The specificity of commercially available hCG and hCG-beta antibodies (anti-hCG-beta from N.I.A.M.D.D. (SB6), Serono (S), Biosigma (B), Union Carbide (UC) (I.R.E.) and anti-hCG from Union Carbide) were compared. Using 125I-hCG (CR 115), the crossreactivity with LH (LER 960) was 5.5% for SB6, 0.25% for S and 0.3% for B. In the homologous hCG system UC, crossreactivity was 0.06% with hCG-beta and 1.2% with LH, in the hCG-beta system UC, it was 2.2% with hCG and 0.01% with LH. Parallel standard curves for hCG, the 2nd I.R.P. hCG and hCG-beta were found exclusively with the 125I-hCG, anti-hCG-beta system S. Consequently accurate estimates of the total hCG or hCG-beta content in serum and standardization with the 2nd I.R.P. is possible. This system is the most useful for clinical purposes. The serum half-life of hCG was calculated in 10 pregnant women after delivery and was found to be 10 to 34 h. AFP half-life time, determined by the double antibody radioimmunoassay was 4.0 +/- 1.8 (+/- SD) days in pregnant women after delivery (n = 60) and 3.8 +/- 0.9 days in neonates during the first 16 days of life (n = 26). Altogether, apparent half-life determinations were carried out in 29 patients with nonseminomatous germ cell tumors. The results indicate that half-life determinations of hCG and AFP may have a predictive value with respect to the monitoring of therapy.