GEOLOGY OF MOUNT MARCY 4I 



growths, repeatedly observed, of fine lamellae of monoclinic augite 

 in these enstatites." Plate i8A is a photograph of an area of pyrox- 

 ene, in the ground rhass of anorthosite, which is not surrounded by 

 garnets, but which shows a good development of a green amphibole 

 thought to be uralite. 



From one of the specimens with a favorable reaction rim, enough 

 of the garnet was sorted out for the making of duplicate chemical 

 analyses which give the following as an average: 



Si02 40.11 



AI2O3 22.90 



Fe203 .60 



FeO 25.31 



MgO 4.46 



CaO 6.19 



MnO trace 



Ti02 none 



99-57 



An attempt to recast this analysis indicates that the elimination 

 of quartz and feldspar was not complete. It shows, however, the 

 presence in the garnet of the almandite, pyrope and grossularite mole- 

 cules. Their ratio to one another is probably about 



3 parts almandite, 3FeO, AI2O3, sSiOg 



I part pyrope — sMgO, AlgOg, 3Si02 



I part grossularite 3CaO, Al O3, 3Si02 

 The amount of grossularite is open to question since some of the 

 calcium may have come from labradorite. 



A summary of the facts shown by megascopic, microscopic and 

 chemical examination follows: 



There are in this rock aggregates of pyroxene made up mostly of 

 a variety of augite and partly of hypersthene with accessory magne- 

 tite and interstitial quartz. This quartz shows no strain shadows. 

 These aggregates may or may not be surrounded by garnet rims. 

 Where the garnet rims are lacking, there is a greater development of 

 uralite than when they are present. 



The garnet rims are composed of garnets containing the almandite, 

 pyrope and grossularite molecules, and of unstrained quartz grains. 

 The ground mass is as a rule a normal granulated anorthosite. In one 

 case it is an aplite showing strain and containing grains of broken 

 labradorite. Unfortunately, the hand specimen is not sufficiently 

 large to show the relationship between the aplite and anorthosite, 

 and this relationship must be left to inference on a very meager 

 basis of fact. 



