Average Rainfall Supply.

161

level of the surrounding lands, and the adjacent public road. Of this area six biggahs was marked out for the tank and catch-water surface. The tank or reservoir was then excavated to a total depth of twenty-seven feet with a surface area of three biggahs and three and-a-half cottahs. The banks were formed with a slope of two to one, with a four feet berm at eighteen feet; where the footings of the ghats were laid. The remaining land was raised several feet above the surrounding level, with a gentle slope from the edge of the tank outwards to an artificial stone channel on all sides of the conserved area, the water falling over this area being caught by the stone channel and led by masonry shoots at the corners into the reservoir; by this means erosion and guttering of the banks is effectually prevented. The supply of water derived from the direct rainfall is supplemented to some extent by four deep wells in the bottom of the reservoir, which is calculated to hold over four millions of gallons of water. The reservoir was planned on the following calculation: The rainfall in Bengal for the ten years previous to construction had varied from 45.55 inches to 93.31 inches, the average, therefore, was taken at 70 inches per annum. It had been ascertained by experiment that, with a rainfall of 70 inches, the amount of water falling over a measured area of 1,200 square feet amounted to 43,628 gallons. Taking the catch-water area, therefore, to be six biggahs, or 86,400 square feet, the amount of water falling over it with the same rainfall would amount to nearly 3 millions: we may assume the loss by evaporation and absorption to be counterbalanced by the springs and wells, and as the entire supply is not consumed, the rainfall is ample

L

162

Soil and Surroundings.

to maintain the water level. Although we have had, during the past year, the shortest rainfall known for some twenty years, this reservoir has never had less than seventeen feet of water in it at its lowest, and its water is in use by a large neighbouring population.

In constructing tanks due heed must be had to the soil and to the surroundings. A tank, wholly or partly excavated in what is termed made-up ground, that is, originally low land raised in the course of years by refuse deposits, will hardly ever yield good water, and the subsoil of nearly all our old Indian towns is so fouled by the soakage of the sewage or filth of generations, that great care is necessary in selecting a site. If the soil be loose and sandy, and there be a number of filthy ponds or cesspools in the vicinity, there will always be the danger of sewage percolation from them into the new excavation. This may, of course, be prevented to a great extent by well puddling the bottom and sides of the reservoir, or by the use of a puddle core in the embankment; but this adds greatly to the cost of the work, is not always easily effected, and from the slipping of the banks or the disturbance of the puddle coating is not always effectual in preventing filtration. A new tank, therefore, should always be dug, where possible, in fresh new ground, and all cesspools and filthy ponds in the vicinity should be emptied, cleansed, and filled up with earth. This can generally be easily effected whilst the excavation is in progress, and the spoil earth is available.

Where a tank is to be excavated near a river and in sandy soil, advantage may sometimes be taken of the tendency to filtration to secure a constant supply of

Filtration and Percolation.

163

water partially filtered, and therefore cleared of the silt and mud, held in suspension by the river water.

A rather curious instance of this exists in a fine large reservoir constructed by me, some years ago, in Garden Reach. There had existed, for many years, at the back of the Government Dockyard premises, a very extensive jheel or swamp, and which, as its name, the Adhygunga, implies, was, in former years, a portion of the original bed of the Ganges. At what period. it was cut off from the river I do not know; but a good many years ago, when the Dockyard premises were enlarged, the old Garden Reach road was diverted and carried across a portion of the swamp, which had been, in the course of years, filled up by the deposit of sand from the docks, mixed with the cinders and slag from the furnaces and cupolas. The swamp was long an eyesore and a source of malaria and sickness to this part of the suburbs. In 1875, the Government offered the site, some fifteen biggahs, to the Municipality free, on condition, that a tank should be formed in one portion of the ground, and the rest of the swamp raised and reclaimed. The work was accordingly taken in hand, a portion of the native village adjoining purchased and removed, and a fine rectangular reservoir, with a water surface of nearly six biggahs surrounded by walks, grass, and ornamental shrubs, now occupies the site of the old, unsightly, and pestiferous swamp. Owing to the nature of the soil, a wet sand, it was found impossible to get deeper down with the excavation than seventeen feet, the water below this depth rising fast in the cuttings. Whilst making the excavation, undeniable proof was discovered of the swamp having formed in

bygone days a portion of a river bed, for, at a depth of ten feet below the surface of the bed of the jheel, several portions of sunken boats came to light, and in one spot an entire boat, measuring some thirty feet in length, was disinterred, containing, besides the remains of firewood, rice-bags, earthen utensils, &c., the complete skeleton of a man, portions of two other human skeletons being found underneath and close to the boat.

At one corner of the tank, a most persistent spring of clear water interfered greatly with the work, the water filling up the cuttings as fast as it could be baled out; and, on investigating the cause, the spring was traced back to the bed of cinders and slag underlying the Garden Reach road. Adjoining this is a large settling tank for the river water used for the foundry boilers, and which is pumped daily from the river, the water passing through the thick bed of cinder, slag, and sand, is, therefore, effectually filtered. A pipe culvert was laid to conduct this water into a settling well, from whence it passes into the reservoir through an overflow pipe. There is thus not only a constant accession of fresh filtered water into the reservoir, but a continual, though imperceptible, current through it, which assists in keeping the water clear and fresh. The supply is so constant, that, in the hottest and driest seasons, the water level only sinks from two to three feet.

The two examples given will serve to show what can be done to improve a tank water-supply and at a comparatively small cost, the two tanks mentioned having been completed, inclusive of purchase of land, construction of ghats, drains, railings, and turnstiles, for something less than Rs. 8,000.

Effects of Water-plants and Fish.

165

In all tanks or reservoirs, water plants and fish in moderate quantities are a necessity: the former exert a distinct chemical influence on the water, while the latter not only prevent overgrowth of vegetable matter, but consume a large quantity of insects, crustaceæ, and organic matter.

That well-known Indian Chemist, Dr. David Waldie, of Cossipore, says: "Some of the best tanks, General's tank more particularly, are probably equal to the river water in purity at some seasons, and superior to it at others. Tank water deteriorates in the hot season from putrefactive fermentation, the river water proper improves from oxidation, but near Calcutta deteriorates from sewage and tidal water. Tank water improves during the rains by dilution with rain water, and the animal and vegetable life in it preserves the proper balance, removes decaying matters, and prevents putrefaction to any great extent; at least this is the case in good tanks. General's tank seems a well kept aquarium: it abounds in animal life.”—(Asiatic Society's Journal.) Dr. Wilson says, "the purifying process is aided to some extent by the presence of fresh water plants;" and again, “among other purifying agents may be mentioned the effects of plants and fish." In store reservoirs, the presence of a moderate quantity of living plants exerts a decidedly purifying influence, while the destruction of fish has been followed by an excessive multiplication of the small crustacean animals on which the fish had lived, thereby rendering the water nauseous and impure. The remedy was found in re-stocking the reservoir with fish (Rankine).-(Wilson's Hygiene.)

Of the water-plants best suited for preservation of