I raised an issue in my recent e-mail message ("RE: Why Name a Child James Arthur BROWN?", Wed 6/8/2016 1:10 AM) which I want to share with a larger possible constituency of researchers.
I had previously understood that researcher Anne Lee had identified the marriage of a Robert BROWN and an Elizabeth REED on 18 Mar 1828 in Jessamine, KY, as possibly the marriage record of Robert and Mary Ann BROWN. (I haven't seen the underlying primary record.) The record Anne Lee has identified also seems to be about RIGHT as to TIME and would seem to fully explain the known children, so there is much to recommend it.
In my message, I identified another marriage as possibly pertaining to Robert BROWN (b 11 Mar 1800 - PA, d 31 Aug 1853 - Livingston, IL) and his wife Mary Ann:
Robert R. BROWN m Mary Ann KINDALL/KENDALL on 14 Feb 1833 in Bourbon, Kentucky. Joseph KENDALL was a bondsman for this marriage.
See: "Kentucky, County Marriages, 1797-1954," database with images, FamilySearch (https://familysearch.org/pal:/MM9.3.1/TH-1-9999-17962-71?cc=...
: accessed 8 June 2016), 004542764 > image 106 of 343; Madison County Courthouse, Richmond.
See also: "Kentucky, County Marriages, 1797-1954," database with images, FamilySearch (https://familysearch.org/pal:/MM9.3.1/TH-1-9981-55669-75?cc=...
: accessed 8 June 2016), 004542762 > image 564 of 729; Madison County Courthouse, Richmond.
* * *
Anne's identified record begins with the ex ante assumption that Mary Ann's maiden name was "REED," based upon published accounts mostly written after Robert and Mary Ann's death. Then, Anne rationalizes that Elizabeth and Mary Ann might be the same person.
The record I suggest for consideration does NOT make ANY assumptions as to Mary Ann's maiden name (or prior married name, if any), but rather relies upon the reasonably well documented primary evidence that Robert's wife identified by name in the 1850 Census record was named "Mary" with possibly a middle name of "Ann."
This does NOT mean that Robert BROWN didn't marry a REED or that he didn't marry Elizabeth REED. These records are NOT really mutually exclusive.
One bare possiblity is that Robert first married Elizabeth REED and then later married Mary Ann KENDALL after Elizabeth's death. This is admittedly a very speculative ascription.
One of the "facts" which would seem to preclude this construction is the "established" date of birth of Alexander BROWN, who seems to be shown to be born on 21 Aug 1833 in Batavia, Ohio. If this date of birth is correct, this would seem to preclude the possibility that Mary Ann KENDALL could be Alexander's mother, since the Robert BROWN-Mary Ann KENDALL marriage shown in PRIMARY records to have taken place on 14 Feb 1833. This marriage date also would preclude Elizabeth REED from being the mother if she was already dead by the 14 Feb 1833 date of this possible second marriage.
Alexander would have had to have been born earlier or later, for example in 1832 or 1834 for this construction to work. Reexamination of EACH primary record pertaining to Alexander's age and year of birth would seem to be indicated.
Without expressly examining and assessing each primary record regarding Alexander's age and date of birth here, I would at least point to the 25 Aug 1863 Civil War Draft registration record which show Alexander BROWN to then be age 33 (b abt 1830) as indicative that there was some disagreement in the primary evidence.
* * * * *
There is actually a VERY SCIENTIFIC and RELIABLE WAY to explore whether Robert BROWN had one wife or more than one using autosomal DNA testing. This testing is exceptionally PRECISE and would reveal to a very high degree of certainty whether there was more than one wife.
What is needed is an autosomal DNA test from a descendant of one of Robert's eldest two children's descendants -- Nancy Sophia BROWN or James Arthur BROWN -- and one or more of Robert's youngest children: Ezekiel, Elizabeth, William, George, or Alonzo.
Since Alexander's date of birth seems to be the subject of some uncertainty, his descendants are less than ideal test subjects to determine whether there was one wife or two, but testing a descendant of Alexander would establish whether he was related to the SAME degree as the elder or younger children.
* * * * *
Recognizing that the testing would probably need to be done using third or fourth cousins, let me explain the approach using children and first cousins by way of example.
Each child of Robert and Mary Ann would receive 1/2 of his or her DNA from each parent, but each children would get a different mix of each parent's DNA.
If Robert or Mary Ann could be directly tested, each of their children would share 1/2 of his or her DNA with each parent. Each brother and sister would also share ONE HALF of the DNA with a full sibling. However, a child would share only 1/4 of their DNA with a half brother or sister.
Thus, if we could directly test Nancy Sophia and James Arthur, if full brother's and sisters, they would each share 1/2 of their DNA. If the younger siblings were also full brother's and sisters byt he same two parents (Robert and Mary Ann), then the elder children would also share 1/2 DNA (a differet half in each case) with the younger children two.
Thus, if ALL of the children shared rather precisely one half of each other's DNA, then there would be ONE WIFE (Mary Ann). If, by contrast, the elder children had a different mother, such as Elizabeth REED, they might share only 1/4 of the DNA with the youngest children.
Pushing this down another generation, assuming ONE WIFE, then ALL of the grandchildren would be first cousins. First cousins share 1/8 of each other's DNA -- 12.5% -- while half first cousins share 6.25% of their DNA. If we could reach back in time and test the grandchildren through both the elder and younger children, a 12.5% match would support the one wife hypothesis while a 6.25% match would show either that there had to have been more than one wife OR that NOT all of the identified children of Robert are actually children after all (that some are related in another way).
So if we could test Samuel A. BROWN (b 1866, d 1942) and James W. BROWN (b 1886, d 1960) or others of that generation (first cousins) a 12.5% match would prove ONE WIFE and a 6.25% match would prove more than one wife.
So if we could test Hazel M. BROWN (b 1898, d 1969) and James W. BROWN, Jr. (b 1919, d 2005) or others of that generation (second cousins), a 3.125% match would prove ONE WIFE and a 1.563% match would prove more than one wife.
* * *
It appears from your posted tree that there may be some living second cousins known to descend from Robert BROWN. It would seem likely that there are many living and readily identifiable third cousins.
The precision of the autosomal DNA testing is such that it will give a near certain result whether the persons tested are second, third or fourth cousins.
Another advantage of the autosomal testing rather than the Y-Chromosomal testing is that the test subject need not be male and no patrilineal relationship is necessary. The test subjects can be related in essentially ANY WAY. What is most helpful, but not essential, is some understanding of each test subject's family history going back to some point of uncertainty.
* * * * *
A similar test subject recruited from a candidate REED family or KENDALL family would be a means of establishing with an exceptionally high degree of certainty whether the older or younger children of Robert BROWN were also related to that test subject.
* * * * *
It is unclear to me whether you realize that autosomal DNA testing should also show the DEGREE OF KINSHIP you share with your fellow BROWN researchers descended from Robert BROWN as well as the degree of kinship to others known to be related, such as Carol Brown HOSKINS. The Y-Chromosomal DNA testing does NOT show the precise degree of kinship in the same way.
You may be a 67-marker match to Carol, but this still doesn't tell you whether you are 3rd, 4th or 5th cousins. The autosomal DNA testing shows much more precisely the degree of kinship.