360 Dr. Baltli. van der Pol on Measuring without Electrodes 



switched on and off' (by closing and breaking the filament 

 circuit), and no change in the position of the various parts of 

 the glow-discharge was observed, neither did the steady 

 current through the discharge vary. Even when the 

 Faraday dark space was brought between the Lecher wires, 

 which part of the discharge was found before to be the most 

 sensitive one, no disturbing effect of the waves on the dis- 

 charge could be detected. 



On the other hand, the effect of the discharge on the waves 

 was very marked, and the absorption of the latter in the 

 various parts of the discharge could easily be measured. An 

 exception is to be made, however, for the brighter part of the 

 negative glow and the regions near the electrodes. The elec- 

 trodes, namely, were turned out of brass, covered with 

 aluminium foil, and fitted closely inside the discharge-tube. 



When now the latter was lowered till one of the electrodes 

 was in the plane of the Lecher wires, the big capacity be- 

 tween the electrode inside the tube and the circular arcs 

 directly outside the tube, being only separated by the glass 

 wall of the vessel, upset the tuning and the slider K had to 

 be moved over several centimetres, and the Lecher wires 

 therefore shortened, in order to attain resonance again. 

 The same effect of detuning was caused by the big conduc- 

 tivity of the brighter part of the negative glow. However, 

 the complete positive 'column, whether striated or not (with 

 the exception of the region close to the anode), together 

 with the Faraday dark space, could easily be measured out. 

 The wave-field was further sufficiently localized to find the 

 variation of absorption over the distance of one stria, as 

 shown by fig. 6 (these observations were taken without the 

 bent wires of fig. 4). 



Here the discharge passed through hydrogen taken from 

 a steel cylinder with a little turpentine vapour added to get 

 the striations well marked. The striae had a sharpty defined 

 concave boundary facing the cathode and the parts between 

 the successive striations were very dark. The discharge 

 current was z = 4 , 0xl0 -3 amp., while the pressure was 

 p=zO m 50 mm., as read from a MacLeod gauge. 



The galvanometer current is again represented by the 

 ordinates, and the abscissae show that a region round 

 the central part of the discharge was measured out, the 

 anode being placed at the point of the abscissa? scale, 

 while the cathode is found at the point 42*2. 



The resonance length of the parallel wires was first care- 

 fully determined and found to be constant over the range of 

 the discharge represented in fig. 6. The big absorption (high 



