716 



Prof. L. T. More on the Fatigue of 



constant secondary radiation to be set np by the trial plate in 

 hydrogen, and at the same time the quantity of air remaining 

 in the box to change slightly and slowly. Then we should 

 have all the appearance of a change in this constant secondary 

 radiation. Or, if the quantity of air changed quickly, we 

 should naturally attribute the consequent change in the ioni- 

 zation current to irregularity in the working of the Rontgen- 

 ray bulb. But if we keep the standard plate in air, the latter 

 error would be correctly interpreted by the constancy of the 

 readings of its ionization current. The former difficulty can- 

 not be certainly prevented. The best I could do was to make 

 it strongly probable that slow changes in the gas did not 

 occur. This was done by taking a set of readings for both 

 the standard and the trial plates. Immediately afterwards, 

 the lead shutters on both boxes were closed and remained so for 

 half an hour. If another set of readings for both plates were 

 then taken, the ratio should be the same as before. Previous 

 to beginning a test for fatigue, this process was repeated two 

 or three times at half- hour intervals, or until the ratios did 

 remain constant within the limit of accuracy of reading. In 

 some cases it was necessary to continue the flow of gas for 

 two or three days before a steady state was reached. 



Table IV. 



Exposure. 



IrONj. 



Ratio to 

 Standard Plate. 



Iron 2 . 

 Ratio. 



hours 



100 

 94 

 95 

 03 



100 

 103 

 100 

 101 



1 „ 



2 



3 , 





The behaviour of the two samples of iron is different, as the 

 first shows a slight fatigue and the second a small increase in 

 activity. But I am led to believe that the effect of the 

 hydrogen is practically to prevent change in the activity. 

 The separate readings of the ionizations in the trial box, in 

 spite of great care, were not so accurate as in the other ex- 

 periments. And this was to be expected for the reasons given 

 above. Even if the preliminary experiments remained constant 

 and indicated that all the air was expelled, yet it is very 

 possible that long continued use of the Rontgen rays may 



