450 Motion of the Needle of q Quadrant Electrometer. 



series, the other pole being earthed. The quadrant was 

 released from earth, and observations were taken every three 

 seconds, the present table representing the mean results of: 

 three experiments. The effect of the periodic term is clearly 

 shown, the first hump being very marked. The first maximum 

 of the periodic term occurs when 



tan(^-s)=3, 



or on substitution of the constants deduced below, at £ = 4*2. 

 The steady deflexion 6 X corresponding to two cadmium 

 cells was 552'8, so that taking the case where 0/0, =0-25 for 

 which £ = 24-5, we readily calculate 7i = ! 01]7. Since 

 T ^= 14*2 and A, = I'll, the correcting factor (1 — fc) is 

 l-0-0164^0'984, and S = 0-104(2tt), so that according to 

 our rule for taking the observations, we should take the first 

 at t=sl m 5, the second at £==15% and so on. In order to test 

 the theory, let us take the readings corresponding to the 

 above times for the curve A f , multiply bv the correcting 

 factor 0-984, and add on the quantity (0'0164)(552'8). We 

 thus obtain the corrected values 0' given in Table II. 





Tablf, II. 





Time 



t. 



Actual reading 



from curve 



9. 



Corrected 



reading 



0\ 



Calculated 



value of 



9'. 



Vb 



12 



10-3 



10- 1 



15*7 



90 5 



931 



96-8 



29-9 



163«5 



1G9-9 



169-9* 



i 



The values of 6' should satisfy the equation 

 <9' = 552-8(l -<?-*'), 



near the agree- 



and the third and fourth columns show how 

 ment is. The number marked with an asterisk represents 

 the observations used to calculate h, and B is the theoretical 

 curve drawn to correspond to this value of A. The points 

 marked on the curve B represent the corrected readings 

 given in the third column of the table, and it is obvious that 



f The scale of the actual curve used was much larger than that of the 

 reduced reproduction here shown. 



