264 Prof. E. Edlund on the Cause of the Phenomena 



the current, were not more than might be assumed to arise 

 from the differences in temperature at the points of contact. 

 The experiments thus did not at all prove that the cooling 

 and heating observed by Peltier had anything to do with the 

 voltaic conducting-power. It is also clear that if the voltaic re- 

 sistance were indeed different with the direction in which the 

 current traversed the point of contact, it would certainly follow 

 that the degree of heating might vary with the direction of the 

 current, but in no case could there be a cooling or real absorp- 

 tion of heat. Becquerel was of opinion, however, that these 

 experiments indicated another connexion between the pheno- 

 mena in question and previously well-known voltaic phenomena ; 

 for he found that when the voltaic current which traverses the 

 place of contact has the same direction as the thermoelectric 

 current which would be formed by heating the junction, 

 the temperature diminishes at the place of contact, but that 

 when the current is in the opposite direction there is a rise in 

 temperature. The phenomena in question would thus have a 

 connexion with the thermoelectric properties of bodies. How 

 far this conclusion is right or not under all circumstances can 

 only be definitely settled when a larger number of metals and 

 alloys have been investigated with this view. 



G. v. Quintus Icilius has made careful examination of the 

 quantitative relations of these phenomena, from which it resulted 

 that the difference in temperature produced by the current at 

 the junctions of a thermoelectric pile of bismuth and antimony 

 was proportional to the intensity of the current. Hence these 

 phenomena follow a totally different law from that of the ordi- 

 nary thermal action of the voltaic current; for while the former 

 are simply proportional to the intensity ; the latter thermal action 

 is proportional to the square of the intensity. The accuracy of 

 this result has been confirmed by Frankenheinr's investigation*", 

 which was made in a manner totally different from the above. 

 Hence it maybe regarded as demonstrated that the variations in 

 temperature in question are proportional to the intensity of the 

 current by which they are caused. 



It is in itself a very remarkable fact that under certain cir- 

 cumstances the voltaic current can produce an absorption of heat ; 

 for its ordinary action is to produce heat. Hence I have thought 

 that an account of the cause of this deportment might have 

 some interest; for, as will afterwards be shown, Peltier's phe- 

 nomena of cooling and heating may be easily deduced from the 

 idea of electromotive force; their existence may be proved to 

 be absolutely necessary ; so that they might have been disco- 

 vered a priori if their existence had not previously been demon- 

 * Pogg. Ann. vol. xci. p. 161. 



