40 
MANGANESE DEPOSITS OF INDIA : MINERALOGY. [Part I : 
veinlets, although as far as I could determine, by careful examination of 
the specimens to be tested under the microscope, they seemed quite pure. 
The manganese-ore quarried at Garbham is frequently seen to contain 
scattered shining specks which at first sight one takes to be braimite. 
Judging, however, from the magnetism test, these bright specks consist 
more often of manganmagnetite than of braunite at Garbham, thus 
accounting for the high percentage of iron found in these ores. 
At Avagudem a rather common form of ore is one composed of a 
mixture of psilomelane and pyrolusite Avith abundant scattered specks 
of a very magnetic mineral. As at Garbham this magnetic mineral 
is traversed by veinlets of psilomelane, but to a still greater extent. 
With the greatest difficulty a piece was picked out which seemed to be 
free from these veinlets ; this gave only a very weak reaction indicat- 
ing not much more than a trace of manganese. At Kodur the patches 
and specks of the shining mineral that is so often seen in the mauga- 
nese-ores are found to consist more often than not of braunite ; but 
sometimes manganmagnetite traversed by psilomelane veinlets is found. 
As regards the origin of this manganmagnetite it is obvious that 
when it forms a part of the manganese-ores — which, as is explained 
on pages 262 to 272, are of secondary 
origin formed by the chemical alteration, 
with solution and re-deposition, of various constituents of the 
rocks of the kodurite series — it must also have been formed during 
the series of chemical changes by which the ores were formed. In 
the case of the spandite-rock at Garbham in which it occurs it seems, 
however, possible that it may be an original mineral formed at the same 
time as the enclosing rock. Microscopic examination does not 
definitely settle this question. Under the microscope it is seen that the 
garnet is undoubtedly being replaced by manganese-ore (psilomelane). 
In places in the rock there are curious rims or borders of garnet to the 
opaque minerals consisting of manganmagnetite and psilomelane. 
Reflected light shows that o:i one side of the rim is mangamnagnetite 
(grey-black)^ and on the oth(!r psilomelane (duW black) evidently replacing 
the garnet. The manganmagnetite does not show this evident replace- 
ment ; hence it seems possible that the original rock was composed mainly 
of spandite and manganmagnetite and that manganiferous solutions 
have attacked the rock and replaced the garnet, in some places com- 
pletely and in some cases only partially, so as to leave an unaltered rim 
of garnet between the manganmagnetite and the secondarily formed 
manganese-ore ; whilst the manganmagnetite has escaped alteration, 
