The Devil's Punchbowl, near the Castle of Comfort Inn on the Mendips, is, in all probability, a collapse of the remarkable Lias beds which there occur into some pre-existing cavity in the Mountain Limestone below, somewhat in the same manner as the Shake Holes in the Glacial Drift on the Yorkshire moors were formed. No one questions the existence of the cavities beneath before the deposit of the Drift, neither do I doubt the existence of swallets beneath the Trias and Lias before these were deposited on the Mendips. The question naturally arises, Why do we not find in our caverns remains of all the ages that have elapsed since that time? Why are only Pleistocene remains discovered? Surely, because we have not found them it does not follow that they are nonexistent. The recent discovery of Pliocene remains in a cavern at Doveholes, near Buxton (Derbyshire), is clear proof that we may search hopefully for similar remains in the Mendips. It must be borne in mind, that the further we go back in time, the more certain we are to find that the contents of any Limestone cavern would be completely mineralised, until the whole of the contents may have become cemented into a solid mass. Where running water is present, attrition may have destroyed them, or borne them onwards to those great depths where, constantly submerged as they must be, we can never hope to penetrate. I am aware, however, of the existence, in the Eastwater Cavern, of very ancient chokes of water-borne material, from which I have some hope of obtaining remains. I might mention the demonstrated antiquity of the bosses of stalagmite in Kent's Cavern at Torquay, and from it argue the immense age of the great masses of stalagmite in the Mendip Caves, but, recognising the variable rate of deposit of the carbonate of lime in different caverns, and indeed in different parts of the same cavern, no useful purpose would be served thereby. The huge Beehive of Lamb's Lair at Harptree, the large boss in the first great chamber at Wookey Hole, Gough's "Niagara" at Cheddar, the tall and slender pillars in Cox's Cave at Cheddar, and the taller "Sentinel" pillar at Wookey Hole, all demand for their formation a prodigious length of time, which it is but folly to attempt to compute with our present information. Certainly many thousands of years are required for some of them, and it should be remembered that we have then arrived merely at the time when the floor upon which they stand had received its final form, the action of running water having ceased.[2] Who can doubt then, that, as we stand in the great waterways of the profound depths of our hills, we are looking upon scenes which have varied little since remote ages, and that in some form or other these waterways played an important part in the degradation of the earlier and loftier Mendip range?