MEASUREMENT OF THE PELTIER EFFECT. 439 



heat, and that the energy of the current IH is equal to the algebraical 

 sum of these works. 



The measurement of the heat disengaged on the electrode is very 

 difficult, owing to its dissemination by conductivity and convection in 

 the surrounding liquid. Bouty got over this difficulty very happily 

 by taking as electrode the thermometer itself. In order to graduate 

 the thermometer in thermal units, he winds round the bulb a care- 

 fully insulated wire, and, the apparatus being immersed in the liquid, 

 he passes a current of known strength through the wire ; the ascent 

 of the column, corresponding to a known quantity of heat disengaged 

 in each second at the surface of the bulb, is thus measured. The 

 spiral being removed, the surface of the bulb is silvered, and a thin 

 layer of copper, for instance, is deposited on it by electrolysis. The 

 thermometer may then be used with another similar thermometer as 

 electrodes in a copper bath. 



At the contact of a copper electrode, for example, with a solution 

 of copper sulphate, a disengagement of heat is observed if the 

 electrode is positive, and a cooling if it is negative. As these 

 quantities of heat are proportional to the intensity of the current, 

 the phenomenon has all the characteristics of a Peltier effect. 

 Bouty* has observed that, with such metals as copper, zinc, and 

 cadmium, the effect is independent of the nature of the acid, and of 

 the degree of concentration of the solution, provided they are not 

 too dilute. For copper, the quantity of heat per coulomb is 0-05078, 

 which would represent a difference of potential of 0*212 volt; zinc 

 gives 0-241 volt. These numbers greatly exceed those which are 

 found by the same method for the contact of metals with each 

 other. 



The positive electrode is the seat of the positive chemical work. 

 The formation of an equivalent of hydrated cupric oxide disengages 

 19,000 thermal units, and the combination of this oxide with dilute 

 sulphuric acid 9,200, or 28,200 in all. The quantity of heat pro- 

 duced at the electrode by chemical action is thus 0-292 for a 

 coulomb, which corresponds to a difference of potential of i'2i7 

 volt. The heat disengaged is the excess of the chemical energy 

 over the heat absorbed by the rise of the electrical level. The 

 excess of potential of the solution over that of the metal, or the 

 electromotive force of contact, is thus 



H = i "2 17-0 "2 12 = 1 -005 volt. 

 * BOUTY. Journal de Physique [i], Vol. IX., p. 229. 1880. 



