Who was Huldah's father?

Though I have previously discussed the question of who Huldah's father might have been, the topic needs to be revisited now that we have clearly tied her to the line of Andrew Harwood of Dartmouth, Devonshire, England.

The Harwood genealogy written by Watson H. Harwood, M.D. attempted, in several volumes, to account for all colonial Harwood lines in America. His first volume, published in 1911, detailed his own line, that of Andrew Harwood of Dartmouth. Andrew, born about 1627, was, according to this genealogy, the only one found to have come to America.   Andrew married Elizabeth Bowden in 1648 in Dartmouth and came to America; Watson Harwood notes that he likely left his wife and children behind. He died in America in 1658 and English records show that he had three minor children, two girls and a boy named James. 

As a young man, James came to America and was a soldier in King Phillip's War. He married Lydia Barrett of Chelmsford, Middlesex, MA, daughter of John and Sarah Barrett, and initially lived in Chelmsford. James had four male children, all born in Chelmsford:

> Andrew, b. 2 Sep 1692. Watson Harwood notes that he likely died early.

> Twins James and John, b. 30 Sep 1695. John died in infancy.

> John, b. 27 May 1703

Watson Harwood found no records of Andrew, b.1692, nor did I when I used FamilySearch "Full Text" search to scour the early records from 1690 to 1780. If Watson Harwood did not miss anything, James, b.1695 and John, b.1703, are the only candidates to be Huldah's father. 

Huldah, b. Abt. 1633 (per her tombstone), did not have a birth record. If she had been a child of John Harwood, b. 1703, all indications are that her birth would have been recorded along with John's other children in Hardwick, Worcester, MA, born in the same timeframe as Huldah. 

James, on the other hand, had two sons with no birth record, but known to be born between 1728 and 1734, when they lived in the vicinity of Groton, Middlesex, MA; this leaves room for at least one other child to be born in that timeframe without a birth record.

James had moved to Littleton, Middlesex, MA with his mother and father, where his father died in 1719. In 1727, he sold land in Littleton jointly with his brother John. I found that, in 1729, James and family were warned away from Chelmsford, after apparently trying to return there. 

"To either of the Constables of Chelmsford in the County, Greeting

Whereas information has been given unto the Selectmen of the town above that James and Lydia Harwood his wife and three of there Children -- Andrew, Mary and Unice have about three weeks ago came from the Town of Littleton and resided in teh Town of Chelmsford These are therefore to Command you forthwith to give warning to the above to James Harwood and Lydia his wife and Andrew Mary & Eunice there children that they  forthwith depart & Leave the Town of Chelmsford for which this shall be your sufficient Warrant. Dated at Chelmsford, the [...] day of February in the thirteenth year of his Maj's Reign Anno Domine 1729. By order of the Selectmen of the town, Benj. Adams Town Clerk" (Source: Middlesex Co. Court Records, 1652-1798, FamilySearch.org, image 887)

Per the Watson Harwood genealogy, the family was living on a farm near Groton, MA in 1731 and I found a court record indicating that James lived in Groton in 1736. The genealogy states that a 1737, document listed their residence as Concord and that they moved to Dunstable, NH sometime after 1740 (in an area now part of Nashua, NH). 

I've found court records involving disputes over debts that add to my impression that the family was not well settled and was likely struggling during their time in Groton; this may have made the recording of their children's births less likely. 

What if there was a brother of Andrew Harwood who came to America, whose descendant fathered Huldah? That scenario further increases the genetic distance and reduces the odds of finding triangulated segments. All three triangulated segments we've found that have been passed down through Harwood descendants have come through the family of James and Lydia (MNU), making James the most likely father of Huldah. Despite a shorter genetic distance and plenty of descendants, we've seen no indication of DNA segments shared with descendants of James' brother John, b. 1703.

Triangulated segments tip the scale to the Andrew Harwood line

A triangulated segment of DNA refers to an overlapping segment of DNA that is shared by three or more individuals who are all related to a common ancestor. There are three triangulated DNA segments that we have been able to identify by chromosome number and segment start/stop position that triangulate with descendants of the Andrew Harwood line and descendants of more than one son of John and Huldah Jacobs (isolating the source to John or Huldah). These DNA segments are the Chr#14 segment I've previously written about, plus segments from Chr#5 and Chr#20. This link shows two tables:

1) Top table: the matches that we have to each of these triangulated segments (color coded) and the Harwood lines they descend from.  

2) Bottom table: the Jacobs matches associated with each triangulated segment. 

No triangulated DNA segments have been found with descendants of the Concord or Salem lines. Though the Chr#14 line has Concord-line Harwoods in the tree, there is no evidence that they sourced the segment. In the cases above, multiple lines carrying the segment converging on a common Harwood ancestor descended from James Harwood, b.1695, whose wife was Lydia (MNU - "maiden name unknown"). 

There is a fourth segment listed for which no chromosome or segment data has yet become available. Triangulation is suggested by ProTools data showing the same set of matches shared at about the same cM value (~ 20cM) with mockvet, a descendant of Annis Harwood, daughter of Archibald Harwood (son of James and Lydia (MNU). We hope to gain the cooperation of one of the matches so that we can identify the segment and prove that it triangulates. It would also bolster the case for this segment if another Harwood match were to be found that traced to Annis Harwood via a different child.

This data is the centerpiece of the case for Huldah Harwood being a descendant of the Andrew Harwood line. 

Our Hoover line is now known

 After meeting with my Hoover Y-DNA match and his daughter, who is a professional genealogist, I can now say with certainty that our Jacobs line is connected to that of Johann Michael Huber, b.1737 in Lancaster Co., PA, d. 1816, Dauphin Co., PA. Their research had also led to Johann Michael Huber, though she had some residual uncertainty; that uncertainty was erased by the Y-DNA data and the autosomal DNA data that I presented to her. 

The Y-DNA data shows a four-marker difference on a 111-marker comparison of her father's data to our Jacobs line; this result puts the median time to most-recent-common ancestor (MRCA) at 1706, right in the ballpark of the expected birth year of our John Jacobs (b.1731/32). This would suggest that, most likely, our John was a brother or first-cousin to Johann Michael Huber. 

PA Mennonite records show the father of Johann Michael Huber to be Hans Jerg Huber; Hans Jerg is buried in the same Spring Hill "Huber cemetery" that Johann Michael Huber is buried in. Researchers have Hans Jerg Huber, also recorded as Hans George Huber, as born in 1691. Jerg is equivalent to Georg/George in German, but is usually written Jorg. Those researchers also have his wife in America as Anna  Maria Hoos, who he married in Blankenloch, Baden-Wurttemberg, Germany (just north of Karlsruhe) 11 Aug 1733 before immigrating; she is recorded as the third wife of Hans Jerg Huber. There seems to be little controversy over this research, however I plan to do my own validation of it. I have found the website of a Kris Hocker , who has done in depth research on our Huber line, and am attempting to get in contact with him.

It is very important that we understand, if possible, whether Hans Jerg Huber had any brothers who immigrated to America before 1743. If he did not, then our John was likely a brother of Johann Michael Huber, born to the second wife of Hans Jerg.

There is research claiming a pedigree for Hans Jerg that continues in Germany, though I don't want to get into that at this time; I want to validate what is available on Hans Jerg first. Most likely his pedigree eventually goes back to Switzerland, known to be the origin of most Huber's. 

Y-DNA alone would have eventually led to this conclusion. The autosomal DNA alone would have certainly strongly suggested it, but there would have been some significant residual uncertainty (paternal vs. maternal line). The combination of Y-DNA and autosomal DNA, along with the stories from the Chester Historical Society's "History of Chester, Vt" and the History of Rutland County, Vt., leave no doubt that our Jacobs line originally came from Germany as Huber's.  

All roads lead to Chelmsford, MA

Despite the evidence discussed in the last blog post connecting our Bowker family to the Concord Harwood line, the evidence pointing to the Andrew Harwood line is actually more significant at this point in time. A major source of confusion is that both lines went through Middlesex Co., MA and had a presence in Chelmsford and its vicinity. Both lines had a good chance of connecting to the same family and picking up common segments. In fact, the Chr#14 segment is evidence of that, being present in known descendants of the Andrew Harwood line and showing the potential of being in the Concord line by them being represented (via Nathaniel Harwood, who married Mary Barron) in the Josiah Proctor tree; but did it come from the Nathaniel Harwood's or Mary Barron's line (actually the data is more suggestive of Mary Barron's line). That there is only the potential of the Chr#14 segment being transferred via a Harwood line, with no proof yet of it actually being transferred by a Harwood, is an important fact in favor of Huldah being part of the Andrew Harwood line. We have identified several Andrew Harwood descendants, via James Harwood, Jr. (who married Lydia MNU... Maiden-Name-Unknown). James Harwood Jr. was born in 1695, such that one of his sons could have been the father of Huldah. Not only that, between 1728 and 1734, his children's birth dates and birthplaces have no written records supporting them; this suggests that Huldah, b.1733, could be explained as a child of James Harwood, Jr. (as I've discussed before). 

There is circumstantial evidence in the fact that Archibald Harwood, a descendant of James, Jr. lived in Windsor Co., Vt near  Chester, where John and Huldah lived. Daniel Jacobs had children named James and Lydia, while John, Jr. named a daughter Lydia. One of John Jr.'s children was named Cyrus and this name has turned up too in the lines of Daniel and possibly Samuel (if he is in fact, as I theorize, the father of the Bolton, Warren Co., NY, Jacobs family). The name Cyrus (not a common one) is also found in the Archibald Harwood line. 

Also, new segments of interest have been discovered since the last post. One, a Chr#5 segment matched to CJK,ND,DAJ and DFJ of the John Jacobs, Jr. line and also to multiple descendants of the Daniel Jacobs line, shows multiple shared matches with descendants of Archibald Harwood. Along with the Archibald Harwood matches, there are sets of matches whose pedigrees converge of Blanchards and Chamberlains, both families that have Chelmsford, MA connections. We are still researching these lines at this time, trying to find a common ancestor between the different sets of matches having this Chr#5 segment. 

We've seen DNA matches suggesting that Frank Gurly Jacobs, b. 1842 in Grafton, Windham, Vt, could be a part of our Jacobs family and we recently proved it with paper-trail evidence. Frank Gurly Jacobs was the son of a Joseph, b. abt. 1806 in Dorset, Bennington, Co., Vt, who was the son of Daniel and Abigail (Lawrence) Jacobs of Dorset. There is a Vermont Vital Statistics record for Joseph that states he was the son of John and Abigal Jacobs of Dorset, however this was clearly a mistake, as there was no John in Dorset married to an Abigail... only Daniel. Of the multiple DNA matches we have to Frank Gurly Jacobs, we were fortunate to have one (CTV) that was willing to share her matches with us and also uploaded her data to Gedmatch. With this access, we found more descendants of Frank Gurly Jacobs as well as other Jacobs cousins in her matches; we also obtained the matching chromosome details from GedMatch. Cousin LT has two segments matching to CTV and both are on Chr#10. Again, we are still in the process of researching pedigrees, but one of them clearly leads to the Barrett family of Chelmsford. This is significant because James Harwood, Jr.'s mother was Lydia Barrett of Chelmsford, daughter of John and Sarah (MNU). There are also some pedigrees converging on Butterfields, yet another Chelmsford family. This is all very significant, but with both Concord Harwood and Andrew Harwood lines going through Chelmsford, how do we know which family to attribute them too? At his point, I can't answer that question. The evidence continues to accumulate and the hope is that an answer with high credence eventually emerges from it. 

More J.Michael Huber connections

The discovery of the J.Michael Huber connection on Chr#13 prompted a search for more connections; that search was successful. Of course, when you do a search on Ancestry you can come up with just about anyone you want in the list of matches; just because J. Michael Hoover is in the pedigree of a match doesn't mean he is necessarily the one you connect with. I use two criteria to establish the likelihood that the match is valid: 1) Are there shared matches that also lead to the same person? and 2) Are any of your cousins from different lines also a match? 

For item 1), the first thing to do is to use the search utility to look for key surnames and placenames in the trees of the shared matches. In some of the J.Michael Huber matches, this quickly identified Hoovers/Hubers in Dauphin Co., PA. Also, if the search doesn't turn up anything, ProTools is extremely helpful in identifying those who are closely related, facilitating the search of the the pedigrees for  common ancestors.

For item 2), having a large number of cousins identified in the Ancestry database is essential. SA/SO/LT/BD cousin matches were all in my John Jacobs, Jr. line, leaving open the possibility that his wife's line sourced the segment. Though having BD in the list meant the probability of the segment coming from John Hover was 25% (great for someone born in 1731!), it was nice in some cases to see that Daniel Jacobs descendants also shared the match (as they do on the Chr#13 segment).

Having access to view the matches of key individuals is also helpful. Ancestry gives you information for whether your matches are paternal or maternal, making it easy to eliminated invalid matches from the irrelevant side; it does not provide a means of eliminating invalid paternal matches. For Chr#13, I have access to view ALL of the matches (even those < 21cM) for each of my cousins having the match. If a shared match is not in the match list of all of the cousins, it does not share the same segment. 

Of course it is also very helpful if one of the matches you are investigating is willing to go to her DNA settings and give you permission to view the all of her matches. Check the profile of the match and some of them include in their profile picture "Willing to help". On Chr#13, one of the matches did this for me and it allowed me to confirm that J.Michael Huber descendants were among the < 21cM matches as well as the ones that Ancestry allowed me to see > 20cM. 

So, how about those new connections to J. Michael Huber? In the figure below, you see most of the ones I've identified. Jacobs cousins are in the left column and column headers across the top identify the J.Michael Huber descendant and how they connect. Colored box outlines identify groupings of matches have at least some common shared matches (indicative of possibly sharing the same segment). 

Green shaded cells indicate a match and, other than Chr#13 (red box), they show the cM value. All with colored boxes have indications of validity; others are more uncertain. Those corresponding to CTV are especially interesting because CTV is from the Daniel Jacobs line and was nice enough to both allowing viewing of her matches and upload her data to GedMatch (with huge benefits also to the search for Huldah's parentage - yet to be covered in this blog). Though we do not know the Chr# and segment start/stop yet, the large cM values and commonality in the shared matches is intriguing; also, DH and SW are in common with the Chr#13 segment.  

The yellow boxed cells relate to a Chr#16 segment with high confidence based upon knowledge of the share segments of GM, SW and SO. This is a small segment and has only two with connections to J.Michael Hoover, but among the shared matches on MyHeritage, two had deep trees. One tree included a Samuel Good Huber, b. 1725 in Lancaster Co., PA as the most distant known ancestor. The other pedigree was totally German and included Hubers from Kappelrodeck, Ortenaukreis, Baden-Wurttemburg, Germany. with the most distant known ancestor Nicolaus Huber, b. 1690. Being a small cM segment, it would not be surprising to see it come from a line other than J. Michael Huber's; it likely means that the two lines connect in the more distant past. What makes J. Michael Huber intriguing is that some of the cM values are relatively large, suggesting a chain of sub-segments and a more recent connection to our family. This is likely true also of the small segment on Chr#10 that we have seen connecting to Solomon Hoover of York Co., PA and Lincoln Co., NC. 

Finally, some really exciting news:  we have a new Y-DNA match that is really close and it is a Hoover. The data is only 37 markers but has only a 2-step different from my Jacobs cousin, KJ; this is the closest match we've seen so far. Even more exciting is that this match seems willing to expand his test to 111 markers (with me gladly paying for it!). In preparation, we are getting KJ's 67 marker test extended to 111 markers. Because these other markers tend to mutate more slowly, the results could result in refining the distance to the most-recent-common-ancestor (MRCA) and showing a closer match to that family. In what little conversation I've had with the representative of the person involved, it sounds like there is some chance he is connected to J. Michael Hoover. 

New Autosomal DNA evidence connecting another Hoover/Huber line

 We have now connected our Jacobs line to another Hoover line via an autosomal DNA match, this one on Chr#13. This match, which I'll refer to as "SM" is on Ancestry so no details of SM's chromosome data was available. However, because we have the data on the Jacobs cousins who match (SA, SO, LT and BD), we know it is this particular Chr#13 segment because: 1) SA, SO and LT only share one segment with BD and it is this Chr#13 segment, and 2) There is a set of several shared matches with SM that are all in common SA, SO, LT, and BD - indication triangulation. BD comes and SA/SO/LT come from different lines from John Jacobs, Jr., such that triangulating segments have a 25% chance of coming from our John Jacob Hoover. The segments matching SM range from 13cM to 18cM, so are clearly above "the noise" and signfiicant. 

Among these shared matches is another Jacobs cousin (a Jacobs-descendant relative of BD), but the rest ALL have paths to Michael J. Hoover/Huber, b. 22 Sep 1737, Lancaster, Lancaster, PA, d. 26 Jun 1816 in Dauphin Co., PA. The new Ancestry "ProTools" feature was very helpful in facilitating this discovery. In all, there are three very different paths to Michael J. Hoover among these shared matches. Other researchers have found that Michael J. Hoover's father was Hans Jerg Huber, b. 1691, Switzerland, d. 1747, Dauphin Co., PA. He is buried in the same cemetary as Michael J. Hoover (Spring Creek Cemetary in Dauphin Co.). 

We have two Hoover/Huber lines now that we can attempt to trace back to Europe. If we can find a proven descendant of Michael J. Hoover we can attempt to get a Y-DNA test that will give us an idea of how closely he is related to our line and to the Solomon Hoover line. 

More Bowker-Harwood connections

In the wake of finding that Elias Bowker, brother of Daniel Bowker (father of Hannah, who married John Jacobs Jr.) married Sarah Harwood, daughter of Nathaniel Harwood, son of Peter and Mary (Fox) Harwood, two more Bowker-Harwood connections have been found. 

A match on Ancestry.com to someone I'll refer to as "JB", shows her connection to the same Nathaniel Harwood via his son James. James Harwood had a daughter Hannah, who married Asa Bowker, a grandson of our Edmund Bowker (father of Daniel). This match was common to our Jacobs cousin SA and my brother, DFJ. I was able to identify the DNA segment that SA and DFJ share by a process of elimination, narrowing it down to a segment on Chr#6. Triangulating that segment with them on GedMatch identified other matches to the same segment, among which was a match I'll refer to as "CB", who was on Ancestry.com. CB also had ties to the same Nathaniel Harwood via his daughter Hannah, b.1732, who married a Micah Whitney. Micah and Hannah (Harwood) Whitney had a son Nathaniel Whitney, who married Eunice Bowker, b. 1768, who happens to be the sister of our Hannah Bowker, who married John Jacobs, Jr.  One of the shared matches involved here is an LW, who has ties to Bowker's, but not Harwood's; because of this and the involvement of Bowker's in the other matches, we cannot say for sure at this time which family the Chr#6 gene segment came from. In any case, this finding demonstrates a close relationship between not only the Concord Harwood line and our Bowker and Jacobs families, but to a specific family, that of Nathaniel Harwood, son of Peter and Mary (Fox) Harwood.

Nathaniel Harwood is not likely to be the father of our Huldah Harwood, who was born in 1733. Not only was that only a year after the birth of his daughter Hannah, it is wedged between that birth and that of his daughter Sarah's in 1735 and both of those births were recorded (as were the births of all of his children). If Huldah was born to Nathaniel Harwood, why wouldn't her birth also be recorded? All of Nathaniel's children born 1745 and before were born in Lunenburg, Worcester Co., MA. 

In the post "Possible Parents of Huldah", the most plausible son of Peter and Mary (Fox) Harwood to be the father of Huldah was Joseph, who married Lidia Brooks in 1732 and died in November of that year, seven months after their marriage. It seems likely that Lidia died in childbirth and that, if she did, the father would not have kept the child (Joseph married again, but not until at least 1736). In this scenario, perhaps Nathaniel Harwood and his wife Hannah adopted Huldah.  Of course another major question is: Could a premature baby, born at seven months, have survived in the year 1732?