During work for the construction of Rome’s new structure for trade fairs in the locality of Ponte Galeria, preliminary investigations undertaken by the Archaeological Superintendency of Ostia uncovered the ancient Via Portuense, for an overall length of 1100 m. This road, which linked Rome with _Portus_ (modern Fumicino), was constructed on a viaduct due to the impracticability of the valley floor. The building technique and the stamps on the bricks in the arch lintels suggest that the road was built sometime between the end of the 1st century A.D. and the early 2nd century A.D. Within the fluvial sediment covering the structure there were areas of calcareous incrustation, interpreted as traces of the emission of hydrothermal waters from the Tiber valley bottom. In one case, the removal of the surface layer resulted in an emission of gas and water. From stratigraphic and archaeological data it is known that this tract of the valley bottom was already characterised, before the construction of the Roman road, by a series of slight depressions, each one containing one or more thermal pools of water, both active and extinct. These particular environmental conditions seem to have conditioned the building of this tract of the Via Portuense, and to have specifically led to its passing over numerous bridges with one or two arches. The distribution of the thermal pools in relation to the road suggests activity, probably discontinuous, from before the Roman period to the present day. Four different levels of incrustation were identified: the first cut by the foundations of the Roman road; at least another two alternating with levels relating to phases in which the road was damaged; a fourth, the latest, post dating the road’s abandonment following the flooding of the Tiber in the 16th century. The radiometric dating undertaken with the U/Th method on the incrustations below the Roman road gave a result of between 900 and 400 B.C. and on the latest level, contemporary with the roads use, gave a period between 750 – 1300 A.D. Overall the data attests that the hydrothermal activity, documented in the Ponte Galeris area for a period of at least 2,500 years and still active, relates to the final activity of the volcano of the Alban hills, which is still effected by an intense out put of gas, both localised and diffused, thus supplementing the historical record available to date. (Paola Tuccimei, Michele Soligo, Antonia Arnoldus-Huyzendveld, Cinzia Morelli, Andrea Carbonara, Marilena Tedeschi, Guido Giordano)