Maternal caretaking behavior towards a dead juvenile in a wild, multi-level primate society

Maternal caretaking and transport of dead infants are widespread among nonhuman primates, having been reported in numerous species of monkeys and apes. By contrast, accounts of such behaviors toward dead juveniles are scarce. Here, we describe responses by the mother and other group members to the death of a juvenile in a wild, multi-level group of Sichuan snub-nosed monkeys (Rhinopithecus roxellana). Following the juvenile’s fatal accident, his mother transported and cared for the corpse for four days. Immature monkeys belonging to the same one-male unit, and some individuals from other social units also showed interest in and tended the corpse. Comparisons of this case with those involving the deaths of infants and an adult female in the same population highlight possible effects of physiological, psychological and emotional factors in primate thanatological responses, and provide an additional perspective on the origin and evolution of compassionate acts.

Responses to dying and dead conspecifics across animal species reveal a continuum ranging from hard-wired fixed-action patterns to varied and flexible behaviors with cognitive and emotional correlates 1 . Describing and studying these responses can clarify adaptive and possible maladaptive behaviors in various species, as well as the underlying biological (including psychological) mechanisms in death-related responses [1][2][3] . The comparative thanatology literature includes many reports of responses toward corpses in nonhuman primates: prosimians, monkeys and apes [2][3][4] . In particular, care and transport of dead infants by their mother and sometimes other individuals has been described in a range of primate species including, among others, capuchin monkeys (Sapajus sp.) 5 , macaques (Macaca sp.) 6,7 , baboons (Papio sp.) 8,9 , geladas (Theropithecus gelada) 10 , chimpanzees (Pan troglodytes) [11][12][13][14][15] , and mountain gorillas (Gorilla beringei beringei) 15,16 . Based on behaviors such as social withdrawal and reduced interest in the environment, many authors have suggested that mothers continuing to care for and look after their lifeless infants are experiencing grief. The mechanisms underlying these behaviors are still debated, but the interaction of maternal physiological and emotional factors, including the strength of the bond with the offspring seems highly likely to be involved 14,17,18 .
Although the phenomenon of mothers continuing to carry and care for dead infants is widespread among anthropoid primates, accounts of such responses to dead juveniles are notably lacking. For example, researchers studying Yunnan snub-nosed monkeys (Rhinopithecus bieti) 19 and Sichuan snub-nosed monkeys (Rhinopithecus roxellana) 20,21 reported carrying and care of dead infants, but not juveniles, and dead infant geladas (Theropithecus gelada) received more attention than dead juveniles and adults 10 . However, in chimpanzees, dead adults and subadults may receive more attention than dead infants 22 , although the larger corpses are not carried. Mortality in infant primates is much higher than in juveniles, which can at least partly explain the prevalence of accounts of infant deaths and postmortem care. More descriptions of responses to dead juveniles are needed, to compare and contrast responses to corpses belonging to different age categories across species.
Here, we report the first known case of maternal transport and caretaking of a dead juvenile, along with responses by other individuals, in a wild group of Sichuan snub-nosed monkeys. These monkeys live in temperate forests on mountain plateaux in central and southwestern China, at 1500-3400 m. They are described as living in a multi-level society, consisting of several one-male units (OMU) and associated all-male units (AMU) 23 www.nature.com/scientificreports/ and sub-adult females, juveniles and infants [24][25][26][27][28][29][30] . Within this society the monkeys have multiple levels of social interactions and relationships [23][24][25][26][27] . Although most affiliative behaviors occur within OMUs, infants and juveniles often visit other units and form play groups of 10-30 individuals; other behaviors such as grooming and cofeeding may also occur [23][24][25][26][27] . Endemic to China, Sichuan snub-nosed monkeys are seasonal breeders, in which mating occurs from September to December, and births occur from March to May 21,28-30 . Newborn infant snub-nosed monkeys are at risk of dying from cold weather, congenital abnormality or disease, and other, less common causes such as accidents or infanticide 20,21,31 . To contextualize the responses observed toward the dead juvenile described here, we compare and contrast this case with previously described responses toward dead infants and and an adult female in the same population 20,21,32 . This account highlights the multiplicity of factors involved in primate thanatological behaviors. It thus contributes further evidence for variability in primate reactions, and to the debate surrounding death-related compassion and grieving in nonhuman species 18,[32][33][34][35][36] .
The study group consisted of 92 individuals organized in seven one-male units (OMUs) and one associated all-male unit (AMU) in June 2016. The OMU in which the death occurred contained an adult male (SBW, three adult females (AF1: CM, AF2: XW, AF3: JD), five juveniles (J1, J2, J3, J4, and CM's offspring: CM-J) and one infant (JD-I). We recorded the reaction of the OMU members to the dying juvenile using focal-animal, ad-libitum sampling 37 . We observed through binoculars when necessary (Nikon 7245 Action Ex Extreme 10 X 50 mm), recorded times with a stopwatch (Casio Men's AE1000W-1B, wrote behaviors in a notebook, and took photographs (Canon cameras EOS-1D X, EOS 700D, and EOS-5D Mark II). The death occurred at the Daping valley (107°59′6.5688"E, 33°40′19.0164"N), well inside the group's home range. The study adhered to the legal requirements of the Guanyinshan National Nature Reserve, China, and the guidelines of the International Primatological Society for the ethical treatment of primates. The study was approved by the animal care committee of the Wildlife Protection Society of Shaanxi Province, China (permit number: SX43537ACC).

Results
Here, we present the main events observed over a 5-day period covering the juvenile's death, maternal care and transport of the corpse, and ending on the day following abandonment of the corpse. Based on the ad-libitum observations, Table 1 summarizes selected behaviors of the mother before the death of the juvenile and during the days when she carried and cared for the corpse. Table 2 summarizes other group members' responses to the juvenile before and after his death, and after abandonment by the mother, and Table 3 summarizes their responses to the mother. The descriptions below focus on some of the most notable events observed. A comprehensive photo collage with a more complete timeline including all behavioral descriptions extracted from notes and photos is presented as online supplementary material.  Figure 1A shows some of the behaviors seen during the next few minutes. CM-J's mother (CM) immediately ran over, held and inspected the inert juvenile, and then mouthed and groomed him, while intermittently emitting occasional alarm calls. Juve- Table 1. Comparison of mother's behavior before and after the day that CM-J died.

Behavior
Before CM-J dead After CM-J dead www.nature.com/scientificreports/ nile J3 climbed down the tree, then approached, looked at, and lightly touched CM-J; he then gently mouthed CM-J's face, and groomed the body. Other members of the OMU looked on from slightly farther away. Adult female JD stopped grooming SBW (the presumed father of CM-J), who yawned rapidly in succession while sitting on a tree branch above the dead juvenile. Another adult female (XW) arrived, and approached SBW. Juveniles J1 and J2 watched CM and the corpse from slightly farther away, embracing each other. JD resumed grooming SBW, and XW moved closer to them. Social grooming among other OMU members farther away appeared exaggerated, with some individuals appearing anxious or disconcerted. Over the next two hours, CM alternated between sitting silently by CM-J's corpse, and grooming it. At 08:19 she picked it up, carried it for about 30 m, then placed it on the ground and started feeding. At 10:01, she carried it into a tree where she rested, and groomed it. At 12:59 she carried it down from the tree, and at 13:01, gripping its shoulder with one hand she brought its face against her chest (Fig. 1B). At 13:03, she turned the corpse to face away from her, and held it against her chest with one hand as she clumsily walked forward. At 13:07 she again brought its face against her chest, and struggled to walk while clasping the corpse (Fig. 1C), before laying it on the ground and feeding. At 14:52, as CM fed about 2 m away from the corpse, an extra-unit juvenile touched and groomed it. An extra-unit infant approached, looked at, touched and briefly groomed it, then left. At 15:05, CM carried the corpse into a nearby tree, and alternated between grooming it and sitting motionless.
At 17:05 the group started to leave the area and move up into the mountain toward a sleeping site. CM carried the corpse down from the tree in which she had been resting. At 17:09, she held the corpse (facing outward) against her chest with both hands, and emitted "Wa Wa Wa" calls. A juvenile from another OMU approached and looked at the mother-corpse pair. Starting at 17:10 and for approximately 7 min CM alternated between holding the corpse to her chest with one hand and dragging it along the ground as she walked, and placing it on the ground. She then put the corpse on the ground and vocalized, possibly in response to people nearby (Fig. 1D).
At 17:17 she put the corpse on the ground, and was left alone as other monkeys moved farther away. SBW looked and called towards CM, who simply looked back at him. The entire group appeared to move more slowly than normal, and their sleeping site for this night was less far up the mountain than normal.  (Fig. 1E). Extra-OMU juveniles (teal arrow, cream arrow) approached and/or looked, while an extra-unit adult female also watched (light blue arrow).  www.nature.com/scientificreports/ At 06:20, CM laid the corpse down on the large stone on which she was sitting. She picked up a few leaves from the ground and placed them on the corpse's back. A juvenile approached (Fig. 1F)  www.nature.com/scientificreports/ at 17:21 she stopped, put it on the ground, and looked in the direction where the group had gone (Fig. 1I). As on the previous day, the group moved notably slower than usual; this was especially true for CM's unit.
Day 4, 1 July: abandonment. At 11:06, having eaten, the group was resting in the feeding area. However, CM was still on the road approximately half-way up the mountain. As she moved she tried repeatedly in vain to pull CM-J's corpse, which had further decomposed; it had lost more fur, some of which was scattered across the ground, its exposed flesh wet and slippery, and emitting a pungent odor. CM sat beside it, then moved to rest in a nearby tree, where she self-groomed. The now almost furless corpse was infested by maggots, and the head was unrecognizable (Fig. 1J). At 11:27 CM continued sitting in the tree and self-groomed, occasionally looking at the corpse and in the direction of the group. Shortly thereafter she rejoined her unit, leaving the corpse where it was. At 11:56 CM sat next to SBW, with XW alongside. At 11:57 she ate while sitting beside the male. At 12:00 and still feeding, CM occasionally looked in the direction of the hillside (out of view) where she had left the corpse, and occasionally uttered the plaintive contact call "A wo wo" (Fig. 1K). This is a different kind of sound from the usual contact sound made while foraging, typically used when an individual gets separated from the group or searches for a specific individual. At 12:10 CM approached and groomed SBW, while a juvenile groomed CM. At 12:18 she sat alone while the other members of her unit fed. She stood up and walked to a large nearby boulder and sat down again, facing and occasionally looking in the direction of the hillside as before; she sometimes self-groomed. At 12:23 CM joined SBW and groomed him (Fig. 1L).
At 12:56, due to the unpleasant odour and concerns about possible disease risk, a field assistant started to bury CM-J's remains. Some monkeys became aware of the burial, approached (CM was not among them), and occasionally alarm called from up in the trees. Later that afternoon, CM groomed with other members of her unit. At 17:00, the group moved up the mountain, notably faster than the previous day, and covered a greater distance to reach their sleeping site. Day 5, 2 July: post-abandonment. The group's overall behavior appeared normal; in particular, CM's locomotion, feeding and socializing appeared unremarkable, including grooming and being groomed by her unit's females. However, she occasionally emitted plaintive contact calls; for example, at 12:40 when, after eating alone, she emitted a series of: "A wo wo" vocalizations.
The distance traveled by the mother and the group was calculated for the 3 days before CM-J's death (mean: 3,427.0 m; SD = 95.9 m), the 3 days following the death (mean: 1,310.7 m, SD = 117.0 m), and the 3 days after CM abandoned the corpse (mean: 3,699.3 m, SD = 182.5 m) (Fig. 2). These distances reflect overall slower movement and and reduced ranging by the group during the days when the mother was carrying the juvenile's corpse, compared to the pre-death and post-abandonment periods.

Discussion
Reports of responses to dead conspecifics in primates indicate wide variability, with the setting, cause and context of death, age and sex of the dead individual and the nature of the social relationships between the latter individual and other group members all potentially likely to influence those responses 2,38 . The most striking feature of the present case, the first concerning a dead juvenile Sichuan snub-nosed monkey, was the mother's heavy investment in carrying and caring for her juvenile's corpse; she was in physical contact with or proximity to the corpse before abandoning it on the fourth day after death. Despite the corpse being considerably larger and heavier than that of an infant, several of the mothers' actions toward the corpse-carrying, peering, gently mouthing face, grooming, embracing and gentle pulling-are highly similar to those reported in mothers of dead infants, and strongly suggest a continuing emotional attachment. Furthermore, the mother periodically gave alarm calls, usually emitted in response to some kind of danger such as the approach of a potential predator, suggesting protectiveness.
Although she was clearly protective of the corpse, the mother did allow infants from her unit and other immatures to approach and contact it ( Table 2). Mothers in other primate species are reported to tolerate some close inspection and manipulation of their dead infant by others, especially kin, but they also spend more time Figure 2. The average distances covered by the group and mother before, the juvenile's death, after death, and after corpse abandonment by the mother. www.nature.com/scientificreports/ alone-perhaps partly due to others avoiding the smell of the decaying corpse 7 -and may reject some approaches and attempts to contact it 6,7,9,18,39,40 . Intra-and inter-species variability in maternal tolerance of access to dead offspring by other members of the group remains to be more thoroughly studied. The overall demeanor of the bereaved mother was consistent with being in a state of grief, consistent with other reports of bereavement in primates 11,14,[33][34][35][36] . For long periods she simply sat alone with the corpse, in silence but for occasional contact vocalizations usually made while searching for a missing individual. Furthermore, in contrast to most mothers with a live offspring, she generally avoided other individuals; she was less often in proximity to them and received less grooming than before the death. The amount of grooming received by the mother recovered after she abandoned the corpse, similar to but faster than in mothers with dead infants 20 . This might suggest disruption to social interactions due to the mere physical presence of the corpse, or simply a faster recovery from any grief response to the loss of a juvenile compared to an infant; further data are required to clarify this issue. Shortly after finally abandoning the corpse, she repeatedly looked back to where she had left it, emitting contact calls. However, by the following day her behavior appeared to have returned to normal in terms of movement, feeding, and social interactions. We have no physiological data covering the period of this study, which is unfortunate as maternal hormonal status can provide valuable information about the bereaved mother's emotional state 41,42 . Some of the mother's behaviors were untypical of mothers with live offspring, such as placing leaves on the corpse, and manually bringing its head to her nipple. The function of the former, if any, is unclear, but it might be similar to Taï chimpanzees' behavior of dropping vegetation onto some corpses, perhaps testing for a reaction 43 . Holding the dead juvenile's face against her nipple recalls the observation of an adult female Hanuman langur with a dead 8-month-old infant sucking her nipple until milk was expressed and then bringing the infant's head into contact with the nipple 44 . These acts appear to be attempts by these mothers to elicit suckling, which would suggest that they did not fully understand that the juvenile was dead, or that their understanding of death lacked the non-functionality and/or irreversibility subcomponents of the death concept typical of humans 1,3,38 . Additonally, the continued contact calling by the mother even after she abandoned the corpse might also indicate lack of its non-functionality and irreversible condition.
In contrast to some of the younger monkeys, adults generally showed no increased affiliative behavior toward the bereaved mother while she carried the corpse (Table 3), as well as relatively little interest in the corpse. However, they travelled more slowly than normal and occasionally vocalized towards the mother, suggesting that there was not total indifference toward the mother-dead juvenile pair. Nonetheless, this general lack of interest contrasts with interest and affiliation towards a dying adult female seen in other adults of the same OMU in this population of Sichuan snub-nosed monkeys 32 . In that case, the OMU adult male was particularly strongly bonded with the female, and for around 10 min after she died he remained with her corpse, gently manipulating and pulling at it. He abandoned the corpse hesitatingly, and gazed alternately between the corpse and the rest of the group before finally leaving.
Starting at about 3 months of age, infant snub-nosed monkeys can leave their mother and unit to form play groups with immature members of other family units [23][24][25][26] . By about 10-12 months age they are almost fully independent: most are weaned, they spend more time playing with unit and extra-unit juveniles and infants, and are carried and groomed by the mother much less than before [23][24][25][26] . Presumably, infants and other juveniles in the group had become friends through growing up and playing with the CM-J before his fatal accident, and so were naturally curious about what had happened to their companion. By approaching and engaging with the corpse youngsters may learn about death. Older individuals, by contrast, are more likely to have already experienced death events and so are less curious about it 34,[39][40][41]45,46 ; hence the overall lack of interest shown in the corpse by most adults. The strength of the mother's bond with her juvenile, however, caused her to continue transporting and caring for him even after his death.
Previous reports of responses to dead newborn infants in snub-nosed monkeys indicate that interactions with the corpse were restricted to only the mother and siblings 20,21 . By contrast, several immature individuals with close social ties with CM-J when he was alive were seen to gently contact his corpse. These approaches continued up until the mother abandoned the corpse, after which they ceased, but some alarm calls continued to be emitted, possibly also reflecting the influence of social bonds.
Climate may influence corpse-carrying behavior 10 . The death-related events described here took place during the summer, when the combination of rapid decomposition, flies, and maggots invading the increasingly naked and wet corpse conceivably hastened its abandonment by the mother. Systematic comparisons of reactions to corpses at different times of year are required to better understand the influence of climatic factors on responses to corpses.
In human societies, even genetically unrelated people may care for sick or dying others, especially if they are socially close. What appears clear from the accumulating comparative literature, including the present report, is that compassionate caretaking is unlikely to be unique to humans; at least some elements may be found in other primates. Although we need to guard against anthropomorphism when interpreting reactions other species 47 , arguably especially those living in multi-level societies, we can conclude that deaths in those species appears to activate some of the behavioral and affective mechanisms that underlie the varied and complex thanatological practices that characterize human societies. Case-studies that are largely descriptive, such as the present one, are important for building a more complete picture. Where possible, however, more quantitative data are desirable: multiple variables including social interactions, other behavioral activities, age-sex class of dead individuals and surviving kin, social status, maternal parity, and climatic factors are all of interest. Supplemented with behavioral and physiological measures, such information will help to further develop the growing field of comparative thanatology, clarifying other species' conceptions of death.