Ancestors reunited: 10 numbers that can trace your family tree

The unwelcome sensation of "hitting the wall" is an occupational hazard for genealogists. After weeks, months or even years spent searching public records for clues to the identity of distant ancestors, it is the moment the written trail goes cold. The roots of the family tree have gone as far as they will go.

The unwelcome sensation of "hitting the wall" is an occupational hazard for genealogists. After weeks, months or even years spent searching public records for clues to the identity of distant ancestors, it is the moment the written trail goes cold. The roots of the family tree have gone as far as they will go.

Thousands of people who have taken up the increasingly popular pursuit of genealogy have hit the wall. Family trees tend to die out around the 1790s when the industrial revolution and growing population movements meant records were either lost or not kept.

But genealogy is undergoing a revolution of its own thanks to DNA technology. It will soon be possible to overcome research problems that would normally take years or decades to settle simply by typing a DNA code into an internet site. The technology, which is being developed by a company spun off from Oxford University, will allow genealogists to get past the point where the written trail dies and will give millions of amateur historians a head start never available before.

George Redmonds, a historian specialising in surnames, caught the mood at a get- together of more than 100 amateur genealogists at the Public Record Office (PRO). He had thought he had taken his studies as far as possible, he said. Now, because of the DNA breakthrough, he would have to go back to the drawing board.

"It is the biggest step forward for generations. I thought I had finished what I had to say about surnames but now I find I have to go down a new corridor," Mr Redmonds said. "We are at a key time for the study of genes, genealogy and surnames. We are looking at a number of revolutions."

The centre of the revolution is an anonymous business park in Kidlington, outside Oxford, where a staff of four are developing a new service to go live on the internet before the end of the year.

It is based on the Y chromosome, which is held only by men, who pass it down unchanged from father to son, like a surname, and which has already been used by Brian Sykes, professor of human genetics at Oxford University, to establish the origins of his own surname. The company the professor founded, Oxford Ancestors, has collected thousands of samples of Y chromosomes to build the database that will be made available on the internet.

So far the company has been providing Y chromosome analysis for genealogists unable to resolve gaps in their ancestral history through the written records. In most cases, the process was used by two people who suspected they were descended from the same person but could not prove it through the records. If their Y chromosomes were tested and were identical, their question was answered. Interest has been mushrooming and Oxford Ancestors' customers have grown from about a thousand to 4,000 over the last year. Once the internet service is running, its chief executive, David Ashworth, expects the numbers to double every year.

The service can be used by anyone who knows their 10-digit code for their Y chromosome, which is easily found by sending the company a swab taken from the inside of your cheek and £150. When the code is typed into the website, it will bring up the matches for a person's surname and DNA and the country where they live. Once some confidentiality issues are settled, it may narrow the location down to a county.

The snag appears to be that it only works with others registered on the site; there may be those out there with your DNA who are not registered and cannot be found. However, if you find someone by conventional means who you think might share your DNA, the match can be confirmed through the website.

There is immense demand from Americans, Canadians and Australians who believe they are descended from a particular Briton but have been unable to make the transatlantic link. If a Bennett from Massachusetts types in their DNA and finds it is an identical match to a Bennett from Lincolnshire, their search is over.

It can also help those who have done no previous research beyond providing a swab. They may instantaneously find they have blood relatives across the Atlantic.

Mr Ashworth warns against assuming DNA is a replacement for painstaking research, and says that even when a match is established, relevant documentation may be needed to prove it beyond all doubt. "DNA is a tool to allow you to do something. It's not an end in itself, it is a means to an end," he said. "However, genealogists talk about hitting the wall, when you have the circumstantial evidence to say that you have the same ancestor but you can't prove it.

"Now they can have the genetic analysis done and they might find it supports their theory. It helps genealogists to find the links that they are unable to prove using written records, so when they hit the wall we might be able to help them get over it."

The service can also destroy illusions genealogists might have about the nobility of their ancestry. Mr Ashworth added: "It could be that two people have traced their history to the same person, we do the analysis and find their DNA is completely different. That means one of the paper records is incorrect. Some people say it is a disaster and they have wasted 20 years' work; others are happy because they had got stuck and now they know to go off in a different direction and they have another 20 years' work ahead.''

Sometimes the truth may hurt as family historians can find that the piles of documents they have collected over years to prove their lineage conceal an illicit encounter.

As Mr Redmonds told scores of researchers at the PRO anxious for a way round individual historical dead ends: "The potential that DNA has through the Y chromosome is enormous for genealogy. It will not replace it but it establishes a context within which you tie up all the leads and establish your own family history.

"The negative side is that you might not like the results of the test. They might show that a female member of the family was playing away from home and covered it up."

How to find out your DNA code

By Charles Arthur

Every man (but no woman) has a Y chromosome, whose unique characteristics can be expressed as a 10-number code. Only close relatives have a Y chromosome with the same code. The code can be identified by taking a swab from the inside of the cheek that will pick up DNA-bearing cells. Their DNA can be extracted and put in a test tube after a few hours of work in the laboratory. To that is added a cocktail of 10 DNA "primers", which each look along the double helix (the shape of our DNA) for a specific sequence of the four "base pairs" that stand like steps on the staircase of the helices. The primers cut out those sections, known as "Short Tandem Repeats" (STRs); the cocktail is then mixed to produce thousands of those 10 sections, each a perfect copy of the original. Standard lab work then measures the length in base pairs of each particular section. The length of the STRs varies subtly between men of different descent. At each point it can range from zero to an unlimited number – though in practice the difference tends to range from about eight to a maximum of around 30. A typical reading might be 12-14-10-23-13-13-11-16-12-12. For someone unrelated to the person whose DNA provided that reading, the figure might be 12-14-11-24... and so on. The difference may be tiny but it's enough to prove whether two people are related or not.