American Journal of Primatology 37:263-269(1995) Drinking Tools of Wild Chimpanzees at Bossou YUKIMARU SUGIYAMA Kyoto University Primate Research Institute, Inuyama, Aichi, Japan Use of drinking tools by wild chimpanzees (Pan troglodytes) and the context in which the tools were used were studied a t Bossou, Republic of Guinea, West Africa. During the middle to late dry season and early wet season liquids are available occasionally in the holes of trees. Chimpanzees drank water or sap using a leaf (or fiber) as a sponge or spoon. When the chimpanzees were on the ground, they tended to use one of a few kinds of soft, hairless leaves, if they were available nearby. Females, particularly juveniles and adolescents, were thought to be the main users of the drinking tool. In a few episodes, a tool set was used to procure liquid. Once a chimpanzee used a stick to push a leaf sponge into a water hole and to pull it out from the hole. In addition, three chimpanzees used a pestle to squeeze sap from an oil-palm tree before using a fiber sponge. 0 1995 Wiley-Liss, Inc. Key words: chimpanzee (wild), drinking tool, leaf spoon, leaf sponge, pestle pounding, tool set INTRODUCTION Chimpanzees (Pan troglodytes) are well known to remove objects from their natural substrate and use them as tools. Sometimes chimpanzees manufacture tools for convenient and effective use in order to collect food more easily and for other purposes. Tool-using behaviors, however, vary greatly among local populations due to differences in environment and traditions [McGrew, 1992; Sugiyama, 19931. Use of leaves as a drinking tool is one example of tool use among wild chimpanzees in their natural habitat and was first recorded in the earliest stages of the study of wild chimpanzees at Gombe, western Tanzania [Goodall, 19641. Since th a t time, many long- and short-term field studies have been carried out on chimpanzees in their natural habitats. However, using tools for drinking has been reported at only a few chimpanzee habitats [McGrew, 19921. Moreover, this behavior and the circumstances under which it occurs have been described only briefly [Goodall, 1968; McGrew, 19771. The aim of this report is to present basic data on the use of drinking tools and and compare these data with the context of tool use for chimpanzees of BOSSOU, those in other populations, although observations are still low in number. Received for publication February 23, 1994; revision accepted January 18,1995. Address reprint requests to Yukimaru Sugiyama, Kyoto University Primate Research Institute, Inuyama, Aichi, 484 Japan. 0 1995 Wiley-Liss, Inc. 264 I Sugiyama MATERIALS AND METHODS Bossou is located in the southeastern corner of the Republic of Guinea, West Africa. The core-area of a chimpanzee (P. t. uerus) group is covered with dense primary and open secondary forests surrounded by cultivated and abandoned fields. All individuals have been identified since 1976 and the group size fluctuated between 22 and 16 individuals up until 1994. Within this period intermittent observations were carried out from 2 weeks to 6 months every year (except 1978, 1981, and 19841, between September and May, and data is available for over 1,000 days in total. Chimpanzees were observed without provisioning except between 1990 and 1993when about 10 bananas or oranges were given each day to the group for 1 to 2 months a year during field experiments [Matsuzawa, 19941. RESULTS Between mid 1977 and early 1994 a total of 40 chimpanzees (including some repeated performances of the same individuals) in 25 episodes of the use of drinking tools were recorded in their natural environment. Each episode consisted of either observation from discovery of drinking or related behaviors by one or more chimpanzees until they left the site, or the finding of a trace of drinking with abandoned tools. Out of the 25 episodes, 20 were direct observations of drinking behavior using tools as performed by a total of 35 chimpanzees. The other 5 episodes were findings of traces of tool use within a few hours after chimpanzees had left. Season of Tool Use for Drinking Figure 1 depicts the number of tool users for drinking in each month per 100 observation days. The figure also shows the mean monthly rainfall recorded for 8 years between 1957 and 1964 a t the Geologists’ Camp of Monts Nimba which is about 12 km from Bossou [Adam, 19711. Chimpanzees used drinking-tools in months when rainfall was low; however, a significant correlation was not found (Pearson’s correlation coefficient, r = -0.216). They began to use drinking tools when rainfall was at a minimum (January), and did so most frequently a t the beginning of the wet season (March) when rain was slightly increasing. The use of drinking-tools was usually seen within a few days after a heavy rain, when the forest floor was still dry but water was stored in the holes of large trees [McGrew, 19771. Water was usually deep inside the hole and chimpanzees could not drink directly with their mouths. At these times, they used a tool for acquiring water or other liquids. They continued to do so until May but with less frequency. No observation of tool use for drinking in April may reflect a n insufficient number of observation days. Location of the Liquid Ten species of host trees where liquid was stored were identified (Table I). Most of the host trees were more than 40 cm in diameter at the height of water holes. Twenty-eight water holes were less than 1.5 m from the base of the tree trunk. Another 12 were above 5 m in height (Table I). In 34 cases water was stored in the hole of the tree. Six others differed in that the liquid was sap squeezed out by a pestle from the tree crown of oil-palm trees (Elaeis guineensis). Tool Materials When a chimpanzee found a water hole high in a tree, she or he always searched for a nearby leaf or leaves in the tree. Three chimpanzees picked from one Drinking Tools of Chimpanzees I 265 frequency / rainfall 35 I 30 25 20 15 10 5 0 Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul month tool-use1100 ob.day Fig. 1. Seasonal frequency of tool use for drinking. Tool use frequency in each month is drawn as a number of performers and traces of tool use per 100 observation days. Number of observation days in each month is shown above the rainfall column. Data were not obtained between June and August. Rainfall is the mean of 8 years, 1957-1964, recorded a t Geologists’ Camp of Monts Nimba, about 12 km from Bossou [data from Adam, 19711. It is drawn x 0.5 cm, e.g., it rained 49 cm in August. to four small leaves in the tree where there were no large ones [Sugiyama, 19891. Chimpanzees a t low-level water holes usually used only one large leaf at a time. When water was found a t a height of less than 1.5 m, chimpanzees searched for a leaf on the ground around the tree. Nineteen out of 28 leaf sponges used a t low-level water holes were represented by only two species; ten were young leaves of Carapa procera and nine were mature leaves of Hybophrynium braunianum (Table I). Both have no hair and are soft. Details of selectivity of a particular kind of leaf from a field experiment are to be reported elsewhere. Three chimpanzees, when extracting sap from the oil-palm tree, used a mass of fibers, possibly detached by repeated pestle poundings, as a sponge. Age and Sex of Performers From 1976 to 1994, 35 chimpanzees (except infants of less than 4 years) have been recorded in one of four age-sex classes of the Bossou group: 11 males and 11 females as juvenile-adolescent (JA = 4-11 years), 5 males and 8 females as adult (A = 1 2 + years). Out of these 35, neglecting the difference of their observation frequency, 17 were observed to use drinking tools at least once; 4, 5, 1, and 7 individuals in each age-sex class, respectively. Differences between the age-sex classes were statistically significant (x2test, x = 3.34, df = 3, P < 0.05),with more adult females and less adult males a s the drinking-tool users. However, I hesitate to emphasize this result because most adult females stayed for more than 15 years in their class, whereas, many juvenile-adolescents remained in their classes for a only few years and then disappeared [Sugiyama, 1994bl. Out of 40 drinking-tool users, 34 in 20 episodes were identified in their age-sex 266 I Sugiyama TABLE I. Host Tree and Tool Material for Drinking* Plant sp. Highllow Host Tool” H L H L H L H L H L H L H L H L H L H L H L 12, L 3 15 0 2 0 0 0 1 0 3 1 0 0 0 1 0 1 0 10 6 0 1 0 0 9 0 3 0 2 Aningeria altissima (Sapotaceae) Antiaris africana (Moraceae) Blighia zuelwitschii (Sapindaceae) Bosquiea angolensis (Moraceae) Carapa procera (Meliaceae) Elaeis guineensis‘ (Palmae) Ficus umbellata? (Moraceae) Hybophrynium braunianum (Marantaceae) Myrianthus arboreus (Moraceae) Ricinodendron heudelotii (Euphorbiaceae) ? Total H = 4 6 1 0 0 0 0 0 1 0 4 = 28 3 0 2 40 40 Part used for tool Mat leafb Yg leaf Leaf Leaf Yg leaf Petiole, fibre Leaf Mat leaf Yg leaf Yg leaf 1 *High/low = height of the water (sap) source, higher or lower than 1.5 m. Mat = mature; Yg “Two or more leaves of the same species used as a sponge was counted as a single tool. bOnce a leaf was used as spoon (see text). ‘Sap was produced by pestle pounding (see text). = young. classes. Figure 2 shows the observed frequency of tool use and its expected frequency for each age-sex class. The latter was calculated from the accumulated number of chimpanzees of each age-sex class in the observed party for the 20 episodes (JAM = 49, J A F = 34, AM = 29, AF = 112, total = 224). Juvenileadolescent females used the drinking tool more than expected whereas adult males were seen only once. Adult females and juvenile-adolescent males, however, were observed to use it almost as frequently as expected by chance. My impression of observations coincides with this result as shown below. When a 7-year-old female was getting water from a tree hole with leaves, an adult female looked a t her but expressed little interest. An adult male displayed interest and approached; however, he went away after looking into the water hole [see Sugiyama, 19891. Drinking Method In seven out of ten episodes which were observed from start to finish, at least the first chimpanzees dipped a hand into the hole and licked it. After doing this one to three times they went to collect a leaf and returned. A 7-year-old female, and a 4-year-old male put an uncrumpled leaf into the tree hole, pulled it up and licked it as if it were a spoon. They repeated this more than three times before they left. At least 22 out of 32 performers (except episodes of leaf-spoon use and those at oil-palm trees) went to search farther than 2 m and came back to the same place, putting a leaf into their mouth. When they pulled it out from the mouth it was Drinking Tools of Chimpanzees I 267 frequency *O 15 i I 10 5 0 4-11 M >12 M 4-11 F >12 F age-sex class tool-user 0 expected Fig. 2. Tool users in each age-sex class. Expected frequency was calculated from the accumulated number of chimpanzees observed in tool-use episodes (n = 224) and that of tool users (n = 34)for each age-sex class. 4-12 = age in years; M = male; F = female. crumpled. They pushed it into the hole for 2-10 sec and pulled out a drenched, crumpled leaf before hastening to put it into their mouth. They then chewed it as if it were a drenched sponge. After sucking water from the leaf, they repeated a series of the soaking-and-sucking behaviors. In 12 episodes, two or more chimpanzees took turns performing the behavior. Mean duration of an episode was 13.3 min (n = 20, sd = 6.9). The longest duration of a chimpanzee’s work was 15 minutes and the largest number of consecutive soaking-and-sucking events for a chimpanzee was more than 42 by an adolescent male. The true mean duration of a continuous drinking performance must be longer than observed because several observations began when chimpanzees were already working. In all low-level tree holes, including 5 episodes where traces of drinking were found, the water surface was more than 20 cm below the mouth of the hole. In all holes, except two, it was difficult for me to touch my mouth directly to the water, either because the water was too low, the mouth of the hole was too small, or the hole was inaccessible because it was covered by a tree trunk. In two cases, I could place my mouth onto the water surface, but this forced me into an unbalanced position. Twenty-eight crumpled leaf sponges were collected around low-level tree holes. Mean length of a (extended) tool leaf was 13 cm (range = 8-20 cm); the diameter of a crumpled leaf sponge was about 3.5 cm. Use of a Tool Set Occasionally two different kinds of tools [a tool set; Brewer & McGrew, 19901 were used for obtaining liquid. In one episode, a 4-year-old female repeatedly soaked and sucked a crumpled leaf. Then, she broke a dead twig and used it to push the leaf-sponge deep into the hole. She pulled up the leaf with the stick and sucked the leaf. She repeated this process several times [Matsuzawa, 19911. Three chimpanzees in one episode a t an oil-palm tree behaved differently. After taking off a long, thick leaf stalk and biting and chewing its base, chimpan- 2681 Sugiyama zees pounded the center of the tree crown with the base of the leaf stalk a s if they were pounding grain with a “mortar-and-pestle.” Through more than 10 repeated poundings, a hole was made in the center of the tree crown and sap was squeezed into the hole. The chimpanzees then dipped their hands and licked them before gathering fragments of fiber from the bottom of the hole, perhaps detached by such poundings, and bringing them to their mouth to suck the sap. They repeated soaking-and-sucking of the fiber sponge [Sugiyama, 1994al. DISCUSSION During the dry season it is hot and desiccated a t Bossou. Chimpanzees must get thirsty as there is little water in their core area. There are, however, opportunities to find water in the holes of large trees. This kind of situation may not occur frequently in some other chimpanzee habitats. In wet and cool rain forests, such as Tai of Ivory Coast, chimpanzees may not suffer a water shortage and use of drinking tools is yet to be reported. Thus, the physical condition of chimpanzees and their environment may confine the use of drinking tools to only particular populations and restricted seasons. At a woodland in Gombe this behavior was observed more frequently [McGrew, 19771than in BOSSOU, yet, it is rare a t Mahale, located only 100 km from and in similar environment to Gombe [Nishida, 19941. Use of drinking tools, at a dry deciduous forest in Kibale, Uganda, was reported only recently even though the study has been conducted over many years [Ghiglieri, 1984; Wrangham, 19921. No other observation of this behavior has been reported from any study site on wild chimpanzees. Even if it occurs, the chimpanzees may perform it infrequently. Using a leaf sponge for drinking water appears to be one of the most simple cases of tool-using behavior in chimpanzees. Such a local difference may depend not only on the environment but also on local tradition. Chimpanzees of Gombe use a handful of dry dead blossoms, pieces of grass and twigs as well as leaves, numbering from one to eight, a t a time. On the other hand, those of Bossou use only leaves except in some special cases at the top of the oil-palm tree and a large leaf for making a sponge is more common than two or more leaves. Also, if on the ground, chimpanzees at Bossou tended to use a few special kinds of leaves (Carapa procera and Hybophrynium braunianum), although availability of these leaves was not assessed. Among chimpanzees of Bossou, use of drinking tools tends to be concentrated among females and immatures, particularly in juvenile-adolescent females. This trend is also evident among the chimpanzees of Gombe [Goodall, 19861. Infant chimpanzees of Gombe sometimes used the drinking tool at a stream where they could easily drink without any tool [Goodall, 19861. Similar behavior of leaf-spoon use in infants was also recorded at Gombe [McGrew, 19771. These behaviors possibly occur out of curiosity and may form part of the training course of infants and juveniles. chimpanzees usually crumple leaves Interestingly, both at Gombe and BOSSOU, in their mouth before soaking in water. Individually or traditionally they know the function of the “sponge.” This behavior, as well as the use of tool sets, suggests the high cognitive ability of wild chimpanzees. ACKNOWLEDGMENTS Throughout the study periods Direction Nationale de la Recherche Scientifique et Technique, Republique de Guinee, supported the field research. It was financed by grants under the Monbusho International Scientific Research Program. The field research has been carried out with J. Koman, K. Kumazaki, and Drinking Tools of Chimpanzees / 269 other colleagues, who provided me with unpublished data. W.C. McGrew and T. Matsuzawa kindly read and commented on an earlier draft. I express my heartfelt thanks to the above persons and organizations. REFERENCES Adam, J.G. FLORE DESCRIPTIVE DES MONTS NIMBA. Paris, Editions du Museum, 1971. Brewer, S.M.; McGrew, W.C. Chimpanzee use of a tool-set to get honey. FOLIA PRIMATOLOGICA 54:lOO-104, 1990. Ghiglieri, M.P. THE CHIMPANZEES OF KIBALE FOREST. New York, Columbia University Press, 1984. Goodall, J. Tool-using and aimed throwing in a community of free-living chimpanzees. NATURE 201:1264-1266, 1964. Goodall, J.v.L. The behaviour of free-living chimpanzees in the Gombe Stream Reserve. ANIMAL BEHAVIOUR MONOGRAPHS 1:161-311, 1968. Goodall, J. THE CHIMPANZEES OF GOMBE. Cambridge, Belknap Press, 1986. Matsuzawa, T. CHIMPANZEE MIND. Tokyo, Iwanami, 1991 (in Japanese). Matsuzawa, T. Field experiments on use of stone tools by chimpanzees in the wild. Pp. 351-370 in CHIMPANZEE CULTURES. R.W. Wrangham, et al., eds. Cambridge, MA, Harvard University Press, 1994. McGrew, W.C. Socialization and object manipulation of wild chimpanzees. Pp. 304309 in PRIMATE BIOSOCIAL DEVELOPMENT. S. Chevalier-Skolnikoff; F.E. Poirier, eds. New York, Garland, 1977. McGrew, W.C. CHIMPANZEE MATERIAL CULTURE. Cambridge, Cambridge University Press, 1992. Nishida, T. CHIMPANZEE OMOSIRO KANSATUKI. Tokyo, Kinokuniya, 1994 (in Japanese). Sugiyama, Y. Description of some characteristic behaviors and discussion on their propagation process among chimpanzees of Bossou, Guinea. Pp. 43-76 in BEHAVIORAL STUDIES OF WILD CHIMPANZEES AT BOSSOU, GUINEA. Y. Sugiyama, ed. Inuyama, KUPRI, 1989. Sugiyama, Y. Local variation of tools and tool use among wild chimpanzee populations. Pp. 175-187 i n THE USE OF TOOLS BY HUMAN AND NON-HUMAN PRIMATES, A. Berthelet; J. Chavaillon, eds. Oxford, Clarendon Press of Oxford University Press, 1993. Sugiyama, Y. Tool use by wild chimpanzees. NATURE 367:327,1994a. Sugiyama, Y. Age-specific birth rate and lifetime reproductive success of chimpanzees at Bossou, Guinea. AMERICAN JOURNAL OF PRIMATOLOGY 32:311318,1994b. Wrangham, R.W. Living naturally: Aspects of wild environments relevant to captive chimpanzee management. Pp. 71-81 in CHIMPANZEE CONSERVATION AND PUBLIC HEALTH, J. Erwin; J.C. Landon, eds. Rockville, Diagnon, 1992.