Chap. XVII.] gondite series : alteration. 
361 
Mn203 in the R2O3 group would be attacked, with the solution of the MnO 
portion and the leaving behind of the Mn02 portion. Hence the result 
of the attack of solutions carrpng sulphuric acid would be to remove both 
the alumina and the sUica, and also to remove all the manganese present 
in the protoxide condition, and a portion of that present in the sesquioxide 
condition, the residue being the peroxide of manganese. Since the pro- 
toxide portion of the spessartite is far more important in quantity than 
the sesquioxide portion, the result of the attack of sulphuric acid solu- 
tions would be the removal of the greater portion of the manganese as 
manganous sulphate. 
Such attack, with the solution of manganese as sulphate, might go 
on until the solution was saturated with the various substances dissolved. 
We must then suppose that on coming in contact with a fresh portion of 
the gondite or spessartite- rock the tendency would be for a further portion 
of the alumina and silica of the garnet, and of the silica of the quartz, to 
pass into the solution, with the deposition of a corresponding quantity 
of manganese oxide, which would be added to the manganese oxide, and 
other constituents necessary for the formation of manganese-ore, already 
in the spessartite. In the presence of oxygen in the attacking solutions 
the formation of the manganese-ores may have taken place according 
to the equation shown on page 270, in which the AI2O3 and Si02 passed 
into solution, with the precipitation of an equivalent amoimt of MnO, 
which was converted into a higher state of oxidation by the oxygen. It 
might be objected that if the attacking solutions contained both 
manganous sulphate and oxygen, the manganese should have been 
deposited as oxide without the intervention of the garnet. It must be 
noticed, however, that the MnS04 was supposed to have passed into 
solution by the reverse reaction, which would not be imdone except 
with a change in the conditions ; and that the experiments of 
F. P. Dunnington^ show that the manganese of manganous sulphate 
is not precipitated by the action of oxygen alone, but that the presence 
of some such substance as CaC'03 (which in Nature would probably 
be found in the form of limestone) is necessary, the deposition then 
taking place according to the following equation : — 
JlnSOi T CaCOs + 0 = CaS04 + ilnOa + CO2. 
n this particular case the constituent to take the place of the 
manganese in the sulphate would be aluminium instead of calcium. 
1 Amer. Jour. Sci., 3rd Series, XXX\a, p. 177, (1888). 
