362 PEOf. T. G. BONNET ON EOCK-SPECIMENS 



recognized), and a fair proportion of a greenish micaceous mineral 

 (altered biotite). One slide shows much crushing, the rock being 

 in places brecciated in situ. It may be a granite, but the structure is 

 a little abnormal. We have in it the same difficulty as is presented 

 by the Dimetian of St. David's, and by the granitoidites of Twt Hill, 

 Ty Croes, and Ercal (Wrekin). Under these circumstances, though 

 we cannot prove the Brawdy rock to be of Archiean age, this 

 identification is the more probable. 



Cwm Bach, near Newgale. — A pale greenish-grey, flinty-looking- 

 rock, with lighter specks, resembling an altered ash, such, for 

 example, as some of that near Clegyr Bridge. There are crystals of 

 felspar (some oligoclase) which are evidently broken and corroded 

 by the ground-mass, which exhibits a minute devitrified structure, 

 with a few grains of iron oxide and one or two of a green serpen- 

 tinous mineral. The microscopic examination does not give a 

 result which is absolutely beyond dispute. There is just a possi- 

 bility that the rock may be a rhy elite with flow-brecciation ; but 

 I have little doubt that it is a true indurated ash. 



Conclusions. 

 The following conclusions, then, appear to me to follow from the 

 above investigations : — 



A. When the Chanter's Seat conglomerate was formed the fol- 



lowing rocks were undergoing denudation : — 



(1) Granitoid rocks, identical with the existing Dimetian. 



(2) Trachytic rocks, among which were probably true lava- 



flows *. 

 (8) Quartzites and schists, the latter resembling those which 

 in many districts occur rather high in the Archaean series, 

 (4) Ordinary sedimentary rocks. 

 Hence there was in this district a series of rocks, some much 

 older than others, which contributed to the formation of 

 the Cambrian conglomerate. 



B. The conglomerate above the Trefgarn series is formed from 



rocks which occur in the latter. 



C. The peculiar characteristics distinctive of certain members of 



the Trefgarn series had been assumed by them when the 

 conglomerate was formed. 



D. Either the Dimetian is a member of an old gneissoid series or^ 



if it is the core of a volcanic group from which the trachytic 

 lavas had been ejected, this had been laid bare by denu- 

 dation before the Cambrian conglomerate was formed. 

 Hence in either case both the Dimetian and the felstones 

 a*re Pre-Cambrian. 



* It is quite true that we cannot, in our present stage of knowledge, fix upon 

 afly structure as distinctive of a lava-flow, but there are many which only occur 

 in rocks cooled under conditions very similar to those of fl.ows ; and when we 

 find fragments exhibiting these occurring in great variety and abundance, the 

 probability is very strong that they are derived from subaerially consolidated 

 masses. Indeed, as similar structures could only occur elsewhere in dj-kes 

 and small intrusions, it is not likely that in this ease they could furnish many 

 fragments. 



