326 



NA y UKh 



[September i, 1923 



which has been observed cxj' ly by Sten- 



strttm and also by Duane and I \Phys. Rev., 



16, 532). The latter find that the dulcrence between 

 the values of X' when the tungsten line 1-473 A.U. 

 is reflected in the first and second orders from calcite 

 is 0-00015 A 0-00009 A.U. Formula (4) gives a differ- 

 ence of 0-00007 A.U., which is within experimental 

 error of the observed difference. However, this 

 difference may also be explained on the assumption 

 of a refractive index for X-rays. 



X-ray wave-lengths are also measured by observing 

 the angle of deviation (^1 -^-^^ between the reflected 

 and incident ray. This is particularly the case when 

 the photographic method is used (Siegbahn, Dershera, 

 Ovem and others). Let X' be the apparent wave- 

 length when 9^ + 0^ is observed so that nX' = ai sin 

 (*i +^a)/2- We now have a difference between X' 

 and X on our theory given by 



(5) 



X'-X'=yCOS«0, 



to the first power of 7. For the lower orders of 

 reflection this difference is approximately 0-024 A.U., 

 which should be easily observable. Ovem {Phys. 

 Rev., 14, 137) has found X' for the above line. Com- 

 paring with Duane and Patterson's value of X' for 

 the same line we find the experimental value of X' -\* 

 to be 0-0005 A.U., which is within experimental error 

 of zero. This would seem to be decisive evidence 

 that there is no change of wave-length when X-rays 

 are reflected from a crystal. 



G. E. M. Jauncey. 



Carl H. Eckart. 

 Physics Laboratory, Washington University, 

 St. Louis, Mo.', U.S.A., July 3. 



On the Structure of the Molecule. 



The difficulty of reconciling the atomic systems of 

 Bohr and of Langmuir, and of accounting for the 

 attraction between atoms to form molecules and 

 chemical compounds, might perhaps be elucidated in 

 the following way. 



If the analogy between atomic structure and 

 astronomical planetary systems holds good, the atom 

 is essentially a two-dimensional figure, while matter, 

 which is composed of atoms, is essentially three- 

 dimensional. 



If then combination takes place between two or 

 more atoms, it would be reasonable to suppose that 



-X 



® @ 



® 



Fig. I. 



this does not take place in the same plane as the 

 electronic orbit — an idea which is borne out by the 

 work of Bragg on crystal structure. 



We may suppose combination to take place some- 

 what as follows, in the formation of HjO : — 



If the plane of electrons be represented in Fig. i 

 by the dots, it would be quite possible, if an electron 

 were to be drawn out of the normal plane of each 

 atom as indicated by the arrows, for it to form part 

 of two atoms while revolving in a similar orbit to 

 the original one but, owing to its divided allegiance, 

 in a different plane. It would seem, however, that 

 it would very soon take up a position directly between 

 the two nuclei, when it would become static. 



NO. 2809, VOL. I I 2] 



In the case of hydrogen, which has only a sufli' 

 positive charge in the nucleus to hold one no;'; 

 electron, if we suppose two electrons to l>e <! 

 of the plane — one from each atom — the j 

 from the oxygen atom would be negative m ;..!!. 

 forming a negative link between two positive nu' It : 

 This would perhaps explain the quite unique position 

 of hydrogen in chemical combinations. 



The system could be applied quite readily to rr- r. 

 complex molecules. Hragg's model of the tar' 

 acid molecule (see Nature of June 9, Supplenieni, 

 p. ix) is readily amenable to this way of treatment. 





<o) (§ 



G Q 



Q 



Fig. ». 



as the accompanying diagram (Fig. 2) will show, the 

 electrons coming out of their respective systems being 

 shown surrounded by a square. 



It will be seen too from Sir W. Bragg's drawings 

 that the links may easily be conceived as being in 

 planes which would not interfere with the orbital 

 motions of the remaining electrons. 



The suggestion is then that, in the formation of 

 the molecule, both dynamic and static electrons have 

 their place, and this will throw considerable light on 

 the nature of the links between the two or more 

 positive nuclei. A. Pearsk Jfnkin. 



Trewirgie, Redruth, July 20. 



A Primitive Lens. 



If a wire of J to J mm. diameter be bent into a 

 closed circular loop of about 8 mm. diameter and 

 dipped in water, or a transparent oil such as castor 

 oil, a stable liquid film can be readily obtained 

 covering the loop. A thin dished metal disc with a 

 circular hole in the centre is a convenient alternative 

 to the wire loop. Liquid can be easily added or 

 removed without breaking the film, so as to vary the 

 curvature of the liquid lens so formed. Such a lens, 

 though far from perfect, may be made to give a 

 magnifying power of nearly 5 over a small field. 



It is conceivable that some of the very fine work 

 done in Egypt, long before the invention of " optical " 

 glass, may have been made possible by the use of a 

 liquid lens of this kind. The phenomenon might 

 easily have been accidentally observed ; for even a 

 drop of water lying on a greasy surface gives a small 

 but appreciable magnification of the surface which 

 it covers. 



By using a thicker wire (about 2 mm. diameter) 

 and less liquid, a diminishing lens may be made in 

 the same way. 



R. A. S. Paget. 



East India House, 74 Strand, 



London, W.C.2, 



August 14. 



