354 Mr. J. Parker on Thermoelectric Phenomena. 



Take, then, two plates, A, B, of two different metals, say 

 oopper and iron, and connect them by a wire of either copper 

 or iron. Parallel and close to these two plates place equal 

 plates of any metal whatever, as zinc, and connect the zinc 

 plates by long zinc wires with a large distant mass of the same 

 metal in the neutral state, so that the zinc plates are constantly 

 at the zero of potential. 



Let the mass of the metal A be M a , its specific heat at con- 

 stant pressure c«, its charge E a , and its potential V«. Let 

 Mp, cp, Ep, Yp denote similar quantities for the metal B at the 

 same absolute temperature t\ Also put V^— V a =<5. 

 >. By a theorem, due originally to Helmholtz, but lately ob- 

 tained by Duhem as a strict consequence of thermodynamical 



£ B 



CL. 



principles, the energy of the electrification of the system may 

 be written 



iE a .V a + JE^.V^+E a .H a + E^H^. 



As we are only going to consider the case of constant pressure, 

 this may be written 



iE a . V„+iE„ . V e + E« ./„(«) + E„ .Mt). 



Now let our apparatus be made to undergo the following 

 cycle of reversible operations, 



(1) By slowly bringing the plates B nearer together, and 

 slowly separating the plates A, let a quantity of electricity q 

 be made to pass from A to B against the rise of potential 8 

 without altering the potential of either A or B. The work 

 done on the system will be 



The only thermal effect produced will be at the junction, and 

 the heat absorbed there may be represented by II . q in mecha- 

 nical units. 

 Hence 



n-*8=KV„-v.) +/*-/,, 



or 



n=s+>-/. (i) 



