ROCKS OF PEMBROKESHIRE, 521 
special bands referred to are important as showing the main sources 
of derivation. The ashy particles are minute fragments derived by 
denudation from the Pebidian series; and the arkose material can 
with equal certainty, as will be pointed out by Mr. Davies in his 
notes, be shown to have been derived from the denudation of the 
Dimetian. Foreign fragments are found here, as in all conglome- 
rates; but the matrix, like the succeeding grits and sandstones, 
points unmistakably to the true source of derivation of the sediments. 
It is very remarkable that the matrix of the conglomerates, and 
the succeeding grits and sandstones, should have been so completely 
neglected in the Director-General’s inquiries. Supposing it possible 
that his view of a natural succession of the Cambrian on the 
Pebidian beds could be the true one, whence, under the circumstances, 
could the materials which build up the many hundreds of feet of the 
immediately succeeding sandstones and grits, which he admits are 
true sediments and not volcanic deposits, have been derived ? 
In this conglomerate many fragments of the underlying porcel- 
lanites, much in the condition in which they are now found, occur. 
Many years ago Prof. Hughes’s keen and experienced eye detected 
in it one fragment of the porcellanite which could not be less than 
9 inches across. Smaller fragments are abundant, and some have 
been cut for examination with the microscope (Note 61). Pebbles of 
quartz-felsites from the Arvonian and of volcanic rocks from the 
Pebidian, are also to be readily found. Quartzites and vein-quartz 
are also abundant as large pebbles. Whether the quartzites and 
the fragments of true quartz-schists which are occasionally found 
were derived from some Pre-Cambrian rocks in the neighbourhood, 
now covered up by the newer sediments or under the sea to the 
south and west, is a question which has not as yet been solved. 
b. General Description of the Cambrian Succession, and of the 
Mineral Composition of the Beds.—The Cambrian succession, as it is 
exhibited towards the $.E. from this point, is the most complete in 
the neighbourhood, and as the different members in it will have to 
be referred to further on as they are brought by faults into contact 
with the Dimetian, it will be well now to describe them briefly. The 
rocks which immediately succeed the conglomerates are the greenish 
flagey sandstones (2 on map, Pl. XXIV.). On microscopical exami- 
nation these are found to be made up mainly of quartz—the dirty 
quartz of the Dimetian, and quartz such as may be derived from 
quartz-felsites or veins—and of broken bits of decomposed felspar. 
A dull green mineral similar to that found so abundantly in the 
Dimetian is also present in a broken condition, and there are also 
frequently present bits of mica, magnetite, &c. Secondary quartz 
and some chloritic and micaceous minerals are also developed in 
fissures and along the lines of lamination. This may be taken as 
the condition of these rocks throughout the district. Where crushed, 
however, in the immediate neighbourhood of faults, the fissured 
conditions become of course much more marked, and consequently 
there is a greater proportion of the secondary minerals present. The 
actual rock itself, however, remains much the same, and there is no 
