Discharge from an Electrified Point. 587 



at all currents there is a back discharge from the plate 

 increasing with current and becoming very intensified 

 at particular points on the plate under certain special 

 conditions. 



Experimental Eesults. 



The experimental results may be divided into two groups : 

 (1) Experiments on the strength of field at the surface of a 

 point ; and (2) Experiments on the pressure of the Electric 

 Wind. 



(1) Experiments on the strength of field at the 

 surface of a point. 



Most of the original theories were apparently verified by 

 some experiments, in which the field at a hemispherically 

 ended point P was determined in terms of the mechanical 

 pull on its surface. 



If r is the radius of the point, / the field at the centre of 

 its surface, and P the mechanical pull due to the lines of 

 force which start from the end of P, then 



/= — — x constant. 



The value of the constant is 2'83 for a positive point and 

 about 3*07 for a negative point. The particular value of / 

 at which under given conditions glow first appears at P is 

 referred to below as f . 



The supply of "external ions'' was obtained by spraying- 

 ions on to the point P from a point N in its neighbourhood. 

 N was a point of smaller radius than P, so that JM discharged 

 more readily than P. The tendency of N to start first was 

 further increased by causing P to project through a plate, 

 the plane of which was only a few millimetres behind P. 

 P and N were both horizontal and in the same vertical plane, 

 and they were so arranged that the vertical component x and 

 the horizontal component y of the distance between them 

 could be varied at will. 



When P was positive and external ions were supplied to 

 it from N placed opposite to it, and at a distance y greater 

 than about 2 cms.,/ was reduced to about 0*6 of the value it 

 had when N was replaced by a plate. On the above theories 

 this field when N was present was /- , and when N was 

 absent /+; /— was about 0*8 /+. 



As y was decreased below 2 cms. / rapidly decreased, and 

 apparently the curve of fa and y cut the axis of f at a positive 



2Q2 



