104 On Measuring the Electrical Capacity of a Condenser. 



Resistance observations: — 



a=10; d=1000; c = 232'5. 



Hence * 



C 12 8 = *3301 ■ w -. r rr microfarad. 



±5. A. unit 



The sensitiveness throughout these observations was such 



that an alteration of 1 in 2000 in the resistance produced a 



deflection of about one scale-division in the spot of light. 



Hence the greatest difference in the observations amounts 



to about 5 in 3300, or about 1 in 660 ; and there is no 



trace visible of variation in the capacity with the time of 



charging. 



We will collect the numbers together: — 



C 16 -3300 



C 32 -3304 



C 64 -3299 



C 64 (second arrangement) *3299 



C 128 . -3301 



Mean value of capacity = '33006 ^5— r-yy- microfarad. 



Taking the Cavendish-Laboratory experiments, we have 

 1 B.A. U. = -9867 ohm. 

 n -33006 . n , 

 '"" ~ 7 M67 microfarad 



b= *3344 microfarad. 

 If we take the legal ohm and the value of the specific 

 resistance of mercury in B.A. units, adopted by the B.A. 

 Committee, 



1 B.A. unit =-9889 ohm, 



and C = '3336 microfarad. 



This value supposes that the various coils used have their 

 nominal resistance in B.A. units. And this assumption is 

 correct, at any rate to 1 in 1000. Thus it seems that the 

 method gives satisfactory results, and may safely be used to 

 determine the capacity of a condenser. 



But the converse of the method may be even more useful. 

 The fundamental equation gives us C if we know n and the 

 resistances; it will equally give us n if we know C and the 

 resistances; and in many cases this may be the readiest 

 method of rinding ft, especially if the period is too long to 

 give an audible note. Thus in the above experiments I require 

 to adjust a fork to vibrate about 16 times a second, the lowest 

 I could obtain being one of a frequency of almost 20. 



