COSMIC RAYS MILLIKAN AND CAMERON" 219 



to see whether the Milky Way is more or less effective than other 

 portions of the sky in sending these rays into the earth. 



On all these four jjoints we obtained, despite unfortunate accidents 

 with tAvo of the electroscopes, satisfactory and definite information. 



As to (1), we obtained on the surface of Lake Titicaca (altitude 

 12,540 feet or 3,822 meters) readings which corresponded very nicely 

 with similar ones taken at ]\Iuir Lake, Calif. Also, in Lake MiguilUi, 

 near Caracoles, Bolivia (altitude 15,000 feet or 4,500 meters), we 

 obtained readings which fell satisfactorily on the extrapolated Muir 

 Lake curve. If, then, there are any geographical differences in the 

 altitude-ionizatidn curve, they are beyond the limits of our present 

 observational technique. 



As to (2), Lake Miguilla is a small lake surrounded on all sides by 

 mountains several thousand feet high. It would be completely 

 shielded from rays having their origins in thunderstorms anywhere 

 on the earth. Further, off the coast of Central America we took a 

 long series of readings in the wireless room on shipboard on a night 

 on wdiich a brilliant display of lightning was going on along the 

 coast, and we compared the results with readings taken on the 

 California coast, which is almost entirely free from thunderstorms, 

 without bringing to light the slightest difference. The C. T. E,. 

 Wilson hypothesis is therefore quite definitely eliminated. 



As to (3), we took the zeros of tw^o of our electroscopes by sinkmg 

 them to sufficient depths and then made an elaborate series of sea- 

 level observations on tlfe ship all the way from Mollendo, Peru, to 

 Los Angeles. We found no variation, in sea-l^vel reading with geo- 

 graphical position^ and but slight differences between the ionizations 

 in different instruments, though they had volumes nearly in the ratio 

 1 to 2 and different sorts of walls. The mean value of the sea-level 

 ionization thus directly observed was but a few tenths of an iou 

 above the mean of the sea level cosmic ray ionizations given by the 

 two curves of our jjreceding report. These curve-values wwe 1.4 

 for electroscope No. 1, and 1.6 for electroscope Xo. 3 — mean value 

 1.5, which is thus checked approximately, though not yet accurately 

 (see below), since the ionization due to the radioactive matter in the 

 air above the ocean must be very small. The main uncertaintj'' in 

 this present value 1.5 for the sea-level ionization lies in the determi- 

 nation of the capacities of the electroscopes, and in uncertainties in 

 the effect of electroscope' walls. Upon the latter effect Ave shall make 

 a later report. 



As to (4), we took two long series of observations, each lasting 3 

 days, at an altitude of 15,400 feet (4,G20 meters) in a deep valley 

 from which the Milky Way was in sight for a period of 5 or 6 hours 

 and then practically out of sight for another 6 hours. The value 

 of the cosmic rays which entered our electroscopes in this valley 



