400 THE ADDRESSES, LECTURES, ETC., OF 



out to the electric light or to the place where the electric effect is 

 to be produced, and the other flowing round the bar-magnets and 

 back again to the machine. If this arrangement is applied to a 

 machine wound for a continuous circuit, it will not produce any 

 useful result, but if it is modified that is to say, if, as I showed 

 in a Paper before the Eoyal Society in 1880, the resistance of the 

 wire on the field-magnets is increased a hundred-fold, then we get 

 a machine capable of very sustained action, and this form of shunt 

 dynamo machine, as it is called, has been since largely adopted in 

 the production of electric light. The greatest uniformity of 

 current is produced, however, in combining the shunt with the 

 old method of winding the field-magnets by furnishing them with 

 two separate coils, the one of low resistance forming part of the 

 outer circuit, and the other of high resistance consisting of thin 

 wire forming a shunt circuit. 



There is another form of dynamo machine which differs essen- 

 tially from those I have as yet described, viz., the alternate 

 current machine. This differs from Holmes's type, chiefly in the 

 substitution of electro-magnets for permanent magnets, although 

 De Meritens in his ingenious modification of Holrnes's machine 

 continues to give the preference to permanent magnets. The 

 modification adopted by my firm consists in substituting mere 

 coils of wire for the armatures, rotating through the magnetic 

 fields produced by electro-magnets which are excited by means of 

 a dynamo machine of the original type. The principal advantages 

 obtained in suppressing the iron armatures, are, that less weight 

 has to be put into rapid motion, and that less energy is converted 

 into heat by eddies set up within the iron. This machine repre- 

 sents in fact a return to Faraday's original demonstration of the 

 current set up in a conductor passed through a magnetic field, 

 and it is interesting to observe by inspection of the Table on 

 page 401, that these machines give the highest yield of electrical 

 energy for a given expenditure of mechanical power. This Table 

 contains a considerable amount of practically useful information. 

 The machines marked D in the first column are of the self- 

 exciting type first described ; those marked S D (S meaning 

 shunt) are wound in the manner of a parallel circuit as next 

 described, and those marked W are the alternating current 

 machines to which I have just referred. The second column gives 



