372 Mr Wilson, On the Measurement of the Earth-Air 



20 mm. of the scale. The first six observations* were made by 

 the first method described above, the rest by the second method. 

 The compensator was graduated in mm., and the tenths of a mm. 

 were estimated. 



Charge 



Compensator 

 reading 



Charge 



Compensator 

 reading 







30-2 



24 



15-7 



1 



26-1 



30 



135 



2 



24-6 



3G 



11-4 



3 



23-8 



42 



9-3 



4 



23-2 



48 



7-2 



5 



22-7 



54 



5-0 



6 



22-2 



60 



2-9 







30-2 



66 



0-8 



6 



22-2 







30-2 



12 



20-0 



6 



22-2 



18 



17-9 



1 



26-1 



The charge on the test-plate when at zero potential and with 

 the condenser plate at a potential of two volts, the unit of charge 

 in the above table, is given sufficiently nearly by 



Q= 



+ ri V 

 ~2T~ X 45' 



where r 1} r 2 are the radii of the test-plate and the aperture 

 surrounding it, V is the potential difference between the plates 

 in electrostatic measure and d is their distance apart. Inserting 

 the actual values r 1 = 2'o, r 2 = 3, V = 2/300, d = 1*28, we obtain 

 Q = 0'99 x 10~' 2 electrostatic unit. We have thus the data for 

 measuring the charge held on the compensator rod, kept at zero 

 potential, for any given compensator reading, i.e. so long as the 

 potential of the compensator tube remained the same. If this 

 potential were altered the factor by which the ordinates of the 

 calibration curve had to be multiplied to give the charge in 

 electrostatic units would also be different. The potential of the 

 compensator tube in the observations of the above table was such 

 that with the ordinary cover over the test-plate the compensator 

 had to be drawn out through 63 divisions to cause a movement of 

 the gold-leaf through 30 scale divisions, the compensator being 

 pushed in nearly to its full extent before the displacement. Thus 

 the same condition of the compensator could be recovered at any 



* In the curve some additional points have been inserted corresponding to 

 observations made with the aid of a potentiometer. 



