SECTIONAL TEANSACTI0N8.— A. 301 



Thirdly there is the exchange principle. Electrons may interchange their patterns, 

 and as all electrons are identical, it is impossible to know which electrons occup\- 

 which patterns. This exchange phenomenon has important consequences. The 

 charge distribution of an atom can only be expressed by the use of six dimensions, 

 iind this six-dimensional character is important in calculating the energy of atoms or 

 the energy of interaction of electrons. 



To these new principles must be added an old one, viz., the principle of minimum 

 energy : an atomic or electronic system tends to assure the state of lowest energy. 



Cohesive forces may be classified under five headings, viz. : (i) Van der Waals ; 

 (ii) homopolar ; (iii) ionic : (iv) metallic ; and (v) adamantine. 



The principles enumerated above may be applied to an interpretation of tlieso 

 types of cohesive forces. It is becoming clear that the first may now be calculated 

 owing to the smearing-out process described above. Van der Waals' fields seem to 

 be due to a di/namic polarisation. Homopolar binding, on the other hand, seems to 

 be due to the exchange phenomenon, and ionic binding due to the principle of 

 minimum energy. Metallic cohesion, though not yet investigated in detail, seems 

 to be due to the interpenetration of atomic electric charge, which leads to greater 

 electrostatic attraction, and due also to the exchange principle. The question as to 

 whether atoms will form a metal-like sodium or a collection of molecules like hydrogen 

 seems to be determined bj' the principle of minimum energy. Adamantine cohesion 

 is probably' due to the exchange phenomenon and thus the same as homopolar binding, 

 but this is not yet definiteh' established. 



It seems as though the principles underlying the nature of cohesion are no^\- 

 understood, and the immediate need of the future is the discovery of a mathematical 

 technique, which will permit of their application to particular cases. 



Prof. W. L. Bragg, F.R.S. — The Structure of the Solid State : Inorganic 

 Compounds. 



Recent 3-ears have witnessed a great advance in our understanding of the way in 

 which the physical properties of matter are explained by its atomic arrangement. 

 Progress has been made possible both by the closer understanding of interatomic 

 forces which we o^ve to recent developments in mathematical physics, and by X-ray 

 analj'sis in revealing crystalline arrangement. The latter in particular has passed 

 from a technique which could onl}- deal with the simplest compounds, to one by 

 which the most complex inorganic crystals can be analysed. The present paper 

 attempts a general survey of inorganic compounds. 



Of the four main types of interatomic binding, ionic, homopolar, molecular and 

 metallic, the first two are of prime importance in typical inorganic compounds, though 

 examples can be given showing a continuous transition towards the other types. A 

 great deal of experimental material has now been collected, and presents very interest- 

 ing problems for further theoretical treatment. The simplest compounds are the 

 associations of ions forming structures of great regularitj'. A number of workers, in 

 particular in the school of V. M. Goldschmidt, have shown how the compounds fall 

 into types determined by the factors of ionic charge, size and polarisation. In the 

 salts with complex acid radicles, composed of outer electronisation atoms surrounding 

 an inner acid-forming atom, it would be verj- interesting to know the type of binding 

 within the group. So many of its physical properties can be given a semi-quantitative 

 explanation on the assumption that outer and inner atoms are charged ions, yet this 

 can hardly be the true picture. Much experimental work remains to be done on the 

 configuration of the more complex groups, most of those so far examined being the 

 simple tetrahedral or threefold types. The silicates present an interesting inter- 

 mediate stage between the simple compounds with continuous ionic lattices, and the 

 salts with independent complex ions. SiOy groups can be independent (orthosilicates), 

 or link in chains (pyroxenes and amphiboles) or in sheets (micas) or in three- 

 dimensional complexes (silica, zeolites). They can be regarded as acid radicles with 

 indefinite extension in one, two, or three spatial dimensions. 



The laws of co-ordination in simple compounds outlined by Goldschmidt can be 

 extended in a very satisfactory way to the more complex types. Pauling has framed 

 a striking series of general rules for such structures, which goes far to explain their 

 forms as representing an atomic arrangement which gives a minimum value to the 

 potential energy. 



