Denudation aided by recently ejected castings flowing down inclined grass-covered surfaces • The amount of earth which annually flows downwards • The effect of tropical rain on worm castings • The finest particles of earth washed completely away from castings • The disintegration of dried castings into pellets, and their rolling down inclined surfaces • The formation of little ledges on hill-sides, in part due to the accumulation of disintegrated castings • Castings blown to leeward over level land • An attempt to estimate the amount thus blown • The degradation of ancient encampments and tumuli • The preservation of the crowns and furrows on land anciently ploughed • The formation and amount of mould over the Chalk formation.
We are now prepared to consider the more direct part which worms take in the denudation of the land. When reflecting on sub-aerial denudation, it formerly appeared to me, as it has to others, that a nearly level or very gently inclined surface, covered with turf, could suffer no loss during even a long lapse of time. It may, however, be urged that at long intervals, debacles of rain or water-spouts would remove all the mould from a very gentle slope; but when examining the steep, turf-covered slopes in Glen Roy, I was struck with the fact how rarely any such event could have happened since the Glacial period, as was plain from the well- preserved state of the three successive "roads" or lake-margins. But the difficulty in believing that earth in any appreciable quantity can be removed from a gently inclined surface, covered with vegetation and matted with roots, is removed through the agency of worms. For the many castings which are thrown up during rain, and those thrown up some little time before heavy rain, flow for a short distance down an inclined surface. Moreover much of the finest levigated earth is washed completely away from the castings. During dry weather castings often disintegrate into small rounded pellets, and these from their weight often roll down any slope. This is more especially apt to occur when they are started by the wind, and probably when started by the touch of an animal, however small. We shall also see that a strong wind blows all the castings, even on a level field, to leeward, whilst they are soft; and in like manner the pellets when they are dry. If the wind blows in nearly the direction of an inclined surface, the flowing down of the castings is much aided.
The observations on which these several statements are founded must now be given in some detail. Castings when first ejected are viscid and soft; during rain, at which time worms apparently prefer to eject them, they are still softer; so that I have sometimes thought that worms must swallow much water at such times. However this may be, rain, even when not very heavy, if long continued, renders recently-ejected castings semi-fluid; and on level ground they then spread out into thin, circular, flat discs, exactly as would so much honey or very soft mortar, with all traces of their vermiform structure lost. This latter fact was sometimes made evident, when a worm had subsequently bored through a flat circular disc of this kind, and heaped up a fresh vermiform mass in the centre. These flat subsided discs have been repeatedly seen by me after heavy rain, in many places on land of all kinds.
On the flowing of wet castings, and the rolling of dry disintegrated castings down inclined surfaces.—When castings are ejected on an inclined surface during or shortly before heavy rain, they cannot fail to flow a little down the slope. Thus, on some steep slopes in Knole Park, which were covered with coarse grass and had apparently existed in this state from time immemorial, I found (Oct. 22, 1872) after several wet days that almost all the many castings were considerably elongated in the line of the slope; and that they now consisted of smooth, only slightly conical masses. Whenever the mouths of the burrows could be found from which the earth had been ejected, there was more earth below than above them. After some heavy storms of rain (Jan. 25, 1872) two rather steeply inclined fields near Down, which had formerly been ploughed and were now rather sparsely clothed with poor grass, were visited, and many castings extended down the slopes for a length of 5 inches, which was twice or thrice the usual diameter of the castings thrown up on the level parts of these same fields. On some fine grassy slopes in Holwood Park, inclined at angles between 8 degrees and 11 degrees 30 seconds with the horizon, where the surface apparently had never been disturbed by the hand of man, castings abounded in extraordinary numbers: and a space 16 inches in length transversely to the slope and 6 inches in the line of the slope, was completely coated, between the blades of grass, with a uniform sheet of confluent and subsided castings. Here also in many places the castings had flowed down the slope, and now formed smooth narrow patches of earth, 6, 7, and 7.5 inches in length. Some of these consisted of two castings, one above the other, which had become so completely confluent that they could hardly be distinguished. On my lawn, clothed with very fine grass, most of the castings are black, but some are yellowish from earth having been brought up from a greater depth than usual, and the flowing- down of these yellow castings after heavy rain, could be clearly seen where the slope was 5 degrees; and where it was less than 1 degree some evidence of their flowing down could still be detected. On another occasion, after rain which was never heavy, but which lasted for 18 hours, all the castings on this same gently inclined lawn had lost their vermiform structure; and they had flowed, so that fully two-thirds of the ejected earth lay below the mouths of the burrows.
These observations led me to make others with more care. Eight castings were found on my lawn, where the grass-blades are fine and close together, and three others on a field with coarse grass. The inclination of the surface at the eleven places where these castings were collected varied between 4 degrees 30 seconds and 17 degrees 30 seconds; the mean of the eleven inclinations being 9 degrees 26 seconds. The length of the castings in the direction of the slope was first measured with as much accuracy as their irregularities would permit. It was found possible to make these measurements within about of an inch, but one of the castings was too irregular to admit of measurement. The average length in the direction of the slope of the remaining ten castings was 2.03 inches. The castings were then divided with a knife into two parts along a horizontal line passing through the mouth of the burrow, which was discovered by slicing off the turf; and all the ejected earth was separately collected, namely, the part above the hole and the part below. Afterwards these two parts were weighed. In every case there was much more earth below than above; the mean weight of that above being 103 grains, and of that below 205 grains; so that the latter was very nearly double the former. As on level ground castings are commonly thrown up almost equally round the mouths of the burrows, this difference in weight indicates the amount of ejected earth which had flowed down the slope. But very many more observations would be requisite to arrive at any general result; for the nature of the vegetation and other accidental circumstances, such as the heaviness of the rain, the direction and force of the wind, appear to be more important in determining the quantity of the earth which flows down a slope than its angle. Thus with four castings on my lawn (included in the above eleven) where the mean slope was 7 degrees 19 seconds, the difference in the amount of earth above and below the burrows was greater than with three other castings on the same lawn where the mean slope was 12 degrees 5 seconds.
We may, however, take the above eleven cases, which are accurate as far as they go, and calculate the weight of the ejected earth which annually flows down a slope having a mean inclination of 9 degrees 26 seconds. This was done by my son George. It has been shown that almost exactly two-thirds of the ejected earth is found below the mouth of the burrow and one-third above it. Now if the two- thirds which is below the hole be divided into two equal parts, the upper half of this two-thirds exactly counterbalances the one-third which is above the hole, so that as far as regards the one-third above and the upper half of the two-thirds below, there is no flow of earth down the hill-side. The earth constituting the lower half of the two-thirds is, however, displaced through distances which are different for every part of it, but which may be represented by the distance between the middle point of the lower half of the two- thirds and the hole. So that the average distance of displacement is a half of the whole length of the worm-casting. Now the average length of ten out of the above eleven castings was 2.03 inches, and half of this we may take as being 1 inch. It may therefore be concluded that one-third of the whole earth brought to the surface was in these cases carried down the slope through 1 inch. 
It was shown in the third chapter that on Leith Hill Common, dry earth weighing at least 7.453 lbs. was brought up by worms to the surface on a square yard in the course of a year. If a square yard be drawn on a hillside with two of its sides horizontal, then it is clear that only 1/36 part of the earth brought up on that square yard would be near enough to its lower side to cross it, supposing the displacement of the earth to be through one inch. But it appears that only of the earth brought up can be considered to flow downwards; hence 1/3 of 1/36 or 1/108 of 7.453 lbs. will cross the lower side of our square yard in a year. Now 1/108 of 7.453 lbs. is 1.1 oz. Therefore 1.1 oz. of dry earth will annually cross each linear yard running horizontally along a slope having the above inclination; or very nearly 7 lbs. will annually cross a horizontal line, 100 yards in length, on a hill-side having this inclination.
A more accurate, though still very rough, calculation can be made of the bulk of earth, which in its natural damp state annually flows down the same slope over a yard-line drawn horizontally across it. From the several cases given in the third chapter, it is known that the castings annually brought to the surface on a square yard, if uniformly spread out would form a layer 0.2 of an inch in thickness: it therefore follows by a calculation similar to the one already given, that 1/3 of 0.2x36, or 2.4 cubic inches of damp earth will annually cross a horizontal line one yard in length on a hillside with the above inclination. This bulk of damp castings was found to weigh 1.85 oz. Therefore 11.56 lbs. of damp earth, instead of 7 lbs. of dry earth as by the former calculation, would annually cross a line 100 yards in length on our inclined surface.
In these calculations it has been assumed that the castings flow a short distance downwards during the whole year, but this occurs only with those ejected during or shortly before rain; so that the above results are thus far exaggerated. On the other hand, during rain much of the finest earth is washed to a considerable distance from the castings, even where the slope is an extremely gentle one, and is thus wholly lost as far as the above calculations are concerned. Castings ejected during dry weather and which have set hard, lose in the same manner a considerable quantity of fine earth. Dried castings, moreover, are apt to disintegrate into little pellets, which often roll or are blown down any inclined surface. Therefore the above result, namely, that 24 cubic inches of earth (weighing 1.85 oz. whilst damp) annually crosses a yard- line of the specified kind, is probably not much if at all exaggerated.
This amount is small; but we should bear in mind how many branching valleys intersect most countries, the whole length of which must be very great; and that earth is steadily travelling down both turf- covered sides of each valley. For every 100 yards in length in a valley with sides sloping as in the foregoing cases, 480 cubic inches of damp earth, weighing above 23 pounds, will annually reach the bottom. Here a thick bed of alluvium will accumulate, ready to be washed away in the course of centuries, as the stream in the middle meanders from side to side.
If it could be shown that worms generally excavate their burrows at right angles to an inclined surface, and this would be their shortest course for bringing up earth from beneath, then as the old burrows collapsed from the weight of the superincumbent soil, the collapsing would inevitably cause the whole bed of vegetable mould to sink or slide slowly down the inclined surface. But to ascertain the direction of many burrows was found too difficult and troublesome. A straight piece of wire was, however, pushed into twenty-five burrows on several sloping fields, and in eight cases the burrows were nearly at right angles to the slope; whilst in the remaining cases they were indifferently directed at various angles, either upwards or downwards with respect to the slope.
In countries where the rain is very heavy, as in the tropics, the castings appear, as might have been expected, to be washed down in a greater degree than in England. Mr. Scott informs me that near Calcutta the tall columnar castings (previously described), the diameter of which is usually between 1 and 1.5 inch, subside on a level surface, after heavy rain, into almost circular, thin, flat discs, between 3 and 4 and sometimes 5 inches in diameter. Three fresh castings, which had been ejected in the Botanic Gardens "on a slightly inclined, grass-covered, artificial bank of loamy clay," were carefully measured, and had a mean height of 2.17, and a mean diameter of 1.43 inches; these after heavy rain, formed elongated patches of earth, with a mean length in the direction of the slope of 5.83 inches. As the earth had spread very little up the slope, a large part, judging from the original diameter of these castings, must have flowed bodily downwards about 4 inches. Moreover some of the finest earth of which they were composed must have been washed completely away to a still greater distance. In drier sites near Calcutta, a species of worm ejects its castings, not in vermiform masses, but in little pellets of varying sizes: these are very numerous in some places, and Mr. Scott says that they "are washed away by every shower."
I was led to believe that a considerable quantity of fine earth is washed quite away from castings during rain, from the surfaces of old ones being often studded with coarse particles. Accordingly a little fine precipitated chalk, moistened with saliva or gum-water, so as to be slightly viscid and of the same consistence as a fresh casting, was placed on the summits of several castings and gently mixed with them. These castings were then watered through a very fine rose, the drops from which were closer together than those of rain, but not nearly so large as those in a thunderstorm; nor did they strike the ground with nearly so much force as drops during heavy rain. A casting thus treated subsided with surprising slowness, owing as I suppose to its viscidity. It did not flow bodily down the grass-covered surface of the lawn, which was here inclined at an angle of 16 degrees 20 seconds; nevertheless many particles of the chalk were found three inches below the casting. The experiment was repeated on three other castings on different parts of the lawn, which sloped at 2 degrees 30 seconds, 3 degrees and 6 degrees; and particles of chalk could be seen between 4 and 5 inches below the casting; and after the surface had become dry, particles were found in two cases at a distance of 5 and 6 inches. Several other castings with precipitated chalk placed on their summits were left to the natural action of the rain. In one case, after rain which was not heavy, the casting was longitudinally streaked with white. In two other cases the surface of the ground was rendered somewhat white for a distance of one inch from the casting; and some soil collected at a distance of 2.5 inches, where the slope was 7 degrees, effervesced slightly when placed in acid. After one or two weeks, the chalk was wholly or almost wholly washed away from all the castings on which it had been placed, and these had recovered their natural colour.
It may be here remarked that after very heavy rain shallow pools may be seen on level or nearly level fields, where the soil is not very porous, and the water in them is often slightly muddy; when such little pools have dried, the leaves and blades of grass at their bottoms are generally coated with a thin layer of mud. This mud I believe is derived in large part from recently ejected castings.
Dr. King informs me that the majority of the before described gigantic castings, which he found on a fully exposed, bare, gravelly knoll on the Nilgiri Mountains in India, had been more or less weathered by the previous north-east monsoon; and most of them presented a subsided appearance. The worms here eject their castings only during the rainy season; and at the time of Dr. King's visit no rain had fallen for 110 days. He carefully examined the ground between the place where these huge castings lay, and a little watercourse at the base of the knoll, and nowhere was there any accumulation of fine earth, such as would necessarily have been left by the disintegration of the castings if they had not been wholly removed. He therefore has no hesitation in asserting that the whole of these huge castings are annually washed during the two monsoons (when about 100 inches of rain fall) into the little water-course, and thence into the plains lying below at a depth of 3000 or 4000 feet.
Castings ejected before or during dry weather become hard, sometimes surprisingly hard, from the particles of earth having been cemented together by the intestinal secretions. Frost seems to be less effective in their disintegration than might have been expected. Nevertheless they readily disintegrate into small pellets, after being alternately moistened with rain and again dried. Those which have flowed during rain down a slope, disintegrate in the same manner. Such pellets often roll a little down any sloping surface; their descent being sometimes much aided by the wind. The whole bottom of a broad dry ditch in my grounds, where there were very few fresh castings, was completely covered with these pellets or disintegrated castings, which had rolled down the steep sides, inclined at an angle of 27 degrees.
Near Nice, in places where the great cylindrical castings, previously described, abound, the soil consists of very fine arenaceo-calcareous loam; and Dr. King informs me that these castings are extremely liable to crumble during dry weather into small fragments, which are soon acted on by rain, and then sink down so as to be no longer distinguishable from the surrounding soil. He sent me a mass of such disintegrated castings, collected on the top of a bank, where none could have rolled down from above. They must have been ejected within the previous five or six months, but they now consisted of more or less rounded fragments of all sizes, from 0.75 of an inch in diameter to minute grains and mere dust. Dr. King witnessed the crumbling process whilst drying some perfect castings, which he afterwards sent me. Mr. Scott also remarks on the crumbling of the castings near Calcutta and on the mountains of Sikkim during the hot and dry season.
When the castings near Nice had been ejected on an inclined surface, the disintegrated fragments rolled downwards, without losing their distinctive shape; and in some places could "be collected in basketfuls." Dr. King observed a striking instance of this fact on the Corniche road, where a drain, about 2.5 feet wide and 9 inches deep, had been made to catch the surface drainage from the adjoining hill-side. The bottom of this drain was covered for a distance of several hundred yards, to a depth of from 1.5 to 3 inches, by a layer of broken castings, still retaining their characteristic shape. Nearly all these innumerable fragments had rolled down from above, for extremely few castings had been ejected in the drain itself. The hill-side was steep, but varied much in inclination, which Dr. King estimated at from 30 degrees to 60 degrees with the horizon. He climbed up the slope, and "found every here and there little embankments, formed by fragments of the castings that had been arrested in their downward progress by irregularities of the surface, by stones, twigs, One little group of plants of Anemone hortensis had acted in this manner, and quite a small bank of soil had collected round it. Much of this soil had crumbled down, but a great deal of it still retained the form of castings." Dr. King dug up this plant, and was struck with the thickness of the soil which must have recently accumulated over the crown of the rhizoma, as shown by the length of the bleached petioles, in comparison with those of other plants of the same kind, where there had been no such accumulation. The earth thus accumulated had no doubt been secured (as I have everywhere seen) by the smaller roots of the plants. After describing this and other analogous cases, Dr. King concludes: "I can have no doubt that worms help greatly in the process of denudation."
Ledges of earth on steep hill-sides.—Little horizontal ledges, one above another, have been observed on steep grassy slopes in many parts of the world. The formation has been attributed to animals travelling repeatedly along the slope in the same horizontal lines while grazing, and that they do thus move and use the ledges is certain; but Professor Henslow (a most careful observer) told Sir J. Hooker that he was convinced that this was not the sole cause of their formation. Sir J. Hooker saw such ledges on the Himalayan and Atlas ranges, where there were no domesticated animals and not many wild ones; but these latter would, it is probable, use the ledges at night while grazing like our domesticated animals. A friend observed for me the ledges on the Alps of Switzerland, and states that they ran at 3 or 4 ft. one above the other, and were about a foot in breadth. They had been deeply pitted by the feet of grazing cows. Similar ledges were observed by the same friend on our Chalk downs, and on an old talus of chalk-fragments (thrown out of a quarry) which had become clothed with turf.
My son Francis examined a Chalk escarpment near Lewes; and here on a part which was very steep, sloping at 40 degrees with the horizon, about 30 flat ledges extended horizontally for more than 100 yards, at an average distance of about 20 inches, one beneath the other. They were from 9 to 10 inches in breadth. When viewed from a distance they presented a striking appearance, owing to their parallelism; but when examined closely, they were seen to be somewhat sinuous, and one often ran into another, giving the appearance of the ledge having forked into two. They are formed of light-coloured earth, which on the outside, where thickest, was in one case 9 inches, and in another case between 6 and 7 inches in thickness. Above the ledges, the thickness of the earth over the chalk was in the former case 4 and in the latter only 3 inches. The grass grew more vigorously on the outer edges of the ledges than on any other part of the slope, and here formed a tufted fringe. Their middle part was bare, but whether this had been caused by the trampling of sheep, which sometimes frequent the ledges, my son could not ascertain. Nor could he feel sure how much of the earth on the middle and bare parts, consisted of disintegrated worm-castings which had rolled down from above; but he felt convinced that some had thus originated; and it was manifest that the ledges with their grass-fringed edges would arrest any small object rolling down from above.
At one end or side of the bank bearing these ledges, the surface consisted in parts of bare chalk, and here the ledges were very irregular. At the other end of the bank, the slope suddenly became less steep, and here the ledges ceased rather abruptly; but little embankments only a foot or two in length were still present. The slope became steeper lower down the hill, and the regular ledges then reappeared. Another of my sons observed, on the inland side of Beachy Head, where the surface sloped at about 25 degrees, many short little embankments like those just mentioned. They extended horizontally and were from a few inches to two or three feet in length. They supported tufts of grass growing vigorously. The average thickness of the mould of which they were formed, taken from nine measurements, was 4.5 inches; while that of the mould above and beneath them was on an average only 3.2 inches, and on each side, on the same level, 3.1 inches. On the upper parts of the slope, these embankments showed no signs of having been trampled on by sheep, but in the lower parts such signs were fairly plain. No long continuous ledges had here been formed.
If the little embankments above the Corniche road, which Dr. King saw in the act of formation by the accumulation of disintegrated and rolled worm-castings, were to become confluent along horizontal lines, ledges would be formed. Each embankment would tend to extend laterally by the lateral extension of the arrested castings; and animals grazing on a steep slope would almost certainly make use of every prominence at nearly the same level, and would indent the turf between them; and such intermediate indentations would again arrest the castings. An irregular ledge when once formed would also tend to become more regular and horizontal by some of the castings rolling laterally from the higher to the lower parts, which would thus be raised. Any projection beneath a ledge would not afterwards receive disintegrated matter from above, and would tend to be obliterated by rain and other atmospheric agencies. There is some analogy between the formation, as here supposed, of these ledges, and that of the ripples of wind-drifted sand as described by Lyell. 
The steep, grass-covered sides of a mountainous valley in Westmoreland, called Grisedale, was marked in many places with innumerable lines of miniature cliffs, with almost horizontal, little ledges at their bases. Their formation was in no way connected with the action of worms, for castings could not anywhere be seen (and their absence is an inexplicable fact), although the turf lay in many places over a considerable thickness of boulder- clay and moraine rubbish. Nor, as far as I could judge, was the formation of these little cliffs at all closely connected with the trampling of cows or sheep. It appeared as if the whole superficial, somewhat argillaceous earth, while partially held together by the roots of the grasses, had slided a little way down the mountain sides; and in thus sliding, had yielded and cracked in horizontal lines, transversely to the slope.
Castings blown to leeward by the wind.—We have seen that moist castings flow, and that disintegrated castings roll down any inclined surface; and we shall now see that castings, recently ejected on level grass-covered surfaces, are blown during gales of wind accompanied by rain to leeward. This has been observed by me many times on many fields during several successive years. After such gales, the castings present a gently inclined and smooth, or sometimes furrowed, surface to windward, while they are steeply inclined or precipitous to leeward, so that they resemble on a miniature scale glacier-ground hillocks of rock. They are often cavernous on the leeward side, from the upper part having curled over the lower part. During one unusually heavy south-west gale with torrents of rain, many castings were wholly blown to leeward, so that the mouths of the burrows were left naked and exposed on the windward side. Recent castings naturally flow down an inclined surface, but on a grassy field, which sloped between 10 degrees and 15 degrees, several were found after a heavy gale blown up the slope. This likewise occurred on another occasion on a part of my lawn where the slope was somewhat less. On a third occasion, the castings on the steep, grass-covered sides of a valley, down which a gale had blown, were directed obliquely instead of straight down the slope; and this was obviously due to the combined action of the wind and gravity. Four castings on my lawn, where the downward inclination was 0 degrees 45 seconds, 1 degree, 3 degrees and 3 degrees 30 seconds (mean 2 degrees 45 seconds) towards the north- east, after a heavy south-west gale with rain, were divided across the mouths of the burrows and weighed in the manner formerly described. The mean weight of the earth below the mouths of burrows and to leeward, was to that above the mouths and on the windward side as 2.75 to 1; whereas we have seen that with several castings which had flowed down slopes having a mean inclination of 9 degrees 26 seconds, and with three castings where the inclination was above 12 degrees; the proportional weight of the earth below to that above the burrows was as only 2 to 1. These several cases show how efficiently gales of wind accompanied by rain act in displacing recently ejected castings. We may therefore conclude that even a moderately strong wind will produce some slight effect on them.
Dry and indurated castings, after their disintegration into small fragments or pellets, are sometimes, probably often, blown by a strong wind to leeward. This was observed on four occasions, but I did not sufficiently attend to this point. One old casting on a gently sloping bank was blown quite away by a strong south-west wind. Dr. King believes that the wind removes the greater part of the old crumbling castings near Nice. Several old castings on my lawn were marked with pins and protected from any disturbance. They were examined after an interval of 10 weeks, during which time the weather had been alternately dry and rainy. Some, which were of a yellowish colour had been washed almost completely away, as could be seen by the colour of the surrounding ground. Others had completely disappeared, and these no doubt had been blown away. Lastly, others still remained and would long remain, as blades of grass had grown through them. On poor pasture-land, which has never been rolled and has not been much trampled on by animals, the whole surface is sometimes dotted with little pimples, through and on which grass grows; and these pimples consist of old worm- castings.
In all the many observed cases of soft castings blown to leeward, this had been effected by strong winds accompanied by rain. As such winds in England generally blow from the south and south-west, earth must on the whole tend to travel over our fields in a north and north-east direction. This fact is interesting, because it might be thought that none could be removed from a level, grass- covered surface by any means. In thick and level woods, protected from the wind, castings will never be removed as long as the wood lasts; and mould will here tend to accumulate to the depth at which worms can work. I tried to procure evidence as to how much mould is blown, whilst in the state of castings, by our wet southern gales to the north-east, over open and flat land, by looking to the level of the surface on opposite sides of old trees and hedge-rows; but I failed owing to the unequal growth of the roots of trees and to most pasture-land having been formerly ploughed.
On an open plain near Stonehenge, there exist shallow circular trenches, with a low embankment outside, surrounding level spaces 50 yards in diameter. These rings appear very ancient, and are believed to be contemporaneous with the Druidical stones. Castings ejected within these circular spaces, if blown to the north-east by south-west winds would form a layer of mould within the trench, thicker on the north-eastern than on any other side. But the site was not favourable for the action of worms, for the mould over the surrounding Chalk formation with flints, was only 3.37 inches in thickness, from a mean of six observations made at a distance of 10 yards outside the embankment. The thickness of the mould within two of the circular trenches was measured every 5 yards all round, on the inner sides near the bottom. My son Horace protracted these measurements on paper; and though the curved line representing the thickness of the mould was extremely irregular, yet in both diagrams it could be seen to be thicker on the north-eastern side than elsewhere. When a mean of all the measurements in both the trenches was laid down and the line smoothed, it was obvious that the mould was thickest in the quarter of the circle between north- west and north-east; and thinnest in the quarter between south-east and south-west, especially at this latter point. Besides the foregoing measurements, six others were taken near together in one of the circular trenches, on the north-east side; and the mould here had a mean thickness of 2.29 inches; while the mean of six other measurements on the south-west side was only 1.46 inches. These observations indicate that the castings had been blown by the south-west winds from the circular enclosed space into the trench on the north-east side; but many more measurements in other analogous cases would be requisite for a trustworthy result.
The amount of fine earth brought to the surface under the form of castings, and afterwards transported by the winds accompanied by rain, or that which flows and rolls down an inclined surface, no doubt is small in the course of a few scores of years; for otherwise all the inequalities in our pasture fields would be smoothed within a much shorter period than appears to be the case. But the amount which is thus transported in the course of thousands of years cannot fail to be considerable and deserves attention. E. de Beaumont looks at the vegetable mould which everywhere covers the land as a fixed line, from which the amount of denudation may be measured.  He ignores the continued formation of fresh mould by the disintegration of the underlying rocks and fragments of rock; and it is curious to find how much more philosophical were the views maintained long ago, by Playfair, who, in 1802, wrote, "In the permanence of a coat of vegetable mould on the surface of the earth, we have a demonstrative proof of the continued destruction of the rocks." 
Ancient encampments and tumuli.—E. de Beaumont adduces the present state of many ancient encampments and tumuli and of old ploughed fields, as evidence that the surface of the land undergoes hardly any degradation. But it does not appear that he ever examined the thickness of the mould over different parts of such old remains. He relies chiefly on indirect, but apparently trustworthy, evidence that the slopes of the old embankments are the same as they originally were; and it is obvious that he could know nothing about their original heights. In Knole Park a mound had been thrown up behind the rifle-targets, which appeared to have been formed of earth originally supported by square blocks of turf. The sides sloped, as nearly as I could estimate them, at an angle of 45 degrees or 50 degrees with the horizon, and they were covered, especially on the northern side, with long coarse grass, beneath which many worm-castings were found. These had flowed bodily downwards, and others had rolled down as pellets. Hence it is certain that as long as a mound of this kind is tenanted by worms, its height will be continually lowered. The fine earth which flows or rolls down the sides of such a mound accumulates at its base in the form of a talus. A bed, even a very thin bed, of fine earth is eminently favourable for worms; so that a greater number of castings would tend to be ejected on a talus thus formed than elsewhere; and these would be partially washed away by every heavy shower and be spread over the adjoining level ground. The final result would be the lowering of the whole mound, whilst the inclination of the sides would not be greatly lessened. The same result would assuredly follow with ancient embankments and tumuli; except where they had been formed of gravel or of nearly pure sand, as such matter is unfavourable for worms. Many old fortifications and tumuli are believed to be at least 2000 years old; and we should bear in mind that in many places about one inch of mould is brought to the surface in 5 years or two inches in 10 years. Therefore in so long a period as 2000 years, a large amount of earth will have been repeatedly brought to the surface on most old embankments and tumuli, especially on the talus round their bases, and much of this earth will have been washed completely away. We may therefore conclude that all ancient mounds, when not formed of materials unfavourable to worms, will have been somewhat lowered in the course of centuries, although their inclinations may not have been greatly changed.
Fields formerly ploughed.—From a very remote period and in many countries, land has been ploughed, so that convex beds, called crowns or ridges, usually about 8 feet across and separated by furrows, have been thrown up. The furrows are directed so as to carry off the surface water. In my attempts to ascertain how long a time these crowns and furrows last, when ploughed land has been converted into pasture, obstacles of many kinds were encountered. It is rarely known when a field was last ploughed; and some fields which were thought to have been in pasture from time immemorial were afterwards discovered to have been ploughed only 50 or 60 years before. During the early part of the present century, when the price of corn was very high, land of all kinds seems to have been ploughed in Britain. There is, however, no reason to doubt that in many cases the old crowns and furrows have been preserved from a very ancient period.  That they should have been preserved for very unequal lengths of time would naturally follow from the crowns, when first thrown up, having differed much in height in different districts, as is now the case with recently ploughed land.
In old pasture fields, the mould, wherever measurements were made, was found to be from 0.5 to 2 inches thicker in the furrows than on the crowns; but this would naturally follow from the finer earth having been washed from the crowns into the furrows before the land was well clothed with turf; and it is impossible to tell what part worms may have played in the work. Nevertheless from what we have seen, castings would certainly tend to flow and to be washed during heavy rain from the crowns into the furrows. But as soon as a bed of fine earth had by any means been accumulated in the furrows, it would be more favourable for worms than the other parts, and a greater number of castings would be thrown up here than elsewhere; and as the furrows on sloping land are usually directed so as to carry off the surface water, some of the finest earth would be washed from the castings which had been here ejected and be carried completely away. The result would be that the furrows would be filled up very slowly, while the crowns would be lowered perhaps still more slowly by the flowing and rolling of the castings down their gentle inclinations into the furrows.
Nevertheless it might be expected that old furrows, especially those on a sloping surface, would in the course of time be filled up and disappear. Some careful observers, however, who examined fields for me in Gloucestershire and Staffordshire could not detect any difference in the state of the furrows in the upper and lower parts of sloping fields, supposed to have been long in pasture; and they came to the conclusion that the crowns and furrows would last for an almost endless number of centuries. On the other hand the process of obliteration seems to have commenced in some places. Thus in a grass field in North Wales, known to have been ploughed about 65 years ago, which sloped at an angle of 15 degrees to the north-east, the depth of the furrows (only 7 feet apart) was carefully measured, and was found to be about 4.5 inches in the upper part of the slope, and only 1 inch near the base, where they could be traced with difficulty. On another field sloping at about the same angle to the south-west, the furrows were scarcely perceptible in the lower part; although these same furrows when followed on to some adjoining level ground were from 2.5 to 3.5 inches in depth. A third and closely similar case was observed. In a fourth case, the mould in a furrow in the upper part of a sloping field was 2.5 inches, and in the lower part 4.5 inches in thickness.
On the Chalk Downs at about a mile distance from Stonehenge, my son William examined a grass-covered, furrowed surface, sloping at from 8 degrees to 10 degrees, which an old shepherd said had not been ploughed within the memory of man. The depth of one furrow was measured at 16 points in a length of 68 paces, and was found to be deeper where the slope was greatest and where less earth would naturally tend to accumulate, and at the base it almost disappeared. The thickness of the mould in this furrow in the upper part was 2.5 inches, which increased to 5 inches, a little above the steepest part of the slope; and at the base, in the middle of the narrow valley, at a point which the furrow if continued would have struck, it amounted to 7 inches. On the opposite side of the valley, there were very faint, almost obliterated, traces of furrows. Another analogous but not so decided a case was observed at a few miles' distance from Stonehenge. On the whole it appears that the crowns and furrows on land formerly ploughed, but now covered with grass, tend slowly to disappear when the surface is inclined; and this is probably in large part due to the action of worms; but that the crowns and furrows last for a very long time when the surface is nearly level.
Formation and amount of mould over the Chalk Formation.—Worm- castings are often ejected in extraordinary numbers on steep, grass-covered slopes, where the Chalk comes close to the surface, as my son William observed near Winchester and elsewhere. If such castings are largely washed away during heavy rains, it is difficult to understand at first how any mould can still remain on our Downs, as there does not appear any evident means for supplying the loss. There is, moreover, another cause of loss, namely, in the percolation of the finer particles of earth into the fissures in the chalk and into the chalk itself. These considerations led me to doubt for a time whether I had not exaggerated the amount of fine earth which flows or rolls down grass-covered slopes under the form of castings; and I sought for additional information. In some places, the castings on Chalk Downs consist largely of calcareous matter, and here the supply is of course unlimited. But in other places, for instance on a part of Teg Down near Winchester, the castings were all black and did not effervesce with acids. The mould over the chalk was here only from 3 to 4 inches in thickness. So again on the plain near Stonehenge, the mould, apparently free from calcareous matter, averaged rather less than 3.5 inches in thickness. Why worms should penetrate and bring up chalk in some places and not in others I do not know.
In many districts where the land is nearly level, a bed several feet in thickness of red clay full of unworn flints overlies the Upper Chalk. This overlying matter, the surface of which has been converted into mould, consists of the undissolved residue from the chalk. It may be well here to recall the case of the fragments of chalk buried beneath worm-castings on one of my fields, the angles of which were so completely rounded in the course of 29 years that the fragments now resembled water-worn pebbles. This must have been effected by the carbonic acid in the rain and in the ground, by the humus-acids, and by the corroding power of living roots. Why a thick mass of residue has not been left on the Chalk, wherever the land is nearly level, may perhaps be accounted for by the percolation of the fine particles into the fissures, which are often present in the chalk and are either open or are filled up with impure chalk, or into the solid chalk itself. That such percolation occurs can hardly be doubted. My son collected some powdered and fragmentary chalk beneath the turf near Winchester; the former was found by Colonel Parsons, R. E., to contain 10 per cent., and the fragments 8 per cent. of earthy matter. On the flanks of the escarpment near Abinger in Surrey, some chalk close beneath a layer of flints, 2 inches in thickness and covered by 8 inches of mould, yielded a residue of 3.7 per cent. of earthy matter. On the other hand the Upper Chalk properly contains, as I was informed by the late David Forbes who had made many analyses, only from 1 to 2 per cent. of earthy matter; and two samples from pits near my house contained 1.3 and 0.6 per cent. I mention these latter cases because, from the thickness of the overlying bed of red clay with flints, I had imagined that the underlying chalk might here be less pure than elsewhere. The cause of the residue accumulating more in some places than in others, may be attributed to a layer of argillaceous matter having been left at an early period on the chalk, and this would check the subsequent percolation of earthy matter into it.
From the facts now given we may conclude that castings ejected on our Chalk Downs suffer some loss by the percolation of their finer matter into the chalk. But such impure superficial chalk, when dissolved, would leave a larger supply of earthy matter to be added to the mould than in the case of pure chalk. Besides the loss caused by percolation, some fine earth is certainly washed down the sloping grass-covered surfaces of our Downs. The washing-down process, however, will be checked in the course of time; for although I do not know how thin a layer of mould suffices to support worms, yet a limit must at last be reached; and then their castings would cease to be ejected or would become scanty.
The following cases show that a considerable quantity of fine earth is washed down. The thickness of the mould was measured at points 12 yards apart across a small valley in the Chalk near Winchester. The sides sloped gently at first; then became inclined at about 20 degrees; then more gently to near the bottom, which transversely was almost level and about 50 yards across. In the bottom, the mean thickness of the mould from five measurements was 8.3 inches; whilst on the sides of the valley, where the inclination varied between 14 degrees and 20 degrees, its mean thickness was rather less than 3.5 inches. As the turf-covered bottom of the valley sloped at an angle of only between 2 degrees and 3 degrees, it is probable that most of the 8.3-inch layer of mould had been washed down from the flanks of the valley, and not from the upper part. But as a shepherd said that he had seen water flowing in this valley after the sudden thawing of snow, it is possible that some earth may have been brought down from the upper part; or, on the other hand, that some may have been carried further down the valley. Closely similar results, with respect to the thickness of the mould, were obtained in a neighbouring valley.
St. Catherine's Hill, near Winchester, is 327 feet in height, and consists of a steep cone of chalk about 0.25 of a mile in diameter. The upper part was converted by the Romans, or, as some think, by the ancient Britons, into an encampment, by the excavation of a deep and broad ditch all round it. Most of the chalk removed during the work was thrown upwards, by which a projecting bank was formed; and this effectually prevents worm-castings (which are numerous in parts), stones, and other objects from being washed or rolled into the ditch. The mould on the upper and fortified part of the hill was found to be in most places only from 2.5 to 3.5 inches in thickness; whereas it had accumulated at the foot of the embankment above the ditch to a thickness in most places of from 8 to 9.5 inches. On the embankment itself the mould was only 1 to 1.5 inch in thickness; and within the ditch at the bottom it varied from 2.5 to 3.5, but was in one spot 6 inches in thickness. On the north-west side of the hill, either no embankment had ever been thrown up above the ditch, or it had subsequently been removed; so that here there was nothing to prevent worm-castings, earth and stones being washed into the ditch, at the bottom of which the mould formed a layer from 11 to 22 inches in thickness. It should however be stated that here and on other parts of the slope, the bed of mould often contained fragments of chalk and flint which had obviously rolled down at different times from above. The interstices in the underlying fragmentary chalk were also filled up with mould.
My son examined the surface of this hill to its base in a south- west direction. Beneath the great ditch, where the slope was about 24 degrees, the mould was very thin, namely, from 1.5 to 2.5 inches; whilst near the base, where the slope was only 3 degrees to 4 degrees, it increased to between 8 and 9 inches in thickness. We may therefore conclude that on this artificially modified hill, as well as in the natural valleys of the neighbouring Chalk Downs, some fine earth, probably derived in large part from worm-castings, is washed down, and accumulates in the lower parts, notwithstanding the percolation of an unknown quantity into the underlying chalk; a supply of fresh earthy matter being afforded by the dissolution of the chalk through atmospheric and other agencies.