13 



which shows that the pegmatite is of primary origin i. e. formed 

 simultaneously with the main mass of the rock. In this first 

 stage of mineral formation the primary minerals, felspar and 

 ægirine have in most cases grown so as to meet , leaving no 

 open space between them. But nearly as often the formation of 

 the these minerals has been interrupted , before the space 

 between them had been filled; in such cases open drusy cavities 

 have been the result. In these cavities a number of new minerals 

 have then formed during successive generations. Thus a great 

 number of the pegmatite formations met with italics on the 

 Narsarsuk plateau are quite solid and consist entirely of coarse 

 crystalline masses of the primary minerals. But in other places 

 numerous open drusy cavities are met with, the walls of which 

 are frequently lined with crystals representing later stages of 

 formation. The locality has as yet been very little worked for 

 minerals, and the drusy cavities that have been found have all 

 been small. To judge, however, from the considerable size of 

 many crystals found among the loose gravel, very large cavities 

 must have existed in the parts of the rock which have weath- 

 ered into gravel. The spaces between the more recent crystals 

 are usually quite open. Sometimes they are filled with a black, 

 earthy substance containing a large percentage of manganese. 

 Lastly, several drusy cavities have been filled with some hard 

 mineral substance as eudialyte, fluorite, calcite, etc. 



All the minerals found on Narsarsuk in a quantity large 

 enough to allow of a tolerably satisfactory investigation will be 

 included in the following description. Only the feldspars are 

 excluded, because they are to be examined and described, 

 together with other Greenland feldspars , by another mineral- 

 ogist. Most of the new minerals have been found in small 

 quantities, on account of which the analyses, mostly very 

 difficult, have in many cases been made on very scanty ma- 

 terial. This difficult and important work could , however, 

 thanks to the great kindness and liberaUty of Professor Hj. 



