Estradiol-17 beta (E2) and progesterone (P) levels have been effectively reduced in pregnant rats by injecting them with antibodies which bind E2 or P (A-E2- or A-P) with high specificity. Before, and at daily intervals after treatment with A-E2 or A-P, blood was collected sequentially from the tail vein and the levels of circulating A-E2, A-P total E2 (E2t) and total P (Pt) determined. Having established the relationship between A-E2, E2t, bound E2 (E2b) and unbound E2 (E2u), as well as between A-P, Pt, bound P (Pb) and unbound P (Pu), the concentrations of E2u and Pu could be calculated reliably. Treatment of gestating rats with A-E2 and A-P lowered E2u and Pu levels significantly (P less than 0.001). In comparison with the 14 controls, the 17 A-E2 treated (and thus E2u-deficient) animals showed significant increases in placental weight (P less than 0.001). This effect of A-E2 was readily prevented by replacement therapy with diethylstilbestrol (DES), since A-E2 does not bind DES. In 21 pregnant rats, treatment at day 10 of gestation with only 2 ml A-P (total "Binding Capacity" = 6 mug P) critically lowered the Pu levels and provoked abortion. In contrast, the same treatment in 12 rats at day 6 of gestation (when the P levels are higher than at day 10) failed to provoke the appropriate degree of P-withdrawal (Pw) and abortion, illustrating the quantitative nature of the relationship between effective A-P treatment, Pw and the termination of pregnancy.