Mitochondrial DNA Variability in Grauer's Gorillas of ... - CiteSeerX

Mitochondrial DNA Variability in Grauer’s Gorillas of Kahuzi-Biega National Park K. Saltonstall, G. Amato, and J. Powell

Eastern lowland gorillas (Gorilla gorilla graueri) are the least studied of the three gorilla subspecies, particularly at the molecular level. We sequenced an internal region of the mitochondrial DNA cytochrome oxidase subunit II (COII) region and a hypervariable portion of the mitochondrial DNA control region (D-loop) from wild gorillas in both the montane and lowland habitats of Kahuzi-Biega National Park, Democratic Republic of Congo. All individuals (n 5 38) were identical at the COII region; this sequence indicates that diagnostic sites previously suggested for gorilla subspecies may be valid. Low variability was found within the D-loop region from a subset of the individuals (n 5 15) sequenced for COII. Haplotype frequencies differed between the two habitats, suggesting a level of population subdivision that may have demographic consequences. These results also support the distinction of two distinct clades of gorillas comprised of western populations (G. g. gorilla) and eastern populations (G. g. graueri and G. g. beringei). Future management of Kahuzi-Biega National Park should ensure that sufficient habitat remains to prevent further genetic isolation of gorillas in the montane section of the park.

From the Yale School of Forestry and Environmental Studies (Saltonstall) and the Department of Ecology and Evolutionary Biology (Saltonstall and Powell), Yale University, New Haven, Connecticut, and the Wildlife Conservation Society, Bronx, New York (Amato). This project was completed as part of the requirements for the Masters of Forest Science degree at the Yale School of Forestry and Environmental Studies. Funding for this research came from the Wildlife Conservation Society Research Fellows Program. The acquisition and maintenance of facilities was supported by grants from the National Science Foundation. We would like to thank Jefferson Hall and all participants in the Grauer’s Gorilla and Eastern Forest Survey who assisted in hair sample collection. Tony Goldberg provided advice on sample extraction procedures and Gisella Caccone gave technical advice on laboratory procedures. We thank James Gibbs, Rob DeSalle, and several anonymous reviewers for comments on the manuscript. Address correspondence to K. Saltonstall, Department of Ecology and Evolutionary Biology, 427 Osborn Memorial Laboratories, P.O. Box 208104, Yale University, New Haven, CT 06520–8104 or e-mail: kristin.saltonstall@. yale.edu. Journal of Heredity 1998;89:129–135

Three subspecies of gorillas are currently recognized: western lowland gorillas (Gorilla gorilla gorilla), eastern lowland or Grauer’s gorillas (G. g. graueri), and mountain gorillas (G. g. beringei; Groves 1970). Of the three subspecies, Grauer’s gorilla is the least studied. An estimated 8150 individuals (4500–11,800) of this subspecies inhabit the lowland forest parks and reserves of eastern Democratic Republic of Congo ( Hart and Hall 1996). Gorillas in all regions of Africa are threatened by human encroachment in the forms of hunting for food, capture of infants for sale, and land clearing. Eastern Democratic Republic of Congo has some of the highest human densities in Central Africa, with a population density of 300 individuals/km2 in some areas ( Institut National de la Statistique 1984). This human density has recently been augmented by the influx of refugees from Rwanda into eastern Democratic Republic of Congo, creating additional pressures on the available forest habitat. Such threats have led the IUCN to declare the graueri subspecies as endangered. Knowledge of the genetic structure of populations of endangered species is important for any conservation program targeting individual species (Amato and Gatesy 1994; Avise 1996; Hedrick and Miller

1992). Important genetic issues include the degree of evolutionary differentiation and novelty represented by a population, as well as the relative relatedness of individuals and possible effects of inbreeding (Amato 1991; Templeton 1986). Little is known about genetic diversity in G. g. graueri. Previous studies used limited samples and have indicated an evolutionary split between eastern and western gorillas (Garner and Ryder 1992, 1996; Ruvolo et al. 1994). The degree to which graueri are evolutionarily distinct from mountain gorillas and estimation of the levels of gene flow that occur between populations of the subspecies warrants further study. Such information is needed to establish conservation priorities for eastern gorillas. The issue of intraspecific variability in hominoids has been well studied. For gorillas, molecular and morphological data currently support two distinct clades— one composed of western lowland populations and the other including the two eastern subspecies (Garner and Ryder 1992; Ruvolo et al. 1994). Much of this work has been done, however, using samples from zoo animals whose origins are unknown. Of the three subspecies, Grauer’s gorilla has been the least represented in such studies, presumably due to

129

their scarcity in zoos and the lack of field research focused on the subspecies. Recently developed techniques using hair samples for DNA extraction ( Higuchi et al. 1988; Morin et al. 1993; Woodruff 1993) and the polymerase chain reaction (PCR; Innis et al. 1988) now allow wild individuals to be studied using noninvasive techniques for sample collection. Although consensus has been reached that all gorilla populations west of the African Lakes are distinct from the mountain gorillas inhabiting the Virunga Volcano region (Groves and Stott 1979), the systematic designation of individual populations under the name of ‘‘eastern gorilla’’ has been subject to much debate. One such population inhabits the forests surrounding Mt. Kahuzi in Kahuzi-Biega National Park. Groves (1967, 1970) assigned them to the beringei subspecies based on postcranial features but considered them to be transitional between the true graueri of the nearby Utu lowland region and the beringei populations of the Virunga region. In revisiting this designation, Groves and Stott (1979) concluded that the Kahuzi gorillas belong to the subspecies graueri but have some beringei features, suggesting that at some time in the past they were in reproductive contact with the Virunga population but since then have been more influenced by the true graueri of the Utu region. This classification holds today and indicates that clinal variation may exist between gorillas living in different habitat types. Morphological differences between the gorillas in the montane section of Kahuzi-Biega National Park and those in the lowlands are well documented, but these populations have never been examined genetically (Groves 1970; Groves and Stott 1979). This study compares patterns of variation in mitochondrial DNA between the two populations. We amplified DNA from hair samples obtained from wild gorillas in Kahuzi-Biega National Park. Due to the degraded state of many of the samples, we amplified only short regions (250–300 base pairs) of DNA. We conducted an analysis of a 252 base pair internal region of the mitochondrial protein-coding gene for cytochrome oxidase subunit II (COII) and a 277 base pair hypervariable portion of the mitochondrial DNA displacement loop ( D-loop) control region to examine patterns of genetic diversity within the graueri subspecies and for comparison with other gorilla subspecies. First, we used the mtDNA data to evaluate the hypothesis that within the different habitats of Kahuzi-Biega National

130 The Journal of Heredity 1998:89(2)

Figure 1. Map of Kahuzi-Biega National Park and the surrounding area. Zone I refers to the lowlands and zone II represents the montane sector of the park.

Park, gorillas may have different histories of reproductive contact with other populations. Second, we compared diversity within graueri to other subspecies of gorilla and examined it in relation to previously suggested diagnostic sites which indicate genetic differentiation from other gorilla subspecies (Garner and Ryder 1996; Ruvolo et al. 1994). Finally, we discuss the implications of these results for Grauer’s gorilla conservation and management.

Materials and Methods Study Area Kahuzi-Biega National Park is located in the Kivu region of eastern Democratic Republic of Congo (278339 288519 west and 18529 28299 south; Figure 1). The park is divided into a 600 km2 montane sector and a 5400 km2 lowland sector which are connected by a narrow corridor. Although it is forested, the corridor is surrounded by a dense human population and the degree to which it is used by gorillas is unknown. In the past solitary gorilla nests have been found in the corridor region (Refisch 1991), however, today forest cover has been destroyed and much of the area is used for agriculture ( Inogwabini et al., in

preparation). We divided the park into two sampling zones: zone I in the lowland extension of the park and zone II in the montane sector of the park ( Figure 1). Sample Preparation Hair samples were collected from vacated day or night nests of wild individuals and stored in paper envelopes at ambient temperatures. Samples from a total of 16 family groups in zone I and 3 family groups ( Nindja, Maheshe, and Mushamuka) in zone II were sequenced ( Table 1). Samples in zone I were collected during a survey of large mammals in 1994 in which line transects were walked and gorilla nest sites identified and mapped. The nature of the survey technique and the means of data collection allowed gorilla groups to be identified based on group size. Samples collected from nest sites thought to have been created by the same family group were not included in this analysis in order to avoid duplicate sampling of individuals. In some cases, multiple individuals from a family group were sequenced to increase the total sample size ( Table 1). Where multiple individuals within a group were found to have the same haplotype for the loci examined here, only single represen-

quencing reactions were performed using the Dye Terminator Cycle Sequencing kit (Perkin Elmer) and an automated sequencer (ABI 373). Samples were sequenced in both directions and individuals were sequenced twice to detect Taq DNA polymerase errors. Where differences were observed, individuals were resequenced from new PCR products to resolve any discrepancies.

Table 1. Family group, identification, and geographic origin of Gorilla gorilla graueri individuals in Kahuzi-Biega National Park Family group

Individuals

Zone

Location

A B C D E F G H I J K L M N O P Mushamuka Nindja Maheshe

G11, G14 G115 G136 G142, G143 G23 G27 G212 G216 G219 G223 G232 G31 G33 G34 G39 G413, G417, G420 G51, G52, G53, G54, G55 G57, G59, G510, G513, G514, G515, G516, G517, G520 G524, G525, G528, G530, G531

I I I I I I I I I I I I I I I I II II II

Busakala Busakala Busakala Busakala Mankese Mankese Mankese Mankese Mankese Mankese Mankese Mankese Mankese Mankese Mankese Nzovu Tshivanga Tshivanga Tshivanga

tatives of each group were used for comparisons. Extractions were performed using 250 ml 5% Chelex solution, with 1.5 ml 20 mg/ml proteinase K and 7 ml 1 M DTT (Goldberg T, personal communication; Walsh et al. 1991). One to five hairs were used per individual, depending on the number of hairs collected. Following overnight digestion at 568C, samples were vortexed, placed in a boiling water bath for 8 min, and then centrifuged for 3 min so that the resulting supernatant could be removed. The Chelex resin was then resuspended in 200 ml of water, vortexed, and centrifuged again to elute any remaining DNA. The resulting supernatant was removed and added to the previous solution. This final DNA template was concentrated using Microcon 30 microconcentrators to a volume of 20 to 50 ml. PCR Conditions The polymerase chain reaction (PCR) was used to create double-stranded DNA using

Results A total of 39 individuals were sequenced for the COII gene, all of which were identical. Of these 39, 20 (representing 16 family groups) came from the zone I lowlands and 19 (representing three family groups) from the montane zone II. This sequence is also identical to the one obtained by Ruvolo et al. (1994) for the graueri subspecies, based on sequencing of one zoo individual of unknown origin. The sequence differs from the other eastern gorilla subspecies (G. g. beringei) at two positions (Ruvolo et al. 1994; Table 2a). D-loop sequences were obtained from 15 individuals. These samples represent a subset of those analyzed with the COII gene, with 5 from zone I and 10 from zone II ( Table 3). Although attempts were made to obtain sequences from all individuals at this region, the degraded nature of the samples prevented us from successfully amplifying all of them. Six mtDNA haplotypes were found, differing by 0.4 to 1.8% ( Figure 2 and Table 4). Of these, haplotypes designated EL1, 2, 3, and 6 were previously identified in another study, which also found two other haplotypes, here designated as EL5 and EL8, respectively ( Table 4; Garner and Ryder 1996). This analysis also includes a variable internal region of approximately 30 base pairs cen-

reaction conditions described by Innis et al. (1988). Primer sequences for the COII fragment were C187G: 59 TCAGACGCCCAAGAAATAGAGA 39, and D402G: 59 TCGGTTGTCGACGTCAAGGAGT 39. These primers were modified from those used by Ruvolo et al. (1994) for specificity with G. g. graueri based on published sequence (Ruvolo et al. 1994). Primer sequences for the D-loop region are described in Garner and Ryder (1996). The reaction cycle consisted of denaturation at 948C for 30 to 60 s, annealing at 508C–558C for 30 to 60 s, and DNA extension at 728C–748C for 1 to 5 min. Samples were amplified for 40 cycles. In most cases, reamplification using 1 ml of PCR product was necessary to produce sufficient quantities of DNA for sequencing. DNA Sequencing PCR products were purified for sequence analysis with the GeneClean III kit ( Bio101) using the directions supplied by the manufacturer. Single-stranded se-

Table 2. Diagnostic positions which distinguish the three subspecies of gorilla (A) within COII, (B) within D-loop (A)

WL MT EL

a

22

39

72

105

108

117

181

192

195

204

A G G

C A A

T C C

T C C

C C T

C T T

A A G

C T T

T C C

C T T

( B)

WL MT EL

b

1

11

17

27

53

105

106

117

122

123

124

135

136

141

143

144

145

151

153

154

161

199

220

226

241

A G G

A G G

C C T

C T T

G G T

A C C

A C C

C T T

A T T

C — —

A — —

T C C

C A A

C — —

C A A

C T T

C T C

T T C

T A A

G C C

G G A

A A G

A C C

C T C

T C C

Site numbers refer to positions within the gene. Western lowland (WL) gorilla (Gg04; Ruvolo et al. 1994). Substitutions are shown in mountain gorilla (MT; Gg06; Ruvolo et al. 1994) and eastern lowland gorilla ( EL; Gg05; Ruvolo et al. 1994; this study) sequences. b Reference sequence is a western lowland gorilla (WL191; Garner and Ryder 1996). Mountain and eastern lowland gorilla substitutions are based on sequences described in Garner and Ryder (1996) and this study.

a

Saltonstall et al • Mitochondrial Variation in Grauer’s Gorillas 131

Table 3. D-loop haplotypes identified for Grauer’s gorillas D-loop haplotype EL1 EL2b EL3c EL4 EL6d

a

EL7

Individual(s) G223 G31 G115 G33 G34, G51, G54, G59, G513, G514, G515, G516, G517, G530 G531

Corresponds to Corresponds to c Corresponds to d Corresponds to

a

b

EL058 EL094 EL055 EL099

(Garner (Garner (Garner (Garner

and and and and

Ryder Ryder Ryder Ryder

1996). 1996). 1996). 1996).

tering on a string of C’s not included in the phylogenetic analysis by Garner and Ryder (1996). Because gorillas are the species of interest here and intraspecific alignment of this region is possible, we included it in our analysis. The six D-loop haplotypes identified here were not distributed randomly throughout the two study zones. Haplotype EL6 predominated in zone II and was found in all three family groups that were sampled. One individual in zone I also carried this sequence ( Table 3). One other sequence, EL7, was found in one individual of zone II. The remaining four haplotypes described here were each found in one individual from zone I ( Table 3). Three of these ( EL1, 2, and 3) were previously identified by Garner and Ryder (1996), who also found one other sequence ( EL5) in this habitat region that was not identified in this study.

Discussion Genetic Differences Within Gorillas in Kahuzi-Biega National Park Previous morphological studies on the gorillas in Kahuzi-Biega National Park have distinguished the montane from the lowland populations on the basis of both cranial and postcranial indices (Casimir 1975; Goodall and Groves 1977; Groves 1970; Groves and Stott 1979). Inclusion of the montane population under both the graueri and beringei designations has been called for, however, it has never been suggested that gorillas in the montane sector be given a subspecies rank despite their distinctive morphological characters (Groves and Stott 1979). This study affirms this conclusion by showing that mitochondrial genetic data clearly links the two populations. The protein-coding COII region is invari-

132 The Journal of Heredity 1998:89(2)

Figure 2. Alignment showing variation in the eastern lowland gorilla mitochondrial D-loop gene within a 277 base pair internal fragment. Sequences EL5 and EL8 are from Garner and Ryder (1996).

ant between the study zones indicating relatively recent common ancestry for the two populations. It also shows no difference with previously published sequences for this gene in a graueri individual of unknown origin (Ruvolo et al. 1994). In that

study, graueri were found to differ from beringei at only two sites in the region examined here ( Table 2a). Using a moderately conserved gene region such as COII results in fewer variable sites for character analysis (Ruvolo et al. 1993). However, it

Table 4. Percent nucleotide sequence differences between a 277 base pair region of D-loop sequences determined from mountain and eastern lowland gorillas

MT045 MT074 MT374 MT504 EL1 EL2 EL3 EL4 EL5 EL6 EL7 EL8

MT045

MT074

MT374

MT504

EL1

EL2

EL3

EL4

EL5a

EL6

EL7

EL8b

—

1.1 —

1.4 0.4 —

1.1 0.7 1.1 —

6.1 7.2 7.6 6.5 —

6.9 7.9 8.3 7.2 0.7 —

6.1 7.2 7.6 6.5 0.7 0.7 —

6.1 7.2 7.6 6.1 0.7 0.7 0.7 —

6.5 7.6 7.9 6.9 1.1 1.1 0.4 1.1 —

6.1 7.2 7.6 6.5 1.4 1.4 1.4 0.7 1.8 —

6.5 7.6 7.9 6.9 0.4 0.4 0.4 0.7 0.7 1.1 —

7.2 7.9 8.7 7.6 1.1 1.1 1.1 1.1 1.4 1.8 0.7 —

All mountain gorilla samples are of wild origin (Garner and Ryder 1996). Samples EL1–EL5 are wild individuals living in the zone I lowlands of Kahuzi-Biega National Park. EL6 and EL7 are wild individuals living in the montane zone II. EL8 is a captive individual of unknown origin (Garner and Ryder 1996; this study). a Corresponds to EL086 (Garner and Ryder 1996). b Corresponds to EL057 (Garner and Ryder 1996).

is likely that the two phylogenetically diagnostic characters supported by this expanded dataset are less likely to be affected by problems with homoplasy than are characters in other regions, such as the hypervariable D-loop region (Amato 1994; Amato et al. 1995; Ruvolo 1994). For these reasons, inclusion of the population in the montane zone II in the graueri designation rather than beringei seems warranted. In contrast to the COII fragment, the hypervariable portion of the D-loop control region analyzed here showed several base changes throughout the samples ( Figure 2). This region has previously been observed to evolve on the order of 10 times faster than COII (Garner and Ryder 1996; Ruvolo et al. 1993). Several haplotypes were found, but their distribution and frequency varied by zone. Zone II was dominated by sequence EL6 (all groups, 9 of 10 individuals), previously identified by Garner and Ryder (1996). While only three family groups were sampled here, it is probable that a greater number of matrilineal lines are actually represented here due to female transfer between groups. Recent census information indicates that the gorillas in this region of the park are now confined to a 200 km2 central region due to human encroachment ( Inogwabini et al., in preparation; Vedder et al., unpublished data). Within this small area, at least 25 groups are found ( Vedder et al., unpublished data), making it likely that contact between groups occurs regularly. Thus EL6 may be widespread throughout zone II. Evidence for contact between gorillas in both zones also exists. Sequence EL6 also appeared in a lone male sample (G34) collected in zone I. Further, Garner and Ryder (1996) identified sequence EL2 in an individual from the Mushamuka group found

in zone II. In our study, we found this sequence in a lone male from zone I (G31). This suggests that in the recent past, if not still today, gorillas have traversed the corridor connecting the montane and lowland habitats bringing the two populations into reproductive contact. Given the population size of the montane area (n 5 243; Vedder et al., unpublished data), had gene flow ceased between the two populations, greater subdivision between the two zones might have occurred due to genetic drift. Diversity Within and Differences Between Gorilla Subspecies An issue of importance for both conservation and evolutionary biologists regarding variability within a species is that of variation within populations and between populations. Without knowledge of sample origin, it is impossible to make such distinctions. Ruvolo et al. (1994) have distinguished four haplotypes within western lowland gorillas based on COII sequences from zoo individuals of unknown origin. Variability in both mountain and eastern lowland gorillas at the COII gene appears to be much lower than in western gorilla populations (Ruvolo et al. 1994). Today graueri are found in 11 distinct populations which have little or no contact with each other due to human presence ( Hall et al., in press a) while mountain gorillas are found in only two populations which also are geographically isolated. Within graueri, this study demonstrates that little to no variability exists at COII among the gorillas of Kahuzi-Biega National Park, but no statement can be made about levels of variation within the subspecies as a whole. Among mountain gorillas the same holds true. Within western lowland gorillas, recent data have shown variability

within regional populations to be much less than in the subspecies as a whole (Garner and Ryder 1996). Until more geographically isolated populations have been sampled, the level of differentiation at the molecular level within eastern gorilla populations remains unclear. For conservation purposes, correlation of genetic variation with geographic location is essential (Amato and Gatesy 1994; Avise 1989). These results also support previous designations of phylogenetic clades in the gorilla species (Garner and Ryder 1992; Ruvolo et al. 1994). Based on an earlier study using COII sequences from six individuals (four western lowland, one eastern lowland, and one mountain gorilla; Ruvolo et al. 1994), diagnostic sites for the three subspecies of gorilla may be identified ( Table 2a). Eastern gorillas (graueri and beringei) can be distinguished from western gorillas (gorilla) at eight positions. Two additional sites are unique to graueri. This study greatly expands the dataset for eastern lowland gorillas at this site by confirming marker positions for the two populations examined here. However, additional study on other populations of Grauer’s gorilla that are geographically isolated is warranted to validate the sites for the subspecies as a whole. Garner and Ryder (1996) identify a number of positions within the D-loop fragment that are consistent within the subspecies based on their samples. Our results also concur with their findings. Eastern gorillas can be distinguished from western gorillas at 18 positions within this region. The graueri subspecies can be distinguished from beringei at another seven positions ( Table 2b). Differences between individual eastern gorillas range up to 8.7% in this region ( Table 4), making the

Saltonstall et al • Mitochondrial Variation in Grauer’s Gorillas 133

D-loop a more useful mitochondrial region for estimation of genetic variability between populations of a subspecies. While these diagnostic positions are based on limited samples from two known graueri populations as well as several individuals of unknown origin, they provide information on regional if not subspecific differentiation for Grauer’s gorillas. These diagnostic markers could be extremely useful in identifying the origins of animals captured for illegal trade (Garner and Ryder 1996). Management Implications Our results clearly demonstrate that despite their morphological differences, the gorillas inhabiting the montane and lowland regions of Kahuzi-Biega National Park are closely related. In spite of the intense human pressure in the regions surrounding them, gorillas in the two habitat regions are possibly in reproductive contact even today. However, the levels of contact may be low and may not be significant from a demographic perspective. Low levels of gene flow and dispersal rates, as well as geographical structuring in mtDNA, can imply demographic independence among populations (Avise 1995). This is particularly true in species such as gorillas where females, although they may transfer between family groups ( Yamagiwa 1983), do not travel great distances to mate ( Harcourt 1978). The only D-loop sequences common to both zones were found in lone males from zone I (this study; Garner and Ryder 1996). Within Kahuzi-Biega, gorilla populations in the two study zones are separated by nonforested land that is inhabited and used by humans. Such breaks in forest cover probably hinder movement of gorillas between populations. Dispersal between the gorilla populations of zones I and II appears to be occurring at an evolutionarily significant level of gene flow. However, at the demographic level it may not be enough to sustain the small montane population in the long term, due to low recruitment rates of females from the nearby lowlands (Avise 1995; Taylor and Dizon 1996). Despite this, every effort should be made to restore and maintain the corridor connecting the two sections of the park so that gorillas, and other forest dwelling species, may continue to travel between the two habitats. Since the montane population is one of the few gorilla populations known to have remained relatively stable in size (Murnyak 1981; Vedder et al., unpublished data; Yamagiwa et al. 1993)

134 The Journal of Heredity 1998:89(2)

and the lowland population makes up the core of the largest population of Grauer’s gorillas today ( Hall et al., in press b), it is critical that they remain protected. Noninvasive Sampling Finally, we would like to comment on the difficulties of working with hair samples, especially samples such as those analyzed in this study that have been exposed to the elements and are in a somewhat degraded state. Because of the difficulties in obtaining adequate amounts of DNA, the sample size analyzed here is small. DNA extractions were attempted from 104 hair samples, but sequencing was possible in only 39 of these samples. Length of time before collection of hair samples appears to be important in the success of DNA amplification, as samples from the montane area which were collected from nests created by gorillas the previous night had much higher rates of successful amplification than nests from the lowlands where length of exposure was not known. However, several samples for which the age of the nest from which they were collected was estimated to be 4 to 6 weeks were successfully amplified and sequenced, indicating the usefulness of this noninvasive method of obtaining DNA samples from wild individuals. This dataset is the largest to analyze relationships between graueri populations. Since all of the individuals tested here are also of known origin, we have been able to draw some conclusions about the extent of variability within gorillas of KahuziBiega National Park. With threatened or protected species, such data are essential, as are noninvasive methods of obtaining samples. Further studies are necessary to assess the variability of the subspecies as a whole as well as to confirm the diagnostic marker positions that we identify. Molecular studies of mitochondrial as well as nuclear DNA (e.g., microsatellites) can provide relevant information to a range of disciplines, including conservation and management, evolutionary biology, and taxonomy.

References Amato G, 1991. Species hybridization and protection of endangered animals. Science 253:250. Amato G, 1994. A systematic approach for identifying evolutionarily significant units for conservation: the dilemma of subspecies (PhD dissertation). New Haven, Connecticut: Yale University. Amato G and Gatesy J, 1994. PCR assays of variable nucleotide sites for identification of conservation units. In: Molecular ecology and evolution: approaches and

applications (Schierwater B, Streit B, Wagner GP, and DeSalle R, eds). Basel, Switzerland: Birkhauser Verlag. Amato G, Wharton D, Zainuddin ZZ, and Powell JR, 1995. Assessment of conservation units for the Sumatran rhinoceros (Dicerorhinus sumatrensis). Zoo Biol 14: 395–402. Avise JC, 1989. A role for molecular genetics in the recognition and conservation of endangered species. Trends Ecol Evol 4:279–281. Avise JC, 1995. Mitochondrial DNA polymorphism and a connection between genetics and demography of relevance to conservation. Conserv Biol 9:686–690. Avise JC, 1996. Introduction: the scope of conservation genetics. In: Conservation genetics: case histories from nature (Avise JC and Hamrick JL, eds). New York: Chapman and Hall; 1–9. Casimir MJ, 1975. Some data on the systematic position of the eastern gorilla populations of the Mt. Kahuzi region (Republique du Zaire). Z Morph Anthropol 66:188– 201. Garner KJ and Ryder OA, 1992. Some applications of PCR to studies in wildlife genetics. Symp Zool Soc Lond 64:167–181. Garner KJ and Ryder OA, 1996. Mitochondrial DNA diversity in gorillas. Mol Phyl Evol 6:39–48. Goodall A and Groves CP, 1977. The conservation of eastern gorillas. In: Primate conservation. (Prince Rainier III and Bourne GH, eds). London: Academic Press; 450–479. Groves CP, 1967. Ecology and taxonomy of the gorilla. Nature 213:890–893. Groves CP, 1970. Population systematics of the gorilla. J Zool Soc Lond 161:287–300. Groves CP and Stott KW, 1979. Systematic relationships of the gorillas from Kahuzi, Tshiaberimu, and Kayonza. Folia Primatol 32:161–179. Hall JS, Saltonstall K, Inogwabini BI, and Omari I, in press a. Distribution, abundance, and conservation status of Grauer’s gorilla, Gorilla gorilla graueri. Oryx. Hall J, White LJT, Inogwabini BI, Omari I, Simons-Morland H, Williamson EA, Walsh P, Saltonstall K, Sikubwabu C, Bonny D, Kaleme KP, Vedder A, and Freeman K, in press b. A survey of Grauer’s gorillas (Gorilla gorilla graueri) and chimpanzees (Pan troglodytes schweinfurthi) in the Kahuzi-Biega National Park lowland sector and adjacent forest in eastern Zaire. Int J Primatol. Harcourt AH, 1978. Strategies of emigration and transfer by primates, with particular reference to gorillas. Z Tierpsychol 48:401–420. Hart JA and Hall JS, 1996. Status of eastern Zaire’s forest parks and reserves. Conserv Biol 10:316–324. Hedrick PW and Miller PS, 1992. Conservation genetics: techniques and fundamentals. Ecol Appl 2:30–46. Higuchi R, von Beroldingen CH, Sensabaugh GF, and Erlich HA, 1988. DNA typing from single hairs. Nature 332:543–546. Innis MA, Myambo KB, Gelfand DH, and Brow MD, 1988. DNA sequencing with Thermus aquaticus DNA polymerase and direct sequencing of polymerase chain reaction-amplified DNA. Proc Natl Acad Sci USA 85:9436– 9440. Institut National de la Statistique, 1984. Recensement scientifique de la population 1984: projections demographiques Zaire et regions, 1984–2000. Ministere du Plan et Amenagement du Territoire, Kinshasa. Morin PA, Wallis J, Moore JJ, Chakraborty R, and Woodruff DS, 1993. Non-invasive sampling and DNA amplification for paternity exclusion, community structure and phylogeography in wild chimpanzees. Primates 34: 347–356. Murnyak DF, 1981. Censusing the gorillas in Kahuzi-Biega National Park. Biol Conserv 21:163–176. Refisch J, 1991. Pre´sence des grands mammife`res dans le Parc National de Kahuzi-Biega et l’influence humaine

sur les populations des animaux. Rapport dans le cadre du Projet IZCN/GTZ—Conservation de la Nature Inte´gre´e, L’Est Zaire.

Taylor BL and Dizon AE, 1996. The need to estimate power to link genetics and demography for conservation. Conserv Biol 10:661–664.

Ruvolo M, 1994. Molecular evolutionary processes and conflicting gene trees: the hominoid case. Am J Phys Anthro 94:89–113.

Templeton AR, 1986. Coadaptation and outbreeding depression. In: Conservation biology: the science of scarcity and diversity (Soule ME, ed). Sunderland, Massachusetts: Sinauer; 105–116.

Ruvolo M, Pan D, Zehr S, Goldberg T, Disotell TR, and von Dornum M, 1994. Gene trees and hominoid phylogeny. Proc Natl Acad Sci USA 91:8900–8904.

Walsh PS, Metzger DA, and Higuchi R, 1991. Chelex 100 as a medium for simple extraction of DNA for PCRbased typing from forensic material. Biotechniques 10: 506–513.

Ruvolo M, Zehr S, von Dornum M, Pan D, Chang B, and Li J, 1993. Mitochondrial COII sequences and modern human origins. Mol Biol Evol 10:1115–1135.

Woodruff DS, 1993. Non-invasive genotyping of primates. Primates 34:333–346.

Yamagiwa J, 1983. Diachronic changes in two eastern lowland gorilla groups (Gorilla gorilla graueri) in the Mt. Kahuzi region, Zaire. Primates 24:174–183. Yamagiwa J, Mwanza N, Spandenberg A, Maruhashi T, Yumoto T, Fischer A, and Steinhauer-Burkart B, 1993. A census of the eastern lowland gorillas Gorilla gorilla graueri in Kahuzi-Biega National Park with reference to mountain gorillas G. g. beringei in the Virunga region, Zaire. Biol Conserv 64:83–89. Received December 4, 1996 Accepted July 10, 1997 Corresponding Editor: Stephen J. O’Brien

Saltonstall et al • Mitochondrial Variation in Grauer’s Gorillas 135