Our analysis of the results of this study show that the two proposed treatments (eCG Treatment, CIDR® for 8 days + EB D0 + 700UI eCG D4 + cloprostenol D8; and FSH Treatment, CIDR® for 7.5 days + EB D0 + 130 mg FSH D4.5/D8 + cloprostenol D7.5) were capable of promoting a superovulatory response; however, eCG Treatment produced the greatest number of corpora lutea, while FSH Treatment promoted only a minimal response (2.6 ± 0.8 CL), sufficient to be considered superovulated.
The best mean ovulation obtained in eCG Treatment (7.0 ± 1.8 CL) was similar to the means obtained in the species D. dama (6 to 10 CL [11, 13, 18]) and C. elaphus, which have already been selected for MOET programs (mean of 5 to 12 CL [10–13,15-17 ]). Moreover, this mean was higher than the ovulation means obtained for other species, such as E. davidianus (CL 3.83 ), O. virginianus (3.10 CL ) and C. elaphus, which were not selected for a MOET program (2.15 CL ). Higher mean ovulation (3.40 ± 0.68 CL) has already been obtained for the species M. gouazoubira, in a previous experiment which used similar SOV protocols with modifications in the EB dosage (0.5 mg) used to induce ovulatory follicular atresia and synchronize the follicle wave in conjunction with the CIDR® and eCG (600UI) used to induce SOV .
These two modifications in the protocol could begin to explain the differences between the ovulatory rates obtained in eCG Treatment and in the previous study with the same species. The use of a lower dose of EB and a higher dose of eCG seem to have been a positive factor. It has already been shown that high levels of estradiol at the moment of insertion of the progesterone implant and early administration of gonadotropins have a negative impact on superovulatory response and on follicular wave synchronization [23, 42] and that the ovulation rate increases in a dose-dependent manner  in relation to an increase in the eCG dosage . However, even using a lower dose of EB, it is probable the protocol is not yet well adjusted to the needs of this species, since the hinds presented behavioral estrus following the administration of EB. Another factor that could explain the lower ovulation rate obtained in the previous study are changes in luteinizing hormone (LH) pulsatility caused by stress induced by manual physical restraint  during the hormone applications, which in this present study were substituted by leading the deer to a restraint box.
Although considered low, the ovulation rate obtained in FSH Treatment was three times higher than the rate obtained in a previous study (0.80 ± 12:49 CL) involving the species M. gouazoubira, which used 250 IU FSH (Pluset®, Calier®, Spain) administered in a single application, together with a synthetic organic polymer (polyvinylpyrrolidone), used in the processing of long-acting drugs . The regime of multiple FSH applications seems to be the most suitable for stimulating SOV in cervids; however, despite the way it was conducted, it still seems inadequate to provide good ovulatory stimulus, since the total follicular stimulation of eCG and FSH Treatments were similar, but the ovulation rate of eCG Treatment was approximately 2.5 times greater than that for FSH Treatment.
In general, the system of multiple FSH applications is influenced by the effects of stress management [24, 25], as suggested by the significant increase in FGM concentrations after day 4 of FSH Treatment, when the FSH applications were initiated. The FGM levels increased during period of intensive management of the deer, as previously reported for Gazella dama mhorr. Although the FGM levels of FSH Treatment were higher, no significant difference was verified between the treatments, which may indicate that only one application of eCG on day 4 was sufficient to increase FGM concentrations and that the use of eCG provided extra gonadotrophic stimulation for ovulation  in eCG Treatment, as reported in studies with other deer species that combine low doses of eCG at the end of treatments with FSH  due to the lack of SOV response when FSH was used alone . The reinforcement of eCG on the preovulatory peak in eCG Treatment can also be inferred from shorter synchrony period of this treatment, suggesting that stimulation of ovulation of a greater number of follicles occurred in a shorter period of time .
The use of eCG has been associated with the occurrence of asynchronous follicular development and anovulatory follicles in numbers proportional to the doses used . However, the occurrence of anovulatory follicles in this study does not seem to be exclusively related to the use of eCG, since FSH treatment also resulted in numerous anovulatory follicles. The high levels of FGM during the follicular growth phase should also be considered as an important factor in the occurrence of these follicles, due to endocrine disorders caused by the release of corticosteroids  and/or progesterone production by the adrenal gland , which can inhibit the LH peak and/or affect oocyte competence , resulting in lower receptor acquisition and, consequently, lower responsiveness to LH .
Other factors, including the high concentrations of estradiol secreted by anovulatory follicles, can also significantly reduce the rates of ovulation and total ovarian stimulation, as well as promoting premature regression of approximately 10% of the CL, similar to that previously reported for C. elaphus[8, 16, 49] and other domestic species [21, 50]. Although both treatments resulted in a large number of anovulatory follicles, in eCG Treatment, four hinds (66.7%) presented repetition of behavioral estrus and signs of premature regression of CL, whereas in FSH Treatment, only two hinds (33.3%) showed the same signs, suggesting that premature regression is more closely associated with the use of eCG and its prolonged action than the mere occurrence of anovulatory follicles.
In addition to these factors, it is important to consider that the ovulation rates obtained in this study may have been influenced by the administration of gonadotropins at an inappropriate time in the follicular wave, since the first study that induced follicular wave emergence in cervids was developed in C. elaphus, during anestrus, and demonstrated the onset of a new wave 5.2 ± 0.2 days following the administration of progesterone and estradiol-17β . Moreover, the use of an ovulation inductor could have improved the ovulation rates and should be considered in future protocols.
The profiles of FPM concentration during the luteal phase associated with the surgical view of corpora lutea revealed a luteal source of secretion following by ovulation in both treatments. The increase in the ovulation rate was associated with increased progesterone secretion as a result of the significant increase in luteal tissue mass [7, 13]. The secretion of progesterone during this period was apparently not influenced by the secretion of glucocorticoids in the final phase of the treatments, as confirmed by the correlation value obtained and as previously reported by González et al. . However, this result could have been influenced by the production of progesterone in the adrenal gland  and/or individual variations among the small sample of deer used in the study.
As previously reported for the species M. gouazoubira, administration of cloprostenol 14 days after the completion of treatment resulted in rapid destruction of luteal tissue and a return to baseline concentrations of progesterone within 72 hours. This fact is relevant in embryo transfer programs to prevent the implantation of embryos that may remain following collection and to resynchronize the females for natural mating .