ELECTROMETERS. 219 



vertical glass tube at the top. The disc, M M', graduated at 

 the circumference, is supported on the tube N, which fits 

 into the cylinder H. K is a button which fits into the hole 

 in the disc M M'. The glass fibre which carries the torsion 

 arm is attached to the button K ; and there is also an index 

 i, which indicates on the graduated scale of MM' the angle 

 through which the button K has been turned with respect 

 to it. 



In Coulomb's ordinary method of using the instrument, 

 the torsion ball is completely diselectrified. The index i is 

 put at zero on the torsion head. The tube N, carrying, as 

 I have explained, the scale, button, and fibre, is turned 

 round so that the middle of the torsion ball shall be at zero 

 on the lower scale. The carrier ball is now electrified and 

 put into its place, and the torsion ball comes in contact 

 with it. The charge is thus divided between the two balls 

 halved if the balls are exactly equal and similar and the 

 torsion ball is repelled from the other. The displacement 

 of the torsion arm is opposed by the torsion of the fibre, 

 and finally we have equilibrium between the couple, as 

 it is called in dynamics, due to the repulsion of the balls 

 acting at the end of the torsion arm, and the couple due 

 to torsion. By turning the button K of the torsion 

 head, the torsional couple can be increased, which will 

 have the effect of forcing the torsion ball nearer to the 

 carrier ball. 



Coulomb, as I have said, first undertook the determination 

 of the laws of electric attraction and repulsion. He began 

 by examining very carefully the laws of torsional elasticity 

 of the wires that he employed, and established the im- 

 portant law that the torsional couple, for a given wire, is 

 in simple proportion to the angle of torsion ; or, in other 

 words, that if a certain couple be required to turn the lower 

 end of the wire through a certain angle relatively to the 

 upper end, the couple required to turn it through twice that 

 angle will be double ; the couple required to turn it through 

 three times that angle will be triple of the first, and so on. 

 He tested this law to high angular displacements, and found 

 that it holds with great exactness unless the torsion is so 

 great that the wire receives a permanent set, and when 

 released from torsion does not return to its original 

 condition. 



