Performance Trade-offs Driven by Morphological Plasticity Contribute to Habitat
Specialization of Bornean Tree Species
Daisy H. Dent1 and David F. R. P. Burslem
Department of Plant and Soil Science, School of Biological Sciences, University of Aberdeen, St Machar Drive, Aberdeen, AB24 3UU, UK
ABSTRACT
Growth-survival trade-offs play an important role in niche differentiation of tropical tree species in relation to light-gradient partitioning. However, the mechanisms
that determine differential species performance in response to light and soil resource availability are poorly understood. To examine responses to light and soil nutrient
availability, we grew seedlings of five tropical tree species for 12 mo at o 2 and 18 percent full sunlight and in two soil types representing natural contrasts in nutrient
availability within a lowland dipterocarp forest in North Borneo. We chose two specialists of nutrient-rich and nutrient-poor soils, respectively, and one habitat
generalist. Across all species, growth was higher in high than low light and on more nutrient rich soil. Although species differed in growth rates, the ranking of species,
in terms of growth, was consistent across the four treatments. Nutrient-rich soils improved seedling survival and increased growth of three species even under low light.
Slower-growing species increased root allocation and reduced specific leaf area (SLA) and leaf area ratio (LAR) in response to decreased nutrient supply. All species
increased LAR in response to low light. Maximum growth rates were negatively correlated with survival in the most resource-limited environment. Nutrient-poor soil
specialists had low maximum growth rates but high survival at low resource availability. Specialists of nutrient-rich soils, plus the habitat generalist, had the opposite
suite of traits. Fitness component trade-offs may be driven by both light and belowground resource availability. These trade-offs contribute to differentiation of tropical
tree species among habitats defined by edaphic variation.
Key words: Biomass allocation; Dipterocarpaceae; edaphic variation; niche partitioning; Southeast Asia.
ECOLOGICAL TRADE-OFFS may permit closely related tree species to
coexist in highly diverse tropical tree communities (Kitajima 1994;
Baraloto et al. 2005, 2006; Fine et al. 2006; Poorter & Kitajima
2007). Trade-offs occur when the traits that maximize the growth rate
or fitness of plants in one context are inappropriate for another
(Dalling & Burslem 2005). There are two main views on how trade-
offs promote the coexistence of species with different maximum
growth rates (Baraloto et al. 2006). On the one hand, traits that
sustain high maximum growth rates may limit allocation to storage
and defense, and reduce the likelihood of survival (Kitajima 1994,
Kitajima & Bolker 2003, Baraloto et al. 2005, Poorter & Kitajima
2007). Alternatively, traits that confer rapid growth rate at high re-
source availability may be disadvantageous to growth at low resource
availability, and vice versa (Sack & Grubb 2001, 2003). The first me-
chanism predicts a negative relationship between growth and survival
across species, and the second predicts rank shifts in species growth
rates from high to low resource availabilities (Baraloto et al. 2006).
Coexisting tropical tree species vary in their light requirements
for regeneration at the seedling stage (Veneklaas & Poorter 1998,
Montgomery & Chazdon 2002, Bloor & Grubb 2003). A trade-off
between rapid growth in high light and survival in the shaded for-
est understory may be an important mechanism underlying these
differential light requirements (Kitajima 1994, Sterck et al. 2006,
Poorter & Kitajima 2007). In addition, there is substantial evi-
dence of habitat partitioning of tropical tree species associated with
fine-scale edaphic variation (Harms et al. 2001, Palmiotto et al.
2004, Paoli et al. 2006, John et al. 2007). A recent study in Amazo-
nian Peru indicated that a trade-off between seedling growth rates
and survival, driven by a growth/defence allocation trade-off,
determined tree species distributions between clay and white-sand
soils, which differ markedly in nutrient availability (Fine et al. 2006).
However, uncertainties remain as to the mechanisms of species-
habitat partitioning in response to more fine-scale variation of soil
resource availability (Dalling & Burslem 2005, John et al. 2007).
Tests of the hypothesis that tropical tree species respond dif-
ferentially to nutrient supply and irradiance, resulting in niche-
separation, have largely been addressed by addition of inorganic
nutrients to seedlings growing in shade-houses that manipulate the
light environment (Burslem et al. 1995, 1996; Raaimakers &
Lambers 1996; Gunatilleke et al. 1997; Lawrence 2001). These ex-
periments have provided valuable information on the extent and
nature of nutrient limitation of tree seedling growth. However, the
addition of inorganic nutrients may elevate nutrient availability
beyond conditions that are likely to occur naturally. More recent
studies have employed natural variation in soil fertility (Veenendaal
et al. 1996; Palmiotto et al. 2004; Baraloto et al. 2005, 2006) and
soil dilution techniques (Metcalfe et al. 2002). Although these
studies have addressed seedling responses to more realistic gradients
in soil nutrient availability, they have not adequately explored
responses to combinations of nutrient availability and light. Seed-
lings were either grown in field-scale reciprocal transplant experi-
ments, where light could not be controlled independently of soil type
(Palmiotto et al. 2004, Baltzer et al. 2005, Baraloto et al. 2005), in
just one light environment (Baraloto et al. 2006), or in light treat-
ments that do not represent fully the range of light environments
that occur in natural forests (Veenendaal et al. 1996, Metcalfe et al.
2002). Therefore, there remains a necessity for experimental studies
that investigate growth rate trade-offs across environments that apply
combinations of light and soil resource availability in realistic
settings.
Received 3 March 2008; revision accepted 1 December 2008.
1Corresponding author; current address: Smithsonian Tropical Research
Institute, Unit 0948, APO AA 34002, U.S.A. e-mail: daisy.h.dent@gmail.com
BIOTROPICA 41(4): 424?434 2009 10.1111/j.1744-7429.2009.00505.x
424 r 2009 The Author(s)
Journal compilationr 2009 by The Association for Tropical Biology and Conservation
Tropical and temperate tree species differ markedly in the min-
imum irradiance at which they respond to increased nutrient avail-
ability (Latham 1992, Burslem et al. 1996, Canham et al. 1996).
Nutrient additions generally increase growth and survival rates at
4 5 percent daylight (Coomes & Grubb 2000). However, in deep
shade (o 2% daylight) increased nutrient supply has been reported
either to increase growth rates (Burslem et al. 1996) or to reduce
growth rates and survival (Coomes & Grubb 2000). A reduction
in growth and survival might occur because respiration rates are
upregulated by increased nutrient supply without any compensat-
ing increase in photosynthetic rates, thus inducing net carbon loss
(Lambers & Poorter 1992, Reich et al. 1996). Frequently, species
that are most responsive to nutrient addition, even at relatively low
irradiance, have the highest relative growth rates at moderate nutri-
ent supply (Latham 1992, Huante et al. 1995, Denslow et al. 1998;
but see Veenendaal et al. 1996).
To understand the mechanisms that underlie growth rate trade-
offs it is necessary to investigate the morphological and physiological
determinants of growth rate, which are intercorrelated. In general,
species that are capable of attaining high maximum growth rates
have high values of specific leaf area (SLA), leaf area ratio (LAR), leaf
mass ratio (LMR) and maximum rate of photosynthesis, and low
root mass ratio (RMR) (Reich et al. 1997, Poorter 2005). Plasticity
in these traits may provide seedlings with a competitive advantage in
environments with substantial temporal and spatial variability of
essential resources (Rice & Bazzaz 1989, Bloor & Grubb 2003).
High plasticity in these traits in response to variation in light is
characteristic of fast-growing woody tropical species (Poorter 1999,
Valladares et al. 2000). However, a recent study in French Guiana
found a negative relationship between species phenotypic plasticity
and maximum growth rates in comparisons across a soil fertility
gradient (Baraloto et al. 2006). Therefore, differences in phenotypic
plasticity between species may be determined by the resource that
is most limiting to growth.
In this study, we grew seedlings of five tropical tree seedlings
under controlled conditions of light and soil resource availability to
address the following questions: (1) In terms of growth, survival,
and biomass allocation, do species respond in a similar way to
variation in light availability as they do to nutrients? (2) Do species
rankings, in terms of growth, change with altered resource avail-
ability? (3) Are high maximum growth rates associated with low
survival? (4) Do fast-growing species exhibit greater plasticity in the
morphological components of growth than slow-growing species?
METHODS
STUDY SITE.?The study site was Sepilok Forest Reserve (51100 N,
1171560 E), hereafter SFR, which is on the east coast of Sabah,
Malaysia. The reserve is a 4475 ha patch of lowland dipterocarp
and heath forest at 0?170 m asl (Fox 1973, DeWalt et al. 2006).
Mean annual rainfall during 1976?1995 was 2975 mm, with no
month receiving o 100 mm on average (Malaysian Meteorological
Department, unpublished data). However, through the year there is
distinct variation in rainfall distribution; April is generally the driest
month and December and January the wettest with 45 percent of the
annual precipitation falling from early November to mid-February
(Fox 1973).
SFR supports two lowland dipterocarp forest communities
(alluvial forest and sandstone hill forest) that differ significantly in
their species composition and occur in association with changes in
the underlying soil and geological substrate (Fox 1973, Nilus
2004). Alluvial forest occurs on ultisols overlying alluvial flats and
gently sloping, low mudstone and sandstone hills (Fox 1973), while
the sandstone hill forest occurs on welldrained ultisols and lithosols
on steeply sloping sandstone ridges and valleys interbedded with
mudstone (for further details see Fox 1973, Baltzer et al. 2005,
Dent et al. 2006, DeWalt et al. 2006). Tree species diversity and
basal area are lower in the sandstone hill forest than in the alluvial
forest, and stem density is greater (Nilus 2004).
Alluvial forest soils have significantly greater concentrations
of total P and N than the sandstone hill forest soils (Table 1; see
also Dent 2004, Baltzer et al. 2005, Dent et al. 2006). Concen-
trations of nitrate and the base cations are also significantly greater
in alluvial than sandstone soils, while pH and concentrations of
phosphate, ammonium and Al do not differ. Monthly sampling of
surface soil matric potential and gravimetric water content over one
year revealed no consistent difference in matric potential between
alluvial and sandstone soils (Dent 2004). However, gravimetric soil
water content and soil water storage were significantly lower in
the welldrained sandstone derived soils than in alluvial soils (Dent
2004). The alluvial forest understory has lower mean irradiance
than the sandstone forest understory (2.81  0.15 vs. 3.21  0.19
mol/m2/day respectively), which suggests that canopy structure and
light environments differ (Baltzer et al. 2005).
STUDY SPECIES.?Five species were selected for the study: two species
native to alluvial forest (Dryobalanops lanceolata Burck and Shorea
leprosula Miq.), two species found exclusively on sandstone soils
(Hopea beccariana Burck and Shorea multiflora (Burck) Sym.), and
one ubiquitous species (Shorea smithiana Sym.). Dryobalanops lance-
olata is a shade-tolerant species that is widespread on fertile soils in
north Borneo (Zipperlen & Press 1996, Bungard et al. 2002). Shorea
leprosula is a common and widespread emergent dipterocarp species
in Malaysian lowland forests. It is a relatively light-demanding
nonpioneer species and exhibits rapid growth rates (Zipperlen &
Press 1996, 1997). Shorea multiflora and H. beccariana are com-
mon in the sandstone hills and ridges of SFR (Nicholson 1965,
Burgess 1966). Shorea multiflora is reported as having shade-tolerant
seedlings (Turner 1990). Shorea smithiana is one of only two dipte-
rocarp species that has approximately equal abundance in both the
alluvial and sandstone hill forest areas of SFR. It can grow to a very
large size, is one of the most common Shorea species in Borneo, and
is frequent on undulating land (Meijer & Wood 1964). There is no
published information on the shade-tolerance of H. beccariana and
S. smithiana, but seedlings of both species persist in small numbers in
the shaded understory at our study site (D. F. R. P. Burslem, pers.
obs.).
Habitat Specialization of Tropical Tree Species 425
GROWTH CONDITIONS.?Seedlings of D. lanceolata, S. leprosula,
and S. smithiana were grown from seed in the nursery of the For-
est Research Centre, Sabah, adjacent to SFR. Seeds were collected
during January 2002 from primary forest in SFR. Seedlings were
grown initially in cylindrical polyethylene bags containing 100 ml
of clay-rich forest soil. Seedlings of H. beccariana and S. multiflora
were collected from the understory of sandstone hill forest in SFR,
during 22?26 January 2002, and were transplanted immediately
into 100 ml polyethylene bags containing sandstone hill soil. The
seedlings were stored for 3?4 weeks in a shade house transmit-
ting approximately 18 percent of full daylight prior to their use in
the experiment. At the time of transplantation the mean height of
seedlings was 20 cm, 11 cm, 14 cm, 11 cm, and 10 cm for D. lance-
olata, S. leprosula, S. smithiana, H. beccariana, and S. multiflora,
respectively. Mean leaf number was six for D. lanceolata, nine for H.
beccariana, and three for S. leprosula, S. smithiana, and S. multiflora.
Seedlings of each species were transplanted into 1.2 l plastic
pots containing either alluvial or sandstone hill forest soils in a
balanced factorial design comprising all combinations of the five
species and two soil types. Soil was collected from ca. 5?50 cm
depth to exclude soils highly enriched in organic matter close to the
soil surface. Soils were collected at three alluvial and three sand-
stone primary forest sites in SFR and passed through a 2  2 mm
sieve. On 26 March 2002, seedlings were distributed among seven
tables shaded by two layers of shade cloth (high-light environment)
and seven tables with four layers of cloth (low-light environment,
see below). On each table, plants were randomly arranged within
two blocks of 10 seedlings, with one replicate of each species by
soil type treatment per block. Hence, five species were grown in
two light environments each with seven shade tables that had two
soil types randomly arranged in two blocks (five species  two
light levels  seven tables  two soil types  two blocks = 280
plants, with 14 replicates of each species/ treatment combination).
Seedlings were grown for 12 mo and watered daily with tap
water until the soil in the pots was saturated to minimize differences
in water availability. Although differences in moisture availability
between the two soil types were not incorporated into this experi-
mental design, it is recognized that traits related to this resource
are potentially important for growth. Plants were re-randomized within
the shade tables on 26 June and 28 September 2002, and 4 January
2003.
On 12 March 2002 two photosynthetically active radiation
(PAR) sensors (sensor model SKP 215 attached to two Datahog
model 2, SDL 5000 Series, Skye Instruments Ltd., Llandrindod
Wells, UK) were placed on the floor of each of two shade tables
within each light treatment. Over the subsequent 15 d the light
sensors were moved to randomly selected shade tables to record the
light environment of five high-light and five low-light shade tables.
The same procedure was repeated from 1 April 2003 to as-
sess whether the shade cloth had degraded and to quantify the extent
of this degradation. In March 2002, the sensors recorded a mean
PAR of 8.71  0.214 mol m2/day (mean  SE) in the high-light
and 0.659  0.0241 mol m2/day in the low-light treatment. These
values correspond to 18.4  1.1 percent and 1.4  0.05 percent
PAR in full sunlight in the high-light and low-light treatments,
respectively, when compared with an adjacent unshaded sensor. In
April 2003, the sensors recorded 23.5  0.7 percent and 1.8  0.04
percent of full daylight PAR in the same two treatments. Measure-
ments of the R:FR ratio were taken in April 2003 using the same
protocol (Skye Instruments Ltd. Llandrindod Wells, UK). Sensors
recorded R: FR ratios of 0.79  0.01 in the low-light treatment and
1.14  0.01 in the high-light treatment. The two light treatments
imposed in this experiment simulate the PAR of a small gap and the
forest understory in SFR (D. H. Dent, unpublished data), but the
TABLE 1. Environmental characteristics of experimental gap and understory sites in alluvial and sandstone hill forests at SFR, Malaysia, including total daily PPFD
(mol/m2/day; Baltzer et al. 2005), yearly mean soil gravimetric water content (g/g) at 12?17 cm below the litter layer, and pH and concentrations of major
nutrients (mg/kg) at 0?5 cm below the litter layer. All values represent means ( 1 SE) based on N = 5 (PPFD measurements), N = 12 (soil water content), and
N = 10 (soil chemical analyses). Means with the same superscript are not significantly different (P 4 0.05) using Tukey?s HSD range test. For details of sampling
and analytical techniques see Dent (2004).
Alluvial Sandstone
Gap Understory Gap Understory
Total daily PPFD (mol/m2/day) 14.9 0.10 2.81 0.15 11.9 0.10 3.21 0.19
Gravimetric water content soil (g/g) 0.38 0.03a 0.28 0.02b 0.32 0.03b 0.21 0.02c
Total P 281.1 22.39a 343.7 70.50a 53.01 12.11b 74.82 14.11b
Total N 2882 247a 3306 276a 1046 218b 1517 161b
Exchangeable P 0.83 0.17a 0.96 0.22a 0.88 0.35a 2.12 0.87a
NO3 17.3 3.13a 13.1 3.82a 5.51 0.73b 5.15 1.44b
NH4 31.4 5.28a 22.7 2.67a 20.1 4.77a 23.2 3.12a
Exchangeable K 0.135 0.023a 0.125 0.022a 0.060 0.012b 0.056 0.007b
Exchangeable Ca. 0.446 0.059a 0.354 0.040a 0.137 0.021b 0.111 0.012b
Exchangeable Mg 0.148 0.050a 0.134 0.045a 0.038 0.011b 0.039 0.006b
pH 4.65 0.05a 4.67 0.17a 4.54 0.03a 4.45 0.14a
426 Dent and Burslem
R:FR in both treatments was higher than would be encountered in
these forest environments.
SEEDLING MEASUREMENTS.?After 12 mo (27?28 March 2003)
seedlings were harvested, divided into leaf, stem, and root fractions,
dried to constant mass at 601C for 48 h and weighed. Mass ratios
(mass of a plant part divided by total plant mass) were calculated for
leaves (LMR), stems and petioles (stem mass ratio, SMR), and roots
(RMR), as described by Evans (1972). Prior to drying, all leaves
(with petiole removed) were photocopied, and the area of each leaf
was then calculated by weighing the photocopies and a known area
of paper. Mean SLA (leaf area divided by total leaf dry-mass
excluding petiole) and LAR (total plant leaf area divided by total
plant dry-mass) were calculated per plant. Plasticity in relation to
soil or light was defined as the magnitude of change in allocation to
specific plant parts in response to these environmental variables and
was thus computed as the difference in the mean value of biomass
ratios between the two light environments or the two soil types. For
all seedlings, diameter at 5 cm above soil surface, seedling height
and leaf number were recorded on 25?26 March 2002, and 27?28
March 2003. Relative growth rates (RGR) of stem diameter, height,
and leaf number were calculated as follows (Evans 1972):
RGR ? logeW2  logeW1=t2  t1;
where W2 and W1 are final and initial growth measurements and
t2  t1 was 12 mo.
STATISTICAL ANALYSIS.?To examine variation in survival we derived a
linear mixed-effects model with binomial errors using the glmmPQL
function, MASS library, R version 2.5.1. Light, species, and soil type
were treated as fixed factors, and block nested within light as a
random factor (Sokal & Rohlf 1995). It should be noted that the
number of replicates per treatment combination (14) provides
limited statistical power for the detection of treatment effects and
interactions in the survival analysis. Prior to analysis propor-
tional values of leaf, stem and root dry-mass were arcsine trans-
formed and total dry-mass and relative growth rates were log trans-
formed if residuals were not normally distributed (Sokal & Rohlf
1995). Relative growth rate, final dry-mass, and dry-mass alloca-
tion ratios were analyzed using a linear mixed effects model, with
light, species, and soil type as fixed factors and block nested within
light as a random factor. Initial seedling height was included as a
covariate for the analysis of RGRheight, and initial diameter was used
as a covariate for all other variables. These analyses indicated that
RGRdiameter and RGRheight responded similarly to the effects of soil,
light, and species (Table 2). Therefore RGRdiameter is presented as
a surrogate for relative growth rates in this paper and RGRmax is
defined as RGRdiameter on alluvial soil in high light. Multiple com-
parisons among means were made using Tukey?s honest significant
difference tests with the error rate corrected to 0.05. The signifi-
cance of differences within species was tested using analogous mixed
models to those described above. Pearson correlations were fitted to
describe the trends of association between growth components and
both RGRmax and survival, and to test the statistical significance of
those relationships (Sokal & Rohlf 1995). All statistical analyses
were conducted using R 2.5.1 (The R Foundation for Statistical
Computing, 2006).
RESULTS
SEEDLING GROWTH AND SURVIVAL.?Species, soil type, and light
environment all significantly affected measures of relative growth
rate and final dry-mass (Table 2). Across all species, relative growth
rates were higher in high than low light and on alluvial rather than
sandstone soil (Table 2; Fig. 1). Growth rates varied significantly
among species but the ranking of species, in terms of growth rate,
did not change with soil type (i.e., there was no significant species
 soil type interaction; Table 2). In contrast, there was evidence
of an interaction between species and light environment for both
TABLE 2. Summary of ANOVA for seedling growth rates and morphological traits, and analysis of deviance for seedling survival. df, F-statistics, and degrees of significance
are reported. Growth rate data are calculated over 1 yr and all other traits were measured at the final harvest of five dipterocarp species planted into pots of
either alluvial or sandstone derived soil, in either high (18% PAR) or low-light (o 2% PAR) shade house environments. Prior to analysis, relative growth rates,
total dry-mass, and SLA were log-transformed and mass ratios were arc-sine transformed. Degrees of significance: P o 0.001; P o 0.01; P o 0.05.
Factor df RGRheight RGRdiameter Dry-mass LMR SMR RMR SLA Survival
Initial seedling size 1 34.9 140 3.82 3.79 5.21 0.36 2.30 1.71
Soil 1 12.6 11.0 55.6 15.3 1.52 14.3 1.78 4.42
Light 1 8.11 9.98 26.0 46.6 1.49 36.6 212 1.45
Species 4 72.3 71.6 163 4.33 7.81 6.82 32.8 0.96
Soil  light 1 0.07 2.05 3.22 1.02 0.01 0.70 0.09 0.01
Soil  species 4 1.21 0.69 6.89 4.17 2.76 7.42 0.86 0.43
Light  species 4 3.03 4.07 5.81 1.52 1.39 2.38 1.74 0.32
Soil  light  species 4 0.80 0.98 1.51 1.15 0.72 1.07 1.34 0.87
Block (light) 12 1.23 2.12 1.97 1.15 1.65 1.48 0.60 ?
Model R2 0.66 0.65 0.88 0.38 0.43 0.60 0.77 ?
Habitat Specialization of Tropical Tree Species 427
RGRdiameter and RGRheight (Table 2). In all environments, S. leprosula
and S. smithiana grew significantly faster than the three other species,
but growth rates within these two groupings did not differ for
RGRheight (Fig. 1). Seedling survival was high in all treatments but
was significantly higher in alluvial (97%) than sandstone soil (87%),
and did not vary significantly with irradiance or species (Table 2).
Survival was 95 percent in the high light treatment and 89 percent in
low light.
Species differed in their dry-mass growth in response to soil
type and light environment (Table 2). Mean seedling dry-mass was
significantly higher at high than at low light availability for all
species except H. beccariana and greater in alluvial than sandstone
soil for all species except S. smithiana. The relative effect of soil type
on final dry-mass was greater than the effect of light environment
for the sandstone specialist species, S. multiflora and H. beccariana
(Table 3). In contrast, dry-mass of seedlings of the alluvial specialist
D. lanceolata was more strongly affected by light than soil type,
and growth of soil generalist S. smithiana was affected only by
light. The nature of the interaction between soil-type and light-
environment on dry-mass varied among species: in S. leprosula and
S. multiflora the increase in dry-mass in response to growth in
alluvial soil was greater at high light than low light, while there was
FIGURE 1. The effects of light environment and soil type on RGRheight and RGRdiameter of seedlings of five dipterocarp species planted into pots of either alluvial or
sandstone derived soil, in either high (18% PAR) or low light (o 2% PAR) shade house environments. All bars represent means ( SE; N = 14). Within each
treatment group species sharing the same superscript letter are not significantly different, P 4 0.05, Tukey?s HSD test.
428 Dent and Burslem
no such interaction for the other three species. In sandstone soils
only species native to alluvial forest responded to the higher light
environment by increasing seedling dry-mass (Table 3).
MORPHOLOGICAL TRAIT VARIATION IN RESPONSE TO ENVIRONMENTAL
VARIATION.?Species differed in their dry-mass allocation in
response to soil type and light environment (Table 3). Seedlings
of all species except S. multiflora decreased dry-mass allocation to
leaves, by 28?43 percent of the low light value, in high light com-
pared to low light (Table 3). For all species, values of SLA and LAR
increased significantly at lower light availability. Only seedlings of
the two sandstone specialists exhibited plasticity in mass allocation
in response to soil type by increasing RMR when grown in sand-
stone soil, relative to their values in alluvial soil, and this came at the
TABLE 3. The effects of light environment and soil type on mean final dry-mass (g ), leaf (LMR), stem (SMR) and root mass ratios (RMR), specific leaf area (SLA, cm2/g ) and
leaf area ratio (LAR, cm2/g ) for seedlings of five dipterocarp species planted into pots of either alluvial (A) or sandstone (SS) derived soil, in either high or low light
shade house environments (F and P values from ANOVA presented by variable). Species soil affinity is indicated after the species name. Degrees of significance: NS
= not significant; P o 0.001; P o 0.01; P o 0.05. ( N = 14).
Light treatment
High light Low light
Soil Light Soil  light
Soil type A SS A SS F F F
Dryobalanops lanceolata?Alluvial
Final dry-mass 14.8a 11.4ab 8.33b 7.57b 5.35 20.4 NS
LMR 0.290a 0.296a 0.470b 0.381ab NS 16.8 5.22
SMR 0.313ab 0.346a 0.257b 0.302ab 4.81 4.61 NS
RMR 0.397a 0.359a 0.273b 0.317ab NS 12.5 4.76
SLA 128a 109a 155b 161b NS 27.0 4.15
LAR 36.7a 31.6a 72.5b 60.1b 4.05 17.9 NS
Shorea leprosula?Alluvial
Final dry-mass 10.9a 5.19b 5.79b 3.07b 37.1 13.1 5.26
LMR 0.273a 0.290a 0.415b 0.392b NS 14.7 NS
SMR 0.293a 0.342b 0.303a 0.320a 5.28 NS NS
RMR 0.435a 0.368ab 0.282b 0.288 NS 12.2 NS
SLA 154a 170ab 196b 205b NS 27.2 NS
LAR 46.2a 50.0a 82.9b 79.4b NS 17.2 NS
Shorea smithiana?Generalist
Final dry-mass 13.6a 15.6a 7.00b 6.25b NS 44.1 NS
LMR 0.286a 0.291a 0.359ab 0.422b NS 9.4 NS
SMR 0.292a 0.301a 0.328a 0.285a NS NS NS
RMR 0.422a 0.408a 0.313b 0.293b NS 16.3 NS
SLA 131a 115a 192b 182b NS 52.1 NS
LAR 37.2a 33.9a 69.6b 76.2b NS 11.8 NS
Hopea beccariana ?Sandstone
Final dry-mass 0.753a 0.469b 0.971a 0.430b 15.8 NS NS
LMR 0.244a 0.193a 0.470b 0.295a 8.8  30.6 NS
SMR 0.282a 0.289a 0.249a 0.256a NS NS NS
RMR 0.474a 0.518a 0.280b 0.448a 14.3 23.1 NS
SLA 160a 148a 222b 192a NS 19.8 NS
LAR 40.1a 24.2a 104b 56.3a 10.6  44.4 NS
Shorea multiflora?Sandstone
Final dry-mass 3.50a 0.950b 1.44b 0.775b 33.4 5.15 7.04
LMR 0.460a 0.282b 0.474a 0.370ab 13.3 NS NS
SMR 0.223a 0.246a 0.191a 0.222a NS NS NS
RMR 0.317a 0.471b 0.335a 0.409ab 11.1 NS NS
SLA 105a 104a 131b 128b NS 17.1 NS
LAR 48.4ab 32.5a 61.9b 47.4ab 7.94 8.15 NS
Mean values sharing the same superscript letter within a row are not significantly different, P 4 0.05, Tukey?s HSD test.
Habitat Specialization of Tropical Tree Species 429
expense of a lower LMR. Seedlings of H. beccariana decreased LMR
and LAR by 29 and 40 percent, respectively and increased RMR by
27 percent on sandstone compared to alluvial soil, while seedlings
of S. multiflora decreased both LMR and LAR by 32 percent and
increased RMR by 26 percent (all percentages are relative to values
on the alluvial soil). Although these changes were evident in both
light environments, they were less pronounced at low light (o 2%
PAR), where seedlings also increased allocation to leaves and
thereby compromised proportional increases in root mass. The
SMR was generally consistent across treatments, and so the
increased values of LMR in the low light environment resulted in
decreased RMR.
PLASTICITY OF MORPHOLOGICAL TRAITS AND ECOLOGICAL TRADE-
OFFS.?Interspecific patterns of plasticity for dry-mass allocation
to roots, leaves, and total leaf area in response to soil type were
correlated with RGRmax (defined as RGRdiameter on alluvial soil in
high light), but plasticity in morphological traits in response to
light were not correlated with RGRmax (Table 4; Fig. 2A, B).
Species with low maximum growth rates had a greater relative
increase in mass allocation to roots and a greater relative decrease in
allocation to leaf area in response to a reduction in nutrient
availability than fast-growing species. There was no evidence to
suggest that fast-growing species exhibited higher plasticity in
morphological traits than slow-growing species. In cross species
comparisons, RGRmax exhibited a negative relationship with seed-
ling survival in sandstone soil in low light (Fig. 2D). Although this
relationship was marginally nonsignificant (r =  0.84, P = 0.073),
a negative relationship between maximum RGRheight and survival in
sandstone soil in low light was also observed (r =  0.88, P = 0.045,
data not shown). RGRmax did not show any relationship with
seedling survival in any other environment. There was a nonsignifi-
cant negative relationship between plasticity in LAR in relation to
soil type and seedling survival in sandstone soil in low light
(r =  0.71, P = 0.181; Fig. 2C).
DISCUSSION
SEEDLING GROWTH AND SURVIVAL IN RELATION TO LIGHT AND SOIL
RESOURCE AVAILABILITY.?Seedling growth and survival were en-
hanced in alluvial forest soil. The difference between soil type
treatments must reflect the greater concentrations of most nutrients
in alluvial forest soil (Table 1), because differences in soil water
availability would have been eliminated by daily watering. The
faster growth rates and improved survival of seedlings in the alluvial
forest soil provides evidence that variation in soil nutrient avail-
ability at the habitat scale has the potential to influence dipterocarp
seedling performance and may contribute to the species habitat
partitioning observed at this site (Nilus 2004, DeWalt et al. 2006).
Responses to increases in the availability of light and nutrients
support other recent studies. Seedling survival and growth rate both
increased with light availability, except in the case of H. beccariana,
which may have a whole-plant light compensation point for growth
that is lower than the low light treatment in this experiment (i.e., o
2 percent PAR). This study adds to recent work demonstrating that
natural variability in soil nutrient availability may affect tropical tree
seedling growth and survival (Palmiotto et al. 2004; Baraloto et al.
2005, 2006; Fine et al. 2006). In our study, responses in terms of
relative growth rate were estimated from nondestructive measures of
aboveground growth. However, these measures may not correlate
directly to relative growth rate of biomass because allocation of dry-
mass changes in response to environmental changes and through
ontogeny, and species respond differentially (Table 2; Hunt 1982).
Therefore, our conclusions require confirmation by further work to
quantify relative growth rates of biomass from sequential harvests.
Higher nutrient availability increased final dry-mass of four
species (D. lanceolata, S. leprosula, H. beccariana, and S. multiflora)
and increased survival for all five species combined, even in low
light environments. Nutrients stimulated aboveground growth by
increasing LAR without otherwise impacting dry-mass allocation
(in D. lanceolata), or by increasing proportional allocation of dry-
mass to leaves at the expense of roots, as well as increasing SLA (in
H. beccariana and S. multiflora). The positive response to nutrient
supply in low light conditions shown here for dipterocarps is in
marked contrast to the responses of seedlings of other nonpioneer
tropical trees, which typically show a growth response to nutrient
addition under conditions equivalent to our high-light treatment,
but not at low irradiance (e.g., Thompson et al. 1992, Huante et al.
1998). In a study conducted in forest soils from Singapore, Burslem
et al. (1996) found that seedlings of two tree species were limited by
one or more macro-nutrients when grown at o 1 percent daylight.
These findings suggest that seedlings growing in the shaded
understory of lowland dipterocarp forests in SE Asia may be limited
by nutrients, as well as light. In our study, seedling survival was
unaffected by reduced irradiance, whereas seedling mortality is
typically higher in deep shade. However, increased mortality in
forest understory sites may result from factors other than light
limitation, such as herbivory and branchfalls (Myers & Kitajima
2007), and these biotic factors were not present in our study.
Residual effects of the light environment in which seedlings
established or were stored prior to the experiment (18% of full
TABLE 4. Pearson correlation coefficients between RGRdiameter on alluvial soil
in high light and indices of plasticity in five plant traits in response
to soil type and light environment variation. The indices of plas-
ticity are presented for leaf mass (LMR), root mass (RMR), stem
mass (SMR), and leaf area ratios (LAR) and specific leaf area (SLA;
N = 5).
Soil type plasticity R P Light plasticity r P
LMR 0.89 0.044 LMR  0.10 0.875
RMR  0.90 0.036 RMR  0.33 0.588
SMR 0.39 0.523 SMR 0.35 0.563
SLA 0.77 0.127 SLA 0.39 0.513
LAR 0.98 0.004 LAR  0.18 0.766
430 Dent and Burslem
daylight PAR) could have impacted seedling traits. Morphological
traits of tropical rain forest tree species take from a few weeks (SLA)
to many months (LAR) to acclimate to changes in light availability
(Popma & Bongers 1991). However, given the small size of
seedlings, the short period of storage, and the long growth period
it is likely that complete acclimation to experimental growth
conditions had taken place by the final harvest.
PERFORMANCE TRADE-OFFS.?Although seedling growth varied sig-
nificantly with changes in soil type and light availability, the species?
growth rate hierarchy was consistent across the four environmental
treatments. These data do not support the hypothesis that traits con-
ferring rapid growth at high resource availability trade-off against
traits that maintain relatively fast growth even when resources are
low, resulting in shifts in species growth rankings from high to
low resource availabilities (Sack & Grubb 2001, 2003). Similar
patterns of highly consistent species growth rate rankings were ob-
served across eight different soil treatments (Baraloto et al. 2006)
and across gradients of both light and soil resources (Baraloto et al.
2005) in two recent studies of nine tropical tree seedlings in French
Guiana. However, for the Bornean dipterocarps in our study we did
find tentative evidence for a growth/survival trade-off, even over the
relatively short duration of one year. Maximum growth rate was
inversely correlated with survival in the most light and nutrient-
limited environment. This finding parallels the observation that
the traits that maximize growth in high-light trade-off against traits
that confer increased survival in the shade in studies of differential
shade tolerance (Kitajima 1994, 2002; Davies 2001; Poorter &
FIGURE 2. The relationships between (A) RMR plasticity in relation to soil type and RGRmax (defined as RGRdiameter on alluvial soil in high light; r =  0.90, P =
0.036); (B) LAR plasticity in relation to soil type and RGRmax (r = 0.98, P = 0.004); (C) LAR plasticity in relation to soil type and percentage seedling survival in
sandstone soil in low light (r =  0.71, P = 0.181); (D) RGRmax and percentage seedling survival in sandstone soil in low light (r =  0.84, P = 0.073) for five
dipterocarp species comprising two sandstone specialists (SM = Shorea multiflora; HB = Hopea beccariana), two alluvial specialists (DL = Dryobalanops lanceolata;
SL = S. leprosula), and one generalist (SS = S. smithiana).
Habitat Specialization of Tropical Tree Species 431
Kitajima 2007). However, this negative relationship was not evi-
dent in all gap versus understory comparisons, and was found only
when growth in the least resource-limited treatment was correlated
with survival in the most resource-limited treatment. This finding
suggests that fitness component trade-offs may not be driven solely
by differential light availability (Kitajima 1994) but by responses to
the availability of both light and belowground resources (Baraloto
et al. 2005). It is possible that the growth-survival trade-off was not
manifested in all high-low light comparisons because the factors
that reinforce increased seedling mortality in the forest understory,
such as pathogens, falling debris, and herbivores (Myers & Kitajima
2007), were absent from this study.
In our controlled environment experiment the growth versus
survival trade-off was manifested even though the seedlings were not
exposed to herbivory and fungal pathogens, which are often pur-
ported to underlie this trade-off and its link to patterns of adult dis-
tribution (Baraloto et al. 2005, Fine et al. 2006, Poorter & Kitajima
2007). This discrepancy suggests that, in the current study, the
mechanism underpinning the trade-off is unlikely to be connected
to defense allocation. Instead, the growth versus survival trade-off
must be determined by differential physiological tolerance to nu-
trient limitation at low light. A reciprocal transplant experiment of
five dipterocarp species grown in alluvial and sandstone forest
habitats at Sepilok suggested that invertebrate herbivory has a rel-
atively minor impact on seedling growth and survival (Eichhorn
et al. 2006), and that habitat specialist species had contrasting gas
exchange characteristics (Baltzer et al. 2005). These studies lend
support to our interpretation that the growth/survival trade-off ob-
served under controlled environment conditions has significance for
the mechanism of habitat specialization in the context of physiolo-
gical constraints and not differential growth and defense allocation.
PLASTICITY OF MORPHOLOGICAL TRAITS.?Tropical tree seedlings have
been reported to increase SLA and allocate a greater proportion
of dry-mass to leaves to maximize carbon gain and reduce carbon
losses in response to low irradiance (Veneklaas & Poorter 1998,
Poorter 1999, Bloor & Grubb 2003). Consistent with these trends,
in low light, the seedlings in our study altered their morphology to
increase leaf area for light capture and photosynthesis. All species
increased SLA and LAR in low light and seedlings of all species,
except S. multiflora, allocated a greater proportion of dry-mass to
leaves in low light. Increased allocation to leaf area in low light may
increase seedling survival, thus allowing seedling populations to
persist in the forest understory until canopy gaps, and increased
irradiance, support onward growth (Veneklaas & Poorter 1998,
Poorter 1999).
Differential plasticity of morphological traits among species
did not vary according to expectation. High plasticity of morpho-
logical traits in response to variation in light is characteristic of fast-
growing woody tropical species (Poorter 1999, Valladares et al.
2000), but we found that the proportional increase in allocation to
leaf area in the shade was approximately equal across all species.
Therefore, although morphological plasticity is important for sur-
vival in the forest understory, a lack of differentiation among the
dipterocarp species in this study suggests that plasticity in these
traits is not significant for light partitioning at the seedling stage in
this system. In contrast, seedlings with low growth rates, which did
not allocate an intrinsically greater proportion of dry-mass to roots
than faster-growing species (Aerts & Chapin 2000, but see Reich
2002), possessed a greater capacity to increase dry-mass allocation
to roots, and to reduce SLA and LAR, in response to decreased
nutrient supply. Therefore, our data do not support the hypothesis
that fast-growing species exhibit greater plasticity in morphological
traits (Poorter 1999, Valladares et al. 2000). Conversely, we found
a negative relationship between phenotypic plasticity and max-
imum growth rate in response to natural variation in soil nutrient
availability. A similar interpretation was reported by Baraloto et al.
(2006) working with seedlings of nine tropical tree species across a
soil fertility gradient in French Guiana.
Plasticity in morphological traits in response to edaphic varia-
tion may be driven by adaptations to cope with shortage of nutrients
and/or water. At our study site, the surface soil of sandstone forest
has lower concentrations of total P and N, nitrate and available
base cations (K, Ca, and Mg) than that of alluvial forest, and the
fluxes of nutrients via litterfall and litter decomposition are also
lower in the sandstone forest (Dent 2004, Dent et al. 2006). There
is also evidence of limitation by low water availability during
occasional dry spells on the summits of the welldrained sandstone
ridges, and no such evidence for the low-lying alluvial forest (Dent
2004, Dent et al. 2006). Increased dry-mass allocation to roots is
observed frequently as a response to nutrient limitation, and implies
that increased nutrient uptake may improve long-term survival on
nutrient-poor soils (Evans 1972). Similarly, reduced values of SLA
and LAR may be driven by water or nutrient limitation (Aerts &
Chapin 2000; Wright et al. 2001, 2002; Wright & Westoby 2002;
Reich et al. 2003). Lowered SLA and LAR both contribute to a
reduction in relative growth rates, which reduces turnover of carbon
and nutrients in low-nutrient environments (Aerts & Chapin 2000).
However, low SLA and LAR may also reduce the total leaf area
available for water loss and so are consistent with a strategy of water
conservation, which may be an important adaptation to survive
periods of seasonal drought. A field study of seedling physiological
traits at the SFR suggested that divergent water-use strategies may
contribute to the mechanisms underlying differences in edaphic
associations, as sandstone specialist dipterocarp species consistently
exhibited higher water-use efficiencies than alluvial specialist or
generalist species (Baltzer et al. 2005). These studies support the
perspective that both nutrient and water availability contribute to
differentiation of dipterocarp habitat associations at our study site
and elsewhere in Borneo (Palmiotto et al. 2004, Baltzer et al. 2005,
Dent et al. 2006, Paoli et al. 2006).
IMPLICATIONS FOR SPECIES PARTITIONING.?Performance trade-offs
may play an important role in the partitioning of tree species in
response to resource heterogeneity in tropical forests. Our re-
sults indicate that a trade-off exists between traits that confer
high maximum growth rates and traits that confer high survival
when resources are limiting. This finding provides support for the
hypothesis that trade-offs similar to those seen across light gra-
dients (Kitajima 1994) may contribute to species differentiation
432 Dent and Burslem
across soils of different nutrient status (Baraloto et al. 2005, 2006).
Recent studies have identified ecophysiological traits that underlie
the trade-off between growth in high light versus survival in the
forest understory, such as the ability to store nonstructural carbohy-
drates (Myers & Kitajima 2007, Poorter & Kitajima 2007). Results
from existing studies on the mechanisms that underpin growth ver-
sus survival trade-offs in relation to soil resource availability are not
consistent, possibly due to differences in experimental designs. In
this study, slower-growing species were able to withstand limited
soil resource availability by increasing allocation to roots and re-
ducing leaf area to reduce growth and transpiration rates. However,
for tree seedlings growing in French Guiana the increased survival
of slower-growing species in response to limited soil resource avail-
ability was related to greater plasticity in assimilation rates and not
plasticity in allocation to roots (Baraloto et al. 2006). Our study of
Bornean dipterocarps was not designed to address the role of plas-
ticity in assimilation rates in coping with limitations in soil resource
availability, and further research is required to address this issue.
ACKNOWLEDGMENTS
We thank the Natural Environment Research Council (studentship
to DHD) and the British Ecological Society?s Overseas Research
Programme for financial support. The Economic Planning Unit of
the Federal Government of Malaysia kindly granted permission to
conduct research in Malaysia. We also thank R. Nilus and C.
Maycock for advice and O. Johnny, R. Yudot, and B. Seligi for
their contribution to the field and laboratory work in Sepilok.
LITERATURE CITED
AERTS, R., AND F. S. III CHAPIN. 2000. The mineral nutrition of wild plants
revisited: A re-evaluation of processes and patterns. Adv. Ecol. Res. 30:
1?67.
BALTZER, J. L., S. C. THOMAS, R. NILUS, AND D. F. R. P. BURSLEM. 2005.
Edaphic specialization in tropical forest trees: Physiological correlates
and responses to reciprocal transplantation. Ecology 86: 3048?3062.
BARALOTO, C., D. E. GOLDBERG, AND D. BONAL. 2005. Performance trade-offs
among tropical tree seedlings in contrasting microhabitats. Ecology
86: 2461?2472.
BARALOTO, C., D. BONAL, AND D. E. GOLDBERG. 2006. Differential seedling
growth response to soil resource availability among nine Neotropical tree
species. J. Trop. Ecol. 22: 487?497.
BLOOR, J. M. G., AND P. J. GRUBB. 2003. Growth and mortality in high and low
light: Trends among 15 shade-tolerant tropical rain forest tree species.
J. Ecol. 91: 77?85.
BUNGARD, R. A., S. A. ZIPPERLEN, M. C. PRESS, AND J. D. SCHOLES. 2002. The
influence of nutrients on growth and photosynthesis of seedlings of two
rainforest dipterocarp species. Funct. Plant Bio. 29: 505?515.
BURGESS, P. F. 1966. Timbers of Sabah. Sabah Forestry Record: 6, Malaysia.
BURSLEM, D. F. R. P., P. J. GRUBB, AND I. M. TURNER. 1995. Responses to
nutrient addition among shade-tolerant tree seedlings of lowland
tropical rain forest in Singapore. J. Ecol. 83: 113?122.
BURSLEM, D. F. R. P., P. J. GRUBB, AND I. M. TURNER. 1996. Responses to
simulated drought and elevated nutrient supply among shade-tolerant
tree seedlings of lowland tropical forest in Singapore. Biotropica 28:
636?648.
CANHAM, C. D., A. R. BERKOWITZ, V. R. KELLY, G. M. LOVETT, S. V. OLLINGER,
AND J. SCHNURR. 1996. Biomass allocation and multiple resource
limitation in tree seedlings. Can. J. For. Res. 26: 1521?1530.
COOMES, D. A., AND P. J. GRUBB. 2000. Impacts of root competition in forests
and woodlands: A theoretical framework and review of experiments.
Ecol. Monogr. 70: 171?207.
DALLING, J. W., AND D. F. R. P. BURSLEM. 2005. Role of life-history trade-offs in
the equalization and differentiation of tropical tree species. In D. F. R. P.
Burslem, M. A. Pinard, and S. E. Hartley (Eds.). Biotic interactions in
the tropics: Their role in the maintenance of species diversity, pp.
65?88. Cambridge University Press, Cambridge, UK.
DAVIES, S. J. 2001. Tree mortality and growth in 11 sympatric Macaranga
Species in Borneo. Ecology 82: 920?932.
DENSLOW, J. S., A. M. ELLISON, AND R. E. SANFORD. 1998. Treefall gap size
effects on above- and below-ground processes in a tropical wet forest.
J. Ecol. 86: 597?609.
DENT, D. H. 2004. The mechanistic basis of habitat specialization of dipter-
ocarps: Are species differentiated by specialisation to environmental
variation? PhD Dissertation, University of Aberdeen, UK.
DENT, D. H., R. BAGCHI, N. MAJALAP, D. ROBINSON, AND D. F. R. P. BURSLEM.
2006. Nutrient fluxes via litterfall and leaf litter decomposition across a
gradient of soil nutrient supply in a lowland tropical rain forest. Plant
Soil 288: 197?215.
DEWALT, S. J., K. ICKES, R. NILUS, K. E. HARMS, AND D. F. R. P. BURSLEM. 2006.
Liana habitat associations and community structure in a Bornean
lowland tropical forest. Plant Ecol. 186: 203?216.
EICHHORN, M. P., S. G. COMPTON, AND S. E. HARTLEY. 2006. Seedling species
determines rates of leaf herbivory in a Malaysian rain forest. J. Trop.
Ecol. 22: 513?519.
EVANS, G. C. 1972. The quantitative analysis of plant growth. Blackwell
Scientific Publications, Oxford, UK.
FINE, P. V. A., Z. J. MILLER, I. MESONES, S. IRAZUZTA, H. M. APPEL, M. H. H.
STEVENS, I. SAAKSJARVI, J. C. SCHULTZ, AND P. D. COLEY. 2006. The
growth-defense trade-off and habitat specialization by plants in Amazo-
nian forests. Ecology 87: S150?S162.
FOX, J. E. D. 1973. Kabili-Sepilok forest reserve. Sabah Forest Record: 9.
Borneo Literature Bureau, Malaysia.
GUNATILLEKE, C. V. S., I. A. U. N. GUNATILLEKE, G. A. D. PERERA, D. F. R. P.
BURSLEM, P. M. S. ASHTON, AND P. S. ASHTON. 1997. Responses to
nutrient addition among seedlings of eight closely-related species of
Shorea in Sri Lanka. J. Ecol. 85: 301?311.
HARMS, K. E., R. CONDIT, S. P. HUBBELL, AND R. B. FOSTER. 2001. Habitat
associations of trees and shrubs in a 50-ha Neotropical forest plot. J.
Ecol. 89: 947?959.
HUANTE, P., E. RINCON, AND I. ACOSTA. 1995. Nutrient availability and growth
rate of 34 woody species from a tropical deciduous forest in Mexico.
Funct. Ecol. 9: 849?858.
HUANTE, P., E. RINCON, AND F. S. CHAPIN. 1998. Foraging for nutrients,
responses to changes in light, and competition in tropical deciduous tree
seedlings. Oecologia 117: 209?216.
HUNT, R. 1982. Plant growth curves: The functional approach to plant growth
analysis. Edward Arnold, London, UK.
JOHN, R., J. W. DALLING, K. E. HARMS, J. B. YAVITT, R. F. STALLARD, M.
MIRABELLO, S. P. HUBBELL, R. VALENCIA, H. NAVARRETE, M. VALLEJO,
AND R. B. FOSTER. 2007. Soil nutrients influence spatial distributions of
tropical tree species. Proc. Natl. Acad. Sci. USA 104: 864?869.
KITAJIMA, K. 1994. Relative importance of photosynthetic and allocation traits as
correlates of seedling shade tolerance of 15 tropical tree species.
Oecologia 98: 419?428.
KITAJIMA, K. 2002. Do shade-tolerant tropical tree seedlings depend longer on
seed reserves? Functional growth analysis of three Bignoniaceae species.
Funct. Ecol. 16: 433?444.
KITAJIMA, K., AND B. M. BOLKER. 2003. Testing performance rank reversals
among coexisting species: Crossover point irradiance analysis by Sack &
Grubb 2001 and alternatives. Funct. Ecol. 17: 276?281.
Habitat Specialization of Tropical Tree Species 433
LAMBERS, H., AND H. POORTER. 1992. Inherent variation in growth-rate between
higher-plants?a search for physiological causes and ecological conse-
quences. Adv. Ecol. Res. 23: 187?261.
LATHAM, R. E. 1992. Co-occurring tree species change rank in seedling perfor-
mance with resources varied experimentally. Ecology 73: 2129?2144.
LAWRENCE, D. 2001. Nitrogen and phosphorus enhance growth and luxury
consumption of four secondary forest tree species in Borneo. J. Trop.
Ecol. 17: 859?869.
MEIJER, W., AND G. H. S. WOOD. 1964. Dipterocarps of Sabah (North Borneo).
Sabah Forestry Record: 5, Malaysia.
METCALFE, D. J., P. J. GRUBB, AND S. S. METCALFE. 2002. Soil dilution as a
surrogate for root competition: Effects on growth of seedlings of
Australian tropical rainforest trees. Funct. Ecol. 16: 223?231.
MONTGOMERY, R. A., AND R. L. CHAZDON. 2002. Light gradient partitioning by
tropical tree seedlings in the absence of canopy gaps. Oecologia 131:
165?174.
MYERS, J. A., AND K. KITAJIMA. 2007. Carbohydrate storage enhances seedling
shade and stress tolerance in a Neotropical forest. J. Ecol. 95: 383?395.
NICHOLSON, D. I. 1965. Proceedings of the symposium for ecological research in
humid tropics vegetation, Kuching. pp. 67?87. UNESCO, Paris,
France.
NILUS, R. 2004. Effect of edaphic variation on forest structure, dynamics,
diversity and regeneration in a lowland tropical rain forest in Borneo.
PhD dissertation, University of Aberdeen, UK.
PAOLI, G. D., L. M. CURRAN, AND D. R. ZAK. 2006. Soil nutrients and beta
diversity in the Bornean Dipterocarpaceae: Evidence for niche partition-
ing by tropical rain forest trees. J. Ecol. 94: 157?170.
PALMIOTTO, P. A., S. J. DAVIES, K. A. VOGT, M. S. ASHTON, D. J. VOGT, AND
P. S. ASHTON. 2004. Soil-related habitat specialization in dipterocarp
rain forest tree species in Borneo. J. Ecol. 92: 609?623.
POORTER, L. 1999. Growth responses of 15 rain-forest tree species to a light
gradient: The relative importance of morphological and physiological
traits. Funct. Ecol. 13: 396?410.
POORTER, L. 2005. Resource capture and use by tropical forest tree seedlings and
their consequences for competition. In D. F. R. P. Burslem, M. A.
Pinard and S. E. Hartley (Eds.). Biotic interactions in the tropics: Their
role in the maintenance of species diversity, pp. 35?64. Cambridge
University Press, Cambridge, UK.
POORTER, L., AND K. KITAJIMA. 2007. Carbohydrate storage and light require-
ments of tropical moist and dry forest tree species. Ecology 88:
1000?1011.
POPMA, J., AND F. BONGERS. 1991. Acclimation of seedlings of 3 Mexican
tropical rain-forest tree species to a change in light availability. J. Trop.
Ecol. 7: 85?97.
RAAIMAKERS, D., AND H. LAMBERS. 1996. Response to phosphorus supply of
tropical tree seedlings: A comparison between a pioneer species Tapirira
obtusa and a climax species Lecythis corrugata. New Phytol. 132: 97?102.
REICH, P. B. 2002. Root-shoot relations: Optimality in acclimation and
adaptation or the ??Emporor?s New Clothes??? In Y. Waisel, A. Eshel
and U. Kafkafi, (Eds.). Plant roots: The hidden half, 3rd Edition, pp.
205?220. Marcel Dekker Inc., New York, New York.
REICH, P. B., J. OLEKSYN, AND M. G. TJOELKER. 1996. Needle respiration and
nitrogen concentration in Scots Pine populations from a broad latitu-
dinal range: A common garden test with field-grown trees. Funct. Ecol.
6: 768?776.
REICH, P. B., I. J. WRIGHT, J. CAVENDER-BARES, J. M. CRAINE, J. OLEKSYN, M.
WESTOBY, AND M. B. WALTERS. 2003. The evolution of plant functional
variation: Traits, spectra, and strategies. Int. J. Plant Sci. 164: S143?
S164.
REICH, P. B., M. B. WALTERS, AND D. S. ELLSWORTH. 1997. From tropics to
tundra: Global convergence in plant functioning. Proc. Natl. Acad. Sci.
USA 94: 13730?13734.
RICE, S. A., AND F. A. BAZZAZ. 1989. Growth consequences of plasticity
of plant traits in response to light conditions. Oecologia 78: 508?
512.
SACK, L., AND P. J. GRUBB. 2001. Why do species of woody seedlings change
rank in relative growth rate between low and high irradiance? Funct.
Ecol. 15: 145?154.
SACK, L., AND P. J. GRUBB. 2003. Crossovers in seedling relative growth rates
between low and high irradiance: Analyses and ecological potential
(reply to Kitajima & Bolker 2003). Funct. Ecol. 17: 281?287.
SOKAL, R. R., AND F. J. ROHLF. 1995. Biometry. W.H. Freeman and company,
New York, New York.
STERCK, F. J., L. POORTER, AND F. SCHIEVING. 2006. Leaf traits determine the
growth?survival trade-off across rain forest tree species. Am. Nat. 167:
758?765.
THOMPSON, W. A., P. E. KRIEDEMANN, AND I. E. CRAIG. 1992. Photosynthetic
response to light and nutrients in sun-tolerant and shade-tolerant rain-
forest trees. 1. Growth, leaf anatomy and nutrient content. Aust. J.
Plant. Physiol. 19: 1?18.
TURNER, I. M. 1990. Tree seedling growth and survival in a Malaysian rain
forest. Biotropica 22: 146?154.
VALLADARES, F., S. J. WRIGHT, E. LASSO, K. KITAJIMA, AND R. W. PEARCY. 2000.
Plastic phenotypic response to light of 16 congeneric shrubs from a
Panamanian rainforest. Ecology 81: 1925?1936.
VEENENDAAL, E. M., M. D. SWAINE, R. T. LECHA, M. F. WALSH, I. K. ABE-
BRESE, AND K. OWUSU-AFRIYIE. 1996. Responses of West African
forest tree seedlings to irradiance and soil fertility. Funct. Ecol. 10: 501?
511.
VENEKLAAS, E. J., AND L. POORTER. 1998. Growth and carbon partitioning of
tropical tree seedlings in contrasting light environments. In H. Lambers,
H. Poorter and M. M. I. Van Vuuren (Eds.). Inherent variation in plant
growth, physiological mechanisms and ecological consequences. Back-
huys, Leiden, The Netherlands.
WRIGHT, I. J., AND M. WESTOBY. 2002. Leaves at low versus high rainfall:
Coordination of structure, lifespan and physiology. New Phytol. 155:
403?416.
WRIGHT, I. J., P. B. REICH, AND M. WESTOBY. 2001. Strategy shifts in leaf
physiology, structure and nutrient content between species of high- and
low-rainfall and high- and low nutrient habitats. Funct. Ecol. 15:
423?434.
WRIGHT, I. J., M. WESTOBY, AND P. B. REICH. 2002. Convergence towards
higher leaf mass per area in dry and nutrient poor habitats has different
consequences for leaf lifespan. J. Ecol. 90: 534?543.
ZIPPERLEN, S. A., AND M. C. PRESS. 1996. Photosynthesis in relation to growth
and seedling ecology of two dipterocarp rain forest tree species. J. Ecol.
84: 863?876.
ZIPPERLEN, S. A., AND M. C. PRESS. 1997. Photosynthetic induction and
stomatal oscillations in relation to the light environment of two
dipterocarp rain forest tree species. J. Ecol. 85: 491?503.
434 Dent and Burslem