versión española


The usual word in Latin to name an aqueduct is aqua, following by the name of the man who ordered its building; so in Rome we know the aqua Marcia, the aqua Appia, the aqua Claudia, etc.

The building of an aqueduct needed a not too high point –to avoid the interruptions of the supply in summer or in drought- and a not too low point –to avoid the stagnation due to the mud dragged down by the flow-; in addition, if the water came from a dam and after the water entered in a city, it was necessary that water settled previously in a tank (in Latin castellum).  Finally, the engineers had to calculate, first of all, that pitch was always enough to allow that water flowed and did not stagnate and, on the other hand, they had to calculate that the water pressure was enough to be supported by the aqueduct without overflowing.

When the Romans needed to supply water to their towns, from a dam or through a deviation of water from a river, they built covered channels that took water across the terrain (these channels are called in Latin specus); the Romans used lead pipes or tubes in the towns to take water to a particular use or to cross a valley with a siphon.

Sometimes, these channels were dug out in the rocks of the mountains and hill.  In the normal building of this kind of tunnel-channel the diggers progressed in the interior and extracted the rubbish at the same time; so the galleries must have been enough high and wide: in the aqua Claudia the gallery reached 2,3 metres wide.  Also it was frequent to build an elevated gallery; from this the diggers drug out wells to have different accesses to a level; in that way, the works progressed quickly, because the channel could be drug out by some groups of diggers in different sections at the same time, using the wells as tunnel to evacuate the rubbish.  The drug gallery used to be recovered with rough stone, in spite of the impermeability and hardness of the rock, to reinforce the channel and, above all, to control its pitch.  This is the case of the aqueduct between Albarracín and Cella.




Some other times, it was necessary to build the specus completely with opus caementicium, i. e., to make an artificial channel and to cover it with flagstones to protect the water; so, the channel, in some way, became buried.

Finally, to build a simple wall –substructio- or some arches –opus arcuatum- as supporters was enough to elevate the specus in order the water to flow without any problem.

However, when the orography of the terrain didn’t allow the previous options, it was necessary to build aqueducts (Malissard said “together with the amphitheatres, the aqueduct are, of course, the most representative monuments of the power and the permanence of Rome.  Impressive for their number, their height and the apparent hardness of their arches that still stand up straight under all the skies, they mean the showiest and most spectacular part of a whole still more gigantic ” –the whole is all the hydraulic buildings-).


Specus cubierto junto al castellum de Nîmes

(Photo: Roberto Lérida Lafarga 14/8/2007)




The Roman engineers had as challenge not only to cross a depression in the terrain and to make an elevated channel that allowed to divert water, but to keep along all the course of the channel the enough and precise pitch to the advance of the water; if the pitch was excessive, they run a risk:  the water could wear down and drag down the building; but if the pitch was very weak, they ran other risk:  the water could stagnate or facilitate the deposit of mud that obstruct the outlet of water.  So, in the construction of an aqueduct and in the flow of the water through a specus, the Roman engineers used in general the physical principle of the gravity of the water and not the principle of the water pressure.  With this principle they adapted the aqueduct and the channel to the nature of the terrain as far as possible.

So, the pitch of an aqueduct is different from the pitch of other aqueducts and even in the same aqueduct a section could have a different pitch from other sections.  If water flow was very quickly and strong, the Roman engineers made curves or waviness that restrained the water or they made small springs of water or deviations to tanks that reduced the strength of the water.


Arches of the aqueducts Aqua Claudia and Aqua Anio Novus in the southeast of Rome (Photo from CONNOLLY, P. y DODGE, H., La Ciudad Antigua.  La vida en la Atenas y Roma clásicas, Madrid, 1998)




The building of an aqueduct-bridge had another difficulty: the bulging, i. e., a progressive lateral deformation of the pilasters that supported an aqueduct; so, in general, the pilasters were reinforced with braces or tyrants or crosspieces or they were  made lower or they superimposed arches so that they made aqueducts with two levels, like the aqueducts of Segovia or Tarraco (now Tarragona), or with three levels, like the aqua Claudia in Rome or the Pont du Gard in France (48,67 metres high).


Section of the arches and channels of the aqueducts Aqua Claudia and Aqua Anio Novus (according to CONNOLLY, P. y DODGE, H., La Ciudad Antigua.  La vida en la Atenas y Roma clásicas, Madrid, 1998)




First of all the specus was in the open air, but soon the Roman engineers covered it; this covering, with small displaceable flagstones, prevented the channel from being blocked by residues nor too much water vaporized in its flow nor animals drank nor people got out water; in the other hand, when the water did not flow due to the accumulation of residues, it was possible to clean the specus, emptying it, lifting up some flagstone and extracting the residues.  The Roman engineers achieved the hermeticism of the specus and the avoiding of water escapes with the opus signinum, i. e., a red mortar that coated the specus’ bottom and walls with some layers every time thiner.


Arches of the aqueducts Aqua Claudia and Aqua Anio Novus in Fosso della Noce (Reconstruction of CONNOLLY, P. y DODGE, H., La Ciudad Antigua.  La vida en la Atenas y Roma clásicas, Madrid, 1998)


In general a specus reached enough height that a man could walk inside it, but the dimensions changed and depended on the water flow.  In the case of the Pont du Gard, between Avignon and Nîmes, in the French Provence, the specus reached 1,80 metres high and 1,30 metres wide.


Specus of the Pont du Gard

(Photo: Roberto Lérida Lafarga 7/8/2007)




Size of the specus of the Pont du Gard

(Photo: Roberto Lérida Lafarga 7/8/2007)


Specus drug out in the rock; the water passed from here to the specus of the aqueduct of the Pont du Gard

(Photo: Roberto Lérida Lafarga 7/8/2007)




At the end of the IV century b. C., Rome only had water from wells, springs and rain; about the year 312 b. C. the first aqueduct was built, the aqua Appia (so called because the censor Appius Claudius ordered its construction); finally, Rome had 13 aqueducts (10 big aqueducts and 3 minor aqueducts); now we can see the remains of some of them.  The most famous was the aqua Virgo, but we know the name of some of them, like aqua Appia, aqua Anio Vetus, aqua Marcia, aqua Repula, aqua Julia, aqua Alsietina, aqua Claudia, aqua Anio Novus, aqua Traiana, aqua Alexandrina, etc.  If we have quite a lot of information about the water in Rome, it is thanks to Frontinus.  In the age of the emperor Augustus, Agrippa directed the cura aquarum –“the care of the water”, i. e., a department that took care of the maintenance and function of the hydraulic infrastructure in Rome; in the age of the emperor Nerva, Frontinus took care of this charge and wrote a book, De aquis urbis RomaeAbout the water of the city of Rome- where he detailed historical aspects and descriptions of the aqueducts, the supply of water, the problems in maintenance, legal aspects and even affairs of political corruption.

Aqua Claudia, the aqueduct of Claudio in Roma, near the Colosseo

(Photo: Roberto Lérida Lafarga 28/12/2004)


The Aqua Claudia in Roma, near the Colosseo (Photo de G. Rinaldi, from de GABUCCI, Ada: Guía a la Roma antigua, Roma, 2000)




However, probably the most famous and most visited by tourists is the Pont du Gard, in the French Provence, due to its size and its state of preservation.  The aqueduct-bridge over the valley of the Gardon river is 275 metres long and the complete specus that flowed from a spring near the modern Uzès to Nemausus (now Nîmes) reached 50 kilometres.  The researches have calculated that through this aqueduct 20.000.000 litres flowed dialy, supported by a building with up to 6 tons weight ashlars.



Aqueduct-bridge of the Pont du Gard (Francia)

(Photo: Roberto Lérida Lafarga 7/8/2007)







- MALISSARD, Alain: Los romanos y el agua: La cultura del agua en la Roma antigua, Barcelona, 1996

- BELTRÁN LLORIS, Miguel: “El agua profana en la cuenca media del valle del Ebro:  AQUA DUCTA.  La captación del agua, presas, embalses, conducciones”, en AA. VV.: Aquaria: Agua, territorio y paisaje en Aragón, Zaragoza, 2006

- CONNOLLY, P. y DODGE, H., La Ciudad Antigua.  La vida en la Atenas y Roma clásicas, Madrid, 1998