was sufficient to fill all the syphons ; without this precaution the volume of Avater, which might have risen to 4 feet, would have been too great. It is probable that the regulating vane or sluice could be raised, or lowered, at pleasure to allow for the increase of head, and that the superfluous quantity was at times permitted to flow into some other channel. Where the aqueduct was carried above ground, it was buiH on a footing of masonry 6 feet thick, even where the elevation above the surface did not exceed 6 or 7 feet ; but when it was greater, arches were formed, and also piers, when the elevation was consid- erable ; on this elevation depended the span of the arches, the thickness of the piers, and their height. For an opening of 18 feet in height the width is 12 feet, and the piers 6 PRELIMINARY ESSAY. 37 feet, sustaining a semi -circular ; when the inequality of the surface did not allow the piers to have an equal height of 18 feet to each opening, the piers were shortened, and the other parts remained of the same general dimensions. The piers of the arches in some places, are rather less on the face than 6 feet, varying from 5 feet 8 inches to 6 feet ; and in other cases they are 7 feet 9 inches. The arcade which conducts the water into the reservoir called St. Irenasus, is 31 feet high, its width is the half of this, and the faces of the piers are 7 feet 9 inches. As the upper part, containing the canal, is only 6 feet thick, there is an offset of 6 inches on each face at the impost of the arches. On this off- set there is a projection, or pilaster, 10 inches thick, and 3 feet wide, which acts as a counter-fort, to strengthen the sides of the water channel. The foundations of the piers having the smallest elevation, are sunk between three and four feet below the surface of the ground, and between six and eight feet for those of the greatest height. All the different supports of the aqueduct are of the same kind of masonry, formed of small, rough, squared stones, laid in a thick bed of mortar, with the apparent faces of reticulated work. This kind of masonry was bound, at every four feet of its height, by two courses of "great bricks," each brick being 22 inches square, and two inches thick. The angles of the piers, formed of small square slabs of stone, offered, in many instances, an insufficient resistance to the lozenge masses which they terminated, and their displace- ment has been apparently the main cause of the ruin of the greater number of the piers, for these have been formed by a sort of encasements, of the thicknesses of four feet of the opus reticulatum, without being properly bonded by stones large enough at the quoins. The arches are semi-circular ; the arch stones are slabs (thick slates) of stone, three inches thick, alternating with a " great brick ;" the extrados of the arch is finished by a row of bricks, which forms a fillet ; on this fillet is laid a double horizontal row of bricks, which runs through the entire length of the aqueduct, without, however, forming any projection. It is upon these bricks, as a pavement, that the water channel is laid, or bedded. Of the arcades forming that part of the aqueduct called Langoneau, only seven piers remain, arid these of the common reticulated masonry. The valley between Soncieu and Chaponost is about 200 feet deep. Five ranges of arcades, placed one over the other, for a length of 2400 feet, conducted the water across the valley ; the valley through which the river d'Izeron flows, between Chaponost and St. Foi, is nearly 300 feet deep, and was crossed by a series of arcades having eight ranges in height. The third valley, formed by the small hill of St. Foi, and that of Fourvieres, had three ranges of arcades. These prodigious substructions must have occasioned an outlay so enormous, as under almost any circumstances would have completely arrested the completion of the underta- king ; and the more so, as these valleys were neither all, nor the greatest across which th* 10 38 PRELIMINARY ESSAY. water had to be conducted. The resources of the architects here become conspicuous, in their substitution of leaden pipes, forming syphons, already described, which were laid at an expense comparatively trifling, to what must have been incurred by following the other and more usual method. In describing the passage of the valley of the Garon, the aqueduct arriving at the summit of the hill, was stated to deliver