Supplement to '' N attire,''' August 1^, 19: 



509 



hypothetical sheet of solid air would be the same as for 

 mica. 



We have now all the data required to determine the 

 values of Aj and \^ corresponding to a-particles of 

 different velocities. The results are given in the follow- 

 ing table for three different velocities. The mean free 

 paths are expressed in terms of millimetres of air at 

 standard pressure and temperature. Vq, the maximum 

 velocity of the a-particles from radium C^ is 1-9 x 10^ 

 cm. per second. 



It has been seen that the mean free path for loss 

 varies directly as the velocity, and thus only alters in 

 a ratio of about i to 2 over the range of velocities given 

 in the table. On the other hand, the ratio Ag/Aj in- 

 creases very rapidly with diminution of velocity varying 

 approximately as V"^. From this it follows that Aj 

 varies as V^, thus decreasing by a factor of 60 or more 

 when the velocity is halved. 



From these data and relations it can easily be cal- 

 culated that the mean free path for capture should be 

 equal to that for loss for a velocity about 0-3 Vq, and 

 for this speed the numbers of He+ and He++ particles 

 should be equal. 



The actual value of the velocity for equality of the 

 two types in a special experiment was found to be 

 0-29 Vq, in good agreement with the calculated value. 

 It is a difificult matter to determine the values of Aj and 

 A2 for velocities less than 0-3 Vq, for not only are the 

 scintillations weak in intensity and difficult to count 

 with accuracy, but also the issuing rays are very hetero- 

 geneous and no longer show well-defined edges on the 

 high velocity side. It was, however, noted that the 

 ratio N2/N1 rapidly increased below the velocity 



0-3 Vo- 



We have so far dealt with the equilibrium between 

 lie J and lie.,, ^ particles. It is clear, however, that 

 similar considerations apply to the ((luililiritnii hi-twccn 

 singly charged and neutral hiliuin particles at low 

 velocities of the a-parti( U;. It was noted that the 

 neutral pariiih-, a])pear promiiuni 1\- after the rays 

 have passed llinm-h mica of U cm. siop|)iiiL; poWiT, but 

 no doul)t tlicy I ( ,iil(l lie (liicctcd for still lower stopping 

 power. 'rii'M- nciitial pariKlcs, of course, produce 

 sciiu illat ioIl■^. i)Ut ol an inleii,>ity rnrresponding to an 

 '/ ]i,ii-i ii li- ol low \-elocii y. 'I'lie.^c neiil ral particles prob- 

 !i an elei'i fi A\ inan\' t inie> liefore 

 iwLj <tic,,.oppni ,ii Llic zinc bulphidc ur Oilier absorbing 



material. This effect was shown by introducing gas 

 at low pressure into the apparatus, when the scintilla- 

 tions due to the neutral particles diminished in number 

 and ultimately vanished. The explanation of this is 

 similar to that given for the disappearance of the He+ 

 band, for the neutral particles occasionally lose an 

 electron in passing through the gas and are then de- 

 flected away from the zero position by the magnetic 

 field. 



It was estimated that the mean free path in air for 

 conversion of neutral helium particles to singly charged 

 particles was about 1/600 mm. No doubt this is an 

 average for particles of very different velocities which 

 may be present in the neutral band. 



For the higher velocities we have to deal mainly 

 with the interchange IIe++^He_,_. For velocities 

 less than 0-5 Vq the interchange He+^Hco also comes 

 in and becomes all-important for velocities less than 

 0-3 Vq. No doubt, as Henderson has shown, at still 

 lower velocities most of the IIe.^+ particles disappear 

 and the Hcq and He+ particles predominate. 



At these low velocities, counting scintillations be- 

 comes very difficult and uncertain, and the photo- 

 graphic method, as used by Henderson, is preferable. 

 It will be a matter of very great interest to examine 

 whether the relative numbers of the three types of 

 particles alter when the a-particles are slowed down 

 by passage through different materials. This side of 

 the work is being attacked by Mr. Henderson in the 

 University of Saskatchewan. 



There is one very interesting point that may be con- 

 sidered here. It has been shown that these singly and 

 doubly charged a-particles are always present after the 

 a-rays have passed through mica or other absorber, 

 but are there any singly charged particles present when 

 a-particles escape from a wire coated with an infinitely 

 thin deposit of active matter ? This was first tested 

 for a platinum wire coated with a deposit of radium 

 B + C, by exposure to the radium eiAanation, when it 

 was found that singly charged helium atoms were 

 present in about the equilibrium ratio for this velocity. 

 This was a rather surprising observation, but it was 

 thought it might result from the fact that by the recoil 

 from radium A the radium B particles penetrate some 

 distance into the material of the wire. Under these 

 conditions many of the a-particles expelled from 

 radium C have to pass through a small but appreciable 

 thickness of matter before escape from the wire and 

 might tluis capture electrons. This explanation 

 ei ined unlikely because the average distance pene- 

 tiaied by the recoil atom is only a minute fraction of 

 the mean free path for capture at such high xelocities 

 of the a-particle. The experiment was tried with a 

 nickel wire on which radium C had been deposited on 



