=^.0 



NATURE 



[November 26, 1914 



the solar radiation and the number of sun-spots was 

 also traced, an increased spottedness of the sun's disc 

 being found to correspond with an increased radia- 

 tion. It appears, therefore, that the sun, besides 

 varying from day to day, varies also from year to 

 \ car in accordance with the march of the sun-spot 

 cycle. In addition, observations had been made on 

 the contrast of brightness between the centre of the 

 sun's disc and the limb for different wave-lengths; 

 this also appears to vmdergo an irregular variation 

 similar to the variation of the total radiation. Further 

 researches on these problems are in progress, includ- 

 ing some balloon ascents with self-registering appa- 

 ratus, in which a height of twenty-two miles has been 

 obtained. Dr. Abbot urged the importance of secur- 

 ing the co-operation of stations widely distributed 

 over the globe to eliminate any local atmospheric 

 influences, and he suggested that work in this branch 

 of solar physics might well be taken up by the pro- 

 posed new observatory in Australia. A long discussion 

 followed the paper, and, in answer to questions, Dr. 

 Abbot explained the general lines of work which he 

 would like to see carried out in Australia. Dr. Simp- 

 son and Mr. Gold discussed the bearing of the results 

 on meteorology. Dr. Abbot showed that a close 

 parallelism existed between the spottedness of the sun 

 and the mean temperature in the United States, the 

 main disturbing factor being the presence of volcanic 

 dust in the atmosphere after some of the great erup- 

 tions. These two causes accounted for most of the 

 main features of the temperature curve. Prof. Edding- 

 ton suggested that a similar irregular variability 

 might be looked for in other stars having spectra 

 resembling the sun's; apparently the amount of varia- 

 tion would be sufficiently great to be detected by 

 differential measures of stellar magnitudes. 



The morning's proceedings concluded with a paper 

 by Mr. P. Baracchi on the present state of the deter- 

 minations of Australian longitudes. The stations at 

 Port Darwin and at Southport (Queensland) are the 

 terminals of two chains of telegraphically determined 

 longitudes, one carried eastward and the other 

 westward from Greenwich. The connection between 

 these two meridians obtained by means of measured 

 longitudinal arcs completes a whole longitude circuit 

 of the globe with a closing error of less than loo ft. 



On Tuesday, August i8, a joint meeting was held 

 with Section B (Chemistry), for a discussion on the 

 structure of atoms and molecules. Sir Oliver Lodge 

 presided. Sir Ernest Rutherford, who opened the dis- 

 cussion, directed attention to the recent accumulation 

 of evidence of the independent existence of the chem- 

 ical atom. It is no longer merely an hypothesis intro- 

 duced to explain the laws of chemical combination ; 

 we are able to detect and count the individual atoms. 

 We can determine the actual mass of an atom cer- 

 tainly to within a few per cent. The idea that the 

 atom is an electrical structure received a great 

 impetus by the detection of the electron by J. J. Thom- 

 son ; and, moreover, the Zeeman effect showed that 

 it must be in part electrical. The atomic character 

 of negative electricity is well established ; we always 

 find the negative electron, however produced, carrying 

 a definite charge. We have, unfortunately, not the 

 same certainty with regard to the behaviour of positive 

 electricity, for it cannot be obtained except associated 

 with a mass comparable with that of a hydrogen 

 atom. In J. J. Thomson's model of the atom the 

 positive electricity was supposed (for mathematical 

 reasons) to be distributed throughout a large sphere 

 with the negative corpuscles moving inside it. This 

 hypothesis played a useful part in Indicating possible 

 lines of advance ; but it does not fit in with more 

 recent discoveries, which point to a concentrated posi- 



NO. 2352, VOL. 94] 



five nucleus. We have now two powerful methods 

 that aid us in determining the inner structure of the 

 atom — the scattering of high-speed particles in transit 

 through matter, and the vibrations of the interior 

 parts of the atom. In C. T. R. Wilson's photographs 

 of the tracks of the a particles through a gas, we 

 notice many sudden bends in the paths ; in order to 

 account for these deflections it is found necessary 

 to believe that there is a concentrated nucleus atom 

 (having a certain number of units of charge), in which 

 the main part of the mass resides ; outside this there 

 are a number of electrons. The whole dimensions of 

 the nucleus are very small indeed compared with the 

 distance of the outer electrons. From the scattering 

 experiments it appears that the law of force right up 

 to the nucleus is the inverse square law ; no other 

 formula would give accordance with the observations. 

 The radius of the nucleus is of the order lo-^^ cm. 

 in the case of gold, and for a lighter element It Is 

 smaller still. Another fact that appears from the 

 scattering experiments is that the number of electrons 

 (outside the nucleus) is about half the atomic weight. 

 There Is now fairly good evidence that, if the elements 

 are numbered serially In order of atomic weight, 

 the numbers will actually express the charge on the 

 nucleus. The I'ate of vibration of the inner parts of 

 the nucleus can now be measured by means of the 

 characteristic X-rays emitted. Each substance has 

 two strong lines in its X-ray spectrum, and as we pass 

 from element to element in order of atomic weight 

 the frequencies of these change by regular jumps. 

 H. G.-J. Moseley has investigated all the known 

 elements in this way, and he is even able to show at 

 what points elements are missing, because at such 

 points the X-ray frequencies make a double jump. In 

 this way he has found that between aluminium and 

 gold only four elements are now missing. It is 

 deduced from these considerations that there Is some- 

 thing more fundamental than atomic weight, viz., the 

 charge on the nucleus, and that this Is the main factor 

 which controls the frequency of the Interior vibrations, 

 the mass having only a slight Influence. There are 

 certain elements with identical chemical properties, 

 but different atomic weights ; thus radium-B (atomic 

 weight 214) and lead (207) are chemically Inseparable 

 and have the same y-ray spectrum. It Is quite clear 

 that some new conception is required to explain how 

 the atoms, having the structure we have supposed, 

 can hold together. N. Bohr has faced the difficulty 

 by bringing in the Idea of the quantum. At all 

 events, there is something going on which Is Inexplic- 

 able by the older mechanics. 



Prof. Armstrong referred to the new attitude with 

 which the chemical elements were regarded ; he had 

 long been prepared to believe that they may eventually 

 prove to be compounds. With regard to the existence 

 of elements of different atomic weights chemically 

 indistinguishable. It Is too early to conclude that no 

 method of separation and distinction can be devised. 

 The deduction from the X-ray spectra that very few 

 elements are still missing Is scarcely justifiable ; it may 

 well be that most of those that are known belong 

 to a preferred type, and that this particular series is 

 nearly complete. For the chemist any theory of 

 atomic structure must take fully into account the 

 peculiar valency relationships ; the system Is admittedly 

 one of extraordinary perfection and simplicity ; it should 

 be pointed out, however, that so-called structural 

 formulae are to be regarded as condensed symbolic 

 expressions of the general behaviour and not as repre- 

 senting the actual structure. Thus the conventional 

 representation of benzene by a plane regular hexagon 

 is quite impossible if the affinities of a carbon atom 

 act tetrahedrallv. The variable valencv of certain 



