North Carolina and Virginia · 1587 to about 1607, with testimony to 1888
What became of the men, women and children John White left on Roanoke Island in August 1587, whose only message to him when he returned in August 1590 was the word CROATOAN cut on a post?
What happened. On 27 August 1587 John White sailed from Roanoke Island, on the coast of what is now North Carolina, to fetch supplies from England; he was governor of a colony of about 115 men, women and children, the first English settlement in America to include families, sent out that spring by Sir Walter Ralegh, who held the queen's licence to settle the coast, and White's daughter Eleanor and her newborn child Virginia stayed behind. War with Spain kept him away for three years, and when he landed again on 18 August 1590 the houses were down, the site was walled with tree trunks, the boats were gone, heavy iron and lead lay in the grass, and the word CROATOAN was carved on a post: the name of an island fifty miles to the south, now the southern part of Hatteras Island, and of the friendly people who lived there. By a code agreed before he left, a cross over the word would have meant trouble, and there was none. A storm cost his ships their anchors before he could reach the island, and the voyage was given up.
Why it is a mystery. No search reached them, no house, grave or bone of theirs has been identified, and every later report is hearsay from Indians, heard through interpreters at Jamestown, the English colony founded on Chesapeake Bay in 1607, about a hundred miles to the north. The colonists had told White they would move fifty miles inland, to the west, while the word they carved names an island to the south. David Beers Quinn, the historian who edited the colony's documents, held that the main body went to Chesapeake Bay and was killed there in April 1607 by the people of Powhatan, the paramount chief of the Indian nations around Jamestown, and the historians Karen Ordahl Kupperman and Michael Leroy Oberg followed him; since 2010 the usual reading, set out most fully by the historian James Horn and shared by the journalist Andrew Lawler, is that the colony split between Croatoan and the mainland at the head of Albemarle Sound, the shallow water west of the island, where the Chowan River enters it, and was absorbed into the Indian peoples around it, the Algonquian-speaking peoples of the sounds and the Tuscarora, the Iroquoian-speaking people of the interior; and the First Colony Foundation, an archaeological group that has dug on that shore since 2012, announced in June 2026 that it had found the colony. Scott Dawson, a Hatteras native who founded a local archaeological society, and the archaeologist Mark Horton, who have dug on Hatteras since 2009 at Cape Creek, the site of the Croatoan town, hold that everyone went to Croatoan, a reading endorsed in 2025 by Kathleen DuVal, a historian of Native America; Charles Ewen, an archaeologist, and E. Thomson Shields, a literary scholar, both of East Carolina University, hold in their 2024 book Becoming the Lost Colony that no hypothesis is more probable than another; and a carved stone of 1937 claiming to be a message from Eleanor Dare (the first Dare Stone) has its own defenders.
How it was judged. Every surviving document was read in full, in the original spelling where the original text could be reached and otherwise in published translations or transcriptions: the fifty pages in the Principal Navigations of Richard Hakluyt, the clergyman and geographer who collected and printed the English voyages; the Spanish documents; the Jamestown reports of 1607 to 1625, each traced to its first teller, chiefly through the writings of Captain John Smith, the Jamestown leader, and the Historie of William Strachey, the colony's secretary in 1610 and 1611; and the later testimony of 1650 to 1888. The archaeology of three sites was checked against what has been published, as of September 2026, and the possible fates were then set side by side in a ledger, a table of weights that anyone can change, which was run nineteen ways to show how far the result depends on the weights. On that weighing the mainland at the head of Albemarle Sound comes first as the main body's principal seat, with about 68 chances in 100 (range 41 to 77), Croatoan second at 17, somewhere else 7 and Chesapeake Bay fourth at 4, and the answer is a leading candidate rather than a settled one, because no source names the place, no structure or grave has been found, and the two excavations that could decide the question have not been published. A published, sealed (undisturbed) 1580s deposit at Cape Creek with a hearth and slag (the waste from working metal) would put Croatoan first; separately, some colonists were possibly killed by Powhatan's people about 1607 (45 in 100), and the Hatteras Indians who told the surveyor John Lawson, in his book of 1709, that several of their ancestors were white people are more likely than not a trace of them (60 in 100).
The carving names Croatoan as the colonists' seat; their arrival there is inferred, since no witness recorded it. On the weighing of the evidence in the full report, linked below, the main body's principal seat was the Albemarle mainland at about 68 in 100 (range 41 to 77), Croatoan 17, somewhere else 7, the Chesapeake 4.
The possibilities are put in order and none is established. The leading one is a reconstruction from hearsay and from digs that have not been published, and every figure depends on the weights chosen in section 6 of the report, which anyone can change.
The figures are for the principal seat of the main body from about 1588 to 1607, on the weighing set out in section 6 of the report, which anyone can change: the Albemarle mainland (the Chowan mouth and the Roanoke River country) 68 in 100, and between 41 and 77 across nineteen alternative weightings; Croatoan (Hatteras) as the principal seat 17 (11 to 38); somewhere else 7 (5 to 14); the Chesapeake, Quinn's thesis, 4 (2 to 9); an attempt to sail home 3 (0 to 6). Each figure blends in an even split across the five places, at 15 parts in 100, to allow for evidence nobody has seen; the range includes the raw ledger without that blend (mainland 77, Croatoan 16, other 5, Chesapeake 2, sea 0), which is where the lowest figures for the Chesapeake and the sea come from. The mainland leads in every weighting; its lead over Croatoan is narrowest when the announced plan is left out (49 against 32) and when the plan, the Jamestown reports and the Bertie ceramics (the English pottery from Sites X and Y, the First Colony Foundation's two dig sites on Salmon Creek in Bertie County, on the mainland at the head of Albemarle Sound) are all counted at half strength (41 against 38); the lowest weighting still puts the mainland first.
Separate findings, each judged on its own evidence and each depending on the readings stated in section 6: some of the colonists went to Croatoan after August 1587, 80 (65 to 90), a finding that rests on the carving, which names Croatoan as their seat, while their arrival is inferred; the colony divided, 75 (60 to 85); some were killed by Powhatan's people about 1607, 45 (30 to 60), which is why the verdict says "possibly killed"; descendants lived among the Hatteras Indians in 1709, 60 (45 to 75); some tried to sail home, 10 (5 to 20); the first Dare Stone is a sixteenth-century inscription, 20 (15 to 25), which makes it about 80 in 100 a fabrication of 1937.
The strongest rival is the reading of Dawson and Horton, endorsed by DuVal, that the whole colony went to Croatoan and stayed, argued from the carving, from Lawson and from the Cape Creek excavations, now with the hammerscale, the flakes that fall from iron being forged, found there in 2024. Against it: the only destination the colonists stated was "50 miles further vp into the maine", "the maine" meaning the mainland as against the islands, and Croatoan is not the mainland; the Croatoans had asked the English in July 1587 not to take their corn, "for that they had but little"; every Jamestown informant placed the clothed men (men dressed like the English, whom Indian informants reported living in the interior and whom the English took for colonists) inland to the south and south-west, none at Croatoan; and the 1580s dates claimed for the hammerscale rest on interviews, with no section drawing or radiocarbon table in print. Quinn's Chesapeake reading fails on Strachey's own words, that the victims were "seated far from him" and "out of his territory", which the Chesepians, the people of the south shore of Chesapeake Bay, within Powhatan's reach, were not. The leading scenario is itself a reconstruction from hearsay and from digs that have not been published: no source names the place, no structure or grave has been found, and the two excavations that could decide it (Sites X and Y on Salmon Creek; the Cape Creek hammerscale) have no published data as of 14 September 2026.
Chances in 100 that each place was the main body's principal seat from about 1588 to 1607, on the weighing in section 6 of the report, with the range across nineteen alternative weightings in brackets; each figure blends in an even split across the five places, at 15 parts in 100, to allow for evidence nobody has seen, and the range includes one weighting that drops that blend. The mainland reading rests most on the colonists' announced plan and on the Jamestown hearsay; discount those and Croatoan runs close. The separate findings on Croatoan, the division of the colony, the killings of about 1607 and the Dare Stone carry their own figures.
Every surviving document was read in full and in the original spelling: the fifty pages in Richard Hakluyt's Principal Navigations, the collection of English voyages printed in 1598 to 1600; the Spanish documents; the reports of 1607 to 1625 from Jamestown, the English colony on Chesapeake Bay, chiefly in the writings of Captain John Smith, its leader, and of William Strachey, its secretary, each traced to its first teller; and the later testimony of 1650 to 1888. The modern books were read in extracts and reviews rather than cover to cover, and the 2001 article of Thomas Parramore, a North Carolina historian, was not obtained. There was no fieldwork and no laboratory result.
The archaeology of three sites, on Roanoke Island, on the mainland shore of the Chowan River and on Hatteras, was checked against what had been published by September 2026. Straight-line distances from Roanoke Island to every place in play were computed as great-circle distances, the shortest distance over the earth's surface, in statute miles. The colonists' "50 miles further vp into the maine" was checked against the "130 miles" that Ralph Lane, governor of the first colony of 1585 to 1586, gave to the Chesepians on Chesapeake Bay and to Chawanook, the chief town of the Chowanoc people on the Chowan River: in any reckoning of the mile the phrase stays inside the Albemarle basin.
The possible fates were then set side by side in a ledger, a table giving each group of evidence a weight for how well it fits each possibility; the weights are multiplied through. The numbers are judgements rather than measurements, and the ledger was run again nineteen ways with different weights to show how far the result depends on them. The first Dare Stone, a carved rock produced in 1937 that claims to be a message from Eleanor Dare, the daughter of the colony's governor John White, was judged on its own, and its spellings were tested against the narratives of Hakluyt, Smith and Strachey.
Figure 1 shows the chances in 100, on that weighing, that the main body's principal seat was each of five places, a bar for the reported value, a line for the range across the nineteen alternative weightings: the Albemarle mainland 68 (41 to 77), Croatoan, the island 50 miles to the south, 17 (11 to 38), somewhere else 7, Chesapeake Bay 4, the sea 3. Figure 2 plots the straight-line distances from Roanoke Island, with rings at 25, 50, 75 and 100 statute miles and the colonists' "50 miles" as a dashed ring: Cape Creek on Croatoan lies 47 miles away and Edenton at the mouth of the Chowan 51.
On that weighing the mainland at the head of Albemarle Sound comes first, with about 68 chances in 100 (41 to 77); Croatoan is second at 17. The result is a ranking, and every figure depends on the weights in the ledger, which anyone can change. The answer is not settled: no source names the place, no structure or grave has been found, and the two excavations that could decide the question have not been published. Some colonists were possibly killed about 1607 by the people of Powhatan, the paramount chief of the Indian nations around Jamestown (45 in 100), and the first Dare Stone is, at about 80 in 100, a fabrication of 1937.
Already established by others. Quinn, Horn and Philip Barbour, the editor of the Jamestown texts, whose identifications of the place names are the standard ones, read the documents closely first, and section 8 of the report credits each step: Quinn for the reading that Strachey's "at Roanoak" means the colony rather than the island; Horn for the argument that Strachey's own geography rules out the Chesapeake; Helen Rountree, the ethnohistorian of the Powhatans, for the dating of the Chesepians' destruction (Strachey says Powhatan destroyed them for a prophecy; Rountree puts it about 1597 or 1598, before Jamestown); Barbour for the point that the 1609 searchers (the two parties Jamestown sent into the interior in 1608 and 1609 to look for the colonists) never crossed the watershed to the Roanoke River; Lee Miller, a writer (Roanoke: Solving the Mystery of the Lost Colony, 2000), and Kupperman for the reading of the slaughter story (the report that Powhatan's people killed the colonists) as propaganda of the Virginia Company, the London company that financed Jamestown; Oberg for the point that Croatoan could not feed the whole colony; Ewen for the objection that the layers holding the Cape Creek finds are seventeenth-century; Brandon Fullam, an independent researcher, for the caution that the clothed men reported at Ocanahowan, an inland town named in the Jamestown reports, may have been Spaniards. The order of the reports, the single London channel behind the slaughter story (one Powhatan informant, whose story was told in London in the winter of 1608 to 1609 and is repeated by all three London texts that carry it), the late and editorial character of Powhatan's supposed confession of the killing (a note that Purchas printed in a margin in 1625 and that appears nowhere in Smith), the negative result of the 1609 searches (both parties came back with rumours and no survivor), the independence of Lawson's testimony and the weakness of the Lumbee link (the claim, first published in 1888, that the Indians of Robeson County, North Carolina, the Lumbee, descend from the colonists) are all in Quinn, Barbour, Horn, the Lumbee historian Malinda Maynor Lowery, and Lawler.
What historians have said. Quinn (1974, 1984, 1985), followed by Kupperman, Oberg (2008) and Horn (2005), held that the main body went to the Chesapeake and was killed by Powhatan in April 1607. Since 2010 (Horn, A Kingdom Strange, 2010; the First Colony Foundation; Lawler 2018) the usual reading is that the colony split between Croatoan and the Chowan and was absorbed; Thomas Parramore, a North Carolina historian (2001), had already put the main body on the Albemarle mainland, where on his reading it died in warfare between the Tuscarora and the Algonquian peoples on the western shore of the sound. Dawson and Horton (2016 to 2026), endorsed by DuVal (2025), hold that all of them went to Croatoan. Ewen and Shields (2024) hold that no hypothesis is more probable than another.
The verdict agrees with Horn's split-and-absorption reading at lower confidence than he states: the sources support the direction and the dispersal but not the route, the season, the headcount at Croatoan or the Powhatan raid of 400 warriors that his narrative supplies, and the ledger gives the mainland about two chances in three. It moves the weight of the slaughter story: the instructions sent in May 1609 by the London Council of the Virginia Company, the dedication Hakluyt wrote for a book on Virginia in April 1609 and Strachey's Historie are one London report from one Powhatan informant, Machumps, a man brought to England in January 1609, which itself dates the killing to spring 1607 and places it outside Powhatan's territory, so it does not support the Chesapeake; and Powhatan's "confession" with the musket barrel and brass mortar is a 1625 marginal note by Purchas and appears nowhere in Smith. No witness recorded the mainland move or the dispersal.