RECENT PROGRESS IN CRYSTALLOGRAPHY 105 



which one axis is otherwise inclined, the amount of this in- 

 clination changes when one alkali metal is replaced by another 

 to an extent which is directly proportional to the change in 

 atomic weight. 



The result of the investigation of the exterior form of the 

 crystals is thus to reinstate permanently, as a law of nature, 

 Haiiy's view that difference of chemical composition is accoro 

 panied by difference of crystalline form. At the same time, 

 Mitscherlich is shown to be approximately correct, for his 

 isomorphous series are so nearly alike in external morphology 

 that it has required the most careful research to discover and 

 establish the fact that there are real differences, other than for- 

 tuitous ones due to malformation. The three above-mentioned 

 quantitative laws, which have been shown to govern the differ- 

 ences between perfect crystals of the various salts, form the 

 harmonising bond of reconciliation between the work of Haiiy 

 and Mitscherlich. 



Fig. 8. 



Fig. 9. 



Fig. 10. 



A further fact has been observed with regard to the relative 

 development of the various faces of the crystals, which deter- 

 mines what is known as the " habit." It is most marked in the 

 case of the monoclinic double salts, and figs. 8, 9 and 10, 

 representing respectively typical crystals of potassium, rubidium, 

 and caesium salts, will at once make it clear. The faces of the 

 principal forms, c, q, and p, show a definite progression, any 

 one of them on the rubidium salt being developed to an 

 intermediate extent compared with its development on the 

 potassium and caesium salts. 



The question of the internal structure of the crystals must 

 now be introduced. Crystallographers had been somewhat 

 thrown off the scent for a time by the idea, prevalent among 

 physicists, that in solid substances we were dealing with edifices 

 composed of units which were aggregates of chemical molecules, 

 and attempts had been made from time to time to determine the 

 number of chemical molecules contained in the physical one. 



