Molecular studies suggest that the simian polyomavirus SV40 is present in the human population, possibly introduced in contaminated polio vaccine. However, no recent seroepidemiological data exist in England on SV40 or on the two human polyomaviruses, BKV and JCV. A comparative age seroprevalence study was undertaken on 2,435 residual sera from 1991 by haemagglutination inhibition (HI) for BKV and JCV, and virus neutralisation for SV40. The overall rates of seropositivity for BKV and JCV were 81% and 35%, respectively, and each was significantly related to age (P < 0.001). BKV seroprevalence reached 91% at 5-9 years of age, but JCV seroprevalence reached only 50% by age 60-69 years. There was a highly significant association between BKV antibody titre and age (P < 0.001), titres decreasing linearly at a rate of 8.7% per 10 years (95% CI = 7.4-10% drop). Significantly more males than females had antibody to JCV (P = 0.013). In individuals under 40 years of age there was a significant negative association between the presence of antibody to BKV and JCV (P < 0.001). By contrast, the antibody prevalence to SV40 remained at 1.3-5% throughout all age groups and titres were low. There was a significant positive association between the presence of antibody to SV40 and antibody to both BKV (P < 0.001) and JCV (P = 0.009), and also to the geometric mean titre (GMT) of BKV antibody (P = 0.011). The results indicate that BKV and JCV are transmitted by different routes. There is no serological evidence that SV40 entered the human population during the past 80 years, and the possibility of cross-reaction with BKV or JCV antibody must be considered.