American Journal of Primatology 5:171-174 (1983) BRIEF REPORT Chimpanzees, Tools, and Termites: New Record From Gabon W.C. McGREW' A N D M.E. ROGERS' 'Department ofPsychology, University of Stirling, and 'Department Edinburgh, Scotland of Zoology, University of Wild chimpanzees (Pun troglodytes) in northeastern Gabon use tools made of vegetation to obtain termites (Macrotermes Znobilis) for food. They mostly use probes in termite fishing, as recorded elsewhere in eastern and far western Africa. This is the first record of termite fishing by the central west African race of chimpanzees. There are signs that they also use stouter tools, perhaps to perforate the mounds of the termites. These new findings further complicate the status of material culture in this species of ape in nature. Key words: tool use, chimpanzee, termite, Pan troglodytes, Macrotermes, culture INTRODUCTION Chimpanzees in the wild use tools made of vegetation to obtain termites for food. Since this was first found by Goodall  a t Gombe in Tanzania, it has been recorded a t eight other sites in Africa [McGrew et al, 1979; Nishida & Uehara, 19801. This note describes a newly discovered case of this tool use, at Belinga in northeastern Gabon. This has important ethnological implications; for example, it is the first report of termite fishing in primary forest, and it may be the first instance of one population of apes using tools for both fishing and perforating. METHODS We spent 7 weeks (15 July-31 August, 1981) collecting ecological data on chimpanzees (Pun t. troglodytes) and gorillas (GoriZZa g. goriZZa) a t Belinga (1'06' N, 13"12' El. Most of the area is primary rain forest, although the annual rainfall is only about 1,700 mm [Hladik, 19781. Over this period we spent 276 h searching for apes or their signs. The apes of the area are unhabituated to humans, and the vegetation is dense, so it is very difficult to obtain direct sightings. On 3 days spent following chimpanzees, we found tools left by them on the surfaces of the subterranean nests of Macrotermes ?nobilis. Received July 24, 1982; accepted May 23, 1983. Address reprint requests to W.C. McGrew, Department of Psychology, University of Stirling, Stirling, FK9 4LA. Scotland. 0 1983 Alan R. Liss, Inc. 172 McGrew and Rogers As we never saw chimpanzees using these objects, how can we say that they were tools used by chimpanzees? This question assumes even greater importance as the gorillas of the area regularly eat termites [Tutin and Fernandez, in press]. In fact, the evidence is circumstantial, and a previous paper gave eight criteria to be met in considering the validity of such data [McGrew et al, 1979, pp 189-1901. Table I lists these criteria and their definitions. The objects we collected meet seven of the eight criteria with only the year-to-year reliability of the data yet to be tested. Perhaps the crucial criterion is that of close proximity to chimpanzees. On 1 August we tracked and intermittently sighted a group of chimpanzees for almost 4 h. Midway through this period, we found fresh signs of tool use only minutes after chimpanzees had passed that way. There was no sign of gorillas in the area a t the time. One tool was found in situ-ie, inserted into the soil surface of a termites’ nest. The possibility of confusion with traces of gorilla activity is further reduced because the two species of ape feed on different species of termites [Tutin and Fernandez, in press]. Faecal samples show that gorillas at Belinga eat Cubitermes sp., a smaller type of termite which builds tiered, earthen nests above the ground, usually against the trunks of large trees. There are no records of chimpanzees eating this genus, just as there are no records of gorillas eating Macrotermes. Gorillas have not been seen to use tools to obtain the insects, but instead appear to break open the nests by hand. RESULTS We found 30 tools. Nine additional artifacts were discarded as doubtful, and fragments of tools resulting from manufacture were ignored. Twenty-eight of the tools were fishing probes, and all but one of these (made of a liana) were made of twigs stripped of leaves and twiglets. None of the tools were peeled of bark. Sixteen tools were bent or frayed from use a t their distal ends; only five of the tools showed such signs of wear at the proximal ends. (See Table 11). TABLE I. Criteria for Evaluating Circumstantial Evidence of Fishing For Termites by Chimpanzees 1. Alteration 2. Use 3. Site 4. Season 5. Coincidence 6. Association 7. Near-Miss 8. Persistence Objects are unnaturally modified-eg, peeled of bark, stripped of leaves, etc. Objects show signs of wear-eg, frayed from insertion, bitten by termites, etc. Objects found only on termite mounds. Objects found only when mound’s surface is vulnerable to probing. Objects found only at mounds of one species of termites, and subterranean forms of only that species found in the feces of chimpanzees at that time. Objects found in association with other signs of chimpanzees-eg, footprints, hairs, feces. Objects found in close proximity to chimpanzees seen or heard by observers. Criteria 1-7 hold over more than one annual cycle. Chimpanzee Termite-Fishing in Gabon 173 TABLE 11. Characteristics of Fishing Probes Used by Chimpanzeesat Belinga, Gabon Proximal diameter Length N - = 23a N 25 0.4 cm Median = 0.4 cm = x = 37.8 cm x Median = 38 cm Range = 17-59 cm Range = 0.3-0.9 cm = Distal diameter N = 23a x = 0.3 cm Median = 0.3 cm Range = 0.15-0.6 cm Signs of wear = 25 9-Frayed 6-Bent 5 Frayed and bent 5-Neither N aTwo of the tools were unfinished-ie. not modified at the distal end. The other two tools were very different, being stout sticks. They were longer (68 and 76 cm) and thicker at both proximal (1.7 and 1.8 cm) and distal (1.4 and 1.6 cm) ends. They too were stripped of leaves and twigs. Only four sources of raw materials could be identified: Scaphopetalum thonneri (Sterculiaceae), Calpocalyx sp. (Mimosaceae), Drypetes sp. (Euphorbiaceae), and a species of Tiliaceae. DISCUSSION That another population of wild chimpanzees should be found to use tools to obtain termites is not surprising. This is the first record from Gabon, but Hladik  noted that chimpanzees that had been transported and provisioned but which were free-ranging near Makokou (about 100 km southwest of Belinga) ate several other types of termites: Nasutitermes, Microterotermes, Procubitermes, etc. They did so without tools, however and ignored Macrotermes spp. The nearest other records of tool use to obtain termites come from Equatorial Guinea (formerly Rio Muni). There chimpanzees used sticks as perforating tools to break open the mounds of M. rnuelleri and M. lilljeborgi [Jones & Sabater Pi, 19693. However, only two of the tools found at Belinga resembled the sticks found by Jones and Sabater Pi, and we found no fresh evidence of digging a t the nests of termites. It may be significant that, unlike M. muelleri, which builds 1-2 m high, mamillated mounds, the earthen dome over the underground nest of M. nobilis is low, flat, and spread out [Ruelle, 1970, p 4191. Perforating and breaking up the latter type of nest would seem to be a n inefficient way of extracting termites from it, because excavation would be required. The other tools closely resembled the fishing probes found at sites in eastern and far western Africa [McGrew et al, 19791. This is the first record of their use by the central west African race of chimpanzees. Also, all other cases of termite fishing come from chimpanzees living in savanna, woodland, or mixed forest-woodland sites with lower annual rainfall. There is no other evidence of such tool use from studies of forest-living chimpanzees [eg, Ghiglieri, 19791. However, work in Tanzania [Nishida & Uehara, 1980; Uehara, 19821 suggests that chimpanzees in wetter habitats seem to use tools less than those living in drier areas. McGrew et a1  proposed hypotheses about the cultural origins of tool use by chimpanzees to obtain termites. We hypothesised three separate inventions of the custom, with two of these being for fishing in northern Guinea and in western Tanzania, and one being for perforation in Rio Muni or Cameroon. We proposed four ways in which this hypothesis could be falsified. If both fishing and perforation are used in Gabon (and possibly in Tanzania also), the the two methods of obtaining 174 McGrew and Rogers termites cannot be simply correlated with differences in habitat and may not have been invented separately. Much more systematic study is needed, but it seems likely that the material culture of chimpanzees will not be explained simply in terms of environmental constraints. CONCLUSIONS 1. Wild chimpanzees in Gabon use slender probes to fish for termites. 2. They may also use stouter sticks to perforate or excavate termite mounds. ACKNOWLEDGMENTS We thank C. Tutin and M. Fernandez for essential help in the field, P. Hecketsweiler and A. Moungazi for identification of plants, A. Collins and T. Nishida for critical comments on the manuscript, L.S.B. Leakey Trust and the Internal Research Fund of the University of Stirling for financial aid, and the Centre International de Recherches Medicales de Franceville and L’Institut de Recherche sur 1’Ecologie Tropicale for sponsorship in Gabon. REFERENCES Ghiglieri, M.P. THE SOCIOECOLOGY OF CHIMPANZEES IN KIBALE FOREST, UGANDA. Ph.D. thesis, University of California, Davis, 1979. Goodall, J. My life among wild chimpanzees. NATIONAL GEOGRAPHIC 124(2):272308,1963. Hladik, A. Phenology of leaf production in rain forest of Gabon: Distribution and composition of food for folivores, pp 51-71 in THE ECOLOGY OF ARBOREAL FOLIVORES. G.G. Montgomery, ed. Washington, Smithsonian Institution Press, 1978. Hladik, M. Alimentation et activite d’un groupe de chimpanzes reintroduits en foret Gabonaise. LA TERRE ET LA VIE 27:343413, 1973. Jones, C; Sabater Pi, J. Sticks used by chimuanzees in Rio Muni. West Africa. NAh J R E 223:lOO-101,1969, McGrew, W.C.; Tutin, C.E.G.; Baldwin, P.J. Chimpanzees, tools, and termites: Cross cultural comparisons of Senegal, Tanzania, and Rio Muni. MAN 14:185-214, 1979. Nishida, T.: Uehara, S. Chimpanzee, tools, and termites: Another example from Tanzania. CURRENT ANTHROPOLOGY 21: 671-672,1980. Ruelle. J.E. A revision of the termites of the genus Macrotermes from the Ethiopian region (Isoptera: Termitidae). BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), ENTOMOLOGICAL SERIES 24: 365-444,1970. Tutin, C.E.G.; Fernandez, M. Gorillas feeding on termites in Gabon, West Africa. JOURNAL OF MAMMALOGY (in press). Uehara, S. Seasonal changes in the techniques employed by wild chimpanzees in the Mahale Mountains, Tanzania, to feed on termites (Pseudacanthotermes spiniger). FOLIA PRIMATOLOGICA 37:44-76,1982.