154 BELL SYSTEM TECHNICAL JOURNAL 



can say that the residue {Restgan^, the Germans call it) is the joint 

 effect of the causes aforesaid, and must thenceforth be subtracted 

 from every reading made with the same apparatus. Kolhorster did 

 this on a grand scale, going many hundreds of metres down into one 

 of the potash-mines of Stassfurt, where there is a great hollow exca- 

 vated in rocksalt; the thickness of the overlying rock was great, but 

 as a drawback there was a powerful radioactivity of the potassium in 

 the rock-salt, causing hard gamma-rays of which a perceptible frac- 

 tion invaded the chamber. In Java the physicist Clay took his 

 machinery into a tunnel, where he was covered by rock eighty-four 

 metres thick and far less rich than rocksalt in radioactive atoms; in 

 the Sierra Nevada Millikan descended 185 metres into granite. 



More instructive is the customary procedure of Millikan and of 

 Regener. We shall presently consider curves (Figs. 3, 4, 5) which 

 show the rate of discharge of electroscopes in air-chambers sunk under 

 water, plotted against the depth of submergence. The discharge- 

 rate falls off as the air-chamber is lowered, but the decline grows slower 

 and slower; the curve flattens out and seems to approach, seems even 

 to attain, a certain horizontal line. It is the ordinate of this line which 

 is taken as the Restgang; this is subtracted from all the other ordinates, 

 the differences are ascribed to the cosmic rays. Now if one could 

 be absolutely sure that this is the line to which the curve is making 

 asymptotic approach, there would be no uncertainty. But since no 

 measurement is perfectly precise, no one can say absolutely that the 

 curve is not still gently sloping, towards an asymptote distinctly 

 lower than the lowest value measured. Yet it is on some guess 

 as to the answer of this unanswerable question that there rest, not 

 (fortunately) the proof that there are such things as cosmic rays, but 

 the estimates of their amount and of their greatest penetrating power. 

 I revert to this question later; at present I merely mention it, in order 

 to show how difficult these estimates may be. 



A very important point now demands to be noticed. In taking 

 these elaborate precautions to exclude the rays from radioactive 

 substances beyond the walls of the air-chamber, is one not also keeping 

 out some of the cosmic rays themselves? To deny this would be all 

 but impossible; it would amount to assuming that all ionizing rays, 

 apart from those which we recognize as proceeding from known 

 radioactive atoms, are so much more penetrating than these that they 

 pass absolutely undiminished through centimetres of lead and metres 

 of water — an assumption which has only to be stated, to show itself 

 improbable. True, if the unknown rays come altogether from above, 

 and if the radioactivity of the air can be allowed for, one may omit to 



