September 22, 1923] 



NATURE 



435 



Letters to the Editor. 



[ The Editor • does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return., nor to correspond with 

 the writers of., rejected jnanuscripts intended for 

 this or any other part of NATURE. No notice is 

 taken of anonymous communications ^^ 



Recoil of Electrons from Scattered X-Rays. 



In a recent paper before the Royal Society (as 

 reported in Nature, July 7, p. 26), C. T. R. Wilson 

 announced that in his cloud expansion pictures of 

 secondary /3-rays produced by X-rays shorter than 

 0-5 A, tracks of very short range appear. These 

 electrons, he says, " are ejected nearly along the 

 direction of the primary X-rays." 



A quantum theory of the scattering of X-rays, 

 devised primarily to account for the change in wave- 

 length which occurs when X-rays are scattered, led 

 me to predict (Bulletin National Research Council, 

 No. 20, pp. 19 and 27, October 1922) that electrons 

 should be ejected from atoms whenever X-rays are 

 scattered. The idea is that a quantum of radiation 

 is scattered in a definite direction by an individual 

 electron. The change in momentum of the radiation, 

 due to its change in direction, results in a recoil of the 

 electron which deflects the ray. The direction of 

 recoil is not far from that of the primary beam, in 

 accord with Wilson's observation on his short tracks. 



Corresponding to this momentum acquired by the 

 electrpn, it has kinetic energy which varies from o 

 when the scattered X-ray proceeds forward, to a 

 maximum value hf .2a/{i + 2a), when the ray is 

 scattered backward (P. Debye, Phys. Zeitschr. 24, 161, 

 Apr. 15, 1923 ; A. H. Compton, Phys. Rev. 21, 486, May 

 1923). Here a = yj\, where 7 = ^/mc = 0-0242 A, and 

 X is the incident wave-length. The ratio of the 

 maximum energy of a photoelectron excited by an 

 X-ray to the maximum energy of such a recoil 

 electron would thus be (i-l-2a)/2a. But Wilson finds 

 the length of the trails proportional to the square of 

 the energy. The track due to the photoelectron 

 should therefore be (i -f- 20)^/40* times that of the 

 longest recoil electron tracks. 



Taking Wilson's datum that a track of i cm. 

 corresponds to 21,000 volts, the equation Ye — hcjK 

 indicates that a ray of wave-length 0-5 A will eject 

 a photoelectron with a path of 1-4 cm. The recoil 

 electron, taking a = o-0242/o-5, should accordingly have 

 a range of o-ii mm., which should just be visible. 

 For his harder X-rays, with a wave-length for 

 example of 0-242 A (a = o'i), the recoil tracks on 

 Wilson's photographs should be as long as 1-7 mm. 

 The quantum idea of X-ray scattering thus leads to 

 recoil electrons moving in the right direction and 

 possessing energy which is of the same order of magni- 

 tude as that possessed by the electrons responsible for 

 C. T. R. Wilson's very short tracks. 



Arthur H. Compton. 



University of Chicago, 

 August 4. 



As Prof. Compton points out, the phenomena re- 

 lating to the forward directed /3-ray tracks of short 

 range, which appear in air exposed to X-rays of short 

 wave-length, are in agreement with his suggestion that 

 scattering of a quantum may be effected by a single 

 electron. 



That the phenomena are in general accordance with 

 Compton's theory was pointed out in my paper 

 (which has now appeared m the current number of the 

 Proc. Roy. Soc.) ; mention of this was made in my 



NO. 2812, VOL. I 12] 



summary of the paper, but was omitted in the abbrevi- 

 ated report of that summary which appeared in 

 Nature of July 7. 



It is obvious that further observations on the range 

 and direction of tracks of this type produced by 

 homogeneous radiations may throw light on some 

 very fundamental questions. The data thus far 

 obtained by this method are not sufficient to decide 

 without ambiguity whether a quantum of radiation 

 scattered by an electron is emitted in one direction 

 only or with a continuous wave-front. 



C. T. R. Wilson. 



Cambridge, August 24. 



Long-range Particles from Radium-active Deposit. 



With reference to the communication of G. Kirsch 

 and H. Pettersson in the issue of Nature of September 

 15, p. 394, on the " Sources of long-range H-particles," 

 the results of an examination by the scintillation 

 inethod of the particles emitted by radium-active 

 deposit, in which we have been engaged for the past 

 six months, are of interest. 



It was found that the active deposit, radium B -f-C, 

 on a brass disc emitted particles with ranges (in air 

 at 15° C. and a pressure of 760 mm. of mercury) of 

 9-3, II -I, and 13-2 cm. respectively, as well as 

 particles of which the range was considerably greater 

 than 18 cm., which were not further investigated, 

 as they appeared to be H-particles. The particles of 

 range 9-3 cm. were previously observed by Sir Ernest 

 Rutherford {Phil. Mag., xxxvii., 1919, p. '571). 



Although it would not be possible definitely to 

 decide that these particles were a-rays except by their 

 deflexions in electric and magnetic fields, the appear- 

 ance of the scintillations strongly suggests that they 

 are a-rays. The numbers of these additional par- 

 ticles were relatively very small ; for every" 10' 

 a-rays of range 6-97 cm. emitted by the source, there 

 were present 380, 126, and 65 particles of ranges 9-3, 

 ii-i, and 13-2 cm. respectively, together with about 

 160 long-range H-particles. 



To ensure that these long-range particles were not 

 produced by collisions by the 6-97 cm. a-particles with 

 air molecules, the experiments on the ii-i and 13-2 cm. 

 particles were repeated, using carbon dioxide in place 

 of air. In this case the equivalent ranges in air were 

 found to be 11-3 and 13-6 cm. respectively, the 

 agreement being considered satisfactory, as the 

 measurements in carbon dioxide were not made with 

 the same precision as in air. 



Moreover, these particles could not have been 

 excited in the mica sheets which were used to provide 

 screens of various stopping powers, for the majority 

 of the experiments were carried out with air or carbon 

 dioxide gaps between the source and the mica, suffi- 

 ciently large to prevent the 6-97 cm. a-rays from 

 reaching the mica. 



The particles under consideration appeared to be 

 independent of the metal on which the deposit was 

 formed, as a check determination of the range of one 

 set of particles, emitted from an active deposit on a 

 platinum disc, gave a value of 11 -2 cm. 



It seems possible, therefore, that the 12, 13, and 

 10 cm. H-particles which Kirsch and Pettersson con- 

 sidered to arise from the collisions of a-particles from 

 their emanation tubes with atoms of beryllium, 

 magnesium and lithium, respectively, are actually 

 long range a-particles emitted by the active deposit. 

 It is of interest to note that, should the particles of 

 range 13-2 cm. later prove to be a-particles, they will 

 be the longest range a-particles yet discovered. 



Further details of our results and experimental 



