In order to study the association between seropositivity against human papillomavirus-type-16 capsids and CIN 11-111 in the general population in ages at which high-grade cervical dysplasia arises, 90 cases and 216 controls participating in a population-based case-control study of incident CIN 11-111, were analyzed for the presence of HPV antibodies, HPV DNA and for the influence of behavioral factors. A significantly higher proportion of cases than controls were seropositive. Of HPV-16-DNApositive cases and controls, 42 and 14% respectively were seropositive. A similar proportion of seropositivity was found among the I72 cytologically normal, HPV-DNA-negative controls. However, seropositivity was closely linked to the sexual history of the women. Logistic-regression analyses, adjusting for sexual behavior, smoking history and educational level, revealed that CIN 11-111 was associated with HPV-16 seropositivity and with HPV DNA. Controlling for the presence of HPV DNA indicated that antibodies were not independently associated with CIN. The low correlation between the presence of HPV antibodies and DNA, the finding that the association between seropositivity and CIN depended on the presence of HPV DNA, and the association of seropositivity with sexual histoq, may be explained by serology detecting both past and present persistent infections and presence of HPV DNA, reflecting mostly transient infections in controls and persistent infections in