- Top of page
- Materials and Methods
In woody plants, above-ground branching patterns and their repetitive expression during growth directly affect light capture, water and nutrient transport, mechanical support, reproduction, and ultimately, competitive ability and fitness (Horn, 1971; Honda & Fisher, 1978; Küppers, 1989; Farnsworth & Niklas, 1995; Valladares, 1999). A single branch is made up of modules (i.e. shoots with leaves and buds) and is considered as a structural subunit responding independently to environments, although branches cannot be completely autonomous with respect to water and nutrients (Sprugel et al., 1991; Wilson, 2000). Therefore, scaling up from shoots to the whole canopy allows us to reduce the complexity of a tree crown to a much simpler observational unit (Valladares, 1999), and measurements of branching patterns (i.e. shoot architecture and shoot dynamics) may be useful surrogates for evaluating the strategies of whole trees. As branching patterns are strongly controlled by physiological (Pearcy & Valladares, 1999), biomechanical (Givnish, 1995) and environmental (Pickett & Kempf, 1980; King, 1990; Kohyama & Hotta, 1990) factors, as well as by genetic factors under phylogenetic constraints (Halle, 1978; Ackerly & Donoghue, 1998; Valladares et al., 2000), their evolution is the result of reconciling these different design requirements. Thus, there are numerous optimizing branching designs whose probabilities of survival depend on the particular environment inhabited by the plants (Küppers, 1989; Farnsworth & Niklas, 1995).
From the viewpoint of optimal resource acquisition, crucial linkages have been observed between branching design and successonal status or shade tolerance in broad-leaved tree species. Early successional and shade-intolerant tree species generally show a great degree of branching, more explosive and indeterminate growth patterns, and a multilayered canopy, presumably to take greater advantage of the increased light availability. In contrast, late successional and shade-tolerant trees show determinate shoot growth, shoot orientation to avoid self-shading, and densely packed canopies adapted to relatively predictable and shaded environments, such as closed forests (Horn, 1971; Pickett & Kempf, 1980; Shukla & Ramakrishnan, 1986; Küppers, 1989; Sakai, 1990; Ackerly & Donoghue, 1998; Valladares, 1999; Kikuzawa, 2003; but see Poorter & Werger, 1999). In most of these studies, however, optimal branching designs have been evaluated by the static analysis of shoots and branches, not by the dynamic analysis of shoot populations. The optimal design can be fundamentally constructed not only through the continuously or periodically repeated production of new shoots, but also through the mortality of old shoots (Harper, 1977; Wilson, 1987; Remphrey & Davidson, 1992; Umeki & Kikuzawa, 2000). Thus, the adaptive significance of shoot mortality should be evaluated in the context of optimal light-acquisition strategies.
Shoot mortality should be proximately facilitated by the mutual shading of neighboring shoots. Even though many woody species have branching patterns with long shoots at the distal end of each annual increment and short shoots at the proximal end (Wilson, 2000), differences in the branching patterns (i.e. shoot size hierarchy and the spatial distribution of shoots on parent shoots) will affect the extent of mutual shading (i.e. environmental light conditions of individual shoots) and the consequent probability of shoot survival. Branching patterns are closely related to the size, position and activity of winter buds on the parent shoots (Harmer, 1991). In woody plants therefore information on bud activity, subsequent shoot growth, and causes and consequences of shoot mortality should provide considerable insight into our understanding of branching patterns.
In plants, the longevity of resource-acquiring tissues (e.g. leaves and fine roots) is closely associated with environmental resource availability (e.g. light, nutrients and water) and the growth strategy or successional status of the species (Chabot & Hicks, 1982; Kikuzawa, 1983; Koike, 1988; Aerts et al., 1989; Reich et al., 1992; Espeleta & Donovan, 2002; Westoby et al., 2002). A shorter life span for these tissues is usually associated with early successional status or resource-rich environments, presumably because rapid tissue turnover keeps the tissue area in resource-rich environments, where resource gain can be maximized and because it maximizes the efficiency of resource deployment, as a result of nutrient recycling (Hikosaka et al., 1994; Ackerly, 1996; Espeleta & Donovan, 2002). In contrast, a much longer tissue life span is associated with late-successional, shaded, or resource-limited environments, and is the result of meeting the costs of construction and maintenance, and/or the result of compensating for limited carbon gain (e.g. net photosynthesis) by means of a longer tissue-retention time (Chabot & Hicks, 1982; Williams et al., 1989; Kikuzawa, 1991; Seiwa & Kikuzawa, 1996; Walters & Reich, 1999; Reich et al., 2003). As most leaves, which are essentially energy-gaining organs, are attached to current-year shoots, shoot life span may be linked to suites of traits, similar to those found in leaves.
Plants also exhibit a remarkable within-species and within-individual variability in their leaf habits and branching patterns, mostly as a result of plastic responses to local conditions. For example, within a species, leaf life span generally decreases with increasing resource availability (Chabot & Hicks, 1982; Aerts et al., 1989; Seiwa & Kikuzawa, 1996), and branching patterns (e.g. shoot length and shoot angle) also change according to the availability of light (Pickett & Kempf, 1980; Valladares, 1999). Thus, a major problem with comparative field studies of shoot (leaf) life span or branching patterns in habitats with different levels of resource availability (e.g. light) is that phenotypic responses and interspecific differences can easily confound the results. Such phenotypically plastic traits should therefore be compared between species exposed to the same environmental conditions because, owing to an inherited developmental program, most plants attain a characteristic shape when grown alone in the open (Horn, 1971; Halle, 1978; Valladares et al., 2000).
In this study, we investigated shoot survivorship (n = 3938) over 27 months in 15 deciduous broad-leaved tree species co-occurring in a temperate forest. To evaluate the influence of shoot architecture on shoot shedding, we investigated length, light conditions and positions of individual current-year shoots on 1-yr-old shoots. We addressed the following questions.
(1) Does shoot survivorship (i.e. life span) vary among species? If so, is this variation associated with successional status of the species or environmental resource availability of the habitats in which they are common?
(2) To what extent do shoot architecture (branching patterns) and architecture-induced light conditions affect shoot survival?
(3) Is shoot life span an important life history trait for evaluating the optimal strategy of resource acquisition in tree species?