44 2 MEASUREMENT OF ELECTROMOTIVE FORCES. 



For the couple iron-zinc, one of the junctions of which is at zero, 

 the electromotive force, expressed as a function of the temperature / 

 of the hot junction, is 



E = 9i777/- 1-9488/2. 



From this is deduced, for the value of the Peltier effect at 13 '8, the 

 number 0-00235 v lt instead of the number 0*00253 volt, which 

 had been obtained directly (1032). 



For thermoelectric couples metal -liquid, working between the 

 temperatures t and /', Bouty found that the electromotive force is 

 represented by an expression of the first degree 



in which the parameter only depends on the nature of the metal ; it 

 follows that the Peltier effect is proportional to the absolute tem- 

 perature. With copper, taking the EMF of a Daniell's element at 

 i -i volt, experiment gives 



m 0^00078 volt. 



From this is deduced 0*2 1 8 volt for the Peltier effect at 12; 

 direct measurement gave 0*212 volt. 



1036. THERMOELECTRIC DIAGRAM. The quantity ( j has 

 been called by Sir W. Thomson the thermoelectric power of the 

 two metals at the temperature T. 



Clausius gave the name entropy to a function such that the 

 quantity of heat relative to an infinitely small transformation of a 

 body is equal to the product of the corresponding absolute tem- 

 perature by the variations of entropy. The thermoelectric power of 

 the two metals presents a property analogous to the function of 

 Clausius, and Maxwell* has proposed to call it electric entropy. 



We have seen that, taking lead as standard of comparison, its 

 specific heat of electricity being null (284), the curves which 

 represent the thermoelectric power as a function of the temperature 

 are, for most metals, straight lines, the angular coefficient of which 

 represents the specific heat of electricity. The only exceptions 



* MAXWELL. Element. Treatise on Elect., p. 137. 



