Savaria - A Vas Megyei Múzeumok értesítője 30. (2006) (Szombathely, 2007)

Régészet - Anderkó Krisztián: Savaria vízvezetéke

Savaria a Vas megyei Múzeumok Értesítője, 30 (2006) THE SAVARIAN AQUEDUCT The starting point of the water channel was the Bozsoki Valley, where the springs were connected to the channel adapted to the topographic conditions and worked by the principle of gravity feed. Precipitation supplied these water streams of natural origin that has a transit time less than a year both in the surficial and subsurface gathering grounds (PACHNER 1975: 3). The quality of such water gathered from precipitation that came from basalt rocks was excellent as Vitruvius also indicated (VITRUVIUS 8. 1. 2.), therefore this spot was perfectly capable to have been the starting point of an aquaeductus. Natural galleries are the spots where subsurface natural water percolating through the rocks is gathered, which rises to the surface afterwards. If the rectangular structure recovered in 1975 (Fig. 2: no. 5) has Roman origins, then ideally a catchments of spring made by a technical solution having been similar to the one in Köln has been recovered. The measures of the channel in the Austrian zone enlarge similarly to other known aqueductes and as a result, its covering also alters. The size increase primarily correlates to the cleanability of the channel channel and besides statics reasons; the vault applied in lieu of the plain covering increases the penetrability of the interior. At the same time, to be noted that this sort of alteration occurs right before the point the channel would cross Creek Rohonci, although it cannot be confirmed that the creek itself would have been connected to the network. Despite their short distances, the channel sections subject to research, to say, the one in Sé in 1998 and the second one in Bucsu in 2003, were not arranged in one single line. This is due to the fact that the design and implementation processes must have considered the actual topographical conditions, to which the optimal gradient as well as the water delivery capability of the channel working by the principle of gravity feed must have been adapted. Besides to arrange proper gradient, the frostline and the distance between the drain and the surface also required utmost care. Scarcely can the sections recovered in Sé in 1998 and in 2001 be stated as connected ones even if the above is taken into account. The channel sections recovered when Routes U2 and U3 were subject to research (1998) that follow the topographical particularities are more or less arranged in one single line in the direction from the west to east (Fig. 11:24—25). The sections recovered in the excavations on the lots located between Routes U3 and U2, however, stretch from the south to the north (Fig. 11:26). Although, it points at the channel section having been oriented in the opposite direction, the channel recovered in U3 lies in the test pits dug between the two rescue sites. Perhaps, the sondage missed the channel, which hada room of a few meters only from the U3 point to turn sharply to south and by having hada large loop to connect to the section oriented south to north that was recovered in 2001. On the other hand, this solution is hard to accept owing to the negligible distance. The most probable notion is that it is a service connecting channel, which can be confirmed by that there may have been a natural water stream along with the channel (the section stretching from south to north), which was marked in the cadastral map of 1965. Moreover, the bed of such water stream was also recovered during the excavations, thus it is conceivable that this creek was also connected to the network. The section of the water channel recovered in U3 poses further questions. The terrazzo keeping straightforward within the interior of the channel narrows clearance and furthermore, a layer of bricks and breakstone as thick as 15 cms situated between the substructure and the aquifer layer was observed 41