162 



BELL SYSTEM TECHNICAL JOURNAL 



tables) and the ratio of the absorbing-power of air of given density and 

 the absorbing-power of water for the cosmic rays. This last not being 

 known a priori, it is the custom to take the inverse ratio of the densities 

 of the media (page 152) — another instance of assuming that cosmic 

 rays are of the nature of gamma-rays. The assumption is strength- 

 ened by the success of the thus-made allowance in bringing all the 

 points upon a single curve. 



The data of Fig. 4 were obtained in 1925 and 1926; those of the 

 following Fig. 5, in 1927. New readings were made in the waters of 



3 

 2.4 



8 10 

 40 



20 

 50 



22 24 26 28 



30 32 34 36 

 60 



38 40 42 



70 



Fig- 5 — •Ionization ascribed to cosmic rays as function of depth beneath top of 

 atmosphere (later curve of MiUikan and Cameron). 



Arrowhead Lake and in another not previously used, Gem Lake, 2,760 

 metres up in the high Sierras; into this last the ionization-chamber was 

 lowered to a depth as great as sixty metres. The data for the two 

 localities were altered to make allowance in the foregoing way for the 

 different heights of air overlying the two lakes, and the values so ob- 

 tained were plotted against "depth in metres-of-water beneath top of 

 atmosphere." Evidently the data fit better to the curve than in the 

 prior work: this Millikan and Cameron ascribe to improvements in 

 the new ionization-chamber or "electroscope" (outwardly like that in 

 Fig. 6) into which air was compressed to a pressure of eight atmospheres 

 instead of one, as previously. Incidentally, in Fig. 5 the Restgang 

 appears under the title "zero of the electroscope." 



