part 3] METAMOKPHTSM IN THE MONA COMPLEX. 343 



that the crystallization and foliation of the Gneisses is older than 

 the deposition of the Bedded Succession. 



Note. — On the 1-inch map, the colour selected for the Gneisses 

 has unfortunately, in process of printing, come out so as to be 

 almost indistinguishable from that of the New Harbour Beds. 

 The Gneiss of the larger tracts can be distinguished by its symbol, 

 but on the smaller tracts there was no room for a symbol. Most 

 of them can be identified from the Memoir. The tract at the 

 word ' Mynachuy,' however, comes against New Harbour Beds at 

 its north-western end ; and there is a small one beyond it, between 

 Gwna Beds and Serpentine. A distinctive colour-wash might, 

 with advantage, be added to all the gneissic tracts by hand. 



VI. The Origin or the Basic Gneisses. 



Some time ago I was led to suspect that the phenomena cited as 

 evidence of movement after consolidation ('G. of A.' pp. 131-33, 

 903, 944) might not be conclusive ; and that I might have over- 

 looked signs of primary injection-banding, such as that of Skye, 

 the Lizard, and Southern Australia. The development of gneisses 

 has proved on re-examination to be essential]} 7 the same (p. 341), 

 in the Middle as in the Aethwy Region. 



Early stages. — The most primitive condition of the basic 

 rock (' G. of A.' pp. 131-32, 322, 376, &c.) is an assemblage of 

 closely-crowded masses, usually ovoid, but occasionally sub-cubic, 

 or even exhibiting re-entering curves, of unfoliated or faintly- 

 foliated dioritic matter, which (though occasionally) is not pre- 

 valently coarse. Evidently this is (as in the Lewisian complex 

 and elsewhere) a product of early differentiation of the basic 

 magma. Acid albitic matter acts as a matrix to basic lumps ; but 

 this matrix darkens in some directions, and then acts as matrix to 

 acid masses. Further research is. therefore, needed to determine 

 the general order of differentiation, which seems to be less simple 

 than in the Lewisian complex ; this, however, is not essential to our 

 present purpose. In only a few yards (as at the Werthyr sections) 

 the dioritic masses begin to be flattened, and the whole assemblage 

 passes rapidly into a thorough banded gneiss. Except for the 

 differences in the nature and behaviour of the matrix, the pheno- 

 menon resembles, almost exactly, that of pi. ix in the Geological 

 Survey Memoir on the North-West Highlands, which might 

 almost pass for an illustration of it. So far as the foregoing 

 description goes, this process might have taken place before con- 

 solidation : that is, before crystallization of the differentiated 

 magma. The more basic of the unfoliated masses, however, 

 frequently contain groups of large quasi-porphyritic hornblendes. 

 When flattening begins, these groups also flatten ; and in the 

 banded rock they acquire a foliation as well. Now, had the magma 

 been fluid when the flattening took place, these groups would have 

 disintegrated, and their crystals have been floated awaj r one fr< m 



Q. J. G. S. No. 315. 2 ]j 



