172 MR. F. R. C0WPER REED ON THE GEOLOGY OF [May 1 895, 



Small, well-formed felspar-crystals project into the spots from 

 the groundmass, or lie wholly inside them [315]. I have heen 

 unable to decide whether these spots are original or secondary 

 structures ; but the evidence seems rather to indicate that they are 

 original cavities subsequently filled in with silica, or the silica may 

 pseudomorph some pre-existing mineral. Mr. Harker describes an 

 apparently similar type of structure common in the lowest Bala 

 lavas at Y Drosgl, near Dwygyfylchi, and at Penmaenbach and else- 

 where. 1 In these rocks circular, elliptic, or irregularly shaped 

 spots occur, marked out by a dusty-looking border. Within each 

 spot the quartz and felspar are more or less completely individual- 

 ized, the former being in excess of the latter, and the whole of the 

 quartz behaves as a single crystal. Mr. Harker was inclined to 

 think these structures original, but is somewhat doubtful on this 

 point in the case of some Cross Fell rhyolitic rocks. 2 



Reference has frequently been made in this paper to perlitic 

 rocks. But there are some occurring near Bwlch Mawr, Dinas, which 

 are of especial interest. In one case [316] (PL VI. fig. 1) the 

 perlitic structure occurs in narrow bands, suggesting the outpouring 

 of successive thin streams of lava, in each of which the outer surface 

 cooled rapidly into a glass. The perlitic structure is marvellously 

 preserved, and the large curving cracks surrounding spheroidal 

 portions of the groundmass frequently enclose smaller concentric 

 curves, or nests of smaller spheroids in minute imitation of the 

 spheroidal structure in diabase-masses described and figured by 

 Prof. Bonney. 3 In some of the areas within the perlitic curves 

 there are seen in ordinary light under the microscope a dozen or 

 more small, clear, circular spots either isolated from each other or 

 collected in groups which have a reticulate appearance and fill up 

 the whole of the enclosed space. Each minute spot has in the 

 majority of cases a dark dot at its centre. In that portion of the rock 

 less replaced by quartz-mosaic and with the best-preserved perlitic 

 structure these minute spheres are regular, and have well-defined 

 boundaries. Each behaves as a single crystalline individual of 

 quartz. In more altered portions they are irregular in outline and 

 run one into the other in a more or less confused manner. In some 

 cases radially arranged inclusions of very fine dusty material or 

 crystallites occur in them, but no true spherulitic structure is 

 observable. Where a considerable number are crowded together so 

 as to assume polygonal outlines, a reticulate appearance is produced, 

 and this seems to be that structure compared by Miss Baisin to 

 crackled china in a felstone from the Lleyn, in which ' the ground- 

 mass shows perlitic curves, and appears also minutely broken, like 

 crackled china.' 4 This resemblance to crackled china is particu- 



1 ' The Bala Volcanic Series of Caernarvonshire ' (Sedgwick Essay), 1889, 

 p. 22. 



2 Quart. Journ. Geol. Soc. vol. xlvii. (1891) p. 519. 



3 Ibid. vol. xxxii. (1876) p. 151. 



4 ' Nodular JMstones of the Lleyn,' Quart. Journ. Geol. Soc. vol. xlv. (1889) 

 p. 251. 



