KING. — MEASUREMENT OF THOMSON EFFECT IN COPPER. 361 



change of temperature was calculated. The standard low resistance was 

 included, with the experimental wire, between the terminals of the poten- 

 tiometer, as shown in Fig. 4. 



The galvanometer was balanced, see Fig. 4, with the terminal ffi con- 

 nected with 1, 2, 3, 4, 5, etc., successively, by adjusting the position of 

 the contact y., on the potentiometer. The relation of the differences 

 between the potentiometer readings, corresponding to 3, 4, 5, G, etc., to 



i 



STANDARD 

 LOW RCS1STAMCC 



JTBT 



nl 



B. 



t £ 34 5 e 7 a 9 to It It cS 



V 



CYLINDER POTENTIOMETER 



Figure 4. 



the difference between the readings for 1 and 2, expresses the relation 

 of the resistances of sections 3 to 4, 4 to 5, 5 to 6, etc., to the resistance 

 of the standard S. 



Determination of the h Curve,/ g (Fig. 3). 



If the experimental wire a i^ be long enough, the curve of distribution 

 of temperature will be flat for the part corresponding to the middle of the 

 bar. That is, the middle section will not be losing heat by conduction to 

 the neighboring parts of the bar. Therefore the heat generated by the 

 Joule Effect in the middle section will be all got rid of by radiation and 

 convection. By measuring the resistance of the middle section and the 

 electric current in amperes, under the above conditions, the heat generated, 

 as well as the temperature, for the section can be calculated. By using 

 several values for the electric current, different temperatures are obtained 

 for the middle of the bar and corresponding values for the heat dissipated 

 by radiation and convection. This gives all the data required for the 

 determination of the h curve, provided we assume that the condition of 

 the surface of the bar is uniform and that all parts of the bar are affected 

 in the same way by convection currents. 



In the experiments described in this paper the experimental bar was 

 too short to avoid the effect of metallic conduction on the middle section. 



