56 XXXVI. V. Novák a B. Mackfi: 



The électrodes of the coherer were of copper, the layer on the 

 plate was prepaired by the direct contact of the gaseous halogen by 

 method shown in the figur 2. The washed and dry gas was let into 

 the vessel through one of the tubes, the air going out through the 

 other. At first the circular holes were covered with glass plates; 

 when the vessel was füll of halogen the glass-covers wete taken 

 away and the plate-electrodes put on. 



The résistance of the coherer was measured by means of a sen- 

 sitiv galvanometer at three différent intensities of the current, i^ zr 10~^ ; 

 ij :=z 0*2; ig :=z 2.10~^ ampère, using one accumulator as a source. The 

 résistance of a clean contact of two copper-electrodes was about 

 0"5 SI (at a pressure of 1 cm Hg), it was constant when under a con- 

 stant pressure, and it was absolutely insensitiv to the inductive effect 

 of the spark. 



The results of the resistance-measurements made on the coherer 

 with a layer of iodin are in table 2. The first column contains the 

 number of the measurement and a note relating to the nature of the 

 layer, the second the mass of the layer in mg, the pressure of the 

 contact in mm of Hg is in the third column, the other numbers 

 refer to the résistance of the contact, the 4th and 5th column to the 

 initial-resistance at intensities i^ and ig respectively, the 6th and 7th 

 to résistance changed by the inductive effect of the spark. 



The results, taken from these numbers are as follows: 



1. The initial-resistance dépends on the pressure of the contact, 

 it rises with the growing pressure to a maximum. 



2. The initial-resistance at the samé pressure and the samé inten- 

 sity of the current dépends on the thickness of the layer, on the time 

 which elapsed from the préparation of the layer („age of the layer") 

 and on the change of the layer, which was caused by illuminating 

 the layer with sun-beams, With a current of 0"0L milliampère the 

 résistance of a freshly prepaired and thin layer was some thousands SI., 

 this value is rising with time, (with the age of the layer) but it is 

 reduced greatly by ligteniog the layer with sun beams. 



3. The résistance of the coherer is less at stronger current. The 

 200-time8 stronger current reduces the résistance 20--200-times. Ob- 

 servations made on the samé contact but with différent intensities 

 of the current prove that the change is due to the thermal effect of 

 the current. The little résistance at the intensity ig rised to the higíi 

 value when the intensity ig was changed into ij. 



