VOLCANIC SERIES OF ST. DAVIDS. 247 



the cliff into and through the arch, and the junction of the con- 

 glomerate with the underlying ]*ebidian may without difficulty or 

 danger be followed up the cliff to within about 6 feet of the grass 

 slopes. 



The beds immediately beneath the conglomerate very clearly 

 indicate the oncoming of sedimentary conditions, which is also indi- 

 cated by the porcellanites. There are fine-grained pink and white 

 beds, a red felspathic tuff with numerous rounded grains of quartz, 

 greenish shaly bands, white quartzose beds, and a very close-grained, 

 white, porcellanitic material. Upon the somewhat hollowed surface 

 of these beds lies the conglomerate. There is about 8 or 10 feet 

 transgression of the conglomerate across the Pebidian strata. Near 

 the foot of the cliff, for example, the soft pink and white beds are 

 about 6 feet below the conglomerate. Higher up the cliff the con- 

 glomerate rests upon this bed and it is washed in among the pebbles, 

 large stones being well imbedded in this material. I examined this 

 locality very critically and carefully. In my judgment the facts 

 favour Dr. Geikie's view (contemporaneous erosion ) rather than that 

 of Dr. Hicks (systematic unconformity). 



In the westerly creek at the foot of the cliff, below .f, the junc- 

 tion of the conglomerate with the underlying Tcbidian, here grey 

 with small water-worn loebbles from | to ^ an inch long, may be 

 seen. The line of junction is an uneven one, the conglomerate 

 V-ing into the subjacent material. I could see no satisfactory 

 evidence here of a systematic unconformity. 



Nor is such evidence afforded by the detached conglomerate 

 masses between this creek and the arch. The uuderlying beds are 

 similar to the red sandy beds with rolled (quartz pebbles found 

 beneath the conglomerate at 8t. Non's arch, and well seen on the 

 top of the little promontory which the archway pierces. 



In St. Non's Bay, then, together with clear and unmistakable 

 signs of local or contemporaneous erosion, I find the general paral- 

 lelism of the strike most marked. There is no evidence of the 

 bending round of the conglomerate against the strike of the Pebidians 

 drawn by Dr. Hicks in his " Plan of Coast soutli of Nun's Chapel " 

 Q. 0. fig. 3, p. ^36), while the unconformity shown in his tig. 2 

 (p. 525) is much exaggerated. The incoming of sedimentary con- 

 ditions is also noteworthy. 



The contact of the conglomerate with Pebidian is again seen 

 further wo^t at Ogof Llesugn : but a description of this locality may 

 be advantageously deferred. 



We come now to the junctions seen on the northern side of tbe 

 axis. 



d. At Oijof GoJchfa. — This locality is a mile and a half north-west 

 of St. David's city, situate at the soutli end of Whitesand iiay. 



Between Perth Sele and Whitesand Bay the conglomerate is well 

 exposed and is here of considerable thickness. Tracing the cliff 

 from Perth Sele northwards we have tlie Haggy sandstones of the 

 Caerfai series striking N. 15° E. and dipping seawards at about 60^. 

 These extend for two or three hundred yards, when the succession 



Q. J. G. S. No. 182. * s 



