The success of biomedical research is founded largely on the sharing of discoveries amongst members of the scientific community through publication in journals. The free dissemination of information results in scientific advancement and innovation to an extent and at a pace that could never occur if research groups worked independently and in secret.

Even research carried out in the private sector, where information is guarded and products proprietary, is heavily dependent on publicly funded research for success. A scientific manuscript is written in sufficient detail to permit not only the evaluation of the data, but also to allow others to repeat the experiments in order to confirm or refute the published work.

Yet two journals are considering an unprecedented recommendation from the National Science Board on Biosecurity to withhold details of two related manuscripts precisely so that they cannot be repeated. The concern is that information could be used to commit bioterrorism. However, we should consider that attenuating the free exchange of information does not prevent terrorism, but rather is a consequence of it. Terrorism is successful when it elicits fear. Is limiting the creative activities of free people the fearful response that the terrorist seeks?

At issue are two manuscripts that have been accepted for publication that describe how the virus responsible for the bird flu can be genetically altered in the lab in such a way that may make it spread much more easily. The bird flu virus, called A(H5N1), is usually contracted directly from birds. The bird flu is serious and often lethal, but transmission is limited to those in close contact with birds.

By contrast, the genetically modified virus can spread directly between ferrets, the experimental model animal used in the studies, making it much more likely that it can also be transmitted between people directly by coughing or sneezing. This highly transmissible form would spread much faster, possibly resulting in a pandemic. The fear is that terrorists could use the information to reconstruct a deadly, highly transmissible form of the bird flu virus.

Why was this research carried out in the first place, and what are the possible consequences of publishing the protocol in full? Equally important, what may be the consequences of withholding parts of it?

Many types of viruses mutate rapidly, which can alter their virulence properties. This is why, for example, a vaccine that gave us immunity to last year's common flu virus may not protect us this year. The virus has changed sufficiently such that our immune system does not recognize the new variety and cannot protect us from it.

The bird flu is caused by a different viral strain, but the idea is the same. Researchers want to be able to predict in advance the specific genetic mutations that will make the bird flu virus more virulent so that they can be prepared with drugs and vaccines should it arise naturally. The surest way to predict what changes will be dangerous is to create and test them in the lab. Theresearchers found that it was surprisingly easy to do this.

This is not the first time bioterrorism concerns have been raised by working with a highly contagious and lethal microbe. The virus that caused the Spanish flu pandemic of 1918, resulting in about 50 million deaths, was reconstructed in the laboratory in 2005. Before that time, the viral strain no longer existed anywhere, and now is kept at the Centers for Disease Control in Atlanta.

The genetic material of the Spanish flu virus was recovered from the lung of an Alaskan influenza victim buried in permafrost in 1918, and the information used to rebuild the virus. The procedures, as well as the exact genetic sequence of the deadly virus, were published in full in the magazines Science and Nature. The goals of that study are similar to the bird flu research in question now, which is to prepare us for viral strains occurring in nature capable of causing a pandemic.

Under consideration in the bird flu studies is a plan to make the work available to scientists in the field who would likely continue to work on it or related issues. This plan, however, has potential problems. In a New York Times interview, Dr. Ron Fouchier, the lead author of one of the studies, estimated that about 1,000 people would have access to the two uncensored reports. The most obvious problem with this plan is that 1,000 people cannot easily keep a secret, which Fouchier acknowledges. In addition, the follow-up studies would likely be censored as well, resulting in a body of scientific literature in which access is limited. For the authors to come up with a list of scientists to share the information assumes that they know who will best move the field forward. But great ideas sometimes come from unexpected sources. Scientists within a given field often have similar training and background, and therefore they think about problems in a similar way. Innovation often requires someone tangential to the field with a different perspective, someone who can see connections in seemingly unrelated research. Will this person be on the list?

Perhaps the greatest concern is establishing a precedent that changes the rules by which scientists work. Currently, laboratories that work with dangerous biological agents must receive security clearance and are tightly monitored. Will similar clearance be required just to know about that research? It is important to keep in mind that censoring the science in question does not prevent someone from stealing a deadly virus, but rather from building one. The requisite social and intellectual integration into a modern professional community needed to carry out such a deed does not fit the terrorist profile, whether it be the Unabomber, Timothy McVeigh or the 9/1 1 hijackers. By analogy, stealing an airplane to slam into a building is one thing, building that airplane is something else.

High population density and greater mobility increase the likelihood of a disease pandemic. In worrying about terrorism, we should not lose sight of the fact that the bird flu studies suggest that we are at least as vulnerable to the whim of nature. Security comes from having as many creative minds as possible thinking of ways for people to take care of each other. To do otherwise involves risk.

Mark O'Brian is a professor of biochemistry at the University at Buffalo.