Resistance of Metals and Alloys. 279 



experiments the wires of which the resistance was to be deter- 

 mined were wound on a rectangle of mica, having serrations 

 or teeth cut upon the edge, and the ends of the wire were 

 then soldered to stout copper connexions. For many reasons 

 this proved to be an undesirable method. Many wires, such 

 as aluminium, were easily broken by the sharp bend in the 

 wire, and other soft wires, such as lead, were deformed at the 

 same point of bending. 



In preparing coils for the purpose we had in view it was 

 necessary to keep them small, so that they might be easily 

 contained in a moderate-sized vacuum-jacketed test-tube in 

 which the liquid gas was held. For obvious reasons also any 

 large amount of non-conducting material round the wire was 

 objectionable. After some trials the following method of 

 constructing the resistance-coils to be used was adopted, and 

 answered very well. Stout high-conductivity copper wire of 

 4 millim. diameter was cut into lengths, and pairs of these wires 

 bent up and bound together at the middle, but insulated from 

 each other with indiarubber tape. The ends of the wires 

 were well tinned and the bottom ends nicked. 



A very thin sheet of vulcanized fibre was then bound round 

 the bottom part of the copper rods and sewn to them by silk 

 thread. This formed a thin cylinder on which the wire to be 

 tested was carried. A stout silk thread was then wound 

 spiral fashion round the cylinder. The wire to be tested, 

 generally 1, 2, or 3 metres in length, was then care 

 fully wound on the fibre cylinder in between the silk thread, 

 the several turns of which kept the turns of wire from 

 becoming short-circuited. The dimensions of the cylinder of 

 vulcanized fibre varied according to the length of wire used, 

 but was from two to three centimetres in diameter and four 

 to five centimetres in length. A large number of such bobbins 

 having been prepared, the next step was to anneal the wires 

 on these coils. 



The so prepared resistance-coils were immersed in melted 

 paraffin wax and kept at a temperature of 200° C. for some 

 time. These being slowly cooled they were annealed and at 

 the same time the insulating portions of the coils well saturated 

 with paraffin. 



§ 6. The resistance measurements were carried out by 

 means of a Wheatstone's Bridge by Messrs. Elliott Bros. 

 The coils of this bridge were of platinum-silver and adjusted 

 to read in standard ohms. The coefficient of variation of resist- 

 ance of the coils of the bridge with temperature was determined 

 and found to be *031 per cent, per degree Centigrade. The 

 temperature of the bridge-coil box was always taken at the 



