I 92 I. No. 8. ON THE ECLOGUES OF NORWAY. I 03 



The specimens trom Ramsgronaave consist of garnet in ideal dode- 

 cahedra embedded in a medium-grained mass of light green clinopyroxène 

 with pale green hornblende, brown rutile, quartz and apatite. The p\Toxene 

 shows c: Y = 40", and a positive axial angle larger than that in common 

 diopsides. These properties conform, with those found in the chloromelanitic 

 P3Toxene in the eclogite from Duen, ^'anelvsdalen. 



Certain bands in this eclogite contain numerous scales of light brownish 

 \'ellow mica showing a small axial angle. 



The perfect idiomorphism of the garnet is the mst peculiar character 

 of this eclogite. 



The specimens from Hordens, Hellevik, show a very peculiar structure. 

 Visible to the unaided eye are ore crystalline aggregates in an aphanitic pale 

 green mass. This mass consists almost exclusively of finely granular 

 pyroxene, in which flakes of colourless mica occur in the manner of 

 porph^Toblasts. The crN-stalline aggregates consist of granular monoclinic 

 pyroxene forming the central parts, surrounded by a granular mass of 

 garnet, whose individual crN-stals show ideal dodecahedral forms towards 

 the finely crystalline pyroxene mass forming the matrix. The minerals ot 

 these eclosrites show no sigrn of anv alteration. 



GARXET-AMPHIBOLITES .AND G.ARNET-BEARING 



NORITES. 



By studying the available analyses of garnets I had found the garnets 

 in the garnet-amphibolites and gabbroid rocks alwa\-s to contain almandite 

 as the chief component. As such analyses were not ver\' numerous, it 

 seemed desirable to make some new investigations. Further, it could be 

 presumed that garnet-bearing amphibolites had been eclogites before their 

 metamorphism into the amphibolite faciès. I therefore decided to determine 

 whether there could be detected any relict features that would throw light 

 on the question. 



As the separation of the garnet in garnet-amphibolites is usually a 

 difficult task, the mineral being generally stained with very minute inclusions 

 of quartz and iron ore that in the aggregate frequently have the same weight 

 effect as the garnet, I did not take the trouble of separating garnet in a 

 pure form in more than one case, that of the gamet-amphibolite from 

 Kantalahti. In other cases the mineral was used in the state of purity 

 resulting after one separation with the Clerici solution. As the analyses 

 thus have but a diagnostic value, to an approximate determination of their 

 constituent isomorphous compounds, the determination of the state of oxida- 

 tion was here omitted and all the iron was calculated as FeO. 



