400 Edtcarcl T. Sarchnan — On Scsmatite in Tyrone. 



carbonate of lime being formed as already shown (see note ante), 

 and would be carried off by the acid water. 



(2). When alkaline silicates are allowed to act on silicates of 

 alumina, they extract minute quantities of alumina ; and silicate of 

 soda possesses this property to a very sensible degree : nioreover, the 

 alkaline silicates dissolve small portions of the silicate of alumina 

 itself.^ Thus in time a considerable quantity of alumina might be 

 removed. 



Now all these are solutions we should expect to find in the vicinity 

 of strata made up of calcareous and arenaceous rocks. It is of course 

 well known that the water of limestone districts often contains a 

 large amount of sulphate of lime, to which is due what is called its 

 "permanent hardness"; on the other hand, the solution of alkaline 

 silicates may result from the action first of carbonic acid on minerals 

 containing them, and then the subsequent re-action of the alkaline 

 carbonates so obtained on silicate of lime. 



By these agencies then, (1) and (2), the silicates of alumina 

 forming the clay must have been for the most part extracted. It 

 will be hardly necessary to remark that carbonic acid would have no 

 direct effect on them, as it does not combine at all with aluminium. 



Thus the clay-ironstone results in brown haematite more or less 

 free from siliceous and clayey particles, according to the extent to 

 which the alteration has been carried. The final process seems to 

 have been the deposition from solution of the thin external layer of 

 pure fibrous hydrate. 



While it is thus shown how such a change is theoretically possible, 

 it is satisfactory to obtain undoubted evidence that this metamorphosis 

 of clay -ironstones of various kinds really does take place ; and it 

 may, I think, be proved pretty conclusively by the following 

 example. 



On the southern shore of Lough Neagh is found a great thickness 

 of beds of — sometimes very arenaceous — pottery clays, with sand, 

 and occasionally beds of lignite, besides concretionary sandstones, 

 and nodules of clay-ironstone. The whole has been correlated with 

 the Bovey Tracey deposit, but is, I believe, of somewhat more recent 

 date. The ironstones are very hard and compact,^ sometimes of a 

 slightly greenish colour, and seem to consist chiefly of proto-silicate 

 and carbonate of iron ; but the outside is often composed of a thick 

 crust of hard, close, siliceous brown haematite, evidently the fruit of 

 the chemical alteration of the original nodule. One of them is 

 represented in the accompanying Woodcut (see Fig. 3). The ex- 

 ternal shell of haematite (b) is nearly a quarter of an inch thick, 

 and it almost completely incloses the unaltered ironstone, being 

 broken to show the interior. A qualitative examination was made 

 by me of a portion of the unaltered part of one of these nodules 

 with the following result. 



It did not effervesce at all with dilute acid ; but when powdered, 

 and treated with strong hydrochloric acid, and gently heated, violent 



1 Bischof s Chein. Geol., vol. ii. p. 66. 



2 They contain reed-like plants, still retaining the woody tissue. 



