290 



Mr. A. P. Ghattock on the Electrification 



The small amount (3 to 6 per cent.) by which the potential 

 step calculated from V exceeds that calculated from P 

 may be accounted for, I think, by the fact that the lines 

 of force from the needle-point do not diverge quite as 

 rapidly as calculation supposed. The parallel lines of the 

 rest of the field keep them together and so increase the real 

 step of potential at the point without altering P in pro- 

 portion. To test this point I measured P for a needle of 

 about the same size as B, first suspending it in a parallel field, 

 and then with its point at the centre of curvature of a spherical 

 cup. In the first case, yP was about 3 or 4 per cent, less 

 than in the second ; showing, as it seems to me in the light 

 of what follows, that, in the parallel field, / at the point was 

 nearer the average value of the field through a small distance 

 from the point surface than in the radial field of the second 

 case ; and to about the same amount that the above dis- 

 crepancy requires. 



This accounts, moreover, for the fact that needle B is the 

 only one which does not agree with the results of Table II. 

 Its constant (/ x r ' 8 ) being 15*5 at a pressure of 76 centim. 



As regards the measurement of potential in the above, the 

 electrometer described in § b was used ; its constant being 

 determined by comparison with short sparks between brass 

 knobs of 5*3 centim. diameter. 



The following are the readings taken : — 



Spark-length. 



V from spark- 

 length (E.S. 

 units). 





Constant of 

 electrometer. 





^/Electronieter- 

 ^ reading. 



Observed. 



Corrected for 

 flat plates. 



0-127 



0-190 

 » 



0-125 



>> 



0186 



>> 



17-8 



>> 

 24-8 



>> 



6-63 

 6-63 

 9-23 

 9-30 



2-69 

 2-69 

 2-68 

 2-67 



The values of V were taken from Dr. Liebig's paper al- 

 ready referred to. 



I think, then, it may be safely concluded that the attraction 

 of a plate on a needle-point is practically due only to the lines 

 of force ending at its point — even when the needle is not a 

 perfect cylinder ; for a conical point could only have the effect 

 of increasing P and so making the step of potential calculated 



