﻿Vol. 66.] DEDOLOMITlZAIION IN THE HAEBLE OF PORT SHEPSTONE. 517 



granite, which forms the whole country for hundreds of square miles, 

 completely surrounds the dolomite, and, in addition, the latter is 

 traversed by broad dykes, which probably represent the last phase 

 of the eruption. 



The normal metamorphism of the dolomite produced a saccha- 

 roidal marble of very coarse texture, and, as a rule, free from 

 minerals other than carbonates. It is noteworthy that our sections 

 of the normal marble do not disclose the presence of brucite, as 

 in the cases described by Dr. Teall and Mr. Harker in the Durness 

 Limestone of the North-West of Scotland, and in the well-known 

 instance at Predazzo in the Tyrol. 



Where silica could be supplied from extraneous sources, as in the 

 neighbourhood of the included blocks, dedolomitization took place, 

 the following minerals being produced : olivine, forsterite, diopside, 

 scapolite, phlogopite, all of which are silicates. In addition to 

 these there is the non-silicate spinel. A noteworthy feature is the 

 absence of minerals such as garnet and cordierite, which are specially 

 characteristic of cases in which other considerations indicate the 

 occurrence of low-temperature metamorphism. All the facts agree, 

 therefore, in indicating the prevalence of a high temperature during 

 the metamorphism of the dolomite. 



The inclusions are blocks of rock that appear to have been 

 deposited in the limestone or dolomite at the time of its original 

 formation, and the large granite fragment especially bears out this 

 view. W'hile the grey granite of the district, which is respon- 

 sible for the metamorphism of the dolomite, is a potash-granite, the 

 inclusion is a fragment of soda-granite or soda-aplite of wholly 

 different character. So far as our experience goes, no such rock 

 has been observed among the South African granites. Since the 

 dolomite belongs to the oldest group of rocks found in the 

 country (the Swaziland System) into which the known granites are 

 intrusive, this rock if a boulder must have formed part of an 

 igneous complex older than anything yet discovered. 



The occurrence of granite boulders in limestone is not unknown ; 

 they have been found, for instance, in the Carboniferous Limestone 

 which crops out in the neighbourhood of Dublin (near Milltown) 

 and south of Kimmage, 1 and Jukes suggested the possibility of 

 these fragments having been transported by adhesion ' to the roots 

 of plants which were drifted down the water-courses from the 

 neighbouring granite land into the sea of the Carboniferous Period.' 

 Again, a large boulder of granite, together with smaller blocks of 

 greenstone, was found in a chalk quarry at Haling, south of 

 Croydon, and described by It. A. C. Godwin-Austen. 2 In the same 

 paper Godwin- Austen mentions the discovery of a boulder of syenite 



1 S. Haughton, Journ. Geol. Soc. Dublin, vol. v (1850-53) p. 113. See also 

 'The Geology of the Country around Dublin' Expl. Sheet 112, Meiu. 

 Geol. Surv. Ireland (1903) pp. 76-77. 



2 Quart. Journ. Geol. Soc. vol. xiv (1858) p. 252. See also A. J. Jukes- 

 Browne, 'Cretaceous Kocks of Britain: the Upper Chalk of England' Mem. 

 Geol. Surv. vol. iii (1904) p. 178. 



