MAGNETO-ELECTRICAL HEAT. 6 1 



not generation of heat, takes place in the voltaic ap- 

 paratus, the simple conducting parts evolving that which 

 was previously latent in the battery." In the generation of 

 magneto-electricity there is no chemical union to supply the 

 heat, nor any change in the nature of parts of the machine, 

 so that if more heat is evolved in one part than is abstracted 

 from another this heat has for its only source the effect of 

 the magneto - electric current. But in order to ascertain if 

 this were the case it was necessary to take account of the 

 changes of temperature of each part of the machine. This 

 was the investigation Joule now undertook with his mind 

 perfectly open. " I resolved," he says, " to clear up the 

 uncertainty with respect to magneto-electrical heat." 



The research was one of extreme difficulty, and as an 

 experimental feat probably surpasses anything ever 

 accomplished. Joule confined his attention to the action 

 of the magneto-electric current caused in the coil of a 

 revolving armature, or " small compound electro-magnet 

 enclosed, coil and all, in a revolving vessel of water." From 

 the vessel the ends of the coil protruded, and enabled him, 

 by means of a mercury commutator, to include a galvano- 

 meter of small resistance in the circuit. He then measured 

 the current of magneto-electricity generated and the heat 

 developed in the coil — as shown by the rise of temperature 

 of the water — when the armature revolved between 

 the poles of a powerful electro - magnet. He took 

 elaborate precautions to prevent loss of heat by radiation 

 and air currents ; notwithstanding which, he says, " I was 

 led into error by my first experiment ; the water had lost 

 heat even when the temperature of the room was such as 

 led me to anticipate a contrary result." " I did not stop," 

 he continues, " to inquire into the cause of the anomaly, but 



