Several investigators (Tsafriri, Koch & Lindner, 1973;Sato, Taya, Jyujo & Igarashi, 1974) have reported that a single injection of indomethacin, an inhibitor of prostaglandin (PGs) synthesis, does not block luteinizing hormone (LH) release but exerts its anti-ovulatory action directly on the ovary. However, it has been shown (Harms, Ojeda & McCann, 1973) that PGs have stimulatory effects at various levels of the hypothalamic-pituitary system.Adult female rats of the Holtzman strain were used 5 weeks after ovariectomy. Serum or plasma, and pituitary levels of LH were determined using a radioimmuno¬ assay kit (NIAMD-NIH). In Expt 1, 3 mg indomethacin (Sumitomo Kagaku Pharmaceutical Co., Osaka) were injected s.c. in 0-2 ml 15 % gelatin, with or without PGF2ct (Ono Pharmaceutical Co., Osaka) (0-8 mg/rat) daily for 5 days and blood samples were collected 24 h after the last injection. In Expt 2, indomethacin was implanted into the median eminence (ME) of the rats. The cannulae for indomethacin implants were prepared according to the method of Corbin & Cohen (1966), by stand¬ ing 20 gauge stainless steel tubes in a melted mixture of indomethacin (50 mg) and cocoa butter (0-5 g) at 38°C. Seven days after implantation, the rats were anaes¬ thetized with ether and blood samples were collected. The anterior pituitary was homogenized in saline, extracted overnight at 4°C , centrifuged and the supernatant used for LH assay. Localization of the implants was confirmed macroscopically and microscopically. In Expt 3, 100/¿g oestradiol benzoate and 10 mg progesterone (Schering AG Berlin) were injected s.c. in 0-2 ml sesame oil on the morning of day 0. Indomethacin injections (2 mg in 0-2 ml gelatin) were then given daily to 12 rats for 3 days including day 0. Control injections of gelatin alone were given to a further 12 rats. Twenty-four hours after the last injection, 100 ng LH-releasing hormone (Daiichi Pharmaceutical Co., Tokyo) (LH-RH) were injected into the jugular vein of each animal except for six of the gelatin-control rats, which received 1 ml isotonic saline instead. Blood samples (1 ml) were obtained from the jugular vein immediately before and 10 min after the injection of LH-RH. In Expt 4, indomethacin (2 mg in 0-2 ml gelatin) was injected s.c, daily for 4 days. Twenty-four hours after the last injection, 50 µg (in 5 µ\) PGE2 (prepared by dissolving in 95 % ethanol followed by addition of 9 parts 0-02 % Na2C03 to one part solution) were injected into the third ventricle of animals anaesthetized with ether (Hedge, 1971). Injection was made with