Just what do we want to get for the billions the Government spends on research. The increased concentration of research funding in a few universities, highlighted by the imminent increase in research overheads which we report this week (page 3), is opposed vigorously by the heads of the new universities. But it is not only those institutions that have never had research money which stand to lose. Many old universities with solid but not front-rank research reputations stand to lose too, as Frank Gould pointed out in The É«ºÐÖ±²¥S last week.

In 20 years, it may turn out that spreading the money more thickly in fewer places was the right idea when it comes to impressing the Nobel committees. The Laboratory for Molecular Biology in Cambridge, for many years the main British machine for Nobel prize winning, succeeded in large part because the Medical Research Council and others put substantial resources in it that could have gone elsewhere.

The decision to increase overhead payments means that in future years, a small number of universities are more likely to have the resources to compete with the world's best research institutions. Their researchers will also be best placed to join international research teams and make the most of major scientific investments in supercomputers, satellites, ships or telescopes.

However, this does not mean that they will be able to supply all the research the nation needs to ensure its economic future, or to answer key questions in areas such as health and safety, transport or planning. One example of this is Bovine Spongiform Encephalitis and the related diseases of sheep, humans and other species. A premier-league research university might work on the biology of the prions that cause the infection, but where would work on its transmission between species, or on the effectiveness of incinerating affected animals, go on?

É«ºÐÖ±²¥

The right place for such work is surely in either the university system or in government laboratories, and with significant crossover between the two. However, the pressure on departmental budgets has meant that BSE research has only been able to expand within the agriculture ministry's budget at the expense of other lines of work, including the improvement of livestock. "The support of government policy" is one of the declared aims of state science spending, but if science is to be used in this way, politicians and research managers have to be willing to pay for it.

Long-term scientific research that creates new ideas and industries needs commitment over decades. But the strategic research needed when problems arise requires something even more problematic - money that can be found without wrecking ongoing research of lasting value. This is a test which the agriculture ministry has failed in the BSE saga so far.

É«ºÐÖ±²¥

Ministers' insistence that they are doing their best on the basis of the scientific advice seems to many scientists like a tactic for concealment and obfuscation. It only makes matters worse if politicians then fine other scientists to pay for such research instead of finding new money.