6o SCIENCE PROGRESS. 



and enclosing minute sand -grains. These shells were 

 found to be of silica, soluble in caustic soda. With a 

 high magnification they showed very minute rods, which, 

 however, might probably be foreign bodies introduced 

 during the preparation of the specimens. 



It is evident that since bedded iron-ore deposits may 

 admittedly be formed either directly as such or indirectly by 

 the transformation of limestones, the origin of any particular 

 deposit must be a question to be decided by such evidence 

 as is available for the case in question. Some of the criteria 

 which may be applied in discussing the question are well 

 brought out in the papers of Kimball and Smyth on the 

 Clinton ores. For instance, Smyth, in maintaining- the 

 direct formation of these rocks, points to the rarity or 

 absence in the wide area discussed by him of chalybite, 

 which, on the metasomatic theory, must be formed as a 

 stage in the transformation of limestone to haematite. He 

 considers that, if this latter theory were the true one, there 

 should be, at least in some places, ferrous carbonate remain- 

 ing to mark the intermediate stage of the transition. The 

 normal ore is a red haematite, usually of earthy aspect, but 

 sometimes quite compact and with an almost metallic lustre. 

 One type consists largely of fragments of bryozoa, corals, 

 crinoids, brachiopods, etc., coated by, and more or less con- 

 verted to, haematite. The oolitic type is less abundant. 



It will be observed that Smyth himself admits some extent 

 of metasomatic change, at least as regards the organic frag- 

 ments in the Clinton ores. The replacement especially of 

 bryozoan and coral remains is a common feature in many 

 limestones, e.g., in the Carboniferous of the Bristol district. 

 Again, it is not uncommon to find oolitic grains more readily 

 affected than their matrix, and in many of the English 

 oolitic iron-ores we may observe that the grains have been 

 converted into haematite or limonite, while the matrix is only 

 in the condition of chalybite. 



The metasomatic processes in calcareous rocks which 

 result in what may be termed silicification open out a 

 number of interesting questions which cannot conveniently 

 be treated in the present contribution, and the apparent re- 



