A study in Sweden indicates that women are consistently marked down when applying for research funds. Hugh Aldersley-Williams calls for action to dismantle the laboratory's glass ceiling
Female scientists are systematically penalised by their mainly male peers when applying for funds, according to researchers at Goteborg University in Sweden. Christine Wenneras and Agnes Wold used their country's press freedom laws to release peer reviewers' scores for research applications to the Swedish Medical Research Council. "Our study strongly suggests that peer reviewers cannot judge scientific merit independent of gender," they wrote in last week's Nature. Reviewers scored women consistently lower on three criteria: their competence based on published work, their methodology, and the relevance of their research.

Like women in other fields, female scientists have always experienced overt discrimination, but they are often also subject to covert discrimination special to the sciences. This may largely explain both the minority of women going into science (through an implicit male bias in the way it is taught, for example) and the subsequent attrition with each stage of promotion.

In Britain, girls equal boys in achievement at school, but begin to drop science when they can choose their subjects. Women account for 28 per cent of science graduates, but only 20 per cent of researchers and 10 per cent of full-time academic staff, and just 2 per cent of professors are female.

Even in egalitarian Sweden, the situation is similar. In biomedicine, the field examined by the Goteborg survey, women gain 44 per cent of doctorates but hold only 20 per cent of post-doctoral positions and 7 per cent of professorships.

Swedish researchers believe that their findings may hold a complete explanation for the attrition. Women needed to be far more productive than male colleagues - to the tune of three extra papers in a top journal such as Nature, or 20 in a leading specialist journal - to attain the same competence score from the peer reviewers. When competing for research funds, being both female and without the right connections "was a double handicap of such severity that it could hardly be compensated for by scientific productivity alone," Wenneras and Wold concluded.

A cynic might advise a woman should use nepotism to compensate for the sexism, but there are more positive remedies on offer. The Royal Society's Rising Tide report in 1994 sought to encourage girls to stay with science in school and employers to use "family-friendly" policies to retain female researchers. The report also looked to increased media coverage of female role models.

It is recognised that a "peer effect" is important. At Cambridge University, the biochemist Professor Jean Thomas has inspired generations of women to pursue careers in science. "She showed that you could be the kind of woman you might want to be and still be a good scientist," says Dr Fiona Watt, a principal scientist at the Imperial Cancer Research Fund.

Dr Watt finds that she in turn draws a disproportionate number of women to her research group. "As there are more of them, they can help each other. In terms of the management of science, women are often bolder, but they need the support of other women if they are not to feel isolated. When it comes to doing research, though, it's harder to say whether there are differences between men and women."

It is in crystallography where the peer effect is seen most dramatically - and where preconceptions are quickly shattered. For a start, it is not one of the biological sciences, which are patronisingly supposed to attract women. Nor is it a soft option; crystallography demands mathematical dexterity and the ability to think in three dimensions turned inside-out. It shows that the science that women do is the equal of men's. Kathleen Lonsdale used X-ray diffraction to confirm the hexagonal structure of benzene and related molecules. Dorothy Hodgkin worked out the structures of penicillin, vitamin B12 and other biologically important chemicals. Rosalind Franklin played a crucial role in discovering the structure of DNA. Such work was, and continues to be, vital to chemists' efforts to synthesise new drugs.

The legacy of these mid-century pioneers thrives today. Dr Olga Kennard, one of the first female Fellows of the Royal Society, is scientific director of the Cambridge Crystallographic Data Centre. Two of Britain's three female professors of chemistry are crystallographers.

One of them is Judith Howard of Durham University, who worked in Hodgkin's laboratory. "I don't remember ever thinking `I'm going to grow up and be like Dorothy'," she recalls, "but you weren't put off. I've got at least 50 per cent women in my group. They're drawn to it; I don't seek them out." The other is Mary McPartlin of the University of North London. "I can trace my lineage back, as every crystallographer can, to the Braggs. From early on, there have been distinguished women crystallographers in this country." Professor McPartlin believes the fact that women have been at the top of this science since its inception makes it natural that they should continue to be drawn to it.

In other fields - "hard" disciplines such as mathematics, astronomy and spectroscopy, as well as "soft" ones such as botany - women were prominent in the 19th century. Before the emergence of professional laboratories and career scientists, ladies of leisure could contribute to the advancement of knowledge just as gentlemen could. This reinforces the sense that there is something wrong with the institutions and procedures of science rather than anything off-putting about science itself.

Diagnosing the full extent of the problem and prescribing a cure remains a challenge, however. Encouraging female students to take up science is admirable but slow to work through, and ultimately a waste of time if the number of women continues to be decimated at every promotional hurdle.

Tackling the attrition has focused on the less important stages. Last year, the Engineering and Physical Sciences Research Council introduced a five-year part-time research studentship aimed at scientists trying to ease themselves back into a research career. Child-care help and a rethinking of the short-term contract system of employing research scientists comes high on the list of hopes of those who do return. But none of this addresses the most dramatic drop-off in numbers, as the 10 per cent of women academics is culled to only 2 per cent of professors. "Men are very good at helping women to a certain point, but not beyond it," says Professor Howard. Beyond the post-doctoral stage, competitive pressures increase at a time when many people choose to start a family.

Perhaps more radical thinking is needed. The Labour Party's short-living policy of all-female shortlists for some seats has transformed Parliament's composition, after years of slow, minimal change. If it worked for one hidebound institution, might it not work for another? Rosalind Franklin died before James Watson, Francis Crick and Maurice Wilkins won a Nobel prize for their part in elucidating the structure of DNA. Many think she would have shared it. (Nobels are not given posthumously.) Is it now time for Rosalind's List? .