438 DR. CHARLES H. LEES ON THE EFFECT OF TEMPERATURE ON THE 



The equal resistances, M,M 2 , which formed two sides of the bridge in the determi- 

 nation of the difference of resistance of the two spirals, each consisted of 50 centims. 



of No. 29 double silk-covered manganine 

 wire, wound in parallel on a varnished block 

 of wood, adjusted to equality within 1 part 

 in 10,000, and varnished. The resistance 

 of each coil, with its short copper leads, 

 was found to be 4 '8 50 B.A. ohms. 



In series with that spiral PI or P 2 , which 

 in any experiment had the smaller resist- 

 ance, could be placed an external resist- 

 ance N capable of adjustment by means of 

 mercury cups to '01 ohm. This did not 

 admit of an actual balance in the bridge, 



but the deflections of the galvanometer with the resistance adjusted a little above 

 and a little below the value necessary for a balance were obtained, and the latter 

 value calculated. 



To fix the temperature at which each determination was made, the hotter spiral 

 and the manganine coil in series with it could, by means of mercury-cup connections, 

 be made into two of the resistances of a second bridge, the third arm of which was a 

 coil of bare manganiue wire, M 3 , wound on a block of wood and varnished, and the 

 fourth arm, P :j , a resistance box with ohms and tenths, and a mercury switch resist- 

 ance capable of adjustment to '01 ohm. These were so connected that the adjustable 

 resistance was proportional to the resistance of the platinum spiral. The resistance 

 of the coil in the third arm was for convenience taken such that the resistance of the 

 fourth arm in tenth-ohms was approximately the temperature of the platinum spiral. 

 The galvanometer was of the moving coil type, had a resistance of 22 ohms, and 

 was used with a shunt of 10 ohms across its terminals. It was connected to either 

 bridge through a thermo-electric key, K, which carried three fl -shaped conductors, 

 the ends of which dipped into six mercury cups arranged as shown in fig. 4. When 

 connected to the second bridge, a resistance R was placed in series with it to 

 diminish its sensitiveness. 



Method of Experimenting. 



In making a test of a material, the material was melted in the brass tube D, the 

 tube being for this purpose placed in a water or oil bath, the temperature of which 

 was raised till it was just sufficient to melt the substance. The end A^ of the 

 apparatus A^B^ which was to be placed in the material, was surrounded by a 

 tube and placed in the bath to be heated to the same temperature. When the 

 material in D was liquid, the apparatus was immersed to the requisite extent in it. 

 The tube and contents were then removed from the bath, and cooled from under- 



