176 MR. E. K. COWPER REED ON THE GEOLOGY OE [May 1895, 



quartz-crystals in a decomposed, brownish, almost isotropic matrix. 

 The majority are composed almost entirely of very vesicular frag- 

 ments ; the vesicles are filled with calcite or dolomite, and the 

 former exists also in large crystalline plates in the matrix [16], 

 [29], [51], [14], and [105]. 



Another rock resembles a fine-grained bluish-grey slate [68], but 

 under the microscope it is seen to consist of a microcrystalline to 

 isotropic groundmass with vesicular fragments embedded in it. 

 Another similar rock [80] has the appearance of a fine-bedded 

 cleaved slaty ash. Black chloritoid flakes have been developed along 

 the cleavage-planes in some cases [99] by the crushing. 



The tuffs to the north of Carn Pran, on the road to Dinas, have 

 the greenish colour of the lavas of whose fragments they are com- 

 posed, due to the presence of multitudes of small greenish grannies. 

 Near Pwll Deri thin interbedded tuffs occur [102], being the lateral 

 modification of the coarse breccias of Caerlem. Again, at Porth 

 Sychan tuffs occur. The coarse breccias of Goodwick and Caerlem 

 have already been described (p. 159). 



Many of the rocks show their fragmental origin much more clearly 

 in weathered hand-specimens than under the microscope. But still 

 a large number, having suffered almost complete replacement of their 

 groundmass and inclusions by quartz, must remain doubtful [255], 

 [244], [239], and [284]. 



Several, however, of these apparently fragmental rocks, or whose 

 original character it is difficult to decide, are probably contempo- 

 raneously brecciated lava-streams — broken up and re-cemented 

 coulees. They have a peculiar streaky look, or show included 

 angular fragments. The rocks, too, are more compact and homo- 

 geneous than, as a rule, are those composed entirely of pyroclastic 

 materials. 



Such rocks are found between Bwlch Mawr and Pishguard, 

 especially around and on Carn Pica [248], [236], and [260]. The 

 process of their formation must have been somewhat as follows : — 

 The solidified crust of the lava-stream was broken up and shattered 

 by a suddenly increased rapidity of flow or tension due to a fresh 

 outpouring of lava ; the fragments of the crust sank down into the 

 molten material below, and were cemented together in a matrix of 

 fresh lava as a pseudo-breccia. In many cases the outlines of the 

 fragments are extremely indistinct and the patches merge gradually 

 into the surrounding rock ; this points to a partial melting-down of 

 the fragments, and to a greater or less absorption of them into the 

 liquid magma; in the plastic state to which they were thus reduced 

 they were dragged out into long wisps or streaks in the direction of 

 flow. These features are well shown in the Carn Pica rocks [243], 

 [280], and [240]. Such re-cemented breccias have been described 

 by Harker in the Cross Pell inlier, and Brb'gger has ascribed a some- 

 what similar mode of origin to certain rocks in the Christiania 

 district. 



