is four hundred feet long, and the top is covered with large marble slabs of dazzling 11 42 PREL1MINARYESSAY. brilliancy. On the side next the reservoir, a substantial marble balustrade, three feet in height, gives a finish to this Cyclopean undertaking. A tall marble tablet indicates the date of its erection, or more probably of its repair or reconstruction. From the date, 1211, it appears to have been built about forty-six years ago. It is called the Validay Bendt, and is said to have been built by the mother of the reigning sultan. It is furnished with a waste gate, and, at a short distance below, the water from the reservoir is carried across a ravine by a short aqueduct. About two miles from this is another bendt, erected in 1163, which corresponds to the year 1749. This is also a magnificent work, although inferior in size to the preceding. They both supply the aqueduct of Batchikeui, which, as has already been stated, furnishes the suburbs of Pera and Galata with water. Beyond Belgrade are other reservoirs which will be elsewhere noticed. These supply Constan- tinople proper, with water. " In order to convey a clear idea of the direction of these various hydraulic works, it may be advisable to follow each singly. Beyond Belgrade is a large bendt, which sends its waters into a basin already partially supplied from another reservoir. A mile farther on, the water is carried across two aqueducts, the larger of which is known as the aqueduct of Mustapha III. From this it is conveyed into the aqueduct of Justinian. This is twelve miles from Constantinople. It consists of two tiers of arches, each forty-two feet wide. The arches are four in number ; the total length of the aqueduct, with its abut- ments, is seven hundred arid twenty feet, and its greatest height a hundred and ten feet. A gallery pierces the square pillars, forming the first story of arches, and allows a passage through its whole length. There are four small arches at each end of the first story, about twelve feet wide. The precise epoch of the construction of this aqueduct is not known, although it is commonly attributed to the Emperor Justinian II. This aqueduct receives also water from two others, the principal of which is known under the name of Solyman. This is sixteen hundred feet long, and eighty feet high, and consists of two stories of fifty arches each. It is a Turkish work. Another aqueduct also conveys water into that of Justinian, and is generally supposed to be of the age of Constantine. It is three stories high ; the lowest tier consists of thirty-three arches, fifteen feet wide, the second of twelve arches, and the uppermost of four. It is three hundred and fifty feet in length. All these magnificent and costly structures are intended for the supply of Constantinople alone, and we will now trace the course of the water. Leaving the aqueduct of Justinian, it follows the right bank of the Cydaris, and receiving in its course various tributary rivu- lets from the neighboring hills, it enters within the walls of Constantinople near the aygry kapoosi, or crooked gate, whence it is distributed over the city. It was impossible to ascertain the quantity of water furnished throug-h this series of hydraulic works ; but, judging by comparison with that which supplies the suburbs, it cannot be less than fifteen millions of gallons within twenty-four hours. PRELIMINARY ESSAY. 43 " We will now return to the aqueduct of Batchikeui, and follow the direction of its waters. These are carefully brought round the heads of the valleys in covered canals, in which there are at certain intervals, sudden breaks or alterations in the level, which answer the double purpose of agitating the water in contact with air, and of precipitating its impurities. It likewise affords fountains on the road for the use of cattle and weary travellers. When hills intervene, tunnels are boldly driven through, at the depth of fifty, eighty, and in some places a hundred feet. The course of these tunnels may be traced on the road between Pera and Buyukdery by numerous pits, which were about two hundred feet apart. These pits were convenient for giving air and light beneath, and also afforded a ready means of getting rid of the excavated earth and rocks. It is possible, that at the period when these tunnels were made, the pits were previously dug, in order to enable them to give the necessary direction and level to the subterranean passage. Branches from this main stream are continually thrown off to supply the villages, and the palaces of the sultan along the Bosphorus. Notwithstanding all these expensive works, it some- times happens, after long droughts, that the