6 HOLLAND: PECULIAR FORM OF ALTERED PER1DOT1TE 



III.— ORIGIN OF THE ROCK. 



The supposed composition of the magma, estimated by remov- 

 ing the carbonic acid and water from the bulk analysis of the rock, 

 is sufficiently near that of a fresh dunite to render it highly probable 

 that such was the original nature of the magma. The presence of 

 chromium and absence of lime and alkalies are also in agreement 

 with this idea. 



In other areas in South India, where magnesite veins are 

 abundant, we have direct evidence of their formation by the altera- 

 tion of olivine-rock. Picrolite and talc are also found as secondary 

 minerals in such areas ; but, as previously pointed out, there is no 

 general serpentinous alteration of the olivine-rocks. 1 



In the present instance, no unaltered relic of olivine-rock has 

 been preserved, and but for the previous knowledge of such instances 

 as the " Chalk Hills " near Salem, it is highly likely that one would 

 not have suspected such a rock as that described in this note to be 

 a direct product of the alteration of a dunite. 



In the common case known to us in South India the magnesite 

 is distributed along veins, which vary from microscopic infillings of 

 the irregular cracks in an olivine crystal to thick veins a foot or two 

 across. These have been referred to the action of carbonic acid 

 from deep-seated sources attacking the olivine, and such action may 

 have occurred at any time after the solidification of the rock. But 

 the present instance is different and indicates a greater intimacy 

 between the carbonic acid, water and olivine-rock magma ; for the 

 breunnerite crystals have grown outwards in the matrix in a way 

 which indicates a certain degree of free molecular translation. One 

 at first considers the possibility of the magma containing sufficient 

 carbonic acid and water imprisoned so as to give rise in direct 

 combination with the dunite material to a primary formation of 

 these carbonates and hydrous magnesian silicates. Unfortunately 



1 Holland: Geology of the neighbourhood of Salem — the Magnesian Series 

 of the " Chalk Hills." Mem., Geol. Surv., Ind., XXX, p. 133. 

 ( 6 ) 



