On the Cause, of the Phenomena of Voltaic Cooling and Heating. 263 



becomes dissociated, and the carbon set free then gives off the 

 ordinary carbon-spectrum . 



In conclusion, my best thanks are due to Professor Roscoe 

 for valuable advice and assistance rendered me in this investi- 

 gation. 



XXXI. On the Cause of the Phenomena of Voltaic Cooling and 

 Heating discovered by Peltier. By E. Edltjnd*. 



F a voltaic current passes through a metal conductor, heat is 

 developed, and its quantity is proportional to the resistance 

 and the square of the intensity. An exception to this rule, how- 

 ever, is formed by the place of junction of two heterogeneous 

 metals. Peltier showed, so long ago as 1834f, that the solder- 

 ings between two different metals become either colder or warmer 

 than the other parts of the conductor, according to the direction 

 in which the current traverses the places of contact. Peltier 

 found that the strongest action was that between bismuth and 

 antimony. If the current passed through the junction from the 

 bismuth to the antimony there was a fall of temperature, while 

 in the opposite case there was an increase. These experiments 

 were confirmed by Moserj. Lenz subsequently§ gave this 

 experiment an attractive form by showing that at the place of 

 contact between bismuth and antimony water can be made to 

 freeze if a feeble current passes from the former to the latter 

 metal and both have been previously cooled in a mixture of ice 

 and water. 



Peltier was led by his experiments to the view that these phe- 

 nomena of cooling and heating are closely connected with the 

 electrical conductivity of the metals. When the current passes 

 from a worse to a better conductor, in his opinion the tempera- 

 ture at the soldering is higher than when the current goes in 

 the opposite direction. E. Becquerel, however, has shown || 

 that this is not always the case, and that therefore the voltaic 

 resistance is of no importance from this point of view. He 

 made special experiments to ascertain whether at the point of 

 contact the voltaic resistance was in any manner dependent on 

 the direction of the current, so that in one case it should be 

 greater and in another smaller than in the other parts of the 

 circuit. But the experiments gave a negative result ; the ob- 

 served differences in the resistance, according to the direction of 



* Translated from Poggendorff's Annalen, having been read before the 

 Swedish Academy of Sciences at Stockholm, April 14, 1869. 

 t Ann. de Chim. et de Phys. vol. lvi. p. 371. 

 % Repertorium der Physik, vol. 1. p. 349. 

 § Pogg. Ann. vol. xliv. p. 342. 

 || Ann. de Chim. et de Phys. S. 3. vol. xx. p. 53. 



