162 
MANGANESE DEPOSITS OF INDIA : MINERALOGY. [ PaRT I : 
The garnets found in Nature seldom conform exactly to the foregoing 
formulae, but are usually found on analysis to be 
Isomorphous replace- . , t , • •,• i , . p ,i 
,neiit. mtermediate m composition between two oi the 
ideal species. This is usually ascribed to the 
isomorphous replacement of the various metals in the protoxide and 
sesquioxide groups one by the other. Whilst it is convenient to use this 
term ' isomorphous replacement it is as well to remember that in 
all probability this replacement cannot take place within the molecule ; 
but that the phrase really means that there is a molecular mixture of 
molecules of the same structural formulae and the same or nearly the 
same crystalline form. Hence these garnets are best regarded as made 
up of mixtures of the molecules of two or more of the type garnets 
in varying proportions. The garnet receives its name according to the 
predominant molecule. 
All the six species of garnets may contain a certain pro- 
portion of manganese in the RO groupi. But in connection with the 
Indian manganese-garnets it will only be necessary to consider three 
species, namely, spessartite, andradite, and grossularite. 
The published analyses of spessartite often diverge widely from the 
formula given above. The commonest divergence is for the MnO 
to be to a considerable extent replaced by PeO, the mineral then 
gradating towards almandite ; and it is obvious that a point may be 
reached at which it is impossible to decide whether to call the garnet 
spessartite or almandite. In the same way the AI2O3 may be replaced to 
a considerable extent by Fe203. No garnet seems to be known 
conforming to the formula 3MnO.Fe203.3Si02,2 so that such a garnet 
as the above would still be called spessartite. If, however, the replace- 
ment of AI2O3 by Fe203 be accompanied by the replacement of MnO by 
CaO, it is obvious that here also we may have a garnet lying about half- 
way between two type species, in this case spessartite and andradite. 
Of the garnets that have been analysed none seem to occupy this 
intermediate position exactly, there being as a rule little difficulty in 
grouping them either as spessartite or andradite, although the 
mineral so called may differ widely from the theoretical composition. 
To obviate, however, this application of one of these names to a mineral 
that differs considerably in composition from the typical garnets, 
special names have sometimes been given ; consequently there are 
1 What IS really meant in saying this is that mixed with the moleo jles of all the 
garnets miy be a certam proportion of the molecules correspondiiig to .spessartite. 
2 Uujojs calderite be really a proper apeoiea corresponding to this foriauliw 
