438 FKAGMENTS OF SCIENCE. 



passage when the number of intervals is increased. 

 Thus, by augmenting the convolutions of our machines 

 we sacrifice quantity and gain electro-motive force; 

 while by lessening the number of the convolutions, 

 we sacrifice electro-motive force and gain quantity. 

 Whether we ought to choose the one form of machine 

 or the other depends entirely upon the external work 

 the machine has to perform. If the object be to obtain 

 a single light of great splendour, machines of low re- 

 sistance and large quantity must be employed. If we 

 want to obtain in the same circuit several lights of 

 moderate intensity, machines of high internal resistance 

 and of correspondingly high electro-motive power must 

 be invoked. 



When a coil of covered wire surrounds a bar of iron, 

 the two ends of the coil being connected together, every 

 alteration of the magnetism of the bar is accompanied 

 by the development of an induced current in the coil. 

 The current is only excited during the period of mag- 

 netic change. No matter how strong or how weak the 

 magnetism of the bar may be, as long as its condition 

 remains permanent no current is developed. Conceive, 

 then, the pole of a magnet placed near one end of 

 the bar to be moved along it towards the other end. 

 During the time of the pole's motion there will be an 

 incessant change in the magnetism of the bar, and 

 accompanying this change we shall have an induced 

 current in the surrounding coil. If, instead of moving 

 the magnet, we move the bar and its surrounding coil 

 past the magnetic pole, a similar alteration of the mag- 

 netism of the bar will occur, and a similar current will 

 be induced in the coil. You have here the fundamental 

 conception which led M. Gramme to the construction 

 of his beautiful machine. 1 He aimed at giving con- 



1 'Comptes Kendus,' 1871, p. 17G. See also Gaugain on the 

 Gramme machine, 'Ann. de Chem. et de Phys.,' vol. xxviii. p. 324. 



