Month: July 2018

Edinburgh Castle

I have long been a sceptic concerning the use of DNA testing and its place in the practice of family history and genealogy. This was mostly down to ignorance of what was involved. Having been asked to make a presentation on the subject at the U3A group I help lead, I was forced into a deeper exploration of what DNA analysis involved and how it had helped, or hindered, others who have taken the plunge.

Basically a DNA test involves (in the case of Ancestry, who I decided to use as they have the largest number of testers) spitting into a small tube and sending it to Ancestry, who arrange the processing of the DNA and the  analysis of the results. After a few weeks the results appear on the Ancestry website, in the form of a summary of one’s “ethnicity” and a list of potential cousins, in order of closeness, as indicated by the amount of DNA material in common. Of course, only people who have tested with Ancestry are included, and many of those have little or no interest in family history and only tested to discover their ethnic origins. Some of the “cousins” will have their trees on the site and these can be checked for similarities with one’s own.

The ethnicity results can be interesting, but are always to be regarded as general guidelines, rather than 100% accurate. As they are based on the DNA results of others who have tested, they only represent a small proportion of humanity, and can sometimes be misleading. However they do have a bearing on my experience in genetic genealogy.

The actual result of the DNA analysis (known as raw data) is not intelligible to the layman, but can be uploaded to other sites that offer DNA services. In my case I have uploaded mine to My Heritage, Family Tree (FTdna) and finally to Gedmatch, a site that does not carry out tests but processes others’ data. It has been interesting to see that the closeness of relationships has been mirrored, so far, in those cases where the same individuals have also uploaded their data to multiple sites. This seems to boost the authenticity of the degree of relationship as indicated by the sites.

My main reason for testing was to confirm my family tree which I had discovered over many years research. Building a document-based tree is time consuming and can be extremely frustrating when so much is missing from past records. In addition, one is relying on the accuracy of written records that may, or may not be, reliable. However, it is also rewarding, although the niggle remains – “is all this accurate?”. By comparing with others trees, using DNA test results as a guideline for degrees of relationship, one can “prove”, as far as is possible, that the paper tree is a true reflection of one’s heritage. A note of caution must be raised the further back into the past one goes. Some of the “cousins” thrown up by the websites are quite remote – 5th to 8th cousins for instance, and the amount of DNA material that is shared could equally be the result of random chance. In this case, unless a paper-trail connection can be made as well, it is best to put them to one side until more evidence is found. On the other hand, where the amount of matching DNA material is small, but the paper-trail exists, one can be more confident that it confirms the match, as the possiblilty of any chance matching with a non-relative who appears by the record to be  a cousin, however remote, is unlikely.

When a true match is fully backed up by the researched tree, there is little doubt that that portion of the tree is totally correct, in the direct line. So far, I have had several of these matches, which have enabled me to definitely confirm my descent from the following:

Thomas Hall (c1776-1841) and his wife, Betty (c1779-1854, nee Pickford). I am descended from their daughter Elizabeth, and have found a matching cousin who descends from their son, Benjamin.

John Gibbs (1809-1880) and his wife, Amelia (1815-1900, nee Helps). I am descended from their son William, and a matching cousin descends from their son Robert Frederick.

William Noyes (1815-1894) and his wife, Eliza (1811-1891, nee Pritchard). My descent is from their daughter, Maria and my cousins’ (there are two of them) from Henry, their son.

Stephen Bumstead (1778-1841) and his wife Betsy (1782-?, nee Wase). I descend from their son, Stephen and my matching cousin from their son, William Wase Bumstead.

Louisa Bumstead (1842-1923, nee Peters). Louisa was my great grandmother through her son, George Albert Bumstead. My matching cousin is descended from Rosina Fear Peters (1864-1946) an illegitimate daughter of Louisa’s, born before her marriage.

Edward Flexney (1795-1853) and his wife Mary (1796-1878, nee Godfrey). I am descended from their son, John (b 1840) and my cousin from their son, Frederick.

So far, so good, but there are downsides as well. The closest match that has appeared so far – someone who is probably a second or third cousin – was adopted and does not know his immediate forebears. Another close match has a father who was adopted. These will be total stumbling blocks in finding our relationship, unless the individuals wish, and are able to, track down their parentage. There are also several matching cousins whose background takes them to areas where my forebears lived, but without any paper-researched link in their trees. Because the suggested degree of cousinship indicated by the various sites is vague, “4th to 8th cousins” for example, the chance of finding the link is poor. However, there are many hundreds of matches I have not followed up yet, so it’s very early days in my genetic quest.

By far the most exciting possibility for me in the DNA field is the chance of discovering the identity of my “missing” grandfather, which was unknown to anyone in my immediate family. Family stories relate that he was a “very respectible young man”, possibly a soldier, who died in April 1918. He was, most likely, in Bristol in September/October 1917. Carrying out my DNA test has resulted in  two linked clues; firstly, I have mentioned how the ethnicity element of the results is not conclusive, and often vague, yet the interesting figure from the Ancestry analysis is that my background is 34% Scots. Now I have taken most of lines back to the 18th century, and so far it has been wholly southern English with the exception of a ggg grandmother born in Dublin in 1804. The Ancestry estimate indicated I have 36% southern English inheritance, so it would show that I have a major Scottish ancestor not too far back in my lines. Confirming this, is the fact that among my matches are a large number of people with Scots descent, the names Campbell, Buchanan, McFarlane and MacDonald cropping up as the most common. I am convinced that this indicates that my missing grandfather is of Scottish lineage. I am at the stage now of contacting some of these matches to see if any likely candidates turn up. The main problem with this approach is that any common ancestor of both myself and any one matching cousin could be three to seven generations back, and as most family historians only trace back a direct line (and possibly the siblings of that line) so discovering an individual who would have been connected to us both, but possibly several generations down from the common ancestor is difficult to say the least.

My Ancestry ethnicity profile

I do have one strong candidate at present, suggested by one of my Scottish matches. He fits the bill in many ways, but his military record from the First World War is patchy and problematic and I can’t place him in Bristol at the right time at the moment. However it does give a possible line of research; when most paper trails have been exhausted at least the genetic angle of family history  provides new prospects of success as well as confirming much of what has been done so far is correct.