Online today, gone tomorrow.

By the time a research report or publication goes to press, many of its references, in the form of footnotes and endnotes, may no longer exist. That is because more researchers and professionals are relying on information found oil Web sites as sources for their scientific, academic, and white papers.

Today, references are more likely to refer to Web pages than books or journal articles, hut before the ink dries, many of the sites cited will have moved to other locations on the Internet or disappear altogether, rendering them useless to readers.

In research described in a recent edition of Science, a team looked at footnotes from scientific articles in three major journals--The New England Journal of Medicine, Science, and Nature--three months, 15 months, and 27 months after publication. The prevalence of inactive Internet references grew during those intervals from 3.8 percent to 10 percent to 13 percent, respectively.

People are increasingly dependent on the Internet to get information from companies, organizations, and governments. Yet, studies have revealed that the average lifespan of a Web page today is a mere 100 days. For example, of the almost 2,500 British government Web sites, 25 percent change their uniform resource locators (URLs) each year, according to Electronic Publishing Services in London.

A growing number of scientists and scholars are nervous about the increasing reliance on what is proving to be a transient medium. The situation is particularly sobering because some documents exist only as Web pages--for example, the British government's dossier on Iraqi weapons.

Web sites become inaccessible for many reasons--site managers may move on and remove the material, leaving the site to "die," or the site's handlers may move the material to a different URL without providing a direct link to the new site.

Help may be on the way, however. Several companies, including search engine Google and Kahle's Internet Archive (www.archive.org) are taking snapshots of Web pages and archiving them as quickly as possible so they can be viewed even after they die or move. The Internet Archive already contains more than 200 terabytes of information (1 terabyte is 1 trillion bytes)--the equivalent of 200 million books. Each month, it adds 20 more terabytes, which is equivalent to the number of words in the entire Library of Congress. But with an estimated 7 million new pages added to the World Wide Web each day, archivists can do little more than play catch up.