Hourly pulses of gonadotropin-releasing hormone (GnRH) or bi-daily injections of estradiol (E2) can increase luteinizing hormone (LH) secretion in ovariectomized, anestrous pony mares. However, the site (pituitary versus hypothalamus) of positive feedback of estradiol on gonadotropin secretion has not been described in mares. Thus, one of our objectives involved investigating the feedback of estradiol on the pituitary. The second objective consisted of determining if hourly pulses of GnRH could re-establish physiological LH and FSH concentrations after pituitary stalk-section (PSS), and the third objective was to describe the declining time trends of LH and FSH secretion after PSS. During summer months, ovariectomized pony mares were divided into three groups: Group 1 (control, n = 2), Group 2 (pulsatile GnRH (25 micrograms/hr), n = 3), and Group 3 (estradiol (5 mg/12 hr), n = 3). All mares were stalk-sectioned and treatment begun immediately after stalk-section. Blood samples were collected every 30 min for 8 h on the day before surgery (D0) and 5 d post surgery (D5) to facilitate the comparison of gonadotropin levels before and after pituitary stalk-section. Additionally, jugular blood samples were collected every 12 hr beginning the evening of surgery, allowing for evaluation of the gonadotropin secretory time trends over the 10 d of treatment. On Day 10, animals were euthanized to confirm pituitary stalk-section and to submit tissue for messenger RNA analysis (parallel study). Plasma samples were assayed for LH and FSH by RIA. Mean LH secretion decreased from Day 0 to Day 5 in Groups 1 and 3, whereas LH secretion tended (P < 0.08) to decrease in Group 2 mares. On Day 5, LH was higher (P < 0.01) in Group 2 (17.26 +/- 3.68 ng/ml: LSMEANS = SEM), than either Group 1 (2.65 +/- 4.64 ng/ml) or Group 3 (4.28 +/- 3.68 ng/ml). Group 1 did not differ from Group 3 on Day 5 (P < 0.40). Similarly, mean FSH levels decreased in all groups after surgery, yet Group 2 mares had significantly (P < 0.001) higher FSH concentrations (17.66 +/- 1.53 ng/ml) than Group 1 or Group 3 (8.34 +/- 1.84 and 7.69 +/- 1.63 ng/ml, respectively). Regression analysis of bi-daily LH and FSH levels indicated that the time trends were not parallel. These findings indicate: 1) Pituitary stalk-section lowered LH and FSH to undetectable levels within 5 d after surgery. 2) pulsatile administration of GnRH (25 micrograms/hr) maintained LH and FSH secretion, although concentrations tended to be lower than on Day 0, and 3) E2 did not stimulate LH or FSH secretion.