446 H. A. Rowland — Electrical Measurement. 



6. This assumption can be obviated by adjusting 6 = 6' when 

 we have 



1 bap _ (r + H'')(rJl"-rIl t ") 



These can be further simplified by making R" = R/ 7 . 



The method thus becomes very easy to apply and capable of 

 considerable accuracy. As the absolute determination depends 

 on the current period, however, no great accuracy can be ex- 

 pected for absolute values except where this period is known 

 and constant, a condition almost impossible to be obtained. 

 The comparison of condensers or of inductances is, however, 

 independent of the period and can be carried out, however 

 variable the period, by means of a key to make the change 

 instantaneously. 



Method 26. 



Similar results can be obtained by putting the condenser 

 or inductance in R" instead of r, but the current through the 

 electrodynamometer suspension is usually too great in this case 

 unless r is enormous. We have in this case for equal deflec- 

 tions, 



VC" 



orW J "-=.R"(B"+r) ( rR '"-£?" ) 



Where ?\ and R/' are the resistances without condenser or 

 self induction. 



This is a very good method in many respects. 



For using 25 and 26, a key to make instantaneous change of 

 connections is almost necessary. 



To measure resistance by alternating currents, a Wheat- 

 stone bridge is often used with a telephone. 



I propose to increase the sensitiveness of the method by 

 using my method of passing a strong current through the 

 fixed coils of an electrodynamometer while the weaker testing 

 current goes through the suspended system. 



Using non-inductive resistances, methods 10, 13 A, B, C, 

 and 14 all reduce to proper ones. 10 or 14 is specially good 

 and I have no doubt will be of great value for liquid resist- 

 ances. The liquid resistances must, however, be properly de- 

 signed to avoid polarization errors. The increase of accuracy 

 over using the electrodynamometer in the usual manner is of 

 the order of magnitude of 1000 times. 



Since writing the above I have tried some of the methods, 

 especially 6 and 12, with much satisfaction. By the method 



