Missing Links

DNA testing,

I just found out there's a stranger in my gene pool.

It was only one in a series of surprises in recent DNA testing undertaken by three of my distant cousins and me. We had each been tested for DNA patterns at locations called markers on our Y chromosome, which men inherit only from their fathers. Matching patterns can help prove kinship.

THE JOURNAL REPORT

 

See the full Encore report.

The testing would lead me, after a search of DNA databases, to a man identified only as "Rutherford." According to our markers, Mr. Rutherford and I could almost be genetic brothers -- but ones (obviously) with different surnames. This raised a number of questions: Where did he come from? How are we connected? And do I have too much time on my hands?

DNA testing has become the hottest tool in genealogy, allowing amateur sleuths like myself to graft and prune their family trees. The process is simple, involving little more than a swab of the inside of your cheek. Advances in lab technology, meanwhile, have brought the costs down to home-appliance levels. And for your efforts, you can learn, among other things, some of the ancient ethnic and geographic origins of your ancestry.

But beware: DNA can open doors you can't close.

Questions Upon Questions

My own interest in genealogy dates to my childhood in Texas and an early curiosity that never dulled with age. How and why, I wondered, had I come to exist in Texas, and who were my ancestors? My grandfather told me that his American forebears had settled in North Carolina before the American Revolution, and his father, born on a wagon trip to Iowa before the Civil War, had become an itinerant home builder in Texas and the Midwest. But that scanty information only prompted more questions.

Off and on for 50 years, a similarly obsessive uncle and I followed the leads in our family Bibles and letters, and eventually reconstructed a long and detailed, but still incomplete, family story. It came from a variety of documentary sources -- court records, wills, census reports, tax lists and Quaker meeting minutes -- and from recollections of our oldest living relatives. I ordered my first DNA test last fall after learning of my cousins' testing, just to confirm our kinship.

DNA testing for the layperson is only several years old. In 2000, Bryan Sykes, a noted geneticist at Oxford University, published one of the first research papers on DNA and a surname -- his own. As "a bit of fun," he explained, he tested a number of Sykes males in three English counties where the name had also been common 700 years earlier. Nearly half of them shared a haplotype -- or set of DNA patterns -- in Mr. Sykes's tests of a set of four markers on their Y chromosomes. And that made their common ancestry highly probable. That year, he founded Oxford Ancestors Ltd. to offer DNA testing to British and American consumers.

Also in 2000, two American entrepreneurs with keen personal interests in genealogy, Bennett Greenspan and James Levoy Sorenson, separately founded what have become the country's largest genealogical DNA testing laboratories. They are Mr. Greenspan's Family Tree DNA, in Houston, and Relative Genetics, a division of Sorenson Genomics LLC in Salt Lake City. Since 2000, these labs and others have tested or are testing more than 120,000 DNA samples.

With each company, the procedure is much the same. A person orders a kit via the Web or the telephone, does a cheek swab, and mails the kit back to the test lab. The cost ranges from less than $100 to several hundred dollars, depending on the laboratory and the number of DNA markers examined. One caveat: Results can take several months to arrive.

Some laboratories offer testing of the mixtures of DNA you inherit from both your father and your mother in the 22 pairs of autosomal, or non-sex-determining, chromosomes. Those combine at conception to code your physical characteristics. The testing, among other things, can give indications, though often fuzzy and questionable, of the ethnic or geographical origins of your total ancestry.

More definitively, genealogical labs also test for known single nucleotide polymorphisms, or SNPs. An SNP looks like a one-character typographical error in a long string of words. It's a genetic mistake that gets inherited and points to distant paternal- or maternal-line forebears among Stone Age clans in Africa, Asia, Europe or, more recently, in the Americas before 1492. SNP tests confirm the deep ancestry that genealogical haplotypes suggest but don't prove.

A customer usually has to order a separate SNP test. But Family Tree DNA automatically does a SNP test if 12 or more tested markers on the Y chromosome show a pattern that hasn't already been SNP tested, at no extra cost.

Online Connections

One consequence of DNA testing is the growth of Web-based surname projects, which are efforts to connect individuals with near or distant cousins in order to confirm, discredit or expand the links in sometimes error-ridden paper-trail genealogies.

Hosted by testing laboratories and administered by volunteers, the projects recruit participants and sometimes finance tests for the less affluent. (To assuage privacy concerns, most participants are promised anonymity if they wish.) More than 2,800 such projects, with participants ranging from a handful to hundreds, now exist.

Nancy Custer, a California high-school teacher who has a doctorate in biology, founded an early project after DNA testing helped her to solve a mystery that had bedeviled her since childhood. A great-great-grandfather, Harvey Kelley, for reasons undetermined, had changed his name from Dorsey about the time he moved from Tennessee to South Carolina after the Civil War.

Who had he been as Dorsey? Nobody knew. Census and Civil War service records pointed toward a Tennessean named Elisha H. (for Harvey?) Dorsey, whose wife, like the wife of Mr. Kelley, was named Louisa. A few years ago, Ms. Custer tracked down several male Dorsey descendants and persuaded them and a Kelley cousin to submit their DNA for testing. It matched. And Ms. Custer expanded her research into the family with a surname project for Dorseys.

"Even after 35 years as a biologist, it was an incredible experience to see those rows of matching numbers," Ms. Custer says. "That quiet little Y chromosome had stayed true to its secret so many years, and...technology had found a way to bring that secret to light. I was very excited."

Meet the Moffitts/Moffetts

My own DNA test, and those of my distant cousins, revealed two surprises: a new branch on our family tree and the mysterious Mr. Rutherford.

The four of us had met online through genealogy message-board postings and, eventually, through our DNA tests. Three of us -- myself, Kyle Moffitt (a high-school teacher who lives in Australia) and C. Michael Moffitt (an environmental consultant in Austin, Texas) -- had been able to trace our roots to the sons of one Robert Moffitt in Northern Ireland. His sons settled in North Carolina in 1760.

But the fourth player at our table -- John Moffett, a utility executive in Kentucky, who spells his name with an "e" -- originally had no strong reason to suspect he was kin to us. His known American ancestry originated in what is now Kentucky but was Virginia before 1780. There had been no persuasive evidence to connect any of several Moffett families in colonial Virginia with our North Carolina Moffitts.

Our DNA tests provided that evidence.

The tests first showed that all of us shared a straight paternal-line ancestor, perhaps with 100 million or more males in Western Europe and the Americas. The patriarch seems to have fathered a Stone Age clan in northern Spain that survived, grew and drifted northward as the glaciers of the Ice Age began to melt roughly 15,000 years ago. Researchers consider that we, all his descendants -- including many of you who are reading this article -- comprise a population labeled "R1b," which saturates parts of Ireland, Scotland, and the Spanish Basque country but is found all over Europe.

But our tests also showed that the four of us shared an extremely rare mutation along the Y chromosome, a DNA pattern that appears in only a few hundredths of 1% of the R1b population. That match -- which Mr. Greenspan at Family Tree characterized as nothing less than a silver bullet -- was a virtual guarantee of close kinship. The Moffitts of North Carolina and one group of Moffetts of colonial Virginia were family.

Man of Mystery

And then came Mr. Rutherford. Searching publicly accessible DNA databases, I found a man identified only as a Rutherford who matched on 27 of the 28 markers from our DNA tests -- including our rarest, comparable with those of my most extensively tested cousin. The close match indicated a good chance of recent common ancestry without even taking the rarity into account.

Y-chromosome markers, though, usually follow surnames. Phonetic spelling could turn Moffat into Moffett or Moffitt but not into Rutherford. Was this a purely random match -- like that of the same winning lottery number in drawings in two states -- owing to identical but unrelated mutations? Did it result from common biological ancestry before either family began using surnames?

Or did it represent what genealogists call a nonpaternity event, such as an informal adoption, illegitimacy, or deliberate name change, where someone fathered by a Moffitt or Moffett became known as Rutherford or vice versa? If so, when and where could a Moffitt-Rutherford connection have taken place?

The nature of nonpaternity events usually makes them impossible to prove. It's true that Mr. Rutherford's DNA doesn't come close to matching that of other Rutherfords, but only a few have been tested. I don't even know who Mr. Rutherford is. I found his colonial pedigree in his DNA database and wrote a letter of inquiry to the person listed as having provided the pedigree. There's been no response.

Common Ground

That didn't end the search, though. Mr. Rutherford's pedigree identified his earliest known Rutherford forebear as a Henry Rutherford, born in 1801 in Virginia. Expecting nothing, I entered "Moffett," "Rutherford" and "Virginia" in Google.com's search engine. I inspected 200 irrelevant links; No. 201 got me a scholarly account of the surveying and development of a settlement in the part of colonial Virginia that became West Virginia during the Civil War.

Land in the "Patterson Creek Manor" was owned by one of Virginia's main proprietors, Lord Fairfax. In 1748, he commissioned a surveyor, accompanied by Lord Fairfax's young nephew and the nephew's best friend, 16-year-old George Washington, to map it. The survey was unusually erroneous, and in 1762 a team that included a John Moffett, who might or might not have been one of ours, did a resurvey.

The new survey produced a plat map showing the ownership of "lots" within it. Some 500 acres retained by Lord Fairfax occupied a central lot. It abutted the holding of a Solomon Rutherford, and a few tracts away lay one owned by a John Moffitt, probably the surveyor himself. Coincidence piled on coincidence or not, there the mystery stands.