Tree Physiology 20, 987–991
© 2000 Heron Publishing—Victoria, Canada
Shoot growth responses to light microenvironment and correlative
inhibition in tree seedlings under a forest canopy
AKIO TAKENAKA
National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba 305-0053, Japan
Received October 26, 1999
Summary To examine the mechanisms underlying crown
development, I investigated the dependence of shoot behavior
on light microenvironment in saplings of the evergreen broadleaved tree species, Litsea acuminata (Bl.) Kurata, growing on
a forest floor. The local light environment of individual shoots
(shoot irradiance) and plants (plant irradiance, defined as the
shoot irradiance of the most sunlit shoot of a plant) were analyzed as factors affecting shoot behavior. Daughter shoots that
developed under partially sunlit conditions were longer and
less leafy than daughter shoots developed under shaded conditions. Shoot production increased with increasing shoot
irradiance. Terminal shoots receiving 5% or less of full sunlight produced 0.67 daughter shoots on average, whereas
shoots receiving 10% or more of full sunlight produced 1.72
daughter shoots. In terminal shoots receiving 5% or less of full
sunlight, the probability of producing no daughter shoots was
about 63% when other shoots on the plant received 10% or
more of full sunlight, but was < 35% where the rest of the plant
was also shaded. Shoot death was observed only in shoots receiving 5% or less of full sunlight. The mortality of shaded
shoots was higher in plants growing in high irradiance than in
plants growing in low irradiance. The ecological significance
of correlative inhibition (the enhanced mortality and reduced
production of new shaded shoots in the presence of partiallysunlit shoots) is discussed.
Keywords: branching, Litsea acuminata, shoot demography,
shoot structure.
Introduction
The three-dimensional (3-D) branching structure of a tree,
which provides a framework for leaf display, has an important
functional significance, because the spatial distribution of
leaves critically affects the efficiency of light capture and the
competitive ability of the tree (Ackerly and Bazzaz 1995,
Valladares 1999). The complex branching structure of a tree is
formed through the repetitive production of shoots. Spatial
distributions of new shoot formation, shoot mortality and the
structure and geometry of new shoots determine the 3-D
branching structure of a tree. Thus, the process of tree crown
development can be considered as a structural dynamic of the
shoot population (Maillette 1982a, 1982b).
The production of new shoots on parent shoots is affected
by light microenvironment. Shoots growing in high light tend
to make more daughter shoots than shoots growing in shade
(Koike 1989, Stoll and Schmid 1998). Interaction between the
spatial distribution of light and new shoot production plays an
important role in competition among neighboring trees
(Sorrensen-Cothern et al. 1993, Takenaka 1994). Takenaka
(1994) showed that increased new shoot formation in response
to high irradiances enables a tree to increase leaf area in sunlit
spaces. A similar phenomenon has been observed for ramet
production among clonal plants. In many species, more ramets
are produced under resource-rich conditions than under resource-limited conditions (de Kroon and Hutchings 1995).
This environmental dependency of ramet production is considered to be advantageous in the acquisition of heterogeneously distributed resources (Sutherland and Stillman 1988,
Cain et al. 1996).
The autonomy of shoots with respect to their carbon economy (Sprugel et al. 1991) may partially explain the dependence of new shoot formation on light microenvironment in a
tree crown. However, shoots are not completely mutually independent. Stoll and Schmid (1998) showed that sunlit shoots
inhibit new shoot formation on shaded shoots of the same tree.
In herbaceous plants, Novoplansky et al. (1989) reported increased mortality and reduced growth of a shaded shoot when
there is a shoot at higher irradiance. They called this limitation
of shaded shoots “correlative inhibition.” Correlative inhibition may be an important factor affecting the response of tree
crown structure to a spatially heterogeneous light environment. This is because inhibition of the growth and survival of
shoots in the shaded part of a crown is likely to be coupled
with enhanced shoot production in the sunlit part of the crown.
The main objectives of this study were to clarify the dependency of shoot behavior on the light microenvironment
within a tree crown, and to determine if there is correlative inhibition of shoot growth and survival. Specifically, I investigated the light dependency of new shoot production, bud
dormancy, and shoot mortality of seedlings and saplings of an
evergreen broad-leaved tree species, Litsea acuminata (Bl.)
Kurata, growing under a forest canopy. This species was chosen because the ramification of branches is not intense. Saplings of about 3 m in height growing on a forest floor have
988
TAKENAKA
about 30 to 40 terminal shoots. The limited number of shoots
facilitates a whole-crown survey of shoot behavior.
Materials and methods
The study area (30°21′ N, 130°28′ E, about 200 m a.s.l.) is in
an evergreen broad-leaved forest in the western part of
Yakushima, an island located in the warm-temperate zone of
Japan. Mean annual temperature and precipitation during the
period 1992–1996 at the meteorological station on the island
(30°23′ N, 130°40′ E, about 36 m a.s.l.) are 19.2 °C and
4595 mm, respectively. The dominant tree species of the canopy are evergreens such as Litsea acuminata, Neolitsea
aciculata (Bl.) Koidz, Quercus salicina Blume, Ardisia
sieboldii Miq. and Ilex rotunda Thunb. The height of the canopy trees is around 15 m.
Litsea acuminata is a broad-leaved evergreen tree species in
the family Lauraceae. Leaf longevity ranges from 2 to 5 years,
or even more under shaded conditions. Stem extension growth
is determinate. Winter buds break in May and form orthotropic
shoots. Hereafter, a shoot refers to the growth increment originating from a bud.
Shoot length varies over a wide range from a few millimeters to more than 30 cm. Leaves are narrow and lanceolate.
The number of leaves on a shoot varies from a few to more
than 20. Saplings of about 3 m in height growing on the forest
floor have about 30 to 40 terminal shoots. The limited number
of shoots facilitated a whole-crown survey of shoot behavior.
In May 1997, 40 seedlings and saplings of L. acuminata
were chosen. Their heights ranged from 30 cm to about 3 m.
Plants were chosen to cover a range of light environments,
from shaded conditions under a dense, closed canopy to partially sunlit conditions near canopy gaps. All terminal shoots
were marked, and their length, leaf number and the length of
their largest leaf were recorded. Current-year shoots were not
marked because they were not yet developed at this time. The
total area of leaves on a shoot was estimated from the product
of the number of leaves and the squared length of the largest
leaf on the shoot, based on a regression equation generated
from 40 sample shoots (R 2 = 0.97). Photosynthetic photon flux
density (PPFD) at each marked shoot was measured on an
overcast day with a quantum sensor (LI-190SA, Li-Cor, Inc.,
Lincoln, NE). The measurement was repeated three times on
each shoot and averaged. The values are expressed relative to
the PPFD measured at the top of a nearby tower above the forest canopy.
In November 1997, current-year shoots were marked, and
their length, leaf number and the length of their largest leaf
were recorded. The survival of the terminal shoots marked in
May and November 1997 was checked in May 1998.
The shoot census and light measurements were repeated for
the same sample plants from May 1998 to May 1999. Shoots
that died as a result of external factors, such as grazing by deer,
were not taken into account in further analysis.
Results
New shoot production and mortality were not significantly
different between the first and second years of observation.
Therefore, the results of the 2 years were combined for the following analysis.
Shoot irradiance ranged from 0.4 to 21.8% of that above the
forest canopy. Terminal shoots were classified into three
groups according to their irradiance: < 5%, 5–10% and > 10%
of full sunlight. The sample plants were also classified into
three groups according to plant irradiance (Table 1). The behaviors of shoots at different shoot irradiances were compared
on the basis of the irradiance of the plants to which the shoots
belonged.
The number of new daughter shoots produced on a terminal
parent shoot increased with increasing irradiance at the parent
shoot (Figure 1). Mean number of new daughter shoots at the
Table 1. Numbers of plants and shoots in different irradiance classes.
Shoot irradiance is the photosynthetic photon flux density above the
shoot relative to that above the forest canopy. Plant irradiance is the
highest shoot irradiance measured for the plant. Data from 40 samples
over two seasons are pooled. Note that shoot irradiance cannot exceed
plant irradiance.
Shoot irradiance
Plant irradiance
0–5%
5–10%
No. of plants
57
15
0–5%
5–10%
> 10%
No. of shoots
221
75
59
> 10%
Total
8
80
72
48
61
368
107
61
Figure 1. Mean number of daughter shoots formed on individual terminal shoots of L. acuminata at different shoot irradiances. Shoots are
grouped by their light microenvironment. Mean values calculated for
all sample plants (the right-most column of each group of columns)
were significantly different between shoot irradiances (P < 0.01,
Mann-Whitney U-test). Within each shoot irradiance, mean values
calculated separately for each plant irradiance are significantly different when followed by different letters.
TREE PHYSIOLOGY VOLUME 20, 2000
SHOOT GROWTH RESPONSES TO LIGHT MICROENVIRONMENT
highest shoot irradiance was 1.72. Many shoots at this irradiance produced more than one daughter shoot, forming a branching structure. In contrast, branching seldom occurred in
shoots at the lowest shoot irradiance. Among shoots at the
lowest shoot irradiance, those on plants with the highest plant
irradiance produced significantly fewer daughter shoots than
those on more shaded plants.
Relative frequency of shoots that produced no daughter
shoots increased with decreasing shoot irradiance (Figure 2).
Almost all shoots at the highest irradiance produced one or
more daughter shoots. On the other hand, about 40% of shoots
at the lowest irradiance produced no daughter shoots. Among
the shoots at the lowest shoot irradiance, the relative frequency of shoots without daughter shoots was significantly
greater in shoots of plants at the highest plant irradiance.
Mean length of new daughter shoots was positively correlated with shoot irradiance (Table 2). Plant irradiance had no
significant effect on daughter shoot length (data not shown).
Total area of leaves per unit stem length was calculated as an
index of leafiness of a shoot. The ratio was negatively correlated with shoot length (Figure 3): short shoots were relatively
leafy, whereas long shoots had less leaf area per unit shoot
length.
Shoot death was observed only in shoots at the lowest shoot
irradiance (Figure 4). Mortality (relative frequency of shoots
989
Figure 3. Leafiness of a shoot (total leaf area per unit shoot length)
plotted against shoot length.
Figure 4. Mortality (relative frequency of shoots dying in a year) of
terminal shoots at different shoot irradiances. Mortality of shoots at a
shoot irradiance of 0–5% was significantly higher than that of shoots
at the other shoot irradiances (P < 0.01, Fisher’s exact test). Other details are as in Figure 1.
Figure 2. Relative frequency of terminal shoots showing no bud expansion at different shoot irradiances. Mean values calculated for all
sample plants (the right-most column of each group of columns) were
significantly different between shoot irradiances (P < 0.01, Fisher’s
exact test). Other details are as in Figure 1.
Table 2. Mean length of daughter shoots that sprouted from parent terminal shoots at different shoot irradiances. Differences were significant for each pair of comparisons at the 1% level (Mann–Whitney
U-test).
Irradiance of parent shoots
Mean length (cm)
SD (cm)
No. of daughter shoots
0–5%
5–10%
> 10%
6.9
5.3
233
12.7
7.3
128
16.9
7.2
105
that died per year) of the shaded shoots was significantly
higher in plants at the highest irradiance. For plants receiving
10% or more of full sunlight, mortality of the shaded shoots
was 28%. For plants receiving < 10% of full sunlight, it was as
low as 2%.
Few shoots died in their first year (Table 3). Terminal shoots
that did not sprout new shoots the following spring suffered
20.5% mortality. Almost 50% of terminal shoots that did not
sprout new daughter shoots for 2 years died in their third year.
Discussion
Shoot behavior of L. acuminata saplings depended on light
microenvironment. Shoot production was higher and shoot
mortality less in sunlit microenvironments than in shaded
microenvironments. Similar findings have been reported for
shoot production in other tree species (Koike 1989, Stoll and
Schmid 1998) and for ramet production in clonal plants (de
Kroon and Hutchings 1995). Production of more shoots or
ramets under resource-rich conditions than under resourcepoor conditions may be a general phenomenon contributing to
TREE PHYSIOLOGY ON-LINE at http://www.heronpublishing.com
990
TAKENAKA
Table 3. Mortality of terminal shoots at different ages. Current-year
terminal shoots sprouted during the latest spring; 1-year-old shoots
experienced one spring without daughter shoot production; 2-yearold shoots experienced two springs without daughter shoot production. Differences in mortality between shoots of different ages were
significant for each pair of comparisons at the 1% level (Fisher’s exact test).
Age of terminal shoot (years)
No. of shoots
No. of shoots
dying in a year
Mortality (%)
Current-year
One-Year-old
≥ Two-Year-old
445
2
73
15
17
8
20.5
47.1
0.45
for extension are formed under less shaded conditions. Production of long shoots preferentially in sunlit spaces is beneficial in expanding the framework for supporting more leaves to
acquire more light, and also in preparing for future competition for light with neighboring trees. The production of long
shoots preferentially under high-light conditions has also been
reported in other tree species (Koike 1989, Stoll and Schmid
1998). It is concluded that relative investment of resources in
structurally and functionally different shoots that also differ in
their response to the light environment is of great importance
in the development of a tree crown that is efficient at light capture.
Acknowledgments
resource acquisition by plants growing in spatially heterogeneous environments.
Evidence for correlative inhibition of shoot behavior was
obtained. Thus, sunlit shoots inhibited both formation of new
shoots and survival of shaded shoots. Number of daughter
shoots of shaded parent shoots decreased and the mortality of
shaded shoots increased with increasing plant irradiance.
Among the 25 shoot deaths observed, 17 (68%) were on current-year and 1-year-old terminal shoots, and eight (32%)
were on 2-year-old or older shoots (Table 3). Because leaves
of L. acuminata can survive for more than 5 years in the study
area, the observed mortality of terminal shoots cannot be attributed solely to loss of aging leaves.
A probable advantage of the correlative inhibition of shaded
shoots in the presence of sunlit shoots is that it enables a tree to
save the cost of supporting relatively unproductive shoots and
to invest resources preferentially in the development of shoots
in spaces with favorable light conditions. Thus, photoassimilates, water and nutrients are saved or reallocated by casting
off shaded shoots and limiting new shoot formation under
shaded conditions.
Stoll and Schmid (1998) observed correlative inhibition of
new shoot formation in canopy trees of a Pinus sylvestris L.
stand. In the present study, correlative inhibition was observed
in seedlings and saplings growing on the forest floor. Additional studies are needed to determine whether correlative inhibition is a common phenomenon occurring in different
species and life forms at different life stages.
Takenaka (1997) reported within-species variations in shoot
structure in eight broad-leaved evergreen tree species, including L. acuminata, that lack distinct morphological differences
between short and long shoots. Because total leaf area per unit
stem length was greater in relatively short shoots than in long
shoots, it was suggested that shorter shoots were oriented for
leaf display and longer shoots were oriented for extension.
This suggestion is in agreement with both the previous assignment of roles for morphologically differentiated short and
long shoots (Hallé et al. 1978, Jones and Harper 1987), and the
findings of this study that short shoots oriented for leaf display
are formed under shaded conditions and long shoots oriented
This work was partly supported by a Grant-In-Aid from the Ministry
of Education, Science and Culture, Japan (Grant No. 10440233). It
was also partly supported by the Yakushima World Heritage Center,
Environment Agency, Japan. I thank N. Adachi for his help in the
fieldwork and acknowledge N. Kachi, Y. Okabe, A. Sumida and
I. Terashima for their discussion and critical comments on the manuscript.
References
Ackerly, D.D. and F.A. Bazzaz. 1995. Seedling crown orientation and
interception of diffuse radiation in tropical forest gaps. Ecology
6:1134–1146.
Cain, M.L., D.A. Dudle and J.P. Evans. 1996. Spatial models of foraging in clonal plant species. Am. J. Bot. 83:76–85.
de Kroon, H. and M.J. Hutchings. 1995. Morphological plasticity in
clonal plants: the foraging concept reconsidered. J. Ecol. 83:
143–152.
Hallé, F., R.A.A. Oldeman and P.B. Tomlinson. 1978. Tropical trees
and forests. Springer-Verlag, Berlin, 441 p.
Jones, M. and J.L. Harper. 1987. The influence of neighbours on the
growth of trees. II. The fate of buds on long and short shoots in
Betula pendula. Proc. R. Soc. Lond. B 232:19–33.
Koike, F. 1989. Foliage-crown development and interaction in Quercus gilva and Q. acuta. J. Ecol. 77:92–111.
Maillette, L. 1982a. Structural dynamics of silver birch. 1. The fates
of buds. J. Appl. Ecol. 19:203–218.
Maillette, L. 1982b. Structural dynamics of silver birch. 2. A matrix
model of the bud population. J. Appl. Ecol. 19:219–238.
Novoplansky, A., D. Cohen and T. Sachs. 1989. Ecological implications of correlative inhibition between plant shoots. Physiol. Plant.
77:136–140.
Sorrensen-Cothern, K.A., E.D. Ford and D.G. Sprugel. 1993. A
model of competition incorporating plasticity through modular foliage and crown development. Ecol. Monogr. 63:277–304.
Sprugel, D.G., T.M. Hinckley and W. Schaap. 1991. The theory and
practice of branch autonomy. Annu. Rev. Ecol. Syst. 22:309–34.
Stoll, P. and B. Schmid. 1998. Plant foraging and dynamic competition between branches of Pinus sylvestris in contrasting light.
J. Ecol. 86:934–945.
Sutherland, W.J. and R.A. Stillman. 1988. The foraging tactics of
plants. Oikos 52:239–244.
Takenaka, A. 1994. A simulation model of tree architecture development based on growth response to local light environment. J. Plant
Res. 107:321–330.
TREE PHYSIOLOGY VOLUME 20, 2000
SHOOT GROWTH RESPONSES TO LIGHT MICROENVIRONMENT
Takenaka, A. 1997. Structural variation in current-year shoots of
broad-leaved evergreen tree saplings under forest canopies in
warm temperate Japan. Tree Physiol. 17:205–210.
991
Valladares, F. R. 1999. Architecture, ecology, and evolution of plant
crowns. In Handbook of Functional Plant Ecology. Eds. F.I.
Pugnaire and F. Valladares. Marcel Dekker Inc., New York, pp
121–194.
TREE PHYSIOLOGY ON-LINE at http://www.heronpublishing.com