﻿Absorption of Rant gen Rays in Air. GS5 



error due to a change of hardness o£ the bulb. The current 

 was passed for a few seconds and the fall of potential in each 

 electroscope observed. The ratio (moved/fixed) was multi- 

 plied by the square of the distance between electroscope and 

 bulb. The corresponding value for the scattered radiation 

 was deducted, and the natural logarithm of the difference 

 was plotted against, the distance. The result for ranges of 

 4-10 metres gave a straight line, so that the rays obeyed the 

 inverse square law and were also absorbed exponentially. 



Three Rontgen-ray bulbs were used. The first (A) was 

 an exceedingly hard one, used by one of us * some years ago 

 in a comparison of the ionization of gases by Rontgen and 

 7 rays. On the first day X was found to be '00077 in the 

 morning and *00069 in the afternoon ; on the following day 

 '00065. On the third day the bulb was run for some time, 

 and the alternative spark-gap extended to a maximum of 

 30 cm. The bulb became extremely hard, and gave 

 X = '00033 in the morning and '00025 in the afternoon. 

 These values accord well with those found for a very hard 

 bulb by Rutherford and McClung, namely '000279. The 

 rays from such bulbs are too penetrating to give good results 

 for skiagrams for medical work. 



The second bulb (B) tested was a new one, recently 

 purchased by the Physics department. It was fitted with a 

 regulator for varying the hardness, and it proved to be re- 

 markably steady. Values in the morning and afternoon of 

 the same day gave X= '00040 and '00041. The alternative 

 spark gap was 11 cm. 



The third bulb (0) was the property of the Royal Vic- 

 toria Hospital, and was too soft for medical work. Three 

 determinations gave in succession *0018, '0011, '0010, and 

 in the meantime the alternative spark-gap had increased 

 from 1'5 to 5 cm. No satisfactory exact relationship was 

 established between the length of the maximum alternative 

 spark-gap and the value of X. 



A fourth bulb was examined, but although fitted with a 

 regulator, it was extremely erratic, alternating between 

 extreme hardness and softness, so that no reliable results 

 could be obtained. 



As the second bulb (B), \ = '00044, gives excellent shadow 

 pictures, it would seem to be a good value for the coefficient 

 of absorption for apparatus for medical work. 



Long Ranges (20-80 m.). 

 , The hard bulb (A) was placed on a ledge outside a window 

 of the Physics building, 5 metres from the ground* The 



* Phil. Mag-. Nov. 1004. 



