July 24, 1919] 



NATURE 



417 



number diminishes rapidly with velocity, and is very 

 small for o-particles of range 2-5 cm. 



It must be borne in mind that the production of a 

 high-speed H atom by an o-particle is an exceedingly 

 rare occurrence. Under the conditions of the experi- 

 ment the number of H atoms is seldom more than 

 I 30,000 of the number of o-particles. Probably each 

 o-particle passes through the structure of 10,000 

 hydrogen molecules in traversing one centimetre of 

 hydrogen at atmospheric pressure, and only one 

 o-particIe in 100,000 of these produces a high-speed 

 H atom ; so that in 10' collisions with the molecules 

 of hydrogen the o-particle, on the average, approaches 

 only once close enough to the centre of the nucleus 

 to give rise to a swift hydrogen atom. 



We should anticipate that for such collisions the 

 a-particle is unable to distinguish between the 

 hydrogen atom and the hydrogen molecule, and that 

 H .atoms should be liberated ifrom matter containing 

 free or combined hydrogen. This is fully borne out 

 by e.xperiment. 



From the number of H atoms observed it can be 

 easily calculated that the o-particle must be fired 

 within a perpendicular distance of 2-4xio-''^ cm. of 

 the centre of the H nucleus in order to set it in swift 

 motion. This is a distance less than the diameter of 

 the electron, viz. 3-6x10-' cni. The general results | 

 obtained with o-rays of ran^( 7 cm. are similar to | 

 those to be expected if the o-particle behaves like a 

 charged disc, of radius about the diameter of an 

 electron, travelling with its plane perpendicular to the I 

 direction of motion. \ 



It is clear from ili< > \[). nnii n;- with Indrogen that, I 

 for distances of the urdti ut the diameter of the elec- 

 tron, the a-particle no longer behaves like a point 

 charge, but that the o-particles must have dimensions 

 of the order of that of the electron. The closest 

 distance of approach in these collisions in hydrogen is 

 about one-tenth the corresponding distances in the 

 case of a collision of an o-oarticle with an atom of 

 gold. 



The results obtained with hydrogen in no wav 

 invalidate the nucleus theory as used to fxplain \\m- 

 scattering of o-ravs by heavy atoms, but show, a> 

 we should expect, that the theory breaks down when 

 wo approach ver\- close to the nucleus structure. In 

 our ignorance of the constitution of the nucleus of the 

 o-oarticle. we can onh sjxiulah as to its structure 

 and the distribution of tern -. \. i\ close to it. If we 

 take the o-particle of mass 4 to consist of four posi- 

 tivelv charged H nuclei and two negative electrons, 

 we should expect it to have dimensions of the order 

 of the diameter of the electron, supposing, as seems 

 probable, that the H nucleus is "of much smaller 

 dimensions than the electron itself. When we con- 

 sider the enormous maifnitude of the forces between 

 the o-particle and the H nucleus in a close collision 

 —amounting to 6 kg. of weight — it is to be expected 

 that the structure of the o-particle should be much 

 deformed, and that the law of force mav undergo 

 verv marked chansjes in direction and magnitude for 

 small changes in the closeness of approach of the two 

 collidinc nuclei. Such considerations offer a reason- 

 able explanation of the anomali< s -how n in the 

 number and distribution with \.!o(it\ ut ilir 

 H atoms exhibited fm- dilTr,-. nt velocities ef tlie 

 o-particles. 



When we consider tlir . tio-muus forces between the 

 nuclei, it i^ iini s(i much a matter of surprise that the 

 nuclei shmiid be defuimtd as that the structure of 

 the rr-particie or helium nucleus escapes disruption 

 into its constituent narts. Such an effect has been 

 carefuUv looked for, but so far no definite evidence of 



NO. 2595, VOL. 103] 



such a disintegration has been observed. If this is 

 the case, the helium nucleus must be a very stable 

 structure to stand the strain of the gigantic forces 

 involved in a close collision. 



We have seen that the recoil atoms of all elements 

 of atomic mass less than 18 should travel beyond the 

 range of the o-particle, provided they carry a single 

 charge. Preliminary experiments, in which the 

 o-particles passed through pure helium, showed that 

 no long-range recoil atoms were present, indicating 

 that after recoil the helium atom carries a double 

 charge. In a similar way no certain evidence has 

 been obtained of long-range recoil atoms from lithium, 

 Ixjron, or beryllium. It is difficult in experiments 

 with solids or solid compounds to be sure of the 

 absence of hydrogen or water-vapour, which results 

 in the production of numerous swift H atoms. These 

 difficulties are not present in the case of nitrogen and 

 oxygen, and a special examination has been made of 

 recoil atoms in these gases. Bright scintillations were 

 observed in both these gases about 2 cm. beyond 

 the laiii^r nf til. a-parlicir. These scinlinatiun> are, 

 pre>umal)l\, du* to -,\\iti N and t) atoms can-\ing a 

 single charge, for the ranges otjserved are about those 

 to be expected for such atoms. The scintillations due 

 to recoil atoms of N and O are much brighter than 

 H scintillaiions, although the actual energy of the 

 flying atom is j^i . ater in the latter case. This differ- 

 ence in briglitness is probably connected with the 

 much weaker ionisaiii.n yw una of path due to the 

 swifter H atom. 



The corresponding range of the recoil atoms was 

 .about the same in oxvgen, nitrogen, and carbon 

 dioxide. Theoreticallv, it is to be anticipated that the 

 N recoil atom should ifive a somewhat greater range 

 than the O atom. The recoil atoms observed in 

 carbon dioxide are apparentK due to oxygen, 1or if 

 the carbon atoms carried a single charge they should 

 be detected bevond the range of O atoms. 



rh<' iiumher of recoil atoms in nitrogen and oxygen 

 and then absorption indicate that these atoms, like 

 II aU)nl-^, are shot forward mainly in the direction of 

 ih. a-parii(lis. It is clear from the results that the 

 tuulei ot the atoms under consideration cannot be 

 regarded as point charges for distances of the order 

 of the diameter of the electron. Taking into account 

 the close similaritv of the effects produced in hvdrogen 

 and oxvgen, and the greater repulsive forces between 

 the nuclei in the latter case, it seems probable that 

 the abnormal forces in the case of oxvgen manifest 

 themselves at about twice the distance obs<rved m \\\i- 

 case of hvdrogen, /.<•. tor distances les> than 

 7x10-" cm. Such a conclusion is to be anticipated 

 on general grounds, for presumablv the oxvgen 

 nucleus is more complex and has larger dimensions 

 than that of helium. 



In his preliminarN exp.itmen;. Mar^den ol.MTved 

 that the active source alwavs uivcs n-^e 10 a number 

 of scintillations on a zinc sulphide sck en tar bevond 

 the range of the o-particle. I have alwavs lound 

 these natural scintillations p|-e>^ent in the sources_ of 

 radiation emploved. 'lie -witt atoms producing 

 these scintillntior^s are detl,,te,l ;.i a magnetic field, 

 ,,nd have ahoui 1 he ^nne ,, ;■ i-x a- the 



switl If at()m> i.KKluced hx 1 t a-ptriudes 



through hvdrogen. i'lie niimi>e.- ni these n-'Pural 

 scintillations is usuallv small, and it is verv difficult 

 to deride definitelv whether such atoms arise from 

 the (hsintiefatiiMt (d' ihe active matter or are due to 

 the action ot the a-p.irlicle^ on h\(iioL;cn occluded in 

 the source. 



These natuial ^cintillatMn- we,e studied h\ nlacing 

 the source m a closed l.o.x .xliaux.'-l ot an at.out 



