Chimpanzee (Pan troglodytes troglodytes) tool use in the Ngotto Forest Central African Republic.код для вставкиСкачать
American Journal of Primatology 65:221–237 (2005) RESEARCH ARTICLE Chimpanzee (Pan troglodytes troglodytes) Tool Use in the Ngotto Forest, Central African Republic THURSTON C. HICKS, ROGER S. FOUTS*, and DEBORAH H. FOUTS 1 Chimpanzee and Human Communications Institute, Central Washington University, Ellensburg, Washington Over a 7-month period, stick tools constructed by chimpanzees were collected and measured at the Ngotto Forest site in Central African Republic. The chimpanzees were found to use tools to dip for ants and to probe for honey. The basic descriptions of these tools and the contexts in which they were found are presented. The lengths of two of the tool types were compared with the use of a t-test for independent groups. It was found that the lengths of the tools differed significantly depending upon their function. The location and habitat type of each tool site were plotted on a map. The tool types were distributed throughout the southern part of the study area, and with one exception all tool sites were found in the same type of habitat. Two tool sites with two other types of tools (honey hammer/club and ant club) were found. The tool types at Ngotto are compared with those found at other chimpanzee field sites, and the implications for diversity in chimpanzee material culture are discussed. Am. J. Primatol. 65:221–237, 2005. r 2005 Wiley-Liss, Inc. Key words: Pan troglodytes troglodytes; tool use; Ngotto Forest; tool standardization; ant dip; honey probe INTRODUCTION Free-living chimpanzees (Pan troglodytes) have developed a diverse set of tool traditions that are considered by some authors to represent culture [Whiten et al., 2001]. Goodall  first discovered that chimpanzees not only use tools but modify them prior to their use. Since the time of that discovery, each population of chimpanzee that has been studied has been found to possess its own unique tool kit [McGrew, 1998]. Some components of these kits are shared between populations, while others appear to be unique to particular populations. Contract grant sponsor: Arcus Foundation; Contract grant sponsor: Friends of Washoe. n Correspondence to: Roger S. Fouts, Chimpanzee and Human Communications Institute, Central Washington University, 400th East University Way, Ellensburg, WA, 98926-7573. E-mail: email@example.com Received 18 June 2004; revised 1 September 2004; revision accepted 28 September 2004 DOI 10.1002/ajp.20111 Published online in Wiley InterScience (www.interscience.wiley.com) r 2005 Wiley-Liss, Inc. 222 / Hicks et al. Whiten et al.  considered 65 behavioral patterns from nine long-term chimpanzee study sites as candidates for cultural behavior. To be considered a cultural variant, a behavior had to be customary or habitual at one or more sites but absent from others, where the absence of the behavior could not be explained by environmental differences. Of the 65 candidate behaviors, 39 were accepted as cultural variants. The majority of these could be considered tools. Of these, 36 involved the use of an external object not including another chimpanzee, and 26 involved the use of an object by a chimpanzee as an active extension of the body (as opposed to a substrate for an action). The evidence for cultural diversity in chimpanzees has been accumulating so rapidly that Wrangham et al.  suggested it would be productive to construct an ethnography of chimpanzee cultural variants, as is practiced by researchers of human cultures. Few large-scale population patterns of chimpanzee tool use have been discerned [McGrew, 1992]. One of these is the cracking of nuts with wood or stone hammers, which appears to be unique to Western chimpanzees (Pan troglodytes verus), and occurs in many P. t. verus study sites to the west of the Sassandra-Izo River, Ivory Coast. Boesch and Boesch-Acherman  found no evidence of nut-cracking behavior in P. t. verus populations on the east side of the SassandraIzo, despite the availability of hammers and appropriate nut species (but see Joulian ). The behavior has not been observed in the chimpanzee populations that have been studied in Eastern and Central Africa [McGrew et al., 1997]. Another possible large-scale pattern involves the use of brush-sticks [Sugiyama, 1985] or sticks used as picks to open up termite mounds [McGrew & Rogers, 1983; Sabater Pi, 1974], both of which have been found only at Central chimpanzee (Pan troglodytes troglodytes) study sites [McGrew, 1992]. Beyond these two uses, tool traditions seem to be distributed rather idiosyncratically among chimpanzee populations, and no clear patterns between the subspecies are discernible. For example, Eastern chimpanzees (Pan troglodytes schweinfurthii) at Gombe, Tanzania, commonly use sticks to dip for driver ants (Dorylus sp.) [Goodall, 1964; McGrew, 1974]. Chimpanzees of the same subspecies at the nearby long-term research site of Kasoje (Mahale, Tanzania) show no evidence of this behavior, even though ants are plentiful there [McGrew, 1992]. However, the chimpanzees use a different technique (ant fishing) to acquire Camponotus ants [Nishida, 1973]. Likewise, at Okorobikó, Equatorial Guinea, Central chimpanzees use tools to obtain Macrotermes termites [Jones & Sabater Pi, 1969], while at Lopé, Gabon (another Central chimpanzee study site), this insect species is present but is never eaten [Tutin et al., 1995]. Chimpanzee populations at opposite ends of the continent use similar tools to acquire the same species of insect. For example, Macrotermes are fished at Gombe, Tanzania, and Assirik, Senegal [McGrew et al., 1979], as well as at other sites; however, there are often subtle differences in the construction of these tools between sites. For example, Gombe chimpanzees used stick tools with a mean length of 66 cm [McGrew, 1974] to dip for ants, whereas the tools Tai chimpanzees used for the same purpose have a mean length of 24 cm [Boesch & Boesch, 1990] and achieve inferior results. At Bossou, Guinea [Humle & Matsuzawa, 2002], chimpanzees use both long and short tools, each of which is used for different-colored species of driver ants and require different fishing techniques. Previously unknown examples of tool use have continued to be discovered, both at older long-term sites, such as Bossou [Hirata et al., 1998], and in less wellstudied populations, such as at Tenkere, Sierra Leone [Alp, 1997]. Each population of chimpanzees studied offers the potential to expand the database Chimpanzee Tool Use in the Ngotto Forest / 223 of chimpanzee cultural variations, both in the description of new tools and in the documentation of alternative ways of making tools or different uses for tool types that are shared with other sites. In addition to culture, tool preparation provides an ideal opportunity to investigate the flexibility of chimpanzee behavior in selecting and constructing appropriate tools for the job. Boesch and Boesch-Acherman  described standardization of stick foraging tools based on tool function by the Tai Forest chimpanzees. The honey-fishing tools made by these chimpanzees were significantly longer than their ant-dipping tools, and both of these types of tools were significantly longer than the nut-emptying and bone marrow-scraping tools. Similarly, Stanford et al.  reported that Eastern chimpanzees at Bwindi, Uganda, manufactured short stick tools to obtain honey from stingless-bee (Meliponula bocandei) nests, and significantly longer, thicker tools to acquire the honey of African honeybees (Apis mellifera). A population of chimpanzees (estimated density=0.78–1.06 weaned individuals/km2 (Hicks et al., unpublished results)) inhabits the dense, semideciduous, moist, lowland Ngotto Forest on the north bank of the Mbaéré River in Central African Republic (Fig. 1). These chimpanzees belong to the Central African subspecies Pan troglodytes troglodytes, which remains relatively unstudied compared to the Eastern and Western subspecies. This paucity of knowledge extends to their tool-use traditions [McGrew, 1992]. In this paper we provide a detailed description of the different types of chimpanzee tools found during a 7month survey of the Ngotto Forest to the south of the village of Grima, and compare the tools with tool types found at other sites. In addition, the lengths of two of the tool types discovered at Ngotto (ant-dip tools and honey probes) are compared with one another. We expected that the chimpanzees would construct stick tools of different lengths depending on whether they were using them for honey-probing or ant-dipping. MATERIALS AND METHODS Participants The chimpanzees (Pan troglodytes troglodytes) of the Ngotto Forest live in a semideciduous, moist, lowland tropical forest/swamp environment approximately 8 km south of Grima, the village which served as the project base (41010 N, 171040 E) [Brugiére et al., 1999] (Hicks et al., unpublished results). They share the forest with western lowland gorillas (Gorilla gorilla gorilla), and their diet consists primarily of fruits and leaves (personal observation). Roger and Deborah Fouts initiated the Grima project in 2000 in an effort to document the gestural and material traditions of an unstudied population of chimpanzees, as well as to promote the conservation of the chimpanzees and gorillas living in the forest. Throughout the duration of the project they coordinated the research and conservation efforts from the United States. The study was conducted between the dry and early wet seasons, from November 2001 to May 2002. The research team was made up of Thurston Hicks, Research Director Philip Hoffman, and six Bofi-speaking assistants from the village of Grima, who were experts in identifying the flora and fauna of the region. Since this population of chimpanzees was not habituated to humans, only indirect trace evidence, such as tools and feeding remains, could be obtained. During the study period the northeast section of the forest was being selectively logged by the timber company Industrie Forestière du Batalimo (IFB), and logging transects had recently been cut across the northern part of the study site. The southern forests adjacent to the Mbaéré 224 / Hicks et al. Fig. 1. Map of the Ngotto Forest study region between the Lobaye and Mbaéré rivers in Central African Republic. River had not yet been surveyed by the timber company, and this is where the research team made the majority of its ape contacts and found a large number of ape nests. Chimpanzee Tool Use in the Ngotto Forest / 225 Materials The materials used in this study consisted of a steel tape measure with plastic casing (in units of inches), data notebooks and pens, and a GARMIN 2001 model (Garmin AT, Inc., Salem, OR) geopositional satellite (GPS) device that recorded the coordinates of the tool sites. A 35-mm camera was used to photograph the tool sites. The GPS Utility mapping program (version 4.04.0, copyright Alan Murphy, 1998–2002, licensed to Friends of Washoe, CAR Project, Grima) was used to construct the tool map shown in the Results section. Procedure We found tools in the forest while walking along survey transects or taking random forest walks in the 87 km2 study area. For the purposes of our study, the forest types were divided into seven categories, as defined in Table I. In his survey of 92,829 ha of the Ngotto Forest north of the Mbaéré River (which included our study area), Bastin  divided the forest into nine types, only seven of which were present in the study area of the current project. Three of Bastin’s categories corresponded to our dense, mixed, and open forest areas, and made up 21.6%, 31.2%, and 20.8%, respectively, of the forest area surveyed by his team. In all, 13% of the forest surveyed by Bastin was swamp forest, and 1.6% and 1.9% were logged forest and degraded forest, respectively (Bastin’s remaining forest types were rare or absent in our area). Although we made no exact measures of the relative proportions of the forest types that we surveyed, we found (as did Bastin) that mixed forest was the most common vegetation type, followed by dense forest and then open forest, with swamp forest restricted to 1–2 km on the banks of the Mbaéré River. Tool sites were easily recognized by the presence of one or more sticks (or in two cases, vines). Often these sticks had been ripped from nearby small trees and were usually stripped of their leaves, and sometimes of their bark, in association with an underground beehive or ant nest. A stick was classified as a tool if it was associated closely with an insect colony, had been stripped of its leaves, had been broken off from nearby trees, and/or showed blunting at the end (presumably caused by probing). In many cases, clusters of tools were found in or beside the insect nests in piles, and could be dated to several different time periods. We assessed a tool’s age by the freshness of the exposed sections that had been TABLE I. Definition of Forest Types Surveyed in the Ngotto Forest Forest type Open forest–closed-canopy, open-understory, dry-ground forest Mixed forest–intermediate between open and dense dry-ground forest Dense forest–canopy relatively open with dense, closed understory Open swamp forest–relatively closed canopy, open understory, seasonally-flooded forest Closed swamp or river forest–relatively open-canopy, closed understory seasonally flooded swamp or riparian forest Swamp-edge tangle–extremely dense strip of vegetation running along edge of swamp Herb patch (thicket)–open canopy forest with dense covering of Marantaceae herbs on forest floor Logging road–8-m wide logging access road, along which trees are felled Logged forest–dry-ground forest in which tree-cutting operations are underway, often crisscrossed with access roads 226 / Hicks et al. stripped of bark, as well as by using other clues, such as fresh ape prints or nests at the tool site. At three fresh ant sites and four fresh honey sites, chimpanzee prints and/or dung were found in association with the tools, and at the majority of sites one or more tools were found inserted in the hive or mound. No sign of gorillas (e.g., dung, footprints, or food remains) was ever found in association with any of the tool sites. Because tool use is unknown in free-living gorillas [McGrew, 1992], it is assumed that all tools found in this study were used by chimpanzees. At the fresher honey sites, wadges (masses of beeswax that have been sucked on by apes to extract honey) were sometimes found lying beside the open holes, and occasionally showed ape tooth marks. Tools were found scattered on the ground beneath trees containing beehives. Ant-dip tools were usually found inside or next to a subterranean bivouac at the base of a tree, or in association with columns of ants moving away from the nest. Sticks had not been merely selected by the chimpanzees from debris on the ground, but had in many cases been broken off from small trees in the area around the insect colony, which we verified by matching the breaks in the tools with the ends of nearby severed branches or saplings. Depending on whether they were found in association with ant nests or beehives, the stick tools were classified as ant tools or honey tools. Within each category, the tool types were further classified as ant-dip tools and ant clubs, and honey probes and honey hammer/clubs. We measured all tools found in the field lengthwise in inches to the nearest half-inch, using the standard steel tape measure. The measures were later converted to centimeters. The Griman assistants held the tools in place during the measurements. With the exception of four tools brought back to the United States, for which measures of diameter were acquired, no measures of tool weight or diameter were taken. In addition to the measurements of the tools, the following information was recorded at each site: the number of tools at the site, the distance of each tool from an ant or bee hole, the estimated age of the tool based on the freshness of the sap or wood, the Bofi language name for the species of sapling used, the presence or absence of a source tree for the tool within a 25-m2 radius, the type of habitat for the tool site, and any additional evidence of chimpanzee presence (e.g., hairs, dung, prints, etc.). Bofi names were used to identify the bee and ant species, which were identified by T. Hicks. For the most part the ants were Dorylus, and the bees were various stingless species (probably Halictidae and Trigonidae) and, rarely, stinging bees. T. Hicks drew detailed maps of the tool sites, made photographs, and recorded GPS waypoints to provide satellite coordinates for the sites. The waypoints were used to construct a 1:64,308 map of the location of the tool sites relative to our camp waypoints. No GPS waypoints with errors of 413 m were used, and the majority of error readings were r10 m. The research team made a 1-week trip through the forest at the beginning of the project to ensure the accuracy of the instruments and establish the methods used. As a test trial for our methods, tools were identified based on the criteria mentioned above and measurements were made. Beginning with the second forest trip, all tools were included as part of the final data set. RESULTS As shown in Table II, a total of 102 individual tools were found at 25 ant-dip and honey-probe tool sites. With the exception of two that were made from vines, Chimpanzee Tool Use in the Ngotto Forest / 227 TABLE II. Ngotto Tool Sites: Ant Dips and Honey Probes Tool-types Ant Dip Tools Honey probes Total Tool number (n) % of total tool sites % of total tools 13 46 52 12 56 25 102 Tool sites (n) Mean length (cm) SEM 45 84.6 4.8 Dorylus 48 55 58.1 3.0 100 100 na na Terres-trial stingless bees na Insect species the tools discovered were slender branches ripped from nearby small trees. In some cases their ends were blunted, and they were usually found without leaves, which suggests that they had been stripped. Eleven of the ant tools and 11 of the honey tools were found still projecting from insect holes. Two additional kinds of toolsFthe ant club and honey hammer/club–were also found. Because of the small sample size of these tools, they were excluded from the quantitative analysis of tool length, although they are described in the text. All tool sites, with one exception, were found in mixed forest, the most common forest type. The exception was one ant-dip tool found in open forest. Mixed forest was also more likely to contain chimpanzee nests than gorillas: in a transect study conducted in the same forest region by Hicks et al. (unpublished results), 74% of chimpanzee nests but only 12% of gorilla nests were found in mixed forest. Please see Fig. 2 for a map of the distribution of the different tool types across the southern section of the study area. As can be seen, the majority of the tool sites were found in the southern forests adjacent to the Mbaéré River, where no logging or logging surveys had yet occurred. Photographs of the four different tool types can be seen in Fig. 3. Tool Type 1: Ant-Dip Tool Ant-dip tools were typically long, thin wands that apparently had been stripped of their leaves and were found either protruding from or lying next to the subterranean bivouacs of driver ants. The bivouacs were often located in holes or mounds at the base of a tree. Two of the tools were found with leaves still attached. Nine had been ripped from small trees found within 10 m of the ant hole. The mean length for the ant-dip tools was 85 cm (SEM=4.8). The average number of tools per site was 3.54 (SD=2.15), with a minimum of one and a maximum of eight tools at a site. At some sites, the tools clearly dated from different time periods, indicating that the chimpanzees had revisited ant-dipping sites. At one site, following a contact with chimpanzees, the prints of an adult and an immature chimpanzee were found in the dirt around a large driver ant bivouac, and driver ants were swarming aggressively up the several sticks found projecting from and lying around the ant holes. Older tools were found at this site as well. The ant holes were often quite deep. The three holes that were measured were 61, 91, and 91 cm deep, respectively. On two occasions the tools were found in a foraging ant swarm away from the bivouac. On one of these occasions, it was apparent from the footprints that the chimpanzee had bent down a small sapling and sat on it while fishing for ants in the leaves (a ‘‘sapling seat’’?). At this site, along with several stick tools, a vine tool was found projecting from the churned 228 / Hicks et al. Fig. 2. Map of distribution of tool sites in the study area, north of the Mbaéré River. dirt. This was one of only two cases of a vine being used by a chimpanzee to dip for ants. The Bofi names for the types of stick used to make 38 of the ant-dip tools were recorded. Unfortunately, the Latin names were not available. Tool Type 2: Honey Probe The chimpanzees used honey tools to probe the ground and log holes of stingless bees to withdraw wax and honey. Tools were found projecting from or lying to the side of the holes, which often had bees still inside. Two honey tools at two sites were found projecting from holes in rotting logs, and the rest were associated with ground holes. None of the tools had leaves attached, and three were found with stripped leaves lying beside the tool. Two tools at two sites were found with their bark peeled off, strips of which were found lying beside the tools. Seventeen of the tools had been ripped from small trees within 10 m of the honey Chimpanzee Tool Use in the Ngotto Forest / 229 Fig. 3. a: Ant-dip site, with two ant-dip tools projecting from a bivouac at the base of a tree. b: Two ant clubs lying next to a fallen solebumu nest. c, d: Two honey-probe tool sites. Ape hairs were found inside the beehole (c), and bits of chewed wax can be seen around the tool (d). e, f: Two images of the first honey hammer/club described in the text. 230 / Hicks et al. Fig. 3. Continued. site. The mean length of the honey probes was 58 cm (SEM=3.0). The average number of tools per site was 4.67 (SD=3.9), with a minimum of one and a maximum of 15 tools at a site. As with the ant tools, at some sites the tools dated Chimpanzee Tool Use in the Ngotto Forest / 231 Fig. 3. Continued. 232 / Hicks et al. from different time periods, indicating that the chimpanzees revisited honey sites as well. At three fresh honey sites, wadges of beeswax were found lying beside the open holes, and in two cases showed ape tooth marks. One of these ground holes had been dug into, and the extracted dirt lay next to the hole. Local humans, as well as the chimpanzees, harvested honey from the subterranean stingless-bee nests (personal observation). The humans often used their hands to dig a meter or more into the nests to obtain honey. Chimpanzee excavations were shallow in comparison to the human excavations, and were apparently limited to the upper chamber of the hive. The five ground honey holes that were measured were approximately 28, 8, 8, 31, and 25 cm deep, respectively. These holes were much shallower than the ant holes. The chimpanzees used various types of small trees to make honey-probe tools. The Bofi names for the types of stick used to make 51 of the honey tools were recorded. Latin names were not available. Ant-Dip Tool and Honey-Probe Comparison The ant-dip tools were significantly longer than the honey probes: t=4.739, Po.01 (n1=46, n2=56). With the exception of the mobaba tree (which was used to make 10.5% and 13.7% of the ant and honey tools, respectively), there was no overlap in the type of tree used in constructing the ant-dip tools and honey probes. Tool Type 3: Honey Hammer/Club Two examples of the honey hammer/club were found, each at a different site. Although these tools differed from one another in length and shape, they were included in the same category because both apparently were used to forcefully pound into arboreal beehives. On 19 April 2002, immediately following an observation of three adult chimpanzees feeding on figs, a large number of chimpanzees vocalizing excitedly along with a chimpanzee apparently using a honey-hammer/club were heard to our north. For approximately 3 min, the loud hammering sound coming from the upper canopy continued. According to a Griman assistant, a chimpanzee was trying to open up an arboreal beehive, a behavior the assistant said he had seen before. As we approached, we heard a loud crash from the direction of the hammering, followed by the scream of a chimpanzee apparently falling out of the canopy. Upon investigating, we found that an enormous dead tree had just fallen. Wet sap from the vines that had come down with the tree was splattered across the scene. Inside a knot halfway up the tree, bees were visible flying around the entrance to their hive. Beneath the tree, one of the assistants found what appeared to be a tool made from a short section of tree branch. The tool was shaped like a dagger and fit neatly in the hand. It measured 35.6 cm in length and 2.4 cm in diameter. The end of the tool was freshly blunted, with the tip frayed and pushed back, consistent with it having been used to pound against a tree trunk. The second hammer/club was found on another day at the base of a tree, lying on top of a pile of freshly severed vines. The assistants said that a chimpanzee had slashed the vines with his or her teeth to clear the space around the beehive 15 m up in the tree. Apparently the chimpanzee had pounded on the nest with the thick club to open it, and then dropped the club on top of the pile of vines below. Chimpanzee Tool Use in the Ngotto Forest / 233 We could see the stinging bees (possibly Apis mellifera) still swarming around the mouth of their hive. The club measured 88.9 cm in length. Tool Type 4: Ant Club We discovered beneath two recently-made chimpanzee nests a large solebumu ant nest that had been knocked down that day from the limb of an overhead moseteke tree. The solebumu nest was a brown, brittle, football-shaped nest constructed from what we assumed to be leaves. It resembled the arboreal Crematogaster nests described by Goodall [1964, 1986]. A large moseteke stick had been ripped from a fallen branch 5 m to the east, and apparently had been used as a club to punch six deep holes into the ant nest. This club, lying just beside the fallen ant nest, was still swarming with ants, which are quick and painful biters. A second, thinner moseteke stick tool was found lying next to the nest as well, also covered in ants. The larger of the two tools was approximately three times as thick as the other, and measured 91.4 cm in length. The smaller tool, which resembled an ant-dip tool, was 83.8 cm in length. DISCUSSION As has been found in other populations of Pan troglodytes [Boesch & BoeschAcherman, 2000; Humle & Matsuzawa, 2002; Stanford et al., 2000], chimpanzees of the Ngotto Forest selected or constructed tools of different lengths depending on the function of the tool. The tools used to dip for ants were longer than those used to probe for honey. Ecological Factors No difference was found in the habitat types in which the tool types were encountered (they were mixed forest only, with one exception). However, other subtle differences in habitat cannot be ruled out. Other than the mobaba tree (which was used to make 10.5% and 13.7% of the ant and honey tools, respectively), there was no overlap in Bofi tree types used to construct the antdip and honey probe tools. Selection of different plant species as raw materials for the two tool types cannot be ruled out, as was documented at Mount Assirik, Senegal, by McBeath and McGrew , and this may explain the difference in the tool lengths. Between-Site Comparisons: Ant-Dip Tools and Honey Probes As can be seen in Table III, the Ngotto chimpanzees used longer ant-dip tools than the chimpanzees at Tai, Gombe, and Bossou. The chimpanzees of Gombe and Tai use different methods to dip for ants [Boesch, 1996]. The Tai chimpanzees sweep the ants directly from short rods with their lips (the mouthsweep technique). The Gombe chimpanzees pull long wands through their free hand, catching the ants in that hand and then putting them in their mouths (the pull-through technique) [McGrew, 1974]. At Bossou, the chimpanzees use both techniques: wands are used for the pull-through technique, and rods are used for the mouth-sweep technique [Humle & Matsuzawa, 2002]. In that study, the pullthrough technique was used in situations in which the chimpanzees were more likely to get bitten, such as at the ant bivouac. They used this technique most often with the aggressive black driver ants, and the mouth-sweep technique was used with the less-aggressive driver ant species of the area. The length of the 234 / Hicks et al. TABLE III. Driver Ant Dip Tool Between-Site Comparison n Mean length (cm) SEM Range (cm) Reference Ngotto Tai 46 35 84.6 23.9 4.8 – 24–152 11–50 Gombe Bossou 13 189 66.0 53.7 – 1.5 15–113 23–154 This study Boesch and Boesch  McGrew  Humle and Matsuzawa  TABLE IV. Honey Tool Between-Site Comparison Ngotto Tai Boesch Bwindi stinging bees (Apis mellifera) Bwindi stingless bees (Meliponula bocandei) n Mean length (cm) 12 45 SEM Range (cm) Reference 58.1 28.1 3 – 6.5–99 14–60 9 60.0 – 25–85 12 27.0 – 14–70.5 This study Boesch and Boesch  Stanford et al.  Stanford et al.  sticks measured in this study suggests that the Ngotto Forest chimpanzees use the pull-through method. As shown in Table IV, the mean length of the Ngotto honey probes is similar to that of the sticks used by the Bwindi chimpanzees to probe for the honey of stinging bees [Stanford et al., 2000]. Within the Bwindi population, a significant difference exists between the lengths of the longer sticks used to acquire stingingbee honey and the shorter sticks used to acquire stingless-bee honey. The Bwindi short sticks are almost as long as the Tai honey tools. Between-Site Comparisons: Honey Hammer/Clubs, Ant Clubs, and Ant-Fishing Perches Tools resembling the ant clubs and honey hammer/clubs of Ngotto have been observed at other sites. In the extreme southwest corner of Central African Republic, Central chimpanzees were observed by Fay and Carroll  at the Bai Hokou and the Ndakan research sites using wooden hammers of up to 10 cm in diameter to smash open both arboreal and terrestrial nests of meliponine bees. Fay and Carroll also found circumstantial evidence of chimpanzee honeyhammering across the border in the People’s Republic of Congo, indicating that P. t. troglodytes engages in this behavior across a wide geographical area. If the Ngotto Forest chimpanzees, who live 210 km to the north-northeast of Ndakan, are included as honey-hammerers, the area across which this type of tool use occurs becomes enormous, and honey-hammering could be identified as a characteristic large-scale tool use pattern of the Central chimpanzee. There have been few observations of this behavior in the other subspecies of chimpanzee. J. Hart, in a personal communication to McGrew , described finding stick Chimpanzee Tool Use in the Ngotto Forest / 235 clubs in the Ituri Forest that had been used by Eastern chimpanzees to open up beehives in hollow trees. Brewer and McGrew  reported the use of a branch as a chisel by a forest-born, rehabilitated female Western chimpanzee on an island in the Gambia. The chisel, which was part of a tool set of four different-sized branches and vines, was used to pound into the batumen of an arboreal stinglessbee nest to acquire honey. There are very few reports of the use of clubs to punch holes into arboreal ant nests in chimpanzees. Goodall  observed chimpanzees at Gombe using sticks to force openings into the arboreal nests of ants, possibly in the same manner in which the two ant clubs found at Ngotto were used. Only once was a possible example of a ‘‘sapling seat’’ seen in the Ngotto Forest, but the Griman assistants claimed that the chimpanzees of the area commonly bent saplings as perches while fishing for driver ants. This behavior was described in the Gombe chimpanzees by McGrew . Insect Holes With regard to the ant and stingless-bee ground holes, humans sometimes dug down to a meter beneath the ground to excavate the deeper chambers of stingless-bee holes, while the chimpanzees apparently were limited to probing into the upper chamber (the part easily accessible from the surface). The ant holes were much deeper than the upper chambers of the stingless-bee holes. This may be why the chimpanzees were using longer sticks for ants. A future study could compare the depths of insect colonies at different chimpanzee study sites to correlate them with tool length. Termite Fishing Termite fishing is a common practice in many chimpanzee populations across Africa [McGrew, 1992]. No termite-fishing tools were observed during the 7month study at Ngotto. No obvious Macrotermes mounds were observed by T. Hicks in the forest. However, since the study was conducted primarily in the dry/ early wet season, we cannot rule out the possibility that the Ngotto chimpanzees engaged in some sort of termite fishing during the rainy season. Future Research Future studies of the Ngotto Forest chimpanzees should focus on direct observations of the apes constructing and using their tools. This would help determine whether the chimpanzees reuse old tools that have been left at frequently-visited sites (as documented at Gombe by McGrew ), and whether they modify their tools prior to use, as was observed by Goodall  and McGrew . Direct observations of chimpanzees using honey hammer/ clubs and ant clubs would be valuable as well. Unfortunately, with the whole of their forest currently in the process of being logged, the future survival of this population and its tool traditions are not assured. CONCLUSIONS The evidence presented here is a further demonstration of the impressive behavioral diversity of chimpanzee populations [Boesch, 1996; Whiten et al., 2001]. Chimpanzee culture is a powerful explanatory factor for this diversity. By continuing to document the full range of chimpanzee material and behavioral 236 / Hicks et al. culture in both little-known and habituated populations [McGrew et al., 2003], we move toward increasing our understanding of the depth and complexity of tradition beyond Homo sapiens. ACKNOWLEDGMENTS Data from this paper were originally presented to Central Washington University in partial completion of a Master of Science degree. We thank Bernard Voyemakoa, the Ministre et Directeur Général des Recherches Scientifiques et Technologiques of Central African Republic, for allowing us to conduct our research in the Ngotto Forest. We also thank Hillary Fouts, Alain Penelon of ECOFAC, Chief Dakamoli, Dibesco Pilandon, Ngasa Bernard, Pilabo Martin, Wanley Bemina, Samedi Claude, Fami Eve, Justin Kula, Tim Tikouzou, J.-B. Bobi, and Brice for their help with the project. REFERENCES Alp R. 1997. ‘‘Stepping-sticks’’ and ‘‘seatsticks’’: new types of tools used by wild chimpanzees (Pan troglodytes) in Sierra Leone. Am J Primatol 41:45–52. Bastin D. 1996. Foret de N’gotto inventaire forestier: rapport general, zone 1 et 2. Rapport ECOFAC, RCA, Union Européen, Groupement AGRECO/CIRAD Foret. p 15. Article retrieved October 25, 2003 from http://www.ecofac.org/Biblio.htm Boesch C, Boesch H. 1990. Tool use and tool making in wild chimpanzees. Folia Primatol 54:86–99. Boesch C. 1996. The emergence of cultures among wild chimpanzees. Proc Br Acad 88:251–268. Boesch C, Boesch-Acherman H. 2000. The chimpanzees of the Tai Forest: behavioral ecology and evolution. Oxford: Oxford University Press. 316 p. Brewer S, McGrew W. 1990. Chimpanzee use of a tool-set to get honey. Folia Primatol 54:100–104. Brugiére D, Sakom D, Sinassonasibe J. 1999. Estimation des densités et analyse du comportement nidificateur des gorilles et chimpanzes en forêt de Ngotto. Rapport ECOFAC, Bruxelles. 35 p. Article retrieved October 25, 2003 from http://www.ecofac. org/Biblio.htm Fay J, Carroll R. 1994. Chimpanzee tool use for honey and termite extraction in Central Africa. Am J Primatol 34:309–317. Goodall J. 1964. Tool-using and aimed throwing in a community of free-living chimpanzees. Nature 201:1264–1266. Goodall J. 1986. The chimpanzees of Gombe: patterns of behavior. Cambridge: Belknap Press of Harvard University Press. 673 p. Hirata S, Myowa M, Matsuzawa T. 1998. Use of leaves as cushions to sit on wet ground by wild chimpanzees. Am J Primatol 44: 215–220. Humle T, Matsuzawa T. 2002. Ant-dipping among the chimpanzees of Bossou, Guinea, and some comparisons with other sites. Am J Primatol 58:133–148. Jones C, Sabater Pi J. 1969. Sticks used by chimpanzees in Rio Muni, West Africa. Nature 223:100–101. Joulian F. 1995. Mise en évidence de différences traditionnelles dans le cassage des noix chez les chimpanzés de la Cote d’Ivoire, implications paléoanthropologiques. J Afr 65:57–77. McBeath N, McGrew W. 1982. Tools used by wild chimpanzees to obtain termites at Mt. Assirik, Senegal. J Hum Evol 11: 65–72. McGrew W. 1974. Tool use by wild chimpanzees in feeding on driver ants. J Hum Evol 3:501–508. McGrew W, Tutin C, Baldwin P. 1979. Chimpanzees, tools, and termites: crosscultural comparisons of Senegal, Tanzania, and Rio Muni. Man 14:185–214. McGrew W, Rogers M. 1983. Chimpanzees, tools, and termites: new records from Gabon. Am J Primatol 5:171–174. McGrew W. 1992. Chimpanzee material culture: implications for human evolution. Cambridge: Cambridge University Press. 277 p. McGrew W, Ham R, White L, Tutin C, Fernandez M. 1997. Why don’t chimpanzees in Gabon crack nuts? Int J Primatol 18: 353–374. McGrew W. 1998. Culture in nonhuman primates? Ann Rev Anthropol 27:301–328. Chimpanzee Tool Use in the Ngotto Forest / 237 McGrew W, Baldwin P, Marchant L, Pruetz J, Scott S, Tutin C. 2003. Ethnoarchaeology and elementary technology of unhabituated wild chimpanzees at Assirik, Senegal, West Africa. Paleoanthropology 1:1–20. Nishida T. 1973. The ant-gathering behaviour by use of tool among wild chimpanzees of the Mahale Mountains. J Hum Evol 2: 357–370. Sabater Pi J. 1974. An elementary industry of the chimpanzees in the Okorobiko Mountains, Rio Muni (Republic of Equatorial Guinea), West Africa. Primates 15: 351–364. Stanford C, Gambanez C, Bosco J, Goldsmith M. 2000. Chimpanzees in Bwindi-Impenetrable National Park, Uganda, use different tools to obtain different types of honey. Primates 41:337–341. Sugiyama Y. 1985. The brush-stick of the chimpanzees found in southwest Cameroon and their cultural characteristics. Primates 26:361–374. Tutin C, Ham R, Wrogeman D. 1995. Tool use by chimpanzees (Pan troglodytes troglodytes) in the Lope Reserve. Primates 36: 181–192. Whiten A, Goodall J, McGrew W, Nishida T, Reynolds V, Sugiyama Y, Tutin C, Wrangham R, Boesch C. 2001. Charting cultural variation in chimpanzees. Behaviour 138: 1481–1516. Wrangham R, De Waal F, McGrew W. 1994. The challenge of behavioral diversity. In: Wrangham R, McGrew W, De Waal F, Heltne P, editors. Chimpanzee cultures. Cambridge: Harvard University Press. p 1–15.