Different Types of ' y Rays from Radioactive Substances. 529 



The glass cylinder A, 6 cm. in diameter and 20 cm. in length, 

 could be filled by means o£ two taps with the gas whose 

 absorption was under investigation. The ends were closed 

 air-tight by very thin mica sheets (equivalent in stopping- 

 power of ol particles to about 1*5 cm. of air), in order to 

 reduce the absorption of the soft rays to a minimum. 

 The ionization-chamber B was constructed of thick brass 

 (8x8x8 cm.) and lined throughout with cardboard ; the 

 wall nearest the source was made of thin mica covered with 

 a very thin aluminium foil (1'4 x 10~ 4 gr. per uuit area). 

 The central electrode — a brass rod — was connected to a 

 Wilson-Kaye electroscope, and the charging of the leaf- 

 system was observed when the walls of the chamber were 

 raised to a sufficiently high potential to ensure saturation. 

 The ionization in B was measured, first with hydrogen, then 

 with the gas under investigation, all gases being carefully 

 purified and dried. The absorption in hydrogen is known to 

 be negligibly small for rays of the penetrating power used *; 

 we may therefore assume that the ionization with hydrogen 

 in the cylinder is practically the same as if the cylinder had 

 been evacuated (this was impossible, because of the thin mica 

 windows). Let the ratio of leak be I T and I 2 , with hydrogen 

 and the gas respectively, where the natural leak and the very 

 small effect of the hard rays have been subtracted ; then /z,, 

 the absorption coefficient of the soft rays, can be determined 

 from the formula I 2 = I 1 e~ 20 > J ; as the absorption of the hard 

 rays of radium D in a length of 20 cm. is too small to in- 

 fluence the results. The values of the coefficients of absorption 

 defined in this way are given in Table I. 



Table I. 





H at 18° in em.-l. 



fi at C 



(*/p- 



Air 



CO* 



000382 

 000643 

 00318 

 00381 



000407 

 0-00674 

 0-0338 

 00406 



3-15 

 3-42 



220 



13-9 



HoS 



2 

 SOo 





It is of interest to compare the above figures with the data 

 for homogeneous X rays determined recently by several 



* Gowdv, Journ. de Phys. p. 622 (1913). 

 Phil. Mag. S. 6. Vol. 28. No. 166. Oct. 1914. 2 M 



