PEE-CAMBEIAN EOCKS OF ST. DAYH>'s. 321 



petrographical characters with the petrosilex of Sala in Sweden, to 

 which, as already remarked, Beudant gave the name of " adinole " *. 



The microscopic structure of some of the bands has been described 

 by Mr. Davies and Dr. Hicks f. The latier author remarks that 

 the chief peculiarity of the rock consists in the way in which the 

 quartz is separated into nests. 



In the slices which have been prepared in the Geological Survey 

 laboratories the more granular varieties are evidently tine shales or 

 tuffs, in which the clastic materials are quite distinct, though the 

 matrix separating them shows an incipient, feebly developed micro- 

 crystalline structure. The more flint-like varieties display a more 

 perfectly microcrystalline base. The constituents of the base appear 

 to be chiefly felspars and quartz. They are here and there aggregated 

 into patches of coarse crystallization, among which large well-striated 

 plagioclase is occasionally conspicuous. One of the most interesting 

 slides was prepared from a pale, milk-white, flint-like mass, taken 

 from the coast below JN"un's Chapel. Tt presents the usual finely 

 granular microcrystalline base, whieh remains of a pale bluish tint 

 between crossed prisms. The slide is traversed by two parallel veins 

 of quartz, one of which measures . T L of an inch in diameter and is 

 quite visible to the naked eye. Other minute threads of the same 

 substance appear under the microscope. The most remarkable 

 feature in this quartz-vein is the fact that it is crowded with liquid 

 inclusions, arranged in approximately parallel partitions which run 

 across the breadth of the vein. Each inclusion has a bubble; and in 

 some cases the bubbles rotate or vibrate as we watch them in the 

 field of the microscope. 



In trying to realize what has been the origin of these highly 

 compact siliceous aggregates, we see that they have been specially 

 developed among fine tuffs and shales or schists. From the analysis of 

 the acid tuffs above given, it is certain that some bands contain above 

 70 per cent of silica • the specimen of adinole from Xun's Chapel, 

 analyzed by M. Renard, contains, as we have seen, 78*62 per cent. 

 of that ingredient. In some instances the whole of the silica may 

 have been present in the rock before its alteration. 



But whether or not silica has been subsequently introduced into 

 the flinty bands of St. David's, it is evident that they cannot be due 

 entirely to original deposition. In the first place the material of 

 the bauds and of the concretions is so precisely the same as that of 

 the transverse and ramifying veins that we must connect the three 

 modes of occurrence together as parts of one general process of alter- 

 ation, and must conclude that, at least in their present condition, 

 these microcrystalline aggregates must be later in date than the 

 rocks in which they lie. 



In the secoud place, the same flinty substance traverses the quartz 

 porphyry of Nun's Chapel. In the veins which, in connexion 

 with that rock, ramify into the adjacent shale, it is here and there 

 conspicuous. One of my microscopic slides from the elvan itself 



* Traite de Mineralogie (2nd edit., 1832), vol. ii. p. 126. 

 t Quart. Journ. Geol. Soc. vol. xxxv. pp. 285 et seq. 



