Potential limitations of prenatal diagnosis of hemophilia B, as compared to hemophilia A, include (1) occurrence of far more frequent defects with abnormal circulating antigen, (2) lower levels of factor IX in fetal plasma at 16 to 20 weeks gestation, and (3) the presence of factor IX antigen in amniotic fluid. In addition, proteolysis could occur, especially with amniotic fluid contamination of fetal plasma. A sensitive polyclonal immunoradiometric assay for factor IX antigen was used to characterize the range of levels in amniotic fluids and fetal plasma samples. To assess for altered forms, factor IX species were compared to those of a homologous clotting factor, prothrombin. Fourteen postmortem abortus blood samples from fetuses of 14 to 23 weeks gestation had factor IX antigen levels that averaged 5.1 U/dL and ranged from 1.7 to 15 U/dL. Amniotic fluid factor IX antigen averaged 2.9 U/dL, with a range from 1.4 to 8.5 U/dL in 19 separate amniocentesis samples. Thus, in a male fetus at risk of hemophilia B and with a low circulating level of gene product, mixture of fetal plasma with amniotic fluid could severely limit prenatal diagnosis, assuming that the amniotic fluid factor IX is of maternal origin. Despite rapid processing of amniotic fluid samples, the prothrombin was extensively cleaved, suggesting that it had been activated in vivo. On gel electrophoresis of amniotic fluid samples, however, factor IX was only minimally cleaved. In the postmortem fetal blood specimens, prothrombin was partially cleaved. On crossed-immunoelectrophoresis, fetal plasma prothrombin showed decreased migration in calcium, compared to EDTA, indicative of mature gamma-glutamyl carboxylation. The latter presumably resulted from fetal hepatic synthesis.