424 MR, J. E. H. GORDON ON MEASUREMENTS OF ELECTRICAL CONSTANTS. 
deflection would have been in the opposite direction. It is clear that if we insert a 
dielectric, and also move a away from b, we can, by properly adjusting the motion, 
keep the electrometer at zero. 
Fig. 4. 
The amount of motion of a required in each case, will depend on the thickness of 
the dielectric, and on its specific inductive capacity. 
Let us suppose the balance not perfectly established, and a deflection, say, to the 
right—now reverse the poles (we are still supposed to he working with a battery); the 
deflection will be to the left, and with the electrifications which are actually used, and 
which are reversed 12,000 times in each second, no disturbance of equilibrium would 
produce any deflection if the needle were charged in the ordinary way. 
To escape from this difficulty Professor Maxwell devised the following arrange¬ 
ment The needle, instead of being permanently charged, was connected with the 
plate c, and every time the electrification of the plates was reversed, that of the needle 
was reversed also. As the sign of the force between two electrified bodies depends only 
on whether the electrifications are like or unlike, and not on whether they are positive 
or negative, this arrangement caused the force in the needle to depend only on the 
numerical difference between the potentials of the quadrants, and not on the sign of 
the electrification. 
