Shoot growth patterns in saplings of Cleyera japonica in relation to light and architectural position

Transcription

1 Tree Physiology 23, Heron Publishing Victoria, Canada Shoot growth patterns in saplings of Cleyera japonica in relation to light and architectural position ARATA ANTONIO SUZUKI 1,2 1 Laboratory of Forest Biology, Department of Agriculture, Kyoto University, Oiwake-cho, Kitashirakawa, Sakyo-ku, Kyoto , Japan 2 Present address: Department of Biology, Graduate School of Science, Osaka University, Machikaneyama-cho, Toyonaka, Osaka , Japan (eurya@bio.sci.osaka-u.ac.jp) Received March 7, 2002; accepted July 20, 2002; published online December 2, 2002 Summary To gain further insight into crown development, the influences of shoot architectural position (branch order) and light environment on patterns of shoot growth of Cleyera japonica Thunberg (Theaceae) were investigated. Annual shoot length and light environment were positively correlated within same-order branches. Shoot length differed significantly among branch orders: shoot length was greater for the lower-order branches when light environments were comparable. Lower-order branches lengthened to a certain extent even if light availability was relatively low, whereas higher-order branches did not grow vigorously even when light availability was relatively high. Within same-order branches, branching was independent of the light environment of the shoot. Sylleptic shoot production differed significantly among branch orders, with most sylleptic shoots being produced on second-order branches. It is concluded that both light condition and architectural position of shoots must be considered when examining the mechanisms underlying crown development. Keywords: branching, branching order, crown architecture development, shoot length. Introduction The crown structure of a tree results from the repetitive production of shoots (Maillette 1982, Takenaka 2000). Describing shoot growth patterns and unraveling the external and internal factors that affect such patterns are essential to understanding the processes underlying tree crown development. Light environment is one of the most important external factors influencing shoot growth patterns (Steingraeber et al. 1979, Canham 1988, Tolvanen 1995). For example, Canham (1988) found that rates of height growth, lateral growth and new shoot production were up to an order of magnitude greater in Acer saccharum Marsh. and Fagus grandifolia Ehrh. saplings growing in canopy gaps than in saplings growing beneath closed canopies. Several internal factors affect shoot growth patterns, including the architectural positions (branch order) of shoots (Kozlowski and Ward 1961, Steingraeber 1982, Remphrey and Davidson 1994, Sabatier and Barthélémy 1999, Suzuki 2000). For example, Kozlowski and Ward (1961) found that Pinus spp. show strong correlative growth inhibition: the primary axis growing more than a secondary axis; a secondary axis growing more than a tertiary axis; and a tertiary axis growing more than a quaternary axis. Sabatier and Barthélémy (1999) observed that annual shoot lengths of young Cedrus atlantica (Endl.) Manetti ex Carrière trees decrease with increasing branch order. Because architectural position and light environment have not been simultaneously considered as factors influencing shoot growth, little is known about their interaction on growth patterns. The objective of this study was to examine the influences of shoot architectural position and light environment on shoot growth patterns in an evergreen broad-leaved woody plant, Cleyera japonica Thunberg (Theaceae), growing under a forest canopy. Materials and methods Study species Cleyera japonica is a small broad-leaved evergreen tree that is widely distributed in Taiwan, China, South Korea (Cheju Island) and Japan (Hirai 1996). Cleyera japonica has an excurrent branching habit and shows a multi-layer distribution of foliage (Ardhana et al. 1988). Vegetative growth begins between late April and early May and ends in autumn. Study area The study was carried out in a natural forest dominated by Chamaecyparis obtusa Endl. at the Kamigamo Experimental Forest Station of Kyoto University (35 04 N, E), about 12 km north of Kyoto. At this location, C. japonica is a common understory tree. Meteorological records were obtained from the Forest Station. Mean annual precipitation and evaporation are 1678 and 985 mm, respectively. Sampling procedures A m plot was established within the experimental forest. The density of C. japonica trees in the plot was

2 68 SUZUKI 12,300 ha 1. Over 88% of C. japonica trees in the plot were < 1.5 m in height and the tallest was 3.4 m tall. In early April 1999, three trees, m tall, were randomly selected. The branch order of each axis was determined. The trunk was assigned Order 1 axis; branches issuing directly from the trunk were called Order 2 axes; branches growing on Order 2 axes were called Order 3 axes, and so on (Suzuki 2002). The term axis is used here to define each linear succession of shoots of the same branch order. The term branch system denotes the set of lateral axes borne on the main axis, i.e., a branch system represents the set of axes comprising an Order 2 axis plus all of its lateral Order 3, 4 and 5 axes. On each tree, I selected branch systems from the upper part of the crown to minimize size variation among branches. A total of 29 branch systems (10, 10 and 9 branched systems per tree) was selected. Two-dimensional diagrams were drawn to describe the branching structure of each sampled branch system. Because C. japonica bears plagiotropic branch systems (having planar leaf arrangements) on vertical stems, all stems within a branch system are on a nearly horizontal plane. To analyze patterns of shoot growth in relation to branch order, the order number (centrifugal ordering) of each terminal branch was determined from the branching structure diagrams. In this paper, the terminal branch is defined as each terminal unbranched segment of a branch, which is equivalent to the first-order branch in a centripetal ordering system, such as the Strahler system (see Borchert and Slade 1981). Light measurements The light environment of each terminal branch was measured on an overcast day with a quantum sensor (LI-190SA, Li-Cor, Lincoln, NE) positioned immediately above and parallel to the 1-year-old leaf surface. Instantaneous photosynthetic photon flux density (PPFD) data for each terminal branch were measured five times on five cloudy days during September and October At the same time, light environments at open sites were also measured. The relative PPFD (RPPFD) for each terminal branch was calculated at the time of each measurement. The RPPFD values of five measurements for each terminal branch were averaged. Results Shoot length Relationships between shoot length and light environment for each terminal branch order are shown in Figure 1. The analysis of covariance of shoot length, with light as the covariate and branch order as the random factor, is presented in Table 1. Within a branch order, the gradients of shoot length relative to light availability were not significantly different (see the among slopes term of ANCOVA, Table 1), indicating that shoot length exhibited a consistent pattern of change in relation to light availability within branches of the same order. A significant covariate term indicates that the pooled regression of shoot length on light is significant (see covariate term of ANCOVA, Table 1). Within a branch order, shoot length was significantly and positively correlated with light environment. Furthermore, mean shoot length differed significantly among branch orders when the branch light environment was comparable, as indicated by a highly significant among-intercepts term. That is, current-year shoot length was greater in lower-order branches than in higher-order branches when light environments were comparable. There were no significant differences between trees for shoot length for all branch orders combined (ANOVA, P > 0.1). Branching patterns Relationships between numbers of sylleptic shoots per parent shoot and light environment for each terminal branch order are shown in Figure 2. In C. japonica, most lateral annual shoots on the upper parts of the crown are produced as sylleptic shoots; i.e., few proleptic lateral shoots are produced on the upper parts of the crown (unpublished data). Table 1 shows the results of the analysis of covariance of sylleptic shoot production, with light as the covariate and branch order as the random factor. The gradients of sylleptic shoot production in relation Statistical analysis Analysis of covariance (ANCOVA) was used to relate the gradients in shoot growth parameters (current-year shoot length and branching) to gradients of PPFD within each order of the terminal branch (see Ackerly 1992, Sokal and Rohlf 1995). The ANCOVA model was constructed with the order of the terminal branch as a factor and PPFD as a covariate. The interaction between the order of the terminal branch and PPFD was used to test for differences in slopes among orders of the branch. If current-year shoot length gave consistently parallel gradients of PPFD, then there would be a significant covariate effect in the ANCOVAs. A significant interaction term, on the other hand, would indicate that the slopes of the relationships vary among orders of the terminal branch. Figure 1. Relationship between current-year shoot length and relative photosynthetic photon flux density (RPPFD) at the terminal branch level. TREE PHYSIOLOGY VOLUME 23, 2003

3 LIGHT AND POSITION EFFECTS ON SHOOT GROWTH 69 Table 1. Summary of analysis of covariance of shoot growth parameters with light as a covariate and the three terminal branching orders as random factors. A significant interaction term between branching order and the covariate (among slopes) indicates that the slope of the relationship between the shoot growth parameter and the covariate was different on different terminal branching orders. A significant covariate term indicates that there was a significant correlation between the shoot growth parameter and the covariate. Dependent variables MS F P Shoot length Among intercepts < Covariate < Among slopes Branching Among intercepts < Covariate Among slopes to light were not significantly different among branch orders (Figure 2). However, there was no significant light covariate effect for sylleptic shoot production, reflecting little correlation between sylleptic shoot production and the light environment; i.e., branching was independent of the shoot light environment within the same-order branches. Sylleptic shoot production differed significantly among branch orders. There were no significant differences among trees for sylleptic shoot production for all branch orders combined (ANOVA, 2 degrees of freedom, P > 0.1). Relationships between shoot length and branching Relationships between shoot length and numbers of sylleptic shoots per parent shoot for each branch order are shown in Figure 3. Table 2 shows the results of the ANCOVA of sylleptic shoot production, with shoot length as the covariate and branch order as the random factor. The slopes of the regression lines were fitted in each group (branch order) to be parallel. Sylleptic shoot production exhibited a marginally significant shoot-length effect, reflecting a weak correlation with light availability. Discussion Although shoot length was correlated with PPFD within branches of the same order, the response of shoot length to light differed among branch orders. Shoot lengths decreased with increasing branch order when the light environment was comparable. Terminal shoots of the Order 2 axes lengthened to a certain extent even if light availability was relatively low, whereas terminal shoots of the higher-order branches did not grow vigorously even if light availability was high. The significant variation in shoot length on branches of different orders suggests pronounced internal control of growth (Kozlowski and Ward 1961, Brown et al. 1967, Cline 1991). However, the physiological mechanisms of internal control are not well understood (Cline 1991). Most sylleptic shoots were produced on terminal branches on Order 2 axes and few sylleptic shoots were produced on the higher-order branches, even when light availability was relatively high. The correlation between sylleptic shoot production and light environment was weak, even for terminal branches on Order 2 axes. A study of Cedrus atlantica by Sabatier and Barthélémy (1999) demonstrated that the mean number of long sylleptic shoots per parent shoot was greater Figure 2. Relationship between number of sylleptic shoots per shoot and relative photosynthetic photon flux density (RPPFD) for terminal branch Order 2 (a), Order 3 (b) and Order 4 (c). Figure 3. Relationship between the number of sylleptic shoots per shoot and current-year shoot length for terminal branch Order 2 (a), Order 3 (b) and Order 4 (c). TREE PHYSIOLOGY ONLINE at

4 70 SUZUKI Table 2. Summary of analysis of covariance of sylleptic shoot production with shoot length as a covariate and the three terminal branching orders as random factors. Source of variation MS F P Among intercepts < Covariate Among slopes for the main stem, although they did not consider the influence of light conditions on sylleptic shoot production. Hierarchical shoot growth according to branch order is an important factor in the development of a branched system, especially while the framework of a young branched system is developing. In this study, the branched systems were sampled from the upper parts of tree crowns, and therefore were relatively young. The different growth patterns in different orders of branches may be related to the different roles they play in the crown structure. In the tree, a bud forms part of a canopy within which there is a variety of branch orders (Harper 1980). Because the Order 2 axis is the main axis of a branched system, the elongation of Order 2 axes is related to the length of the branched system. Therefore, the terminal growth of the Order 2 axis is important for the expansion of a branched system and, consequently, for the total crown system. Shoot growth may be influenced not only by architectural position (branch order), but also by vertical position within the crown. Goulet et al. (2000) evaluated the effects of vertical position and light availability on shoot growth within the crowns of understory saplings of yellow birch (Betula alleghaniensis Britton.) and found that shoot growth was regulated by light availability, but this was true only for the terminal leader and upper branches. Although the PPFD received by lower branches was as much as 14% of full sunlight, these branches made little growth, and such growth as they made was unrelated to light availability. Therefore, Goulet et al. (2000) suggested that short-shoot formation in yellow birch saplings was partly a result of a low-light environment, but mainly a result of the position within the branch and the crown. Furthermore, Goulet et al. (2000) showed that the effect of position varied greatly among species. Shoot growth in the crown of sugar maple (Acer saccharum) saplings was regulated mainly by light availability and much less or not at all by position within the crown. They concluded that there was little apical control by the terminal leader over the lower branches in sugar maple. The adaptive significance of tree crown architecture has long been a topic of study (Horn 1971). Simulation models have been developed for quantitative and comparative analyses of the adaptive geometry of tree species. Honda (1971) showed that tree-like bodies can be described quantitatively by a set of simple rules of branching geometry. However, Honda (1971) did not consider the influence of the local light environment on branch growth. Takenaka (1994) presented a simulation model of crown architecture development, incorporating the dependence of number and size of new shoots on the photosynthetic production of parental shoots in the previous growing season. The results of the present study, showing variations in shoot growth response to local light environment in relation to branch order, will help to improve such theoretical models. In conclusion, shoot growth in C. japonica is influenced by the architectural position of the shoot, as well as by its light environment. This species preferentially allocates resources to the main axis of the branched system, even when the light environment is comparable among branch orders. To understand the processes underlying crown development, it will be necessary to determine both the relationships among shoots in terms of their architectural positions and to elucidate the responses of individual shoots to environmental conditions. Acknowledgments The author thanks Kihachiro Kikuzawa, Atsushi Takayanagi, Michimasa Yamasaki and other members of the Forest Biology Laboratory for helpful discussions. References Ackerly, D.D Light, leaf age, and leaf nitrogen concentration in a tropical vine. Oecologia 89: Ardhana, I.P.G., H. Takeda, M. Sakimoto and T. Tsutsumi The vertical foliage distribution of six understory tree species in a Chamaecyparis obtusa Endl. forest. Trees 2: Borchert, R. and N.A. Slade Bifurcation ratios and the adaptive geometry of trees. Bot. Gaz. 142: Brown, C.L., R.G. McAlpine and P.P. Kormanik Apical dominance and form in woody plants: a reappraisal. Am. J. Bot. 54: Canham, C.D Growth and canopy architecture of shade-tolerant trees: response to canopy gaps. Ecology 69: Cline, M.G Apical dominance. Bot. Rev. 57: Goulet, J., C. Messier and E. Nikinmaa Effect of branch position and light availability on shoot growth of understory sugar maple and yellow birch saplings. Can. J. Bot. 78: Harper, J.L Plant demography and ecological theory. Oikos 35: Hirai, S Ki no dai hyakka. Asakura shoten, Tokyo, 642 p. Honda, H Description of the form of trees by the parameters of the tree-like body: effects of the branching angle and the branch length on the shape of the tree-like body. J. Theor. Biol. 31: Horn, H.S The adaptive geometry of trees. Princeton University Press, Princeton, NJ, 144 p. Kozlowski, T.T. and R.C. Ward Shoot elongation characteristics of forest trees. For. Sci. 7: Maillette, L Structural dynamics of silver birch. I. The fates of buds. J. Appl. Ecol. 19: Remphrey, W.R. and C.G. Davidson Shoot and leaf growth in Fraxinus pennsylvanica and its relation to crown location and pruning. Can. J. For. Res. 24: Sabatier, S. and D. Barthélémy Growth dynamics and morphology of annual shoots, according to their architectural position, in young Cedrus atlantica (Endl.) Manetti ex Carrière (Pinaceae). Ann. Bot. 84: Sokal, R.R. and F.J. Rohlf Biometry. W.H. Freeman, New York, 887 p. TREE PHYSIOLOGY VOLUME 23, 2003

5 LIGHT AND POSITION EFFECTS ON SHOOT GROWTH 71 Steingraeber, D.A Phenotypic plasticity of branching patterns in sugar maple (Acer saccharum). Am. J. Bot. 69: Steingraeber, D.A., L.J. Kascht and D.H. Franck Variation of shoot morphology and bifurcation ratio in sugar maple (Acer saccharum) saplings. Am. J. Bot. 66: Suzuki, A Patterns of vegetative growth and reproduction in relation to branch orders: the plant as a spatially structured population. Trees 14: Suzuki, A Influence of shoot architectural position on shoot growth and branching patterns in Cleyera japonica. Tree Physiol. 22: Takenaka, A A simulation model of tree architecture development based on growth response to local light environment. J. Plant Res. 107: Takenaka, A Shoot growth responses to light microenvironment and correlative inhibition in tree seedlings under a forest canopy. Tree Physiol. 20: Tolvanen, A Aboveground growth habits of two Vaccinium species in relation to habitat. Can. J. Bot. 73: TREE PHYSIOLOGY ONLINE at