Treasure of the Dokos Shipwreck — One of the World’s Oldest Known Wrecks
Pottery, not gold. The Dokos wreck, found by Peter Throckmorton in 1975 off the Greek island of Dokos in the Argolic Gulf, dates to the Early Bronze Age, roughly 2700–2200 BC. Excavated by the Hellenic Institute of Marine Archaeology in 1989–92, it yielded over 500 clay vases, more than 15,000 sherds, lead ingots, millstones and two stone anchors. It is among the oldest shipwrecks known.
The ship at Dokos left almost no timber, only a mound of pots spread over a rocky slope. That mound is roughly 4,000 years older than the Spanish galleons that define most treasure stories, and for half a century it was widely described as the oldest shipwreck ever found. The “treasure” is not gold or silver. It is the cargo of an Early Bronze Age trading boat, frozen at the moment it went down, and it is one of the few direct pieces of evidence for how goods moved around the Aegean before the age of palaces.
This page looks at the wreck as a cargo: what was on board, what each kind of find tells archaeologists, how the material was recovered and where it is now, and how Dokos compares with the richer and later Bronze Age wrecks found off Turkey. For the island, the dive conditions and the general story of the site, see our companion guide to the Dokos shipwreck.
How old is the Dokos shipwreck?
Early Bronze Age. Published dates range from about 2700–2200 BC for the wider period down to roughly 2250–2050 BC for the wreck itself.
The ship cannot be dated by its hull, because the hull has gone. Its date comes from the pottery. The vessels belong to the Early Helladic II phase of the mainland Greek Bronze Age, which is conventionally placed at about 2650–2200 BC, and the shipwreck is usually described as lying within the broader “second Proto-Helladic” span of 2700–2200 BC. The Hellenic Institute of Marine Archaeology (HIMA), which excavated it, dated the Cycladic-style pottery to around 2200 BC. Other summaries narrow it further: the biography of its discoverer, Peter Throckmorton, gives c. 2250–2050 BC, and a reference entry for Dokos island gives about 2150 BC. Those figures are not really in conflict; they reflect different ways of expressing a date that rests on pottery styles rather than on a laboratory measurement of the ship.
To put that in perspective, Early Helladic II overlaps with the Old Kingdom in Egypt, the era of the great pyramids. The ship sank roughly a thousand years before the Mycenaean palaces at Mycenae and Pylos reached their height, and some 800–900 years before the famous Uluburun ship went down off southern Turkey.
| Date | Event |
|---|---|
| ~2700–2200 BC | Ship sinks off Dokos (pottery date; ~2250–2050 BC in some summaries) |
| Aug 23, 1975 | Cargo mound found by Peter Throckmorton |
| 1989–1992 | Excavated by HIMA under George Papathanasopoulos |
| 2002 | Microscopy study of Early Helladic II wood from the site published |
| July 2026 | Kekova (Turkey) wreck published; now listed by Guinness as the oldest |

Where was the Dokos ship found?
Dokos is a small, rocky island in the Argo-Saronic Gulf, next to Hydra and separated from the Peloponnese by a narrow channel. It covers about 13.5 km², rises to 308 m and had just 18 recorded residents at the 2011 census, mostly monks and shepherds. In antiquity it was called Aperopia. The island lies about 100 km east of Sparta, between the Argolic Gulf and the open Aegean.
The cargo lies on a sloping seabed at roughly 15–30 m (50–100 ft) off the island’s coast; Throckmorton’s own account describes finding pottery at about 20 m. The position matters. Dokos sits on the natural sea route between the Saronic Gulf, the Argolid and the islands to the east, and the bay where the pots were found would have been an obvious place for a small boat to shelter or put in. HIMA also reported traces of two Early Helladic settlements on the island itself, which means the ship may have been heading to, or leaving, a landing place that was in use at the time.
Who found the Dokos wreck?
The wreck was found on August 23, 1975, by Peter Throckmorton, an American photojournalist turned underwater archaeologist who had already changed the field once. In 1958 a Bodrum sponge diver, Kemal Aras, described to him a scatter of metal “biscuits” he had seen off Cape Gelidonya in southern Turkey. Throckmorton recognised it as a Bronze Age wreck, and its excavation in 1960 became the first ancient shipwreck excavated in full on the seabed. He spent years cataloguing wrecks shown to him by sponge divers in Turkey and Greece, and he was also the man who spotted the 1877 iron barque Elissa in a Greek scrapyard, leading to her restoration in Galveston, Texas. He died in 1990, aged 61.
At Dokos, what he saw was not a ship but a dense spread of clay vessels, many still whole, that could only be the remains of a lost cargo. It took another 14 years before a full excavation began.
How was the Dokos cargo excavated?
Between 1989 and 1992 HIMA, led by its president, the archaeologist George Papathanasopoulos, carried out what is described as the first full-scale excavation of an ancient shipwreck in Greek waters. A team of about twenty people worked the site over several seasons. The seabed is irregular and rocky, which makes traditional tape-and-grid recording slow and inaccurate, so the team used SHARPS (Sonic High Accuracy Ranging and Positioning System), an acoustic system that fixes the position of each find by timing sound pulses between transponders. Every pot and sherd could then be plotted in three dimensions before it was lifted.
That care is the real value of the site. A cargo mound with no hull is easy to treat as a pile of loot. Mapped properly, it becomes a record of how the cargo was stowed, which objects were packed together and how the material spread as the ship broke up. The finds were raised and taken to the museum on Spetses, the neighbouring island, for conservation and study.
What was the Dokos ship carrying?
Mainly pottery: more than 500 vessels and over 15,000 sherds and artifacts, with lead ingots, millstones and stone anchors.
The excavation recovered more than 500 clay vessels, many intact, and in total over 15,000 sherds and other artifacts. It is one of the largest single groups of Early Helladic II pottery known from anywhere. The range is wide and very domestic: cups and bowls, jugs, “sauceboats” in seven or eight different shapes and sizes, amphorae, basins, wide-mouthed jars, braziers, baking trays, askoi (small, asymmetric pouring jugs), large pithoi for storage, and everyday kitchenware.
The mixture tells a story. These are not luxury goods for a single elite customer but the kind of tableware, cooking ware and storage jars that any household or small settlement would use. Some of the pottery looks like it was made in the Argolid, the region around Argos and Lerna on the Peloponnese; other pieces are Cycladic in style. The excavators proposed that the ship was a coastal trader carrying household wares from a mainland workshop to small villages around the Gulf of Argos and the Myrtoan Sea, perhaps picking up and dropping off goods along the way.
Why do the Dokos sauceboats matter?
The “sauceboat” is the signature vessel of Early Helladic II: a low, boat-shaped bowl with a long spout at one end and a handle at the other. Nobody knows exactly what it was used for. The modern name is a convenient label, not a statement of function; suggestions include pouring wine or oil, drinking, or ritual libations. What archaeologists do know is that the shape is extremely widespread in southern and central Greece at this time and that its details change from region to region.
That makes sauceboats useful as tracers. The Dokos examples have been compared with sauceboats from Askitario in Attica, from Lerna in the Argolid and from the Cyclades. The similarities suggest the wreck lies on a maritime route running from southern Euboea, through the Saronic Gulf, to the Argolid. A single cargo that combines several regional styles is direct evidence that pots, and presumably the people and food that went with them, travelled along that route as part of one voyage.
The pottery was also made before the potter’s wheel was in general use in this region. That is part of what gives the assemblage its research value: hundreds of near-identical hand-built vessels shipped together say a lot about the scale of production in a mainland workshop more than four thousand years ago.
Were there metals on the Dokos wreck?
Yes, but not much, and not precious. Two lead ingots were found. According to press accounts of the excavation, Papathanasopoulos believed they came from the mines at Lavrion (Laurion) in southern Attica, which were later the source of the silver that financed classical Athens. Lead and silver come from the same ores at Lavrion, and the district is thought to have been worked as early as the Early Bronze Age. If that attribution holds, the ingots tie the ship to one of the oldest mining districts in Europe.
The ingots are a reminder that metal was valuable enough in the third millennium BC to be shipped in small quantities alongside bulk goods. By the Late Bronze Age the picture had changed: the Uluburun ship of the late 14th century BC carried about ten tons of copper in 354 oxhide ingots, plus roughly a ton of tin. Dokos sits at the beginning of that long rise in seaborne metal trade.
What were the millstones and stone anchors for?
Many millstones were found among the pottery. Grinding stones were a basic tool for every household that processed grain, and good stone was not available everywhere, so they could have been part of the cargo. They are also heavy, compact and stable, which makes them ideal ballast. The excavators left both possibilities open, and they are not mutually exclusive: a trader could carry saleable millstones low in the hull precisely because they also steadied the boat.
About 40 m from the main cargo, the team found two large boulders with holes bored through them. These were stone anchors, probably dropped before the ship was lost, perhaps in a failed attempt to hold off the rocks. They are among the oldest stone anchors recorded from an excavated wreck. Their position away from the pottery is consistent with a ship that anchored, dragged or broke free, and then struck the shore while trying to reach the bay. One account of the excavation reports the working theory that the ship ran aground while trying to enter the harbour.
How big was the Dokos ship?
Nobody knows, because the hull has not survived in a form that can be measured. One team member, the marine archaeologist Yiannis Vichos, offered an estimate based on ship images painted or incised on pottery of the period: an oared vessel with a low prow and a high stern, at least 80 feet (about 24 m) long. That is an informed guess from iconography rather than a measurement from the wreck. The newer Kekova wreck in Turkey, by comparison, has been estimated at about 10 m long from the spread of its cargo and anchors.
Early Aegean ships were almost certainly paddled or rowed rather than sailed. Images of Early Bronze Age longboats appear on Cycladic “frying pan” vessels and on rock carvings, showing long, low hulls with many paddlers. A boat of that kind could carry a cargo of pottery and stone along a coast, hopping between beaches and small harbours, which matches what the Dokos cargo suggests about its route.
Did any wood survive?
Most accounts say the sea dissolved everything biodegradable, leaving only ceramics, stone and metal. That is broadly true: there is no hull to study. However, a 2002 study in the journal Microscopy and Microanalysis reported microscopic analysis of Early Helladic II period wood recovered from the underwater site near Dokos. Small fragments trapped under or among heavy objects can survive where an entire hull does not. Such fragments can help identify the timber species used, and potentially the condition of the wood, even if they cannot reconstruct the ship itself.
Was Dokos really a shipwreck?
Most archaeologists accept that the site is the remains of a sunken ship’s cargo. The sheer number of complete vessels, their tight concentration and the presence of anchors, ingots and millstones all fit a single lost cargo. But because no hull survives, the interpretation is not quite as secure as it is for a wreck with timbers. The nautical archaeologist Frederick van Doorninck was reported as suggesting that some of the material could have fallen or been thrown overboard, possibly even as votive offerings, and that no fully convincing argument had settled the question. That caution is healthy: an accumulation of pottery in a sheltered bay can form in more than one way.
Even so, the uniformity of the assemblage is hard to explain as anything other than a cargo. Offerings or accidental losses would be expected to build up over a long time and in a variety of styles and dates; the Dokos pottery is overwhelmingly of one phase. That is why the site is still generally treated as a single shipwreck.
Is Dokos still the oldest shipwreck in the world?
Not any more, on current evidence. A wreck off Kekova in Turkey, published in 2026, is now listed by Guinness World Records as the oldest.
For decades Dokos was routinely called the oldest known shipwreck, a claim repeated in reference books and still in many online articles. In July 2026 Hakan Öniz and Ceyda Öztosun of Akdeniz University published a preliminary report in the Journal of Maritime Archaeology on an Early Bronze Age wreck in the Kekova region of Antalya. Discovered in 2024 at about 30 m, it is estimated at 4,400–4,700 years old, dated by three Bronze Age stone anchors and more than 30 amphorae that formed its cargo. Like Dokos, no wooden structure survives. Guinness World Records now lists the Kekova site as the oldest shipwreck. A second possible Early Bronze Age wreck, at Finike Güvercinada, also off Antalya, was reported by the same team a month later.
Because both sites are dated by their cargo rather than by the ship, and both date ranges are wide, the ranking could change again as more analyses are published. What is clear is that Dokos is no longer alone. It remains one of the two or three oldest shipwrecks known, the oldest excavated in Greece and the best published of the Early Bronze Age cargo sites.
How does Dokos compare with other Bronze Age wrecks?
The best-known Bronze Age wrecks in the Mediterranean are all from Turkey’s south coast and all much later. Cape Gelidonya, which Throckmorton also found, sank around 1200 BC with a cargo of copper and tin ingots, scrap bronze and merchant’s weights. Uluburun, found in 1982 by a sponge diver and excavated by the Institute of Nautical Archaeology between 1984 and 1994, sank at the end of the 14th century BC; its excavators recovered nearly 17 tons of artifacts, including copper, tin, glass ingots, ivory, gold jewellery and exotic woods. A Late Bronze Age wreck with ingots was also reported off Kumluca near Antalya in 2019.
Next to those, Dokos looks modest: no gold, no glass and very little metal. Its importance lies in its age and its ordinariness. Uluburun and Gelidonya show the high-value metal trade of the Late Bronze Age, while Dokos shows the earlier, everyday coastal traffic in pottery and household goods that connected small communities a thousand years before. Read our accounts of the Uluburun treasure and the Antikythera wreck for later chapters in the same story.
What does Dokos tell us about Early Bronze Age Greece?
Early Helladic II was a period of rapid change on the Greek mainland. Bronze working spread, large “corridor houses” such as the House of the Tiles at Lerna were built, and fortified, densely populated settlements grew at places like Kolonna on Aegina and Manika on Euboea, which may have covered 70–80 hectares. Archaeologists have long argued that sea trade helped drive that growth, and the Dokos cargo supplies physical evidence of it: mass-produced pottery from a mainland workshop moving by boat, mixed with Cycladic wares and metal from Attica.
The ship also sank near the end of the period. Around 2200 BC many Early Helladic II sites, Lerna among them, were destroyed or abandoned, and a shift in climate is one of the explanations proposed. The Dokos cargo is a snapshot of the trading world that existed just before that disruption.
Can you see the Dokos treasure today?
The recovered finds were taken to the museum on Spetses for conservation and study, and the main published accounts of the excavation appeared in HIMA’s journal Enalia. The wreck site itself is a protected archaeological area. Greek law tightly controls diving on known ancient sites and forbids removing antiquities from the seabed, so the wreck cannot be visited casually on a recreational dive. What remains on the bottom today is, in any case, largely an excavated slope; the cargo itself is ashore.
If you are interested in where treasure-hunting stops and archaeology starts, Dokos is a good example. Nothing on board would excite a salvor, yet its scientific value is enormous. Compare it with the coin-laden San José galleon, where the arguments are about billions of dollars, to see how differently the word “treasure” can be used.
How old is the Dokos shipwreck?
Early Bronze Age. Most sources place it in Early Helladic II, about 2700–2200 BC; HIMA dated the pottery to about 2200 BC, and some summaries give 2250–2050 BC.
What treasure was found on the Dokos shipwreck?
No gold. More than 500 clay vessels and over 15,000 sherds and artifacts, two lead ingots, millstones and two stone anchors.
Where is the Dokos shipwreck?
At roughly 15–30 m off the small island of Dokos, next to Hydra in the Argo-Saronic Gulf of Greece.
Who found the Dokos shipwreck?
Peter Throckmorton, on August 23, 1975. HIMA excavated it from 1989 to 1992 under George Papathanasopoulos.
Is Dokos the oldest shipwreck in the world?
It was long described that way, but a wreck off Kekova in Turkey, published in 2026 and estimated at 4,400–4,700 years old, is now listed by Guinness as the oldest.
Where are the Dokos finds kept?
They were taken to the museum on Spetses for conservation and study.
Can you dive the Dokos wreck?
Not freely. It is a protected archaeological site, and removing anything from ancient sites in Greece is illegal.
- Dokos shipwreck and Dokos island reference entries — date, depth, discovery, excavation, finds, island area and population.
- Greek City Times (2023), “Maritime History: Dokos, the Most Ancient Shipwreck in the World” — HIMA excavation, SHARPS, Spetses Museum, cargo list.
- Greek Reporter (2024), on the Dokos wreck — sauceboat types, Lavrion lead, Vichos size estimate, van Doorninck’s caution (drawing on the 1989 New York Times report by P. Anastasi).
- Pendleton, Stephens & Ellis (2002), “The Characteristics of Early Helladic II Period Wood Recovered From An Underwater Shipwreck Site Near Dokos, Greece,” Microscopy and Microanalysis. doi:10.1017/s143192760210691x
- Öniz & Öztosun (2026), “Preliminary Report on the Early Bronze Age Shipwreck in the Kekova Region of Antalya,” Journal of Maritime Archaeology. doi:10.1007/s11457-026-09532-3; Guinness World Records, “Oldest shipwreck.”
- Pulak (2012), “Uluburun Shipwreck,” and Bass (2012), “Cape Gelidonya Shipwreck,” in The Oxford Handbook of the Bronze Age Aegean.
- Early Helladic period and Peter Throckmorton reference entries — EH II chronology, Lerna, Kolonna, Manika; Throckmorton’s career.
Blane Perun has been exploring Earth’s oceans and marine life for over 25 years. Founder of TheSea.Org in 1999, underwater photographer, coral aquaculture pioneer, and explorer of 80+ countries.
About Blane Perun → More on Shipwreck Treasure →