306 SCIENCE PROGRESS. 



direction to that shown in Fig. 2. In this latter case their 

 effect was to demagnetise the circuit, but now, of course, it 

 must be the reverse. Thus the ampere-turns between inn 

 and op in Fig. 3 have a forward magnetising effect, and 

 increase the number of lines cut by the active inductors : or, 

 in other words, the armature itself magnetises the field. 

 This result sounds, perhaps, more remarkable than it really 

 is. It must, of course, be remembered that when the 

 brushes are placed at no, and the direction of the current 

 is thus the opposite to that shown in Fig. 2 in the direct 

 magnetising turns, any E. M. F. which is generated in these 

 latter is opposed to the direction of the current flowing in 

 them. The distribution of the field remains the same as in 

 Fig. 2 ; as, however, the position of the brushes does not now 

 coincide with the neutral line of zero field but is behind it, 

 if the wires between the neutral line and the diameter of 

 commutation are cutting any lines, they generate a back 

 E. M. F. It is thus evident that no wire can at one and 

 the same time act as a forward magnetisine turn, and also 

 as a useful inductor : the combination of the two functions 

 is, in fact, entirely impossible. The forward magnetising 

 loops, therefore, which Dr. Hopkinson obtained on the 

 armature, in effect were but the translation of some of the 

 usual stationary field-magnet winding on to the rotating 

 armature. The wires themselves, it is true, move with 

 the armature, but their place is continually taken by fresh 

 wires, so that the current-sheets between mu or op remain 

 stationary in space and magnetise the circuit. The account 

 of the experiment ended with the following pregnant 

 sentence : " If we could put up with the sparking that 

 would ensue, it would be possible to make A negative in 

 a generator of electricity, and thereby obtain by the re- 

 actions of the armature itself all the results usually obtained 

 by compound winding ". This leads us to the method of 

 winding armatures, recently invented by Mr. Sayers 

 (Patent Nos. 16,572, 1 891, and 10,298, 1893), an ^ described 

 by him in a paper read before the Institution of Electrical 

 Engineers. 



1 



x Journal, No. 107, vol. xxii., 1893. 



