International Journal of Primatology

, Volume 31, Issue 1, pp 1–13

Maintenance of Multifemale Social Organization in a Group of Nomascus concolor at Wuliang Mountain, Yunnan, China

Authors

  • Peng-Fei Fan
    • Institute of Eastern-Himalaya Biodiversity ResearchDali University
    • State Key Laboratory of Genetic Resources and EvolutionKunming Institute of Zoology, Chinese Academy of Sciences
    • State Key Laboratory of Genetic Resources and EvolutionKunming Institute of Zoology, Chinese Academy of Sciences
Article

DOI: 10.1007/s10764-009-9375-9

Cite this article as:
Fan, P. & Jiang, X. Int J Primatol (2010) 31: 1. doi:10.1007/s10764-009-9375-9
  • 170 Views

Abstract

Many short-term studies have reported groups of black crested gibbons containing ≥2 adult females (Nomascus concolor). We report the stability of multifemale groups in this species over a period of 6 yr. Our focal group and 2 neighboring groups included 2 breeding females between March 2003 and June 2009. We also habituated 1 multifemale group to observers and present detailed information concerning their social relationships over a 9-mo observation period. We investigated interindividual distances and agonistic behavior among the 5 group members. The spatial relationship between the 3 adult members (1 male, 2 females) formed an equilateral triangle. A subadult male was peripheral to the focal group, while a juvenile male maintained a closer spatial relationship with the adult members. We observed little agonistic behavior among the adult members. The close spatial relationship and lack of high rates of agonistic behavior among females suggest that the benefits of living in a multifemale group were equal to or greater than the costs for both females, given their ecological and social circumstances. The focal group occupied a large home range that was likely to provide sufficient food sources for the 2 females and their offspring. Between March 2003 and June 2009, 1 adult female gave 2 births and the other one gave 1 birth. All individuals in the focal group survived to June 2009. A long-term comparative study focused on females living in multifemale groups and females living in pair-living groups would provide insight into understanding the evolutionary mechanisms of the social system in gibbons.

Keywords

agonisticgibbonmultifemaleNomascus concolorspatial distanceYunnan

Introduction

Gibbons were traditionally believed to live in monogamous groups characterized by 1 adult pair and their offspring (Leighton 1987). Sommer and Reichard (2000) summarized the traditional characteristics of monogamous grouping as strict territoriality, pair formation through natal dispersal, lifelong monogamy, mating exclusivity, nuclear families, and pair bonds advertised through duetting. However, increasing field data from different gibbon species challenges the view that gibbons are monogamous, with observations of multimale, multifemale groups (Fuentes 2000), extrapair copulations (Palombit 1994; Reichard 1995), and partner changes (Brockleman et al.1998).

Researchers have reported groups with 2 adult males in hoolock gibbons (Hoolock hoolock: Siddiqi 1986), white-handed gibbons (Hylobates lar: Brockleman et al.1998; Sommer and Reichard 2000), and siamangs (Symphalangus syndactylus: Lappan 2007), and observed adult males copulating with the group female in 3 groups of white-handed gibbons and 3 groups of siamangs (Brockleman et al.1998; Lappan, 2007; Sommer and Reichard 2000). Researchers have also reported groups with ≥2 females in several species, including the black crested gibbon (Nomascus concolor: Fan et al.2006; Jiang et al.1999), Hainan gibbon (N. hainanus: Liu et al.1989; Zhou et al.2005), white-handed gibbon (Sommer and Reichard 2000), pileated gibbon (Hylobates pileatus: Srikosamatara and Brockelman 1987), and hoolock gibbon (Ahsan 1995). Researchers have previously described the role of intrasexual aggression in maintaining socially monogamous grouping in several primate taxa (Callicebus moloch: Fernandez-Duque et al.1997; Hylobates agilis: Gittins 1979, 1980; Mitani 1987; Hylobates klossii: Tenaza 1975; Tilson 1981; Hylobates lar: Reichard and Sommer 1997; Hylobates muelleri: Mitani 1984; Symphalangus syndactylus: Chivers 1974). A multifemale group should break up if the costs exceeded the benefits for 1 or both females, which should then either emigrate or seek to restore a unifemale grouping by evicting the second female. The prediction is supported by observations of lar gibbons (Sommer and Reichard 2000), pileated gibbons (Srikosamatara and Brockelman 1987), and hoolock gibbons (Ahsan 1995). Although 2 females both carried infants in 1 lar gibbon group, this group existed for only ca. 8 months (Sommer and Reichard 2000). In the multifemale pileated gibbon group, each of 2 adult females carried infants, but the second adult female’s infant disappeared in ≤1 yr (Srikosamatara and Brockelman 1987). Ahsan (1995) speculated the adult male and the 2 adult females of the hoolock gibbon group were siblings; however, the older female expelled the younger one after 14 mo. However, no researchers have yet investigated the social relationships in a multifemale group in detail.

We previously described a black crested gibbon population in Dazhaizi, Wuliang Mountain, central Yunnan, in which ≥3 of 5 groups included 2 breeding females (Fan et al.2006). Here we investigated the long-term stability of these groups, and collected data on spatial proximity and agonistic behavior from 1 multifemale black crested gibbon group on Wuliang Mountain to explore the social relationships between individuals. We adapted Lappan’s (2007) predictions for the social behavior of multimale siamang groups to the study of a multifemale group of black crested gibbons, as follows: 1) if the costs of a multifemale group exceed the benefits for 1 or both females, then emigration or eviction should occur to restore unifemale grouping. High rates of aggression between the 2 females may therefore reflect a conflict of interest between females regarding the maintenance of multifemale grouping. 2) If ≥1 female in a multifemale group benefits from multifemale grouping, then they may seek to avoid aggression by avoiding close proximity with the second female, while maintaining cohesion with the social group. 3) If neither high rates of aggression nor avoidance occur among females, then it is reasonable to conclude that the benefits exceed the costs for each female given their ecological and social circumstances.

Materials and Methods

Study Site

We conducted the study between August 2005 and April 2006 at Dazhaizi (24°21′N, 100°42′E) on the western slope of Mt. Wuliang, a national nature reserve in Jingdong County, central Yunnan, China. The predominant habitat types are primary semihumid evergreen broadleaved forests, mid-mountain humid evergreen broadleaved forests from 1900 to ca. 2700 m elevation, and rhododendron dwarf forests >2700 m elevation. There are also several large secondary forest patches that regenerated ca. 50 yr ago after an extensive forest fire (Fan and Jiang 2008a). The nearest village is 1.5 km from the edge of the focal group’s home range; however, the steep slope and cliffs limit use of the area by people or domestic animals, which seldom appear >2100 m.

The mean temperature during the research period varied seasonally; the lowest was in December (11.8°), and the highest in June (19.6°). The minimum recorded temperature was −2° in March 2005, while the maximum was 31° in October 2005. We recorded precipitation from April 2005 to March 2006 via a SJ1 siphon rainfall recorder situated in the nearest village, 1.5 km from the study area. The total precipitation was 1607 mm, with the wet season from May to October accounting for 96% of the rainfall (Fan and Jiang 2008a).

Focal Subjects

We studied the behavior of black crested gibbons from March 2003 to April 2006. We used the first 2 yr to habituate gibbon groups. Several factors made habituation of these gibbons extremely difficult. First, the gibbons were wary of human interaction owing to serious hunting pressure pre-1997. Second, the groups occupied a large home range of >100 ha, and elevation ranged between 1900 m and 2700 m. Third, there are numerous cliffs throughout the study area, making the full-day follows very difficult. We successfully habituated 1 group (G3) in March 2005 and conducted ecological and behavioral observations for 14 mo until April 2006. We developed and trialed methods for data collection and distance estimation between March and July 2005. Our main research period was from August 2005 to April 2006.

We classified individuals as infants, juveniles, subadults, or adults. Infants were small, with yellow or black coloration, and at least partially dependent on their mothers for transportation. Juveniles were independent, with a smaller body length than adults and black coloring. Subadult males had a body size similar to that of adult males and were able to sing solo bouts, but were not paired with females. Subadult females had a coloration similar to that of adult females, sometimes changing from black to yellow, and were not paired with males. Adult males were completely black with a sharp crest, paired with female(s) with which they defended a territory, duetted, and reproduced. We used body size and crest shape to distinguish black individuals. Adult females were yellow with some darkening on the abdomen and head, and were paired with males. We used body and abdomen color and the size and color of dependent infants to distinguish adult females.

Data Collection and Analysis

We practiced distance estimation between March and July 2005 by estimating the distance between any 2 trees by eye and confirming it using a direct-reading optical range finder (OLC 600XV). We achieved distance estimation of <10 m with error <1 m and 10–30 m estimation with a ca. 3 m error. We then collected systematic data for between August 2005 and April 2006 on all independent individuals. The 2 infants remained dependent or semidependent during the study. We recorded interindividual distance using instantaneous scan sampling at 5-min intervals and recorded all incidents of agonistic behavior ad libitum.

When group members were separated by >30 m, we split into 2 teams to follow different members. We defined subgroups as individuals that maintained a distance of <30 m while maintaining a distance of >30 m with members of other subgroups. In some cases, we were able to follow both subgroups when the focal group split, but in other cases, we could follow only 1 subgroup. In the latter situations, we presumed that members had separated from the group if they were not observed within 1 h (12 successive scans). We recorded the composition and the duration of the subgroups that we followed. If only 1 individual left the group, we recorded the time it left and the time it came back to calculate the time spent separated from the group.

We used radio communication to report the position of adult members on a topography map and estimated the distance between group members based on this map. We never estimated interindividual distances without direct visual contact with both members of a dyad. Map-based data accounted for only 5.8% of the total 15,363 records.

We analyzed spatial relationships between members in 2 ways: 1) mean interindividual distance for each dyad and 2) the frequency of the following 6 categories: <1 m (body contact or within arm’s reach); 1–5 m; 6–10 m; 11–15; 16–20; >20 m. We investigated interindividual distance using daily means of hourly averages. We also examined hourly proportions of time in each category (Lappan 2007). Based on Lappan (2007), we included hours if ≥25% of data were available, and days if ≥5 h of data were available. Sample sizes differed between dyads because the number of individuals missing in instantaneous samples differed (Table I).
Table I

Mean interindividual distance for each dyad during the study period, with the sample size available

Dyad

 

Interindividual distance

Hours/daya

Daysb

FB-FY

Mean

14.9

6.5

39

SD

6.8

1.4

 

FB-male

Mean

15.7

6.7

37

SD

10.3

1.3

 

FY-male

Mean

13.0

6.4

36

SD

7.2

1.5

 

FB-juvenile

Mean

10.0

6.9

24

SD

3.0

1.5

 

FY-juvenile

Mean

13.0

6.0

19

SD

6.2

1.6

 

Male-juvenile

Mean

11.2

7.3

28

SD

5.0

1.4

 

aHours for which ≥25% of instantaneous activity data are available.

bDays for which ≥5 h of activity data are available.

We used an ANOVA with Bonferroni-corrected multiple comparisons to compare mean interindividual distance and percentage of time in different categories among 3 dyads. We used a Mann-Whitney U test to compare mean interindividual distance and percentage of time in different categories between adult-adult dyads and juvenile-adult dyads. The sample size for the 3 adult-adult dyads was 39 + 37 + 36 = 112 (n1), and the sample size for the 3 juvenile-adult dyads was 24 + 19 + 28 = 71 (n2) (Table I).We arcsine transformed all proportional data before conducting parametric statistical tests. All tests were 2-tailed. We used SPSS 11.0® for all analyses.

Results

Group Composition

All 5 groups occupying the study area consisted of 1 adult male, 2 adult females, and 2–5 offspring between August 2003 and August 2005. At least 3 groups had 2 breeding females (Fan et al.2006). One group (G1) disappeared from the site after selective logging (Fan et al.2009b). Our focal group (G3) and 2 neighboring groups (G2 and G4) included 2 adult females from March 2003 until the revision of this manuscript in June 2009. Female FB in G3 and the 2 females in G2 bred repeatedly during this period.

During the main study period from August 2005 to April 2006, G3 comprised 1 adult male, 2 adult females (FY with yellow abdomen and FB with black abdomen), 1 subadult male, 1 juvenile male, 1 infant born to FB in February 2004, and 1 infant born to FY in July 2005. During the study FY’s abdomen blackened. The difference in color between the 2 females is likely to reflect their age difference. Individual social histories for members of G3, and also intra- and intergroup relatedness patterns were unknown before March 2003. During March 2007 the subadult male replaced the resident male of G2, and FB gave birth to a single offspring. The interbirth period for FB was 36 mo. G3 was multifemale and all the members were still alive in June 2009.

Composition and Duration of Subgroups

Our data set included 60 d for which ≥5 h of data were available (mean: 8.4 h, SD = 1.4 h, range: 5–10.5; Table II). In 50 of these days, the focal group separated into subgroups. Subgrouping occurred in every month for which we had data, but was less frequent in August and September 2005 (Table II). Members formed 27 different subgroup structures (Table III). All group members except the juvenile male left the group alone. The subadult male was the most solitary individual: he left the group alone for 1.0–9.6 h in 32 different incidents (Table III). The mean duration of subgrouping did not exceed 3.5 h (Table III).
Table II

Number of observation days with ≥5 h data and number of days in which the focal group separated into subgroups

Month

Observation days with ≥5 h data

Group separated days

Separated days/observation days (%)

August 2005

6

2

33.3

September 2005

8

4

50

October 2005

7

7

100

November 2005

7

6

85.7

December 2005

8

7

87.5

January 2006

7

7

100

February 2006

5

5

100

March 2006

6

6

100

April 2006

6

6

100

Total

60

50

83.3

Table III

Composition and duration of the subgroups in a multifemale black crested gibbon group in 60 observation days with ≥5 h data

Subgroups

Minimum

Maximum

Mean

SD

n

Adult male

1.1

1.5

1.2

0.2

3

Adult male, FB

1.3

1.3

1.3

 

1

Adult male, FB, juvenile

1.0

1.7

1.4

0.4

4

Adult male, FB, juvenile, sub-adult

1.0

3.8

2.1

0.9

6

Adult male, FY

1.0

1.8

1.3

0.4

3

Adult male, FY, FB

1.5

1.5

1.5

 

1

Adult male, FY, juvenile

1.3

4.4

2.7

1.6

3

Adult male, FY, juvenile, sub-adult

1.7

2.0

1.9

0.2

2

Adult male, FY, sub-adult

1.4

4.8

2.7

1.8

3

Adult male, juvenile

1.8

3.5

2.7

1.2

2

Adult male, juvenile, FB

1.3

2.8

2.2

0.8

3

Adult male, juvenile, FY,FB

1.0

9.6

3.1

2.1

31

Adult male, juvenile, subadult

1.0

1.7

1.4

0.3

4

Adult male, juvenile, subadult, FY

1.0

2.1

1.4

0.5

4

Adult male, subadult

1.0

1.5

1.3

0.4

2

Adult male, subadult, FY

1.3

1.3

1.3

0.0

2

FB

1.0

2.1

1.5

0.4

8

FB, juvenile

1.3

2.0

1.8

0.4

3

FB, juvenile, subadult

1.0

2.3

1.6

0.6

4

FY

1.0

3.8

2.1

0.8

8

FY, FB

1.0

5.3

2.3

1.8

6

FY, FB, juvenile

1.0

1.1

1.1

0.1

2

FY, FB, juvenile, subadult

1.5

1.5

1.5

 

1

FY, FB, subadult

1.5

1.5

1.5

 

1

FY, subadult

3.5

3.5

3.5

 

1

Juvenile, subadult

1.5

1.5

1.5

 

1

Subadult

1.0

9.6

3.0

2.1

32

Mean Interindividual Distance

Interindividual distance data was limited for the subadult male because he was peripheral to the study group (2 d with FB; 0 d with FY; 5 d with the adult male). We therefore limited analysis of interindividual distance to 6 dyads: FB-FY, FB-male, FY-male, FB-juvenile, FY-juvenile, and male-juvenile. The distance between the 3 adult members of the focal group as 13.0–15.7 m (Table I). There are no significant differences among the mean interindividual distances of adult dyads (ANOVA: F2, 109 = 1.074, p = 0.345). The juvenile male was 10.0–13.0 m from the adult members (Table I). This was significantly less than the mean interindividual distance among adult dyads (Mann-Whitney U: Z = −3.672, p < 0.001, n1 = 112, n2 = 71).

Percentage of Time in Different Categories

The 3 adult members spent on average 9.3–11.8% of their time <1 m from one another, and 12.6–18.3% of their time >20 m apart each other (Fig. 1). There are no significant differences among adult dyads in the mean percentages of time spent in any distance categories (ANOVA: <1 m: F2, 109 = 1.247, p = 0.291; 1–5 m: F2, 109 = 1.104, p = 0.335; 6–10 m: F2, 109 = 1.195, p = 0.307; 11–15 m: F2, 109 = 0.154, p = 0.858; 16–20 m: F2, 109 = 0.511, p = 0.601; >20 m: F2, 109 = 1.573, p = 0.212).
https://static-content.springer.com/image/art%3A10.1007%2Fs10764-009-9375-9/MediaObjects/10764_2009_9375_Fig1_HTML.gif
Fig. 1

Mean percentages of time at different distance categories for each dyad during the study period.

The juvenile male spent 10.4–13.3% of his time within <1 m of the adult members, and 9.1–15.8% of his time >20 m from the adult members (Fig. 1). The juvenile also spent significantly more time at 1–5 m of FB than at the same distance from FY and adult male (ANOVA with Bonferroni-corrected multiple comparisons: F2, 109 = 6.987, p = 0.002; FB-juvenile and FY-juvenile: p = 0.002; FB-juvenile and male-juvenile: p = 0.023). The juvenile male spent significantly more time at 1–5 m from adults (Z = −2.794, p = 0.005, n1 = 112, n2 = 71), and significantly less time at 11–15 m and  > 20 m from adult members (11–15: Z = −2.784; p = 0.005, n1 = 112, n2 = 71;  > 20: Z = −2.151, p = 0.031, n1 = 112, n2 = 71).

Agonistic Behavior

We observed a total of 38 agonistic incidents in the focal group during 845 observation hours made on 125 d from March 2005 to April 2006 (Table IV). Of these incidents, 37 occurred in feeding trees, and involved food displacement. The winners usually took the feeding places of the losers and the losers changed to other feeding places or moved to a nearby tree. The adult male supplanted the subadult male in 9 cases and juvenile male in 8 cases. Only 1 case was directed at the adult members, when FY was displaced by the adult male. We observed no food displacement cases between the 2 adult females. Open-mouth threats were directed by the adult male toward the juvenile male in 2 cases. The juvenile uttered scream calls twice when FY replaced him, and once when the subadult male replaced him.
Table IV

Incidents of agonistic behavior in a multifemale black crested gibbon group between March 2005 and April 2006

Initiator

Recipient

n

Male

FY

1

Male

Subadult male

9

Male

Juvenile

8

FB

Juvenile

9

FY

Subadult male

1

FY

Juvenile

6

Subadult male

Juvenile

4

Discussion

Many short-term studies have reported groups of black crested gibbons containing ≥2 adult females (Bleisch and Chen 1991; Delacour 1933; Fan et al.2006; Haimoff et al.1986, 1987; Jiang et al.1999). However, ours is the first study to confirm that multifemale grouping can be long term (6 yr), and that multiple females can breed in a group. How these groups formed is unknown. Because it is difficult to observe the formation of multifemale group in the field, studying the kinship of 2 adult females in the same group would be useful in understanding the social relationships between them.

Social Relationships Between Members

In gibbons, subadults are peripheralized and finally excluded from the family groups at about the time of maturity (Srikosamatara and Brockelman 1987). Our observations confirm that subadult individuals become peripheral before dispersal. The juvenile male maintained a closer spatial relationship with adult members than the subadult, especially with female FB. FB is likely to be the mother of this juvenile. Our previous study showed that the juvenile slept with FB on 1 night (Fan and Jiang 2008a) and that she always shared meat with him when she killed flying squirrels (Fan and Jiang 2009). The juvenile male usually slept with the adult male at night (Fan and Jiang 2008a), but this study shows that they did not maintain a close spatial relationship throughout the day.

The most notable characteristic of our focal group was that it comprised 2 breeding females. The spatial relationship among the 3 adult members was nearly an equilateral triangle. The 2 females maintained similar distance (mean ≤20 m) from the adult male. The 2 adult females commonly sang great calls synchronously (Fan et al.2009c). They also exchanged grooming with the adult male and groomed each other (Fan et al.2006). They even shared meat when 1 female killed flying squirrels (Fan and Jiang 2009). We recorded few cases of agonism among adults during the study. One exception occurred when FY was replaced by the adult male while in a feeding tree. We recorded no food cases of food replacement between the 2 adult females. The observations suggest that the relationship between the 2 females was relatively harmonious.

Why Might This Group Be a Stable Multifemale Group?

Our study suggests that the benefits of living in a multifemale group outweigh the costs for both females, given their ecological and social circumstances. The benefits of multifemale grouping may include both mate and territorial defense. We previously reported that we observed a floating female in the home range of G3 from April 15 to 24, 2005, and approached the G3 adults while trying to duet with the male on April 19, 21, 22, and 23. The 2 resident females repeatedly chased the intruder and interrupted her great calls with the male (Fan et al.2006). The floating female left G3’s territory 10 d later.

Although adult females did not usually participate in territorial disputes, the subadult and juvenile males often chased members of other groups during conflicts (unpub. data). Multifemale groups have more subadult and juvenile individuals that can provide territorial defense. Three members (adult male, subadult male, and juvenile male) from G3 and 4 members (adult male, subadult male, and 2 juveniles) from G2 participated in chasing during several intergroup encounters at Dazhaizi (unpub. data). In a pair-living group, it is unlikely that >2 individuals other than females will be involved in territorial defense. However, the more individuals participating in territorial dispute, the more likely the group would effectively defend their territory. For example, in siamangs, larger groups are more likely to win during intergroup interactions (Kinnaird et al.2002).

Even if females do not benefit from mate guarding and territory defense, ecological circumstances may facilitate multifemale grouping in some situations. Animals will defend a territory of the minimum size necessary to satisfy their resource requirements (Macdonald and Carr 1989). However, the minimum territory needed by primary occupants may sometimes be able to support extragroup members. Such secondary members are more easily accommodated into a territory where 1) the pattern of resource availability is heterogeneous and 2) the difference in food security demand from the primary occupants is greater than for the secondary occupants, e.g. for larger generalist species (Macdonald and Carr 1989). The focal group habitat neared the northern extreme of the known distribution of hylobatids, at high altitude with extreme seasonal temperatures and rainfall. The main food species for the focal group, defined as the minimum number of foods comprising 75% of the monthly diet, varied from month to month (Fan et al.2009a). Main food species sites were distributed throughout the home range and varied seasonally. The heterogeneity of food availability in our study site is thought to be higher than that in the rain forest, which is typical habitat for other gibbon species. Our focal group must therefore occupy a large home range to cover the distribution sites of main food species (Fan and Jiang 2008b). The home range size was 151 ha between March 2005 and April 2006 (Fan and Jiang 2008b), and much larger than previously reported for other gibbons (Bartlett 2007). This bigger home range may include enough food sources to support 2 breeding females and their offspring. The black crested gibbon is a generalist species in comparison with other gibbons, spending nearly equal time feeding on leaves and fruit (Fan et al.2009a), which is also likely to facilitate the multifemale grouping in the species.

There is also some preliminary evidence to suggest that neither diet nor daily path length are compromised by living in multifemale groups in this population. First, the group members spent >70% of their feeding time eating figs and fruits (Fan et al.2009a), which was not lower than other gibbons’ diets in the months when fruit was available (Bartlett 2007; Chivers 1984), implying that members of the focal group could find enough fruit when fruit was available. Second, the mean daily path length was 1391 m for the focal group (Fan and Jiang 2008b), which is within the range of other gibbon species (Bartlett 2007; Chivers 1984), suggesting that group members did not forage in larger areas per day than other typical pair-living gibbon groups. Third, the frequency of agonistic behavior in feeding trees (38/125 = 0.30 cases/d) was not higher than in a pair-living siamang group (1.5 cases/d; Chivers 1974), suggesting that food competition in our focal group was not higher than in the pair-living siamang group. Frequent separation into subgroups also decreased the food competition among the members in the group.

In conclusion, we report the stability of multifemale grouping in black crested gibbons over an extended period, and present original information concerning the social relationships between the members of a multifemale group. The spatial relationship among the 3 adult members (1 male, 2 females) formed an equilateral triangle and we observed little agonistic behavior among the adult members. It is still unclear why some groups are multifemale, whereas others contain only 1 female in Wuliang Mountain. Researchers should investigate food availability in gibbon groups with different social organizations to test “food security” hypotheses. A long-term comparative study of females living in multifemale groups and pair-living groups would aid understanding of the social system evolution mechanism of gibbons.

Acknowledgments

We conducted this research at the Gibbon Monitoring Station at Mt. Wuliang, central Yunnan, China, with support provided by The National Basic Research Program of China (no. 2007CB411600), Doctoral Funding from Dali University (no. KY430840), and the National Natural Science Foundation of China (no. 30670270). We thank the 3 anonymous reviewers and the editor for their valuable comments on the manuscript. We thank the staff from the Jingdong Nature Reserve Management Bureau for their needed support. We also thank our field assistants, Mr. Liu Yekun and Mr. Liu Yeyong, for their help.

Copyright information

© Springer Science+Business Media, LLC 2009