VOLCANIC GROUP OF ST. DAVID’S. 309 
association of similar rocks in other volcanic areas of more recent 
origin. It is remarkable also that in the areas elsewhere proved 
pre-Cambrian, the same association of rocks occurs, though the patch 
may be relatively small. Thus in the Wrekin, associated with the 
perlite, Mr. Allport describes both ashes and spherulites, but no gra- 
nitic rocks ; while in Anglesey and Caernarvonshire all three forms, 
according to Dr. Hicks, are associated in no less than three areas. 
Considering the small size of these exposures, the constant association 
is remarkable, and in itself suggests a connexion, especially as, though 
the two upper parts may stand alone, the basal part never seems to 
do so. 
Two general observations are suggested by these results, one with 
regard to the composition, the other with regard to the distribution 
of these volcanic rocks. 
A marked feature in the composition is the highly acid character. In 
the central granitic mass this is shown by the abundance of quartz and 
the almost entire absence of any ferrous or magnesian constituent. 
In the felsites or porphyries it likewise appears in the abundance of 
original quartz and the small quantities of any magnesian mineral, 
pyroxene not being certainly present even in the intermediate rocks. 
That there are basic rocks amongst the ashes in the form of lavas 
has been proved ; but even in this portion the peculiar feature is the 
abundance of porcellanite and other highly acid tuffs. But whatever 
the original nature of the supragranitic rocks, it is certain they have 
undergone great change in the way of addition of silica. This, we 
have seen, has been deposited in cavities of former crystals, in inter- 
stices and in cracks, a phenomenon reaching its maximum develop- 
ment in the porcellanites and the Roche-Castle rock. Whence is 
this silica derived, and when was it introduced? Whatever we may 
answer to this, it must be reckoned with in considering the signifi- 
cance of analyses. It would seem, however, that the rocks of St. 
David’s were at the commencement nearly as acid as now, since the 
relics of decomposition are not grains of iron-oxide, which, as a 
rule, are very rare, but kaolin, which is abundantly distributed in 
every slide; hence the original minerals were mostly felspathic. 
In regard to the distribution, we note first the linear arrangement 
of the centres: an east-and-west line through the central mass of 
St. David’s cuts through Ramsey Island on the one side and the 
mass at Pointz Castle on the other, and nearly cuts what I think is 
probably another centre north of Solva; this is in accordance with 
the modern arrangement of volcanic outbursts. But if these be 
truly centres, how insignificant they are! Elsewhere the centres of 
like age would seem to be equally small, and hence their confusion 
with simple intrusive rocks of subterranean origin. 
As we pass up the geological series, the magnitude of volcanic 
centres, as found in Silurian, Carboniferous or Permian, and Miocene 
times, seems to increase till we reach the magnificent volcanoes of 
to-day, which have mostly not more than a Tertiary origin. But 
while the magnitude has increased, the concentration has pari passu 
progressed ; and instead of the numerous Paleozoic centres, and still 
ess 
= es" Se ee. ee eee oe Se 
ee 
wwe 
_ rhe ee Le: pope | 
> Ee eT SE SS Pe pe — Se 
eee ee 
