Metals subjected to Rbntgen Radiation. 



715 



stream of coal-gas was then passed through it for at least a 

 day and allowed to stand over night and started again. The 

 gas entered by the lower tube, C, and issued from the higher 

 level. During the test a very slow stream of the gas flowed 

 through this box, as it seemed to be the best method of keep- 

 ing the atmosphere uniform and pure, as occluded gas issuing 

 from parts of the box would be removed. 



Table III. 



Exposure. 



Iron. 



Ratio to 

 Standard Plate. 



Nickel. 

 Ratio. 



hours 



100 



105 

 100 

 104 



105 



100 



98 

 99 

 98 



1 „ 



2 „ 



3 „ 



1 

 4 „ 



5 „ 



6 , 





These results are similar in character to those for polished 

 surfaces in air. 



Polished Plates in Hydrogen. 



The effect of immersing the plate in hydrogen was tried 

 with iron, as this metal gives apparently the most characteristic 

 results. The hydrogen was generated from pure zinc and 

 sulphuric acid and passed through drying-tubes before entering 

 the test-box. Considerable difficulty was experienced in 

 getting rid of traces of air. The gas passed through the box 

 for four hours in a slow stream, entering near the top and 

 issuing from the bottom, and the plate was then allowed to 

 stand all night in the atmosphere of hydrogen. In the 

 morning, the stream of gas was again sent through the box 

 continuously until two o'clock when a test was made. During 

 the test also the stream, reduced in quantity, continued to 

 flow. 



It is essential for good results to drive out all traces or air 

 from the trial plate and to keep the standard plate in air. 

 These precautions are necessary because the ionization currents 

 produced in hydrogen and air by the same secondary radiation 

 are not at all equal. So, for example, we might suppose a 



