The Bozburun ("gray cape") site lies at the base of a cliff, called Kucuven Burnu, on a sloping, sandy bottom. The visible portion of the site is a mound of amphoras approximately 20 meters long and 8 meters wide, although there are many artifacts lying in the rocks of the cliff face. The upper end of the primary mound, hard against the base of the cliff, lies at a depth of 26 meters. The last visible amphoras at the far end lie at approximately 34 meters, although it is apparent that material continues under the sand. The lowest datum points currently established lie at 36 meters.

An example of a 'pyriform'  Class 1 amphora. (slide# BK-98.16) Photo: INA.

The amphoras appear to be of two major classes common toward the end of the first millennium AD. The dominant class, are often referred to as "pyriform" (pear-shaped) or "ovoid" (egg-shaped) and are characterized by a tapering, ribbed body with rounded shoulders and round bottom without a foot; the neck is relatively tall and narrow, with a heavy, distinct rim, and the wide handles rise from the upper shoulder to just below the rim. The best parallels found so far are dated to the second half of the ninth century and come from kiln sites in the Crimea. The second class, noticeably shorter, with a wide, flaring neck and flat bottom with central dimple or kick, may be from the Sea of Marmara region (see the section on amphoras for more details).

In addition to the whole and broken amphoras that make up the greatest part of visible material, there are several concretions visible, and a collection of small bricks and building stones in the upper part of the site. Based on similar finds at Yassi Ada, it is currently thought that this may be the remains of the hearth. Anchors, of cruciform type, have been found on the rocks at the top of the wreck and mixed with the tumbled amphoras at the bottom. Based on the finds to date, it is expected that the upper end of the site is the stern.

The site has long been known to Turkish sponge divers and was shown to George Bass during INA's first survey in 1973 by a diver from Bozburun, Mehmet Askin. The site was surveyed in greater detail in 1982, when test pits revealed well-preserved (for Mediterranean waters) hull remains beneath the sand in two places. On the 1982 survey, another local sponge diver was interviewed, and he reported that the site had been known to his colleagues for some time, and perhaps as many as 50 whole jars had been removed over the years for use as water jugs on sponge boats. The site was also visited in 1992, to check out reports that the site was being looted (there is no obvious evidence of disturbance, but only a comparison with photos taken in 1982 will reveal whether significant damage was done), and in 1994 for the purpose of planning the excavation. A total of three amphoras, all of the first, pyriform class, were raised by INA before excavation began in 1995. A few amphoras in the Bodrum museum, donated by sponge divers, may also be from this wreck.

Looking up slope, toward the stern. (slide# BK-50.08) Photo: INA.

During the first season of excavation, initial topographical mapping of the site was completed, along with the establishment of primary datum points and a grid of 2-meter squares. Some excavation was carried out in a few of the squares that had visible remains, but in no place was the bottom of the deposit reached. It is clear that over some of the site, the broken and scattered visible amphoras lay atop a layer of amphoras still stacked as they were in the ship. The upper layers also appear to have slid and rolled down slope somewhat as well, since the squares excavated at the bottom of the mound revealed only tumbled amphoras lying on sand. A large quantity of broken sherds were recovered (most were redeposited on site, in square M9), as well as about forty whole or nearly whole amphoras, a number of partial amphoras, and a small number of non-amphora finds, such as a pitcher, several bricks and tiles, and a few lead fishing weights (the last probably intrusive).

The small hull section that was exposed in 1996. (slide# BK-74.06) Photo: INA.

In 1996, during a long and productive season, work concentrated on the upper half of the site, although Dave Stewart and Korhan Bircan continued their work on two lower squares. In addition to lots of amphoras, including examples of a possible third class, small finds included galley pottery and a glass goblet were recovered. Removal of much of the broken material from the upper end revealed a large area of amphoras still stacked in the bottom of the hold, running from square F11 down to H11. H11 also contained the first coherent hull remains we encountered. Although only a small area was exposed, this included the keel, four frames, four bottom strakes, a stringer and ceiling. The keel and planking are of oak, the remainder of pine. Dendrochronological analysis of two of the frames was inconclusive.

Three examples of galley ware from the stern area. (slide# BK-95.32) Photo: INA.

1997 was the year of the amphora. Primary excavation goals for the season were recovery of as much of the cargo as possible and further investigation of the hull remains. Nearly 500 amphoras were removed from the site, although not all were brought to the surface (approximately 175 were stored on the bottom, off site). Divers looking at the lower half of the site can largely thank Robin Piercy, David Stewart, and Bill Charlton for the moonscape. At the upper end, the after three meters of the hull remains were exposed and some of the timber recovered. More frames were exposed in unit J11, and a concentration of tiles and stones first identified in 1995 was largely mapped and recovered. This area also yielded a collection of cooking pots, as well as other domestic objects. A mate to the glass goblet found in 1996 was recovered in the stern, as was a fine glass flask that may go with the goblets.