There is a well-documented increase in luteolytic failure, resulting in spontaneously prolonged corpus luteum (SPCL) function, during estrous cycles of horses in autumn. The cause of this phenomenon may be due to seasonal alterations in PGF(2alpha) and/or in prolactin (PRL) secretion around luteolysis. To investigate this, progesterone (P4), 13, 14-dihydro, 15-keto PGF(2alpha) (PGFM) and PRL concentrations were compared between summer and autumn estrous cycles during natural luteolysis and luteolysis induced by benign uterine stimulation. A single estrous cycle from mares in June-July (n=12) was compared to multiple estrous cycles from these 12 mares plus 8 additional mares in September through December. Reproductive behavior was monitored by bringing a stallion in close proximity to the mare and ovarian events by ultrasonography. Blood was collected via jugular cannula every 6h from d 13 to 17 post-ovulation in untreated control mares (n=8 summer, n=9 autumn). In treated mares, blood collection occurred at 0, 15, 30, 45, 60, 90, 120, 180 and 240min followed by 6h intervals for a total of 5d following intrauterine saline infusion on d 7 (n=4 summer, n=11 autumn). Mares failing to return to estrus for 30d received intrauterine saline and the described intensive blood sampling protocol on d 30. Progesterone and PRL were determined on daily samples and PGFM on frequent plasma collections by RIA. Duration of ovarian luteal and follicular phases, P4 and PRL concentrations and PGFM secretion around luteolysis were compared between treatments and seasons by ANOVA. Mean P4 declined from June to December in all groups. Pulses of PGFM were detected on d 13-17 in controls and d 7-11 in saline-infused mares. Pulse patterns were not different between groups. The incidence of SPCL increased during autumn in the control group. PGFM pulses were absent on d 13-17 in mares with SPCL, but PGFM pulses could be induced in these mares by saline infusion at d 30. Autumn PGFM profiles were unchanged during spontaneous or saline-induced luteolysis compared with summer. Circulating PRL increased around natural or induced luteolysis. These results provide evidence that changes in luteal function during the autumn transition are not the result of alterations in the ability of the uterus to produce PGF(2alpha) nor due to changed CL sensitivity to PGF(2alpha). We conclude that seasonal changes in luteolytic function are caused by an alteration in the signal for PGF(2alpha) release.