Yet these explanations "remain astonishingly superficial and full of inconsistencies," says Dr. Scott Dowell, director of the Global Disease Protection Program at the Centers for Disease Control and Prevention in Atlanta.

Now Dowell and other researchers are focusing on a provocative new hypothesis that blames annual flu epidemics on something most people don't get enough of this time of year: sunshine.

In a paper scheduled for publication next month in the journal Epidemiology and Infection, a Harvard University-led team proposes that a vitamin D deficiency caused by inadequate winter sun exposure may predispose people to infection.

If this theory proves correct, it would not only solve a long-standing mystery, but could also have major public health consequences.

Influenza kills an average 36,000 people in the U.S. each winter, mostly the very old and very young. If scientists could pinpoint the secret behind its seasonal recurrence and somehow alter it, "the potential impact would be far greater than the current influenza vaccine," says Dowell.

Hippocrates, the Greek physician widely regarded as the father of medicine, was the first to recognize that certain diseases ebb and flow with the calendar.

"Whoever wishes to investigate medicine properly," he wrote around 400 B.C., "should ... consider the seasons of the year."

Epidemiologists, however, have found that this is easier said than done.

"You look at the environment around you and say, 'What's the difference between winter and summer?'" says Dr. David Fisman of the Ontario Provincial Public Health Laboratory in Canada. "There are so many things that are seasonal, it's really hard to tease them apart."

One obvious answer is that it's colder in winter. And for as long as parents have bundled up their babies, there's been an unshakable belief that catching a chill makes a person more susceptible to cold or flu.

Cold flunks tests

Since World War II, scientists have devised numerous teeth-chattering tests of this stubborn wives' tale, dunking volunteers in cold baths or confining them to refrigerated meat lockers while squirting virus-spiked mucus up their noses.

"All attempts at demonstrating some relationship between cold exposure and susceptibility to infection have proved negative," Ron Eccles, director of the Common Cold Centre at Cardiff University in Wales, concluded in a recent published review of such studies.

Experiments on the influenza virus have hinted that the flu bug is more stable in the cool, dry air of winter. But that doesn't solve the mystery, says Fisman, whose summary of efforts to understand seasonal influenza is scheduled for publication next year in the Annual Review of Public Health.

One reason: Influenza surveillance efforts in Southeast Asia and other steamy tropical locales reveal that flu is not only common there but also exhibits seasonal patterns akin to those in colder climates.

"If you go back to the hypothesis about cold temperature and flu, it doesn't hold for the tropics," says epidemiologist Cecile Viboud of the National Institutes of Health's Fogarty International Center in Bethesda.

Then there's the crowding theory.

Because scientists think that the flu spreads only from person to person, most have assumed that the disease ravages in winter because people are cooped up in close quarters.