PKE-CAMBEIAN KOCKS OF ST. DAYID's. 305 



from orthorhombic forms with the incipient development of trans- 

 verse fissures filled with haematite, to others of distorted shapes in 

 which the ferruginous matter occupies the whole or nearly the whole 

 of the mould of the original crystal. 



The rocks now described differ from the Palaeozoic porphyrites 

 with which I am acquainted in the less abundance of their micro- 

 scopic base, in the comparatively inconspicuous development of fel- 

 spars, and the absence of large porphyritic felspars, in the extra- 

 ordinary prominence of the augite, and in the presence of olivine. 

 In composition and structure they are essentially forms of olivine 

 diabase. 



I cannot pretend at present to offer more than a mere outline- 

 sketch of the petrography of the Cambrian volcanic group of St. 

 David's ; but from the data here brought forward it will, I think, 

 be apparent that the rocks of that group possess exceptional interest 

 from the extraordinary combination of modern types of structure 

 with so remote an antiquity. 



2. The Quartz Conglomerate, 



The lithological characters of this band and its stratigraphical rela- 

 tions to the beds beneath it have been sufficiently described in Part I. 

 (p. 286). It is essentially a mass of rolled pebbles of quartzite and 

 quartz, embedded in a reddish ferruginous and quartzose matrix. 

 The pebbles vary up to occasional blocks as large as a man's head or 

 larger ; but their average size is probably less than that of a walnut. 

 The conglomerate band continues as a persistent and easily recog- 

 nizable horizon through the St. David's district, but presents notice- 

 able variations in thickness and in coarseness of materials. The 

 pebbly beds are lenticular, rapidly wedging out and passing into 

 fine grit and sandstone. In some places the total thickness of the 

 band dwindles down to possibly not more than two or three feet ; 

 in others, as on the south-east of St. David's, it swells out to 

 more than one hundred feet. I have alluded in the first part of 

 this paper to the perfect conformabifity of the conglomerate with 

 the top of the volcanic group, and to the intercalation of bands 

 of tuff in it (fig. 8, p. 290). These facts prove that no abrupt break 

 can be traced between the volcanic group and the conglomerate. I 

 have also referred to the presence of occasional seams of rolled quartz 

 pebbles in the tuffs, as indicating that the conditions of deposit to 

 which the conglomerate was due had begun to appear even during 

 the volcanic period. It is obvious, however, that the intercalation 

 of the marked band of quartz conglomerate points to an important 

 change in the sedimentation of the time. It suggests some inter- 

 esting questions of general interest, to which reference may here be 

 made. 



There can be no doubt that conglomerates frequently mark the 

 natural base of a series of sedimentary deposits. They do so more 

 especially where they are formed of materials that have had an 

 obviously local origin, and where they rest unconformably on the 



