396 



BELL SYSTEM TECHNICAL JOURNAL 



of crystallographers on the forms which crystals assume and the ways 

 in which they act on Hght lead to conclusions about the symmetry of 

 these elementary particles; and it is found that while they usually have 

 some degree of symmetry, they do not have that full degree which the 

 sphere represents. In later sketches, therefore, I have followed Bragg 

 in representing the particles by perfectly unsymmetrical figures, like 

 large commas (Figs. 4 and 5). Only, they should be unsymmetrical 

 in three dimensions instead of only two; the reader may conceive each 

 comma as being rough on one side, smooth on the other. 



Fig. 2 — A cubical space-lattice, with particles of full spherical symmetry indicated 



at the lattice-points. 



The next obvious question is: what is the relation between these 

 particles, and the atoms of the crystal if it is a crystal of some chemical 

 element, or the "molecules" if it is a crystallized chemical compound? 

 One might expect that they would be the same; but one would usually 

 be wrong. When for instance gaseous potassium or argon condense 

 into crystals, it takes four atoms of the argon gas to make up the 

 elementary particle of the argon crystal, while that of the potassium 

 crystal consists of two atoms.^ With chemical compounds the crystal 

 particle may be the molecule, part of the molecule or a group of several 

 molecules. I will consequently use for it hereafter the colorless name 

 atom-group, with occasional variation by the conventional but not 

 very descriptive term unit cell. 



Now to visualize the diffraction-pattern, let us conceive the crystal 

 set at the centre of a transparent bulb painted inwardly with some 

 fluorescent matter, so that wherever X-rays or electrons fall upon its 



^ Provided that we conceive the lattices as cubic, which is customary. There 

 happens to be in these cases something like what in other fields of physics is called a 

 "degeneracy": one and the same lattice may be viewed either as cubic or (with a 

 different choice of axes) as non-cubic, and with the latter conception the "particle" 

 turns out to be a single atom. 



