128 RET. G. C. H. POLLEN ON THE [Feb. 1 898, 



separate vertically (see fig. 4, p. 129). In each case we have taken 

 the measurements from the highest point of the roof. 



In the entrance is seen a large block of limestone below, which is 

 the only sign of a floor with which we have met, even if it can truly 

 be said to represent a floor. We hope to be able in the course of 

 next year to excavate at a lower level, and if possible to find the 

 true floor of the tunnel. Up to sect. IV ^ the sequence is quite 

 clear, and in the upper beds to sect. VI also, but the middle and 

 lower beds were too confused to follow exactly. 



From sect. VI to sect. VII we despaired of taking sections 

 which would fairly represent the beds, owing to the presence of large 

 blocks of limestone and extreme dampness. The upper laminated 

 clay and gravel were alone measured accurately. The succession of 

 the other strata is correct, but I caimot say that they are strictly to 

 scale. We can also connect all the beds here with those which 

 extend beyond sect. VII, except the second sandy gravel. This 

 most probably unites with the upper bed at sect. VII, and both 

 continue together, but it may have become more clayey and been 

 merged in the clayey gravel below, since this contains a few pieces 

 of stalagmite in its upper part. A possible explanation would be 

 that all the sandy gravel and laminated clay were originally against 

 the roof of the cave, and that the beds have gradually settled down, 

 the present top part being retained in position by two projecting 

 shoulders of the walls. The sand, on this supposition, would have 

 been introduced later through the shaft above. 



This opening is 16 feet 2 inches high from the roof of sect. VI, 

 but the top is not the original surface, as several feet have been 

 quarried away quite recently. 



At a height of 5 feet from sect. VI, on the south side of the shaft, 

 there is a small gallery, open for 15 feet, but too narrow to explore 

 further. There was about 1 foot of mixed clay and sand on the 

 floor, but no lamination or stratification was visible in this bed. 



From sect. VII onward every line represents a distinct bed, 

 which was not only measured at the subsequent sections, but was 

 also followed up carefully from one point to another. 



In all this portion of the cave there are many large blocks of 

 limestone, especiallj^ in the upper strata. I have not figured these, 

 as they did not prevent us from following up the various strata. 



Bej'ond sect. XI there is a second opening in the roof. This is 21 

 feet high. It is filled with a bed of gravel, of a different character 

 from any with which we have met before, composed of angular lime- 

 stone in sand or clay, with a few minute grains of Denbighshire Grit 

 here and there, and some small patches of blue clay. The top part, 

 for 5§ feet, had some larger stones with very distinct glacial striae, 

 and they include some erratics from the Snowdon district. 



The laminated clays and gravels all pass under the shaft, and are 

 found quite distinct on the farther side, though there has evidently 

 been some confusion immediately below the opening. 



^ The Eoman numerals indicate the various transverse sections taken. 



