1921. No. 8. ox THE ECLOGITES OF NORWAY. qq 



in the omphacite habit, this reel: should be called hypersthene- 

 e c log i te. Here the clinopyroxene is a brown schillerizing variety show- 

 ing most distinct diallage-parting and being filled up with pigmentary 

 inclusions which in part make the mineral almost opaque. 0:7 = 45". 

 Axial dispersion is dictinct : z > j. This pyroxene makes up about two 

 thirds of the rock-mass. 



Next in abundance is the brownish red garnet, perfectly free from 

 any inclusions. 



The third main constituent is the hypersthene. Its colours are: 

 X ^ brownish, ,3 = reddish brown, 7 = greenish grey. The optical cha- 

 racter is negative and the axial angle about 75" or 80'. It should there- 

 fore contain some 25 of 30 per cent FeO. 



The rock is free from feldspar, but there occur spots of saussuritized 

 labradorite-rock which, under the microscope, give the impression of miarolitic 

 cavities, crystal edges of all the minerals protruding from the surrounding 

 pyroxene-garnet-rock towards their centres. Towards these spots the clino- 

 pyroxene has developed thin zones of greenschist products, whilst, on 

 the outer boundaries of the segregations, there are a little thicker zones 

 of amphibolite products. 



As minor constituents there are grains of iron ore with kernels of 

 rutile. The latter is here, as in all rocks of eclogite character, decidedly 

 the first formed mineral preceding the ilmenite which often accompanies it. 

 The iron ore has sharp and clean boundaries towards the pyroxenes, but 

 where it meets the feldspar, either in the miarolitic cavities or at the outer 

 boundaries, there it is surrounded by a zone of posterior garnet in small 

 grains, in part showing a myrmekite-like intergrowth with the feldspar. I have 

 frequently seen this garnet in contact with the primary garnet, but it was 

 not possible to decide the relation of their réfringence. Apparently, how- 

 ever, they are different in composition, as they are in structure. A little 

 reddish brown biotite has also grown around the ore-grains. 



The structure of this rock is most interesting. The sequence of idio- 

 morphism is distinct: diallage, garnet, hypersthene. This is the only rock 

 in which 1 have seen xenomorphic garnet, in a striking manner filling up the 

 interstices between the clinopyroxene-grains (fig. 15, table III). But where 

 the garnet meets a crystal of hypersthene it assumes its usual rounded 

 grain-shape and idiomorphic relation. 



The great importance of this observation lies in the fact that we have 

 a conclusive proof that the garnet is here a mineral that has crystallized 

 direcdy from the magma. The idiomorphism of the garnets in our labradorite- 

 rocks and eclogites, then, need not be regarded as the result of any greater 

 energy of crystallization or other crystalloblasdc properties, but simply to 

 the prior separation, as in the other minerals in the igneous rocks. 



