Biological Conservation 1993, 64, 83-89 :'~ ...~::,~v~II
A CENSUS OF THE EASTERN L O W L A N D GORILLAS GORILLA GORILLA GRA UERI IN KAHUZI-BIEGA N A T I O N A L P A R K WITH R E F E R E N C E TO M O U N T A I N GORILLAS G. G. BERINGEI IN THE V I R U N G A REGION, ZAIRE Juichi Yamagiwa Primate Research Institute, Kyoto University, Inuyama, Aichi, Japan
Ndunda Mwanza Centre de Recherches en Sciences Naturelles, Lwiro, D.S. Bukavu, Zaire
Andrea Spangenberg Department of Forestry, University of Gi~ttingen, Stangengasse 14, W-8172 Lenggries, Germany
Tamaki Maruhashi Department of Human and Cultural Sciences, Musashi University, ToyotamakamL Nerima, Japan
Takakazu Yumoto College of Liberal Arts, Kobe University, Kobe, Nada, Japan
Antje Fischer & Bernd Steinhauer-Burkart Projet IZCN/GTZ, Conservation de la Nature, BP 852 Bukavu, Zaire (Received 23 January 1992; revised version received 11 March 1992; accepted 12 May 1992)
the Park. Reference is made to the more detailed censuses of the mountain gorilla G. g. beringei. More international attention will be needed to increase conservation activity to protect gorillas from the hazards of human disturbances in this area.
A population census of eastern lowland gorillas Gorilla gorilla graueri was conducted in the original part of Kahuzi-Biega National Park, Zaire. At least 258 gorillas ('25 groups and nine solitary males) were estimated by bed counts to have survived within the Park in the 1990 bamboo season. In comparison with results of the previous censuses, conducted between 1978 and 1979, the population showed a slight increase. The percentage of immature individuals in the population indicates that gorillas still maintain a healthy population, but mean group size has decreased markedly. Although gorillas expanded their range in the 11 years between the surveys, they have still tended to concentrate in the central well-protected area of the Park, possibly stimulating frequent transfer of females between groups or from groups to solitary males, ~'esulting in the decrease in group size. Recent social changes recorded from observations of habituated groups tend to support this interpretation. Human disturbance in the bridge zone between highland and lowland forests prevents gorillas from making contact with their neighbouts and reduces the possibility of outbreeding within
Key words: Zaire, gorillas, population census. INTRODUCTION
Eastern gorillas Gorilla gorilla beringei and Gorilla gorilla graueri have been found in small isolated populations scattered throughout areas of eastern Zaire, western Rwanda and south-western Uganda (Schaller, 1963; Goodall & Groves, 1977; Mwanza & Yamagiwa, 1989). Although several National Parks and Reserves have been established to protect them in this century, they are threatened by extinction as a result of increased hazards due to human disturbance such as deforestation, cattle encroachment and poaching. In this respect, they are listed as Endangered by the International Union for Conservation of Nature and Natural resources (IUCN) and are recommended as subjects for urgent research for the development of optimal conservation planning by the International Primatological Society.
Biological Conservation 0006-3207/93/$06.00 © 1993 Elsevier Science Publishers Ltd, England. Printed in Great Britain 83
Juichi Yamagiwa et al.
Reliable information on the size of population and distribution of these gorillas is scant, except in the case of mountain gorillas Gorilla gorilla beringei in the Virunga volcanoes, where a population census has been periodically conducted in the past decade (Harcourt et al., 1981; Aveling & Harcourt, 1983; Weber & Vedder, 1983; Aveling & Aveling, 1987). Gorillas inhabiting both highland and lowland forests of the Kahuzi-Biega National Park (6,000 km 2) were classified as eastern lowland gorillas Gorilla gorilla graueri (Casimir, 1975b; Groves & Stott, 1979). In the original part of the Park (highland forest: 600 km2), Adrien Deschryver, the late Conservator, estimated that 200--500 eastern lowland gorillas survived in the early 1970s (Goodall & Groves, 1977). The first census was conducted by Murnyak (1981) in 1978-79, who counted 223 gorillas and suggested that the population was relatively stable or increasing in the original part of the Park. Yamagiwa (1983) also made a short survey between September and October 1978 and estimated that there were 230-250 gorillas in this part. Since then no census has been made, although conservation activity of the Institut Zairois pour Conservation de la Nature (IZCN) has been increased and improved with the cooperation of the Deutsche Gesellschaft for Technische Zusammenarbeit (GTZ). In 1990, it was still unknown whether the population maintained a healthy size and composition, and whether the gorillas had altered their distribution within the Park. In order to examine the effects of conservation efforts and to determine a plan of action for future conservation, it was necessary to analyse recent changes in the gorilla population in the original part of the Park. We therefore conducted a further census in this area to estimate the present distribution of gorillas, population size and composition, in order to compare the results with those of the previous censuses in the Park and in the Virunga region in general.
METHODS The census was conducted between September and November 1990 in the original part of Kahuzi-Biega National Park, Zaire (Fig. 1), first established as a Forest Reserve in 1960, and declared a National Park in 1970 for the protection of the gorillas (Mankoto, 1988). The Park was enlarged 10-fold to include interconnected montane and tropical forests in 1980. The original part covers an area of 600 km 2 at an altitude of 1800 to 3308 m, made up of bamboo Arundinaria alpina forest from 2350 to 2600 m (37%), primary montane forest in the western and northern parts of the Park (28%), secondary montane forest (20%) in the eastern part, and Cyperus swamp (7%) and other vegetation (8%), as described by Goodall (1977) and Murnyak (1981). Dominant tree species in each vegetation type are Podocarpus sp., Ficus sp. and Symphonia globulifera in primary montane forest, Hagenia abyssinica, Myrianthus holstii and Vernonia sp; in secondary montane forest, and Hypericum lanceolatum and Rapanea pulchra in Cyperus (Cyperus latifolius) swamp (Casimir, 1975a; Goodall, 1977). The census was made within the bamboo season (from September to November), when gorillas tend to move into this type of forest, seeking and feeding on bamboo shoots (Casimir & Butenandt, 1973; Casimir, 1975a; Goodall, 1977). The census initially covered various types of vegetation but no fresh trails of gorillas were found far from bamboo forest. The census area was divided into four sectors (A--D), and three groups
KAHUZ I-BIEGA NATIONAL PARK
,t , I
Fig. l. Map showing the two National Parks (]::::::p)in eastern Zaire and the original part ( l l ) of Kahuzi-Biega National Park.
Fig. 2. Map of vegetation in the Kahuzi-Biega National Park showing the four sectors where the census was conducted. I::::::1, bamboo forest; 7]7~ Cyperus swamp; [~-----], primary or secondary forest.
Eastern gorillas in Zaire
with apparently similar composition were regarded as the same group when the distance between bed sites was less than 10 km or when the groups were not followed simultaneously by the census takers. We used the same procedure to identify different solitary males. In Sector B, t w o groups of gorillas have been continuously monitored by the Park since the early 1970s for visiting tourists, and their age-sex compositions recorded (Table 1). In 1972, the Maeshe and Mushamuka Groups consisted of 18 and 20 individuals, respectively (Goodall, 1977; Casimir, 1979). In 1975, the leader male (silverback) of the Maeshe Group indulged in fierce displays and charges with a solitary silverback male, then fell ill from an infected wound and died. The solitary male took over the Maeshe Group (MacKinnon, 1978). This event stimulated female transfers from the Maeshe to the Mushamuka Group, and resulted in the large size difference between the two groups in 1978 (Yamagiwa, 1983). In the mid 1980s, a solitary male which had ranged around the Maeshe and Mushamuka Groups for several years formed a new group (Mubalala) by gathering females (Mankoto, 1988). This group has been habituated by the Park since 1986. In 1989, a silverback male, which emigrated from the Mushamuka Group, formed another new group (Nindja), to which several females from the Maeshe and Mushamuka Groups may possibly have transferred. These recent social changes caused the reduction in group size of the former established Maeshe and Mushamuka Groups. At the beginning of the census, we examined the bed sites of these groups and measured their faeces. We obtained the same results as those from direct observation, with the exception of the number of infants of less than one year of age.
formed by field workers with Park Guards and field assistants walked simultaneously through each sector, mainly in bamboo forest (Fig. 2). The census did not cover the western half of the northern sector, where the Park Guards had patrolled periodically during the preceding days and found no evidence of gorilla presence. In order to compare the results of this census with those of previous censuses (Murnyak, 1981; Yamagiwa, 1983) or with those of mountain gorillas in the Virunga region (Harcourt et al., 1981; Aveling & Harcourt, 1983; Weber & Vedder, 1983; Aveling & Aveling, 1987), we adopted methods similar to those formerly conducted in both regions. We tried to find fresh gorilla trails (up to two days old) and to follow them to their night bed sites, with simultaneous recording of field signs of any mammalian or recent human activity. Since the size of dung balls is correlated with gorilla age (Schaller, 1963), we counted the number of beds and measured the diameter of faeces left by gorillas inside and beside beds at each bed site. We allocated the size of faeces to five age-sex classes (silverback male older than 14 years; blackback male eight-14 years old; adult female older than eight years; independent immature three to eight years old; dependent immature 0 to three years old). Since faeces of infants of less than one year of age are usually missed because they are small and infants often evacuated on their mothers' bodies, infants were excluded from calculation within the census area. Two or more consecutive night bed sites were examined to obtain accurate data on the composition of the various groups. Although gorillas in the census area tend to shift their range monthly or seasonally (Casimir & Butenandt, 1973; Goodall, 1977; Yamagiwa, 1983), the average daily travel distance is less than 1.5 km (Casimir, 1975a; Goodall, 1977; Yamagiwa, 1988). Therefore, two groups
Table 1. The change in age--sex compositions of monitored groups~ in the Kahuzi-Biega National Park
2 7 3 9
18 19 9 24
Casimir (1979) Lyon (1975) Yamagiwa (1983) This study
20 42 21
Goodall (1977) Yamagiwa (1983) This study Mankoto (1988) This study
MaesheGroup 1972 1974 1978 1990
1 1 1 1
MushamukaGroup 1972 1978 1990
2 1 1
4 4 2
4 17 9
2 9 8
5 I1 1
MubalalaGroup 1986 1990
Nindja Group 1989 1990
SB, silverback male; BB, blackback male; AF adult female; Indep., independent; Dep., dependent.
Park Warden This study
Juichi Yamagiwa et al.
Table 2. Age-sex composition of gorilla groups found in the Kahuzi-Biega National Park during the census period
1 2 3 4 5 6 7 8 9 10 11 12 13a 14a 15a 16a 17 18 19 20 21 22 23 24 25b S1 $2 $3 $4 $5 $6 $7 $8 $9
1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1
2 1 3 1
1 1 1 2 1 6 9 1 5 4 3 2 6 9 9 6 4 5 2 5 1 4 2 2
1 1 1 1 4
5 7 3 4 4 11 20 7 10 15 6 8 11 22 21 13 24 11 5 15 2 7 8 7 4
A A A A A A A B B B B B B B B B B/C C C C B/D D D D D A A A B B B C D D
1 1 1 1
1 2 1 5 3 9 8 3 8 1
1 3 1 2 3 4
1 5 4 3 6 1
--1---1-1 1 1 1 1 1
Abbreviations as in Table 1. a Gorilla groups monitored by the Park: Nindja Group (13), Maeshe Group (14), Mushamuka Group (15), Mubalala Group (16). b Suspected to be an all-male group. RESULTS We c o u n t e d a total o f 258 gorillas (25 groups and nine solitary males) during the census (Table 2). The results are regarded as an almost complete count o f the population. Since the results o f the 1981 Virunga census were assumed to represent at the m o s t a 10% underestimate based on an analysis o f the accuracy o f the census (Aveling & Harcourt, 1983), the size o f the Park popula-
tion was estimated to be between 258 and 284. The population density was 0-43~0.47 individual/km z, and the mean g r o u p size 10.0 individuals (range : 2-24). In comparison with the results o f the 1978-79 census ( M u r n y a k , 1981), the p o p u l a t i o n has increased slightly in size f r o m 223 to 258. The n u m b e r o f g r o u p s has also increased f r o m 14 to 25 and solitary males f r o m five to nine, while the m e a n g r o u p size has decreased f r o m 15-6 to 10.0. In the 1978-79 census, two large groups
Table 3. Comparison of the mean group composition between the 1978-79 and the 1990 censuses.
1978 No. of samples Mean group size Mean no. of Silverback male Adult Immature (Indep.) Immature (Dep.) Source
1-2 6-8 3.4 2.9
1.1 7-1 4.8 3-0
1.1 4.8 3-1 1.3
0.92 0.70 0.91 0-45
1.0 0-68 0.65 0.43
Eastern gorillas in Zaire (34 and 37 individuals) were recorded, and Yamagiwa (1983) also found large groups (35 and 42) between September and October 1978. In the 1990 census, the largest group contained 24 individuals, and most of the groups (84%) consisted of less than 20 individuals. All the groups included a silverback male, and most contained adult females, except for Group 25 which was suspected of being all-male. In order to compare the group composition from the 1990 census with that for 1978 (Yamagiwa, 1983) and 1978-79 (Murnyak, 1981), we excluded data on Groups 1, 2, 7, 8, 9, 10, 19, 23, 24, since 25% or more of the compositions of these groups were unknown. We also excluded data on the apparently all-male Group 25. Data on blackback males and females were combined together as adults. Table 3 shows the mean group composition calculated from 15 groups from the 1990 census and those from the i.978 and 1978-79 censuses. The number of silverback males per group has not changed, while the number of other adults and immature individuals has decreased in the 1990 census, as has group size. The mean number of dependent immature individuals seems to have decreased dramatically. Even if we add the number of infants of less than one year of age, the 1990/1978 to 1990/ 1978-79 ratios of dependent immature individuals may be lower than similar ratios of group size. These results suggest that a reduction in the number of females per group has been the main cause of the decrease in numbers of immature gorillas per group and thus may be responsible for the reduction in group size. In fact, long-term records on the monitored groups indicate that emigration of females from large groups (Maeshe and Mushamuka) and their transfer to other males have resulted in the establishment of new groups (Mubalala and Nindja) and in a reduction of average group size (Table 1). In 18 years, the total population of monitored groups has increased from 38 to 69 and the number of groups from two to four. However, their ranging areas have not been enlarged. In 1990, the four monitored groups still ranged within Sector B where the two previously established groups (Maeshe and Mushamuka) occurred in 1972 and 1978. Figure 3 shows the location of 25 groups and nine solitary males recorded during the census. All were found in and around bamboo forest. From their old trails or bed sites (one to three months old), we estimated that they had moved from primary or secondary forests to bamboo forest at the beginning of the bamboo season. In the north-eastern part of Sector A, the eastem part of Sector B and the western part of Sector C, we found old field signs of groups and solitary males. Compared with the results of the 1978-79 census (Murnyak, 1981), gorillas have recently expanded their range to the north-east, west and south. However, in the western part of Sector A (Park Guards, pers. comm.) and the southern part of Sector D, which is linked with the Park extension, and where Murnyak (1981) had found at least two solitary males, we found no field signs of gorillas. Our observations suggest that
Fig. 3. Distribution of gorilla groups and solitary males (S) during the census. Numbers correspond to those in Table 1. Each arrow shows the direction of the shift in the gorillas' range at the beginning of the bamboo season, as judged from their old bed sites. they tended to concentrate their ranging in the central area (Sector B and C), so that now the gorillas lack a bridge zone through which they can move between highland and lowland forest. DISCUSSION The results of the 1990 census show a slight increase in the gorilla population in the Kahuzi-Biega National Park, as compared with the results of 1978-79 (Murnyak, 1981). This suggests that the recent conservation activity of IZCN/GTZ has been successful. Gorillas have expanded their range to new areas; however, during the bamboo season when our census was conducted, they were still concentrated in the central area where Murnyak found all the groups in the 1978-79 census. Murnyak (1981) attributed this to their preference for 'suitable habitat', made up of regenerating secondary forest with nearby bamboo forest and scattered patches of primary forest far from human disturbance, and in which he calculated the gorilla density to be 0-88 individual/km2. The 1990 census shows that this density has apparently increased to 1.03 individual/km2. Bamboo forest still extends widely over the northern part of Sector A and the southern part of Sector D, where no fresh signs of gorillas were found in the 1990 census. Old field signs of gorillas indicate that those which had ranged in the north-eastern part of Sector A moved to its southern part at the beginning of the bamboo
Juichi Yamagiwa et al.
season, instead of to the nearby bamboo forest, suggesting that their ranging behaviour may be influenced by human disturbance, rather than by their preference for a particular habitat. In the central area, where the Park Guards patrolled frequently and tourists visited four groups of gorillas on a daily basis, they are well-protected. However, in the northern and southern areas the number of guards is small and they cannot patrol the whole area. During the census we found more traps, fireplaces, hunting camps or new poacher trails in these areas than in the central area. Although no gorilla deaths were directly due to poachers, many gorillas were wounded by antelope snares. The infrequent patrols could not reduce poaching activity, and thus probably prevented gorillas from remaining in thenewly acquired habitat. In the Virunga region, where another subspecies of gorillas (Gorilla gorilla beringeO is surviving thanks to the active conservation efforts of three countries (Zaire, Rwanda and Uganda) and international organizations, a population census of gorillas has been conducted periodically since 1973. In 1959-60, Schaller (1963) estimated that there were 400-500 gorillas in 400 km 2 of the Virunga region. The mean group size was 16.9 individuals and the percentage of immature individuals was 44.2%, from calculations made on 10 groups (Table 4). In the 1960s and early 1970s, the Park area was drastically reduced (Rwanda side: 40%) for cultivation, as well as cattle encroachment and poaching, and affected by civil war in Zaire (Fossey, 1983). Only 261-290 gorillas were counted in the 1973 census (Harcourt et al., 1981) and the population had gradually decreased, as observed in the 1978 (Weber & Vedder, 1983) and 1981 (Aveling & Harcourt, 1983) censuses. As a result of improvements in conservation, the population has now increased slightly (Aveling & Aveling, 1987). A change in the Virunga gorilla population in the 1960s and early 1970s was the marked decrease in the Table 4. Comparison of the mean group size and the percentage of immature individuals in groups between different censuses in the Virunga and the Kahuzi regions
Estimated Mean population size group size
Virunga 1959-60 1973 1978 1981 1986
400-500 261-290 252-285 242-266 279-307
16-9 7-9 8.8 8.5 9.2b
44.2 39.8 35.8 39.7~ 48. V
Kahuzi 1978 1978-79 1990
230-250 223 258-284
14.3 15.6 10.8
44-4 48.5 42.5
'Guarded group', 48.5%; 'unguarded group', 33% (Aveling & Harcourt, 1983). b 'Protected group', 10-4; 'non-protected group', 7.1 (Aveling & Aveling, 1987). c 'Protected group', 50-8%; 'non-protected group', 40.8% (Aveling & Aveling, 1987).
size of the population and mean group size. However, since gorillas ceased to range in some areas where they had previously been found (Harcourt & Fossey, 1981), it is likely that the population density within the gorillas' range may not in fact have decreased prominently in the 1970s and 1980s. The concentration of gorillas in the small protected area may possibly have increased contact between groups or between groups and solitary males, and may have stimulated frequent female transfers, resulting in the decrease of mean group size (Yamagiwa, 1988). Aveling and Aveling (1987) regarded an increase in the mean group size and the percentage of immature individuals in the population to be a sign of an increasing population, by comparison with the Virunga results. This conclusion is based on the higher figures, calculated from 'guarded groups' or 'protected groups' which were monitored by the Karisoke Research Center and the Mountain Gorilla Project, rather than analogous figures from 'unguarded groups' or 'non-protected groups' (Aveling & Harcourt, 1983; Aveling & Aveling, 1987). The percentage of immature individuals calculated from our census indicates an increase and suggests that gorillas in the Kahuzi-Biega National Park still maintain a healthy population. However, frequent female transfers possibly caused by the concentration of gorillas in the central protected area might be responsible for the recent reduction of mean group size in this area. Frequent female transfers may result in or be the result of infanticides by silverback males (Fossey, 1984) and may thus be accompanied by a decrease in size of the next generation. The decrease in the mean number of dependent immature individuals per group, as detected in this census, suggests that infanticides may have occurred over the past 10 years. At least one abortion was recorded during several encounters between the Maeshe and Nindja Groups in 1990 (M.O. Mankoto, pers. comm.). More international aid is needed to increase the numbers of Park Guards in the northern and southern parts of the Park, so that more frequent patrols can provide new and safe habitats for gorillas. It is particularly urgent and important to protect the southern part from human disturbances. As Schaller (1963) and Goodall and Groves (1977) have reported, the eastern gorilla population has been divided over the past decades into small isolated 'pockets'. Most gorillas inhabiting montane forest are separated from their neighbours in the lowland forest. Recent studies in the Kahuzi-Biega National Park extension (lowland) imply that gorillas in both habitats show a similar ranging pattern while maintaining different food habits (Yamagiwa et al., 1989; Yumoto et al., 1989). Gorillas in the Kahuzi region also extend over a wider home range than those in the Virunga region (Casimir, 1975a; Goodall, 1977; Yamagiwa, 1983). In order to clarify evolutionary processes and the differentiation of eastern gorillas, and to retain the possibility for outbreeding or genetic and ecological independence, it is necessary for us to establish optimal methods for
Eastern gorillas in Zaire gorilla conservation in this region. Another population census and more detailed research on gorillas will be needed in the near future. ACKNOWLEDGEMENTS This study was financed by the Monbusho (Ministry of Education, Science and Culture, Japan) International Scientific Research Program and by a Grant from G T Z fDeutsche Gesellschaft ftir Technische Zusammenarbeit), in cooperation with CRSN (Centre de Recherches en ~iences Naturelles) and I Z C N (Institut Zairois pour Conservation de la Nature). We would like to express our thanks to Dr Zana Ndontoni, Dr Mankoto ma Mbaelele, Mr Mankoto ma Oyisenzoo, Mr Ngoro Likuwani, Mr Mashagiro Menshi for their administrative help and hospitality. We are also greatly indebted to Mr N d u m b o Bosilana and to all the Guides, Guards, and Field Assistants of the Kahuzi-Biega National Park for their technical help and hospitality throughout the fieldwork. REFERENCES ~veling, C. & Harcourt, A. H. (1983). A census of the Virunga gorillas. Oryx, 18, 8-13. Aveling, R. & Aveling, C. (1987). Report from the Zaire Gorilla Conservation Project. Primate Conserv., 8, 162~1. Casimir, M. J. (1975a). Feeding ecology and nutrition of an eastern gorilla group in the Mt Kahuzi region (R6publique du Zaire). Folia Primatol., 24, 81-136. Casimir, M. J. (1975b). Some data on the systematic position of the eastern gorilla population of the Mt Kahuzi region. Z. Morph. Anthrop., 66, 188-201. Casimir, M. J. (1979). An analysis of gorilla nesting sites of the Mt Kahuzi region (Zaire). Folia Primatol., 32, 290-308. Casimir, M. J. & Butenandt, E. (1973). Migration and core area shifting in relation to some ecological factors in a mountain gorilla group in the Mt Kahuzi region. Z. Tierpsychol., 33, 514-22. Fossey, D. (1983). Gorillas in the Mist. Houghton Mifflin Co., Boston. Fossey, D. (1984). Infanticide in mountain gorillas Gorilla gorilla beringei with comparative notes on chimpanzees. In
Infanticide, ed. G. Hausfater & S. B. Hardy. Aldine, New York, pp. 217-35. Goodall, A. G. (1977). Feeding and ranging bchaviour of a mountain gorilla group Gorilla gorilla beringei in the Tshibinda-Kahuzi region (Zaire). In Primate Ecology, ed. T. H. Clutton-Brock. Academic Press, London, pp. 450-79. Goodall, A. G. & Groves, C. P. (1977). The conservation of the eastern gorillas. In Primate Conservation, ed. H. S. H. Prince Rainier I I I & G. H. Bourne. Academic Press, New York, pp. 599-637. Groves, C. P. & Stott, K. W. (1979). Systematic relationships of gorillas from Kahuzi, Tshiaberium and Kayonza. Folia Primatol., 32, 161-79. Harcourt, A. H. & Fossey, D. (1981). The Virunga gorillas: decline of an 'island' population. Afr. J. Ecol., 19, 83-97. Harcourt, A. H., Fossey, D. & Sabater Pi, J. (1981). Demography of Gorilla gorilla. J. Zool., Lond., 195, 215-33. MacKinnon, J. (1978). The Ape within Us. William Collins, New York. Mankoto, M. O. (1988). La gestion du Pare National du Kahuzi-Biega (Zaire). Cah. Ethol. Appl., 8, 447-50. Mwanza, N. & Yamagiwa, J. (1989). A note on the distribution of primates between the Zaire-Lualaba River and the African Rift Valley. Grant-in-Aid for Overseas Scientific Research Report: Interspecies Relationships of Primates in the Tropical and Montane Forests 1, pp. 5-10. Murnyak, D. F. (1981). Censusing the gorillas in KahuziBiega National Park. Biol. Conservl, 21, 163-76. Schaller, G. B. (1963). The Mountain Gorilla: Ecology and Behavior. University of Chicago Press, Chicago. Weber, A. W. & Vedder, A. (1983). Population dynamics of the Virunga gorillas: 1959-1978. Biol. Conserv., 26, 341-66. Yamagiwa, J. (1983). Diachronic changes in two eastern lowland gorilla groups (Gorilla gorilla grauerl) in the Mt Kahuzi region, Zaire. Primates, 24, 174-83. Yamagiwa, J. (1988). Local differences observed in socioecological characters of highland gorillas. J. Aft. Stud., 33, 19~t4 (in Japanese). Yamagiwa, J., Maruhashi, T., Yurnoto, T. & Mwanza, N. (1989). A preliminary survey on sympatric populations of Gorilla g. graueri and Pan t. schweinfurthii in eastern Zaire. Grant-in-Aid for Overseas Scientific Research Report: Interspecies Relationships of Primates in the Tropical and Montane Forests, 1, pp. 23~10. Yumoto, T., Maruhashi, T., Yamagiwa, J. & Mwanza, N. (1989). Feeding habit of primates in the tropical forests of eastern Zaire. Grant-in-Aid for Overseas Scientific Research Report: Interspecies Relationships of Primates in the Tropical and Montane Forests, 1, pp. 41-51.