The growth and prosperity of the period of peace between 97 and 180 AD directly ben­efits the Roman provinces. All Gallo-Roman cities of importance benefit from one or several aqueducts during this period, under the reigns of Nerva, Trajan, Hadrian, Antonius, and Marcus Aurelius, then until 235 AD under Severus. Aix-en-Provence and Lyon have four aqueducts; later on we will examine those of Lyon in detail. At Vienne, where the sources are very near,[232] we know of eleven aqueducts; at Poitiers (Lemonum) and Perigueux (Vesone), three. Some of these aqueducts are quite simple in design, with a very regular slope, as at Perigueux and Rodez. Others have quite a variable slope, most often steeper upstream than downstream, and with quite a variety of hydraulic works: tunnels, bridges, drops, siphons. The average slope itself can be quite variable from one aqueduct to another; it is usually between 0.2 and 1.6 m/km. One notable singularity is the Craponne aqueduct at Lyon, whose average slope is 16 m/km.

The dimensions and discharges of the aqueducts are also quite variable. The aque­ducts of largest cross-sectional area are at Vienne and Nimes. At Lyon and Nimes, the lengths rival those of the aqueducts of Rome itself (50 to 75 km). What is striking from the outset is, as shown in table 6.3, the diversity of configurations. Yet the dates of con­struction, when known, are most often in the 1st and 2nd centuries AD, the period of prosperity, Roman peace, and the concomitant economic climate favorable for develop­ment of the provinces. Samples of the broad panorama of Roman techniques of this period, adapted to local conditions, can be found in Gaul. We too rarely know the his­tory of these aqueducts, histories that are sometimes quite dynamic[233] as conditions change. The needs for water can increase over time, or the discharge available from the sources can decrease. Calcareous water can clog the conduits with deposits.

At Sens (Agedincum), the Fauconderie spring (on the left bank of the Vanne) is ini­tially the sole water source for the aqueduct. But when the spring dries up, four other springs are tapped successively until the flow is restored to, or slightly exceeds, the ini­tial supply. The aqueduct is increased in length from 6.1 km to 14.2 km, so as to reuse both the canal between the former Fauconderie spring and Sens, and the bridge on which the aqueduct crosses the valley of the Vanne.

At Saintes (Mediolanum), after some fifty years the capacity of the aqueduct is reduced due to the increase in roughness caused by calcium deposits. Therefore the aqueduct is completely rebuilt, being replaced by one of larger cross-sectional area but flatter slope, capturing water from new springs and delivering it to the city at a higher elevation. This reconstruction, like that of Sens, was based on re-use of large art works – the two bridges called the Arcs (27 arches, 298 m long) and the Hautmont (62 arches, 400 m long) as well as the tunnel of Neufs Puits. On these bridges, the canal is simply raised. But along underground segments, major construction efforts are devoted to mod­ification of the cross section. However, the water from new sources is even more cal­careous than the original. According to the estimates of Marcel Bailhache, the discharge, which had been increased from 4,000 to 22,600 m3/day by the reconstruction, subse­quently decreases to a value of only about 8,000 m3/day.

At Toulouse (Tolosa), the capturing of supplementary sources increases the dis­charge of the Ardenne aqueduct. But the resulting water depth (0.66 m) ends up exceed­ing the depth of the 40-cm canal lining, thus causing serious degradation of the canal walls.

Water quality is a strong determinant of the longevity of the aqueducts. If the water is calcareous, lime encrustation can reduce the flow area and increase the roughness to the point of reducing the discharge capacity by a factor of two in a few decades. Without maintenance, the canal can become completely blocked. On the other hand, if the water is not calcareous, the canals can have a very long lifetime. The aqueduct of Sens, for example, will be used up until the middle ages.

The most important works are those of Lyon and Nimes.

Table 6.3 A collection of synthesized data on the principal known Gallo-Roman aqueducts, listed in decreasing order of length.[234]

continued from page 145

The accession of Nero to supreme power marked the beginning of a somber period for Rome.[228] Civil war followed his assassination in 68 AD, then came the too-short reigns of Vespasianus and Titus. After a promising start, Domitian becomes morbidly suspi­cious, and sheds the blood of those near to him and his collaborators only to end up being assassinated himself in 96 AD. Then Nerva, an old senator, is chosen as the new emper­or in a climate of restoration of the old Roman virtues. We have noted earlier that the Romans placed high symbolic value on water; among the early preoccupations of the new emperor is the restoration of water distribution in the city. For this he calls on a for­mer collaborator and experienced administrator: Sextus Julius Frontinus, 61 years old, former governor of Britain, i. e. Great Britain, and former preconsul of Asia. In 97 Nerva names Frontinus as curator aquarum, translated as “high commissioner for water” of the city of Rome. Frontinus is a methodical man who takes his responsibili­ties seriously. At the beginning of his service, Frontinus prepares a written description of the aqueducts of the city for his own information and then continues with written descriptions of his observations, measurements, and actions during his tenure. This trea­tise has come down to us, and we have already cited several extracts. We can learn much about both the characteristics and the history of these aqueducts from Frontinus’ writings (Table 6.2).

The principal aqueducts at the time of Frontinus came from the region of the Sabine mountains; the four largest ones, Anio Vetus, Marcia, Claudia and Anio Novus, came from the valley of the Anio, upstream of Tivoli (the Roman Tibur) (see Figure 6.8). Where this river flows out onto the plain of Rome, at Tivoli, the aqueducts leave the val­ley of the Anio and trace a large arc in the direction of the Alban mountains, thus main­taining their elevation as they near the city of Rome. On the Frascati heights, they are rejoined by the Tepula and Julia aqueducts, and complete their trajectory to the city on arches. At their entrance into Rome, the Anio Novus flows above the Claudia, on the same arches (Figure 6.9). Similarly, the canal of the Julia is above those of the Tepula and the Marcia, all three superposed on the same arches. These five aqueducts flow on a slope of about 1.3 m/km in their final reaches.[229]

What surely motivated Nerva to put things in order was that the amount of legally distributed water was far less what it was expected to be, as dictated by the written impe­rial records. The new curator aquaraum therefore took it upon himself to make dis­charge measurements in the aqueducts, both at their origins and at their points of arrival in the city. To his surprise, he found that now the measured discharges were well above those specified in the official registers. There could only be one conclusion: there had been fraud, greater fraud than had originally been suspected!

Here we must pause to note the significance of discharge for the Romans. Not hav­ing understood or recognized the physical concept of water velocity, the Romans found it sufficient to measure the flow cross-section at spots where the water velocity was

judged to be neither excessive nor too weak. The Roman unit of discharge, the quinar – ia, is actually just a measure of area (4.2 cm2), whereas everyone knows today that vol­umetric discharge is the product of the velocity and the cross-sectional area of the canal. What Frontinus really measured is the depth of water in well-defined locations in the canal, from which he deduces the area and therefore calculates his estimate of discharge in quinariae. The discharges shown in table 6.2 are based on a fixed relation between quinariae and real discharges. We will discuss this further at the end of this chapter, regarding the hydraulic knowledge of the Romans. As imprecise as it was, this system nevertheless enabled Frontinus to detect the important deficits and differences described above:

“I do not doubt that many will be surprised that according to our gaugings, the quantity of water was found to be much greater than that given in the imperial records. The reason for this is to be found in the blunders of those who carelessly computed each of these waters at the outset. Moreover, I am prevented from believing that it was from fear of droughts in the summer that they deviated so far from the truth, for the reason that I myself made my gaugings in the month of July, and found the above-recorded supply of each one remaining constant throughout the entire remainder of the summer. But whatever the reason may be, it has any rate been discovered that 10,000 quinariae were intercepted, while the amounts granted by the sovereign are limited to the quantities set down in the records.

“Another variance consists in this, that one measure is used at the intake, another, consid­erably smaller, at the settling-reservoir, and the smallest at the point of distribution. The cause of this is the dishonesty of the water-men, whom we have detected diverting water from the public conduits for private use. But a large number of landed proprietors also, past whose fields the aqueducts run, tap the conduits; whence it comes that the public water­courses are actually brought to a standstill by private citizens, just to water their gar­dens.”[231]

This observation is rather severe. Numerous wildcat taps provide a clandestine water supply to those living adjacent to the aqueducts, and support a parallel, unofficial water market. Nearly half of the water delivered to Rome by the aqueducts escapes the official accounting. But what also upsets Frontinus is that water from the different aque­ducts is senselessly mixed together without taking into account the quality of the differ­ent sources. The last aqueduct built under Claudius, (predecessor of Nero), the Anio Novus, serves to complete the aqueduct system, and its additional supply was useful in obscuring the pilfering:

“The two Anios (aqueducts) are less limpid, for they are drawn from a river, and are often muddy even in good weather, because the Anio, although flowing from a lake whose waters are very pure, is nevertheless made turbid by carrying away portions of its loose crumbling banks, before it enters the conduits (….)

“One of the Anio (aqueducts), namely Old Anio (Anio Vetus), running at a lower level than most of the others, keeps this pollution to itself. But New Anio (Anio Novus) contaminated all the others, because, coming from a higher altitude and flowing very abundantly, it helps to make up the shortage of the others; but by the unskillfulness of the water-men, who divert­ed into the other conduits oftener than there was any need of an augmented supply, it spoiled also the waters of those aqueducts that had a plentiful supply, especially Claudia, which, after flowing in its own conduit for many milles, finally at Rome, as a result of its mixture with Anio, lost till recently its own qualities. And so far was New Anio from being an advan­tage to the waters it supplemented that many of these were then called upon improperly through the heedlessness of those who allotted the waters. We have found even Marcia, so

charming in its brilliancy and coldness, serving baths, fullers, and even purposes too vile to „75

mention.

Frontinus devotes the three years of his mission to reducing these clandestine diver­sions, and improving the distribution system, both quantitatively and qualitatively. He restores respect for the quality of consumable water, and moves the intake of the Anio Novus further upstream to capture clearer water. He also improves the reliability of sup­ply to the public fountains, supplying each with two outlets so as to assure continuous supply when work is being done on one of the aqueducts.

25 Ibid., 90-91.

Figure 6.9 The Agua Claudia and the Aqua Anio Novus on the same arches, at the Porta Maggiore of Rome. The aqueduct was integrated into the new city wall under the emperor Aurelius, in the 4th century AD (photo by the author).

26 The discharges are those of table 6.2, after Frontinus. The demographic data are taken from Christol andNauny (1997).

Whatever may have been these problems in managing the system, they did not reflect any fragility in the supply of water to the city. Figure 6.10 shows how the increase in discharge delivered to Rome accompanies the demographic growth. With nearly a cubic meter of water per person per day, a Roman inhabitant had nearly a hundred times more water than his or her counterpart in Paris of the 19th century!

Later, two additional aqueducts are constructed, but they are less well known to us since they came after the writings of Frontinus. They are the Trajana, constructed around 110 AD under the grand Emperor Trajan, successor of Nerva; and the Alexandrina, built under Alexander Severus in 226 AD. From the 2nd century AD, Rome falls into a kind of recession, and its population probably stabilizes. This second century is the century of the provinces, insofar as expansion is concerned. The construc­tion of grand aqueducts begins again in these provinces, in particular in Gaul, Spain, and Africa.

The Cretan, Greek, and later the Hellenistic aqueducts primarily use terra-cotta pipes. The Romans, following the Etruscan heritage, build their aqueducts as masonry canals, usual­ly rectangular in section, and covered over by a vault or stone slabs. The aqueducts are fit­ted with openings at regular intervals (from tens to hundreds of meters apart) to facilitate inspection and maintenance. We have seen earlier that the Aqua Appia, the first Roman aqueduct built in 312 BC, is nearly entirely underground. This practice is surely inherited from the Etruscan techniques, but also has a military dimension, since Italy is still far from being pacified at this time – this is only eleven years after the death of Alexander.

The Romans stay with this concept throughout the long period of aqueduct construc­tion in the Empire, i. e. from the 4th century BC to the 3rd century AD. A canal (specus) is, depending on the local topography, sometimes laid on the surface or on a supporting wall, or sometimes buried underground, occasionally passing through true tunnels, and, when necessary, supported on arched structures. The longitudinal profile of an aqueduct is driven by two constraints: first to maintain sufficient slope to convey the water, as regular as possible to avoid local break points; and second to deliver water at a high enough elevation to supply the city’s water tower (castellum) and thus enable distribu­tion of water to the highest areas of the city. Thus the arch structures that one often sees near the cities are needed to keep the canal at a sufficiently high elevation. And it is of course these arch structures, still visible today in numerous locations of the Roman Empire, that one commonly associates with the Roman aqueducts. However one should not forget that these structures represent only a small fraction of the total length of an aqueduct. The slopes of the canals are quite variable, as we will see in examples pre­sented further on, but most often are of the order of one meter per kilometer. Occasionally, where there is a need to drop to a lower elevation, there are local chutes comprising short sections of steep slope, or even true cascades.

Let’s once again listen to Vitruvius:

“Water is conducted in three ways, either in streams by means of channels built to convey it, in leaden pipes or in earthen (terra cotta) tubes (that is, according to the Greek process cited earlier), according to the following rules. If in channels, the structure must be as solid as pos­sible, and the bed of the channel must have a fall of not less than half a foot to a length of one hundred.[222] These channels are arched over at top, that the sun may strike on the water as lit­tle as possible.

“(….) If hills intervene between the city walls and the spring head, tunnels under ground must be made preserving the fall above assigned; if the ground cut through be sandstone or stone, the channel may be cut therein, but if the soil be earth or gravel, side walls must be built, and an arch turned over, and through this the water may be conducted.”[223]

A masonry canal is not inherently watertight. Therefore it is necessary to plaster the useful (or “wetted”) walls of the canal, both to minimize leakage and to protect the masonry walls themselves from infiltration damage. The Roman plaster (opus signinum) is of very high quality. It is a mortar of crushed tile solidified with thick lime, also con­taining crushed brick and pottery shards, and sometimes other additives.[224]

The most imposing and important structures are those necessary to carry the aque­ducts across valleys. These are the famous bridge-aqueducts such as the Pont du Gard near Nimes (47.8 m high), or the bridges of Segovia and Tarragona in Spain or, similar­ly, the bridge of Chabet Ilelouine on the Cherchell aqueduct in Algeria, all three some thirty meters high. There are also the inverted siphons echoing the technology of the Hellenistic world. These expensive siphons are reserved for valleys more than 50 m deep, a depth beyond which the cost of a bridge becomes prohibitive. Further on we describe the siphons of Lyon, the largest ones in the Roman world, but it is first useful to provide a glimpse of this Roman technology.

Roman inverted siphons begin with a head tank, receiving water from the aqueduct’s canal, then distributing the water into one or more parallel lead pipes leading out of the basin. These pipes descend to the bottom of the valley and cross it on a bridge-siphon, and then come back up on the other side to an exit, or escape, basin. This exit basin then delivers water into the continuation of the aqueduct (Figure 6.7). Vitruvius emphasizes the need to provide a rather long and rectilinear length of siphon at the low point of the valley (the “venter”):

“(…) when it arrives at the bottom, let it be carried level by means of a low substruction as great a distance as possible; this is the part called the venter, by the Greeks coelia. When it arrives at the opposite acclivity, the water therein being but slightly swelled on account of the length of the venter, it may be directed upwards.

“If the venter were not made use of in valleys, nor the level substruction, but instead of that the aqueduct were brought to an elbow, the water would burst and destroy the joints of the

pipes. Over the venter long stand pipes (collivaria) should be placed, by means of which, the

violence of the air may escape. Thus, those who have to conduct water through leaden pipes,

may by these rules, excellently regulate its descent, its circuit, the venter, and the compres – 1 8

sion of the air.”

One can appreciate the value of a rectilinear “venter” (usually implemented as a bridge-siphon) in reducing the hydrodynamic force that would be concentrated on a small-radius elbow at the valley floor. But the reasons given by Vitruvius (i. e. to avoid the excessive “swelling” of the water) somewhat miss the mark. There has been consid­erable questioning of what is meant by the colliviaria. Some have interpreted them as purges to eliminate air. According to Henning Fahlbusch,[225] [226] [227] they are towers supporting an elevated intermediate reservoir, acting like a modern surge tank to purge air from the siphon’s “venter”. Such towers are useful when, in an inverted siphon, there are elbows or high points favorable to the formation of air pockets. The Tourillon de Craponne, on one of the aqueducts of Lyon that we describe further on, could serve as an example. Possible remains of others can be found at Aspendos, in the south of Anatolia.

Table 6.1 presents a synthesis of data for a number of known aqueducts in the Roman Empire. It shows that the aqueducts could reach quite significant lengths, often up to 50 or 100 km. But the length of an aqueduct is not, by itself, a good indicator of its grandeur. The construction of bridges, arcades, tunnels, etc. in effect reduces the length of an aqueduct, limiting the numerous detours that would be necessary if the structure had to follow the local topography along the ground surface. Certain aqueducts are actu­ally shortened during renovation, thanks to the construction of such projects. This is notably the case for the Aqua Marcia, one of the aqueducts of Rome:

“But now, whenever a conduit has succumbed to old age, it is the practice to carry it in cer­tain parts on substructures or on arches, in order to save length, abandoning the subterranean 20

loops in the valleys.”