THE AMOUNT OF FINE EARTH BROUGHT UP BY WORMS TO THE SURFACE.
Rate at which various objects strewed on the surface of grass-fields are covered up by the castings of worms--The burial of a paved path--The slow subsidence of great stones left on the surface--The number of worms which live within a given space--The weight of earth ejected from a burrow, and from all the burrows within a given space--The thickness of the layer of mould which the castings on a given space would form within a given time if uniformly spread out--The slow rate at which mould can increase to a great thickness--Conclusion.
We now come to the more immediate subject of this volume, namely, the amount of earth which is brought up by worms from beneath the surface, and is afterwards spread out more or less completely by the rain and wind. The amount can be judged of by two methods,--by the rate at which objects left on the surface are buried, and more accurately by weighing the quantity brought up within a given time. We will begin with the first method, as it was first followed.
Near Mael Hall in Staffordshire, quick-lime had been spread about the year 1827 thickly over a field of good pasture-land, which had not since been ploughed. Some square holes were dug in this field in the beginning of October 1837; and the sections showed a layer of turf, formed by the matted roots of the grasses, 0.5 inch in thickness, beneath which, at a depth of 2.5 inches (or 3 inches from the surface), a layer of the lime in powder or in small lumps could be distinctly seen running all round the vertical sides of the holes. The soil beneath the layer of lime was either gravelly or of a coarse sandy nature, and differed considerably in appearance from the overlying dark-coloured fine mould. Coal- cinders had been spread over a part of this same field either in the year 1833 or 1834; and when the above holes were dug, that is after an interval of 3 or 4 years, the cinders formed a line of black spots round the holes, at a depth of 1 inch beneath the surface, parallel to and above the white layer of lime. Over another part of this field cinders had been strewed, only about half-a-year before, and these either still lay on the surface or were entangled among the roots of the grasses; and I here saw the commencement of the burying process, for worm-castings had been heaped on several of the smaller fragments. After an interval of 4.75 years this field was re-examined, and now the two layers of lime and cinders were found almost everywhere at a greater depth than before by nearly 1 inch, we will say by 0.75 of an inch. Therefore mould to an average thickness of 0.22 of an inch had been annually brought up by the worms, and had been spread over the surface of this field.
Coal-cinders had been strewed over another field, at a date which could not be positively ascertained, so thickly that they formed (October, 1837) a layer, 1 inch in thickness at a depth of about 3 inches from the surface. The layer was so continuous that the over-lying dark vegetable mould was connected with the sub-soil of red clay only by the roots of the grasses; and when these were broken, the mould and the red clay fell apart. In a third field, on which coal-cinders and burnt marl had been strewed several times at unknown dates, holes were dug in 1842; and a layer of cinders could be traced at a depth of 3.5 inches, beneath which at a depth of 9.5 inches from the surface there was a line of cinders together with burnt marl. On the sides of one hole there were two layers of cinders, at 2 and 3.5 inches beneath the surface; and below them at a depth in parts of 9.5, and in other parts of 10.5 inches there were fragments of burnt marl. In a fourth field two layers of lime, one above the other, could be distinctly traced, and beneath them a layer of cinders and burnt marl at a depth of from 10 to 12 inches below the surface.
A piece of waste, swampy land was enclosed, drained, ploughed, harrowed and thickly covered in the year 1822 with burnt marl and cinders. It was sowed with grass seeds, and now supports a tolerably good but coarse pasture. Holes were dug in this field in 1837, or 15 years after its reclamation, and we see in the accompanying diagram (Fig. 5), reduced to half of the natural scale, that the turf was 1 inch thick, beneath which there was a layer of vegetable mould 2.5 inches thick. This layer did not contain fragments of any kind; but beneath it there was a layer of mould, 1.5 inch in thickness, full of fragments of burnt marl, conspicuous from their red colour, one of which near the bottom was an inch in length; and other fragments of coal-cinders together with a few white quartz pebbles. Beneath this layer and at a depth of 4.5 inches from the surface, the original black, peaty, sandy soil with a few quartz pebbles was encountered. Here therefore the fragments of burnt marl and cinders had been covered in the course of 15 years by a layer of fine vegetable mould, only 2.5 inches in thickness, excluding the turf. Six and a half years subsequently this field was re-examined, and the fragments were now found at from 4 to 5 inches beneath the surface. So that in this interval of 6.5 years, about 1.5 inch of mould had been added to the superficial layer. I am surprised that a greater quantity had not been brought up during the whole 21.5 years, for in the closely underlying black, peaty soil there were many worms. It is, however, probable that formerly, whilst the land remained poor, worms were scanty; and the mould would then have accumulated slowly. The average annual increase of thickness for the whole period is 0.19 of an inch.
Two other cases are worth recording. In the spring of 1835, a field, which had long existed as poor pasture and was so swampy that it trembled slightly when stamped on, was thickly covered with red sand so that the whole surface appeared at first bright red. When holes were dug in this field after an interval of about 2.5 years, the sand formed a layer at a depth of 0.75 in. beneath the surface. In 1842 (i.e., 7 years after the sand had been laid on) fresh holes were dug, and now the red sand formed a distinct layer, 2 inches beneath the surface, or 1.5 inch beneath the turf; so that on an average, 0.21 inch of mould had been annually brought to the surface. Immediately beneath the layer of red sand, the original substratum of black sandy peat extended.
A grass field, likewise not far from Maer Hall, had formerly been thickly covered with marl, and was then left for several years as pasture; it was afterwards ploughed. A friend had three trenches dug in this field 28 years after the application of the marl,  and a layer of the marl fragments could be traced at a depth, carefully measured, of 12 inches in some parts, and of 14 inches in other parts. This difference in depth depended on the layer being horizontal, whilst the surface consisted of ridges and furrows from the field having been ploughed. The tenant assured me that it had never been turned up to a greater depth than from 6 to 8 inches; and as the fragments formed an unbroken horizontal layer from 12 to 14 inches beneath the surface, these must have been buried by the worms whilst the land was in pasture before it was ploughed, for otherwise they would have been indiscriminately scattered by the plough throughout the whole thickness of the soil. Four-and-a-half years afterwards I had three holes dug in this field, in which potatoes had been lately planted, and the layer of marl-fragments was now found 13 inches beneath the bottoms of the furrows, and therefore probably 15 inches beneath the general level of the field. It should, however, be observed that the thickness of the blackish sandy soil, which had been thrown up by the worms above the marl-fragments in the course of 32.5 years, would have measured less than 15 inches, if the field had always remained as pasture, for the soil would in this case have been much more compact. The fragments of marl almost rested on an undisturbed substratum of white sand with quartz pebbles; and as this would be little attractive to worms, the mould would hereafter be very slowly increased by their action.
We will now give some cases of the action of worms, on land differing widely from the dry sandy or the swampy pastures just described. The chalk formation extends all round my house in Kent; and its surface, from having been exposed during an immense period to the dissolving action of rain-water, is extremely irregular, being abruptly festooned and penetrated by many deep well-like cavities.  During the dissolution of the chalk, the insoluble matter, including a vast number of unrolled flints of all sizes, has been left on the surface and forms a bed of stiff red clay, full of flints, and generally from 6 to 14 feet in thickness. Over the red clay, wherever the land has long remained as pasture, there is a layer a few inches in thickness, of dark-coloured vegetable mould.
A quantity of broken chalk was spread, on December 20, 1842, over a part of a field near my house, which had existed as pasture certainly for 30, probably for twice or thrice as many years. The chalk was laid on the land for the sake of observing at some future period to what depth it would become buried. At the end of November, 1871, that is after an interval of 29 years, a trench was dug across this part of the field; and a line of white nodules could be traced on both sides of the trench, at a depth of 7 inches from the surface. The mould, therefore, (excluding the turf) had here been thrown up at an average rate of 0.22 inch per year. Beneath the line of chalk nodules there was in parts hardly any fine earth free of flints, while in other parts there was a layer, 2.25 inches in thickness. In this latter case the mould was altogether 9.25 inches thick; and in one such spot a nodule of chalk and a smooth flint pebble, both of which must have been left at some former time on the surface, were found at this depth. At from 11 to 12 inches beneath the surface, the undisturbed reddish clay, full of flints, extended. The appearance of the above nodules of chalk surprised me, much at first, as they closely resembled water-worn pebbles, whereas the freshly-broken fragments had been angular. But on examining the nodules with a lens, they no longer appeared water-worn, for their surfaces were pitted through unequal corrosion, and minute, sharp points, formed of broken fossil shells, projected from them. It was evident that the corners of the original fragments of chalk had been wholly dissolved, from presenting a large surface to the carbonic acid dissolved in the rain-water and to that generated in soil containing vegetable matter, as well as to the humus-acids.  The projecting corners would also, relatively to the other parts, have been embraced by a larger number of living rootlets; and these have the power of even attacking marble, as Sachs has shown. Thus, in the course of 29 years, buried angular fragments of chalk had been converted into well-rounded nodules.
Another part of this same field was mossy, and as it was thought that sifted coal-cinders would improve the pasture, a thick layer was spread over this part either in 1842 or 1843, and another layer some years afterwards. In 1871 a trench was here dug, and many cinders lay in a line at a depth of 7 inches beneath the surface, with another line at a depth of 5.5 inches parallel to the one beneath. In another part of this field, which had formerly existed as a separate one, and which it was believed had been pasture-land for more than a century, trenches were dug to see how thick the vegetable mould was. By chance the first trench was made at a spot where at some former period, certainly more than forty years before, a large hole had been filled up with coarse red clay, flints, fragments of chalk, and gravel; and here the fine vegetable mould was only from 4.125 to 4.375 inches in thickness. In another and undisturbed place, the mould varied much in thickness, namely, from 6.5 to 8.5 inches; beneath which a few small fragments of brick were found in one place. From these several cases, it would appear that during the last 29 years mould has been heaped on the surface at an average annual rate of from 0.2 to 0.22 of an inch. But in this district when a ploughed field is first laid down in grass, the mould accumulates at a much slower rate. The rate, also, must become very much slower after a bed of mould, several inches in thickness, has been formed; for the worms then live chiefly near the surface, and burrow down to a greater depth so as to bring up fresh earth from below, only during the winter when the weather is very cold (at which time worms were found in this field at a depth of 26 inches) and during summer, when the weather is very dry.
A field, which adjoins the one just described, slopes in one part rather steeply (viz., at from 10 degrees to 15 degrees); this part was last ploughed in 1841, was then harrowed and left to become pasture-land. For several years it was clothed with an extremely scant vegetation, and was so thickly covered with small and large flints (some of them half as large as a child's head) that the field was always called by my sons "the stony field." When they ran down the slope the stones clattered together, I remember doubting whether I should live to see these larger flints covered with vegetable mould and turf. But the smaller stones disappeared before many years had elapsed, as did every one of the larger ones after a time; so that after thirty years (1871) a horse could gallop over the compact turf from one end of the field to the other, and not strike a single stone with his shoes. To anyone who remembered the appearance of the field in 1842, the transformation was wonderful. This was certainly the work of the worms, for though castings were not frequent for several years, yet some were thrown up month after month, and these gradually increased in numbers as the pasture improved. In the year 1871 a trench was dug on the above slope, and the blades of grass were cut off close to the roots, so that the thickness of the turf and of the vegetable mould could be measured accurately. The turf was rather less than half an inch, and the mould, which did not contain any stones, 2.5 inches in thickness. Beneath this lay coarse clayey earth full of flints, like that in any of the neighbouring ploughed fields. This coarse earth easily fell apart from the overlying mould when a spit was lifted up. The average rate of accumulation of the mould during the whole thirty years was only .083 inch per year (i.e., nearly one inch in twelve years); but the rate must have been much slower at first, and afterwards considerably quicker.
The transformation in the appearance of this field, which had been effected beneath my eyes, was afterwards rendered the more striking, when I examined in Knole Park a dense forest of lofty beech-trees, beneath which nothing grew. Here the ground was thickly strewed with large naked stones, and worm-castings were almost wholly absent. Obscure lines and irregularities on the surface indicated that the land had been cultivated some centuries ago. It is probable that a thick wood of young beech-trees sprung up so quickly, that time enough was not allowed for worms to cover up the stones with their castings, before the site became unfitted for their existence. Anyhow the contrast between the state of the now miscalled "stony field," well stocked with worms, and the present state of the ground beneath the old beech-trees in Knole Park, where worms appeared to be absent, was striking.
A narrow path running across part of my lawn was paved in 1843 with small flagstones, set edgeways; but worms threw up many castings and weeds grew thickly between them. During several years the path was weeded and swept; but ultimately the weeds and worms prevailed, and the gardener ceased to sweep, merely mowing off the weeds, as often as the lawn was mowed. The path soon became almost covered up, and after several years no trace of it was left. On removing, in 1877, the thin overlying layer of turf, the small flag-stones, all in their proper places, were found covered by an inch of fine mould.
Two recently published accounts of substances strewed on the surface of pasture-land, having become buried through the action of worms, may be here noticed. The Rev. H. C. Key had a ditch cut in a field, over which coal-ashes had been spread, as it was believed, eighteen years before; and on the clean-cut perpendicular sides of the ditch, at a depth of at least seven inches, there could be seen, for a length of 60 yards, "a distinct, very even, narrow line of coal-ashes, mixed with small coal, perfectly parallel with the top-sward."  This parallelism and the length of the section give interest to the case. Secondly, Mr. Dancer states  that crushed bones had been thickly strewed over a field; and "some years afterwards" these were found "several inches below the surface, at a uniform depth."
The Rev. Mr. Zincke informs me that he has lately had an orchard dug to the unusual depth of 4 feet. The upper 18 inches consisted of dark-coloured vegetable mould, and the next 18 inches of sandy loam, containing in the lower part many rolled pieces of sandstone, with some bits of brick and tile, probably of Roman origin, as remains of this period have been found close by. The sandy loam rested on an indurated ferruginous pan of yellow clay, on the surface of which two perfect celts were found. If, as seems probable, the celts were originally left on the surface of the land, they have since been covered up with earth 3 feet in thickness, all of which has probably passed through the bodies of worms, excepting the stones which may have been scattered on the surface at different times, together with manure or by other means. It is difficult otherwise to understand the source of the 18 inches of sandy loam, which differed from the overlying dark vegetable mould, after both had been burnt, only in being of a brighter red colour, and in not being quite so fine-grained. But on this view we must suppose that the carbon in vegetable mould, when it lies at some little depth beneath the surface and does not continually receive decaying vegetable matter from above, loses its dark colour in the course of centuries; but whether this is probable I do not know.
Worms appear to act in the same manner in New Zealand as in Europe; for Professor J. von Haast has described  a section near the coast, consisting of mica-schist, "covered by 5 or 6 feet of loess, above which about 12 inches of vegetable soil had accumulated." Between the loess and the mould there was a layer from 3 to 6 inches in thickness, consisting of "cores, implements, flakes, and chips, all manufactured from hard basaltic rock." It is therefore probable that the aborigines, at some former period, had left these objects on the surface, and that they had afterwards been slowly covered up by the castings of worms.
Farmers in England are well aware that objects of all kinds, left on the surface of pasture-land, after a time disappear, or, as they say, work themselves downwards. How powdered lime, cinders, and heavy stones, can work down, and at the same rate, through the matted roots of a grass-covered surface, is a question which has probably never occurred to them. 
The Sinking of great Stones through the Action of Worms.--When a stone of large size and of irregular shape is left on the surface of the ground, it rests, of course, on the more protuberant parts; but worms soon fill up with their castings all the hollow spaces on the lower side; for, as Hensen remarks, they like the shelter of stones. As soon as the hollows are filled up, the worms eject the earth which they have swallowed beyond the circumference of the stones; and thus the surface of the ground is raised all round the stone. As the burrows excavated directly beneath the stone after a time collapse, the stone sinks a little.  Hence it is, that boulders which at some ancient period have rolled down from a rocky mountain or cliff on to a meadow at its base, are always somewhat imbedded in the soil; and, when removed, leave an exact impression of their lower surfaces in the underlying fine mould. If, however, a boulder is of such huge dimensions, that the earth beneath is kept dry, such earth will not be inhabited by worms, and the boulder will not sink into the ground.
A lime-kiln formerly stood in a grass-field near Leith Hill Place in Surrey, and was pulled down 35 years before my visit; all the loose rubbish had been carted away, excepting three large stones of quartzose sandstone, which it was thought might hereafter be of some use. An old workman remembered that they had been left on a bare surface of broken bricks and mortar, close to the foundations of the kiln; but the whole surrounding surface is now covered with turf and mould. The two largest of these stones had never since been moved; nor could this easily have been done, as, when I had them removed, it was the work of two men with levers. One of these stones, and not the largest, was 64 inches long, 17 inches broad, and from 9 to 10 inches in thickness. Its lower surface was somewhat protuberant in the middle; and this part still rested on broken bricks and mortar, showing the truth of the old workman's account. Beneath the brick rubbish the natural sandy soil, full of fragments of sandstone was found; and this could have yielded very little, if at all, to the weight of the stone, as might have been expected if the sub-soil had been clay. The surface of the field, for a distance of about 9 inches round the stone, gradually sloped up to it, and close to the stone stood in most places about 4 inches above the surrounding ground. The base of the stone was buried from 1 to 2 inches beneath the general level, and the upper surface projected about 8 inches above this level, or about 4 inches above the sloping border of turf. After the removal of the stone it became evident that one of its pointed ends must at first have stood clear above the ground by some inches, but its upper surface was now on a level with the surrounding turf. When the stone was removed, an exact cast of its lower side, forming a shallow crateriform hollow, was left, the inner surface of which consisted of fine black mould, excepting where the more protuberant parts rested on the brick-rubbish. A transverse section of this stone, together with its bed, drawn from measurements made after it had been displaced, is here given on a scale of 0.5 inch to a foot (Fig. 6). The turf-covered border which sloped up to the stone, consisted of fine vegetable mould, in one part 7 inches in thickness. This evidently consisted of worm-castings, several of which had been recently ejected. The whole stone had sunk in the thirty-five years, as far as I could judge, about 1.5 inch; and this must have been due to the brick-rubbish beneath the more protuberant parts having been undermined by worms. At this rate the upper surface of the stone, if it had been left undisturbed, would have sunk to the general level of the field in 247 years; but before this could have occurred, some earth would have been washed down by heavy rain from the castings on the raised border of turf over the upper surface of the stone.
The second stone was larger that the one just described, viz., 67 inches in length, 39 in breadth, and 15 in thickness. The lower surface was nearly flat, so that the worms must soon have been compelled to eject their castings beyond its circumference. The stone as a whole had sunk about 2 inches into the ground. At this rate it would have required 262 years for its upper surface to have sunk to the general level of the field. The upwardly sloping, turf-covered border round the stone was broader than in the last case, viz., from 14 to 16 inches; and why this should be so, I could see no reason. In most parts this border was not so high as in the last case, viz., from 2 to 2.5 inches, but in one place it was as much as 5.5. Its average height close to the stone was probably about 3 inches, and it thinned out to nothing. If so, a layer of fine earth, 15 inches in breadth and 1.5 inch in average thickness, of sufficient length to surround the whole of the much elongated slab, must have been brought up by the worms in chief part from beneath the stone in the course of 35 years. This amount would be amply sufficient to account for its having sunk about 2 inches into the ground; more especially if we bear in mind that a good deal of the finest earth would have been washed by heavy rain from the castings ejected on the sloping border down to the level of the field. Some fresh castings were seen close to the stone. Nevertheless, on digging a large hole to a depth of 18 inches where the stone had lain, only two worms and a few burrows were seen, although the soil was damp and seemed favourable for worms. There were some large colonies of ants beneath the stone, and possibly since their establishment the worms had decreased in number.
The third stone was only about half as large as the others; and two strong boys could together have rolled it over. I have no doubt that it had been rolled over at a moderately recent time, for it now lay at some distance from the two other stones at the bottom of a little adjoining slope. It rested also on fine earth, instead of partly on brick-rubbish. In agreement with this conclusion, the raised surrounding border of turf was only 1 inch high in some parts, and 2 inches in other parts. There were no colonies of ants beneath this stone, and on digging a hole where it had lain, several burrows and worms were found.
At Stonehenge, some of the outer Druidical stones are now prostrate, having fallen at a remote but unknown period; and these have become buried to a moderate depth in the ground. They are surrounded by sloping borders of turf, on which recent castings were seen. Close to one of these fallen stones, which was 17 ft long, 6 ft. broad, and 28.5 inches thick, a hole was dug; and here the vegetable mould was at least 9.5 inches in thickness. At this depth a flint was found, and a little higher up on one side of the hole a fragment of glass. The base of the stone lay about 9.5 inches beneath the level of the surrounding ground, and its upper surface 19 inches above the ground.
A hole was also dug close to a second huge stone, which in falling had broken into two pieces; and this must have happened long ago, judging from the weathered aspect of the fractured ends. The base was buried to a depth of 10 inches, as was ascertained by driving an iron skewer horizontally into the ground beneath it. The vegetable mould forming the turf-covered sloping border round the stone, on which many castings had recently been ejected, was 10 inches in thickness; and most of this mould must have been brought up by worms from beneath its base. At a distance of 8 yards from the stone, the mould was only 5.5 inches in thickness (with a piece of tobacco pipe at a depth of 4 inches), and this rested on broken flint and chalk which could not have easily yielded to the pressure or weight of the stone.
A straight rod was fixed horizontally (by the aid of a spirit- level) across a third fallen stone, which was 7 feet 9 inches long; and the contour of the projecting parts and of the adjoining ground, which was not quite level, was thus ascertained, as shown in the accompanying diagram (Fig. 7) on a scale of 0.5 inch to a foot. The turf-covered border sloped up to the stone on one side to a height of 4 inches, and on the opposite side to only 2.5 inches above the general level. A hole was dug on the eastern side, and the base of the stone was here found to lie at a depth of 4 inches beneath the general level of the ground, and of 8 inches beneath the top of the sloping turf-covered border.
Sufficient evidence has now been given showing that small objects left on the surface of the land where worms abound soon get buried, and that large stones sink slowly downwards through the same means. Every step of the process could be followed, from the accidental deposition of a single casting on a small object lying loose on the surface, to its being entangled amidst the matted roots of the turf, and lastly to its being embedded in the mould at various depths beneath the surface. When the same field was re-examined after the interval of a few years, such objects were found at a greater depth than before. The straightness and regularity of the lines formed by the imbedded objects, and their parallelism with the surface of the land, are the most striking features of the case; for this parallelism shows how equably the worms must have worked; the result being, however, partly the effect of the washing down of the fresh castings by rain. The specific gravity of the objects does not affect their rate of sinking, as could be seen by porous cinders, burnt marl, chalk and quartz pebbles, having all sunk to the same depth within the same time. Considering the nature of the substratum, which at Leith Hill Place was sandy soil including many bits of rock, and at Stonehenge, chalk-rubble with broken flints; considering, also, the presence of the turf-covered sloping border of mould round the great fragments of stone at both these places, their sinking does not appear to have been sensibly aided by their weight, though this was considerable. 
On the number of worms which live within a given space.--We will now show, firstly, what a vast number of worms live unseen by us beneath our feet, and, secondly, the actual weight of the earth which they bring up to the surface within a given space and within a given time. Hensen, who has published so full and interesting an account of the habits of worms,  calculates, from the number which he found in a measured space, that there must exist 133,000 living worms in a hectare of land, or 53,767 in an acre. This latter number of worms would weigh 356 pounds, taking Hensen's standard of the weight of a single worm, namely, three grams. It should, however, be noted that this calculation is founded on the numbers found in a garden, and Hensen believes that worms are here twice as numerous as in corn-fields. The above result, astonishing though it be, seems to me credible, judging from the number of worms which I have sometimes seen, and from the number daily destroyed by birds without the species being exterminated. Some barrels of bad ale were left on Mr. Miller's land,  in the hope of making vinegar, but the vinegar proved bad, and the barrels were upset. It should be premised that acetic acid is so deadly a poison to worms that Perrier found that a glass rod dipped into this acid and then into a considerable body of water in which worms were immersed, invariably killed them quickly. On the morning after the barrels had been upset, "the heaps of worms which lay dead on the ground were so amazing, that if Mr. Miller had not seen them, he could not have thought it possible for such numbers to have existed in the space." As further evidence of the large number of worms which live in the ground, Hensen states that he found in a garden sixty-four open burrows in a space of 14.5 square feet, that is, nine in 2 square feet. But the burrows are sometimes much more numerous, for when digging in a grass-field near Maer Hall, I found a cake of dry earth, as large as my two open hands, which was penetrated by seven burrows, as large as goose-quills.
Weight of the earth ejected from a single burrow, and from all the burrows within a given space.--With respect to the weight of the earth daily ejected by worms, Hensen found that it amounted, in the case of some worms which he kept in confinement, and which he appears to have fed with leaves, to only 0.5 gram, or less than 8 grains per diem. But a very much larger amount must be ejected by worms in their natural state, at the periods when they consume earth as food instead of leaves, and when they are making deep burrows. This is rendered almost certain by the following weights of the castings thrown up at the mouths of single burrows; the whole of which appeared to have been ejected within no long time, as was certainly the case in several instances. The castings were dried (excepting in one specified instance) by exposure during many days to the sun or before a hot fire.
WEIGHT OF THE CASTINGS ACCUMULATED AT THE MOUTH OF A SINGLE BURROW.
(Weight in ounces given in parenthesis--DP.)
(1.) Down, Kent (sub-soil red clay, full of flints, over-lying the chalk). The largest casting which I could find on the flanks of a steep valley, the sub-soil being here shallow. In this one case, the casting was not well dried (3.98)
(2.) Down.--Largest casting which I could find (consisting chiefly of calcareous matter), on extremely poor pasture land at the bottom of the valley mentioned under (1.) (3.87)
(3.) Down.--A large casting, but not of unusual size, from a nearly level field, poor pasture, laid down in a grass about 35 years before (1.22)
(4.) Down. Average weight of 11 not large castings ejected on a sloping surface on my lawn, after they had suffered some loss of weight from being exposed during a considerable length of time to rain (0.7)
(5.) Near Nice in France.--Average weight of 12 castings of ordinary dimensions, collected by Dr. King on land which had not been mown for a long time and where worms abounded, viz., a lawn protected by shrubberies near the sea; soil sandy and calcareous; these castings had been exposed for some time to rain, before being collected, and must have lost some weight by disintegration, but they still retained their form (1.37)
(6.) The heaviest of the above twelve castings (1.76)
(7.) Lower Bengal.--Average weight of 22 castings, collected by Mr. J. Scott, and stated by him to have been thrown up in the course of one or two nights (1.24)
(8.) The heaviest of the above 22 castings (2.09)
(9.) Nilgiri Mountains, S. India; average weight of the 5 largest castings collected by Dr. King. They had been exposed to the rain of the last monsoon, and must have lost some weight (3.15)
(10.) The heaviest of the above 5 castings (4.34)
In this table we see that castings which had been ejected at the mouth of the same burrow, and which in most cases appeared fresh and always retained their vermiform configuration, generally exceeded an ounce in weight after being dried, and sometimes nearly equalled a quarter of a pound. On the Nilgiri mountains one casting even exceeded this latter weight. The largest castings in England were found on extremely poor pasture-land; and these, as far as I have seen, are generally larger than those on land producing a rich vegetation. It would appear that worms have to swallow a greater amount of earth on poor than on rich land, in order to obtain sufficient nutriment.
With respect to the tower-like castings near Nice (Nos. 5 and 6 in the above table), Dr. King often found five or six of them on a square foot of surface; and these, judging from their average weight, would have weighed together 7.5 ounces; so that the weight of those on a square yard would have been 4 lb. 3.5 oz. Dr. King collected, near the close of the year 1872, all the castings which still retained their vermiform shape, whether broken down or not, from a square foot, in a place abounding with worms, on the summit of a bank, where no castings could have rolled down from above. These castings must have been ejected, as he judged from their appearance in reference to the rainy and dry periods near Nice, within the previous five or six months; they weighed 9.5 oz., or 5 lb. 5.5 oz. per square yard. After an interval of four months, Dr. King collected all the castings subsequently ejected on the same square foot of surface, and they weighed 2.5 oz., or 1 lb. 6.5 oz. per square yard. Therefore within about ten months, or we will say for safety's sake within a year, 12 oz. of castings were thrown up on this one square foot, or 6.75 pounds on the square yard; and this would give 14.58 tons per acre.
In a field at the bottom of a valley in the chalk (see No. 2 in the foregoing table), a square yard was measured at a spot where very large castings abounded; they appeared, however, almost equally numerous in a few other places. These castings, which retained perfectly their vermiform shape, were collected; and they weighed when partially dried, 1 lb. 13.5 oz. This field had been rolled with a heavy agricultural roller fifty-two days before, and this would certainly have flattened every single casting on the land. The weather had been very dry for two or three weeks before the day of collection, so that not one casting appeared fresh or had been recently ejected. We may therefore assume that those which were weighed had been ejected within, we will say, forty days from the time when the field was rolled,--that is, twelve days short of the whole intervening period. I had examined the same part of the field shortly before it was rolled, and it then abounded with fresh castings. Worms do not work in dry weather during the summer, or in winter during severe frosts. If we assume that they work for only half the year--though this is too low an estimate--then the worms in this field would eject during the year, 8.387 pounds per square yard; or 18.12 tons per acre, assuming the whole surface to be equally productive in castings.
In the foregoing cases some of the necessary data had to be estimated, but in the two following cases the results are much more trustworthy. A lady, on whose accuracy I can implicitly rely, offered to collect during a year all the castings thrown up on two separate square yards, near Leith Hill Place, in Surrey. The amount collected was, however, somewhat less than that originally ejected by the worms; for, as I have repeatedly observed, a good deal of the finest earth is washed away, whenever castings are thrown up during or shortly before heavy rain. Small portions also adhered to the surrounding blades of grass, and it required too much time to detach every one of them.
On sandy soil, as in the present instance, castings are liable to crumble after dry weather, and particles were thus often lost. The lady also occasionally left home for a week or two, and at such times the castings must have suffered still greater loss from exposure to the weather. These losses were, however, compensated to some extent by the collections having been made on one of the squares for four days, and on the other square for two days more than the year.
A space was selected (October 9th, 1870) for one of the squares on a broad, grass-covered terrace, which had been mowed and swept during many years. It faced the south, but was shaded during part of the day by trees. It had been formed at least a century ago by a great accumulation of small and large fragments of sandstone, together with some sandy earth, rammed down level. It is probable that it was at first protected by being covered with turf. This terrace, judging from the number of castings on it, was rather unfavourable for the existence of worms, in comparison with the neighbouring fields and an upper terrace. It was indeed surprising that as many worms could live here as were seen; for on digging a hole in this terrace, the black vegetable mould together with the turf was only four inches in thickness, beneath which lay the level surface of light-coloured sandy soil, with many fragments of sandstone. Before any castings were collected all the previously existing ones were carefully removed. The last day's collection was on October 14th, 1871. The castings were then well dried before a fire; and they weighed exactly 3.5 lbs. This would give for an acre of similar land 7.56 tons of dry earth annually ejected by worms.
The second square was marked on unenclosed common land, at a height of about 700 ft. above the sea, at some little distance from Leith Hill Tower. The surface was clothed with short, fine turf, and had never been disturbed by the hand of man. The spot selected appeared neither particularly favourable nor the reverse for worms; but I have often noticed that castings are especially abundant on common land, and this may, perhaps, be attributed to the poorness of the soil. The vegetable mould was here between three and four inches in thickness. As this spot was at some distance from the house where the lady lived, the castings were not collected at such short intervals of time as those on the terrace; consequently the loss of fine earth during rainy weather must have been greater in this than in the last case. The castings moreover were more sandy, and in collecting them during dry weather they sometimes crumbled into dust, and much was thus lost. Therefore it is certain that the worms brought up to the surface considerably more earth than that which was collected. The last collection was made on October 27th, 1871; i.e., 367 days after the square had been marked out and the surface cleared of all pre-existing castings. The collected castings, after being well dried, weighed 7.453 pounds; and this would give, for an acre of the same kind of land, 16.1 tons of annually ejected dry earth.
SUMMARY OF THE FOUR FOREGOING CASES.
(1.) Castings ejected near Nice within about a year, collected by Dr. King on a square foot of surface, calculated to yield per acre 14.58 tons.
(2.) Castings ejected during about 40 days on a square yard, in a field of poor pasture at the bottom of a large valley in the Chalk, calculated to yield annually per acre 18.12 tons.
(3.) Castings collected from a square yard on an old terrace at Leith Hill Place, during 369 days, calculated to yield annually per acre 7.56 tons.
(4.) Castings collected from a square yard on Leith Hill Common during 367 days, calculated to yield annually per acre 16.1 tons.
The thickness of the layer of mould, which castings ejected during a year would form if uniformly spread out.--As we know, from the two last cases in the above summary, the weight of the dried castings ejected by worms during a year on a square yard of surface, I wished to learn how thick a layer of ordinary mould this amount would form if spread uniformly over a square yard. The dry castings were therefore broken into small particles, and whilst being placed in a measure were well shaken and pressed down. Those collected on the Terrace amounted to 124.77 cubic inches; and this amount, if spread out over a square yard, would make a layer 0.9627 inch in thickness. Those collected on the Common amounted to 197.56 cubic inches, and would make a similar layer 0.1524 inch in thickness,
These thicknesses must, however, be corrected, for the triturated castings, after being well shaken down and pressed, did not make nearly so compact a mass as vegetable mould, though each separate particle was very compact. Yet mould is far from being compact, as is shown by the number of air-bubbles which rise up when the surface is flooded with water. It is moreover penetrated by many fine roots. To ascertain approximately by how much ordinary vegetable mould would be increased in bulk by being broken up into small particles and then dried, a thin oblong block of somewhat argillaceous mould (with the turf pared off) was measured before being broken up, was well dried and again measured. The drying caused it to shrink by 1/7 of its original bulk, judging from exterior measurements alone. It was then triturated and partly reduced to powder, in the same manner as the castings had been treated, and its bulk now exceeded (notwithstanding shrinkage from drying) by 1/16 that of the original block of damp mould. Therefore the above calculated thickness of the layer, formed by the castings from the Terrace, after being damped and spread over a square yard, would have to be reduced by 1/16; and this will reduce the layer to 0.09 of an inch, so that a layer 0.9 inch in thickness would be formed in the course of ten years. On the same principle the castings from the Common would make in the course of a single year a layer 0.1429 inch, or in the course of 10 years 1.429 inch, in thickness. We may say in round numbers that the thickness in the former case would amount to nearly 1 inch, and in the second case to nearly 1.5 inch in 10 years.
In order to compare these results with those deduced from the rates at which small objects left on the surfaces of grass-fields become buried (as described in the early part of this chapter), we will give the following summary:-
SUMMARY OF THE THICKNESS OF THE MOULD ACCUMULATED OVER OBJECTS LEFT STREWED ON THE SURFACE, IN THE COURSE OF TEN YEARS.
The accumulation of mould during 14.75 years on the surface of a dry, sandy, grass-field near Maer Hall, amounted to 2.2 inches in 10 years.
The accumulation during 21.5 years on a swampy field near Maer Hall, amounted to nearly 1.9 inch in 10 years.
The accumulation during 7 years on a very swampy field near Maer Hall amounted to 2.1 inches in 10 years.
The accumulation during 29 years, on good, argillaceous pasture- land over the Chalk at Down, amounted to 2.2 inches in 10 years.
The accumulation during 30 years on the side of a valley over the Chalk at Down, the soil being argillaceous, very poor, and only just converted into pasture (so that it was for some years unfavourable for worms), amounted to 0.83 inch in 10 years.
In these cases (excepting the last) it may be seen that the amount of earth brought to the surface during 10 years is somewhat greater than that calculated from the castings which were actually weighed. This excess may be partly accounted for by the loss which the weighed castings had previously undergone through being washed by rain, by the adhesion of particles to the blades of the surrounding grass, and by their crumbling when dry. Nor must we overlook other agencies which in all ordinary cases add to the amount of mould, and which would not be included in the castings that were collected, namely, the fine earth brought up to the surface by burrowing larvae and insects, especially by ants. The earth brought up by moles generally has a somewhat different appearance from vegetable mould; but after a time would not be distinguishable from it. In dry countries, moreover, the wind plays an important part in carrying dust from one place to another, and even in England it must add to the mould on fields near great roads. But in our country these latter several agencies appear to be of quite subordinate importance in comparison with the action of worms.
We have no means of judging how great a weight of earth a single full-sized worm ejects during a year. Hensen estimates that 53,767 worms exist in an acre of land; but this is founded on the number found in gardens, and he believes that only about half as many live in corn-fields. How many live in old pasture land is unknown; but if we assume that half the above number, or 26,886 worms live on such land, then taking from the previous summary 15 tons as the weight of the castings annually thrown up on an acre of land, each worm must annually eject 20 ounces. A full-sized casting at the mouth of a single burrow often exceeds, as we have seen, an ounce in weight; and it is probable that worms eject more than 20 full- sized castings during a year. If they eject annually more than 20 ounces, we may infer that the worms which live in an acre of pasture land must be less than 26,886 in number.
Worms live chiefly in the superficial mould, which is usually from 4 or 5 to 10 and even 12 inches in thickness; and it is this mould which passes over and over again through their bodies and is brought to the surface. But worms occasionally burrow into the subsoil to a much greater depth, and on such occasions they bring up earth from this greater depth; and this process has gone on for countless ages. Therefore the superficial layer of mould would ultimately attain, though at a slower and slower rate, a thickness equal to the depth to which worms ever burrow, were there not other opposing agencies at work which carry away to a lower level some of the finest earth which is continually being brought to the surface by worms. How great a thickness vegetable mould ever attains, I have not had good opportunities for observing; but in the next chapter, when we consider the burial of ancient buildings, some facts will be given on this head. In the two last chapters we shall see that the soil is actually increased, though only to a small degree, through the agency of worms; but their chief work is to sift the finer from the coarser particles, to mingle the whole with vegetable debris, and to saturate it with their intestinal secretions.
Finally, no one who considers the facts given in this chapter--on the burying of small objects and on the sinking of great stones left on the surface--on the vast number of worms which live within a moderate extent of ground on the weight of the castings ejected from the mouth of the same burrow--on the weight of all the castings ejected within a known time on a measured space--will hereafter, as I believe, doubt that worms play an important part in nature.
 This case is given in a postscript to my paper in the 'Transact. Geolog. Soc.' (Vol. v. p. 505), and contains a serious error, as in the account received I mistook the figure 30 for 80. The tenant, moreover, formerly said that he had marled the field thirty years before, but was now positive that this was done in 1809, that is twenty-eight years before the first examination of the field by my friend. The error, as far as the figure 80 is concerned, was corrected in an article by me, in the 'Gardeners' Chronicle,' 1844, p. 218.
 These pits or pipes are still in process of formation. During the last forty years I have seen or heard of five cases, in which a circular space, several feet in diameter, suddenly fell in, leaving on the field an open hole with perpendicular sides, some feet in depth. This occurred in one of my own fields, whilst it was being rolled, and the hinder quarters of the shaft horse fell in; two or three cart-loads of rubbish were required to fill up the hole. The subsidence occurred where there was a broad depression, as if the surface had fallen in at several former periods. I heard of a hole which must have been suddenly formed at the bottom of a small shallow pool, where sheep had been washed during many years, and into which a man thus occupied fell to his great terror. The rain-water over this whole district sinks perpendicularly into the ground, but the chalk is more porous in certain places than in others. Thus the drainage from the overlying clay is directed to certain points, where a greater amount of calcareous matter is dissolved than elsewhere. Even narrow open channels are sometimes formed in the solid chalk. As the chalk is slowly dissolved over the whole country, but more in some parts than in others, the undissolved residue--that is the overlying mass of red clay with flints,--likewise sinks slowly down, and tends to fill up the pipes or cavities. But the upper part of the red clay holds together, aided probably by the roots of plants, for a longer time than the lower parts, and thus forms a roof, which sooner or later falls in, as in the above mentioned five cases. The downward movement of the clay may be compared with that of a glacier, but is incomparably slower; and this movement accounts for a singular fact, namely, that the much elongated flints which are embedded in the chalk in a nearly horizontal position, are commonly found standing nearly or quite upright in the red clay. This fact is so common that the workmen assured me that this was their natural position. I roughly measured one which stood vertically, and it was of the same length and of the same relative thickness as one of my arms. These elongated flints must get placed in their upright position, on the same principle that a trunk of a tree left on a glacier assumes a position parallel to the line of motion. The flints in the clay which form almost half its bulk, are very often broken, though not rolled or abraded; and this may he accounted for by their mutual pressure, whilst the whole mass is subsiding. I may add that the chalk here appears to have been originally covered in parts by a thin bed of fine sand with some perfectly rounded flint pebbles, probably of Tertiary age; for such sand often partly fills up the deeper pits or cavities in the chalk.
 S. W. Johnson, 'How Crops Feed,' 1870, p. 139.
 'Nature,' November 1877, p. 28.
 'Proc. Phil. Soc.' of Manchester, 1877, p. 247.
 'Trans. of the New Zealand Institute,' vol. xii., 1880, p. 152.
 Mr. Lindsay Carnagie, in a letter (June 1838) to Sir C. Lyell, remarks that Scotch farmers are afraid of putting lime on ploughed land until just before it is laid down for pasture, from a belief that it has some tendency to sink. He adds: "Some years since, in autumn, I laid lime on an oat-stubble and ploughed it down; thus bringing it into immediate contact with the dead vegetable matter, and securing its thorough mixture through the means of all the subsequent operations of fallow. In consequence of the above prejudice, I was considered to have committed a great fault; but the result was eminently successful, and the practice was partially followed. By means of Mr. Darwin's observations, I think the prejudice will be removed."
 This conclusion, which, as we shall immediately see, is fully justified, is of some little importance, as the so-called bench-stones, which surveyors fix in the ground as a record of their levels, may in time become false standards. My son Horace intends at some future period to ascertain how far this has occurred.
 Mr. R. Mallet remarks ('Quarterly Journal of Geolog. Soc.' vol. xxxiii., 1877, p. 745) that "the extent to which the ground beneath the foundations of ponderous architectural structures, such as cathedral towers, has been known to become compressed, is as remarkable as it is instructive and curious. The amount of depression in some cases may be measured by feet." He instances the Tower of Pisa, but adds that it was founded on "dense clay."
 'Zeitschrift fur wissensch. Zoolog.' Bd. xxviii., 1877, p. 360.
 See Mr. Dancer's paper in 'Proc. Phil. Soc. of Manchester,' 1877, p. 248.