1912] Secondary Rays Excited by the Alpha Rays. 191 



carbon was at a low temperature than when it was maintained at a higher 

 temperature. 



From the foregoing it is clear that this modification of the intensity 

 of the secondary radiation is attributable as in the case of air to an in- 

 crease in the amount of gaseous oxygen and hydrogen occluded in the 

 surface of the carbon by a reduction of the temperature. 



From the experiments just described it. follows that the amount 

 of secondary radiation obtained from an electrode under bombardment 

 by a rays may be taken as a measure of the density of the gaseous layer 

 at the surface of the electrode when the latter is placed in a gas at a very 

 low pressure. If there were no difference in the secondary radiation 

 from a substance at different temperatures under the conditions mention- 

 ed it would indicate either that there was no gaseous layer at the surface 

 of the substance bombarded, or else that the density of the gaseous 

 layer adhering to it was the same at all temperatures. In the following 

 section an experiment is described which makes use of this conclusion. 



IV. — Experiments with Brass. 



A peculiar effect was observed during the course of all the experi- 

 ments, namely, that it took a longer tim.e to reduce the pressure of the 

 gas in the apparatus to a low value when the walls of the vessel were 

 cooled with liquid air, than when the walls were at ordinary room temp- 

 erature. At first this effect was supposed to be due to some leak in the 

 apparatus and the vessel was removed and carefully tested under pres- 

 sure for small holes or porous places in the brass. But invariably it was found 

 that no such holes or porous places could be discovered. It v/as also 

 found that when the apparatus was put back and exhausted at ordinary 

 temperatures the pressure was reduced with the same speed as before 

 the apparatus was cooled with liquid air. There seemed, therefore, to 

 be no leak of air through the walls of the apparatus and the only other 

 explanation that could be offered was that a gas came from the walls of 

 the brass vessel at liquid air temperature which did not come at room 

 temperature. If this were the case there would be less gas in the brass at 

 low temperature than at high, and the method indicated in the previous 

 section of studying the gaseous layer at the surface of a substance by 

 the quantity of secondary rays coming from it could be used to test the 

 above explanation. Accordingly a brass plate was placed at K instead 

 of a carbon one and the amount of a ray excited secondary radiation 

 from the brass plate was determined when it was at room temperature, 

 and when it was at the temperature of liquid air. The series of readings 

 taken are given in Tables XIV and XV. In order that the brass when at 

 room temperature might be as free of air as possible the pump was made 



