8 On some Points in Electrolysis, 



Even if the Yolta effect is due to the presence of traces of 

 an electrolyte, such as water, and not to a non-electrolytic 

 combination with free oxygen, the " seat of the electromotive 

 force " still depends on a mere definition. The term " electro- 

 motive force " is continually used in two senses. It is some- 

 times used to denote the difference between the potentials of two 

 points, and sometimes to denote the rate of fall of potential. 

 Maxwell gives it the latter definition formally, but frequently 

 writes of the electromotive force when he means the line- 

 integral of the electromotive force, that is to say the difference 

 of potential. In connexion with cells, u electromotive force " 

 is always used to mean difference of potential, so the stricter 

 meaning, rate of fall of potential, may be disregarded just now. 

 If we define the difference of potential between two points as 

 proportional to the work done on a unit quantity of electricity 

 moving from one point to the other, we have a clear statement 

 of what we mean. But if we attempt to make the measure- 

 ment, we find the unit quantity of electricity must have some 

 carrier ; and when we use an electrometer, we do not measure 

 the work done on a unit quantity of electricity moving from 

 one point to the other ; we really measure the work done 

 when a particular conductor is used as carrier. It is not 

 necessary to consider any part of the circuit as the seat of the 

 electromotive force. There is a circuit, and the potential is 

 cyclic. Similar cases occur in other branches of electricity. 

 For instance, a "unipolar" dynamo may be made up of a 

 rotating magnet with a stationary circuit. If the resistance of 

 the internal circuit is negligible in comparison with that of the 

 external, the whole expenditure of power is in the external 

 circuit, and the fall of potential over the external resistance is 

 sensibly the whole potential or electromotive force of the 

 machine. There is, however, no way of finding the seat of 

 this electromotive force. If the rotating magnet is sup- 

 posed to carry "lines of induction " round with it, they cut 

 the external circuit and produce electromotive force there, 

 so that becomes the seat of the electromotive force. If the 

 lines of induction are taken as stationary, the rotating part of 

 the circuit cuts them and becomes the seat of the electromotive 

 force. The effects on the external resistance are the same in 

 both cases. If it is a voltmeter the readings are the same, 

 and they remain so if the lines of induction rotate faster or 

 slower than the magnet, or in the opposite direction. If the 

 voltmeter is replaced by an electrometer, it is still impossible 

 to say where the seat of the electromotive force is. The 

 readings would be the same whether the lines of induction 

 were stationary, or revolved and cut the leads to the elec- 



