Contrasting Demographic Structure of Short- and Long-lived Pioneer Tree Species on
Amazonian Forest Edges
Graciliano G. A. Santos1, Bra´ulio A. Santos2,4,5, Henrique E. M. Nascimento3, and Marcelo Tabarelli2
1 Centro de Estudos Integrados em Biodiversidade Amazoˆnica, Universidade Federal do Amapa´, Macapa´, Amapa´, 68902-280, Brazil
2 Departamento de Botaˆnica, Universidade Federal de Pernambuco, Recife, Pernambuco, 50670-901, Brazil
3 Coordenac¸a˜o de Pesquisa emSilvicultura Tropical, Instituto Nacional de Pesquisas da Amazoˆnia, C.P. 478, Manaus, Amazonas, 69011-970, Brazil
ABSTRACT
Although tropical forests have been rapidly converted into human-modified landscapes, tree species response to forest edges remains
poorly examined. In this study, we addressed four pioneer tree species to document demographic shifts experienced by this key ecologi-
cal group and make inferences about pioneer response to forest edges. All individuals with dbh 
 1 cm of two short-lived (Bellucia gros-
sularioides and Cecropia sciadophylla) and two long-lived species (Goupia glabra and Laetia procera) were sampled in 20 1-ha forest edge plots
and 20 1-ha forest interior plots in Oiapoque and Manaus, Northeast and Central Amazon, respectively. As expected, pioneer stem den-
sity with dbh 
 1 cm increased by around 10–17-fold along forest edges regardless of species, lifespan, and study site. Edge popula-
tions of long-lived pioneers presented 84–94 percent of their individuals in sapling/subadult size classes, whereas edge populations of
short-lived pioneers showed 56–97 percent of their individuals in adult size classes. These demographic biases were associated with neg-
ative and positive net adult recruitment of long- and short-lived pioneers, respectively. Our population-level analyses support three gen-
eral statements: (1) native pioneer tree species proliferate along forest edges (i.e., increased density), at least in terms of non-reproductive
individuals; (2) pioneer response to edge establishment is not homogeneous as species differ in terms of demographic structure and net
adult recruitment; and (3) some pioneer species, particularly long-lived ones, may experience population decline due to adult sensitivity
to edge-affected habitats.
Key words: Bellucia grossularioides; Cecropia sciadophylla; edge effects; Goupia glabra; habitat fragmentation; Laetia procera; plant demography.
HABITAT LOSS AND FRAGMENTATION REPRESENT A KEY FORCE on the
ongoing global crisis of biodiversity and associated ecosystem ser-
vices (Laurance & Peres 2006). About 710,000 km2 (17%) of the
Amazon Basin had been deforested by 2006 (Vieira et al. 2008),
especially in the ‘arc of deforestation’ on the southern and eastern
boundaries. New deforestation frontiers have arisen in Central
and Northern Amazonia as well, resulting in ‘archipelagos’ of
small forest fragments and the creation of 70,000 km of new for-
est edges every year (Fearnside 2005, Silva et al. 2005, Broadbent
et al. 2008). These changes in landscape configuration negatively
affect the remaining biota due to area, isolation, and matrix
effects (Ewers & Didham 2006), but edge effects persist as the
main drivers of ecological changes experienced by forest frag-
ments (Laurance et al. 2011).
Edge effects consist of abiotic and biotic processes operating
along forest edges, which result in distorted tree assemblages.
Recent evidence from an aging (>200 yr-old) and hyper-frag-
mented landscape in the Atlantic forest of Northeast Brazil sug-
gests that edge effects represent the main force for driving forest
fragments or edge-affected habitats toward early-successional sys-
tems (Santos et al. 2008, 2010; Tabarelli et al. 2008, Lopes et al.
2009, see also Pütz et al. 2011, but see Borouncle & Finegan
2011). The striking biomass collapse, floristic simplification, and
functional impoverishment observed in recently (<40 yr-old) iso-
lated Amazonian forest fragments indicate that such trajectory
may be operational in the Amazon as well (Laurance et al. 2011
and references therein). Our current understanding on the
demography of individual tropical tree species along forest edges,
however, remains insufficient to make accurate predictions about
population persistence and potential successional trajectory expe-
rienced by such altered habitats (but see Brum et al. 2008 for the
tree palm Oenocarpus bacaba). In the context of demographic
responses to edge effects, several studies have demonstrated that
populations of the perennial herb Heliconia acuminata are skewed
toward smaller size classes in Amazonian forest fragments (e.g.,
Bruna 2002, Bruna & Kress 2002, Bruna & Oli 2005), possibly
due to chronic reduction in plant growth and inflorescence pro-
duction in forest fragments (Gagnon et al. 2011).
It is well known that seedlings, saplings, and adults of some
pioneer tree species are able to dominate forest edges (Sizer &
Tanner 1999, Laurance et al. 2006, Michalski et al. 2007, Santos
et al. 2008), but whether short- and long-lived pioneer species
respond similarly in terms of demographic structure is an open
question. Overall, short-lived pioneers (i.e., lifespan <50 yr) have
their demography completely associated with forest gaps, but
their smaller adults usually thrive along forest edges (Tabarelli
et al. 2010a, Laurance et al. 2011), indicating recent canopy
Received 15 August 2011; revision accepted 28 February 2012.
4Corresponding author; e-mail: braulioalmeida@yahoo.com.br
5Current address: Departamento de Sistemática e Ecologia, Universidade
Federal da Paraíba, João Pessoa, Paraíba, 58051-900, Brazil.
ª 2012 The Author(s) 771
Journal compilation ª 2012 by The Association for Tropical Biology and Conservation
BIOTROPICA 44(6): 771–778 2012 10.1111/j.1744-7429.2012.00882.x
disturbance (Whitmore 1989). Long-lived pioneers (i.e., lifespan
>50 yr) are also gap-dependent, but their taller, emergent adults
generally persist in later successional stages by attaining the upper
strata before canopy closure. Such large trees are relicts of past
forest disturbance (Clark & Clark 1996) and usually vanish from
forest edges due to increased wind turbulence and habitat desic-
cation (D’Ângelo et al. 2004, Laurance & Curran 2008). Based
on such regeneration strategies, it is reasonable to expect that
after some decades of edge creation, differences in adult recruit-
ment and mortality between short- and long-lived pioneers may
result in distinct demographic structures, affecting their relative
contribution to forest structure, composition, and functioning
along forest edges.
In this study, we sampled four pioneer tree species (two
short- and two long-lived) across edge and forest interior plots of
two Amazonian sites to document demographic shifts experi-
enced by this key ecological group and make inferences about
pioneer response to forest edges. We expected to find a prolifera-
tion of pioneer species across edge plots (i.e., a positive response
to edge establishment), with edge populations of short-lived spe-
cies supporting a large number of adults (no adult recruitment
bottlenecks), whereas those of long-lived species being mostly
represented by saplings/subadults due to negative net adult
recruitment. First, we documented cross-habitat shifts on stem
density (edge vs. interior plots) and described size-distribution of
stems (hereafter demographic structure) considering all species,
habitats, and sites. Second, we offered estimates of adult recruit-
ment and mortality over 23–27 yr for four species in one of the
study sites. Finally, we explored the generality of our findings and
provide some insights regarding pioneer demographic responses,
species persistence, and the successional trajectory experienced by
edge-affected habitats in human-modified landscapes.
METHODS
STUDY SITES.—The study was conducted near the cities of Oiapo-
que and Manaus, Northeast and Central Amazon (Fig. S1A). In
Northeast Amazon, the study site refers to a private cattle ranch
located 40 km south of Oiapoque at 100 m asl (3° 33′ 44″ N,
51° 47′ 22″ W; Fig. S1B). The study site is located in the rain
forest eco-region (i.e., terra firme forest) in a sub-region called
‘Superfície Dissecada Guianense’. Annual rainfall averages
2500 mm, with pronounced dry season from September to
November (Souza & Cunha 2010). Average annual temperature
is 26.5°C and soils vary with topography from red-yellow
distrophic latosoil to red-yellow argisoil.
In Central Amazon, the study area is the 1000-km2 experi-
mentally fragmented landscape of the Biological Dynamics of
Forest Fragments Project (BDFFP), located 80 km north of
Manaus at 100–150 m asl (2° 26′ 00″ S, 59° 52′ 25″ W;
Fig. S1C). The mean annual precipitation is 2200 mm with a pro-
nounced dry season from June to September. Average tempera-
ture is 26.7°C and the predominant soil type is clay distrophic
yellow latosoil (Laurance et al. 1999). Terra firme tropical rain for-
ests account for most of the vegetation cover of the study area.
Forest canopy is 30–37 m tall, with some emergent trees reaching
55 m. Tree species richness (
 10 cm diameter at breast height
[dbh]) may exceed 280 species/ha (Oliveira & Mori 1999).
Landscape matrix differs between the study sites. In Oiapo-
que, pastures were established in late 1990s and have been main-
tained through annual burnings. In Manaus, pastures were
opened in the 1980s and abandoned in the early 1990s. Fire was
used in some BDFFP areas and not used in others, resulting in
secondary forest patches dominated by Vismia and Cecropia spe-
cies, respectively (Mesquita et al. 1999, Williamson & Mesquita
2001). The attributes of surrounding matrix have been recognized
to affect tree recruitment along forest edges and may confound
the demographic responses of pioneer trees to edge proximity/
effects (Nascimento et al. 2006).
PLANT SURVEYS.—In this study, we sampled four phylogenetically
unrelated pioneer species  two short-lived (Bellucia grossularioides
and Cecropia sciadophylla) and two long-lived (Goupia glabra and Lae-
tia procera). These four taxa refer to light-demanding, fast-growing
tree species with continuous or annual reproduction, and high
ability to rapidly respond to human changes in the landscape con-
figuration (Laurance et al. 2006, Bentos et al. 2008, Zalamea et al.
2008, 2011); i.e., they are typical early colonizers in the Amazon
Basin (Table 1). Between September and December 2007, we
established 20 1-ha plots (100 9 100 m) in a vast tract of con-
tinuous forest in Oiapoque: ten at 0 m from the edge (edge
plots) and ten in core areas (interior plots) 1000 m away from
the nearest edge. The first edge plot was randomly selected and
the others were located every 300 m along the forest edge. No
edge plot was simultaneously exposed to two edges and interior
plots were disposed in pairs with edge ones. In Manaus, a long-
term study of vegetation dynamics has been conducted in forest
fragments and continuous forests since early 1980s. About
65,000 trees (dbh 
 10 cm) have been periodically monitored in
66 1-ha permanent plots distributed across 1-ha, 10-ha, and 100-
ha forest fragments and several continuous forests (Rankin-
de-Merona et al. 1990, Laurance et al. 1998). Tree mortality,
recruitment, and growth have been recorded every 4–6 yr. More
recently, stems sizing 1–9.9 cm dbh started to be included in
plant inventories within each 1-ha plot. For this study, we ran-
domly selected 20 of the 66 BDFFP plots: ten edge plots and
ten interior plots.
In each of the 40 1-ha plots, all individuals with dbh 
 1 cm
of the four selected species were classified into four size classes
according to their dbh: I (1 to <2 cm), II (2 to <10 cm), III (10 to
<20 cm), and IV (
 20 cm). The proportion of individuals in each
size class was used to characterize the demographic structure of
these edge populations. During the latest BDFFP vegetation cen-
sus, the short-lived B. grossularioides was not present in any plot and
the long-lived L. procera had not been systematically sampled for
saplings and subadults (1.0–9.9 cm dbh). Thus, we excluded these
species from the demographic structure analyses for Manaus.
Finally, we used the BDFFP data to estimate net adult recruitment
along forest edges for each species, calculated by the number of
trees (dbh 
 10 cm) present in the last census after 23–27 yr of
772 Santos, Santos, Nascimento, and Tabarelli
fragmentation (‘final number of trees’) minus the number of trees
established before fragmentation (‘initial number of trees’). Unfor-
tunately, we were not able to examine net sapling recruitment
because they were sampled only once, but we are aware that the
absence of dynamic data on sapling mortality and recruitment (still
rare in the literature) limits the conclusions that could be drawn
from our findings.
STATISTICAL ANALYSES.—All analyses were performed separately
for Oiapoque and Manaus sites. To examine whether or not plant
density differed between edge and interior plots, we ran a gener-
alized linear mixed model with habitat, species, and their interac-
tion as fixed factors and plot nested within habitat as a random
factor. The number of individuals per 1-ha was set as a depen-
dent variable after log (x + 1) transformation. Random factor
was used to increase the explanatory power of the model and to
estimate the magnitude of the unexplained variance attributed to
differences among plots (a measure of spatial variation in plant
density). We used residual maximum likelihood method (REML)
to separate variances of random and fixed effects (Grafen &
Hails 2002) and Tukey–Kramer HSD (honestly significant differ-
ence) tests to compare differences among treatments.
On the edges of both Oiapoque and Manaus sites, popula-
tion size at the 1-ha basis was too small to provide a reasonable
representation of the demographic structure at this spatial scale
(median plant density per species was lower than 15 plants per
hectare in both sites). Thus, we pooled data from all ten edge
plots, classified individuals into the four size classes (I, II, III,
and IV), and then used likelihood ratio Chi-square tests to com-
pare the demographic structure among species. For Oiapoque,
the proportion of individuals in each size classes was compared
not only among the four species but also between pairs of spe-
cies with similar lifespan (i.e., Bellucia vs. Cecropia and Goupia vs.
Laetia). Finally, we used one-way analysis of variance to compare
the net adult recruitment among species after log (x + 1) trans-
formation of initial and final number of trees. All analyses were
performed in JMP 7 (SAS Institute Inc., Cary, NC, USA). We
report mean and standard errors for untransformed data.
RESULTS
EDGE EFFECTS ON STEM DENSITY.—We recorded a total of 643
individuals 
 1 cm dbh in the 20 Oiapoque plots: 74 of B. grossu-
larioides, 195 of C. sciadophylla, 266 of G. glabra, and 108 of L. pro-
cera. As expected, there was a very strong effect of habitat type
on the density of all species considered (F 1,18 = 44.1;
P < 0.0001). Plant density was on average 17-fold greater in edge
than in interior plots (60.8 ind/ha vs. 3.5 ind/ha; Fig. 1A),
despite species identity (F 3,54 = 2.3; P = 0.085; Fig. 1B). In
Manaus, we recorded a total of 283 individuals: 157 of C. sciado-
phylla and 126 of G. glabra. As in Oiapoque, habitat type affected
significantly plant density (F 1,18 = 8.1; P < 0.01), which was on
average 10-fold greater in edge than in interior plots (25.9 ind/ha
vs. 2.4 ind/ha; Fig. 1C), irrespective of species identity (F 1,18 =
1.9; P = 0.185; Fig. 1D).
DEMOGRAPHIC STRUCTURE OF EDGE POPULATIONS.—In Oiapoque,
the demographic structure of edge populations varied not only
among species differing in lifespan (v2 = 410.5; df = 9;
P < 0.0001) but also between species with similar lifespan (short-
lived species v2 = 53.1; df = 3; P < 0.0001; long-lived species
v
2
= 35.8; df = 3; P < 0.0001). Despite these differences, 94 and
84 percent of the total individuals of the long-lived G. glabra and
L. procera were <10 cm (classes I and II), respectively, whereas 80
and 56 percent of the individuals belonging to the short-lived C.
sciadophylla and B. glossularioides were 
 10 cm (classes III and IV)
(Fig. 2A). A similar trend was observed in the edge populations
of Manaus. About 87 percent of the individuals of the long-lived
G. glabra were assigned to classes I and II, whereas 97 percent of
the individuals of the short-lived C. sciadophylla were classified into
categories III and IV (v2 = 220.3; df = 3; P < 0.0001) (Fig. 2B).
NET ADULT RECRUITMENT.—At the edge plots of Manaus, net
adult recruitment differed significantly among species (F 3,36 =
4.6; P = 0.0079), but it was remarkably greater in C. sciadophylla
as compared with the other species (Tukey–Kramer HSD test,
P < 0.05) (Fig. 3). After 23–27 yr of fragmentation, all B. grossula-
rioides adults established before fragmentation died (N = 3) and
were not replaced by any other adults, resulting in the collapse of
this short-lived pioneer along the edges. On the other hand, the
two C. sciadophylla adults established before fragmentation died,
but were replaced by 150 new adults (113 of them in only one
plot [see the size of error bar in Fig. 3]). According to our initial
prediction, the long-lived G. glabra and L. procera exhibited a
negative net recruitment of adults along forest edges. More specif-
ically, there were 15 G. glabra adults established before fragmenta-
tion and only ten remained after 23–27 yr of fragmentation.
TABLE 1. Pioneer tree species studied, their family, lifespan, minimum dbh (cm) for
flowering, reproductive frequency, and known pollinators and seed dispersers.
Modified from Bentos et al. 2008.
Bellucia
glossularioides
Cecropia
sciadophylla
Goupia
glabra
Laetia
procera
Family Melastomataceae Urticaceae Celastraceae Salicaceae
Lifespan Short-lived Short-lived Long-lived Long-
lived
Maximum
height (m)
20–25 30 40 35
Minimum
reproduction
dbh (cm)
7.8 4.8 8.6 15.7
Reproduction
frequency
Continuously Annually Continuously Annually
Pollinator Bees Bats Bees Bees
Disperser Birds,
monkeys
Birds, bats,
monkeys
Birds Birds,
monkeys
Demographic Structure of Pioneer Tree Species 773
Similarly, L. procera was represented by six individuals before frag-
mentation and by four, three decades later.
DISCUSSION
Our population-level analyses, although based on a limited num-
ber of species, support three general statements: (1) native pio-
neer tree species proliferate along forest edges (i.e., increased
density), at least in terms of non-reproductive individuals; (2) pio-
neer response to edge establishment is not homogeneous as spe-
cies differ in terms of demographic structure and net adult
recruitment; and (3) some pioneer species, particularly long-lived
ones, may experience population decline due to adult sensitivity
to forest edges. Briefly, stem density of pioneers increased by
around 10–17-fold along forest edges regardless of species, life-
span, and study site. Edge populations of long-lived pioneers,
however, presented most of their individuals (84–94%, 82–235
individuals) in sapling/subadult size classes, whereas edge popu-
lations of short-lived pioneers showed most of their individuals
(56–97%, 40–148 individuals) in adult size classes, corroborating
our initial prediction. These demographic patterns were consistent
with a positive net adult recruitment experienced by the short-
lived C. sciadophylla, whereas both long-lived pioneer species
exhibited a negative recruitment of adults.
Proliferation of pioneer species on forest edges and small
forest fragments (i.e., edge-affected habitats) is not a novelty in
the Amazon (e.g., Laurance et al. 2006, Michalski et al. 2007) and
FIGURE 1. Plant density (mean ± SE) of Bellucia grossularioides (BG), Cecropia
sciadophylla (CS), Goupia glabra (GG), and Laetia procera (LP) in ten edge and
ten interior 1-ha plots at Oiapoque, Northeast Amazonia (A, B), and Manaus,
Central Amazonia (C, D). In figures A and C, letters E and I represent edge
and interior, respectively.
FIGURE 2. Demographic structure of short- and long-lived pioneer tree spe-
cies along forest edges in (A) Oiapoque (Northeast Amazonia) and (B) Man-
aus (Central Amazonia), Brazil. Population size in Oiapaque was 72, 187, 251,
and 98 for Bellucia grossularioides, Cecropia sciadophylla, Goupia glabra, and Laetia
procera, respectively. In Manaus it was 153 for C. sciadophylla and 103 for G.
glabra. SL, short-lived; LL, long-lived.
774 Santos, Santos, Nascimento, and Tabarelli
Atlantic forest (Santos et al. 2008, Tabarelli et al. 2010b). At Man-
aus, forest regeneration following clear-cutting and cattle raising
(i.e., abandoned pastures) has originated second-growth stands
dominated by either Cecropia or Vismia species (Mesquita et al.
1999, Williamson & Mesquita 2001). This change in landscape
matrix differentially affected the recruitment of certain pioneer
tree species on abutting forest edges (Nascimento et al. 2006). C.
sciadophylla, in particular, accounted for nearly 50 percent of all
pioneer trees in Cecropia-surrounded fragments, indicating that
some pioneers species are proliferating across a wide range of
habitats in human-modified landscapes (Tabarelli et al. 2010b,
Laurance et al. 2011). More than a landscape-level phenomenon,
such proliferation is apparently operating at multiple spatial scales.
In the Atlantic forest of Northeast Brazil, where forest conver-
sion to agriculture dates back to 18th century and forest edges
account for most of the remaining vegetation (Ranta et al. 1998,
Ribeiro et al. 2009), native pioneer tree species have driven
remaining forest patches to biotic homogenization at regional
level (Lôbo et al. 2011). Compared with Atlantic forest land-
scapes, our Amazonian study sites are much younger (<40 yr-old)
and still embedded within less fragmented and deforested land-
scapes, but edge-affected habitats are apparently following the
same successional trajectory owing to the proliferation of pioneer
species. In fact, Amazonian findings reinforce the notion that
edge-induced microclimatic changes can maintain suitable condi-
tions for pioneer recruitment and cycles of pioneer self-replace-
ment (i.e., multi- rather than single regeneration pioneer
assemblages), leading edge-affected habitats toward early-succes-
sional systems; i.e., a permanent shift in the successional trajec-
tory experienced by edge-affected habitats (Tabarelli et al. 2008,
Pütz et al. 2011).
Overall, our results indicate that pioneer species may bene-
fit from edge-affected habitats, at least in terms of juvenile
abundance, but they differ in their response as indicated by the
demographic structures and patterns of adult recruitment/mor-
tality we observed. In fact, the short- and long-lived pioneer
tree species analyzed share several life-history traits, such as
rapid growth, dependence on generalist vectors of pollination
and seed dispersal, and copious seed production, which have
been considered to favor their establishment/proliferation in
human-dominated Neotropical landscapes (Tabarelli et al. 2008,
Lopes et al. 2009). Laurance et al. (2006), for instance, showed
a strong positive relationship between growth rate and adult
net recruitment by using a subset of 30 successional tree spe-
cies established along forest edges in the Manaus site, suggest-
ing that growth rate is a good predictor for the early
establishment success of pioneer tree species in recently frag-
mented forests. Both long-lived species examined here, however,
as well as one short-lived in a particular site, exhibited edge
populations that clearly deviated toward non-reproductive indi-
viduals with a negative adult net recruitment. One potential
process explains this recruitment bottleneck of large trees in
edge-affected habitats: the pervasive wind turbulence in the
upper strata of the forest, which increases mortality of large
trees by uprooting and physiological stress (Ennos 1997, Lau-
rance et al. 2000, D’Ângelo et al. 2004). In addition, this
mechanic force also suppresses the recruitment of new large
individuals, which continue inhabiting the canopy stratum, but
exhibit a depressed height/dbh stem ratio (Oliveira et al. 2008).
This abiotic filtering may prevent saplings and subadults of
long-lived pioneers from attaining the emergent layer of the
forest and becoming adults (Oliveira et al. 2008, Paula et al.
2011), resulting in the negative net adult recruitment and the
demographic bias toward smaller size classes we documented
here. In the long run, increased recruitment of juveniles trig-
gered by edge creation may not compensated for negative net
adult recruitment, making the persistence of long-lived pioneers
in edge-affected habitats dependent on source habitat as pro-
posed elsewhere (Oliveira et al. 2008).
Unlike long-lived pioneers and their more robust adults,
individuals of short-lived pioneers do not need to attain the
emergent stratum of the forest to properly reproduce (Whitmore
1989). Instead, saplings and subadults of short-lived pioneers can
take advantage of the increased number of canopy gaps along
forest edges to rapidly grow, achieve maturity, and leave descen-
dents (Kapos et al. 1993, Laurance et al. 2006). Once these short-
lived trees age and start to die, canopy disturbance increases
again, leaving light available for seed germination and seedling
recruitment along forest edges, including conspecifics (Melo &
Tabarelli 2003). This dynamic, however, may be spatially variable
as it is affected by deterministic and stochastic processes that
jointly determine local species abundance (e.g., dispersal and
recruitment limitation, Hubbell et al. 1999, presence of antago-
nists, Wirth et al. 2008). The excessive recruitment of C. sciadophy-
lla in a single 1-ha plot and the unexpected collapse of B.
grossularioides on Manaus forest edges illustrate the high degree of
spatial heterogeneity in the distribution of short-lived pioneers
along forest edges.
FIGURE 3. Net adult recruitment (mean ± SE) of Bellucia grossularioides, Cecro-
pia sciadophylla, Goupia glabra, and Laetia procera along forest edges after 23–
27 yr of fragmentation near Manaus, Central Amazonia, Brazil. SL, short-
lived; LL, long-lived.
Demographic Structure of Pioneer Tree Species 775
The ontogenetic-related response exhibited by some pioneer
species is apparently shared with other tropical tree species,
although responses may involve more than one ontogenetic stage
and opposite directions. For example, forest fragmentation in the
Manaus site reduced the density of seedlings and juveniles (5–
400 cm tall) of the tree palm Oenocarpus bacaba, possibly due to
failure in reproduction or increase in seed/seedling predation.
However, O. bacaba recruitment into size class 
 10 cm was sig-
nificantly greater in forest edges than interiors, indicating that su-
badults established before forest isolation were favored in forest
fragments (Brum et al. 2008). It is likely that O. bacaba may be
extirpated from the forest fragments in the long-term, but not
owing to the suppression of adult recruitment. We can now sug-
gest the inclusion of lifespan to the list of life-history traits that
confer tree species sensitiveness to human disturbances (Tabarelli
et al. 2004, Lopes et al. 2009, see also Ewers & Didham 2006).
A persistent decline of long-lived pioneer populations in
edge-affected habitats may trigger a cascade of effects on forest
dynamics in human-modified landscapes, e.g., a 30 percent col-
lapse on the aboveground biomass on forest edges (Paula et al.
2011). Large trees, such as the long-lived pioneers considered
here: (1) account for 37 percent of aboveground plant biomass in
Central Amazonia (Nascimento & Laurance 2002); (2) house
many endemic bromelia species in the Brazilian Atlantic forest
(Siqueira-Filho & Tabarelli 2006); and (3) provide sites for harpy
eagle nidification in Mesoamerica (Ridgely & Gwynne 1993) and
food resources for a great variety of vertebrates and invertebrates
throughout the tropics (Gribel et al. 1999, Peres 2000, Arroyo-
Rodríguez et al. 2007). Moreover, they are among the most com-
mercialized goods in the international market of wood products
(FAO [Food & Agriculture Organization of The United Nations]
2009), thus their decline along forest edges may significantly
reduce the economic opportunities offered by forest habitats in
human-modified landscapes.
In synthesis, by assessing the demographic structure of
short- and long-lived pioneer tree species, we were able to docu-
ment demographic responses of pioneer trees to forest edges and
explore their potential impacts on the successional trajectory of
edge-affected habitats. Nevertheless, a long-term monitoring of
long-lived pioneer populations, particularly subadult/adult individ-
uals, is required to offer conclusive evidence about the persis-
tence of this key ecological group in human-modified landscapes.
Also, a phylogenetically controlled approach considering a greater
number of pioneer species is needed to accurately examine the
role of lifespan in species persistence across altered landscapes.
ACKNOWLEDGMENTS
We thank Claudia Funi for helping to produce Figure 1 and Felipe
Melo, Paul-Camilo Zalamea, and an anonymous reviewer for valu-
able comments on previous versions of this manuscript. We are
grateful to the Biological Dynamics of Forest Fragment Project
(BDFFP) for research facilities, Conservação Internacional do Bra-
sil for graduate scholarship to GGAS, and Conselho Nacional de
Ciência e Tecnologia (CNPq) for postdoctoral fellowship to BAS
and research grant to MT.
SUPPORTING INFORMATION
Additional Supporting Information may be found in the online
version of this article:
FIGURE S1. Study area.
Please note: Wiley-Blackwell are not responsible for the con-
tent or functionality of any supporting materials supplied by the
authors. Any queries (other than missing material) should be
directed to the corresponding author for the article.
LITERATURE CITED
ARROYO-RODRÍGUEZ, V., S. MANDUJANO, J. BENÍTEZ-MALVIDO, AND C. CUENDE-
FANTON. 2007. The influence of large tree density on howler monkey
(Alouatta palliata mexicana) presence in very small rain forest fragments.
Biotropica 39: 760–766.
BENTOS, T. V., R. C. G. MESQUITA, AND G. B. WILLIAMSON. 2008. Reproductive
phenology of Central Amazon pioneer trees. Trop. Conserv. Sci. 1:
186–203.
BOROUNCLE, C., AND B. FINEGAN. 2011. Tree regeneration and understory
woody plants show diverse responses to forest–pasture edges in Costa
Rica. Biotropica 43: 562–571.
BROADBENT, E. N., G. P. ASNER, M. KELLER, D. E. KNAPP, P. J. C. OLIVEIRA,
AND J. N. SILVA. 2008. Forest fragmentation and edge effects from
deforestation and selective logging in the Brazilian Amazon. Biol.
Conserv. 141: 1745–1757.
BRUM, H. D., H. E. M. NASCIMENTO, W. F. LAURANCE, A.C. S. ANDRADE, S. G.
LAURANCE, AND R. C. C. LUIZÃO. 2008. Rainforest fragmentation and
demography of the economically important palm in Central Amazonia.
Plant Ecol. 199: 209–215.
BRUNA, E. M. 2002. Effects of forest fragmentation onHeliconia acuminata seedling
recruitment in Central Amazonia. Oecologia 132: 235–243.
BRUNA, E. M., AND W. J. KRESS. 2002. Habitat fragmentation and the demo-
graphic structure of an Amazonian understory herb (Heliconia acuminat-
a). Conserv. Biol. 16: 1256–1266.
BRUNA, E. M., AND M. K. OLI. 2005. Demographic effects of habitat frag-
mentation on a tropical herb: Life-table response experiments. Ecology
86: 1816–1824.
CLARK, D. B., AND D. A. CLARK. 1996. Abundance, growth and mortality of
very large trees in neotropical lowland rain forest. For. Ecol. Manage.
80: 235–244.
D’ÂNGELO, S. A., A. C. S. ANDRADE, S. G. LAURANCE, W. F. LAURANCE, AND R. C.
G. MESQUITA. 2004. Inferred causes of tree mortality in fragmented and
intact Amazonian forests. J. Trop. Ecol. 20: 243–246.
ENNOS, A.R. 1997. Wind as an ecological factor. Trends Ecol. Evol. 12: 108–
111.
EWERS, R. M., AND R. K. DIDHAM. 2006. Confounding factors in the detection
of species responses to habitat fragmentation. Biol. Rev. 81: 117–142.
FAO (Food and Agriculture Organization of The United Nations). 2009. State
of the World’s forest 2009. FAO, Rome, Italy.
FEARNSIDE, P. M. 2005. Deforestation in Brazilian Amazon: History, rates and
consequences. Conserv. Biol. 19: 680–688.
GAGNON, P. R., E. M. BRUNA, P. RUBIM, M. R. DARIGO, R. C. LITTELL, M. URI-
ARTE, AND W. J. KRESS. 2011. Growth of an understory herb is chroni-
cally reduced in Amazonian forest fragments. Biol. Consev. 144: 830–
835.
776 Santos, Santos, Nascimento, and Tabarelli
GRAFEN, A., AND R. HAILS. 2002. Modern statistics for the life sciences.
Oxford University Press, Oxford, U.K.
GRIBEL, R., P. E. GIBBS, AND A. L. QUEIROZ. 1999. Flowering phenology and
pollination biology of Ceiba pentandra (Bombacaceae) in Central
Amazonia. J. Trop. Ecol. 15: 247–263.
HUBBELL, S. P., R. B. FOSTER, S. T. O’BRIEN, K. E. HARMS, R. CONDIT, B.
WECHSLER, S. J. WRIGHT, AND S. L. LAO. . 1999. Light-gap distur-
bances, recruitment limitation, and tree diversity in a neotropical for-
est. Science 283: 554–557.
KAPOS, V., G. GANADE, E. MATSUI, AND R. L. VICTORIA. 1993. o13C as an
indicator of edge effects in tropical rainforest reserves. J. Ecol. 81: 425
–432.
LAURANCE, W. F., J. L. C. CAMARGO, R. C. C. LUIZÃO, S. G. LAURANCE, S. L.
PIMM, E. M. BRUNA, P. C. STOUFFER, C. B. WILLIAMSON, J. BENÍTEZ-
MALVIDO, H. L. VASCONCELOS, K. S. V. HOUTAN, C. E. ZARTMAN, S. A.
BOYLE, R. K. DIDHAM, A. ANDRADE, AND T. E. LOVEJOY. 2011. The
fate of Amazonian forest fragments: A 32-year investigation. Biol.
Conserv. 144: 56–67.
LAURANCE, W. F., AND T. J. CURRAN. 2008. Impacts of wind disturbance on
fragmented tropical forests: A review and synthesis. Austral Ecol. 33:
399–408.
LAURANCE, W. F., P. DELAMÔNICA, S. G. LAURANCE, H. L. VASCONCELOS, T. E.
LOVEJOY, AND T. E. . 2000. Rainforest fragmentation kills big trees.
Nature 404: 836.
LAURANCE, W. F., P. M. FEARNSIDE, S. G. LAURANCE, P. DELAMONICA, T. E.
LOVEJOY, J. M. RANKIN-DE-MERONA, J. Q. CHAMBERS, AND C. GASCON.
1999. Relationship between soils and Amazon forest biomass: A land-
scape-scale study. For. Ecol. Manage. 118: 127–138.
LAURANCE, W. F., L. V. FERREIRA, J. M. RANKIN-DE-MERONA, S. G. LAURANCE,
R. HUTCHINGS, AND T. E. LOVEJOY. 1998. Effects of forest fragmenta-
tion on recruitment patterns in Amazonian tree communities. Conserv.
Biol. 12: 460–464.
LAURANCE, W. F., H. E. M. NASCIMENTO, S. G. LAURANCE, A. C. ANDRADE, P.
M. FEARNSIDE, J. E. L. RIBEIRO, AND R. L. CAPRETZ. 2006. Rain forest
fragmentation and the proliferation of successional trees. Ecology 87:
469–482.
LAURANCE, W. F., AND C. A. PERES. 2006. Emerging threats to tropical forests.
University of Chicago Press, Chicago, Illinois.
LÔBO, D., LEÃO, T., F. P. L. MELO, A. M. M. SANTOS, AND M. TABARELLI.
2011. Forest fragmentation drives Atlantic forest of northeastern Bra-
zil to biotic homogenization. Divers. Distrib. 17: 287–296.
LOPES, A. V., L. C. GIRÃO, B. A. SANTOS, C. A. PERES, AND M. TABARELLI. 2009.
Long-term erosion of tree reproductive trait diversity in edge-dominated
Atlantic forest fragments. Biol. Conserv. 142: 1154–1165.
MELO, F. P. L., AND M. TABARELLI. 2003. Seed dispersal and demography of pio-
neer trees: The case of Hortia arborea. Plant Biol. 5: 359–365.
MESQUITA, R. C. G., P. DELAMONICA, AND W. F. LAURANCE. 1999. Effects of
matrix type on edge-related tree mortality in Amazonian forest frag-
ments. Biol. Conserv. 91: 129–134.
MICHALSKI, F., I. NISHI, AND C. A. PERES. 2007. Disturbance-mediated drift in
tree functional groups in Amazonian forest fragments. Biotropica 39:
691–701.
NASCIMENTO, H. E. M., A. ANDRADE, J. L. C. CAMARGO, W. F. LAURANCE, S.
G. LAURANCE, AND J. E. L. RIBEIRO. 2006. Effects of the surrounding
matrix on tree recruitment in Amazonian forest fragments. Conserv.
Biol. 20: 853–860.
NASCIMENTO, H. E. M., AND W. F. LAURANCE. 2002. Total aboveground bio-
mass in central Amazonian rainforests: A landscape-scale study. For.
Ecol. Manage. 168: 311–321.
OLIVEIRA, A. O., AND S. A. MORI. 1999. A central Amazonian terra firme for-
est. I. High tree species richness on poor soils. Biodivers. Conserv. 8:
1219–1244.
OLIVEIRA, M. A., A. M. M. SANTOS, AND M. TABARELLI. 2008. Profound
impoverishment of the large-tree stand in a hyper-fragmented land-
scape of the Atlantic forest. For. Ecol. Manage. 256: 1910–1917.
PAULA, M. D., C. P. ALVES-COSTA, AND M. TABARELLI. 2011. Carbon storage in
a fragmented landscape of Atlantic forest: The role played by edge-
affected habitats and emergent trees. Trop. Conserv. Sci. 4: 349–358.
PERES, C. A. 2000. Identifying keystone species plant resources in tropical for-
ests: The case of gums from Parkia pods. J. Trop. Ecol. 16: 287–317.
PÜTZ, S., J. GROENEVELD, L. F. ALVES, J. P. METZGER, AND A. HUTH. 2011.
Fragmentation drives tropical forest fragments to early successional
states: A modelling study for Brazilian Atlantic forests. Ecol. Model.
222: 1986–1997.
RANKIN-DE-MERONA, J. M., R. W. HUTCHINGS, AND T. E. LOVEJOY. 1990. Tree
mortality and recruitment over a five-year period in undisturbed
upland rain forest of the central Amazon. In A. H. Gentry (Ed.). Four
neotropical rainforests, pp. 573–584. Yale University Press, New
Haven, Connecticut.
RANTA, P., T. BLOM, J. NIEMELA, E. JOENSUU, AND M. SIITONEN. 1998. The
fragmented Atlantic rain forest of Brazil: Size, shape and distribution
of forest fragments. Biodivers. Conserv. 7: 385–403.
RIBEIRO, M.C., J. P. METZGER, A. C. MARTENSEN, F. J. PONZONI, AND M. M.
HIROTA. 2009. The Brazilian Atlantic forest: How much is left, and
how is the remaining forest distributed? Implications for conservation
Biol. Conserv. 142: 1141–1153.
RIDGELY, R. S., AND J. A. GWYNNE Jr. 1993. Guía de las aves de Panamá; in-
cluyendo Costa Rica, Nicaragua y Honduras. University of Princeton
Press, Princeton, New Jersey.
SANTOS, B. A., V. ARROYO-RODRÍGUEZ, C. E. MORENO, AND M. TABARELLI.
2010. Edge-related loss of tree phylogenetic diversity in the severely
fragmented Brazilian Atlantic forest. PLoS ONE 5: e12625.
SANTOS, B. A., C. A. PERES, M. A. OLIVEIRA, A. GRILLO, C. P. ALVES-COSTA,
AND M. TABARELLI. 2008. Drastic erosion in functional attributes of
tree assemblages in Atlantic forest fragments of northeastern Brazil.
Biol. Conserv. 141: 249–260.
SILVA, J. M. C., A. B. RYLANDS, AND G. A. B. FONSECA. 2005. The fate of
Amazonian areas of endemism. Conserv. Biol. 19: 689–694.
SIQUEIRA-FILHO, J. A., AND M. TABARELLI. 2006. Bromeliad species from the
Atlantic forest of north-east Brazil: Losses of critical populations of
endemic species. Oryx 40: 218–224.
SIZER, N., AND E. V. J. TANNER. 1999. Responses of woody plant seedlings to
edge formation in a lowland tropical rainforest, Amazonia. Biol. Con-
serv. 91: 135–142.
SOUZA, E. B., AND A. C. CUNHA. 2010. Climatologia de Precipitação no
Amapá e Mecanismos Climáticos de Grande Escala. In A. C. Cunha,
E. B. Souza, and H. F. A. Cunha (Eds.). Tempo, clima e recursos
hídricos: resultados do Projeto REMETAP no Estado do Amapa, pp.
177–196. IEPA, Macapá, Brazil.
TABARELLI, M., A. V. AGUIAR, L. C. GIRÃO, C. A. PERES, AND A. V. LOPES. 2010a.
Effects of pioneer tree species hyperabundance on forest fragments in
northeastern Brazil. Conserv. Biol. 24: 1654–1663.
TABARELLI, M., A. V. LOPES, AND C. A. PERES. 2008. Edge-effects drive tropi-
cal forest fragments towards an early-successional system. Biotropica
40: 657–661.
TABARELLI, M, J. M. C. SILVA, AND C. GASCON. 2004. Forest fragmentation,
synergisms and the impoverishment of neotropical forests. Biodivers.
Conserv. 13: 1419–1425.
TABARELLI, M., A. V AGUIAR, M. C. RIBEIRO, J. P. METZGER, AND C. A. PERES.
2010b. Prospects for biodiversity conservation in the Atlantic Forest:
Lessons from aging human-modified landscapes. Biol. Conserv. 143:
2328–2340.
VIEIRA, I. C. G., P. M. TOLEDO, J. M. C. SILVA, AND H. HIGUCHI. 2008. Defor-
estation and threats to the biodiversity of Amazonia. Braz. J. Biol. 68:
949–956.
WHITMORE, T. C. 1989. Canopy gaps and the two major groups of forest
trees. Ecology 70: 536–538.
WILLIAMSON, G. B., AND R. C. G. MESQUITA. 2001. Effects of fire on rainfor-
est regeneration in the Amazon Basin. In R. O. Bierregaard, C.
Gascon, T. E. Lovejoy, and R. C. G. Mesquita (Eds.). Lessons from
Demographic Structure of Pioneer Tree Species 777
Amazonia: The ecology and conservation of a fragmented forest, pp.
325–334. Yale University Press, New Haven, Connecticut.
WIRTH, R., S. T. MEYER, I. R. LEAL, AND M. TABARELLI. 2008. Plant-herbivore
interactions at the forest edge. Prog. Bot. 69: 423–448.
ZALAMEA, P. C., F. MUNOZ, P. R. STEVENSON, C. E. T. PAINE, C. SARMIENTO,
D. SABATIER, AND P. HEURET. 2011. Continental-scale patterns of Cecro-
pia reproductive phenology: Evidence from herbarium specimens.
Proc. R. Soc. B 278: 2437–2445.
ZALAMEA, P.C., P. R. STEVENSON, S. MADRIÑÁN, P. M. AUBERT, AND P. HEURET.
2008. Growth pattern and age determination for Cecropia sciadophylla
(Urticaceae). Am. J. Bot. 95: 263–271.
778 Santos, Santos, Nascimento, and Tabarelli