Chap. V. ] 
RHODONITE. 
143 
practically the whole of the rock. Then there are varieties of rhodonite- 
rock containing small quantities of spessartite or quartz, or of both ; and 
from these there is every gradation through rocks composed of roughly 
equal quantities of rhodonite and spessartite, or more rarely of rhodonite 
and quartz, up to rocks composed practically entirely of spessartite or 
quartz. WTienever the rhodonite occurs in rocks with spessartite 
it seems to give the garnet a good opportunity to assume a crystalline 
form. On rare occasions, however, I have seen under the microscope 
rounded idiomorphic crystals of rhodonite enclosed in spessartite, as for 
example in a rock from Jothvad. As will be seen from the list of rocks 
of the gondite series given on page 329, this mineral may also be 
associated with rhodochrosite, amphibole, orthoclase, magnetite, 
manganese-mica, and barytes. In the rocks of the gondite series the 
rhodonite individuals are sometimes as large as 1 or even 2 inches 
across, then exhibiting well their perfect cleavage, and often enclosing 
other minerals, especially spessartite, poikilitically. 
In the manganese-pyroxenites of the Yizagapatam district the 
rhodonite usually occurs in prismatic individuals together with the other 
pyroxenes in the rock. Owing to the rather loose state of aggregation, 
in some cases, of the granules of these pyroxenites, these prismatic 
individuals are often easily detached and are sometimes as much as 
an inch in length. 
The appearance of a piece of rhodonite-rock under the microscope is 
^. shown in figure 3 of Plate 12, which illustrates why 
the rock often resembles a crystaUine hmestone in 
structure as seen in the hand-specimen. For the rock to be so fresh as 
in this example is, however, not the rule ; for this mineral is very 
prone to alteration. This alteration seems to start along the cleavage 
cracks of the mineral as shown in figure 4 of Plate 12, and then to 
spread in all directions so as to leave smaller and smaller clear areas 
between the altered portions, the final result being a blackened mass 
of manganese and iron oxides. When the alteration is taking place, 
the product seen along the cracks is not aU black in colour, but 
areas and strings of a curious orange to yeUow-brown substance, 
polarizing as an aggregate in the way serpentine sometimes does, are also 
formed. The nature of this substance is not known, but it may be an 
intermediate product between rhodonite and manganese oxide. In some 
cases, moreover, the alteration of the rhodonite does not seem to take 
place along the cleavages in particular, but to spread out irregularly over 
the mineral in moss-like or dendritic forms, and as irregular cloudings, 
