424 The American Geologist. June, 1894 
siderite mingled with carbon, and in some Swedish "Blandstenar," i. e. 
manganiferous iron ores mingled with carbonates and in many cases 
also with organic matter) can we establish the probability of an original 
separation as carbonate of iron. 
By oxidation according to the formula. 2FeC0 8 +0=Fe 2 3 -j-2C02, 
free carbonic oxide is produced. By this: 
(a) The separation of CaC0 3 (and MgCO s ), and at the same time of 
Fe 2 < >.■}. is hindered. 
[I>\ The precipitation of silica is facilitated. 
(c) The phosphoric acid present in the solution is wholly or partly 
thrown down, if iron oxide (or hydrate) is suspended in the water. 
(d) Manganous carbonate is not so quicklj oxidized as the correspond- 
ing salt of iron. Hence the principal part of the manganese will be 
thrown down at a later period, depending on the local conditions, as oxide 
(hyperoxide), hydrous oxide, or carbonate (together with some of the 
CaCO and MgC0 8 ). 
That the precipitations must actually occur in the order given, — 
viz.. first, oxide of iron, with silica and phosphoric acid, together with 
part of the manganese; next, the bulk of the manganese; and finally, 
calcium and magnesium carbonates. — is proven by many concordant 
observations regarding the deposits of hot springs, salt wells, etc. 
For the sake of comparison it max be said that the lake and bog ores, 
which owe l heir origin chiefly to separation by the oxidation of carbon- 
ated solutions, really consist of hydrated iron oxides, with considerable 
silica and P 2 5 , but on the other hand with very little carbonate, man- 
ganese in increasingly greater proportion, and occasionally traces of 
Ti(> 2 . Cu, Xi. Co. Zn, As, etc. Moreover, at the bottom of the ocean 
have been found manganese concretions, comparable to the lake ores, 
and probably formed by analogous processes. 
Those chemical reactions which follow as a direct result of the oxida- 
tion of ferrous (and manganous) carbonates explain why 
(1) The Archean and Cambrian magnetite and hematite (in the Uni- 
ted States also siderite) deposits occur in close relation with limestone 
(and dolomite). 
(2) The precipitation of the iron ore-limestone-dolomite series was al- 
ways in many places accompanied by the separation of the iron ore beds 
(3) The jron ore, although occurring so often in and near limestone 
(or dolomite), nevertheless is not mingled principally with carbonates 
but with silicates, especially quartz and acidic silicates. The silica 
which was separated at the same time with some of the alumina, lime, 
magnesia, etc.. was precipitated under the same conditions as the sili- 
cates present in the granular limestone, and later underwent the same 
metamorphism; hence the mineralogical and structural analogy. 
(4) Tlie iron ores almost always contain some manganese, partly in 
proportionately significant amount, while i'u. Pb, Zn. Xi, Co. etc.. are 
almost wholly absent. 
(5) We occasionally find in one and the same locality (L&ngban, 
Nordmarken) separated independent beds (a) of siliceous and (b) of 
