318 Prof. T. G. Bonney—Pebbles in the Cambrian at St. Davids. 
locking, as if the mass had been reduced to a plastic condition and 
then slightly but rather uniformly compressed in all three dimen- 
sions of space, or as if the grains had been growing gradually 
outwards at the expense of a semi-fluid magma which had filled up 
all the interstices. A rather similar aspect is presented by quartz 
grains, which have formed (often mainly by segregation—though 
no doubt frequently around a clastic nucleus) in a ‘silty’ rock 
which has been much affected by ‘contact metamorphism.’* A 
similar aspect is presented by the quartzes in certain quartzose 
schists, which are presumably metamorphosed sediments, and which 
do not appear to have been subsequently very distinctly modified by 
pressure. This peculiarity—though distinct to an accustomed eye— 
is not very easy to describe in words or even to represent diagram-_ 
matically ; but an idea of it may be obtained by comparing the 
lower half of figure 8 (omitting some of the small interstitial 
granules) with figs. 1 and 2 and the upper halves of 3 and 4 of plate 
xxxi. in the late Prof. R. D. Irving’s excellent paper on the 
Archean Formation of the North-Western States.2 A hard and fast 
line, as indeed these figures show, cannot be drawn between this 
structure and that of an ordinary quartzite—still, I believe, a very 
large number of specimens can be classed under one or the other 
type; and I am quite certain that if these three specimens from St. 
Davids had been sent to me (as specimens often are sent) without 
any clue as to their geological horizon, I should have returned them 
with the remark that they had the appearance of being very ancient 
rocks. 
Here, then, at the base of the British Cambrians, we find, in three 
specimens selected as fair samples of the materials of a conglomerate, 
structures, of which one indicates that the rock is not likely to be 
anything but ancient; the second is more like the structure of the 
Huronian quartzites than of any indubitable Paleeozoic quartzite 
which I have examined; and the third resembles that of quartz-schists, 
which are almost certainly Archean, though they probably do not 
belong to the oldest part of that series. The structures also of these 
two are not such as are suggestive of pressure metamorphism, but of 
slow molecular change, under constraint indeed, in the presence of 
water, and at a fairly high temperature. Thus the evidence of these 
pebbles, so far as it goes, is favourable to the opinion that, as a 
general rule,*? rocks of Archean age may be identified by their 
structures, and that the conditions under which they consolidated 
have recurred, if at all, only rarely and locally. 
1 See, for instance, Q.J.G.S. vol. xliv. p. 16. 
2 United States Geological Survey, Fifth Annual Report, 1883-4. 
3 I must not be supposed to assert that a hard, fast, and universal line can be 
drawn between Archean and Paleozoic (or Cambrian and Pre-Cambrian) any more 
than between Paleozoic and Mesozoic or between any other geological groups, 
systems, or classificatory divisions. 
