Volltext Seite (XML)
THE INSTITUTION OF CIVIL ENGINEERS. April, 1868.—The first Paper read was, On Irrigation in India, by Mr. Allan Wilson, M. Inst. C.E. Having had fourteen years’ experience in the construction and superintendence of irrigation works in the central and southern pro vinces of India, where wet cultivation was extensively practised, the author proposed in this communication mainly to refer to that part of the empire. A brief account was in the first place given of the former and present systems of irrigation, followed by a detailed description of the accessory works; and the cost of providing the water was then noticed. The value of artificial means of irrigation, for increasing the fertility of the soil, was recognized in India at an early date. In the Punjaub, canals for this purpose, as well as for navigation, were constructed as far back as the middle of the fourteenth century. But it was in the southern parts of India, where the rainfall was more precarious, and the river supplies less easily available, that the most extensive works were to be found. It had been estimated that, prior to the establish ment of British rule, there were, in fourteen of the piincipal irrigated districts of the Madras Presidency, upwards of forty-three thousand tanks and channels in repair, besides about ten thousand out of repair, having, probably, thirty thousand miles of embankments, and three hundred thousand separate masonry works. Some of these tanks and reservoirs were on an immense scale, for irrigating many thousand acres, while there were smaller tanks, wells, and springs, which watered only a few acres. It was remarkable that the Government should have allowed so many fine works gradually to fall into decay, without replacing them by others; as great natural facilities existed for storing water, and for forming canals to lead it on to the land. The irrigation works on the Godavery and Kistna rivers, in the Northern Circars, and on the Coleroon, in Tan jure, had only recently been completed ; but many large rivers were still allowed to flow into the ocean, almost unused for agricultural purposes. With regard to the most general and least expensive mode of irriga tion by means of artificial reservoirs, and to the methods adopted in forming such reservoirs, it was stated that, in selecting a site, it was essential to ascertain, in the first place, that the foundation was suit able; the next point to be determined was the extent of land to be irri gated, and the quantity of water necessary for such irrigation. The area of the drainage or gathering grounds could be estimated from the trignometrical survey maps of India, and the quantity of water that would pass into the tank dining floods should be calculated according to the known rainfall, due allowance being made for absorption and evaporation. With these data, the dimensions of the different works could be fixed. It should, however, be borne in mind, that depth of water was of greater importance than a large surface area, as the eva poration would be less in the former case. An examination should also be made of the valleys in the vicinity of the proposed reservoir, with a view to ascertain whether the surplus water flowing through the tank during floods could not be carried across intervening ridges, and be stored in natural basins at a small outlay, so as to fill a chain of tanks. It was explained that a tank was simplv a reservoir formed by throwing an embankment, or bund as it was called in India, across a valley to dam up the drainage. The most simple description of bund was constructed entirely of earth, which was generally dug from the bed of the intended reservoir. The breadth at the top was usually about 12 feet. The inner slope was 3 to 1, and this was faced with a pitching of loose stone, •while the slope of the land side varied from 2 to 1 to 1 to 1. Buddle was seldom, if ever, used; indeed it was not required, as owing to the lodgment of silt, a tank would puddle itself as soon as it had been once filled. In illustration of this fact it was mentioned, that Major-General Sir Arthur Cotton had stated that in a channel cut through loose sand, within a yard of the water’s edge to a depth of 5 feet, not the least moisture was found in the excavation; the lining of silt having rendered it completely water-tight. In addition to this embankment, some of the large Hindoo works had a massive retaining wall of masonry in front. Many of these walls were built of dressed stone, close-jointed, backed with rubble and a rough description of concrete; and flights of steps of cut stone were constructed down to the edge of the watsr. To obviate the danger of an excessive influx of water during floods, most tanks were provided at one end, and not unfrequently at both ends of the embankment, with a waste weir (known in India as a calingalah), to allow the surplus water to escape after the tank had been filled. In constructing a tank, the discharge capacity of the calin- gulah was an essential feature. It was a safe rule to allow one-fourth more than the dimensions obtained by calculation, so that the water might have a free passage in the event of an excessive flood, as other- the earthwork might be entirely destroyed. The author had found that many of the tanks which were now useless, had been breached from no other apparent cause than the want of sufficient outlet to carry off the surplus water during floods. In constructing these calingulahs, upright stones, varying from 2 feet te 5 feet in height, were fixed in the body of the masonry, at intervals of from 3 feet to 4 feet. When the rains were moderate, as the monsoon was declining, these spaces were filled with brushwood, straw, earth, or rubbish, to prevent the further escape of water, and to store up as much as possible. This con trivance showed how highly water was appreciated by the natives of India; but unless the process of filling up was very gradual, and its completion deferred until the rains had entirely ceased, the results were often most disastrous. Many of the ruined tanks appeared to have been, breached solely from the desire on the part of the cultivators to retain the utmost quantity of water. The forms and dimensions of the calin gulahs must, of course, be regulated according to the capacity of the tank, the area of the gathering ground, the rainfall, liability of drought, and other local circumstances; so that it was not possible to lay down any general rule, having regard to the great differences of rainfall, &c ; , in the several districts of India. In order to draw off the water for cultivating the fields, each tank was provided with one or more sluices; their size and number depend ing entirely upon the area of land to be irrigated. The sluice in gene ral use in Southern India was exceedingly simple. It was merely a tunnel through the bank, built of brick, or rubble, either arched at the top or covered with fiat stones, the floor being paved with flat-jointed flags, and the side walls lined with the same. The inner end, or head of the sluice or tunnel, was closed by stones, and the water was allowed to enter through one or more orifices, generally from 3 inches to 5 inches in diameter, cut in these covering stones. The flow of water through these orifices was regulated by means of conical plugs of wood, each of which, being attached to a long handle, could be withdrawn or inserted as required. At the end of the sluice tunnel below the bank, a cistern was generally built, having its sides pierced with holes at different levels, to enable the water to be drawn off at various elevations. From this point the water was carried forward in open ducts of the requisite dimensions, in the sides of which subsidiary or distributing channels were cut, generally at the expense of the farmer, to lead the water on to the land to be irrigated. The mode of preparing land for wet cultivation, was by laying it out in squares, rectangles, &c, each plot being separated from the rest by mud walls of sufficient height and thickness to prevent overflow or percolation. These squares or rect angles were kept, level, and were as large as the nature of the ground would admit. When one was sufficiently flooded, the water was let off at the lower corner of the field into the next division, and so on until the whole area had been irrigated. With a view of showing how favourable some parts of India were for forming reservoirs of large capacity, attention was directed to a design for a large artificial lake, which it was proposed to construct by- damming up the gorge of a valley. This reservoir would be capable of storing sufficient water to irrigate 200,000 acres of land, an an a equal to the county of Buckingham, allowing the usual average of 500 acres to the square mile as being under cultivation. Taking 170,000 acres as the extent of land to be irrigated for a single crop, this would require provision to be made for the discharge of 170,000 cubic yards of water per hour at each end of the tank ; and discussion was invited as to the best description of sluice for discharging such a vast volume of water. Another method of providing a supply of water for artificial irriga tion was by damming up rivers, by anicuts, or masonry dams (also called weirs). The sites for many of ihese had been so judiciously chosen by native engineers, that but little expense had been incurred in building them. They were generally formed of boulders and stones thrown promiseulously thrown together. The side next the water had the outline of a retaining wall with a slight batter, and was laid with some care. On the down-stream side, two rows of stones composing the lower edge of the apron were carefully laid by hand, and the upper courses on both sides were similarly placed. A detailed description was given of some old anicuts across the River Toombuddra, as well as a modern one over the same stream; of an anicut across the Kistna river, near the town of Bizwarrah; and of temporary weirs, of rough stones, which were not now so generally used as they were formerly. These masonry dams were, when practicable, taken advantage of for forming means ©f communication from one bank of the river to another, by building a bridge on the top of the weir. These dams had sluices at each end, and if requisite at other parts, to be kept open at times, so as to prevent the lodgment of silt above the weir. A head sluice should also be constructed, to enable the water in the river during a flood to be shut off. and so prevent sand, &c., from entering the irri gating channel. These sluice openings were closed by native shutters, consisting of teak-wood planks, a native system which had been found to answer so well, that it had been adopted more or less by English engineers. Great differences of opinion existed as to the cost of storing and dis tributing water for irrigation. Thus, in a report made by Colonel Dickens, in 1855, on the Shahabad district, the estimated cost of storing and supplying water from four different tanks, including the cost of the distributing channels, was 400, 352, 350, and 494 cubic yards per rupee respectively, or an average of 400 cubic yards per rupee, giving a cost of 8 rupees 12 annas per acre irrigated with 3,500 cubic yards, a quantity which was amply sufficient to raise more than an average rice crop. Colonel Rundall stated the rates per acre in the Madras Presi dency at 6 rupees and 8 rupees 3 annas, giving an average of 7 rupees 15 annas per acre as the cost of storing water; but it did not appear whether the expense of the engineering establishment and of the dis-