FOR A few days in January it was the hottest property on the Web. Demand was so high that the site had to be closed down for fear it would paralyse the telephone network. It still hasn鈥檛 reopened. So just what was on this site? A century-old British census, the first to go online. The Public Record Office expected 1.2 million hits a day. It got 7 million.
The 1901 census wasn鈥檛 the first genealogical website to be swamped. When the Mormon Church put its vast archives online in 1999 it too had to re-engineer its site to cope with unexpectedly high demand. No one, it seems, bargains for the enormous popularity of genealogy.
For some reason people have a deep-seated fascination for their family histories. Until recently, researching them meant spending hours thumbing through dusty old files in out-of-the-way record offices, trying to make sense of the paper trail left by long-dead relatives. Online services have made the task easier. But there鈥檚 an even more sophisticated way of tracing your family tree. Each of us has a permanent record of our ancestry written into our genes. And now you can access it鈥攆or a price.
Advertisement
The basic technique for tracing your family tree is to start with living relatives and work your way back. At various stages of their lives your ancestors will have had a brush with bureaucracy: a marriage, a census, a court appearance or an immigration hearing. All leave a mark on the documentary record.
But sooner or later the paper trail peters out. In England, for example, national registries of births, marriages and deaths only go back to 1837. Before that, the church recorded christenings, marriages and burials. But few parish records extend back beyond the middle of the 17th century, because of the disruption caused by the Civil War. And your research will probably run into a brick wall long before you get that far. Once you exhaust the written record you鈥檙e often left with loose ends and little idea of how to tie them together.
That鈥檚 where commercial genetic fingerprinting can help. There are two tests on the market already鈥攐ne based on the Y chromosome which traces male lineages, and an equivalent for the female line using mitochondrial DNA. The useful thing about Y chromosomes is that, like surnames, they pass down the male line almost completely unchanged. A son鈥檚 Y chromosome is 99.9 per cent the same as his dad鈥檚. There are mutations in the regions used for fingerprinting, but the rate in each region is only one change per 500 generations. This opens two windows on the past. First, men with a male ancestor in common will have identical or nearly identical Y chromosome fingerprints. Secondly, researchers can use the mutation rate to estimate when two males last shared a common male ancestor.
Mitochondrial DNA is useful for similar reasons. It passes unchanged down the maternal line, so comparing mitochondrial DNA fingerprints can tell you if two people share a common female ancestor. And mitochondrial DNA also accumulates random mutations over time, so you can also estimate when two people鈥檚 last common female ancestor lived.
Both techniques have had famous successes (see 鈥淐elebrity cases鈥). But in many ways they鈥檙e most meaningful when they鈥檙e brought to bear on everyday folk.
One of the companies selling DNA genealogy services is Oxford Ancestors, set up two years ago to capitalise on research carried out by Bryan Sykes at the University of Oxford鈥檚 Institute of Molecular Medicine. For 拢150 the company will put together a 10-marker fingerprint of your Y chromosome. On its own this doesn鈥檛 tell you much. But allied to an existing genealogy project, it can help solve mysteries.
Take the Pomeroy family鈥檚 puzzle. According to a cherished family tradition they were all descended from an 11th-century Norman aristocrat, Ralph de la Pommerai, who arrived in England with William the Conqueror. Dozens of Pomeroys, Pummerys, Pomerys and Pomroys became involved in an attempt to map the family tree back. Some traced their origins back to known descendants of the Norman family, but for most the documentary trail went cold and there was nowhere left to look.
Enter Oxford Ancestors. It analysed Y chromosomes from 51 males representing branches of the family that had yet to be linked by historical records. What they found came as a shock: there were at least seven separate lineages. In other words, for most of the family the tradition was a myth鈥攐nly six of the 51 were potentially related to the Norman family. But the tests had a positive side too, showing probable genetic relationships between males not linked by documentary evidence. Seven of the 51 men, for example, had the same Y chromosome fingerprint, so probably shared a common male relative. 鈥淭he results have given us big clues about where to look for documentary evidence to fill the gaps,鈥 says Chris Pomery, who runs the Pomeroy genealogy project.
It鈥檚 not the only type of puzzle Oxford Ancestors鈥 Y chromosome test can resolve. George Redmonds, a family historian based in Huddersfield, West Yorkshire, for example, used it to build a connection between three families that had no idea they were related.
One family took its name from a Lancashire place name, Ridehalgh. Redmonds found documentary evidence that in some branches of the family, the name gradually transmuted to Reddihalgh, then Redyoke and Redgeck. From this he proposed that the Ridehalghs were linked to two modern families in Yorkshire, the Ridgwicks and the Reddihoughs. The documentary evidence wasn鈥檛 conclusive, but a Y chromosome test clinched it: the families shared a common male ancestor.
Surnames are a constant source of genealogical fascination鈥攅specially their origins. And surprisingly, DNA can help here too. Surnames were first used in the early Middle Ages, when people started taking their names from their trade, their village or a local feature. For many years, Sykes says, people assumed that the Yorkshire surname Dyson was connected with dyeing. The county鈥檚 traditional connection with the wool trade added weight to the derivation.
To check the theory, Redmonds mapped the geographical distribution of the surname, working backwards through time using sources such as returns from the hearth tax of 1672. As he went back so the Dysons became more and more focused on the Huddersfield area.
Eventually Redmonds turned up records of a manorial court case of 1316 where a certain John Dyson and his mother Dionisia鈥攃ommonly known as Dye鈥攐f Lynthwaite near Huddersfield were charged with stealing cattle from a neighbour鈥檚 land. For Redmonds it was a breakthrough. That suggested the surname had nothing to do with dyeing鈥攖he original John Dyson was simply Dye鈥檚 son.
To help resolve the issue Redmonds asked Sykes to help. The Y chromosome test was ideal, because just like surnames the chromosome is passed down the male line. Sykes took DNA samples from dozens of living male Dysons. 鈥淲hen we did the Y chromosome test, 85 per cent of them had the same Y chromosome,鈥 says Sykes. This high proportion was remarkable, not the least because over the centuries adoption and illegitimacy introduce foreign Y chromosomes into any family line. Then Sykes calculated from the rate of mutation when the shared Y chromosome had a common ancestor. The finding was just as remarkable. 鈥淚t was very close to 1316,鈥 says Sykes. It seems that Dye鈥檚 son John was the original Dyson and most Dysons round the world are descended from him.
But surnames don鈥檛 always stay the same, or trace back easily to a single location. If that鈥檚 the case, then another company offering Y chromosome tests, Family Tree DNA of Houston, Texas, can help. Most of its customers are professional genealogists trying to find out which part of Europe an American family came from, says Bennett Greenspan, the company president. Immigrant families often adopted a new surname when they arrived in the US, making it hard to compile family histories. Many families don鈥檛 even know where to start looking. The surname Smith, for example, could be an anglicised version of the Germanic Schmidt.
Together with anthropologists from the University of Arizona in Tucson, Family Tree DNA has gathered a database of 6000 DNA samples from men all over the world and used it to identify region-specific markers on the Y chromosome. 鈥淚t鈥檚 pretty representative,鈥 says Greenspan. By comparing clients鈥 Y chromosome fingerprints with those on the database, the company claims it can advise families where to concentrate their search.
Commercial mitochondrial DNA testing isn鈥檛 as immediately useful as Y chromosome testing, largely because there鈥檚 no link between maternal lines and surnames. But there is one service the amateur genealogist can buy. From the rate of mutation of mitochondrial DNA Sykes has been able to trace back the maternal ancestry of almost all Europeans to just seven women who lived in different parts of the continent between 45,000 and 10,000 years ago.
For 拢150 Oxford Ancestors will test your mitochondrial DNA to determine which of these women you鈥檙e descended from. If, like 47 per cent of Europeans, your mitochondrial DNA comes from a woman Sykes has named Helena, then you will find out that your great-great-great-etcetera grandmother lived 20,000 years ago at the end of the last ice age in what is now the Dordogne in south-west France. Of course the results are of little practical use, because not many people can trace their family tree back more than a few hundred years, let alone 10,000. Yet people want the test. 鈥淚t was a real surprise,鈥 says Sykes, but he thinks he understands the appeal. 鈥淚t鈥檚 telling you that you do fit in somewhere, giving you context and a deep history.鈥
Certainly the desire to find out more about your ancestors touches a deep chord in people. 鈥淚t鈥檚 odd that people want this information,鈥 says Ashworth. 鈥淏ut you don鈥檛 just give 拢150 away. People tell us that they get emotional satisfaction from getting this piece of the puzzle.鈥
The search is now on for other genetic tests that will help people plug gaps in their family histories. Western Europe has seen a succession of invasions starting with the spread of the Roman Empire. Many of the invaders settled down and had children. Already Oxford Ancestors鈥 Y chromosome test can tell whether or not your forefathers were Norse Vikings, based on a comparison with the Y chromosomes of probable Viking descendants living in Scandinavia. And last year researchers showed that it was possible to detect Viking or Gaelic ancestry in the Western Isles and Orkney from variations in mitochondrial DNA (American Journal of Human Genetics, vol 68, p 723).
Sykes and his colleagues are currently building a database, the Oxford Genetic Atlas, which will extend the scope of the Viking test. Eventually, says Sykes, it should help people find out whether they were descended from Ancient Britons or Saxons. Similar Y chromosome tests provided by Family Tree DNA can determine if you鈥檙e descended from Native Americans.
Meanwhile, researchers at the molecular genealogy research group at Brigham Young University in Salt Lake City, Utah, are about a quarter of the way towards building a global DNA database with 100,000 samples. Everyone giving a sample has to be able to trace their family back for four generations, which means knowing who all 16 of their great-great grandparents were and where they were born. The researchers are looking for genetic variations that are associated with a specific geographic origin. 鈥淲e admit that this is a tremendous job, but we have been successful in several small-scale trials,鈥 says Scott Woodward, who heads the group.
The database鈥攚hich they plan to make publicly available鈥攚ill be particularly useful for immigrants who have lost all contact with the old country. When it is finished in about four years鈥 time, people will be able to send off a DNA sample and may find out that their ancestors came, for example, from a particular part of Russia, so narrowing down the search for their family history.
The use of DNA tests in genealogy is still in its infancy. But there is little doubt that it will become more important. 鈥淚 think genetics will become a standard tool, it is such a powerful technique,鈥 says Sykes. For there鈥檚 a story in your genes, and it鈥檚 just waiting to be told.


Celebrity cases
DNA genealogy made the headlines in 1998 when it seemingly confirmed a persistent rumour that American President Thomas Jefferson had fathered a child, Eston, by one of his slaves, Sally Hemings.
Jefferson didn鈥檛 leave a legitimate male heir, so researchers compared the Y chromosomes of five male descendants of his paternal uncle, Field Jefferson, with Hemings鈥檚 great-great-great grandson (see Diagram). The genetic fingerprints were the same, apart from one that had a minor mutation. It didn鈥檛 definitely prove that the president was the father, but if he wasn鈥檛 then one of his close male relatives was (Nature, vol 396, p 27).FIG-mg23343702.jpg
The mitochondrial test also has its celebrity case. When the supposed remains of the Russian royal family, the Romanovs, who had been executed by the Bolsheviks in 1918, were dug up at Yekaterinburg in 1991, forensic scientists took samples of mitochondrial DNA from the bodies. From this they were able to show that one of them was Nicholas II鈥檚 Tsarina, Alexandra. How did they know? Alexandra鈥檚 maternal grandmother was Queen Victoria, who was also a direct female ancestor of the Duke of Edinburgh. Mitochondrial DNA from one of the adult bodies in the grave matched the Duke鈥檚 exactly showing it was almost certainly Alexandra (see Diagram).FIG-mg23343701.jpg
Do try this at home
Curious as to what Oxford Ancestors could tell me about my own family history, I called to request a test. An envelope containing a DNA sampling kit arrived the next day.
I washed out my mouth and cut open the sterile envelope to remove a small brush. I then brushed the inside of each cheek ten times, avoiding my teeth so that I didn鈥檛 contaminate the sample with bacteria. With a few hundred cheek cells safely on the brush I popped it back in the envelope, sealed it up and posted it back.
While the lab worked on my DNA, I received the first results: the geographical distribution of people with the surname Hamer.
The result was a minor surprise. I already knew that at the beginning of the 19th century my ancestors lived in a small village in central Wales called Llanidloes. This was where I expected to find the greatest concentration of Hamers. There was indeed a cluster of Hamers in central Wales, but by far the biggest concentration was in Lancashire.
So does the family have several origins or just one? A Y chromosome test could answer the question鈥攂ut I鈥檇 need samples from a few dozen Hamers.
The mitochondrial DNA results also had an element of surprise. My maternal line goes back to a woman who lived 17,000 years ago in what is now Tuscany. Hers is one of the more uncommon lines, representing only 9 per cent of Europeans. Her descendants are now concentrated around the Mediterranean and in Ireland.
Then my personal trail goes cold鈥攗ntil Annie, my great-great-great grandmother, turns up in Kent during the Napoleonic Wars.