1176

BIOLOGY OF REPRODUCTION 64, 1176–1182 (2001)

Annual Ovarian Cycles in an Aseasonal Breeder, the Springbok(Antidorcas marsupialis)1

Donal C. Skinner,2,3 Trevor A. Richter,4 Benoit Malpaux,5 and John D. Skinner4

Department of Clinical Veterinary Science,3 University of Bristol, Langford House,Langford BS40 5DU, United KingdomVeterinary Wildlife Unit,4 Faculty of Veterinary Science, Onderstepoort 0110, South AfricaInstitut National de la Recherche Agronomique,5 Physiologie de la Reproduction des Mammiferes Domestiques,37380 Nouzilly, France

ABSTRACT

The springbok is an arid-adapted antelope inhabiting the de-sert and semidesert regions of southern Africa. Because it thrivesin these sparsely vegetated areas, the springbok is of potentialagricultural importance and the prospect of domestication hasbeen speculated for many years. However, apart from observa-tional studies on its breeding in the wild, suggesting it is anaseasonal breeder, little is known about the underlying repro-ductive endocrinology of this species. In this study, biweeklyperipheral blood samples were collected from eight captivespringbok ewes from October 1995 until September 1998 andanalyzed for progesterone. At the start of the study, six eweswere prepubertal and cycling commenced spontaneously be-tween November 1995 and June 1996. Cycling had alreadycommenced in two ewes. At the end of November 1996, estrouscycles ceased abruptly in all ewes and restarted in April 1997.Cycling ceased again between December 1997 and February1998 and restarted in June 1998 in six ewes; there was no ces-sation of estrous cycles in two ewes. Thus, although some indi-viduals cycle continuously, there is a clear endocrine anestrusof between 4 and 5 mo in springbok, the timing and durationof which is synchronized between some individuals but the timeof onset and cessation is variable from year to year. To ensurethat the fluctuations we observed in progesterone levels werereliable indicators of changes in the estrous cycle, blood sampleswere collected every 6 h for 16 days in August 1998. A surge inLH secretion was observed in all ewes 55 6 5 h after the fall inprogesterone. Progesterone levels increased again 45 6 8 h afterthe surge. A final study showed that the pattern of melatoninrelease in springbok exhibits a normal day/night profile, andthus photoperiodic information is transformed into an endocrinecode to springbok but does not appear to affect reproduction.Rather, our data raise the possibility that the prevailing ambienttemperature may influence the onset of ovarian activity in thisspecies.

ovulatory cycle, pineal, seasonal reproduction

INTRODUCTION

The springbok, Antidorcas marsupialis, is an arid-adapt-ed gazelle confined to southwestern Africa [1]. In commonwith most wild antelope, little is known about reproductionin this species, not least because of the harsh environment

1Funded by the National Research Foundation, South Africa.2Correspondence. FAX: 44 117 928 9582;e-mail: donal.c.skinner@bristol.ac.uk

Received: 15 September 2000.First decision: 10 October 2000.Accepted: 16 November 2000.Q 2000 by the Society for the Study of Reproduction, Inc.ISSN: 0006-3363. http://www.biolreprod.org

it inhabits and its tendency to disperse, making prolongedobservations in the wild impractical. Pressures exerted onanimals inhabiting arid environments are extreme and anyincrement can lead to extinction in the wild, as occurredwith the mhorr gazelle, Gazella dama mhorr [2], the Ara-bian oryx, Oryx leucoryx [3], and the scimitar-horned oryx,Oryx dammah [4]. Indeed, ex situ breeding programs havebeen initiated on several arid-environment inhabiting spe-cies, although basic data remain limited and, to our knowl-edge, nothing has been published on reproductive cycles ingazelles. Not surprisingly, the optimal periods for manip-ulating fertility remain largely unidentified and artificial in-semination programs have met with limited success [4]; inspringbok, a single reported artificial insemination attemptwas unsuccessful [5] and, in the mhorr gazelle, 4 of 13artificially inseminated ewes became pregnant, althoughonly 1 carried the fetus to term [6]. Understanding the re-productive endocrine profile will improve our ability toconserve a species in the wild and, with an increased com-mercial market for venison, particularly from springbok,knowledge of reproduction will also assist in their hus-bandry.

To our knowledge, only a single other long-term studyover more than a year on individual wild antelope has beenreported [7] and most studies concluding that a speciesbreeds aseasonally have relied on the distribution of births.In the wild, it is probable that a ewe will conceive at thefirst cycle, and thus it is also not possible to establishwhether individuals or populations have periods of repro-ductive quiescence or, in the event of nonconception,whether estrous cycles continue unabated throughout theyear. For this, it is essential to carry out studies in the ab-sence of males and, to allow for repetitive sampling of spe-cific individuals, in captivity. Thus, the first study investi-gated the duration and occurrence of estrous cycles in acaptive population of springbok over 30 mo. We hypothe-sized that as in the more equatorially located dik-dik, Ma-doqua guentheri [7], and sable antelope, Hippotragus niger[8], these springbok would either remain reproductively ac-tive throughout the year or would exhibit unsynchronizedperiods of reproductive activity with reproductive quies-cence, as found in some sheep, Ovis aries, on a fixed pho-toperiod [9]. Either of these hypotheses would explain thefield observations of lambs being born throughout the year(i.e., aseasonality).

There is considerable evidence that photoperiodic infor-mation is the primary proximate factor determining the pe-riods of sexual activity and quiescence in seasonal breeders(see reviews [10, 11]). In the sheep, photoperiodic infor-mation is transduced to the reproductive neuroendocrineaxis by the nocturnal increase in the pineal hormone, mel-

1177ANNUAL ESTROUS CYCLICITY IN SPRINGBOK

atonin. In an aseasonal species it is possible that melatoninsecretion is impaired and that daylight, therefore, does notget transformed into a biochemical signal. To address thisquestion, a further study investigated whether springbok se-creted melatonin in accordance with a normal day (lowmelatonin):night (high melatonin) profile.

MATERIALS AND METHODS

Animals and Husbandry

Juvenile springbok ewes (2–3 mo; n 5 8) were capturedat the same time near Hutchinson, South Africa, and heldin a paddock and hand-reared on cow’s milk for 3 mo be-fore being weaned onto alfalfa. At approximately 6 mo ofage, the springbok were transferred to the experimentalfarm of the University of Pretoria, Pretoria (258S, 288E) ina 500 m2 enclosure with 3-m-high sides. They were feddaily on alfalfa, cottonseed, and antelope pellets (Epol Ltd,Pretoria, South Africa). This diet was supplemented withfreshly cut Salix mucronata, Morusmesozygia leaf, or both,and mineral licks (Kynoch, South Africa). Water was avail-able ad libitum.

All the springbok were released at the end of the studyonto a game farm and have since bred successfully.

Blood Sampling

For collection of blood samples, a collapsible chutewithin the enclosure was opened and the springbok wereconfined into a small space where they were manually re-strained. Blood was taken by venipuncture from the jugularvein and collected into heparinized venoject tubes (BectonDickinson, Meylan Cedex, France) and kept on ice for 30min until centrifuged. Plasma was harvested and frozen at2208C for later radioimmunoassay analysis. In the firststudy (November 1995–July 1998), sampling was under-taken biweekly between 0800 and 1000 h. To ensure thata preovulatory LH surge was occurring, blood sampleswere collected every 6 h for 16 days and to investigatewhether springbok expressed a nocturnal melatonin incre-ment, blood samples were collected every 4 h and every 2h at dusk and dawn.

It took approximately 20–25 min to collect samples fromall eight springbok.

Radioimmunoassays

Tests for parallelism between springbok plasma and therespective assay standards were carried out on all assaysprior to analysis. Specifically, serial dilutions of sampleswith elevated hormone concentrations yielded curves par-allel to those obtained with hormone standards.

The progesterone level in biweekly plasma samples wasestablished using an immunologic radioimmunoassay [12],which revealed whether the plasma sample progesteroneconcentration was above 1 ng/ml. This assay, which hasbeen used extensively in the sheep, goat, and cow is semi-quantitative and several quality controls, particularlyaround 1 ng/ml, were used. We verified that the distributionof counts per minute (cpm) for all springbok plasma sam-ples was bimodal (populations of samples with low value5 high progesterone, and high value 5 low progesterone)and that the cpm value corresponding to the 1 ng/ml qualitycontrol allowed us to discriminate between these two pop-ulations. To eliminate intraindividual variation with time,all samples from a ewe were analyzed in the same assay.

For the more frequent blood sampling, progesterone lev-

els were estimated in a single immunometric radioimmu-noassay [13]. The intraassay coefficient of variation was10% and assay sensitivity was 0.05 ng/ml. The standard inboth progesterone assays was pure progesterone (SteraloidsInc., Newport, RI).

The 6-h blood samples were assayed for LH in duplicate100-ml aliquots of plasma using the radioimmunoassaymethod of Pelletier et al. [14] as modified by Montgomeryet al. [15], and all samples were tested in a single assay.The intraassay coefficient of variation averaged 6% andsensitivity was 0.15 ng/ml of standard 1051-CY-LH (1 ng5 2.08 ng NIH-LH-S1; a gift of Y. Combarnous, INRA,France).

Jugular plasma melatonin concentrations were deter-mined in 100-ml aliquots using a single-well-validated ra-dioimmunoassay [16]. Assay sensitivity averaged 16 pg/mland the intra-assay coefficient of variation was 4%. Stan-dards were made from pure melatonin (Fluka, Saint-Quen-tin Fallavier, France).

Analysis

The immunologic assay was used to determine whethera ewe was in the follicular or luteal phase of the estrouscycle and also whether the ewe was in the breeding seasonor anestrus. Specifically, if progesterone concentrationswere below 1 ng/ml a ewe was noted to be in the follicularphase, but if concentrations were above 1 ng/ml for at leasttwo sequential samples the ewe was recorded to be in theluteal phase of the estrous cycle. The first sample to exceed1 ng/ml was recorded as the start of the luteal phase andthat ovulation had occurred. If progesterone levels re-mained basal for more than a month, the ewe was noted asanestrus. To estimate the mean duration of ovarian cyclesfor the group of springbok, the mean cycle duration foreach springbok was first calculated.

To determine the time when maximum and minimumambient temperatures changed significantly, the month inwhich temperature rose above the mean 11 SD of the pre-ceding month was marked as the onset of the increase andthe month in which temperature fell below the mean 21SD of the preceding month was marked as the onset of thedecrease.

RESULTS

Springbok exhibit a mixture between continuous repro-ductive cycling and periods of synchronized anestrus thatare not predictable from year to year. The occurrence ofluteal phases expressed by the eight individual springbokewes is depicted in Figure 1 and the annual change in thepercentage of ewes expressing estrous cycles, as detectedby changes in progesterone concentrations, in Figure 2.During the first year of the study, when springbok werebecoming sexually mature, seven had started to exhibit es-trous cycles by January 1995, which continued to June1996, whereupon the eighth also displayed ovarian activity.All ewes became quiescent between November 1996 andMarch 1997 (range, 124–161 days; mean, 140 6 5 days)and a synchronous onset in estrous cycles was noted be-tween 25 March 1997 and 22 April 1997. The period be-tween luteal phase onsets, which is equivalent to the lengthof the estrous cycle, averaged 16.4 1 0.4 days. In six ewes,cycling ceased again between 26 December 1997 and 27January 1998 (range, 73–179 days; mean, 132 6 15 days),but two ewes continued to cycle until the end of the study.

1178 SKINNER ET AL.

FIG. 1. Top: Progesterone values from two ewes between 1997 and 1998, which were revealed as either high (luteal phase) or low (follicular phaseor anestrus) using the radioimmunologic assay. The first elevated progesterone sample was taken as the start of the luteal phase. Bottom: A compositediagram of the detected luteal phases of all springbok ewes from October 1995 to September 1998. The horizontal dashed line indicates the 1st ofJanuary. The annual change in photoperiod is also shown.

The six anestrous ewes started to cycle again between 19May 1998 and 5 June 1998.

It is apparent that the fluctuations observed in circulatingprogesterone were associated with distinct preovulatory LHsurges (Fig. 3). A surge in LH secretion was observed inall ewes 55 6 5 h after the fall in progesterone. Progester-one levels increased again 45 6 8 h after the surge. Pho-toperiodic information is transformed into a biochemicalsignal in the springbok ewe as there was a clear differencein the release of melatonin during the day and night (Fig.4).

DISCUSSION

This study provides a definitive endocrinological de-scription of the annual reproductive cycle of the springbok,a species known to breed continuously throughout the year.Importantly, it reveals that there is a distinct anestrous pe-riod in most ewes but some individuals do not always ceaseovarian activity. It is arguable, therefore, that there is anelement of seasonal breeding in the springbok ewe. Like‘‘typical’’ seasonal breeders, such as sheep and goats, theseperiods of anestrus are significantly synchronized betweenindividuals but, in contrast, the time of year is not predict-able. This suggests strongly that some proximate factor,other than photoperiod, is determining estrous cyclicity inthe springbok ewe.

Our finding that most springbok exhibit an anestrous pe-riod agrees with studies on other species considered asea-sonal: the addax, Addax nasomaculatus [17] and scimitar-horned oryx [18]. Although the study on addax was muchshorter (late autumn to early spring), anovulatory periods

of between 39 and 131 days were detected in individualanimals. Some addax may have expressed periods of syn-chronized anestrus (possibly four of six), but the durationof the study did not allow for full analysis of this charac-teristic. More recently, the scimitar-horned oryx was foundto have a synchronized anestrus that ranged from 35 to 95days, but whether this anovulatory period was repeated be-tween years was not investigated [4]. In contrast to thesethree antelope species, no evidence for periods of repro-ductive anestrus was detected during a 3-yr study on dik-dik [7], nor during a study over 1 yr on sable antelope [8].Similarly, estrous cycles continue unabated in farmed chitaldeer (Axis axis) [19]. It is of interest that in the aseasonallybreeding sambar deer, Cervus unicolor, there is also an an-ovulatory period but, like springbok, some individuals cy-cle continuously [20].

In sheep, genetic parameters have been shown to affectthe timing of the onset of ovulatory activity after anestrus[21]. Thus, the interindividual differences that we have ob-served in springbok may be genetic, with some ewes ableto cycle continuously, whereas others are responsive tosome as yet unidentified, proximate factor. However, as sev-en ewes were cycling continuously at the start of the study,it is possible that all springbok are able to remain repro-ductively active throughout the period when anestrus mayoccur. This point should be viewed in context with the factthat because this first year encompassed the transition pe-riod between prepuberty and sexual maturity in most of ourspringbok, any pattern of breeding may not have emerged.Alternatively, it is possible that when springbok cycle con-tinuously, some ovarian factor dampens the expression of

1179ANNUAL ESTROUS CYCLICITY IN SPRINGBOK

FIG. 2. The mean monthly changes in annual photoperiod, minimum and maximum temperatures, and rainfall. The percentage of springbok cyclingthrough the duration of the study (dashed bars) are shown to facilitate comparison. The time of increment and decrement in the maximal (closed circles)and minimal (open circles) temperatures is highlighted.

FIG. 4. Mean (1 SEM) melatonin concentrations detected in jugularplasma over the course of two nights. The duration of the night is indi-cated by the black bar.

FIG. 3. Mean (1 SEM) LH levels standardized to the time of the peakLH concentration and the corresponding mean progesterone concentra-tion.

1180 SKINNER ET AL.

FIG. 5. The mean monthly changes in annual minimum and maximum temperatures, and rainfall at Nossob, in the Kalahari Desert, from 1994 to1998. The time of decrement in the maximal (closed circles) and minimal (open circles) temperatures is highlighted. The predicted time (if temperaturewas used as a proximate factor in timing reproductive activity) when ewes would start to give birth, is noted by arrows.

a reproductive rhythm. For example, the rhythm of somehormone, such as prolactin, in the ewes that are continu-ously cycling may still be in synchrony with the release ofthat hormone in the quiescent ewes [22, 23]. Due to thenature of the present study, we could not investigate pro-lactin changes as springbok were acutely stressed prior toblood sample collection.

It is apparent that the absence of a seasonally predictivereproductive pattern in springbok is not due to the inabilityof this species to secrete melatonin. However, whetherspringbok are able to interpret the melatonin signal remainsunknown. It is of interest that a similar interindividual dif-ference was noted in sheep subjected to a constant photo-period (12L:12D) for 4 yr, with some sheep expressing con-tinuous estrous cycles, whereas others had periods of re-productive activity and quiescence [9]. However, thosesheep with intermittent reproductive activity were not syn-chronized. Although the proximate factors responsible forthe commencement and cessation of springbok reproductiveactivity are not known, our data provide some evidenceabout the relative importance of specific factors. Becausethe diet remained unchanged for the duration of the study,this could not have contributed to the observed synchronyof cycles in the springbok. It is possible in the wild thatdiet may affect ovarian activity as previously suggested[24]. Rainfall may have been responsible for inducing thesynchronous onset of cycling in 1997, as all ewes startedto cycle immediately after an exceptionally high rainfall inMarch. However, no rain had fallen before a similar degreeof synchrony was evident in the six anestrous ewes in June1998. Rather, our study raises the possibility that the annualchange in temperature may play a critical role in initiatingthe onset of reproductive activity in the springbok. Specif-ically, both synchronous onsets in April 1997 and June1998 were coincident with a significant decline in both theminimum and maximum ambient temperature.

There are numerous studies on domestic livestock dem-onstrating that high ambient temperatures depress repro-ductive efficiency (e.g., [25–27]). In the natural habitat ofspringbok, the Kalahari, the mean maximal temperaturefrom November to March exceeds 358C, with the averageabove 39.58C in some months (source: South AfricanWeather Bureau). Indeed, one speculation is that an impor-tant ultimate factor determining when springbok enter the

breeding season is to avoid the extremely high summertemperatures in their natural habitat. There are, however,extremely few studies investigating ambient temperature asa proximate regulator of reproduction. This is not surprisingbecause such studies are difficult to conduct in the absenceof a suitable model because it is well established in sheepand goats that photoperiod is the primary proximate factoraffecting ovarian activity. Responses to temperature, there-fore, may be heavily masked by responses to photoperiod.Although some ovine studies [28] have reported a link be-tween the onset of ovarian activity and lowered tempera-ture, others have found no effect [29]. Studies on goats [30]and sheep [31] suggest that temperature may provide a fine-tuning mechanism to arrest ovarian activity at the end ofthe breeding season.

In the natural habitat of springbok, food availability andquality increase dramatically in response to rainfall [1]. Be-tween 1994 and 1998 in the Kalahari, rainfall averaged 146mm with 80% falling between November and March, al-though in 1994–1995, only 12 mm of rain fell. If ambienttemperature is a principal proximate factor driving repro-ductive activity in wild springbok, then most offspring con-ceived in response to a drop in ambient temperature wouldbe born during the hot-wet season, when food availabilitywas most abundant (Fig. 5). Further long-term research willbe required to establish whether temperature is a proximatefactor affecting reproduction in other species inhabiting aridenvironments.

Clearly, the different adaptations we proposed to explainaseasonal breeding are not upheld by the present study. Al-though some ewes were able to cycle continuously, the ma-jority had a clear synchronous anestrus. How can we rec-oncile this finding with what is observed in the wild, whenlambs are recorded throughout the year? Although it is pos-sible that in their natural environment more ewes cycle con-tinuously and thus conceive at any time in the year, webelieve that the male rut may be critical in this regard. Therut is a period of intense territorial male behavioral activitycharacterized by loud vocalizing, increased boundary mark-ing, and aggression toward other males and decreased for-aging, lasting between 5 and 23 days [24]; the proximatefactor or factors initiating the rut in springbok is unknown.Springbok rutting in the wild has only been observed fromJanuary to August [24], but this period coincides with the

1181ANNUAL ESTROUS CYCLICITY IN SPRINGBOK

occurrence of anestrus in the present study. Following rutsin April and July, there were definite increases in the pro-portion of lambs to ewes in a herd [1], but whether theewes were already cycling or anestrus at the time of the rutcould not be established. During the rut in many seasonalungulate species (e.g., impala [32]), most anestrous femalescome into estrus and, indeed, estrus has become synony-mous with rutting during the mating season in these spe-cies. Thus, one function of the male rut may be to initiatecycling in anestrous ewes, and this could maximize thebreeding capacity of springbok. Rutting has been equatedto the ‘‘ram effect’’ in sheep, which is used to induce anearlier onset of the breeding season. Extrapolating fromovine data [33] to springbok suggests that if springbok eweswere anestrous and relied solely on the ram effect to induceovulation, then only ruts lasting longer than 3 wk wouldresult in maximal conception.

Combining our data with observations in the wild, wherespringbok clearly breed throughout the year, although withdefinite lambing peaks, suggests that 1) if the rut occursduring the ‘‘breeding season’’, the biggest increase in thelamb-to-ewe ratio occurs; 2) if the rut does not coincidewith the ‘‘breeding season’’, then the number of femalesimpregnated will be lower; 3) ewes that are able to cycleat any time of the year can get impregnated at other timesof the year, thus giving the ‘‘appearance’’ that springbokbreed aseasonally, which would mask any underlying sea-sonal effect. These ewes would be receptive during theshorter ruts; 4) during the ‘‘breeding season’’, when allewes are cycling, opportunistic mating may occur outsideof the rut so that some females are impregnated, includingpubertal females coming into estrus for the first time. Thiswould also give the appearance of aseasonal breeding.

In summary, our study provides novel insights into thereproductive endocrine changes occurring in a wild ante-lope, the springbok, which is considered to be an aseasonalbreeder. Furthermore, it demonstrates that an aseasonallybreeding species may express periods of anestrus, as in theaddax and scimitar-horned oryx, but may also cycle con-tinuously as in the dik-dik, which would allow for oppor-tunistic breeding. Our study suggests that some proximatefactor, possibly ambient temperature, may time the onset ofestrous cyclicity in the springbok. Synchronous seasonalbreeding alone is too risky in arid regions, as the ultimatereason for breeding is survival of offspring [34]. Thus,combining synchronous breeding with opportunistic breed-ing means that mortality rate, although probably higher, isa necessary expense for exploiting the arid niche. Thisgambling is essential because the springbok ewe is unableto predict climatic conditions prevailing 6 mo after concep-tion.

ACKNOWLEDGMENTS

We thank Martin Haupt and Dominic Moss for the considerable assis-tance they gave in maintaining the animals and collecting blood samples,without whom this project would not have been possible. We thank theSouth African Weather Bureau for the meteorological data presented inthis paper.

REFERENCES

1. Skinner JD, Louw GN. The Springbok. Pretoria: Transvaal MuseumMonograph; 1996.

2. Wiesner H, Muller P. On the reintroduction of the Mhorr gazelle inTunisia and Morocco. Naturwissenschaften 1998; 85:553–555.

3. Sempere AJ, Ancrenaz M, Delhomme A, Greth A, Blanvillain C.Length of estrous cycle and gestation in the arabian oryx (Oryx leu-

coryx) and the importance of the male presence for induction of post-partum estrus. Gen Comp Endocrinol 1996; 101:235–241.

4. Morrow CJ, Wolfe BA, Roth TL, Wildt DE, Bush M, Blumer ES,Atkinson MW, Monfort SL. Comparing ovulation synchronizationprotocols for artificial insemination in the scimitar-horned oryx (Oryxdammah). Anim Reprod Sci 2000; 59:71–86.

5. Boever J, Knox D, Merilan C, Read B. Estrus induction and artificialinsemination with successful pregnancy in Speke’s gazelle. In: Pro-ceedings of the 9th International Congress on Animal Reproductionand Artificial Insemination 2; 1980; Madrid, Spain. pp. 565–569.

6. Holt WV, Abaigar T, Jabbour HN. Oestrous synchronization, semenpreservation and artificial insemination in the Mohor gazelle (Gazelladama mhorr) for the establishment of a genome resource bank pro-gramme. Reprod Fertil Dev 1996; 8:1215–1222.

7. Robeck TR, Fitzgerald LJ, Hnida JA, Turczynski CJ, Smith D, Krae-mer DC. Analysis of urinary progesterone metabolites with behavioralcorrelation in Guenther’s dik-dik (Madoque guentheri). J Zoo WildlMed 1997; 28:434–442.

8. Thompson KV, Mashburn KL, Monfort SL. Characterization of es-trous cyclicity in the sable antelope (Hippotragus niger) through fecalprogestagen monitoring. Gen Comp Endocrinol 1998; 112:129–137.

9. Jansen HT, Jackson GL. Circannual rhythms in the ewe—patterns ofovarian cycles and prolactin secretion under 2 different constant pho-toperiods. Biol Reprod 1993; 49:627–634.

10. Williams LM, Helliwell RJA. Melatonin and seasonality in sheep.Anim Reprod Sci 1993; 33:159–182.

11. Malpaux B, Chemineau P, Pelletier J. Melatonin and reproduction insheep and goats. In: Yu HS, Reiter RJ (eds.), Melatonin: Biosynthesis,Physiological Effects, and Clinical Applications. Boca Raton, FL:CRC Press; 1993: 253–281.

12. Terqui M, Thimonier J. Nouvelle methode radioimmunologique rapidepour l’estimation du niveau de progesterone plasmatique. Applicationpour le diagnostique precoce de la gestation chez la brebis et lachevre. CR Acad Sci (Paris) 1974; 279:1109–1112.

13. Saumande J, Tamboura D, Chupin D. Changes in milk and plasmaconcentration of progesterone in cows after treatment to induce su-perovulation and their relationship with the number of ovulations andembryos collected. Theriogenology 1985; 23:719–731.

14. Pelletier J, Kann J, Dolais J, Rosselin G. Dosage radioimmunologiquedel’hormone luteinisante plasmatique chez le mouton Mise au pointde la technique de dosage. CR Acad Sci (Paris) 1968; 266:2291–2294.

15. Montgomery GW, Martin GB, Pelletier J. Changes in pulsatile LHsecretion after ovariectomy in Ile-de-France ewes in two seasons. JReprod Fertil 1985; 73:173–183.

16. Tillet Y, Ravault JP, Selve C, Elvin G, Castro B, Dubois MP. Condi-tions d’utilisation d’anticorps specefiques pour la visualisation imu-nohistochimique de la serottonin et de la melatonine dans la glandepineale du mouton. CR Acad Sci (Paris) 1986; 303:77–82.

17. Asa CS, Houston EW, Fischer MT, Bauman JE, Bauman KL, HagbergPK, Read BW. Ovulatory cycles and anovulatory periods in the addax(Addax nasomaculatus). J Reprod Fertil 1996; 107:119–124.

18. Morrow CJ, Wildt DE, Monfort SL. Reproductive seasonality in thefemale scimitar-horned oryx (Oryx dammah). Anim Conservation2000; 2:261–268.

19. Mylrea GE, Mulley RC, English AW, Evans G. Reproductive cyclesof farmed female chital deer (Axis axis). Reprod Fertil Dev 1999; 11:167–173.

20. Asher GW, Muir PD, Semiadi G, O’Neill KT, Scott IC, Barry TN.Seasonal patterns of luteal cyclicity in young red deer (Cervus ela-phus) and sambar deer (Cervus unicolor). Reprod Fertil Dev 1997; 9:587–596.

21. Hanocq E, Bodin L, Thimonier J, Teyssier J, Malpaux B, Chemi-neau P. Genetic parameters of spontaneous spring ovulatory activityin Merinos d’Arles sheep. Genet Selection Evol 1999; 31:77–90.

22. Jackson GL, Jansen H. Persistence of a circannual rhythm of plasmaprolactin concentrations in ewes exposed to a constant equatorial pho-toperiod. Biol Reprod 1991; 44:469–475.

23. Karsch FJ, Robinson JE, Woodfill CJI, Brown MB. Circannual cyclesof luteinizing hormone and prolactin secretion in ewes during a pro-longed exposure to a fixed photoperiod: evidence for an endogenousreproductive rhythm. Biol Reprod 1989; 41:1034–1046.

24. Skinner JD, van Aarde RJ, Knight MH, Dott HM. Morphometrics andreproduction in a population of springbok, Antidorcas marsupialis, inthe semi-arid southern Kalahari. Afr J Ecol 1996; 34:312–330.

25. Thwaites C. Embryo mortality in the heat stressed ewe. 1. The influ-ence of breed. J Reprod Fertil 1967; 14:5–14.

26. Shelton M, Huston JE. Effects of high temperature stress during ges-

1182 SKINNER ET AL.

tation on certain aspects of reproduction in the ewe. J Anim Sci 1968;27:153–158.

27. Badinga L, Collier RJ, Thatcher WW, Wilcox CJ. Effects of climaticand management factors on conception rate of dairy cattle in subtrop-ical environment. J Dairy Sci 1985; 68:78–85.

28. Dutt RH, Bush LF. The effect of low environmental temperature oninitiation of the breeding season and fertility in sheep. J Anim Sci1955; 14:885–898.

29. Wodzicka-Tomaszewska M, Hutchinson JCD, Bennett JW. Control ofthe annual rhythm of breeding in ewes: effect of an equatorial day-length with reversed thermal seasons. J Agric Sci 1967; 68:61–67.

30. Gebbie FE, Forsyth IA, Arendt J. Effects of maintaining solstice light

and temperature on reproductive activity, coat growth, plasma prolac-tin and melatonin in goats. J Reprod Fertil 1999; 116:25–33.

31. Jackson GL, Jansen H, Kuehl DE, Shanks RD. Time of the siderealyear affects responsiveness to the phase-resetting effects of photope-riod in the ewe. J Reprod Fertil 1989; 85:221–227.

32. Murray MG. The rut of impala—aspects of seasonal mating undertropical conditions. Z Tierpsychol 1982; 59:319–337.

33. Martin GB, Scaramuzzi RJ. The induction of oestrus and ovulation inseasonally anovular ewes by exposure to rams. J Steroid Biochem1983; 19:869–875.

34. Baker JR. The evolution of breeding seasons. In: de Beer GR (ed.),Evolution: Essays on Aspects of Evolutionary Biology. Oxford, Unit-ed Kingdom: Oxford University Press; 1938: 161–177.