Influenza victims crowd into an emergency hospital at Camp Funston, a subdivision of Fort Riley in Kansas, in 1918. The flu killed at least 20 million people worldwide.
A novel flu circulated in some American cities in April and May of 1918, causing mild illness and going largely unnoticed. It returned in September, and again in January...
Another new flu appeared in Louisiana schoolchildren in early summer 1957 and smoldered for several months before turning virulent; more than 60,000 people died.
While the public may be over the initial scare from Mexico last month, it is these longer-term scenarios that keep pandemic experts up at night. A seeming resurgence of swine flu in New York over the past several days and its continuing spread worldwide haven't helped.
Despite an explosion of genetic information over the past decade, strikingly little is known not just about pandemic flu but about the seasonal variety as well.
Scientists don't even understand why it is seasonal. Indoor crowding, humidity and temperature are all possible explanations for the November-to-April season in North America; some tropical countries have year-round, low-level flu with two peaks rather than one.
But an out-of-season appearance is a red flag for public health officials
-- and one that, in hindsight, was not heeded amid the chaos of World War I.
"The epidemics which occurred in the spring," Army Sanitary Corps Major George A. Soper wrote in the Nov. 8, 1918, issue of the journal Science, near the peak of the pandemic, "were like those which are taking place now, except that the disease was milder and there was less pneumonia. ... Something seems to have occurred during the summer greatly to increase the virulence of the disease."
That so-called Spanish flu turned out to unfold in four major waves in the United States
: February to May 1918, September to December 1918, January to April 1919 and December 1919 to March 1920.
The first wave was the weakest, and until recently was largely unnoticed by scientists and historians who focused on the far more deadly second wave. Death data from that era are abundant, but illness statistics are not.
Stephen S. Morse, a professor of clinical epidemiology at Columbia University's Mailman School of Public Health, co-authored a 2005 paper that analyzed death records for different age groups -- pandemics disproportionately hit young adults -- to show that an early wave had hit New York City in the spring of 1918.
"But why very often the subsequent waves are more virulent,
" Dr. Morse said in an interview, "is unknown."
What is known is that influenza A circulates around the world, constantly undergoing small mutations known as "drift." People who are exposed retain some immunity depending on the extent of the drift, but it is enough to require new vaccines every year.
Every once in a while, a major change -- usually a genetic reassortment, known as a "shift" -- creates a form of virus that is virulent, highly contagious and novel enough that people have little immunity. This causes a pandemic, often in several waves. The population gradually builds immunity and it persists as the seasonal flu until the next pandemic comes along to displace it.
Species preferences and other differences are not reflected in the H and N designations. The current H1N1 is believed to be a distant relative of what began in 1918.
The link may help explain the current pattern of disease among younger people since anyone born before 1957 would have residual immunity. Some protection may also be conferred by forms of H1N1 that have been circulating for unknown reasons since 1977.
Scientists speculate that, unlike the seasonal flu, pandemic flu might first appear in any season because people have little or no immunity; it strikes an immediately susceptible population.
"I think we already know that this virus has pandemic potential," said Edwin D. Kilbourne, an emeritus professor of microbiology and immunology at New York Medical College.
So far, though, the new H1N1 has proved only slightly more deadly than the seasonal flu, Anne Schuchat, a CDC infectious-disease expert, told reporters last week. Yet it's an unusual season, with particularly late outbreaks in schools.