252 Prof. E. Taylor Jones on a 



to have rather more effect on the period in this case. The 

 resistance of the tube was, however, probably underestimated. 



A cubic similar to (4) can be applied, neglecting the tube, 

 to the case o£ the two coils A and B, the latter being closed. 

 This leads to (2) if the resistance of B is negligible. The 

 solution of the cubic shows that the resistance of B has only 

 an extremely small effect on the period of the oscillations. 



In the Table given below the calculated periods are those 

 given by the formulee (1), (2), or (3), the effect of the brass 

 tube being neglected. 



(10) Results. 



The results are given in the following Table, in which the 

 numbers in the third column give the capacity in microfarads 

 of the leyden-jar and electrometer corrected for the capacity 

 of the coil. The primary coil was in all cases the coil A, of 

 self-inductance 70*15 x 10 9 cm. 



Condenser. 



Secondary 

 Circuit. 



Capacity. 



Calculated 

 period. 

 Seconds. 



Observed 

 period. 

 Seconds. 



II 



None. 



•002002 

 •005092 



•002356 

 003762 



•002353 

 •003760 



IV 





I 



Ill 



B open. 



•001065 

 •002390 

 •005094 



•001718 

 •002575 

 •003762 



•001716 

 •002534 

 •003760 



IV 





I 



B closed. 



•001065 

 •002004 



•001603 

 •002200 



•001610 

 •002211 



II 





n { 



n { 



B with 1-83 

 microfarads 

 B with 3-83 

 microfarads 



j -002006 

 j -002006 



•002383 

 •002418 



•002381 

 •002410 



With the exception of the case of condenser III., which is 

 discussed in section 5 above, the calculated and observed 

 periods agree on the average to about one-fifth per cent. In 

 the simpler cases, where the secondary coil was either open 

 or absent, the agreement is closer. On the whole, therefore, 

 the method may be said to give reliable measurements of the 

 frequency of electrical oscillations up to 600 per second. 



