538 SCIENCE PROGRESS 



curve, Prof, Bragg gives an estimate of the diameters of the 

 outer electron shells of these elements. Since these shells 

 correspond, on Langmuir's theory, to the outer shells of Neon, 

 Argon, Krypton, and Xenon, the values 1-30 A, 2-05 A, 2*35 A, 

 and 2-70 A are given for the outer shells of these respective 

 gases. The relations hold most accurately for compounds and 

 electronegative elements ; they do not hold accurately for 

 crystals of the electropositive elements ; this is consistent with 

 Langmuir's conception that a metallic crystal consists of a 

 number of positive ions bound together by electrons whose 

 positions are not fixed, whereas crystals of electronegative 

 elements consist of atoms bound together by sharing electrons. 

 In continuation of his work on the influence of molecular 

 onstitution and temperature on magnetic susceptibility, of 

 which a resume is to be found in Science Progress, No. 56 

 (March 1920), Dr. A. E. Oxley publishes a paper on magnetism 

 and atomic structure in the Proc. Royal Society, No. A, 692, in 

 which he gives a conception of the electron structure of matter 

 in the light of recent experiments on magnetic effects. A 

 crystalline plate of naphthalene sets with its principal cleavage 

 perpendicular to the lines of a uniform magnetic field, and 

 experiments on piles of naphthalene plates— each pile being 

 constructed of selected plates cemented together with Canada 

 balsam at different orientations, but with the principal cleavages 

 of the plates parallel to one another — show that a pile always 

 sets with the principal cleavage perpendicular to the magnetic 

 field, however thick the pile may be. The diamagnetic effect 

 is, therefore, a maximum in a direction parallel to the principal 

 cleavage. Now Dr. Oxley attributes the cohesive force, which 

 is a maximum parallel to the same cleavage, to the highly 

 localised mutual induction between a pair of electron orbits, 

 and he gives diagrams which illustrate the nature of this form 

 of coupling which would account for the diamagnetic and 

 paramagnetic properties of molecules and the cohesive force of 

 non-ionised compounds. In particular, he suggests a model of 

 the diamagnetic hydrogen molecule. Crystallisation is imagined 

 to take place under the influence of the Laplace intrinsic 

 pressure, which Dr. Oxley suggests may arise from electrostatic 

 doublets of atomic dimensions, and the directive magnetic 

 couplings which are responsible for the crystalline symmetry of 

 a given molecule and the rigidity of the crystalline structure. 

 Thus, the electron coupling is a maximum parallel to the 

 principal cleavage, so that paramagnetic and diamagnetic 

 crystals should set with this cleavage parallel and perpendicular 

 to the field respectively ; and this is verified in the case of the 

 naphthalene piles. Simple cubic crystals, which we know from 

 X-ray analysis to possess an ionised atomic structure, possess 



