H 



N.-l TURE 



[NOVEMDER 5, 1908 



This, as we have seen, can be done in two waj's, one 

 electrical and the other optical. The possibihty of 

 detection of a single atom of matter is in this case, 

 of course, due to the great energy of motion of the 

 a particle. 



In the second paper, an account is given of experi- 

 ments to measure the charge carried by the a par- 

 ticles. Since the number of o particles is known from 

 the counting experiments, the charge on each a par- 

 ticle can be determined by measuring the charge 

 carried by the a particles expelled from a known 

 quantity of radium. As in the counting experiments, 

 radium C was used as a source of rays. It was 

 found that each a particle carried a positive charge 

 of 9'3xio-'" electrostatic units. Now the charge 

 carried by an ion in gases has been determined by 

 several observers, using the well-known method of 

 making each ion the nucleus of a visible drop of water 

 by a sudden expansion. J. J. Thomson obtained a 

 value 34x10-", H. A. Wilson 31x10-'°, and 

 Millikan and Begeman 406 x 10- '°. 



The mean of these three determinations of e is 

 3"Sxio-'°. The charge E on an a particle on this 

 data thus lies between le and 3<'. 



Some calculations of the value of E and e are' 

 then made from radio-active data based on simple 

 and very probable assumptions. Taking the half- 

 period of transformation of radium as 2000 years — the 

 value found by direct measurement by Boltwood — it 

 is shown, on the assumption that each atom of radium 

 in breaking up emits one o particle, that the charge e 

 carried by a hydrogen atom comes out to be 4'! x 10-'". 

 Similarly, supposing that the heating effect of radium 

 is a measure of the kinetic energy of the a particles, 

 the charge carried by an a particle comes out at 

 9' I X 10-'° — a value close to that found experimentally. 

 .\ discussion is then given of the methods employed 

 in the previous determination of e, and it is shown that 

 in consequence of certain sources of error which are 

 very diflficult to eliminate, the values previously obtained 

 tend to be too small. It is concluded that the unit 

 charge e is not very different from E/2 or 465 x lo"'", 

 and that an a particle carries twice the unit charge. 

 From the previous discussion of the interpretation of 

 tlie value of E/.M for the a particle, it follows that an 

 a particle must be an atom of helium carrying a 

 double charge, or, in other words, that an a. particle 

 when its charge is neutralised is a helium atom. 



It seems at first sight contradictory that an atom of a 

 monatomic gas like helium can carry two unit charges. 

 It must be borne in mind that in this case the a par- 

 ticle plunges at a great speed through the molecules 

 of matter, and must itself be ionised by collision. If 

 two electrons can be removed by this process, the 

 double positive charge is at once explained. 



We thus see that by a direct method we have been 

 enabled to count the number of a particles and to 

 determine tlie charge caused by each, and from other 

 evidence to deduce that the unit charge e is half the 

 charge carried by the a particle. 



With the aid of this data we can at once deduce 

 the magnitudes of some important atomic quantities. 

 The value of e/m for the hydrogen atom is 288 x 10'' 

 electrostatic units. Substituting the value of 

 t' = 4'65 X 10-'°, it follows that the mass of a hydrogen 

 atom is r6i x lo-^* gram. From this it follows that 

 there are 6'2 x 10°' atoms in one gram of hydrogen, 

 and that there are 2'72 x 10" molecules in a cubic 

 centimetre of any gas at standard pressure and tem- 

 perature. 



From the data already given we can pre- 

 determine the magnitude of some important radio- 

 active quantities. Let us first consider the rate of 

 production of helium by radium. One gram ot 



NO. 2036, VOL. 79] 



radium in equilibrium contains four a-ray products, 

 each of which expels 3'4 x 10'° a. particles, i.e. atoms 

 of helium, per second. Consequently, since there are 

 2'72 X 10^' atoms of helium in a cubic centimetre, the 



volume of helium produced per second is — — ^-3 —-, 



272x10" 

 or 50x10-' c.mm. per second. This corresponds to 

 a production of helium of 043 c.mm. per day, or 

 158 c.mm. per year. 



In a similar way, the maximum volume of the 

 emanation in one gram of radium can be calculated. 

 Since one atom of radium in breaking up emits one 

 a particle and gives rise to one atom of emanation, 

 the volume of emanation produced per second is one- 

 quarter the volume of helium, or i"25 x 10-* c.mm. 

 per second. Since the average life of the emanation 

 is 468,000 seconds, the maximum volume of the 

 emanation comes out to be o'585 c.mm. In a recent 

 paper Rutherford 'J'hil. Mag., August) has mea- 

 sured the volume of the emanation and obtained 

 a value not very different from the calculated volume. 

 In a similar way, it is not difi'icult to calculate the 

 period of transformation of radium and the heating 

 effect of radium. The former comes out at 1750 years, 

 which is somewhat shorter than the value 2000 years 

 found experimentally by Boltwood. As Boltwood 

 points out, however, the probable experimental errors 

 are such as to tend to give too high a value for the 

 period. The latter is deduced on the hypothesis that 

 the heating effect is a measure of the kinetic energy of 

 the expelled a particles. The heating effect is calcu- 

 lated to be about 113 gram calories per gram per hour, 

 while the observed healing effect of the sample of 

 radium from which the standard preparation was 

 taken was found to be no gram calories per hour. 

 For convenience, the data obtained in this paper are 

 collected below : — 



Charge carried by a hydrogen 1_ ^ j^^,„ ^i^.t^^^t^j;^ ^^-^^^ 



atom J 



Charge carried by a particle =9'3 x lo"'" electrostatic units. 



Mass of H atom = 161 x lo"-^ gram. 



Number of atoms per gram of) -6-2 x lo'-^ 



Number of molecules per cc. j 



of any gas at standard pres- - = 272 x 10''' 



sure and temperature ... ) 

 Number of a particles expelled I 



per sec. per gram of 'adium - = 3'4 x 10'" 



itself j 



Number of atoms breaking up| _,., ^ ._m 



per sec. jier gram uf radumi j " ^ ^ 

 Calculated volume of emana-1 ._o. 



tion per gram of radium ) ■' ■' 

 Production of helium perl ,,0 



gram of radium per year ) •' 

 Calculated healing effect o'U ,13 g^. cal. per hour. 



radium per gram J -■ " '^ 



Calculated period of radium... =1750 years. 



We have already seen that there is a substantial 

 agreement between the calculated values of the heat- 

 ing effect, the life of radium and the volume of the 

 emanation, and the experimentally determined values. 

 .\ still further test would lie in a comparison of the 

 calculated and observed rates of production of helium 

 by radium. Data on this subject will probably soon 

 be forthcoming.' 



Some very important consequences follow from the 

 proof that the a particle is a helium atom. It must be 

 concluded that the atoms of the known radio-active 

 elements are in part at least constituted of helium 

 atoms which are liberated at definite stages during 



1 (Footnote, added September 12, 1908.) In a paper just to hand(Proc. 

 Roy. Soc, A., vol. Ixxxi.,p. 280) Sir James Devvar has shown experimetitallv 

 that o'.q? c.mm. of helium is produced per gram of radium per day. This is 

 in excellent agreement with the calculated rate, 0*43 c.mm. per day. 



