52 



NATURE 



[September 9, 1920 



all the properties of matter should be reproduced 

 by the interactions of electrons and positive 

 charges, the latter at that time quite unknown. 

 It is not possible to dwell here on the earlier atom 

 models (such as those of Thomson or Ritz) which 

 were devised on these lines. They had a certain 

 measure of success in explaining some of the pro- 

 perties of matter ; but we can now see that there 

 was insufficient experimental knowledge at that 

 time, and also that no theory could possibly work 

 which was based on the old classical ideas of 

 dynamics, as these are certainly inadequate to 

 deal with atomic problems. 



As electrons were known to be common to all 

 matter, it was natural to ask how many were 

 contained in each type of atom. The mass of an 

 electron is about 1/1840 of that of a hydrogen 

 atom, and therefore this atom cannot contain more 

 than 1840 electrons. For other elements the cor- 

 responding upper limit is given by multiplying the 

 atomic weight by 1840. Thus in all cases a very 

 wide field of choice is left. The attack on the 

 problem was not possible until the development of 

 X-rays and radio-activity had reached a certain 

 stage of advancement, for ordinary chemical and 

 physical methods are quite incapable of penetrat- 

 ing beyond the surface layers of the atom, whereas 

 both X-rays and the radio-active particles are so 

 powerful that they can go right through the atom, 

 and from their behaviour in passing we can deduce 

 a certain amount about its interior. With both 

 the method consists in observing the phenomenon 

 of scattering ; that is to say, a narrow parallel 

 beam is sent through a plate, and the amounts 

 are found which are thrown off at various angles. 

 Theory tells us how much a single electron should 

 scatter, the number of atoms in the plate is known 

 with some accuracy, and so we can count the 

 number of electrons in each atom, if we assume 

 that the scattering is due to them alone. When 

 this method is applied with either X-rays or 

 )3-particles (the latter are simply electrons moving 

 at a high speed), certain theoretical complications 

 make the results rather inaccurate, but in both 

 cases the experiments clearly indicated that the 

 number of electrons in the atom was of the same 

 general magnitude as the atomic weight. This 

 was found for several elements ranging in weight 

 between carbon and gold. So our first upper 

 limit to the number of electrons in the atom was 

 vastly in excess, and no considerable fraction of 

 the mass of an atom is contributed by its elec- 

 trons. 



The experiments with o-particles were far more 

 definite. An a-particle is simply a helium .atom, 

 carrying a positive charge equal to that of two 

 electrons, and moving at a very high speed. As 

 the atomic weight of helium is 4, it is seven 

 thousand times as heavy as an electron. When an 

 o-particle passes an electron the attractive force 

 sets the electron in motion, and the reaction of this 

 force gives the a-particle a small deflection and 

 reduces its velocity slightly. On account of the 

 smallness of the effect of a single electron, great 

 uniformity is obtained in the average, and the 

 NO. 2654, VOL. 106] 



number of electrons in an atom can be counted by 

 observing either the scattering of the beam or else 

 the rate at which the particles lose their velocity. 

 Both give the same result, that the number of elec- 

 trons in an atom is approximately half its atomic 

 weight. 



But the experiments on the scattering of 

 o-particles contained the germ of far more import- 

 ant information. For in the course of them it was 

 observed that a certain small fraction of the 

 o-particles underwent large deflections. A few 

 even were thrown right backwards. Now this 

 fraction, though small in itself, was out of all pro- 

 portion large compared with what could be given 

 by the cumulative effect of a large number of col- 

 lisions with electrons. Rutherford showed that 

 the only reasonable hypothesis to account for this 

 type of scattering was to suppose that the large 

 deflections were produced in a single step. In 

 this way he was led to put forward the nucleus 

 theory of the atom, now universally accepted. 

 According to this theory practically the whole 

 mass of the atom is carried by a nucleus of very 

 small dimensions (at most lo-^^ j.^ j which has 

 a charge of positive electricity equal to some 

 multiple of the electronic charge. This nucleus is 

 surrounded by a cloud of electrons of just such a 

 number as to neutralise its charge. Each element 

 has a different nuclear charge, and (to anticipate 

 some results that we shall come to later) the value 

 of this nuclear charge completely determines all 

 the chemical characters of the element. Radio- 

 active properties belong to the nucleus, as well 

 as mass, while chemical and spectroscopic proper- 

 ties are attributed to the electrons of the cloud, 

 and are only indirectly controlled by the nucleus 

 because it determines the number and arrange- 

 ment of these surrounding electrons. The dy- 

 namical structure of this planetary system was left 

 unspecified, and remains to this day almost un- 

 known. 



Now consider the bearing of this theory on the 

 question of scattering. The a-particle is itself the 

 nucleus of a helium atom. In passing through 

 matter most of the particles will not go very near 

 any nucleus, and so will behave in the way we 

 have already discussed. But the paths of a few 

 will take them near some nucleus, and large re- 

 pulsive forces will be developed between the two 

 positive charges. Most atoms are much heavier 

 than a-particles, and so the latter will describe 

 hyperbolas according to Newton's laws, and pass 

 off in new directions. It is a simple matter to 

 calculate the number of particles to be expected 

 at any inclination to the original beam in terms 

 of their initial velocity and the charge on the 

 nucleus. The comparison with experiment, there- 

 fore, first furnishes a test as to whether the law 

 of force has been taken correctly, and then pro- 

 vides a value for the nuclear charge. In both par- 

 ticulars the experiments fully bore out the pre- 

 diction, and it was found that the nuclear charge 

 (measured with the electronic charge as unit) was 

 about half the atomic weight for the lighter 

 elements, and rather less than half for the heavier. 



