Mr Wilson, On a Sensitive Gold-Leaf Electrometer. 139 



be further increased the deflection now increases less and less 

 rapidly with the potential difference. If we retrace the curve by 

 diminishing the potential difference, the sensitiveness continually 

 increases till it becomes very great near C, and there follows a 

 sudden drop to A, the lower branch of the curve being then 

 retraced from this point as the potential is further diminished. 



If the tilt of the apparatus be increased beyond the critical 

 amount, the curve has the form shown in fig. 4 b. Here the deflec- 

 tion for a given potential is single-valued, and there is no part of 

 the curve indicating instability. Exactly at the critical angle 

 there will be a portion of the curve vertical, indicating infinite 

 sensitiveness, and the nearer this is approached the more sensitive 

 will the instrument be, when the plate is kept at the corresponding 

 potential. 



The practical limit to the sensitiveness is probably fixed by the 

 degree of constancy which can be attained in the potential of the 

 plate. 



The instrument may be carried about without risk of injury to 

 the gold-leaf; it is only necessary to charge up the plate to such a 

 potential that the gold-leaf (connected to the case) is stretched 

 straight out towards the plate, which it is just too short to touch. 

 The apparatus may then safely be inverted or carried in any 

 position ; the gold-leaf remains steadily pointing towards the 

 plate. 



