18 Dadourian — Radio-activity of Underground Air. 



battery being to earth. The central rod of the testing con- 

 denser and the pair of quadrants it was connected with were 

 earthed ordinarily, but could be iusnlated at will by pressing a 

 key. The electrometer was very steady throughout the follow- 

 ing experiments and had a sensitiveness of 250 cm per volt with 

 100 volts on the needle and the scale at one meter. 



The observations were taken in the following manner : The 

 zero position of the needle was observed, then the central rod 

 of the testing vessel and the quadrants it was connected with 

 were insulated. Observations of the ionization current were 

 taken at the end of half a minute, one minute and two minutes 



T^K 















q\ 



\ 













4 



\^ 



A 

















\ 





\I 









.3 







V 



\* N 

















\A 



\ 







2 









\ 



A 



















\ 





/ 













\\ 



\ 















\ 



\\ 









/ 





2 





3 



Mourh 



Fig. 2. 



after the quadrants and the central rod were insulated ; then 

 these were earthed again. This was done at six-minute inter- 

 vals for five hours, after which the activity of the wire decayed 

 very slowly, and it was not necessary to take observations so 

 frequently. The decay of the activity for the first four hours 

 is given by curve I of figure 2, where the time is plotted as 

 abscissae and the natural logarithms of the ionization currents 

 as ordinates. The experiment was continued for three days, 



