Museum Trail – 1905 Edenton Teapot

The Edenton Teapot

Commemorative Sculpture

1905

 

Bronze teapot - front side

Bronze teapot (front) – photo William Ahearn

Bronze teapot - back side

Bronze teapot (back)

 

The bronze teapot sitting atop a Revolutionary War era cannon commemorates the first political action instigated by a group of North Carolina women.  Not as well known as the Boston Tea Party (where demonstrators wore disguises), Edenton’s version, carried out ten months later on October 25, 1774, was a bold revolutionary statement, unique in a male-dominated political era and public society. Under the leadership of Penelope Barker, fifty-one women publicly signed a petition resolving to abstain from purchasing English tea and cloth.  Once signed, the petition was mailed to King George and its contents made known to the British public.

British cartoon belittling the "Ladies" of Edenton

British cartoon belittling the “Ladies” of Edenton

An account of the event appeared in the Morning Chronicle and London Advertiser on January 16, 1775, along with a drawing portraying the women in a less than flattering light, where the action was satirized in a  political cartoon entitled “A Society of Patriotic Ladies.”.

The bronze teapot is also a testament to one man’s vision and another’s gratitude. Frank Wood, owner of Hayes Plantation, long wished to have an appropriate marker to commemorate the action of the town’s ancestresses. He had just the model, a silver teapot once owned by Samuel Johnston, one of North Carolina’s eighteenth century governors. He was at a loss, however, as to an artisan who could reproduce a larger version of it in bronze.

Frank Baldwin with Mr. & Mrs. Jim Hassett

Frank Baldwin with Mr. & Mrs. Jim Hassett

Providentially, an expert craftsman came into Wood’s life in time to make his dream a reality. In 1905 Frank Baldwin, a foundry man from Watertown, Connecticut and former Union sailor, came to Edenton to revisit the site of the 1864 Civil War Battle of the Albemarle Sound.

Quite by chance he ran into Frank Hassell, Wood’s farm manager, who invited Baldwin to spend some time on Greenfield Farm, which overlooked the scene of the battle in which Baldwin had fought. Hassell turned away Baldwin’s efforts to repay his hospitality, but at the foundry man’s insistence in finding a means to give compensation, put him in touch with Frank Wood. When Wood learned of Baldwin’s expertise, the two struck a deal: a replica of the teapot in honor of the famous Tea Party as an expression of Baldwin’s gratitude.

Although a segment of the teapot’s inscription regarding the exact physical location of the signing of the petition has been called into question it stands, nevertheless, as a beautiful monument to the patriotic courage of Edenton’s first ladies

From NC highway marker info pages:

…There has been much confusion about the Edenton Tea Party, primarily because the event went unrecorded in North Carolina. It remained unknown until 1827, when a North Carolina native naval officer purchased a rendering of the cartoon in a shop abroad. Following his discovery citizens tried to piece together what they believed must have happened in Edenton in 1774.

One of the primary errors is the belief that there was, in fact, a party of 51 women gathered at the home of Mrs. Elizabeth King, a prominent member of the Edenton community. The King home was too small for such an assembly. The wording of the resolution, too, does not indicate a gathering, but rather an agreement. There is no doubt, however, that the ladies of Edenton sent the document to England in 1774, making the resolution among the first public political acts by women in America.”

There is no evidence the event was referred to as a “tea party” until an article was published in the U.S. almost 60 years after it occurred. It was intended to be a courageous show of opposition to the British effort to dominate life in the colonies that would become the United States of America after the brewing revolution.  The first call for revolution occurred in North Carolina, but that is a different story.

Visit and learn more about other sites on the Edenton Museum Trail:

(1) Barker House

(2) Old Colonial Wharf

(3) Joseph Hewes & 1778 Cannons

(4) Hugh Williamson Monument

(5) 1905 Edenton Teapot

(6) 1767 Chowan County Courthouse (no QR code on pedestal sign)

(7) Old Jail

(8) Oldest House in North Carolina

(9) Cotton Mill Village

(10) 1800/1827 James Iredell House (no QR code on pedestal sign)

(11) Kadesh Church

(12) 1736 St. Paul’s Church

(13) 1758 Cupola House

(14) Josephine Leary Building 

(15) 1886 Roanoke River Lighthouse (no QR code on pedestal sign)

Museum Trail – 1800 Iredell House

James Iredell House

U. S. National Registry of Historic Place

James Iredell House

James Iredell House – photo William Ahearn

The home of James Iredell, Sr. (1751-1799) was built in 1773 and was most recently enlarged in 1810, over 200 years ago.  Iredell was the youngest (38) appointee in 1790 when President George Washington named him as a Justice on the first United States Supreme Court. He was confirmed by the U. S. Senate the following day.

Years later, his son, James Iredell, Jr. also resided in the house and became Governor of North Carolina.
The dwelling was erected 1773 by Joseph Whedbee and consisted of the two-story gable front section on the east.
Library in Iredell House

Library in Iredell House – photo Kip Shaw

The interior was architecturally side­-hall in plan. James Iredell, Sr. bought the house in 1778. During the early nineteenth century (1810), lredell’s widow Hannah Johnston Iredell enlarged the house to the present two-story center-hall floor plan.  Covered with beaded weatherboards, the one-bay building had a single exterior chimney. The interior side-hall plan house may have been an expansion of Wilkins’ circa 1756 house, thus explaining the gable’s atypical orientation to the street.

James Iredell came to Edenton as the Deputy Collector for the North Carolina Port of Roanoke at age 17, studied law under Samuel Johnston (a future Governor and U. S. Senator of North Carolina) and married his mentor’s sister, Hannah. He was very active in political matters and was Attorney General of North Carolina by age 28.    Iredell died in 1799 and his wife died in 1826. The house remained in the Iredell family until 1870, then was sold to the McCoy Family.  In 1949 the property was sold to the Edenton Tea Party Chapter of the Daughters of the American Revolution.
During the early nineteenth century, Iredell’s widow added the two-story center-hall plan block on the west. The house remained in Mrs. Iredell’s possession until her death in 1826, and in her daughter’s estate until 1870. Saved in 1948 by the Edenton Tea Party Chapter, National Society Daughters of the American Revolution, the house is owned and managed by the State of North Carolina. The James Iredell Association assisted in restoring and furnishing the building.  The circa 1829 Bandon Plantation office, now furnished as a school, was moved in 1964 in appreciation of novelist Inglis Fletcher, who wrote part of the Carolina series here. Although the Bandon Plantation house was destroyed by fire in 1964, the kitchen was deconstructed and moved here in 2009. Reconstruction was completed in 2013.
Edenton Tea Party Chapter DAR history:
The Edenton Tea Party Chapter DAR was chartered on February 27, 1948 and their first act of business was to acquire and restore the home of James Iredell. These 20 women signed bank notes to make this acquisition a reality and to allow preservation of this residence. In 1951 the house was turned over to the State of North Carolina Cultural Resources and today the house and Edenton Visitor Center are all part of the Edenton Visitor Site. The house has been completely furnished by the James Iredell Association.
You asked for a funny story!!!!!
These 20 ladies are often compared to the 51 Tea Party Ladies. They went to the Banker told him their business and signed the bank notes to purchase the house. The next morning there were 20 husbands at the bank wanting to know why this banker had let these 20 ladies sign bank notes. The banker’s wife was one of these ladies. There were almost 20 divorces. But the Edenton Tea Party saved the house and it is an architectural gem that Edenton is proud of today.
There is a DAR marker on the lawn of the James Iredell House in recognition of the Charter member’s effort in 1948 to save the house.

 

Visit and learn more about other sites on the Edenton Museum Trail:

(1) Barker House

(2) Joseph Hewes & 1778 Cannons

(3) 1905 Edenton Teapot

(4) 1767 Chowan County Courthouse (no QR code on pedestal sign)

(5) 1800/1827 James Iredell House (no QR code on pedestal sign)

(6) Kadesh Church

(7) 1736 St. Paul’s Church

(8) 1758 Cupola House

(9) Josephine Leary Building 

(10) 1886 Roanoke River Lighthouse (no QR code on pedestal sign)

Historic Trail Map WA

Museum Trail – 1767 Courthouse

1767 Chowan County Courthouse and Jail

 

Edenton Court House with CrowdIn November of 1712, the Colonial Assembly passed an act “to promote the building of a courthouse to hold the assembly in, at the fork of Queen Anne’s Creek”, essentially establishing the town to be the seat of the provincial government.   By 1718, the first courthouse was completed, apparently unsatisfactorily, and a second building was constructed in 1724. The location of that 1724 building was on the north side of East King Street, the location of the new courthouse built in 1767.

Color Photo of Chowan County Courthouse

The Chowan County Courthouse is considered to be the finest Georgian courthouse in the South, with a central courtroom and side appendages and one of the most important public buildings in colonial America. It is the oldest public building in North Carolina. The design and construction, while unconfirmed, are attributed to John Hawks, the architect of Tryon Palace, who was active in the period and region.

1767 Courthouse Courtroom

1767 Courthouse Courtroom

Based on classical British architecture and built by local artisans using local materials, the T-plan building is laid in Flemish bond brick over an English bond water table. The entrance leads directly into the courtroom. The assembly room that is sometimes called the ballroom occupies the central portion of the second story.  The second floor is furnished with a large assembly room appointed with raised paneling woodwork and fireplaces on either side.

Standing at the head of the Courthouse Green since 1767, the Chowan County Courthouse is one of the most important surviving buildings from the colonial period in North Carolina.  The Courthouse was the most important building for public and community activities in Chowan County and Edenton from the late 18th century into the 21st century.

The oldest courthouse still in active use the state, it has always been used for many purposes. Here Edenton citizens met during the struggle for independence. It was the county’s seat of government for over two hundred years, and the venue for many public meetings, religious services, and a wide range of social activities.

Patriots who founded the American Republic, relied on this courthouse for public business and community affairs. Joseph Hewes was one of the commissioners appointed to raise money for its construction. James Iredell and Samuel Johnston pleaded their cases before judges presiding here.  Joseph Hewes was also a signer of the Declaration of Independence.  Hugh Williamson, signed the U.S. Constitution and was later a member of the U. S. House of Representatives in the 1st and 2nd Congresses.  Samuel Johnston served as Governor of North Carolina from 1787 – 1789 and U.S. Senator from 1789 – 1793. James Iredell, used his knowledge of the law and Constitution as one of George Washington’s  appointees to the first U. S. Supreme Court from 1790-1799,

In 1792, Congress established Edenton as a place for holding U.S. District Court. In April 1819, President James Monroe visited Edenton and attended a public dinner at the Courthouse.

Chowan County Courthouse and the North Carolina Courthouse in Raleigh are the only places where the North Carolina Supreme Court can convene.

Chowan County Courthouse Green

Chowan County Courthouse Green

Courthouse Green

East King Street

The Courthouse Green, public property since the town of Edenton was laid out in 1712, continues as a location for memorials, celebrations, and public events. To the north is the Historic Chowan County Courthouse; at the southern end is a large marble memorial to Edenton merchant Joseph Hewes (1730-1779), one of the three North Carolina signers of the Declaration of Independence.

To the west across Colonial Avenue, a handsome bronze teapot commemorating the Edenton Tea Party of October 25, 1774 sits on a pedestal beside the fence surrounding the Homestead. Frank Baldwin, a skilled craftsman in a Watertown, Connecticut brass foundry, made the teapot circa 1905 for Frank Wood, owner of the Homestead. The design was based on a teapot that had been in the Wood family for years.

Public events occur frequently on the Green

Public events occur frequently on the Green – photo Kip Shaw

At the foot of the green are three mounted Revolutionary Was cannons acquired by American arms merchants in France. These were among forty-five pieces a Swiss captain brought to Edenton in 1778 since the British blockade made delivery to the intended ports too dangerous. Twenty-two made their way up the Chowan River to Virginia.

 

1905 Jailer’s House with 1825 Jail in background

Jailer's House with Jail in background

Jailer’s Residence and Jail (in background) – photo William Ahearn

The Chowan County Jail is the oldest documented jail in North Carolina. Built in 1825, it was the fifth jail in the county, with earlier prisons erected around 1722, 1741, 1788, and 1809.

The brick structure, raised in English bond, has front and rear walls which are 29 inches thick on the first story and 24 inches thick on the second story; the end walls of each story are 18 inches thick. Steel cells, added in 1905, remain essentially unaltered. This jail was used until the new regional jail was built in 1979-1980.

The Jailer’s Residence is a modestly finished brick Victorian dwelling that was erected by the county as the home of the keeper of the Chowan County Jail at a cost of $2,050. Conveniently located between the jail and the Courthouse, the jailer occupied the building until the early 1970s.

The Historic Edenton State Historic Site offers daily tours of a remarkable collection of Georgian, Federal, and Greek Revival buildings including the 1767 Courthouse, the Lighthouse and the Iredell House.

Visit and learn more about other sites on the Edenton Museum Trail:

(1) Barker House

(2) Old Colonial Wharf

(3) Joseph Hewes & 1778 Cannons

(4) Hugh Williamson Monument

(5) 1905 Edenton Teapot

(4) 1767 Chowan County Courthouse (no QR code on pedestal sign)

(7) Old Jail

(8) Oldest House in North Carolina

(9) Cotton Mill Village

(10) 1800/1827 James Iredell House (no QR code on pedestal sign)

(11) Kadesh Church

(12) 1736 St. Paul’s Church

(13) 1758 Cupola House 

(14) Josephine Leary Building 

(15) 1886 Roanoke River Lighthouse (no QR code on pedestal sign)

 

 

 

Museum Trail – 1782 Barker House

U. S. Department of Interior designation plaque

The Penelope Barker House is Edenton & Chowan County’s Welcome Center – Admission is FREE. Keep up on activities at the Barker House on Facebook. While there, sign up to become a Fan of the Barker House.

The House:

* Thomas and Penelope Barker built the house in 1782. It originally consisted of a parlor wing and half of the hall. The house had three additions between 1782 and the 1840s.

* The house was sold in 1830 to Augustus Moore who added onto it three times.

* The woodwork is Federal in the street-front parlor, Georgian in the sound-front parlor, and Greek Revival in the dining room and kitchen.

* The house has three floors and eight fireplaces.

* Originally located two blocks north, the house was rolled to its present site in 1952. Once it arrived, it was turned 180 degrees so the front on the house now faces the water. Visitors enter through the back door, which in the South means you are very welcome and expected to come back often.

* The Moore family and descendants lived in the Barker House until 1952 when it was sold to Haywood Phthisic who bought it with intentions of donating it to an organization for renovation and preservation.

* That same year, the Junior Chamber of Commerce, Business and Professional Women’s Club, and Edenton Woman’s Club raised monies to purchase the home, move it to its present site, and restore it for public use. The Edenton Historical Commission maintains and operates the Penelope Barker House Welcome Center.

The Edenton Tea Party:

* On October 25, 1774, 51 women signed their names to a petition resolving not to buy or use British goods on which they had to pay taxes.

* It is thought that they signed at the home of Elisabeth King, who did not sign.

* Edenton was a cosmopolitan town as evidenced by the fact that 51 women could read and write. At that time, only one-quarter of colonists were literate and women were even less likely than men to be educated.

* This was the first political action by women in the Colonies and caused quite a stir at home and abroad! A contemporary account, along with the resolution and its signer’s names, appeared in the January 16, 1775 Morning Chronicle and London Advertiser. A caricature of the Edenton women appeared in a London newspaper with English politicians’ heads on women’s bodies. A copy of the cartoon hangs today in the entry hall of the Barker House.

* Their resolution, to show their support of North Carolina’s Provisional Assembly, read:

“As we cannot be indifferent on any occasion that appears nearly to affect the peace and happiness of our country, and as it has thought necessary, for the public good, to enter into several particular resolves by a meeting of Members deputed from the whole Province, it is a duty which we owe, not only to our near and dear connections who have concurred in them, but to ourselves who are essentially interested in their welfare, to do everything as far as lies in our power to testify our sincere adherence to the same: and we do therefore accordingly subscribe this paper, as a witness of our fixed intention and solemn determination to do so.”

Penelope Barker:

* Born in Chowan County in 1728, the daughter Samuel Padgett and Elizabeth Blount.

* Penelope and her sister Sarah inherited Blenheim Manor upon the death of their father. Artifact found at the former site of Blenheim Manor have brought more attention to the ruins.

* Penelope married John Hodgson at 19 and was expecting her second child when Hodgson died in 1747. She inherited substantial property from Hodgson.

* In 1752, she married James Craven. Before marrying, he purchased Penelope’s property from her. Craven died in 1755, leaving Penelope all his property, including what he had originally bought from her.

* In 1756, when she was just 28, she married Thomas Barker, then 44 years old. They had three children, all of whom died young.

* On October 25, 1774, Mrs. Barker organized fifty-one Edenton-area women and all signed a petition in open protest to the British Parliament’s Tea Act of 1773, which was sent to King George. Almost 60 years later, the event was recognized as the “Edenton Tea Party” but it was no party, it was a visible act of protest.

* When Thomas died, he left Penelope all his property, which included town lots, two plantations, 33 mahogany chairs, 53 slaves, watches, horses and 400 books.

* Penelope lived through many personal tragedies, showed unbelievable courage in organizing the Edenton Tea Party  and died in 1796.

* Penelope and Thomas are buried in the Johnston family graveyard at Hayes Plantation, Edenton.

Thomas Barker:

* Thomas was born 1713 in Massachusetts or Rhode Island.

* He moved to Bertie County in 1734 to read law. When he traveled the “circuit,” he spent much time in Edenton and Chowan County.

* In 1738, he married Ferebee Savage Pugh, and they had one child, Betsy.

* In 1756, he married Penelope Padgett Hodgson Craven.

* In 1761, he left for England and, because of the American Revolution,  was forced to stay there there until 1778, when he was able to escape via France. He had been away for seventeen years.

* He died in 1789.

Penelope Barker portrait- photo William Ahearn

Penelope Barker

Thomas Barker portrait (showing forehead indentation from horse accident)

Barker House being moved down Broad Street

Barker House at foot of Broad Street after being turned 180 degrees so front now faces the water

A rare Edenton sight, snow at the Barker House

Edenton’s Living Room (aka Penelope Barker House Welcome Center’s Parlor) – photo Kip Shaw

Museum Trail – 1758 Cupola House

The Cupola House

National Historic Landmark (1970) with National Significance

 

For over 250 years, the Cupola House has stood watch over the Edenton Bay. Built in 1758 by Francis Corbin, land agent for John Carteret, Earl of Granville; one of the last of the famous Lords Proprietors.  Granville, in the 17th century, acquired vast territories south of Virginia from King Charles II.

1900 Cupola House

1900 Cupola House

Corbin died in 1767, and Dr. Samuel Dickinson purchased the house in 1777. His descendants called the Cupola House home for over 141years. The house was originally clad in rusticated siding-wood carved to look like stone. There was also a Widow’s Walk around the massive Cupola where the Dickinson-Bond women had tea.

The wealth of the family declined to where the last of the family were unable to properly maintain the house. The Cupola House’s once formal gardens were sold for commercial development until only 10 feet of space remained beside the house. Exterior paint was worn away, and the building was suffering from disrepair. Its loving, but impoverished, owners found no recourse but to sell off family treasures; even the magnificent first floor Georgian woodworks were sold to the Brooklyn Museum. The house was threatened, and a historic landmark was near death.

In 1918, citizens rallied to form an organization to save the Cupola House. This organization eventually became The Cupola House Association, still today dedicated to its protection.

Today the Cupola House stands proudly watching over the bay. The formal colonial-style gardens have been restored. Inside the magnificent woodwork has been restored and local citizens have returned many of the Dickinson artifacts. The interior is furnished with fine furniture dating from 1680 to 1820- the 1750s London tall case clock has been in the house since 1777,the 1760s whale oil lamp is the first in Edenton, and the Dickinson china is secure in the barrel-back cupboard.

It is an anchor to the lovely town of Edenton and a memorial to those who cared and took action to save this architectural treasure.

Spring party on the lawn of the designated National REgistry of National Treasurer with Significance

Spring party on the lawn of the Cupola House, designated by the US Department of Interior as a National Historic Landmark (1970) with National Significance – photo Kip Shaw

The house is of Jacobean design and is known as the finest example of this architectural type south of New England. Corbin died in 1776, and the Cupola House was sold to Dr. Samuel Dickinson who, was married to Elizabeth Penelope Ormond, one of the women who signed the “Edenton Tea Party” resolutions. (Another stop on the “Museum Trail”)

Best Dining Room in Edenton

Best Dining Room in Edenton – photo Bob Quinn

Historically accurate color

Historically accurate colors – photo Kip Shaw

Decked out for the annual Edenton Christmas Tour

Decked out for annual Edenton Christmas Tour – photo Kip Shaw

 

 

 

 

 

 

 

 

The local citizen’s endeavor to save the Cupola House is the earliest community preservation effort in North Carolina to save an historic structure. The Cupola House today is a true treasure of colonial history. Its beautiful woodwork has been reproduced exactly as the original by local craftsmen and women — the Cupola House is a delight to visit.

In 2008 the Cupola House experienced its 250th anniversary and a total restoration, inside and out. The Association leadership submitted grant request to the “Save America’s Treasures” Program of the National Park Service and awarded a matching grant, that grant along with 6 others, local contributions and matching-in-kind funds, paid for a $339,000 make over. Today the house is in superior shape including a shingle roof of Alaskan cut pine hand cut in Virginia by craftsmen who specialize in historic properties such as Mt Vernon, Monticello, and Williamsburg

Annually, visitors from around the world come to Edenton to view the Cupola House and enjoy its colonial heritage gardens. The gardens are part of the Heritage Gardens of America and are tended with loving care by volunteers who call themselves, “Weeders”. The guided tour of the elaborate interior and its important collections of period furniture and textiles is truly breathtaking!

These collections, in accompaniment with the extraordinary architectural aspects of the building are crucial to enable an understanding of the Colonial era history. Students and professors visit regularly to study the collections and the building’s unique architecture featuring its exceptional interior finishes.  You may join them most weekends or with a tour guided by docents of the Historic Edenton State Historic site.  Tickets available at: 108 N Broad St, Edenton, NC 27932.

Visit and learn more about other sites on the Edenton Museum Trail:

(1) Barker House

(2) Joseph Hewes & 1778 Cannons

(3) 1905 Edenton Teapot

(4) 1767 Chowan County Courthouse (no QR code on pedestal sign)

(5) 1800/1827 James Iredell House (no QR code on pedestal sign)

(6) Kadesh Church

(7) 1736 St. Paul’s Church

(8) 1758 Cupola House 

(9) Josephine Leary Building 

(10) 1886 Roanoke River Lighthouse (no QR code on pedestal sign)

 

Historic Trail Map WA

 

Essay by Trinity Copeland

The Hero in a Dress

By Trinity Copeland

 

 

 

There are a few reasons why I chose to write about Penelope Barker . I chose to write about her because I wanted to learn about a girl

in history, for a change, instead of a boy. I also chose her because she is a girl and I am a girl!  I wanted to feel like I could be a hero!

Penelope Barker was born in Chowan County, North Carolina in 1728. While she was still in her teens, her father and sister both died . Penelope began to handle her father’s plantation and took over her sister , Elizabeth ‘ s spot, caring for her children.

She married John Hodgson at a young age and was widowed at age 19. When her husband, John, died, she was left with two children of her own and a niece and two nephews, from her husband’s previous marriage. Penelope also had a large estate to manage.

Not too long after, Penelope married James Craven and they had three children. When she was 27 years old James died . He had no other family , so she got all of his estate and became the richest woman in North Carolina.

Thomas Barker was Penelope’s third husband. They had three children, but none of them lived for more than 11months. Penelope Barker was a one-of-a-kind woman in lots of ways, but she is mostly known for one thing . She is famous for hosting the Edenton Tea Party. She and 50 other women signed a declaration. The declaration protested unfair British taxes on the colonists. There was a Boston Tea Party, which inspired Penelope to host the one in Edenton. Penelope helped to make a change in our community and she is still a hero today because of what she did.

Penelope and the 50 women said this: “We, the ladyes of Edenton do hereby solemnly(sol-um-ly)  engage not to conform to ye pernicious (per­ ni-shus)  custom of drinking Tea or that we, the aforesaid (a-for-sed) Ladyes, will not promote ye wear of any manufacture from England until such time acts which ten to enslave this our Native country shall be repealed.” Barker said, “Maybe it has only been men who have protested the king up to now.  That only means we women have taken too long to let our voices be heard.  We are signing our names to a document, not hiding ourselves behind costumes like the men in Boston did at their tea party. The British will know who we are.”

Penelope Barker had lots of great character traits. Here are some:

She was caring because she took care of her sisters. She was brave, because she was the first woman to stand up against politics. She was courageous because she stood up for what she believed. She believed in fair government taxes. Penelope showed responsibility when she took over her father’s plantation, property and children. She also had to take care of her sister’s children when she died. Penelope could be called spunky, daring, bold, strong, resilient and irrepressible. Penelope died in 1796, she had lived a good life and made a change in the lives of her community. That makes her a hero!

Penelope Barker House

The Penelope Barker House, Edenton & Chowan County’s Welcome Center – Admission is FREE
Barker House NewKeep up on activities at the Barker House on Facebook.  While there, sign up to become a Fan of the Barker House.

  • Thomas and Penelope Barker built the house in 1782
  • It originally consisted of a parlor wing and half of the hall
  • The house had three additions between 1782 and the 1840s.
  • The woodwork is Federal in the street-front parlor, Georgian in the sound-front parlor, and Greek Revival in the dining room and kitchen.
  • The house has three floors and eight fireplaces.
  • Originally located two blocks north, the house was rolled to its present site in 1952. Once it arrived, it was turned 180 degrees so the front on the house now faces the water. Visitors enter through the back door, which in the South means you are very welcome and expected to come back often.

THOMAS BARKER

  • Thomas was born 1713 in Massachusetts or Rhode Island.
  • He moved to Bertie County in 1734 to read law. When he traveled the “circuit,” he spent much time in Edenton and Chowan County.
  • In 1738, he married Ferebee Savage Pugh, and they had one child, Betsy.
  • In 1756, he married Penelope Padgett Hodgson Craven.
  • In 1761, he left for England and stayed there until 1778, the last few years as colonial agent for the colony. Because of the American Revolution, Thomas was forced to remain in England until 1778.
  • He died in 1789.

PENELOPE BARKER

  • She was born in Chowan County in 1728, the daughter Samuel Padgett and Elizabeth Blount.
  • Penelope and her sister Sarah inherited Blenheim Manor upon the death of their father. Recent artifact finds at the former site of Blenheim Manor have brought more attention to the ruins.
  • Penelope married John Hodgson at 19 and was expecting her second child when Hodgson died in 1747. She inherited substantial property from Hodgson.
  • In 1752, she married James Craven. Before marrying, he purchased Penelope’s property from her. Craven died in 1754, leaving Penelope all his property, including what he had originally bought from her.
  • In 1756, when she was just 28, she married Thomas Barker then 44 years old. They had three children, all of whom died young.
  • When Thomas died, he left Penelope all his property, which included town lots, two plantations, 33 mahogany chairs, 53 slaves, watches, horses and 400 books. Penelope lived through many personal tragedies, showed unbelievable courage in organizing the Edenton Tea Party (below) and died in 1796.
  • Penelope and Thomas are buried in the Johnston family graveyard at Hayes Plantation, Edenton.

THE EDENTON TEA PARTY

  • On October 25, 1774, 51 women signed their names to a petition resolving not to buy or use British goods on which they had to pay taxes.
  • They signed at the home of Elisabeth King, who did not sign.
  • Edenton was a cosmopolitan town as evidenced by the fact that 51 women could read and write. At that time, only one-quarter of colonists were literate and women were even less likely than men to be educated.
  • The first political action by women in the Colonies caused quite a stir at home and abroad! A contemporary account, along with the resolution and its signer’s names, appeared in the January 16, 1775 Morning Chronicle and London Advertiser. A caricature of the Edenton women appeared in a London newspaper with English politicians’ heads on women’s bodies. A copy of the cartoon hangs today in the entry hall of the Barker House.

Their resolution, to show their support of North Carolina’s Provisional Assembly, read:

“As we cannot be indifferent on any occasion that appears nearly to affect the peace and happiness of our country, and as it has thought necessary, for the public good, to enter into several particular resolves by a meeting of Members deputed from the whole Province, it is a duty which we owe, not only to our near and dear connections who have concurred in them, but to ourselves who are essentially interested in their welfare, to do everything as far as lies in our power to testify our sincere adherence to the same: and we do therefore accordingly subscribe this paper, as a witness of our fixed intention and solemn determination to do so.”

Penelope Barker

Mother of Women’s Political Liberation

Penelope Barker may well have been one of the most courageous women in US history. At a time when women simply did not publicly engage in politics, she held a tea party. But it was not really a tea party; it turned into the first woman’s political activity in colonial American history — in fact, in western history. At the time of Barker’s tea party, October 25, 1774, women in Europe did NOT engage in political discourse either. For many reasons, Hillary Clinton, Condoleezza Rice, Sarah Palin, Margaret Thatcher and every woman who has ever expressed an opinion at a school board or town hall meeting or in the halls of the Congress owes gratitude to Penelope Barker for what she did in Edenton, NC, to advance the rights of women to engage in politics.

Early Life

Young Penelope

In the years 1714-1722, Dr. Samuel Pagett received about 2,000 acres in Proprietary Land Grants, including or what was already Blenheim Plantation. This area is listed on Edward Mosley’s map of 1733. It is possible that the land was granted about the same time Gov. Charles Eden (Town of Edenton namesake) appointed Pagett Justice of the Peace. Dr. Pagett was married to Elizabeth Blount, whose father was Thomas Blount, Justice of Chowan Precinct Court.(1) They had three daughters, Elizabeth, Penelope, and Sarah. Penelope was born at Blenheim Manor in Chowan (Cho-WAN) County on June 17, 1728.

Marriages

While still a teen, Barker’s father and sister Elizabeth died in the same year, leaving Barker to help raise Elizabeth’s children, Isabella, John, and Robert. Barely seventeen years old, Barker married her sister’s widowed husband, John Hodgson, in 1745.(2) She was pregnant with their second of two sons (Samuel and Thomas) when John died just two years later, leaving her with five children to raise.(3) Penelope qualified as Hodgson’s administratrix, with her uncles, John and James Blount, as sureties. Her uncle, Charles Blount, panicked when, in checking her accounts, he discovered that she had little cash on hand to pay her husband’s debts. Moreover, the court threatened to take both the children and their property away from her, believing that she, a twenty-one-year-old widow with five children ranging from ages five to twelve, was not rearing and educating them satisfactorily. Subsequently, Peter Payne was appointed guardian of John Hodgson and Charles Blount of Robert Hodgson. In October 1751, the guardianship of Isabella, John, and Robert was restored to Penelope.(4)

For six years, Penelope managed her affairs by herself. In 1751, she remarried to a farmer and politician, James Craven, a native of Droughton in Yorkshire, England. Although the date of his immigration to North Carolina is unknown, he was serving as Clerk of Court in Pasquotank Precinct in 1734. He was sworn in as Clerk of Court in Chowan County in January 1740, and the following month, he was elected to the Lower House of the Assembly from Edenton. He continued in that office for six years. They had no children together before he died in 1755. Before marrying, he purchased Penelope’s property from her, which he left to her upon his death, along with all of his other property.(5) Read his will HERE. At this point, Penelope Padgett Hodgson Craven was, no doubt, one of the wealthiest women in North Carolina, if not all the Colonies.

Thomas Barker was born in the colony of Rhode Island and came to North Carolina “as the skipper of a New England bark”. In all probability, Barker was serving as the on-board factor for a New England commercial firm, sailing with a cargo to supervise its sale and to oversee the purchase of goods for the firm. Barker arrived in Edenton in 1735 and immediately entered upon a career in public service. In April 1741, he secured from Governor Gabriel Johnston a six-hundred-acre land grant in Bertie County and soon thereafter crossed the Roanoke River from Edenton to take possession of this land. Early in 1742, having settled his plantation, he married Pheribee Pugh (née Savage), widow of Francis Pugh, wealthy planter, shipowner, and member of the governor’s council. To this marriage, one daughter, Betsy, was born. Early in the 1750s, following his wife’s death, he reestablished his residence in Edenton and was elected as that borough’s representative to the Assembly in 1754. His devotion to his constituents in Edenton was evidenced by his bills to improve the inspection of exports at the port, to confirm existing land patents in the area, to repair the county’s courthouse, and to erect St. Paul’s Church in the parish. Barker’s high standing in the Assembly is best shown by his repeated selection to chair the committee of the whole and by his appointment as a member of the Committee of Correspondence in December 1758. By the late 1750s, Barker had established himself as a successful attorney in Edenton and a political leader of prominence throughout the colony. In Edenton, a thriving commercial center during the 1750s, Barker met and was attracted to Penelope Craven.(6) In 1757, at 28 years of age and having been widowed twice, Penelope Craven married Thomas Barker, who was 16 years her senior.

Trials

Thomas Hodgson

Together, the Barkers had three children, all of whom died before reaching their first birthdays. They were Penelope (survived less than two months), Thomas (less than nine), and Nathaniel (less than ten). In fact, by 1761, seven of the nine children she had either borne or for whom she was responsible had died, and in 1772 her son, Thomas Hodgson, died at the early age of twenty-five. The lone surviving child, Betsy Barker, also left Penelope’s care through marriage to Colonel William Tunstall, a prominent planter of Pittsylvania County, Va.(7)

In 1761, just four short years after Thomas and Penelope were married, Penelope found herself with only two of the nine children still living, and her husband leaving the country. A controversy erupted between the Assembly and Governor Dobbs over the selection of an agent to represent the colony in London. Thomas Barker was the overwhelming choice of the lower house, while the governor and council countered with Samuel Smith. Efforts at compromise proved futile, and both men were ultimately sent to London, Smith to represent the council and Barker the Assembly. The period must have proved a trying one for Barker, separated from his wife and friends in a strange land.

As the Anglo-American problems deepened and the movement toward open colonial rebellion accelerated, Barker’s role in London became increasingly more difficult, just as his concern for his wife and friends in North Carolina grew more acute. In June 1775, Barker and Elmsley achieved for North Carolina an exemption from the Trade Restraining Act, which had been passed by Parliament two months earlier. However, news of this exemption was received with hostility by North Carolinians, who had forced the flight of Governor Josiah Martin and who had recently heard of the opening skirmish of the Revolution at Lexington and Concord. No doubt disillusioned by the totally unexpected, bitter denunciations of his efforts and deeply concerned about his wife’s welfare, Barker, in 1776, wanted only to return to America. However, the outbreak of open warfare in the colonies and the British blockade of America made his immediate return impossible. In late spring of 1778, he was able to leave England for France and on July 8, 1778, sailed from Nantes for North Carolina.(8)

Triumph

During her husband’s 17-year absence, Penelope managed their affairs, while keeping an eye on what was happening between the Colonies and England. As the colonial independence movement grew in intensity, with an increase in riots and other extralegal activities in opposition to British taxation, Penelope, because of her husband’s position, probably realized as much as any other person in North Carolina the potential costs of such actions.9 Throughout the colonies, patriot leaders urged women, in their role as consumers, to support the rebellion by boycotting British imports such as cloth and tea, in keeping with the non-importation resolutions passed by the First Continental Congress in 1774. On October 25, 1774, it is thought that Barker gathered 50 women in the home of Elizabeth King, where they signed a resolution supporting the boycott as they drank tea made from mulberry leaves, lavender, and other local herbs. The event, dubbed the Edenton Tea Party, is historically significant because it is the first recorded women’s political demonstration in America.(10)

The better-know Boston Tea Party was conducted by men, wearing costumes to protect their identity. At the time, according to Diane Silcox-Jarrett, Barker said:

“Maybe it has only been men who have protested the king up to now. That only means we women have taken too long to let our voices be heard. We are signing our names to a document, not hiding ourselves behind costumes like the men in Boston did at their tea party. The British will know who we are.” (11)

Barker rejected the notion of hiding and instead publicly affixed signatures to the proclamation, which was published in London newspapers. The actions of the 51 ladies in Edenton were reported in a very negative fashion, even alleging the women were bad mothers and/or loose women. An enlarged copy of one famous cartoon hangs today in the Barker House in Edenton. Other women in the Colonies thought otherwise and emulated Barker and the women of Edenton by boycotting British goods, which soon got the attention of the Crown.

Penelope’s proclamation read:

“As we cannot be indifferent on any occasion that appears nearly to affect the peace and happiness of our country, and as it has been thought necessary, for the public good, to enter into several particular resolves by a meeting of Members deputed from the whole Province, it is a duty which we owe, not only to our near and dear connections who have concurred in them, but to ourselves who are essentially interested in their welfare, to do everything as far as lies in our power to testify our sincere adherence to the same; and we do therefore accordingly  subscribe this paper, as a witness of our fixed intention and solemn determination to do so.”

The Barker House

Upon her husband’s return from England, Penelope and Thomas built the house that would become what we know today as the Penelope Barker House Welcome Center. Life came somewhat full circle for Penelope, in that the land where they built their home belonged to her previous husband. In the area that is now 209 South Broad Street, there were five lots. James Craven brought them all under single ownership in 1748. He sold them to, and then bought them back from the widow, Penelope Hodgson, whom he later married. After his death in 1756, Penelope inherited the property back and sold it to Thomas Barker, whom she later married. The 1769 Sauthier Map shows on these lots an earlier dwelling, numerous ancillary buildings, and a formal arrangement of gardens. Evidence is mounting that this earlier structure was actually incorporated into the current house when it was built in 1782.(12) You can read more about the house and the changes it has seen on the house’s Museum Trail page.

Thomas and Penelope had eight years together before he passed. Widowed for the third time, Penelope lived out her years in Edenton in the house they built together. She passed away in 1796 and then was buried next to Thomas in the Johnston family graveyard at Hayes Plantation, Edenton.
Penelope Barker’s former home now serves as the home of the Edenton Historical Commission, 505 South Broad St., Edenton, NC. If you have an ancestor among the signers of Penelope Barker’s proclamation, thank them for advancing the cause of women’s right to engage in politics, not to mention that matter of Independence for the United State.

The 51 signers included several relatives with comparable or similar names (as published in the Virginia Gazette, the London Packet, and the Morning Chronicle and London Advertiser):

Abigail Charlton

F. Johnston

Margaret Cathcart

Anne Johnston

Margaret Pearson

Penelope Dawson

Jean Blair

Grace Clayton

Frances Hall

Mary Blount

Elizabeth Creacy

Elizabeth Patterson

Jane Wellwood

Mary Woolard

Sarah Beasley

Susannah Vail

Elizabeth Vail

Elizabeth Vail

Mary Jones Anne Hall

Rebecca Bondfield

Sarah Littlejohn

Penelope Barker

Elizabeth P. Ormond

M. Payne

Elizabeth Johnston

Mary Bonner

Lydia Bonner

Sarah Howe

Lydia Bennet

Marion Wells

Anne Anderson

Sarah Matthews

Anne Haughton

Elizabeth Beasley

Mary Creacy

Mary Creacy

Ruth Benbury

Sarah Howcott

Sarah Hoskins

Mary Littledle

Sarah Valentine

Elizabeth Crickett

Elizabeth Green

Mary Ramsay

Anne Horniblow

Mary Hunter

Teressa Cunningham

Elizabeth Roberts

Elizabeth Roberts

Elizabeth Roberts

Endnotes

1. James Williams, Chowan County NC Archives Photo Place….Blenheim Manor House, http://files.usgwarchives.net/nc/chowan/photos/blenheim983ph.txt, 2009.

2. Debra Michals, PhD, Penelope Barker, www.womenshistory.org, 2015.

3.Donna Campbell Smith, Penelope Barker: First American Female Political Activist, https://hubpages.com/education/Penelope-Barker-and-the-Edenton-Tea-Party, 2019.

4. Vernon O. Stumpf, Hodgson, John, https://www.ncpedia.org/biography/hodgson-john, 1988.

5. William S. Price, Jr., Craven, James, https://www.ncpedia.org/biography/craven-james, 1979.

6. Michael G. Martin, Jr., Barker, Thomas, https://www.ncpedia.org/biography/barker-thomas, 1979.

7. Michael G. Martin, Jr., Barker, Penelope, https://www.ncpedia.org/biography/barker-penelope, 1979.

8. Ibid., Martin, Barker, Thomas.

9. Ibid., Martin, Barker, Penelope.

10. Ibid., Michals, Penelope Barker.

11. Diane Silcox-Jarrett, “Penelope Barker, Leader of the Edenton Tea Party,” Heroines of the American Revolution, America’s Founding Mothers, Chapel Hill, North Carolina: Green Angle Press, 1998.

12. Thomas R. Butchko, Edenton An Architectural Portrait …, Edenton, NC: Edenton Women’s Club, 1992, p.120-121.

Museum Trail – Hugh Williamson Monument

Edenton was home to several colonial-era leaders with many talents, such as:

  • A delegate to the Constitutional Convention from North Carolina and one of 39 signers of United States Constitution in 1787, exactly 200 years after the “Lost Colony” disappeared at Manteo;
  • Author, in 1770, of one of the first books on climate change: (Observations on the Climate in Different Parts of America, Compared With the Climate in Corresponding Parts of the Other Continent: To Which Are …);
  • Was ordained a Presbyterian Minister;
  • Studied medicine at Edinburgh, London, and earned his MD from the Univ. of Utrecht, Netherland and practiced medicine in Edenton, NC and served as the Surgeon General of the NC Militia during the Revolutionary War;
  • Was a member of American Philosophical Society with Benjamin Franklin;
  • Wrote “An Essay on Comets” for which he was awarded the Doctorate of Law degree by University of Leyden;
  • Happened to be in Boston on his way to London when he witnessed the Boston Tea Party. Immediately afterward, John Hancock – the wealthy ship owner and statesman from Massachusetts – put him on one of his fastest ships to report news of the event to King George III and the Privy Court, which he did warning that repression would lead to revolution;
  • During the Revolutionary War he went behind enemy lines to care for the wounded and also convinced the British to inoculate against smallpox;
  • His speech, “Remarks on New Plan of Government” promoting ratification of draft Constitution was distributed around the nation and a version delivered on the steps of the still standing 1767 Historic Courthouse at Edenton convinced NC to endorse the new Constitution, with a Bill of Rights, even after the state had initially declined to do so;
  • Served as the first Sec. Board of Trustees of University of NC, America’s first State University;
  • Named trustee NY College of Physicians, Founder of NY’s Literary and Philosophical Society; and
  • Authored “History of North Carolina”, which was published in 1812.

The amazing thing about this list is that one man, Hugh Williamson, possessed all of these attributes. He may well be the “most fascinating founding father forgotten by history”, according to Dr. George Sheldon. Sheldon, like Williamson, a recognized leader in medicine and professor of surgery and social medicine in the School of Medicine, produced a remedy for the historic slight with his 2009 biography, “Hugh Williamson: Physician, Patriot and Founding Father.”

As reported in the University Gazette, Vol. 35, No. 8 story about Dr. Sheldon: “In 1774, Williamson co-wrote with (Benjamin) Franklin and Hunter a paper on the electric eel that was presented to the Royal Society of London, a learned society for science founded in 1660 by King Charles II.
Just two years later, Franklin endorsed the false charge that Williamson was a British spy. The charge, leveled in a letter by Silas Deane, the first official envoy to France from the Continental Congress, was made as Williamson returned to the colonies that October.
Off the Delaware coast, Williamson’s ship was captured by the British. Williamson managed to escape by rowboat and make his way to the Continental Congress where he applied for, but was denied, a military commission because of the charge that he was a spy. Under this cloud of suspicion, Williamson left for Charleston, S.C., to join his brother in the business of shipbuilding and commercial trading.
Williamson planned to center his commercial operations in Philadelphia, but a British blockade in the Chesapeake Bay forced him to dock his ship in the port of Edenton off the North Carolina coast.
For whatever reason, Sheldon said, Williamson stayed in the Tar Heel state – and remained loyal to the cause of independence. In a 1778 letter, Williamson chafed at the question of his loyalty: “There was not in America a man who served it more faithfully or disinterestedly.”
The Gazette goes on to report: “That service to country found deep and multifaceted expression in Williamson’s adopted home of North Carolina from 1777 to 1793, Sheldon said.
In 1779, a year after affirming his loyalty to the colonies by signing the Book of Allegiance in Edenton, Williamson was appointed surgeon general of the North Carolina Revolutionary War militia. As an army surgeon, he recommended inoculation against smallpox for civilians and military troops before they entered active service.
In 1782, Williamson returned to Edenton and was elected to the N.C. House of Commons. In 1787, the governor appointed Williamson to serve as a delegate to the Constitutional Convention. And on Sept. 17, 1787, he was one of the 39 delegates (out of the 55 delegates in attendance) who signed the United States Constitution in the city where a decade earlier he had been accused of being a Tory spy. “

In the famous Howard Chandler Christy painting of the signing of the Constitution, Hugh Williamson is directly in front of the desk with one foot on the first step of the dais. He is wearing a brown coat and looking directly at General Washington.

As noted by American History (http://www.let.rug.nl/usa/biographies/hugh-williamson/): Williamson, after attending preparatory schools he entered the first class of the College of Philadelphia (later part of the University of Pennsylvania) and took his degree in 1757. He then trained for the ministry, and became a licensed Presbyterian preacher but was never ordained. He taught mathematics at his alma mater before studying, in 1764, medicine, eventually obtaining a degree from the University of Utrecht. In 1768 he became a member of the American Philosophical Society. The next year, he served on a commission that observed the transits of Venus and Mercury. In 1771 he wrote An Essay on Comets, in which he advanced several original ideas. As a result, the University of Leyden awarded him an LL.D. degree.
Williamson also authored, in 1770, of one of the first books on climate change, (Observations on the Climate in Different Parts of America…)
Williamson witnessed the Boston Tea Party as he departed for London where he then reported on the uprising to the Privy Council. He warned the Crown that harsh reactions would lead to revolution, as it did.
The term polymath is frequently linked to Williamson and a review of his contributions to knowledge as well as his leadership illustrates the accuracy of the term in his case. See the chronological listing below to understand that point:

Timeline Hugh Williamson – Expanded

  • 1735 Born in Pennsylvania of Scotch-Irish descendent. Parents hoping, he would become a Presbyterian Minister. Attended first class of the College of Pennsylvania, later made part of Univ. of Pennsylvania.
  • 1757/60 College of Philadelphia, Master’s Degree in Mathematics, Professor of Mathematics
  • 1762 Studies Theology – Licensed as Presbyterian Minister
  • 1764/68 Studies Medicine at Edinburgh, London, obtains degree from Univ. of Utrecht, Netherlands. Returns to Philadelphia.
  • 1768 Becomes member of American Philosophical Society with Benjamin Franklin and John Bartram, among other notables.
  • 1770 Wrote paper on Climate Change suggesting that clearing land of trees let to global warming
  • 1771 Wrote “An Essay on Comets” for which he was awarded Doctorate of Law degree by University of Leyden, oldest university in the Netherlands.
  • 1773 In route to England witnesses Boston Tea Party. In London summoned to Privy Council to testify on this act of rebellion. Testifies, “Repression would provoke rebellion”
  • 1775 Develops a successful company importing medicines from West Indies, Indies, Martinique and America’s coastline.
  • 1777 Arrives in Edenton, NC, eluding British blockades. Signs “Book of Allegiance” to Cause of Independence. Operates his trading company through Roanoke Inlet.
  • 1779 Appointed Surgeon General NC Militia. Required inoculation against smallpox, many lives saved.
  • 1780 Went behind enemy lines to treat Americans wounded in Battle of Camden. Convinced British to inoculate against smallpox.
  • 1782/88 Elected NC House of Commons. Continental Congress 1782-88, NC delegate to Constitutional Convention, and in
  • 1787 was one of 39 signers of United States Constitution.
  • 1788 Published Speech, “Remarks on New Plan of Government” promoting ratification of Constitution and made key speech on the steps of the 1767 Courthouse at Edenton that convinced the state to ratify the new Constitution after it first refused to take action.
  • 1789/01 Member of the University of NC board and first Secretary of the Board of Trustees of America’s first State University.
  • 1790‘s Named trustee NY College of Physicians, Founder of NY’s Literary and Philosophical Society.
  • 1812 Authored, “History of North Carolina”
  • 1819 Died on May 22, age 83 in New York, buried in Trinity Courtyard.

This is a very good PowerPoint on Williamson’s life, assembled by Chris Grimes of the Albemarle Chapter of the Sons of the American Revolution:

Hugh Williamson per grimes.ppt

For a printable .pdf of this page, click here:

Hugh Williamson

(1) Barker House
(2) Old Colonial Wharf
(3) Joseph Hewes & 1778 Cannons
(4) Hugh Williamson Monument
(5) 1905 Edenton Teapot
(6) 1767 Chowan County Courthouse (no QR code on pedestal sign)
(7) Old Jail
(8) Oldest House in North Carolina
(9) Cotton Mill Village
(10) 1800/1827 James Iredell House (no QR code on pedestal sign)
(11) Kadesh Church
(12) 1736 St. Paul’s Church
(13) 1758 Cupola House
(14) Josephine Leary Building
(15) 1886 Roanoke River Lighthouse (no QR code on pedestal sign)

Blades in VA and NC – From Stones to Steel

By Jim Melchor and Tom Newbern

From the dawn of mankind, humans have needed blades for hunting, chopping, cutting, and fighting. In this article, we will focus on the evolution of blades in what is now Virginia and North Carolina from pre-historic times into the early nineteenth century. 1

The length of time humans have inhabited North America is still in question. Estimates range upward in the tens of thousands of years with no clear resolution is sight. However, we do know humans have lived in Virginia and North Carolina for at least the past 12,000 years. There is clear evidence of this in the plethora of artifacts left by these people, primarily stone tools including, but not limited to, projectile points, knives, scrapers, choppers, axes, and the stone waste (reduction flakes and broken rocks) resulting from the manufacture of these tools.

In Virginia and North Carolina, as elsewhere in eastern North America, there are basically three, recognized, distinct periods of early man, Paleo-Indian (circa 12,000 years BP and before), Archaic (circa 10,000 years BP to circa 3,000 years BP), and Woodland (circa 3,000 years BP to contact with Europeans). Some overlap from period to period did occur, and there were developmental stages within each period. Full discussions of these periods are, and have been, argued in detail in the print media and online.

The Paleo-Indian people consisted primarily of small family groups of nomadic hunters who followed and killed large and small animals for their primary subsistence. However, they undoubtedly utilized other food resources they encountered. Archaic people were primarily larger groups of migratory hunters and gatherers whose subsistence depended both on the killing of animals and on the gathering of other wild edibles such as shellfish, fish, nuts, berries, fruits, vegetables, roots, etc. The Woodland people also hunted and gathered, but they were primarily farmers living in larger settlements and even “cities”. These are the people the first European explorers and settlers encountered in the new world and called “Indians” in the mistaken belief that they had discovered the East Indies. Unlike elsewhere in the world where early man transitioned from the Stone Age to the Bronze Age to the Iron Age, these Native Americans, encountered by the first Europeans, were still in the Stone Age.

The most recognizable tool of the Paleo-Indians in Virginia and North Carolina is the finely made, fluted, lanceolate point or blade (Fig. 1, Four fluted points found in eastern Virginia, top left – Surry County, top right – City of Williamsburg, center – Brunswick County, and bottom – Williamson Paleo-Indian Site, Dinwiddie County). This type of fluted point resembles similar points found near Clovis, New Mexico, and the term, Clovis point, has been generically applied to this form. Clovis points most likely were used as spear points and hafted knives. However, fine points are not the only Paleo-Indian artifacts encountered in Virginia and North Carolina. Indeed, more mundane stone tools are also found on Paleo-Indian sites such as the scraper and preform in Figure 2 (Fig. 2, Scraper on left and preform on right, both found on the Williamson Paleo-Indian Site, Dinwiddie County). Scrapers were used for scraping flesh and hair from hides as well as for scraping wood, bone, etc. Preforms are larger pieces of suitable stone that were reduced in size for easier transport. They could be used as a scraper or chopper, as is, or flaked into a projectile point, knife, or other tool as needed, the prehistoric equivalent of a Swiss army knife. The Williamson Paleo-Indian Site and the adjoining Ampy Farm are literally covered in stone artifacts and debris, every piece of which has been worked to some extent by early man. The readily recognizable, bluish, stone material encountered here is known as Little Cattail Creek chalcedony, a cryptocrystalline form of quartz. Archaeologist Floyd Painter coined the name. 2 The actual source of the stone on this extensive workshop site has not been found, but it probably was an outcrop in one of the deep ravines on the property that was mined out by the Paleo-Indians occupying the site. Because of the relatively small population of Paleo-Indians in Virginia and North Carolina, their artifacts are quite rare compared with those of the subsequent two cultures. Though heavily picked for many years, the Williamson Paleo-Indian Site remains an important research location for those studying early man. A major collection of points and other tools from the Williamson Paleo-Indian site resides at the College of William and Mary.

Figure 1

Figure 1

Figure 2

Figure 2

As a basic overview, three techniques were incorporated in the manufacture of stone tools of the Paleo-Indians; percussion flaking, pressure flaking, and grinding. Percussion flaking involves the striking of a stone workpiece to remove large chips or flakes from the workpiece for further shaping, to reduce the size of the workpiece, or to refine the shape of the workpiece. The striking implement can be another stone (hammer stone) or heavy baton of antler or wood (Fig. 3, Hammer stone, workpieces, flakes). Pressure flaking involves precisely removing small flakes in the process of shaping a larger flake into a finished tool. Pressure is applied to the edge of a flake with a pointed antler, bone, or possibly wood tool (Fig. 4, Deer antler, flakes being worked, and small flakes removed). Fluted points required careful pressure flaking in shaping the point overall. Also, pressure flaking was incorporated in preparing a blade for percussion removal of the long flakes of the flutes. The arched base and sides of a fluted point near its base were ground smooth with an abrasive stone to prevent sharp edges from cutting the binding material holding the blade to its haft. 3

Fig 3

Figure 3

Fig 4

Figure 4

While the above explanation of the flaking process might sound simplistic, nothing could be further from the case. Producing stone blades, most especially Clovis points, took considerable skill, precision, and time of an accomplished knapper or stone worker, even in the production of basic tools. Also, stone material was carefully selected for its flaking characteristics (workability) and often for its aesthetics. The survival of the Paleo-Indians depended on their blades. It should be noted here that once a blade was produced, it would have been reworked as necessary as it sustained damage until such time it was lost or no longer useful. As can be seen in Figure 1, the top two fluted points have been substantially reduced in length through reshaping as their tips were damaged in use. Also seen in Figure 1, the tip of the point in the center has been reshaped into a scraper, probably as a result of damage to the original pointed tip.

Archaic artifacts are plentiful due to the longevity of the period, the vastly increased size of the population, and the spatial distribution of the population. In fact, numerous subcultures developed within the Archaic through time and over space, and these subcultures are reflected in the varied styles of blades they produced. Archaic blades range from very finely executed to rather crude. Figs. 5, 6, and 7 show an assortment of well made points that cover the entire temporal span of the Archaic period from its overlap with the Paleo-Indian period on its early end to its transition into the Woodland period at its later end (Figs. 5, 6, and 7, Archaic points). Interestingly enough, all of these blades, which represent thousands of years of development, were found within a limited geographical area in the vicinity of Stony Creek in Sussex County, Virginia. While this might seem highly unusual, it is, in fact, rather commonplace. Collectors and archaeologists have been making similar mixed finds in eastern Virginia and North Carolina for over a century. Not all Archaic points are as refined as those in Figs. 5, 6, and 7. Figure 8 shows an assortment of points from the middle Archaic. These points, locally known as Guilford points from their type location in Guilford County, North Carolina, range from well made to crude (Fig. 8, Guilford points). The points in the upper portion of Figure 8 are from the same area of Sussex County, Virginia, as the more refined points discussed above. The two points in the bottom of Figure 8 are actually from Guilford County, North Carolina. The top blade is very well done with carefully controlled flakes, whereas the bottom one is clunky. The salient points of the discussion in this paragraph are a) the Archaic subcultures were visiting the same resource areas through time and some maybe at the same time, b) all the blade makers were not equally talented, and c) not all cultures cared that much about the aesthetics of their blades.

figures 5-7

Fig 8

Figure 8

Archaic blades were produced essentially the same way Paleo-Indian blades were made using percussion and pressure flaking. However, the Archaic cultures rarely ground the bases of their blades. The exception is that blades produced in the Paleo-Indian to Archaic transition period sometimes exhibit basal grinding. 4

As discussed above, producing stone blades took considerable skill and time, and the Archaic people, too, reworked their blades to keep them in service as long as possible. Figure 9 shows an assortment of Archaic points from eastern Virginia and illustrates how the full-size points likely would have been reworked, and thus shortened, to keep them in service (Fig. 9, Archaic points, full size and reworked). The top right blade was apparently broken in half and reworked into a hafted scraper.

Figure 9

Figure 9

Around the middle of the Archaic period, our early Virginia and North Carolina residents developed the need for heavier blades in the form of axes. While these axes were no things of beauty, they were functional. They were simply percussion flaked from small boulders (Fig. 10, Four flaked axes from Nansemond County {now City of Suffolk}, Virginia).

Figure 10

Figure 10

As the Archaic period progressed and transitioned into the Woodland period, the axes became more refined. Furthermore, the manufacturing process for these axes changed from simple percussion flaking to pecking, grinding, and polishing (Fig. 11, Four polished, grooved axes, bottom right example from Sunbury Area of Gates County, North Carolina, other three from Nansemond County, Virginia). This newly developed process involved repeatedly pecking the stone workpiece with another pointed stone, thus pulverizing and reducing the surface of the workpiece until it took the shape of the desired axe form. Intermittent grinding of the surface of the workpiece with another stone also was incorporated in this shaping process. Finally, the finished bit of the axe, and sometimes the entire axe, was polished smooth, probably with sand. Polished, grooved axes generally were of substantial size. The lower left example in Figure 11 is six inches long, has lost at least two inches of its bit, and still weighs three pounds. Of course, there are always exceptions to the rule. The diminutive axe in Figure 12 is barely three and a quarter inches long and weighs only six ounces (Fig. 12, Small axe). Was this a toy for a young family member? Probably not. Obviously, making polished, grooved axes was a time-consuming, labor-intensive process. The resulting axes were very effective tools and likely were treasured items. As with the blades discussed above, these axes were reshaped when damaged to extend their useful lives as long as possible (see Fig. 11, lower right axe from Sunbury was extensively reworked and is now much shorter than when made originally).

Figure 11

Figure 11

Fiigure 12

Fiigure 12

The Woodland Indians continued making polished, grooved axes throughout their period. In addition, they introduced a new variety of smaller chopping/cutting tool more akin to a modern hatchet. These polished stone tools are presently known as celts (Fig. 13, Four celts from Nansemond County, Virginia). Celts were likely hafted and used primarily for mundane, light, chopping chores; however, they probably were employed as weapons also. The same techniques used in the manufacture of polished axes were applied in making celts.

Figure 13

Figure 13

The bow and arrow was another major technological advancement of the Woodland Indians. This tool supplanted the spear as the primary hunting and fighting weapon, though the spear did not become obsolete. The arrow necessitated a shift in stone-blade morphology. Colloquially, all of the above-discussed projectile points of the Paleo-Indian and Archaic periods have been collectively called “arrowheads”, when, in fact, they were spear points. The Woodland Indians actually made arrowheads. Very generally speaking, these arrowheads were smaller, triangular blades, some very well executed and some not so much (Fig. 14, Assorted Woodland arrowheads). Interestingly, all of these points were found in the Stony Creek area of Sussex County as were those Archaic points in Figures 5, 6, 7, and 8. This group of Woodland points from essentially the same general location as the Archaic points further reinforces the argument made above that both Archaic and Woodland subcultures, and likely Paleo-Indians as well, were visiting the same resource areas through time (and some groups maybe at the same time), that all the blade makers were not equally talented, and that not all subcultures or even total cultures cared that much about the aesthetics of their blades.

Figure 14

Figure 14

One thing, however, did remain constant through all three of the major periods of American Indians in our study area of what is now Virginia and North Carolina. The basic manufacturing techniques for stone blades (percussion flaking, pressure flaking, and grinding) remained essentially the same. Two hammer stones are shown in Figure 15 (Fig. 15, Hammer stones). They, along with several other hammer stones, assorted flakes, scrapers, and a couple of broken and reworked Woodland arrowheads were found in close proximity on a workshop site on Knee Branch in Hertford County, North Carolina. Similar artifacts have been found on other Woodland workshop sites as well as on Archaic and even Paleo-Indian sites, such as the Williamson Paleo-Indian Site. Remember, these stone-working techniques were in use in our study area pushing 12,000 years and in other Stone Age cultures worldwide for even longer. Curiously, early humans, separated spatially and temporally, independently developed these basic techniques worldwide.

Figure 15

Figure 15

When the early European explorers and settlers first came to our study area, they encountered an advanced culture of Stone Age people. These first encounters, subsequent contacts, and eventually temporary and permanent settlements in the New World, would dramatically and forever change the culture and lives of the Woodland Indians. There were two significant, English contact locations in our study area. 5 The first, sponsored by Sir Walter Raleigh, was a couple of failed attempts on Roanoke Island in what is now eastern North Carolina. In 1585, Raleigh named Sir Richard Grenville as Governor-General of Virginia and sent him to Roanoke Island to establish a colony. Grenville delivered Ralph Lane, his assistant, Thomas Hariot, the surveyor/historian of the colony, and the others to America to establish the colony. Grenville did not remain, leaving Lane in command. This attempt did not fare well, and a year later, Sir Francis Drake evacuated those remaining alive, including Lane and Hariot.

In 1587, Raleigh tried again. This resulted in Governor John White’s unsuccessful colony, later known as the Lost Colony. During White’s brief stay on Roanoke Island, he produced a portfolio of watercolors to document the local Indians and their lives. White returned to England to seek additional support for his colony, which was in dire straits. Fortunately, he carried his portfolio of watercolors with him. These original watercolors are now preserved in the British Museum in London. White was unable to return to his colony until 1590. When he arrived, he discovered the colony was gone. Because of adverse weather and time constraints, he was unable to locate any of its members, including his granddaughter, Virginia Dare, the first English child born in America.

In London in 1588, Hariot published his account for Raleigh of Grenville’s attempt in the New World in his book, Briefe and True Report of the New Found Land of Virginia. In 1590, Theodore De Bry in Frankfurt, Germany republished Hariot’s work in his Grands et Petits Voyages, Volume 1, and he illustrated it with engravings he made from White’s watercolors. A comprehensive and very readable book on all of this information on the Roanoke Island attempts, including illustrations of White’s watercolors and De Bry’s engravings, is The New World by Stefan Lorant. There are two De Bry illustrations pertinent to our blades discussion. One is a front and back view of an Indian armed with a bow and arrows and of other Indians hunting deer with bows and arrows (Fig. 16, 1590 De Bry engraving, armed Indians). Here, the triangular, Woodland arrowheads are clearly visible. The second engraving shows Indians fishing using various methods (Fig. 17, 1590 De Bry engraving, Indians fishing). In the background, the Indians are using spears. This is proof that spears were not abandoned with the advent of the bow and arrow.

Figure 16

Figure 16

Figure 17

Figure 17

The second significant contact location in our study area, sponsored by the Virginia Company of London, was John Smith’s colony on Jamestown Island in Virginia in 1607. Here Smith and the colonists constructed James Fort. While this colony had a very shaky start beginning with the “starving time” the first year or so, followed by the massacres of 1622 and 1644, it nevertheless was the first, permanent, English colony in the New World. White man’s settlement in the New World was traumatic for both the settlers and the Indians, particularly for the settlers in the beginning but devastating for the Indian populations later.

As strange as it might seem, we know far more about the manufacturing of blades in what is now Virginia and North Carolina during the pre-contact Indian periods than we do after the arrival of the first Europeans. Concerning the Native American blades, we know the Indians were living here, as we have found millions of examples of their various blades and other artifacts. In addition, we have found numerous workshop and other sites where they lived and worked. If that is not enough evidence, live Indians equipped with their Stone Age tools greeted the early Europeans upon their arrival. Conversely, the iron and steel blades used in the study area after the arrival of the Europeans included both blades imported from Europe and blades made here. Distinguishing between the two can be challenging.

To begin our look at iron and steel blades, we must first distinguish the differences between iron and steel. An in-depth discussion on processing iron ore into iron and further into steel is well beyond the scope of this article. Iron is a naturally occurring element, usually found as iron oxide in nature. Iron oxide (ore) was converted through a melting process into elemental iron, which contained high levels of impurities such as slag, sand, and carbon. Further early refinement of this iron by heating and hammering produced usable wrought iron that could be fashioned by blacksmiths and other artisans into utilitarian objects. Steel is basically an alloy of iron and small, controlled amounts of carbon. 6 Depending upon the carbon content in the alloy, mild steel and high-carbon or tool steel could be produced. Iron is a strong, easily worked metal with many applications; however, iron cannot be hardened without further processing into steel and will not hold a sharp edge on blades. Mild steel is a stronger metal than wrought iron, and it works under the hammer about the same as wrought iron. Similar to wrought iron, mild steel cannot be hardened without further processing. High-carbon or tool steel is tougher still, and it can be hardened by heating, quenching, and tempering. Tool steel can be, and was, fashioned into blades that hold sharp edges, springs, and other objects benefiting from its properties. Also on this website, in our article, That Rusty Object – Is It Wrought Iron or Steel?, we provide a simple technique for determining whether a ferrous metal object is made of wrought iron, mild steel, or high-carbon steel. This methodology is mildly destructive and is not recommended for well-preserved artifacts.

In searching for examples of iron and steel blades made and/or used by the earliest English settlers in Virginia and North Carolina over 400 years ago, one needs to look no farther than Jamestown Island. In 1994, the Association for the Preservation of Virginia Antiquities (APVA, now Preservation Virginia) launched its Jamestown Rediscovery project to search for any evidence of the 1607 site of James Fort. Conventional wisdom (rumors) persisted that the James Fort sight on Jamestown Island had eroded into the James River long ago. Archaeologist Bill Kelso (now Director of Archaeology at Historic Jamestowne), archaeologist Bly Straube (now Senior Archaeological Curator for Historic Jamestowne), and a team of other archaeologists and researchers, undertook the extensive research and excavation project. Remarkably, they discovered that only a small portion of the fort had eroded into the river and that most of the site remains on dry land! Extensive excavations of the site continue, and over a million artifacts have been recovered, including blades used by the first colonists. 7 Many of these artifacts are on display in the Historic Jamestowne Archaearium.

Edged blades known to have been used by the early English settlers in Virginia and North Carolina include: a) Swords of all types, b) Belt knives and daggers, both for practical use and show, c) Folding knives, d) Axes, both broad and felling, e) Halberds – ceremonial, axe-shaped polearms, used to designate military rank, f) Pikes – military, spear-shaped polearms used as weapons, g) Spontoons – ceremonial, spear-shaped polearms used to designate military rank, and h) Bills – sharp, curved, axe-like, military polearms often with a spike for thrusting, used as weapons. Here, let us distinguish the basic difference between bills and billhooks for purposes of this article. Billhooks are agricultural tools for chopping brush, saplings, hedges, etc. They date back at least to the ancient Greeks, and they are still in use today, known as bush axes or brush hooks. Around the fifteenth century, billhooks were “militarized” into polearms known as bills and used as weapons for combat. This is not to say that agricultural billhooks were not used as weapons also. During the sixteenth century, English troops were particularly fond of bills, especially for removing mounted cavalrymen from their horses. However, by the end of the sixteenth century, bills were obsolete. Nevertheless, they were still supplied to the Jamestown settlers as weapons, where apparently many reverted back for use as billhooks. 8

The following blades from the Jamestown Rediscovery, Preservation Virginia Collection are representative of what was being used by the early settlers in our study area at least through the mid seventeenth century and quite likely through the end of the century (Fig. 18, Scottish, basket-hilt broadsword), (Fig. 19, Hilt from Scottish broadsword), (Fig. 20, Halberd associated with Sir Thomas West, 12th Baron De La Warr who visited Jamestown in 1610, held by Michael Lavin, Jamestown Rediscovery), (Fig. 21, Felling axe), (Fig. 22, Broad axe), (Fig. 23, Dagger in sheath), (Fig. 24, Bill). Similar forms continued in use throughout the eighteenth century. Because these actual blades were found in situ in the excavations on Jamestown Island, we know for a fact that they were here. What we do not know for certain is where they were made. Most likely, the settlers imported all of them. Straube has postulated that most of the iron and steel blades used by the early settlers in our study area were imported from Europe and that the local blacksmiths and armorers were focused on repairing damaged weapons. 9 She would know better than anyone else.

 

figures 18-21figures 22-24Figure 25 is a halberd that dates to the mid seventeenth century, and it is of German origin (Fig. 25, Seventeenth century German halberd). While this halberd was purchased in eastern Virginia, there is no firm evidence that it was ever used here. Figure 26, on the other hand, is a poleax that was discovered in 1992 by Chris Calkins in the Petersburg, Virginia, area near the traditional, circa 1645 site of Fort Henry (Fig. 26, Poleax found in Petersburg, VA area). It most certainly was used here. Archaeologist Ivor Noel Hume of Colonial Williamsburg examined this object, and he identified it as being of Spanish origin. Further, Noel Hume opined that it was brought to the location where it was found by remnants of the Lost Colony. 10 Considering this poleax was found near the site of Fort Henry, a frontier post of the Jamestown colony, this postulated scenario of Noel Hume is not likely. A poleax is a traditional battleaxe, similar in shape to a ceremonial halberd; however, it was an actual combat weapon. The poleax was obsolete well before the beginning of the seventeenth century; nevertheless, apparently some were supplied to the Jamestown colonists along with obsolete bills. This particular poleax is owned by the Tourism Department of the City of Petersburg.

Figure 25

Figure 25

Figure 26

Figure 26

Figure 27 is an American-made ceremonial halberd in the Colonial Williamsburg Foundation Collection. It dates circa 1720-40 (Fig. 27, American-made Halberd). Figure 28 is a crude, blacksmith-made knife found in the vicinity of Menola in Hertford County, North Carolina. This field find was made and used in our study area, and it likely dates to the seventeenth century (Fig. 28, Early NC wrought knife). This knife filled an immediate need for someone, and little time was taken in its manufacture. The wrought pike in Figure 29 (Fig. 29, Early NC wrought pike) also was found in eastern North Carolina, and it was used and likely made in the area. This pike dates from the late seventeenth century to mid eighteenth century. Two broad axes of seventeenth-century form were found in eastern Virginia (Fig. 30, Broad axe, found along the Nansemond River in Nansemond County) (Fig. 31, Broad axe, found along the James River in James City County). These were probably imported, but they could have been made locally as well. Both certainly were used in Virginia in the seventeenth century and possibly into the early eighteenth century.

figures 27-31

As previously noted, blade forms, similar to those of the seventeenth century, continued in use throughout the eighteenth century and even later. These three billhooks (Fig. 32, Billhooks), all found in Virginia, are carryovers from the earlier military bills. The top billhook is British and dates to the mid eighteenth century. The middle example is late eighteenth century French. The bottom billhook is early nineteenth century Virginia made. Another carryover example is this c. 1725, imported, Scottish broadsword with a Virginia background (Fig. 33, Scottish broadsword). This style sword developed directly from its seventeenth century and earlier counterparts such as the James Fort examples (see Figs. 18 and 19). Into the eighteenth century, civilian use of swords, daggers, and knives to show status began to wane somewhat; however, the military and functional uses of these blades obviously continued. Both Virginia and North Carolina were British colonies, and there was a thriving trade between the colonies and Great Britain up until the Revolutionary War. Consequently, most actual military arms and ceremonial arms of the British Army and the militias continued to be imported from the homeland. There was no exigent need for extensive manufacture of these military items in our study area; though some were undoubtedly made. Two major exceptions to the primary import of blades were axes and utilitarian knives. While not everyone in the colonies needed swords or ceremonial blades, virtually every family, if not every man, needed a knife and an axe. As a result, these blades were made here in substantial quantities by blacksmiths and gunsmiths to supplement imports, and uniquely American forms evolved. An excellent, comprehensive, illustrated source for information on blades used in the American colonies from the seventeenth century through the Revolutionary War is Swords and Blades of the American Revolution by George C. Newmann.

Figure 32

Figure 32

Figure 33

Figure 33

The typical axes seen in Figure 34 were used extensively in Virginia and North Carolina throughout the eighteenth century and well into the nineteenth (Fig. 34, Felling axes). They were made primarily in Europe and, to a limited extent, in our study area. These four, eighteenth-century examples, as well as many others, were recovered all over Virginia. Most are less expensive iron axes without steel bits. (We will discuss steel bits later in this article.) It is difficult to say where these axes were actually made, but since they are lower-end examples and very formulaic, they probably were made in Europe and shipped here for trade with the Indians and for sale to budget-minded colonists. The one possible exception is the broad axe with the flared blade. Of the four shown, this one probably was made here rather than being imported. Note two of the axes have bogus touch marks to hype their quality to the unsuspecting. The felling axe in Figure 35, found in the Valley of Virginia, does have a steel bit and is of higher quality; however, it still has a bogus touch mark. This axe, too, was probably an import from Europe (Fig. 35, Imported felling axe with steel bit).

Figure 34

Figure 34

Figure 35

Figure 35

The two pipe tomahawks in Figure 36 were imports specifically for use in the Indian trade (Fig. 36, Trade pipe tomahawks). Both were found in the Valley of Virginia, and neither have steel bits. Virginia and North Carolina gunsmiths produced extremely well made pipe tomahawks for discerning clients, mainly on the frontier. Two such examples are included here. The first is a remarkable, eighteenth-century tomahawk made for Virginia, North Carolina, Kentucky, and Tennessee frontiersman, Indian fighter, Revolutionary War notable, surveyor, statesman, U. S. Senator, and William and Mary graduate, Daniel Smith (Fig. 37 and Fig. 38, show both sides of this tomahawk). They do not get any better than this example. On March 15, 1780, Smith recorded in his diary that he lost his tomahawk. There is a secondary name, James Stephenson, inscribed on the blade of this tomahawk above and below the spear (see Fig. 38). James Stephenson was a militiaman guarding Smith’s party at the time of the loss. Likely, Stephenson found and kept Smith’s tomahawk. 11 The second is a “spontoon”-type, pipe tomahawk (Fig. 39, Spontoon tomahawk). This eighteenth-century tomahawk descended in a Kentucky family. Of course, when it was made and used, Kentucky was still part of Virginia. Both the Smith and the “spontoon” pipe tomahawks have steel bits. While not common, riflemen and also Indians, when they could get them, used fighting tomahawks in our study area in the late eighteenth century and early nineteenth. Two Virginia, spiked tomahawks are shown in Figure 40 (Fig. 40, American spiked tomahawks). Both have steel bits.

figures 36-40

Around the middle of the eighteenth century, the American pattern of axe, a highly efficient chopping tool, was developed in the colonies, and it remains in use today (Fig. 41, American-pattern felling axes). All three of these eighteenth-century felling axes were found in Virginia and were likely made by blacksmiths in Virginia. They are of higher quality with steel bits. The curved, weld joint of the steel bit is clearly visible on the top axe. The joints on the other two axes are not as distinct, but they are there.

Figure 41

Figure 41

Throughout the eighteenth century and well into the nineteenth century in our study area, belt axes were in widespread use by both civilians and many soldiers and militiamen who preferred belt axes to swords. The Indians also included belt axes in their arms inventories. These belt axes followed the basic form of European and American-pattern felling axes. An even smaller axe, the bag axe, another American innovation, likewise followed the two basic felling axe forms (Fig. 42, European form belt axes) (Fig. 43, American-pattern belt and bag axes). These latter diminutive axes were designed for carrying in a rifleman’s bag. In Figure 42, only the right example has a steel bit. The others are entirely of wrought iron. The seven belt and bag axes in Figure 43 have steel bits. The belt axe, second in from the right side, has decorative file work on its lower edge and a genuine touch mark, “MJ” (Fig. 44, Details of belt axe). All of the above belt and bag axes were found in Virginia or North Carolina, and some were still associated with rifleman’s bags or frontier families. Unlike the imported felling axes with their cookie-cutter profiles, belt and bag axes are more individual, like snow flakes. We have not encountered any belt or bag axes recovered in our study area that we thought were imported.

Figure 42

Figure 42

Figure 43

Figure 43

Figure 44

Figure 44

Generally speaking, fashioning blades of iron and steel in the seventeenth and eighteenth centuries required a skilled artisan with the knowledge of the “art and mystery” of his trade and only basic tools – a forge, anvil, hammer, tongs, files, and sometimes a grinding wheel. Until the middle of the nineteenth century, steel was difficult, time consuming, and expensive to produce. Consequently, blacksmiths, armors, and gunsmiths used it only sparingly in applications where absolutely required, such as in springs, areas of high wear, and blades needing tough, sharp edges. For most other iron implements, wrought iron would suffice. In discussing axes, we have mentioned steel bits. In this application, a small piece of high-carbon steel was forge welded to a wrought-iron axe to provide it with a tough bit that would retain its sharp edge. Ken Schwarz, Master Blacksmith, Anderson Blacksmith Shop and Public Armoury, Colonial Williamsburg Foundation, demonstrates how to make an axe with a steel bit (Fig. 45, Iron body of the axe is forged to shape) (Fig. 46, Iron body of axe is folded and forge welded together) (Fig. 47, Steel bit is fitted into place for welding) (Fig. 48, Steel bit is forge welded into wrought iron axe body, weld joint is visible). Figure 49 is another application where a steel bit was forge welded into an eighteenth century blacksmith’s cold chisel that was later used as a wedge. The weld joints in the wrought-iron body and along the steel bit are clearly visible (Fig. 49, Cold chisel with steel bit).

Figure 45

Figure 45

Figure 46

Figure 46

Figure 47

Figure 47

Figure 48

Figure 48

Figure 49

Figure 49

Knives in eighteenth- and early nineteenth-century Virginia and North Carolina were ubiquitous. There was relatively little demand for imported knives, as local blacksmiths, gunsmiths, and even individuals made them by the thousands to satisfy the needs for a utilitarian cutting and/or fighting blade. Knives were made of steel to retain a sharp cutting edge. Of course, traditional knives and daggers were made such as the formal dagger in Figure 50. This dagger from the Valley of Virginia has a whitesmithed guard, iron ferrule, and fitted leather sheath (Fig. 50, Formal VA-made dagger). The less formal, but still traditional, dagger in Figure 51 was fashioned from a spent file. It, too, is from the Valley of Virginia (Fig. 51, VA dagger made from a file). The pewter band around the hilt was a period repair to the hilt. In addition to the traditional forms, a vernacular form of knife also evolved. These knives were generally fitted with bone or deer-antler hilts. While many were professionally made, many more were homemade by the owners. The mid eighteenth-century belt knife with a bone hilt in Figure 52 descended in a Shenandoah County, Virginia family. It was probably blacksmith made, and it was found along with a rifleman’s bag and horn (Fig. 52, Early Virginia belt knife). Conversely, a gunsmith likely made the large, Valley of Virginia, belt knife in Figure 53 for a customer. This knife dates to the late eighteenth-early nineteenth century (Fig. 53, Gunsmith-made belt knife). A more diminutive and often more rustic style of these vernacular knives is the patch knife (Fig. 54, six examples of patch knives). Patch knives were carried in a leather sheath usually attached to a rifleman’s bag. They were used for small cutting jobs but primarily for cutting cloth patches for covering the balls fired from a rifle. Rifle balls fit snugly in a rifle barrel. The cloth patch around the ball actually engages the rifling or spiral grooves in a rifle barrel, thus imparting a stabilizing spin on the ball when fired from a rifle. The top knife and the knife in the center with the iron ferrule probably were professionally made. The rest likely were products of their owners, fashioned from files, broken saw blades, other knives, scrap pieces of steel, etc. All of these illustrated patch knives, dating from the early nineteenth century, were found in the Valley of Virginia, but other similar ones were made and used all over the frontier areas of Virginia and North Carolina. Wherever and whenever there was a man with a longrifle, there was a small knife for cutting patches.

Figure 50

Figure 50

Figure 51

Figure 51

Figure 52

Figure 52

Figure 53

Figure 53

Figure 54

Figure 54

Folding knives also were commonplace in our study area. Often they were used in lieu of patch knives and were carried in a rifleman’s bag. However, they certainly were not limited to use by riflemen. Unlike belt and patch knives of the period in our study area, folding knives were made both locally and imported in considerable quantities from at least the middle of the eighteenth century (Fig. 55, Folding knives from various locations in our study area). The top and bottom two knives were locally made. The second knife from the top is French, and the third knife down is probably Spanish. The basic forms of these knives remained unchanged from at least the seventeenth century.

Figure 55

Figure 55

Numerous, inexpensive, French, folding knives, known as “penny knives”, were imported into Virginia and North Carolina. They were popular with soldiers and militiamen during the French and Indian Wars and the Revolutionary War, as well as with long hunters and other frontiersmen. Bivouacked soldiers and encamped frontiersmen often had time on their hands and sometimes occupied themselves with craft projects, the period equivalent of producing World War I trench art. One such project was the carving of a “penny knife” into the full figure of a woman, including tiny, black, glass, trade beads for her eyes (Fig. 56, “Penny Knives” carved and un-carved). What once was a trifle for a friend or loved one, this carved knife is now a serious piece of folk art. Another less artistic, locally made example is a small knife with a whistle cut into its applewood handle (Fig. 57, Whistle knife).

Figure 57

Figure 56

Figure 57

Figure 57

Functional pikes and their ceremonial forms, spontoons, carried over from their seventeenth and early eighteenth century origins in Virginia and North Carolina well into the last half of the eighteenth century. The decorative spontoon in Figure 58 was made in Germany and is of the type carried by Hessian mercenary troops supporting the British during the Revolutionary War (Fig. 58, Hessian spontoon). 12 Americans captured a large number of Hessian, as well as British, soldiers in the decisive Battle of Yorktown, Virginia. This spontoon was purchased in eastern Virginia. The American pike/spontoon in Figure 59 was acquired from a family in Norfolk, Virginia, and it was likely made in that city for use in the Revolutionary War against the British (Fig. 59, American spontoon). This polearm still retains a portion of its original wood haft in its socket. The British attacked Norfolk on December 1, 1776. The city was essentially destroyed by fire from British naval bombardment and by rowdy local residents.

Figure 58

Figure 58

Figure 59

Figure 59

The above Hessian spontoon has decorative rings around its socket (see Fig. 58). It might be of interest to some to know how this was accomplished. In the following figures, master blacksmith, Peter Ross, formerly of Colonial Williamsburg, demonstrates the process of forge welding and shaping a similar ring on an iron shaft (Fig. 60, First a band of bar stock is partially fitted around the iron shaft) (Fig. 61, The cut bar stock ring is tightened onto the shaft) (Fig. 62, The shaft and ring are brought to welding heat in a forge and hammered in a swage to make the weld) (Fig. 63, Ring is welded to shaft) (Fig. 64, Ring is forged to near final shape) (Fig. 65, Ring and shaft are reheated and placed in a spring swage for final shaping) (Fig. 66, Ring and shaft are hammered into final shape in the swage) (Fig. 67, Ring removed from the spring swage). Note that Ross welded and finished the ring onto a solid iron shaft. For attempting the same work on a hollow socket as on the spontoon, a solid, well fitted, iron mandrel must be inserted into the socket to keep it from collapsing under the hammer blows. Final shaping and finishing were done with files.

Figure 60

Figure 60

Figure 61

Figure 61

Figure 62

Figure 62

Figure 63

Figure 63

Figure 64

Figure 64

Figure 65

Figure 65

Figure 66

Figure 66

Figure 67

Figure 67

By their very nature of use, sword blades must be sharp, flexible, and tough. These characteristics required the use of steel over wrought iron in their manufacture. Quality swords took considerable skill to make, were expensive, and were well regarded by their owners.

As previously discussed, Virginia and North Carolina were British colonies, and the full range of British military swords and other blades required by the British Army and colonial militias throughout the eighteenth century up to the Revolutionary War were imported. 13 With the advent of the Revolutionary War, all that changed drastically. There simply were not enough swords, blades, and other weapons on hand to arm the Continental Army and the state militias to fight the British. Also, the industrial base to produce the needed weapons was sorely lacking in the colonies. Virginia, North Carolina, and their sister colonies had to scramble to overcome these deficiencies. The most logical sources of supply for needed weapons were other European countries (France in particular), and the Americans began importing from them in quantity. To supplement the imports and to repair damaged weapons to keep them in service, new homeland industries were required and existing industrial concerns had to shift over to support the war effort. A prime example of this shift of a private industry was James Hunter’s Rappahannock Forge near Falmouth, Virginia across the Rappahannock River from Fredericksburg. 14 Also, Virginia established the Point of Fork Arsenal in Fluvanna County at the confluence of the James and Rivanna Rivers as a depot for collecting, repairing, manufacturing, storing, and distributing arms to Virginia troops. 15 Schwarz, in his on-going research, found documents indicating that the Continental Congress established a similar depot, the New London Armory, in Bedford County, Virginia to serve the arms needs of the Continental Army. However, there is no evidence of new manufacturing at New London. 16

Of note, American, James Potter, in New York City made particularly fine cavalry sabers during the war years that were desired and used by both Continental and British dragoons alike. Unfortunately, Potter was a Tory, New York City was under British occupation at the time, and his swords went to support the British cause. The Continental dragoons used Potter swords only when they could capture them. 17

Virginia’s first, major, Revolutionary War, sword purchase has been well researched by Giles Cromwell, and all the following information about this purchase and the swords comes from him. 18 To make a long story short, Governor Patrick Henry of Virginia directed that assorted military stores, including swords, be procured for the state in France. While this procurement effort was shrouded in intrigue and questionable pricing, a contract was let in 1778 with Klingenthal near Strasbourg, France. The swords were completed in 1779 and shipped to Virginia. There were three types of swords produced by Klingenthal under this contract that numbered roughly 1500-2000 in total. The first two, grenadier and artillery, were similar, hanger-type swords for foot troops. The blade on the grenadier sword was 26 ½” long and on the artillery sword 23 ¾”. The third was a longer, dragoon sword of a different design for mounted troops with a blade 36” long. All the Klingenthal swords for Virginia were variously marked on both sides of their blades. In addition, all were marked “Klingenthal” on the top edges of their blades (Fig. 68, Klingenthal mark). The following figures illustrate the markings on the three types of swords (Fig. 69, Overall of grenadier sword) (Fig. 70, Detail of markings on right side of grenadier sword) (Fig. 71, Detail of markings on left side of grenadier sword) (Fig. 72, Touch mark on right side of grenadier sword, found on all Klingenthal swords) This sword is in the Colonial Williamsburg Collection. (Figs. 73, Detail of markings on right side of artillery sword, overall sword shape and blade markings on left side similar to grenadier sword) (Fig. 74, Overall comparison of artillery sword on left and grenadier sword on right) (Fig. 75, Markings on left side of artillery swords) (Fig. 76, Overall of dragoon sword) (Fig. 77, Hilt and markings on right side of dragoon sword, note American hilt on this sword, others had brass French hilts) (Fig. 78, Detail of markings in Fig. 77) (Fig. 79, Markings on left side of dragoon sword) (Fig. 80, Detail of markings in Fig. 79). At the end of the Revolutionary War, all the Klingenthal swords not needed by Virginia militia troops were placed in storage at Point of Fork Arsenal. While there are a few grenadier swords surviving, Cromwell reports that there are only four artillery swords and five dragoon swords presently known, and one of these dragoon swords is a dug relic.

Figure 68

Figure 68

Figure 69

Figure 69

Figure 70

Figure 70

Figure 71

Figure 71

Figure 72

Figure 72

Figure 73

Figure 73

Figure 74

Figure 74

Figure 75

Figure 75

Figure 76

Figure 76

Figure 77

Figure 77

Figure 78

Figure 78

Figure 79

Figure 79

Figure 80

Figure 80

 

Home-grown weapons and other wares were being made in Virginia on an industrial scale at Hunter’s Rappahannock Forge. Extant records show that he was producing muskets, pistols, wall guns, and swords for the American cause. He was also repairing weapons. 19 & 20 Very few Rappahannock Forge firearms remain. However, they are easily identified as most of Hunter’s guns were marked “Rappa Forge”. The swords are a different matter. We know for a fact from the records that he was making them; however, they were not marked “Rappa Forge” as on the guns. So, where are they, and what did they look like? There is compelling evidence that Hunter’s swords are the few remaining ones, mostly with Virginia provenances, marked on their hilts with an “H” (Fig. 81, Overall of Hunter sword) (Fig. 82, Hilt of Hunter sword) (Fig. 83, Hilt markings on Hunter sword). This sword is in the Colonial Williamsburg Collection. In Figure 83, the “H” mark is clearly visible. Additional markings are “1 T P L D N 22” which mean “this weapon was number 22 in the first troop of Pulaski’s Light (or ‘Legion’ of) Dragoons. Another “H” sword, number 17 of Pulaski’s 3rd troop, is in a private collection in Connecticut. This unit existed from 1778 until early 1780, when it was incorporated into Armand’s Legion after Pulaski’s death at Savannah the previous October. Adding credence to the interpretation of these markings are the identically styled markings for both Pulaski’s and Armand’s Legions found on some pistols made by Hunter, also for the Continental Light Dragoons”. 21 Several of these swords have imported blades that would indicate Hunter was refurbishing swords with new hilts as well as making new swords. 22

Figure 81

Figure 81

Figure 82

Figure 82

Figure 83

Figure 83

Casimir Pulaski, a Polish emigrant and cavalry expert, was a general in the Continental Army and a Revolutionary War hero. Pulaski County, Virginia, is named for him.

There is considerable archaeological evidence in the form of numerous incomplete blade and socket forgings that, in addition to repair work at Point of Fork Arsenal, socket bayonets were being made there as well. These bayonets were likely being made from near the end of the Revolutionary War until Point of Fork ceased operation in 1801. 23 Figure 84 is a Point of Fork Bayonet in the Colonial Williamsburg Collection (Fig. 84, Point of Fork bayonet).

Figure 84

Figure 84

Revolutionary War weapons manufacturing in North Carolina was not as extensive as in Virginia, but the effort was significant nevertheless. It consisted primarily of small contract operations. For example, in 1776, North Carolina directed Ambrose Ramsey to establish a gun manufactory in the District of Hillsboro and to produce 200 muskets, bayonets, etc. This effort was under funded and met with other environmental difficulties. It failed after only partially filling the order. Likewise, in 1778, North Carolina contracted with James Ransom in Bertie County to make muskets and bayonets. He only produced 36 complete guns, 24 bayonets, and assorted other parts. In 1776, the North Carolina Committee of Safety contracted with Timothy Bloodworth to manufacture muskets and bayonets. His production is not known. In 1776, North Carolina contracted with John DeVane and Richard Herring of the Wilmington Gun Factory. They produced 100 muskets and bayonets before the British destroyed their operation. 24

 

The Revolutionary War taught the fledging federal and state governments an important lesson about the necessity of an arms-making industrial base. Toward the end of the eighteenth century, federal armories were established at Harpers Ferry, Virginia and Springfield, Massachusetts. In the same timeframe, Virginia decided to consolidate its state-owned arms industry in Richmond. In 1798, the Virginia General Assembly authorized the establishment of an armory, thus the Virginia Manufactory of Arms in Richmond was born. Initial, limited production of muskets began in 1802. Around 1801, Point of Fork closed, and its materials, including its remaining stock of Klingenthal swords, were shipped to and stored in various locations in Richmond, including in the Capitol loft. The Point of Fork property was sold by the state in 1809. Cromwell has covered in detail all aspects of the Virginia Manufactory of Arms in his landmark book on the subject. This information comes from Cromwell. 25 The primary mission of the Virginia Manufactory of Arms was to produce for Virginia’s militia troops quality military arms, namely muskets, bayonets, rifles, pistols, and swords. First model rifles and pistols are extremely rare, on par with the “Rappa. Forge” wall guns, muskets, pistols, and “H” swords. The later model Virginia Manufactory firearms and all swords are more common survivors. For our article, we are focusing on the blades. Virginia Manufactory firearms were clearly marked on their lock plates as can be seen on this second model pistol made in 1813 (Fig. 85, Virginia Manufactory of Arms markings on firearms – Virginia, Richmond, and date made). The blades did not have the identifying, Virginia markings as found on the firearms; however, the iconic blade shapes and hilts make identifying Virginia Manufactory blades easy. The Virginia Manufactory produced four distinct models of swords: first, second, third, and artillery (Fig. 86, First model sword in its original, black-Japanned scabbard) (Fig. 87, Second model sword) (Fig. 88, Third model sword) (Fig. 89, Artillery model sword). All the first and second model swords originally were made with the long, curved blades as seen on the second model sword in Figure 87. These long, curved blades were generally not well received by the Virginia cavalry units, and many, along with their scabbards, were returned to the Virginia Manufactory of Arms for shortening. The first model sword in Figure 86 is an example of this modification. The third model sword (see Fig. 88) was made with a slightly shorter, less-curved blade with a clipped point. There were two slight variants of the artillery sword. One has a curved blade as seen in Figure 89. The other variant has a somewhat straighter blade (Fig 90, Artillery model sword with straighter blade). This artillery sword is in its original leather scabbard. Only two of these artillery swords with their original scabbards are known. Not unlike the Indians in prehistoric times reworking their broken blades into other usable tools, broken swords sometimes were reworked to extend their lives. The knife in Figure 91 is just such an example. It was made from a broken third model Virginia Manufactory sword (Fig. 91, Knife made from sword). Many, but not all, of the swords and firearms made at the Virginia Manufactory of Arms were stamped with regimental marks like “32, V’A REG’T” which is found on the sword in Figure 89 (Fig. 92, Detail of regimental marks on Fig. 89). Bayonets also were made at the Virginia Manufactory. Unfortunately, these blades were not marked as to manufacturer. They also underwent several design changes from their inception in 1802 to end of production around 1821. The most recognizable Virginia Manufactory bayonets are the particularly long ones standardized in 1807. These had blades 24” long and were 28 ¼” long overall (Fig. 93, Virginia Manufactory bayonet with its original 1807 musket) (Fig. 94, Close-up photo of 1807 bayonet in Fig. 93). Controversy surrounded these long bayonets as it did with the long swords, and production reverted back to shorter bayonets. The Virginia Manufactory of Arms ceased all weapons production in 1821 until the start of the Civil War. When Richmond fell to union forces, the Virginia Manufactory of Arms was destroyed. 26

Figure 85

Figure 85

Figure 86

Figure 86

Figure 87

Figure 87

Figure 88

Figure 88

Figure 89

Figure 89

Figure 90

Figure 90

Figure 91

Figure 91

Figure 92

Figure 92

Figure 93

Figure 93

Figure 94

Figure 94

For our final blade discussion, it is only fitting that we return to a recycled example from Cromwell’s research. According to Cromwell, in 1806, the governor of Virginia ordered that 187 of the remaining 600 Klingenthal grenadier swords stored in the Capitol be taken to the Virginia Manufactory of Arms. There they were to be cleaned, and all previous markings were to be ground off. New regimental markings for Norfolk County, “7, V’A REG’T” and for the Borough of Norfolk, “4, V’A REG’T” were to be stamped on the top of the blades (Fig. 95, Overall of refurbished Norfolk County sword) (Fig. 96, Regimental mark, “7, V’A REG’T” on Norfolk County sword in Fig. 95). In July 1807, the refurbished swords were placed in a hogshead and shipped on the sloop, Nancy, to Norfolk County. There, the regimental commander took his 80, Norfolk County- marked swords for his non-commissioned officers, and he delivered the remaining, 107, Norfolk-marked swords to Norfolk commanders. Only two of these 187, refurbished, and marked swords are currently known. Interestingly, there are also two other marked “7, V’A REG’T” swords known, one grenadier-style sword and one artillery-style sword. In addition to the Norfolk County regimental markings, these two swords still retain all their earlier Klingenthal marks as well. Were these two swords part of the 187 ordered but were not ground for some reason? Why is there an artillery sword in the mix when the governor ordered 187 grenadier swords? Were they Virginia militia swords already in possession of the Norfolk County regiment that had been previously marked for the regiment? Answers to these questions will have to await future research. 27

Figure 95

Figure 95

Figure 96

Figure 96

In conclusion, this article was never intended to be an all-inclusive study of the blades used in Virginia and North Carolina from stone to steel. That would not have been possible for us, especially considering our advanced ages and the enormous body of material. Certainly, there are stone, iron, and steel blades of various types in our study area that we have missed. For that, we apologize. However, we have tried to provide a general overview of the major types of blades used in what is now Virginia and North Carolina from the Paleo-Indians through around 1815.

 

Published on: Feb 13, 2015

Endnotes:

 

  1. Unless noted otherwise in the text or Photo Credits below, all artifacts illustrated in this article are privately owned, and the photographs are by the authors. When a scale is included in a photograph, the scale is six inches long. This research was conducted for the Department of Collections and Conservation, Colonial Williamsburg Foundation in an effort to take the thoughts that have been floating in our brains for years and reducing them to written words to share with others. Several of our sources, old personal friends and colleagues, are now deceased and many more are aging. These scholars were and still are pioneers in their respective fields. In this article, we have attempted to share that which we have learned from them and to provide a partial bibliography of their publications relevant to our study.

 

  1. Personal communication with Floyd Painter (now deceased).

 

  1. Following extensive personal communications with Floyd Painter and Dr. Ben McCary (both now deceased), we based these observations on our personal examination of numerous stone tools and reverse engineering for likely manufacturing techniques. We have tried these techniques to satisfy ourselves. We are, by no means, the first to propose these manufacturing methods, nor are we proficient at them. Numerous earlier researchers, including Painter and McCary, came to these same conclusions long before us and have published extensively on them.

 

  1. Personal communication with Floyd Painter (now deceased).

 

  1. Earlier Spanish and French explorers undoubtedly made incursions into what is now North Carolina and Virginia prior to the arrival of the English.

 

  1. We are discussing early steel here. In modern steel, additional elements are added to change steel properties to accommodate different uses.

 

  1. Personal communications with Dr. Bill Kelso and Dr. Bly Straube; For complete information on this monumental find, visit the Historic Jamestowne website at: http://apva.org/rediscovery/page.php?page_id=4 , or better still, go to Historic Jamestowne and see the James Fort excavations and Archaearium for yourself. You will not be disappointed!

 

  1. Personal communication with Dr. Bly Straube.

 

  1. Ibid.

 

  1. Personal communications with Mark Wenger and Willie Graham, including Graham’s recollections of conversations with Chris Calkins, who discovered the poleax, and with Ivor Noel Hume, who examined the object.

 

  1. Personal communication with Bill Guthman (now deceased) and from his article in Man At Arms.

 

  1. A nearly identical example is illustrated in Newmann.

 

  1. Newmann covers the full range of British swords and blades in this period, and we will not re-discuss them here.

 

  1. Personal communication with Jerrilynn Eby MacGregor and from her book on the Rappahannock Forge, Laying The Hoe.

 

  1. Personal communication with Giles Cromwell.

 

  1. Personal Communication with Ken Schwarz.

 

  1. Personal communication with Erik Goldstein, Curator of Mechanical Arts and Numismatics (Curator of Cool Guy Stuff), Colonial Williamsburg Foundation.

 

  1. Personal communication with Giles Cromwell and from his privately published treatise on the French swords for Virginia. Note, this treatise is currently being revised, updated, and edited, and it will be republished soon on this website.

 

  1. Personal communication with Jerrilynn Eby MacGregor and from her book on the Rappahannock Forge, Laying The Hoe.

 

  1. Personal communications with Giles Cromwell and Gordon Barlow.

 

  1. Quoted directly from Erik Goldstein’s description in the Colonial Williamsburg file on this sword.

 

  1. Personal communications with Giles Cromwell and Erik Goldstein.

 

  1. Personal communication with Giles Cromwell.

 

  1. All from the research notes of Garland Wood.

 

  1. Personal communication with Giles Cromwell and from his book on the Virginia Manufactory of Arms.

 

  1. Ibid.

 

  1. Personal communication with Giles Cromwell and from his privately published treatise on the French swords for Virginia.

 

 

 

 

Photo Credits:

 

Figs. 18-24 Courtesy of Jamestown Rediscovery, Preservation Virginia.

 

Fig. 26 Courtesy of Willie Graham and Colonial Williamsburg Foundation.

 

Figs. 37-39, 68, 73-80 Courtesy of Giles Cromwell.

 

Figs. 45-48 Courtesy of Ken Schwarz.

 

Figs. 27, 69-72, 81-84 Courtesy of Colonial Williamsburg Foundation.

 

 

 

References/Bibliography:

 

Books

 

Cromwell, Giles. French Swords for Virginia, 1779. Ellicott City, MD: Courtney B. Wilson & Associates, 1995.

 

Cromwell, Giles. The Virginia Manufactory of Arms. Charlottesville, VA: University Press of Virginia, 1975.

 

Eby (now MacGregor), Jerrilynn. Laying the Hoe, A Century of Iron Manufacturing in Stafford County, Virginia. Westminster, MD: Heritage Books, Inc., 2007.

 

Goldstein, Erik. The Socket Bayonet in the British Army 1687-1783. Lincoln, RI: Andrew Mowbray Publishers, 2000.

 

Goldstein, Erik and Mowbray, Stuart. The Brown Bess. Woonsocket, RI: Mowbray Publishing, 2010.

 

Hranicky, Wm Jack. Indian Stone Tools of the Commonwealth of Virginia. Alexandria, VA: Virginia Academic Press, 2002.

 

Kelso, William M. Jamestown Rediscovery II, Search for 1607 James Fort. Richmond, VA: Association for the Preservation of Virginia Antiquities, 1996.

 

Lorant, Stefan. The New World. New York: Duell, Sloan, & Pearce, Inc., 1946.

 

Newmann, George C. Swords and Blades of the American Revolution. Harrisburg, PA: Stackpole Books, 1973.

 

Swayze, Nathan L. The Rappahannock Forge. Saratoga Springs, NY: American Society of Arms Collectors, 1976.

 

Periodicals

 

Babits, L. E. “The Evolution and Adoption of Firearm Ignition Systems in Eastern North America: An Ethnohistorical Approach.” The Chesopiean, June-August (1976).

 

Bottoms, Edward. “The Paleo-Indian Component of the Richmond Site, Chesterfield County, Virginia.” The Chesopiean, August (1972).

 

Bottoms, Edward. “Survey of North Carolina Paleo-Indian Projectile Points, Reports No. 1-5, Points 1-129.” The Chesopiean, (2008).

 

Eckard, Christopher S. “Hammerstone Types Found in the Coastal Plain of Southeast Virginia.” ASV Quarterly Bulletin, September (1998).

 

Egloff, Keith and McAvoy, Joseph M. “A Tribute to Ben McCary.” ASV Quarterly Bulletin, December (1998).

 

Guthman, William H. “Daniel Smith – Frontier Surveyor.” Man At Arms, March-April (1979).

 

Hranicky, Wm Jack and Painter, Floyd. “Projectile Point Types in Virginia and Neighboring Areas.” ASV Special Publication, Number 16 (1988).

 

Hranicky, Wm Jack. “Projectile Point Typology and Nomenclature for Maryland, Virginia, West Virginia, and North/South Carolina.” ASV Special Publication, Number 26 (1991).

 

Hranicky, Wm Jack. “Prehistoric Axes, Celts, Bannerstones, and Other Large Tools in Virginia and Various States.” ASV Special Publication, Number 34 (1995).

 

Hranicky, Wm Jack. “McCary Fluted Point Survey: Points 1000-1008.” ASV Quarterly Bulletin, March (2004).

 

Johnson, Gerald H. “An Analysis of Lithic Material from the Williamson Site.” ASV Quarterly Bulletin, December (1978).

 

Johnson, Michael F. and Pearsall, Joyce E. “The Dr. Ben McCary Virginia Fluted Point Survey, Nos. 846-867.” ASV Quarterly Bulletin, June (1991).

 

Land, Howard D. “The Flint Perspective.” The Chesopiean, January-June (1979).

 

McCary, Ben C. “The Williamson Paleo-Indian Site Dinwiddie County, VA.” The Chesopiean, June-August (1975).

 

McCary, Ben C. “Bannerstones from the Dismal Swamp Area and Nearby Counties in Virginia and North Carolina.” ASV Quarterly Bulletin, September (1975).

 

McCary, Ben C. “Semi-Lunar Knives from Tidewater Virginia.” ASV Quarterly Bulletin, March (1977).

 

McCary, Ben C. “Three Uncommon Atlatl Weights.” ASV Quarterly Bulletin, March (1977).

 

McCary, Ben C. and Bittner, Glenn R. “Excavations at the Williamson Site, Dinwiddie County, Virginia.” ASV Quarterly Bulletin, December (1978).

 

McCary, Ben C. and Bittner, Glenn R. “The Paleo-Indian Component of the Mitchell Plantation Site, Sussex County, Virginia.” ASV Quarterly Bulletin, September (1979).

 

McCary, Ben C. “Survey of Virginia Fluted Points, Nos. 537-603.” ASV Quarterly Bulletin, March (1980).

 

McCary, Ben C. “Survey of Virginia Fluted Points, Nos. 604-679.” ASV Quarterly Bulletin, September (1982).

 

McCary, Ben C. “The Paleo-Indian in Virginia.” ASV Quarterly Bulletin, March (1983).

 

McCary, Ben C. “Survey of Virginia Fluted Points.” ASV Special Publication, Number 12 (1984).

 

McCary, Ben C. “Survey of Virginia Fluted Points, Nos. 733-770.” ASV Quarterly Bulletin, March (1986).

 

MacCord, Howard A., Sr. and Hranicky, Wm Jack. “A Basic Guide to Virginia Prehistoric Projectile Points.” ASV Special Publication, Number 6 (1979).

 

Melchor, James R. “Steel-Bitted Tools and Pattern-Welded (Damascus) Blades.” The Chesopiean, Winter (1999).

 

Melchor, J. R. and Ross, P. M. “Knobs, Knops, Finials, and Drops.” Anvil Magazine, v. 27, no. 1 (2002).

 

Melchor, J. R. and Ross, P. M. “Spring Swages.” Anvil Magazine, v. 27, no. 2 (2002).

 

Melchor, James R. “An Iron and Steel Primer.” The Chesopiean, Summer-Fall (2004).

 

Melchor, Jim, Newbern, Tom, and Melchor, Marilyn. “That Rusty Object – Is It Wrought Iron or Steel?” Edenton Historical Commission www.ehcnc.org, (2013).

 

Painter, Floyd. “Lancets, Unusual Items From Paleo Man’s Tool Kit.” The Chesopiean, February (1973)

 

Painter, Floyd. “One Man’s Trash Is Another Man’s Treasure: A Study in Discarded Tools and Weapons.” The Chesopiean, October-December (1982).

 

Painter, Floyd. “Paleo-Indian Cutting Blades: The Two Basic Traditions (A Revolutionary, Evolutionary Theory).” The Chesopiean, January-March (1983).

 

Painter, Floyd. “Have We Been Mutilating the Wrong Elephants?”

The Chesopiean, Spring (1988).

 

Patterson, Leland W. “Diffusion of Technologies in the Southeastern Archaic.” The Chesopiean, Winter-Spring (1994).

 

Patterson, Leland W. “The Clovis Point: A Changing Picture.” The Chesopiean, Spring-Summer (2001).

 

Peck, Rodney and Painter, Floyd. “The Baucom Hardaway Site: A Stratified Deposit in Union County, North Carolina.” The Chesopiean, Spring (1984).

 

Perrin, Ellen S. “Analysis of Endscrapers from the Williamson Site, Dinwiddie County, Virginia.” ASV Quarterly Bulletin, March (1977).

 

Rountree, Helen. “A Guide to the Late Woodland Indians’ Use of Ecological Zones in the Chesapeake Region.” The Chesopiean, Summer (1996).

 

Sasser, Ray R., Jr. “Regional Adaptation in the Eastern Clovis.” The Chesopiean, January-June (1979).