Nickel begins to lose suddenly its Magnetic Properties. 373 



round a brass tube slit throughout its entire length, and having 

 an internal diameter of 2'5 centim. Before being wrapped 

 round the brass tube the wire was well coated with pipeclay 

 moistened in water. Round 10 centim. of the central portion 

 of the solenoid were wrapped 240 turns of another piece of the 

 same wire, also coated with pipeclay, to serve as a secondary 

 coil. The whole was then introduced into an air-chamber and 

 the wet pipeclay allowed to dry, at first slowly at the tempe- 

 rature of the air, and afterwards more quickly at temperatures 

 which were gradually raised to 400° C. The air-chamber 

 consisted of two concentric copper cylinders 40 centim. in 

 length, enclosing between them an annular space about *5 

 centim. thick which was filled with fine sand, and was heated 

 by a row of burners placed underneath. The temperature of 

 the air-chamber was calculated from the alteration of electrical 

 resistance of a coil of platinum wire, whose resistance had 

 been very carefully determined at different temperatures up 

 to 100° C., and expressed in terms of the temperature by a 

 formula of the form 



R^^l + at-bt 2 ), 



where R t and R are the resistances at t° C. and 0° C. respec- 

 tively, and a and b are constants. The platinum coil was 

 wound double and was placed inside the slit brass tube close 

 to the nickel wire, but insulated from it and the tube by 

 asbestos ; the length of the coil was 10 centim., and it occupied 

 the central portion of the air-chamber. To the ends of the 

 coil were hard-soldered stout copper terminal rods, which 

 passed through a wooden cap closing one end of the air- 

 chamber, and served to connect with a Wheatstone-bridge 

 arrangement employed for determining the resistance of the 

 platinum. The other end of the air-chamber was also closed 

 with a wooden cap, through holes in which passed the ends of 

 the wires of the magnetizing solenoid and the secondary coil. 

 After the pipeclay coating had become thoroughly drj' the 

 resistances of the primary and secondary coils were tested, and 

 found to be sensibly the same as before winding. The object 

 of coating the cotton covering of the wires with pipeclay was 

 to maintain the insulation, for at the temperatures reached in 

 some of the experiments the cotton became charred. Whilst 

 the coils lay undisturbed the pipeclay coating served the pur- 

 pose for which it was intended very well ; but when they 

 were removed from the air-chamber on the conclusion of the 

 experiments, both the charred cotton and the pipeclay easily 

 came away from the wire, and so rendered the coils useless. 



A pair of primary and secondary coils exactly similar to the 

 Phil. Mag. S. 5. Vol. 25. No. 156. May 1888. 2 C 



