The UK government took a brave step earlier this week when it decided to make known publicly the precise locations of four new trial sites for tests of genetically modified (GM) crops, despite the increasing numbers of attacks on such sites by environmental activists. The four farm-scale trials of GM oilseed rape will begin this autumn at sites in Lincolnshire, Nottinghamshire and Hertfordshire, and will bring the total number of test sites to 75 next year.

The announcement, from the Department of the Environment, Transport and the Regions, quickly came under fire from farmers and commercial groups, who argued that it was an open invitation to protesters to destroy the trials (significantly, groups such as Greenpeace have declined to promise not to take such action). The British Plant Breeders Association, for example, says that it had recommended to the government that only the counties in which the trials were being held should be revealed, as is the practice in Germany and France.

But given that the issues at stake in the conflict over GM crops are as much to do with trust in public institutions as they are with rational action — or even scientific data — the government's decision in favour of openness must be applauded. The nuclear industry has already suffered the consequences of combining rigid secrecy with overstated claims to safety; the one can all too easily feed on the other, and attempts to break this particular vicious circle are therefore welcome.

They are also timely. In a week that has seen the company Monsanto heavily criticized by the UK Advertising Standards Authority for exaggerating some claims about the safety of its products in a public relations blitz in Britain last summer, the need to restore public credibility in the whole handling of the GM issue remains high on the political agenda (see page 702). Openness will not achieve this on its own. But the alternative is doomed to fail.