GENERATORE SECONDARIO (iAfLARD E GIBBS 753 



done at that instant l)etween tbe points I) and C. Hence it 

 follows that the deflection of the rieedle multiplied (*) by the 

 resistance r is a correct measure of tlie work done in unit time 

 between i) and C. 



« After this explanation I think you will agree with me 

 that the method I adopted was correct and that I bave net 

 made the error you not unnaturally supposed I had made. 



« Owing to the considerations which you point out, the effi- 

 ciencies ]\1. Uzel deduces from bis experiments are ali somewhat 

 lower than those experiments really indicate. ]\ry own conclusions, 

 viz. , an efficiency of 79.3 per cent for the old type of 

 instrument and 80.1 per cent for the new type are the correct 

 deductions from the observations which I made. My use of the 

 electro-dynamometer in the case in which I used it for verification 

 was also strictly correct for the self-induction of the portion 

 of the circuit in which I proposed to measure the work done 

 was practically niì. 



« I send herewith a copy of my lecture before the Insti- 

 tution of Civil Engineers and would ask you to refer to the 

 remarks on page 1 1 , also a copy of a recent paper of mine on 

 alternate currents in which the subject of secondary generators 

 is touched upon. From these I think you will see that I was 

 hardly likely to make the fundamental error which you bave 

 attributed to me 



« I am sending copies of this letter to the Englisb Journals 

 in which abstracts of your paper bave appeared. 



« I am Sir 



« Yonr rrspectfuììy 



« J. HOPKINSON. » 



(•) Si voleva dire » divided ». 



