Severe acute respiratory syndrome (SARS), which was responsible for about a hundred deaths by the time Nature went to press, was already roaming the globe by airliner by the time medical science realized that it was dealing with something new.

Asked at a press conference whether SARS serves as a fire drill for a bioterrorist attack, James Hughes, director of the National Center for Infectious Diseases at the US Centers for Disease Control and Prevention in Atlanta, Georgia, agreed that it does. But he pointed out that it also serves as a fire drill for another grim possibility: the next influenza pandemic.

The parallels are clear. The leading theory is that SARS is caused by a virus that jumped from animals to humans (see page 547), much like new strains of flu. And the mortality rate is similar: around 3% for SARS, compared with about 2.5% for the Spanish flu of 1918. Thankfully, today we are better equipped to respond to the threat. The genome sequence of the prime suspect, a new strain of coronavirus, will become available any day now. This should help to reveal where the virus came from, suggest reasons for its lethality, and speed the development of rapid tests for its presence.

The genomic information would be nearly useless, however, were it not for the basic research into coronavirus biology carried out since this family of viruses was first found to infect humans in the 1960s. In the United States, vast sums of money are now being ploughed into biodefence, and nearly $300 million of the 2003 allocation is designated for basic research, including genomics. SARS reminds us of the value of ensuring that this research is 'dual use' — relevant to fighting not just bioterrorism, but also naturally emerging diseases, which may ultimately prove more dangerous.