538 Henry C. Richards : 



From this table it will be seen how close the microscopically 

 estimated cheniical composition is to that determined by 

 chemical analysis. 



The silica has been slightly under-estimated, but as about 

 50 per cent, of the rock is a fine-grained giouiidinass with grains 

 averaging only .01 mm. in diameter, their discrimination by 

 means of the difference of refractive index was a matter of some 

 difficulty. 



The discrepancies between the A].A);h percentages and those 

 of the FeO and MgO suggest that the analysis of the por- 

 phyritic biotite recorded above is not representative of all the 

 biotite in the rock. 



Under the microscope, besides the large phenocrysts of bio- 

 tite one sees developed granular masses of what appears to be a 

 secondary biotite around the ilmenite, and it has been suggested 

 by Pr'ofessor Skeats that this secondary mica arises as a result 

 of interaction between the ilmenite and felspar of the ground- 

 mass under certain conditions, and also between the hyper- 

 sthene and groundmass less commonly. If that is the case it 

 cannot be assumed that ;ill the biotite throughout the rock is of 

 uniform composition, but that at least we have two different 

 kinds of biotite. 



With biotite arising from ilmenite and the groundmass, one 

 would expect a low MgO percentage, as the ilmenite is very low 

 in this constituent, while the groundmass felspar has none ; if 

 this percentage were low the Al.X):^ value would probably be 

 high, so that if one knew what composition to assign to this 

 secondary biotite it is highly probable that the Al./).,, FeO and 

 MgO percentages would be more comparable. 



The alkalies and lime are remarkably close, so that the 

 felspars appear to have been correctly determined. 



The ferric-iron and water are close, while disci-epancies arise 

 between the TiO.^ and MnO values. On the whole, however, 

 the analyses are as close as one could expect to get them, con- 

 sidering the structure of the rock and its fine-grained ground- 

 mass, and if one could estimate the composition of the secondary 

 biotite it is thought that the I'oiuparison would be still closer. 



Another point showing the closeness of tlie two analyses is 

 that the position of the rock, according to the American classi- 



