60 



[American Journal of Science [4], IV, 429-448, 1897 ; Philosophical Magazine [5], XL V, 



66-85, 1898] 



The electrical quantities pertaining to an electric current which it 

 is usually necessary to measure, outside of current, electromotive force, 

 watts, etc., are resistances, self and mutual inductances and capacities. 

 I propose to treat of the measurement of alternating currents, electro- 

 motive force and watts in a separate paper. Eesistances are ordinarily 

 best dealt with by continuous currents, except liquid resistances. I 

 propose to treat in this paper, however, mainly of inductances, self and 

 mutual, and of capacities together with their ratios and values in abso- 

 lute measure as obtained by alternating currents. I also give a few 

 methods of resistance measurement more accurate than usually given 

 by means of telephones or electrodynamometers as usually used and 

 specially suitable for resistances of electrolytic liquids. 



I have introduced many new and some old methods, depending upon 

 making the whole current through a given branch circuit equal to zero. 

 These always require two adjustments and they must often be made 

 simultaneously. However, some of them admit of the adjustments 

 being made independently of each other, and these, of course, are the 

 most convenient. But all these zero methods do not admit of any 

 great accuracy unless very heavy currents are passed through the 

 resistances. The reason of this is that an electrodynamometer cannot 

 be made nearly as sensitive for small currents as a magnetic galvano- 

 meter. The deflection of an electrodynamometer is as the square of 

 the current. To make it doubly sensitive requires double the number 

 of turns in both the coils. Hence we quickly reach a limit of sensitive- 

 ness. It is easy to measure an alternating current of -0001 ampere and 

 difficult for -00001 ampere. A telephone is more sensitive and an 

 instrument made by suspending a piece of soft iron at an angle of 45, 

 as invented by Lord Eayleigh, is also probably more sensitive. 



For this reason I have introduced here many new methods, depend- 

 ing upon adjusting two currents to a phase-difference of 90 which I 

 believe to be a new principle. This I do by passing one current through 



