DISCOVERY 



A MONTHLY POPULAR 

 JOURNAL or KNOWTLLDGE 



No. 2. FEBRUARY 1920. 



PRICE «d. NET. 



DISCOVERY. A Monthly Popular Journal of Know- 

 ledge. 



Edited by A. S. Russell, RI.C, D.Sc, S Severn Road, 

 Sheffield, to whom all Editorial Communications should be 

 addressed. 



Published by John Murray, 50.^ Albemarle Street, 

 London, \V. i, to whom all Business Communications 

 should be addressed. 



Advertisement Office : 

 S.W.I. 



16 Regent Street, London, 

 Single 



Annual Subscription 65., or post free 7s. 6d, 

 numbers 6d. net ; post free j^d. 



Crystal Structure 



By W. L. Bragg, M.C., M.A. 



Nobel Laure.Ue : Langioorlhg Professor 0/ Physics in the 

 Universily 0/ \Ianchester 



When a liquid solidifies on cooling, or when a solid is 

 deposited from a solution, the attractions between the 

 molecules cause them to pack themselves together 

 into a rigid structure. This can happen in two ways : 

 the molecules may come together in any position, 

 or they may arrange themselves in a regular pattern. 

 In the first case the solid is said to be amorphous ; in 

 the second case it is called crystalline. The amorphous 

 and crystalline states may be compared to a number 

 of men forming a crowd, and to the same number 

 drawn up in military' formation. 



In the great majority of cases, the molecules, when 

 forming a solid, arrange themselves in a regular 

 crystalline structure, and it is only under exceptional 

 circumstances that amorphous solids are produced. 

 The mineral constituents of the earth's crust are almost 

 entirely crj'stalline : metals are composed of a mass of 

 small crystals, solutions deposit the dissolved body as 

 crj'stals. Ice consists of a mass of w-ater crystals, 

 frost of the same crystals deposited by the condensation 

 of vapour. When the molecules are given time, as 

 it were, to arrange themselves regular!}', there is 

 always a tendency to form a crystalline structure. 



Glass, on the other hand, is a typical example of a 

 class of solids called " vitreous " which show no 

 evidence of crj-stalline structure. 



When a crystal is not disturbed in its growth, its 

 crj-stalline nature is shown by its having a number of 

 plane surfaces meeting in straight edges, which are 

 often s\Tnmetrically disposed. Although the extent 

 to which these surfaces are developed may differ 

 greatly from crystal to crystal, the angles between 

 corresponding faces are always the same for any given 

 substance. This can only be due to there being for 

 every substance a regular pattern in which the mole- 

 cules are always arranged when it crystallises. 



This is illustrated by the two-dimensional model of 

 Fig. I, where the circles represent the molecules of 

 a crystalline structure. 



The figure shows how the two-dimensional " crystal " 

 will have straight edges, formed by adding rows of 



FIG. J.— TWODIMENSlON.iL MODEL OF CRYSTAL STRUCTURE. 



molecules as the crystal develops, and if we imagine 

 the same process to take place in three dimensions, the 

 way in which the plane faces of a crj-stal grow, by the 

 adding of layer after layer of molecules, can be pic- 

 tured. It is not the possession of plane faces which 



35 



