COLLECTED BY BR. HICKS IX N.W. i>EMBROKESHlllE. 361 



underlyino- series. I may mention that, in the case of (1-1-), the 

 hand-specimen presents no marked indication that it is from a 

 pebble, and I examined it for some time under the impression that 

 it was from the subjacent rock, and did not find out my mistake 

 until I noticed the label *. 



OrIGM OE THE " HaLLEFLINTAS " OF TrEEGARN AND ROCH. 



I have compared these slides from the rocks of Trefgarn and Roch 

 with specimens in my own collection of obsidians and pitchstones 

 from various localities, and with slides of devitrified rhyolitic rocks 

 from below the Cambrian conglomerate near Llanberis, from the 

 Wrekin, and from large fragments in the volcanic ashes of Charn- 

 wood. To many of these they present close resemblances ; the 

 belonites and the arrangement of the ferrite and opacite can be 

 paralleled by the former (glassy) rocks t; the devitrification-structure 

 by that in the latter group, though sometimes there are differences due 

 to silicification. But while this structure is more normal in these 

 rocks than in those from Trefgarn and Roch, the microporphy- 

 ritic and amygdaloidal characters are more conspicuous in the latter, 

 and the fragments in the ash-beds of Trefgarn are of a perfectly 

 normal character. Hence I regard the series, on the whole, as 

 indubitably a volcanic one, consisting of acid lavas and their asso- 

 ciated ashes, some, especially of the former, having been, as sug- 

 gested by Prof. Blake, subsequently permeated by hot water con- 

 taining silica in solution, which has silicified the rock, replacing the 

 felspars and, in part, even the felspathic constituent, by chalcedonic 

 quartz and filling up the cavities with the same. This process 

 need not have occurred long after the emission of the lava, though 

 I believe it was subsequent to the devitrification of the rock ; but, 

 in any case, the ash-beds of Trefgarn (which are not silicified) 

 must have become very hard, b}^ means of ordinary microminera- 

 logical change, before the pebbles were made from them, and so must 

 be much older than the overlying conglomerate. 



The general similarity of the unsilicified rocks of Trefgarn — 

 and to some extent even of the silicified — to the rhyolitic, trachytic, 

 or andesitic rocks which abound in the Cambrian conglomerate and 

 the underlying breccia, is an argument for the fragments in these 

 being derived from lavas of about the same age as those at Trefgarn. 

 Moreover the Koch and Trefgarn rocks are practically identical : the 

 latter cannot be intrusive, the former (even if there be no ash-beds 

 there) has all the structure of a flow as opposed to an intrusive mass : 

 hence we may safely say that the series, as a whole, cannot be 

 intrusive in beds of '" Harlech " or later date. 



PvOCKS FROM BrAWDY AND CwMBACH. 



Granitoid rock from Brcnvdi/ (2 slides). — This rock is holo- 

 crystalline, consisting of quartz, decomposed felspar (among which 

 plagioclase, probably oligoclase, and orthoclase may be occasionally 



* To avoid tany prejudice I usually, in the first stage of my examination 

 of specimens collected by others, abstain from looking at labels or notes. 

 t E.g. by specimens from Mexico, Arran, Meissen. 



' 2c2 



