Twiddy is a volunteer participating in this investigation funded by the nonprofit Lost Colony Center for Science and Research in Williamston, North Carolina. Professional archaeologists have been sporadically investigating the sprawling Cape Creek site since 1938.

    Heath, co-director of the nonprofit foundation Olde Albemarle Research (OAR), has been retained by the Lost Colony Center to direct the investigations of Cape Creek and other sites in Buxton Village, which is roughly midpoint on Hatteras Island along the Outer Banks of North Carolina. He’s a veteran explorer of this maritime forest, having worked here half a dozen times while a graduate student, and later as a research associate with the Coastal Archaeology Office at East Carolina University, in nearby Greenville. That makes Heath and his co-investigator, Clay Swindell, also of OAR, two of a long line of archaeologists who have come here hoping to learn about Croatan lifeways before and after the time of Contact, and to perhaps find evidence that could help solve one of the most tantalizing mysteries of Colonial times -- the fate of the Lost Colony, the more than 100 English and Irish who landed on nearby Roanoke Island in 1587 and later abandoned their settlement.

    The Lost Colony resulted from England's second attempt to colonize the shores of northeastern North Carolina. The first attempt in 1585 and 1586 failed after its members, led by Ralph Lane, warred with local Indians and were unable to establish the means to sustain themselves. Commissioned by Sir Walter Ralegh, this expedition generated ethnographic reports, maps, and dramatic paintings by Thomas Harriot and John White. Its members signaled English ships passing off the shoals on Hatteras, and they sailed back to England with Sir Francis Drake's privateer fleet in 1586. A small party of soldiers left by Sir Richard Grenville in the same year to garrison Lane’s abandoned fort at Roanoke Island disappeared in 1586 and 1587. Their troubles did not dissuade the second group of settlers, who were borne by three ships that arrived in July 1587.

- Photo by Michael Halminski

Project co-director Clay Swindell writes field notes on his impressions of a newly discovered midden associated with a late 17th- to-early 18th-century European house site in Buxton Village.

    Although the colonist planned to settle on the Chesapeake Bay, circumstances forced their disembarkation at Roanoke Island, where they first occupied the rough cottages built by Lane's soldiers. But they arrived too late in the season to plant crops and recent research indicates that the region was in the midst of an extended drought. In late August, John White, governor of the colony, sailed to England to secure more supplies. Among the settlers left behind was his daughter, Elenor White Dare, who gave birth to Virginia Dare, the first reported child born of English parents in the New World. He would not return until August 1590, long delayed by sea battles between the English and the Spanish -- all English ships were held back by Queen Elizabeth I for the defense of the realm -- and the inability to organize a relief effort in London. Eventually, White found transport on a privateer headed for the West Indies to loot Spanish shipping.

    Upon returning to Roanoke Island, White and a party of sailors climbed the embankment adjacent the settlement and found “CRO” carved on a tree on the bluff. When they reached the area of the cottages, they found instead a palisade of tree trunks. One pale near the gate had the word “CROATAN” carved in capital letters. White wrote that the carvings were “to signifie the place, where I should find the planters seated, according to a secret token agreed upon betweene them & me at my last departure from them…for at my coming away, they were prepared to remove from Roanoak 50 miles into the maine.” But there was no accompanying Maltese cross symbol, a sign of distress the parties agreed to use. White and the privateers also found the previously cleared settlement area overgrown, suggesting it had been abandoned for some time. The colonists’ small boats and light artillery pieces were missing as well.

    The Lost Colony mystery now spans more than four centuries. White planned to sail down to Croatoan Island (now Hatteras Island) to look for the colonists, but a fierce storm came up and blew the privateers’ ships out into the Atlantic. The fleet commanders refused to further delay their expedition and White was forced home to England, never to see the colonists, his daughter, or granddaughter again.

    Like White, many historians interpret the cryptic carvings to mean some of the Lost Colonists may have migrated to Hatteras Island to live with the local natives, known to the English as the “Croatoans.” While there are innumerable theories about the fate of the group, Heath and others believe that at least some of the colonists likely made their way down the Outer Banks to take refuge among the Croatans (the modern spelling), who were known at the time to be on friendly terms with the English. That's why the archaeologists are digging in Buxton, hoping to find evidence of late 16th-century English occupation among the Contact-period Croatan settlements.

John White, the governor of the Lost Colony, painted this watercolor of Native Americans fishing in 1585. The painting reflects the Croatan Indians' intimate relationship with the rugged barrier island environment encompassed by their chiefdom in the 16th century.

    White’s maps of coastal North Carolina depicted numerous coastal Native American towns and chiefdoms, including Croatoan, offering tantalizing clues for others who later tried to find the Lost Colony. Sir Walter Raleigh, organizer of the failed 1587 colony, continued to look for the lost expedition, although he never left the comforts of England himself. Raleigh sent a ship commanded by Samuel Mace in 1602 that didn't reach Croatoan. In 1603 Mace returned, exploring the Chesapeake Bay area. His sailors kidnapped several Indians along the coast, but they learned nothing of the fate of their countrymen.

    Spanish and French reports during the same era further suggested that some of the English colonists were living among the Croatans. English settlers who reached Jamestown, Virginia, in 1607 later searched for the Roanoke colonists without success, thinking they had perhaps relocated near the Chesapeake Bay. Captain John Smith and others reportedly queried the local Powhatan Indians about the missing colonists. Chief Powhatan and other Indian contacts indicated that Powhatan's warriors destroyed an enclave of English refugees living among the Chesapeake tribe, when they subjugated the Chesapeakes sometime before 1607.

    In the early 1700s, explorer and writer John Lawson visited Roanoke Island, where he spent some time among the Hatteras Indians (then the English name for the Croatans). He wrote in 1709 that “several of their ancestors were white people and could talk in a book as we do, the truth of which is confirmed by gray eyes being found infrequently among these Indians and no others” -- suggesting the gray-eyed people descended from the Lost Colonists. Similar speculations continued to abound. In the 1880s, with the approach of the Roanoke Colony's 300th anniversary, a North Carolina attorney, Hamilton MacMillan, said he knew Pembroke or Robeson Indians -- now self-identified as the Lumbee -- who claimed their ancestors came from “Roanoke in Virginia”, which is what the English called the colony. According to MacMillan, some Pembrokes spoke pure Anglo-Saxon English and bore the last names of many of the Lost Colonists. The Pembrokes also had European features: fair eyes, light hair, and an Anglo bone structure. MacMillan's speculations, published in 1888, renewed interest in the Lost Colony.

    Over the centuries the mystery grew, spawning one theory after another about the fate of those early colonists. Did they move 50 miles inland, into what is now Tidewater, North Carolina, as White noted? Did they assimilate with the Croatans, whose primary settlement was present-day Buxton? Were they killed by the Roanoke Indians, who were angered by Ralph Lane’s 1586 attacks, or later by the Powhatan Indians in Virginia? Did they simply starve? Did the warring Spanish swoop down and capture them? Heath believes the theory that some of the colonists split off and came to Cape Hatteras, hoping to signal a passing English ship as Lane’s soldiers did in 1586, is reasonable. Other colonists, he says, may have gone north to the Chesapeake Bay knowing the English intended to found a colony there.

- Photo by Michael Halminski

Student Catie Galloway records field notes on stratigraphic profile drawings during the 1998 investigations. Wind-twisted maritime forest vegetation and wind-blown dune sand cover a thin, but stratified midden (dark soil bands with marine shell and other artifacts) at the northeast end of the Cape Creek site.

    If part of the Lost Colony lived with the Croatans and blended into their society in the modern Buxton, Avon, Hatteras, or Ocracoke village areas of the ancient Croatan chiefdom, their archaeological legacy may not look too different from that of the 16th-century Croatans, according to Heath and Swindell, because the refugees would have likely brought little with them from Roanoke Island. The colonists simply lacked sufficient transport and White reported finding English trunks and goods scattered around the abandoned Citie of Ralegh. “One question we have as we're out here day to day is, if we found the Lost Colony, what would it look like? It's open to supposition," Heath says. “They would be essentially a refugee colony, for lack of a better term, probably very hungry and carrying with them personal belongings and a minimal quantity of household goods. Heath’s hypothesis is that they would have “gone native” with some attempts to maintain their Englishness. “They certainly would be eating the local food resources,” he adds. “To a certain extent, they would be reliant on the local technologies, perhaps with a European flair, because they were not being resupplied (with European goods).”

    To distinguish archaeologically between assimilated Lost Colonists and the natives requires an understanding of Croatan society during the critical mid-16th through mid-17th century period. “Without a good solid foundation, it's actually difficult to understand what the Lost Colonists living among the Croatan Indians would look like,” Heath says. Thanks to the work of William G. Haag of the Louisiana State University, and David S. Phelps, an archaeologist now retired from East Carolina University, researchers have a general understanding of the prehistoric Croatan, as well as knowledge of their activities during specific time periods after Contact. But they need to better understand the period just before and after English contact, according to Heath, and that's what has drawn them to Cape Creek.

    Heath notes that his and Swindell's investigations are often “one step ahead of the bulldozer,” because development on the island is occurring at dramatic speed. Nonetheless, Fred Willard, founder and director of the Lost Colony Center for Science and Research, says he has permission from many property owners to dig on their land. “We have decades and decades of fieldwork to do before we can begin to make sense of this site,” Heath notes. “Through systematic survey and testing we’re hoping to find out what the limits of the Cape Creek site are, how large it was, and how it changed over time,” Heath says. “We'll look at the settlement patterns over time, get a sense of which areas were most intensively occupied in a particular time slot, and then we can target areas in the period we’re most interested in, the 1580s up to about 1600.”

Fred Willard inspects a 17-century hand-blown glass bottleneck discovered by Junius Jennette, a Buxton Village landowner, after potions of his property were cleared and prepared for house construction in the early 1990s.

    This investigation is part of an evolving multidisciplinary approach sponsored by the Lost Colony Center. It is also a resurrection of an earlier investigation, the Croatan Project, first supported by local enthusiasts who helped fund several subsequent digs, directed by David Phelps through the 1990s. The most enthusiastic of the locals is the 66-year-old Willard, who became so enraptured by archaeology that he returned to school at ECU to earn a bachelor's degree in anthropology and hopes to eventually earn a doctorate.

    In 1993, Willard and Barbara Midgett, a long-time project supporter and volunteer, discovered shell midden deposits containing Croatan artifacts exposed by Hurricane Emily’s scouring of the coastline a few weeks after the storm hit. That revived professional interest in the site. For more than a decade since, Willard has used primary sources including maps, deeds, and artifacts, as well as satellite remote-sensing images, and genealogical records -- he and fellow researchers have found Indian descendents with 168 English surnames including Elks, Squires, Gibbs, and Dunbar -- to determine possible places to search for the Lost Colony. His central premise is that the colonists later migrated 50 miles inland from Roanoke Island, though he agrees with Heath and Swindell that they may have spent time in the Buxton area.

    Willard has authored or co-authored several papers about his quest. In one study, he and colleagues from nearby Elizabeth State City University correlated remote-sensing data that identified regional soils suitable for corn production -- which the colonists would have desperately needed -- with Indian settlements noted on historic maps by White and others. They concluded that five potential locations for a Lost Colony settlement could be identified, the most promising being a place called Buck Ridge along the Alligator River near Roanoke Island. That location roughly corresponds with an Indian village mapped by White. Buck Ridge is some 50 miles from Roanoke Island, Willard notes. But, Heath and Swindell note, so are a number of other places that they think are more plausible candidates.

    Willard hopes to do a survey at Buck Ridge next year. He also expects to continue supporting work at Cape Creek, which has a long history of intermittent investigation. Researchers did a surface collection of artifacts there in 1938 and 1939, and Haag conducted an extensive survey of the Outer Banks in 1954. In 1958, he reported that the “Cape Creek site is the best midden found on the Outer Banks,” and suggested the site was the “only contender“ for being the Croatan capital.

    Nearly 30 years later, Phelps directed follow-up test excavations at the site as part of a 400th anniversary project commemorating the Roanoke Voyages. Supported by the Croatan Group, he later directed several intensive excavations in the late-1990s. Phelps reported that his research confirmed that significant archaeological deposits from the town of Croatan, capital of the chiefdom, were still intact under the dunes. The digs uncovered both prehistoric and historic period features (cooking pits, structural postmolds) and diverse artifacts, including quantities of food remains, primarily fish and shellfish, pottery, stone tools and tool making debitage, copper beads, 17th century English coins, lead musket shot, musket flints, shell and glass beads, wrought iron nails and tools, as well as European and Croatan smoking pipes.

- Photo by Michael Halminski

Archaeologist David S. Phelps collects topographic data at the northeast end of the Cape Creek site in 1998, the year he discovered the early gunlock.

    In 1998, Phelps’ crew discovered an early gunlock and a gold signet rain adorned with a rampant lion, a symbol of English nobility. At the time, Phelps said the ring was the discovery of his career. Analysis of the rating revealed that its design is typical of the 16th century, and it prompted speculation about a connection with the Lost Colony.

    But subsequent research found the ring’s familial crest is likely that of the Kendall family, a surname not included on the 1587 list of colonists. There are two specific references to early English adventurers surnamed Kendall -- a “Master Kendall” and an “Abraham Kendall.” Swindell notes that several Kendalls, one of whom was executed for treason in 1607, were also associated with both the Jamestown Colony and later Virginia settlements, further complicating the issue. Master Kendall was a gentleman explorer with the 1585-1586 Lane expedition and Abraham Kendall was one of Sir Francis Drake's officers who were involved in the recovery of Lane and his men. Consequently, one or both of these men may have visited Croatan at some point. Moreover, Master Kendall may have been one of the tiny group of men who lived among the Croatans in the months before their rescue by Drake.

    After 1607, Virginia colonists traded with Indians along the Carolina coast and the ring could have been lost there in the early 17th century, an event that would correspond with the dates of diagnostic artifacts found in direct association with the ring. “The ring is an enigma,” Heath notes. “How it came to be where it was found is really open to supposition. But the archaeological context suggests the ring was lost by one of the Croatan Indians between 1660 in 1690.”

- Photo by Michael Halminski

Student Laurie Lucas sorts cleaned artifacts that will later be inventoried, catalogued, and analyzed.

    The musket lock was also found in a late 17-century context. Heath’s research, which includes correspondence with museum curators in England, concludes it was manufactured between 1610 and 1620, and found its way to North Carolina sometime later in the century. Willard, however, has consulted with a historical firearms specialist who believes the musket lock dates to the 16th century.

    So the researchers have yet to uncover the conclusive material link between the Croatans and the Lost Colony. “If you don't have fun with this, you're gone,” Willard says. “This whole thing is about the hunt. You work 10 years and you get one piece of the puzzle.” The pieces they're finding now -- abundant marine shells, fish and mammal bone, a triangular arrow point, and an abundance of Croatan pottery sherds -- indicate a likely late prehistoric occupation circa A.D. 1400-1585. The limited density of the midden deposit and the artifact distribution patterns suggest the activities of a single household that was occupied for only one or two seasons. During the last few days of the 2006 dig, the archaeologists uncovered evidence of a late 17th-century European house site, situated atop an earlier Croatan habitation shell midden deposit, perhaps one of the earliest permanent European structures in the Buxton Village area.

    But there are no traces of the Lost Colony this time. Heath is philosophical about the mystery. “The Lost Colony is sort of a talisman. It engages people's imagination,” he says. “We would like to find the Lost Colony, sure. That would be a great archaeological and historical coup.” However, that's not to say Heath considers the search for the Lost Colony to be more important than the search for knowledge of the Croatan. “We're most interested in the total picture, and it (the Lost Colony) is just a part of that picture.”

    Jim Morrison is a Virginia-based writer whose work has appeared in Smithsonian, The New York Times, National Wildlife, and numerous other publications.