﻿Vol. 66J] DEDOLOMITIZATION IN THE MARBLE OP POET SHEPSTONE. 521 



constituted his osmotic theory of metamorphisrn. He was 

 surprised that the Authors gave no account of the contacts of the 

 dolomitic limestone and the great mass of granite. He combated 

 the idea that this granite block was a sedimentary erratic in the 

 ancient limestone. It might be in rare cases, but he had often 

 seen evidence of such dolomitic limestones having been fused and 

 injected ; and as a dyke of igneous rock could detach and carry on 

 pieces of its wall, this could do likewise. In fact, it was only in 

 such a way that the eozoonal masses of Canada and Monte Somma 

 could be explained. 



The speaker likewise believed that the smaller inclusions were 

 originally the same rock, but altered by paramorphism. At Monte 

 Somma one could trace a piece of leucitite through a series of ultra- 

 basic rock right on to masses or cavities lined with meionite, 

 anorthite, and perhaps a little pyroxene. 



He also enquired what was the insoluble residue of the dolomite 

 analysis, which he thought of great importance. 



Dr. J. W. Evans believed that the presence of the blocks of 

 soda- granite might be accounted for by the breaking-up of an 

 intrusive mass into lenticles by earth-movements and the removal 

 of all traces of these disturbances from the limestone by its re- 

 crystallization. He also spoke of the power, which crystalline 

 limestone possessed, of accommodating itself to pressure like a 

 fluid, as a result of movements along the gliding-planes of indi- 

 vidual grains. 



Mr. T. Crook remarked that the fundamental cause of dedolo- 

 mitization was to be found in the fact that magnesium carbonate 

 had a much lower dissociation-temperature than calcium carbonate. 

 It was for this reason that magnesia took precedence over lime 

 in the reactions which accompanied the thermal metamorphisrn 

 of dolomite. It was noteworthy that the secondary minerals of 

 contact-metamorphosed dolomites were usually rich in magnesia, 

 lime often taking no vital part in the reaction ; as, for example, 

 spinel, the olivine-group, and phlogopite. Presumably, these secon- 

 dary minerals could form at a lower temperature than those 

 requiring the interaction of lime. 



Perhaps the most novel and interesting feature of the paper was 

 the Authors' contention that the absence of garnet and cordierite 

 proved high-temperature metamorphism in this particular case. 

 The speaker was unable to agree with this conclusion, which 

 seemed to him quite untenable. The garnets of metamorphic 

 limestone belonged to the lime-alumina-iron-oxide group, and were 

 typically void of magnesia. Therefore they could not be expected 

 to form under conditions of metamorphism in which the reaction 

 of magnesia preceded that of lime. Similar reasoning applied to 

 cordierite, which was comparatively poor in magnesia. The facts 

 of paragenesis led to the same conclusion. 



Dr. Hatch, in replying for the Authors, said that the main facts 

 were indisputable, namely : blocks of granite in the dolomite were 

 surrounded by rings of silicates and other minerals, which were 



