An epidemic of dengue type 2 infection occurred in North Queensland during 1992 and 1993. A random serosurvey of 1,000 residents of a population that experienced this epidemic only during 1993 was conducted to determine the proportion of the population at risk for secondary infection in the event of another epidemic with a different serotype. The ability of an ELISA to detect prior exposure to the dengue virus was compared with the hemagglutination inhibition assay. Dengue 2 virus plaque-reduction neutralization assays were performed to evaluate the specificity of the antibody response. Antibodies to dengue virus, or closely related flaviviruses, were detected in 61.9%. Seroprevalence increased with age and correlated well with known previous epidemics in the region. The sensitivity and specificity of the ELISA was 99.2% and 96.2%, respectively. An estimated 26% of the population was infected during the 1993 epidemic.