rizontally, aud even upwards, the last arising from the general statical law, that "action and reaction are equal and in opposite directions." It follows, from these hydrostatical laws, that the lowermost portion of each side of our supposed cistern will suffer a pressure from the water equal to that which acts upon the bottom, hence, taking the lowermost inch in the height of the sides of this cistern, it will be pressed with a force of 523 lb. or thereby, or 44 lb. on the square inch, and each of the four sides will suffer the same pressure. Suppose, now, that the cistern is elongated in one direction to any number of feet, and again filled to the depth of 10 feet, the pressure on each square foot of the bottom remains the same as before, and so in like manner does it remain the same upon the sides, for the pressure is not altered in any direction, although the proportion of the cistern has been changed. Keeping this in view, it will be seen that length and breadth produce no effect on the pressures that a fluid exerts against the vessel or body that retains it, and that, in calculating the resistance to sustain such pressures, depth is the only element requiring to be taken into account. It is also to be kept in view, that pressure on the bottom or sides is directly as the depth; thus, if our supposed cistern were reduced to 5 feet in depth, the pressure on the bottom would only be one-half, or 24 lb. on each square inch. (2048.) The conclusion to be drawn from these remarks is, that a cistern in the form of a pit or well should be always avoided, unless it can be formed in a natural bed of impervious clay. When such a substratum can be attained, a pit may be adopted, but not otherwise. If such has been found, and the pit dug out, it should be lined with brick or with stone built in mortar, the bottom being first lined with the same material. When the building approaches to the surface, the wall can be gradually reduced in diameter to a small compass, leaving only an opening of 2 to 3 feet square, which is covered in at small expense, and the saving in this last item is the only apparent advantage that seems to attend the practice of pit cisterns. Deep cisterns are liable to another inconvenience, of their becoming recipients of spring or of drainage-water; and it is sometimes more difficult to keep such water out than to keep the proper liquid in, for if springs and their origin lay at considerable heights, their hydrostatic pressure may be so great as to render the prevention of access to their products a process of great difficulty. (2049.) A cistern of moderate depth, not exceeding 4 feet below the out-fall of the drains, may be constructed in any situation, whether in gravel or in clay, and its length can be extended so as to afford any required capacity; the breadth being restricted to that for which materials for covering it, can be most easily obtained, which may be from 3 to 4 feet. Whatever be the stratum in which such a cistern is to be formed (unless it be perfectly impervious clay), it should be puddled to the thickness of at least 1 foot with the best clay that can be procured. For this purpose, the earthy matters are to be dug out to a depth of 14 foot lower than the intended sole, and to a width of 4 feet more than that proposed for the cistern. Two or three thin layers of the prepared clay are then to be compactly laid over the whole breadth of the excavation, and beaten firmly together at all points, making up the depth to 1 foot, and the surface of it brought to a uniform level. Upon this the side-walls are to be founded, and these may be of brick 9 inches in thickness, or of flat bedded rubble stone 14 inches. The wall should be built in successive courses of about 1 foot in height, the whole being bedded in mortar, and, as each course is completed, the puddle is to be carefully laid and beaten in behind, in layers of 6 inches or thereby, the first layer being properly incorporated with the foundation puddle, and each succeeding layer with the one immediately preceding it. To prevent the side-walls from being pushed inward by the pressure of the puddle or of the bank, tie-walls of brick or of stone should be formed at every 5 feet of the length of the cistern, these may be 9 inches of brick or 14 inches of stone; and they must have conduits formed at the level of the sole, to allow the liquid to run towards the pump. The sole should be laid all over with brick set on edge, or with strong pavement, the whole having a slight declivity towards one end, where a small well-hole of 9 inches in depth is to be formed to receive the bottom of the pump. The brick or pavement, as the case may be, is to be bedded on the puddle, and grouted flush in the joints with mortar; and when the walls and sole are built up, they should then be pointed in every joint with Roman cement. The covering is to be effected with strong pavement, of length sufficient to rest on the side-walls, laid and jointed with mortar; or with rough found-stones, where such can be procured, and if neither can conveniently be found, a beam of sound Memel fir may be laid along the middle of the cistern resting on the tie-walls, and, with this bearer, stones of half the length will be sufficient to form a cover. A thin layer of clay may be laid over the stone covers, and upon that a coat of gravel, by which means carts may be allowed to pass over it. To prevent accident, however, it is always desirable to construct the cistern in a situation where it will be as little as possible exposed to the transit of carts; and this may be always obtained at a small additional expense of covered drain to convey the manure from the dunghills to the cistern. (2050.) The pump for lifting the liquid from the cistern to the cart may be either of wood or cast-iron, but the latter is preferable. A common suckingpump of 3 inches chamber is quite sufficient; the chamber should be bored out, and the pump-boxes, for durability, should be also of metal, with leathern flap-valves. The height of the pump should be such as to deliver the liquid freely into the funnel of the barrel, or tank; but if this height is found to raise the pump-lever above the reach of a man's hand, it is only necessary to joint a light connecting-rod to the lever, its lower end being furnished with a crosshandle, and by these means the pump-man will be able to work the pump in the same manner as the lower end of the common pit-saw. Forcing and lifting pumps have been proposed and even employed for the purpose we have here in view, though with questionable propriety; and here it may be proper to explain, that by the term force-pump is to be understood a pump that raises water to any height above the point where the power is applied, by the descent of a solid piston acting in the chamber of the pump, sending the liquid into an ascending pipe, which springs from below the piston. The lifting pump differs from this in having a valved piston through which the liquid passes, as in the sucking-pump, on the descent of the piston; and, on its ascent, the valve being now closed, the liquid is lifted and forced into the ascending pipe, which, in this case, springs from above the piston, the chamber being closed at top with a water-tight stuffingbox. From this brief description, the simplicity, both in construction and in management of the sucking or common pump, as compared with the other two, will be obvious; the cost being also in favour of the first.-J. S.] 45. OF WINTER IRRIGATION. "Hence Irrigation's power at first was learnt, In cold and watery climes; though even there GRAHAM. (2051.) It is not my purpose here to describe the mode of making irrigated meadows; that being a summer occupation, it will be attended to in due time. At present, I shall only describe the watering of meadows in winter. Suppose, then, that the meadows have been formed in a proper manner, and every channel cleaned with proper care, let us at once proceed to let on the water, and regulate its duration so as not only to preserve the vitality of the plants, but to promote their vegetation at an earlier period than the natural call of the season would arouse into action. As the late Mr George Stephens knew the irrigation of land well, I shall use his words in describing the management of water on meadows in winter. I shall only premise that there are 2 kinds of water-meadows, one called the bed-work, in which the ground is made to decline gently, and the water to flow in the direction of the inclination. The other kind is called catch-work, which is only suited to ground having a considerable declivity, and by which the water is brought across the face of the declivity. The object aimed at in both kinds is the greater production of grass. Irrigation, therefore, is one mode of manuring grass-land. (2052.) "The whole works," as Mr Stephens directs, "being repaired, and there being generally water enough at this season either for the whole or for part, the sluice should be drawn, when, in the course of half an hour, the conductor and the upper part of the feeders will be nearly filled. The first operation of the irrigator is to adjust the water in the conductor; or, if the meadow is in more parts than one, the water in each conductor must be first regulated. Then he commences a-new, by regulating the stops in the first feeder; but should there not be sufficient water in the feeder, a little more must be let in, by making the aperture wider or deeper, till the water flows regularly over the sides from one end to the other. From the first he proceeds to the second feeder, and so on, until the water in all the feeders is adjusted. Let the beds of a water-meadow be ever so well formed, yet, by some places sinking more than others, or by the ice raising the surface of the ground, although the water along the banks of the feeders has been ever so nicely adjusted, it often happens that there may be some places between the feeders and drains with too little water, when it will be advisable for the manager to make a third round, redressing inequalities of the surface so as to give every spot 1 inch deep of water. Every part of the works being regulated, the water should be allowed to run through the whole of October, November, December, and January, from 15 to 20 days at a time without intermission. At the expiration of each of these periods, the ground should be made completely dry for 5 or 6 days, to give it air; for there are few species of the grasses which form the most nutritive part of the herbage of watermeadows, that will long exist under an entire immersion of water. Moreover, if the frost should be severe and the water begin to freeze, the watering must be discontinued, otherwise the whole surface will become one sheet of ice; and whenever the ice takes hold of the ground, it will undoubtedly draw it into heaps, which is very injurious to plants." "The object of this early watering of the meadows is to take advantage of the autumnal floods, which bring along with them a variety of putrescent matter, which is found very enriching to land. It is the chief object of the irrigator in those months to collect as much of this manure as possible, and at the same time to shelter the land from the severity of frosty nights. It is therefore requisite to use as much water as the land will carry without guttering. I believe it would be difficult to give land, with a dry subsoil and considerable descent, too much water before the weather begins to get warm. It is necessary, in those months, that the meadows be inspected at least once in 3 or 4 days, to see that the equal distribution of the water is not obstructed by the accumulation of weeds," &c. (2053.) Simple as these directions are, yet the actual management of the water of meadows is not unattended with difficulty, but requires the exercise of good judgment and great attention. "The adjustment of water flowing over the surface of land," remarks Mr Stephens, "for the purpose of improving the herbage, is a very nice operation; it requires a perfect knowledge of levels and the vegetation of grasses, and ought never to be entrusted to an unskilful manager. When the supply of water is, in any state of the stream that supplies it, sufficient for the whole or one-half of the meadow at once, the management becomes pretty easy; for after the works are cleaned, and the water regulated in the autumn, the sluices should be fixed at such a height as to let in the exact quantity of water required, when it is allowed to run, according to the state of the weather and the season of the year, for 2, 6, 10, 15 days, without any alteration; and it will be found (unless the water has carried along with it weeds, sticks, or wreck of any kind) to run dur or ing that whole period nearly as equally over the surface as when first put on. But when the stream is small, and rising and falling with every shower of rain, the management becomes so much the more difficult, that it will require every possible attention of the irrigator to watch and change the water from one part of the meadow to another, or from one bed to another, according to its abundance or deficiency. Such meadows are indeed ill managed, although half an hour's work in a day would put every thing to rights. Indeed, let the formation of the meadow be ever so perfect, and the supply of water constant and uniform, yet it is necessary that the manager should survey the whole every 3 or 4 days, to remedy any defect occasioned by the accumulation of weeds, or by a stop being washed away, and thereby cause some places to have too much water, and others too little; so that, in the former case, the grasses might either be killed or very much injured by the generation of scum, or, in the latter case, there would be little or no produce of grass. Small streams are certainly much more at command than large; but if the manager, as is too often the case with a young practitioner, vainly endeavour to water too much ground at a time, he may give one part too much water, and another too little; for on the alteration of the apertures, and adjustment of the water, greatly depends not only the quality but the quantity of the crop." (2054.) There are many ways of mismanaging water-meadows, such as retaining a moist subsoil, or allowing the grass to stand too long before cutting; but there is an error committed at this period of the year, to which I wish to direct your attention, and which is thus characterized by Mr Stephens. "Another great error generally committed is, allowing the water to run too long at a time, without properly drying the ground. I know some instances where the ground is not attempted to be dried from the time the water is put on the meadows in autumn till 8 or 10 days before the cutting of the hay; the consequence is, that the grass is of the coarsest quality, and the ground becomes so very boggy, that the whole crop of grass is obliged to be carried by people to some other place to be made into hay. A water-meadow," adds Mr Stephens," like a garden, will be good for little without due attention. All dry soils require more attention than moist ones; for if the water in moist soils should not be so nicely regulated as on sandy or dry land, the crop of grass will not be so defective as on porous soils, where the management has been neglected. I presume that all dry land that has been converted into water-meadows, in countries where the art of irrigation is not well known, and the supply of water not abundant or regular, is liable to more injury, from imperfect treatment, than land |