Prof. E. Edlund's Investigation of the Electric Light. 105 



heat depends, have not been altered. In a former investigation* 

 I have experimentally proved that in the present case the heat 

 produced in the primary circuit is in fact less ; and in all cases 

 the sum of the heat produced by the principal current, and of 

 that developed in the path of the induction-current, is equal to 

 the heat which would be produced by the principal current 

 alone if there were no induction. But as the heat which the 

 induction-current developes is justly regarded as the thermal 

 equivalent of this current, it follows that the heat produced in 

 the primary current during induction has been diminished by 

 the quantity of heat which corresponds to the work which the 

 principal current consumes in producing induction. As I have 

 shown in the above-mentioned research, this diminution in the 

 production of heat in the principal current is produced by the 

 circumstance that the induction-current developes induction- 

 currents of higher order in the principal conduction, which add 

 to or subtract from the principal current in such a manner that 

 the heat produced is always diminished. Thus in this case the 

 quantity of heat produced would be diminished by means of the 

 induction- current. We assume, finally, that the conducting-wire 

 between the poles of the battery is connected with a coil of 

 covered wire, and that in this coil there is a soft-iron cylinder 

 whose keeper is removed from the iron cylinder as soon as the 

 magnetism developed is too feeble to attract it to the poles of 

 the electromagnet. Now if the current be opened and closed at 

 definite intervals during a certain time, a definite quantity of 

 work is produced. If, then, the keeper while it is being attracted 

 towards the magnet be caused to perform mechanical work (by 

 raising a weight for instance), it is clear that the heat produced 

 will be diminished by an amount corresponding to the work 

 performed, which, indeed, Eavre has experimentally proved f is 

 actually the case. On the other hand, neither the resistance in 

 the battery nor the electromotive force, the points on which the 

 heat developed depends, has been changed. Yet it may easily 

 be seen that the different behaviour of the induction-currents 

 in the two cases accounts for this discrepancy ; for part of the heat 

 developed is due to the induction- currents produced in the 

 electromagnet and its keeper by the motion of the keeper. The 

 quantity of electricity here set in motion is indeed the same 

 whether the keeper is loaded or not ; but if the keeper is free, it 

 moves quicker than if it performs work, and therefore the in- 

 duction-currents in question have in the first case a shorter 

 duration than in the latter. The same quantity of electricity 

 performs its circulation in a shorter time when the keeper is free 



* Phil. Mag. S. 4. vol. xxxi. p. 253. 

 f Comptes Rendus, vol. xlv. p. 56, 



