For several years our small garden near the cathedral in Norwich has been hung with metal canisters that dispense sunflower seeds, peanuts and various grains to the birds. Only when there were as many rats eating the spilled seed on the ground and clambering over the trees from which the seeds hung, as there were birds feeding, did the family complain.

The rat catcher was booked. He came immediately, laid down bait to accustom the rats to the taste and said he would be back next day to put poison on it. He didn’t count on my wife’s soft heart. She is a Scot and, like many Scots, has an otherworldly streak. She was determined to emulate the famous Lady Dowding and communicate with the rats. She suggested to them, just as Lady Dowding had, that unless they left the garden promptly and disappeared before the rat catcher returned, they would be poisoned.

If my wife wanted to wander around the garden talking to the rats that was all right by me, but I didn’t think that her words would be as effective as a hefty dose of warfarin. I was wrong. Next day the rats that had been running all over the garden for weeks had gone. The rat catcher put down his poisoned bait but not a bite of it was taken.

Last year we worried about the myriad organisms that the rats attracted by the scattered bird seed might be carrying. This year it seems a trivial concern compared with the avian flu virus that migratory birds will mutate and spread from person to person.

Autumn is now upon us. Should we, with the hundreds of thousands of other people who feed the birds in our cities and suburbs, change the habits of a lifetime to reduce an as yet unquantified risk of avian flu? Ought we to remember that British bird life increasingly depends on bird-table handouts? Last week the virulent H5N1 strain of avian flu was carried by birds migrating from Russia to Turkey, the traditional bridge between Asia and Europe.

The H5N1 strain of flu has been causing concern in the Far East since 1997 and in Europe for the past two or three years, during which time experts have been troubled by the lack of appreciation that this virus would inevitably mutate and eventually spread from person to person.

An international alert was sounded only when the World Health Organisation (WHO) said in July that when the virus was transmitted to humans in the unmutated form it was associated with a mortality that made the Spanish flu outbreak of 1918 look relatively benign. The death rate from Spanish flu was about 1 per cent; the avian flu outbreak in 1997 had a 30 per cent mortality.

Even worse has been the outbreak over the past 18 months: at the time of the WHO report in July, 40 of the 55 people who had caught avian flu from their poultry died.

Professor John Oxford, the flu expert who can be relied upon to give a balanced opinion, has supported the popular and reassuring suggestion that it is likely when the avian flu virus mutates that there will be a shift in its virulence and in its infectiousness. It is thought that it will be likely to become more infectious and less lethal.

Meanwhile Professor Oxford supports the call for laying as great a stock as can be manufactured of vaccines against H5N1, which are already available in some other countries. Until the mutation occurs it will be impossible to produce the perfect vaccine, but if the authorities wait until this happens it may be that they will have waited for too long. It will take at least six months to get even near adequate supplies for a chosen minority. We should plan to have enough for everyone.

Roche is manufacturing Tamiflu, the antiviral agent against influenza, as fast as it can. The manufacturing process is a long one, with ten steps, three of which have potential bottlenecks. The first step and bottleneck has been a dependency on star anise, a plant grown in China. It is harvested for only four months a year and there is a limit to the amount that can be grown.

Roche has found a way to manufacture the relevant chemical artificially in the lab. This has helped to speed up another of the bottlenecks, a hazardous step which, if carried out inexpertly or with too much material, may cause an explosion. The third bottleneck is the stage at which the final product has to be prepared so that it can be loaded into capsules. The simpler, safer steps are already being farmed out to other com- panies so as to reduce production time. The amount of Tamiflu being ordered may well be too late and far too little.

The Royal Society for the Prevention of Birds has issued a statement on its position in relation to avian influenza. It maintains that the risk of migratory birds bringing avian flu into the UK and the risk of it spreading to humans if this does happen is low. The National Farmers’ Union, concerned about poultry farming, is equally confident and reassuring.

At this stage all opinions are no more than conjectures but such optimism seems foolhardy. If migratory birds from Russia can carry avian flu from there to Turkey and Rumania, why not to The Wash, Morecambe Bay and the Solway Firth? The usual reason given is that the infected birds all die en route when passing over mainland Europe.

This would be believable only if the birds never rested, or didn’t feed while migrating (and so couldn’t spread the virus to others). Once at their winter destination the huge flocks of migrating waterfowl frequently feed with the resident smaller bird population, and could also infect local poultry.

The culling of wild birds is, as the RSPB says, unlikely to help and might make matters worse. The obvious steps to take are those that keep wild birds away from poultry. What about bird tables? If Norwich was closer to the Wash I wouldn’t be feeding birds, once migration starts. Nor would I if avian flu became established in the UK. Until then I would. Would I go wildfowling this year? No.