244 



William E. Ford, 



considered, — .003; for the extraordinary index the average 

 difference is .005 or — .001. If Analysis 17 could also be 

 eliminated it would make these differences still smaller, but 

 including this, the averages are well within the allowable limits. 

 There can be, therefore, no doubt but that the refractive indices 

 for these minerals do vary directly with the composition and 

 can be calculated from the analyses with considerable precision. 



^t^/MCO, 



Fig. 5. 



It may be pointed out here that the small variations in molec- 

 ular volumes that these minerals show and which have a distinct 

 influence upon their specific gravities, have no apparent effect 

 upon their refractive indices. The latter would, therefore seem 

 to be controlled by the elements present and their arrangement 

 in the molecular net-work but not by the spacing of the molecular 

 planes. 



It will be of interest to combine the relations existing between 

 the chemical composition of these minerals and both their specific 

 gravities and refractive indices in order to see how far we can 



i 



