The goal of this study was to determine the effects of short-term feed withdrawal on reproductive and metabolic hormones during the luteal phase of the estrous cycle in mature ewes. Mature ewes observed in estrus were assigned randomly to control and fasted groups (n = 10 per group Trials 1 and 2). For Trials 1 and 2, control ewes had ad libitum access to feed, whereas fasted ewes were not fed from d 7 through 11 of their estrous cycle; on d 12, all ewes were treated with 10 mg of PGF(2alpha) and fasted ewes were gvien ad libitum access to feed. For Trial 1, blood samples were collected daily through fasting and at 2-h intervals following PGF(2alpha) for 72 h. Serum concentrations of insulin (P less than or equal to 0.002) and IGF-I (P less than or equal to 0.01), but not GH (P greater than or equal to 0.60), were decreased during fasting compared with fed ewes. Serum concentrations of 29 (P = 0.02) and 34 kDa (P = 0.04) IGFBP were greater in fasted ewes at 96 h after initiation of fasting than in control ewes. Two control and four fasted ewes in Trial 1 did not exhibit a preovulatory surge release of LH by 72 h. Therefore, Trial 2 was conducted so that the timing of the LH surge could be predicted following the collection of blood samples at 2-h intervals for 112 h and then at 6-h intervals until 178 h following PGF(2alpha) administration and realimentation. The magnitude of the preovulatory LH surge in Trial 2 was decreased (P = 0.009) and delayed (P = 0.04), and serum concentrations of estradiol were diminished (P less than or equal to 0.03) 12 h before the LH surge in fasted ewes. Ovulation rates were not influenced (P greater than or equal to 0.32) by fasting in Trials 1 and 2. Serum concentrations of progesterone in both Trials 1 and 2 were, however, greater (P < 0.001) in fasted than in control ewes. A third trial with ovariectomized ewes was conducted to determine whether the increased serum concentrations of progesterone observed in fasted ewes during Trials I and 2 were ovarian-derived. Ovariectomized ewes were implanted with progesterone-containing intravaginal implants and allotted to control (n = 5) or fasted (n = 5) treatment groups and fed as described for Trials 1 and 2. Similar to intact ewes, serum concentrations of progesterone were approximately twofold greater (P < 0.001) in fasted than in control implanted ovariectomized ewes. In summary, feed withdrawal for 5 d during the luteal phase of the estrous cycle increased serum concentrations of progesterone and evoked endocrine changes that could perturb the subsequent estrous cycle.