in this construction were one foot nine inches PRELIMINARY ESSAY. 33 long, one foot broad, and one and a half inches thick ; the cement of one of the aqueducts at the bottom is six inches thick, and one and a half thick on the sides ; about two feet above the floor of the canal were fixed, on each side, cramps of three lines square, at two and a half feet distance from each other. The utmost breadth of the piers of the aque- duct of Chaponost, which carried a canal of three feet broad, by six feet high, is not more than six feet, while the breadth of the aqueduct which passed over the river Baunan, arid which has no canal, is 24 feet broad, consisting of two piers, each five feet, supporting an arch 14 feet in diameter. M. Delorme, in his account, (Seance de 1'Academie, 1759,) traced three of the aqueducts of ancient Lugdunum to their source, in three several tracks, of many miles each. He examined their general level, and the level of each part, as they ran above and under ground, along the sides of the mountains, and sides of valleys, and over the bridges where they passed the valleys ; he observed the apparent care which the archi- tects of these edifices took to avoid the building of works, enormous in bulk, height, and expense, by carrying the walls up into the narrower and shallower parts of the valleys. Where smaller bridges would serve, and where they could carry the waters over a bridge by a rectilinear canal, they always built up bridges to that level, but where that would become too high, and yet where a bridge was necessary, they built bridges of a height sufficient to carry the water over in syphons of easy curvature. The sources of supply of the aqueduct of Mont de Pile, were from the waters of the river Gievre, from the rivulet of Sauon, and probably from the river Tanon, to which were joined those of the rivulet Langoneau ; after these waters were united in one stream at the aqueduct bridge of the little Varizelle, they make a long detour on the sides of the mountains and hills, till they arrive at the valleys, which they must pass ; yet here they are seen trained along the sides of these valleys, until they come to situa- tions which are not so deep or so wide. It is then that the architects built bridges across the valleys, over which to conduct the waters, either in rectilinear canals, or in syphons ; had this latter precaution not been taken, the construction of such bridges would have been of necessity so high as to become enormous, both in work and expense ; yet, notwithstanding all these precautions, there were in the aqueduct which takes its sources in Mount Pile, and determines at the gates St. Irenaeus, nine bridges carrying aqueducts, and three calculated to carry syphons. The ninth is in a very deep and wide bottom, on the heights of Soncieu. The aqueduct, when it arrives at this bottom, is terminated with a reservoir at the south edge of the valley of the river Garon. The mode by which the water passed this pro- found chasm, was by causing it to flow from a reservoir on the one side, in leaden pipes^ bedded in the sides of the valley along part of the descent ; it then flowed in continued 34 PRELIMINARY ESSAY. pipes of the same sort, bedded on a bridge, whose top course was built in a descending or reversed curve ; having thus passed over this bridge, when they came at a certain height, on the opposite side of the valley, they were protruded up in pipes, bedded as before, on the opposite sides of the valley, and the water was delivered in another reser- voir on the top of this opposite hill, called the reservoir of Chaponost. From this reservoir, the water entered into the aqueduct of Chaponost, which runs under ground along the west side of the village. It emerges on the north, and flows over a bridge composed of ninety arches, of which more than sixty, in Delorme's time, were remaining. This was terminated by a reservoir, whence the water, in like manner as before, descended by pipes into another valley, and in part passed it and the river Baunan, over a bridge of a reversed curvature, and mounted again on the opposite side, there entering a second reservoir at St. Foi. The waters flowed hence in a canal, carried by a bridge for some way above ground, and then became subterraneous, and continued thus along the heights to the point near the gate of St. Irenseus. Here another reservoir was situated