Current Biology Magazine

Correspondence receptivity and attractiveness are 2007 and 2013, the time period more extended in bonobos [2], males for which we sampled all potential Male reproductive compete less intensely for each mating sires, in the Bompusa community of opportunity. This would reduce the bonobos at LuiKotale, Democratic skew is higher strength of selection for traits that lead Republic of the Congo. Skew was to success in direct contest competition calculated from published paternity in bonobos than between males and in sexual coercion data from 41 temporally overlapping chimpanzees of females, thus increasing the potential 7-year time periods in fi ve chimpanzee for female choice [3]. Accordingly, it communities, which represented a total has been predicted that the infl uence of of 191 paternities occurring over 70 Martin Surbeck1,6,*, male dominance rank on reproductive independent chimpanzee community- 2,3,6, Kevin E. Langergraber *, success and the extent of male years (Table 1). Nonac’s binomial skew Barbara Fruth4,5, Linda Vigilant1 reproductive skew should be lower (B) index [6], where –1 indicates an equal and Gottfried Hohmann1 in bonobos than in chimpanzees [1]. distribution of reproduction, 0 a random Although relevant for understanding distribution, and 1 total monopolization The two closest living relatives of the evolution of the unusual levels of by one individual, was higher in bonobos humans, bonobos (Pan paniscus) and egalitarianism and cooperation found in (0.22) than in any chimpanzee period chimpanzees (Pan troglodytes), share human hunter-gatherers [4], comparative (range: –0.03–0.14, mean: 0.03; Table 1). many traits that are common in humans analyses in the genus Pan have been The percentage of paternities achieved but rare in other mammals, including limited by the scanty paternity data by the most reproductively successful societies with high fi ssion–fusion available for wild bonobos [5]. Here, male was also higher in bonobos (62%) dynamics, male philopatry, female we show using the largest sample of than in any chimpanzee 7-year period dispersal and extensive social bonding paternity data available that, contrary (range: 7–56%, mean = 26%; Table 1). among unrelated individuals [1]. The to expectation, male bonobos have a We also calculated skew in bonobos major difference between these two higher reproductive skew and a stronger after extending the analysis to a 12-year species is that male aggression is more relationship between dominance period (2002–2013) where we had a frequent and intense in male-dominated rank and reproductive success than larger sample of genotyped offspring chimpanzees than in bonobos, where chimpanzees. (n = 24) but lacked genotypes from the highest-ranking individuals are We compared skew in bonobos an estimated one to three candidate female [1]. One potential explanation is using paternities for 13 offspring sire males present at the beginning that because periods of female sexual conceived in the seven years between (2002 – 2006). Even if we assume

Table 1. Overview of male reproductive competition in one bonobo (bold) and fi ve chimpanzee communities over 7-year and 12-year time periods.

Average Most Number Average rank Time number of Number of Nonac’s B successful Species Community of time of sires period males per paternitiesa indexa sire’s sharea periods (95% C.I.) conceptiona [%] 2007– 0.81 Bonobos Bompusa 1 7.4 13 0.22 62 2013 (0.63–0.95) 1987– 6.3 10.5 0.09 51 0.77 Chimpanzees Taï North 2 1994 (6.1–6.5) (10–11) (0.07–0.11) (45–56) (0.60–0.91) 1985– 11.6 12 0.05 31 0.65 Chimpanzees Kasakela 21 7-year 2011 (11.3–12.3) (8–16) (-0.03–0.14) (13–45) (0.53–0.76) periods 1999– 0.91 Chimpanzees M group 1 11.6 11 0.06 36 2005 (0.84–0.97) 1993– 16.0 11 0.01 34 0.80 Chimpanzees Sonso 4 2002 (15.7–16.2) (11–11) (0.01–0.02) (27–36) (0.66–0.91) 1996– 41.8 42 0.01 11 0.66 Chimpanzees Ngogo 13 2014 (40.2–43.9) (30–54) (0–0.02) (7–13) (0.62–0.71)

2002– Bonobos Bompusa 1 - 23 0.08 39 - 2013 12-year 1996– 72.3 0.01 7 Chimpanzees Ngogo 8- - periods 2014 (62–80) (0.00–0.01) (6–8) 1985– 20.7 0.04 27 Chimpanzees Kasakela 16 - - 2011 (15–25) (-0.01–0.12) (16–41) avalues represent the average across all time periods for each group and the values in brackets indicate the range of values of all time periods for each group.

R640 Current Biology 27, R623–R641, July 10, 2017 © 2017 Elsevier Ltd. Current Biology Magazine

that all 3 offspring with unassigned lower estrous synchrony among females ACKNOWLEDGEMENTS paternity were sired by different does not appear to explain the stronger ungenotyped males, the B-index was infl uence of dominance rank on male Thanks to C. Rowney and A. Nicklisch for still higher in bonobos (0.083) than in reproductive success in bonobos than in performing genetics lab work, to R. Mundry all but two of 24 12-year periods in chimpanzees. for statistical advice, to P. L. Hooper and two chimpanzees (range = -0.010–0.123, If not an effect of the numbers of reviewers for helpful comments. The Congolese wildlife authority (ICCN) gave permission to mean = 0.028; Table 1). Additionally, competing males or simultaneously conduct fi eld research and export fecal samples the most successful male’s share of receptive females, how did high- for genotyping. Funding for bonobo fi eld and reproduction, which is unaffected by the ranking males more effectively genetics research was provided by the Max presence of ungenotyped males, was monopolize reproduction in bonobos Planck Society, the National Geographic Society higher in bonobos (39%) than in all but than chimpanzees? One possibility is and the Wenner-Gren Foundation. M.S. was 1 of 24 chimpanzee 12-year periods that higher levels of male aggression partially supported by SNF. K.E.L. is supported by (range = 6–41%, mean = 20%; Table 1). in chimpanzees actually decrease NIH RO1AG049395 and Arizona State University Comparative research in group- reproductive skew. Although rank and President’s Strategic Initiative Fund. The authors living primates indicates that male sexual coercion are positively correlated have no fi nancial confl icts of interest that reproductive skew is high when the in male chimpanzees, coercion by infl uence the results or interpretations contained number of competing males is low low-ranking males may also result in this paper. [7]. However, this does not appear in paternities that are unavailable for to explain why skew was higher in low-ranking bonobos [9]. Within-group REFERENCES bonobos than chimpanzees. Although infanticide also appears to be absent in 1. Hare, B., Wobber, V., and Wrangham, R. (2012). The we found that the chimpanzee periods bonobos. The need to confuse paternity self-domestication hypothesis: Evolution of bonobo with skew values most similar to those through promiscuous mating may thus psychology is due to selection against aggression. of bonobos also had a small average lower for female bonobos. Rather, the Anim. Behav. 83, 573–585. 2. Douglas, P.H., Hohmann, G., Murtagh, R., Thiessen- number of males present at the time low levels of sexual coercion and the Bock, R., and Deschner, T. (2016). Mixed messages: of each conception (Table 1), both the comparatively higher dominance of wild female bonobos show high variability in the timing of ovulation in relation to sexual swelling B-index and the most successful male’s female bonobos may allow them to patterns. BMC Evol. Biol. 16, 1–17. share of reproduction for bonobos were selectively mate with high-ranking males 3. Furuichi, T. (1992). The prolonged estrus of females outside the 95% confi dence intervals of during periods when they are likely to and factors infl uencing mating in wild groups of bonobos (Pan paniscus) in Wamba, Zaire. In the model prediction for a chimpanzee conceive, thereby gaining ‘good genes’ Topics in Primatology. 2. Behavior, Ecology, and community with the same number or some other benefi ts [10]. A further, Conservation, Y. Itoigawa, Y. Sugiyama, G. P. Sackett, and R. Thompson, eds. (University of of competing males (Supplemental non-exclusive possibility is that the Tokyo Press), pp. 179–190. information). more cohesive society of bonobos, with 4. Boehm, C. (2012). Moral Origins: Social Selection Male reproductive skew was parties consisting of a higher proportion and the Evolution of Virtue, Altruism, and Shame (New York: Basic Books). higher in bonobos than chimpanzees of the total community, makes it easier 5. Gerloff, U., Hartung, B., Fruth, B., Hohmann, G., because high-ranking bonobo for high-ranking males to monopolize and Tautz, D. (1999). Intracommunity relationships, dispersal pattern and paternity success in a wild males more effectively monopolized reproduction, as it limits the use of living community of Bonobos (Pan paniscus) reproduction. The average standardized alternative mating strategies that low- determined from DNA analysis of faecal samples. dominance ranks of sires was higher in ranking male chimpanzees use, such Proc. R. Soc. B Biol. Sci. 266, 1189–1195. 6. Nonacs, P. (2000). Measuring and using skew in the bonobos (0.81) than in four out of fi ve as sneaky copulation or consortship. In study of social behavior and evolution. Am. Nat. chimpanzee communities (mean = 0.77, addition to conducting paternity studies 156, 577–589. 7. Ostner, J., Nunn, C.L., and Schulke, O. (2008). range = 0.65–0.91). According to the in other bonobo groups to establish the Female reproductive synchrony predicts skewed priority-of-access model, the infl uence generality of our results, an important paternity across primates. Behav. Ecol. 19, of male dominance rank on reproductive area of future research will be to collect 1150–1158. 8. Altmann, S.A. (1962). A fi led study of the success decreases with the degree of the detailed behavioral data necessary sociobiology of rhesus monkeys, Macaca mulatta. estrous synchrony: when one female for evaluating the specifi c ways in Ann. N. Y. Acad. Sci. 102, 338–435. 9. Muller, M.N., and Wrangham R. W. (2009). Sexual is maximally tumescent, the alpha which male competition and female Coercion in Primates and Humans: An Evolutionary male will sire her offspring, when two choice interact to result in the large Perspective on Male Aggression against Females females are simultaneously maximally rank effects and high reproductive skew (Cambridge: Harvard University Press). 10. Møller, A.P., and Alatalo, R.V. (1999). Good-genes tumescent, the alpha and beta male reported here. A clearer understanding effects in sexual selection. Proc. R. Soc. Biol. Sci. will each sire one offspring, and so on of the phylogenetic building blocks 266, 85–91. [7,8]. However, the average number and functional mechanisms underlying 1 of maximally tumescent females at the evolution of the social and mating Primatology Department, Max Planck Institute for Evolutionary Anthropology, Leipzig, 04103, each conception was actually higher systems of humans is likely to emerge Germany. 2School of Human Evolution and Social in bonobos (mean = 4.9, range = 1–9) as a consequence. Change and 3Institute of Human Origins, Arizona than in chimpanzees (mean = 3.8, State University, Tempe, AZ 85281, USA. 4Centre for Research and Conservation, Royal Zoological range = 1–11). In addition, in bonobos, SUPPLEMENTAL INFORMATION but in none of the chimpanzee Society of Antwerp; 2018 Antwerp, Belgium. 5 communities, the alpha male sired School of Natural Sciences and Psychology, Supplemental Information including Liverpool John Moores University, Liverpool, L3 signifi cantly more offspring than experimental procedures and one table can 3AF, United Kingdom. Co-fi rst authors expected under the priority-of-access be found with this article online at *E-mail: [email protected]; model (Supplemental information). Thus, http://dx.doi.org/10.1016/j.cub.2017.05.039. [email protected]

Current Biology 27, R623–R641, July 10, 2017 R641