Chap. XVII.] gondite series : alteration. 
363 
Whether the reagent was carbon dioxide or sulphuric acid, there 
Rarity of open should be loose or cavernous masses of rock left 
rwk^." '° *^ ^ wherever the solvents attacked and dissolved with- 
out depositing something in return. But, except for occasional spaces 
in the ore-body, such cavities are not foimd. This may possibly mean 
that the changes took place prior to the last set of compressing earth 
movements that affected the rocks of this area. If the earth movements 
that faulted and folded the Gondwana rocks can be supposed to have 
affected the ancient crystalline rocks to any serious extent, then they 
would possibly have been sufficient to close up the spaces hj'pothecated 
above. But the probability is that these movements would have 
been insufficient to close up spaces formed in narrow bands of rock 
that formed but a very small proportion of the already highly com- 
pressed and therefore highly resistant rocks of the Archaean complex. 
Hence if these spaces really did exist and have since been closed up by 
compression, it is necessary to go back to the series of earth movements 
that affected all the Archaean rocks at the end of the Dharwar period. 
At first sight this may seem to be inconsistent, because these move- 
ments have been already called in to explain the metamorphism of the 
manganese- oxide sediments with the formation of the gondite series. 
It does not follow, however, that the folding that took place at the end 
of the Dharwar period of sedimentation, consisted of one prolonged 
compressive movement. We can imagine a period of compression 
during which the more deeply lying portions of the manganiferous 
sediments were metamorphosed to the condition of the gondite series, 
followed by a period in which the pressure was released. On referring 
to page 291 it will be seen that the formation of the manganese silicates, 
Further reasons for spessartite and rhodonite, involved the liberation 
trn'tor'ikof'tn Of oxygen. It is not necessary that this oxygen 
Arehaan times. should have been removed very far from the 
compressed rocks, and this may be the very oxygen that, aided by 
carbon dioxide or sulphuric acid, produced the conversion of a portion 
of the manganese siUcates back into manganese-oxide ores, according to 
the equations given on pages 357 and 362. This period of diminished 
pressure may have been of some duration, gi^ang time for the conver- 
sion of a considerable proportion of the manganese-silicate-rocks into 
manganese-ore. Any open spaces thus formed would be closed up 
by another period of compression, sufficient to close the spaces and 
yet not sufficient to again convert manganese oxides into sUicates. If 
there be any truth in these considerations then we arrive at the 
