Is there evidence of Glacial Action in the Valleys of Dawlish and Ashcombe, South Devon? (1872)

By George Pycroft. (Read at Exeter, July, 1872.)

The little stream which falls into the sea at Dawlish, and which the inhabitants of that place have made the most of by widening and damming with artificial weirs, is so small and unimportant that it has never received a name. It is about six miles long, four feet broad, and deep enough in some places to allow a duck to float without touching bottom. But insignificant as this streamlet is, it has in the lapse of ages cut its way so deeply through the rocks over which it originally flowed, and has thus exposed such interesting geological sections, that I have carefully traced it from its one mouth to its five sources, and have endeavoured to learn from babbling brook the secrets of the valley. The vale of Dawlish is a continuation of that of Ashcombe. They are valleys of denudation, scooped to a depth of seven hundred and odd feet in the new red sandstone. They are shut in from the south, west, and north-west, by Little and Great Haldon hills, which rise to the height of 850 feet. The new red sandstone of these hills is capped with gault about ninety feet in thickness, and this again by a layer of flints without chalk. In the valley itself the new red sandstone is covered with a layer of superficial deposits, through which the brook has cut its way, and of which I shall have more particularly to speak.

The stream rises from five sources. The principal source is at Grammercombe, in Great Haldon; a smaller source exists at the Thorn’s Plantation; a third at Charnwood; a fourth at the side of Little Haldon, below the Tower Plantation; and the fifth at Lidwell Chapel. There is also a very small tributary at Dawlish water. These sources have certain characters in common. They all arise at the junction of the permeable green sand and the denser new red sandstone, flow precipitously down the sides of Haldon, cutting their way deeply through the new red sandstone rock, then through a mass of superficial deposits resting on the red, and so find their way to the bottom of the valley.

Commencing from the mouth of the stream, and tracing it upward, I find the first mile converted into an artificial piece of water, and containing nothing to chronicle. Above this, at the commencement of Luscombe grounds, the water has cut through the rock, leaving escarpments above twelve feet in depth, showing a confused mass of gravel, with very little signs of stratification. Here in the stream itself, and the fields adjoining, many boulders of porphyritic trap are met with, some three feet across, some less. At this point the tributary from Lidwell Chapel joins the main stream, and tracing this upwards, I find for the first two miles the streamlet flowing in an artificial channel. At Higher Southwood Farm blocks of trap are very common in the farm-yard and in the road adjoining – one block, about four feet in length, has been hollowed out, and converted into a drinking trough for cattle. Farther up the stream, at a height of 271 feet above the sea level, is situated Lidwell Farm, and in a field just above is the second largest boulder I have met with, measuring 15 feet in circumference. As the stream is followed up, many more are observed in the fields; and far above Lidwell Chapel, at a height of 702 feet, is a large rounded boulder, the highest I have met with anywhere in the valley. A little lower down another lies on the surface, measuring about 12 feet in girth. I met with no limestone in any part of this stream; but some parts are so difficult to get at, that I cannot say with certainty that none exists there. Returning to the main stream at Luscombe, I find it flowing through an alluvial valley, showing nothing worthy of mention, except that at Dawlish water there are a great number of the trap boulders in the banks, roads, and fields. A small tributary here joins the stream, and on following it up to its source on Little Haldon, I find that it has cut its way through a ravine 20 feet and upwards in depth by 50 in width; but as its sides are clothed with vegetation, a good section is not anywhere to be obtained. At one spot, however, I was able to see that the boulders are embedded in a red clay. In this stream are a few hard specimens of water-worn limestone; but as the remains of a limekiln are to be found high up the ravine where Chudleigh limestone used formerly to be burnt with wood, I cannot say that the limestones may not have found their way hither in a waggon. All through the ravine, and in the roads and fields near, are to be found porphyritic blocks.

The main stream flows through flat grass lands, and exhibits nothing worth mentioning till the ford at Ashcombe schoolhouse is reached. Here the valley narrows considerably, and the wilder part of the stream (so to speak) commences. A few years ago the stream was blocked up by huge boulders of trap; but I am sorry to say that most of these monuments of the past have been removed or broken up, to admit of the building of a small weir. There are still remaining about two dozen blocks, the largest of which measures 14 feet in girth, and 2 1/2 feet in height, and possesses the peculiarity that I have frequently observed in other specimens of having one flat side.

Higher up, between the school and Midtown, are numerous boulders; and near Midtown, projecting from the bank, entirely covered and concealed by a coating of moss, I discovered a block of limestone, the part of which was above ground measured five feet by three.

At Midtown, where the stream receives the waters of its tributaries from Haldon Tower Plantation and Charnwood, the boulders are very numerous in the stream and the adjoining roads and fields. Indeed, they are not the exception, but the rule. Walls, cottages, and bridges are built of them; they are hollowed out for cattle-troughs, and the village appears as if built in a granitic country. Wherever there is a road, section, or a little landslip, they are found abundantly; they are by no means confined to the streams, but exist in all parts of the valley. No doubt formerly they were to be found in much greater numbers; even in the last 20 years many have been removed, but they are still in sufficient numbers to arrest the attention of the most ordinary observer.

From this point upwards all the streams contain limestone blocks.

By the river side, about half a mile above Midtown, I observed, on a small scale, what I afterwards found on a much larger one, that the following is the order of stratification in this valley. Deepest down the ordinary new red sandstone conglomerate of a fine grain. Resting on this lay a bed or stratum of coarse conglomerate or breccia, containing limestone and trap pebbles, and cemented together, probably, by lime – a regular breccia, in fact, formed of the materials of the deposit above the new red sandstone. Above this lay a stratum of red earth, containing large blocks of limestone, trap, and flints, from the Haldon gravel, lying higgledy piggledy, perfectly unstratified, the blocks not even arranged according to gravity.

To avoid repetition, I may as well say at once, that where I found a section the same order of things appeared – always the red sandstone at bottom, always a breccia of trap and limestone resting upon it, always red mud above this containing boulders. I never found the boulders resting upon the red rock, although they may do so. The breccia higher up the valley is very coarse, sometimes containing stones as large as one’s head, sometimes larger. In one place I found the breccia surmounted by about 10 feet of fine red sand, evidently a re-deposit; and further on I observed this passing into a red mud or earth, out of which emerged a huge trap boulder. It was not in the red rock, nor even on the red rock, but eight feet above it, lying in a superficial deposit.

The mud or clay in which the boulders are embedded proved, on examination, to consist mainly of the débris, or the result of decomposition of trap stones. Many of these soften under the influence of water, and can be spread like mud or clay. The red earth evidently did not derive its origin from the sandy rock beneath, but was the result of the disintegration and decomposition of the igneous rocks.

I remarked that the red earth in some good sections contained no other stone but trap boulders alone, with the limestone boulders at its base between it and the new red sandstone, and in these cases not infrequently the red earth was capped with Haldon flint gravel. In the lowest part, however, of the deposit, even when the three deposits were distinct, I detected numerous pieces of Haldon green sand.

I did not observe any fragments of limestone in the underlying new red sandstone, except in one instance, when a piece of limestone, about 18 inches in length, was firmly embedded in the surface of the red rock, underneath the breccia.

I followed up this, and afterwards the Charnwood branch, to their sources in the green sand, and have nothing more to say of them than that they contained many blocks of trap and limestone.

Returning to the valley by the road which leads from Haldon to Midtown, I observed a solitary boulder of no great size projecting from the red earth of the road-side cutting. It had the peculiarity previously referred to, of having one flat side; but in this case in addition, the flat base was as smooth as a table, and as nearly polished as the nature of the stone would admit.

The stream which rises from Thorn’s Plantation, and falls into the main stream at Court Farm, is crowded with trap and limestone boulders – from its source to its mouth. One trap boulder measured 10 feet by 5. A weathered block of limestone, from the number of fossils it contained, of what precise kind I could not determine, with encrinite stems, and numerous spiral shells, reminded me of the blocks found at Babbacombe, and another closely resembled the limestone of the Bradley valley.

From the bridge near the church to the river’s source at Grammercombe, the exploitation of the river is most difficult, it being so full of brambles and covered in by trees; indeed, there is no way of seeing it except by walking in the water. Frequently I was reduced to crawling on my hands to enable me to make any progress at all. At one point I almost gave up further travel in despair, when I was encouraged by finding a block of porphyritic trap lying across the stream of unusual magnitude. Part was buried in the bank; the part visible measured 13 feet by 8. Many parts of the stream showed good bank sections, and all parts, to within half-a-mile of its source, were filled with limestone boulders from two to five feet in diameter, mingled with rounded masses of trap. The appearance of the river’s bed was that of a steam flowing through a limestone country.

Throughout my paper I have mentioned the boulders as found in the streams; this is simply because they are there most easily seen. They are, in fact, scattered all over the whole valley, so much so that if the earth could be removed, and all the boulders brought to light, the valley would appear as great a wilderness of stones as Lustleigh Cleave.

Boulders are found in the valley to the west of Little Haldon, but I have never seen one on any part of the summit of either Great or Little Haldon. There is one, however, which has been pointed out to me by Mr. Hutchinson, of Sidmouth, lying on the green sand at Holcombe Down. This, however, being only about 2 1/2 feet long by 18 inches wide, and about the same height, may, as he suggest, have been brought hither by human hands. I have not seen any in the Exe valley, or even on the other side of the low ridge of hills to the east of the vale of Dawlish. There are numerous trap stones of a small size, and a large quantity of ash lying superficially, as we approach Haldon Belvidere, and the Dunchideoch end of the Kenn Vale; but the close proximity of the Haldon, Dunchideoch, Idestone, and Pocombe igneous rocks will easily account for this. Although I have been continually riding in every part of the right bank of the Exe from Dawlish to Dunchideoch, I have never in one instance detected a block of limestone.

What conclusions may be drawn from these observations? We have a valley six miles in length by about two in width, the bottom and sides of which are filled with unstratified red earth or mud, resting on a breccia formed of stones immediately above, and in this red earth many thousands of erratic boulders of trap and limestone weighing from one to six tons. We find some trap boulders with smooth sides, and frequently both trap and limestone blocks mingled in confusion with Haldon flints.

Two questions arise from this :—
1st. Whence did these erratics come?
2nd. How did they come?

1st. Whence did they come?
The nearest point at which limestone is now found in situ is near Ideford, about three and a half miles as the crow flies to the west. There are also the Chudleigh and Lyndridge rocks, about four miles to the west, with Haldon Hill intervening. The nearest known point at which igneous rocks at all resembling the trap boulders is found is at Idestone and Knowle, about six miles to the north, and Pocombe and Dunchideoch. De la Beche instances the rocks of the two former places as nearest resembling the trap boulders, as they contain crystals of quartz and felspar;* but whatever actual spot they came from, they must, I presume, have come from the junction of the new red sandstone and carboniferous rocks, which takes place at the N. and W. Probably, therefore, both limestone and trap boulders were brought from the N. and W.

* Mr. Vicary, of Exeter, has shown in a previous number of the Devon Association Transactions that these porphyritic blocks are in truth granite. If this be the case, their original home was Dartmoor, and their journey hither far longer.

2nd. How did they come?
Not scattered by the shot of a volcano, for volcanoes do not distribute limestones. By water? Without insisting upon the difficulty and the enormous time required to shift from two to six ton blocks so many miles, I think that had water been the agent there would have been more appearance of stratification, the blocks would have arranged themselves somewhat according to their gravity, and the bed in which they are found would be shingly or sandy, and not red mud or clay, formed of the breaking up and decomposition of the transported blocks themselves.

They might have been brought hither and scattered broadcast by ice, and this, I suggest, is probably the solution of the difficulty. I submit that, when we consider the great weight of the erratics, the distance from which they must have been transported, the irregular way in which they are distributed, the unstratified, confused way in which they are deposited, the red mud in which they lie, and the finding them occasionally with flat polished sides, and the breccia between them and the subjacent rock, we are justified in raising the question whether their long travel was not due to glacial action.