70 BELL SYSTEM TECHNICAL JOURNAL 



scattering particle, the greater this maximum possible transfer of 

 energy and this maximum possible frequency-shift become; and for a 

 collision between an X-ray quantum and a free electron, it attains the 

 order 10"^ of the primary frequency, and is very appreciable. In fact, 

 the frequency-shift occurring when X-ray quanta transfer energy to 

 free electrons and these employ it as kinetic energy was the first of all 

 to be observed. It is simply the Compton effect. It was noticed first 

 towards 1904 and was described as "softening of the scattered X-rays," 

 and in 1922 was for the first time properly measured and properly 

 interpreted by Arthur Compton. The scattered rays include every 

 frequency from that of the primary rays, «o let us call it, downward to 

 the lower limit (1 — 2hn/MC^)no, as they should." 



If the primary quantum breaks the particle which it strikes into two 

 or more fragments — as for instance when an atom is ionized or a 

 molecule dissociated — the requirement of conservation of momentum 

 no longer limits the amount of energy which it may pass to these. 

 It must give at least enough energy to ionize or to dissociate the 

 particle; beyond this, so far as we know a priori, any extent of transfer 

 is permitted. Hence we should expect to observe in the spectrum 

 of the scattered light a continuous band, commencing at the frequency 

 which is less than that of the primary light by the quotient of h into 

 the ionizing-potential or the dissociation-potential, and extending 

 towards lower frequencies indefinitely far. More precisely, we should 

 expect to observe as many of these bands as there are modes of dis- 

 sociation or modes of ionization feasible by light. 



No one, so far as I know, has yet observed any bands corresponding 

 to dissociation of molecules or to the detachment of loosely-bound 

 electrons by visible or ultra-violet light. In the X-ray region, however, 

 it is different. In the spectra of scattered X-ray bands answering to 

 this description, and suggesting that X-ray quanta have extracted 

 deep-lying tightly-bound electrons from atoms and have conferred 

 kinetic energy upon them, have in fact been reported. Several such 

 spectra were depicted in 1923 and 1924 by G. L. Clark and W. Duane. 



If it should turn out in any special case that quanta could extract 

 electrons from atoms, but could not confer extra kinetic energy of 

 translatory motion on them — a restriction which there is no evident 



^ This is disguised by the fact that the rays scattered in any one direction (rela- 

 tively to the primary beam) are of a single frequency. If we observed simultaneously 

 rays scattered in all directions, we should see a continuous band of light extending 

 between Wo and the stated lower limit. This condition was approached, though not 

 purposely, in some of the earlier researches on the Compton effect. 



The shift which should occur if quanta of the visible spectrum are scattered by 

 free electrons is very small, but sufficiently large to be appreciable; however, this 

 type of scattering does not seem to occur to a perceptible extent, for it has been sought 

 in vain (P. A. Ross). 



