1871J Heat, and Force. 91 



them in the circuit as to act on the needle in opposition to 

 one another., The piatinum evolved very much more heat 

 than the copper wire, and, indeed, soon became red hot ; but 

 they both acted on the needle with exactly equal forces. So 

 that clearly heat is not necessarily any measure of magnetic 

 force, though, under certain accidental circumstances, it 

 may be so. 



9. The state of things revealed by comparing M. Soret's 

 experiments with M. Favre's is something very remarkable. 

 M. Soret has confirmed (by numerous experiments) M. 

 Jacobi's conclusion that when a magnet is doing actual work 

 it increases the resistance, and consequently diminishes the 

 consumption of zinc; while M. Favre's experiments clearly 

 show that a magnet whilst doing work absorbs less of the 

 total heat of the circuit than whilst it is doing no actual 

 work. These two facts, apparently both thoroughly' well 

 established, seem utterly inconsistent. How are they to be 

 explained ? Is it that M. Favre's magnets by working 

 produced cold, and so diminished the calories shown in their 

 own calorimeter, whilst, at the same time, by the repeated 

 approach of the magnetised soft iron armatures to the 

 magnets they produced counter currents of electricity, 

 and so produced heat in the circuit, of which the battery, by 

 its higher resistance, took the lion's share, and consequently 

 exhibited in its own calorimeter ? At any rate, I think it 

 is quite plain that we have not yet got to anything like the 

 bottom of the subject, and that our present theories cannot 

 account for the facts revealed. 



10. Take another case in which we make a battery do an 

 unlimited amount of work. Put in the circuit a cell or 

 voltameter, having no electro-motive power of its own, such 

 as a solution of nitrate of silver with silver poles. Then, 

 for every equivalent of zinc consumed, an equivalent of 

 silver will be carried from one pole to the other. Now put 

 two such cells in the circuit, and then every equivalent of 

 zinc will convey two equivalents of silver the same distance ; 

 and by repeating the process we can make an equivalent of 

 zinc move any weight of silver a certain distance. 



It may be said we lose time : Yes ; but the- only time we 

 lose is the time which it takes for the electric force in the 

 first instance to traverse the circuit. The current once 

 established, equal amounts of silver are conveyed an equal 

 distance in equal times, by a consumption of zinc which may 

 be diminished to any extent. Does not this prove, then, 

 that we may make a certain amount of zinc do any amount 

 of work, with this only condition, that it takes so much 



