TRANSACTIONS OF SECTION C. 607 



-uniformly, in a grit from near the ruins, Bradgate, whicli also contains grains of com- 

 pound structure. In conclusion, I must briefly notice the so-called Torridon sand- 

 stone of North-western Scotland, which is in many respects invaluable to the student. 

 That it is not later than the base of the Ordovician is indisputable ; that it is under- 

 lain by and derived from a mass of Archaean rocks, gneisses, more or less granitoid, 

 with occasional schists, is universally admitted. Its coarser basement beds are 

 ■crowded with fragments of the underlying gneisses and schists, and since the epoch 

 of their formation no important change has taken place in either the one or the other. 

 The finer beds, though other materials occasionally occur, are largely, sometimes 

 almost exclusively, composed of grains of quartz and of felspar identical in every 

 respect with those in the underlying series. It may be a fact of some significance, 

 for it agrees with what I have elsewhere noticed in very old fragmental rocks, that 

 the felspar appears to have been broken ofl:" from the parent rock while still unde- 

 composed, and in many cases is even now remarkably well preserved. It would 

 ^eem, therefore, as if the denudation of the granitoid rock had been accomplished 

 without material decomposition of its felspar, but I must not allow myself to 

 digress into speculations on this interesting and suggestive fact. While referring to 

 this district I may mention the quartzites, though, strictly speaking, they are 

 Ordovician in age. These in some cases consist all but exclusively of quartz 

 grains derived from the Archaean series, which, however, are generally smaller than 

 those in the Torridon ; it would seem as if the felspar of the parent rock had either 

 ■decomposed in situ, or had been broken up in consequence of the longer distance 

 from the source of supply. This quartzite is sometimes of singular purity, contain- 

 ing little or no earthy material, and only rarely a flake of mica or a grain of felspar, 

 tourmaline or epidote (?). 



Onlovician-Silurian.^ln regard to the earlier of these formations I am better 

 acquainted with the volcanic than with the non-volcanic fragmental beds. Still, 

 so far as I have seen, we find among the latter frequent indications of a supply of 

 materials from regions of crystalline as well as of ordinary sedimentary rocks. The 

 quartzite of the Stiper Stones (possibly earlier than the Arenig) appears to have 

 derived most of its grains from granitoid rocks, and probably the same is true of 

 many of the coarser beds in the Caradoc group of Shropshire and Eastern Wales. 

 The Garth grit of Portmadoc appears to have derived much of its quartz from a 

 like source as the Stiper Stones, but it also contains bits of a fine grained quartzose 

 schist and of older clastic rocks. A grit from the Borrowdale series of Chapel-le- 

 dale contains, in addition, bits of old andesite and probably diabase, with fragments of 

 a rather granitoid gneiss and quartzose schists. Fragments of crystalline rock, both 

 small and large, abound in the Upper Llandovery beds at Howler's Heath, at 

 Ankerdine HiU, in the Abberley district, on the west flank of the Malveras, and at 

 May Hill, thus indicating that early in Silurian times far greater outcrops of 

 crystalline rock existed than are now visible west of the Severn. Mr. W. Keeping ^ 

 <jalls attention to the abundance of fragments of quartz, felspar, and mica in the 

 ■* greywackes ' of the Aberystwith district, which give the rock sometimes quite a 

 granitoid appearance, and adds that, in his opinion,^ 'the abundance of felspar 

 ■crystals, so general in the Silurian rocks (Upper Silurian of North Wales, South 

 Wales, and the Lake district), points to the neighbouring presence of a vast mass 

 ■of early, perhaps primeval, igneous rocks as the great source of sediment supply in 

 Silurian times.' What I have seen of the Denbigh grit of North Wales and of the 

 Coniston beds of the Lake district confirms this conclusion. It is true that some 

 of the material may have been supplied by Ordovician volcanic rocks, and that the 

 quartz grains in the specimens which I have examined are not large. But we must 

 remember that the latter can hardly have been furnished by the lavas of the Lake 

 district, and those of North Wales, though richer in silica, do not, so far as I know, 

 generally contain large quartzes. These, indeed, may have been derived from the 

 denudation of Cambrian rocks, but I should doubt the sufficiency of such an ex- 

 planation. In one specimen, a Denbigh grit from Pen-y-glog, near Corwen, there 

 occurs, besides one of smaller size, a fragment about -1" in diameter, exhibiting a 



' QiMT. JouTn, Geol. Soc. vol. xxxvii. p. 149, &c. - lUd. p. 150. 



