ESSAY-REVIEWS 91 



lightened by numerous well-executed drawings of solid models, 

 some of which, by the courtesy of the publishers, are here 

 reproduced. On the other hand, the complementary task of 

 giving the crystallographer the necessary introduction to the 

 science of the X-rays is less successful. For example, we read 

 in Chapter IV. — "The next table is well known." Yes! but to 

 whom ? One unfamiliar with the subject is plunged into very 

 intricate phenomena at this point, with the minimum of help 

 and guidance. 



The first application of the new method made to the problem 

 of crystal structure at once showed its power. A set of 

 isomorphous cubic crystals, sodium chloride or rock-salt, NaCl, 

 potassium chloride or sylvine, KC1, potassium bromide, KBr, 

 and iodide, KI, gave, by reflection from the (100) face, a set of 

 values for the various crystals of d, the spacing between the 

 planes, proportional to the cube-root of the molecular volumes 

 of the crystals. But potassium chloride alone gave for the 

 spacings for the (100), (1 10), and (1 1 1) faces the ratio i:v / 2: v/3, 

 required of a simple cubic space-lattice. An examination of the 

 other crystals showed that potassium chloride was apparently 

 simpler than the other salts because its constituent atoms, 

 potassium and chlorine, being nearly identical in mass, act 

 indistinguishably towards the X-rays. For the other salts the 

 behaviour was completely explained on the view that each of 

 the two sets of atoms, of alkali-metal and halogen respectively, 

 lie on two different face-centred cubic space-lattices intersecting 

 in such a way that when the two sets of atoms are alike the 

 form reduces to the simple cubic space-lattice. This showed at 

 once that it is the individual atoms, and not the molecules, which 

 act as the diffracting points in the crystal. 



This set of crystals is depicted in fig. 2. The black and 

 white dots represent the two kinds of atoms. The (100) planes 

 are those parallel with ADHE, the (no) planes are parallel to 

 that containing the points CDEF, and the (hi) planes are 

 parallel with that containing the points BDE. It will be seen 

 that in the (100) and the (no) planes the atoms consist of equal 

 numbers of both kinds, but the (in) planes consist alternately 

 of all halogen and all alkali-metal atoms, the one kind of plane 

 being midway between two of the other kind. This is shown 

 in the figure in the diagrams below the crystal-model. 



Another cubic crystal, zincblende, ZnS, was found also to 



