﻿Vol. 64.] STRUCTUEE OF THE ST. DAVID's AEEA. 367 



The lowest bed shown in the district (A 1) is composed of purple- 

 red and grass-green halleflintas/ occasionally with coarser bands. 

 It is well developed at both ends of the bridge north of the 

 cathedral, and at the point west of Porthlisky. It undoubtedly 

 exists, as shown by the numerous blocks in the soil, east of Carn 

 Poeth, but it is not there seen in situ. 



In microscopic section the absence of felspars such as are found 

 in the later silicified rocks of D 4 and F 2 is noticeable. The rock 

 consists of sericite and quartz, with grains of chlorite and opaque 

 matter and occasional epidote. Quartz largely predominates, 

 forming a fine-textured aggregate. 



The next subdivision (A 2) is best exposed at the fine sections in 

 the western clifi" of Porthlisky ; and again in the quarries of Penrhiw.. 

 north of the cathedral. The identity of the beds at these two 

 localities, fully 2 miles apart, is singularly striking. The sub- 

 division consists throughout of alternations of red and green gritty 

 tuff's. The lower bands are fine-grained and sheared, the upper 

 part is coarser. The red tufi's are often somewhat basic, and 

 weather in a rusty manner. A distinctive characteristic is the 

 presence of bluish fragments which weather white. In microscopic 

 section, these fragments are seen to be bits of devitrified glasr. 

 crowded with globulites (?). The outline of the fragments is a suc- 

 cession of little concavities, but it is not clear whether these are 

 due to original gas-bubbles or to the weathering-out of a perlitic 

 structure. A certain number of true vesicles filled with siliceous 

 matter undoubtedly occur, but they do not seem sufficiently 

 numerous to account for the peculiar outline of the fragments. 

 The matrix of the red tufi" is ferruginous, and includes felspars 

 and decomposition-products. 



In the green tufl^s felspars are more plentiful, and the matrix is 

 nearly isotropic, apparently consisting chiefly of scales of a pale 

 chlorite. The included fragments are mostly olive-green flakes of 

 a felsite much like that of the red tuffs, but containing a consider- 

 able amount of chlorite. Fragments of vesicular trachyte resembling 

 that presently to be described in connexion with B 2 (p. 369) 

 occasionally occur. 



A noticeable point in the felsite-fragments of both red and green 

 tuffs, is the occurrence in the devitrified ground-mass of elongated 

 areas with ragged margins, which extinguish straight. This 

 peculiarity in structure may be due to an original trachytic com- 

 position. The green tuffs are progressively more felspathic and 

 gritty towards the top, passing gradually into the next subdivision. 



(A 3) — Wherever the bed just described is seen, it is overlain by 

 a felspathic grit (A 3), which is one of the most useful horizons in 

 the Pebidian. In the extreme south-west of the district, near 



1 I use the term halleflinta as indicating a homogeneous silicified rock 

 with conclioidal fracture, and porcellanite as indicating a similar rock 

 hut with splintery fracture. 



Q. J. G. S. No. 255. 2 b 



