Chap. XIV.] dharwaks: crystalline limestones. 
303 
was probably produced at the time of metamorphism of these rocks, 
by an interaction between the surface shell of the original manganese- 
oxide segregation and the surrounding calcareous sediment. I have 
not yet found evidence that any appreciable quantity of the manganese- 
Alteration of man- found ill the crystalline limestones has been 
ganiferoiis limestones, produced by the Subsequent chemical alteration 
of the piedmontite, so that it can for the present be assumed that by far 
the larger proportion of the ores occurring in limestones are the product 
of the direct compression of the originally deposited oxides of manganese. 
In cases where spessartite and rhodonite were formed, however, subse- 
quent alteration has often taken place with the production of inferior 
ores. But as these two silicates have been luucli less frequently formed 
than piedmontite the ores so produced are not in sufficient quantity to 
be of much commercial value. Devi in the Chhindwara district may be 
instanced as a case of the formation of ores in this way. Subsequent 
changes have, however, often affected the piedmontite-bearing rocks. 
At every place where piedmontite limestones are found there is also 
found the black limestone, to the colouring of which by the deposition 
of manganese oxide along cleavage and twinning planes reference has 
already been made on page 300. In all these cases there seems to be a 
certain amount of spessartite and more rarely of rhodonite present, 
but it is not known whether all the manganese secondarily deposited in 
these black limestones has been derived from the spessartite or rhodonite, 
or whether the piedmontite has also contributed its quota. In one case, 
namely at Pali, in the Nagpur district, the limestone seems to have 
been bodily dissolved away with the formation of cavities in which 
Secondary forma- beautiful crystals of pyrolusite have been deposited, 
tion of pyrolusite. manganese required for this process having 
no doubt been derived from the manganese silicates contained in the 
adjacent limestones. In other parts of the same mass of rock at Pali 
Manganiferous cherts, ^^^e whole rock has been replaced by chert, in 
which are often found shining particles of man- 
ganese-ore formed previous to the silicification of the limestone by the 
conversion into manganese-ore of whatever manganese silicate happened 
to be in the hmestone, more probably rhodonite or spessartite than 
piedmontite. 
The nodules of manganese-ore are not usually composed of one 
.Tile mineral compo- mineral onlv. Analysis shows that they are as a 
sition 01 the nian£;anese- r i • i i 
ores. rule mixtures of braunite with a predominating 
amount of one of the manganates, hoUandite or psilomelane. From 
