Those Green Leaves


Anyone who has taken the AncestryDNA test is aware of the iconic green leaf, known officially as the ‘shared ancestor hint.’ This is a tool that reveals a potential shared line of descent with your DNA matches. This does not conclusively demonstrate that the shared DNA is from those ancestors, but there is a likelihood that it might be the case.

My grandmother, Margaret Pittman, passed away in February 2016 at the age of 97. I had her tested several years ago using AncestryDNA though, and I can see from her kit that she has 32 Shared Ancestor Hints. I recently had my DNA analyzed by AncestryDNA too. The admix results were pretty good, and I will get to that later, but for now I will focus on the Shared Ancestor Hints of which I had only 17, just 10 of which I have in common with my grandmother (and two of those are my grandmother herself, as well as my father John).

That leaves eight Shared Ancestor Hints in common between my kit and her kit. Mathematically, this makes sense. I only inherited at most a quarter of my DNA from my paternal grandmother. So approximately a quarter of 32 is 8. The more interesting question is which quarter did I inherit, or from which ancestors did I inherit DNA?

Four of my eight matches (50 percent) descend from the marriages of Graham Collins and Nancy Arline, or the marriage of Thomas R. Collins and Polly Arline, in Gates County, North Carolina, in the early 1820s. It’s hard to tease these lines apart, because you have an example of a set of siblings marrying another set of siblings.

The remainder of my common matches with my grandmother descend from associated families — the Arlines, either through Jesse Arline, the father of Nancy and Polly, or through his father, James Arline, the Revolutionary veteran, or his mother, Mary Kittrell’s family.

This is really fascinating to me. It shows that my blog has been focused on the very people whose DNA I inherited.

I also match someone who descends from my ancestor William Howell and his wife Jerusha Rawls of Nansemond County, Virginia. I was aware of the Howell match through a relative who contacted me via 23andMe. However, I do not match my grandmother’s Rawls, Pittman, Scoggins, Maget, Cross, or other matches. I assume that a) I did not inherit DNA from those ancestors or b) There are other descendants, carrying other segments of their genomes who have not yet tested or identified their ancestors by building trees online. You can get a better sense of these families here.


Anyway, this has been a fun exercise in an entertaining way to use AncestryDNA. As for the admixture results, this time I was rather pleased. My main way of gauging results is not necessarily if they match my tree, but if they match the results of other family members testing by the same service. For instance, 23andMe does not give my mother or me any South Asian ancestry. However, AncestryDNA showed that both my mother and I carry about 2 percent South Asian, or Asian Indian, ancestry, which is a result that multiple GEDmatch calculators have produced. DNA.Land also said I had 2.3 percent Gujarati ancestry and assigned my mother’s kit 3.3 percent Gujarati ancestry. I tend to believe these results have some validity and are not noise.

Here are the breakdowns for my father, mother, and me, according to AncestryDNA. They are pretty accurate. Obviously, if my father is 51 percent Italy/Greece and my mother is 33 percent, I should be about 46 percent. 23andMe showed me as 49 percent “Southern European” which seems somewhat in line with these results. If you add up the 46 percent Italy/Greece, with the 3 percent Middle East, plus the other non-Northern European ancestry, you get about 50 percent. I am curious though about results like European Jewish, Finnish, or Europe East. I can imagine that for my parents, this represents some kind of Balkan ancestry, inherited from the Italian sides of their families. The Finnish though is more mysterious — our Scandinavian results are so small that it’s odd that some part of it would be identified as Finnish, not to mention that Finns and Scandinavians are not actually genetically similar. Maybe it’s noise, or maybe some component of any Amerindian ancestry is being read as such. DNA.Land, for instance, gave my father about 2 percent Native American ancestry, but my grandmother, his mother, about 7 percent of something called ‘North Slavic.’ Family Tree DNA also gave my grandmother a 40 percent (!) Eastern European result. When I have looked through the admix matching populations I can see that the Amerindian result grows higher as you head east. So that Ukrainians, or Mordovinians, have amount the same amount of Native American ancestry as my grandmother, who was primarily descended from British colonial settlers, does.

Here are the results. Perhaps this is helpful or interesting to some of you.

                               My father               My mother          Me

Italy/Greece       51%                          33%                         46%

Europe West      27%                           2%                           6%

Ireland                8%                             15%                         15%

Scandinavia      7%                             4%                           6%

Great Britain     3%                            25%                         9%

Europe East        2%                            6%                           7%

Middle East        <1%                             –                             3%

Caucasus            <1%                           7%                           <1%

Euro – Jewish      –                                 5%                           3%

Asia South           –                                 2%                           2%

Africa North       –                                  1%                          <1%

Finland                –                                    –                                1%

Posted in Uncategorized | Leave a comment

Hidden Stories


Chromosome 20

A section of my grandmother’s chromosome 20, showing Amerindian, Siberian, and Oceanian segments.

This may be one of the most difficult posts I have attempted to write, because it concerns DNA, and where it can lead you.

Most of us who have been doing genealogy for decades were used to the paper game. We were given interesting-looking pedigree charts and family sheets, and the objective was to fill them out with as much corroborating evidence, records, as possible.

For a certain kind of mind, such a game is as much fun as trying to solve a Rubik’s Cube. Autosomal DNA testing, however, changed all of that. For once, we could see, more or less, what DNA we had actually inherited from our ancestors. It could reveal unknown relatives to you, or verify or discredit long-cherished family traditions of descent.

My objective with my grandmother’s ancestry was to use DNA to see deep into her past. Her ancestors had described themselves as “English,” but I had found multiple instances of relatives being described as “mulatto” or “black” in records. Her admixture results from the major providers — AncestryDNA, or Family Tree DNA — were mostly useless. Was I really to believe that a woman with colonial British ancestry was 4 percent Greek? AncestryDNA said she was only 6 percent British. Family Finder said that she was 40 percent Eastern European and 17 percent British. Which was an impossibility. Although I have a feeling her Native American ancestry was causing the Eastern European result.

Like a lot of people, I sent my kit to Doug McDonald, who did the analysis for free. When he analyzed her Geno 2.0 kit, it reported something back called Na-Dene, at 1.7 percent. Mayan was a second small signal, at 1.3 percent. The Na-Dene sample is found in northern Canada and represents an indigenous population, as does the Mayan of course. After that, I turned to using various GEDmatch programs to explore this signature, which was confirmed using the various Eurogenes tools (K13, K9b), Dodecad World9, HarappaWorld, PuntDNAL, Gedrosia, and even MDLP World (which splits the signal into 0.8 percent Arctic_Amerind and 0.5 percent Mesoamerican).


My grandmother’s kit analyzed using the Eurogenes K13 tool on GEDmatch

Using chromosome painting tools, I set out to discover where these Native segments were, and if I could find anyone else who matched them. I decided to use Eurogenes K13 to paint the chromosomes according to ancestral origin, in part because the Indian and Siberian segments are painted shades of blue, which makes them easier for me to see. I thought, initially, that I would be able to find one line of my family that was Indian, to find others who descended from that line, and that the Native American DNA would be spread evenly among those descendants. What I found, however, is that the Native segments were quite random, and scattered among people not from a particular family, but who had ancestors in a certain location.


Moreover, the DNA could be found in diverse people with whom I shared no obvious common ancestors, from Choctaw Indians in Oklahoma, to people in Texas who had some tradition of Cherokee ancestry. Yet when I traced these families back, I inevitably found links to people who had once been living along the Roanoke River in Bertie County, North Carolina. Names like Whitmell, Harrington, Blount, Craft, Owens, Butler, Hollom, Lloyd, and Bryant, to name a few. In short, the characters found on these Bertie County Bastardy Bonds from the 18th century. Some members of these same families — Collins, Lloyd, Cane, and Lumpkin — later applied for Cherokee benefits from Georgia, where they were denied, as having no known link with the Eastern Cherokee.

One record that I found most interesting was this one:

16 Dec 1741 Bertie County, NC Deed Book F, p. 368: Susannah COLLSON and John COLLSON (her son) to Thomas BLOUNT and Thomas WHITMELL, 16 December 1741. 19 June 1742. 500 pds. for 600 acres “…Executors of the Last will and Testament of John COLLSON, Sen Dec’d . . . except for thirty two pounds quit rents Deducted to us paid by Thomas COLLINS . . . ” Land on SS Rocquis Creek.

This placed the name “Thomas Collins” on the periphery of the Tuscarora Indian Reservation in the 1740s, along with the Colsons, Blounts, and Whitmells. All of these families were involved with the Harringtons, a name that continued to surface among matches, and most of these families were involved in the Indian trade, from dealing with the Tuscarora to trading among the Chickasaw, Cherokee, Choctaw, and Creek. While it did not mean that I descended from that specific Thomas Collins, who was living in Bertie in the 1740s, it was possible that my branch of that family was connected to him in some way, considering the DNA matches coming from that same neighborhood.

The really fascinating aspect of this journey, is that it has been almost impossible to document on paper. This really questions some of the underlying principles of modern genealogical practice. While the process of digging through archives, exploring associations between people of certain surnames, or tracking land transactions can form the bones upon which family histories can be constructed, DNA can point us in the direction of other, hidden stories in our family’s ancestries. Many of my matches from this community, which originally was located on the Roanoke River around Indian Woods, have admixture results suggesting deep mixed ancestry. Most have majority African or European ancestry, but autosomal testing allows us to see those distant Indian, African, European, or even South Asian ancestors who have been obscured by racist laws, record loss, and, in some cases, purposeful erasure of different aspects of family history.

Posted in Uncategorized | Leave a comment

Custalow and Dennis

Members of the Pamunkey Dennis family in the 19th century.

The following is a pretty good assessment of the amount of guesswork that goes into picking apart these would-be/could-be Indian families.

I recently came across a marriage record for a woman named Margaret Dennis and a man named Edward Custalow in Gates County. Actually, the man’s last name was transcribed as “Cutteller.” The marriage bond is dated 2 September 1825, and the bondsman was Charles E. Sumner.

Edward Custalow signed with a mark. His race is not recorded, but the race of the men and women described as Chowan Indians in deeds at this time was not recorded either.

Charles E. Sumner comes from a family that interests me. There were two Sumner-Beasley marriages in Gates County in the early 1800s, and a Josiah Sumner married Molly Collins in Gates County in 1804. Josiah Sumner appeared on the estate record of James Beasley in 1815, as did my relatives James Russell and James Collins. This suggests kinship between the Beasley, Russell, Collins, and Sumner families in Gates County.

Custalow and Dennis are not Gates County names, however. They do not appear in censuses at that time and I cannot trace the trajectory of this particular couple at all.

I can tell you that Custalow, or Castellaw, was the name of a Bertie County family though. Heinegg has drawn up a genealogy for this family based on the common law marriage of William Castellaw and Martha Butler, a mulatto woman, in Bertie County in the 1750s. He also suggests that Martha Butler was the head of a household of 10 “other free” in Gates County in 1820, even though she would have been between 80 and 90 years old. Heinegg subsequently shows how members of this family eventually emigrated north to King William County, Virginia.

K. Paul Johnson in his excellent book Pell Mellers: Race and Memory in a Carolina Pocosin, discusses the Tuscarora Indian ancestry of the Butler family. Another book, Ross Baptist Church, The First Seventy-Five Years, 1800-1875, also makes this claim.

Actually, it’s unlikely that the Martha Butler in the 1820 Gates County census is the same woman who was married to William Castellaw in the 1750s. The Martha Butler in Gates County actually inherited 10 acres from Sarah Butler when Sarah Butler stopped appearing on Gates County tax lists in 1816. Sarah Butler gained this property in an 1803 deed from John Lang, witnessed by my ancestor Thomas Collins and James Ransome. Still, it’s possible that if Butlers moved into Gates County from Bertie, Castellaws, or Custalows, might have migrated in that direction too. Perhaps Edward Custalow was one of the 10 “other free” members of Martha Butler’s household in 1820. This is speculative.

The name Dennis is interesting because “Billy Dennis” and “Sarah Dennis” appeared on the Tuscarora Indian Woods land deeds between 1766 and 1777. While I can find no Dennis family in Gates County at that time, a “Devere Dennis” is listed as the head of a household of 9 “other free” in Bertie County in 1800. “Littleton Dennis” is listed as the head of a household of 2 “whites” in 1790. I refer to these categories with caution because, as I have shown, they were assigned arbitrarily.

In the 19th century, both Dennis and Custalow were to emerge as common surnames among the Pamunkey and Mattaponi Indians in Virginia. One question that remains is whether or not they were Pamunkey Indians who had been living among the Tuscarora and returned once the community there fell apart, or if they were Tuscarora who married into other groups. The arrival of Dennis and Custalow to Gates County, though, along with Butler, adds some more names to the list of potential mixed-race Scratch Hall residents who reportedly moved into the area in the latter part of the 18th century.

As you can see, all of this is guesswork. One more interesting source though is this invaluable list of Bertie County bastardy bonds from the 18th century. Many of these families can be found on this list, including multiple Castellaws and Butlers. To me, as a researcher, this shows that these were large families, with numerous members who may or may not appear in the records from that time.

Posted in Uncategorized | Leave a comment

Yeopim Indian Town

It’s just a few miles from Indian Creek in Norfolk County, Virginia, to Indiantown, which sits on the border between Camden County and Currituck County, North Carolina.

Indiantown was once the site of the Yeopim Indian Reservation. Whitney Petrey does an excellent job of discussing the acculturation of the Indiantown community in a 2014 master’s thesis from East Carolina University.

I have become more interested in Indiantown because one of my great great grandmother’s sisters married a man who had roots in that community. Lydia Collins (1877-1945), my paternal grandmother’s grandmother, had multiple siblings, including Mariah Collins (1868-1932).

Mariah Collins married Albert Sidney Dozier in Norfolk on January 11, 1888. She was 20. He was seven years older, and a tin smith. His parents were James Dozier and Ann Ayers. I haven’t researched the Ayers family, but the name Dozier caught my eye because Noah Robbins, from the Chowanoke family, married Mary Dozier in Gates County in 1825.

Researching further, I found that James Dozier’s father was named Absalom Sawyer Dozier, and descended from the Dozier, Sawyer, and Lurry families, all from the Indiantown area in Camden/Currituck in North Carolina. The Sawyers and Lurrys, along with the Brays, were among the earliest settlers noted living on the Indian patent.


In this 1780 map of Indiantown, one can find the Dozier, Sawyer, and Lurry families.

Interestingly, Lydia’s other sisters Ruth Collins and Genevra Collins married into the Leggett and Robinson families, both of which were present at Indiantown. However, I have not yet been able to nail down the relationships between all of those people.

I have discussed the Bray family in an earlier post. While not all of the people in the Collins family married into other Indian families in the region, some did. In my previous post, I mentioned the marriage of Burwell Collins and Jane Ellis. Their granddaughter, Mamie Wright (1887-1956), married George Weaver from Norfolk. However, I cannot find a record of this marriage and do not know more about who George Weaver was.

Mamie Wright

Mamie and George Weaver, circa 1911

All of these records seem to tie the Collins family that was living in Gates County and Nansemond County more closely to families in Norfolk, Princess Anne, Camden, and Currituck counties. The fact that Collins appears on the 1907 Nansemond Indian census, as do Sawyer, Weaver, and Price,  affirms this connection.

I have not yet found a hot spot for Collins families in this region, but this may just be, as was once suggested to me, a sort of Anglo-Indian fringe family that originated on the Eastern Shore and married into a variety of communities in Virginia and North Carolina. It’s worth noting that in that 1907 Nansemond Indian census, many of the people are listed with European (white) spouses. One can imagine this process of ‘spin out’ occurred over centuries.

The typical way people in modern-day Indian communities define themselves is through their surname. You will hear that a certain name is affiliated with a particular group. However, in looking at these kinds of patterns, it seems more plausible that any connection to Native Americans was through the maternal line. Which is to say that the reason the Collins family arrived to Gates County was through intermarriage, in my case, with the Russell and (probably) Beasley families that were living in Scratch Hall Pocosin.

Posted in Uncategorized | Leave a comment

Indian Creek Canton

norfolk-indian creekAs far as ‘mystery’ populations are concerned, you can’t get more ‘mystery’ than Indian Creek in Norfolk, Virginia. This meandering water body must have just worked its way just north of the state line and fed into the Back Bay area around Knotts Island, where an “Indian Pond” is located. I can find no modern-day maps that reference Indian Creek, however there is an Indian Creek Road that moves through southern Norfolk and into Virginia Beach.

This is an interesting area, because a number of Indian ‘pilot’ families moved through here in the mid-to-early 18th century. One can find Hall, Bass, Nickens, and others in this region, some of whom later became associated with the Nansemond Indians. Names like Price, Sawyer, and Collins, found on the 1907 census of the Nansemond are also found in this neighborhood.

But who was living in this Back Bay area? Colonial era maps show no Native American villages in the border lands between what became North Carolina and Virginia. Some believe the Chesapeake, or Chesepians, were inhabiting this place, although it is said that Powhatan’s forces eliminated the Chesapeake in the first decades of the 17th century. Moreover, there are no colonial references to Chesapeake people. One can find “Indian Creek” mentioned in Norfolk records as early as the 1680s, and well into the 18th century. There survive some tax records for an “Indian Creek Canton” from the 1730s. Unfortunately, I have not yet seen them.

The other Native American group described in this region was the Weapemeoc, or Yeopim. The etymology of the Yeopim name, the “dawn land people,” is fascinating. This is the same name that the Wabanaki (Abenaki) of Maine and Wampanoag people of Massachusetts called themselves, and demonstrates the links among the coastal peoples.

Both the Chesapeake and Yeopim were Carolina Algonquians, and not entangled in any political alignment with Powhatan’s confederacy. There is a “Monkey Island” in Currituck Sound that is named for the Pamunkey Indians. It was apparently used as a summer refuge for those Native Americans, which brings the Chesapeake, Nansemond, Pamunkey, and Yeopim all into the area. The recurring intermarriage between the group at Indian Creek and the Yeopim in Currituck County and Camden County in North Carolina, the Chowan in Gates County, the Nansemond in Hertford County and Norfolk County, and the Machapunga at Mattamuskeet in Hyde County, North Carolina, leads me to wonder if these “tribal” or “national” names carried much weight among the Indians of the coast.

The one surviving Native American demonym in this area, around modern- day Saint Brides, is “Pungo,” which could refer both to Matchipongo and Pungoteague in Northampton County, Virginia, on the Eastern Shore, and the Machapunga Indians in Hyde County.  Or it just might be a common Algonquian word for “dirt” or “dust.”

Into the 19th century though, marriages did occur between Eastern Shore, Norfolk-Currituck area, and Hyde County families. I have found a marriage record for instance, for a Thomas Collins, from Northampton County, Virginia, who married Missema Chance, of Hyde County, in Norfolk in 1867. There is also a marriage between James Collins, of Hyde County, and Alice Cuffee, of Deep Creek, the traditional Nansemond Indian hot spot, in Norfolk in 1899. Interestingly, though both Collins and Cuffee were from ‘mulatto’ families they were listed as ‘white’ on their marriage record. There were many of these unions.

This shows continuous kinship links among the descendants of the Algonquian groups living on the Eastern Shore and down into the Inner and Outer Banks of North Carolina. I am interested in this area because of the following document:


My ancestor Thomas Collins (1769-1849) had a daughter named Temperance Collins (1793-1868) who married a cousin named William Collins. Burwell Collins (1822-1877), the son of Temperance, married Mary Jane Ellis. So I was looking into any Ellis-Collins connections in the region when I found this will from Norfolk in 1803. It’s possible that this Thomas Collins and my ancestor are the same person. The name “Stephen Price” also brings to mind the Prices on the 1907 Nansemond census. In looking at these names, it appears that all of these people were living in and around Indian Creek in Norfolk. But who were the Indian Creek people? Were they displaced Yeopim, “Dawn Land People”?

Posted in Uncategorized | Leave a comment

The Coastal Fringe

AskiminokonsonIf you have been reading this blog for a while, you can see how I have pinged back and forth, around and around, between different theories as to the origins of this branch of my family — the Collinses and Russells — and its arrival to the Scratch Hall area of Gates County, North Carolina. So far there are four main theories of arrival:

  • Refugees from Tuscarora Indian Woods. Local historians talk about an influx of mixed-race, Tuscarora-descended people into the Hall area during the course of the 18th century. My DNA matches seem to bear this out, as I match families in and around Indian Woods — Bazemore, Castellow, etc. — that were not present in Gates County. Moreover, the people who match my grandmother on her Indian segments typically have ancestors from southern Bertie County. However, this may reflect the deep shared ancestry of the people in the Roanoke-Chowan area. This theory is supported by some documents, local history, and DNA.
  • In marriages from Norfolk-Currituck-Hyde.  The first three personal names connected to the Collinses in Gates County are Thomas, William, and Lemuel. There is a 1771 apprenticeship in Portsmouth for a Lemuel Collins to learn the trade of a blacksmith, whose father was a Thomas Collins. One also finds men named Lemuel Collins in Saint Bride’s Parish in Norfolk County in the 18th century. The personal name “Lemuel Collins” is also found in a number of other Indian hot spots — among the mixed-race people of Tennessee, at Mattamuskeet in Hyde County, and, especially, in and around Snow Hill, Maryland, near the site of the old Askiminokonson Indian Town, which some believe is the origin of the Indian-related Collins families. This was a gathering place for a number of Eastern Shore peoples, such as the Nanticoke, Pocomoke, and Assateague. This theory is supported by some 19th century marriages in my family into the Norfolk Weaver and Dozier families, for instance.  It’s worth noting that three men named “Graham Collins” — the name of my ancestor — were alive in the 19th century. In addition to my ancestor, who married Nancy Arline, a descendant of the local Williams and Vann families, there was a ‘mulatto’ Graham Collins in Hyde County in 1860, and another ‘free colored’ Graham Collins who married Matilla Barrow — from another Mattamuskeet Indian family — in Pamlico County in 1881. Moreover Russell is a family name on the Mattamuskeet deeds in the 1740s and 1750s. In Gates County, the Collins family also married into the Pierce and Morris families, other names associated with the Mattamuskeet. What you could imagine is a preserved kinship group among the Chowan, Mattamuskeet, Nansemond, and Yeopim Indian peoples in the Outer and Inner Banks.
  • Pamunkey in marriages. The given names of the Collins men in Gates County — Thomas, James, William, and Elijah — match quite nicely with the men recorded as free colored in New Kent, King William, and King & Queen counties Virginia in the 1780s, alongside the Bird, Langston, Sweat, Holmes, and other Indian families. These same names are found among the Collinses and Russells in the 1850s in Nansemond County, along with the Halls, Sawyers, and Reads, other families with a Chowanoke or Yeopim and Nansemond connection, but also the Copelands, Ellises, Boons, Wigginses, and Butlers, who seem to be more tied to the Tuscarora. Another interesting name is Austin. The Austins intermarried with the Collinses and Russells and were their neighbors. They can be decisively traced back to New Kent County, as evidenced by the appearance of “Claiborne Austin” in New Kent County in 1791, followed by his purchase of land in Scratch Hall late that year. Did the Collinses, like the Austins, move from New Kent into Gates County?
  • Fringe Chowanoke.  There is a fourth theory, and that is that the Collinses, despite not appearing in militia and tax lists in Gates County and its predecessor counties in the 18th century, had actually been living at Indian Town the whole time, and only emerged into the historical record once they left the Reservation. The Bennetts and Robbinses also weren’t listed in those records. This is further demonstrated by the fact that the Collinses, like the Pierces and Morrises, were living side by side with the Robbinses in the 1850 census of Gates County. The 1779 Chowan County marriage of John Collins to Sarah Hinton, of a family that had long resided next to the Chowanoke, shows that the family was there at least four years before its first appearances in the Gates County tax lists in the mid-1780s. That John Collins must have died soon after, because Sarah Collins was the mother of one of my ancestor James Arline’s bastard children (his other mistress was named Charity Russell). The fact that the Arlines owned land both around Bennetts Creek and at Mills Swamp on the border, and that James Arline’s granddaughter, Nancy Arline, married Thomas Collins from this Indian-descended family, hints at the idea that the Collinses, as well as some of the Russells, were actually laborers on the Arline plantations. Indeed, throughout his lifetime, my ancestor Thomas Collins (1769-1849) was listed as a laborer, even while the size of his properties grew from nothing to several hundred acres.

So let me know what theory you believe. As you can see, different theories can be supported by different information. One thing seems certain, and that’s that Collins was a family that married into various Algonquian kinship groups over the centuries. I would also speculate that the Collins who arose among the Saponi/Catawba in the 18th century probably also arrived there via an Algonquian kinship connection.

Posted in Uncategorized | 1 Comment



The ancient tartan of Clan Napier

Napier is one of the more unique surnames that has popped up in my research into the Scratch Hall Folk of Gates County. It’s a Scottish surname, like many – Hamilton, Delday, Vann, Russell – in the area. Sometimes rendered as “Napper,” and at least on one occasion, “Knapper,” it belonged to perhaps several generations of men named Robert Napier who were living in what was, at the time, a frontier area (and still is).

There are a smattering of records that provide some clue as to the trajectory of this family. The first is a 1707 record in North Carolina for a “Robert Napier late of New Kent County, Va.” Looking into this record, we find it refers to court case involving “Drew alias Robert Napier,” and was initiated by John Pettiver, who had received land patented by Napier in St. Peter’s Parish in New Kent County.

It is unclear why he prosecuted the case in North Carolina.

This man, Robert Napier, Sr., was born in Virginia in 1660, and was the only son of Patrick Napier, an Edinburgh doctor, and his wife Elizabeth Booth. He owned large tracts of land in Pamunkey Neck, including land still under the administration of the Pamunkey Indians. He also commanded four companies of Virginia Rangers during the Tuscarora War in 1712. He returned to Virginia though, and died there in 1731 leaving behind seven children, none of whom relocated to the Chowan River area.

The next record that shows a Robert Napier in the area comes several decades later, when a “Robert Napier” is recorded in a 1759 tax list for Hertford County. At that time, the western part of present-day Gates County, including Scratch Hall, was part of Hertford County. A review of William Murfree’s tax books for 1768-1770 does not show this name, though these cover the same area. Interestingly, Robert Napier does not appear in any of the colonial era Chowan County militia lists. He is also not in the 1786 state census for Gates County, nor in the 1790 US Federal Census for the county.

There is no explanation for where the Napiers were therefore between 1759 and 1790. They may have been there, and simply not counted. There may be a reason for that, because “Robert Napier” did not become a landowner in Gates County until 16 February 1791, when my ancestor James Arline granted him 50 acres on Beach Swamp, near the Virginia state line, witnessed by James Russell, Thomas Ellen, and Mills Odom, all of whom signed with a mark, meaning they were probably illiterate.

The land, James Arline noted, had been received from his father John Arline, who had died the previous year. James Arline, a Revolutionary War veteran, died later in 1791. And Napier sold the land, however, to William Dilday in 1794 for $36.

Why is this man of interest? In 1791, Robert Napier appeared on the estate record of my ancestor George Russell, who also died that year. He also appeared in the 1800 census for Gates County as the head of a household of six people. One white male under age 10, one white male aged 26 to 44, one free white female aged 16 to 25, another white female aged 26 to 44, one white female 45 and over and one free colored person, age not given. However, based on the fact that three people are listed as being over age 25, the free colored person was under age 25.

One might assume that Robert Napier is the white male aged 26 to 44, giving him a birth year between 1756 and 1774.

There is also a marriage record for Robert Napier from 27 February 1800 to Rachel Griffin. She was probably the daughter of Henry Griffin, the only Griffin listed in the 1790 US Federal Census for the county.  Henry Griffin acquired 100 acres in 1783 from Henry Hill and his wife Elizabeth Russell Hill on Bennetts Creek, bordering the land of the Chowanoke Indian woman Nan Robbins. Robert Napier also served as a bondsman for the marriage of James Collins and Katherine Russell in Gates County in 1801.

Now, Henry Hill’s wife, Elizabeth Russell, was the daughter of William Russell and Sarah Armistead of Cumberland County, Virginia. Their daughter, also named Elizabeth, married James Robbins, chief man of the Chowan Indians. The Hills had been in the area since at least the 1730s, when Thomas Hoyter, chief of the Chowan Indians, deeded 50 acres on Bennetts Creek to Henry Hill, John Freeman, Michael Ward, and William Flemmons witnesses.  Flemmons (Fleming) is another name of interest, because John Flemmons also appears on George Russell’s 1791 estate record. I’ll look deeper into that family in a future post.

This shows that these families were living near Chowan Indian Town by the 1730s.

The relationships between the Russells and Armisteads of Cumberland County, the Hills and Russells of Gates County, and the Griffins and Napiers of the same county, as well as the Chowan Indian Robbinses are unclear. Yet it’s clear that the Hills, Russells, Napiers, Griffins, and Robbinses were all related either by blood or by marriage.

After that, Robert Napier appears in a February 1810 deed in Gates County, but is not there at the time of the census. He surfaces in Marion County, South Carolina in 1820 as the head of a household of 10 people, including two free colored persons. He is still there in 1830 as the head of a household of 10. This time, though, there are no free people of color listed. There is one slave though. We also get, for the first time, a better idea of his age, which is listed as being between 60 and 69, meaning that he was born between 1761 and 1770. His wife, Rachel Griffin Napier presumably, is listed as being between 50 and 59, meaning she was born between 1771 and 1780.

Napier was dead by 1840 though. None of his descendants appear as anything but white in records. His son, Robert Napier was born in 1806 and lived until 1891 and listed his parents’ birthplaces as “Virginia.” He was also a baptist preacher in South Carolina. Robert Napier is not a relative, but he was a neighbor, and his ancestry sheds light on longstanding migration patterns between the people on Pamunkey Neck and those in Gates County, North Carolina, and Nansemond County, Virginia.

To summarize:

  • The Napier family was Scottish in origin, and settled in the Pamunkey Neck area during the mid-17th century, leasing land from the Pamunkey Indians, and acquiring large tracts of land, particularly in St. Peter’s Parish.
  • Robert Napier (1660-1731) was a major landowner in Pamunkey Neck and “gentleman” who commanded several companies of Virginia Rangers during the Tuscarora War in 1712.
  • He apparently was in the Chowan River area around 1707, where he was the defendant in a court case initiated by a New Kent man named John Pettiver.
  • Fifty years elapse until another man, named Robert Napier, is named in a 1759 tax list in Hertford County.
  • It’s another 30 years before we meet this family again in a 1791 land deed. In 1800, another Robert Napier, by this time living on the Virginia-North Carolina border, married a woman named Rachel Griffin from a family on Bennetts Creek in the southern part of Gates County, who was living alongside the Chowan Indians but not named as such.
  • This family later relocated to Marion, South Carolina, by the 1810s.


Posted in Uncategorized | Leave a comment