient 

 rises rapidly, reaching a maximum as the metal becomes red 

 hot and then rapidly falls. 



Following the suggestions made in Rowland's paper, the 

 present series of experiments upon nickel was undertaken with 

 a view to secure complete curves at a number of temperatures, 

 showing the relation between the permeability and the mag- 

 netization, and thus if possible, determine the law by which it 

 changed on passing from one temperature to another. 



The rings used were from commercial rolled nickel§ and 

 therefore not as pure as those used by Rowland. Still they 

 were quite pure, as a chemical examination showed. 



Their physical and chemical proporties were as follows : 



Spec. Cross Mean 



Ring. grav. Weight. sec. diam. 



I. 8-731 34-855 '196 6 -440 ") Contain some C and Si. 



n Q HAa n _ _, . , AO i ona I A small precipitate was 

 8-746 25'ol4 -198 4 - 69b > i -i i tt a 



| produced by H 2 b. 



III. 8-394 7-814 -0624 4-427 J Contain no Co or Fe. 



Rings I and II were from the same piece. Before using 

 they were carefully annealed by heating to redness in a cruci- 

 ble filled with sand, which was allowed to cool as the fire died 

 out. 



As the temperatures employed were so high as to destroy 

 ordinary insulation the rings were covered with a thin layer of 

 asbestos paper, then a coil of wire was wound, then another 



* Pogg. Ergzbd., vol. vii. \ Ann. der Phys. u. Chem., vol. xi. 



f Sitzb. der k. Akad., vol. lxxxii. § Made by Wharton, Philadelphia. 



