Download Primate assemblage structure in amazonian flooded and

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Fauna of Africa wikipedia , lookup

Tropical rainforest wikipedia , lookup

Kasanka National Park wikipedia , lookup

Forest wikipedia , lookup

Old-growth forest wikipedia , lookup

Tropical Africa wikipedia , lookup

Transcript
American Journal of Primatology 67:243–258 (2005)
RESEARCH ARTICLE
Primate Assemblage Structure in Amazonian Flooded
and Unflooded Forests
TORBJØRN HAUGAASENn and CARLOS A. PERES
Centre for Ecology, Evolution and Conservation, School of Environmental Sciences,
University of East Anglia, Norwich, United Kingdom
There is considerable variation in primate species richness across
neotropical forest sites, and the richest assemblages are found in western
Amazonia. Forest type is an important determinant of the patterns of
platyrrhine primate diversity, abundance, and biomass. Here we present
data on the assemblage structure of primates in adjacent unflooded (terra
firme) and seasonally inundated (várzea and igapó) forests in the lower
Purús region of central-western Brazilian Amazonia. A line-transect
census of 2,026 km in terra firme, 2,309 km in várzea, and 277 km in
igapó was conducted. Twelve primate species were recorded from 2,059
primate group sightings. Although terra firme was found to be
consistently more species-rich than várzea, the aggregate primate density
in terra firme forest was considerably lower than that in the species-poor
várzea. Consequently, the total biomass estimate was much higher in
várzea compared to either terra firme or igapó forest. Brown capuchin
monkeys (Cebus apella) were the most abundant species in terra firme,
but were outnumbered by squirrel monkeys (Saimiri cf. ustus) in the
várzea. The results suggest that floodplain forest is a crucial complement
to terra firme in terms of primate conservation in Amazonian forests.
Am. J. Primatol. 67:243–258, 2005. r 2005 Wiley-Liss, Inc.
Key words: Amazonia; Brazil; New World primates; platyrrhines;
primate community
INTRODUCTION
Primates are often the most conspicuous arboreal consumers in paleotropical
[Emmons et al., 1983; Leighton & Leighton, 1983] and neotropical forests [Peres,
1999a]. In many areas they comprise a large proportion of the resident vertebrate
biomass [Oates et al., 1990; Terborgh, 1983a]. Primate community richness is also
Contract grant sponsor: Wildlife Conservation Society; Contract grant sponsor: WWF-US; Contract
grant sponsor: Amazon Conservation Team.
n
Correspondence to: Torbjrn Haugaasen, Centre for Ecology, Evolution and Conservation, School
of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, United Kingdom.
E-mail: [email protected]
Received 7 February 2005; revised 23 March 2005; revision accepted 24 March 2005
DOI 10.1002/ajp.20180
Published online in Wiley InterScience (www.interscience.wiley.com).
r
2005 Wiley-Liss, Inc.
244 / Haugaasen and Peres
highly variable. The most species-rich forest sites in South America are located in
the western Amazon, with as many as 14 sympatric species occurring along the
Rio Juruá [Peres, 1988], and 13 sympatric species inhabiting Ponta da Castanha
[Johns, 1985] and the Urucu [Peres, 1993a].
Amazonia is characterized by a wide range of vegetation types that affect the
ecological distribution of the regional biota. While some species exploit a variety of
habitats, others are primarily or entirely restricted to particular vegetation types.
However, the main determinants of the considerable variation in primate species
richness between sites are currently under debate [Eeley & Lawes, 1999; Peres &
Janson, 1999]. Peres [1997a] found that on a local scale, habitat heterogeneity
within and between different sites (a-diversity and b-diversity, respectively) is the
best determinant of primate species richness, and this has also been shown for
several other taxa on a regional scale [Gascon et al., 2000]. Therefore, the size and
variety of habitats can affect both the number of species and their abundance in a
particular area.
The two most widespread forest environments in Amazonia are unflooded
forests (hereafter termed ‘‘terra firme’’) and seasonally flooded forests, which
represent the main difference in terrestrial macrohabitats across the region.
Terra firme forests account for 83% of the central Amazon basin and are situated
on land that never floods [Hess et al., 2003]. Consequently, these unflooded
forests are situated on well drained terrain that tends to be heavily leached and
nutrient-poor because it has long been deprived of alluvial sediments.
The remaining 17% of the central Amazon basin consists of wetlands, 70% of
which is seasonally flooded forest [Hess et al., 2003]. Different types of flooded
forest can also be distinguished on the basis of hydrochemical [Sioli, 1968] and
floristic [Prance, 1979] differences. Flooded forests along white-water rivers are
known as várzea forests. These rivers originate in the Andes or pre-Andean
regions, which are geologically young and easily erodable landscapes [Räsänen
et al., 1987]. They therefore carry large amounts of nutrient-rich suspended
sediments, and the seasonal influx of nutrients makes the várzea forests
eutrophic.
Igapó forests are inundated on a seasonal basis by black or clear water that
originates in the Amazonian tertiary lowlands or Guiana Shield, and often drain
sandy soils of low-nutrient content. Igapó forests are therefore oligotrophic, carry
less suspended inorganic elements than várzea forests, and contain elevated
concentrations of dissolved organic material, such as humic and fulvic acids.
Few studies have presented data on the primate richness and abundance of
adjacent flooded and unflooded Amazonian forests, and no study to date has
investigated the structure and dynamics of primate assemblages in adjacent terra
firme, várzea, and igapó forests. In this study we investigate primate assemblage
structure and composition in such a forest macro-mosaic in the lower Rio Purús
region of central-western Brazilian Amazonia, and try to pinpoint possible
habitat-related determinants of assemblage biomass and diversity. Primate group
sizes, population densities, and biomass estimates are reported for individual
species in the three forest types, and discussed in relation to other studies. We
also include brief notes on each species and their habitat associations and use.
MATERIALS AND METHODS
Study Area
The study took place at Lago Uauaçú in the lower Rio Purús region of
central-western Brazilian Amazonia (041 200 S, 621 280 W; Fig. 1). This area
Primate Assemblages in Amazonian Forests / 245
Fig. 1. Map of the study area in the lower Rio Purús region of central-western Brazilian Amazonia.
The main forest types that occur in the region are terra firme, várzea, and igapó forest. Trails A–M
(lines) represent the main focal areas censused in these environments.
remains largely undisturbed and incorporates a large landscape mosaic of terra
firme and floodplain forests that are inundated by either white water (várzea) or
black water (igapó [sensu Prance, 1979]) on a seasonal basis for as long as 6
months of the year. Lago Uauaçú itself is a 32-km-long, crescent-shaped, blackwater lake. A human population of 30 caboclo families live near the lake and
depend on the harvesting of Brazil nuts from natural castanhais (stands of Brazil
nut trees (Bertholletia excelsa–Lecythidaceae)), in addition to some small-scale
commercial fishing, logging, and hunting, for their annual income. However, the
forest in the study region remains almost entirely intact, and subsistence hunting
was recently restricted to very light offtakes of large-bodied game species such as
large rodents and ungulates. Subsistence hunters at Lago Uauaçú do not take
primates, and hunting had not taken place in the census areas. We therefore
assume that the primate assemblages at different sites are largely representative
of their respective habitats, and reflect intrinsic differences in forest structure
and composition, flooding regimes, and soil fertility rather than varying degrees
of anthropogenic disturbance. A more detailed description of the study site is
provided elsewhere [Haugaasen, 2004].
Census Techniques and Data Analyses
Data on different primate assemblages at Lago Uauaçú were obtained during
surveys conducted in terra firme, várzea, and igapó forests in 2002 and 2003. All
sites were surveyed using a standardized line-transect census protocol, as
described in Peres [1999b]. In brief, census walks were conducted from 0700–
246 / Haugaasen and Peres
1100 hr and 1300–1700 hr. Census walks were terminated during rain. When a
group was detected, the time, species identity, group size, perpendicular distance
to the approximate geometric center of the group (or subgroup/foraging parties),
sighting location along the transect, and detection cue were recorded. Because
nocturnal surveys were not carried out, the data presented here focus on diurnal
primates, and thus account for all but one species (the southern red-necked night
monkey (Aotus cf. nigriceps)) that occur at Lago Uauaçú.
This study is based on data from a two-way census of 2,026 km in terra firme,
2,309 km in várzea, and 277 km in igapó. Given the smaller sampling conducted
at igapó, the data presented for this forest type should be considered preliminary.
The censuses were carried out along 13 transects that were marked with flagging
tape every 50 m and ranged from 4,000 to 5,000 m in length. This amounts to a
total census effort of 4,612 km walked along 62.5 km of transects. Replicate
census walks within each sampling block were systematically alternated between
terra firme and floodplain forests to minimize any confounding effects of
seasonality. In the high-water season, censuses along the várzea and igapó forest
transects were conducted with the use of unmotorized dugout canoes, and used
the same paths as censuses on foot during the low-water season. Flags along
high-water transects had to be frequently renewed because of changes in the
water level.
The census data were analyzed with the use of DISTANCE 4.0 software
[Thomas et al., 2002]. All visual and acoustic detection events were included in
the analyses. In some cases, however, the perpendicular distance data were
truncated to remove any outliers from the analysis, which provided a better fit to
the various model estimators used to derive detectability functions. Group density
estimates were derived from half-normal models with either a cosine or a
polynomial adjustment. Population density estimates were then calculated using
mean group sizes obtained from reliable group counts during visual detection
events. Crude biomass estimates were calculated using body mass data compiled
by Peres and Dolman [2000] and Peres [2001] (Table 1).
Detection data for the two sympatric tamarin speciesFthe black-capped
moustached tamarin (Saguinus mystax ssp.) and the Avila Pires saddleback
tamarin (Saguinus fuscicollis avilapiresi)Fwere pooled because the exact
composition of mixed-species groups of tamarins was lacking for a number of
observations. In such cases, the individual density and biomass estimates were
extrapolated from the total sample of observations in which reliable group counts
were available for each species. We also note that although moustached tamarins
at Lago Uauaçú occur within the geographic range of the red-capped moustached
tamarin (S. m. pileatus), they are more similar to the epiphenotype of blackcapped tamarin (S. m. mystax), which previously were not known to occur east of
the Juruá River [Peres, 1993b].
Nonmetric multidimensional scaling (MDS) ordination based on the BrayCurtis similarity measure was used to further explore within- and betweenhabitat differences in community structure at the transect level. With the use of
the computer software PRIMER v. 5 [Clarke, 1993], all species were entered in
the ordination as single entities. However, detection data for the two sympatric
tamarin species were again pooled together and the night monkey was excluded
from the analysis. Red titi monkeys were also excluded from the analysis because
they were abundant only in secondary forest throughout the study area. Forest
sites were thus positioned in the ordination space according to their primate
species composition and abundance. The abundance measure used in the
ordination was the number of groups observed per 10 km walked. An index of
0.15
0.39
0.51
0.94
1.05
1.05
1.2
2.2
2.7
2.91
6.5
8.71
9.02
Golden-backed squirrel monkey
Southern red-necked night monkey
Red titi monkey
Collared titi monkey
Buffy saki monkey
White–fronted capuchin
Brown capuchin
Howler monkey
Woolly monkey
Black spider monkey
Mean body mass (kg)a
Pygmy marmoset
Avila Pires saddleback tamarin
Black-capped moustached tamarin
Common name
Fr/in
Fr/in
Fr/sp
Fr/in
Sp/fr
Fr/sp/in/vp
Fr/sp/in/vp
Fo/fr
Fr/fo
fr
Ex/in
Fr/in
Fr/in
Trophic classificationa,b
a
Species are ordered according to body mass within families.
Mean body mass, trophic classifications and stratification use were derived from Peres and Dolman [2000] and Peres [2001].
b
Diet; ex, exudate feeder; fr, frugivore: in, insectivore; sp, seed predator; vp, vertebrate predator; fo, folivore. Listed in order of preference.
c
Forest levels; c, canopy; sc, sub-canopy; u, understorey. Listed in order of preference.
n
Primates
Callitrichidae
Cebuella pygmaea
Saguinus fuscicollis
Saguinus mystax
Cebidea
Saimiri cf. ustus
Aotus cf. nigriceps
Callicebus cupreus
Callicebus torquatus
Pithecia albicans
Cebus albifrons
Cebus apella
Alouatta seniculus
Lagothrix lagotricha
Ateles chamck
Species
TABLE I. Characteristics of the Different Primate Species at Lago Uauaçú, Central Amazonia, Braziln
Sc, C
Sc, C
U, Sc
Sc, U
Sc, C
Sc, U
Sc, C, U
C, Sc
C, Sc
C, Sc
U, Sc
U, Sc
Sc, U
Forest levela,c
Primate Assemblages in Amazonian Forests / 247
248 / Haugaasen and Peres
multivariate dispersion [Warwick & Clarke, 1993] was calculated to investigate
the variability among replicate terra firme, várzea, and igapó forest samples.
RESULTS
Species Richness
A total of 2,059 primate sightings of 12 primate species were recorded at Lago
Uauaçú (Table I). These included the Avila Pires saddleback tamarin (Saguinus
fuscicollis avilapiresi), black-capped moustached tamarin (Saguinus mystax ssp.),
golden-backed squirrel monkey (Saimiri cf. ustus), night monkey (Aotus cf.
nigriceps), red titi monkey (Callicebus cupreus), collared titi monkey (Callicebus
torquatus purinus), buffy saki monkey (Pithecia albicans), white-fronted
capuchin (Cebus albifrons versicolor), brown capuchin (Cebus apella), red howler
monkey (Alouatta seniculus seniculus), gray woolly monkey (Lagothrix lagotricha
cana), and black spider monkey (Ateles chamek). Additionally, the study area is
within the geographic range of the pygmy marmoset (Cebuella pygmaea), and
several local reports confirmed that this species occurred in fringe forest habitat
characterized by high levels of disturbance. However, if indeed it is present, this
species must be very patchily distributed, because we have no records of pygmy
marmosets being found in the main forest matrix during more than 3 years of
field work at Lago Uauaçú.
Spatial Organization of the Primate Assemblage
All 12 primate species encountered during the censuses occurred in terra
firme forest, whereas the maximum number of sympatric species observed in the
várzea and igapó forest sites was 8 and 11 species, respectively. However, a
maximum of only three primate species–C. apella, A. seniculus, and Saimiri cf.
ustus–occurred in várzea sites F, G, I and J, which were isolated from neighboring
terra firme by a ~50-m-wide river (paranã) channel, whereas as many as five
additional species (S. fuscicollis, S. mystax, P. albicans, C. albifrons, and
A. chamek) were found at várzea site K, which was physically connected to the
terra firme forest matrix. An effective fluvial barrier within várzea forest
therefore exacerbated the difference in species richness between the forest types.
This also indicates that the five species that occur in várzea site K but not in
unconnected várzea sites cannot be entirely restricted to várzea, and depend on
terra firme forest habitat for at least part of their year-round metabolic needs.
On the other hand, a total of four primate species that occur in terra firme
were never observed in várzea, whereas only one terra firme forest primateFthe
red titi monkey–was consistently absent from igapó forest. In contrast, all primate
species encountered in várzea forest were also found in adjacent terra firme
forest. The most marked contrast in species composition was therefore found
between the primate assemblages of terra firme and várzea forest, whereas igapó
forest exhibited a very similar set of primate species compared to terra firme. This
is not surprising given the spatial configuration and extent of the igapó forest,
which formed narrow corridors of seasonally flooded forest that interdigitate the
terra firme forest matrix along perennial streams in the shady area.
These observations are supported by the MDS ordination (Fig. 2). On the
basis of the primate species composition and group encounter rates observed at
different sites, terra firme and várzea forests form two clearly distinct clusters on
the left and right sides of the MDS ordination plot, respectively, and diverge
primarily along the first axis (Fig. 2). As expected, the igapó forest sites (L and M)
Primate Assemblages in Amazonian Forests / 249
Fig. 2. MDS ordination of the primate community at 13 terra firme (circles), várzea (triangles), and
igapó (diamonds) sites at Lago Uauaçú, central-western Brazilian Amazonia. Stress = 0.02.
and várzea site K were intermediate between terra firme and várzea (Fig. 2) due
to their interdigitation with terra firme forest terraces. Terra firme site H is
portrayed as an outlier in the ordination, primarily because of the high encounter
rates of brown capuchins and squirrel monkeys at this site compared to the other
terra firme forest sites.
The index of multivariate dispersion shows that igapó forests are more
dispersed (1.259) than terra firme and várzea forests (1.032 and 0.926,
respectively), which show more cohesion despite the outlier found within each
group of sites. The high dispersion value observed between the igapó plots is
probably best explained by the small sampling effort made at these sites, which
affected the number of species observed and their encounter rates between the
two sites.
Patterns of Abundance and Biomass
The aggregate population density of all primate species in terra firme and
igapó forest was considerably lower than that in várzea forest, despite the lower
species-richness of the latter (Table II). The overall density of primate groups in
várzea forest was more than twice that of terra firme, resulting in a similar
difference in densities of individuals. Consequently, the aggregate primate
biomass in várzea (364.1 kg/km2) was considerably higher than that of terra firme
(277.6 kg/km2) and igapó forest (283.4 kg/km2). However, this difference was not
as great as could be expected from the difference in numerical abundance because
the small-bodied squirrel monkeys were the most common primates in both types
of flooded forest (especially in várzea forest). In summary, the várzea forests were
6.4
3.8
9
2
48
21
20
p
1
7
197
pef
197
TF
603
264
26
6
475
–
–
–
11
–
11
VZ
7
p
–
4
1
1
43
24
12
9
–
IG
4.3
0.01*
0.9
0.5
0.8
–
0.05*
0.1*
2.9
–
2.9
TF
9
3.2
0.3
0.1*
9.4
–
–
–
0.2
–
0.2
VZ
6.7
2.3
1.2
0.7*
2.5*
–
–
0.5*
0.1*
–
0.1*
IG
Groups (km2)c
38.3
0.02
3.3
7.8
16.8
–
0.3
0.4
26.8
–
15.2
TF
–
0.5
189.8
57.6
12
0.8
0.6
116.9
–
–
–
0.9
–
0.5
VZ
0.5
p
126.0
45.8
10.5
4.6
6
56.8
–
–
1.8
–
–
–
IG
Individuals (km2)
39
–
1 0.7
–
0.02*
11.7
1
3
p 0.05* 0.04* p
0.3
570 1388 101 10.2
22.2
13.9
120.9
6.8 236
4.6
1
3.9
8.5
–
30
12.6 –
2.7
5.8
3.7
15
18.9
6
?
–
?
IG
12.4 22.7
–
–
–
–
–
3.5
4.3
9.25
22.3
–
6
3.3
–
2.4
VZ
–
5.25
TF
Nb
101.9
2.7
280.8
111.5
0.2
7.3
21
15.8
–
0.3
0.5
13.7
–
5.9
TF
–
4.5
364.1
167.6
78
1.8
1.6
109.9
–
–
–
0.5
–
0.2
VZ
4.7
p
288.1
133.3
68.3
10.1
16.2
53.4
–
–
2.1
–
–
–
IG
Biomass (kg km2)d
b
Number
c
are ordered according to body size within families.
Mean group size obtained from all visual detection events.
of individual detection events.
All densities derived with DISTANCE [Thomas et al., 2002] using the half normal model except n where King’s method was used. In this case, the density
estimate was calculated on the basis of the effective strip width in other forest types.
d
Biomass estimates calculated using data presented in Peres and Dolman [2000] and Peres [2001].
e
Althogh moustached tamarins at Lago Uauaçú occur within the geographic range of the reddish!capped Saguinus mystax pileatus, these are more similar
to the epiphenotype of the black-capped Saguinus mystax mystax.
f
p=present in this forest type but population density estimate is unavailable
w
Species
a
Callitrichidae
Cebuella pygmaea Spix, 1823
Saguinus fuscicollis avilapiresi
Hershkovitz, 1966e
Saguinus mystax pileatus
I. Geoffroy & Deville 1848e
Cebidae
Saimiri cf. ustus I. Geoffroy, 1843
Aotus cf. nigriceps Dollman, 1909
Callicebus cupreus Spix, 1823
Cellicebus torquatus purinus
Hoffmannsegg, 1807
Pithecia albicans Gray, 1860
Cebus albifrons versicolor
Pucheran, 1845
Cebus apella Linnaeus, 1758
Alouatta seniculus seniculus
Linnaeus, 1766
Lagothrix lagotricha cana
Ateles chamek Humboldt, 1812
Total
Species
Group sizea
TABLE II. Summary of Primate Species data in Terra Firme, Várzea, and Igapó at Lago Uauaçúw
250 / Haugaasen and Peres
Primate Assemblages in Amazonian Forests / 251
characterized by low primate species richness and high assemblage biomass
compared to the adjacent terra firme and igapó forests.
Habitat Differences at the Species Level
Despite the similarities between terra firme and igapó in species richness,
abundance, and biomass, there were marked differences in population densities
at the species level. Golden-backed squirrel monkeys were the most abundant
primate species in várzea and igapó, whereas brown capuchins, closely followed
by moustached tamarins, were the most abundant species in terra firme. Howler
monkeys were also far more common in várzea and igapó than in terra firme
forest, whereas both of the tamarin species and woolly monkeys were common in
terra firme. Collared titis, buffy sakis, and white-fronted capuchins were
primarily found in terra firme and igapó, and the two former species were
observed in surprisingly high densities in igapó forest. The only true habitat
generalist with high population densities in all forest types was the brown
capuchin. In fact, this ubiquitous, midsized species had the highest biomass
density of all primate species in all three forest types, closely followed by woolly
monkeys in terra firme, squirrel monkeys in várzea, and howler monkeys in igapó
forest (Table II). In summary, várzea and igapó converged in terms of primate
densities at the species level, while terra firme and igapó converged in richness,
abundance, and biomass. However, more extensive surveys in igapó are needed to
better elucidate the perceived differences between forest types.
DISCUSSION
Species Richness and Spatial Characteristics of the Primate
Assemblage
In terms of species number and composition, the primate assemblage at Lago
Uauaçú is similar to those documented at other central and western Amazonian
forest sites, where as many as 13 [Peres, 1993a, 1997a] or 14 [Johns, 1985]
sympatric primate species can be found. Our findings closely match those
reported for primate assemblages along the Juruá River, where várzea forest is
consistently species-poor compared to the adjacent terra firme forest [Peres,
1997a]. Peres [1997a] also estimated the biomass of the primate assemblage in the
várzea along the Juruá to be more than twice that of terra firme, which is similar
to the results reported here. However, our aggregate primate biomass estimates
for várzea forest are significantly lower than those obtained along the Rio Juruá,
primarily because of the lower howler monkey densities at Lago Uauaçú.
Likewise, the primate biomass noted at a similar terra firme site in centralwestern Brazilian Amazonia (Urucu [Peres, 1993a]) was higher than our
estimates, mainly because of the lower densities of woolly monkeys at Lago
Uauaçú.
The spatial characteristics of the primate assemblages appear to be largely a
function of physical connectivity and proximity to adjacent terra firme forests.
There was a considerable reduction in species richness across the river channel
from the várzea transect K, which abutted terra firme terraces, to unconnected
várzea transects (F, G, I, and J). This suggests that the ~50-m-wide paranã
channel provides an effective fluvial barrier to dispersal and movements by
primates, which is consistent with evidence presented elsewhere [Ayres &
Clutton-Brock, 1992; Heymann et al., 2002]. The close resemblance in species
composition between igapó and terra firme forest is similarly due to the year-
252 / Haugaasen and Peres
round physical connection between these forest types, which ensures that lateral
movements by several species are not hindered.
However, despite the high species richness exhibited by the igapó forest and
várzea transect K, it is clear from data presented elsewhere [Haugaasen, 2004]
that apart from the three primate species (C. apella, A. seniculus, and Saimiri
cf. ustus) that consistently occur in unconnected várzea sites, all of the other
species appear to use várzea and igapó only sporadically. Thus the species
richness of these areas was only temporarily boosted at certain times of the year.
In fact, the sporadic use of floodplain forests by some terra firme species (e.g.,
P. albicans, C. albifrons, A. chamek, and L. lagotricha) appear to be intimately
related to ephemeral pulses in fruit availability in these environments
[Haugaasen, 2004].
Few specific attempts have been made to document the use of várzea or igapó
forest by primates; however, extensive use of seasonally flooded forest appears to
be a consistent trait throughout the Amazon basin [e.g., Boubli, 1999; Branch,
1983; Defler, 1996; Peres, 1993a; Trolle, 2003] (M.G.M van Roosmalen, personal
communication). The general consensus is that the use of these environments is
largely driven by spatial and temporal variability in food availability [CluttonBrock, 1977; Janson & Emmons, 1990; Terborgh, 1983a]. The ability to select and
move across habitats, or include within the home range different habitats that
exhibit distinct temporal peaks in food resources, may have positive effects on
population densities by reducing the impact of local food shortages in any one
habitat [Janson et al., 1981]. For example, spider monkeys (Ateles belzebuth) that
use home ranges consisting of 50% flooded forest can live at higher densities than
conspecifics in other home ranges that contain either greater or smaller
proportions of flooded forest [Ahumada et al., 1998]. It is therefore likely that,
all other things being equal, a mosaic of floodplain and terra firme forests has a
positive impact in buffering local resource scarcity, and can also boost the
abundance of the primary consumer fauna. Evidently, more work is needed to
further map the extent to which flooded forest complements terra firme in terms
of resources, given the community-wide patterns of plant phenology found in each
forest type.
In any case, floodplain forests should not be seen as just marginal habitats
that occasionally subsidize terra firme forest specialists. Interestingly, mounting
evidence suggests that species that are frequently regarded as inhabitants of
flooded forest may perform movements in the opposite direction, i.e., into
unflooded terra firme forest. In particular, this has been observed in large groupliving pitheciine primates such as black uakaries (Cacajao melanocephalus)
[Barnett & Brandon-Jones, 1997; Barnett & de Castilho, 2000; Boubli, 1999;
Defler, 1989, 2001] (C.A. Peres, unpublished data), white uakaries (Cacajao
calvus) [Peres, 1997b], and squirrel monkeys (Saimiri spp.) appear to follow a
similar pattern [Peres, 1993a; Terborgh, 1983a] (T. Haugaasen, personal
observation).
Environmental Factors That Shape the Primate Assemblage
Apart from habitat connectivity, differences in floristic composition, habitat
structure, and soil fertility may provide the most plausible explanations for the
dramatic differences in primate assemblage structure found between terra firme,
várzea, and igapó forests. For example, the lower species richness of várzea and
igapó may be due to their lower floristic diversity compared to terra firme [Ayres,
1993; Balslev et al., 1987; Campbell et al., 1986; Haugaasen & Peres, 2004]. As a
Primate Assemblages in Amazonian Forests / 253
result of flooding, várzea environments also have a considerably simplified forest
structure, because the understory and ground layers are completely submerged
for much of the year. This may prohibit a suite of species from surviving in these
environments on a year-round basis. Indeed, small-bodied understory insectivores, such as the pygmy marmoset (Cebuella pygmaea), tamarin (Saguinus spp.),
and Goeldi’s monkey (Callimico goeldii), do not occur in large expanses of
annually flooded várzea forests along the Rio Juruá, presumably because of the
seasonal submersion or lack of understory foraging substrates [Peres, 1997a].
The apparent success of the three ‘‘true’’ várzea and igapó species may
therefore be best explained by their better adaptation to the seasonal disruption
of certain food supplies by flooding. Brown capuchins and squirrel monkeys use
all forest strata from the ground layer to the canopy, and feed on a mixture of
fruit and arthropods [e.g., Lima & Ferrari, 2003; Zhang, 1995]. Brown capuchins
also consume a variety of flowers and other nonreproductive plant materials, such
as the basal leaf of bromeliad rosettes (C. Peres, personal observation),
understory palm pith [Terborgh, 1983a, b], young bamboo shoots, and new leaf
shoots of the Jauarı́ palm (Astrocaryum jauari) in várzea forest (T. Haugaasen,
personal observation).
The success of howler monkeys in várzea forest is probably due to their
ability to consume and digest foliage. Chemical analyses of the soil macronutrients that are crucial to plants (N, P, K, Na, Ca, and Mg) consistently showed
that the alluvial soils of várzea forest at Lago Uauaçú were far more fertile than
those of terra firme and igapó forests [Haugaasen & Peres, in review], which is in
agreement with other findings from Amazonian forests [Irion, 1978]. The levels of
phytochemical defenses may therefore decrease in várzea forest because of less
limiting nutrient uptake. Foliage quality may thus be more favorable to arboreal
folivores, since the plant metabolic costs of replacing leaf tissue lost to herbivores
are relatively low [Janzen, 1974]. This is consistent with the overwhelmingly
greater biomass density of many folivore species in várzea forest, such as howler
monkeys, two genera of sloths (Bradypus and Choloepus), hoatzins (Opisthocomus hoazin), horned screamers (Anhima cornuta), and iguanas (Iguana iguana)
[Haugaasen & Peres, 2005; Peres, 1997b, 1999a; Queiroz, 1995]. Haugaasen and
Peres [in press] also showed that várzea and igapó forests are more deciduous
than terra firme forest. This results in lower leaf longevity and a higher
abundance of young foliage, which has a more favorable nutrient level and
contains reduced levels of chemical defenses compared to mature foliage [Klinge
et al., 1983]. However, it is less apparent why howlers are so abundant in igapó
forest despite the geochemical similarities between this forest type and terra
firme forests. High levels of tree deciduousness and close proximity to terra firme
forest may be possible explanations.
Furthermore, the high fertility of the young alluvial soils renewed in várzea
forest every year may increase overall fruit production in this forest type. Indeed,
comparable data sets from terra firme, várzea, and igapó sites at Lago Uauaçú
show that várzea on average have a higher number of trees bearing fruit during
any given month of the year [Haugaasen & Peres, in press]. Similarly, the flood
disturbance may generate a more heterogeneous environment with a greater
diversity of successional habitats, which tend to fruit at different times of the
year. This may allow higher densities of primates to persist in várzea forest,
whereas habitat diversity in the terra firme and igapó forest matrix tends to be
lower, and the lack of alternative fruit resources during periods of scarcity may
severely limit the carrying capacity of these forest types. These are probably the
primary explanations as to why terra firme and igapó converge in terms of
254 / Haugaasen and Peres
primate group densities, abundance, and biomass. If nutrient input is an
important determinant of primary productivity, the lack of seasonal influx of
nutrient-rich sediments should limit the availability of macronutrients to plants
on terra firme and igapó soils, and in turn affect the overall primary productivity
available to primary consumers in these environments [Peres, 1997a].
Patterns of Habitat Use
The habitat plasticity shown by brown capuchins may be attributed to their
generalist diet, highly destructive manipulative foraging (especially during the
dry season), and heavy reliance on palm resources [Peres, 1994; Spironello, 1991;
Terborgh, 1983a, b]. This plasticity, combined with their apparently small home
ranges and the nonexistent hunting pressure at Lago Uauaçú, could explain their
relatively high densities in all forest types. The contribution of brown capuchins
to the assemblage-wide biomass estimate lies within the range of other studies.
However, the estimates for Lago Uauaçú are among the highest found at similar
forest types elsewhere in Amazonia [Peres, 1997a].
In contrast to forest sites along the Rio Juruá that were sampled during the
low-water season [Peres, 1997a], white-fronted capuchins were more abundant in
terra firme than várzea forest on a year-round basis. This is in accordance with
Rylands [1987], who stated that this species generally avoids flooded forest.
Observations of white-fronted capuchins in várzea forest directly adjacent to
terra firme coincided with the height of the fruiting season [Haugaasen, 2004],
which suggests that some groups may travel long distances to take advantage of
abundant fruit resources. Defler [1979] made similar observations of C. albifrons
in Colombia, which heavily utilized seasonally flooded forest at the peak of the
Goupia glabra (Celastraceae) fruiting season. The generally low rate of sightings
of this large group-living primate may be explained by their low group density,
relatively large home ranges, and heavy reliance on widely dispersed fruit crops,
such as figs [Terborgh, 1983b].
The poorly studied buffy saki [cf., Peres, 1993c] was observed in all three
forest types, which is consistent with observations of this species entering
inundated forest on a seasonal basis elsewhere in central-western Amazonia
[Rylands, 1987] (C. Peres, unpublished data). However, buffy sakis were entirely
restricted to várzea sites that extensively interfaced with abutting terra firme
forest, and these sightings coincided with the height of the várzea fruiting season.
This suggests that this species could not subsist on a year-round basis in home
ranges consisting of várzea forest only, but could take advantage of abundant
fruit resources in flooded forest.
The two tamarin species occurred at higher densities in terra firme forest
sites at Lago Uauaçú than at the nearest site where tamarins have been studied
(Urucu [Peres, 1993a]). At Lago Uauaçú, moustached tamarins were the second
most abundant species (following brown capuchins) in the terra firme forest
matrix. Although mixed-species groups of tamarins occurred primarily in terra
firme, those occupying home ranges along the várzea–terra firme interface made
frequent incursions into várzea forest. However, they used only fringe várzea
habitats that were not far from terra firme forest, which is consistent with
observations made by Peres [1997a].
Howler monkeys (Alouatta spp.) are arguably the best studied of all
platyrrhine primates. Distance from white-water rivers, soil fertility, forest
structure, and rainfall seasonality have all been shown to be good predictors of
howler monkey densities in neotropical forests [Peres, 1997b]. Their consistent
Primate Assemblages in Amazonian Forests / 255
rarity in terra firme forest at Lago Uauaçú is typical of other central-western
Amazonian forests, and the density we report for várzea forest is considerably
lower that found at most other sites [Peres, 1997b], and is on a par with
moderately hunted várzea sites along the Rio Juruá [Peres, 1997a]. However, the
densities of howler monkeys are highly variable throughout Amazonia [Peres,
1997b], albeit less so within broad classes of forest types.
Woolly monkeys were never observed in várzea forest at Lago Uauaçú,
despite the proximity of this habitat to the home-range boundaries of several
groups. This is consistent with the findings of Peres [1997a], who classified
Lagothrix lagotricha as a hallmark terra firme species, but contrasts with
Rylands’ [1987] report of this species entering inundated forest at times of high
fruit abundance. However, Rylands [1987] did not specify what type of inundated
forest was referred to in that study. Local reports at Lago Uauaçú suggest that
this species uses extensive portions of igapó forest at the height of the fruiting
season, although woolly monkeys were seen using this habitat only once. Further
investigations into the ranging movements of the woolly monkey, and additional
censuses in igapó forest are clearly needed.
The squirrel monkey was by far the most abundant primate species found at
Lago Uauaçú, considering all forest types. Their terra firme forest densities were
remarkably similar to those of congeners in the same habitat along the Rio Juruá
[Peres, 1997a]. On the other hand, their densities in várzea have been observed to
be even higher than those reported here [Peres, 1997a]. It has been suggested that
the high abundance of this species in várzea may result from a lack of competition
from other small-bodied insectivorous primates [Rylands, 1987]. Although both
tamarins (Saguinus spp.) and marmosets (Callithrix spp.) can occur in sympatry
with squirrel monkeys, these callitrichids never occur in large tracts of várzea
forest that are far removed from terra firme and subjected to an annual flood
pulse. Squirrel monkeys in terra firme were never sighted far away from várzea or
igapó forest, which suggests that they require proximity to seasonally inundated
forest. All but two observations coincided with the onset of the fruiting season in
terra firme forest (T. Haugaasen, unpublished data), and thus this species appears
to exhibit reverse lateral movement compared to the movements of other
primates (i.e., occasionally squirrel monkeys tracks terra firme forest resources
within large home ranges that predominantly consist of várzea forest).
The limited, patchy distribution of spider monkeys at Lago Uauaçú may result
from small-scale habitat variation within terra firme forest [Klein & Klein, 1976]
and competition with woolly monkeys [Rylands, 1987]. In our study area, spider
monkeys were most commonly encountered along the broad interface between
várzea and terra firme forest. This underlines the possibility that foraging
incursions into várzea may play a vital role in the ecology of black spider monkeys
at Lago Uauaçú. Furthermore, congeners in central Suriname perform similar
seasonal foraging forays into different forest types at times of fruit scarcity [van
Roosmalen, 1985]. Indeed, both our observations and local reports suggest that
this species enters várzea on a seasonal basis at the height of the fruiting season.
Observations of collared titi monkeys (Callicebus torquatus) were rare, but
these monkeys appeared to express a consistent habitat preference for terra firme
and igapó forest, confirming earlier reports [Defler, 1994]. The myth that this
species is an edaphic specialist restricted to Amazonian forests on white-sand soils
[Kinzey & Gentry, 1979] has been solidly refuted [Defler, 1994; Peres, 1997a], and
is also contradicted by observations of the species in this study.
The rarity of red titi monkeys (Callicebus cupreus) in the overall primary
forest matrix at Lago Uauaçú is explained by the strong preference of this species
256 / Haugaasen and Peres
for secondary growth close to human habitation. Their preference for secondary
or disturbance-dependent habitats is well documented [Peres, 1993a; Rylands,
1987].
Conservation Implications
The consistently lower species richness observed in várzea forest should not
imply that this major Amazonian forest type should be neglected in future
conservation planning [cf., Goulding et al., 2003; Peres & Terborgh, 1995]. We
have demonstrated that várzea forests differ profoundly from terra firme forests
in terms of primate species composition, abundance, and biomass, even in areas
where the two forest types are juxtaposed. The data provided here and elsewhere
also show that several unflooded forest primate species may use flooded forests on
a seasonal basis, which suggests that landscape-scale habitat heterogeneity is
crucial to the long-term population viability of some species. Failure to protect the
greatest expression of seasonally flooded forests on Earth may also alter or
disrupt the seed dispersal services provided by primates. These results suggest
that várzea and igapó forests are important complements to terra firme forest in
terms of regional-scale primate conservation, and that conservation planning
in Amazonian forests must be considered on a landscape scale.
ACKNOWLEDGMENTS
We thank the Amazon Association for the Preservation of Areas of High
Biodiversity (AAP) for logistical help. We especially thank Marilene, Edivar, Zé,
and Evineu for their help with the field work. P. Judge prepared Fig. 1.
REFERENCES
Ahumada JA, Stevenson PR, Quinones MJ.
1998. Ecological response of spider monkeys
to temporal variation in fruit abundance:
the importance of flooded forest as a keystone habitat. Primate Conserv 18:10–14.
Ayres JM, Clutton-Brock TH. 1992. River
boundaries and species range size in Amazonian primates. Am Nat 140:531–537.
Ayres JM. 1993. As matas de várzea do
Mamirauá. Belém: Sociedade Civil Mamiraua. 96 p.
Balslev H, Luteyn J, Øllgard B, Holm-Nielsen
LB. 1987. Composition and structure of
adjacent unflooded and floodplain forest in
Amazonian Ecuador. Opera Botanica 92:
37–57.
Barnett AA, Brandon-Jones D. 1997. The
ecology, biogeography and conservation of
the uakaris, Cacajao (Pitheciinae). Folia
Primatol 68:223–235.
Barnett AA, de Castilho C. 2000. Report on a
short study of the dry season feeding
ecology and habitat preferences of the golden-backed uacari or bicó, Cacajao melanocephalus ouakary (Cebidae: Pitheciinae), on
the lower Rio Jaú, Amazonas, Brazil. Manaus. 20p.
Boubli JP. 1999. Feeding ecology of blackheaded uacaris (Cacajao melanocephalus
melanocephalus) in Pico da Neblina
National Park, Brazil. Int J Primatol 20:
719–749.
Branch LC. 1983. Seasonal and habitat differences in the abundance of primates in the
Amazon (Tapajós) National Park, Brazil.
Primates 24:424–431.
Campbell DG, Daly DC, Prance GT, Maciel
UN. 1986. Quantitative ecological inventory
of terra firme and várzea tropical forest on
the Rio Xingú, Brazilian Amazon. Brittonia
38:369–393.
Clarke KR. 1993. Nonparametric multivariate
analyses of changes in community structure. Austral J Ecol 18:117–143.
Clutton-Brock TH, editor. 1977. Primate
ecology: studies of feeding and ranging
behaviour in lemurs, monkeys and apes.
London: Academic Press. 631p.
Defler TR. 1979. On the ecology and behavior
of Cebus albifrons in eastern Colombia.
Primates 20:475–490.
Defler TR. 1989. The status and some ecology
of primates in the Colombian Amazon.
Primate Conserv 10:51–56.
Primate Assemblages in Amazonian Forests / 257
Defler TR. 1994. Callicebus torquatus is not a
white-sand specialist. Am J Primatol
33:149–154.
Defler TR. 1996. Aspects of the ranging
pattern in a group of wild woolly monkeys
(Lagothrix lagothricha). Am J Primatol
38:289–302.
Defler TR. 2001. Cacajao melanocephalus
ouakary densities on the lower Apaporis
River, Colombian Amazon. Primate Rep
61:31–36.
Eeley HAC, Lawes MJ. 1999. Large-scale
patterns of species richness and species
range size in anthropoid primates. In:
Fleagle JG, Janson CH, Reed KE, editors.
Primate communities. Cambridge: Cambridge University Press. p 191–219.
Emmons LH, Gautierhion A, Dubost G. 1983.
Community structure of the frugivorous
folivorous forest mammals of Gabon. J Zool
199:209–222.
Gascon C, Malcolm JR, Patton JL, da Silva
MNF, Bogart JP, Lougheed SC, Peres CA,
Neckel S, Boag PT. 2000. Riverine barriers
and the geographic distribution of Amazonian species. Proc Natl Acad Sci USA
97:13672–13677.
Goulding M, Barthem R, Ferreira EJG. 2003.
The Smithsonian atlas of the Amazon.
Washington, DC: Smithsonian Institution
Press. 256p.
Haugaasen T. 2004. Structure, composition
and dynamics of a central Amazonian forest
landscape: a conservation perspective.
Ph.D. thesis, University of East Anglia,
Norwich, UK. 227p.
Haugaasen T, Peres CA. 2005. Mammal
assemblage
structure
in Amazonian
flooded and unflooded forests. J Trop Ecol
21:1–13.
Haugaasen T, Peres CA. 2005. Floristic,
edaphic and structural characteristics of
flooded and unflooded forests in the lower
Purús region of central Amazonia, Brazil.
Acta Amazonica (in press).
Haugaasen T, Peres CA. 2005. Patterns of tree
phenology in adjacent Amazonian flooded
and unflooded forests. Biotropica (in press).
Hess LL, Melack JM, Novo E, Barbosa CCF,
Gastil M. 2003. Dual-season mapping of
wetland inundation and vegetation for the
central Amazon basin. Remote Sensing
Environ 87:404–428.
Heymann EW, Encarnacion F, Canaquin JE.
2002. Primates of the Rio Curaray, northern Peruvian Amazon. Int J Primatol
23:191–201.
Irion G. 1978. Soil infertility in the Amazonian
rain forest. Naturwissenschaften 65:515–
519.
Janson CH, Terborgh J, Emmons LH. 1981.
Non-flying mammals as pollinating agents
in the Amazonian forest. Biotropica
13(suppl):1–6.
Janson CH, Emmons CH. 1990. Ecological
structure of the nonflying mammal community at Cocha Cashu Biological Station,
Manu National Park, Peru. In: Gentry
AH, editor. Four neotropical rainforests.
New Haven: Yale University Press. p 314–
338.
Janzen DH. 1974. Tropical blackwater rivers,
animals, and mast fruiting by the Dipterocarpaceae. Biotropica 6:69–103.
Johns AD. 1985. Primate and forest exploitation at Tefé, Brazilian Amazonia. Primate
Conserv 6:27–29.
Kinzey WG, Gentry AH. 1979. Habitat utilization in two species of Callicebus. In: Sussman RW, editor. Primate ecology: problemoriented field studies. New York: John
Wiley & Sons.
Klein LL, Klein DJ. 1976. Neotropical primates, aspects of habitat usage, population
desity, and regional distribution in La
Macarena, Columbia. In: Thorington RW,
Heltne PG, editors. Neotropical primates:
field studies and conservation. Washington,
DC: National Academy of Sciences. p 70–78.
Klinge H, Furch B, Harms E, Revilla JD. 1983.
Foliar nutrient levels of native tree species
from central Amazonia. I. Inundation forest. Amazoniana 8:19–45.
Leighton M, Leighton DR. 1983. Vertebrate
responses to fruiting seasonality within a
Bornean rain forest. In: Sutton SL, Whitmore TC, Chadwick AC, editors. Tropical
rain forest: ecology and management. London: Blackwell Scientific. p 181–196.
Lima EM, Ferrari SF. 2003. Diet of a freeranging group of squirrel monkeys (Saimiri
sciureus) in eastern Brazilian Amazonia.
Folia Primatol 74:150–158.
Oates JF, Whitesides GH, Davies AG, Waterman PG, Green SM, Dasilva GL, Mole S.
1990. Determinants of variation in tropical
forest primate biomass–new evidence from
west Africa. Ecology 71:328–343.
Peres CA. 1988. Primate community structure
in western Brazilian Amazonia. Primate
Conserv 9:83–87.
Peres CA. 1993a. Structure and spatial organization of an Amazonian terra firme forest
primate community. J Trop Ecol 9:259–276.
Peres CA. 1993b. Notes on the primates of the
Juruá River, western Brazilian Amazonia.
Folia Primatol 61:97–103.
Peres CA. 1993c. Notes on the ecology of buffy
saki monkeys (Pithecia albicans, Gray
1860)–a canopy seed predator. Am J Primatol 31:129–140.
Peres CA. 1994. Primate responses to phenological changes in an Amazonian terra firme
forest. Biotropica 26:98–112.
258 / Haugaasen and Peres
Peres CA, Terborgh JW. 1995. Amazonian
nature reserves–an analysis of the defensibility status of existing conservation units
and design criteria for the future. Conserv
Biol 9:34–46.
Peres CA. 1997a. Primate community structure at twenty western Amazonian flooded
and unflooded forests. J Trop Ecol 13:
381–405.
Peres CA. 1997b. Effects of habitat quality and
hunting pressure on arboreal folivore densities in neotropical forests: a case study
of howler monkeys (Alouatta spp.). Folia
Primatol 68:199–222.
Peres CA. 1999a. The structure of nonvolant
mammal communities in different Amazonian forest types. In: Eisenberg JF, Redford
KH, editors. Mammals of the neotropics: the
central neotropics. Chicago: University of
Chicago Press. p 564–578.
Peres CA. 1999b. General guidelines for
standardising line-transect surverys of tropical forest primates. Neotrop Primates 7:
11–16.
Peres CA, Janson CH. 1999. Species coexistence, distribution and environmental
determinants of neotropical primate
richness: a community-level zoogeographic
analysis. In: Fleagle JG, Janson CH,
Reed KE, editors. Primate communities.
Cambridge: Cambridge University Press.
p 55–74.
Peres CA, Dolman PM. 2000. Density compensation in neotropical primate communities: evidence from 56 hunted and
nonhunted Amazonian forests of varying
productivity. Oecologia 122:175–189.
Peres CA. 2001. Synergistic effects of subsistence hunting and habitat fragmentation
on Amazonian forest vertebrates. Conserv
Biol 15:1490–1505.
Prance GT. 1979. Notes on the vegetation of
Amazonia. III. The terminology of Amazonian forest types subject to inundation.
Brittonia 31:26–38.
Queiroz HL. 1995. Preguiças e guaribas:
os mamiferos folivoros arboricolas do
Mamirauá. Belém: Sociedade Civil Mamirauá. 161p.
Räsänen ME, Salo JS, Kalliola RJ. 1987.
Fluvial perturbance in the western Amazon
basin–regulation by long-term sub-Andean
tectonics. Science 238:1398–1401.
Rylands AB. 1987. Primate communities in
Amazonian forests–their habitats and food
resources. Experientia 43:265–279.
Sioli H. 1968. Hydrochemistry and geology in
the Brazilian Amazon region. Amazoniana
1:267–277.
Spironello WR. 1991. Importância dos frutos
de palmeiras (Palmae) na dieta de um grupo
de Cebus apella (Cebidae: Primates) na
Amazônia central. Primatol Brasil 3:285–
296.
Terborgh J. 1983a. The ecology of Amazonian
primates. In: Prance GT, editor. The Amazon rainforest. Oxford: Pergammon. p 284–
304.
Terborgh J. 1983b. Five New World primatesFa study in comparative ecology. Princeton: Princeton University Press. 260p.
Thomas L, Laake JL, Strindberg S, Marques
FFC, Buckland ST, Borchers DL, Anderson
DR, Burnham KP, Hedley SL, Pollard JH.
2002. DISTANCE 4.0 release 1. University
of St. Andrews, Research Unit for Wildlife
Population Assessment. http://www.ruwpa.
st-and.ac.uk/distance/.
Trolle M. 2003. Mammal survey in the Rio
Jauaperi region, Rio Negro Basin, the
Amazon, Brazil. Mammalia 67:75–83.
van Roosmalen MGM. 1985. Habitat preferences, diet, feeding strategy, and social
organization of the black spider monkey
(Ateles p. paniscus Linnaeus 1758) in Surinam. Acta Amazonica 15:1–238.
Warwick RM, Clarke KR. 1993. Increased
variability as a symptom of stress in marine
communities. J Exp Marine Biol Ecol 172:
215–226.
Zhang SY. 1995. Activity and ranging patterns
in relation to fruit utilization by brown
capuchins (Cebus apella) in French Guiana.
Int J Primatol 16:489–507.