1 92 1. No. 



ON THE ECLOGITES OF NORWAY. 



Sp.gr. = 3.146 It = 12 I. 



Ill, 4, 4, 3, Koghose. 



In the present case it would be useful to have a direct bulk analysis 

 of the rock to check this calculation whose premises are not quite accu- 

 rate. Meanwhile, however, we may use the figures arri\ed at and we will 

 find some peculiar features that can be stated with full confidence. Such 

 is the great preponderance of ferrous oxide over magnesia, appearing in 

 both the chief mafic constituents. Another peculiar characteristic of this rock 

 is the comparatively high percentage of silica combined with a low proportion 

 of alumina, appearing mineralogically in the large quantities of quartz, in 

 a rock whose other main constituents are hornblende and almandite. The 

 arrangement of the quartz in the form of elongated stringers would perhaps 

 suggest the idea that silica might have been added during some metamorphic 

 processes whereby the bulk composition would have undergone consider- 

 able changes from an original composition more normal gabbroic. 



The following considerations, however, do not seem to be in favour 

 of such an hypothesis. For the first, doubtless primary volcanic and abyssal 

 rocks sharing this chemical character are not by any means rare, and there 

 are, in Washington's tables, closely sim'ilar analyses of koghoses. For the 

 second, it is known that, in a northwesterly direction from the Kantalahti 

 district, almandite-bearing granitic rocks, so-called granulites or leptynites, 

 underlie immense areas in Finnish Lapland and Norwegian Finnmark. These 

 facts suggest a genetic connection between these granulites and the Kanta- 



