498 KEY. J. F. BLAKE ON THE 



in the unaltered sediment, and these would gradually increase in 

 number till, leaving no interval between them, they occupied the 

 ■whole rock. It is an approach to this that we see in the case of 

 contact-metamorphism. If, on the other hand, the crystalline rock 

 was the original, and the other produced from it by brecciation, we 

 should find the cracks increasing as we approached the margin, till at 

 last the rock was a mass of crystalline fragments, which might get 

 smaller and fewer as the sedimentary rock received other ingredients. 

 Between these two phenomena there is a very clear difference, 

 and in the present case the answer of the microscope is entirely 

 in favour of the latter. This, however, does not dispose of the 

 idea of the "dark schist" being an altered sedimentary rock of 

 an older series. Such a contention would have to be proved on the 

 clearest stratigraphical evidence, for beyond being both " schists " 

 and both " dark," there is nothing lithologically in common between 

 these and the Holyhead rocks ; the one set are composed of quartz 

 and chlorite, and the other of plagioclase and hornblende, j^o stra- 

 tigraphical evidence, however, can here be brought forward, since 

 the diorite only approaches the grey gneiss in one spot, near the 

 stream north of Llangwllog, and then it is on the wrong side. 



There are four areas in which these dioritic rocks occur. The 

 best known is that round Craig-yr-allor. Here Dr. Callaway states 

 that " grey gneiss passes up through the dark tj'pe into the grani- 

 toidite." I could find, however, no grey gneiss here at all. The 

 country is rather low and marshy, and out of it stand up several 

 bosses of rock. Some of these are composed of the ashy pelite con- 

 taining granite veins, and in one place there is apparently an intrusion 

 of the diorite. This rock is mostly confined to the centre, where it is 

 schistose in some bosses, and not so in others. In the southern boss 

 it is veined by granite, as before described. In parts the hornblende 

 and plagioclase are so intimately mixed that the rock is uniformly 

 dark; in others the lighter-coloured felspar has segregated into 

 lenticular fiakes, but in such cases there are no signs of movement 

 in the material of the rock. J^othing is ever seen below the diorite, 

 of which the boundary is fairly marked on the Survey map. AU 

 round it are the clastic rocks, some, as in the northern tongue, being 

 produced by its brecciation, the hornblende having mostly turned 

 into chlorite and calcite. The rocks to the west are of the same 

 ashy character, but usually develop more mica. 



A second area runs from near Pont-rhyd-defaid as far as Lle- 

 cheynfarwy. This consists partly of good crystalline diorite, partly of 

 altered varieties of the same, in which a few particles of quartz and 

 the alteration of hornblende into chlorite begin an approach to rocks 

 of another type, from which, however, they are still distinguished, 

 by their build, by the prevalence of plagioclase and the presence of 

 sphene. 



A third area, connected with the first, according to Dr. CaUaway, 

 by exposures of diorite near Pentrefelin, runs from Llandrygarn 

 farm to Mynydd-Mawr farm, thus escaping from the surrounding 

 granite, and being overlain by the Ordovician grits. The arrange- 



