Chap. IV. 2.] metamorpiiic docks. 103 



mag-nesite^ resemLling- on a small scale those at the " Chalk Hills^' in 

 every respect. The exact extent of gi'ound occupied by these magnesite 

 veins could not be determined in the thick thorny jungle^ but cannot be 

 much less than an acre in area. There can be no doubt tbat these mag- 

 nesite veins are due to a further metamorphic action on the bed of stea- 

 tite, the associated beds of hornblendic gneiss also showing sign of alter- 

 ation in the shape of numerous little veins and thin coatings of pistacUe ; 

 and the existence of the heat requisite for such changes being proved by 

 the iiTuption of the basaltic-trap-dyke. 



These phenomena of re-metamorphism of metamorphic rocks are of 

 such a peculiar character that there can be but 



Probahle metamorplio- 



sing agency, thermal small grounds for hesitation in ascribing them to 



springs. 



the action of water, (probably of such elevated tem- 

 perature as to have acted in some cases in the form of steam,) aided by 

 the presence of corrosive gases, especially carbonic acid gas. By the long- 

 continued action of steam and hot water, the disintegration of the rock 

 affected by them w^ould be very thorough and extensive, enabling the cor- 

 rosive gases to get access to the very inmost recesses of the mass, and the 

 greatest changes of composition and texture of the original hornblendic, 

 chloritic, and talcose rocks would take place, as we see they have done at 

 the so-called '' Chalk Hills.^^ Not the least perceptible effect on the physi- 

 cal position of these rocks relatively to the surrounding strata has resulted. 

 A decrease of temperature reducing the steam to water would give 

 rise to the formation of numerous springs of acidulated warm water, which 

 would dissolve out the magnesia and lime from the well-steamed horn- 

 blendic and talcose beds in immense quantities, with smaller proportions 

 of iron and silica. Near the surface these minerals would, on the cooling 

 of the mother liquid, be deposited by precipitation in the open cracks 

 and joint fissiu'es of the rocks, either alone or in combination with each 

 other, as we have seen them to occur at the various localities enumerated 

 and described. In the case of the intersection of a bed of steatite by 



