Football matches and pop concerts could be banned and people asked to stay at home if a flu pandemic sweeps the country. But how likely is it that this will happen? Nick Morrison looks at the danger lurking in chickens.

BEHIND it all is the spectre of 1918, but it is not the horrors of the battlefield. The Great War was drawing to a close when a flu pandemic swept the world, claiming more lives than had been lost in the trenches and the mud.

And it is not just the number of deaths - around 220,000 in Britain and 50 million worldwide - it is the nature of the victims. The old and the frail are those normally vulnerable to flu, but this Spanish flu also took the young and the otherwise healthy.

What makes it more worrying is that even digging up corpses to try and isolate the virus has failed to provide a satisfactory explanation as to why it proved so virulent.

It is fear of a pandemic of similar power that underpins anxiety about the danger of avian, or bird, flu. It may so far have been confined to south-east Asia, and have claimed relatively few lives, but it is the prime suspect for what the World Health Organisation (WHO) has warned is the "inevitable" forthcoming pandemic.

Acting on the WHO's advice, the Government is drawing up contingency plans to deal with a pandemic, including banning football matches and pop concerts, and confining people to their homes. Mass evacuations of towns and cities have been ruled out.

What makes avian flu a candidate for a devastating pandemic is the high mortality rate. Since the Spanish flu of 1918, there have been other flu epidemics, most notably Asian flu in 1957 and Hong Kong flu in 1968, but most of those infected recovered, and the deaths were largely among the elderly.

Although the number of people known to have been infected with avian flu is low, the majority of those have died. Of the 52 cases reported in Asia in the last 12 months, 39 have proved fatal, and have included the young and otherwise healthy.

But that is not reason enough to panic yet, says Geoffrey Toms, senior lecturer in virology at Newcastle University.

"By and large it is not easily transmissible from human to human; the majority of cases have been people who have contracted it from birds," he says. "This is not a virus that is going to sweep the world in its present form."

Most cases appear to have been in people who have come into contact with live chickens, which carry the virus, known as H5N1, in their gut. The danger, says Dr Toms, is if the virus mutates.

Other viruses have followed this route. Ebola, Aids and Sars are all thought to have originated in animals before mutating so they could be passed from human to human. And viruses which originate in animals pose a particular threat, according to Dr Toms.

"We have lots of human viruses, which we have had since the year dot and which have evolved with us," he says. "In the long term, it is not good for a virus to make you terribly ill. It wants to hang about and infect your grandchildren, but if you don't live to be 40, you don't live to infect your grandchildren.

"And with normal flu, you have got to be walking around in order to transmit it efficiently, so there is a lot of pressure on the virus not to make you too ill.

"When a virus and host live together for a long time, they evolve to a state where the virus replicates well and the host remains well."

The herpes virus which produces cold sores is a classic example. It does not stop its host from getting about, so it can more easily pass from one person to another. Animal viruses are a different matter. They have not evolved with humans over thousands of years, so when they jump the species barrier they can cause severe infections.

Rhesus monkeys are prone to a cold sore virus similar to the one we get. But if that virus is passed to humans, it has always proved fatal, even though the two viruses are closely related, as are humans and monkeys.

Flu itself was originally a bird virus, found in ducks and shore birds. New strains periodically jump over to humans, combining with existing human flu strains to create a new virus capable of causing epidemics. Most of the major pandemics of the last century were caused by bird flu combining with existing human viruses.

These pandemics have occurred at fairly regular intervals for the last 100 years. As well as 1918, there were outbreaks in 1902, 1936, 1948, 1956, 1968 and 1977, every ten to 15 years or so. But 1977 was the last one, leading many experts to believe another one is long overdue.

And when you put the fact a pandemic is overdue with the emergence of a virulent strain of flu in the Far East, that can be cause for concern.

"It looks as though there has been a fairly well-established pattern of influenza epidemics, so we should really be getting another one," says Dr Toms.

"The conventional epidemic, that we're almost definitely going to get, is going to hit the very old and the very young, the less than twos and the over 60s.

"In order for a pandemic to occur, there has to be the recombination of an old human virus with a new influenza from birds. If there is a lot of bird flu about, which there is, the chance of a co-infection, of people with bird flu and the old human flu, is high. It seems this may be the crucible of a new pandemic."

The standard defence against a virus is a vaccine, but this could only be developed once the new strain has emerged, and then will take some time. The spread two years ago of the Sars virus, which leapt from Hong Kong to Canada in a matter of days, suggests that by the time a vaccine is available, the virus could already cover the globe.

The second line of defence is anti-viral drugs, which limit the symptoms and lessen the chances of the disease spreading. But making large quantities available, when there is a possibility a pandemic might not occur for 30 years, is an expensive risk for both drug companies and governments.

But the doom-mongers are not having everything their own way, says Dr Toms, at least not yet.

"The virus must change in order to transmit from human to human, and in that changing it may become less harmful. Viruses which kill are not generally good at transmitting themselves through a population. You need to be out mixing with people to be transmitting it, not lying in your bed sweating," he says.

A virus has to be present in high enough levels in coughs and sneezes and other discharges to be able to infect another person, as well as being able to stimulate sneezes while not incapacitating its host, in order to spread. All of this means that even if avian flu is able to combine with human flu, the end result will not necessarily be devastation.

"Nobody knows whether it will happen or not, but it could," says Dr Toms. "I would not say it won't happen, although I won't lose a lot of sleep over it."