Discharge from an Electrified Point. 425 



stated below ; but the results are not without interest, and 

 they are therefore here described. 



That point discharge at atmospheric pressure and moderate 

 currents is not solely a one-way discharge has been sug- 

 gested by various experimenters. Thus Campbell * has 

 shown by the condensation method that negative ions are 

 present during positive discharge. Secondly, the author f 

 found that there is an apparent increase — when the current 

 increases — of the specific velocity of the ions in air, as mea- 

 sured by the wind-pressure method ; this is readily explained 

 by back discharge. The discharge has since been examined 

 under better conditions, and it has been found that at the 

 currents at which this effect is appreciable the whole dis- 

 charge region is very faintly luminous. Thirdly, there is 

 definite evidence % for back discharge in exceedingly pure 

 hydrogen, since in the purest samples of that gas the direction 

 of the electric wind is actually from plate to point. 



On the other hand, Zeleny concludes from some experi- 

 ments on the rate of leak of a charged conductor placed in 

 the discharge, that back discharge, if present, is exceedingly 

 small in hydrogen and air at atmospheric pressure. 



Recently, Moore § has shown that calculating the velocity 

 of the ions from the reaction at a discharging point, on th« 

 assumption of one-w T ay discharge, values are obtained about 

 five times greater than those generally accepted. He point* 

 out that this may be explained either by the presence of 

 discharge from the plate or of a region of ionization ex- 

 tending some distance from the point* That the nature of 

 the discharge within a few millimetres of a point is different 

 from that beyond that distance has been shown in the various 

 measurements of the electric wind. It is not necessary, 

 therefore, to suppose a general back discharge to explain 

 his results. 



The method of investigation in the present work depends 

 upon some observations made by Chattock ||, who found that 

 the area over which the current spreads is .slightly greater 

 in negative than in positive discharge. Assuming that back 

 discharge is absent, or at any rate small, this leads to a 

 simple method of comparing the specific velocity of ions 

 under different physical conditions. If this method be tested 

 for a known case and it be found that the calculated value 



* Campbell, Phil. Mag. vol. vi. p. 618, 



t Phil. Mag. vol. xxi. p. 585 (1911). 



X Chattock & Tvndall, Phil. Mag. vol. xix. p, 449 (1910). 



§ Phvs. Kev. Feb. 1912. 



j| Phil. Mag. vol. xlviii. p. 401 (1899). 



