[ 376 ] 



XLII. Thermoelectrics of Platinum-indium and of Platinum- 

 rhodium. By Carl Barus*. 



1. JN TROD UC TORY. — Before proceeding further it 

 seemed fit to clinch the remarks of my preceding 

 paper t by making an actual comparison of the thermoelec- 

 tromotive forces of iridioplatinura- platinum and rhodioplati- 

 num-platinum within an interval of temperature of about 

 1700° C I was quite prepared to find a simple relation 

 between the corresponding data of these two couples ; but a 

 relation so nearly, if not actually linear, is a striking result, 

 and was not anticipated. 



2. The Thermocouples. — The first of these consisted of soft 

 platinum combined with an alloy of platinum with 20 per cent, 

 (nearly) of iridium. The material was purchased in Paris in 

 1883, and used throughout my high-temperature work. 



The second couple consisted of pure platinum with an alloy 

 of platinum with 10 per cent, (nearly) of rhodium. It was 

 obtained from Paris (Carpentier) in 1892. 



3. The Method. — The hot junctions of the two couples were 

 fused into a single globule before the oxyhydrogen blowpipe. 

 The wires were left bare for about 5 centim. above this, and 

 the remaining lengths of the wires surrounded by a pair of 

 fire-clay insulators. Thus they could be secured by a clamp 

 on the outside of the furnace. The cold junctions were sub- 

 merged, side by side, in a bath of petroleum, at known tem- 

 peratures (thermometer) . Here they communicated with the 

 terminals leading to the apparatus for the measurement of 

 electromotive force by the zero method. 



In order to insure identity of exposure and slow changes 

 of temperature, the common globular junction of the two 

 thermocouples was imbedded in a mass of calcined lime 

 contained in a large platinum crucible (6 centim. high, 

 5 centim. in diameter). At the close of my work I found that 

 large holes had been fused through the walls of the crucible. 



4. Results. — Table I. contains a series of results obtained 

 as follows : — The furnace (with all parts in place) was first 

 heated to above 1600°. Then the burner was removed, and 

 the furnace closed up and allowed to cool without further 

 interference. 



Observations of the electromotive forces of the two couples 

 were made alternately, every two minutes, during the time of 

 cooling. The electromotive forces of the platinum-rhodium 

 couple were then plotted graphically as a function of time 

 (showing a remarkably uniform curve), from which the elec- 

 tromotive forces (platinum-rhodium), for the times at which 



* Communicated bv the Author. 

 t This Magazine, July L892, § 2. 



