510 



Mr. H. A. Wilson on the Variation of the 



Fisr. 4. 



To £x£cr/roMC7<eA 



diameter and 025 millim. bore, and was covered with a very 

 thin coating of glass. 



The apparatus is shown in fig. 4. The two tubes W, W 

 were sealed to glass tubes provided with taps which commu- 

 nicated with small insulated mercury- 

 reservoirs. The positive electrode M 

 was formed of mercury which could be 

 raised and lowered, and into which the 

 mercury from the two jets fell. The 

 negative electrode N was kept fixed. 

 The two tubes were connected to the 

 electrometer. 



With this arrangement it was found 

 that the electric intensity indicated was 

 not affected by running the mercury 

 provided the stream of mercury was 

 strong enough to go for a short distance 

 horizontally before breaking into drops. 

 Of course if big drops of mercury formed on the ends of the 

 jets, these caused an alteration of the electrometer-reading : 

 but otherwise running the mercury had no effect either in the 

 positive column or very near to the positive electrode, where 

 the intensity was found as before to be very small, or even 

 negative. 



The apparently negative intensity near the positive electrode 

 was found to be most marked at low pressures and with small 

 currents. Increasing either the current or the pressure dimi- 

 nished the length of the region of apparently negative 

 intensity. 



Since it is probable that the mercury -jets really took up 

 the actual potential in the gas, we must conclude either that 

 the presence of the wires or jets disturbs the discharge to 

 such an extent as to reverse the sign of the intensity near the 

 positive electrode, or that the intensity there is really negative. 

 Such a region of negative intensity in the gas indicates the 

 existence of a source of E.M.F. acting in the gas throughout 

 the region in question. Such a source of E.M.F. may be 

 imagined to be provided by positive ions shot off from the 

 positive electrode by the excessively sudden potential-fall 

 which occurs at its surface. Since the rise of potential in 

 the region of negative intensity is usually only 1 or 2 volts, it 

 is conceivable that the positive drop of 25 volts may project 

 the positive ions from the positive electrode with enough 

 energy to carry them against a negative intensity for a short 

 distance, or they may be supposed to get across the region by 

 diffusing simply. 



