Evolutionary Anthropology 22:226–238 (2013)

ARTICLE

Reproductive Suppression in Primates: A Review

JACINTA C. BEEHNER AND AMY LU

Reproductive performance is the currency of . All things being equal, A comprehensive review of the an organism should reproduce as often as possible. The puzzling questions in evo- reproductive suppression model is lutionary biology, therefore, are not how and why an organism does reproduce, not possible here. Therefore, we but rather how and why an organism does not reproduce. It is difficult to under- restrict this review in three impor- stand why any individual, particularly a female, might forestall reproduction when tant ways. First, we focus on within- one of the biggest limitations for female mammalian reproduction is time (that is, population comparisons and exam- reproductive lifespan).1 The answer, now widely cited throughout behavioral ecol- ine why one female might reproduce ogy is quite simple: Reproductive suppression can be an adaptive strategy.2 successfully while another in the same population might not. Between-species comparisons that Mammalian female reproduction may increase their lifetime explore life-history evolution across is a costly endeavor that may have a reproductive success by restricting taxa have received excellent coverage 1 high chance of failure under subopti- reproduction to coincide with favor- elsewhere. mal circumstances. Therefore, over able conditions. Thus, the general Second, although we briefly review evolutionary time, natural selection rule of thumb is this: If prevailing the energetic causes of reproductive should have favored females that are conditions are poor, delay reproduc- suppression, we focus primarily on sensitive to environmental cues and tion to preserve energy for future social causes (Fig. 1). Certainly, a able to suppress their own reproduc- reproduction; if conditions are likely female’s physiology should be sensi- tion when current prospects for to remain the same or get worse, tive to all cues that indicate harsh reproductive success are poor. This carry on with reproduction. reproductive conditions: their own idea, known as the reproductive sup- Several assumptions are associated energy balance and health, the physi- pression model,2 maintains that with this model.2 First, female repro- cal and genetic status of their or fetus, and the local envi- duction should be sensitive to all 3 cues that are predictive of poor ronmental conditions. But social reproductive outcomes and natural pressures are among the most dra- Jacinta C. Beehner is an Associate Pro- selection should favor females that matic and least understood factors fessor in the Departments of Anthropol- are able to translate these cues cor- affecting female reproduction. ogy and Psychology at the University of Third, in the spirit of this special Michigan. She is broadly interested in rectly into reproductive decisions. the interactions between hormones and Second, females should suppress issue, we would like to highlight the animal behavior, with a particular empha- work by Jeanne Altmann and Sarah sis on reproductive strategies in female reproduction as soon as a cue is primates. available because the earlier the loss, Hrdy, who have substantially furth- Amy Lu is a postdoctoral fellow in the the less it affects a female’s subse- ered our understanding of reproduc- Department of Anthropology at the Uni- tive suppression. In her versity of Illinois Urbana-Champaign. quent reproduction. Third, females She is interested in socioendocrinology, should use only cues that reliably groundbreaking work with wild yel- sexual selection, and development. Her low baboons in Amboseli National current research focuses on socially predict poor reproductive perform- mediated maturation and associated life- ance because it is costly to suppress Park, Kenya, Jeanne Altmann has history tradeoffs in wild geladas. a reproductive event that would oth- demonstrated that living in a erwise have been successful. Fourth, resource-poor and unpredictable Key words: anovulation; energetic suppression; reproductive suppression should be environment can have dramatic con- fertility; ovulatory suppression; loss; regulated by the female herself, and sequences on female fitness across 4–9 reproductive failure; social suppression thus differs from the more general the life span. Similarly, Sarah Hrdy term “reproductive failure,” which has established much of the ground-

VC 2013 Wiley Periodicals, Inc. also includes externally imposed work for testing hypotheses related to DOI: 10.1002/evan.21369 sources of loss, such as disease, social suppression in female primates Published online in Wiley Online Library (wileyonlinelibrary.com). injury, or . through her empirical work with the ARTICLE Reproductive Suppression in Female Primates 227

ENERGETIC SUPPRESSION Mammalian female reproduction requires a great deal of energy.16 When food is limited or entails high acquisition costs, organisms must make allocation decisions to priori- tize the partitioning of metabolic fuels. For many long-lived organisms, individual survival is paramount thus, when energy is at a minimum, basic cellular processes critical for survival take precedence over proc- esses that can be compromised, such as reproduction.17 However, even when survival is not at stake, females should abandon reproductive efforts as soon as they detect that they do not have enough energy to complete them. Energetic suppression is defined as the self-imposed inhibition of reproductive physiology in other- wise fertile individuals in response to an internal metabolic cue that indi- Figure 1. Routes to reproductive suppression. It has been notoriously difficult to separate cates a poor reproductive outcome. rank-related reproductive dysfunction caused directly by elevated levels of social stress To be clear, “self-imposed inhibition” from that caused indirectly by nutritional insufficiencies. Here, we focus mainly on empiri- does not involve any conscious deci- cal studies in primates that have demonstrated a direct link between the social environ- sion by the female, but is mediated by ment and female reproductive failure (red arrow). Note that although dominance rank a metabolic signal triggered in is a component of the social environment, it can mediate fitness in two ways: through the ecological environment (for example, by restricting energy intake, increasing energy response to an appropriate cue. The expenditure, or both) and through the social environment (for example, via social stress specifics that underlie this metabolic or social suppression of subordinates). [Color figure can be viewed in the online issue, pathway have yet to be fully identi- which is available at wileyonlinelibrary.com.] fied, but it is becoming increasingly clear that integrative control of energy balance and reproduction is Hanuman langurs of Mt. Abu, pression via neuroendocrine mecha- carried out by multiple metabolic and India,10,11 and her theoretical ideas nisms. In some cases, these are one neuroendocrine signals. Research on about infanticide.12–14 and the same; the factors that serve how females suppress reproduction In what follows, we provide an as proximate cues indicating a sub- in response to energetic signals is still overview of reproductive suppres- optimal environment are the same rapidly developing, and we refer read- sion in female primates, paying ones that trigger reproductive sup- ers to recent reviews of this topic.17,18 particular attention to two different pression (for example, poor ener- The question of why females sup- levels of analysis: an ultimate (func- getic condition). In most cases, press reproduction in response to tional) and a proximate (mechanis- however, the cues that reliably pre- limited energy is easier to tackle. tic) perspective (Fig. 2). From a dict reproductive failure are not Energetic suppression is common in functional perspective, we ask why inherently linked to reproductive females suppress their own repro- suppression (for example, the species in which physical condition arrival of a new male). In the inter- must meet a threshold before initiat- duction. That is, what are the fit- 19 ness benefits? To answer this est of bridging function and mecha- ing reproduction (capital breeders). question, we review the consequen- nism, we focus on both the cues as This energetic threshold presumably ces of reproducing under conditions well as the specific neuroendocrine indicates whether a mother will be in which the prospects for success mechanisms. We ask “why” and able to partition the necessary are poor. Second, from a proximate “how” questions as two related lines energy to reproduction on an perspective, we ask how females of inquiry. Just as the blind man ongoing basis. In contrast, other spe- suppress their own reproduction. cannot completely describe an ele- cies exclusively rely on external cues That is, what mechanisms are phant by holding onto its tail, we to initiate reproduction (income involved? There are two ways to specifically hope to emphasize how breeders). These external cues enable answer this question. We can integrating answers at different lev- females to schedule more energeti- explore the cues themselves or how els can help us emerge with a bet- cally demanding stages of reproduc- we can examine these cues are ter understanding of reproductive tion with a reliable and predictable translated into reproductive sup- suppression in female primates. abundance of resources.19 Few 228 Beehner and Lu ARTICLE

Figure 2. Overview of proximate (orange and purple) and functional (green) pathways that produce social suppression of reproduction in female primates. In this model, selection should favor females that translate external social cues appropriately. Note that the costs caused by reproductive suppression could lead to lower reproductive success if social cues become unreliable and females that “reproduce anyway” are successful. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] primates are strictly capital or another group, severe habitat degrada- ting the link between energetic condi- income breeders, and primates with- tion, forced individuals to move to a tion and reproductive rate, one study out a strict breeding season probably new home range with more abundant on female yellow baboons reported use a combination of internal and food resources.9 In both cases, female that (for females of all ranks), success- external cues. Therefore, with all the baboons in the improved foraging ful implantation was facilitated by rel- uncertainty that comes with a environments matured earlier and atively high levels of lengthy investment period, most pri- exhibited higher fertility.8,9 Another shortly after ovulation.33 However, mates should be able to abandon study on the Amboseli baboons pro- the progesterone threshold for proper reproduction at any time. vided direct support for the role of implantation was higher in subordi- Captive studies of primates have energetics in female reproductive sup- nate females than in dominants. unambiguously confirmed that inad- pression. During drought periods, Thus, an elevated threshold for equate maternal energy is associated females were much less likely to ovu- implantation may actually serve as a with reproductive dysfunction. For late; if they ovulated, they were less “reproductive filter,” reducing the example experimental manipulations likely to conceive and if they con- probability of conception in females involving food deprivation were able ceived, they were less likely to carry most likely to be affected by to induce amenorrhea in common the pregnancy to term.24 harsh environmental conditions (i. e. marmosets20 or pregnancy loss in Studies from Amboseli and other subordinate). rhesus macaques.21 Similarly, strenu- wild primate populations have dem- The relationships among group size, ous exercise without a corresponding onstrated that female fertility declines dominance rank, and fertility may be increase in caloric intake resulted in as the number of competitors for indirectly mediated by access to amenorrhea in rhesus macaques.22 resources goes up.24–26 For example, resources, directly mediated by social However, captive studies do not pro- female yellow baboons in larger stress, or a combination of both. Sepa- vide the multi-dimensional aspects of groups exhibited longer interbirth rating the relative contributions of the natural lives of female primates intervals than did females in smaller each mechanism is critical for making where a multitude of social and envi- groups.9 Similarly, during drought predictions about which populations ronmental factors contribute to vari- periods, the probability of conception are likely to exhibit rank-related or able reproductive outcomes. Some of decreased as the number of females group-size differences in female repro- the best examples of the adaptive sig- in the group increased.24 duction and under what conditions. nificance of energetic suppression in Numerous studies have shown that Social variables are likely to interact wild primates derive from Jeanne Alt- the costs of living in a primate group with energetic ones to generate an mann’s long-term studies on Amboseli may not affect all females equally, adaptive reproductive “decision.” To yellow baboons. Altmann and her col- particularly in species with strict dom- date, social suppression of reproduc- leagues found that a resource-rich inance relationships, such as cercopi- tion has largely been ignored in many environment can dramatically improve thecines. Across cercopithecines, low primates, despite evidence that social female reproductive outcomes. In one maternal rank is often associated with interactions can have major impacts group, baboons supplemented their a later age at maturation,27–29 longer on female fitness. For this reason, we diet with scraps from the garbage of a interbirth intervals,9,26,31,32 and later focus the remainder of our review on nearby tourist lodge23,whilein age at first birth.26,28 Possibly media- social suppression. ARTICLE Reproductive Suppression in Female Primates 229

SOCIAL SUPPRESSION energy from growth and the benefits ing with a novel male,40 which pre- of reproducing earlier. However, sumably is too short a time to Social suppression is said to occur social factors, such as the availability encompass the entire maturational when reproduction in otherwise fer- of suitable mates, may influence this process. Moreover, ovulating females tile individuals is delayed, inhibited, calculus. When only kin are will sometimes quickly revert to an or terminated in response to social “available” as mates, the costs of anovulatory condition at the first cues.2 It is independent from other inbreeding might shift the balance signs of social subordination.45 Con- factors that may influence reproduc- toward investing in growth and delay- sequently, it has been proposed that tion, such as energy balance, physi- ing the onset of reproduction. In suppressed females are not delaying cal health, and day length. Sarah , the presence of unrelated puberty at all; that is, the reproduc- Hrdy was one of the first to highlight males is able to accelerate puberty in tive axis undergoes sexual matura- the fact that social factors can have juvenile females,36 a phenomenon tion, but puberty is masked by dramatic effects on female reproduc- known as the . immediate ovulatory suppression.40 tive success. Her landmark work on The Vandenbergh effect encompasses Thus, a more accurate characteriza- infanticide in langurs and other pri- two levels of control: the inhibition of tion in callitrichids may be that sub- mates10,12,13 stimulated interest in reproduction when social cues indi- ordinate females are released from or how females can reduce the costs of cate high competition or no appropri- inhibited by family-induced fertility such interactions. ate mates, and the stimulation of suppression. reproduction once these inhibitory Among catarrhines, we know of A FUNCTIONAL PERSPECTIVE ON cues are lifted. Such social inhibition only two cases suggesting that males and stimulation of puberty clearly have an effect on female puberty. The SOCIAL SUPPRESSION: WHY IS operate in callitrichids and may first is an anecdote from captive ham- REPRODUCTIVE SUPPRESSION explain the timing of puberty in some adryas baboons, a species that lives in ADVANTAGEOUS? catarrhines, like geladas. relatively “closed” one-male polygy- Delayed Maturation to Avoid Among callitrichids (with a few nous units. The arrival of a new male exceptions37,38), female daughters in a captive group was followed by Inbreeding that remain in their natal group the sudden, early maturation of four In most taxa, males or females do exhibit some degree of ovulatory sup- young females, as determined by their not remain in their natal group long pression in the presence of the domi- first signs of perineal sexual swel- enough for inbreeding to occur. nant breeding female (usually the lings.46 The second case is from our However, among primates there are mother).34 Subordinate females population of geladas, who also some notable exceptions. For exam- remain anovulatory until social con- exhibit closed reproductive one-male ple, in the cooperatively breeding ditions change. The social conditions units.47 When we examined the tim- callitrichids (marmosets and tamar- necessary for the full onset of puberty ing of puberty for gelada females, we ins), daughters routinely remain in in subordinates, including ovulation, found two striking patterns. First, their natal groups well past the age mating, and conception, varies by puberty was disproportionately more of puberty,34 possibly due to a short- species. For example, puberty in at likely to occur the week after the age of productive territories and/or least two species of marmosets arrival of an unrelated male.48 An the inclusive fitness benefits or expe- requires only the removal of the dom- examination of the timing of female rience gained from helping to raise inant female;38,39 while puberty in maturation found that there was an siblings in the natal group.35 Another cotton-top tamarins requires the 11-fold increase in maturations the example comes from our own introduction of a novel male as week after a new male arrived as com- research on a wild population of well40,41 (but see French, Abbott, and pared to the background rate in the geladas in the Simien Mountains Snowdom42and Heistermann and population (N 5 33 females across 17 National Park, Ethiopia. Geladas live coworkers43). Occasionally, subordi- units, Fig. 3). Second, no females in polygynous one-male units in nate marmoset daughters ovulate in matured while their fathers were the which leader males occasionally the presence of their mothers.39 only breeding male, even though maintain extensive reproductive ten- Remarkably, the best predictor of this many were old enough to do so ures (maximum mate tenure, 6.9 “escape” from reproductive suppres- (N 5 40). Yet 60% of these females years) that are well past the age at sion is the relationship of these subsequently matured in the two puberty for females in this popula- daughters with their mother. Daugh- months after an unrelated male tion (manage at puberty 4.7 years). ters that do not show submission to arrived (Fig. 4). Taken together, these Yet for both callitrichids and gela- their mothers are the ones that data strongly suggest that social con- das, females are able to avoid ovulate.39 ditions can affect the timing of inbreeding because (for the most Puberty is a complex process that is puberty in geladas; new males appear part) they are able to delay puberty affected by both life-history parame- to trigger puberty and father-only under certain conditions. ters and social cues. For instance, groups appear to delay it. Life-history theory predicts that the female callitrichids are able to ovulate To our knowledge, these are the timing of puberty should reflect a bal- only a few short weeks after removal only two reports of delayed puberty in ance between the costs of diverting from the dominant female44 and pair- a catarrhine primate in association 230 Beehner and Lu ARTICLE

to suppress reproduction until compe- tition becomes less fierce, (e.g. when food is abundant or when the domi- nant is not breeding). In some soci- eties, the predictable nature of social inequalities between dominants and subordinates may have selected for complete or partial ovulatory suppres- sion in subordinates.49

Infanticide. Females are known to kill other Figure 3. Wild female geladas (Theropithecus gelada) exhibit external signs of maturation females’ offspring across a wide vari- (based on the first signs of sexual swelling) disproportionately more often during the ety of mammalian taxa.12,50 However, week after a male replacement. Shown are the number of females reaching maturity in female infanticide is relatively rare groups where males were replaced relative to the timing of those replacements (N533 among primates, having been females). The zero point on the x-axis indicates the week after a male replacement. reported only in ring-tailed and brown lemurs,51 yellow baboons,52,53 with a social factor. As yet, we have Reproductive suppression to chimpanzees,54,55 and common mar- no data on the sexual maturation of Minimize Female Competition mosets.56 Female infanticide is the reproductive axis relative to the hypothesized to result from competi- Primate females may also suppress putative suppression imposed by the tion for scarce resources. That is, by reproduction to minimize competition female’s father. Furthermore, because killing the young of other females, the with other females. In some species, many catarrhine primates, (e.g. yel- perpetrator and her offspring elimi- 4 dominants may reduce the fitness of low baboons), have a period of nate current and future competitors subordinates by imposing costs (for “adolescent sterility” between puberty for alloparents and/or food.13,50 Such example, through harassment) or and first reproduction, we do not yet intense competition among females reducing the likelihood of success (for know whether accelerated or delayed has almost certainly had strong selec- example, through infanticide). Thus, puberty in a catarrhine necessarily tion on the evolution of reproductive dominants can create an environment corresponds to an earlier or later age suppression in subordinates.50 in which subordinates would do better at first reproduction. The callitrichid social system may provide the most extreme example of social suppression because, generally, only the dominant female breeds. Most callitrichid females give birth to twins, which must be carried continu- ously for the first few weeks of life. Soon after, conception occurs again.57 Such high rates of reproductive invest require that nonbreeding group mem- bers (“helpers”) aid in rearing the dominant female’s offspring.57 The ubiquity of helpers has led some to suggest that they are mandatory for successful rearing of twins.58 Competi- tion for helpers is therefore intense. Perhaps as a result, among marmosets, dominant females kill the offspring of subordinate females59 and among tam- arins, dominant females evict subordi- nate females that give birth.60,61 These actions allow the dominant female to maintain “reproductive sovereignty.” As in many cooperatively breeding mammals, selection appears to have Figure 4. Shown is the number of female geladas reaching maturation (based on the first favored predictable ovulatory sup- signs of sexual swelling) relative to the timing of their genetic fathers being replaced as the leader male (N510 females across 20 groups). The dark vertical line represents the pression in subordinate callitrichid father’s replacement. Only females whose fathers were present are shown. Note that no females, sometimes even in the pres- females matured while their father was still the dominant male. ence of an appropriate male mate.39 ARTICLE Reproductive Suppression in Female Primates 231

Harassment this threat, females have developed a called a “Bruce effect,” a phenom- variety of counter-strategies.71 Some enon in which pregnant females ter- In contrast with the predictable of these strategies appear to prevent minated their after ovulatory suppression of callitri- infanticide altogether, while other exposure to novel males.73 The Bruce chids, individual females in many strategies appear to mitigate the costs effect was first documented in mice other primate taxa suppress ovula- of infanticide. and later extended to numerous tion in a more probabilistic fashion. Prevention strategies typically other species.74 However, all Is in callitrichids most of these cases involve behavioral tactics by females of these demonstrations were con- are associated with high levels of that act as either deterrents (for exam- ducted under laboratory conditions, female competition when resources ple, enlisting the protection of other leading to speculation that the Bruce are scarce. Females that are persis- males) or manipulations (for example, effect might simply be a laboratory tently harassed may delay reproduc- obscuring the cues that trigger infanti- artifact.75 However, our recent tion until a more opportune time. cidal behavior by males). Prevention research on geladas has since pro- The most widely cited study of strategies per se are not examples of vided clear evidence of an adaptive female social harassment documents female reproductive suppression. Mit- Bruce effect in a wild popuation.72 female coalitionary attack behavior in igation strategies, in contrast, cause Using five years of demographic and a wild population of yellow baboons.53 vulnerable females to immediately ter- hormonal data from geladas living in Such attacks were disproportionately minate all further investment in their the Simien Mountains National Park, directed toward in reproductive current infant to allow investment in a Ethiopia, we found that more than phases most susceptible to disruption future and presumably more success- 80% of pregnant females spontane- (i.e., ovulation and early pregnancy). ful infant. Such “loss-cutting” strat- ously miscarried when the dominant However, the results of this study have egies fall squarely within the male was replaced by another male been criticized, mainly because the framework of the reproductive sup- (“takeover”). Because nearly half of all frequency of attacks (rather than indi- pression model. Mitigation strategies infants fall victim to infanticide after vidual females) was the unit of analy- inflict reproductive costs on females takeovers,76 the context and high fre- sis.63 Other results consistent with the because all previous investment in the quency of the Bruce effect in geladas social suppression hypothesis derive current infant is lost. support its connection to infanticide. from studies on captive rhesus maca- Most surprising, females miscarry the ques,64 wild chacma baboons,65 and same day that the new male arrives. wild geladas.66,67 In each case, ovula- Strategies for pregnant females Because of the immediacy of the effect tory females received the highest rates Pregnancy termination may be one and because males rarely harass or of aggression from other females. of the most dramatic mitigation attack the females that miscarry, the However, the study on yellow baboons strategies available to primate Bruce effect in geladas is unlikely to be was the only one to also record females that encounter the threat of mediated by generalized stress. changes in reproductive parameters infanticide during the prepartum Anecdotal observations from cap- following aggression, with females period. If females that maintain their tive hamadryas baboons also suggest receiving the most attacks exhibiting pregnancies after a new dominant that a Bruce effect may be a female more cycles to conception and a male enters the group routinely lose reproductive strategy to mitigate higher rate of than their their infants to infanticide, then infanticide. A pregnant female spon- dominant counterparts.68 Moreover, a selection should weed out this strat- taneously aborted immediately after follow-up study on captive geladas egy. Alternatively, if females that mis- the introduction of a new male, even confirmed that attacks by dominant carry following the arrival of a new though this male exhibited no aggres- females can indeed directly inhibit male are able to quickly and success- sion towards her.46 In contrast, with ovulation in subordinates.69 fully reproduce with the new male, geladas and hamadryas baboons, in then females with the physiological at least two other cases, pregnancy Reproductive suppression to ability to terminate pregnancies in loss following male takeovers was response to a novel male will pre- directly associated with heightened Counter Infanticide by Males dominate until such a trait becomes aggression and harassment from the Male infanticide has been docu- relatively fixed in a population. In incoming male. For example, in yel- mented in nearly every major taxon of support of this scenario, our data low baboons, although primates and is perhaps the most per- from geladas indicates that terminat- are rare and infanticide is uncom- vasive and extreme threat to female ing a pregnancy is indeed an adapt- mon,24,77 three were reproductive success.70 Incoming ive strategy for females whose infants observed within two weeks of the dominant males kill offspring that are might be susceptible to infanticide.72 immigration of an extremely aggres- not their own. When they do, the The hypothesis that pregnancy ter- sive adult male.78 Similarly, spontane- mothers of these infants return to a mination might be adaptive for ous abortions following takeovers in fertile state, allowing the infanticidal females builds on decades of experi- Hanuman langurs were accompanied male to sire offspring much sooner mental studies of rodents. In 1959, by harassment and aggression than if he had to wait for the previous Hilda Bruce published the first dem- directed at pregnant females or their infants to be weaned. In response to onstration of what would later be infants.79 Therefore, we suspect that 232 Beehner and Lu ARTICLE there may be two different mecha- short-term, costs for females, (i.e., regulates an organism’s short-term nisms mediating pregnancy termina- delay in reproduction, that neverthe- response to stress.89 Although con- tion following male takeovers. One is less reap the long-term benefits of clusive studies from wild primate an adaptive strategy to a highly pre- avoiding harassment or infanticide), populations are rare, experimental dictable threat as, for example, in but are there other options? Cer- work has documented the negative geladas and hamadryas; the other is tainly, there is ample evidence that effects of stress on reproductive an outcome of a generalized stress many female primates can use function in a variety of primate and response to a novel threat as, for behavioral strategies in lieu of repro- other mammalian species.17,90 example, in yellow baboons. ductive suppression, particularly All vertebrate reproductive activity with respect to infanticide avoid- is coordinated through the hypo- Strategies for lactating females ance.71 Although such “prevention” thalamic–pituitary–gonadal (HPG) Lactating females may also use strategies are beyond the scope of axis (Box 1). Gonadotropin-releasing strategies to mitigate the effects of this review, we do focus on one such hormone (GnRH) neurons in the infanticide. Mothers could abandon strategy. release GnRH into infants that are vulnerable to infanti- Across many primate species, the median eminence, from which cide, thus reducing the costs of pregnant and lactating females it is transported through the “wasted” investment and hastening exhibit nonconceptive sexual behav- hypothalamic-hypophysial portal sys- the resumption of fertility. However, ior or display external signs of fertil- tem to the anterior pituitary. There, abandonment is uncommon and it ity such as sexual swellings. These GnRH regulates the secretion of the has only been reported for Hanuman false signals of fertility may manipu- gonadotropins, follicle-stimulating langurs12 and ursine colobus mon- late or obscure cues that infanticidal hormone (FSH), and luteinizing hor- keys.80 In contrast to complete aban- males use to identify their victims. mone (LH), which are released into donment, mothers could wean older For example, males may refrain general circulation, where they act infants somewhat earlier than usual from killing infants of females they on the ovary to stimulate ovulation. after a takeover. Early weaning might have mated with in the past. Routine Although activation of the stress reduce the infant’s risk of being killed nonconceptive mating is common in axis could potentially influence 71 if it hastens the mother’s return to fer- many primates. However, reproduction at any level of the HPG tility or prevents males from using “conditional” nonconceptive mating axis, the majority of studies suggests nursing behavior as a cue in their observed after male takeovers has that the effects of stress downregu- decision to attack infants. Accelerated been reported only for Hanuman late the release of GnRH from neu- 17 weaning after the arrival of a new langurs, sifakas, brown capuchins, rons in the hypothalamus. 71 male has been documented in several and blue monkeys. In addition, Identifying the precise signals that 85 primate species, including wild ursine among chimpanzees, red colobus relay HPA activity to GnRH neurons 86 87 colobus monkeys,81 wild white- monkeys, hamadryas baboons, has been challenging, but recent 88 headed leaf monkeys,82 and captive and geladas such nonconceptive research points toward corticotropin vervet monkeys.83 Premature weaning mating is accompanied by releasing factor (CRF) as one likely 91 has also been observed in wild gela- “deceptive” sexual swellings. Intrigu- candidate. Other hormones das, but it is unclear whether this was ingly, for hamadryas baboons and released into circulation during associated with an earlier return to geladas, the infants of lactating stress, such as glucocorticoids, have fertility.84 Finally, following the intro- females that produced postpartum been proposed to mediate stress duction of new males in a captive col- sexual swellings were less likely to be effects to the hypothalamic-pituitary 90 ony of hamadryas baboons, all six killed by new males than were the unit. Lending support to this idea, 87,88 lactating females resumed fertile sex- infants of other females. How- several studies have shown that ual cycles within two weeks, regard- ever, false signals do not seem to administration of glucocorticoids less of the age of their infants, provide effective protection to can, at least in part, inhibit the 71 (suggesting accelerated weaning) infants in all cases. secretion of LH from the anterior 90 although only the oldest infant pituitary. Although the precise sig- (nearly one year old) survived.46 A PROXIMATE PERSPECTIVE ON nals that relay the effects of stress to the HPG axis are not completely Thus, the combined data from wild SOCIAL SUPPRESSION; HOW DO and captive studies lends support to known, they are likely to include FEMALES SUPPRESS the hypothesis that accelerated wean- multiple neuroendocrine pathways ing can be a successful strategy, par- REPRODUCTION? that vary depending on the species, ticularly for mothers of older infants. Activation of the Stress sex, and hormonal or social status of the subject.17,90 Response Many mammalian studies show Alternatives to Reproductive Almost 75 years ago, Hans Selye that early pregnancy can be termi- discovered that reproduction in nated by stress.92 Under stressful Suppression female rats could be inhibited conditions, the adrenal cortex Based on the preceding examples, through activation of the (HPA) axis, secretes substantial quantities of reproductive suppression can entail the neuroendocrine pathway that , which readily metabolize ARTICLE Reproductive Suppression in Female Primates 233

Box 1. Reproductive Function and Dysfunction in Female Primates

Reproductive function in female primates. The reproductive cycle of female primates is controlled by the HPG axis. Release of GnRH from the hypothalamus into the median eminence and its transport through the hypothalamic-hypophysial portal system controls the secretion of gonado- tropins, (FSH, and LH) from the anterior pituitary. (FSH and LH) travel via systemic circulation to the ovaries, where they stimulate follicle growth (which secretes ) and eventual ovulation. After ovulation, the ruptured follicle tissue becomes the , which secretes progesterone. The HPG-axis is involved in each reproductive stage: puberty followed by regular ovulatory cycles (ovulation) and finally off- spring production. Puberty. The normal process of sexual maturation in most female pri- mates is tightly linked with growth and energy balance. Female puberty occurs shortly after a transition in energy allocation: Energy previously invested in growth is shunted to reproduction. On a physiological level, the HPG-axis inhibits reproduction in pre-pubertal females at both the pituitary and gonadal level, keeping reproductive hormones (that is, estra- diol and progesterone) low. As growth slows, increasing energy reserves in the peri-pubertal female cause this inhibition to lift, leading to a grad- ual rise in hormones. The rise in , in particular, stimulates devel- opment of the reproductive tract and eventually facilitates functional menstrual cycles and sexual behavior. Ovulation. This is the process whereby a mature egg is released from the female ovary into the oviduct, where it is available to become fertilized. For successful ovulation, GnRH is released by the hypothalamus in brief pulses, stimulating the anterior pituitary to release gonadotropins. Gonda- dotropins cause one or more ovarian follicles to develop. As the follicle develops, it produces estrogens and progestins. Through a positive feed- back mechanism, increasing levels of estrogens stimulate more gonadotro- pins to be released. Ovulation occurs when a spike in the amount of gondadotropins causes eggs to be released from the developing follicles. Conception. If females successfully ovulate, the released eggs still need to be fertilized and undergo success- ful implantation. Thus, conception necessarily involves both a behavioral component (mating with a fertile male) and a physiological one (successful implantation). Implantation is an early stage of pregnancy at which the embryo must successfully adhere to the wall of the . This process is highly dependent on progester- one coming from the corpus luteum. Gestation. If successful implantation occurs, the embryo proceeds to develop within the protective confines of the uterus. Gestation begins the moment of implantation and ends with parturition, and is the second most energetically expensive reproductive phase for mammalian females. The maintenance of gestation is dependent on threshold levels of progesterone and estrogens. At the beginning of pregnancy, these hormones are secreted by the ovary; however, once pregnancy is established, the fetal placenta takes over this process. Minute changes in the concentrations of either steroid can disrupt gestation and lead to pregnancy failure. Lactation. Parturition is immediately followed by lactation, during which the neonate is dependent on only maternal milk as a source of nutrition. Lactation is the most energetically expensive reproductive phase for mammalian females. The ability to sustain the costs of lactation and avoid infant mortality is therefore highly dependent on maternal body condition. The HPG axis is quiescent during this period, returning to cyclic activ- ity only when suckling bouts have subsided and maternal energy balance has returned to positive. Reproductive suppression. This is defined broadly as the inhibition of reproductive physiology and/or repro- ductive behavior in an otherwise fertile individual in response to specific environmental or physiological condi- tions. Reproductive suppression generally results from external and internal cues that predict poor reproductive outcomes. Such cues may come in the form of negative energy balance, in which, energy output exceeds energy input; physiological stress, or elevated circulating concentrations of glucocorticoids; or either chemosignals or social signals from conspecifics. These signals are mediated by specialized neuroendocrine mechanisms. 234 Beehner and Lu ARTICLE

Generally, the common endpoint for all of these mechanisms is disruption of the secretion of GnRH in the hypothalamus, which curtails all downstream products of the HPG axis. Reproductive suppression has been documented in females across a wide range of primate taxa and may involve inhibition at any reproductive stage. Importantly, the costs and benefits of reproductive suppression change with respect to the stage at which it occurs, with suppression at early stages entailing fewer costs and more benefits, while suppression at late stages generates higher costs and fewer benefits. Reproductive suppres- sion can be divided into two “types” based on the cues thought to mediate reproductive inhibition. Energetic suppression: The inhibition of reproductive physiology in otherwise fertile individuals in response to inadequate availability of maternal energy. Social suppression. The inhibition of reproductive physiology or behavior in otherwise fertile individuals directly in response to their interactions with other conspecifics.

to estrogens. Consequently, in addi- societies, subordinate individuals at times when reproduction is most tion to glucocorticoids, acute stress tend to be characterized by glucocor- probable and suppress it when it is leads to an elevation in estradiol. ticoid levels that are equal to or not. For example, chemosignals Although estradiol is an essential lower than those of dominants.97 As within the of male rodents can hormone for preparing the uterus for a result, it is now accepted that the accelerate puberty in juvenile pregnancy, even very small eleva- extreme reproductive suppression in females (the Vandenbergh effect)101 tions above optimal levels can inter- cooperatively breeding species is and stimulate ovulation in anovula- fere with blastocyst implantation.93 probably not mediated by social tory adult females (the Whitten Elevations in estradiol may also be stress. effect).102 This pro-ovulatory “male an important component of repro- effect” has also been found in anes- ductive control outside of physiologi- trous domestic sheep and goats.103 cal stress.93 Chemosensory or Other Signals The same chemosignals that stimu- Although most of what we know Chemical signals (or chemosignals) late reproduction in cycling females about the effects of stress on repro- also alter the reproductive physiology may also disrupt gestation in preg- 73 duction comes from captive studies and behavior of conspecifics. nant females (the Bruce effect), on rodents and domestic animals, a Although this term is often used inter- suggesting that each of these effects few primate studies have successfully changeably with “,” it is may be controlled by the same 104 connected the dots between a stress- rare among mammals for a chemosig- mechanism. Remarkably, after ful social environment, a neuroendo- nal to meet the strict criteria for a females learn the olfactory signature crine stress response, and pheromone: (which is a unitary, of a male, (that is, they recognize his reproductive suppression in one form species-typical substance that is both odor) his urine no longer induces or other. For example, high rates of necessary and sufficient for an such effects. aggression and elevated levels of cor- experience-independent behavioral or Currently, two hypotheses exist for tisol in subordinate females were physiological response).98 In contrast, the “stimulating agent” within male 98 linked to failure to produce the LH mammalian chemosignals contain urine. The longstanding hypothesis surge required for ovulation in cap- multiple compounds that interact to is that a combination of volatile 94 tive talapoin monkeys and a longer produce a “chemical image” that gen- compounds in male urine activates a time to conception in captive gela- erally contains priming or releasing neuroendocrine pathway in the vom- 69 das. Further, subordinate female agents (serving essentially as phero- eronasal olfactory system (VNS) of cynomolgus monkeys experienced mones)99 as well as a chemical signa- females and inhibits the higher rates of aggression, lower rates ture used to recognize an animal as release necessary for maintaining 99 of grooming, increased cortisol an individual or as a member of a par- pregnancy. The alternative hypoth- responses to an ACTH infusion, and ticular social group.100 Therefore, in esis does not require mediation by 95,96 impaired reproductive function. mammals, social information has the neural events but is instead directly These results suggest that many potential to dramatically alter the mediated by estradiol in an unknown examples of social suppression in effects of particular chemosignals. male urine. Exogenous estradiol is female primates may be mediated by directly absorbed by females and binds to receptors along the repro- a generalized stress response. Chemosignals in nonprimates ductive tract, inducing puberty or Remarkably, physiological stress In mammals, chemosignals have disrupting pregnancy.93 Both does not appear to play a role in cal- the potential to prime reproductive hypotheses support the idea that the litrichid reproductive suppression.49 physiology so that individuals signal is -dependent and Indeed, in cooperatively breeding become sexually mature and active must be bound to nonvolatile ARTICLE Reproductive Suppression in Female Primates 235 proteins to facilitate odor individual- ineffective at maintaining ovulatory gonadotropins and ovarian hor- ity.98 The VNS detects and translates suppression, suggesting that chemi- mones.49,111,118 Further, at the first chemical signals in urine that indi- cal signals from dominants are effec- signs of subordination, previously vidually identify males.105 Impor- tive only if they are paired with cycling females exhibit a sudden tantly, in most rodents, these urinary recognition of the signaler.115 In this drop in gonadotropins, followed constituents are both necessary and regard, reproductive control by dom- shortly afterward by complete ovula- sufficient for inducing these inants may be mediated simply tory suppression.45,119 Although the effects.106 through the learned association by precise mechanism for this is subordinates between the dominant unknown, it has been suggested, in Urine of group-housed female female’s cues and the behavioral sub- common marmosets, that social sub- mice can delay puberty in juvenile ordination she imposes.116 ordination causes a dissociation females (another form of the Van- In addition, scent secretions from between GnRH and pituitary secre- 107 denbergh effect) and prolong or dominant females are not necessary tion of gonadotropins, possibly via completely suppress ovarian cyclicity for maintaining ovulatory suppres- reduced pituitary sensitivity to in adult females (the Lee-Boot sion in callitrichids. In another ver- GnRH.116 108 effect). In both cases, the sion of the experiment above, In catarrhine primates, we under- “delaying agent” in female urine is subordinate marmoset females ren- stand almost nothing of the proxi- adrenal-dependent. Females suppress dered anosmic remained anovulatory mate mechanisms stimulating ovulation when signals from as long as they were housed with a changes in female reproductive con- conspecifics indicate a high popula- known dominant female.114 Further- dition in response to social cues. In tion density, suggesting increased more, even after physical separation addition to a Bruce effect and a possi- 98 competition for resources. from the dominant female, her visual ble Vandenbergh effect in female cues alone were enough to maintain geladas, we also see signs of false fer- anovulation in anosmic subordi- tility in lactating females after new Chemosignals in primates nates.114 Yet neither visual nor males enter the group. (Intriguingly, In primates, chemosignals are chemical signals were able to main- all three have also been reported for used in a variety of contexts.109 tain suppression indefinitely, indicat- hamadryas baboons.)46,87 None of However, most of what we know ing that social cues from the these is consistent with a stress- about the role of chemosignals in dominant female may be the pre- mediated mechanism. In both gelada reproductive suppression derives dominant stimulus controlling ovula- and hamadryas studies, new males from research on captive callitri- tory suppression.49 did not appear to harass, intimidate, chids. It has become clear that che- The role that novel males (or their or injure females on takeover,76 and mosignals in the scent secretions of chemosignals) might have in stimu- the speed with which pregnant a dominant female play a key role in lating ovulation in callitrichid females females terminated their pregnancies both initiating and maintaining is less straightforward. First, whether does not suggest the outcome of reproductive inhibition in subordi- males themselves are even necessary stress-induced inhibition of the HPG nate marmosets and tamar- for stimulating ovulation is variable axis. Instead, it resembles the ins.41,110,111 (Additionally at least across species, with only cotton-top chemosensory-mediated Bruce and one study on cotton-top tamarins tamarins requiring both the removal Vandenbergh effects in rodents. How- suggests that related males can have of the dominant female and the intro- ever, the VNS in catarrhine primates the same suppressive effect.112) For duction of a novel male.40,41 Second, has become vestigial, indicating a example, common marmoset in cotton-top tamarins, males were diminished role for olfaction and che- females rendered anosmic (unable to required for initiating ovulation in mosignals.99 Without a functional detect chemosignals) were able to previously acyclic females, but were VNS, catarrhines presumably lack the avoid ovulatory suppression after not required for maintaining.117 ability to detect non-volatile chemo- exposure to a dominant female.113 Finally, male chemosignals alone sensory compounds. So how are these Furthermore, anovulatory marmoset were not sufficient for stimulating effects mediated? and tamarin females kept only ovulation in anovulatory female tam- At present, several possibilities within scent contact of the dominant arins,112 although combining these may explain the reproductive female exhibited a significant delay chemosignals with visual and audi- changes observed in female geladas in the onset of ovulation.41,110,111 tory cues did stimulate cyclicity.117 and hamadryas baboons following However, despite a delay in ovula- To our knowledge, no study has takeovers. First, although unlikely, tion, the effectiveness of the odor examined whether male chemosig- male chemosignals could be trans- cues alone expired after a few nals are necessary for ovulatory mitted to females via the main olfac- weeks,114 indicating that chemosig- stimulation. tory system, dermal contact, or nals from the dominant female are We know even less about the spe- semen. Second, as in callitrichids, not sufficient for maintaining ovula- cific neuroendocrine mechanisms the trigger for reproductive suppres- tory suppression. Supporting this involved. Anovulatory, mature, sion could come from visual, tactile, argument, odor from an unfamiliar female callitrichids dwelling in their or other social cues. These cues dominant female was completely natal groups exhibit acyclic levels of alone or in combination with 236 Beehner and Lu ARTICLE chemosignals could stimulate neuro- data emerging from geladas, where ACKNOWLEDGMENTS endocrine activity within the female, systematic abortions following take- We thank Joan Silk for inviting us to causing prolactin inhibition and/or a overs appear to be the rule rather participate in this special issue and to rise in estradiol, both of which can than the exception. So how do we thank Jeanne Altmann and Sarah significantly effect a developing distinguish adaptive reproductive Hrdy for their inspiration to both of us fetus. Regardless of the mechanism, suppression from more general mal- across the years. Further, we thank it is possible that the same neuroen- adaptive reproductive failure? two anonymous reviewers for sub- docrine pathway controls all three We suggest three ways to resolving stantially improving a previous ver- different reproductive responses to this. First, we place reproductive fail- sion of the manuscript. Funding for novel males. Understanding the ures along a continuum that ranges the gelada work was provided by the details of this pathway is therefore from adaptive strategies at one end National Science Foundation (BCS- crucial for determining how such (for example, a Bruce effect in gela- 0715179, BCS-0824592, BCS- responses have been shaped by das) to non-adaptive outcomes at the 0962118), the Leakey Foundation, the evolution. other (for example a, infant death by National Geographic Society (Grant disease). The adaptive end requires no. 8100-06), the Bronx Zoo, and the reliable and predictable cues; the CONCLUSIONS University of Michigan. non-adaptive end is imposed by Throughout this review, we have purely stochastic events. Stress- taken the stance that reproductive mediated reproductive failure falls in REFERENCES suppression is an evolved strategy the middle of this continuum, where 1 Stearns SC. 1992. The evolution of life histor- mediated by females. Females trans- the generalized stress response is ies. Oxford: Oxford University Press. late predictable and reliable signals adaptive, but the events triggering 2 Wasser SK, Barash DP. 1983. Reproductive suppression among female mammals: implica- into a “decision” to continue or the response are stochastic. tions for biomedicine and sexual selection abandon reproduction at a mecha- Second, we emphasize the impor- theory. Q Rev Biol 58:513–538. nistic level. But to what extent is this tance of identifying the proximate 3 Bronson FH. 1989. Mammalian reproductive assumption justified? In other words, cues associated with the immanent biology. Chicago: University of Chicago Press. 4 Altmann J. 1980. Baboon mothers and could reproductive suppression sim- failure. If the proximate cue, (such infants. Cambridge: Harvard University Press. ply be the outcome of physiological as the arrival of a new male) is not 5 Altmann J, Alberts S. 1987. Body mass and constraints in the same way that also the mechanism for reproductive growth rates in a wild primate population. genetic abnormalities may cause a failure, this is more likely to be Oecologia 72:15–20. 6 Altmann J, Muruthi P. 1988. Differences in pregnancy to fail? evolved reproductive strategy than if daily life between semi-provisioned and wild- Reproduction can fail in one of the cue is intricately tied to the feeding baboons. Am J Primatol 15: 213–221. two ways: indirectly, as the outcome mechanism itself (such as negative 7 Altmann J, Schoeller D, Altmann SA, et al. of an evolved mechanism that energy balance then this more 1993. Body size and fatness of free-living baboons reflect food availability and activity responds appropriately to a proxi- likely). levels. Am J Primatol 30:149–161. mate cue indicating suboptimal con- Third, we encourage research that 8 Altmann J, Alberts SC. 2003. Variability in ditions, (as has been suggested for merges mechanism and function to reproductive success viewed from a life-history perspective in baboons. Am J Hum Biol 15: social suppression), or directly, as the test adaptive hypotheses. For exam- 401–409. outcome of a mechanism that ple, we know a great deal about the 9 Altmann J, Alberts SC. 2003. Intraspecific responds to the suboptimal condi- mechanisms mediating reproductive variability in fertility and offspring survival in a nonhuman primate: behavioral control of eco- tions themselves, (as has been sug- suppression in callitrichids, but we logical and social resources. In: Wachter KW, gested for energetic suppression). know much less about their adaptive Bulatao RA, editors. Offspring: the biodemogra- For example, high rates of aggres- significance. Why do most subordi- phy of fertility and family behavior. Washing- ton, D.C.: National Academy Press. p 140–169. sion inflicted by one unusually nate callitrichid females completely 10 Hrdy SB. 1974. Male-male competition and aggressive yellow baboon male in suppress their own reproduction? infanticide among the langurs (Presbytis entel- Amboseli78 almost certainly caused What is the likelihood that they will lus) of Abu, Rajasthan. Folia Primatol 22:19– the subsequent abortions observed in eventually become the dominant 58. 11 Hrdy SB. 1977. The langurs of Abu: female several females, perhaps as a result breeding female? Or is their advant- and male strategies of reproduction. Cam- of hemorrhage or stress. One could age solely one of kin selection? In con- bridge: Harvard University Press. argue that an acute stress response trast, in catarrhines we are beginning 12 Hrdy SB. 1977. Infanticide as a primate by these females acts the same way to understand of why some forms of reproductive strategy. Am Sci 65:40–49. 13 Hrdy SB. 1979. Infanticide among animals: that injury, disease, or pathology reproductive suppression are adapt- a review, classification, and examination of the does to cause reproductive failure (a ive, but we have little to no data on implications for reproductive strategies of non-adaptive response). Alterna- the mechanisms mediating these females. Ethol Sociobiol 1:13–40. tively, one could argue that females strategies. It is paramount that pri- 14 Hrdy SB, Janson CH, van Schaik CP. 1995. Infanticide: let’s not throw out the baby with have evolved to suppress reproduc- mate biologists integrate both func- the bath water. Evol Anthropol 3:151–154. tion in response to high levels of tional and mechanistic approaches 15 Hrdy S. 1976. The care and exploitation of generalized stress, regardless of the for us to fully understand the why nonhuman primate infants by conspecifics other than the mother. Advances in the Study of nature of this stress. Both of these and how behind reproductive sup- Behavior, vol. 6. Academic Presss: New York, arguments stand in contrast to the pression in female primates. pp 101–158. ARTICLE Reproductive Suppression in Female Primates 237

16 Gittleman JL, Thompson SD. 1988. Energy 32 Smuts BB, Nicolson N. 1989. Reproduction sets: what can proximate mechanisms tell us allocation in mammalian reproduction. Am in wild female olive baboons. Am J Primatol about ultimate causes? Proc R Soc Lond B 276: Zool 28:863–875. 119:229–246. 389–399. 17 Uphouse L. 2011. Stress and reproduction 33 Wasser SK. 1996. Reproductive control in 50 Digby L. 2000. Infanticide by female mam- in mammals. In: Norris DO, Lopez KH, editors. wild baboons measured by fecal steroids. Biol mals: implications for the evolution of social Hormones and reproduction of vertebrates, vol. Reprod 55:393–399. systems. In: van Schaik CP, Janson CH editors. 5: mammals. San Diego: Elsevier. p 117–138. 34 Abbott DH, Barrett J, George LM. 1993. Infanticide by males and its implications. Cam- 18 Garcia-Garcia RM. 2012. Integrative control Comparative aspects of social suppression of bridge: Cambridge University Press. p 423–446. of energy balance and reproduction in females. reproduction in female marmosets and tamar- 51 Jolly A, Caless A, Cavigelli S, et al. 2000. Int Scholarly Res Network Vet Sci 2012: 1–13. ins. In: Rylands AB, editor. Marmosets and Infant killing, wounding and predation in Eule- 19 Brockman DK, van Schaik CP. 2005. Sea- tamarins: systematics, behaviour and ecology. mur and Lemur. Int J Primatol 21:21–40. Oxford: Oxford University Press. p 152–163. sonality and reproductive function. In: Brock- 52 Shopland JM, Altmann J. 1987. Fatal intra- man DK, van Schaik CP, editors. Seasonality in 35 Goldizen AW. 1987. Tamarins and marmo- group kidnapping in yellow baboons. Am J Pri- primates: studies of living and extinct human sets: communal care of offspring. In: Smuts matol 13:61–65. and non-human primates. New York: Cam- BB, Cheney DL, Seyfarth RM, et al., editors. bridge University Press. p 269–305. Primate societies. Chicago: University of Chi- 53 Wasser SK, Starling AK. 1988. Proximate cago Press. p 34–43. and ultimate causes of reproductive suppres- 20 Lujan ME, Krzemien AA, Reid RL, et al. sion among female yellow baboons at Mikumi 36 Vandenberg JG. 2006. and 2006. Developing a model of nutritional amen- National Park, Tanzania. Am J Primatol 16:97– mammalian reproductive function. Knobil and orrhea in rhesus monkeys. Endocrinology 147: 121. 483–492. Neill’s physiology of reproduction, 3rd ed. San 54 Goodall J. 1986. The chimpanzees of 21 Tardif S, Power M, Layne D, et al. 2004. Diego: Elselvier.p 2041–2058. 37 French JA, Stribley JA. 1985. Patterns of uri- Gombe: patterns of behaviour. Cambridge: Har- Energy restriction initiated at different gesta- vard University Press. tional ages has varying effects on maternal nary oestrogen excretion in female golden weight gain and pregnancy outcome in com- tamarins (Leontopithecus rosalia). J Reprod 55 Townsend SW, Slocombe KE, Emery mon marmoset monkeys (Callithrix jacchus). Br Fertil 75:537–546. Thompson M. 2007. Female-led infanticide in J Nutr 92:841–849. 38 Smith TE, Schaffner CM, French JA. 1997. wild chimpanzees. Curr Biol 17:R355–R356. 22 Williams NI, Caston-Balderrama AL, Social and developmental influences on repro- 56 Digby L. 1995. Infant care, infanticide, and Helmreich DL, et al. 2001. Longitudinal ductive function in female Wied’s black tufted- female reproductive strategies in polygynous changes in reproductive hormones and men- ear marmosets (Callithrix kuhli). Horm Behav groups of common marmosets. Behav Ecol strual cyclicity in cynomolgus monkeys during 31:159–168. Sociobiol 37:51–61. strenuous exercise training: abrupt transition 39 Saltzman W, Pick RR, Salper OJ, et al. 57 Tardif SD. 1996. The bioenergetics of paren- to exercise-induced amenorrhea. Endocrinology 2004. Onset of plural cooperative breeding in tal behavior and the evolution of alloparental 142:2381–2389. common marmoset families following replace- care in marmosets and tamarins. In: Solomon 23 Muruthi P, Altmann J, Altmann SA. 1991. ment of the breeding male. Anim Behav 68:59– NG, French JA, editors. Cooperative breeding Resource base, parity, and reproductive condi- 73. in mammals. Cambridge: Cambridge University tion affect females’ feeding time and nutrient 40 Ziegler TE, Savage A, Scheffler G, et al. Press. p 11–32. intake within and between groups of a baboon 1987. The endocrinology of puberty and repro- 58 Bales K, Dietz J, Baker A, et al. 2000. population. Oecologia 87:467–472. ductive functioning in female cotton-top tamar- Effects of allocare-givers on fitness of infants 24 Beehner JC, Onderdonk DA, Alberts SC, et ins (Saguinus oedipus) under varying social and parents in callitrichid primates. Folia Pri- al 2006. The ecology of conception and preg- conditions. Biol Reprod 37:618–627. matol 71: 27–38. nancy failure in wild baboons. Behav Ecol 17: 41 Savage A, Ziegler TE, Snowdon CT. 1988. 59 Digby LJ, Saltzman W. 2009. Balancing 741–750. Sociosexual development, pair bond formation, cooperation and competition in callitrichid pri- 25 Borries C, Larney E, Lu A, et al. 2008. Costs and mechanisms of fertility suppression in mates: rxamining the relative risk of infanticide of group size: lower developmental and repro- female cotton-top tamarins (Saguinus oedipus across species. In: Davis LC, Ford SM, Porter ductive rates in larger groups of leaf monkeys. oedipus). Am J Primatol 14:345–359. LM, editors. The smallest anthropoids: the mar- Behav Ecol 19:1186–1191. 42 French JA, Abbott DH, Snowdon CT. 1984. moset/callimico radiation. New York: Springer. 26 van Noordwijk MA, van Schaik CP. 1999. The effect of social environment on p 135–153. excretion, scent marking, and sociosexual The effects of dominance rank and group size 60 Price EC, McGrew WC. 1991. Departures behavior in tamarins (Saguinus oedipus). Am J on female lifetime reproductive success in wild from monogamy in colonies of captive cotton- Primatol 6:155–167. long-tailed macaques, Macaca fascicularis. Pri- top tamarins. Folia Primatol 57:16–27. mates 40:105–130. 43 Heistermann M, Kleis E, Prove E, et al. 1989. Fertility status, dominance, and scent 61 Henry MD. 2011. Proximate mechanisms 27 Bercovitch FB, Strum SC. 1993. Dominance and ultimate causes of female reproductive rank, resource availability, and reproductive marking behavior of family-housed female cotton-top tamarins (Saguinus oedipus) in skew in cooperatively breeding golden lion tam- maturation in female savanna baboons. Behav arins, Leontopithecus rosalia. Ph.D. thesis, Uni- Ecol Sociobiol 33:313–318. absence of their mothers. Am J Primatol 18: 177–189. versity of Maryland, College Park. 28 Altmann J, Hausfater G, Altmann SA. 1988. 62 Solomon NG, French JA. 1996. Issues in the Determinants of reproductive success in savan- 44 Evans S, Hodges JK. 1984. Reproductive status of adult daughters in family groups of study of mammalian cooperative breeding. In: nah baboons, Papio cynocephalus. In: Clutton- Solomon NG, French JA, editors. Cooperative Brock TH, editor. Reproductive success: studies common marmosets (Callithrix jacchus jac- chus). Folia Primatol 42:127–133. breeding in mammals. New York: Cambridge of individual variation in contrasting breeding University Press. p 1–10. systems. Chicago: University of Chicago Press. 45 Abbott DH, Hodges JK, George LM. 1988. p 403–418. Social status controls LH secretion and ovula- 63 Harcourt AH. 1987. Dominance and fertility tion in female marmoset monkeys (Callithrix among female primates. J Zool, Lond 213:471– 29 Wasser SK, Norton GW, Kleindorfer S, jacchus). J Endocrinol 117:329–339. 487. et al. 2004. Population trend alters the effects of maternal dominance rank on lifetime repro- 46 Colmenares F, Gomendio M. 1988. Changes 64 Wallen K, Tannenbaum PL. 1997. Hormonal ductive success in yellow baboons (Papio cyno- in female reproductive condition following modulation of sexual behavior and affiliation cephalus). Behav Ecol Sociobiol 56:338–345. male takeovers in a colony of hamadryas and in rhesus monkeys. In: Carter CS, Lederhendler hybrid baboons. Folia Primatol 50:157–174. II, Kirkpatrick B, editors. Integrative neurobiol- 30 Cheney DL, Seyfarth RM, Fischer J, ogy of affiliation, vol. 807. New York: New Beehner JC, Bergman TJ, Johnson SE, Kitchen 47 Snyder-Mackler N, Alberts SC, Bergman TJ. York Academy of Sciences. p 185–202. DM, Palombit RA, Rendall D, Silk JB. Factors 2012. Concessions of an alpha male? Coopera- affecting reproduction, and mortality among tive defence and shared reproduction in multi- 65 Huchard E, Cowlishaw G. 2011. Female- baboons in the Okavango Delta, Botswana. male primate groups. Proc R Soc Lond B 279: female aggression around mating: an extra cost International Journal of Primatology 25:401– 3788–3795. of sociality in a multimale primate society. 428. 48 Lu A, Beehner JC. 2012. Vandenbergh effect Behav Ecol 22:1003–1011. 31 Setchell JM, Wickings EJ. 2004. Social and in wild geladas? Am J Phys Anthropol S54:196– 66 Dunbar RIM. 1980. Determinants and evolu- seasonal influences on the reproductive cycle in 197. tionary consequences of dominance among female mandrills (Mandrillus sphinx). Am J 49 Saltzman W, Digby LJ, Abbott DH. 2009. female gelada baboons. Behav Ecol Sociobiol 7: Phys Anthropol 125:73–84. Reproductive skew in female common marmo- 253–265. 238 Beehner and Lu ARTICLE

67 Dunbar RIM, Dunbar EP. 1977. Dominance 86 Struhsaker TT, Leland L. 1985. Infanticide sory signals in the . Science and reproductive success among female gelada in a patrilineal society of red colobus monkeys. 306:1033–1037. baboons. Nature 266:351–352. Z Tierpsychol 69:89–132. 106 Drickamer LC. 1980. Social cues and 68 Wasser SK, Norton GW, Rhine RJ, et al. 87 Zinner D, Deschner T. 2000. Sexual swel- reproduction: rodents and primates. In: Cohen 1998. Aging and social rank effects on the lings in female hamadryas baboons after male MN, Malpass RS, Klein HG. editor. Biosocial reproductive system of free-ranging yellow take-overs: “deceptive” swellings as a possible mechanisms of population regulation. New baboons (Papio cynocephalus) at Mikumi female counter-strategy against infanticide. Am Haven: Yale University Press. p 37–53. National Park, Tanzania. Hum Reprod Update J Primatol 52:157–168. 107 Colby DR, Vandenberg JG. 1974. Regula- 4:430–438. 88 Roberts EK, Beehner JC. 2011. A tale of tory effects of urinary pheromones on puberty 69 McCann CM. 1995. Social factors affecting deception: pseudo-sexual swellings in a wild in the mouse. Biol Reprod 11:268–279. reproductive success in female gelada baboons primate. Soc Behav Neuroendocrinol. 108 Van Der Lee S, Boot LM. 1956. Spontane- (Theropithecus gelada). New York: City Univer- 89 Selye H. 1939. Effects of adaptation to vari- ous pseudopregnancy in mice II. Acta Physiol sity of New York. ous damaging agents on the female sex organs Pharmacol Neerlandica 5:213–215. 70 van Schaik CP, Janson CH, editors. 2000. in the rat. Endocrinology 25:615–624. 109 Drea CM, Boulet M, Delbarco-Trillo J, Infanticide by males and its implications. Cam- 90 Tilbrook AJ, Turner AI, Clarke IJ. 2000. et al. 2013. The “secret” in secretions: methodo- bridge: Cambridge University Press. Effects of stress on reproduction in non-rodent logical considerations in deciphering primate 71 Palombit RA. 2012. Infanticide: male strat- mammals: the role of glucocorticoids and sex olfactory communication. Am J Primatol 75: egies and female counterstrategies. In: Mitani differences. Rev Reprod 5:105–113. 621–642. JC, Call J, Kappeler PM, et al., editors. The evo- 91 Chand D, Lovejoy DA. 2011. Stress and 110 Barrett J, Abbott DH, George LM. 1990. lution of primate societies. Chicago: University reproduction: Controversies and challenges. Extension of reproductive suppression by pher- of Chicago Press. p 432–468. Gen Comp Endocrinol 171:253–257. omonal cues in subordinate female marmoset 72 Roberts EK, Lu A, Bergman TJ, Beehner JC. 92 de Catanzaro D, MacNiven E. 1992. Psycho- monkeys, Callithrix jacchus. J Reprod Fertil 90: 2012. A Bruce effect in wild geladas. Science genic pregnancy disruptions in mammals. Neu- 411–418. 335:1222–1225. rosci Biobehav Rev 16:43–53. 111 Epple G, Katz Y. 1984. Social influences 73 Bruce HM. 1959. An exteroreceptive block 93 de Catanzaro D. 2011. Blastocyst implanta- on oestrogen excretion and ovarian cyclicity in to pregnancy in the mouse. Nature 184:105. tion is vulnerable to stress-induced rises in saddle back tamarins (Saguinus fuscicollis). Am endogenous estrogens and also to excretions of J Primatol 6:215–227. 74 Becker SD, Hurst JL. 2008. Pregnancy block estrogens by proximate males. J Reprod Immu- 112 Widowski TM, Ziegler TE, Elowson AM, from a female perspective. In: Hurst JL, editor. nol 90:14–20. et al. 1990. The role of males in the stimulation Chemical signals in vertebrates, vol. 11. New of reproductive function in female cotton-top York: Springer. p 141–150. 94 Bowman LA, Dilley SR, Keverne EB. 1978. Suppression of oestrogen-induced LH surges by tamarins, Saguinus oedipus. Anim Behav 40: 75 Wolff JO. 2003. Laboratory studies with social subordination in talapoin monkeys. 731–741. rodents: facts or artifacts? Bioscience 53:421– Nature 275:56–58. 113 Abbott DH, Faulkes CG, Barrett J, et al. 427. 95 Kaplan JR, Adams MR, Koritnik DR, et al. 1993. Social control of female reproduction in 76 Beehner JC, Bergman TJ. 2008. Infant mor- 1986. Adrenal responsiveness and social status marmoset monkeys and naked mole-rats. In: tality following male takeovers in wild geladas. in intact and ovariectomized Macaca fascicula- Lehnert H, Murison RH, Hellhammer D, et al. Am J Primatol 70:1–8. ris. Am J Primatol 11:181–193. editors. Endocrine and nutritional control of basic biological functions. Seattle: Hogrefe and 77 Beehner JC, Nguyen N, Wango EO, et al. 96 Adams MR, Kaplan JR, Koritnik DR. 1985. Huber. p 475–490. 2006. The endocrinology of pregnancy and fetal Psychosocial influences on ovarian endocrine loss in wild baboons. Horm Behav 49:688–699. and ovulatory function in Macaca fascicularis. 114 Barrett J, Abbott DH, George LM. 1993. 78 Pereira ME. 1983. following the Physiol Behav 35:935–940. Sensory cues and the suppression of repro- duction in subordinate female marmoset mon- immigration of an adult male baboon (Papio 97 Creel S. 2001. Social dominance and stress keys, Callithrix jacchus. J Reprod Fertil 97: cynocephalus). Am J Primatol 4:93–98. hormones. Trends Ecol Evol 16:491–497. 301–310. 79 Agoramoorthy G, Mohnot SM, Sommer V, 98 Petrulis A. 2013. Chemosignals, hormones 115 Abbott DH, Saltzman W, Schultz-Darken et al. 1988. Abortions in free ranging Hanuman and mammalian reproduction. Horm Behav 63: NJ, et al. 1997. Specific neuroendocrine mecha- langurs (Presbytis entellus): a male induced 723–741. strategy? Hum Evol 3:297–308. nisms not involving generalized stress mediate 99 Swaney WT, Keverne EB. 2009. The evolu- social regulation of female reproduction in 80 Sicotte P, Teichroeb JA, Saj T. 2007. Aspects tion of pheromonal communication. Behav cooperatively breeding marmoset monkeys. of male competition in Colobus vellerosus: pre- Brain Res 200:239–247. Ann NY Acad Sci 807:219–238. liminary data on male and female loud calling, 100 Wyatt TD. 2010. Pheromones and signa- 116 Abbott DH, Saltzman W, Schultz-Darken and infant deaths after a takeover. Int J Prima- ture mixtures: defining species-wide signals tol 28:627–636. NJ, et al. 1998. Adaptations to subordinate sta- and variable cues for identity in both inverte- tus in female marmoset monkeys. Comp Bio- 81 Teichroeb JA, Sicotte P. 2008. Infanticide in brates and vertebrates. J Comp Physiol A 196: chem Physiol 119(Part C):261–274. ursine colobus monkeys (Colobus vellerosus) in 685–700. 117 Widowski TM, Porter TA, Ziegler TE, et al. Ghana: new cases and a test of the existing 101 Vandenbergh JG. 1969. Male odor accler- 1992. The stimulatory effect of males on the hypotheses. Behaviour 145:727–755. ates female sexual maturation in mice. Endo- initiation but not the maintenance of ovarian 82 Zhao Q, Borries C, Pan W. 2011. Male take- crinology 84:658–660. cycling in cotton-top tamarins (Saguinus oedi- over, infanticide, and female countertactics in 102 Whitten WK, Bronson FH, Greenstein JA. pus). Am J Primatol 26:97–108. white-headed leaf monkeys (Trachypithecus leu- 1968. Estrus-inducing pheromone of male 118 Tardif SD. 1984. Social influences on sex- cocephalus). Behav Ecol Sociobiol 65:1535– mice: transport by movement of air. Science ual maturation of female Saguinus oedipus 1547. 161:584–585. oedipus. Am J Primatol 6:199–209. 83 Fairbanks LA, McGuire MT. 1987. Mother- 103 Galez H, Fabre-Nys C. 2004. The "male 119 Ziegler TE. 2013. Social effects via olfac- infant relationships in vervet monkeys: effect" in sheep and goats: a review of the tory sensory stimuli on reproductive function response to new adult males. Int J Primatol 8: respective roles of the two olfactory systems. and dysfunction in cooperative breeding mar- 351–366. Horm Behav 46:257–271. mosets and tamarins. Am J Primatol 75:202– 84 Dunbar RIM. 1984. Reproductive decisions: 104 Guzzo AC, Jheon J, Imtiaz F, et al. 2012. 211. an economic analysis of gelada baboon social Oestradiol transmission from males to females strategies. Princeton: Princeton University in the context of the Bruce and Vandenbergh Press. effect in mice (Mus musculus). Reproduction 85 Wallis J. 1982. Sexual behavior of captive 143:539–548. chimpanzees (Pan troglodytes): pregnant versus 105 Leinders-Zufall T, Brennan P, Widmayer P, VC 2013 Wiley Periodicals, Inc. cycling females. Am J Primatol 3:77–88. et al. 2004. MHC class I peptides as chemosen-