column and oilier phenomena connected with the discharge. 315 



positive column is the position in the tube where the oppositely 

 charged ions from the electrodes meet. The magnetic field 

 deflects these streams, probably by different amounts since the 

 positive and negative ions may differ in mass, charge and velocity, 

 and so the point of meeting is shifted nearer the kathode. 



When an electrode is a disc the ions produced there are pro- 

 pelled along the axis of the tube by the action of the electric 

 field and a less number shot at the sides than would be the case 

 if it were a piece of wire ; hence the position in the tube where 

 the oppositely charged streams meet will, other things being 

 equal, be further from the disc than from the point, bearing out 

 the observations above on a tube with a point and disc for elec- 

 trodes. 



In the same way it was found that when both electrodes are 

 discs, the pressure being a few tenths of a mm., then the positive 

 column was longer than if the anode was a point, but shorter 

 than if both electrodes were points. 



11. Other causes influencing the distance between the striae 

 are a transverse magnetic field and gas occluded by the electrodes. 

 When a magnetic field is produced at right angles to the tube, 

 more striae appear from the anode. If the current is kept con- 

 stant these striae are nearer together than they were originally. 



Measurements could not be taken concerning this effect as 

 the striae are distorted when the field is not uniform throughout 

 the length of the tube. 



From observations on the lengths of the positive column and 

 the Faraday dark space it appears that these varied irregularly 

 with the current and pressure. 



12. Experiments were made to see if the distance between 

 the striae at a given pressure was equal to the distance between 

 the electrodes for which the sparking potential is a minimum, 

 Peace 1 having shown that this minimum sparking potential 

 depends on the distance apart of the electrodes. 



The apparatus shown in fig. 5 was used for this purpose. 



The top electrode was an aluminium disc and was fixed, the 

 bottom electrode was a clean mercury surface whose position could 

 be varied by means of the reservoir A, a wire dipping in this was 

 connected to one of the poles of a battery. The distance between 

 the electrodes could be measured by a small cathetometer or by 

 placing a ruled glass mirror close behind the tube 



Measurements of the sparking potential for varying distance 

 of the electrodes were first tried but very irregular results were 

 obtained, no doubt owing to the change in the surface of the 



1 Proc. Roy. Soc. 1893. Recent Researches, pp. 86 et seq. 



