A sandwich ELISA was developed to detect HPV antibodies using HPV 1 capsids that were purified from recombinant vaccinia virus-infected cells and a monoclonal antibody to the HPV 1 L1 protein. Sera from 91 college-aged women who had been previously screened for HPV 1 antibodies by immune precipitation of capsid proteins were tested by ELISA. A cutoff point was established independently of other criteria based on the assumption that the ELISA values came from a mixture of two Normal distributions representing seropositive and seronegative individuals. It was found that the data fit this model best when the natural log of the ELISA (+0.5 to make all of the values positive) was used. Positive sera were shown to react with a conformational epitope(s) on the L1 protein. In the population reporting foot warts, 16 of 18 (89%) had ELISA values above the cutoff. This compared to 38 of 73 (53%) positives in the population reporting no history of foot warts. The odds ratio for the association of the ELISA reactivity with foot warts was 7.23 (95% CI 1.53, 69.4; P < 0.01). There was no significant association between the ELISA reactivity and wart infections reported at other sites. The average of the log ELISA values for individuals never reporting foot warts was -0.223 (SD 0.468), whereas the average value for individuals reporting foot warts within 10 years was 0.191 (SD 0.450) (P = 0.001). There was a negative correlation between the magnitude of ELISA reactivity and the time elapsed since the last appearance of foot warts. This apparent loss of seroreactivity over time may indicate that HPV 1 is usually eliminated from the host after infection or that inadequate levels of HPV 1 capsid antigen are produced during latent foot warts to maintain antibody levels.