Conductivity along a Glow Discharge and in Flames. 363 



still further. With a- still further reduction of pressure it 

 was uot possible to keep the discharge current to the same 

 value as before. 



Fig. 8. 



32 



24 



15 



a 



Ol40h^ 



0400 



0-2/5 



26 



30 



34 



33 cm 



It appears to us that the chief interest of these experi- 

 ments lies in the variation of the conductivity found over 

 the striations. 



With our electrodeless method the minima of conductivity 

 were always observed close to the brightest part of the stria- 

 tions, while on the other hand, with the electrode-cross- 

 current method maxima of conductivity are obtained at 

 these points. It is however possible, as suggested by H. 

 A. Wilson, that in the electrode-transverse-current method 

 secondary ionization due to the illumination of the testing 

 electrodes might increase the cross currents at the luminous 

 parts of the striae *. This possible cause of error is of course 

 eliminated in the electrodeless method. Moreover, the results 



* See H. A. Wilson, Phil. Mag. [5] xlix. p. 505 (1900); also J. J. 

 Thomson, Cond. of Electr. thr. Gases (2nd Ed.) p. 502. 



