﻿part 3] 



VOLCANIC EOCKS OF MOZAMBIQUE. 



273 



presumably of basic composition, carbon dioxide may be present in 

 primary calcite and other carbonate minerals. If this were so, 

 then a sufficient relief of pressure, such as may have initiated the 

 Tertiary volcanic activity of East Africa, would inevitably bring- 

 about reaction between natural carbonate and silicate minerals — 

 reactions which, if they were of the kind described by Daly, woidd 

 lead on to the evolution of an alkali series of rocks accompanied 

 by a cafemic complementary series. 



Methods of desilication by water and by natural carbonates 

 have now been considered, 1 but there remains a third possibility 

 that has not been previously suggested. There is reason to 

 believe that below the granites and gneisses of the continents' 

 rocks of basaltic composition fall naturally into place, and that 

 below these occur rocks of peridotitic composition. 2 If igneous 

 activity in any region were initiated, at or through such a depth 

 that the magmas formed included one of peridotitic composition, 

 it seems likely that during the upward progress of the latter' it 

 would interact with basaltic material, taking up silica to form 

 enstatite or hypersthene, and abstracting lime to form monoclinic 

 pyroxenes. Such minerals, being heavy, would tend to sink (or 

 at least not to rise) leaving above a magma relatively impoverished 

 in silica, but enriched in alkalies and perhaps in a new series of 

 volatile fluxes. 



The occurrence of kimberlite pipes in South and Central Africa 

 favours the view that igneous activity may begin in a zone of 

 peridotitic composition ; but, as it stands, the theory still throws 

 no light on the origin of the associated melilite-basalts. On the 

 other hand, if melilite-basalt is always a cafemic complement to 

 an alkali series, it is remarkable that no alkali rocks should 

 have been discovered in South Africa. The calciferous dykes of 

 the Premier Mine, 3 and the great abundance of calcite in many 

 kimberlite pipes 4 and in melilite-basalt, again suggest that carbon - 

 dioxide was an important constituent of the deep-seated materials 

 whence these rocks came. 5 It is difficult to explain the presence 

 of so much calcite, either as an alteration-product or as an infil- 

 tration from surrounding sediments, and we are thus led from 

 another point of view to consider the possibility of calcite being- 

 present as a primary constituent of very deep-seated rocks. A 

 source for the high percentage of lime in melilite-basalts and of 

 calcite in kimberlite pipes would thus be provided. Moreover, 

 melilite-basalt could then be directly formed by the interaction of 



1 Dr. H. I. Jensen's suggestion of desilication by Pre-Cambrian saline beds, 

 Proc. Linn. Soc. N.S.W. vol. xxxiii (1908) p. 522, has obviously no application. 



2 A. Holmes, Geol. Mag. dec. 6, vol. ii (1915) pp. 63- 64. 



:i P. A. Wagner, ' The Diamond- Fields of South Africa' 1914, p. 98. 



4 Ibid. p. 75. 



5 Ibid. p. 117. Wagner ascribes the initiation of the igneous activity 

 represented by kimberlite pipes to the relief of pressure consequent upon the 

 isostatie uplift of the continent that began in Upper Cretaceous times, and 

 suggests that the rocks must have come from unusually great depths. 



