Download Scale dependent relationships between native plant

Weed Science, 53:600–604. 2005
Scale dependent relationships between native plant
diversity and the invasion of croftonweed (Eupatorium
adenophorum) in southwest China
Zhijun Lu
Laboratory of Quantitative Vegetation Ecology,
Institute of Botany, Chinese Academy of Sciences,
20 Nanxincun, Xiangshan, Haidian District, Beijing
100093, China; Graduate School of Chinese
Academy of Sciences, 19 Yuquanlu (Jia), Shijingshan
District, Beijing 100039, China
Keping Ma
Corresponding author. Laboratory of Quantitative
Vegetation Ecology, Institute of Botany, Chinese
Academy of Sciences, 20 Nanxincun, Xiangshan,
Haidian District, Beijing 100093, China;
[email protected]
Croftonweed is an invasive plant in southwest China. We examined the relationships
between its invasion patterns and native plant diversity at different spatio-temporal
scales. At the 25 m2 scale, invasion success was negatively correlated with native
plant diversity, indicating that resource availability might be the dominant factor
regulating community invasibility. At the 400-m2 scale, both negative and positive
relationships were detected, possibly identifying a spatial scale threshold where extrinsic environmental factors became more important to community invasibility. At
the vegetation province scale, variations in physical environment outweighed the
importance of intrinsic biotic factors and positive relationships between diversity and
invader success were found. Native plant diversity also inhibited croftonweed over
the course of community succession and at the early stages of invasion at local spatial
scales. However, the changing relationship might be an artifact of sampling at different spatial scales.
Nomenclature:
Croftonweed, Eupatorium adenophorum Sprengel EUPAD.
Key words: Biodiversity, biological invasions, community succession, invasive species, spatio-temporal scale.
Croftonweed is a many-stemmed erect perennial shrub,
which is native to Mexico (Auld 1969, 1970; Liu et al.
1985; Papes and Peterson 2003). Its high reproductive capacity and windborne seeds are particularly adapted to colonizing bare or intermittently bare areas (Auld and Martin
1975) which make it a troublesome weed of crops, plantations, and pastures in many parts of the world (Wang et al.
1997). The weed was first recorded in China in the 1940s
(Qiang 1998; Zhao and Ma 1989). Since its invasion, croftonweed has spread rapidly across southwest China causing
great economic losses in both cropland and rangeland (Liu
et al. 1985; Zhao and Ma 1989).
Elton (1958) hypothesized that plant communities with
high diversity should be less susceptible to invasion. At small
spatial scales, this hypothesis has been supported by theoretical (Case 1990; Law and Morton 1996) and experimental studies (Dukes 2001; Hector et al. 2001; Kennedy et al.
2002; Knops et al. 1999; Levine 2000; McGrady-Steed et
al. 1997; Naeem et al. 2000; Symstad 2000; Tilman 1997,
1999). However, measured at large regional scales (Levine
and D’Antonio 1999; Planty-Tabacchi et al. 1996; Stohlgren et al. 1999), there is also convincing evidence that the
relationship between native species diversity and invading
species is positive, and furthermore, that the relationship
between species diversity and community invasibility can
change over the course of invasion (Wiser et al.1998). However, few studies have examined the effects of spatio-temporal scales on the relationship between biodiversity and
community invasibility.
We analyzed vegetation data from across the distribution
range of croftonweed in Yunnan and Sichuan Province of
southwest China to (1) test the effect of spatial scale on the
relationship between biodiversity and community invasibility; (2) determine how the relationship between biodiversity
600
•
Weed Science 53, September–October 2005
and community invasibility changes over the course of community succession; and (3) examine the relationship between biodiversity and invasibility in communities with different invasion histories.
Materials and Methods
We analyzed native plant diversity and community invasibility relationships at five spatial scales ranging from 25
m2 to 2.53 by 1011 m2 (Table 1) and over the course of
community succession and invasion by croftonweed. Vegetation data was gathered at the local scale in 400-m2 plots
using a nested design (Table 2) in 2003 and 2004. At each
site, three 20-by-20-m plots were randomly established in
representative vegetation types as described in China Vegetation (Wu 1995), including low, middle, and high canopy
densities (0–30%, 31–70%, and 71–100%; Parendes and
Jones 2000). In each plot, three 5-by-5-m subplots were
placed in alternate corners and at the plot center, and one
1-by-1-m sub-subplot was placed in the right lower corner
of each 5-by-5-m subplot. Percent cover of each tree species
in the overstory was visually estimated in the 20-by-20-m
plots; cover and abundance of each shrub, including croftonweed, was estimated in the 5-by-5-m subplots; forb cover
and abundance was estimated in the 1-by-1-m sub-subplots.
Abundance refers to the visual estimation of the number of
shrubs and forbs in each sample unit according to BraunBlanquet’s five abundance classes (Li 1993). Longitude, latitude, elevation, slope and aspect data were recorded for each
plot. Plant species that could not be identified in the field
were collected and identified at the herbarium of the Institute of Botany, Chinese Academy of Sciences.
We combined data collected at the local scales to comprise vegetation data at the regional scales. Wu (1995) clas-
TABLE 1. The size (m2) and number of vegetation sampling units
used to test the effects of spatial scale on the relationships between
native plant diversity and community invasibility.
Regional scalea
Local scale
Subplot
Size
Number
25
108
Plot
400
36
Province
13
5
1011
Zone
1.68 3
3
1011
Region
2.53 3 1011
2
a The size of the regional scales was calculated using the China Vegetation
District Map (Wu 1995).
sified the vegetation districts of China into eight vegetation
regions, 18 vegetation zones, and 85 vegetation provinces.
The large-scale level in our study was comprised of two
vegetation regions, three vegetation zones, and five vegetation provinces: Subtropical evergreen broad-leaved forest
vegetation region (I); Central subtropical evergreen broadleaved forest vegetation zone (IA); Central Yunnan plateau
and basin, Cyclobalanopsis glaucoides–Castanopsis–Pinus yunnanensis forest vegetation province (IA-1); Sichuan and Yunnan Jinshajiang River gorge, Pinus yunnanensis–dry, hot valley vegetation province (IA-2); South subtropical monsoon
evergreen broad-leaved forest vegetation zone (IB); Yunnan,
Guizhou and Guangxi calcareous mountain, Machilus pingii–Cyclobalanopsis glauca–Pinus yunnanensis var. tenuifolia
forest vegetation province (IB-1); Central and south Yunnan
midmountain, Castanopsis–Schima wallichii–Pinus khasya
forest vegetation province (IB-2); Tropical seasonal rainforest
and rainforest vegetation region (II); North tropical seasonal
rainforest and semievergreen seasonal rainforest vegetation
zone (IIA); Southwest Yunnan valley and mountainous area,
semi-evergreen seasonal rainforest vegetation province (IIA1).
Sites in the vegetation provinces IA-1, IB-1 and IB-2, and
IIA-1 were sampled from March 20 to April 20, 2003 and
August 2004. Sites in province IA-2 were sampled from August 2–21, 2003.
All statistical analyses were conducted using Origin 6.0
(Microcal Software Inc. 1999) at P , 0.05 significance. We
assessed the relationships between native plant richness, and
cover and abundance of croftonweed using linear and second-order polynomial regression models at the local scales
including 25 m2 and 400 m2. There were not enough data
points to conduct the regression analysis at the large spatial
scales of vegetation province, zone, and region. Instead, oneway analyses of variances (ANOVA) were used to analyze
these data following arcsine square root transformation of
all percentage values (Gelbard and Belnap 2003). To test
FIGURE 1. The relationships of cover and abundance of croftonweed to
native plant species richness at different spatial scales (25-m2 subplot, 400m2 plot, vegetation province). Determination coefficients (R 2), equations,
and significance levels of linear and second order polynomial regressions are
shown. In the panels, each black dot represents data from plots, subplots,
or combined data that comprise the vegetation province. The Y axis for
Figures 1b, 1d, and 1f reflects estimated abundance of croftonweed, ranging
from very abundant (5) to very rare (1).
the significant difference of cover, abundance of croftonweed, and native plant richness between successional stages,
we used Fisher’s LSD for paired multiple comparison procedures.
Results and Discussion
At the local scale of 25 m2, linear regression analysis demonstrated significant negative correlations (P , 0.01) between native plant species richness and cover and abundance
of croftonweed (Figures 1a and 1b). At the 400-m2 scale,
second-order polynomial regressions showed a negative relationship up to a critical threshold of native species richness
TABLE 2. General information about the plots and subplots.a
Vegetation
province
Location
Extreme
minimum
temperature
Annual mean
temperature
C
IA-1
IA-2
IB-1
IB-2
IIA-1
a
258–278N, 998–1068E
258309–298459N, 1008–1068E
238–258109N, 1018309–1068209E
228309–268N, 988109–1038E
238–258289N, 978509–998139E
14–17
17
17–21
17–19
19–20
Annual mean
rainfall
Field sites
mm
26
23.4
24.4
0
2–3
700–1200
1040
900–1200
1200–1600
1200–1600
Plots
Subplots
No.
3
2
2
4
1
9
6
6
12
3
27
18
18
36
9
The climatic and geographic data is from China Vegetation (Wu 1995) except for the location of IIA-1, which was calculated from our field investigation.
Lu and Ma: Croftonweed (Eupatorium adenophorum) in southwest China
•
601
FIGURE 2. Correlation coefficients of (a) cover and (b) abundance of croftonweed to native plant species richness at different spatial scales (25 m2,
400 m2, province). Total, total number of native plant species; Tree, number of native tree species; Shrub, number of native shrub species; Forb,
number of native forb species).
after which the relationship between croftonweed and native
species richness turned positive. Forty-five percent of the
variance of cover and 42% of the variance of abundance of
croftonweed were explained by native plant richness (Figures
1c and 1d). At the vegetation province scale, cover (F 5
3.51, P 5 0.01, df 5 4) and abundance (F 5 4.24, P 5
0.003, df 5 4) of croftonweed were significantly different
between vegetation provinces, and the cover of croftonweed
was positively correlated with native plant richness (Figure
1e). At the zone and region level, there were no significant
differences in cover and abundance of croftonweed between
zones and regions (P . 0.05), indicating that there was no
relationship between the invasion of croftonweed and native
plant richness at these spatial scales.
At the 25-m2 scale, the relationship between cover and
abundance of croftonweed and native plant species richness
was negative; i.e., the invasion by croftonweed decreased
with increasing native plant species richness. Communities
with high diversity appeared to be more resistant to invasion, supporting the hypothesis of Elton (1958). The physical environment such as climate was nearly homogeneous
at the 25-m2 scale, and native plant diversity appeared to
play an important role in regulating community invasibility.
Competition for available resources might be the most important process influencing invasibility at this spatial scale.
Higher species diversity could increase the probability of
including species with traits similar to croftonweed and
strongly competitive species, resulting in a greater utilization
proportion of the potentially available resources (Hooper et
al. 2005), thereby interfering with the ability for crofton602
•
Weed Science 53, September–October 2005
FIGURE 3. The relationships between cover and abundance of croftonweed
and native plant species richness at four stages in two kinds of typical
community successions in Kunming at the 25-m2 scale. Figure 3a shows
community succession series from meadow to shrubland, coniferous forest,
and evergreen broad-leaved forest; Figure 3b shows community succession
from meadow to shrubland, coniferous forest, and deciduous broad-leaved
forest; X Axis: I 5 meadow stage; II 5 shrubland stage; III 5 coniferous
forest stage; IV 5 evergreen broad-leaved forest stage, and IV0 5 deciduous
broad-leaved forest stage. Y Axis includes cover and abundance of croftonweed and the number of native plant species. Error bars represent 1 SE.
Different letters for a particular parameter indicate significant differences
(P , 0.05) between successional stages based on LSD. Cover, cover of
croftonweed; Abundance, abundance of croftonweed; Total, total number
of native plant species; Tree, number of native tree species; Shrub, number
of native shrub species; Forb, number of native forb species.
weed to establish. Higher biodiversity might also enhance
the probability that generalist natural enemies should occur
(Hooper et al. 2005), but this mechanism might not necessarily inhibit invaders (Shea and Chesson 2002).
However, at the regional scale of vegetation province, the
relationship between cover of croftonweed and native plant
species richness was positive (Figure 1e), which indicated
that species-rich communities were vulnerable to the invasion of croftonweed. At large spatial scales, variability in
environmental factors such as soil fertility, propagule input,
and disturbance regimes apparently outweighed the effects
of species diversity on community invasibility (Hooper et
al. 2005). Thus, extrinsic factors that contributed to spatial
variation in the environment and promoted diversity in
communities were more important to invader success than
the direct effects of diversity (Foster et al. 2002). Temperature and moisture were the main climatic factors that influenced the occurrence of croftonweed (Zhao and Ma
1989); hence, microsites with favorable temperature and
moisture conditions would likely favor the establishment of
croftonweed. Interestingly, at the 400-m2 scale, both nega-
TABLE 3. The correlation coefficient matrix of cover and abundance of croftonweed and native plant species richness in 400-m2 plots
with three kinds of croftonweed invasion histories (short, intermediate, and long) from three sites (Xichang, Kunming, and Cangyuan).
Cover
Abundance
Total
Tree
Shrub
Forb
1.00
0.80
0.98**
20.20
1.00
0.73
20.69
1.00
20.20
1.00
1.00
0.60
0.84**
0.70*
1.00
0.36
0.47
1.00
0.21
1.00
1.00
0.51
0.92**
0.66
1.00
0.30
0.09
1.00
0.43
1.00
Xichang)a
Short-term invasion (about 20 yr,
1.00
Cover
0.82*
Abundance
1.00
Total
20.76
20.95**
Tree
20.87*
20.93**
Shrub
20.68
20.89*
Forb
0.50
0.43
Intermediate-term invasion (about 30 yr, Kunming)
1.00
Cover
0.99**
Abundance
1.00
Total
20.78*
20.81**
Tree
20.26
20.33
Shrub
20.66
20.67*
Forb
20.59
20.63
Long-term invasion (about 50 yr, Cangyuan)
1.00
Cover
0.90**
Abundance
1.00
Total
0.002
20.36
Tree
0.03
0.01
Shrub
20.50
20.17
Forb
0.35
20.04
* Significant at the 0.05 level (2-tailed); ** Significant at the 0.01 level (2-tailed).
a Cover: cover of croftonweed; Abundance: abundance of croftonweed; Total: the total number of native plant species; Tree: the number of native trees;
Shrub: the number of native shrubs; Forb: the number of native forbs.
tive and positive relationships were found, possibly identifying a spatial scale threshold where species richness became
less important and the physical environment more important in regulating invader success in communities. With increasing spatial scales, variability in the physical environment played a stronger role, whereas the role of competition
and resource availability was weaker.
Among the three life forms (trees, shrubs, and forbs), the
richness of the forb layer had the greatest correlation coefficient to croftonweed cover and abundance, which might
indicate that the habitat requirements of croftonweed were
similar to that of the forbs or that there was less competition
between them as compared to shrubs and trees (Figure 2).
Thus, diversity of the forb layer was least likely to inhibit
croftonweed invasion. This result was consistent with the
findings of Hobbs (1989) and Wiser et al. (1998), but contrasted with the view that communities tend to be more
readily invaded by the invader representing a new morphological type (Johnstone 1986; Mooney and Drake 1989).
To determine how the relationship between native plant
species richness and cover and abundance of croftonweed
changes during succession, we simulated the community
successional process using vegetation data from temporary
25-m2 sample subplots in Kunming, Yunnan. The typical
successional sequence in this district included two types: (1)
from meadow to shrubland to coniferous forest to evergreen
broad-leaved forest, and (2) from meadow to shrubland to
coniferous forest to deciduous broad-leaved forest. Both the
cover and abundance of croftonweed peaked at the shrubland stage in the deciduous and evergreen broad-leaved forest successional series (Figure 3). In the deciduous broadleaved forest successional series, cover and abundance of
croftonweed also showed a second peak at the climax forest
stage (Figure 3b), which might indicate that this forest had
a lower resistance to this invader than coniferous and evergreen broad-leaved forests. Greater light availability in the
deciduous broad-leaved forests might contribute to this pattern. Changes in native species richness over successional
stages were similar in both series and showed an inverse
pattern to croftonweed invasion. The lowest diversity occurred at the shrubland stage, demonstrating the resistance
of native plant diversity to croftonweed invasion over the
course of community succession.
To examine the relationship between invader success and
native species richness during the course of invasion, we
studied vegetation data at the 400-m2 scale from three sites
(Xichang, Kunming, and Cangyuan) with different histories
since time of invasion (Table 3). The time span was determined from the literature (Wang et al. 1994; Wu et al.
1984; Xue et al. 1979). At the early stages of invasion (;20
yr), abundance of croftonweed was significantly negatively
correlated with total native plant species richness (P ,
0.01), tree richness (P , 0.01), and shrub richness (P ,
0.05). Cover of croftonweed also showed a negative correlation with tree richness (P , 0.05). After about 30 yr of
invasion, native plant diversity was still negatively correlated
within cover (P , 0.05) and abundance (P , 0.01) of croftonweed. After about 50 yr of invasion, the cover and abundance of croftonweed was not correlated with native plant
diversity (P . 0.05). Based upon these results, native plant
diversity apparently played an important role and enhanced
community resistance to invasion at the early stage, but its
effects diminished over the course of invasion. At the later
stage (;50 yr), the invasion of croftonweed might be mainly
correlated with abiotic factors, and the predictability of croftonweed occurrence from physical environment should improve (Wiser et al. 1998).
The inconsistency in the relationship between species
Lu and Ma: Croftonweed (Eupatorium adenophorum) in southwest China
•
603
richness and community invasibility might undermine the
general application of Elton’s hypothesis when making comparisons at multiple spatio-temporal scales (Stohlgren et al.
1999). The richness-invasibility relationship comparisons at
different spatial scales demonstrated that intrinsic factors
such as resource availability influence plant community invasion susceptibility at small spatial scales, and extrinsic factors such as physical environment influence plant community invasibility at larger scales. However, the changing relationship might be an artifact of sampling at different spatial scales. At local spatial scales, the ability to detect sites
susceptible to invasion is diminished and the effects of biotic
resistance to invasion are better detected. At regional spatial
scales, the ability to detect habitable sites is improved. Temporal changes in species richness-invasion relationships were
examined at small spatial scales and clearly showed the role
of species diversity on invader success. The change in the
relationship over time suggests that intrinsic factors such as
plant competition are very important during the early stages
of invasion. However, at later stages, environmental factors
become more important. These results indicate the importance of identifying the stage of invasion relative to time
(e.g., early, mid, or late stages) when studying invasion biology.
Acknowledgments
We thank T. G. Gao, S. G. Lu, C. L. Dang, X. L. Liu, H. P.
Tian, L. Q. Sun, Z. H. Lu, and the Division of Agriculture, Forestry, Grazing, and Plant Protection of local governments in Sichuan and Yunnan Province for assistance in the field; R. E. Sherman,
D. Flynn, Q. B. Zhang, P. Lu and J. Liu for helpful reviews; F. L.
Liu for figure construction; and two anonymous reviewers for helpful suggestions on an early version of the paper. This research was
supported by the Major Project of Knowledge Innovation of Chinese Academy of Sciences (KSCX1-SW-13-0X-0X).
Literature Cited
Auld, B. A. 1969. The distribution of E. adenophorum Spreng. on the Far
North Coast of New South Wales. J. Proc. Royal Soc. New South
Wales 102:159–161.
Auld, B. A. 1970. Eupatorium weed species in Australia. Pest Articles and
News Summaries 16:82–86.
Auld, B. A. and P. M. Martin. 1975. The autecology of E. adenophorum
Spreng. in Australia. Weed Res. 15:27–31.
Case, T. J. 1990. Invasion resistance arises in strongly interacting speciesrich model competition communities. Proc. Natl. Acad. Sci., USA 87:
9610–9614.
Dukes, J. S. 2001. Biodiversity and invasibility in grassland microcosms.
Oecologia 126:563–568.
Elton, C. S. 1958. The ecology of invasions by animals and plants. London,
Great Britain: Methuen. 181 p.
Foster, B. L., V. H. Smith, T. L. Dickson, and T. Hildebrand. 2002. Invasibility and compositional stability in a grassland community: relationships to diversity and extrinsic factors. Oikos 99:300–307.
Gelbard, J. L. and J. Belnap. 2003. Roads as conduits for exotic plant
invasions in a semiarid landscape. Conserv. Biol. 17:420–432.
Hector, A., K. Dobson, A. Minns, E. Bazeley-White, and J. H. Lawton.
2001. Community diversity and invasion resistance: an experimental
test in a grassland ecosystem and a review of comparable studies. Ecol.
Res. 16:819–831.
Hobbs, R. J. 1989. The nature and effect of disturbance relative to invasions. Pages 389–540 in J. A. Drake, H. A. Mooney, F. di Castri, R.
H. Groves, F. J. Kruger, M. Rejmánek, and M. Williamson, eds. Biological Invasions: A Global Perspective. Chichester, Great Britain: J.
Wiley.
Hooper, D. U., F. S. Chapin, III, J. J. Ewel, et al. 2005. Effects of biodi-
604
•
Weed Science 53, September–October 2005
versity on ecosystem functioning: a consensus of current knowledge.
Ecol. Monogr. 75:3–35.
Johnsone, I. M. 1986. Plant invasion windows: a time-based classification
of invasion potential. Biol. Rev. 61:369–394.
Kennedy, T. A., S. Naeem, K. M. Howe, J.M.H. Knops, D. Tilman, and
P. Relch. 2002. Biodiversity as a barrier to ecological invasion. Nature
417:636–638.
Knops, J.M.H., D. Tilman, N. M. Haddad, et al. 1999. Effects of plant
species richness on invasion dynamics, disease outbreaks, insect abundances and diversity. Ecol. Lett. 2:286–293.
Law, R. and R. D. Morton. 1996. Permanence and the assembly of ecological communities. Ecology 77:762–775.
Levine, J. 2000. Plant diversity and biological invasions: relating local process to community pattern. Science 288:852–854.
Levine, J. and C. M. D’Antonio. 1999. Elton revisited: a review of evidence
linking diversity and invasibility. Oikos 87:15–26.
Li, B. 1993. General Ecology. Huhhot, China: Inner Mongolia University
Press. 107 p.
Liu, L. H., S. C. Xie, and J. H. Zhang. 1985. Studies on the distribution,
damage and control of E. adenophorum Spreng. Acta Ecol. Sinica 5:
1–11.
McGrady-Steed, J., P. M. Harris, and P. J. Morin. 1997. Biodiversity regulates ecosystem reliability. Nature 390:162–165.
Microcal software, Inc. 1999. Origin. Version 6.0. Northampton, MA: Microcal Software.
Mooney, H. A. and J. A. Drake. 1989. Biological invasions: a SCOPE
program overview. Pages 491–506 in J. A. Drake, H. A. Mooney, F.
di Castri, R. H. Groves, F. J. Kruger, M. Rejmánek, and M. Williamson, eds. Biological Invasions: A Global Perspective. Chichester, Great
Britain: J. Wiley.
Naeem, S., J.M.H. Knops, D. Tilman, K. M. Howe, T. Kennedy, and S.
Gale. 2000. Plant diversity increases resistance to invasion in the absence of covarying extrinsic factors. Oikos 91:97–108.
Papes, M. and A. T. Peterson. 2003. Predicting the potential invasive distribution for E. adenophorum Spreng. in China. J. Wuhan Bot. Res.
21:137–142.
Parendes, L. A. and J. A. Jones. 2000. Role of light availability and dispersal
in exotic plant invasion along roads and streams in the H. J. Andrews
Experimental Forest, Oregon. Conserv. Biol. 14:64–75.
Planty-Tabacchi, A. M., E. Tabacchi, R. J. Naiman, C. Deferrari, and H.
Decamps. 1996. Invasibility of species-rich communities in riparian
zones. Conserv. Biol. 10:598–607.
Qiang, S. 1998. The history and status of the study on croftonweed (E.
adenophorum Spreng.) a worst worldwide weed. J. Wuhan Bot. Res.
16:366–372.
Shea, K. and P. Chesson. 2002. Community ecology theory as a framework
for biological invasions. Trends Ecol. Evol. 17:170–176.
Stohlgren, T. J., D. Binkley, G. W. Chong, et al. 1999. Exotic plant species
invade hot spots of native plant diversity. Ecol. Monogr. 69:25–46.
Symstad, A. J. 2000. A test of the effects of functional group richness and
composition on grassland invasibility. Ecology 81:99–109.
Tilman, D. 1997. Community invasibility, recruitment limitation, and
grassland biodiversity. Ecology 78:81–92.
Tilman, D. 1999. The ecological consequences of changes in biodiversity:
a search for general principles. Ecology 80:1455–1474.
Wang, F., B. A. Summerell, D. R. Marshall, and B. A. Auld. 1997. Inoculum production of Phaeoramularia, a potential mycoherbicide for
croftonweed. Aust. Plant Pathol. 26:173–178.
Wang, H. J., P. He, and J. L. Ma. 1994. An investigation and research
report on the dissemination of A. adenophora on rangeland areas in
Liangshan District of Sichuan Province. Grassl. China. 87(1):62–64.
Wiser, S. K., R. B. Allen, P. W. Clinton, and K. H. Platt. 1998. Community structure and forest invasion by an exotic herb over 23 years.
Ecology 79:2071–2081.
Wu, R. R., D. Y. Zhang, and X. S. Lu. 1984. On the distribution, perniciousness and control of the two weeds, Chromolaena coelestina and
C. odorata in Yunnan Province of China. Grassl. China. 77(2):17–22.
Wu, Z. Y. 1995. Vegetation of China. Beijing: Science Press. 1375 p.
Xue, J. R., S. R. Dong, J. H. Yin, and S. D. Hong. 1979. E. coelestium
L.: its adaptability, damage and control. Acta Botanica Yunnaica 1:
106–114.
Zhao, G. J. and Y. P. Ma. 1989. Investigation and study on the distribution
and damage of E. adenophorum in Yunnan Province. J. Weed Sci. 3:
37–40.
Received November 9, 2004, and approved April 27, 2005.