114 



ROGERS— THE VALIDITY OF 



[March i, 



than the b-a.xis. Out of 59 monoclinic minerals with a less than 

 unity, (010) occurs on 56, but (100) on only 46 while (120) occurs 

 on 23, but (210) on only 13. Out of 64 monoclinic minerals with 

 a greater than unity, (100) occurs on 60, but (010) on only 43 while 

 (210) occurs on 16, but (120) on only 13. If a is shorter than b, 

 the molecules are more closely packed along (010) and (120) than 

 along (100) and (210), but if a is longer than b the reverse is true. 

 Out of 168 combinations of orthoclase (including microcline) 

 crystals given in Hintze's " Handbuch der Mineralogie," (010) 



100 



Fig. 4. The probable structure of orthoclase. 



occurs on 133, but (100) on only 22. The form (130) occurs 70 

 times but (120) occurs only once! This remarkable case is ex- 

 plained by assuming the structure to be that of the monoclinic or 

 clinorhombic prism, one of the space-lattices of Bravais. It can be 

 seen from Fig. 4 that the molecules are more closely spaced along 

 (130) than along (120). 



It is certainly true that the form-series differs for various crys- 

 tals, but according to the law of complication the form-series should 

 be alike for all crystals and all systems. ^^ 



For those who are familiar with chemistry, the whole matter 

 of indices, rationality, crystal structure, and relative frequency of 

 crystal faces may be cleared up by considering the analogy between 

 the fundamental laws and theories of chemistry and those of crys- 

 tallography. 



" Goldschmidt (Joe. cif.) explains the differences in the form-series by 

 assuming outer disturbing influences. These undoubtedly have an effect but 

 certain discrepancies are more easily explained by the law of maximum 

 reticulate density. 



