504 Prof. E. Edlund on the Thermal Phenomena of the 



accounted for in a simple manner according to the mode of 

 representation No. 2, without it being necessary, as in No.^ 1, 

 to take to our aid causes the magnitude of the actions of which 

 cannot be directly determined, and which, moreover, closer 

 examination shows, lead to results that are incompatible with 

 one another. We might have carried out several more calcu- 

 lations and comparisons of this sort, if only the treatises which 

 communicate the results of the experiments instituted with 

 the view in question had contained the necessary data ; but 

 what we have given from older experiments may be sufficient 

 for the present purpose. 



7. By the following experiment which I made, an unequi- 

 vocal proof is obtained that heat is developed when the current 

 passes through an electromotor in the opposite direction to 

 the acting electromotive force. 



Two equal test-tubes of glass (A and B) were filled with 

 the same volume of water containing sulphuric acid ; and in 

 each of these tubes two platinum wires and a thermometer 

 were immersed. All four wires were cut off from the same 

 piece ; and the two thermometers were also alike. The wires 

 and the thermometers passed through the corks with which 

 the tubes were closed, and were thereby kept undisturbed in 

 position during the experiment. One of the wires in the tube 

 A was connected with one pole of a Bunsen series of from 

 four to six elements ; the other wire in the same tube was put 

 into connexion with a commutator which I have described in 

 a previous memoir. The principal constituent of this com- 

 mutator consists of a circular disk, upon the periphery of 

 which two brass springs slide. When one of these springs is 

 connected to one end, and the other spring to the other end 

 of a conductor, the current in this circuit is incessantly 

 changing its direction as soon as the disk is set in rotation *. 

 A conducting-wire united one slide-spring of the commutator 

 to one of the platinum wires in the tube B ; from the other 

 platinum wire in the same tube w T ent another conducting-wire 

 to the other slide-spring ; and, lastly, the other pole of the 

 series was connected with the commutator. If now the disk 

 of the commutator was put in rotation, the currents which 

 went through the liquid in tube B incessantly changed their 

 direction. When the disk made one rotation in a second, the 

 number of current-changes was 24 in the same time. The 

 same currents went also through tube A ; but here they all 

 had the same direction. The currents which traversed the 

 two tubes were therefore in number, intensity, and duration 

 perfectly equal ; the sole difference consisted in this — that in 

 * Pogg. Ann. vol. clvi. p. 251. 



