Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

1.1 Fan Et Al. (2012)

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

Journal of Ecology 2012, 100, 732–741 doi: 10.1111/j.1365-2745.2011.01939.

Hydraulic conductivity traits predict growth rates and


adult stature of 40 Asian tropical tree species better
than wood density
Ze-Xin Fan1, Shi-Bao Zhang1, Guang-You Hao2, J.W. Ferry Slik1 and Kun-Fang Cao1*
1
Key Laboratory of Tropical Forest Ecology, Xishuangbanna Tropical Botanical Garden, Chinese Academy of
Sciences, Menglun, Mengla, Yunnan 666303, China; 2Department of Organismic and Evolutionary Biology, Harvard
University, Cambridge, MA 02138, USA

Summary
1. Stem xylem characteristics have a great impact on growth and adult stature of trees because of
their role in mechanical support, long-distance water transport and whole-plant carbon allocation.
Yet, despite the potential causal link between xylem traits and plant growth ⁄ adult stature, most
studies have tried to link wood density, an indirect but easy to measure proxy for wood properties,
to tree growth and size.
2. To determine whether xylem traits outperform wood density as predictors of tree growth and
stature, we evaluated the covariation among wood density, xylem anatomical traits, tree diameter
growth rate and adult stature in 40 Asian tropical tree species through principal component analy-
ses and through bivariate correlation, both across species and across phylogenetically independent
contrasts.
3. Vessel diameter exhibited a tight negative relationship with vessel frequency. Wood density
showed a significant correlation with vessel diameter and density, but not with vessel fraction. Most
correlations between functional traits indicate adaptive coordination, demonstrated by significant
correlations between phylogenetically independent contrasts.
4. Across species, diameter growth rate and adult stature were positively correlated with vessel
lumen diameter and potential hydraulic conductivity, but not with wood density. Thus, our results
suggest that xylem anatomical traits that are linked to hydraulic conductivity are better predictors
of tree growth rate and adult stature than wood density.
5. Synthesis. We found that xylem anatomical traits have a more significant influence on whole-
plant performance due to their direct association with stem hydraulic conductivity, whereas wood
density is decoupled from hydraulic function due to complex variations in xylem components.
Key-words: long-term tree growth, phylogenetically independent contrasts, plant develop-
ment and life-history traits, tropical rain forest, vessel diameter, whole-plant performance,
xylem anatomy

2007), whereas high wood density is associated with greater


Introduction
stress tolerance and survival because of its association with
Wood characteristics, especially wood density, have been tra- higher hydraulic safety (Hacke et al. 2001; Sperry, Meinzer &
ditionally regarded as core plant functional traits because of McCulloh 2008) and stronger biomechanical strength (van
their importance for mechanical support, defence, architec- Gelder, Poorter & Sterck 2006; Jacobsen et al. 2007). Thus,
ture, hydraulics, carbon gain and growth potential of plants several lines of evidence suggest that wood density is related to
(Castro-Dı́ez et al. 1998; King et al. 2006; Chave et al. 2009; a fast-to-slow growth-rate continuum, whereby species with
Onoda, Richard & Westoby 2010; Zanne & Falster 2010). denser wood tend to have slower growth, higher survival and
Low wood density is often associated with fast growth greater shade tolerance than species with softer wood (King
(Enquist et al. 1999; Muller-Landau 2004; Osunkoya et al. et al. 2005, 2006; Chave et al. 2006).
However, across a wide range of species, wood density is
*Correspondence author. E-mail: caokf@xtbg.ac.cn usually weakly correlated with growth rate (Chave et al. 2009;

 2011 The Authors. Journal of Ecology  2011 British Ecological Society


Xylem traits and tree growth 733

Poorter et al. 2010). For example, Nascimento et al. (2005) negative correlations between wood density and adult stature
found that the association between wood density and growth were found within shade-tolerant species (Falster & Westoby
rate disappeared altogether when a few gap-demanding species 2005; Poorter, Bongers & Bongers 2006), but this relationship
were removed from the analysis. These mixed results may arise vanished when pioneer and shade-tolerant species were pooled
because wood density is a complex trait and the same wood (van Gelder, Poorter & Sterck 2006; Aiba & Nakashizuka
density can be achieved by various combinations of vessel 2009). On the other hand, maximum height has been found to
diameter, vessel frequency and wood fibre traits (Zanne et al. be associated with vessel size and frequency across a wide
2010), each of these traits has different functional consequences range of woody plants (Preston, Cornwell & DeNoyer 2006;
for growth (Preston, Cornwell & DeNoyer 2006; Russo Martı́nez-Cabrera et al. 2011). Chave et al. (2009) also
et al. 2010). suggested that taller angiosperms have wider vessels because
Wood density in vessel-bearing angiosperms is determined they are more efficient for water transport per unit area, leav-
by properties of the vessels, the matrix of fibres and the paren- ing more space for fibres for mechanical support.
chyma cells surrounding them (Baas et al. 2004). Wood com- Trait correlations observed across taxonomically diverse spe-
posed of closely spaced large vessels will have low density with cies do not necessarily reflect coordinated evolutionary changes
high hydraulic conductivity, whereas wood composed of small or adaptive functional trade-offs (Felsenstein 1985; Ackerly &
vessels spaced widely within a fibre matrix will be dense with Reich 1999). Some observed cross-species correlations between
low hydraulic conductivity (Preston, Cornwell & DeNoyer traits may simply be due to a common ancestry (non-indepen-
2006). Wood density is expected to be inversely related to total dence of species) (Garland, Harvey & Ives 1992). Thus,the inclu-
fibre area (Preston, Cornwell & DeNoyer 2006; Jacobsen et al. sion of phylogenetically independent contrasts provides a more
2007), but this relationship has not been found by other studies powerful quantitative test for correlated evolutionary change
(Martı́nez-Cabrera et al. 2009; Poorter et al. 2010). Therefore, underlying contemporary trait variation and covariation (Ack-
it is especially instructive to divide variation in wood density erly & Donoghue 1998; Maherali, Pockman & Jackson 2004).
into its component parts and examine their ecophysiological In the present study, we examined the covariation in wood
consequences for whole-plant performance. However, few density and xylem anatomical traits, together with tree growth
studies have investigated more detailed xylem anatomical traits rate and adult stature among 40 coexisting tree species in an
to link wood properties with whole-plant performance (Cas- Asian tropical forest, southwestern China. By using phyloge-
tro-Dı́ez et al. 1998; Preston, Cornwell & DeNoyer 2006; netically independent contrasts, we examined whether these
Poorter et al. 2010; Martı́nez-Cabrera et al. 2011). traits show coordinated variation at multiple divergences
It is well known that canopy photosynthesis is largely under across the phylogenetic tree. We aim to (i) evaluate the covaria-
the control of hydraulic architecture (Hubbard et al. 2001; tion of xylem anatomical traits among co-occurring tree
Meinzer 2003; Santiago et al. 2004; Ishida et al. 2008). Vessel species, and its potential in constraining variation in high-order
lumen area strongly affects sapwood-specific conductivity traits, such as tree growth rate and adult stature, and (ii) test the
because conductive capacity scales with vessel diameter to the hypothesis that xylem anatomical traits are better predictors of
fourth power (Tyree & Ewers 1991). Thus, a positive correla- tree growth rate and adult stature than wood density.
tion between vessel size and growth rate is expected, because
wider vessels allow greater rates of transpiration and photo- Material and methods
synthesis (Brodribb, Holbrook & Gutierrez 2002; Zhang &
Cao 2009; Meinzer et al. 2010). Meanwhile, wider vessels, STUDY SITE AND SPECIES
which are associated with lighter wood, are more vulnerable to
This study was conducted at Xishuangbanna Tropical Botanical Gar-
cavitation (Davis, Sperry & Hacke 1999). An important mech- den (XTBG, 2141¢ N, 10125¢ E, altitude 570 m) in the southern
anism facilitating drought tolerance is a ‘safe’ hydraulic system Yunnan Province, Southwestern China. The region has a typical
consisting of smaller and denser vessels (Baas et al. 2004). tropical monsoon climate and thus has a distinct dry season from
However, these traits compromise the hydraulic efficiency via November to April. Mean annual precipitation is about 1560 mm, of
increased resistance to water transport (Hacke et al. 2006; which about 80% falls during the rainy season from May to October.
Pratt et al. 2007). In this way, safety from embolism limits can- Mean annual temperature is 21.7 C with a monthly mean tempera-
opy productivity and should result in lower growth rates. ture being 15.9 C during the coldest month (December) and 25.7 C
However, comparative studies combining growth data from during the warmest month (June).
long-term plots with xylem traits are rare. In this study, we selected 40 co-occurring tree species belonging to
39 genera, 24 families and 11 orders (Fig. 1). This set of species was
Adult stature, or maximum plant height, is another complex
well distributed over the phylogeny and covered different leaf habits
trait that is associated with plant light interception (King et al.
(30 evergreen and 10 deciduous species) and successional stages (nine
2005, 2006). It is often hypothesized that there should be a early-successional and 31 late-successional species). Most (38) of the
systematic relationship between adult stature and ecophysio- species had diffuse-porous woods (Table S1 in Supporting Informa-
logical traits (e.g. wood properties and hydraulic xylem net- tion). All these species are common in the natural forests around
work), particularly in a closed canopy system like tropical rain XTBG. During the 1990s, XTBG conducted a comprehensive investi-
forest where competition for light is severe (Thomas 1996; gation of wood properties for over a hundred broadleaf tree species
King et al. 2006). However, reported relationships between growing in the Xishuangbanna region. The investigated properties
adult stature and wood density are mixed. For example, include species distribution and habitats, macroscopic and micro-

 2011 The Authors. Journal of Ecology  2011 British Ecological Society, Journal of Ecology, 100, 732–741
734 Z.-X. Fan et al.

Fig. 1. Phylogenetic relationship of the 40 tropical tree species. The backbone is based on the Angiosperm Phylogenetic Group III classification
(APG III; Webb, Ackerly & Kembel 2008). Family divisions are indicated by the broken line on the right side of the figure.

scopic structures, and wood density and mechanical properties sel diameters were averaged for each image field. Vessel density (VD)
(Zhang et al. 1989; Ye, Mo & Zou 1999). was calculated as number of vessels per unit wood area (mm)2). Ves-
sel fraction (VF, %) was determined as the ratio of total vessel lumen
area to xylem area. According to the Hagen–Poiseuille law (Tyree &
WOOD DENSITY AND ANATOMICAL TRAITS
Ewers 1991), the potential hydraulic conductivity (Kp,
Wood material of the 40 studied species was collected from XTBG or kg m)1 MPa)1 s)1) was calculated as Kp = (pq ⁄ 128gA)[RDi4] for
nearby natural forests for three individuals per species. One sample i = 1 to n vessels in each field, where A is field area, q is the density of
per tree was taken from the main trunk at breast height (1.3 m) with a water (998.2 kg m)3 at 20 C) and g is the viscosity of water
chisel. Wood fresh volume was measured using the water displace- (1.002 · 10)9 MPa s)1 at 20 C).
ment method, after which the samples were oven-dried for at least
48 h at 70 C. Wood density (WD, g cm)3) was determined as wood
DIAMETER GROWTH RATE AND ADULT STATURE
dry mass per unit of fresh volume. Wood samples were then softened
by soaking in a solution of glycerine and ethanol (1 : 1 by volume) for We selected forestry census data of the 40 tree species from the two
three to four months. One transverse section (15–50 lm thick) per permanent plots dedicated to long-term ecological research managed
wood sample was made with a sliding microtome and then stained by XTBG since the mid-1990s (Table S2 in Supporting Information).
with safranin and embedded in neutral balsam. Cross-sections were Plot I (1 ha) was established in 1994, while plot II (0.5 ha) was estab-
made and dry-preserved by XTBG staff (Zou SQ, pers. comm.). lished in 2002. The two plots are adjacent to each other (<10 km)
Cross-sections of the 40 tree species were retrieved from XTBG and have similar climate and soil conditions, as well as tree species
and then imaged using a digital camera (DFC 295, Leica, Germany) diversity. At each plot, trees of diameter at breast height
mounted on a microscope (DM 2500, Leica, Germany). Two to three (DBH) ‡ 5 cm were mapped, tagged, identified and measured yearly
slides per species were taken at 50· magnification and three to five for their DBH (mm) and height (m).
images were taken per slide. Images were analysed using the imagej For each tree, the DBH at consecutive censuses was regressed
program (http://rsb.info.nih.gov/ij/index.html). The lumen diameter against the corresponding measurement date. The corresponding
(major and minor axis) and total lumen area of all individual vessels slope was used to obtain an annual diameter growth rate (GR,
in each cross-sectional image were measured. For each species, at mm yr)1). Trees included in this analysis had at least six DBH mea-
least 50 vessels (>550 for small vessels) were analysed. Since vessels surements. Trees with negative or anomalous growth, assuming here
are not exactly circular but mostly elliptical, the diameter of each ves- to be caused by measurement error, were excluded from the analyses.
sel was calculated as D = [32(ab)3 ⁄ (a2 + b2)]1 ⁄ 4, where a and b are Species-specific growth rate means were obtained for all 40 species
the major and minor axis dimensions, respectively (Lewis 1992). Ves- from a total of 795 individuals (Table S3 in Supporting Information).

 2011 The Authors. Journal of Ecology  2011 British Ecological Society, Journal of Ecology, 100, 732–741
Xylem traits and tree growth 735

For the selected species, maximum (potential) plant height (Hmax) undergone changes in similar direction and magnitude across the
was estimated from H = Hmax[(1)exp()aDBHb)], where H is tree phylogeny and supports a possible functional link between these
height in metres, DBH is diameter at breast height in centimetres and traits (Jacobsen et al. 2007).
a and b are allometric coefficients that approach values of standard
allometric constants for small values of H (Thomas 1996). For tree
species that did not show a clear asymptote, we used height at the Results
maximum observed DBH as predicted by species-specific regression
equations, rather than the estimated asymptote, as the maximum ASSOCIATIONS AMONG WOOD TRAITS
height (Poorter, Bongers & Bongers 2006; Aiba & Nakashizuka
2009). Xylem traits varied substantially across the 40 tree species
(Tables S1 and S4 in Supporting Information): wood density
(WD) varied 2.5-fold (0.34–0.86 g cm)3), vessel diameter (D)
CONSTRUCTION OF THE PHYLOGENETIC TREE
varied 3.7-fold (61.8–226.2 lm), vessel density (VD) varied
A phylogenetic tree at family level was first constructed using PHY- 12-fold (2.2–24.2 vessels per cm)2) and potential hydraulic con-
LOMATIC, which is based on the Angiosperm Phylogeny Group III ductivity (Kp) varied 35-fold (9.5–332.5 kg m)1 s)1 MPa)1).
classification of angiosperms (APG III; http://www.phylodiversity. Vessel fraction (vessel ⁄ xylem ratio, VF) only showed a moder-
net/phylomatic/). To resolve the phylogenetic relations of species
ate variation (4–20%) despite a wide range of vessel densities.
belonging to the same families, nuclear ITS sequences were retrieved
Early-successional species had, on average, wider vessels,
for 12 tree species belonging to three families (5.8s ribosomal RNA
for Lauraceae and Sapindaceae, 18s ribosomal RNA for Euphorbia-
lighter wood, higher Kp and growth rates compared with late-
ceae) from the GenBank (http://www.ncbi.nlm.nih.gov/genbank/). successional species. However, only the differences in Kp and
The ITS sequences were aligned, and the phylogenetic relationships WD were statistically different between these two successional
were built by neighbour-joining (NJ) method using the mega program stage categories (P < 0.05).
(version 4.0; Tamura et al. 2007). The final phylogenetic tree included Results of the principal component analysis showed that the
all the 40 tree species (Fig. 1). Conservative branch lengths of the first axis explained 45.4% of the variation among the seven
phylogenic tree were assigned using the BLADJ function of phylocom traits and was mostly associated with wood density (WD), ves-
(Webb, Ackerly & Kembel 2008). sel density (VD) and vessel lumen diameter (D) (Fig. 2a). The
second axis explained 22.8% of the variation and was associ-
STATISTICS AND PHYLOGENETIC COMPARATIVE ated with water transport capacity (D, Kp), growth rate (GR)
METHODS and adult stature (Hmax) (Table S5 in Supporting Informa-
tion). The results of the PCA based on phylogenetically inde-
To increase the normality of distributions, the original data were
log10-transformed (Kerkhoff & Enquist 2009). Pearson’s correlations
pendent contrasts (PICs) were generally consistent with the
were calculated for all pairwise combinations of traits. Bivariate rela- results of the conventional PCA (Fig. 2b).
tionships were analysed using standardized major axis (SMA) regres- Species with higher vessel density (frequency) had signifi-
sion, or model II regression, using the smatr software (http:// cantly narrower vessel lumens and greater vessel fractions (ves-
www.bio.mq.edu.au/ecology/SMATR/). The SMA is more suitable sel area per unit xylem area), whereas vessel diameter and
than regression model I for examining the scaling relationship vessel fraction were unrelated (Table 1, Fig. S1 in Supporting
between two variables that are both measured with error (Warton Information). Potential hydraulic conductivity (Kp) was posi-
et al. 2006). To describe correlation patterns among multiple traits, tively related to vessel lumen diameter and vessel fraction, but
principal components analyses (PCA) were performed for seven traits negatively related to vessel density (Fig. S2 in Supporting
of all the 40 tree species, using both original trait data and phyloge-
Information). Wood density was correlated negatively with
netically independent contrasts (PICs).
vessel lumen diameter and positively with vessel density, but
We tested the phylogenetic signal using the K-statistic and a
randomization test implemented via phylogenetically independent
not with vessel fraction (Fig. S3 in Spporting Information).
contrasts (Blomberg, Garland & Ives 2003) using r v2.12.0 (R Most of the correlations based on PICs were consistent with
Development Core Term 2008, package ‘picante’). Phylogeneti- the cross-species correlations, except that the negative correla-
cally independent contrasts (PICs) for all the functional traits tion between WD and Kp only became significant when phy-
were calculated using the developed reference phylogeny of the logeny was incorporated into the analysis (Table 1, Figs S1–
40 species (Felsenstein 1985; Garland, Harvey & Ives 1992). Inde- S3 in Supporting Information).
pendent contrasts were calculated as the difference in trait means
of two daughter nodes or tips divided by the expected amount of
change, which is the square root of the branch length separating CORRELATIONS BETWEEN XYLEM TRAITS AND
the two taxa. This provides N)1 contrasts, where N is the num- WHOLE-PLANT PERFORMANCE
ber of tips in the phylogeny (N = 40). All trait data were log10-
Stem diameter growth rate was positively correlated with ves-
transformed prior to contrast calculation. To examine the coordi-
nation among functional traits throughout their phylogeny, we
sel diameter and potential hydraulic conductivity, both for
performed a correlation analysis on calculated PICs by forcing cross-species means and PICs (Table 1, Fig. 3). Growth rate
lines through the origin (Garland, Harvey & Ives 1992), using was inversely correlated with vessel density, but this correlation
cactus 1.13 (comparative analysis of continuous traits using sta- became non-significant when analysed using phylogenetically
tistics, Schwilk & Ackerly 2001). A significant correlation independent contrasts (Fig. 3b,f). Adult stature was positively
between PICs for two traits indicates that these traits have related to vessel diameter and potential hydraulic conductivity,

 2011 The Authors. Journal of Ecology  2011 British Ecological Society, Journal of Ecology, 100, 732–741
736 Z.-X. Fan et al.

(a) species and PICs comparisons (Figs 3 and 4). Growth rate
showed a marginal positive correlation (R = 0.26, P = 0.1)
with adult stature, and this relationship became more signifi-
cant (R = 0.28, P < 0.05) when phylogenetically independent
contrasts were used (Fig. S4 in Supporting Information).

Discussion

ASSOCIATIONS BETWEEN WOOD DENSITY AND


VESSEL TRAITS

Wood density was expected to decrease with increasing vessel


cross-sectional area (vessel fraction), as more open vessel
lumen spaces should lead to less dense material (Preston, Corn-
well & DeNoyer 2006). However, no significant correlation
(b) between wood density and vessel fraction was found across the
40 species in the present study (Fig. S3 in Spporting Informa-
tion). Similarly, wood density of 42 Bolivian tree species was
not correlated with either vessel fraction or fibre cross-sectional
area (Poorter et al. 2010). Jacobsen et al. (2007) proposed that
variation in wood density might be driven by fibre cell wall
thickness, rather than by variations in fibre tissue area.
Accordingly, Martı́nez-Cabrera et al. (2009) found that wood
density of 61 shrub species was mainly driven by fibre charac-
teristics, that is, denser woods were associated with smaller
fibre lumens and higher fibre wall to lumen ratios. Wood den-
sity was negatively correlated with vessel diameter but posi-
tively correlated with vessel density (Fig. S3 in Supporting
Information), which is consistent with a study based on 135
tree species in North America (Martı́nez-Cabrera et al. 2011).
Fig. 2. Principal component analyses (PCA) on cross-species means
(a) and phylogenetically independent contrasts (b) for seven traits CORRELATIONS BETWEEN XYLEM TRAITS AND
among 40 tropical tree species. Grey (early-successional) and white
GROWTH
(late-successional) circles indicate the loadings of individual species
on these axes. D, vessel diameter; VD, vessel density; VF, vessel frac- Contrary to the hypothesis that wood density is a good proxy
tion; Kp, potential hydraulic conductivity; WD, wood density; GR,
of tree species’ position along a fast-to-slow growth continuum
diameter growth rate; Hmax, adult stature.
(Chave et al. 2006), we did not find significant correlation
but not with vessel density or vessel fraction (Table 1, Fig. 4). between wood density and diameter growth rate in the present
On the other hand, neither growth rate nor adult stature was study (Fig. 3). Although species with the lowest wood densities
significantly correlated with wood density, both for cross- (early-successional) tend to grow faster, many species with

Table 1. Pearson’s correlation coefficients and independent contrast correlation coefficients between xylem anatomical traits, wood density,
growth rate and adult stature of 40 tropical tree species

D VD VF Kp WD GR Hmax

D )0.831** 0.171 0.876** )0.401** 0.515** 0.300*


VD )0.785** 0.405** )0.467** 0.428** )0.435** )0.190
VF 0.116 0.523** 0.617** 0.104 0.077 0.160
Kp 0.835** )0.325* 0.633** )0.261 0.440** 0.321*
WD )0.402** 0.351* 0.016 )0.292* )0.126 )0.104
GR 0.379** )0.253 0.112 0.339* 0.169 0.260
Hmax 0.321* )0.060 0.345* 0.422** )0.197 0.280*

Data were log-transformed before analysis of Pearson’s correlation and phylogenetically independent contrasts.
Correlations between species are given upper right of the diagonal, and correlations between phylogenetically independent contrasts
(PICs) forcing lines through the origin are given lower left. D, vessel diameter (lm); VD, vessel density (mm)2); VF, vessel fraction (%);
Kp, potential hydraulic conductivity (kg m)1 s)1 MPa)1); WD, wood density (g cm)3); GR, diameter growth rate (mm yr)1); Hmax, adult
stature (m). N = 40 for cross-species trait data and 39 for PICs. Traits that are significantly correlated are marked: **P < 0.01;
*P < 0.05.

 2011 The Authors. Journal of Ecology  2011 British Ecological Society, Journal of Ecology, 100, 732–741
Xylem traits and tree growth 737

(a) (e)

(b) (f)

(c) (g)

(d) (h)

Fig. 3. Relationship between diameter growth rate (GR) and wood density (WD; a, e), vessel density (VD; b, f), vessel diameter (D; c, g) and
potential hydraulic conductivity (Kp; d,h) for 40 tropical tree species. Black and white circles represent early-successional and late-successional
species, respectively. Note the log10 scales in panels from (a) to (d). Standardized major axis (SMA) regression lines are shown for significant rela-
tionships only. Pearson’s cross-species (Rc) and phylogenetically independent contrast (Rp) correlation coefficients are shown. ns, P > 0.05;
*P < 0.05; **P < 0.01.

intermediate wood density had variable growth rates. More- Information) constrains the variability of vessel fractions, and
over, no significant relationships between wood density and thus fibre cross-sectional area. Moreover, density of the matrix
growth rate were found within different groups of successional outside of the lumens may contribute considerably to variation
stages (earlier- or late-successional species) (Table S6 in in wood density, which allow our studied species to partially
Spporting Information). Similarly, wood density of terminal escape from the putative trade-offs involving wood density
branches from 17 dipterocarp species was not correlated with and conductivity (Hacke et al. 2006; Russo et al. 2010).
either mean diameter or height growth rates (Zhang & Cao In this study, we found xylem anatomical traits were much
2009). Even where a negative correlation between wood density better predictors of growth rate than wood density, which is
and growth was found, it was often relatively weak (Nasci- consistent with their more direct and stronger correlation with
mento et al. 2005; Osunkoya et al. 2007; Chave et al. 2009). hydraulic conductivity (Castro-Dı́ez et al. 1998; Russo et al.
The lack of a strong correlation between wood density and 2010). The coordination of hydraulics and growth rate has
growth rates is likely to have arisen from the fact that wood been observed within dipterocarp species growing in a com-
density is determined by both vessel and fibre properties mon garden (Zhang & Cao 2009) as well as species growing in
(Martı́nez-Cabrera et al. 2009). The tight negative association native rain forest (Poorter et al. 2010), which is due to a strong
between vessel size and vessel density (Fig. S2 in Supporting hydraulic–photosynthetic coordination (Hubbard et al. 2001;

 2011 The Authors. Journal of Ecology  2011 British Ecological Society, Journal of Ecology, 100, 732–741
738 Z.-X. Fan et al.

(a) (e)

(b) (f)

(c) (g)

(d) (h)

Fig. 4. Relationship between adult stature (Hmax) and wood density (WD; a, e), vessel density (VD; b, f), vessel diameter (D; c, g) and potential
hydraulic conductivity (Kp; d,h) for 40 tropical tree species. Black and white circles represent early-successional and late-successional species,
respectively. Note the log10 scales in panels from (a) to (d). Standardized major axis (SMA) regression lines are shown for significant relationships
only. Pearson’s cross-species (Rc) and phylogenetically independent contrast (Rp) correlation coefficients are shown. ns, P > 0.05; *P < 0.05;
**P < 0.01.

Santiago et al. 2004; Ishida et al. 2008). Tree species with lar- associated with successional gradients (Falster & Westoby
ger but less dense vessels are generally associated with high 2005). In each case, the relationship is mediated by a trade-off
potential hydraulic conductivity (Fig. S2 in Supporting Infor- between wood density and growth rate. However, we found no
mation). Higher water transport efficiency through the stem significant correlation between wood density and adult stature,
allows for higher stomatal conductance and crown photosyn- which is consistent with several other studies (Osunkoya et al.
thetic rates, and hence higher potential growth rates (Brodribb, 2007; Russo et al. 2010; Martı́nez-Cabrera et al. 2011). We
Holbrook & Gutierrez 2002; Meinzer et al. 2010). also found no evidence that the correlation between wood den-
sity and adult stature was mediated by a relationship between
wood density and growth rate (Preston, Cornwell & DeNoyer
CORRELATIONS BETWEEN XYLEM TRAITS AND ADULT
2006). Considering the large discrepancies among different
STATURE
studies, wood density may not necessarily have an intrinsic
Both negative and positive interspecific relationships between association with growth rate and adult stature. Instead, we
wood density and stature have been reported. Negative corre- found xylem anatomical traits directly associated with hydrau-
lations typically arise across light gradients within tropical for- lic conductivity are better predictors of growth rate and adult
ests (Thomas 1996), whereas positive correlations are stature of tropical forest tree species.

 2011 The Authors. Journal of Ecology  2011 British Ecological Society, Journal of Ecology, 100, 732–741
Xylem traits and tree growth 739

Adult stature was positively correlated with vessel diameter which determine niche differentiation and functional diversity
and potential hydraulic conductivity (Fig. 4), which implies of tropical forest tree species.
that tree species achieving greater heights require greater
hydraulic capacity (Koch et al. 2004). Similarly, Zach et al.
Acknowledgements
(2010) found that twig vessel diameter and sapwood-area-spe-
cific hydraulic conductivity significantly increased with tree We are grateful to Prof. Zou SQ who kindly provided us wood cross-sections.
height across eight common tropical rain forest tree species in We appreciate Prof. Zhu H, Mr. Wen B and Mr. Tan YH for their help on plant
identification. Plots data were provided by the Xishuangbanna Station for
Indonesia. A recent model has shown that optimal carbon allo- Tropical Rain Forest Ecosystem Studies (XSTRFE) of Chinese Ecosystem
cation would produce stems that maximize conductivity (via Research Network (CERN). We also thank Prof. J.H.C. Hans Cornelissen,
large vessels) and minimize wood density while just maintain- Dr. William K. Cornwell and three anonymous referees for their helpful com-
ments on the manuscript. This work was financially funded by the National
ing adequate mechanical support (Taneda & Tateno 2004). Natural Science Foundation of China (No. 31000237, 31170315) and West
Variation in vessel density may allow species to maintain suffi- Light Foundation of the Chinese Academy of Sciences.
cient wood density even while increasing lumen area (Preston,
Cornwell & DeNoyer 2006).
References
Ackerly, D.D. (1999) Comparative plant ecology and the role of phylogenetic
PHYLOGENETIC SIGNAL AND CORRELATED information. Physiological Plant Ecology (eds M. Press, J.D. Scholes &
EVOLUTION M.G. Barker), pp. 391–413. Blackwell Scientific Press, Oxford, UK.
Ackerly, D.D. & Donoghue, M.J. (1998) Leaf size, sapling allometry, and Cor-
Our analysis with and without the incorporation of evolu- ner’s rules: phylogeny and correlated evolution in maple (Acer). American
Naturalist, 152, 767–791.
tionary relationships was mostly consistent, indicating that Ackerly, D.D. & Reich, P.B. (1999) Convergence and correlations among leaf
the detected patterns reflect coordinated evolution. More- size and function in seed plants: a comparative test using independent con-
over, the relatively low levels of phylogenetic signal of the trasts. American Journal of Botany, 86, 1272–1281.
Aiba, M. & Nakashizuka, T. (2009) Architecture differences associated with
seven studied traits (Blomberg’s K-values <1, Table S4 in adult stature and wood density in 30 temperate tree species. Journal of Ecol-
Supporting Information) may result from the fact that the ogy, 23, 265–273.
studied species are phylogenetically diverse (from many Baas, P., Ewers, F.W., Davis, S.D. & Wheeler, E.A. (2004) Evolution of xylem
physiology. The Evolution of Plant Physiology (eds A.R. Hemsley & I.
families), and thus common ancestral effect is minimal Poole), pp. 273–295. Elsevier Academic Press, London, UK.
(Ackerly 1999). Vessel lumen diameter showed a strong Blomberg, S.P., Garland, T. J. & Ives, A.R. (2003) Testing for phylogenetically
negative contrast correlation with vessel frequency (Fig. 2), signal in comparative data: behavioral traits are more labile. Evolution, 57,
717–745.
indicating that increases in vessel size during evolution Brodribb, T.J., Holbrook, N.M. & Gutierrez, M.V. (2002) Hydraulic and pho-
have generally been accompanied by decreases in vessel tosynthetic co-ordination in seasonally dry tropical forest trees. Plant, Cell
density (Preston, Cornwell & DeNoyer 2006; Russo et al. and Environment, 25, 1435–1444.
Carlquist, S.W. & Hoekman, D.A. (1985) Ecological wood anatomy of the
2010; Zanne et al. 2010). Recently developed metrics indi- woody southern Californian flora. IAWA Bulletin, 6, 319–347.
cate that both vessel area and wood density show strong Castro-Dı́ez, P., Puyravaud, J.P., Cornelissen, J.H.C. & Villar-Salvador, P.
phylogenetic conservatism among 17 Quercus species in (1998) Stem anatomy and relative growth rate in seedlings of a wide range of
woody plant species and types. Oecologia, 116, 57–66.
Florida (Cavender-Bares, Kitajima & Bazzaz 2004). Mean- Cavender-Bares, J., Kitajima, K. & Bazzaz, F.A. (2004) Multiple trait associa-
while, correlation between vessel density and growth rate tions in relation to habitat differentiation among 17 Floridian oak species.
became non-significant when PICs were used (Fig. 3b,f). Ecological Monographs, 74, 635–662.
Chave, J., Muller-Landau, H.C., Baker, T.R., Easdale, T.A., Steege, H.T. &
These results are consistent with the findings of Carlquist Webb, C.O. (2006) Regional and phylogenetic variation of wood density
& Hoekman (1985), who argued that vessel density across 2456 neotropical tree species. Ecological Applications, 16, 2356–2367.
changes more rapidly or is more evolutionarily labile, than Chave, J., Coomes, D., Jansen, S., Lewis, S.L., Swenson, N.G. & Zanne, A.E.
(2009) Towards a worldwide wood economics spectrum. Ecology Letters, 12,
vessel area. 351–366.
Davis, S.D., Sperry, J.S. & Hacke, U.G. (1999) The relationship between xylem
conduit diameter and cavitation caused by freeze-thaw events. American
Conclusion Journal of Botany, 86, 1367–1372.
Enquist, B.J., West, G.B., Charnov, E.L. & Brown, J.H. (1999) Allometric scal-
Our analyses indicate that tree growth rate and adult stature ing of production and life-history variation in vascular plants. Nature, 401,
are associated with xylem anatomical traits determining stem 907–911.
Falster, D.S. & Westoby, M. (2005) Alternative height strategies among 45
hydraulic capacity, but not with wood density. Our results also dicot rain forest species from tropical Queensland, Australia. Journal of
suggest that wood density is not a good proxy for tree species’ Ecology, 93, 521–535.
position along a fast-to-slow growth continuum, which results Felsenstein, J. (1985) Phylogenies and the comparative method. American Nat-
uralist, 125, 1–15.
from the fact that wood density is decoupled from hydraulic Garland, T.J., Harvey, P.H. & Ives, A.R. (1992) Procedures for the analysis of
function due to complex variations of xylem components (ves- comparative data using phylogenetically independent contrasts. Systematic
sel size, vessel density, fibre area and density). In contrast, Biology, 41, 18–32.
van Gelder, H.A., Poorter, L. & Sterck, F.J. (2006) Wood mechanics, allome-
xylem anatomical traits have a more significant influence on try, and life-history variation in a tropical rain forest tree community. New
whole-plant performance due to their direct association with Phytologist, 171, 367–378.
stem hydraulic conductivity and their important role in inter- Hacke, U.G., Sperry, J.S., Pockman, W.T., Davis, S.D. & McCulloh, K.A.
(2001) Trends in wood density and structure are linked to prevention of
mediating a suite of evolutionarily based functional trade-offs, xylem implosion by negative pressure. Oecologia, 126, 457–461.

 2011 The Authors. Journal of Ecology  2011 British Ecological Society, Journal of Ecology, 100, 732–741
740 Z.-X. Fan et al.

Hacke, U.G., Sperry, J.S., Wheeler, J.K. & Castro, L. (2006) Scaling of angio- Schwilk, D.W. & Ackerly, D.D. (2001) Flammability and serotiny as strategies:
sperm xylem structure with safety and efficiency. Tree Physiology, 26, 689–701. correlated evolution in pines. Oikos, 94, 326–336.
Hubbard, R.M., Ryan, M.G., Stiller, V. & Sperry, J.S. (2001) Stomatal conduc- Sperry, J.S., Meinzer, F.C. & McCulloh, K.A. (2008) Safety and efficiency con-
tance and photosynthesis vary linearly with plant hydraulic conductance in flicts in hydraulic architecture: scaling from tissues to trees. Plant, Cell and
ponderosa pine. Plant, Cell and Environment, 24, 113–121. Environment, 31, 632–645.
Ishida, A., Nakano, T., Yazaki, K., Matsuki, S., Koike, N., Lauenstein, D.L., Tamura, K., Dudley, J., Nei, M. & Kumar, S. (2007) MEGA4: molecular evo-
Shimizu, M. & Yamashita, N. (2008) Coordination between leaf and stem lutionary genetics analysis (MEGA) software version 4.0. Molecular Biology
traits related to leaf carbon gain and hydraulics across 32 drought-tolerant and Evolution, 24, 1596–1599.
angiosperms. Oecologia, 156, 193–202. Taneda, H. & Tateno, M. (2004) The criteria for biomass partitioning of the
Jacobsen, A.L., Agenbag, L., Esler, K.J., Brandon Pratt, R., Ewers, F.W. & current shoot: water transport versus mechanical support. American Journal
Davis, S.D. (2007) Xylem density, biomechanics and anatomical traits corre- of Botany, 91, 1949–1959.
late with water stress in 17 evergreen shrub species of the mediterranean-type Thomas, S.C. (1996) Asymptotic height as a predictor of growth and allometric
climate region of South Africa. Journal of Ecology, 95, 171–183. characteristics in Malaysian rain forest trees. American Journal of Botany,
Kerkhoff, A.J. & Enquist, B.J. (2009) Multiplicative by nature: why logarith- 83, 556–566.
mic transformation is necessary in allometry. Journal of Theoretical Biology, Tyree, M.T. & Ewers, F.W. (1991) The hydraulic architecture of trees and other
257, 519–521. woody plants. New Phytologist, 119, 345–360.
King, D.A., Davies, S.J., Nur Supardi, M.N. & Tan, S. (2005) Tree growth is Warton, D.I., Wright, I.J., Falster, D.S. & Westoby, M. (2006) Bivariate line-
related to light interception and wood density in two mixed dipterocarp for- fitting methods for allometry. Biological Review, 81, 259–291.
ests of Malaysia. Functional Ecology, 19, 445–453. Webb, C.O., Ackerly, D.D. & Kembel, S.W. (2008) Phylocom: software for the
King, D.A., Davies, S.J., Tan, S. & Noor, N.S.M. (2006) The role of woody analysis of phylogenetic community structure and trait evolution. Bioinfor-
density and stem support costs in the growth and mortality of tropical trees. matics, 24, 2098–2100.
Journal of Ecology, 94, 670–680. Ye, R.X., Mo, S.M. & Zou, S.Q. (1999) Atlas of Major Breadleaf Woody Spe-
Koch, G.W., Sillett, S.C., Jennings, G.M. & Davis, S.D. (2004) The limits of cies in Yunnan. Yunnan University Press, Kunming.
tree height. Nature, 428, 851–854. Zach, A., Schuldt, B., Brix, S., Horna, V., Culmsee, H. & Leuschner, C.
Lewis, A.M. (1992) Measuring the hydraulic diameter of a pore or conduit. (2010) Vessel diameter and xylem hydraulic conductivity increase with
American Journal of Botany, 79, 1158–1161. tree height in tropical rainforest trees in Sulawesi, Indonesia. Flora,
Maherali, H., Pockman, W.T. & Jackson, R.B. (2004) Adaptive variation in the 205, 506–512.
vulnerability of woody plants to xylem cavitation. Ecology, 85, 2184–2199. Zanne, A.E. & Falster, D.S. (2010) Plant functional traits – linkages among
Martı́nez-Cabrera, H.I., Jones, C.S., Espino, S. & Schenk, H.J. (2009) Wood stem anatomy, plant performance and life history. New Phytologist, 185,
anatomy and wood density in shrubs: responses to varying aridity among 348–351.
transcontinental transects. American Journal of Botany, 96, 1388–1398. Zanne, A.E., Westoby, M., Falster, D.S., Ackerly, D.D., Loarie, S.R., Arnold,
Martı́nez-Cabrera, H.I., Schenk, H.J., Cevallos-Ferriz, R.S. & Jones, C.S. S.J. & Coomes, D.A. (2010) Angiosperm wood structure: global patterns in
(2011) Integration of vessel traits, wood density, and height in angiosperm vessel anatomy and their relation to wood density and potential conductiv-
shrubs and trees. American Journal of Botany, 98, 915–922. ity. American Journal of Botany, 92, 207–215.
Meinzer, F.C. (2003) Functional convergence in plant responses to the environ- Zhang, J.L. & Cao, K.F. (2009) Stem hydraulics mediates leaf water status, car-
ment. Oecologia, 134, 1–11. bon gain, nutrient use efficiencies and plant growth rates across dipterocarp
Meinzer, F.C., McCulloh, K.A., Lachenbruch, B., Woodruff, D.R. & Johnson, species. Functional Ecology, 23, 658–667.
D.M. (2010) The blind men and the elephant: the impact of context and scale Zhang, J.L., Wu, D.Q., Zou, S.Q. & Liu, J.H. (1989) Major Wood in Xishuangb-
in evaluating conflicts between plant hydraulic safety and efficiency. Oecolo- anna. Forestry Press of China, Beijing, China.
gia, 164, 287–296.
Muller-Landau, H.C. (2004) Interspecific and inter-site variation in wood spe- Received 11 July 2011; accepted 15 November 2011
cific gravity of tropical trees. Biotropica, 36, 20–32. Handling Editor: Hans Cornelissen
Nascimento, H.E.M., Laurance, W.F., Condit, R., Laurance, S.G., D’Angelo,
S. & Andrade, A.C. (2005) Demographic and life-history correlateds for
Amazonian trees. Journal of Vegetation Sciences, 16, 625–634.
Onoda, Y., Richard, A.E. & Westoby, M. (2010) The relationship between stem
Supporting Information
biomechanics and wood density is modified by rainfall in 32 Australian The following Supporting Information is available for this article:
woody plant species. New Phytologist, 185, 493–501.
Osunkoya, O.O., Sheng, T.K., Mahmud, N.-A. & Damit, N. (2007) Variation
in wood density, wood water content, stem growth and mortality among Figure S1. Cross-species (a) and phylogenetically independent con-
twenty-seven tree species in a tropical rainforest on Borneo Island. Austral trast (b) correlations between vessel diameter and vessel density
Ecology, 32, 191–201. across 40 tropical tree species.
Poorter, L., Bongers, L. & Bongers, F. (2006) Architecture of 54 moist-forest
tree species: traits, trade-offs, and functional groups. Ecology, 87, 1289–1301.
Poorter, L., McDonald, I., Alarcón, A., Fichtler, E., Licona, J.-C., Peña-Cla- Figure S2. Relationships between potential hydraulic conductivity
ros, M., Sterck, F., Villegas, Z. & Sass-Klaassen, U. (2010) The importance (Kp) versus vessel fraction (VF), vessel density (VD) and vessel diame-
of wood traits and hydraulic conductance for the performance and life his- ter (D) for 40 tropical tree species.
tory strategies of 42 rainforest tree species. New Phytologist, 185, 481–492.
Pratt, R.B., Jacobsen, A.L., Ewers, F.W. & Davis, S.D. (2007) Relationships
among xylem transport, biomechanics and storage in stems and roots of nine Figure S3. Relationships between wood density and potential hydrau-
Rhamnaceae species of the California chaparral. New Phytologist, 174, 787– lic conductivity (Kp), vessel fraction (VF), vessel density (VD) and
798. vessel diameter (D) for 40 tropical tree species.
Preston, K.A., Cornwell, W.K. & DeNoyer, J.L. (2006) Wood density and ves-
sel traits as distinct correlates of ecological strategy in 51 California coast
range angiosperms. New Phytologist, 170, 807–818. Figure S4. Cross-species and phylogenetically independent contrast
R Development Core Term. (2008) R: A Language and Environment for Statisti- correlations between growth rate and adult stature across 40 tropical
cal Computing. R Foundation for Statistical Computing, Vienna, Austria. tree species.
ISBN 3–900051–07–0.
Russo, S.E., Jenkins, K.L., Wiser, S.K., Uriarte, M., Duncan, R.P. & Coomes,
D.A. (2010) Interspecific relationships among growth, mortality and xylem Table S1. List of wood density, wood anatomical and mechanical
traits of woody species from New Zealand. Functional Ecology, 24, 253–262. traits of 40 tropical tree species.
Santiago, L.S., Goldstein, G., Meinzer, F.C., Fisher, J.B., Machado, K.,
Woodruff, D. & Jones, T. (2004) Leaf photosynthetic traits with hydraulic
conductivity and wood density in Panamanian forest canopy trees. Oecolo- Table S2. Plots information for growth data.
gia, 140, 543–550.

 2011 The Authors. Journal of Ecology  2011 British Ecological Society, Journal of Ecology, 100, 732–741
Xylem traits and tree growth 741

Table S3. List of growth rate, adult stature and their calculation Table S6. Pearson’s correlation coefficients among xylem traits,
information for 40 tropical tree species. growth rate and adult stature within different groups of successional
stages.
Table S4. Statistics of the seven traits of 40 tropical tree species, as
well as different successional stages (early- or late-successional). As a service to our authors and readers, this journal provides support-
ing information supplied by the authors. Such materials may be
Table S5. Eigenvalues and factor loadings of conventional and phylo- re-organized for online delivery, but are not copy-edited or typeset.
genetic principal component analysis on seven traits of 40 tropical Technical support issues arising from supporting information (other
tree species. than missing files) should be addressed to the authors.

 2011 The Authors. Journal of Ecology  2011 British Ecological Society, Journal of Ecology, 100, 732–741

You might also like