Xeiylibourhood of Electrodes in Point-plane Discharge. 615 



marked change in slope quite close to the plate, the potential 

 drop there being, in some cases, as much as one fourth of 

 the total fall of potential between point and plate. Further- 

 more, since the completion of the author's experiments on 



Curves III. 



this part of the work. Hovda (Phys. Rev. Jan. 1912) has 

 published a curve of potential gradient for the much greater 

 distance of 50 cm., and this exhibits, near the plate, a drop 

 which is only a very small fraction of the total drop between 

 the electrodes. 



The author's curves, for various shorter distances, agree 

 in shape with that of Hovda for a great distance, and not 

 with Warburg's earlier ones. 



The following considerations present themselves : — 



(i.) The existence of a true drop in potential near the 

 plate, such as occurs at the electrodes in glow discharge at 

 low pressure. This is unlikely, since the observed effect 

 extends at least a millimetre from the plate, and there was 

 no glow at the plate. 



(ii.) The presence of free ions in the space between the 

 electrodes. These should cause an increase in the potential 

 gradient as the plate is approached, provided the divergence 

 of the lines of force between point and plate is not too great. 

 But such an effect would not be confined to the region within 



2 8 2 



