In order to study the biological effects of estriol in women 20 mg estriol was administered daily to 7 young women. Plasma luteinizing hormone (LH), estradiol (E2), progesterone (Pg) and prolactin (Prl) were measured during a treatment and a control cycle every second or third day. Further 3, 6 or 20 mg estriol was administered in a single dose to 5 women and plasma Prl, unconjugated and conjugated estriol (E3) measured over 24 h at 2-3 intervals. In 2 experiments with 20 mg E3, blood samples were taken more frequently, over 6 h. When 20 mg E2 was administered daily, 2 of the 7 young women had anovulatory cycles. The mean plasma E2 was lower during the follicular and ovulatory phases (P less than 0.025) and mean plasma LH was higher (P less than 0.005) during the luteal phase, when E3 was given. Because of the 2 anovulatory cycles the mean Pg value during the luteal phase was lower (P less than 0.05) during treatment. There was a slight decrease in mean Prl in 5 out of 6 women (P less than 0.0005), but in only 1 woman was this decrease substantial (from a mean value of 27.6-18.9 ng/ml; P less than 0.01). When 6 or 20 mg E3 was administered orally in the morning a significant negative correlation (P less than 0.01) between plasma Prl and unconjugated E3 was found. The correlation coefficient was highest (r = -0.74) with 6 mg E3. When 3 mg was administered no obvious effect on Prl section was seen. However, when results from all experiments with identical time schedules were pooled (two with 3 mg, two with 6 mg and one with 20 mg E3) and the mean values for plasma Prl calculated and compared with the mean values obtained in 7 control experiments, it was found that E3 administration in the morning almost abolishes the Prl rise during the following night. There was a statistically significant (P less than 0.0125) decrease in the difference between the maximum value during the night and the minimum value during the day. The minimum value was significantly higher (P less than 0.01) and the maximum value significantly lower (P less than 0.025) after E3 treatment, compared to the control values. It is concluded that long-term administration of 20 mg E3 usually has only a slight but significant decreasing effect on mean plasma Prl concentration measured in the morning, before the next dose is taken.(ABSTRACT TRUNCATED AT 400 WORDS)