364 Prof. Owens on Thorium Radiation. 



Similarly for the nitrate : 



Plates 5 cm. apart. Lower plate 95 volts + . 



Time of 100 div. in sec. 

 65 - 1st observation. 

 63-6 2nd 



>j 



62-0 3rd 



62*2 4th „ constant. 



and for the sulphate : 



Plates 5 cm. apart. Lower plate 95 volts +. 



Time of 100 div. in sec. 

 61*0 1st observation. 

 60-2 2nd „ 

 58*6 3rd „ 

 58'0 4th „ constant. 



This time-effect is very marked with the oxide, and small 

 and nearly equal for the nitrate and sulphate. When thin 

 layers of the active material formed by sifting a little of the 

 salt through a fine wire gauze were used, the effect was 

 comparatively small, even with the oxide. 



As the only apparent effect of waiting some minutes after 

 placing the active material in position before taking observa- 

 tions was to allow the air in the box to come to a steady 

 state, the effect of air-currents through the box was naturally 

 suggested. 



To investigate the effect of air-currents, a small box similar 

 to that shown in fig. 1 was used, but made air-tight. In 

 each of two opposite sides small tubes were fitted above the 

 active material. By means of an air-pump or bellows currents 

 of air of different velocities could be passed through the box. 

 Then by inserting plugs of glass wool, vessels containing 

 water, drying-mixtures, and other materials before the inlet 

 tube, the effect of dust-particles, moisture, &c. could be 

 studied. A thick layer of thorium oxide was placed on the 

 lower plate and charged to 95 volts positive. 



After allowing time for the air in the box to become quiet, 

 the conduction-current was repeatedly measured and found 

 to be constant. Take this steady current as 100. A water- 

 pump was then used to draw air through the apparatus at a 

 fairly rapid and constant rate. The conduction-current fell 



to 33. 



The air was next made to bubble through a vessel of water 

 before passing into the apparatus, but the conduction-current 



