The Netherlands is this week counting the cost of the latest outbreak of avian flu. The flu, which is sweeping through the country's chicken flocks, has already infected more than 50 farms and resulted in the culling of over 2 million chickens.

The economic implications of the outbreak, which began on 28 February, are severe — the Netherlands is Europe's largest exporter of chickens. But the cost could have been much greater.

Initial fears that the disease was linked to cases of avian flu in Hong Kong (see Nature 422, 6; 2003) were quashed early on, thanks to the rapid detective work of a team at Erasmus University in Rotterdam. The Hong Kong strain, known as H5N1, sparked widespread alarm after it killed a man and infected other members of his family.

Within 48 hours of receiving samples of the virus that caused the Dutch outbreak, the Rotterdam team had identified it as an H7N7 strain, revealing that its genetic characteristics are similar to those of a strain identified in mallard ducks in 2000.

Nevertheless, the Dutch virus can infect humans, and as of 18 March it had caused conjunctivitis in at least 32 people, some of whom have flu-like symptoms. Anxious researchers are not ruling out the possibility that the chicken flu could trigger a human flu pandemic. But for this to happen, the virus would have to infect someone infected with a normal human flu virus, so that the two strains could swap genes to produce a more virulent humanized virus.

This is thought to be highly unlikely. But to reduce any chances of such co-infection, Dutch doctors are pursuing aggressive drug treatment of those infected, in an attempt to reduce their viral load.

Avian flu outbreaks are rare, but can kill entire flocks in a few days. Transmission of the virus from wildfowl to domestic chickens may have occurred through free-range hens that spend time outdoors, suggests Albert Osterhaus, who heads the Rotterdam team.