Historians, like other researchers, increasingly cite sources from the Internet. Scholars in the sciences, however, have raised alarms about the frequency with which Internet sources have disappeared after their citation in journals. Do humanists and social scientists face the same problem? This article examined the reliability of World Wide Web citations in two leading history journals over seven years and found that 18 percent of web links referenced over that period were inactive, and that the problem increased over time. These findings suggest that historians (and probably other humanists) face a major problem in scholarly practice: we are citing Internet sources as though they were permanent, when in fact they are ephemeral.