part 1] AND ASSOCIATED KOCKS OF MOZAMBIQUE. 63 



^5) Origin of the Gneisses. 



The question of the origin of the gneisses, so far as it can be 

 discussed here with the data available, reduces itself to that of the 

 origin of the biotite-gneisses. ,The hornblende-gneisses and amphi- 

 bolites, and some of the garnetiferous rocks, have already been 

 interpreted as products of interaction between granitic magma and 

 crystalline limestones. Other members of the garnetiferous rocks 

 cannot yet be so readily disposed of. They may be in part com- 

 posite rocks derived from the contact with, or assimilation of, 

 limestones or former calcareous sediments which have not been seen, 

 or of which, no other relics remain. In others, the garnets may be 

 derived in part by the action of high pressure l from biotite, and in the 

 case of eclogite the rock may be due to high-pressure metamorphism 

 of gabbro or other igneous rock of similar composition. In the 

 absence of detailed field evidence, these possibilities cannot yet be 

 profitably discussed. The biotite-gneisses, however, have been 

 extensively studied, and, as they constitute the chief formation of 

 the country, the question of their origin calls for some attention, 

 even though a final conclusion cannot be arrived at with demon- 

 strable certainty. 



Summing up the facts already recorded in the preceding pages, 

 we have the following data on which to build :— - 



(o) Near the coast the gneisses are interfoliated with schists and sediment- 

 gneisses that are the representatives of ancient arenaceons rocks. 



(b) Near the coast (in the north only) and in various localities inland, the 



gneisses are interfoliated with bands of crystalline limestones pre- 

 sumably of exogenetic origin. 



(c) Except in the neighbourhood of certain inselberg peaks, the banding or 



foliation of the gneisses almost everywhere follows linear or broadly 

 curving planes, neither strike nor dip (which is usually low) being 

 generally subject to sudden variations. 



(d) On the slopes of certain inselberge, the foliation sweeps round the hills 



in closed curves and dips quaquaversally from the summit. 



(e) In a few cases it has been observed that the banding dies away 



as inselberg slopes are approached, the gneisses becoming glomero- 

 plasmatic, and finally passing into gneissose granite (porphyritic or non- 

 porphyritic), or into granulitic granite devoid of foliation, the change 

 being accompanied by a marked loss of biotite. 

 (/) The biotite-rich gneisses are (so far as measurements have been made) 

 poorer in zircon and radium than their leucocratic associates. 



The presence near the coast of metamorphosed rocks that are 

 referred to arenaceous sediments of various types, combined with 

 the occurrence of limestones in the north and interior, suggests 

 that, unless the land from which they were derived was of very 

 different average composition from those of later periods, large 

 quantities of argillaceous sediments must have been deposited 

 concomitantly. The absence of any direct metamorphic repre- 

 sentatives of such rocks at once raises the question whether they 



1 The condition suggested is one in which hydrostatic pressure is high, 

 while shearing stress, which presumably would favour the continued existence 

 or production of biotite, is relatively unimportant. 



