The present study characterizes the relationship between the levels of eCG, ovarian morphology, resumption of cyclicity, and fertility in postaborted embryo transfer recipient mares. A total of 32 pregnant recipient mares carrying a male fetus were aborted at approximately 65 days of gestation by single transcervical administration of cloprostenol. In addition, 25 gestation age-matched mares were used as nonaborted controls. The concentration of progesterone, but not of eCG, differed significantly between controls and aborted mares 48 hours after abortion. Of treated mares, 84.4% (27 of 32) expelled the fetus within 48 hours of treatment. The eCG concentration and the number of supplementary luteal structures were lower in mares aborted in November (equivalent to May in Northern Hemisphere) than in January. A total of 6.2%, 37.5%, and 56.2% of the mares entered anestrus, ovulated normally, and had 1 to 2 consecutive anovulatory cycles, respectively. The mean interval from abortion to the first ovulation was 28.5 ± 3.3 days (range, 5-65 days). The correlation between the levels of eCG at abortion and the interval to the first ovulation was poor (r = 0.38; P = 0.03). Of aborted mares, 90% (18 of 20) were reused and became pregnant after embryo transfer at a mean of 57.6 ± 4.4 days after abortion (range, 19-103 days) and eCG concentration of 0.9 ± 0.3 IU/mL (range, 0.1-3.6 IU/mL). In conclusion, the levels of eCG at the time of abortion were extremely variable and did not correlate well with the number of luteal structures or the interval from abortion to the first ovulation.