全 文 :生物量分配影响三种不同海拔起源的松树生长∗
张石宝
(中国科学院昆明植物研究所资源植物与生物技术所级重点实验室ꎬ 云南 昆明 650201)
摘要: 松属的思茅松 (Pinus kesiya var. langbianensis)、 云南松 (P. yunnanensis) 和高山松 (P. densata) 是
组成中国西南不同海拔针叶森林的主要树种ꎬ 然而这三个树种在发育速度尤其是高生长方面表现出明显的
差异ꎮ 为了弄清引起这些变异的生理和形态学原因ꎬ 本文将三种松树种植于同一环境下ꎬ 对其光合作用、
生物量分配、 生长速率和叶片性状进行了研究ꎮ 研究发现ꎬ 与来源于高海拔的树种相比ꎬ 低海拔的树种有
更高的株高、 以及更大的干物质重量、 相对生长速率、 叶质比、 茎质比和比叶面积ꎬ 但叶片氮含量、 碳含
量和根质比较低ꎮ 高海拔树种的光合速率并不明显低于低海拔树种ꎮ 相对生长速率和树高均与叶质比呈显
著正相关ꎬ 与根质比负相关ꎬ 但与最大光合速率没有显著关系ꎮ 这些结果表明ꎬ 生物量的分配式样和长期
的形态特性能够更好地预测不同海拔松树的生长表现ꎮ
关键词: 生物量分配ꎻ 海拔ꎻ 生长ꎻ 光合作用ꎻ 松树
中图分类号: Q 948 11ꎬ Q 945 文献标识码: A 文章编号: 2095-0845(2014)01-047-09
Biomass Partitioning Affects the Growth of Pinus
Species from Different Elevations
ZHANG Shi ̄Bao
(Key Laboratory of Economic Plants and Biotechnologyꎬ Kunming Institute of Botanyꎬ
Chinese Academy of Sciencesꎬ Kunming 650201ꎬ China)
Abstract: The conifer forests in southwestern China are mainly dominated by three vicariant species within Pinus:
P. kesiya var. langbianensisꎬ P. yunnanensisꎬ and P. densata. Their sites range from lower to higher elevationsꎬ re ̄
spectivelyꎬ and each species shows differences in rates of developmentꎬ especially with regard to height. To identify
the physiological and morphological causes of this inherent variationꎬ photosynthesisꎬ biomass partitioningꎬ growth
rates and leaf traits were investigated of plants cultivated under the same environmental conditions. Trees of the spe ̄
cies native to the lower elevation were tallerꎬ and had higher values for dry weightꎬ relative growth rate (RGR)ꎬ leaf
mass fraction (LMF)ꎬ stem mass fraction (SMF)ꎬ and specific leaf area per unit mass (SLA)ꎬ relative to those
from the higher elevations. Howeverꎬ their leaf N and C contents per unit areaꎬ and their root mass fraction (RMF)ꎬ
were smaller than those of high ̄elevation trees. Photosynthetic capacity in species from high elevations was not signif ̄
icantly reduced from the level calculated for trees from lower elevations. Both RGR and tree height were positively
correlated with LMF and negatively with RMFꎬ but no significant positive correlations were found with maximum pho ̄
tosynthetic rate determined on both an area ̄basis (Amax) and mass ̄basis (Amass). These findings suggest that the
patterns of biomass partitioning and long ̄term morphological traits are better predictors of performance among trees of
different Pinus species growing along an elevational gradient.
Key words: Biomass partitioningꎻ Elevationꎻ Growthꎻ Photosynthesisꎻ Pinus
植 物 分 类 与 资 源 学 报 2014ꎬ 36 (1): 47~55
Plant Diversity and Resources DOI: 10.7677 / ynzwyj201413054
∗ Funding: National Natural Science Foundation of China (31170315)ꎬ and the West Light Foundation of the Chinese Academy of Sciences
Received date: 2013-03-21ꎬ Accepted date: 2013-08-09
作者简介: 张石宝 (1970-) 男ꎬ 研究员ꎬ 主要从事植物生理生态研究ꎮ E ̄mail: sbzhang@ mail. kib. ac. cn
Wide variations in growth rates among trees
across species are closely correlated with investments
of carbon resources and ecological distributions
(Wright and Westobyꎬ 2000ꎻ Poorter and Garnierꎬ
2007ꎻ King et al.ꎬ 2013). Fast ̄growing species are
generally found in relatively more fertile habitatsꎬ
whereas species that occupy infertile environments
tend to have low growth rates (Biereꎬ 1996ꎻ Poorter
and Garnierꎬ 2007). Those rates can affect seedling
survivalꎬ reproductionꎬ productivityꎬ competitionꎬ
and forest structures (Lambers and Poorterꎬ 1992ꎻ
Biereꎬ 1996ꎻ Coomes and Allenꎬ 2007). Thereforeꎬ
examining the factors that influence tree growth is
essential for modeling forest productivity and its
functioning under climatic change (Coomes and Al ̄
lenꎬ 2007ꎻ King et al.ꎬ 2013).
Relative growth rate (RGR) is a complex trait
that is determined by differences in physiologyꎬ mor ̄
phologyꎬ and biomass partitioning (Shipleyꎬ 2006ꎻ
Poorter and Garnierꎬ 2007ꎻ Poorter et al.ꎬ 2012).
This trait encompasses three variable components:
net assimilation rate ( NAR)ꎬ specific leaf area
(SLA)ꎬ and leaf mass ratio (LMR) (Poorter and
Garnierꎬ 2007ꎻ Poorter et al.ꎬ 2012ꎻ Tomlinson et
al.ꎬ 2012). For exampleꎬ species from humid envi ̄
ronments partition more biomass to the roots and less
to the stems than those from semiarid environments
in Australiaꎬ Africaꎬ and South America (Tomlinson
et al.ꎬ 2012). Poorter and Remkes ( 1990) have
foundꎬ in a controlled experimentꎬ that the RGRs of
69 plant species are most strongly correlated with
SLA and LMRꎬ but are also positively associated
with biomass allocation to the leaves and NAR. How ̄
everꎬ based on the results of a meta ̄analysis with
614 species from 83 different experiments in Eu ̄
ropeꎬ Americaꎬ and Australiaꎬ Shipley (2006) has
suggested that LMR is never strongly related to RGR
butꎬ insteadꎬ NAR is the best general predictor of
variations in interspecific RGRs. Consequentlyꎬ the
relative contribution of SLAꎬ NARꎬ and LMR to
RGR varies among species or when plants are grown
in contrasting environments.
Trees in the Pinus genus cover a large geographi ̄
cal area and a wide variety of habitats in southwes ̄
tern Chinaꎬ where they are of great economic and ec ̄
ological importance. As the dominant component of
conifer forests in that regionꎬ the three vicariant spe ̄
ciesꎬ from low to high elevationꎬ are P. kesiya Royle
ex Gordon var. langbianensis Gaussenꎬ P. yunnanensis
Franch.ꎬ and P. densata Mast. (Wuꎬ 1990). These
species exhibit different rates of development in their
habitatsꎬ especially in height incrementꎬ with P.
kesiya being the fastest grower (Dai et al.ꎬ 2006).
Howeverꎬ the physiological and morphological causes
underlying those differences are not completely known.
Plant heights and biomass production in trees
generally decline with increasing elevationꎬ as shown
in both common garden studies and natural forests
(Oleksyn et al.ꎬ 1998ꎻ Li et al.ꎬ 2003ꎻ Coomes and
Allenꎬ 2007). Howeverꎬ the extent of the correlation
between elevation and annual stem elongation can dif ̄
fer among species ( Angertꎬ 2006). For exampleꎬ
highland plants might display slower growth rates be ̄
cause of morphological or physiological reasons (At ̄
kin et al.ꎬ 1996aꎻ Oleksyn et al.ꎬ 1998ꎻ Angertꎬ
2006). Hoch et al. (2002) have suggested that a low
temperature ̄driven sink explains the lag in growth by
P. cembra at higher elevations at the tree line in the
Swiss Alps. Although some studies demonstrated that
leaf thicknessꎬ leaf N contentꎬ and photosynthetic
rates are greater in highland plants than in lowland
plants on the island of Hawaii or the Northern Ameri ̄
ca continent (Cordell et al.ꎬ 1999ꎻ Hultine and Mar ̄
shallꎬ 2000)ꎬ conflicting results from other investiga ̄
tions shown that a general trend in photosynthetic ca ̄
pacity does not occur across elevations in the tropical
high Andes and southwest China ( Cabrera et al.ꎬ
1998ꎻ Zhang et al.ꎬ 2007ꎻ 2011). For exampleꎬ
clones of spruce from the tree line have a 4. 3 ̄fold
lower growth rate and they contain 60% less chloro ̄
phyll per unit mass than trees from valleys in Europe
(Polle et al.ꎬ 1999). Westbeek et al. (1999) have
reportedꎬ from a common gardenꎬ that RGR is nega ̄
tively correlated with leaf N content per unit areaꎬ as
84 植 物 分 类 与 资 源 学 报 第 36卷
well as with chlorophyll and Rubisco contentsꎬ for
Poa species in both alpine and lowland regions. De ̄
spite greater photosynthetic rates in high ̄elevation
populationsꎬ seedling heights and dry masses meas ̄
ured in common garden studies decline with the ele ̄
vation at which their seed originated. Proportional dry
mass partitioning in the roots nearly doubles with in ̄
creasing elevation of origin in controlled experiments
(Oleksyn et al.ꎬ 1998). Consequentlyꎬ researchers
still do not have a clear understanding of the causes
for low relative growth rates by highland plants.
Hereꎬ the growth ratesꎬ leaf N contentsꎬ spe ̄
cific leaf areasꎬ biomass partitioningꎬ and photosyn ̄
thetic rates of P. kesiyaꎬ P. yunnanensis and P. den ̄
sata were monitored in a common garden at Kunming
Botanical Gardenꎬ China. The objective was to iden ̄
tify the factors affecting the development of pines
that originate from different elevations. We hypothe ̄
sized that there is an inherent difference in the rela ̄
tive growth rate of these three speciesꎬ and that this
difference is correlated with leaf physiology and bio ̄
mass partitioning.
1 Materials and methods
1 1 Plant species and sites
Three species in Pinus—P. kesiyaꎬ P. yunnanen ̄
sisꎬ and P. densata—were selected. All are evergreen
species that dominate conifer forestsꎬ from low to
high elevationꎬ respectivelyꎬ in southwestern China
(Table 1). Seeds of P. kesiyaꎬ P. yunnanensisꎬ and
P. densata were collected from Puwen (elev. 1 377
m)ꎬ Jianchuan (2 196 m)ꎬ and Deqing (3 447 m)ꎬ
respectivelyꎬ from October to December in 2006.
They were sown at Kunming Botanical Garden (elev.
1 990 mꎻ E 102 74°ꎬ N 25 15°) on 22 March 2007
and had germinated by 9 April. 70 seedlings per
species were transplanted on 7 May into plastic pots
filled with a peatꎬ forest soilꎬ and humus mixture
(1 ∶ 3 ∶ 1ꎻ v ∶ v ∶ v). The initial dry mass of sample
seedlings from each species was measured 1 d before
transplanting occurredꎬ and the final harvest was
made on 12 October 2007. After transplantingꎬ the
seedlings were grown under full sunlightꎬ and were
fertilized monthly with a liquid nutrient solution and
watered every 2 to 3 d. From May to Octoberꎬ the mean
monthly temperature ranged from 13 2 ℃ to 19 9 ℃
(mean ±SDꎬ 17 6 ± 2 4℃) while mean monthly pre ̄
cipitation was 19 6 mm to 204 0 mm (mean ± SDꎬ
115 7±74 6 mm).
1 2 Physiological measurements
Gas exchange in response to light and CO2 con ̄
centration was recorded on 7 to 9 October 2007 from
fully expanded leavesꎬ using a LI ̄6400 portable
photosynthesis system (LI ̄CORꎬ Lincolnꎬ NEꎬ USA).
Data were recorded from 10 plants per species. Be ̄
fore the measurements beganꎬ each sample leaf was
exposed to actinic light of 1 200 μmol m-2s-1 (10%
blueꎬ 90% red) for 15 min to induce maximum
stomatal opening. Curves for the photosynthetic light
response (A ̄PPFD) were made using an automated
protocol built into the LI ̄6400. The program was
configured to advance to the next step if the sum of
three coefficients of variation ( CO2ꎬ water vaporꎬ
and flow rate) was less than 0 3%. The minimum
waiting time was 3 min. Each leaf was equilibrated to
initial conditions by waiting at least 15 min before
executing the automated protocol. A ̄PPFD curves
were generated for 6 leaves per speciesꎬ at light in ̄
tensities of 2 000ꎬ 1 600ꎬ 1 200ꎬ 1 000ꎬ 800ꎬ 600ꎬ
400ꎬ 300ꎬ 200ꎬ 100ꎬ 50ꎬ and 0 μmol m-2s-1 . Other
test conditions included a controlled level of CO2
(380 μmol mol-1)ꎬ a flow rate of 500 μmol s-1ꎬ a
Table 1 Origins of the three Pinus species used in this study
Species Code Elevation range / m
Site for seed collection
Longitude Latitude Elevation / m
P. kesiya Pk 600-1600 E100°55′ N22°22′ 1377
P. yunnanensis Py 1000-3200 E99°54′ N26°32′ 2196
P. densata Pd 2600-3600 E98°52′ N28°27′ 3447
941期 ZHANG Shi ̄Bao: Biomass Partitioning Affects the Growth of Pinus Species from Different Elevations
leaf temperature of 25 ℃ꎬ and a vapor pressure defi ̄
cit (VPD) of 0 6 to 1 0 kPa.
Photosynthetic CO2 response curves (A ̄C i) and
A ̄PPFD curves were determined for the same leaves.
After measurements were completed for the A ̄PPFD
curveꎬ each leaf was exposed for 15 min to a light
intensity of 1 200 μmol m-2s-1 and a CO2 concentra ̄
tion of 380 μmol mol-1 . Other conditions were a leaf
temperature of 25 ℃ and VPD of 1 0 to 1 5 kPa. The
A ̄C i curve measurement was started at ambient CO2
concentrationꎬ which decreased gradually to 0 μmol
mol-1ꎬ returned to 380 μmol mol-1ꎬ and then in ̄
creased to a higher concentration to ensure that the
stomata remained open throughout the recording pe ̄
riod. Photosynthetic rates were measured at different
CO2 concentrationsꎬ using the automated protocol
built into the LI ̄6400.
After completing the gas exchange measure ̄
mentsꎬ we recalculated the photosynthetic rate based
on actual leaf area valuesꎬ which were estimated by
the method of Johnson (1984):
LA = 2L(1 + π / n) (nv / πL) (1)
where LA is leaf area ( cm-2)ꎬ L is needle length
(cm)ꎬ n is the number of needles per fascicle (n=
3 for P. kesiya and P. yunnanensisꎻ n= 5 for P. den ̄
sata)ꎬ and needle volume (v) is estimated via water
displacement (Johnsonꎬ 1984).
Photosynthetic light response curves were fitted
with non ̄rectangular hyperbola. The maximum pho ̄
tosynthetic rate ( Amax ) and respiration rate ( Rd )
were determined via Photosyn Assistant v1 1 (Dund ̄
ee Scientificꎬ Dundeeꎬ Scotlandꎬ UK) according to
the method of Prioul and Chartier (1977). Using A ̄
C i curvesꎬ we calculated the maximum carboxylation
rate by Rubisco (Vcmax) and light ̄saturated electron
transport ( Jmax) with Photosyn Assistantꎬ based on
the photosynthetic model of von Caemmerer and Far ̄
quhar (1981).
Mesophyll diffusion conductance (gm) from the
internal air space to the chloroplasts was estimated
according to the method of Harley et al. (1992):
gm =
A
C i -
Γ∗[Jmax + 8(A + Rd)]
Jmax - 4(A + Rd)
(2)
where the value for Rd is found from the A ̄PPFD
curveꎬ and Г∗ is the hypothetical CO2 compensation
point in the absence of Rd (42.75 μmol mol ̄1 at 25℃)
(Bernacchi et al.ꎬ 2001). Values for gm were calcu ̄
lated from our measurements of photosynthesis at in ̄
ternal CO2 concentrations of 100 to 300 μmol mol
-1ꎬ
with the average value of gm being determined for
each leaf.
Chlorophyll was extracted per the technique of
Moran and Porath ( 1980)ꎬ and its concentration
was analyzed on a UV ̄2550 spectrophotometer (Shi ̄
madzuꎬ Japan) before being calculated according to
the equations of Inskeep and Bloom (1985).
1 3 Growth rate and biomass partitioning
After the photosynthetic data were obtainedꎬ the
plants were harvested and divided into rootꎬ stemꎬ
and leaf portions. Ten plants were measured for each
species. The dry mass of each portion was recorded
after the tissues were dried for 48 h at 80 ℃. After ̄
wardꎬ the leaf mass fraction (LMF)ꎬ stem mass frac ̄
tion (SMF)ꎬ and root mass fraction (RMF) were de ̄
termined in proportion to total dry mass per plant.
The values for relative growth rate were calculated by
the following formula (Poorter and Garnierꎬ 2007).
RGR =
lnM2 - lnM1
t2 - t1
(3)
where M1 and M2 are the biomass at time t1 and t2ꎬ
respectivelyꎻ and t2 - t1 is the time span between
measuring events. Hereꎬ that span was 185 d.
Leaf area per plant was determined according to
the method of Johnson (1984). From thisꎬ SLA was
calculated as the specific leaf area per unit mass
(m2 kg-2). Leaf N content was assessed with a Leco
FP ̄428 nitrogen analyzer (Leco Corporationꎬ St. Jo ̄
sephꎬ MIꎬ USA).
1 4 Statistical analysis
Statistical analysis was performed with SPSS 13 0
(SPSS Inc.ꎬ Chicagoꎬ ILꎬ USA). We used one ̄way
ANOVA and LSD multiple comparison tests to esti ̄
05 植 物 分 类 与 资 源 学 报 第 36卷
mate the differences in RGRꎬ biomass partitioningꎬ
physiological parametersꎬ and leaf N content. Rela ̄
tionships between RGR and leaf traits were assessed
through Pearson’ s regression analysis. A principal
component analysis (PCA) was performed to char ̄
acterize the associations among 17 leaf traits.
2 Results
2 1 Interspecific variations in growthꎬ biomass
partitioningꎬ and physiology
Tree height (H) and dry mass per plant varied
significantly across species ( Fig. 1). For exampleꎬ
plants of P. kesiya were taller than those of P. yun ̄
nanensis and P. densata (P<0 001) whereas the mean
basal diameter from P. yunnanensis (1 179±0 140 cm)
was larger than from P. densata (0 747±0 073 cm).
Howeverꎬ the dry mass per plant and the RGRs of
P. kesiya and P. yunnanensis were greater than those
of P. densata (Fig 1). Consequentlyꎬ species from
the lower elevations grew more rapidly than the one
Fig 1 Differences in plant heightꎬ dry mass weight (DW)ꎬ relative
growth rate (RGR)ꎬ leaf mass fraction ( LMF)ꎬ stem mass fraction
(SMF)ꎬ and root mass fraction (RMF) for Pinus kesiya ( Pk)ꎬ P.
yunnanensis (Py)ꎬ and P. densata (Pd) grown under the same envi ̄
ronmental conditions. Different letters above bars for each component
indicate statistically significant differences in mean values (P≤0 05)ꎬ
as determined by LSD multiple comparison tests
from the highest elevation.
Biomass partitioning to the organs also differed
significantly among species (Fig 1). Whereas the frac ̄
tions of dry matter produced in the leaf and stem were
higher in P. kesiya and P. yunnanensis than in P. densa ̄
taꎬ the root mass fraction showed an opposite trend.
Significant differences among species were found
for the area ̄based respiration rate (Rd) and stomatal
conductance (gs)ꎬ but not for the area ̄based photo ̄
synthetic rate ( Amax )ꎬ mass ̄based photosynthetic
rate (Amass)ꎬ mass ̄based respiration rate (Rd ̄mass)ꎬ
maximum carboxylation rate (Vcmax)ꎬ light ̄saturated
electron transport rate ( Jmax)ꎬ and mesophyll con ̄
ductance (gm) (Fig 2).
Fig 2 Comparisons of area ̄based photosynthetic rate (Amax)ꎬ mass ̄
based photosynthetic rate ( Amass )ꎬ maximum carboxylation rate
(Vcmax)ꎬ light ̄saturated electron transport rate ( Jmax )ꎬ area ̄based
respiration rate (Rd)ꎬ mass ̄based respiration rate (Rd ̄mass)ꎬ stomatal
conductance ( gs )ꎬ and mesophyll conductance ( gm ) among Pinus
kesiya (Pk)ꎬ P. yunnanensis (Py)ꎬ and P. densata (Pd). Different
letters above bars for each component indicate statistically significant
differences in mean values (P≤0 05)ꎬ as determined by LSD multi ̄
ple comparison tests
151期 ZHANG Shi ̄Bao: Biomass Partitioning Affects the Growth of Pinus Species from Different Elevations
Specific leaf area per unit mass ( SLA) was
lower in species from the higher elevationꎬ but no
significant difference in chlorophyll content was de ̄
tected among species (Fig 3). Both leaf N content
and C content per unit area increased generally with
elevation (Fig 3). Howeverꎬ the mass ̄based N con ̄
tents did not differ significantly among species.
Fig 3 Differences in specific leaf area ( SLA)ꎬ chlorophyll content
per unit mass (Chl ̄mass)ꎬ leaf N contents on area ̄basis (Narea ) and
mass ̄basis (Nmass)ꎬ and leaf C contents on area ̄basis (Carea ) and
mass ̄basis (Cmass) from Pinus kesiya (Pk)ꎬ P. yunnanensis (Py)ꎬ
and P. densata (Pd). Different letters above bars for each component
indicate statistically significant differences in mean values (P≤0 05)ꎬ
as determined by LSD multiple comparison tests
2 2 Determinants of growth rate
Tree height was correlated significantly and posi ̄
tively with LMFꎬ SMFꎬ SLA and RGRꎬ but nega ̄
tively with RMF (Fig 4). No significant correlation
existed between H and Amax .
Dry weight per plant was correlated positively
with LMF but negatively with RMF. Neither Rd nor
Amax was correlated with dry weight ( Fig 5). RGR
was significantly and positively correlated with LMF
but negatively with RMF. No significant correlations
were found between RGR and Amax or Amass (Fig 6).
Finallyꎬ Amax was positively correlated with Vcmaxꎬ
Jmaxꎬ gsꎬ gmꎬ and leaf N content per unit areaꎬ but
negatively with SLA (data not shown).
Fig 4 Pearson correlations of plant height with leaf mass fraction
(LMF)ꎬ stem mass fraction (SMF)ꎬ root mass fraction (RMF)ꎬ rel ̄
ative growth rate (RGR)ꎬ maximum photosynthetic rate (Amax)ꎬ and
specific leaf area (SLA) for 3 Pinus species
Fig 5 Pearson correlations of dry mass weight (DW) with leaf mass
fraction (LMF)ꎬ root mass fraction (RMF)ꎬ maximum photosynthetic
rate (Amax)ꎬ and respiration rate (Rd) for 3 Pinus species
Principal component analysis showed that Hꎬ
RGRꎬ LMFꎬ RSFꎬ RMFꎬ and SLA loaded mainly
on the first PCA axisꎬ explaining 36 6% of the total
variationꎻ Amaxꎬ Jmaxꎬ Vcmaxꎬ gsꎬ gmꎬ and leaf N con ̄
tent per unit area (Narea) loaded on the second axisꎬ
explaining 26 3% of the total (Fig 7). The first axis
of the PCA was mainly associated with plant growth
and biomass partitioning while the second axis was
associated with photosynthetic carbon assimilation.
25 植 物 分 类 与 资 源 学 报 第 36卷
Fig 6 Pearson correlations of relative growth rate (RGR) with leaf
mass fraction (LMF)ꎬ root mass fraction (RMF)ꎬ mass ̄based maxi ̄
mum photosynthetic rate (Amass)ꎬ and area ̄based maximum photosyn ̄
thetic rate (Amax) for 3 Pinus species
Fig 7 Principle component analysis of 17 leaf traits from 3 Pinus
species. Agrossꎬ gross photosynthetic rateꎻ Amaxꎬ area ̄based maximum
photosynthetic rateꎻ DWꎬ dry mass weight per plantꎻ gmꎬ mesophyll
conductanceꎻ gsꎬ stomatal conductanceꎻ Hꎬ tree heightꎻ Jmaxꎬ light ̄
saturated electron transport rateꎻ LMFꎬ leaf mass fractionꎻ Nareaꎬ ni ̄
trogen content per unit areaꎻ Rdꎬ dark ̄respiration rateꎻ RGRꎬ relative
growth rateꎻ RMFꎬ root mass fractionꎻ Rlꎬ photorespiration rateꎻ
RSFꎬ root ̄shoot ratioꎻ SLAꎬ specific leaf area per unit massꎻ SMFꎬ
stem mass fractionꎻ Vcmaxꎬ maximum carboxylation rate
3 Discussion
3 1 Growth rate in relation to elevation
Seedlings of Pinus species originating from the
highest elevation were shorter and had smaller values
for RGR than congeneric species from the lowest ele ̄
vation. This indicated that the former grew more
slowly. These findings are in line with previous re ̄
ports from common gardens and natural forests that
tree height and biomass production are reduced as
elevation increases (Oleksyn et al.ꎬ 1998ꎻ Li et al.ꎬ
2004ꎻ Coomes and Allenꎬ 2007). Li et al. (2003)
have found that the mean annual biomass increment
per tree in the eastern Himalayas is less as elevation
increasesꎬ i.e.ꎬ reduced by 476 g per 100 m between
1 680 and 1 810 m and by 103 g between 1 810 and
1 940 m. Howeverꎬ Angert ( 2006) has shown in
common garden experiments that trees within a given
species accumulate their greatest aboveground bio ̄
mass when grown under a temperature regime that is
characteristic of the center of its natural elevation
range. Thusꎬ the effect of seed origin or elevation on
RGR is considered species ̄specific (Angertꎬ 2006).
3 2 Growth rate in relation to biomass partitioning
The decline in growth rate with increasing eleva ̄
tion can be caused by both environmental and genetic
factors (Körnerꎬ 2003ꎻ Li et al.ꎬ 2003ꎻ King et al.ꎬ
2013). For exampleꎬ in natural forestsꎬ a shorter
growing season that results from low temperatures at a
higher elevation can reduce a tree’s accumulation of
C and its growth rate (Oleksyn et al.ꎬ 1998ꎻ Körnerꎬ
2003). Because seeds of the trees used in our study
were collected at different elevations but were then
exposed to the same experimental conditionsꎬ any
interspecific variations in tree growth rates would
have reflected inherent genetic differences.
Present study demonstrated that tree heightꎬ dry
weightꎬ and relative growth rate were correlated posi ̄
tively with the leaf mass fractionꎬ but negatively with
the root mass fraction. Previous studies under con ̄
trolled conditions have also revealed that RGR is pos ̄
itively associated with biomass partitioning in the
leaves (Poorter and Remkesꎬ 1990ꎻ Wright and Westo ̄
byꎬ 2000). By contrastꎬ Shipley (2006) has noted
no significant correlation between RGR and LMF in
83 different experiments. The essential resources
needed by plants are acquired by different organsꎻ
more resources allocated to the leaves can increase
light captureꎬ while more resources transported to
the roots can improve the uptake of water and miner ̄
al elements (Taubꎬ 2004). Because plants are com ̄
posed largely of materials derived via photosynthesisꎬ
an increased partitioning to non ̄photosynthetic organs
would reduce their growth rate in general (Mooneyꎬ
1972). Consequentlyꎬ the functional balance between
351期 ZHANG Shi ̄Bao: Biomass Partitioning Affects the Growth of Pinus Species from Different Elevations
leaves and roots is an important determinant of inter ̄
specific differences in RGR (Wright and Westobyꎬ
2000ꎻ Osone et al.ꎬ 2008).
Biomass partitioning in different organs may re ̄
flect a survival strategy that is adjusted to meet cur ̄
rent environmental conditions (Poorter et al.ꎬ 2012).
The process of allocating C to either the roots or
shoots is more sensitive to temperatureꎬ shadeꎬ and
water deficit (Poorter et al.ꎬ 2012ꎻ Tomlinson et al.ꎬ
2012). Generallyꎬ plants in cold environments tend
to accumulate more dry matter in the rootsꎬ boosting
their production of those tissues ( Oleksyn et al.ꎬ
1998ꎻ Li et al.ꎬ 2004)ꎬ while also reducing their
stem and leaf fractions to increase RMF values
(Poorter et al.ꎬ 2012). Plants that decrease the pro ̄
portion of leaves and partition more biomass to their
roots at higher elevations or on less productive sites
will benefit from faster recovery of their growth rates
the following yearꎬ and will be better able to adapt to
unfavorable climatic conditions ( Oleksyn et al.ꎬ
1998ꎻ Körnerꎬ 2003). Howeverꎬ improved survivabil ̄
ity comes at the expense of slower growth because
plants must respond to environmental gradients by ad ̄
justing their pattern of partitioning biomass to maxi ̄
mize growth performance (Wang et al.ꎬ 2008).
3 3 Correlation of growth rate with photosyn ̄
thetic rate
Maximum photosynthetic rate was not correlated
with tree heightꎬ plant dry weightꎬ or RGR in our
three Pinus species. These results contradict previous
findings from controlled experiments that RGR is cor ̄
related with the photosynthetic rate ( Lambers and
Poorterꎬ 1992ꎻ Atkin et al.ꎬ 1996bꎻ Shipleyꎬ 2006).
Villar et al. (2005) have also suggested that RGR is
correlated with NAR during a short growing periodꎬ
but with morphological traits over the long term. How ̄
everꎬ maximum photosynthetic rates for the three spe ̄
cies did not vary with elevationꎬ but were positively
correlated with CO2 diffusive conductance and leaf N
content per unit area while negatively correlated with
SLA. Earlier researchers have implied that no gener ̄
al trend in photosynthetic rate and leaf traits exists
across elevations in natural habitats (Cabrera et al.ꎬ
1998ꎬ Zhang et al.ꎬ 2007ꎻ 2011). Villar et al. (1998)
have suggested that the photosynthetic rate in Ae ̄
gilops species is negatively correlated with SLA in a
common gardenꎬ thereby accounting for the absence
of a correlation between SLA and RGR.
In conclusionꎬ the pines grown from seed gath ̄
ered at higher elevations had slower growth ratesꎬ
but their photosynthetic capacity was not significantly
inferior to that of trees representing lower elevations.
We believe thatꎬ at higher elevationsꎬ plants tend to
allocate more biomass to the roots while reducing
their leaf productionꎬ which ultimately leads to slo ̄
wer growth. The pattern of biomass partitioningꎬ
rather than photosynthetic capacityꎬ determines any
interspecific variations in growth rates among Pinus
species originating from different elevations. Thusꎬ
long ̄term morphological traits are better predictors of
growth rates for trees along an elevational gradient.
References:
Angert ALꎬ 2006. Growth and leaf physiology of monkeyflowers with
different altitude ranges [J] . Oecologiaꎬ 148: 183—194
Atkin OKꎬ Botman Bꎬ Lambers Hꎬ 1996a. The causes of inherently
slow growth in alpine plants: an analysis based on the underlying
carbon economies of alpine and lowland Poa species [ J] . Func ̄
tional Ecologyꎬ 10: 698—707
Atkin OKꎬ Botman Bꎬ Lambers Hꎬ 1996b. The relationship between
relative growth rate and nitrogen economy of alpine and lowland
Poa species [J] . Plantꎬ Cell and Environmentꎬ 19: 1324—1330
Bernacchi CJꎬ Singsaas ELꎬ Pimentel C et al.ꎬ 2001. Improved tem ̄
perature response functions for models of Rubisco ̄limited photo ̄
synthesis [J] . Plantꎬ Cell and Environmentꎬ 24: 253—259
Biere Aꎬ 1996. Intra ̄specific variation in relative growth rate: impact
on competitive ability and performance of Lychnis flos ̄cuculi in hab ̄
itats differing in soil fertility [J]. Plant and Soilꎬ 182: 313—327
Cabrera HMꎬ Rada Fꎬ Cavieres Lꎬ 1998. Effects of temperature on
photosynthesis of two morphologically contrasting plant species a ̄
long an altitudinal gradient in the tropical high Andes [J] . Oeco ̄
logiaꎬ 114: 145—152
Coomes DAꎬ Allen Rꎬ 2007. Effect of sizeꎬ competition and altitude
on tree growth [J] . Journal of Ecologyꎬ 95: 1084—1097
Cordell Sꎬ Goldstein Gꎬ Meinzer FC et al.ꎬ 1999. Allocation of nitro ̄
gen and carbon in leaves of Metrosideros polymorpha regulates car ̄
boxylation capacity and δ13C along an altitudinal gradient [ J] .
Functional Ecologyꎬ 13: 811—818
45 植 物 分 类 与 资 源 学 报 第 36卷
Dai KJ (戴开杰)ꎬ He F (何方)ꎬ Shen YX (沈有信) et al.ꎬ 2006.
Advances in the research on Pinus yunnanensis forest [ J] . Jour ̄
nal of Central South Forestry Universityꎬ 26: 138—143
Harley PCꎬ Loreto Fꎬ di Marco G et al.ꎬ 1992. Theoretical considera ̄
tions when estimating the mesophyll conductance to CO2 flux by
analysis of the response of photosynthesis to CO2 [ J] . Plant
Physiologyꎬ 98: 1429—1436
Hoch Gꎬ Popp Mꎬ Körner Cꎬ 2002. Altitudinal increase of mobile
carbon pools in Pinus cembra suggests sink limitation of growth
at the Swiss treeline [J] . Oikosꎬ 98: 361—374
Hultine KRꎬ Marshall JDꎬ 2000. Altitude trends in conifer leaf morphology
and stable carbon isotope composition [J]. Oecologiaꎬ 123: 32—40
Inskeep WRꎬ Bloom PRꎬ 1985. Extinction coefficients of chlorophyll
a and b in Nꎬ N ̄dimethylformamide and 80% acetone [J] . Plant
Physiologyꎬ 77: 483—485
Johnson JDꎬ 1984. A rapid technique for estimating total surface area
of pine needles [J] . Forest Scienceꎬ 30: 913—921
King GMꎬ Gugerli Fꎬ Fonti P et al.ꎬ 2013. Tree growth response a ̄
long an elevational gradient: climate or genetics? [ J] . Oecolo ̄
giaꎬ 173: 1587—1600
Körner CHꎬ 2003. Alpine Plant Life: Functional Plant Ecology of
High Mountain Ecosystems [M]. Heidelberg: Springer ̄Verlag
Lambers Hꎬ Poorter Hꎬ 1992. Inherent variation in growth rate between
higher plants: a search for physiological causes and ecological con ̄
sequences [J]. Advances in Ecological Researchꎬ 23: 187—261
Li Cꎬ Liu Sꎬ Berninger Fꎬ 2004. Picea seedlings show apparent accli ̄
mation to drought with increasing altitude in the eastern Himalaya
[J] . Treesꎬ 18: 277—283
Li MHꎬ Yang Jꎬ Kräuchi Nꎬ 2003. Growth responses of Picea abies
and Larix decidua to elevation in subalpine areas of Tyrolꎬ Austria
[J] . Canadian Journal of Forest Researchꎬ 33: 653—662
Mooney HAꎬ 1972. Carbon balance of plants [ J] . Annual Review of
Ecology and Systematicsꎬ 3: 315—346
Moran Rꎬ Porath Dꎬ 1980. Chlorophyll determination in intact tissue using
Nꎬ Nꎬ  ̄dimethylformamide [J]. Plant Physiologyꎬ 65: 478—479
Oleksyn Jꎬ Modrzynski Jꎬ Tjoelker MG et al.ꎬ 1998. Growth and
physiology of Picea abies populations from elevational transects:
common garden evidence for altitudinal ecotypes and cold adapta ̄
tion [J] . Functional Ecologyꎬ 12: 573—590
Osone Yꎬ Ishida Aꎬ Tateno Mꎬ 2008. Correlation between relative
growth rate and specific leaf area requires associations of specific
leaf area with nitrogen absorption rate of roots [J] . New Phytolo ̄
gistꎬ 179: 417—427
Polle Aꎬ Baumbusch LOꎬ Oschinski C et al.ꎬ 1999. Growth and protec ̄
tion against oxidate stress in young clones and mature spruce trees
(Picea abies L.) at high altitudes [J]. Oecologiaꎬ 121: 149—156
Poorter Hꎬ Garnier Eꎬ 2007. Ecological significance of inherent varia ̄
tion in relative growth rate and its components [A]. In: Vallada ̄
res Fꎬ Pugnaire FI (eds.)ꎬ Handbook of Functional Plant Ecolo ̄
gy [M]. London: CRC Press
Poorter Hꎬ Remkes Cꎬ 1990. Leaf area ratio and net assimilation rate
of 24 wild species differing in relative growth rate [ J] . Oecolo ̄
giaꎬ 83: 553—559
Poorter Hꎬ Niklas KJꎬ Reich PB et al.ꎬ 2012. Biomass allocation to
leavesꎬ stems and roots: meta ̄analyses of interspecific variation
and environmental control [J] . New Phytologistꎬ 193: 30—50
Prioul JLꎬ Chartier Pꎬ 1977. Partitioning of transfer and carboxylation com ̄
ponents of intracellular resistance to photosynthetic CO2 fixation: a criti ̄
cal analysis of the methods used [J]. Annals of Botanyꎬ 41: 789—800
Shipley Bꎬ 2006. Net assimilation rateꎬ specific leaf area and leaf
mass ratio: which is most closely correlated with relative growth
rate? A meta ̄analysis [J] . Functional Ecologyꎬ 20: 565—574
Taub DRꎬ 2004. A meta ̄analysis of studies on plant growth rate and
allocation to roots vs. shoots [A]. Brown Working Papers in the
Arts and Sciencesꎬ Southwestern University [M]ꎬ Vol 4: 1. A ̄
vailable from: http: / / www. southwestern. edu / academics / bwp /
pdf / 2004bwp ̄taub.pdf
Tomlinson KWꎬ Sterck FJꎬ Bongers F et al.ꎬ 2012. Biomass partitio ̄
ning and root morphology of savanna trees across a water gradient
[J] . Journal of Ecologyꎬ 100: 1113—1121
Villar Rꎬ Veneklaas EJꎬ Jordano P et al.ꎬ 1998. Relative growth rate
and biomass allocation in 20 Aegilops ( Poaceae) species [ J] .
New Phytologistꎬ 140: 425—437
Villar Rꎬ Marañón Tꎬ Quero JL et al.ꎬ 2005. Variation in relative
growth rate of 20 Aegilops species (Poaceae) in the field: the im ̄
portance of net assimilation rate or specific leaf area depends on
time scale [J] . Plant and Soilꎬ 272: 11—27
von Caemmerer Sꎬ Farquhar GDꎬ 1981. Some relationships between
the biochemistry of photosynthesis and the gas exchange rates of
leaves [J] . Plantaꎬ 153: 376—387
Wang Xꎬ Fang Jꎬ Zhu Bꎬ 2008. Forest biomass and root ̄shoot alloca ̄
tion in northeast China [ J] . Forest Ecology and Managementꎬ
255: 4007—4020
Westbeek MHMꎬ Pons TLꎬ Cambridge ML et al.ꎬ 1999. Analysis of
differences in photosynthetic nitrogen use efficiency of alpine and
lowland Poa species [J] . Oecologiaꎬ 120: 19—26
Wright IJꎬ Westoby Mꎬ 2000. Cross ̄species relationships between
seedlings relative growth rateꎬ nitrogen productivity and root vs
leaf function in Australian woody species [J] . Functional Ecolo ̄
gyꎬ 14: 97—107
Wu ZL (吴兆录)ꎬ 1990. Geographical distribution and leaf form of
Pinus in Yunnan [J] . Yunnan Forestry Science and Technology
(云南林业科技)ꎬ 3: 46—51
Zhang SBꎬ Zhou ZKꎬ Hu H et al.ꎬ 2007. Gas exchanges and resource uti ̄
lization of two alpine oaks at different altitudes in the Hengduan
Mountains [J]. Canadian Journal of Forest Researchꎬ 37: 1184—1193
Zhang SBꎬ Zhou ZKꎬ Xu Kꎬ 2011. Effects of altitude on photosyn ̄
thetic gas exchange and the associated leaf trait in an alpine oakꎬ
Quercus guyavifolia ( Fagaceae) [ J] . Plant Diversity and Re ̄
sourcesꎬ 33: 214—224
551期 ZHANG Shi ̄Bao: Biomass Partitioning Affects the Growth of Pinus Species from Different Elevations