MOLECULAR ARCHITECTURE 695 



The slight difference in size between the volumes of the 

 spheres of influence of the atoms of two kinds in a binary 

 compound would account for the slight divergence in hexagonal 

 crystals of the axial ratio a : c from the ideal value. 



Barlow and Pope {Trans. Chem. Soc. 1907, 1170 ct seq.) have 

 worked out in great detail the spacial distribution of the atoms 

 in a number of binary compounds, particularly in the cases of 

 silver iodide and a number of halogen salts of the alkali metals. 

 Only a few of the results of this investigation can be quoted. 

 The case of silver iodide is particularly interesting. This 

 substance is dimorphous, crystallising at the ordinary tem- 

 perature in the hemimorphous pyramidal class of the hexagonal 

 system, the ratio «: <: being i :o'8i96; whilst above 147° C. it 

 is stable in a cubic modification. These tw^o modifications of 

 course correspond to the hexagonal and cubic closest-packed 

 assemblages of equal spheres and since the axial ratio a:c oi 

 the hexagonal form is very nearly identical with the ideal 

 value I :o"8i65, it is highly probable that the silver and iodine 

 atomic spheres of influence are very nearly equal in size. 



It has long been known that when a substance is dimorphous, 

 its two crystalline forms are only stable the one above and 

 the other below a certain temperature termed the transition 

 temperature. If the modification stable at lower temperatures 

 be gradually heated, when the transition temperature is reached 

 and passed a change occurs — sometimes suddenly and with 

 violence as in the case of silver iodide— into the modification 

 stable at the higher temperature. It is of great theoretical 

 importance that Barlow and Pope are able to show that in the 

 case of silver iodide, the two modifications, one hexagonal and 

 the other cubic, are very closely related. A cubic assemblage 

 of spheres is deduced representing silver iodide and it is 

 shown that by a slight shear of alternate layers this assem- 

 blage becomes transformed into a hexagonal form possessing 

 hemimorphous symmetry corresponding to the observed hemi- 

 morphism of silver iodide. Similarly, of course, a shear in the 

 reverse direction changes the hexagonal back into the cubic 

 assemblage. It may well be that when hexagonal silver iodide 

 is heated, the change in temperature produces a small change 

 in the relative volumes of the spheres of atomic influence of the 

 two kinds, as a result of which, above 147° C, change takes 

 place by means of the slight shear already described into the 



