A. Trevor Hodge
At all periods of Antiquity, wells were a prime, if not the prime resource. Aqueducts, which we often think of as an essential feature of any ancient city—certainly of any Roman one—were often very late in arriving on the scene, for it was usually the opening of large bath complexes, voracious consumers of water, that spurred their construction. Previously, all water supplies had to come from cisterns and wells, and even after the advent of the aqueduct these would remain in service. Indeed, in most cities it is a good question whether one should think of domestic wells as supplementing the aqueducts or the other way round. And of course some cities, such as Ampurias and London, never had an aqueduct at all, and relied on cisterns and well-water all through their history.
In sinking a well the first step was to determine a likely spot, where it would tap a good supply of water. In domestic wells this was not usually a problem, for there was simply no choice: the well had to be sunk somewhere within the limits of the house or the plot of land it occupied, which, in a city, would be quite small. Presumably one knew from neighbours' wells how far down the water table was liable to be, and if it seemed too far one would construct a cistern instead. Indeed, variations in the water table level sometimes led to variations in technique. In Athens at first the residents used wells. Then, in the fourth century B.C., the water level fell out of reach, causing a general switch to cisterns. Next, in the third century B.C., the water table came back up again and the city went back to wells, some of which were actually dug through the floor of the now-disused cisterns.2
Outside of this restrictive urban context, if one had enough land to offer a choice, the first step was to decide on the best spot to dig. This was done by consulting an aquilex, a water-diviner; the best
1 Early use of wells: Paus. I 14,1 (Athens); Thuc. II 48 (Peiraeus); Fron. Aq. I 4
2 Hodge 1992, 60; Miller 1974, 194–9, 228–9.