Structure of Brazilian Atlantic forests with occurrence of the woolly spider monkey (Brachyteles hypoxanthus)

Text-only Preview

Ecol Res (2010) 25: 25–32
DOI 10.1007/s11284-009-0626-1
Wilson Marcelo da Silva Ju´nior
Fabiano Rodrigues de Melo
Leandro Santana Moreira
Elaine Ferreira Barbosa
Joa˜o Augusto Alves Meira-Neto
Structure of Brazilian Atlantic forests with occurrence
of the woolly spider monkey (Brachyteles hypoxanthus)
Received: 21 October 2008 / Accepted: 2 June 2009 / Published online: 18 July 2009
Ó The Ecological Society of Japan 2009
Abstract This work examines patches of Atlantic forests
spider monkeys than the Sossego forest sample. The
in Minas Gerais State where woolly spider monkeys
lowest density of these primates was in the PESB forest
(Brachyteles hypoxanthus) are inhabitants in order to
sample, the earlier successional stage of secondary suc-
understand the effects of forest structure, diversity, and
cession of this study.
succession in B. hypoxanthus abundance. This study was
carried out in three reserves: Rio Doce State Park
Keywords Primate conservation Æ Diversity Æ Tropical
(PERD), Mata do Sossego (Sossego), and Serra do
forests Æ Plant functional types for succession
Brigadeiro State Park (PESB). Structural studies were
carried out by means of point-centered quarter samples
in areas used by woolly spider monkeys. Evenness was
lower in PERD than in the other two communities. By
means of successional functional groups analysis and
The reduction and fragmentation of the Atlantic Forest
diametric structure, the PERD forest sample was shown
is, in addition to hunting, the main threat for the con-
to be in the later, and PESB forest sample in the earlier,
servation of the woolly spider monkey, Brachyteles
stages of secondary succession. We found that tree
hypoxanthus (Strier 1991). This monkey is the largest
species evenness and structural traits driven by succes-
primate in the Americas and is endemic to this vegeta-
sion were important factors in determining spider
tion. Small forest areas are incompatible with B. hypo-
monkey abundance patterns. The PERD forest sample
xanthus conservation, even though these monkeys are
had lower evenness and had lower density of woolly
exclusively herbivorous and because big primates have
high energetic requirement (Clutton-Brock and Harvey
1977). The four major patches of the Atlantic Forest
Electronic supplementary material The online version of this article
that conserve woolly spider monkeys in Minas Gerais
(doi:10.1007/s11284-009-0626-1) contains supplementary material,
which is available to authorized users.
State (MG), Brazil, have different areas, structures,
richness, and are in different stages of secondary suc-
W. M. da Silva Ju´nior Æ J. A. A. Meira-Neto (&)
cession. Almost all the information about theses pri-
Botanic Post-Graduate Program of Universidade
mates comes from research in Reserva Biolo´gica de
Federal de Vic¸osa, UFV, Vic¸osa 36570-000, Brazil
E-mail: [email protected]
Caratinga, MG (Mendes et al. 2005). Additional infor-
Tel.: +55-31-38991955
mation about other Atlantic Forest patches with woolly
Fax: +55-31-38992580
spider monkeys in MG allows for a correlation of
structural and diversity traits of these forests with B.
F. R. de Melo
Center of Ecological Studies and Environmental Education,
hypoxanthus abundance.
CECO, Universidade Federal de Goia´s,
Abundance is a central issue for ecology and con-
UFG, Campus Jataı´, Jataı´, Brazil
servation (Rosenzweig 1995; Hubbell 2001). An under-
standing of the variables that are related with different
L. S. Moreira
Animal Biology Post Graduate Program of UFV,
densities of B. hypoxanthus in Atlantic Forest patches is
Vic¸osa, Brazil
fundamental for the conservation management of this
species. One of these variables, which is positively cor-
E. F. Barbosa
related with folivorous primate abundance, is tree
Animal Behavior Post-Graduate Program
diversity. Phytodiversity increases the probability of
of Universidade Federal de Juiz de Fora,
UFJF, Juiz de Fora, Brazil
finding alternative food items during times of scarcity

(Ganzhorn 1995; Chiarello 1999; Strier 2000). Another
4 months of drought during winter (IBGE 2001), and
explanatory variable is that big trees are large sources of
have the same general type of forests, Seasonal Tropical
energy that allow formation of tight primate groups,
Forests (Veloso et al. 1991) in the Atlantic Forest do-
decreasing predation (Gaulin et al. 1980; Lemos-de-Sa´
and Strier 1992; Petroni 2000) and improving locomo-
PERD is located in Marlierea, Timoteo, and Dionisio
tion security because of their strong branches (Fimbel
municipalities (19°37¢S and 42°33¢W). It is the biggest
patch of the Atlantic Forest in MG, with a total area of
This work aims to study tree diversity and succes-
360 km2 and is surrounded by Eucalyptus plantations,
sional traits of forest structures of all major patches of
pastures, factories, and cities. The altitudinal range is
Atlantic Forest in MG State where woolly spider mon-
between 230 and 515 m a.s.l. and the climate is humid
keys occur, except Reserva Biolo´gica de Caratinga. The
tropical with dry winters. The annual mean precipitation
goal is to understand the explanatory variables of forest
is 1,480.3 mm, and the annual mean temperature is
structure and diversity for the abundance of B. hypo-
22°C. This forest is a semideciduous seasonal tropical
xanthus for conservational purposes.
forest (Veloso et al. 1991). The sample was located at
400 m a.s.l.
Sossego is located at Simonesia municipality (20°04¢S
Materials and methods
and 42°05¢W) and has 8 km2 of interconnected smaller
forest patches. Its vegetation is semideciduous seasonal
Study sites
montane forest on steep slopes and mountaintops. The
altitudinal range is 1,300–1,600 m a.s.l. The sample was
This study was carried out in three reserves in MG State,
located between 1,400 and 1,500 m a.s.l. The climate is
Brazil (Fig. 1): Rio Doce State Park (PERD), Mata do
humid tropical to humid sub-tropical with dry winters.
Sossego Natural Property Private Reserve—RPPN
The annual mean rainfall is 1,350 mm and the annual
(Sossego), and Serra do Brigadeiro State Park (PESB).
mean temperature is 18°C.
All of these reserves are under the influence of general
PESB is located in Erva´lia, Fervedouro, Sericita,
central Brazilian humid tropical climate type, with 3 or
Araponga, Miradouro, Pedra Bonita, Muriae´, and
Divino municipalities (20°43¢S and 42°01¢W) and has an
area of 132 km2. Most of the area is covered by forests.
Rocky fields cover the bare slopes, crests, and tops of the
mountains (Caiafa and Silva 2005). The altitudinal
range is between 1,000 and 2,000 m a.s.l. The sample
was located between 1,200 and 1,300 m a.s.l. The main
vegetation is semideciduous seasonal montane forest
(Veloso et al. 1991). The climate is humid tropical to
humid subtropical with dry winters. The annual mean
rainfall is 1,300 mm and the annual mean temperature is
For the study of forest structure, 200 point-centered
quarter samples were applied in each of the three sites
(Cottam and Curtis 1956; Newton 2007). Each forest
unit sample was spaced 20 m from another along both
sides of a grid path used for studies of the woolly spider
monkeys. There were 800 sampled trees with a diameter
of at least 5 cm at breast height (dbh) in each forest
For relating structural traits of the vegetation with B.
hypoxanthus abundance was used the number of obser-
vations of woolly spider monkeys of these forests as a
proxy of monkeys density at the same sampled areas for
the same period (Biodiversitas 2006) by means of scan
sampling (Strier 1987). These primatological samplings
were carried out by Biodiversitas (2006) (supplementary
material). The number of observations at each site, only
into the samples, during the years of 2004 and 2005
Fig. 1 Study areas: 1 Rio Doce State Park (PERD), 2 Mata do
Sossego (Sossego) and 3 Serra do Brigadeiro State Park (PESB)
(24 months) was used as an abundance value of woolly

Table 1 Diversity and structure estimates from forests of this study, PERD forest, PESB forest, and Sossego forest
Species richness
0.770* (0.0152)a
0.658* (0.0161)b
0.666* (0.024)b
Log c
0.0635* (0.03)a
0.2429* (0.047)b
0.2931* (0.047)b
Area (ha)

BA (m2 ha-1)
Diam. intercept
5.361 (0.23)a**
6.015 (0.59)a**
5.424 (0.29)a**
Diam. slope
À2.603 (0.131)a**
À3.028 (0.38)b**
À2.598 (0.177)a**
Z (slope) = b-diversity (±SD); Log c (intercept) = a-diversity (±SD); Area = sample area; H¢ = Shannon diversity index;
J = evenness; BA = basal area; Diam. intercept (diameter intercept) and Diam. slope (diameter slope) of diametric frequency distri-
bution (±SD)
Different letters are significant difference * P < 0.001; ** P < 0.05
spider monkey populations. Quite different numbers of
performed using a Chi-square test based on univariate
monkey observations were registered for those three
categorical data (Devore and Farnum 2005). For this
forests during that period. PERD, Sossego, and PESB
analysis, those species with unknown classification in a
had 71, 273, and 22 records, respectively. Before 2004,
successional functional group were not included.
those samples were plotted supposing that the chosen
Another statistical analysis evaluated the succes-
sites were the best sites for monkeys observations by the
sional phase among these forests by using linear
primatological studies team (Biodiversitas 2006).
regression of vegetation structure (log-diameter classes
Curves of diametric distribution of stems abundance
and log-individual number) of each community, per-
were used for phytocoenosis size structure comparison
forming F-test for regressions and calculating standard
(Mueller-Dombois and Ellenberg 1974; Newton 2007),
deviations (Bourg 2006) of slopes and intercepts. The
performed using FITOPAC (Shepherd 2002) and Di-
assumption is that the community with negative stee-
amfito (Mota 1995) software.
per slopes is in an earlier successional phase because
The successional phase was analyzed by the propor-
classes of smaller plants with thinner diameters are
tion of species of seral functional groups of Atlantic
expected to be denser in early successional communi-
Forest species (Martins and Rodrigues 1999; Nakazono
ties than in late ones.
et al. 2001; Nunes et al. 2003; Oliveira-Filho et al. 2004;
Paula et al. 2004; Ribas et al. 2003; Rolim et al. 1999;
Salimon and Negrelle 2001; Silva et al. 2003) using the
groups pioneers, early successionals and late succes-
sionals (Leita˜o Filho et al. 1993). The number of species
in each functional group was the variable used to com-
pare the communities within the secondary succession
For the forest sample of Rio Doce State Park (PERD),
range, using a Chi-square test. A diametric analysis of
185 species of 42 botanical families were reported. The
tree communities by means of linear regression of log-
richest families were Fabaceae (32), Myrtaceae (18),
diameter and log-individuals plots was used as a com-
Lauraceae (13), Annonaceae, Moraceae, and Sapotaceae
plementary method (Bourg 2006). The regression slopes
(eight species each). The floristic checklist is presented as
and intercepts were indicators of successional phase,
supplementary material.
with more initial phases associated with those with
Comparing PERD with the other two communities,
steeper negative slope, reinforced by higher intercepts.
the Shannon diversity index (H¢ = 3.651) and evenness
Comparisons of spatial structures of diversity were
(J = 0.707) showed the lowest values (Table 1). This
made using species-individual relations (Rosenzweig
lowest diversity was reinforced by the lowest intercept of
1995; Condit et al. 1996) by means of F-test and stan-
linear regression of rarefaction (log c = 0.0635 ± 0.03)
dard deviations (Bourg 2006) of z-slopes (b-diversity)
(Table 1; Fig. 2).
and of c-intercepts (a-diversity) from linear regressions
However, PERD had the steepest slope of log spe-
of log-individuals and log-species plots (Hubbell 1997;
cies–log individual linear regression. Hence, PERD had
Magurran 2004; Lomolino et al. 2006). The model was
a higher b-diversity because more species are pooled by
the linearized power-law:
added individuals (Table 1; Fig. 2).
LogS ¼ Logc þ zLogA
The Sossego forest sample contained 134 species of 44
families. Myrtaceae was the richest family, with 19
of Arrhenius (1922).
species, followed by Melastomataceae (14), Lauraceae
Statistical comparisons between forests of number of
(13), Fabaceae (7), Solanaceae (6), Annonaceae, and
species in functional secondary succession groups were
Euphorbiaceae (five each) (supplementary material).

PERD y = 0.7704x + 0.0635
PERD y = -2.6034x + 5.3618
R2 = 0.9973
R2 = 0.973
PESB y = 0.6584x + 0.2429
R2 = 0.9958
PESB y = -3.0281x + 6.0154
R2 = 0.9271
Sossego y = 0.6665x + 0.2931
R2 = 0.991
Sossego y = -2.5982x + 5.4238
R2 = 0.9686
Linear (PERD)
Linear (PERD)
Linear (PESB)
Linear (PESB)
Linear (Sossego)
Linear (Sossego)
Fig. 3 Double-logarithmic linear regressions of decreasing number
Fig. 2 Double-logarithmic rarefactions of species-area power-
of individuals with increasing diameter for the three samples of
function model fitted by linear regressions for the three samples
forests: PERD, PESB, and Sossego. Significance of F-test for
of forests: PERD, PESB, and Sossego. Significance of F -test for
intercepts (Diam. intercept) and slopes (Diam. slope) in Table 1
slopes (z) and intercepts (Log c) in Table 1
supplementary material. A structural synthesis is in
The Sossego forest sample had the highest Shannon
Table 1.
diversity index (H¢ = 4.343) and evenness (J = 0.873).
The most important species of Sossego forest sample
These high a-diversity estimates were reinforced by the
were Euterpe edulis, Nectandra cuspidata, Cyathea del-
highest intercept (b = 0.02931) of species–individual
gadii, Clusia insignis, Marlierea suaveolens, Guatteria
relations from the log–log linear regression. Neverthe-
schomburgkiana, Campomanesia guaviroba, Clethra sca-
less, b-diversity estimated as slope of linear regression
bra, Couepia venosa, and Calyptranthes clusiaefolia
(m = 0.666) was lower than that of the PERD forest
(supplementary material). These species had 35.15% of
sample and was not significantly different from that of
IV of the sample. Euterpe edulis, the ‘‘Jussara palm’’ or
the PESB forest sample (Table 1; Fig. 2).
the ‘‘heart of palm’’ plant, was the most important
PESB forest sample had 128 species of 38 botanical
population in the Sossego forest sample, with 70 sam-
families. The richer families were Melastomataceae, with
pled individuals, 6.82% of IV and with a density of trees
13 species, Lauraceae (12), Myrtaceae and Rubiaceae
of 135 individuals per hectare. It is a shade-tolerant
(eight each), Meliaceae (seven), Asteraceae and Sapot-
species (Nakazono et al. 2001) and an indicator of
aceae (six each) (supplementary material).
habitat quality.
The Shannon index (an alpha-diversity estimator) for
The most important species in the PESB forest sam-
the PESB forest sample was higher than that for the
ple (IV) were Sapium glandulatum, Croton floribundus,
PERD forest sample and lower than the Sossego forest
Solanum leucodendron, Alchornea triplinervea, Solanum
sample. Nevertheless, the intercept of the species–indi-
cinnamomeum, Cyathea delgadii, Psychotria capitata,
vidual linear regression (another a-diversity estimator)
Guapira opposita, Vernonia diffusa, and Myrsine umbel-
of the PESB forest sample, was lower than the Sossego
lata (supplementary material).
intercept, but not significantly different. The slopes of
Analyzing succession by means of diametric struc-
these regressions, used here as b-diversity estimators,
ture, Sossego forest sample showed the least negative
were lower than that for the Sossego forest sample, but
steep slope for regression of log-diameter classes with
not statistically different (Table 1; Fig. 3). The PESB
log-individual numbers, and was significantly different
and Sossego samples were similar in diversity structure.
from PERD and PESB slopes. Structurally, Sossego was
The Sossego forest sample was somewhat more diverse,
in later secondary succession than the other two forests.
but not significantly different.
There was no significant difference among intercept
values (Table 1; Fig. 3).
The numbers of species in the functional groups
Succession and structure
among these three forests were analyzed by a Chi-square
test. Only PERD was different (P = 0.02), with less
At PERD forest sample, the species with highest
pioneer species and more late secondary species than
importance values (IV) were Senefeldera multiflora,
expected. The observed values of Sossego and PESB
Neoraputia alba, Ficus gomelleira, Lecythis pisonis, Vi-
samples were not significantly different from expecta-
rola gardneri, Guatteria acutipetala, Pterygota brasilien-
tions (Table 2). Based on the floristic composition, the
sis, Rollinia sylvatica, Trichilia silvatica, and Cariniana
PERD forest sample was in later secondary succession
legalis. The structural table of this community is in the
than the other two forests.

Table 2 Observed (obs) and expected (exp) values of successional
The basal area and a-diversity values of Sossego
functional group species (only species classified by the literature: P
forest sample (Table 1) were higher than most common
pioneers, ES early successional, LS late successional) in the PERD
values found in Atlantic Forest patches in MG (Meira-
forest, Sossego forest and PESB forest
Neto and Martins 2000). The intermediate rate of late
successional and pioneer species and the high rate of
early successional species (Table 2) indicate that this
PERD (obs)
3 a
41 a
52 a
forest was between the successional stages of the PERD
Sossego (obs)
7 b
31 b
19 b
and PESB samples.
PESB (obs)
10 b
32 b
17 b
Total (obs)
The basal area of the PESB forest sample was the
PERD (exp)
lowest and close to values found in ordinary patches of
Sossego (exp)
semideciduous seasonal forests landscapes (Lopes et al.
PESB (exp)
2002). Its richness was also the lowest. The a-diversity
Different letters are significant difference P < 0.02
and b-diversity were intermediate. The rate of pioneer
species was the highest and that of late successional
species the lowest (Table 2). In view of these results, the
PESB successional stage has been considered the earliest
Plant functional types rates and structural results of
among these forests.
the successional evaluation were contradictory between
Despite some subjectivity of successional functional
the PERD and Sossego samples. Nevertheless, they were
group classification found in the literature (Clark and
clear about the earlier successional phase of the PESB
Clark 1992; Dirzo et al. 1992; Ellison et al. 1993; Den-
forest sample. These results were reinforced by the less
slow et al. 1998; Ribas et al. 2003), diametric structure is
dense tree stratum in PESB sample (supplementary
objective and is widespread in forests studies (McElh-
material). The successional stage of the PESB forest
inny et al. 2005). Thorough comparisons of diametric
resembled the Sossego forest sample in terms of func-
structure were used for succession categorization (Koop
tional groups of species, without significant difference
et al. 1994; Tyrrel and Crow 1994). Diametric compar-
(Table 1; Fig. 3).
isons are considered to be one of the best methods for
measurement of succession (Van Den Meersschaut and
Vandekerkhove 1998; Dewalt et al. 2003; Parkes et al.
2003; McElhinny et al. 2005). Therefore, we are con-
sidering the PERD forest sample in latest stage of suc-
Succession of Atlantic Forests
cession because of plant functional type rate, richness,
and basal area. Nonetheless, the Sossego sample could
Succession of Atlantic Forests appears as an important
be considered at the same successional stage or even
process determining abundance of woolly spider mon-
more advanced because of the diametric distribution of
keys. By means of successional functional groups anal-
ysis and diametric structure, the PERD forest sample
Floristically, the PERD, Sossego, and PESB samples
was shown to be in the later, and PESB forest sample in
are distinct, with less than 25% similarity among them.
the earlier, stages of secondary succession.
Many characteristics are key factors to these floristic
In spite of the lower a-diversity (expressed by H¢)
differences among communities: history of disturbance,
and the lower rarefaction intercept values in the
distance between them (Tabarelli 1997; Zahawi and
PERD sample (Table 1; Fig.2), b-diversity and rich-
Augspurger 1999), edaphic factors (Carvalho et al. 1995;
ness of PERD were the highest. Diversity is supposed
Clark et al. 2002), climatic factors, distance from the
to be higher in later stages of succession in forests on
ocean, altitude, latitude and seasonality (Fontes 1997;
poor and acid soils (Grime 1987), as are the soils of
Oliveira-Filho and Fontes 2000, Meira-Neto and Mar-
these three samples of forest. The basal area value of
tins 2002). Alone, none of these factors is likely be the
the PERD forest sample was the highest among these
major influence on similarity or dissimilarity among
areas. The slope of diametric distribution was the least
these areas because the forests are geographically closed.
steep and the intercept was the lowest among the
forests in this study (Fig. 3). This indicates a lower
number of individuals in the class of thinner stems
Abundance of woolly spider monkeys
and a higher number of individuals in classes of
thicker stems. All these features indicate a later suc-
Structural traits driven by succession and evenness are
cessional stage. Among species whose successional
key factors for woolly spider monkey abundance pat-
functional group classification is in the literature
terns. Abundance and their determining factors are
(Table 2), the late successional group was the richest
among the most important ecological issues (Rosen-
in species number in PERD forest sample. The pio-
zweig 1995; Hubbell 2001; Chapman and Peres 2001;
neers were the poorest functional group in this forest.
Magurran 2004). Despite the fact that Brachyteles hyp-
Thus, we are considering PERD as the latest succes-
oxanthus, the woolly spider monkey, is a critically
sional forest among those investigated.
endangered species of primate (IUCN 2007), the struc-

tural traits of the forests and the minimal area necessary
be opportunists, seem to hinder succession, and are
for its occurrence are still being determined. Aguirre
interesting for further study in sites of occurrence of
(1971) wrote that woolly spider monkeys occur mainly
woolly spider monkeys. Nevertheless, occasional falling
in areas of the Atlantic Forests that are mature; we
is a death risk for big primate individuals (Peres C 2007,
understand this to be a reference to late successional
pers. comm.). A negative correlation is expected for
ones. Recently, most results relating to B. hypoxanthus
adult primate weight and the number of younger trees in
are from a secondary Atlantic Forest in the Caratinga
a secondary successional forest because thicker branches
Biological Reserve (Strier 2000; Strier et al. 2002;
are required to support heavier monkeys (Fimbel 1994).
Mendes et al. 2005), where this forest is not a proper
Thus, older forests mean late successional phases and
example of a late successional Atlantic Forest, at least in
support bigger populations of heavier primates, as ver-
most of its area. Thus, the preference of woolly spider
ified by Fimbel (1994) in western tropical forests in
monkeys for late successional forests has become a de-
Africa. Hence, late successional phases are more suitable
bated issue, and one crucial for its conservation. Could
for woolly spider monkeys.
forest structure be an explanatory variable in woolly
Late successional phases of tropical forests accumu-
spider monkeys’ abundance patterns? We found results
late biomass and basal area, according to most succes-
that showed species evenness (of trees) as a potential
sion models (Miles 1987). Communities accumulate
factor determining woolly spider monkey densities.
biodiversity beyond biomass through time (Hubbell and
Evenness values range from 0 to 1, from absolutely
Foster 1987; Grime 1987; Rosenzweig 1995; Hubbell
unequal distribution of individuals to complete equality
2001). The reports of higher densities of folivorous pri-
of this distribution, respectively (Pielou 1975; Magurran
mates in forests richer in tree species (Chiarello 1999;
2004). Hence, if a denser tree species, which causes de-
Ganzhorn 1995; Strier 2000) reinforce Aguirre’s (1971)
creased equability, is used as a food source by woolly
observations and our interpretation of our results.
spider monkeys, density of this primate is supposed to be
Recently proposed actions for woolly spider monkey
higher. On the other hand, if the densest plant species is
conservation have prioritized adding forested areas to
not a food item, the woolly spider monkeys are forced to
patches where these primates occur, after research to
move longer distances through canopies seeking food,
develop appropriate reforestation techniques (Mendes
resulting in diffuse populations. An important example
et al. 2005). As our results show, such added areas must
of a dense tree species that cause decreased evenness in
have a high richness of indigenous tree species, although
our results is Senefeldera multiflora of the PERD forest
evenness can be low, since the densest species is a pre-
sample. This tree species is not a food source for the
ferred food item for woolly spider monkeys. The added
woolly spider monkeys. Thus, the lower evenness results
restored forests must be structured mainly by late suc-
in higher locomotion requirements for seeking food and
cessional tree species and the areas must have also high
the lower density of B. hypoxanthus. Therefore, in the
tree densities to induce the canopy intersections required
PERD sample, woolly spider monkeys spend more en-
for secure locomotion of these big primates. Not only
ergy in horizontal locomotion to find food items than
amendments of disturbed areas, but also efforts to avoid
they do in that Sossego forest sample.
hunting pressure, are of fundamental importance.
The sampled area of the PERD forest had the lowest
Hunting pressure is the most powerful factor of distur-
evenness among tree species and had a lower number of
bance history in those studied areas and is potentially
observations of woolly spider monkeys (71 observations)
stronger at Sossego forest because of the small size, as a
than the sampled area of the Sossego forest for the same
result of higher perimeter/area ratio. However, the
period (273 observations) (Biodiversitas 2006). In the
densest population of woolly spider monkeys at the
sampled area of the PESB forest, the number of obser-
Sossego forest sample shows that this pressure is not
vations of woolly spider monkeys was the lowest for the
defining the abundance patterns as a main factor.
same period (22 observations) (Biodiversitas 2006). This
Besides hunting, disturbance of vegetation is a major
area had a higher evenness than the PERD forest sam-
factor for woolly spider monkey conservation and can
ple, but had the lowest tree density and basal area. In
be evaluated directly by means of structural studies of
this patch of the PESB forest, the locomotion of these
vegetation, as did this work.
primates must often be vertical because tree canopies are
not connected enough to build secure paths for these
Acknowledgments The authors especially thank The Fundacao de
heavy monkeys. This structural feature is due to past
Amparo a Pesquisa de Minas Gerais—FAPEMIG for grants and
support, as well as the PROBIO/Environment Ministry of Brazil
disturbance, when patches of PESB forest were logged
and the Biodiversitas Foundation’s ‘‘Conservac¸a˜o do Muriqui em
for charcoal production. This allowed bamboo species,
Minas Gerais’’ project, which partially sponsored this paper. In
mainly Chusquea capitata Nees, to dominate the
addition, we thank two anonymous reviewers for helpful comments
understory of some sites, as in the case of the sampled
on earlier draft of the manuscript. Luiz Dias, Carlos Leandro,
Sergio Mendes, Marcio Assis, Amilcar Saporetti, and Michellia
area. The bamboo seem to be specialized to specific
Soares aided in fieldwork. The paper is dedicated in memoriam of
microsites with high canopy openness, an important
Eduardo Veado who fought and died for conservation of woolly
factor for plant populations in Atlantic Forest under-
spider monkeys and of the Atlantic Forest. This work was spon-
story (Souza and Martins 2004). The bamboos seem to
sored by FAPEMIG #1540/05 and #61540/05.

Grime JP (1987) Dominant and subordinate components of plant
communities: implications for succession, stability and diver-
sity. In: Gray AJ, Crawley MJ, Edwards PJ (eds) Colonization,
Aguirre AC (1971) O mono Brachyteles arachnoides (E. Geoffroy).
succession and stability. Blackwell Scientific, Oxford, pp 413–
Anais da Academia Brasileira de Cieˆncias, Rio de Janeiro
Arrhenius O (1922) A new method for the analisys of plant com-
Hubbell SP (1997) A unified theory of biogeography and relative
munities. J Ecol 10:185–199. doi:10.2307/2255740
species abundance and its implication to tropical rain forests
Biodiversitas (2006) Conservac¸a˜o e manejo do muriqui em Minas
and coral reefs. Coral Reefs 16(Suppl):9–21. doi:10.1007/
Gerais: relato´rio final. Fundac¸a˜o Biodiversitas, Belo Horizonte
Bourg, DM (2006) Excel scientific and engineering cookbook.
Hubbell SP (2001) The unified neutral theory of biodiversity and
O’Reilly Media (eds), Sebastopol
biogeography. Princeton University Press, Princeton
Caiafa AN, Silva AF (2005) Composic¸a˜o Florı´stica e Espectro
Hubbell S, Foster RB (1987) The spatial context of regeneration in
Biolo´gico de um Campo de Altitude no Parque Estadual da
a neotropical forest. In: Gray AJ, Crawley MJ, Edwards PJ
Serra do Brigadeiro, MG—Brasil. Rodriguesia 56:163–173
(eds) Colonization, succession and stability. Blackwell Scien-
Carvalho DA, Oliveira-Filho AT, Vilela EA, Gavilanes ML
tific, Oxford, pp 395–412
(1995) Estrutura fitossociolo´gica de mata ripa´ria do Alto Rio
IBGE (2001) Mapa de climas do Brasil. Instituto Brasileiro de
Frande (Bom Sucesso, estado de Minas Gerais). Rev Bras
Geografia e Estatı´stica, Rio de Janeiro
Bot 18:39–49
IUCN 2007 (2007) IUCN Red List of Threatened Species.
Chapman CA, Peres CA (2001) Primate conservation in the new Downloaded on 10 September 2008
millennium: the role of scientist. Evol Anthropol 10:16–33. doi:
Koop H, Rijksen HD, Wind J (1994) Tools to diagnose forest
integrity: an appraisal method substantiated by Silvi-Star
assessment of diversity and forest structure. In: Boyle TJB,
Chiarello AG (1999) Effects of fragmentation of the Atlantic forest
Boontawee B (eds) Measuring and monitoring biodiversity in
on mammal communities in South-eastern Brazil. Biol Conserv
tropical and temperate forest. CIFOR, Chaing Mai, pp 309–
89:71–82. doi:10.1016/S0006-3207(98)00130-X
Clark D, Clark DB (1992) Life history diversity of canopy and
Leita˜o Filho HF, Pagano SN, Cesar O, Timoni JL, Rueda JJ (1993)
emergent trees in a Neotropical Rain Forest. Ecol Monogr
Ecologia de Mata Atlaˆntica em Cubata˜o. Editora da Unesp;
62:315–344. doi:10.2307/2937114
Editora da UNICAMP, Sa˜o Paulo
Clark JS, Grimm EC, Donovan JJ, Fritz SC, Engstrom DR,
Lemos-de-Sa´ RM, Strier KB (1992) A preliminary comparison of
Almendinger JE (2002) Drought cycles and landscape responses
forest structure and use by two isolated groups of woolly spider
to past aridity on prairies of the Northern Great Plains, USA.
monkeys, Brachyteles arachnoides. Biotropica 24:455–459. doi:
Ecology 83:595–601
Clutton-Brock TH, Harvey PH (1977) Primate ecology and social
Lomolino M, Riddle BR, Brown JH (2006) Biogeography. Sinauer
organization. J Zool 183:1–39
Associates, Sunderland
Condit R, Hubbell SP, Lafrankie JV, Sukumar R, Manokaran N,
Lopes WP, Silva AF, Souza AL, Meira-Neto JAA (2002) Estrutura
Foster RB, Ashton PS (1996) Species-area and species indi-
fitossociolo´gica de um trecho de vegetac¸a˜o arbo´rea no Parque
vidual relationships for tropical trees: a comparison of three
Estadual do Rio Doce, Minas gerais. Acta Bot Brasilica 16:443–
50-ha plots. J Ecol 84:549–562. doi:10.2307/2261477
456. doi:10.1590/S0102-33062002000400007
Cottam G, Curtis JT (1956) The use of distance measures in phy-
Magurran AE (2004) Measuring biological diversity. Blackwell,
tossociological sampling. Ecology 3:451–460. doi:10.2307/
Martins SV, Rodrigues RR (1999) Produc¸a˜o de serapilheira em
Denslow J, Ellison AM, Sanford RE (1998) Treefall gap size effects
clareiras de uma Floresta Estacional Semidecidual no Municı´-
on above- and below-ground processes in a tropical wet forest.
pio de Campinas, SP. Rev Bras Bot 22:405–412. doi:10.1590/
J Ecol 86:597–609. doi:10.1046/j.1365-2745.1998.00295.x
Devore J, Farnum N (2005) Applied statistics for engineers and
McElhinny C, Gibbons P, Brack C, Bauhus J (2005) Forest and
scientists. Thomson Brooks/Cole, Belmont
woodland stand structural complexity: its definition and mea-
Dewalt SJ, Maliakal SK, Denslow JS (2003) Changes in vegetation
surement. For Ecol Manage 218:1–24. doi:10.1016/j.foreco.
structure and composition along a tropical forest chronose-
quence: implications for wildlife. For Ecol Manage 182:139–
Meira-Neto JAA, Martins FR (2000) Estrutura da Mata da Silvi-
151. doi:10.1016/S0378-1127(03)00029-X
cultura, uma floresta estacional semidecidual montana no mu-
Dirzo R, Horvitz CC, Quevedo H, Lopez MA (1992) The effects of
nicı´pio de Vic¸osa-MG. Rev Arvore 24:151–160
gap size and age on the understorey herb community of a
Meira-Neto JAA, Martins FR (2002) Composic¸a˜o florı´stica de uma
tropical Mexican rain forest. J Ecol 80:809–822. doi:10.2307/
Floresta Estacional Semidecidual Montana no Municı´pio de
Vic¸osa-MG. Rev Arvore 26(4):437–446. doi:10.1590/S0100-
Ellison AM, Denslow JS, Lioselle BA, Brenes DM (1993) Seed and
seedling ecology of neotropical Melastomataceae. Ecology
Mendes LS, Melo FR, Boubli JP, Dias LG, Strier KB, Pinto LPS,
74:1733–1749. doi:10.2307/1939932
Fagundes V, Cosenza B, De-Marco P (2005) Directives for the
Fimbel C (1994) Ecological correlates of species success in the
conservation of the northern muriqui, Brachyteles hypoxanthus
modified habitats may be disturbance and site-specific: the
(Primates, Atelidae). Neotrop Primates 13(Suppl):7–18
primates of Tiwai Island. Conserv Biol 1:106–113. doi:
Miles J (1987) Vegetation succession: past and present perceptions.
In: Gray AJ, Crawley MJ, Edwards PJ (eds) Colonization,
Fontes MAL (1997) Ana´lise da composic¸a˜o florı´stica das florestas
succession and stability. Blackwell Scientific, Oxford, pp 1–30
nebulares do Parque Estadual do Ibitipoca, Minas Gerais. MS
Mota LP (1995) Diamfito: programa de ana´lise de distribuic¸a˜o
thesis, Universidade Federal de Lavras, Lavras
diame´trica. Universidade Federal de Vic¸osa, Vic¸osa
Ganzhorn JU (1995) Low-level forest disturbance effects on pri-
Mueller-Dombois D, Ellenberg H (1974) Aims and methods of
mary production, leaf chemistry, and lemur population. Ecol-
vegetation ecology. Wiley, New York
ogy 76:2084–2096. doi:10.2307/1941683
Nakazono EM, Costa MC, Futatsugi K, Paulilo MTS (2001)
Gaulin SJC, Knight DH, Gaulin C (1980) Local variance in
Crescimento inicial de Euterpe edulis Mart. em diferentes re-
Alouatta group size and food availability on Barro Colorado
gimes de luz. Rev Bras Bot 24:173–179. doi:10.1590/S0100-
Island. Biotropica 12:137–143. doi:10.2307/2387729

Newton AC (2007) Forest ecology and conservation: a handbook
Shepherd GJ (2002) Fitopac: manual do usua´rio. UNICAMP,
of techniques. Oxford University Press, Oxford
Nunes YRF, Mendonc¸a AVR, Botezelli L, Machado ELM,
Silva AF, Oliveira RV, Santos NL, Paula A (2003) Composic¸a˜o
Oliveira-Filho AT (2003) Variac¸o˜es da fisionomia, diversidade e
florı´stica e grupos ecolo´gicos das espe´cies de um trecho de
composic¸a˜o de guildas da comunidade arbo´rea em um frag-
floresta semidecı´dua submontana da Fazenda Sa˜o Geraldo,
mento de Floresta Semidecidual em Lavras, MG. Acta Bot
Brasilica 17:213–229. doi:10.1590/S0102-33062003000200005
Oliveira-Filho AT, Fontes MAL (2000) Patterns of floristic dif-
Souza AF, Martins FR (2004) Microsite specialization and spatial
ferentiation among Atlantic Forests in southeastern Brazil and
distribution of Geonoma brevispatha, a clonal palm in south-
the influence of climate. Biotropica 32(4b):793–810
eastern Brazil.
Ecol Res 19:521–532. doi:10.1111/j.1440-
Oliveira-Filho AT, Carvalho DA, Vilela EA, Curi N, Fontes MAL
(2004) Diversity and structure of the tree community of a
Strier KB (1987) Activity budgets of woolly spider monkeys, or
fragment of tropical secondary forest of the Brazilian Atlantic
muriquis (Brachyteles arachnoides). Am J Primatol 13:385–395.
Forest domain 15 and 40 years after logging. Rev Bras Bot
27:685–701. doi:10.1590/S0100-84042004000400009
Strier KB (1991) Diet in one group of woolly spider monkey, or
Parkes D, Newell G, Cheal D (2003) Assessing the quality of native
Muriquis (Brachyteles arachnoides). Am J Primatol 23:113–126.
vegetation: the ‘‘habitat hectares’’ approach. Ecol Manage
Restor 4:S29–S38. doi:10.1046/j.1442-8903.4.s.4.x
Strier KB (2000) Population viability and regional conservation
Paula AP, Silva AF, De-Marco P, Mae¨s-Dos-Santos FA, Souza
priorities for muriquis (Brachyteles arachnoides) in Brazil’s
AL (2004) Sucessa˜o ecolo´gica da vegetac¸a˜o arbo´rea em uma
Atlantic Forest. Biotropica 32:903–913
Floresta Estacional Semidecidual, Vic¸osa, MG, Brasil. Acta
Strier KB, Boubli JP, Guimara˜es VO, Mendes SL (2002) The
Bot Brasilica 18:407–423. doi:10.1590/S0102-330620040003
muriquis of the Estac¸a˜o Biolo´gica de Caratinga, Minas Gerais,
Brazil: updates. Neotrop Primates 10:115–119
Petroni LM (2000) Caracterizac¸a˜o da a´rea de uso e dieta do nono
Tabarelli M (1997) A regenerac¸a˜o da floresta Atlaˆntica montana.
carvoeiro (Brachyteles arachnoides, Cebidae-Primates) na Mata
Doctorate thesis, Instituto de Biocieˆncias, Universidade de Sa˜o
Atlaˆntica, Serra de Paranapiacaba, SP. Doctorate thesis, Uni-
Paulo, Sa˜o Paulo
versidade de Sa˜o Paulo, Sa˜o Paulo
Tyrrel LF, Crow TR (1994) Structural characteristics of old-growth
Pielou EC (1975) Ecological diversity. Wiley, New York
hemlock-hardwood forests in relation to age. Ecology 75:370–
Ribas RF, Meira-Neto JAA, Silva AF, Souza AL (2003) Com-
386. doi:10.2307/1939541
posic¸a˜o florı´stica de dois trechos em diferentes etapas serais de
Van Den Meersschaut D, Vandekerkhove K (1998) Development
uma Floresta Estacional Semidecidual em Vic¸osa, Minas
of a stand-scale forest biodiversity index based on the State
Gerais. Rev Arvore 27:821–830. doi:10.1590/S0100-6762200
Forest Inventory. In: Hansen M, Burk T (eds) Integrated tools
for natural resource inventories in the 21st century. USDA,
Rolim SG, Couto HTZ, Jesus RM (1999) Mortalidade e recruta-
Boise, pp 340–349
mento de a´rvores na Floresta Atlaˆntica em Linhares (ES).
Veloso HP, Rangel Filho ALR, Lima JCA (1991) Classificac¸a˜o da
Scientia Forestalis 55:49–69
vegetac¸a˜o brasileira adaptada a um sistema universal. IBGE,
Rosenzweig ML (1995) Species diversity in space and time. Cam-
Rio de Janeiro
bridge University Press, Cambridge
Zahawi RA, Augspurger CK (1999) Early plant succession in
Salimon CI, Negrelle RB (2001) Natural regeneration in a
abandoned pastures in Ecuador. Biotropica 31:540–552. doi:
quaternary coastal plain in Southern Brazilian Atlantic Rain
Forest. Braz Arch Biol Technol 44:155–163. doi:10.1590/S1516-

Document Outline

  • Structure of Brazilian Atlantic forests with occurrence of the woolly spider monkey \(Brachyteles hypoxanthus\)
  • Abstract
  • Introduction
  • Materials and methods
  • Study sites
  • Methods
  • Fig1
  • Results
  • Diversity
  • Tab1
  • Succession and structure
  • Fig2
  • Fig3
  • Discussion
  • Succession of Atlantic Forests
  • Abundance of woolly spider monkeys
  • Tab2
  • Acknowledgments
  • References
  • CR1
  • CR2
  • CR3
  • CR4
  • CR5
  • CR6
  • CR7
  • CR9
  • CR10
  • CR11
  • CR12
  • CR13
  • CR14
  • CR15
  • CR16
  • CR17
  • CR19
  • CR20
  • CR21
  • CR22
  • CR23
  • CR24
  • CR26
  • CR27
  • CR28
  • CR29
  • CR25
  • CR30
  • CR32
  • CR33
  • CR34
  • CR35
  • CR36
  • CR37
  • CR38
  • CR39
  • CR40
  • CR41
  • CR42
  • CR43
  • CR44
  • CR45
  • CR46
  • CR47
  • CR48
  • CR49
  • CR50
  • CR51
  • CR52
  • CR53
  • CR54
  • CR55
  • CR56
  • CR57
  • CR58
  • CR60
  • CR61
  • CR62
  • CR63
  • CR64
  • CR65
  • CR66
  • CR67
  • CR68
  • CR69
  • CR70
  • CR71