lowlands. Underground passages, high above the present water-channels, were swept clean by the body of water forced through them under enormous pressure. Caves that had been sealed up for years with barriers of stalagmite, which one would have thought might have defied the rush of any flood, were burst open. Most of this débris-all, in fact, that was moved by the first rush of water-was carried down the valley. Some remained around the mouth, and some in embayed corners in the caves. Here we saw fragments of stalagmitic floors, mixed up with débris washed in from the swallow-holes above. Some might have seen here evidence that, after the cave had been formed and occupied and gently filled by earth, and coated and partitioned by stalactite and stalagmite, there came an age of flood, perhaps of submergence, when the old deposits were re-sorted, the old floors broken up, and that the cave then entered upon another phase of its history. How different the facts! saw this revolution taking place. It was all over in three short hours. It was another illustration of the great law of Uniformitarianism, which I have heard the Duke of Argyll well state thus: Local catastrophic action is not inconsistent with continuity of causation.

We must bear these things in mind when we are examining cave-deposits.

The peat torn away from the mountain-side above was so beaten up in this great natural churn that the water of the tarn did not get clear for months. The sediment did not settle for three weeks in a long glass which I filled during the flood. There must have been a layer of fine carbonaceous clay formed over the bottom of the tarn and in many a deep cave-pool after that storm. When the rain ceased, the water soon ran off the mountain-side, and I went up to examine Gaping Gill, the great swallow-hole that feeds the cave. I found a passage opened out among some blocks on one side of the stream a little above the chasm. I thought I might perhaps find a zigzag descent, which would lead me down into Gaping Gill Hole. So I crept in.

I soon got beyond the light, and therefore took the precaution of throwing stones in front of me before I advanced. I found the slope increased rapidly, and then all of a sudden the stones dropped into a deep hole, down which they whirred, knocking the sides here and there till they dropped, with a booming noise, into deep water below. I wriggled out, and returned another day, with friends and candles and string, for I could not drop the stones straight so as to clear the sides, and so estimate the depth by the time they took in

falling. Sometimes the weight I attached to the string was too small, so that the increased weight of the string itself, when wetted by the splash of underground waters, prevented my being able to judge whether my plummet had touched the surface of the water below or not. Sometimes the jagged rocks cut my string, and I lost hundreds of feet in this way. At last, however, I got the right sort of string and a convenient weight, and I found that the water here plunged into a vertical hole 360 feet from the grass-covered turf above.

This was not, however, the principal chasm, and I saw a curious sight on the southern, or lowest, face of the great chasm beyond: it was battered and bruised as if it had been bombarded for hours, and so it had. In that flood hundreds of boulders, carried forward by the rush of water, were hurled against the opposite face of rock, and then, dropping into the great chasm, were hurried away through the subterranean watercourses and caves down to the valley far below, where they still rolled on with a noise like thunder over the smooth, rocky bed of the stream, till arrested when the velocity of the water was checked in the wider spaces, or finally stopped in the little tarn below.

Here was the whole story of the formation and infilling of limestone caves, and the sudden breaking up of all the older deposits and the return of tranquil deposition, to be read in Nature's clearest writing.

First we saw the results of the chemical action of the acidulated water running off the peaty moor, and opening out the crevices in the jointed limestone.

Then there was the mechanical action observed on a grand scale in storm, the boulders and pebbles pounding away the solid rock. And next there were the sand and mud left as the water subsided, and the old state of things returned.

Another curious fact I noticed, which shows how the fragmentary rock is rubbed down into mud by the action of running water. There was a fetid smell arising from this flood water, such as the people about there said they had not perceived before. I followed up the stream, and noticed a great quantity of black sand thrown down here and there along its course. This was derived from the bituminous limestones of the lower part of the Yoredale rocks and the upper part of the mountain limestone, and I at once suspected the cause of the smell. When I rubbed a handful of this sand together there was the same fetid smell at once produced. The air tangled in the seething flood was carried down the valley, and, when released, gave off the gases caught up from the pounded rock.

As we cannot follow these watercourses down from above through all their subterranean wanderings, let us go down into the valley below where the water comes down, and see if we can work our way back into the hill towards the foot of the great chasm, and see what is going on there. It is here we find what is more properly a cave being formed. The water drops from one level to another, then runs along between the beds, and drops again. By putting your ear to the fissured rock in one place, you can hear, from the deep recesses of the earth, the sound of a waterfall that man has never seen. Not far off, a beautiful clear river flows out of the lower cave. This is 600 feet below the swallow-hole, where the water enters on the hill above. When the rain floods the stream above, this, too, runs turbid. Some 20 feet above it is the entrance to the other cave, the celebrated Ingleborough Cave, a more ancient outfall for the water, which now runs at the lower level.

This cave was explored many years ago by Mr. James Farrer. I have followed it for about a quarter of a mile, and, with some others, been let down to a lower level at the end. We squeezed our way along till we came to a long, deep cave, full of water, which seemed to flow gently towards the mouth of the lower cave. In the great flood of 1872, all the subterranean caves and fissures were filled, and the water spouted out of the upper cave, carrying along with it great masses of rock, which helped to break up the stalagmitic floors and barriers. This flood was This flood was so exceptional that most of the débris was carried clean away; but we saw, when we examined the ground round the mouth of the cave, and the well-known passages inside, what had been going on; how stalagmitic floors had been undermined, broken up, and re-deposited, and how the torrent débris was sometimes left in the embayed corners of a limestone cave. But this was a cave not far above the existing watercourse. When a cave has been formed in the side of a rapidly-deepened gorge, where, however high the flood may rise, the water can never sweep it out with a rush, gentler processes of denudation and deposition still go on. The débris that falls about the mouth ponds back the rain, and gathers in the fissured rock, and turns in the rivulet that would have trickled down the hill. The damp clay clings to the rock and frets away its surface, and things washed in work their way down along the face of the opening, gradually-weathered limestone, and lie in clay washed down with them.

It is easy to distinguish the chemically-fretted rock from that which has been worn, smoothed, and rounded by the

mechanical action of the sand and pebble-laden water; as you can distinguish the pholas-bored rock from that in which the holes are due to weathering. On the chemically-weathered surface the less soluble grains and bands stand out. This is a useful test.

When any partly-closed cave is invaded by periodic rushes of rain-water, the débris is carried down from above through fissures, or washed in from the mouth, and so we find re-sorted drift and the material of the rainwash from the surface-soil outside the cave occurring also in layers in the cave; and if the cave happens to be occupied by wild animals when not flooded, we find their bones and the remains of their food scattered over the floor or buried in the rainwash.

But when the turbid water fills a pool in the cave or a pond outside it, and the mud is allowed to settle down quietly, the coarser falls first and the finest last. Then the water evaporates or soaks through the sides, or perhaps remains clear and tranquil till the next rain carries in a flood of muddy water. The deposit so formed will have a tendency to split along the layers of coarser sand or loam which first · settled down after flood; that is, it would be a laminated clay. As long as the pool was about the same depth, and the amount of mud carried in suspension in the water was the same, the thickness of the lamina would be practically the same, representing just the mud in one pondful of turbid water, whatever the interval between the refilling of the pond might be. The turbid water may come from the bottom of a glacier, or from melting snow, or from a heavy rainfall; but it certainly has no necessary connexion with glacial action. We see laminated clay so formed commonly in the corner of any old quarry, in ditches, or in caves.

In Chapel le Dale, a valley on the west side of Ingleborough, there is a beautiful chasm which has been so opened out by the action of the torrent that you can get down to the bottom, where the water plunges on to a bed of broken rock and pebbles, through which it passes, as through a sieve or very coarse filter, into the water-courses that carry it off down Chapel le Dale. This great chasm is probably a fair representative of all the large swallow-holes. Hull Pot and Hunt Pot, on the flanks of Whernside, are of the same kind. Probably there is in Gaping Gill somewhere a place where the water in ordinary weather filters through coarse gravel, for I have sent down many boards with a notice on each that I would reward any person who brought it back to me, but I have never heard of one of my notices being found. Yet at times great boulders do get through, so it may be that the

paint of my notices was destroyed in the subterranean waterfalls and rapids.

These chasms or funnel-shaped holes are the feeders of the caves. They are only vertical caves formed in the horizontal surface of the rock. They are known as Swallow-holes, Potholes, Sink-holes, and in Italy as Dolinas. They have various local names, expressing the idea that they are not part of the more regular and common operations of nature: the Devil's Chaldron, as in French, Chaldrons du Diable, Marmites des Géants, Bêtoires, or, more simply named, they are the Katabothra of the Greeks.

They begin sometimes under the covering of drift, and, when the opening grows too large, or the covering soil is sodden and will not hold its own weight together, the surface breaks in. Mr. Haythornthwaite, of Kirkby Lonsdale, told me that on a farm of his above Wethercote Cave, after wet weather, he once saw one fall in.

How swallow-holes are formed in chalk has been described by Prestwich.*

The age of the cave-deposits is quite a separate question from that of the caves themselves. The formation of the caves was a time of destruction; but the infilling of the caves belonged to a time of accumulation—when there was no great scour through the caves, but the rain carried in earth and stones, if there was loose drift above, or only muddy water if the cave was nearly closed, or perhaps nothing was deposited but the fine unctuous clayey residuum of the chemicallydecomposed limestone itself. Angular fragments disengaged by frost or heat formed a barricade about the mouth. Bones were washed in or carried there by beasts of prey— and man. Bucklandt referred most of the caves that he explored to hyena-dens. Constant Prevost‡ thought the bones that occurred so thickly in the cave-earth in Franconia were all washed in by torrents. This explanation will hold only in exceptional cases. The bones may have been washed from one part to another of a cave, and a few do get washed in from above. I have seen three sheep being carried down towards a swallow-hole, and have found two drowned rabbits and some dead trout on the gravel at the bottom of Hunt Pot, on the flanks of Whernside. But we never see the ground so covered with bones of various animals that a flood

* Q. J. G. S. vol. x. 1854, p. 222.
+ Reliquiæ Diluvianæ.

Mem. Soc. d'Hist. Nat. Paris, t. iv.