John White attempted to establish the colony of Roanoke on Roanoke Island, in what is now Dare County, North Carolina, inside the Outer Banks barrier system. The 1587 expedition landed there in July 1587, reoccupying the site of the failed 1585–86 English garrison. White returned to England for supplies that August and could not get back until 1590, when he found the settlement dismantled and the colonists gone.
The 1587 Landing and the Mechanics of Settlement
The Roanoke colony was not planted on the mainland. Roanoke Island sits between the Outer Banks and the North Carolina mainland, roughly 12 miles long and 3 miles wide, separated from the Atlantic by a thin barrier of sand. That geography mattered. The island offered deep water access through Croatan Sound, protection from Atlantic storms, and a defensible position, but it also isolated the settlers from the mainland groups they needed for food and trade.
White's fleet left Plymouth, England, on April 26, 1587, with three ships and about 150 people, including 17 women and 9 children. The original plan called for a settlement near the Chesapeake Bay, roughly 100 miles north, where the water was deeper and the location more strategic. That plan collapsed when the fleet's commander, Simon Fernandes, refused to carry the colonists further and forced them ashore at Roanoke on July 22, 1587. The colonists had not chosen the site themselves. A privateering captain's profit motive overrode the expedition's stated destination.
The settlers rebuilt on the footprint of the earlier military garrison, repairing structures abandoned in 1586. Within weeks, colonist George Howe was killed by local Algonquian people while crabbing in shallow water. The killing signaled that the earlier garrison's violent reputation had poisoned relations with the Secotan communities, and it drove the colonists to press Governor White to return to England for help.
Why the Location Failed: Quantified Context
The Roanoke settlement's collapse was a logistics problem measured in months and miles. White sailed for England on August 27, 1587, expecting to return within a season. The voyage round trip was roughly 6,000 nautical miles. Instead, England's war with Spain intervened: the Spanish Armada forced the crown to requisition every seaworthy ship in 1588, and White could not secure passage until 1590, a delay of nearly three years.
The colony's supply window was far shorter. A 16th-century colonial expedition typically carried 6 to 12 months of provisions, and Roanoke's 1587 store was thinner than the 1585 garrison's because the fleet had diverted supplies for privateering. The settlers arrived in July, too late to plant and harvest a reliable corn crop before winter. That left hunting, fishing, and trade with local communities as the only food sources, and the Howe killing had already degraded the trade relationship.
The 1585–86 garrison that preceded White's colony had survived on imported provisions and gifts from the Roanoke, Croatoan, and Secotan peoples. When the garrison's commander, Ralph Lane, executed a Secotan leader and attacked a neighboring village over a stolen silver cup, he destroyed the alliance that had fed his men. White inherited that damage. Early English reports of the region's fertility, some describing it as comparable to England's best farmland, understated how dependent any settlement would be on Indigenous trade networks.
Edge Cases: Croatoan, Hatteras, and the Chesapeake Option
The word "Croatoan" carved on a post at the abandoned fort, and "CRO" cut into a tree, gave the disappearance its name. White had agreed with the colonists that if they moved, they would carve the name of their destination and add a cross if they were forced to leave by attack. No cross appeared. Croatoan was the name of an island about 50 miles south, home to Manteo's people, who had been friendly to the English. A move there was plausible and prearranged.
Modern archaeology has not settled the question. Excavations on Hatteras Island since 1998 have produced 16th-century English and European artifacts, including a sword hilt and a gold ring, that suggest at least some colonists integrated with Croatoan communities. A site on nearby Croatoan Sound has yielded additional period objects. The evidence supports assimilation or relocation, not proof of it.
The Chesapeake option remains an open edge case. A 2012 analysis of a map drawn by White, held by the British Museum, found a hidden patch concealing what may be a fort symbol near the Chesapeake Bay's Albemarle Sound entrance. Some researchers interpret this as evidence that a faction of colonists relocated north, the original intended destination. Others read the patch as a drafting correction. No large-scale archaeological trace of a 1587 English settlement has been found at either location.
A third edge case involves the 1585–86 garrison, not White's colony. Fifteen men left behind by Sir Richard Grenville when the garrison evacuated in 1586 were never accounted for. Their fate, possibly killed or absorbed into local communities, is separate from the 1587 colonists' disappearance and often conflated with it.
Misconceptions Worth Correcting
The first misconception is that the colony was "lost" as a single event. White's 1590 search was brief and constrained; he and his crew had a storm-damaged ship, a captain unwilling to winter over, and orders to return to England. The colony's disappearance likely unfolded over months or years, and English ships visited the Outer Banks region repeatedly between 1590 and the 1607 Jamestown landing without conducting a systematic search. A 1602 expedition under Samuel Mace was reportedly sent specifically to find survivors and instead spent its time gathering sassafras, a valuable medicinal commodity.
The second misconception is that the colonists were certainly massacred. No bodies, mass grave, or evidence of a battle has been found. The absence of the agreed cross code is a strong argument against a violent end at the site itself.
The third is that Roanoke was England's first attempt at American colonization. England's first permanent claim in the region was the 1583 Newfoundland venture, and the 1585 Roanoke garrison preceded White's colony by two years. Roanoke was England's first sustained attempt at a family settlement.
A professional caveat: the "lost colony" framing comes largely from White's own narrative and from later promotional literature, including his published account of the 1590 voyage. White had personal and financial reasons to emphasize mystery over negligence. Historians treat his account as a primary source shaped by its audience, not a neutral report.
Comparison of English Settlement Options, 1585–1607
| Option | Distance from Roanoke | Strategic rationale | Why it was chosen or rejected |
|---|---|---|---|
| Roanoke Island (1585 garrison) | 0 miles | Deep sound access; defensible; site already cleared | Chosen first, abandoned after supply failure and conflict |
| Roanoke Island (1587 colony) | 0 miles | Reuse existing structures; proximity to Croatoan allies | Chosen under duress by Fernandes, not planned |
| Chesapeake Bay | About 100 miles north | Deeper water; better defense; mainland position | Original 1587 plan; abandoned mid-voyage |
| Croatoan Island | About 50 miles south | Friendly Indigenous community; prearranged refuge | Probable destination of survivors; unconfirmed |
| Jamestown, Virginia | About 150 miles northwest | Inland river defense; deeper water; larger expedition | Settled 1607, 20 years later, with more state support |
| Newfoundland | Far northeast | Fishing grounds; earlier English claim | 1583 claim; not a family settlement |
Timeline of the Settlement and Search
| Date | Event |
|---|---|
| 1584 | Amadas and Barlowe expedition reaches Roanoke; reports favorable conditions |
| 1585 | Grenville's fleet lands 107 men as a military garrison under Ralph Lane |
| June 1586 | Garrison evacuates with Francis Drake's fleet after conflict with Secotan communities |
| July 22, 1587 | White's settlers land on Roanoke Island after Fernandes refuses to go north |
| August 18, 1587 | Virginia Dare, first English child born in the Americas, is baptized |
| August 27, 1587 | White sails for England |
| 1588 | Spanish Armada ties up English shipping; White cannot return |
| August 18, 1590 | White lands at Roanoke, finds the settlement dismantled and "Croatoan" carved |
| 1602 | Samuel Mace expedition sent to find survivors; diverted to sassafras collection |
| 1607 | Jamestown founded; search for Roanoke survivors largely ends |
| 1998–present | Hatteras Island excavations recover 16th-century English artifacts |
Practical Steps for Researching the Evidence
Anyone tracing the question, whether for a school project, a family history, or a site visit, benefits from going to primary and archaeological sources rather than summaries.
Read White's own account of the 1590 voyage in "The Principal Navigations" by Richard Hakluyt, available through university libraries and public-domain archives. Consult the First Colony Foundation and the British Museum's records on the White map and its concealed patch. Check North Carolina State Historic Sites and the National Park Service for the Fort Raleigh National Historic Site, the current protected location on Roanoke Island. Distinguish the 1585 garrison from the 1587 colony in every source; casualty, supply, and command details belong to different expeditions. Treat artifact dates carefully: a 16th-century object on Hatteras supports presence or trade, not a specific colonist's identity.
Limitations: What the Evidence Cannot Establish
The central limitation is archaeological. Roanoke Island's shoreline has eroded by an estimated several hundred feet since 1587, and the original fort site may lie underwater or beneath later construction. Hatteras Island artifacts establish a plausible Indigenous-English contact zone but not the fate of a named individual among about 115 colonists.
The documentary record is also thin and self-interested. White's account is the only eyewitness description of the 1590 landing, and no colonist's written record survives. Spanish and French sources from the period mention rumors of English castaways but do not substantiate them.
Finally, the question of the colony's exact site within the island, separate from the island itself, is unresolved. Fort Raleigh National Historic Site protects the most likely location of the 1585 garrison earthworks, but the 1587 colony's precise footprint has not been identified. Any claim to have found the colonists' houses or graves lacks published archaeological confirmation.
The answer to where John White attempted to establish the colony is firm and well documented: Roanoke Island, in present-day Dare County, North Carolina, in July 1587. What happened to the people he left there remains genuinely open, and responsible accounts should say so.