Looking for His glorious appearing! Titus 2:13

The discovery has prompted fears within the US Government that the knowledge will fall into the hands of terrorists wanting to use it as a bio-weapon of mass destruction.

Some scientists are questioning whether the research should ever have been undertaken in a university laboratory, instead of at a military facility.

The US Government is now taking advice on whether the information is too dangerous to be published.

To see the graphic: The last outbreak - A deadly virus even before the latest twist

"The fear is that if you create something this deadly and it goes into a global pandemic, the mortality and cost to the world could be massive," a senior scientific adviser to the US Government told The Independent, speaking on condition of anonymity.

"The worst-case scenario here is worse than anything you can imagine."

For the first time the researchers have been able to mutate the H5N1 strain of avian influenza so that it can be transmitted easily through the air in coughs and sneezes. Until now, it was thought that H5N1 bird flu could only be transmitted between humans via very close physical contact.

Dutch scientists carried out the controversial research to discover how easy it was to genetically mutate H5N1 into a highly infectious "airborne" strain of human flu. They believe that the knowledge gained will be vital for the development of new vaccines and drugs.

But critics say the scientists have endangered the world by creating a highly dangerous form of flu which could escape from the laboratory – as well as opening a Pandora's box for fanatical terrorists wishing to make a bio-weapon.

The H5N1 strain of avian influenza has killed hundreds of millions of birds since it first appeared in 1996, but has so far infected only about 600 people who came into direct contact with infected poultry.

What makes H5N1 so dangerous, though, is that it has killed about 60 per cent of those it has infected, making it one of the most lethal known forms of influenza in modern history – a deadliness moderated only by its inability (so far) to spread easily through airborne water droplets.

Scientists are in little doubt that the newly created strain of H5N1 – resulting from just five mutations in two key genes – has the potential to cause a devastating human pandemic that could kill tens of millions of people. The study was carried out on ferrets, which when infected with influenza are the best animal "model" of the human disease.

The details of the study are so sensitive that they are being scrutinised by the US Government's own National Science Advisory Board for Biosecurity, which is understood to have advised American officials that key parts of the scientific paper should be redacted to prevent terrorists from using the information to reverse-engineer their own lethal strain of flu virus.

In an unprecedented move, the Biosecurity board is believed to have told the US Government that there is a serious possibility of potentially dangerous information being misused if the full genetic sequence of the mutated H5N1 virus were to be published in open scientific literature.

A senior source close to the Biosecurity board, who wished to remain anonymous, told The Independent that the National Institutes of Health, which funded the work, is about to make a decision on how much of the scientific paper on the H5N1 super strain should be published, and how much held back.

"There are areas of science where information needs to be controlled," the scientist said. "The most extreme examples are, for instance, how to make a nuclear weapon or any weapon that is going to be used primarily to kill people. The life sciences really haven't encountered this situation before. It's really a new age."

The study was carried out by a Dutch team of scientists led by Ron Fouchier of the Erasmus Medical Centre in Rotterdam, where the mutated virus is stored under lock and key, but without armed guards, in a basement building.

Dr Fouchier, who declined to answer questions until a decision is made on publication, said in a statement released on the university's website that it only took a small number of mutations to change the avian flu virus into a form that could spread more easily between humans.

"We have discovered that this is indeed possible, and more easily than previously thought. In the laboratory, it was possible to change H5N1 into an aerosol-transmissible virus that can easily be rapidly spread through the air," Dr Fouchier said. "This process could also take place in a natural setting.

"We know which mutation to watch for in the case of an outbreak and we can then stop the outbreak before it is too late. Furthermore, the finding will help in the timely development of vaccinations and medication."

A second, independent team of researchers led by Yoshihiro Kawaoka of the universities of Wisconsin and Tokyo is understood to have carried out similar work with similar results, which has underlined how easy it is to create the super virus with a combination of deliberate mutations and random genetic changes brought about by passing avian flu manually from the nose of one ferret to another.

Some scientists have privately questioned whether such research should have been done in a university department that does not have the sophisticated anti-terrorist security of a military facility. They also point out that experimental viruses kept in seemingly secure laboratories have escaped in the past to cause human epidemics – such as a 1977 flu outbreak.

"There are people who say that the work should never have been done, or if it was done it should have been done in a setting where the information could be better controlled," said the source close to the biosecurity board.

"With influenza now it is possible to reverse engineer the virus. It's pretty common technology in many parts of the world. With the genomic sequence, you can reconstruct it. That's where the information is dangerous," he said.

"It's scary from a number of different angles. You want to have the vaccines and therapeutics in place, and you need to have a much information as you can about a particular virus, but you also worry about it from a biosecurity perspective."