Wednesday, November 3, 2021

Disproving an AncestryDNA ThruLine

 I have over 450 AncestryDNA ThruLines listed in my DNA Matches, and have entered many of them into my RootsMagic family tree over the last several years.  If the ThruLine provides enough information linking my DNA Match to a known common ancestor, then I add the information for every family in the DNA Match's "string" to RootsMagic (and eventually into my Ancestry Member Tree).  Once you realize that the ThruLines depend on the accuracy of one or more Ancestry Member Trees, then you know that the DNA Match's "string" of ancestors may not be correct and you have to critically evaluate every link of that "string."  The error my be in your tree, your DNA Match's tree, or in another Ancestry Member Tree, or even in Ancestry's hidden "Big Tree."

Such is the case for ones of my DNA Matches with 16 cM in 1 segment (so probably a 4th to 6th cousin) with potential common ancestors of my 5th great-grandparents Burgess Metcalf (1741-1816) and his wife Jerusha --?-- (1750-1817) of Piermont, New Hampshire.  

1)  Here is the ThruLines chart for Burgess Metcalf:

The ThruLines in question is the descent from Melatiah Metcalf (1779-1838, the daughter of Burgess and Jerusha (--?--) Metcalf (the second "string" from the right).  Melatiah Metcalf married Salmon Niles (1768-1852) in 1798 in Piermont, New Hampshire and they had at least 13 children, including a Levi Niles (1814-1867), all born in Grafton County, New Hampshire.  

2)  Here is my Ancestry Tree page for Melatiah (Metcalf) Niles before I added all of the children:

Here is my RootsMagic 8 screen for the Salmon and Melatiah (Metcalf) Niles family after I added all of the children (not all children shown):

3)  The ThruLines for my DNA Match gives the following "string" from Burgess/Jerusha and Melatiah Metcalf:

*  Levi Niles (1800-1865)

*  Moses Niles (1820-1906)

*  Lewis Erwin Niles (1864-1944).

*  and three more generations to my DNA Match.

Several alarm bells went off when I saw this "string," including:

*  The "string" says Levi Niles, son of Salmon and Melatiah (Metcalf) Niles was born in 1800, but the records say 1814.  The tree of the DNA Match had Levi without parents, and a birth year of 1800.  Apparently, Ancestry found my tree (or someone else's tree) with Levi Niles born in 1814 in it and matched it to the Levi Niles born in 1800 without supporting evidence of parents.

*  Moses Niles was born in Steuben County, New York, not in Grafton County, New Hampshire.  

4)  The information in my RootsMagic tree for Levi Niles shows that he married Mary P. Pool (1818-1884) in 1837 in Grafton County, New Hampshire.  They had at least 7 children according to the 1850 and 1860 U.S. Census records.  Here is the 1860 census, which confirms Levi's age in 1860 consistent with an 1814 birth year (as do other records):

Because Levi, grandson of Burgess and Jerusha (--?--) Metcalf, was born in 1814, it is really not possible for him to father Moses Niles born in 1820 (several AMTs say he was born in Bath, Steuben County, New York).  So the first conclusion I can draw is that Levi Niles (1814-1867) is not the right "connection" for this ThruLine.

5)  So who is the "right" common ancestor for this probable DNA match of 16 cM in 1 segment with one shared match?  There is no guarantee that the Niles line from the DNA Match is the correct line.  If  my DNA Match is a 5th or 6th cousin, then there are hundreds of possible strings from him back to one of my ancestors.  My DNA Match has a Private Member Tree with 11 persons in that tree.  So the common ancestor could be any one of my 64 4th great-grandparents or my 128 5th great-grandparents.

I have marked my Note for this DNA Match with "XXX - 16 cM in 1 segment - private tree (11 people) - one shared match with unknown line; NOT a ThruLine to 5th great-grandparents Burgess Metcalf and Jerusha --?--.

I should note that only about 10 of my AncestryDNA Matches are noticeably wrong - many of them are correct.  My conclusion is that the AncestryDNA Thruline feature is very useful, but occasionally incorrect.  Users need to critically evaluate each element of the lines provided back to the common ancestors in my tree.


Disclosure:  I received a complimentary test kit from AncestryDNA a long time ago.  I have a complimentary all-access subscription from, for which I am thankful. has provided material considerations for travel expenses to meetings, and has hosted events and meals that I have attended in Salt Lake City, in past years.

Copyright (c) 2021, Randall J. Seaver

Please comment on this post on the website by clicking the URL above and then the "Comments" link at the bottom of each post.  Share it on Twitter, Facebook, or Pinterest using the icons below.  Or contact me by email at


Linda Stufflebean said...

I have two types of Ancestry ThruLines. Either the line is accurate and I can document the people OR, in the case with my mystery lines. suggestions are either made up people for whom I can't find a single bit of documentation or the suggested person is so wrong that 5 minutes of searching would determine there is no way for it to be possible. I check back every so often hoping that a real clue will emerge!

Marshall said...

Yes, but how do you tell Ancestry "I believe this connection is incorrect; please don't match me through this person"?

I have several Ancestry DNA matches for my wife who connect through "Robert Argue (b1811)" who was born in Ireland and emigrated to Canada about 1828, and then married a woman named Elizabeth Armstrong (b 1815) in Canada.

However, I believe that my wife is descended from "Robert Argue (b 1794)" who was born in Ireland, married Elizabeth Gilliland (also b 1815) in Ireland, and came to a different part of Canada in 1848.

But Ancestry keeps serving up "George Argue (1811)" as a common ancestor.

Marcia Crawford Philbrick said...

Thank you for this post! It reinforces the idea that I need to continue researching the descendants of my ancestors. Unfortunately, I'm not as far back as you in this effort.

Danine Cozzens said...

I have a somewhat different case: ThruLines matched me with "DC" at 1 seg/8cm, with my gggf as MRCA. So far so good. Then ThruLines connects him as a son to someone born in 1805. [I'm old but not that old.] From your post, would filling out my own tree to give grandchildren to that gggf might help clean this up from my end?

Randy Seaver said...


Yes, finding all descendants of your gggf might be helpful - you can probably find his great-grandchildren using online research and your family information.

Ancestry is using someone else's tree, or your tree, to identify the 3rd great-grandfather. Is that father of your gggf wrong? Somebody's tree is probably wrong but that is tough to fix in ThruLines.