430 Dr. A. M. Tyndall on the Spread of the 



The results are shown in Curves I. and II. Curves I. are 

 for positive discharge, and in them the currents arriving at 

 A, B, and C are plotted as ordinates with the total current 

 arriving at the plate as abscissae (circles). The curves with 

 crosses give the results when no point was projecting, and 

 hence no glow present. Curves II. give the curves for 

 negative discharge under the same conditions. Glow was 

 observed to start at 10 microamperes for positive, and at 

 about 12 m.a. for negative discharge. The ordinates are in 

 arbitrary units. 



It will be seen that particularly in positive discharge 

 there is a marked rapid increase in the current between P 

 and A after glow has started at N. 



Both the glow and the increase in current set in much 

 more gradually in negative discharge. This is readily 

 explained if, when N is negative, it is more efficient as a 

 source of back discharge than when it is positive. 



For a given current flowing, the current at B, C, and D 

 after glow has started at A, must be less than the current at 

 these when no glow is present. This is also shown in 

 Curves I. and II., the Curves for B and C alone being plotted. 

 The main part of the increase in current passes between the 

 two points. 



Moreover, it was previously observed that with increasing 

 current the distance between point and plate when the 

 character of the wind changed decreased. In the present 

 work it was found that the current at which glow appeared 

 also increased as the distance between point and plate 

 decreased. This was to be expected if the glow made its 

 appearance at a certain critical value of the field near the 

 plate. 



That the presence of the projecting point has no effect 

 upon the geometrical conditions is shown by the fact that 

 the distribution of current over the plate, before glow, is the 

 same whether the point is there or not. 



Lastly, the apparatus was used to clear up another question 

 outstanding from previous work. 



It was found * that in very pure hydrogen the plate was 

 covered by a velvety glow which underwent curious fluctu- 

 ations in size, sometimes extending over the whole plate and 

 sometimes contracting to the size of a threepenny piece ; 

 the changes were in all cases rapid and irregular. This was 

 discussed in the light of the back discharge which the wind 

 measurements showed to be present in that gas. 



* Oliattock & Tyndall, Phil. Mag. vol. xix. p. 449 (1910). 



