138 STATIC ELECTRICITY 



If, for example, the lines of force pass normally from air into a 

 dielectric of specific inductive capacity K the electrical tension 



x 



in air is ^TrD 2 , while that in the other medium is , so that 



TT _ 1 



for equilibrium a force 2?rD 2 ^ must be supplied to keep the 



surface layer in equilibrium. The medium is no doubt elastically 

 strained, and the corresponding stress supplies the required force. 



If the lines of force are parallel to the surface separating one 

 dielectric from another, the level surfaces are normal to that surface, 

 and the electric intensity is the same in each medium. If the electric 



KE 2 



pressure in air is E 2 /87r, that in the other medium is . Hence 



E 2 



(K 1) is the pressure needed on the surface of the other 



medium in addition to the electric pressure to maintain equilibrium. 

 Quincke's experiments. Quincke made a scries <>i experi- 

 ments of historical interest to test the existence of these electrical 

 tensions and pressures in dielectric liquids. He used a condenser * 

 with horizontal plates, the upper being hung insulated from a 

 balance beam, while the lower was insulated and so arranged that 

 it could be connected to a battery of Leyden jars charged to given 

 potential. The condenser being charged, the pull G, on the upper 

 plate was first determined in air. Then, the condenser being sur- 

 rounded with the liquid to be experimented on, the pull G 2 was 

 determined, when the potential difference, and, therefore, the elect He 

 intensity, was the same as before, the plates being the same distance 

 apart. Taking the area as A and neglecting the edge effect, we 

 have by the foregoing theory : 



G f _AKE / AK 2 _ 



~ = 



or the pulls are in the ratio of the specific inductive capacity. 



In each case the condenser was discharged through a ballistic 

 galvanometer. The quantity flowing through the galvanometer 

 and indicated by the kick-off' was proportional to the capacity of the 

 condenser, and, therefore, to the specific inductive capacity of the 

 medium. This, of course, gave another determination of K. 



At first the values by the two methods were somewhat widely 

 apart, owing to the neglect of the capacity of the connecting wire 

 and key. When this was taken into account the determinations 

 were fairly in accord. f 



The existence of the pressure at right angles to the lines of 

 force was verified as follows : The upper suspended plate was 

 replaced by a fixed plate with a short vertical tube passing up 



* Phil. Mag., vol. xvi. (1883). 



f Xalure, vol. xxxv. (1887), p. 334. 



