I might be having a breakthrough with regard to Francis Bell, the brick wall ancestor I called the Mysterious Continental when I first wrote about him. This morning on FamilySearch.org I found a very intriguing record for a man of that name, born 1793 in “Wittaman, Saxony,” in the British War Office Registers 1772-1935 (National Archives reference TNA WO 25).
Francis Bell’s birthplace as recorded in the 1860 census for Ausable, Clinton County, New York.
Unfortunately the digital image of the original document can only be viewed at a Family History Center or affiliate library, and does not seem to be on the National Archives website or among the military records on FindMyPast.com. I will locate the nearest FHC to see it in case there is additional information, and to decipher the town better because I can find no German town of this name. Could Wittenberg, in Saxony be meant? Or perhaps more likely, Wittmund in Lower Saxony?
And was this Francis a member of the King’s German Legion (Des Königs Deutsche Legion,) which, according to Wikipedia, was “a British Army unit of mostly expatriated German personnel during the period 1803–16? The legion achieved the distinction of being the only German force to fight without interruption against the French during the Napoleonic Wars.”
Soldiers’ uniforms, King’s German Legion 1812.
The legion was disbanded in 1816. Perhaps the date on this record, 3 Feb 1816, is the date Francis was discharged. This gives him plenty of time to emigrate, and to be living in Clinton County, New York by 1825. Though of course these may be two different men, and even if they are one and the same, there may be no way to tie the soldier to my New York ancestor.
Compared to finding out that your Dad isn’t your biological Dad, as so many people have recently, my DNA discovery was nothing. Still I was fairly upset when I realized I had no biological link to my supposed agnate (male line) immigrant ancestor, John Thurber (c. 1625-1705). I had been looking at charts of the generations between me and John since childhood, and was attached to my birth surname even though I dropped it when I married. (Having the name Thurber is a little annoying–everyone who has heard of James Thurber asks if you are related, and everyone who hasn’t needs you to spell it for them.)
James Milford Thurber about 1890.
Based on Y-DNA and autosomal test results, I now believe my Dad’s paternal grandfather, James Milford Thurber (1865-1958), was not fathered by his “paper trail” father, Robert Thurber (1835-1896), even though Robert was married to James’ mother, Cordelia Terry, at the time of his conception, James being the fourth of six children registered to this couple. So who could have been his biological father and how do I know this? First we have to go back to the infancy of genetic genealogy.
Cordelia Terry (1841-1937). Although she looks very serious in the picture, my grandfather told me she had a lot of personality, and wanted very much to ride in an airplane in her later years. I believe she was James’ mother, but that his father was not her husband.
In 2005 I heard about Y-DNA testing at FamilyTreeDNA, and I thought it would be a good idea to start a Thurber Family Surname Project. Since the Y chromosome is passed from fathers to their sons unchanged except for occasional mutations, most living men surnamed Thurber and purportedly descended from the immigrant should carry very similar Y-chromosomes. The Y-DNA chain from John would of course be broken wherever there were Non-Paternity Events (NPEs), typically adoptions or infidelities, so I expected that a few participants would not match the majority.
James Milford Thurber’s Vermont birth record.
Of the 13 participants to date, seven are identical matches on all markers tested, and I believe their Y-DNA haplotype is that of the immigrant or very similar to it. One other participant differs only slightly from these seven. The other five, including my Dad, do not match the “real” Thurbers at all, nor do any of them match one another, though one is a Thurber by adoption so would not be expected to match anyone. My disappointment in my Dad’s being one of the outliers was real, and I soon lost interest in the project. (If anyone is interested in taking the reins, let me know.) I wanted to know when the NPE had occurred and, if possible, what had happened.
Geographical distribution of haplogroup R-M269, by Patricia Balaresque, Georgina R. Bowden, Susan M. Adams, Ho-Yee Leung, Turi E. King, Zoë H. Rosser, Jane Goodwin, Jean-Paul Moisan, Christelle Richard, Ann Millward, Andrew G. Demaine, Guido Barbujani, Carlo Previderè, Ian J. Wilson, Chris Tyler-Smith, Mark A. Jobling via Wikimedia Commons.
My Dad’s haplogroup is now known as R1b-M269, and is the same as the matching Thurbers, though he differs from them on too many markers to have a common agnate ancestor within a genealogical timeframe. R1b-M269 is the most common Western European haplogroup, and is believed to be associated with the migration of Indo-European-speaking people into Western Europe several thousand years ago.
Recently I had my Dad test with the British company LivingDNA, and according to them he belongs to a subgroup of R1b-M269 currently known as L21 or the Atlantic Celtic branch. This pattern is most common in Scotland, Ireland, Wales and Brittany and is believed to be associated with Bronze Age people who spoke Celtic languages and migrated to the British Isles about 4000 years ago.
Bandirran Stone Circle, near Collace and Coupar Angus, Perthshire, Scotland. Photo by Snaik via Wikimedia Commons.
My Dad’s exact haplotype–his particular constellation of markers–seems to be quite rare however, to the point that he has never had even one exact match to this day. However I noticed early on that many of his fairly close matches were to men with Scottish surnames. My Dad’s Thurbers had lived in the Eastern Townships of Quebec for a few generations, where there was heavy Scottish settlement in the 1800s, so it seemed likely that the NPE was there.
Finally in 2016 he had a very close match. Both he and this other gentleman tested at 67 markers, and there was only one marker that was off, and that was only off by one number. According to the FamilyTreeDNA Time Predictor (TiP), there is about a 90% chance that this man and my Dad share a common agnate ancestor within four generations, and a 99% chance they share one within eight generations.
This man’s surname was Robertson, the fifth most common Scottish surname, and his agnate line traced back to a family that lived in the same small village–Tingwick, Quebec–where the Thurbers had lived in the mid 1800s. If the Most Recent Common Ancestor (MRCA) they shared was the Scottish immigrant Leonard Robertson (1790-1863), this is five generations back from Mr. Robertson, and I now believe four generations back from my Dad. So far, so good.
I was already very familiar with this Robertson family, because Cordelia Terry’s three eldest siblings married children of the immigrant Leonard Robertson in 1857: Amy Emerilla Terry married Robert Kennison Robertson, Nancy Terry married John Kane Robertson, and Galin Terry married Alice Robertson. So Cordelia was already a triple sister-in-law to the Robertsons well before 1865 when James Milford Thurber was born.
Royal Sappers and Miners in uniform in Leonard Robertson’s time.
The immigrant Leonard Robertson was born 14 Jul 1790 in the parish of Collace in Perthshire, and served quite a few years in the British army’s Royal Sappers and Miners. He married a Frenchwoman named Catherine Mabille around 1818, and they began their large family in Scotland, but lived in the Waterloo Settlement at Lac Beauport, Quebec by 1831.
The Robertsons moved to Tingwick too late (1840 at the earliest) for Leonard to have fathered Robert Thurber, born in 1835. In other words, I think it is unlikely the NPE is in the preceding generation, because the Thurbers and Robertsons would not even have known each other in 1835, living over 100 miles apart. And Leonard died 9 May 1863, too early to have fathered James Milford Thurber himself. The most likely scenario is that one of his five sons who lived to 1864 did–James (1819-1874), Robert Kennison (1826-1907), Thomas Kennison (1831-1893), John Kane (1838-1890) or Leonard R. (1841-1904).
When Ancestry began to offer autosomal testing, I bought my Dad a kit. If my Dad is a Robertson descendant, his autosomal DNA should match some of the many living descendants of Leonard Robertson and Catherine Mabille at levels that are within the possible ranges for their ostensible relationships. Since he is also a Terry descendant, let’s keep it simple by ignoring his double-cousin matches to descendants of Robertson-Terry marriages.
He has 28 easily verifiable and centimorgan-appropriate matches to descendants of Leonard Robertson and Catherine Mabille who are not also Terry descendants, though there are probably many more who have incomplete trees on Ancestry. Also he has no matches to any Thurber descendants who are not descended from his own grandfather, James Milford, or great grandmother Cordelia Terry.
I feel it is very likely that my Dad is an agnate descendant of one of the sons of Leonard Robertson, though I will never know which one or the exact circumstances. I think the most likely candidate is Leonard R. Robertson (1841-1904) because he was closest in age to Cordelia and was also the only unmarried brother at the time of conception–James Milford would have been conceived around 20 April 1864, and Leonard R. Robertson married Matilda Almira Barlow two months later.
I believe Cordelia Terry must be James’ biological mother because there is no other Terry woman who could have been, and so no other way to account for my Dad’s autosomal matches to Terry descendants who are not also Robertson descendants: Robert Kennison Robertson and Amy Emmarilla Terry had no children, and John Kane Robertson and Nancy Terry had a daughter in November of 1864, so could not have had a son in January of 1865.
Yes, I lost a long branch on my family tree, one stretching to 17th century Massachusetts, with interesting people and places. And Leonard Robertson is pretty much a brick wall, though interestingly his wife is not, and her father was among the French soldiers who fought in the American Revolution. Still, the Robertsons’ roots likely stretch deep into Scottish history, even pre-history, and few places are as beautiful or romantic to me as Scotland. My husband and I spent time in the part of Perthshire where Leonard was born in 2017, Collace being the closest village to Dunsinane Hill.
Several years ago I saw a woven coverlet made by David D. Haring of Rockland County, New York and Bergen County, New Jersey at the American Folk Art Museum in New York City. I knew that my grandmother was descended three different ways from the immigrant Jan Haring (1633-1683) and his wife Grietje Cosyns Van Putten (1641-1724) who settled in what is now Rockland County, so I was sure I must be related to him in some way. (An interesting book about the Harings is Firth Haring Fabend’s A Dutch Family in the Middle Colonies 1660-1880)
When I got home I determined that David was the brother of my fourth great-grandmother Elizabeth Haring (1795-1870), their parents being David A. Haring (1760-1801) and Maria Alyea (1775-1863). Like many 19th century Americans, David was a farmer but also had a “side hustle,” in his case weaving custom coverlets on a jacquard loom. A jacquard loom uses a system of punch cards to simplify the weaving of complex patterns. You can see one in action here.
Jacquard coverlets were very popular in the United States during the early to mid 1800s, and were usually woven with one or two colors of handspun wool on a ground of white cotton. The customer selected the color(s) and patterns, and many were personalized with the owner’s name or initials and sometimes the date of manufacture. This information would usually be woven twice, both backwards and forwards, to be legible from either side. David’s “signature” was a rose with four leaves in the corners of many of his pieces.
The Metropolitan Museum of Art in New York has at least two of David’s coverlets, the Art Institute of Chicago has one, the Henry Ford Museum in Dearborn, Michigan has another, and the McCarl Coverlet Gallery in Latrobe, Pennsylvania has five. One sold at auction in 2016 for $1100.
Crib quilt woven by David D. Haring, from the collection of the Metropolitan Museum of Art.
David’s estate was valued at over $45,000, or about $1.5 million in today’s money, so his work (possibly combined with rising land values) paid off.
Plough and Anchor Genealogy 