BY D. MAWSON. 597 



land material, should give the best results on analysis. Extreme 

 variation in the sesquioxide constituents in the two analyses 

 quoted in Dana* makes it plain that this latter material was not 

 homogeneous. 



Decomposition products. — The decomposition products vary 

 with the conditions operating. Carbonation is most in evidence 

 though in some situations, where carbon dioxide supply has been 

 inefficient for complete carbonation, oxidation is prominent. 

 Illustrative examples were noted in one of the slides in which 

 several small arfvedsonite fragments appeared included in sani- 

 dine crystals; in one case, where the sanidine envelope was 

 specially perfect, the amphibole remained fresh; where the sani- 

 dine proved a less perfect shield, oxidation was prominent; else- 

 where the soda-iron silicates were completely carbonated. For 

 illustrations showing an early and a late stage in primary 

 arfvedsonite decomposition, see Pl.li., figs. 3 and 4. 



Very commonly, especially in veins of class (6), subclass 2. 

 type /3, carbonation and oxidation are both evidenced, the product 

 being an iron oxide-stained siderite. 



Examined critically under a high power, the first stage in the 

 breaking down of the amphibole appears to be the separation of 

 magnetite in dust-like particles, and an irresolvable base which 

 is capable of taking a stain after treatment with hydrochloric 

 acid; it seems likely that this latter substance is a sodium silicate. 

 The attack has, in all probability, been made by solutions of the 

 alkali carbonates, and the effect has been, first, the separation of 

 magnetite, sodium silicate, and iron carbonate followed closety 

 by complete carbonation, when practically all the soda and silica 

 were removed from the original site, the latter largely precipitated 

 later in trunk channels. In some cases, even after the whole indi- 

 vidual has suffered alteration, stainable matter remains in small 

 quantity amongst the debris. Chalcedony is likewise found to a 

 small extent scattered sporadically amongst the carbonates, 



* Eecent literature on the alkali-iron amphiboles quoted by Murgoci is not 

 available here in Adelaide for further reference. 



