178

HISTORY OF ENGINEERING.

Book I.

this kind of construction is unique; the voussoirs seem to be carried side by side, and thesoffites of the arches exhibit the joints not broken, but continuing round in one line. Thethickness of the arch consists of three distinct arches not tied or l>onded together.

The whole is of freestone, from the foundations to the third course above the cymatium,which covers the piers of the third story. Moellon or rubble was employed for the fillingin of the piers, spandrills, and haunches, of the first and second stories.

The stones are laid without cement, and owe their solidity to the weight of each block,and the nicety observed in the working of their beds and joints. Each stone was droppedinto its place by means of the lewis, the holes for the insertion of which arc still seen exactlyplaced over the centre of gravity of each. The work above the small arches of the thirdstory is of masonry filled in with rubble work : here a considerable quantity of cement isused, which has become as hard as the stone itself, forming one impermeable mass, and pre-venting any filtration, which would, if suffered, have been detrimental to the structure.

The channel for the water is placed between two walls of masonry, 2 feet 9 inches thick ;the bottom has an inverted arch, which with the walls are covered with a cement 2 inchesin thickness, composed of fine sand, powdered brick and quick-lime. Its strength is equalto that of the hardest stone, and there is no apparent change in it, since it was first laidon. This layer of cement is covered with a fine mastic, very thin, and of a deep red colour,laid on with a brush, giving the whole a face as smooth as highly polished marble.

The waters of the Eure and Airan were conducted along this course, which was 4 feet inwidth, and 4 feet 9 inches in height. The walls are of moellon or rubble, and had overthem a covering or slab, 2 feet 6 inches in height, formed of two courses of freestoneprojecting 2 feet; over these were laid slabs of freestone, which covered in the aqueduct,12 feet in length, and 3 feet 3 inches in width, projecting over the walls below 9inches. The fall given to the water is four-hundredths of a foot for every hundred feet,throughout the entire length of the aqueduct, which brought the waters to the inhabitantsof Nismes, from the sources of the Eure and Airan in the valley of Uzes ; after these rivershave supplied many mills, they fall into the Gardon, near the Pont du Gard .

Several portions of the two aqueducts which conveyed their waters respectively to thegreat aqueduct still remain, between Uzes and the village of S. Maximin.

The commencement of the aqueduct was to the south of S. Maximin ; it followed allthe sinuosities of the hill, so that the level was always maintained ; it was entirely under-ground, and often cut in the rock itself. The Romans constructed small bridges over therivulets, which were permitted to pursue their course to the Gardon . One of these, thrownacross the cascade of the Bord Negre, remains. After having passed the Chateau d’Argelliesthe side of the hill is lower, and the summit is below the level of the waters conducted tothis point; here the aqueduct is carried on a series of arches, similar to those of the upperpart of the Pont du Gard , with which they unite, passing behind the village of Vers, anddescribing on their plan considerable bendings, that they may follow the crest of the hill,which precaution was necessary, to economise the height of the piers. Thus the waters ofthe Eure and Airan arrived at the Pont du Gard by a circuitous flow of 50,855 feet; afterit had passed the Pont du Gard , the aqueduct was lost in the sides of the mountain as far asLafoux ; it then reappeared, and was carried on arches more or less elevated across thegorges met with in its passage, some of which remain. It was then carried on the easternside of the hill, behind St. Bonnet, under the village of Sarnhac, and crossed the great roadto Nismes near Besonce, to arrive at the mountain of St. Gervasy, which it followed as faras Nismes. The whole distance from the Pont du Gard is 83,665 feet: whence it resultsthat the total length of the aqueduct between the extreme points was 134,515 feet.The relations of both Vitruvius and Pliny on the construction of aqueducts, the engineerhere finds carried into effect in the most admirable manner, and he must be convinced thatthe knowledge possessed by the ancients of the laws of hydrodynamics was far greater thanthey have received credit for. If their writers have not handed down to us the calculationsmade previous to the commencement of these vast and laborious undertakings, the worksshow practically their engineers thoroughly understood the art of levelling, and the lawsby which water in its course was governed : though the fall given to the slope of thechannel of this aqueduct is hut small, it is regular throughout, and from the masterly mannerin which it is conducted, we cannot but give the constructors the credit of having under-stood thoroughly the nature and properties attendant upon running water, and that theymust have observed and studied the slopes and beds of the rivulets and larger streams beforethey could have arrived at the knowledge displayed in these surprising monuments ; as muchcare was required, or perhaps more, to set out one of these water conduits, than is demandedfor a modern railway. We find the principles to perform both nearly similar, the locomotive,in one case, is to he moved along a level or inclined plane at a certain angle, while the wateris to keep one uniform and constant course; both in mountainous districts must be con-ducted over valleys on artificial constructions, and along the sides of steep hills, and thismust be performed without creating any abrupt turns or angles. In drawing a comparisonbetween the experience of the men employed in engineering works in times past and at the