It is ironic that, only weeks after politicians so eagerly celebrated the imminent completion of a draft of the human genome sequence, science programmes at the Department of Energy (DoE) should find themselves under pressure again (see page 221).

Ironic, because it was the DoE, with its traditional strength in ‘big science’, that got the Human Genome Project under way in the first place, in the face of tenacious resistance from the National Institutes of Health (NIH), which later acquired leadership of the project.

The NIH is now enjoying a major build-up of funds that is likely to take its budget above $20 billion next year. Growing numbers of its investigators rely on large scientific facilities, such as supercomputers or synchrotron light sources, to conduct their research at the frontiers of molecular biology.

But the NIH doesn't operate any large facilities; they are all maintained by the DoE. It has recently agreed to help equip some of the DoE's synchrotrons, in response to the fact that biologists are among their largest users. But the synchrotrons themselves will continue to rely on the DoE for their operation, maintenance and replacement.

The DoE has many enemies in Washington, but even the most bitter of them expresses no desire to damage the US scientific infrastructure. Sound science policy demands a balance between the support of individual investigators — by the NIH and the National Science Foundation — and the maintenance of major facilities, and their associated research programmes, at DoE laboratories.