全 文 ：Received 27 Oct. 2003 Accepted 19 Apr. 2004
Supported by the Chinese Ministry of Education (2002-247) and Chongqing Science and Technology Commission (2002-7471).
* Author for correspondence. Tel: +86 (0)23 68254263; E-mail: .
http://www.chineseplantscience.com
Acta Botanica Sinica
植 物 学 报 2004, 46 (8): 907-914
Position-dependent Shoot Production of Two Subtropical Fig Tree
Species Following Crown Damage
ZENG Bo*, ZHONG Zhang-Cheng, ZHANG Xiao-Ping
(Key Laboratory of Eco-environments of Three Gorges Reservoir Region (Ministry of Education); Chongqing Key Laboratory of
Plant Ecology and Resources Research for Three Gorges Reservoir Region; Faculty of Life Sciences,
Southwest China Normal University, Chongqing 400715, China)
Abstract: In Three Gorges reservoir region, a great many of trees are needed for vegetation restoration
and land greening following the massive constructions (e.g. construction of roads, highways, buildings)
associated with the great dam project at Three Gorges of Yangtze River. Ficus microcarpa L. and F. virens
Ait. var. sublanceolata (Miq.) Cornor (Moraceae) are chosen and widely planted in this region as ornamental
trees and/or shade trees due to their shapely crowns and ability of growing on soils with low fertility.
Vegetative multiplication, which uses branch cuttings to cultivate saplings, is the main way for tree
propagation of the two species in Three Gorges reservoir region. Obtaining branch cuttings causes the
damage of tree crown and probably affects the growth of trees. In this study, the shoot production pattern
of two Ficus tree species following crown damage, which is crucial to the regrowth of trees, was investigated.
Data from a crown damage experiment with two damage seasons and a series of damage intensities were
analyzed. It was shown that crown damage, regardless of damage intensity and damage season, had no
effect on the shoot production of lateral branches of both species. However, the shoot production on the
main stem was position-dependent in both F. microcarpa and F. virens trees. Crown damage, conducted
either in spring or in autumn, did not affect the number and density of new shoots on the newly grown
upper stem parts and the branched stem parts within the residual crown, but facilitated the shoot production
on the bare stem parts beneath the residual crown in terms of both shoot number and density. Shoot
production on the bare stem parts increased with damage intensity. In addition, it was found that damage in
autumn led to a stronger emergence of shoots from the bare stem parts than spring damage. Some
mechanisms which could be involved in these results are discussed. Based on the experimental results, it
is suggested that among all investigated variables, only the enhanced shoot production on the bare stem
parts may increase the biomass partitioning to leaves and benefit the regrowth of damaged trees.
Key words: crown damage; Ficus microcarpa ; Ficus virens ; shoot production; Three Gorges reservoir
region
Construction of the great dam at Three Gorges of
Yangtze River is one of the biggest construction projects in
the world. The dam is 2 309 m long and 185 m high. This
project leads to the construction of many roads, highways,
railways, buildings, and even new towns and cities. Along
with the various construction in Three Gorges reservoir
region, a great many of trees are needed for vegetation
restoration and land greening. Ficus microcarpa L. and F.
virens Ait. var. sublanceolata (Miq.) Cornor (Moraceae)
are two fig tree species which are naturally distributed in
subtropical China. Trees of these two species have shapely
crowns and are able to grow on soils with low fertility. Due
to these merits, they are chosen and widely planted in Three
Gorges reservoir region, especially in cities, towns and
along roads as ornamental trees and/or shade trees. For the
two species, vegetative multiplication is the main means of
tree propagation, and usually branch cuttings are used to
cultivate saplings. Generally, the common mode of getting
branch cuttings is to remove relatively large branches from
the lower crown, leaving the upper parts of trees
untouched. Undoubtedly, branch removal damages tree
crown, which may affect the growth of trees due to the loss
of photosynthetic leaves. Therefore, the biomass partition-
ing pattern is of much importance for the regrowth and
recovery of damaged trees. Obviously, larger partitioning
of biomass to leaf production benefits the regrowth of a
damaged tree because of enhanced photosynthesis.
For a plant, it may be envisaged as an aggregation of
many basic structural units, each of them consisting of a
module (defined as the axis initiated from a bud (Prévost,
Acta Botanica Sinica 植物学报 Vol.46 No.8 2004908
1967)) and some appending vessels extending to the roots
(Franco, 1985; Sachs, 1991; Sprugel et al., 1991). According
to this structural unit theory, which has been developed
based on the pipe unit concept of Shinozaki et al. (1964a;
1964b) and the integrated physiological unit concept of
Watson and Casper (1984), modules which are nearer to the
base of a plant tend to have shorter appending vessels.
Thus, units with modules nearer to the base of a plant are
relatively cheaper in construction costs, and more resources
in the plant can be partitioned to leaf production when
units of this type are constructed. For a tree subjected to
crown damage, it is likely that its biomass partitioning to
leaves will be increased if the production of modules (which
can be regarded as shoots actually) on the stem and in the
lower crown is enhanced after crown damage.
This study aims to investigate whether the shoot pro-
duction patterns of F. microcarpa and F. virens trees
change following crown damage. To get a better under-
standing of the response of these two Ficus tree species to
crown damage in respect of shoot production, a series of
damage intensities and two seasons in which crown dam-
age was conducted were incorporated in the experiment.
Specifically, the following questions are addressed for the
study:
(1) Does crown damage decrease the production of
shoots in the upper crown and increase it in the lower
crown?
(2) Is the shoot production on the main stem enhanced
by crown damage?
(3) Are the traits mentioned in (1) and (2) affected by
damage season and damage intensity?
1 Materials and Methods
1.1 Study area and species
This study was conducted in an experimental garden
(ca. 210 m a.s.l.) established at the foot of the National
Nature Reserve of Jinyun Mountain (29º 50 N, 106º 26 E),
which is ca. 40 km north of Chongqing City, China. The
climate of this region is predominantly monsoonal, result-
ing in hot, humid summers and chilly but mostly frost-free
winters (Cornelissen, 1992; Zeng and Zhong, 1997; Li et
al., 1998). Details of the climate in this region are given by
Zeng and Zhong (1997), and Li et al. (1998). Soils in this
region are loamy, acidic and yellowish (Zeng and Zhong,
1997; Li et al., 1998).
Ficus microcarpa L. and F. virens Ait. var.
sublanceolata (Miq.) Cornor are broad-leaved tree species
with entire leaves; the former is evergreen and the latter is
deciduous. Individuals of both species can reach a height
of 20 m (Institute of Botany of The Chinese Academy of
Science, 1980). In the Three Gorges reservoir region, trees
of these two species start their growth in early spring (early
March), and new leaves and shoots can occur during the
whole growing season. No apparent growth in these spe-
cies can be observed during winter. Due to the high pro-
pensity of cut branches to root, it is quite easy to vegeta-
tively propagate these two species.
1.2 Tree preparation and crown damage treatment
In early 1996, small trees （ca. 1.2 m）of each of F.
microcarpa and F. virens species were planted in the ex-
perimental garden. Trees of these two species were grown
with enough spacing to avoid mutual shading during the
whole experiment. Weeding, watering, and insecticide spray-
ing were applied to trees of both species when necessary.
In early 1997, after one year of growth and acclimation,
20 blocks were established in both the F. microcarpa and
the F. virens stands. Each block contained seven trees,
one of which was subjected to each of the following
treatments: 0% (control); 20%, 50%, or 70% crown damage
in spring; 20%, 50%, or 70% crown damage in autumn. Trees
in each block were assigned to the treatments randomly.
Crown damage treatment was conducted in May 1997 for
spring-treated F. microcarpa and F. virens trees after their
spring flush and in October 1997 for autumn-treated F.
microcarpa and F. virens trees before the natural leaf shed-
ding of F. virens trees. The tree heights of F. microcarpa
and F. virens were ca. 1.6 and 1.8 m at spring treatment,
and ca. 2.0 and 2.3 m at autumn treatment, respectively. To
simulate the mode of crown damage usually adopted by
local people in Three Gorges reservoir region, branches in
the lower crown of trees were removed, leaving the branches
in the upper crown unaffected. As a result, crown damage
reduced the crown depths (the distance of the apical mer-
istem to the insertion point of the lowest branch) of trees
by 0%, 20%, 50%, and 70%, respectively (Fig.1).
Consequently, crown-damaged trees had longer bare stem
parts without any lateral branches. Henceforth, the stem
parts within and beneath the residual crown of each tree
after damage treatment are referred to as branched stem
parts and bare stem parts, respectively (Fig.2). The elon-
gated upper stem parts produced after the treatment are
referred to as new stem parts (Fig.2).
1.3 Determination of shoot production
For both F. microcarpa and F. virens trees, the crowns
of 0%, 20%, 50%, and 70% damaged trees, just after
treatment, were visually separated into 4, 3, 2, and 1 layers.
For a specific tree, each layer had approximately equal
depth. Layers from the top of the crown downwards were
ZENG Bo et al.: Position-dependent Shoot Production of Two Subtropical Fig Tree Species Following Crown Damage 909
labelled as Ⅰ, Ⅱ, Ⅲ, Ⅳ for the 0% damaged trees (control);
Ⅰ, Ⅱ, Ⅲ for the 20% damaged trees; Ⅰ, Ⅱ for the 50%
damaged trees; and Ⅰ for the 70% damaged trees (Fig.1).
For either F. microcarpa or F. virens, The height and the
light environment of any crown layer with the same label
(Ⅰ, Ⅱ or Ⅲ) were approximately similar in all trees, regard-
less of damage intensity (data not shown). In each of Ⅰ, Ⅱ
and Ⅲ layers of all trees, one vigorous south-facing first-
order lateral branch (starting from the stem) was chosen
and marked. Thus, each of the 0%, 20%, 50%, and 70%
crown-damaged trees had 3, 3, 2, and 1 marked first-order
lateral branches, respectively. Shoot production on the main
axes of all marked lateral branches was followed for one
year after treatment. The elongation of the main axes was
measured. The number of new shoots (viz. second-order
twigs) on the newly elongated parts of the lateral branches
was recorded, and the density of new shoots (number of
new shoots per unit length of axis) was calculated. Similarly,
the growth of the main stem was also followed for one year
after treatment. The number and density of newly grown
shoots on the newly elongated, the branched, and the bare
stem parts of the main stem were recorded. New shoots
were counted if they were longer than 3 cm.
1.4 Data analysis
For the elongation of the main axes of lateral branches,
and the density of new second-order shoots on the main
axes of lateral branches, one-way ANOVAs were used to
check the effects of crown layer. Since the elongation of
the main axes of lateral branches, and the density of new
second-order shoots on the main axes of lateral branches
did not differ among crown layers, for each of these two
variables, data of all crown layers in a tree were pooled and
two-way ANOVAs were used to evaluate the effects of dam-
age intensity and damage season.
The number and density of newly emerged shoots on
the elongated new stem parts, the branched, and the bare
stem parts of treated trees were explored by applying two-
way ANOVAs (factors are damage season and damage
intensity). Differences in these two traits between different
damage intensities were evaluated with Duncan’s multiple
range test. Logarithmic transformation was performed if
needed to improve the equality of variances prior to statis-
tical analysis.
Fig.1. Illustration of crown damage on Ficus microcarpa and F. virens trees and the separation of remaining crowns. Crown damage was
simulated by removing the branches from the lower tree crown. Crown damage intensities were 0%, 20%, 50%, and 70%, which implies
that the crown depth was reduced by 0%, 20%, 50%, and 70%, respectively. Trees subjected to heavier damage have longer bare stem
parts. The remaining crowns of 0%, 20%, 50%, and 70% damaged trees were visually separated into 4, 3, 2, and 1 layers, respectively.
For a specific tree, each layer of the remaining crown has approximately equal depth.
Fig.2. Diagramatic illustration of crown enlargement due to tree
growth. Crowns indicated with solid lines and dash lines refer to
the crown status instantaneously after damage treatment and some
time following damage treatment respectively. S1, newly grown
upper stem part following treatment; S2, stem part with lateral
branches (branched stem part) instantaneously after treatment;
S3, stem part without branches (bare stem part) instantaneously
after treatment; B1, the main axis of a lateral branch instanta-
neously after treatment; B2, the newly elongated main axis part.
Acta Botanica Sinica 植物学报 Vol.46 No.8 2004910
2 Results
For both F. microcarpa and F. virens, as far as the lat-
eral branches are concerned, new second-order shoots
chiefly emerged on the elongated parts of main axes of
lateral branches, the old parts of main axes hardly had new
second-order shoots. It was found that crown layer and
branch removal did not affect the elongation of the main
axes of lateral branches and the production of new shoots
on the elongated parts of main axes (data not shown). The
density of new shoots on the elongated main axes of lateral
branches was not affected by crown layer, damage inten-
sity and damage season (data not shown).
The number and the density of newly emerged shoots
on the new stem parts of both F. microcarpa and F. virens
trees were not affected by crown damage, either conducted
in spring or in autumn (Fig.3). No shoot sprouted from the
branched stem parts of damaged F. microcarpa and F. virens
trees one year after treatment. However, some shoots
sprouted from the bare stem parts of damaged trees in these
two species (Table 1), and their number and density in-
creased with damage intensity in both F. microcarpa and
F. virens trees (Table 1; Fig. 4). With respect to the effect of
dammage season, it was found that autumn-damaged trees
of both species had larger numbers and higher densities of
sprouted shoots on the bare stem parts than spring-dam-
aged trees (P < 0.001) (Table 2).
3 Discussion
It was found in this study that, for both F. microcarpa
and F. virens, no difference in shoot production of lateral
branches in the remaining crown of trees was detected, in
terms of either the density or the number of shoots (data
not shown). Crown-damaged trees did not present lower
shoot production in the upper crown and higher shoot pro-
duction at the lower crown as expected. Moreover, no ef-
fect of crown layer on shoot production was found in trees
subjected to any damage intensity (data not shown). As to
the shoot production on the tree stem of two species, it
was revealed that, crown damage, either conducted in spring
or in autumn, had no effect on the number and density of
newly emerged shoots on the newly grown upper stem
parts of both species (Fig.3). For the stem parts within the
residual crown in damaged F. microcarpa and F. virens
trees, no new shoots emerged there during one year after
the treatment. However, on the bare stem parts beneath the
Fig.3. Number and density of newly emerged branches (mean ± SE) on the new stem parts of Ficus microcarpa and F. virens trees one
year after crown damage treatment. Crown damage treatment was done in spring or in autumn by removing branches from the lower tree
crown, at the intensities of 0% (control), 20%, 50%, and 70%. The effects of damage season and damage intensity were evaluated with
two-way ANOVAs.
ZENG Bo et al.: Position-dependent Shoot Production of Two Subtropical Fig Tree Species Following Crown Damage 911
residual crown of both damaged F. microcarpa and F. virens
trees, some dormant buds broke and developed into shoots
within one year after the damage treatment, their numbers
increasing with damage intensity (Table 1). It is obvious
that shoot production on the tree stem of these two Ficus
species following crown damage was position-dependent.
Many studies have demonstrated that apical meristems
(Rinne et al., 1993; Wang et al., 1997; Sundberg and Uggla,
1998) and young leaves (Rinne et al., 1993; Wang et al.,
1997; Kotov and Kotova, 2000) are able to produce auxin,
which is transported downwards (Sundberg and Uggla,
1998; Kotov and Kotova, 2000) and plays an important role
in the exertion of apical dominance. The quantity of auxin is
thought to be related to the degree of apical dominance
(Romano et al., 1993; Cline, 1997). Some studies have re-
vealed that removal of apical meristems frees the lower
(proximal) axillary buds from apical dominance (Ouellette
and Young, 1994; Chaar et al., 1997; Lortie and Aarssen,
Table 1 The number of sprouted shoots (mean ± SE) on the bare stem parts (beneath the residual crown) of two Ficus tree species
within one year after crown damage treatments

Species
Damage intensity
Control 20% 50% 70%
Ficus microcarpa
Spring 0 (a) 0.25 ± 0.20 (a) 1.05 ± 0.22 (b) 2.40 ± 0.51 (b)
Autumn 0 (a) 0.95 ± 0.29 (a) 4.95 ± 0.91 (b) 13.8 ± 1.69 (c)
Damage intensity: P < 0.001; damage season: P < 0.001; interaction: P < 0.001
F. virens
Spring 0 (a) 0.55 ± 0.28 (a) 2.35 ± 0.69 (b) 4.25 ± 0.91 (c)
Autumn 0 (a) 3.3 ± 0.80 (a) 6.85 ± 1.00 (b) 12.2 ± 1.75 (c)
Damage intensity: P < 0.001; damage season: P < 0.001; interaction: P < 0.01
Crown damage was done in spring or autumn, at the intensities of 0% (control), 20%, 50%, 70%. Two-way ANOVAs were applied to evaluate
the effects of damage season and damage intensity. For trees damaged either in spring or in autumn, means followed by the same letter do not
significantly differ with each other (Duncan’s multiple range test).
Fig.4. Density (mean ± SE) of sprouted branches on the bare stem parts beneath the residual crown of two Ficus trees one year after
crown damage treatment. Crown damage was conducted in spring or in autumn by removing branches from the lower tree crown, at the
intensities of 0% (control), 20%, 50%, and 70%. The effects of damage season and damage intensity were evaluated with two-way
ANOVAs. For each species, at each damage season, means with the same letter do not significantly differ from each other (Duncan’s
multiple range test).
Table 2 Stem height and basal diameter (cm) (mean ± SE) of Ficus microcarpa and F. virens trees at the time of crown damage treatment
in spring or in autumn
Spring-damaged trees Autumn-damaged trees t-test
F. microcarpa
Stem height 157.8 ± 4.5 201.3 ± 4.8 P < 0.001
Stem basal diameter 1.70 ± 0.05 2.49 ± 0.07 P < 0.001
F. virens
Stem height 174.7 ± 4.1 232.0 ± 6.5 P < 0.001
Stem basal diameter 2.52 ± 0.07 3.58 ± 0.09 P < 0.001
t-test was applied to evaluate the difference among trees when they experienced crown damage in spring or in autumn.
Acta Botanica Sinica 植物学报 Vol.46 No.8 2004912
1997; Venecz and Aarssen, 1998). A bud on the main stem of
a tree is influenced by the confluent auxin produced by all
apices and vigorous leaves above it, which are located on
both the lateral branches and the stem. In our experiment,
crown damage only led to the removal of lower branches
from the stem without touching the upper branches. As a
consequence, the flux of auxin in the stem parts within the
residual crown (viz. branched stem parts) after treatment
could not have been altered. In contrast, the flux of auxin to
buds on the bare stem parts beneath the residual crown
after treatment would be diminished compared to buds in
similar positions on control trees, because the treatment
had removed part of the sources of auxin. The differential
bud burst pattern of different stem parts found in this study
seems due to this altered auxin flux along the stem. It is
obvious in this study that heavy crown damage created
longer bare stem parts which accommodated more dormant
buds (Fig.1). Here, a question is arising: was the stronger
shoot production on the bare stem parts of heavily dam-
aged trees simply caused by the fact that the bare stem
parts of heavily damaged trees had more dormant buds
than those of lightly damaged trees or by a more strongly
reduced apical dominance? Since the density of new shoots
on the bare stem parts increased with damage intensity
(Fig.4), we suggest that the enhanced shoot production on
the bare stem parts of heavily damaged trees was not merely
caused by the larger supply of dormant buds, but was mainly
caused by a stronger release of dormant buds from apical
dominance, so that more dormant buds per unit length of
bare stem parts were able to burst.
It was shown in the experiment that the shoot produc-
tion of lateral branches in the crown was not affected by
crown damage. Apparently, the shoot production pattern
of lateral branches in crown cannot be simply explained by
structural unit theory or pipe unit concept, some other fac-
tors may be involved in the shoot production in crown.
Because the shoot production of lateral branches in the
remaining crown of damaged F. microcarpa and F. virens
trees were not affected by the treatment, it appears that
only the enhanced shoot production on the bare stem parts
may contribute to the increase in biomass partitioning to
leaf growth in damaged F. microcarpa and F. virens trees
and benefit the regrowth of damaged trees. Comparatively,
less carbohydrates are needed for constructing shoots
emerging directly from the main stem than for shoots emerg-
ing from the lateral branches (Cannell et al., 1988; Ford et
al., 1990), due to the shorter appending vessels to the roots.
Therefore, with a given amount of carbohydrates for new
growth, more of it can be invested in leaf growth if new
shoots emerge from the main stem rather than from the
lateral branches.
It was shown in this study that damage season affected
the shoot production on the bare stem parts. In both F.
microcarpa and F. virens trees, more shoots per unit length
of the bare stem parts grew out after autumn treatment as
compared to spring treatment (Table 1; Fig.4). It is likely
that this difference in shoot production was not due to the
different weather conditions between spring and autumn
when trees were damaged, but was caused by the availabil-
ity of nutrients and carbohydrates for shoot development.
In this study, all autumn-treated trees were five months
older than the spring-treated trees when they experienced
crown damage treatment. Both the heights and diameters
of autumn-treated trees were larger than those of spring-
treated trees (P < 0.001) (Table 2). Presumably, the enhanced
shoot production on the bare stem parts of autumn-treated
trees could be caused by the larger nutrient and carbohy-
drate reserve in the roots and stems of the trees at the time
of treatment.
F. microcarpa and F. virens are different in leaf habits:
the former is an evergreen species, and the latter is
deciduous. However, as regards the shoot production
pattern, these two species responded similarly to crown
damage, either conducted in spring or in autumn. This im-
plies that leaf habits are not crucial in affecting the response
of shoot production to crown damage, as least for these
two studied species.
Acknowledgements: We are grateful to Prof. Marinus J.
A. Werger and Dr. Heinjo During in Utrecht University,
The Netherlands, and Dr. Frank Sterck in Wageningen Uni-
versity in The Netherlands for their helpful comments on
an earlier version of the manuscript.
References:
Cannell M G R, Morgan J, Murray M B. 1988. Diameters and
dry weights of tree shoots: effects of Young’s modulus, taper,
deflection and angle. Tree Physiol, 4: 219-231.
Chaar H, Colin F, Leborgne G. 1997. Artificial defoliation, de-
capitation of the terminal bud, and removal of the apical tip of
the shoot in sessile oak seedlings and consequences on subse-
quent growth. Can J For Res, 27: 1614-1621.
Cline M G. Concepts and terminology of apical dominance. 1997.
Am J Bot, 84: 1064-1069.
Cornelissen J H C. 1992. Seasonal and year to year variation in
performance of Gordonia acuminata seedlings in different
light environments. Can J Bot, 70: 2405-2414.
ZENG Bo et al.: Position-dependent Shoot Production of Two Subtropical Fig Tree Species Following Crown Damage 913
Ford E D, Avery A, Ford R. 1990. Simulation of branch growth in
the Pinaceae: interactions of morphology, phenology, foliage
productivity and the requirement for structural support on
the export of carbon. J Theor Biol, 146: 15-36.
Franco M. 1985. A modular approach to tree production. White
J. Studies on Plant Demography: A Festschrift for John L.
Harper. London: Academic Press.
Institute of Botany of The Chinese Academy of Sciences. 1980.
Iconographia Cormophytorum Sinicorum. TomusⅠ. Beijing:
Science Press. 473-485. (in Chinese)
Kotov A A, Kotova L M. 2000. The contents of auxins and
cytokinins in pea internodes as related to the growth of lateral
buds. J Plant Physiol, 156: 438-448.
Li R, During H J, Werger M J A, Zhong Z C. 1998. Positioning of
new shoots relative to adult shoots in groves of the giant
bamboo Phyllostachys pubescens. Flora, 193: 315-322.
Lortie C J, Aarssen L W. 1997. Apical dominance as an adapta-
tion in Verbascum thapsus: effects of water and nutrients on
branching. Int J Plant Sci, 158: 461-464.
Ouellette D R, Young E. 1994. Branch inducement in apple
stoolbed shoots by summer leaf removal and tipping.
HortScience, 29: 1478-1480.
Prévost M F. 1967. Architecture de quelques Apocynacées
ligneuses. Mémoires de la Société Botanique de France.
Colloque sur la physiology de l a´rbre. 23-36.
Rinne P, Tuominen P, Sundberg B. 1993. Growth patterns and
endogenous indole-3-acetic acid concentrations in current-year
coppice shoots and seedlings of two Betula species. Physiol
Plant, 88: 403-412.
Romano C P, Cooper M L, Klee H J. 1993. Uncoupling auxin and
ethylene effect in transgenic tobacco and Arabidopsis plants.
Plant Cell, 5: 181-189.
Sachs T. 1991. Pattern Formation in Plant Tissues. Cambridge:
Cambridge University Press.
Shinozaki K, Yoda K, Hozumi K, Kira T. 1964a. A quantitative
analysis of plant form — the pipe model theory Ⅰ. Basic
analyses. JPN J Ecol, 14: 97-104.
Shinozaki K, Yoda K, Hozumi K, Kira T. 1964b. A quantitative
analysis of plant form — the pipe model theory II. Further
evidence of the theory and its application in forest ecology.
JPN J Ecol, 14: 133-139.
Sprugel D G, Hinckley T M, Schaap W. 1991. The theory and
practice of branch autonomy. Ann Rev Ecol Syst, 22: 309-
334.
Sundberg B, Uggla C. 1998. Origin and dynamics of indoleacetic-
acid under polar transport in Pinus sylvestris. Physiol Plant,
104: 22-29.
Venecz J I, Aarssen L W. 1998. Effects of shoot apex removal in
Lythrum salicaria (Lythraceae): assessing the costs of repro-
duction and apical dominance. Ann Bot Fen, 35: 101-111.
Wang Q, Little C H A, Oden P C. 1997. Control of longitudinal
and cambial growth by gibberellins and indole-3-acetic acid in
current-year shoots of Pinus sylvestris. Tree Physiol, 17: 715-
721.
Watson M A, Casper B B. 1984. Morphogenetic constraints on
patterns of carbon distribution in plants. Ann Rev Ecol Syst,
15: 233-258.
Zeng B, Zhong Z-C . 1997. Polyacrylamides membrane chroma-
tography of flavonoids in Gordonia acuminata. Acta Phytoecol
Sin, 21: 90-96. （in Chinese with English abstract）
(Managing editor: HAN Ya-Qin)