全 文 ：Received 10 Oct. 2003 Accepted 27 Nov. 2003
Supported by the Knowledge Innovation Program of The Chinese Academy of Sciences (KSCX1-08) and the State Key Basic Research and
Development Plan of China (G2000018607).
* Author for correspondence. E-mail: .
http://www.chineseplantscience.com
Acta Botanica Sinica
植 物 学 报 2004, 46 (3): 259-270
Differential Responses of N:P Stoichiometry of Leymus chinensis and Carex
korshinskyi to N Additions in a Steppe Ecosystem in Nei Mongol
ZHANG Li-Xia, BAI Yong-Fei, HAN Xing-Guo*
(Laboratory of Quantitative Vegetation Ecology, Institute of Botany, The Chinese Academy of Sciences, Beijing 100093, China)
Abstract: The typical steppe ecosystems in China are now being increasingly degraded due mainly to
overgrazing. To determine the limiting nutrients is of significance in order to find out ways of successfully
restoring the degraded steppe. In addition to field fertilization experiments, N:P stoichiometry is an alternative,
but argumentative tool to study nutrient limitation. In this study, we used these two approaches to identify
the most limiting nutrient element at the species level. Furthermore, nutrient addition experiment
provides an effective means to test our hypothesis that N:P stoichiometry will remain constant because
relatively narrow range of N:P ratio in tissues of the terrestrial plants is an important adaptive mechanism
for plants to survive on earth. For these purposes, we designed a field experiment to examine the
responses of biomass and N:P stoichiometry of the two dominant species — Leymus chinensis (Trin.)
Tzvel. and Carex korshinskyi Kom. — to N fertilization at rates of 0, 5, 15, 30, 50 and 80 g NH4NO3.m-2.a-1
in two adjacent sites, one being excluded animal grazing for 22 years (site A), and another being free of
grazing for only two years (site B) before the experiment was carried out. No effects of N fertilization were
detected in the first year as reflected by the aboveground biomass and P concentrations of the two
species. The regression analysis showed that N:P ratios of two species of both sites remained constant in
the second year. N fertilization significantly increased the N concentrations of two species in both years,
while only significantly increased the P contents of the two species in the second year. N and P contents
of the two species were significantly correlated in all cases in 2001. Our results suggest that the
L. chinensis was in short of N in site B while the growth of C. korshinskyi was limited by P in site A, and
there is a significant synergistic relationship between tissue N and P concentrations in 2001. Our hypothesis
was valid on the species level since N:P ratio of the two species remained constant with increasing N
application rates after two years of fertilization. We argue that it may be inappropriate to define an ecosystem
which is limited by certain nutrient elements since the responses of coexisting species present in a
community to nutrient additions can vary tremendously.
Key words: typical steppe; nutrient addition; nutrient limitation; biomass
Leymus chinensis steppe is widely distributed in the
Eurasia steppe zone (Wu, 1980) and is one of the dominant
steppe types in the Xilin River basin, Nei Mongol, China.
In the past two decades, L. chinensis steppe ecosystem is
increasingly degraded due mainly to overgrazing (Li et al.,
2003). Overgrazing has not only drastically decreased eco-
system productivity (Li, 1997; Chen and Wang, 2000), al-
tered the species composition (Wang et al., 1996a, 1996b;
Wang et al., 1998), resulted in greater interannual variabil-
ity of biomass production (Li et al., 2002), but also greatly
decreased the site fertility because of increased nutrient
output through herbivory (Li et al., 2003), water and wind
soil erosion (Batjes, 1996; Mainguet and Da Silva, 1998),
increased nutrient mineralization rates (Glaser et al., 2001;
Brockway et al., 2002) or diminishing litter input (Facelli
and Pickett, 1991). Practically, destocking can be an
effective way to restore the degraded steppe ecosystems
of low or intermediate severity (Chen and Baoyin, 1997;
Chen and Wang, 2000). However, if the ecosystems are
extremely severely degraded other measures that can ac-
celerate the restoration processes should be adopted. We
hypothesized that because N might be the most limiting
nutrient element in arid and semi-arid regions, as reported
elsewhere (Vitousek and Howarth, 1991), N fertilization
would greatly enhance the total biomass production in this
region, which would in turn speed up the restoration
processes.
Nitrogen is one of the most limiting nutrients in most
terrestrial ecosystems (Vitousek and Howarth, 1991). Since
fertilization is the best approach to test the nutrient limita-
tion (Bedford et al., 1999), we try to determine the nutrient
status by using a field experiment. Being a new tool to
Acta Botanica Sinica 植物学报 Vol.46 No.3 2004260
study ecology from genes to the biosphere, N:P stoichiom-
etry can be applied widely in ecological research (Zhang et
al., 2003). Plants tend to keep a relative constant N:P ratio
to optimize metabolism, protein synthesis and tissue pro-
duction (Garten Jr., 1976; Duarte, 1992), thus a given N:P
ratio would determine N to P availability. N:P stoichiometry
in plant tissues can predict the potential responses of plants
to nutrient enrichment (Koerselman and Meuleman, 1996;
Bedford et al., 1999). Estimation of nutrient status with N:P
stoichiometry, which just considering relative proportions
(ratios) of N and P in organisms, is easier, more effective
and less laborious than field fertilization experiment.
However, the application of this method is disputable as
some researchers argue that N:P ratio may only reflect the
accumulation of the non-limiting nutrient rather than the
scarcity of the limiting one (Wassen et al., 1995; Pegtel et
al., 1996). In this article, we wanted to compare these two
approaches at the species level. Furthermore, fertilization
could change the nutrient availability, which affects
biomass, species composition and vegetation succession
(Chen et al., 1985; Chapin et al., 1986; Fisher et al., 1988;
DiTomasso and Aarssen, 1989; Story et al., 1989; Wilson
and Tilman, 1993; Chapin et al., 1995; Gough et al., 2000).
But studies of responses of the separate species of natural
ecosystem to nutrient addition in semi-arid region are scarce.
To test the effect of N fertilization on the biomass and
plants chemistry, we designed a field experiment to exam-
ine the responses of biomass and N:P stoichiometry of the
two ecologically significant species, L. chinensis and Carex
korshinskyi, to N fertilization at rates of 0, 5, 15, 30, 50 and
80 g NH4NO3 m-2.a-1 in two adjacent sites, one being
excluded animal grazing for 22 years (site A), and another
being free of grazing for only two years (site B) before the
experiment was carried out. The objectives of this study
were to address the following questions: (1) Does fertiliza-
tion change the N:P ratio of the plant tissue? (2) Can we
predict the limiting nutrient by referring to plant N:P
stoichiometry?
1 Materials and Methods
1.1 Site description
Our study was carried out in a typical steppe ecosystem
in proximity to the Inner Mongolia (Nei Mongol) Grassland
Ecosystem Research Station (IMGERS), which lies between
43o26-44o08 N and 116o04-117o05 E, with an average el-
evation of 1 200 m. The continental middle temperate semi-
arid climate dominates the area, characterized by a cold and
dry winter but a warm and moist summer (Chen, 1988). The
average annual precipitation was 345 mm and the mean
monthly temperature ranged from –21.5 ℃ (January) to 18.
9℃ (June), respectively, based on the meteorological record
of 1982-2001. It clearly demonstrated that the temperature
and precipitation during the time when this study was con-
ducted did not deviate significantly from normal fluctua-
tions based on our 20-year meteorological record (Fig.1).
Greening time of plants occurs at the end of April, and the
plants turn senescenced at the early October. Thus the
plant-growing season is from April to September, lasting
about 150 d (Jiang, 1985). About 88.0% of annual precipita-
tion was concentrated in the growing season (April to
September) during 1982-2001. Dark chestnut soil is the
major soil type, with an average depth of about 100 cm
(Wang and Cai, 1988) and a soil bulk density of 1.16 g/cm3
(Jia et al., 1997). The pH ranged between 7.22-8.71 (Gu and
Li, 1997).
1.2 Experimental design
The first experimental site (400 m ´ 600 m) was built in
1980 with an iron fence to exclude animal grazing. In 2000, a
new permanent experiment site (30 hm2) was set up stretch-
ing from the western side of the old one to further west (500
Fig.1. Means of A temperature and B precipitation per month during the growing periods in 2000 and 2001 in the study area compared
with the long-term mean values of 1982-1999.
ZHANG Li-Xia et al.: Differential Responses of N:P Stoichiometry of Leymus chinensis and Carex korshinskyi to N Additions
in a Steppe Ecosystem in Nei Mongol 261
m ´ 600 m), and also fenced for exclusion of animal grazing.
The experiments were carried out at these two adjacent
permanent experiment sites. We used these two adjacent
sites because they appeared relatively homogeneous in
topography, original species composition, and soil types,
but represented a climax community and the early stage of
succession, respectively. The study began at July of 2000.
Two sites that appeared uniform in plant species richness
and evenness in 1999 were selected for arrangement of the
plots (5 m ´ 5 m) with a buffer belt of 1 m.
1.3 Fertilization
Surface application of fertilizers was carried out at the
beginning of July of 2000 and 2001, because we believed
that high temperature and rainfall at this time would syn-
chronize the large requirement of nutrients by the fast grow-
ing plants. Nitrogen was manually applied to the plots as
commercial ammonium nitrate (NH4NO3) at rates of 0, 5, 15,
30, 50, 80 g NH4NO3·m-2·a-1 respectively, equivalent to 0,
17.5, 52.5, 100.5, 175.0 and 280.0 kg N·hm-2·a-1, respectively.
The experiment was repeated for nine times. We also added
phosphorus (10 g P2O5·m-2·a-1) and trace elements (Zn:
0.001 9 g·m-2·a-1, Mn: 0.001 9 g·m-2·a-1, B: 3.12×10-5
g·m-2·a-1) to make the N be the sole limiting factor. Bulk
trace element solution was made by adding ZnSO4·7H2O
(109.75 g), MnCl2·4H2O (77.00 g), and H3BO3 (3.00 g), and
finally made to a volume of 1 000 mL with deionized water.
The trace element solution was made by mixing 120 mL
of the stock solution with 33.60 g of citric acid dissolved in
deionized water, and brought to a total volume of 1 000 mL
with deionized water to form the working solution. The
working solution was applied at a rate of 0.062 5 mL·m-2·a-1.
This amount of working solution was first mixed with about
25 g of sand, and then thoroughly mixed with N fertilizer.
For each treatment, we also had a control adding neither N
nor trace element solution.
1.4 Field sampling and analysis
The net primary production was estimated by clipping
off aboveground biomass in the 1 m× 0.5 m sampling
quadrat of each plot to soil surface in the early September
of 2000 and 2001 when maximum aboveground biomass
was reached. The sampling quadrat was 0.5 m apart from
the plot boundary in order to minimize the edge effect. The
height and the number of the plant species in each quadrat
were recorded at the same time. The clippings were hand-
sorted to plant species. The total tissues of each plant spe-
cies was dried at 65 ℃ in an oven for 48 h to constant mass,
weighted, grounded to 80 mesh and stored for chemical
analysis. For each plant species except a few species with
dry weight less than 1 g, the total N and P concentrations
were analyzed. To analyze the total N (TKN) and P of plant
tissues, we first digested 150 mg ground material in 5 mL 30
N sulfuric acid and H2O2 (with less than 2 mL as oxidant)
with a 2-stage increase in temperature to 360 ℃. The diges-
tion was completed with dropwise addition of H2O2 as de-
scribed by Lowther (1980). Each digest was cooled and
diluted in deionized water to make up to a known volume.
The diluted digests were then filtered through P-free filter
paper to get rid of the suspending material. The digested
solutions were diluted to concentrations appropriate for
analysis. We analyzed the diluted digestions for total N by
employing the micro-Kjeldahl method and colorimetrically
using the ammonium molybdate method for P (Rice, 1964;
Bradstreet, 1965; Horwitz, 1975). Nutrient concentrations
were expressed as %DW.
1.5 Statistical analysis
The effects of nutrient fertilization on the aboveground
biomass, N, P concentrations and N:P ratios in the plant
species were quantified as the fit (R2 adjusted for small
sample size) and as the slope of linear regressions based
on experimental variables (Güsewell and Koerselman, 2002).
We determined the effects of fertilization with significance
level P = 0.05, and the significance values were reported
when P < 0.05. We calculated the mean values of N, P con-
centrations and N:P ratio in each site and were then tested
for significant differences with t -test (only data from con-
trol plots).
2 Results
2.1 Aboveground biomass
N fertilization had insignificant effect (P > 0.05) on the
aboveground biomass of the two species in 2000 in both
sites. The aboveground biomass of L. chinensis in site B
increased, while that of C. korshinskyi in site A decreased
in the second year after N fertilization applied (Fig.2). In
2001, N fertilization accounted for 11.90% of the variation
in the aboveground biomass of C. korshinskyi in site A,
and explained 15.85% of the variation in the aboveground
biomass of L. chinensis in site B.
2.2 Variations of nitrogen and phosphorus concentra-
tions
N concentrations of L. chinensis and C. korshinskyi in
sites A and B significantly (P < 0.05) increased with in-
creasing N fertilization rates for two years (Fig.3), while P
concentrations of these two species showed slightly dif-
ferent responses. N fertilization had insignificant effect on
P concentration in both sites in 2000, whereas significantly
increased P concentration with increasing N fertilization
rates in 2001 (Fig. 4).
Acta Botanica Sinica 植物学报 Vol.46 No.3 2004262
In the first year following N fertilization, the mean N
concentration of L. chinensis under each treatment was
lower than that of C. korshinskyi in both sites. The mean N
concentration of L. chinensis was higher than that of C.
korshinskyi from 30 g NH4NO3/m2 in site B in 2001, while in
site A, the average N concentration of L. chinensis was
higher than that of C. korshinskyi in all treatments. P con-
centration of C. korshinskyi under N treatments was higher
than that of L. chinensis.
2.3 Variations of N:P ratios
N:P ratios of the two species kept relative constant with
increasing N fertilization rates in both sites, with the excep-
tion that C. korshinskyi in site A and L. chinensis in site B
increased significantly (P < 0.05) with increasing N fertiliza-
tion rates in 2000 (Fig.5).
N:P ratio of C. korshinskyi of the control was higher
than that of L. chinensis in both sites except site A in 2001
(Table 1), suggesting that N:P ratio might be a species-
specific value. t -test showed that N:P ratios of two species
differed significantly in site A in 2000 and site B in 2001.
2.4 Relationship between N and P concentrations
There existed a strong linear relationship between N
and P concentrations of the two species in sites A and B in
2001. Whereas in 2000, N and P contents of the two species
were poorly correlated in both sites except that of L.
chinensis in site A.
3 Discussion
3.1 Nutrient limitation of the two species as referred to
from fertilization experiment
In this experiment, we selected two different sites to test
the effect of N fertilization on the performance of plants,
and their tissue chemistry. Site A had been banned from
large animal (cattle, sheep, goats and horses) grazing for 22
years prior to the commencement of the experiment, and
could be regarded as a climax community while the neigh-
boring site B was at the nascent stage of restoration be-
cause it had been only fenced for two years when our ex-
periment was started. Due to overgrazing, site B was highly
degraded according to the biomass production. Therefore,
Fig.2. Responses of aboveground biomass of L. chinensis and C. korshinskyi to N fertilization rates for two years in sites A and B. Solid
line indicates the linear regression line of L. chinensis, while dashed line indicates that of C. korshinskyi.
ZHANG Li-Xia et al.: Differential Responses of N:P Stoichiometry of Leymus chinensis and Carex korshinskyi to N Additions
in a Steppe Ecosystem in Nei Mongol 263
it is reasonable to predict that there should be differential
responses of the same plant species to N fertilization ow-
ing to the differences in site fertility.
Well-controlled and well-replicated nutrient addition
experiment was the only way to determine the nutrient limi-
tations both at population and community levels (Bedford
et al., 1999). Nutrient status could be determined by the net
primary production (NPP), and an increase in NPP defines
the nutrient limitation (Vitousek and Howarth, 1991). NPP is
correlated with aboveground biomass. Since NPP is very
difficult to measure, we attempted to determine the nutrient
limitation by responses of aboveground biomass and nu-
trient concentrations to N fertilization. We assumed that
species was N-limited when its aboveground biomass and
nutrient concentration (we considered P in this paper) other
than N increased as responses to N fertilization.
Responses of biomass of single species to N fertiliza-
tion were not always consistent with that of the total
communities. In this study, aboveground biomass of L.
chinensis and C. korshinskyi kept relative constant with
increasing N fertilization rates in site A in 2000 (Fig.2), dif-
ferent with that of the total aboveground biomass, which
was significantly increased with N fertilization application
(data not shown). This suggested that N might still be the
limiting nutrient for the entire ecosystem in site A, however,
this could not be inferred to form the responses of the two
dominant species. This implies that species other than the
two species we selected could account for the increased
biomass production in response to N fertilization. We ex-
pected that disturbed site (site B) might be in greater short-
age of N, and should therefore be more responsive to N
fertilization than in site A. The insignificant response of
total maximum community biomass and the biomass of single
species might have been resulted from factors other than
the limiting N alone. For example, the thinner litter layer and
the very low nutrient availability in site B might have caused
the added N lost through leaching and volatilization, or to
be temporarily stored in the plant tissue without stimulat-
ing much biomass increment. In the second year after N
fertilization application, the total aboveground biomass
Fig.3. Concentrations of aboveground tissue nitrogen of L. chinensis and C. korshinskyi as a function of N fertilization rates in two sites
in 2000 and 2001. Solid line indicates the regression line of L. chinensis, while dashed line indicates that of C. korshinskyi.
Acta Botanica Sinica 植物学报 Vol.46 No.3 2004264
significantly increased with increasing N fertilization rates
in both sites, indicating that N is the most limiting factor at
the ecosystem level. This did not necessarily mean that N
application would increase the biomass of all the species,
since the biomass of C. korshinskyi significantly decreased
in site A. Thus other elements (P, for example) could limit
the growth of C. korshinskyi in the well-protected site.
This result was different from that of Theodose and Bow-
man (1997), who found that N addition could increase the
biomass production of grasses whereas sedges were unaf-
fected by fertilization in alpine tundra communities.
Plant nutrient concentrations differ widely among spe-
cies (McJannet et al., 1995). The fact that N concentration
between the two species in site A and P concentration in
site B in 2001 differed significantly (Table 1) supported this
assertion. Other than soil analysis, plant nutrient measure-
ment was an additional method to determine nutrient limita-
tion since plant nutrient content reflects the nutrient sup-
ply in the soil and can act as an integrator of all growth
factors (Tisdale et al., 1985). If an element is limiting, the
concentration of this element in the plant tissue will tend to
positively correlate with nutrient supply rates unless no
other nutrients limit the growth of plants (Garnier, 1998).
N and P concentrations of the two species increased
significantly with increasing N fertilization rates in 2001
(Figs.3, 4), while P concentration remained constant regard-
less of N fertilization in 2000, indicating that plant species
would accumulate one nutrient element as it was provided
in pulses. N was the main growth-limiting factor of grasses
but was not of sedge in our study site according to the
nutrient concentration responses in 2001, being indepen-
dent of the successional stages. The increases in the N and
P concentrations in C. korshinskyi and a corresponding
decrease in biomass with increasing N fertilization rates
suggest that factors other than N might have been limiting.
C. korshinskyi had lower N and P contents among the two
dominant species and was more responsive to P fertiliza-
tion (data not shown), indicating that phosphorus was the
main limiting factor for the growth of C. korshinskyi. The
aboveground biomass of L. chinensis increased most
Fig.4. P concentrations of Leymus chinensis and C. korshinskyi as a function of N fertilization rates in site A and site B for two years.
Solid line indicates the regression line of L. chinensis, while dashed lined indicates that of C. korshinskyi.
ZHANG Li-Xia et al.: Differential Responses of N:P Stoichiometry of Leymus chinensis and Carex korshinskyi to N Additions
in a Steppe Ecosystem in Nei Mongol 265
drastically with N fertilization rates and decreased along P
fertilization gradients, suggesting that N was the main lim-
iting factor for the growth of L. chinensis.
Fertilization could increase the production of sedge by
increasing its ramets (Shaver and Chapin, 1995). However,
in our study, a decrease of ramets and aboveground biom-
ass of C. korshinskyi was found with N fertilization rates in
site A in 2001, in which the number of individuals (data not
shown) and aboveground biomass of C. korshinskyi sig-
nificantly declined with increasing N fertilization rates. This
might have been resulted from the fact that the results of
Shaver and Chapin (1995) was from tundra ecosystems where
sedges were the dominant species, while in our experimen-
tal site sedges contributed only a relative minor fraction of
biomass to the community.
3.2 N:P stoichiometry and nutrient limitation
It is suggested that inferences from N:P stoichiometry
should be verified by the fertilization experiment (Bedford
et al., 1999). N and P concentrations correlated significantly
under different fertilization rates in 2001 (Fig. 6), indicating
Fig.5. Responses of N:P stoichiometry of aboveground tissues of L. chinensis and C. korshinskyi. to nitrogen fertilization in two sites.
Solid line indicates the regression line of L. chinensis, while dashed line indicates that of C. korshinskyi.
Table 1 Mean (±1 SE) N, P concentrations and N:P ratios of Leymus chinensis and Carex korshinskyi in the control in site A and site
B for two years
Site A Site B
2000 2001 2000 2001
　 L. chinensis C. korshinskyi L. chinensis C. korshinskyi L. chinensis C. korshinskyi L. chinensis C. korshinskyi
N% 1.19± 0.06a 1.30± 0.14a 1.70± 0.10a 1.43± 0.05b 1.54± 0.09a 1.72± 0.07a 1.74± 0.11a 1.62± 0.10a
P% 0.078± 0.005a 0.069± 0.004a 0.090± 0.005a 0.080± 0.005a 0.085± 0.006a 0.082± 0.005a 0.107± 0.006a 0.069± 0.005b
N/P 15.33± 0.64a 18.83± 1.08b 19.33± 1.59a 18.12± 1.03a 18.19± 0.99a 21.39± 1.32a 16.41± 0.72a 23.79± 1.62b
Numbers followed with different superscript letters between L. chinensis and C. korshinskyi in the same year indicate significantly different at
P = 0.05 (t-test).
Acta Botanica Sinica 植物学报 Vol.46 No.3 2004266
that plants would regulate the uptake of the two nutrient
elements so as to maintain the inner balance. The insignifi-
cant relationship between N and P concentrations in C.
korshinskyi in site A and in the two species in site B in 2000
(Fig. 6) might be due to the different accumulation rate of N
and P concentrations. N:P ratio in plant tissue has been
used to study the mineral balance since a functional rela-
tionship exists between N and P in plant (Redfield, 1958;
Garten Jr, 1976; Reiners, 1986). Koerselman and Meuleman
(1996) reported results of experiments conducted in the
wetland of the Netherlands. They proposed that N:P ratio
could be a good tool to detect nutrient limitation, and sug-
gested that aboveground biomass would be limited by ni-
trogen when N:P ratio was low (<14), by P when N:P was
high (>16) and by the both when N:P was between 14 and
16. However, using results from bogs in Sweden, both van
den Driessche (1974) and Aerts et al. (1992) have shown
that low N:P ratios (<10) indicate plant growth is N-limited,
high N:P ratios (>14) reflect plant growth is P-limited. We
argue that for different types of ecosystems, the limitation
of N or P can be revealed by different N:P ratios. Moreover,
even in the same areas, coexisting species could be limited
by different nutrient elements. Our results indicated that
nutrient limitation was different among species and among
sites of different fertility. If we accept the N:P thresholds
used by Koerselman and Meuleman (1996), it can be in-
ferred that the two dominant species were in short of phos-
phorus according to their high N:P ratios in the control (N:
P>16) in 2001. This conclusion is in contradiction with the
results from our fertilization experiment. Thus we might pro-
pose a new threshold of N:P ratio for N and P limiting in our
study sites. If N:P is < 21, the species is N limited, while N:
P > 23 might indicate P limiting. This new threshold was in
accordance with the fertilization results, but need further
experiment.
The threshold proposed by Koerselman and Meuleman
(1996) was obtained from a summary of 40 filed experiments
conducted in the Netherlands, and the threshold of N:P
ratio for whether N or P limiting was obtained from the
averaged values of different species in a community.
Fig.6. Relationships between N and P concentrations of L. chinensis and C. korshinskyi in the same site in 2000 and 2001. Solid line
indicates the regression line of L. chinensis, while dashed line indicates that of C. korshinskyi.
ZHANG Li-Xia et al.: Differential Responses of N:P Stoichiometry of Leymus chinensis and Carex korshinskyi to N Additions
in a Steppe Ecosystem in Nei Mongol 267
However, the responses of individual plant species to fer-
tilization may not be predicted from their N:P ratios. Güsewell
and Koerselman (2002) also suggested that N:P should only
be used to assess nutrient limitation at the vegetation level.
In addition, when N:P ratios of different species among
different sites are compared, the variation may more likely
to reflect the variation of nutrient availability among sites
rather than differences among species (Güsewell and
Koerselman, 2002). The variation of N:P ratios of species
indicates that the two sites differ in their nutrient availability.
Even in the same site (site A), variation of N:P ratios of the
two species also existed. Because we chose only two
species, it might be immature to draw a solid conclusion on
what nutrient is the most limiting in different sites.
Nevertheless, a strong response of two dominant herba-
ceous species to N fertilization does indicate that both sites
are N-limited.
One of the objectives of this study was to test the null
hypothesis that fertilization did not change the N:P ratio of
plant species in typical L. chinensis steppe ecosystems.
Our assumption was proved to be valid by the results of
2001 that N addition did not change the N:P ratio of two
species in both site, however, this was not true for the N:P
ratios of C. korshinskyi in site A and of L. chinensis in site
B in the first year, which increased significantly with in-
creasing N fertilization (Fig.5). We believe that this discrep-
ancy was due to the differences of successional stages of
the two sites. The assumption is that plants require a simi-
lar balance of resources to maintain optimal growth (Chapin
et al., 1987). We selected the two species because the sum
of their biomass accounted for more than 36% of the total
at both sites. Because our experiment only lasted for two
years, the long-term responses could not be inferred to
from this short-term study. Moreover, further study is defi-
nitely required to examine the responses of both dominant
species and the redundant ones (minor species).
Aerts et al. (1992) found that N fertilization increased N:
P ratio and P fertilization decreased N:P ratio of the Sphag-
num capitula significantly at the high-N site. Güsewell and
Koerselman (2002) also verified that fertilization with N in-
creased N concentration and N:P ratio, whereas P fertiliza-
tion increased P concentration and reduced the N:P ratio.
Our result was different from theirs on species responses
to nitrogen fertilization. This discrepancy may be due to
the fact that lower plant species such as Sphagnum may
differ in part from higher plants in their responsiveness to
nutrient additions, or the intrinsic difference in site quality
may also account for these observed inconsistencies.
To sum up, the differential responses of different
species in the typical steppe ecosystem suggest that it
may be improper to define a limited nutrient element be-
yond the species level. The relative constant N:P ratios of
some species, and a changing N:P ratio of others not only
suggest that plants could have evolved some internal
mechanisms to maintain a balanced nutrient uptake, but
also this effect could vary dramatically among species. In-
creased tissue P concentration due to increasing N fertili-
zation rates may indicate that there exists a synergistic rela-
tionship between these two elements. However, the under-
lying mechanisms need to be further tested. We hypoth-
esize that in a long run, species that can drastically change
their N:P ratios rather than N and P concentrations in re-
sponse to external input of certain element would become
locally extinct. Because different plant species co-exist in a
community as a result of long-time evolutionary adapta-
tions or the niche availability, the use of N:P ratio to detect
the status of nutrient availability or limitation should be
carefully chosen, especially when the responses at differ-
ent hierarchical levels (tissue, organ, species, population,
community) and the initial site quality are to be considered.
References:
Aerts R, Wallen B, Malmer N. 1992. Growth-limiting nutrients in
Sphagnum-dominated bogs subject to low and high atmospheric
nitrogen supply. J Ecol, 80:131-140.
Batjes N H. 1996. Global assessment of land vulnerability to
water erosion on a 1/2 by 1/2 grid. Land Degrad Develop, 7:
353-365.
Bedford B L, Walbridge M R, Aldous A. 1999. Patterns in nutri-
ent availability and plant diversity of temperate North Ameri-
can wetlands. Ecology, 80:2151-2169.
Bradstreet R B. 1965. The Kjeldahl Method for Organic Nitrogen.
New York: Academic Press.
Brockway D G, Gatewood R G, Paris R B. 2002. Restoring
grassland savannas from degraded pinyon-juniper woodlands:
effects of mechanical overstory reduction and slash treatment
alternatives. J Environ Manage, 64:179-197.
Chapin F S Ⅲ, Bloom A J, Field C B, Waring R H. 1987. Plant
responses to multiple environmental factors. BioScience, 37:
49-57.
Chapin F S Ⅲ, Shaver G R, Giblin A E, Nadelhoffer K J, Laundre
J A. 1995. Responses of arctic tundra to experimental and
observed changes in climate. Ecology, 76:694-711.
Chapin F S Ⅲ, Vitousek P M, van Cleve K. 1986. The nature of
nutrient limitation in plant communities. Am Nat, 127:48-58.
Acta Botanica Sinica 植物学报 Vol.46 No.3 2004268
Chen M, Baoyin T. 1997. Study on the effect of different im-
proving techniques on the degenerated grassland in the typical
steppe region. Inner Mongolia grassland ecosystem research
station . Research on Grassland Ecosystem (No.5). Beijing:
Science Press. 100-110. （in Chinese with English abstract）
Chen Z-Z, Sheng X-W, Yang Z-G, Huang D-H. 1985. The eco-
logical effects of fertilization during rainy season on various
types of steppe on the Xilin river valley, Inner Mongolia.
Inner Mongolia grassland ecosystem research station . Re-
search on Grassland Ecosystem (No.1). Beijing: Science Press.
225-232. （in Chinese with English abstract）
Chen Z-Z, Wang S-P. 2000. Typical Steppe Ecosystem of China.
Beijing: Science Press. 307-319. （in Chinese）
Chen Z-Z. 1988. Topography and climate of Xilin river basin.
Inner Mongolia grassland ecosystem research station . Re-
search on Grassland Ecosystem (No.3). Beijing: Science Press.
3-22. （in Chinese with English abstract）
DiTomasso A, Aarssen L W. 1989. Resource manipulation in
natural vegetation: a review. Vegetatio, 84:9-29.
Duarte C M. 1992. Nutrient concentration of aquatic plants: pat-
terns across species. Limnol Oceanog, 37:882-889.
Elser J J, Dobberfuhl D R, MacKay N A, Schampel J H. 1996.
Organism size, life history, and N:P stoichiometry. BioScience,
46:674-684.
Facelli J M, Pickett S T A. 1991. Plant litter: its dynamics and
effects on plant community structure. Bot Rev, 57:1-32.
Fisher F M, Zak J C, Cunningham G L, Whitford W G. 1988.
Water and nitrogen effects on growth and allocation patterns
of creosotebush in the northern Chihuahuan Desert. J Range
Manage, 41:387-391.
Garnier E. 1998. Interspecific variation in plasticity of grasses in
response to nitrogen supply. Cheplick G P. Population Biol-
ogy of Grasses. Cambridge: Cambridge University Press. 155-
181.
Garten Jr C T. 1976. Correlations between concentrations of
elements in plants. Nature, 261:686-688.
Glaser B, Lehmann J, Führböter, Solomon D, Zech W. 2001.
Carbon and nitrogen mineralization in cultivated and natural
savanna soils of Northern Tanzania. Biol Fertil Soils, 33:301-
309.
Gough L, Osenberg C W, Gross K L, Collins S L. 2000. Fertiliza-
tion effects on species density and primary productivity in
herbaceous plant communities. Oikos, 89:428-439.
Gu X-Y, Li S-Q. 1997. Influence of grazing intensity on the sub-
microscopic features in steppe soils. Inner Mongolia grass-
land ecosystem research station . Research on Grassland Eco-
system (No.5). Beijing: Science Press. 80-87. （in Chinese
with English abstract）
Güsewell S, Koerselman W. 2002. Variation in nitrogen and phos-
phorus concentrations of wetland plants. Perspect Pl Ecol,
Evol Syst, 5:37-61.
Horwitz W. 1975. Methods of Analysis. 12th ed. Washington D
C: Association of Official Analytical Chemists. 11-17.
Jia S-H, Cui X-M, Li S-L, Chen Y-J, Wang F-J. 1997. Changes of
soil physical attributes along grazing gradient. Inner Mongolia
grassland ecosystem research station . Research on Grassland
Ecosystem (No.5). Beijing: Science Press. 12-16. （in Chi-
nese with English abstract）
Jiang S. 1985. An introduction on the Inner Mongolia grassland
ecosystem research station, Academia Sinica. Inner Mongolia
grassland ecosystem research station. Research on Grassland
Ecosystem (No.1). Beijing: Science Press. 1-11. （in Chi-
nese with English abstract）
Koerselman W, Meuleman A F M. 1996. The vegetation N:P
ratio: a new tool to detect the nature of nutrient limitation. J
Appl Ecol, 33:1441-1450.
Li B. 1997. The rangeland degradation in North China and its
preventive strategy. Sci Agr Sin, 30:1-9. （in Chinese with
English abstract）
Li Q-F, Hu C-Y, Wang M-J. 2003. Analysis on the causes of eco-
environmental deterioration in Xilinguole typical grassland
region and counter measures. Acta Sci Nat Univ Nei Mongol
(Nat Sci) , 34:166-172. （in Chinese with English abstract）
Li Q-F, Li F-S, Wu L. 2002. A primary analysis on climatic change
and grassland degradation in Inner Mongolia. Agr Res Arid
Areas, 20:98-102. （in Chinese with English abstract）
Lowther J R. 1980. Use of a single sulphuric acid-hydrogen per-
oxide digest for the analysis of Pinus radiata needles. Commun
Soil Sci Plan, 11:175-188.
Mainguet M, Da Silva G G. 1998. Desertification and drylands
development: what can be done? Land Degrad Develop, 9:
375-382.
McJannet C L, Keddy P A, Pick F R. 1995. Nitrogen and phos-
phorus tissue concentrations in 41 wetland plants: a compari-
son across habitats and functional groups. Funct Ecol, 9:231-
238.
Pegtel D M, Bakker J P, Verweij G L, Fresco L F M. 1996. N, K
and P deficiency in chronosequential cut summer-dry grass-
ZHANG Li-Xia et al.: Differential Responses of N:P Stoichiometry of Leymus chinensis and Carex korshinskyi to N Additions
in a Steppe Ecosystem in Nei Mongol 269
lands on gley podzol after cessation of fertilizer application.
Plant Soil, 178:121-131.
Redfield A C. 1958. The biological control of the chemical factors
in the environment. Am Sci, 46:205-230.
Reiners W A. 1986. Complementary models for ecosystems. Am
Nat, 127:59-73.
Rice E L. 1964. Physiological Ecology Laboratory Procedures.
Norman, Oklahoma: University Oklahoma Bookstore.
Shaver G R, Chapin F S Ⅲ. 1995. Long-term responses to facto-
rial NPK fertilizer treatment by Alaskan wet and moist tundra
sedge species. Ecography, 18:259-275.
Story J M, Boggs K W, Graham D R. 1989. Effects of nitrogen
fertilization on spotted knapweed and competing vegetation
in western Montana. J Range Manage, 42:222-225.
Theodose T A, Bowman W D. 1997. Nutrient availability, plant
abundance, and species diversity in two alpine tundra
communities. Ecology, 78:1861-1872.
Tisdale S L, Nelson W L, Beaton J D. 1985. Soil Fertility and
Fertilizers. 4th ed. New York: Macmillan Publishing Company.
van den Driessche R. 1974. Prediction of mineral nutrient status
of trees by foliar analysis. Bot Rev, 40:347-394.
Vitousek P M, Howarth R W. 1991. Nitrogen limitation on land
and in the sea: how can it occur? Biogeochemistry, 13:87-
115.
Wang J-W, Cai W-Q. 1988. Studies on genesis, types and charac-
teristics of the soils of the Xilin River Basin. Inner Mongolia
grassland ecosystem research station . Research on Grassland
Ecosystem (No.3). Beijing: Science Press. 23-83. （in Chi-
nese with English abstract）
Wang S-P, Li Y-H, Wang Y-F, Han Y-H. 1998. The succession of
Artemisia frigida rangeland and multivariation analysis under
different stocking rates in Inner Mongolia. Acta Agrest Sin , 6:
299-305. （in Chinese with English abstract）
Wang W, Liu Z-L, Hao D-Y, Liang C-Z. 1996a. Research on the
restoring succession of the degenerated grassland in Inner
Mongolia Ⅰ. Basic characteristics and driving force for resto-
ration of the degenerated grassland. Acta Phytoecol Sin, 20:
449-459. （in Chinese with English abstract）
Wang W, Liu Z-L, Hao D-Y, Liang C-Z. 1996b. Research on the
restoring succession of the degenerated grassland in Inner
Mongolia Ⅱ. Analysis of the restoring processes. Acta
Phytoecol Sin, 20:460-471. （in Chinese with English
abstract）
Wassen M J, Olde Venterink H G M, de Swart E O A M. 1995.
Nutrient concentrations in mire vegetation as a measure of
nutrient limitation in mire ecosystems. J Veg Sci, 6:5-16.
Wilson S D, Tilman D. 1993. Plant competition and resource
availability in response to disturbance and fertilization.
Ecology, 74:599-611.
Wu Z-Y. 1980. Vegetation of China. Beijing: Science Press. 519-
533.（in Chinese）
Zhang L-X , Bai Y-F , Han X-G. 2003. Application of N:P sto-
ichiometry to ecology studies. Acta Bot Sin , 45:1009-1018.
(Managing editor: HAN Ya-Qin)