Apeil 23, 1915] 



SCIENCE 



593 



as Fedaroff, had sought to attach impor- 

 tance to these numbers, but the efforts had 

 little to commend them. Lately it has been 

 suggested by van den Broek that this is a 

 fundamental and important number. 

 Beginning with 1 for K, the numbers would 

 be 2 for He, 3 for Li, 4 for Be, etc. The 

 question then naturally arises, can these 

 numbers be reliably determined without 

 reference to the atomic weights and cor- 

 recting the manifest mistakes made in fol- 

 lowing the simple order of these weights? 



One method for doing so, though with 

 limitations, lies in the measuring of the 

 scattering of the a particles when passing 

 through different kinds of matter. Geiger 

 found that the angle of the scattering 

 seemed to depend very largely upon the 

 atomic weight of the scattering metal. A 

 very small fraction are scattered through 

 such a large angle that they return on the 

 side of incidence. This deflection is, of 

 course, both a volume and surface effect. 

 For equal thickness of screen calculations 

 based on Rutherford's conception of the 

 atom and his belief that this large angle 

 scattering is due to the near approach of 

 the positively charged a particle to the 

 positive nucleus of the atom of the screen 

 would make the scattering vary as the 

 product of the density by the atomic weight. 

 Thus Eutherford calculated that the scat- 

 tering by gold should be about fifty times 

 that by aluminium. This has been eon- 

 firmed by the experiments of Geiger and 

 Marsden, and the relative scattering has 

 been determined for a large number of ele- 

 ments. The phenomenon is manifestly 

 one determined by the electrical content of 

 the atom. 



The nuclear charge of the Rutherford 

 atom can be calculated from the a particle 

 scattering at various angles. This charge 

 is found to be one half the atomic weight 

 multiplied by the charge of an electron. 



The same value was reached by Barkla by 

 observations on X-rays. Soddy concludes 

 that it is the nuclear charge rather than the 

 atomic mass which fixes the position of the 

 element, basing his conclusion largely upon 

 the work of Barkla, Sadler and Moseley, 

 which will be briefly outlined further on. 

 This in reality agrees with the hypothesis 

 of van den Broek that the number of elec- 

 trons in an atom in the neutral state deter- 

 mines the place of the element if hydrogen 

 has one electron and one nuclear unit 

 charge, helium two electrons and two nu- 

 clear unit charges, etc. 



The direct method then is a combination 

 of the work of Bragg, Barkla and Sadler, 

 and Moseley. Making use of the work of 

 those first mentioned, Moseley photo- 

 graphed the spectra obtained by the 

 cathode-ray bombardment of a number of 

 elements, the X-rays thus produced being 

 reflected and defined from a crystal face. 

 The frequencies of the vibrations could be 

 determined and this frequency was found 

 proportional to the square of the atomic 

 number. That is, there was a definite 

 shifting in the direction of shorter wave- 

 length in the spectrum of an element from 

 that of the one next above it in the list. 



The graphic representation of the system 

 has never been satisfactory in spite of the 

 many efforts to solve it. It is especially 

 difficult to bring out the facts by any rep- 

 resentation on a plane surface. The faults 

 of the Mendeleeff table can readily be seen, 

 and they make it very desirable to secure a 

 better mode of expression. And yet it is 

 difficult to use the three dimensions of space 

 so that the average student can grasp the 

 whole. Soddy 's lemniscate curve certainly 

 has its good points. This may be compared 

 with the arrangement of Rydberg. It can 

 not be claimed yet, however, that the law or 

 laws underlying this system are known and 

 well understood, and until such time a com- 



