590 BELL SYSTEM TECHNICAL JOURNAL 



the two (not that of the alpha-particle alone !) which must be taken 

 into account.^^ Denote by C/i and U^ the values of this sum for the 

 faster and for the slower alpha-particles. Does the gamma-ray 

 spectrum then consist of a single line of which the photon-energy hv is 

 equal to ( t/i - f/2) ? 



In the case of AcC, the difference {Ui - U2) is 0.35 or 0.36 MEV. 

 There is an intense gamma-ray line proceeding from actinium active 

 deposit (comprising AcC), and the energy of its photons is concordant. 

 In the case of RaC, the difference {Ui — U2) is only 0.04 MEV, and 

 the search for so relatively soft a radiation of photons is difficult. 

 In the case of Ra the conditions are more favorable, and here the 

 history is worth retelling. Long before the earliest analysis of alpha- 

 ray spectra, radium was known to emit feeble gamma-rays of photon- 

 energy about 0.19 MEV. An estimate of their intensity was made in 

 1932 by Stahel; he concluded that the photons are less than one- 

 tenth as numerous as the alpha-particles already known. Search was 

 thereupon made by Rosenblum for fine-structure in the alpha-ray 

 spectrum of radium. Two lines appeared on the plate after five 

 minutes' exposure: they were due to groups proceeding one from 

 radium and the other from its daughter-element radon. On plates 

 exposed for hours there appeared yet another line. The values of 

 Ui and U2 being computed for this and for the stronger radium group, 

 the difference was found to be close to 0.185 MEV, with a sufficient 

 latitude to be concordant with the estimate for the photons. 



As the number of alpha-ray lines increases beyond two, the prospects 

 rapidly become formidable; for a spectrum of n such lines suggests n 

 possible states of the residual nucleus, and every one of these might 

 "combine" (in the technical sense of the word) with every one below 

 it in the energy-scale, making a total of n{n — l)/2 gamma-ray lines 

 to be expected. Even so, anyone acquainted only with optical spectra 

 might think it no difficult matter to photograph (say) the gamma-ray 

 spectrum of ThC, and see whether it consists in just 15 lines in just 

 the right places to correspond with the six alpha-particle groups. 

 But one does not photograph gamma-ray spectra — one photographs 

 the beta-ray spectra of the electrons ejected by the gamma-rays from 

 atoms, and tries to deduce the photon-energies hv from the electron- 

 energies.^^ The atoms may be those of the radioactive substance 



" By multiplying the kinetic energy of the alpha-particle by the factor (1 -|- m/M), 

 where m stands for the mass of the alpha-particle and M for that of the recoiling 

 nucleus. This point was overlooked by a number of people before it was noticed by 

 Feather. 



1* I have dealt with this procedure at length in the article "Radioactivity," No. 

 XII of this series (Bell Sys. Tech. Jour., 6, 55-99, 1927). 



