Wild Chacma Research Articles

Wild Baboons Groom Objects with Fur: Implications for Infant Corpse Carrying in Primates?

Nonhuman primates demonstrate great behavioral diversity in their responses to death, yet the most prevalent of these—infant corpse carrying—remains poorly understood. While mothers are the most frequent carriers, corpse carrying can be performed by non-mothers, and heterospecifics’ corpses are also carried. We propose a new hypothesis to explain, in part, non-maternal corpse carrying, the ‘mammalian cues hypothesis,’ and experimentally test it in a population of wild baboons. The mammalian-cues hypothesis argues that evolutionary pressures on maternal responsiveness have created a generalized attraction towards mammalian-typical traits in mammals of both sexes. This attraction may drive interest in inanimate objects (including corpses) that display mammalian cues. To test this, we presented objects that varied in mammalian traits to wild chacma baboons (Papio ursinus) and recorded their behavioral responses. Contrary to our hypothesis, baboons handled all objects (i.e., with or without mammalian traits) for similar lengths of time. However, in partial support of our hypothesis, baboons groomed the objects with fur—even those without any other mammalian traits—and explored them in a less destructive manner. We caution that infant corpse grooming by non-mothers in primates may be indicative of the visual and tactile stimuli the corpse presents rather than an interest in death or a reflection of social bonds.

Effects of accelerometry-derived physical activity energy expenditure on urinary C-peptide levels in a wild primate (Papio ursinus)

Animals have finite energy reserves for growth, survival, and reproduction and must maintain a stable energy balance. Measuring energy balance in the wild, however, is beset with methodological challenges. Quantification of urinary C-peptide (uCP), a proxy for insulin secretion, has enabled researchers to non-invasively estimate energy balance, and positive relationships between uCP levels and energy intake have been documented in numerous non-human primates. Comparatively few studies show that, consistent with insulin physiology, energy expenditure also alters levels of uCP. The timescale and extent of this relationship, however, remains unclear given the reliance on crude measures of activity and inferred energy expenditure. Here, for the first time, we test for effects of accelerometer-derived Vectorial Dynamic Body Acceleration (VeDBA) - a continuous measure of physical activity energy expenditure - on urinary C-peptide (uCP) levels in n = 12 wild chacma baboons (Papio ursinus). Applying a model selection approach, we show that VeDBA summed over short timescales (30 min to 1 h) prior to urine collection was negatively associated with uCP levels. Using the acceleration-based time individuals spent ‘non-stationary’ (i.e. locomoting) prior to urine collection as a predictor - instead of summed VeDBA - revealed similar but less clear results. Overall, the negative relationship between VeDBA and uCP levels highlights the importance of quantifying physical activity energy expenditure when using uCP measures to estimate energy balance and has potential implications for the field of energetics accelerometry.

Quantifying allo-grooming in wild chacma baboons (Papio ursinus) using tri-axial acceleration data and machine learning.

Quantification of activity budgets is pivotal for understanding how animals respond to changes in their environment. Social grooming is a key activity that underpins various social processes with consequences for health and fitness. Traditional methods use direct (focal) observations to calculate grooming rates, providing systematic but sparse data. Accelerometers, in contrast, can quantify activity budgets continuously but have not been used to quantify social grooming. We test whether grooming can be accurately identified using machine learning (random forest model) trained on labelled acceleration data from wild chacma baboons (Papio ursinus). We successfully identified giving and receiving grooming with high precision (81% and 91%) and recall (87% and 79%). Giving grooming was associated with a distinct rhythmical signal along the surge axis. Receiving grooming had similar acceleration signals to resting, and thus was more difficult to assign. We applied our machine learning model to n = 680 collar data days from n = 12 baboons and found that grooming rates obtained from accelerometers were significantly and positively correlated with direct observation rates for giving but not receiving grooming. The ability to collect continuous grooming data in wild populations will allow researchers to re-examine and expand upon long-standing questions regarding the formation and function of grooming bonds.

Evolutionary Determinants of Nonseasonal Breeding in Wild Chacma Baboons.

AbstractAnimal reproductive phenology varies from strongly seasonal to nonseasonal, sometimes among closely related or sympatric species. While the extent of reproductive seasonality is often attributed to environmental seasonality, this fails to explain many cases of nonseasonal breeding in seasonal environments. We investigated the evolutionary determinants of nonseasonal breeding in a wild primate, the chacma baboon (Papio ursinus), living in a seasonal environment with high climatic unpredictability. We tested three hypotheses proposing that nonseasonal breeding has evolved in response to (1) climatic unpredictability, (2) reproductive competition between females favoring birth asynchrony, and (3) individual, rank-dependent variations in optimal reproductive timing. We found strong support for an effect of reproductive asynchrony modulated by rank: (i) birth synchrony is costly to subordinate females, lengthening their interbirth intervals; (ii) females alter their reproductive timings (fertility periods and conceptions) in relation to previous conceptions in the group; and (iii) the reported effect of birth synchrony on interbirth intervals weakens the intensity of reproductive seasonality at the population level. This study emphasizes the importance of sociality in mediating the evolution of reproductive phenology in group-living organisms, a result of broad significance for understanding key demographic parameters driving population responses to increasing climatic fluctuations.

Stable isotopes reveal the effects of maternal rank and infant age on weaning dynamics in wild chacma baboons

Maternal strategies reflect the trade-off between offspring needs and maternal ability to invest, a concept described by the evolutionary theory of parent–offspring conflict. In mammals this conflict has often been investigated by studying weaning, the transition from maternal milk consumption to dietary independence. An investigation of individual variation in weaning can provide information on the adaptive significance of maternal strategies in relation to social and biological variables. We analysed stable nitrogen isotopes of hair samples collected from 22 mother–infant dyads in a wild population of chacma baboons, Papio ursinus, in conjunction with behavioural data on suckling, to explore the temporal dynamics of weaning, as well as the extent and determinants of individual variation in these dynamics. The weaning pattern suggested by isotope values and behavioural data were congruent. The difference between infant and mother stable nitrogen isotope values decreased faster with age in infants of low-ranking mothers, which suggests a faster progression towards weaning, perhaps due to subordinate females experiencing lower resource availability and so being less able to bear the costs of lactation over prolonged periods. Additionally, within-infant variation in stable nitrogen isotope values showed an increase with age (which was not detectable between infants), potentially highlighting the nutritional costs that weaning imposes on offspring. Our combination of isotope analysis and behavioural data from a wild population provides insight into the evolution of maternal strategies. In particular, it suggests that the quantity of care a mother can provide is affected by her rank, with subordinate females possibly not able to lactate for as long and perhaps benefiting from weaning earlier.

Simultaneous investigation of urinary and faecal glucocorticoid metabolite concentrations reveals short- versus long-term drivers of HPA-axis activity in a wild primate (Papio ursinus)

Glucocorticoids (GCs), a class of steroid hormones released through activation of the hypothalamic–pituitary–adrenal (HPA) axis, perform many vital functions essential for survival, including orchestrating an organism’s response to stressors by modulating physiological and behavioural responses. Assessing changes and variation in GC metabolites from faecal or urine samples allows for the non-invasive monitoring of HPA-axis activity across vertebrates. The time lag of hormone excretion differs between these sample matrices, which has implications for their suitability for studying effects of different temporal nature on HPA-axis activity. However, simultaneous comparisons of predictors of faecal and urinary GC metabolites (fGCs and uGCs, respectively) are lacking. To address this gap, we employ frequent non-invasive sampling to investigate correlates of fGCs and uGCs in wild chacma baboons (Papio ursinus) (n = 17), including long-term (dominance rank, season, female reproductive state) and short-term (time of day, daily weather conditions) factors. Correlated with increasing day length, fGCs gradually decreased from winter to summer. No seasonal effect on uGCs was found but ‘rain days’ were associated with increased uGCs. Pregnant females had significantly higher fGCs compared to cycling and lactating females, whereas uGCs were not statistically different across reproductive states. A circadian effect was observed in uGCs but not in fGCs. Dominance rank did not affect either fGCs or uGCs. Our study highlights the difference in inherent fluctuation between uGCs and fGCs and its potential consequences for HPA-axis activity monitoring. While uGCs offer the opportunity to study short-term effects, they undergo more pronounced fluctuations, reducing their ability to capture long-term effects. Given the increasing use of urine for biological monitoring, knowledge of this potential limitation is crucial. Where possible, uGCs and fGCs should be monitored in tandem to obtain a comprehensive understanding of short- and long-term drivers of HPA-axis activity.

Definition and interpretation effects: how different vigilance definitions can produce varied results

Animals use vigilance to detect or monitor threats. While numerous aspects of vigilance have been studied across a wide range of species, little work has explored the methodological variation that has emerged across these studies. Different approaches in sampling designs, statistical analyses and definitions can make cross-study comparisons challenging and potentially obscure our understanding of animal vigilance. In this study we explore two important components of vigilance definitions and ask (1) whether definitions vary in their interobserver agreement, and (2) whether using different definitions can create varied results within and across observers. Separate groups of ‘experienced’ and ‘inexperienced’ observers extracted data from video focal observations of wild chacma baboons, Papio ursinus, using four different definitions representative of the variation found within primate vigilance literature. In the first stage of analysis, we found that the four definitions varied in their interobserver agreement, with only an operational-looking definition performing well across both duration and frequency assessments, and an experienced/inexperienced dichotomy. This suggests definitions vary in how well observers can converge on similar interpretations of the same definition. The second part of the analysis used the experienced group's data in a typical primate vigilance analysis and found results varied within observers across definitions, i.e. definition effects, and across observers within definitions, i.e. interpretation effects. Together these results suggest that variation in definitions and their interpretation could have a fundamental role in producing between-study differences in results. Future vigilance research must consider these factors and explore working towards a single framework for studying vigilance, particularly within taxonomic families. Without consistency, cross-study comparisons are likely to be challenging and future observational work on other behaviours may also benefit from exploring these types of definitional issues. For baboons, operationalized definitions appear the most consistent across observers; however, future research should explore its application in other taxa.

The Relationship Between GPS Sampling Interval and Estimated Daily Travel Distances in Chacma Baboons (Papio ursinus)

Modern studies of animal movement use the Global Positioning System (GPS) to estimate animals’ distance traveled. The temporal resolution of GPS fixes recorded should match those of the behavior of interest; otherwise estimates are likely to be inappropriate. Here, we investigate how different GPS sampling intervals affect estimated daily travel distances for wild chacma baboons (Papio ursinus). By subsampling GPS data collected at one fix per second for 143 daily travel distances (12 baboons over 11–12 days), we found that less frequent GPS fixes result in smaller estimated travel distances. Moving from a GPS frequency of one fix every second to one fix every 30 s resulted in a 33% reduction in estimated daily travel distance, while using hourly GPS fixes resulted in a 66% reduction. We then use the relationship we find between estimated travel distance and GPS sampling interval to recalculate published baboon daily travel distances and find that accounting for the predicted effect of sampling interval does not affect conclusions of previous comparative analyses. However, if short-interval or continuous GPS data—which are becoming more common in studies of primate movement ecology—are compared with historical (longer interval) GPS data in future work, controlling for sampling interval is necessary.

First report of Mycobacterium bovis in wild chacma baboons (Papio ursinus) at the human-wildlife interface area in Zambia.

Mycobacterium bovis (M.bovis) causes tuberculosis in mammals and is a major public health threat worldwide. While M.bovis has been reported in humans, domestic and wild ruminants at the human-wildlife-livestock interface area in Zambia, there is paucity of information on the role of primates as reservoir hosts. We screened seven wild chacma baboons (Papio ursinus) for tuberculosis at the human-wildlife interface area in Lochinvar National Park in the Kafue Flats, Zambia. Following necropsy, lung tissue and associated lymph nodes with tuberculous-like lesions collected from four adult male baboons were prepared for Mycobacterium culture. The isolates were initially typed using the Mycobacterium tuberculosis complex-discrimination multiplex PCR assay and further characterized by spoligotyping and 26-loci MIRU-VNTR. Mycobacteria were isolated from all four animals and identified as M.bovis by PCR. On Spoligotyping, all isolates belonged to SB 0120 spoligotype, which is similar to what was previously reported in humans, cattle and Kafue lechwe antelopes in Kafue Flats ecosystem. Furthermore, on MIRU-VNTR typing, the baboon isolates clustered with cattle and Kafue lechwe isolates from the same catchment area. This finding intimates probable cross-species transmission of M.bovis in the Kafue Flats ecosystem. Due to the close interaction of baboons and humans at interface areas in Zambia, our results have potential implications for public health. Equally, this finding raises concerns for conservation.

Birth timing generates reproductive trade-offs in a non-seasonal breeding primate.

The evolutionary benefits of reproductive seasonality are often measured by a single-fitness component, namely offspring survival. Yet different fitness components may be maximized by different birth timings. This may generate fitness trade-offs that could be critical to understanding variation in reproductive timing across individuals, populations and species. Here, we use long-term demographic and behavioural data from wild chacma baboons (Papio ursinus) living in a seasonal environment to test the adaptive significance of seasonal variation in birth frequencies. We identify two distinct optimal birth timings in the annual cycle, located four-month apart, which maximize offspring survival or minimize maternal interbirth intervals (IBIs), by respectively matching the annual food peak with late or early weaning. Observed births are the most frequent between these optima, supporting an adaptive trade-off between current and future reproduction. Furthermore, infants born closer to the optimal timing favouring maternal IBIs (instead of offspring survival) throw more tantrums, a typical manifestation of mother-offspring conflict. Maternal trade-offs over birth timing, which extend into mother-offspring conflict after birth, may commonly occur in long-lived species where development from birth to independence spans multiple seasons. Our findings therefore open new avenues to understanding the evolution of breeding phenology in long-lived animals, including humans.

Individual differences in task participation in wild chacma baboons

Despite the controlled testing conditions that are typical of captive environments, many evaluations of animal cognition fail to ensure that all tested individuals participate. This is even more evident under wild conditions, as animals are not restricted in their movement or social interactions and have other activities available. In this study, we aimed to understand variation in cognitive task participation in wild chacma baboons, Papio ursinus . We quantified individual differences in the latency and likelihood to approach and explore two types of stimuli for two cognitive tests: a set of coloured paper bags (in an associative learning test) and a blue cardboard square (in a second-order conditioning test). We evaluated whether participation in each task was predicted by individuals’ phenotypic traits/states, as well as by two additional aspects of their behaviour: (1) the availability of competing activities at the time of testing and (2) their propensity to exploit social information. We found consistent results for both types of stimuli regarding the effect of age: juveniles were more likely to contact the stimuli and explore them for longer. Similarly, for both tasks, individuals involved in an activity at the time of testing were less likely to contact the stimuli and had a lower exploration time. Finally, juveniles and females with a high propensity to use social information (i.e. scrounge) were more likely, and had shorter latencies, to contact the paper bags. Our findings not only highlight the potential bias cognitive studies conducted in the wild can have, but also some of the individual attributes and external factors that determine task participation. • Cognitive testing requires animals to engage with tasks, but not all do. • Phenotypic traits/states and current behaviour may affect participation. • We tested participation by wild chacma baboons in two cognitive tasks. • Age, competing activities and social information use determined participation. • Participation rates in cognitive studies should be addressed to avoid sampling bias.

Energetics at the urban edge: Environmental and individual predictors of urinary C-peptide levels in wild chacma baboons (Papio ursinus)

As human-modified landscapes encroach into natural habitats, wildlife face a reduction in natural food sources but also gain access to calorie-rich, human-derived foods. However, research into the energetics of wildlife living within and adjacent to urban and rural landscapes is lacking. C-peptide - a proxy for insulin production and a diagnostic tool for assessing pancreatic function in humans and domestic animals - can be quantified non-invasively from urine (uCP) and may provide a way to investigate the energetic correlates of living in human-altered landscapes. UCP is increasingly used in studies of primate energetics, and here we examine predictors of variation in uCP levels in n = 17 wild chacma baboons (Papio ursinus) living at the urban edge on the Cape Peninsula, South Africa. We find that uCP was positively associated with food provisioning and negatively with night fasting. UCP levels were comparable between winter and summer but significantly lower during spring, possibly driven by consumption of energy-rich seeds during summer and more human-derived foods during winter. UCP was elevated in pregnant females and similar for lactating and cycling females. We find no effect of dominance rank on uCP. Samples collected with synthetic Salivettes had significantly lower uCP levels than directly pipetted samples. Overall, our results indicate that uCP is a reliable, non-invasive measure of energy balance and intake in baboons, and suggest potential energetic benefits of living at the urban edge. More broadly, studies of uCP may offer unique insight into the environmental control of hormone-behaviour relationships in species crossing natural and urban environments.

Exploring individual variation in associative learning abilities through an operant conditioning task in wild baboons.

Cognitive abilities underpin many of the behavioural decisions of animals. However, we still have very little understanding of how and why cognitive abilities vary between individuals of the same species in wild populations. In this study, we assessed the associative learning abilities of wild chacma baboons (Papio ursinus) across two troops in Namibia with a simple operant conditioning task. We evaluated the ability of individuals to correctly associate a particular colour of corn kernels with a distasteful flavour through repeated presentations of two small piles of corn dyed different colours, one of which had been treated with a non-toxic bitter substance. We also assessed whether individual variation in learning ability was associated with particular phenotypic traits (sex, social rank and neophilia) and states (age and prior vigilance). We found no evidence of learning the association either within each trial or across trials, nor any variation based on individuals' phenotypes. This appeared to be due to a high tolerance for bitter foods leading to similar acceptance of both palatable and unpalatable kernels. Earlier avoidance of the bitter kernels during pilot trials suggests this higher tolerance may have been largely driven by a drought during the experiments. Overall, our findings highlight the potential influence of current environmental challenges associated with conducting cognitive tests of animals in the wild.

Friend or foe: reconciliation between males and females in wild chacma baboons

Male aggression towards females is a common and often costly occurrence in species that live in bisexual groups. But preferential heterosexual relationships are also known to confer numerous fitness advantages to both sexes—making it of interest to explore how aggression is managed among male–female dyads through strategies like reconciliation (i.e. postconflict affiliative reunions between former opponents). In this study, we build on the traditional postconflict matched-control (PC-MC), time rule and rate methods to validate a novel methodological approach that tests for the presence and form of reconciliation between male and female wild chacma baboons, Papio ursinus. We show that heterosexual opponents exhibit friendly postconflict reunions, further demonstrating that reconciliation occurs almost exclusively between males and pregnant/lactating females who form tight social bonds. Such ‘friendships’ represent stable associations offering proximate and ultimate benefits to both parties—mainly improving (future) offspring survival. This aligns our findings with the ‘valuable relationship hypothesis’, which predicts rates of reconciliation to increase with the fitness consequences of the opponents' bond. Moreover, patterns concerning the initiative to reconcile reveal that males are as likely as females to initiate reconciliation, suggesting that males play a heretofore underappreciated role in maintaining heterosexual friendships. Beyond proposing a multivariate methodological technique applicable to other long-term observational data sets, the present research illuminates how male–female aggression in promiscuous societies may be mitigated via relationship repair strategies like reconciliation, the balance in those efforts between partners shedding new light on the mutual investment in such bonds.

Oestrous females avoid mating in front of adult male bystanders in wild chacma baboons.

In social species, female mating strategies can be constrained by both male and female groupmates through sexual conflict and reproductive competition, respectively. This study tests if females adjust their sexual behaviour according to the presence of male and female bystanders in wild chacma baboons (Papio ursinus) and assesses their relative importance. Our results show that oestrous females initiate fewer copulations in the presence of adult male bystanders, irrespective of whether they are mate-guarded or not. This inhibitory effect probably reflects a response to indirect sexual coercion by males, whose close proximity may dissuade females to initiate copulations with rival males to avoid punishment and/or aggressive mating interference. By contrast, females initiate more matings with their mate-guard in the presence of higher-ranking female bystanders, which may reflect an attempt to secure bodyguard services from their mate when they feel threatened. These results emphasize the importance of intra- and intersexual conflicts in shaping female sexual behaviour in this promiscuous society.

Context dependence of female reproductive competition in wild chacma baboons

Recent research reveals that female reproductive competition is common and may shape the social and reproductive strategies of female mammals. This study explores the determinants and intensity of female intrasexual conflicts in a wild promiscuous primate, the chacma baboon, Papio ursinus. We tested a suite of hypotheses to assess whether female–female aggression was primarily driven by instantaneous competition for food, mates or paternal care, or aimed at reducing future competition among offspring via reproductive suppression. Behavioural data were gathered from 53 females in two groups over two contrasting 2-year periods (2005–2006, 2013–2014): the first characterized by stability in the male dominance hierarchies, the second by instability induced by several immigration events and male take-overs. In both periods, we found that sexually receptive females received high levels of aggression from other sexually receptive females, consistent with competition over mating opportunities. In the unstable period, females exchanged higher rates of aggression than in the stable period, regardless of reproductive state, but lactating females received most aggression, consistent with higher competition over social access to male partners in response to increased infanticide risk. There was no evidence that aggression between females reflected either competition over food or reproductive suppression. These findings indicate that patterns of aggression between females fluctuate with sociodemographic factors affecting sexual and social access to males and reflect reproductive competition more closely than resource competition in this promiscuous primate society.

Male Violence and Sexual Intimidation in a Wild Primate Society

Sexual violence occurring in the context of long-term heterosexual relationships, such as sexual intimidation, is widespread across human populations [1-3]. However, its evolutionary origins remain speculative because few studies have investigated the existence of comparable forms of sexual coercion in animals [4, 5], in which repeated male aggression toward a female provides the aggressor with delayed mating benefits [6]. Here, we test whether male aggression toward females functions as sexual coercion in wild chacma baboons (Papio ursinus). We found support for all three main predictions of the sexual coercion hypothesis [7]: male aggression (1) is greatest against cycling females, (2) is costly and represents the main source of injuries for cycling females, and (3) increases male mating success with their victims in the future. Detailed analysis of chronological sequences between aggression and matings ruled out other coercive mechanisms, such as short-term harassment and punishment, by showing that aggression and matings are temporally decoupled. This decoupling may explain why some forms of sexual violence have been largely overlooked in well-studied animal populations despite their likely impact on the fitness of both sexes. Finally, we found no support for alternative hypotheses such as a female preference for aggressive males[8, 9]. This new, detailed study of the forms and intensity of sexual intimidation in a wild primate suggests that it may be widespread across mammalian societies, with important implications for understanding the evolution of mate choice and sexual conflict in mammals, as well as the origins of human sexual violence.

Identification of behaviours from accelerometer data in a wild social primate

BackgroundThe use of accelerometers in bio-logging devices has proved to be a powerful tool for the quantification of animal behaviour. While bio-logging techniques are being used on wide range of species, to date they have only been seldom used with non-human primates. This is likely due to three main factors: the long tradition of direct field observations, a difficulty of attaching bio-logging devices to wild primates and the challenge of deciphering acceleration signals in species’ with remarkable locomotor and behavioural diversity. Here, we overcome these aforementioned obstacles and provide methodology for identification of behaviours from accelerometer data of wild chacma baboons (Papio ursinus) in Cape Town, South Africa.ResultsWe apply machine learning techniques to process complex accelerometer data, collected by bespoke tracking collars to quantify a range of behaviours (focusing on locomotion and foraging behaviour). We successfully identify six broad state behaviours that represent 93.3% of the time budget of the baboons. Resting, walking, running and foraging were all identified with high recall and precision representing the first classification of multiple behavioural states from accelerometer data for a wild primate.ConclusionOur ‘end to end’ process—from collar design and build to the collection and quantification of acceleration data—provides advantages over gathering data by traditional observation, not least because it affords data collection without the presence of an observer which may affect an animal’s behaviour. Furthermore, our methodology and findings open new possibilities for the fine-scale study of movement and foraging ecology in wild primates, and in particular our baboon study population which is in conflict with people.

Episodic-like memory in zebrafish.

Episodic-like memory tests often aid in determining an animal's ability to recall the what, where, and which (context) of an event. To date, this type of memory has been demonstrated in humans, wild chacma baboons, corvids (Scrub jays), humming birds, mice, rats, Yucatan minipigs, and cuttlefish. The potential for this type of memory in zebrafish remains unexplored even though they are quickly becoming an essential model organism for the study of a variety of human cognitive and mental disorders. Here we explore the episodic-like capabilities of zebrafish (Danio rerio) in a previously established mammalian memory paradigm. We demonstrate that when zebrafish were presented with a familiar object in a familiar context but a novel location within that context, they spend more time in the novel quadrant. Thus, zebrafish display episodic-like memory as they remember what object they saw, where they saw it (quadrant location), and on which occasion (yellow or blue walls) it was presented.

Stability and strength of male-female associations in a promiscuous primate society

Close associations between adult males and lactating females occur in several promiscuous primate species. Benefits gained by males from such bonds may include increases in offspring fitness through paternal effort (the “mate-then-care” hypothesis) and/or subsequent mating opportunities with the female (the “care-then-mate” hypothesis). Heterosexual friendships between males and females have been described as tightly linked to the presence of a dependent infant, but few studies have investigated whether these associations may be longer, starting before an infant birth and lasting after its independence, and whether they may vary in strength across time and dyads. We investigated the stability and strength of heterosexual bonds in two groups of wild chacma baboons (Papio ursinus) to test whether male-female associations (1) last from offspring conception to independence, as expected under the mate-then-care hypothesis, (2) last after maternal cycling resumption and increase male mating success, as expected under the care-then-mate hypothesis, and (3) vary in strength depending on female reproductive state, age and rank. Our results show that most male-female bonds were already present at offspring conception, were maintained during pregnancy and lactation and disappeared when mothers resumed cycling. Moreover, heterosexual bonds were most intense during early lactation and for high-ranking females, and did not predict subsequent mating activity. Those findings favour the mate-then-care hypothesis, and confirm that male paternity assessment may, at least partially, rely on mating history. Overall, stable heterosexual friendships exist outside early lactation and last longer than previously thought, suggesting that male-female bonds could play an important and under-appreciated role on the social structure of groups in promiscuous primate societies.