THE continuing BSE took yet another twist this week, though oddly time there was both and bad news. The positive news was the prediction by an international team of scientists that the disease will disappear by the year 2001. The bad news was that they also calculate that 730,000 infected animals ended up in the British food chain during the last 10 years.

British farming representatives - have called on their government to review its culling policy as a result of the (good) news, or at least to consider other options. The European Commission will also review the findings.

But concerns were again raised about the public health risks posed by BSE, and the many other unanswered questions that remain, including the possibility of maternal transmission. The news earlier this month that cows may be able to pass the disease to their calves means the spotlight may yet turn on milk.

The more we learn about BSE, it seems, the more we have to change previously well held positions. And it may yet be that our historic complacency about scrapie in sheep has to go as well.

What future for farmers then?

For the moment, however, perhaps the most important point raised by the latest report is that it highlights once again the need for a test that can diagnose the disease in live animals before they show any symptoms.

This week's conclusions are the latest in a long line of developments, as scientists struggle to understand an unusual disease, caused by a novel and little understood agent, the prion protein. When it comes to BSE there are still few certainties, and our knowledge - is changing all the time.

The latest study - the basis for a major 10 page article published in Nature - was conducted by an international team of 15 scientists who specialise in predicting trends in the incidence of infectious diseases, and who have in the past advised the British government on the AIDS epidemic. Led by Prof Roy Anderson from Oxford University, they used the latest statistical techniques to calculate what the incidence of BSE was, historically, and what it might be in the future.

In the absence of a diagnostic test there is no information on the true incidence of BSE in Britain, but making various assumptions the scientists calculate that, overall, 903,000 animals were infected between 1974 and 1995. Of these, only 161,412 (or just 18 per cent) were diagnosed, the remaining 730,000 went undetected into the food chain.

This is because, although most animals are probably infected in their first year, it takes about five years for the disease to develop, and most beef cattle are slaughtered before they are three years old. Even, if they are infected they are unlikely to show any signs of the disease at that age.

These figures are yet another setback for the officials and others who maintained for years that infected animals were not entering the food chain. Not only does it now seem that they were, but also that the many cases diagnosed were merely the tip of an iceberg.

Most worrying, according to Prof Anderson and his team, are any animals with full blown BSE that might have entered the food chain in the very early years before it was properly recognised: although the link between BSE and the new variant CJD in humans is not clear, the most infectious animals are probably those that are already showing symptoms, rather than those that show no sign of the disease.

It is still not possible to calculate how many truly sick (and therefore infectious) animals might have been slaughtered for consumption, but the researchers say that the emergence of v-CJD cases in 1994 would be consistent with eating infected beef slaughtered in the late 1980s (that is, an incubation period in people of four to six years, similar to that seen in cattle.)

The latest calculations also take account of the possibility that cows can pass BSE to their calves - earlier this month, a small study by the British Ministry of Agriculture suggested that maternal transmission may happen in 10 per cent of cases.

Prof Anderson's team estimates that maternal transmission may account for 5,000 infected animals, or just 0.5 per cent of the total 903,000.

Looking forward, they predict that at most there will be 20,000 infected animals, and 7,000 cases diagnosed between now and 2001. The ban on infected feed was effective, they say, and so any new cases will be due to maternal transmission, but this will not sustain the epidemic. They predict the disease will then disappear "within about six years, even without any culling.

The team also used its predictions to evaluate various culling options, but concluded that only a massive slaughter policy would make any significant impact, and that this would require culling at least one million animals this year.

The most effective strategy, it suggests, would be to target herds, with a threshold level of incidence, (at least one case per 50 cows), as, well as any calves born to cows that: subsequently developed BSE. Even that policy, they say, would affect some 25,000 herds, and would not necessarily be any more efficient than letting the disease take its course, or at least the course that the team predicts it will take.

Ultimately, however, since, there are still few scientific certainties and no live animal test, the decision regarding culling policy will have to be a political one, weighing the concerns regarding public health against consumer confidence and pressure from farmers.

But this is not the first time the end of BSE has been predicted. Prof David Skegg from Otago University in New Zealand, asked by Nature to comment on the latest research, remembers that in 1991 scientists predicted that the disease would disappear in 1995. Yet last year 14,000 new cases were confirmed. The epidemic may well have passed its peak: but the history of BSE, he says, is a "continual retreat from entrenched positions".

Officials also maintained for years that no infected animals had entered the food chain. Now it seems that the majority of them did, Prof Skegg, says. They also asserted that maternal transmission was rare, if it happened at all. Now it seems that, too, can occur, although the latest calculations suggest that maternal transmission will make only a small contribution to the overall number of cases.

The big questions now, Prof Skegg says, are whether maternal transmission takes place before, during or after birth, and more importantly, whether it is via the embryo, placenta, blood or milk.

Prof Skegg also points out that tests with mice found no evidence that placental material taken from infected cows was infectious. This could mean, he says, that there is something wrong with the mouse based tests.

Answering these and all the other, questions that remain regarding BSE, CJD and scrapie, will take years of detailed research. In the meantime, all we have are best estimates, while governments and consumers want certainties.

Moreover, that detailed research will take not just time, but money. Yet, according to Prof Skegg, the British government spent just £3 million sterling on BSE related research during the crucial three years, 1988-91.

It is easy to be wise in hindsight, he admits, but it is also a salutary warning to all who would skimp on funding new areas.

More research in the early years of BSE could have produced the much needed live animal test. And then not all of the 730,000 infected animals would have been eaten.