129 



continuous type, seemed to him might be 

 made more useful if they were more 

 separated. The constant K appeared to 

 contain three things. First, it contained 

 a multiplier of 2, introduced because 

 alternate-current machines generally had 

 only a single circuit, riot two circuits, but 

 this was not always the case, and the 

 multiplier 2 should be put separately. 

 Secondly, the constant included a quan- 

 tity depending on the shape of the field 

 and the arrangement of the armature; 

 and thirdly, the quantity depending on 

 the ratio of the square root of the mean 

 square to the mean. All those three 

 quantities appeared to be mixed up in 

 one number, which might be useful in 

 the particular cases given, but he thought 

 that if they were dealt with as he sug- 

 gested they would be more useful to 

 subsequent designers. Nearly all cases 

 which the author had treated so lucidly 

 as to connecting machines in parallel, 

 had been by the graphic method, which 

 depended entirely upon the assumption 

 that the electromotive force followed a 



