The recall notice from the Food Safety and Inspection Service of the U.S. Department of Agriculture duly stated that it was "voluntary" (all recalls are), and that it was Class II, low health risk. To the USDA, recall No. 067-2003, of "approximately 10,410 pounds of raw beef that may have been exposed to tissues containing the infectious agent that causes bovine spongiform encephalopathy (BSE)," involved a lesser threat than, say, recall No. 055-2003, on Oct. 28, of 26,000 cans of mislabeled chicken-and-pasta soup. There was nothing wrong with the soup, but the list of ingredients omitted the eggs in the pasta; the danger that someone with an allergy might eat it made for a Class I (high risk) recall. Let a noodle fall into a vat of soup, and America's food-safety net is there to catch it. But let a cow too sick to walk be slaughtered and shipped off to supermarkets--well, that's all perfectly legal under the watchful eye of the USDA.

It will be some time before the entire five tons of beef that passed through Verns Moses Lake Meats in Washington state on Dec. 9 are accounted for, but it seemed likely that by the time the recall was announced two days before Christmas, at least some of it had already gone where no USDA inspector could retrieve it--and what the agency considered a "low risk" didn't seem that way at all to anyone who had eaten a hamburger in the third week of December.

The episode cast a harsh light on a food-safety system patched together out of a dozen overlapping jurisdictions, operating on laws in some cases nearly a century old. It's a system that seems dangerously unprepared to cope with the kinds of threats likely to arise in the 21st century, even those as predictable as the eventual discovery of BSE in American beef. Industry critics have long asserted this was only a matter of time--occasionally adding that the USDA's testing regimen seemed almost designed not to find it. In one week, Agriculture Secretary Ann Veneman managed to re-enact nearly a decade's worth of dithering by the British government, which was forced to destroy 5.5 million head of cattle after admitting in 1996 that eating infected meat appeared to cause a fatal brain disease in humans. She proceeded from hollow assurances ("We remain confident in the safety of our beef supply," she said on Dec. 23) to panic over lost exports (40,000 tons were left marooned on foreign docks or aboard ship when 48 countries, led by Japan, banned imports of American beef soon after the recall), and at last to the enactment of reforms that had been deemed unnecessary just a week before (with the grand promise on Dec. 30 that "these additional actions will further strengthen our protection systems").

To be sure, America's food-safety system does the job for which it was designed, well enough that we can take for granted the conquest of diseases such as trichinosis. "As late as the 1940s and 1950s, we still had parasites in the food supply," says Dr. Lester Crawford, deputy commissioner of the Food and Drug Administration. Today, he says, "there's no doubt we have the most perfect system in the world." It's true that as many as 76 million Americans get sick from food each year--and 5,000 die--but most cases are mild, sporadic and/or a result of the food industry's preferred culprit, poor "consumer education." (Let he who is without a package of two-year-old cream cheese in his refrigerator cast the first scone.) But the system has been slow to respond to the changing nature of threats to the food supply. Only over the last six years has the time-honored "poke and sniff" standard for inspecting meat been gradually supplanted by laboratory testing, intended to catch such subtle pathogens as salmonella--"much more of a risk" than BSE, in Crawford's view--or the deadly variety of E. coli from fast-food burgers that killed four children in 1993. And while hepatitis A, which killed three people and sickened more than 600 who ate at a Pennsylvania restaurant last October, can be controlled by elementary sanitation, the government cannot stand over every cook in the nation--much less every field hand in Mexico--to make sure they all wash their hands before touching a scallion. A report by the Centers for Disease Control counted 2,751 outbreaks of foodborne disease between 1993 and 1997--and by far the most dangerous category of food, with 12,537 cases, was fruits and vegetables.

By contrast, the number of Americans with variant Creutzfeldt-Jakob disease, the brain disorder linked to BSE, is essentially zero, and few scientists expect that finding a single mad cow will cause that number to go up much, if at all. In Britain, where hundreds of thousands of cows were diagnosed with BSE beginning in 1986, the number of new human cases has leveled off at fewer than 20 a year, not the tens of thousands that some scientists once feared--although researchers caution that so little is known about the disease that they can't be sure the worst is over. But not since one kicked over a lantern and burned down most of Chicago has an American cow caused such consternation. The insidious nature of vCJD, which can incubate for years before erupting into dementia, paralysis and death, loomed larger in some consumers' minds than the infinitesimal chance of catching it. "I'll go back when I see Dick Cheney eating steak tartare on Fox News," said Mark McGroarty, a 35-year-old Angeleno who had his last hamburger just before the recall was announced.

But McGroarty seemed to be in the minority. "I was quite amazed at the resilience consumers showed to the news," said Mitchell Speiser, senior restaurant analyst at Lehman Brothers. That was true even in Seattle, the big city closest to the outbreak. The mix of orders at Jak's Grill, an upscale steakhouse, was "90 percent" the same as before the recall, said owner Ken Hughes, and the owners of a taco truck in the Latino neighborhood of White Center said they hadn't noticed any falloff in the sales of tacos de sesos--or brains, the part of the animal most likely to harbor BSE. In Cleveland, Jennifer Cherni, who lost 50 pounds on the Atkins diet in 2003, weighed the risk of getting fat by going back to spaghetti against the remote chance of losing her mind in 10 years, and came to a predictable decision: "I think mad cow must be pretty tough to catch."

The news resounded loudly on Wall Street and the campaign trail. McDonald's, after a brief sell-off, recovered nicely; the company took the occasion to issue a press release about its longstanding policy against using "downer" cattle, like the one implicated in last month's recall. One of the biggest steps Veneman took last week was to ban the sale of meat from these animals. They accounted for 150,000 to 200,000 of the 35 million cattle sent to slaughter last year, of which only 20,000 were tested for BSE. The ban on downers was endorsed by the National Cattlemen's Beef Association, which had opposed the idea in Congress for years. "I don't think there was anything wrong with the systems we had in place," the association's president, Idaho rancher Eric Davis, told reporters. The cattlemen contend that most downer animals are merely injured, not sick, and safe to eat. Another prominent cattleman, Jim McAdams of Adkins, Texas, urged the public to remember that "the goal is to provide the world not only a safe beef product, but also [one that is] abundant and affordable."

To consumer advocates and Democratic candidates, the episode was Exhibit 1 in their case against the Bush administration's reflexive opposition to government regulation, except in a crisis. Democrats in Congress, led by Rep. Gary Ackerman of New York and Sen. Daniel Akaka of Hawaii, had been trying for years to enact a ban against downer meat and been thwarted by Republican majorities--and opposition from the USDA itself, several of whose top administrators come from the cattle industry.

Veneman took a number of other actions, including tougher regulations on what is known as "advanced meat recovery"--machinery that strips every scrap of protein off the bones of a carcass, sometimes inadvertently harvesting scraps of nerve tissue along with the muscle. But critics like Marion Nestle, a leading nutritionist, say what's really needed is an overhaul of the food-inspection system. As she demonstrates in her book "Safe Food," American food is regulated by 12 different agencies, from the Customs Service to the Environmental Protection Agency; pizza falls under either the USDA or the FDA depending on the topping. Why wasn't every downer cow tested for BSE, and why did the suspect cow go into the food supply while inspectors awaited test results? Because testing for mad-cow disease falls under the USDA's Animal and Plant Health Inspection Service (APHIS), not the Food Safety and Inspection Service (FSIS)--and APHIS's job is to keep animals healthy. "Obviously it stands to reason that healthy animals are better for food safety than sick animals, but our mission and our authority does not cross into the arena of food safety," APHIS spokeswoman Anna Cherry told NEWSWEEK. Only when the tests came back positive, after the meat was already in markets, did FSIS get involved. In fact, the decision to make APHIS the lead agency on mad cow was done deliberately, according to one official who was involved in the discussions in 1989, because it actually has stronger statutory authority for enforcement than its sister agency--in imposing quarantines, for instance. The law is in some respects tougher on protecting livestock from foot-and-mouth disease than people from E. coli. But what made technical sense was awfully hard to justify to the American people, let alone the Japanese.

Short of rewriting thousands of pages of legislation, though, critics say there are additional steps the government can take now, and it has so far refused to take them. The biggest concern involves cattle feed; most scientists believe BSE infects cows when they are fed the ground-up carcasses of animals--including other cows. Both the United States and Canada banned this practice in 1997--significantly, after the Washington state suspect cow is believed to have been born--but consumer groups say the ban is full of loopholes and poorly enforced in any event. For example, cattle that don't make it into the food supply--such as those that die on the farm--can still be fed to other animals. Together with all the parts left over from slaughtering, they are "rendered"--boiled down and dried to a powder--into chicken feed, say. But what happens at the henhouse? "Whatever the chickens don't eat becomes waste material that is combined with dead chickens [and] fecal material... and rendered to feed back to cattle," says Caroline Smith DeWaal, director of food safety at the Center for Science in the Public Interest.

"We can look at what happened in Britain and learn from their mistakes," says John Stauber, executive director of the Center for Media & Democracy and coauthor of the book "Mad Cow U.S.A." The two critical steps he identifies are a total ban on rendered feed for all food animals, and testing for BSE every cow that enters the food supply. Japan already does that, and Europe tests every animal older than 30 months (the age at which the incidence of BSE starts to climb), says Dr. Stephen De Armond, a neuropathologist at the University of California, San Francisco, and an authority on BSE. De Armond's colleague Stanley Prusiner won a Nobel Prize for identifying the suspected infectious agent in mad cow and related diseases, an aberrant protein he named the prion. Prusiner's theory explains one of the most confounding things about BSE. You can't get rid of it by cooking--the way you can viruses or bacteria--because the infectious agent isn't alive in the first place. So it can be stopped only at the source. To test every head of cattle in the country "would make me feel a lot better," De Armond says--at a cost he estimates at no more than 10 or 20 cents a pound. Is your brain worth any less?