Neighbourhood of Electrodes in Point-plane Discharge. 613 



has disappeared. It was, therefore, evidently an effect due 

 to the space between the glow and E. 



The tact that the variation does not occur for negative 

 discharge may result from two possible causes: — (1) That 

 the glow, with its accompanying wholesale ionization, extends 

 further from a negative point than f ro n a positive ; this is 

 indicated by the fact that the glow seen on a negative point 

 when examined through a microscope is trumpet-shaped, 

 whilst the positive glow appears as a velvety layer. (2) That 

 the space between the glow and E is filled, in the case of 

 positive discharge, with slow moving H- ions, and in the case 

 of negative, with the faster — ions. Eliminating this region, 

 the potential drop near P for positive discharge, as well as 

 for negative, is thus shown to be independent of current. 

 This, again, is consistent with the results of "pull" determi- 

 nations of the field at the surface of the point. 



The influence of the sign of the point on the potential drop 

 necessary to start discharge is small, though the value for P — 

 appears to be consistently a little greater than that for P + . 

 This result is also in agreement with the measurements of 

 the field under the same conditions by the pull method. 



(b) External ions supplied. — To study the effect of sup- 

 plying to P ions of a given sign from an external source, 

 N was introduced (fig. 1). P was viewed, in the dark, 

 through a microscope, and the current from N was adjusted 

 to a value just sufficient to cause P to discharge (as indicated 

 by the appearance of a glow). Readings of this current and 

 the corresponding potential of E were taken for different 

 values of x. The potential drop (V ) between P and E, at 

 the instant of discharge from P, was thus obtained, and is 

 shown plotted with x in Curves I. (filled circles), for positive 

 and negative discharge from P. A comparison of these with 

 the curves obtained with no supply of external ions (open 

 circles) shows that (except for small values of #, in which 

 case the readings were unreliable because the three points 

 P, N, and E were all so close together) the drop in potential 

 near P, for both P+ and P — , is increased by supplying 

 external ions of the opposite sign ; the effect being more 

 marked in the case of P — . 



This admits of the following explanation : — With a plate 

 opposite P (open circles), there are, just at the commencement 

 of the discharge, practically no ions present between P and E. 

 With a point opposite P (filled circles) there are definitely, at 

 the instant of P's discharge, ions of one sign between P and E 

 (viz. those supplied by N, which discharges before P), and these 

 ions contribute their lines of force to the field already present. 



Phil. Mag. S. 6. Vol. 24. No. 142. Oct. 1912. 2 S 



