In 2006 a rural complex was partially excavated in contrada Verdura di Sciacca (Agrigento), situated on the south-western coast of Sicily, close to the mouth of the river Verdura.

The site was linked to one of the main routes crossing Sicily in the late antique period, that which from Lilybaeum led to Agrigentum, in the proximity of the important statio of Aquae Labodes, and close to the mansio of Allava. The settlement was preserved over an area of about one hectare. In fact, a wide strip of terrain had collapsed into the sea due to the erosion of the cliff. Three buildings were identified and one, denominated alpha, was excavated. This structure was an elongated rectangle in plan, comprising at least nine rooms arranged in two rows. The walls had dry-stone built footings with a double facing of cobbles and mud brick standing structures. Only two of the rooms had tiled roofs.

It is only possible to suggest the function of some of the rooms. In fact, a kitchen-pantry and a living-dining room were identified, together with others destined for rural activities.

Among the finds there was a large quantity of African material, amphorae, ARS, cooking and coarse wares. On the basis of the clay fabric diverse coarse ware forms, such as jugs, mortars and basins can be attributed to ateliers in the region of Hammamet-Nabeul. The majority of the African amphorae from the site were also produced in this region. In particular two stubs from cylindrical amphorae type Keay XXXVa and b, the lower part of an amphora type Keay XXXVIII and a number of spatheion stubs can be attributed to an atelier in the city of Nabeul, that of Sidi Zarhuni, where mortars and some ARS forms, including the Hayes 50B, from the site were also produced. It seems important to note that the same type of African materials are among the cargo of the Dramont E wreck. This in fact indicates that the south-western coast of Sicily was inserted into the commercial routes sailed by ships like that wrecked off Cap Dramont at Saint-Raphael, whose cargo is considered to have come mainly from the atelier of Sidi Zahruni.

The data presented suggests that research should continue on the contacts between the region of the Gulf of Hammamet and the south-western coast of Sicily, where the site of Verdura was probably a stopping place for small boats which, departing form larger ports ( Lilibaeum, Aquae Labodes, Agrigentum ) used a series of small harbours. This hypothesis is also supported by the recent discovery of other sites of Roman date on the coast, west of Verdura, including the site in contrada Carabollace in the territory of Sciacca, also situated at a river mouth by the sea.