314 Mr Willows, On the distance between the striae in the positive 



From numerous curves for different gases and different tubes 

 it appears that the limiting distance between the striae is equal 

 to the diameter of the tube. 



Influence of length of tube and shape of electrodes. 



10. The distance between the striae in two tubes similar in 

 all respects except length was the same in each case provided the 

 external resistance be so altered that the same current passes 

 through both. If the gas be pure, the striations are equally 

 spaced throughout the whole column, except that that one next 

 to the Faraday space is farther removed from the next than the 

 mean distance. 



If the gas be impure they may be very unevenly distributed. 

 When the pressure in a tube is reduced, the positive column 

 always becomes striated first at that end nearest the kathode, 

 and the pressure at which this takes place seems fixed for the 

 same tube. 



The shape of the electrodes was found to have a slight in- 

 fluence. This is shown in curves A, G (fig. 2) where A was 

 obtained from measurements taken on a tube having aluminium 

 wires for electrodes and G from an exactly similar tube with discs. 

 The striae are thus seen to be further apart in a tube having 

 points for electrodes ; the difference, however, is very small. 



The discharge is different according as points or discs are used 

 when the pressure is a few tenths of a mm. Thus in a tube with 

 points eight striae were visible, while in an exactly similar tube 

 in communication with it with disc electrodes only two could be 

 seen when the current was strong, the negative glow and Faraday 

 space filling nearly all the tube, but if the current was considerably 

 weakened then the negative glow was considerably reduced and 

 four striae could be seen. 



A similar effect was produced in a tube 15 cm. in length having 

 one electrode a point and the other a disc, e.g. at a pressure of 

 "4 mm. there were four striae present when the point was kathode, 

 but if it was anode and the current was strong, there was only a 

 glow at the extreme tip of the wire to represent the positive 

 column, while the negative glow extended from the disc half way 

 down the tube. If the current was considerably reduced then the 

 negative glow was weakened and two striae appeared. 



These effects as well as the well-known lengthening of the 

 positive column by a transverse magnetic field are easily explained 

 by the ionization theory. 



It has been shown 1 that in all probability the end of the 



1 J. J. Thomson, Phil. Mag. March, 1899, p. 267. 



