November 3, 1923] 



NA TURE 



651 



plexes," found in folklore, by reference to the organisa- 

 tion of a given society. Inversely it allows us also to 

 trace the pattern of instinctive and emotional ten- 

 dencies in the texture of the social fabric. By making 

 the theories .somewhat more elastic, the anthropologist 

 can not only apply them to the interpretation of 

 certain phenomena, but also in the field he can be 

 inspired by them in the exploration of the difficult 

 borderland between social tradition and social 

 organisation. How fruitful Freud's theories are in 

 this respect I hope to demonstrate clearly in the 

 pending publication previously mentioned. 



Bronisi AW Malinowski. 

 Department of Ethnology, 

 London School of Economics, 

 University of London. 



Spectra of Isotopes. 



The quantum theory of line spectra developed by 

 Bohr has been most successful in explaining the 

 spectrum of hydrogen and helium, and hy a further 

 hypothesis the spectra of the alkali metals. By 

 asserting that elliptic orbits are possible, as well as 

 circular orbits, Sommerfeld succeeded in explaining 

 the Stark and Zeeman effects and the fine structure of 

 spectral lines. It is significant that Bohr's equation 

 for the frequency of the spectral series also explains 

 the difference between the series spectra of isotopes 

 of the same element. The experiments of Aronberg 

 and Merton on the structure of 4058 A. of isotopes of 

 lead showed, however, a remarkable discrepancy 

 between the shift predicted by the theory and that 

 actually observed. Similarly Merton's experiments 

 on the line 6708 of lithium showed that the line 

 consisted of two components 0-151 A. apart, while 

 the theoretical shift was 0-087 A. The quantum 

 theory is unable to account for this large separation 

 observed. 



Recently Ehrenfest, commenting upon the validity 

 of the simple Bohr equation, remarked that the 

 equation cannot be true in general for atoms with 

 several electrons, as in this case the radiating electron 

 compels the remaining electrons to execute the motions 

 of reaction which influence the nucleus. Moreover, 

 Nicholson has shown that, by the choice of simpler 

 orbits and by the supposition made by Sommerfeld, 

 as to invariability of energy W for all possible orbits, 

 the inner orbit has a radius of about one-tenth of 

 that of the outer orbit. It has thus been shown that 

 the external electron moves in the field of the nucleus, 

 which is asymptotically a Coulomb field, and that 

 Bohr's formula cannot be far wrong for a. rough 

 determination of the separation to be looked for in 

 the spectra of isotopes. 



Prof. McLennan, however, in an account of interest- 

 ing experiments (Proc. Roy. Soc. A 714, p. 33, and 

 A 711, p. 342) on the structiire of the line 5460 of 

 mercury and the line 6708 of lithium, and isotope 

 displacement, has found that when the radiation 

 constituting the green line of mercury is passed 

 through moderately luminous vapour, the main 

 component and components + i and - i are distinctly 

 absorbed. In an attempt to explain the complex 

 structure of the lines from the point of view of the 

 i.sotopic structure of the elements, the view is put 

 forward that the spectral displacement for isotopes 

 should be given by the atomic number multiplied by 

 the displacement calculated on liohr's theory, and. the 

 main components of 54<'io are attributed to isotope 

 200, and the component -1- 1 and - i to isotopes 198 

 and 202 respectively. 



In the light of the recent experiments of Bronsted 

 and Hevesy, who succeeded in separating the isotopes 



NO. 2818, VOL. 1 12] 



of mercury, and showed also that the isotopic com- 

 position of mercury of terrestrial origin is the same, 

 it is difficult to conceive why, in Prof. McLennan's 

 experiments, the lines corresponding to isotopes 198, 

 200 and 202 should alone be absorbed, while the lines 

 corresponding to the other isotopes are not absorbed. 

 Further, if, according to Aston's experiments, isotopes 

 197-200 exist in mercury in largest proportion, one 

 would naturally expect that the most intense com- 

 ponent of 5460, that is, the main component, should 

 naturally correspond to isotope 197. Similarly in 

 the case of lithium, he found that the line 6708 

 consists of a quartet, the average displacement of 

 one doublet being about 3 to 4 times as great as 

 the calculated separation, namely, 0-087. ^'^'^ gener- 

 ally it is found that enhanced lines are developed 

 when an arc is operated in vacuo, thereby showing that 

 it cannot be supposed that these lines are true arc 

 lines, which is in conformity with Nicholson's view 

 that the radiation 6708, which McLennan examined, 

 might be the principal spark line of lithium, which has 

 a value very close to 6708. 



The.se facts naturally lead one to question whether 

 -McLennan 's view has real physical significance. To 

 settle this point a careful examination of the structure 

 of some bright line spectra was undertaken in this 

 laboratory. The most recent experiments of Aston 

 {Phil. Mag., May 1923, p. 934) have definitely estab- 

 lished that tin is a highly complex element, being a 

 mixture of eight isotopes of atomic weights 120, 118, 

 116, 124, 119, 117, 122 and 121, in which case the 

 isotope displacement for 5631, for isotopes 120 and 

 124, and 120 and 116, is roughly equal to 1:0-0007614, 

 that is, for it6 and 124, 0001523 ; while according 

 to McLennan's view it is equal to 0-03807 and 0-07614, 

 which is well within the limits of resolution of an 

 ordinary Lummer plate or Fabry Perot etalon. There- 

 fore the structure of the lines 5631 and 4524 was 

 carefully examined by a Lummer plate, the R.P. of 

 which for 5631 =250,000. In these experiments the 

 arc was enclosed in a chamber surrounded by a water 

 jacket, and the radiation from tlie arc was examined 

 at different pressures. It was found that even when 

 the pressure was low (that is, of the order of i mm.) 

 both the lines were simple in structure, especially the 

 line 4521, which was very sharp. These experiments 

 do not, therefore, support the view put fonvard by 

 McLennan. A. L. Narayan. 



M. R. College, \ i/ian,igaram. 

 South India. September 1 1 . 



A Substitute for the McLeod Gauge. 



Although numberless accounts have appeared of 

 the precautions necessary in the obtaining of high 

 vacua, some serious workers seem still to imagine that 

 they can reach " a perfect vacuum " or " a pressure of 

 o-ooi mm." in an apparatus from which absorbed 

 water has not been removed. The lingering of this 

 ancient superstition is due to the prevalence of the 

 McLeod gauge ; if any gauge which indicates vapours 

 as well as permanent gases had been in general use, 

 it could never have arisen. Historians may di.spute 

 whether the invention of the McLeod gauge ha-; 

 advanced or retarded the development of science ; 

 but there is no doubt that to-day, though it may 

 have special uses (such as the calibration of other 

 gauges) under rigidly controlled conditions, it is 

 usually a mere relic of the past. 



Compared with its adequate substitutes the McLeod 

 gauge has not even the merit of convenience. In 

 particular, although some workers who are perfectly 

 aware of its limitations continue to make it a normal 

 component of any pumping system, it is not the most 



