Two distinct clusters of artifacts are seen in the photomosaic of Skerki D. The forward cluster includes a hand-rotated stone quern and a large roof tile, suggesting the location of the galley. A gap of 3.6 m separates this heap from a second larger cluster of artifacts at the aft end of the wreck site. The discharge pipes of the bilge pump are visible at the forward end of this group. Why there are two groups of amphoras remains a question. Did the hull fracture across the middle as it sank or were there two separate holds? Or perhaps a cargo had just been unloaded or the cargo in the central area was biodegradable, such as woven goods, grains or hides.

The amphoras are surprisingly varied and include at least 13 different forms originating in Italy, North Africa and Greece. These are mainly wine jars but also some for oil and garum, the prized fish sauce of the ancient Romans made from the guts of fish. Garum was used as both a popular condiment as well as for medicinal needs and brought high prices. Concentrated in nature, only a few garum jars are usually found with any shipment of wine.

By far the most numerous amphora shape on Wreck D is that of Dressel Form IB (Will Type 4b), a popular wine amphora dated from about 80 to 30 B.C. (Fig.4) 8 27 of these amphoras can be identified in the photomosaic and two were lifted.9 Petrological analysis of these by D.F. Williams at the University of Southampton identifies the clay with an amphora kiln in Albinia, next to the Roman port of Cosa10 on the Tyrrhenian coast of Italy in ancient Etruria. Cosa is the earliest Roman port thus far known dating back to 273 B.C. Since these tall necked Dressel IB amphoras are by far the most numerous shape in the cargo, it is reasonable to suggest that Wreck D was loaded at Cosa on its way south. These Dressel IB amphoras held about 26 liters of wine. Their long handles and solid spiked toes made them easy to be lifted onto the shoulder of a stevedore as well as efficiently stacked in the hold of a ship.

An easy day's sail from Rome, the port of Cosa enjoyed the protection of one of the few promontories along this predominantly sandy coastline. The limestone promontory is strongly fractured and in antiquity its long channels served as conduits for gushing fresh water springs that served both the port and a rich, brackish fishing lagoon just behind it. Providing the only source of fresh water in the area, the port with its rich fishing lagoon, was undoubtedly the reason for the founding of the colony whose fortified hill town lies above.

Now the fishing lagoon is silted over and used as pasture land. But in Roman times a remarkable commercial complex existed at Cosa including: the harbor with concrete piers, a lighthouse, a commercial fish farm, factories for making amphoras, wine and fish products along with machinery for lifting fresh water from a gushing spring – all protected by a temple to Neptune on the hill above. Controlled and probably financed by the famous Sestius family who had an estate in the area and also probably owned ships, their wine and garum jars dominated trade in the western Mediterranean in the last quarter of the second and early first centuries B.C. The Sestii were one of the earliest exporters of Italian wine in the western Mediterranean and their enterprise goes back at least to the latter third century B.C.11 Skerki Wreck D provides startling new evidence for the importance of the port of Cosa and the export trade in wine and fish sauce by the Sestius family.

Page 1 2 3 4 5 6 7


8. Will in McCann 1987, pp.182-201.

9. SK97.066 and .067.

10. McCann 1987; McCann 1988.

11. The earliest known Sestius stamp was also found at Skerki Bank on a Graeco-Italic amphora (Will type 1d) from a different wreck site. This amphora can be dated in the early second century B.C. See McCann and Freed 1994, MJ89-29, p. 67, figs 46, 47 a-b, color fig. 16; A.M. McCann, "Lamps and the Dating of Roman Ports and Ships," in N. Goldman, ed., New Light from Ancient Cosa: Studies in Honor of Cleo Rickman Fitch (N.Y.,2001) pp. 13-34. For further chronology and recent discussion of the Sestius enterprise see E.L. Will, "Defining the Regna Vini of the Sestii", in Goldman, ibid., pp. 35-47.