The neuroendocrine control of prolactin synthesis and secretion before birth is not well understood. We have measured the changes in the level of prolactin mRNA in the anterior pituitary of the fetal sheep throughout the last 15 days of pregnancy (term = 147 +/- 3 days gestation). We have also investigated the effects of surgical disconnection of the fetal hypothalamus and pituitary (HPD) with or without long term cortisol infusion on pituitary prolactin mRNA levels and plasma prolactin concentrations in the late gestation sheep fetus. Prolactin mRNA levels were measured in anterior pituitaries collected from a series of fetal sheep (130-134 days, n = 6; 135-140 days, n = 6; 141-145 days, n = 6) in late gestation. HPD was carried out in ten fetal sheep at 105-115 days gestation and five intact fetal sheep were used as controls. In the HPD group, either saline (HPD + saline group, n = 5) or cortisol was infused (3.5 mg/24 h) for 5 days from 134-136 days gestation (HPD + cortisol group, n = 5). There was an increase in the ratio of prolactin mRNA: 18S rRNA in the fetal pituitary between 130-134 days (0.46 +/- 0.08, n = 6) and 135-140 days (1.27 +/- 0.17 n = 6) which was maintained after 141 days gestation, (1.27 +/- 0.11, n = 6). The mean prolactin mRNA: 18 S rRNA ratio was significantly higher (P < 0.05) in intact fetal sheep (1.41 +/- 0.16, n = 4) than in the HPD fetal sheep after either saline (0.54 +/- 0.14, n = 4) or cortisol (0.74 +/- 0.24, n = 5) administration. The mean plasma concentration of prolactin was also higher in the intact group (28.3 +/- 3.9 ng/ml) when compared with the HPD + saline group (8.0 +/- 3.3 ng/ml) or the HPD + cortisol group (5.6 +/- 1.9 ng/ml). We have demonstrated that there is a strong hypothalamic drive to prolactin synthesis and secretion in the fetus and that cortisol does not act directly at the fetal pituitary to stimulate prolactin synthesis and secretion in late gestation.