Stretching a AncestryDNA match?

Question

Trying to decide whether or not to make a parent/child connection in a couple of profiles.

I have a set of 3rd great-grandparents that have been a brick wall for many, many years, William Smith (Smith-132204) and his wife Delila (Newton-7890).  Little is known of them other than dates that appear to be from a gravestone and a bit of census records.  I did my AncestryDNA in early 2017 and although it didn't really tell me much that was new in my direct lineage, it did show a couple of interesting cousin matches.  One was a cousin who was a paternal descendant of William's and I will address that in another question some day.

My situation deals with Delila.  On Ancestry, she had been identified as a Yates (wrong!) for many years and it had been copied from tree to tree many times.  If the people had ever actually looked at the sources they were copying, at least some of them should have realized that they were combining two couples of the same names but who lived in different counties in Tennessee.  Anyway, year or two ago, I finally found her actual LNAB, which was Newton; however this was just as much a brick wall.  Research into the Newton family in Giles County, Tennessee turned up several males who could be a father and several sons.  However, as she was born and married before all family members were enumerated on the census, I have not been able to make a direct connection.  I even created a descendancy tree on Ancestry for them in the hopes that others would contact me, no luck. its called "Descendents of William and Delila Smith of Giles County, Tennessee" in case anyone is interested; it was quickly assembled from census records on Ancestry primarily and is available for public view.

Continuing, not much came from using the AncestryDNA matches, over 2000 of them and few tools to help organize them.  So not long ago, I decided to add the person I thought to be her father to a hidden tree I have on Ancestry (for testing relationships) and within a day or two, more than a half dozen members were added to the DNA/GED match list, all of whom had the individual as our common ancestor.

Now to the point!  This looks like a pretty good indication to me, but I also know that if everyone who was matched, listed Mickey Mouse as an ancestor then it would show that all of us were Mouse descendants, perhaps explaining why everyone in my family loves Jarlsberg cheese!  Anyway, I recently added the potential parent, John Newton (Newton-8908), to Wikitree but I have not yet connected Delila to him.  So, the big question is, "Should I link them?" even if there are no sources to document their relationship.  If I do, I intend to put an obvious notice in their biographies (to be written) that the connection is based on my interpretation of a very meager historical record and autosomal DNA matches, and that no definitive proof has yet been found of their relationship.  So, your thoughts or policy references would be appreciated.

Thank you for sticking with me this far, verbosity problem.  Would also love to hear from anyone with any knowledge of these people.  Always glad to hear from potential cousins.

Answer 1

I'd say link them, but set the relationship to uncertain, and then put a little blurb in a research notes section of each profile explaining why you linked them and why it's uncertain. I'd also suggest you point out why Delila isn't a Yates and include that in the research notes section for her profile as well. I've had a similar issue with an ancestor of mine, but due to paternity issues rather than records being lost to time, and that's what I did for the profiles with the conundrum.

Comments

Thanks for the reply!  Thats pretty much what I will do, but as you said, I'll put it in a Research Notes section rather than the biography like I had planned.  I've already added some sources on her profile for her actual maiden name.  Unfortunately, there are only two sources, one of which, a daughters death certificate, was the clue that finally broke through the brick wall.  The daughter, not my direct line, had moved with her family to Texas in the later 1800s and lived long enough to have a standardized Texas death certificate with her mothers maiden name itemized on it.  
Again, thanks for answering.  I've been doing my family genealogy since my early teen years back in the 1950s and only found Wikitree a little over a year ago.  Slow going for me on here but wanting to get as much as I can of the collected info (with sources) on here before my eyes give out.