Myrtles Plantation Ghost Stories: The Truth About Chloe

Trigger warning: slavery, racial violence, and depiction of lynching.

This post contains affiliate links. If you book or buy through them, I may earn a commission at no extra cost to you. As an Amazon Associate I earn from qualifying purchases.

Search “most haunted house in America” and the Myrtles Plantation comes up first, or close to it, more often than anywhere else in the country. A white bed-and-breakfast in St. Francisville, Louisiana, wraparound veranda, live oaks, the kind of Southern-Gothic postcard imagery that sells rooms on its own before a single ghost story gets told. It doesn’t need the ghost stories. It has them anyway — ten or more, by the property’s own count, and one above all the rest.

Her name, in most tellings, is Chloe. An enslaved woman, the story goes, forced into her enslaver’s bed, caught eavesdropping, punished with a severed ear she hid under a green turban. To win her way back into the house, she baked a cake laced with crushed oleander leaves — meant to sicken the children just enough that she’d be needed to nurse them back to health. The dose was wrong. The children died. Her fellow enslaved people, the story finishes, dragged her from her bed that same night, hanged her from a tree on the property, and threw her body in the river. Her ghost, green turban and all, is said to walk the grounds still — the plantation’s single most bankable feature, printed on tour brochures, narrated nightly, captured, allegedly, in a photograph that’s been reprinted for thirty years.

Here is what should change how you read everything above: researchers examining the surviving Woodruff family and estate records have found no contemporary evidence that a woman named Chloe or Cleo was enslaved here. Federal census schedules count five enslaved people in the Woodruff household in 1820 and thirty-two in 1830, but those schedules did not record their names. The family members Chloe supposedly poisoned did die young, but the documented cause was yellow fever, not cake laced with oleander. As Troy Taylor’s archival account explains, the detailed Chloe narrative appears much later, as the house’s reputation as a tourist attraction grew.

That’s not a footnote to the Chloe story. It’s the actual story behind the Myrtles Plantation ghost stories that circulate today. A plantation that enslaved real people, whose names were mostly never written down in the first place because the systems that kept records didn’t think enslaved people’s names were worth keeping — produced, decades later, one vividly named, vividly detailed woman who never existed, and made her famous. The invented one got a name, a motive, a face, a green turban, and thirty years of ghost tours. The real ones got a number in a census column.

That’s the shape of this location. Not a haunted house. A house that manufactured exactly the ghost it needed.

The Myrtles Plantation

The Myrtles sits on a low rise above the Mississippi River bluffs outside St. Francisville, Louisiana — West Feliciana Parish, roughly thirty miles north of Baton Rouge, the kind of ground that was prime real estate for a plantation precisely because it sat above the floodplain. The house itself is a long, low, twenty-two-room Creole cottage, wrapped almost entirely by a shaded veranda that later Greek Revival additions only made grander. Live oaks, heavy with Spanish moss, line the approach; a pond with a small gazebo island sits behind the house, ornamental rather than practical, added when the property was remodelled into something closer to a showpiece than a working plantation seat.

It functions today as a bed-and-breakfast, a restaurant, a gift shop, and a tour operation running day and night circuits through the same rooms — the French Room, the ladies’ and men’s parlors, the game room with its infamous mirror. Guests can book named suites: the Judge Clark Woodruff Suite, the General David Bradford Suite. It is, by any measure, a successful small hospitality business, and has been for decades. The ghost stories are not incidental to that success. They are, by a wide margin, the product.

The game room holds the property’s other signature relic: a large gilt-framed mirror. According to the legend, it was left uncovered after a death in the family, allowing spirits to become trapped in the glass. Visitors report handprints and strange marks that return after cleaning. These are claims attached to the object, not documented history. Today the house offers daytime historical tours and separate Evening Mystery Tours through six ground-floor rooms, while Restaurant 1796 and the accommodation business occupy the wider estate.

The History of the Myrtles

David Bradford built the house in 1796, calling it Laurel Grove. He’d fled Pennsylvania two years earlier, one of the more prominent figures in the Whiskey Rebellion — the 1794 uprising against a federal tax on distilled spirits that briefly had Pennsylvania farmers arming themselves against Washington’s own government — and built a life in Spanish Louisiana rather than face what was waiting for him back home; President John Adams pardoned him in 1799, by which point he’d already put down roots that would outlast him by two centuries. He built his fortune, and later his family’s, on the same foundation every plantation in the region ran on: enslaved labor. His daughter Sara married one of his own law students, Clark Woodruff, and the couple ran the plantation — and the people enslaved on it — after Bradford’s widow, Elizabeth, died in 1831. Federal census records show the Woodruffs holding five enslaved people in 1820 and thirty-two by 1830. Sara Woodruff died in July 1823. Her children, James and Cornelia, followed roughly a year later, a month apart — all three deaths attributed at the time, and consistently since, to yellow fever, which moved through households like this one on a near-annual basis in antebellum Louisiana. A third child, Mary Octavia, survived into her sixties.

Clark Woodruff sold the plantation to Ruffin Gray Stirling in 1834; the sale is recorded plainly, in the language of the time, as the land, the house, and the enslaved people together — people as part of the inventory being transferred, not incidental to it. Stirling had made his money in the cotton trade further north, and the remodeling he financed at the Myrtles was cotton money on display: the house nearly doubled in size, filled with furniture shipped in from Europe, renamed after the crape myrtles planted around the newly grand grounds. It’s this version of the house, built on the proceeds of one cash crop and the enslaved labor that worked it, that every photograph of the Myrtles today is actually a photograph of. Stirling died in 1854, leaving the estate to his wife; it eventually passed to their daughter Sarah, who married William Drew Winter in 1852. In 1871, Winter was shot on the property by an unidentified assailant — an unsolved murder, genuinely documented, the one violent death in the Myrtles’ catalogue with real evidentiary weight behind it, whatever embellishment the “died on the seventeenth step” detail has picked up in the retelling since. The house survived the Civil War physically intact, though accounts describe it being stripped by looters, and local legend — considerably less documented than Winter’s death — claims Union soldiers were killed inside it.

Ownership after Winter’s murder stayed unsettled for a stretch. His widow, Sarah, remained at the house with her mother and siblings until her own death in 1878; her grandmother, Mary Cobb Stirling, died in 1880, passing the estate to her son Stephen, who found it heavily in debt and sold it in 1886. It changed hands twice more in quick succession before Harrison Milton Williams bought it in 1891, and the Williams family held it into the twentieth century, dividing the surrounding land among heirs as the working plantation era wound down for good. That’s the ownership gap between the house’s violent Reconstruction-era history and the version of it that starts generating ghost stories in the 1950s — forty-some quiet years, no recorded hauntings, just a family estate slowly shrinking.

The turn towards what the Myrtles is now began in the 1950s, when owner Marjorie Munson heard a Williams family story about the ghost of an unnamed older woman in a green bonnet or beret. The house changed hands again in the 1970s, and James and Frances Kermeen Myers opened it as a bed-and-breakfast. Later retellings—especially Kermeen’s book—gave the apparition a name, motive and dramatic backstory, while the label “the most haunted house in America” became part of the property’s identity. Television appearances and a National Geographic feature carried the fully formed legend to new audiences. The publicity grew; the supporting evidence did not.

The Evidence Behind the Ghost Stories

There isn’t much left to skeptically dismantle about whether Chloe existed — that ground is already covered, deliberately, up in The History rather than saved for a reveal here. What’s worth spending real time on instead is the mechanism: how a nameless figure in a green turban, first reported by one homeowner in the 1950s, became a fully named woman with a motive, a mutilation, a murder weapon, and a lynching, inside about forty years.

The pattern is traceable almost stage by stage. The 1950s story associated with Marjorie Munson gives you an unnamed older woman in green — no Chloe, no poisoning and no lynching. That is the seed later versions build from. By the time the house changes hands again in the 1970s and opens as a commercial bed-and-breakfast, a business incentive enters the story for the first time: a hotel with a marketable ghost outperforms a hotel with an unexplained shape on the lawn. Frances Kermeen’s book is the point where the story acquires everything a ghost tour actually needs — a name, a face, a grievance, a villain’s arc, and a martyr’s death — plus the marketing masterstroke of the “most haunted house in America” title, which turns a specific claim about one ghost into an unfalsifiable claim about the whole property. Once the story has a title behind it, television does the rest almost automatically: Unsolved Mysteries, Ghost Hunters, Ghost Adventures, each one a fresh audience introduced to the fully-formed version, none of them incentivized to check the Woodruff family papers before airing.

It’s worth saying plainly that the Myrtles isn’t unique in reaching for this particular title. “Most haunted house/place in America” is a crowded marketing category — dozens of properties across the country claim some version of it, and the claim itself functions less as a factual ranking than as a proven booking driver, borrowed by whichever property gets there first with the best story. What sets the Myrtles apart isn’t the ambition of the claim. It’s what the story underneath this particular claim is made of.

What’s notable is what the story borrows to do its work. It isn’t a generic haunting. It’s built from the specific, real vocabulary of chattel slavery — punishment, mutilation, sexual coercion, extrajudicial killing by a mob — and it recruits that vocabulary to build a villain, not a victim. In the version told on the tour, an enslaved woman becomes a murderer, and her death at the hands of other enslaved people becomes just deserts rather than what a lynching actually is. That’s doing something more specific than generic ghost-story exaggeration. It’s borrowing the real horror of slavery for atmosphere, and then quietly inverting who the story asks you to fear.

The Chloe story is not the only piece of Myrtles folklore built this way. The property is also said to stand on a Tunica burial ground, but I could find no published archaeological or archival evidence supporting that claim. It belongs to the later folklore, not the documented history of the site. The Tunica are not a vanished people or a Gothic device: the Tunica-Biloxi Tribe of Louisiana is a present-day federally recognised nation. Turning an Indigenous nation into “cursed ground” repeats the same mechanism as the Chloe story: real historical violence is reduced to atmosphere while the people themselves disappear.

The marketing claim of “twelve ghosts” and “ten murders” deserves the same scrutiny. Historical records support exactly one murder at the Myrtles: William Winter, shot on the front porch on the night of January 26, 1871, while he taught a Sunday school lesson inside. Contemporary newspaper coverage is specific and consistent — a rider approached on horseback, called Winter to the door, and shot him at close range; he died within minutes, not on a dramatic climb to a precise seventeenth stair, though that detail has become fixed in every retelling since. A suspect, reported in period papers only as E.S. Webber, was said to be standing trial; no outcome was ever recorded, and the case remains formally unsolved.

Winter’s murder took place during Reconstruction, when political and racial violence was widespread across Louisiana. Period newspaper reports describe the shooting but do not establish a motive, and no surviving evidence ties it to a ghost story. Later theories about robbery, personal grievance or political violence remain theories. That uncertainty is historically honest; the precise “seventeenth stair” is storytelling.

It’s also worth being honest about why the invented version had so much room to grow in the first place. Enslaved people were not reliably recorded by name in federal records before 1870 — the census counted them, by age and sex, under their enslaver’s name, not their own. That isn’t a minor archival inconvenience. It means the real people enslaved at this plantation are, for the most part, literally absent from the record as individuals, while a woman who never existed at all got a name, a face, and a thirty-year media career. The empty space in the record didn’t stay empty. It got filled with fiction, and the fiction was more useful to a tourism business than the harder work of researching who was actually there ever would have been.

It’s also worth naming the business plainly, the same way it was worth naming at the Tower: this is a functioning hospitality company, and the ghosts are a substantial part of its product line. Overnight guests can book rooms tied to specific legends at a premium over standard rates; the day and evening tours are ticketed separately and run on a fixed schedule; Frances Kermeen’s book, and every television segment that followed it, functioned as free advertising for a brand the property still leans on today. None of that proves the ghosts aren’t real. It does mean the institution telling you these Myrtles Plantation ghost stories, much like the one at the Tower of London, has no real incentive to have this conversation as bluntly as this piece just did.

None of this requires bad faith from every single person who’s told the Chloe story since. Most tour guides are repeating what they were told, in good faith, by guides before them. The bad faith, if it’s anywhere, sits further back — with whoever first decided a nameless shape needed a name, a motive, and a death, and found that version sold better than the truth: that this was a working plantation where real people were enslaved, most of whose names nobody bothered to keep.

My Thoughts

I keep thinking about the empty column in the census.

Every other location this season, the ghost story is layered on top of something the record actually preserves — a name, a date, a documented death, even when the ghost part itself doesn’t hold up. Here, the thing underneath the ghost story is an absence. The Woodruffs enslaved thirty-two people by 1830, and the record that survives them is a number in a column, not a name. Somebody decided that absence needed filling, and instead of doing the harder work — the genealogical digging, the parish records, the kind of research that eventually does recover real names for real enslaved people at other plantations, when someone bothers — they filled it with a story that was easier to tell and much easier to sell.

It isn’t only Chloe, either, once you start looking. The Tunica burial ground claim does the same thing to a different erased history — a real, present-day nation reduced to unnamed atmosphere. Even the one murder here that’s genuinely documented, William Winter’s, gets told with its most disturbing possible context — Reconstruction-era political violence, a Klan-adjacent theory nobody’s proven but nobody’s ruled out either — quietly dropped in favor of a cleaner story about a mystery man on horseback and a body count on the stairs. Three different real histories, three different times the version that made it into the tour script was the one with the politics and the people sanded off.

What gets me isn’t just that the story is false. It’s what the false version chose to be. It didn’t invent a victim. It invented a villain — a woman who murders children, who gets what’s coming to her from a mob of her own people. Whoever built that story took the actual, real vocabulary of slavery’s violence — mutilation, coercion, extrajudicial killing — and pointed it in a direction that let visitors feel a shiver without feeling implicated. You can tour the Myrtles, hear about Chloe, and never once have to sit with the fact that this was a plantation, that people were enslaved here, that most of them left no trace at all. The invented story does the opposite of what an honest one would do. It gives you a villain to be afraid of instead of a system to reckon with.

And it worked. Decades of it. A book title, television features, travel coverage and a business built substantially on the strength of a woman for whom no contemporary evidence has been found. I don’t think most of the guides who tell this story now know they’re repeating a fabrication — I think most of them were told it in good faith, the same way I was, by every single source I read researching this piece. That’s almost the more unsettling part. The story doesn’t need anyone to be lying anymore. It just needs everyone downstream to keep not checking.

So here’s what I can’t get past: a plantation that enslaved real people, most of whose names nobody thought worth recording, has spent seventy years being famous for one who never existed at all. If the industry built on this house’s reputation had to choose between the true story and the marketable one, and it chose the marketable one without most people even noticing there was a choice — what does that tell you about which version of this history was ever actually for sale?

Visiting the Myrtles Plantation

The Myrtles is on US Highway 61 outside St. Francisville, Louisiana, roughly thirty miles north of Baton Rouge and around two hours by car from New Orleans. Public transport is not a practical option. Daytime historical tours and Evening Mystery Tours run throughout the week, but schedules and prices change: check the official tour page before travelling.

Restaurant 1796, Elta Coffee and a gift shop are on site, and rooms and cottages can be booked through the official Myrtles website. You can also check prices and availability for The Myrtles on Trip.com. Several rooms are named after people or legends discussed above. Choose the daytime historical tour for the property’s documented past; the Evening Mystery Tour is explicitly built around its ghost stories.

Before You Go

This is the eighth stop in The Dark Atlas. Continue with the truth behind Borley Rectory, browse more haunted places and folklore, or explore all writing.

Sources

Books and Academic Works

Troy Taylor, “Myrtles Plantation: Legends, Lore and Lies” — archival research on the Woodruff family papers and estate records.

Journalism and Long-form Reporting

Tiya Miles, “Chloe and Cleo: Louisiana Plantation Phantoms”, in Tales from the Haunted South (Amazon UK).

Southern Spirit Guide, “The Rigors of (In)fame(y) — Myrtles Plantation”

American Ghost Walks, coverage of the William Winter murder and contemporary newspaper accounts

Malorie’s Adventures, “The Myrtles Plantation: History vs. Haunting Claims”

Official and Institutional Sources

U.S. Federal Census records, Woodruff household, 1820 and 1830.

Contemporary newspaper coverage of William Drew Winter’s murder, January 1871.