Throughout the animal kingdom, females and males have evolved different strategies to optimize their own reproductive outcome (Trivers, 1972; Clutton-Brock, 1989). Reproductive strategies, in fact, are often conflicting between the two sexes, leading to an “arms race” between females and males (Engelhardt et al., 2006). Depending on the ecology and mating system of species, various behavioral, anatomical, and physiological traits can be observed with which both sexes try to increase their own fitness (reviewed in Stumpf et al., 2011). For example, in multi-male multi-female groups with male dominance, males compete over access to females by trying, e.g., to monopolize fertile females and females aiming to confuse paternity by mating with several males, while at the same time reproducing with the “best” potential mates (e.g., in terms of high genetic quality or paternal investment).

To achieve their respective reproductive aims, both sexes use cues or signals of reproductive quality and reproductive state in multiple modalities, acting alone or as multimodal cues. While “cues” did not evolve to serve a communicative function, “signals” did evolve to cause a behavioral or physiological change in the recipient (Smith and Harper, 2003). It is well-known that visual and acoustic traits such as sexual swellings or copulation calls can indicate the fertile period in female primates (Cebuella pygmaea: Converse et al., 1995; Microcebus murinus: Buesching et al., 1998). Also olfactory cues can transport a wealth of information about individuals but studies investigating olfactory traits of female fertility are comparatively underrepresented in primates (Drea, 2015). For non-human primates, evidence for olfactory fertility cues predominantly comes from scent-marking strepsirrhine primates and New World monkeys. For instance, when males encountered the odor of fertile females, they reacted with adapted behavior like increased rates of investigation (Galago crassicaudatus: Clark, 1982; Saguinus oedipus: Ziegler et al., 1993; Callithrix jacchus: Smith and Abbott, 1998; Lemur catta: Scordato and Drea, 2007), neuronal activity and increased hormone levels (Callithrix jacchus: Ziegler et al., 2005; Snowdon et al., 2006). In common marmosets, the chemical composition of anogenital odors varied with female fertility and according bioassays provided evidence that males prefer the odor of fertile females compared to non-fertile females (Callithrix jacchus: Kücklich et al., 2019). The few existing studies on Old World monkeys suggest that males can perceive olfactory cues of female fertility and inspect females more often during the fertile period (Papio ursinus: Clarke et al., 2009; Papio anubis: Rigaill et al., 2013; Macaca arctoides: Cerda-Molina et al., 2006a,b; Macaca mulatta: Michael et al., 1971). In humans, men were able to distinguish fertile from non-fertile women using axillary body odors, with the odors from fertile women perceived as most attractive and least intense (e.g., Havliček et al., 2006; Gildersleeve et al., 2012). Furthermore, Michael et al. (1974, 1975) analyzed vaginal secretions of women and discovered a change in volatile aliphatic acids over the menstrual cycle, which may thus underlie men’s odor preferences for fertile women.

To date, our knowledge about the importance of olfactory communication is particularly scarce in great apes. It is known that great apes use their sense of smell to collect information about conspecifics (Matsumoto-Oda et al., 2007; Hepper and Wells, 2010; Klailova and Lee, 2014). A comparative study (Jänig et al., 2018) showed that male chimpanzees sniff more often at conspecifics than females or males of other great ape species (gorillas, orangutans, and bonobos). Analysis of great ape body odor revealed that the richness, intensity, and chemical composition varies between species (bonobos, chimpanzees, gorillas, and orangutans), but also with age (Jänig et al., 2019). However, olfactory fertility cues have not been examined well. It was observed that male chimpanzees inspect females’ genitals (Nishida, 1997; Emery Thompson and Wrangham, 2008) and it was suggested that they might try to collect olfactory cues on the fertility status of the females (Wallis, 1992; Nishida, 1997). Chemical analyses of vaginal secretions of female chimpanzees revealed the presence of substances which were previously found to vary during the menstrual cycle in rhesus macaques and humans (Macaca mulatta: Michael et al., 1971; humans: Michael et al., 1974, 1975), but cycle-related variation could not yet be confirmed in chimpanzees (Fox, 1982; Matsumoto-Oda et al., 2003). Overall, the few existing observations suggest that olfactory cues might be of relevance in great apes and play a particular role for communication in chimpanzees.

Chimpanzees (Pan troglodytes) are particularly interesting to study cues and signals of reproduction involving multiple modalities (or multimodal communication). They form complex fission-fusion groups (Nishida, 1968; Goodall, 1986) and females mate promiscuously during an extended period of receptivity (Tutin, 1980). It has been suggested that the predominant female reproductive strategy is to distribute the probability of paternity across several mating partners (Wrangham, 2002; Kappeler and van Schaik, 2004) to benefit from a reduced risk of infanticide (Stumpf and Boesch, 2006). This may be achieved best when the displayed information on when ovulation actually occurs is imprecise. Indeed, female chimpanzees show several cues and signals with different levels of precision that could potentially be indicative of their fertility, including copulation calls (Townsend et al., 2011), proceptive behavior as well as mating resistance behavior (Stumpf and Boesch, 2006). The most prominent advertisement of fertility in female chimpanzees, however, is an exaggerated anogenital swelling. Sexual swellings depend on estrogen levels and last 10–12 days (Tutin, 1980; Wallis, 1997). However, ovulation may occur within several days of the phase of maximal swelling and importantly, ovulation cannot be determined exactly from swelling size alone (Emery Thompson, 2005). Even though the frequency of copulations is highest when females are fully swollen, mating also occurs outside this period (Wallis, 1982).

The competition between male chimpanzees over access to females is intense (Goodall, 1986; Muller, 2002) and high-ranking males monopolize most copulations (Klinkova et al., 2005). Several females of a group can ovulate at a time (Goodall, 1986), limiting opportunities of male monopolization. Chimpanzee party size was found to be influenced, amongst other factors, by the presence of swollen females (Tutin, 1979), with males preferably joining parties with fully swollen females (Matsumoto-Oda, 1999), and male sexual behavior changed in correspondence to the size of the sexual swelling (Klinkova et al., 2005). On average, mating frequency closely follows the estrogen profile: it increases with increasing estrogen levels as well as swelling size and decreases with decreasing estrogen levels and swelling size, thus correlating with the most fertile days (Deschner et al., 2004; Emery Thompson, 2005). Interestingly, the rate of genital inspections increases before and with the onset of the maximal sexual swelling, suggesting that males might try to perceive olfactory cues in addition to the visual cue (Wallis, 1992; Nishida, 1997). To optimize their mating efforts and increase the chance of siring offspring, males would benefit from investing into mating with females primarily during their most fertile days, which could be achieved when a more precise indicator of fertility than the visual signal is available.

Indeed, observations suggest that, in addition to the imprecise visual signal of swellings, a more precise cue of ovulation could exist (Emery Thompson and Wrangham, 2008). Some males, particularly high-ranking males, seem to adjust their behavior more precisely to the actual time of ovulation. Alpha males were found to increase copulation frequency 1 day before the start of the peri-ovulatory period and copulation frequency was highest during the fertile days within the phase of maximal swelling (Deschner et al., 2004). Although females mate with several males during their maximal swelling stage, alpha males sire the majority of offspring (Constable et al., 2001; Klinkova et al., 2005). This skewed distribution of mating success suggests that high-ranking males may have better access to females, but possibly also to more detailed information on female fertility. Male chimpanzees may have evolved abilities to pinpoint an approaching ovulation using cues and signals of multiple sensory modalities with varying precision. If this is the case, the visual signal could first be used as an obvious but only approximate advertisement available for all males. High-ranking males, which are able to approach and monopolize females, may then use olfactory cues when inspecting the sexual swelling and gain more precise information. Since male chimpanzees frequently inspect swollen females (Matsumoto-Oda et al., 2007; Jänig et al., 2018), we suggest that males indeed could use olfactory cues. If only high-ranking males get access to these precise information, thus increasing their probability of paternity success, females would also benefit, because their offspring acquires the best genes (Kappeler and van Schaik, 2004) while they still maintain a high level of paternity confusion.

Our general aim of the current study was to provide an insight into olfaction in the reproductive context of chimpanzees in order to fill the existing gap of knowledge. In particular, we intended to test the suggestion that olfactory fertility cues might exist in addition to the exaggerated visual signals by investigating males’ sniffing behavior directed toward females. Accordingly, we examined three properties that might affect the actual fertility of females: their (1) swelling stage, (2) cycle regularity, and (3) age. The swelling stage represents the females’ actual reproductive state within the menstrual cycle, with ovulation most likely occurring within the maximal swelling period (Deschner et al., 2004). The regularity of females’ menstrual cycle can be assumed to be an indicator for receptivity, with non-cycling females (e.g., lactating females or females receiving contraceptives) being less or non-receptive. We further wanted to consider age of the female in our study, because male chimpanzees copulate more often with older than with younger females or preferably join parties with older females (Muller et al., 2006). Age seems to reflect reproductive experience (Muller et al., 2006), with older females being more experienced. We observed the frequency at which male chimpanzees sniff at females and examined whether male sniffing frequency varied with the described reproductive properties of females. Since male chimpanzees compete intensively and the access to females is limited, we furthermore argue that the number of males present at a time influences the rate of olfactory inspection. We suggest that the levels of competition rise when more males are around and that each male thus has fewer opportunities to sniff at a female. Assuming that males try to optimize their mating efforts, we predicted that male chimpanzees sniff more often at females (1) during their maximal swelling period, (2) when they are cycling regularly (are not pregnant or lactating), (3) when they have more reproductive experience (at an advanced age), and (4) when fewer competitors are around.

In total, we recorded 1817 sniffs directed to conspecifics during 184 h of observation, with an overall mean sniffing frequency of 0.64 sniffs per individual per hour. More than half of these (53.8%) were directed by males toward females. When males sniffed at females, they focused on the genital region in 58.1% of all events, but inspected other body parts in 41.9% of all cases [mainly back (∼18%), head/face/mouth (∼14%), arm (∼7%), hands (∼2%), and chest (∼1%)]. We observed 21 sniffing events (∼2%) where individuals clearly sniffed body fluids (vomit, urine, saliva) or feces. While in 95.2% of the cases individuals directly sniffed at a conspecific, in 4.8% the individual touched the conspecific with its hand and sniffed the own hand afterwards.

In the model containing swelling stage, cycle regularity, female age and male presence we found a clear impact of the test predictors on the number of male sniffs directed toward females (full-null model comparison (LRT): χ² = 19.174, df = 6, P = 0.0039). Specifically, swelling stage significantly affected sniffing frequencies (Table 2), whereby male chimpanzees preferably sniffed females that had an increasing or maximal swelling rather than females with a flat or decreasing swelling (Figure 1A). Males did not sniff regularly cycling females more often than females without a regular cycle (Table 2), but sniffed older females more often than younger females (Figure 1B and Table 2). Finally, the fewer male competitors were simultaneously present in the group (i.e., when some of them left the group to participate in behavioral studies), the more often males sniffed at females (Figure 1C and Table 2).

The results of our study indicate that sniffing behavior of chimpanzee males toward females is influenced by different female reproductive properties. Males sniffed more often at females with increasing swelling size, higher age, and when fewer male competitors were present, while cycle regularity did not influence male sniffing rates. Hence, our study strongly supports the idea that male chimpanzees use both visual and olfactory cues to assess female fertility.

The present study is in line with earlier observations in the wild that male chimpanzees use their sense of smell to inspect females (Matsumoto-Oda et al., 2007). The main finding of our study was that olfactory behavior of male chimpanzees was associated with a visual signal—the female anogenital swelling. Male sniffing frequency rose when female swelling size increased and decreased again after the maximal swelling period ended. Our findings are in line with patterns of male sociosexual behaviors, including olfactory inspection, that became more frequent with increasing sexual swellings of females in chimpanzees (Klinkova et al., 2005). Our result supports the hypothesis that males, in addition to using a visual signal (as indicated in Deschner et al., 2004), sniff at females according to their fertile states. Interestingly, male olfactory inspection was not restricted to the genital area of females, as males in our study sniffed at female body parts others than the genital area in half of the events we observed. This olfactory inspection of body and genital odor could potentially be used to pinpoint the timing of ovulation more precisely than is possible by visual inspection alone. Male chimpanzees preferentially join parties including females with increasing swellings (Matsumoto-Oda et al., 1998; Matsumoto-Oda, 1999; Boesch and Boesch-Achermann, 2000). Our results are supported by the findings of studies investigating the use of olfactory cues to assess female fertility in other primate species (Saguinus oedipus oedipus: Ziegler et al., 1993; Washabaugh and Snowdon, 1998; Macaca arctoides: Cerda-Molina et al., 2006b; Callithix jacchus: Kücklich et al., 2019).

Male chimpanzees altered their olfactory behavior also in relation to the age of females, i.e., males sniffed at older females more frequently than at younger females. Female reproductive potential of chimpanzees has been described to peak between the age of 14 and 24 years (representing the “younger” individuals in our study) and to sharply decline at the age of 25–30 years (Thompson et al., 2007). Based on this age-related decline in fertility, it would be reasonable that males prefer the younger and more fertile females. Nonetheless, in the wild, male chimpanzees seem to favor older over younger females, as suggested by higher copulation rates or greater likelihood to join a party in which older females are present (Muller et al., 2006), probably because they have more reproductive experience. Moreover, Thompson et al. (2007) found that chimpanzees staying healthy seemed to maintain a high fertility until later ages and the observed population-wide decline in fertility with increasing age was based on an increasing number of unhealthy individuals. In our study, age was correlated with parity (younger females were mostly nulliparous or primiparous, while all older females were multiparous) and thus the test predictor age comprises reproductive potential as well as reproductive experience at the same time. Indeed, previous studies found that nulliparous females were less attractive to males (Tutin, 1979; Hasegawa and Hiraiwa-Hasegawa, 1983). Multiparous females might be more interesting to the males because they could have observed these females reproducing and raising offspring successfully. Thus, for males, reproductive experience may be more important than reproductive potential, which remains to be investigated further in populations in which parity and age are less confounded than in our study group.

Relying on more than one sense is likely to be advantageous. Male chimpanzees might use both visual and olfactory cues to inspect females. The visual signal could act as an initial long-distance cue about the fertile state of a female, which may then elicit increased olfactory inspection from a short distance to gain additional and probably more precise information (Matsumoto-Oda et al., 2007). The olfactory information is potentially non-redundant to the visual one, because the information content might differ (see also Higham and Hebets, 2013). In olive baboons (Papio anubis), where females show variation in swelling stage, behavior, and copulation call rates during the fertile phase, males inspected female genitalia through sniffing more frequently in the fertile period as well (Rigaill et al., 2013). Based on these results, it was suggested that male baboons gather olfactory information on the onset of ovulation in addition to visual cues of the sexual swelling. The authors assumed that, while all males can access visual or auditory signals, only male consort partners are able to gain more information on the basis of odor cues obtained from the female in close spatial proximity to the male. Also in other primate species, it has been suggested that males that stay close to females, e.g., through mate-guarding, can obtain more and maybe more precise information about the onset of ovulation by using olfactory cues (Alouatta pigra: Van Belle et al., 2009; Papio anubis: Rigaill et al., 2013). In line with these suggestions, male chimpanzees regularly sniffing females might get more detailed information about the fertile state of the swollen female and thus could gain an advantage in siring offspring. However, it needs to be examined further, whether being close to females per se (triggered by the visual signal or not) elicits an increased rate of olfactory inspections or if males actually sniff more often at fully swollen females, compared to other (not fully swollen) females nearby. Moreover, it is not clarified whether the olfactory changes that males might gather by sniffing at females are “cues” or “signals.” Responses to cues are believed to need learning based on prior experience or conditioning, whereas signals provoke innate behavioral responses, although those innate responses could as well be modified through learning (reviewed in Beny and Kimchi, 2014). However, to unravel the underlying mechanisms, chemical analyses and bioassays would be needed.

Enhanced opportunities to gather olfactory information about females may be one reason why we observed individual males sniffing females more often when fewer male competitors were simultaneously present in the group. During our observations, the number of male competitors was sometimes reduced when the subjects volunteered in experimental studies outside the main enclosure. The resulting situation of leaving from and returning to the group can be regarded to reflect fission-fusion events similar to those in the wild. The temporal absence of certain males may have allowed the remaining males to get closer to females than in the presence of all male competitors. In this case, we would expect to observe increased rates of sniffing in lower-ranking males when higher-ranking males are absent. However, as we had no systematic dominance rank data available, further studies are necessary to examine this hypothesis.

Against our expectation, we found that sniffing frequencies of male chimpanzees did not differ between non-cycling and cycling females. Presumably, the visual signal, i.e., the presence (or absence) of a swelling, may have sufficed to induce high (or low) olfactory interest toward females irrespective of whether they showed full menstrual cycles or not. In fact, this supports the suggestion that the swelling provokes an increase in olfactory inspection, while information on whether a female is actually fertile or not may only become available from the olfactory inspection itself.

Overall, our study provides evidence that male olfactory behavior is associated with visual signs of female fertility in chimpanzees and therewith brings new insights into the field of great ape olfactory communication. However, certain aspects could not be addressed in the present study and these limitations should be investigated in future studies. First and foremost, this relates to hormone measurements which would allow us to pinpoint the timing of ovulation and thus refine our analysis. During the data collection for this study, we were not able to accomplish the extensive training of the individuals required for the collection of urine or saliva samples to investigate hormones. Second, sniffing data were collected from a single chimpanzee group with only six adult males. Given this limited sample size and scarce knowledge about olfactory fertility cues in catarrhine primates, further observations of other chimpanzee groups would help to compare to and confirm our findings. Additionally, systematic data on male and female dominance ranks or spatial proximity would allow differentiating between possible explanations for the results obtained here, and to widen the scope beyond the reproductive context. Despite these limitations, the data indicate that in a great ape species with an obvious visual signal of female fertility, males additionally appear to aim at collecting olfactory information, which sets a starting point for ongoing research and raises further questions about great ape olfactory communication in general and chimpanzee reproduction in specific. Finally, the chemical composition of female chimpanzees’ body odor, specifically perceivable chemical cues of female fertility, would complement the limited knowledge about great ape body odor.

In conclusion, our study associates olfactory behavior with female reproductive properties of a great ape species. Our results provide evidence that male chimpanzees sniffing behavior is closely linked to visual information on female fertility, which may allow males to collect information about female reproductive status using multiple modalities. Males in promiscuous species with varying group composition, where females provide only imprecise visual signals of fertility, could use additional olfactory cues to pinpoint the timing of ovulation and thus maximize their mating efforts to ultimately increase their reproductive success.

The original contributions presented in this study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by the Ethics Commission of the Department of Psychology of the Max Planck Institute for Evolutionary Anthropology and the Leipzig Zoo.

SJ, MK, BW, and AW conceived and planned the study. SJ and MZ carried out the observations and prepared the data. SJ, MK, LK, and BW planned and carried out the statistical analyses. SJ, MK, LK, BW, and AW contributed to the interpretation of the results. SJ and MK took the lead in writing the manuscript. All authors provided critical feedback and helped shape the research, analyses and manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.