SCOPE 56 - Global Change: Effects on Coniferous Forests and Grasslands

8.1 INTRODUCTION

The term 'allocation' refers to the ways in which plants distribute elemental and energy resources among the various plant functions including growth, reproduction, maintenance, storage, defense, and uptake of additional resources. The relative distribution of plant resources among these functions (i.e. the allocation pattern) varies greatly among species, and it is also responsive to environmental change and developmental changes within individual plants. This chapter provides a brief review of the variation in plant allocation patterns and their controls in terrestrial ecosystems, along with a discussion of the implications of variation in plant allocation for vegetation and ecosystem responses to climate change. 

From a plant perspective, allocation is important because it is closely related to the plant's long- and short-term survival, productivity, and reproductive output. In this view, controls on allocation are a major component of a plant's overall adaptive 'strategy' (e.g. Harper and Ogden 1970; Grime 1979; Tilman 1988). Plant allocation is thus viewed as an optimization process in which survival, productivity, or reproduction are maximized at some 'optimum' allocation pattern. However, the 'optimum' allocation pattern may differ depending on the environment, the time scale of interest, and the variable being maximized. For example, short-term productivity might be maximized when a plant allocates most of its resources into growth of organs, such as leaves and roots, that allow further uptake of resources and thus further growth. Long-term productivity, on the other hand, might be maximized by greater allocation to storage or to defense against herbivores, which would increase long-term plant survival and decrease losses of resource-acquiring organs like leaves. An entirely different pattern of allocation might be considered optimal if reproductive output were the variable to be maximized.

From an ecosystem perspective, plant allocation patterns have profound implications for overall element cycling and especially for the accumulation and storage of organic matter in soils (Field et al. 1992). Because plant allocation pattern influences total primary productivity, it also influences total litter inputs to the soil. Because allocation determines the relative amounts of leaves, stems, roots, and other plant parts in vegetation, it also determines the relative amounts of different litter types and the location where the litter is added to the soil. Finally, allocation controls the chemical composition of litter as well and thus has a major influence on its rate of decomposition.

Effects of plant allocation pattern on soil organic matter dynamics may have important feedbacks on future plant allocation patterns (Field et al. 1992). The feedbacks result from the effects of variation in litter quality on soil resource recycling and future resource availability to plants. If resource availability changes as a result of plant allocation patterns, then the 'optimum' plant allocation pattern to resource uptake must also change.

8.2 ALLOCATION WITHIN PLANTS

Most of our understanding of allocation within plants is phenomenological, not mechanistic. That is, the broad patterns of allocation in relation to environment have been described well enough to allow useful but rough predictions about the types of allocation patterns one should expect in a given environment and how those patterns might change in response to environmental change (e.g. Grime 1979). One particularly useful approach has been to apply the principles of cost-benefit analysis and other economic analogies to develop such predictions (Bloom et al. 1985; Chapin et al. 1990). However, the underlying physiological mechanisms that drive allocation are still largely unknown (Wilson 1988; Ingestad and Ågren 1991); many of these mechanisms appear to be hormonally mediated (Chapin 1991) but are rarely studied by ecologists. It is largely because our mechanistic understanding is lacking that predictions often differ depending on assumptions about the 'goal' of allocation.

8.2.1 Allocation among tissues

One of the oldest and still most commonly used assumptions about within-plant allocation is that its 'goal' is to help achieve a more optimal functional balance among the organs responsible for resource uptake, particularly roots and shoots (Brouwer 1966). For example, Davidson (1969) proposed that the allocation of photosynthate to production of new roots and shoots should be controlled so as to maintain a constant proportional relationship between whole-plant photosynthesis and root absorption, i.e.

Root mass x rate_(absorption) αLeaf mass x rate_{(photosynthesis)}

where the rates of absorption or photosynthesis are expressed per unit root or leaf mass and the constant of proportionality represents some 'optimal' nutritional balance. Thus, plants compensate for changes in carbon or nutrient availability by increasing allocation to either tissue mass or biochemical uptake machinery in the tissue responsible for the resource with lower availability.

Functional balance models like Davidson's (1969), and more detailed analyses based on economic theory (Bloom et al. 1985) are often useful in making predictions about and interpreting variation in root:shoot ratio among sites, species, and vegetation types. They are certainly more general than simple allometric models that do not allow for plant responses to variation in resource availability (Wilson 1988). However, functional balance models are not mechanistic and they typically do not consider allocation to defense, storage, or reproduction.

Mechanistic models of root-shoot allocation are most often based on assumptions about the supply, internal concentration gradients, and consumption of labile carbon (derived from leaf photosynthesis) and labile nitrogen (N) or 'nutrients' (derived from root uptake). Many of these models are based on an original formulation by Thornley (1972; reviewed by Wilson 1988). In these models, transport of labile carbon to roots and labile nutrients to shoots is a function of the concentration differences between roots and shoots. The concentration gradients are determined by the rate of photosynthesis in shoots or nutrient uptake by roots, followed by consumption of labile carbon and nutrients at both ends of the gradient due to growth of roots and shoots. The growth rates of these organs are determined by the available concentrations of both substrates. In effect, roots and shoots compete for labile carbon and nutrient substrates to support growth, with roots having the advantage with respect to nutrient supply and shoots with respect to carbon supply. The result is that root growth is favored over shoot growth when nutrient supply is low, and shoot growth is favored over root growth when carbon supply is low.

Thornley-type models of root-shoot allocation often do a good job at predicting changes in allocation in response to a wide range of environmental conditions, including changes in light, water, CO₂ concentration, temperature, and pruning or defoliation (Wilson 1988). However, Ingestad and Ågren (1991; Ågren and Ingestad 1987) have recently argued that root-shoot allocation patterns are in fact less sensitive to external factors affecting carbon supply than predicted by Thornley-type models (e.g. Johnson and Thornley 1987) and by 'optimal allocation' theory (e.g. Bloom et al. 1985). At least in part, the differences between these models may be explained by their differing assumptions about external controls on nutrient supply (Rastetter and Shaver 1992). Thornley and others usually assume that nutrients are available at constant concentrations that are unaffected by root uptake, and thus, additional nutrients can always be taken up if allocation to root growth is increased. Ingestad and Ågren, on the other hand, assume that nutrient supply rate and uptake rate are equal, and thus additional allocation to roots has no effect on uptake unless the supply rate is increased. Reconciliation of these two approaches depends on testing them under conditions of both constant external nutrient concentrations and constant nutrient supply rates (Rastetter and Shaver 1992).

8.2.2 Structural and maintenance costs

Plant organs vary widely in their biochemical composition, i.e. in the relative abundance of lignin, cellulose, tannin, protein, sugar, starch, nucleic acids, and other essential molecules (Mooney 1972; Penning de Vries et al. 1974; Chapin 1989). This biochemical composition is dependent not only on the organ's principal function (e.g. photosynthesis, support, nutrient uptake) but also on other factors such as its expected longevity, exposure to environmental extremes, and exposure to herbivory. Because of this wide variation in composition among tissues, and because the costs of synthesis of specific molecules vary widely, one might expect the overall resource requirements for allocation to growth to be similarly variable. In other words, the same amount of plant resources allocated to growth might end up producing very different amounts of growth.

In fact, the overall carbon and energy costs of organ growth are fairly similar both within and among tissues, so that organ mass alone is a reasonable indicator of these allocation costs (Abrahamson and Caswell 1982). The reason for this similarity in carbon and energy costs is that almost all plant parts have fairly high concentrations of some energetically expensive compounds like lignin, protein, or tannin, even though the relative abundances of these compounds is variable (Chapin 1989; Ryan 1991b).

Nitrogen and phosphorus costs of growth, on the other hand, are much more variable among plant species and organs, and organ mass alone is not a good indicator of these costs. Variation in N and P costs is in general related to variation in N and P concentration, with leaves typically being more costly than other tissues. Leaves of deciduous tree and shrub species and forbs have higher nutrient costs that those of graminoids or (especially) evergreens (Chapin 1989; Reich et al.1992).

Once a tissue or organ is produced, at least some plant resources must be allocated to its maintenance (Penning de Vries 1975; Ryan 1991a). Maintenance allocation is related to the overall metabolic activity of an organ and especially to the need for replacement and resynthesis of its protein or N content. The respiratory costs of maintenance are also strongly affected by temperature, in contrast to respiration associated with growth which is more related to the specific biochemical products being synthesized than to temperature. The temperature dependence of allocation to maintenance respiration is particularly important in that it is a major component of any plant's carbon budget that is not subject to control by the plant.

Figure 8.1 Allocation of resources to the three major classes of storage, and to defense, growth, and maintenance involves a number of interrelated resources pools and fluxes (from Chapin et al. 1990). Accumulation occurs when resources acquisition exceeds resources utilization in reserve formation, defense, and growth and maintenance (i.e when flux 1 > flux 2). Reserve formation (flux 5), on the other hand, is more strongly metabolically regulated and competes directly with defense and growth (fluxes 3 and 4) for resources. Accumulation, reserves and defense (flux 8) may be used in future growth (fluxes 6 and 7), as can recycled materials (flux 9)

8.2.3 Allocation to storage and defense

Storage and defense against herbivory are often major components of a plant's overall allocation pattern (Figure 8.1). Storage includes a wide range of chemical compounds such as sugars, starches and other polysaccharides, some lipids, proteins or free amino acids, and both soluble organic and inorganic forms of p (Chapin et al. 1990). Many of these compounds also perform other, nonstorage functions as well, making it difficult to estimate the total allocation to storage. However, at times (especially during dormant periods) these storage compounds may account for more than half of the mass of perennial, nonwoody tissues in many plants, and even in woody tissues storage compounds may form the major portion of the mobile, high-turnover chemical fractions.

Chapin et al. (1990) defined three major classes of storage: accumulation, reserve formation, and recycling (Figure 8.1). Accumulation is simply storage that results from excess resource supply relative to immediate demands for growth or maintenance. For example, accumulation of one resource commonly occurs when its availability relative to other resources increases, and the plant cannot instantaneously reallocate its uptake effort to maintain a perfectly balanced nutrition. This form of storage may or may not be used for future growth and maintenance; if it is not it may be lost in litterfall (with consequent effects on litterfall chemistry and decomposition). Reserve formation, on the other hand, is the metabolically regulated production of storage products that might otherwise be used in growth or other plant functions. An example is the autumnal buildup of starch in roots and other perennial tissues of many plants, presumably to meet winter maintenance needs and to support growth the following spring. Recycling involves the breakdown of compounds already functioning to support growth, defense, or other plant processes, and storage of the breakdown products for reuse at a later time. Because most of the materials recycled come from senescent tissues, recycling is a major control over litter chemistry.

The broad patterns of storage and its variation among plants and ecosystems have been reasonably well-described although again the mechanisms controlling storage are poorly understood (Chapin et al. 1990). Accumulation is most common in plants with conservative growth patterns that are relatively unresponsive to short-term environmental fluctuation; these are often slow-growing plants found in more stressful environments. Recycling may be most important in species from infertile environments although within-species variation in recycling across fertility gradients does not show a consistent pattern. Recycling is generally more important for nutrients such as N or p than for carbon. Reserve formation is difficult to distinguish from accumulation or recycling although it clearly is important in many plants with long dormant periods or in plants with a predictable asynchrony in the timing of resource uptake and demand for resources in growth. Time scale is also important, with accumulation generally more important on daily or weekly periods, and storage or recycling more important on annual cycles (Chapin et al. 1990).

Defense includes structures such as thorns or spines, but more importantly it includes the wide array of 'secondary' plant metabolites such as tannins, alkaloids, phenolics, many resins, and more esoteric compounds that serve to reduce or inhibit herbivory. These defense compounds may account for 20-25% ( or more) of the mass of many plant organs, such as leaves and young stems, that would otherwise be especially vulnerable to herbivory (Chapin 1989). Current theories of control over allocation to defense emphasize the importance of carbon and nutrient resource availability to plants and the relative degree of limitation by different resources (Bryant et al. 1983; Coley et al. 1985; Bazzaz et al. 1987). Allocation to defense is also influenced by a host of additional factors, most importantly including the coevolutionary history of specific plant-herbivore interactions (Ehrlich and Raven 1964; Feeney 1975; Rhoads and Cates 1976). The theory is continually evolving (Stamp 1992). However, from the ecosystem perspective of this book, a resource-based approach (Bazzaz et al.1987; Field et al. 1992) to allocation to defense probably has the greatest utility.

In the resource-based approach, plants growing in low-resource environments are expected to allocate more to defense than plants in resource-rich environments because (1) carbon and nutrient resources lost to herbivory are more costly to replace than they are to defend in low-resource environments, and (2) plants growing in low-resource environments tend to grow slowly and to produce tissues that are relatively costly to build and that have slow turnover rates, making allocation to their defense even more advantageous relative to replacement (Coley et al. 1985). Plants from high-resource environments, on the other hand, allocate less to defense and more to resource acquisition, grow rapidly, and are much more responsive to variation in resource availability.

Not only the amount but also the chemical form of defense is expected to vary with resource availability, with carbon-based defenses (e.g. tannins) being favored when nutrients are strongly limiting and N-based defenses (e.g. alkaloids) when carbon or light are strongly limiting (Bryant et al. 1983; Coley et al. 1985). Both the amount allocated to defense and its chemical form may change over a plant's life history. For example, juvenile trees are often more heavily defended than adult plants, probably because juveniles are at greater risk to herbivory than adults (Bryant et al. 1983).

8.2.4 Allocation to reproduction 

The literature on allocation to reproduction is particularly vast and varied (Harper 1977; Willson 1983; Bazzaz and Reekie 1985; Bazzaz et al. 1987). From a long-term, population/evolutionary perspective, reproductive allocation and its controls are clearly key questions. From an ecosystem/biogeochemical perspective, however, allocation to reproduction where the vegetation is dominated by long lived, perennial plants usually amounts to only a small portion of annual carbon and nutrient budgets (the major exceptions being vegetation such as bamboo forests, with occasiona, extreme mast-seeding). In many cases reproductive allocation is driven by storage from previous years (Chapin et al. 1990), thus effectively buffering its impact on allocation of currently acquired resources. For lack of space, we will not consider reproductive allocation further in this chapter.

8.3 ECOSYSTEM LEVEL EFFECTS

8.3.1 The need for a general theory

To develop a truly global, physiologically based model of terrestrial ecosystem response to global change, we must search for general patterns which emerge from the physiological literature, including the literature of allocation. Allocation presents a particular problem in this regard in that the basic patterns of allocation in relation to environment have been described, but the physiological mechanisms controlling allocation are generally unknown.

In nearly all of the models presented in this book, vegetation is described not in terms of species composition, but rather in terms of physiological responses to climate drivers. This is in part because there are no global data bases of species composition to provide data at this level of detail. Models applied globally will have to generalize to the level of detail available as input, including generalizations about allocation. In many cases, this is only to the level of biome or physiognomic group. To match this coarse level of resolution on descriptors of vegetation, we must generalize input parameters of vegetative response to the same level. Either physiognomic groups (e.g. broad-leaved deciduous forests versus annual herbaceous grasslands) or physiological groups (e.g. C₃ versus C₄ plants, symbiotic N fixers versus nonfixers) may be useful. Fortunately most of the major generalizations about allocation are at this level.

Such generalizations seem crude at the physiological level, given the depth of understanding of critical processes at the cellular and biochemical level that has emerged over the last two decades. It will also result in less accurate predictions of ecosystem dynamics at any one location than could be obtained using site- and species-specific data for anyone area. However, given that only a very small fraction of the earth will ever be intensively studied, the intensity of model input demands must be relaxed to a level that is realizable for the entire globe.

8.3.2 An ecosystem-level theory of plant allocation

8.3.2.1 Above ground patterns

The last 10 years have seen tremendous progress in the development of precisely the kind of generalized theory of allocation required for global-scale modeling. Since 1972, Mooney and colleagues have published a series of papers linking several previously distinct physiological, morphological and life-history characteristics into a unified theory based on the fitness of plants in sites of different resource 'richness' (e.g. Mooney 1972; Mooney and Gulmon 1982; Coley et al. 1985; Bloom et al.1985; Field and Mooney 1986; Bazzaz et al. 1987; Chapin et al. 1990; Chapin 1991; Field et al. 1992). This theory can be represented as a circular set of interactions resulting in positive feedbacks between site quality, plant response, and future site quality (Figure 8.2). Using the example of N as the growth limiting resource, this theory can be summarized as follows (see citations above for references). Sites with higher N availability will support foliage with higher N concentrations. There is a positive relationship between the N content of foliage and its maximum rate of photosynthesis. A higher N content in foliage therefore means that less time will be required for a unit of leaf mass to 'repay' the cost of production of that leaf (replacing the carbon that was required for structure and growth respiration). This shorter repay period, along with the high cost of producing antiherbivore compounds, creates selective pressures for low defense allocation and a shorter foliar retention time. All of the interactions above combine to produce high inputs of relatively rich litter materials which decay rapidly and reinforce high N cycling, or a 'richer' site.

Figure 8.2 A generalized theory of interactions between resource availability (nitrogen in this case), maximum rates of photosynthesis, levels of investment in herbivore defense, and foliar longevity (see text for description and references)

The power of this theory is that it unifies the processes of photosynthesis (and transpiration, see Field and Mooney 1986; Aber and Federer 1993), allocation, foliar longevity, herbivory and element cycling in a single interactive whole (Field et al. 1992). It is also well supported by a wealth of field data.

Recently, Reich, Gower and colleagues have elaborated on this theory and provided a rich and broad, cross-biome data set in support. Building on within-biome patterns shown by Reich et al. (1991) and Gower et al. (1993), Reich et al. (1992) show the following set of patterns (R² values are for log-log plots): Maximum net photosynthesis is strongly correlated with N concentration (mg g^-1, R² = 0.74) but not with N per unit leaf area. Leaf N concentration is also correlated with leaf lifespan (R² = 0.52), as is maximum leaf conductance (R² = 0.59). Other morphological and life-history interactions include strong relationships between leaf N concentration and leaf specific area (cm² g^-1, R² = 0.54) and also between leaf specific area and maximum photosynthetic rate. Not surprisingly, these are all then related in turn to relative growth rate and height growth.

Table 8.1 Characteristics of plants with long and short leaf lifespans (adapted from Reich et al. 1992)

At the whole stand level, additional relationships emerge between leaf lifespan and total stand foliar biomass (R² = 0.74) and also aboveground net primary production normalized for leaf mass (inverse relationship, R² = 0.78), while there is no relationship between leaf lifespans and total above ground net primary production. Finally, there are significant relationships between leaf lifespan and both litterfall N use efficiency (l/[N in litterfall], R² = 0.27) and total N retranslocated from foliage during senescence (inverse, R² = 0.57), but not with the fraction of N retranslocated. The differences between N-poor and N-rich end points can be described, according to these results, as in Table 8.1.

It is important to realize that none of these relationships would necessarily be expected to hold within a single community. At that level, interspecific and intergeneric variation might be expected to swamp the very broad and general patterns shown here. Natural selection and evolution cannot be expected to have the precision (or monotony) of human engineered systems, and there are other selective pressures besides resource acquisition and allocation which drive the evolutionary process.

Still, taken together, these relationships describe a global pattern of carbon and N allocation, and carbon assimilation, that reinforces the modelling utility of the theories of Mooney, Field, Chapin et al. with some important system-level additions. First, the higher leaf specific area in N-poor foliage, along with the higher total foliar mass in N-poor systems, suggest that these result from the dilution of a relatively fixed amount of canopy N by different amounts of carbon, in order to provide protection against herbivory and as a result of differences in allocation to structural components. The notion that total canopy N should be held fairly constant across sites of differing N availability is an intriguing one, and worth further measurement. This would be consistent with optimization of canopies for full utilization of incoming radiation, with the amount of foliage required to accomplish this varying with the amount of carbon dilution required for foliar protection, in turn a function of site richness, foliar N concentration and leaf lifespan.

8.3.2.2 Below ground patterns

Far less is known about plant allocation and metabolism below ground. Studies of fine root dynamics are far fewer than for foliage, and methods are less precise and more subject to interpretation. The greatest wealth of information available on fine root biomass under different types of vegetation and site conditions comes from direct measurements of root mass in soil cores. This is a tremendously tedious and time-consuming task, which explains, in part, the lack of data. Estimating production and turnover of root biomass from sequential cores has proven controversial (Vogt et al. 1986; Lauenroth et al. 1986; McClaugherty et al. 1982), and in conflict with estimates derived from system-level measurements of N and carbon cycling (Raich and Nadelhoffer 1989; Nadelhoffer and Raich 1992). Two conflicting theories regarding relationships between foliar and root production have arisen from the existing data. Both agree that fine root biomass tends to increase with declining site quality. Although developed mainly for forest systems, this would extend to semiarid and arid systems as well, where the depth of root penetration is related to exploitation of deep soil horizons for water. The controversy arises over the rates of turnover of roots. The first theory suggests that root turnover is fairly constant and thus fine root production is higher in relation to foliar production on poor sites ( e.g. Keyes and Grier 1981 ). The second theory suggests that fine root turnover is higher on richer sites and, thus, fine root production is relatively constant in proportion to foliar production across sites of different qualities (e.g. Aber et al. 1985).

In a recent paper, Hendricks et al. (1993) have reviewed the current literature and presented arguments in favor of the second theory, that fine root production is relatively constant in relation to foliar production. A key data set in this area is that of Raich and Nadelhoffer (1989), who did not use any direct fine root data at all! Instead, they summarized all the available studies in which aboveground carbon input to soils and total annual soil carbon efflux (as CO₂ ) were measured. They assumed, if soils were near steady-state for total carbon storage, that the difference between these two was due either to root respiration or the decomposition of root litter, the sum of which is total plant carbon allocation below ground. What they found was a very significant linear relationship between the two, suggesting relatively constant allocation and higher turnover rates on richer sites. A similar analysis of N cycling limitations on fine root production yielded similar conclusions (Nadelhoffer and Raich 1992; Hendricks et al. 1993).

Hendricks et al. (1993) have incorporated these results into the general theory in Figure 8.2 by extending it to the below ground part of the system (Figure 8.3). Here, analogous positive feedback loops operate in both halves of the system, again reinforcing the rich or poor status of a site. It should be stressed that this second half of the theory has not yet been supported by extensive field data. Studies are currently underway to test the patterns proposed here.

Figure 8.3 Extension of the generalized theory in Figure 8.1 to belowground tissues (after Hendricks et al. 1993)

8.3.2.3 The cost of producing and maintaining biomass

Accurate predictions of carbon allocation in ecosystems must include losses due to respiration. While the carbon cost of growth respiration is generally agreed to be about 25% of the mass of carbon incorporated into tissues (Penning de Vries et al. 1974; Ryan 1991a), maintenance respiration can vary widely depending on the amount of live biomass maintained, the level of physiological activity in that biomass, and the temperature regime of the site. System-level measurements of maintenance respiration are even fewer than for fine root production, and generalizations have only recently begun to emerge.

Ryan (e.g. 1991 a, b) has pioneered the development of a generalized method for estimating rates of maintenance respiration at several levels of organization. At the tissue level, he summarizes previous studies on the relationship between N content and respiration rates and presents a new data set covering tissues between 0.04 and 6% N by mass and showing a highly significant relationship between the two. He then develops a methodology for applying this to whole systems using stem biomass (heartwood and sapwood), N content, seasonal changes in temperature and a Q_l0 factor (Ryan 1991a). These methods are incorporated into a full carbon balance algorithm which is validated against whole ecosystem carbon balances as predicted by an ecosystem model (Ryan 1991 b). An additional step to an even coarser resolution is the development of a relationship between mean annual temperature and the fraction of total gross photosynthesis expended in maintenance respiration for whole forest ecosystems. Preliminary estimates (M. Ryan, pers. comm.) suggest that maintenance costs vary between 7 and 15% of gross photosynthesis as mean annual temperature increases from 5 to 21°C. At this level, the question is whether there is a higher order set of plant-level carbon allocation optimization 'rules' that constrain maintenance respiration to these levels, or if this is just a chance correlation resulting from similarity in structure and function between the forests studied.

8.3.2.4 Modelling whole ecosystem carbon budgets

The generalized theories of carbon gain and allocation described here offer, to the extent that they are validated by future research, tremendous power for modelling vegetation responses within global-scale models of terrestrial ecosystems. As these relationships cross biome boundaries, they overcome the limitations that occur in applying biome-specific parameters at or near ecotones that might be altered by climate change. They would predict smooth transitions between end-states in response to chronic atmospheric deposition. They could offer a very interesting set of predictions on the expected distribution of major physiognomic groups (e.g. grasslands, deciduous forests, evergreen forests) which could be validated using existing vegetation distribution data.

However, it should be remembered that this combination of algorithms is not likely to produce an accurate carbon balance at any particular site. The significance of the relationships described here results entirely from the wide range of conditions over which the measurements were accumulated. At anyone site, greater precision could easily be achieved using more detailed models requiring more specific input data and taking into account that site's recent disturbance and vegetation history. The tradeoff between generality and accuracy is very clear in this case.

8.3.3 Conclusions

The major implication of the set of theories presented here is that the interaction between vegetation and site quality should be that of a positive feedback (accentuating either the richness or poorness of a site) rather than a negative feedback. The other factors which eventually limit the extent to which this positive feedback operates (e.g. limitations by other factors [water, radiation, climate] on rich sites, and system 'resetting' by fire on poor sites) should be better understood. The direct interactions with processes such as decomposition are explored in other chapters in this volume.

8.4 REFERENCES

Aber, J. D. and Federer, C. A. (1993) A generalized, lumped-parameter model of photosynthesis, evapotranspiration and net primary production in temperate and boreal forest ecosystems. Oecologia 92, 463-474.

Aber, J. D., Melillo, J. M., Nadelhoffer, K. J., McClaugherty, C. A. and Pastor, J. (1985) Fine root turnover in forest ecosystems in relation to quantity and form of nitrogen availability: a comparison of two methods. Oecologia (Berlin) 66, 317-321.

Abrahamson, W. G. and Caswell, H. (1982) On the comparative allocation of biomass, energy, and nutrients in plants. Ecology 63, 982-991.

Ågren, G. and Ingestad, T. (1987) Root/shoot ratios as a balance between nitrogen productivity and photosynthesis. Plant Cell Environ. 10, 579-586.

Bazzaz, F. A., Chiariello, N. R., Coley, P. D. and Pitelka, L. F.(1987) Allocating resources to reproduction and defense. BioScience 37,58-67.

Bazzaz, F. A. and Reekie, E. G. (1985) The meaning and measurement of reproductive effort in plants. In: White, J. (Ed.) Studies in Plant Demography: A Festschrift for John L. Harper, pp. 259-275. Academic Press, London.

Bloom, A. J., Chapin III, F. S. and Mooney, H. A.(1985) Resource limitation in plants-an economic analogy. Ann. Rev. Ecol. Syst. 16, 363-392.

Brouwer, R. (1966) Root growth of grasses and cereals. In: Milthorpe, F. L. and Ivins, D. J. (Eds) The Growth of Cereals and Grasses, pp. 153-166. Butterworths, London. 358 pp.

Bryant, J. P., Chapin F. S.III, and Klein, D. R. (1983) Carbon/nutrient balance of boreal plants in relation to vertebrate herbivory. Oikos 40, 357-368.

Chapin F. S. III (1989) The costs of tundra plant structures: evaluation of concepts and currencies. Am. Nat. 133, 1-19.

Chapin F. S. III (1991) Integrated response of plants to stress. BioScience 41, 29-36.

Chapin F. S.III, Schulze, E.-D. and Mooney, H. A. (1990) The ecology and economics of storage in plants. Annu. Rev. Ecol. Syst. 21, 423-427.

Coley, P. D., Bryant, J. P. and Chapin III, F. S. (1985) Resource availability and plant antiherbivore defense. Science 230, 895-899.

Davidson, R. L. (1969) Effect of root/leaf temperature differentials on root/shoot ratios in some pasture grasses and clover. Ann. Bot. 33, 561-569.

Ehrlich, P. R. and Raven, P. H. (1964) Butterflies and plants: a study in coevolution. Evolution 18, 586-608.

Feeney, P. (1975) Biochemical coevolution between plants and their insect herbivores. In: Gilbert, L.E. and Raven, P.H. (Eds) Coevolution of Animals and Plants, pp. 3-19. University of Texas, Austin.

Field, C. B., Chapin F. S. III, Matson, P. A. and Mooney, H. A. (1992) Responses of terrestrial ecosystems to the changing atmosphere. Ann. Rev. Ecol. Syst. 23, 201-235.

Field, C. J. and Mooney, H. A. (1986) The photosynthesis-nitrogen relationship in wild plants. In: Givinish, T. J. (Ed.) On the Economy of Plant Form and Function, pp. 25-55. Cambridge University Press, New York.

Gower, S. T., Reich, P. B. and Son, Y. (1993) Canopy dynamics and aboveground production for five tree species with different leaf longevities. Tree Physiol. (in press).

Grime, J. P. (1979) Plant Strategies and Vegetation Processes. John Wiley & Sons, New York. 222 pp.

Harper, J. L. (1977) Population Biology of Plants. Academic Press, London. 892 pp.

Harper, J. L. and Ogden, J. (1970) The reproductive strategy of higher plants: 1. The concept of strategy with special reference to Senecio vulgaris. J. Ecol. 58, 681-698.

Hendricks, J. J., Nadelhoffer, K. J. and Aber, J. D. (1993) The role of fine roots in energy and nutrient cycling. Trends Ecol. Evol. 8, 174-178.

Ingestad, T. and Ågren, G. (1991) The influence of plant nutrition on biomass allocation. Ecol. Appl. 1, 168-174.

Johnson, I. R. and Thornley, J. H. M. (1987) A model of root:shoot partitioning with optimal growth. Ann. Bot. 60, 133-142.

Keyes, M. R. and Grier, C. C. (1981) Above- and below-ground net production in 40 year old Douglas-fir stands on low and higher productivity sites. Can. J. For. Res. 11, 599-605.

Lauenroth, W. K., Hunt, H. W., Swift, D. M. and Singh, J. S. (1986) Reply to Vogt et al. Ecology 67, 580-582.

McClaugherty, C. A., Aber, J. D. and Melillo, J. M. (1982) The role of fine roots in the organic matter and nitrogen budgets of two forested ecosystems. Ecology 63, 1481-1490.

Mooney, H. A. (1972) The carbon balance of plants. Ann. Rev. Ecol. Syst. 3, 315-346.

Mooney, H. A. and Gulmon, S. L. (1982) Constraints on leaf structure and function in relation to herbivory. BioScience 32, 198-206.

Nadelhoffer, K. J. and Raich, J. W. (1992) Fine root production estimates and below-ground carbon allocation in forest ecosystems. Ecology 73,1139-1147.

Penning de Vries, F. W. T. (1975) The cost of maintenance processes in plant cells. Ann. Bot. 39, 77-92.

Penning de Vries, F. W. T., Brunsting, A. H. M. and van Laar, H. M. (1974) Products, requirements and efficiency of biosynthesis: a quantitative approach. J. Theor. Biol. 45, 358-377.

Raich, J. W. and Nadelhoffer, K. J. (1989) Belowground carbon allocation in forest ecosystems: global trends. Ecology 70, 1346-1354.

Rastetter, E. B. and Shaver, G. R. (1992) A model of multiple element limitation for an acclimating plant. Ecology 73,1157-1174.

Reich, P. B., Uhl, C., Walters, M. B. and Ellsworth, D. S. (1991) Leaf lifespan as a determinant of leaf structure and function among 23 Amazonian tree species. Oecologia 86, 16-24.

Reich, P. B., Walters, M. B. and Ellsworth, D. S. (1992) Leaf lifespan in relation to leaf, plant and stand characteristics among diverse ecosystems. Ecol. Monogr. 62, 365-392.

Rhoads, D. F. and Cates, R. G. (1976) Toward a general theory of plant antiherbivore chemistry. Rec. Adv. Phytochem. 19, 168-213.

Ryan, M. G. (1991a) Effects of climate change on plant respiration. Ecol. Appl.1, 157-167.

Ryan, M. G. (1991b) A simple method for estimating gross carbon budgets for vegetation in forest ecosystems. Tree Physiol. 9, 255-266.

Stamp, N.E. (1992) The theory of plant-insect herbivore interactions on the inevitable brink of re-synthesis. Bull. Ecol. Soc. Am. 73, 28-34.

Thornley, J. H. M. (1972) A balanced quantitative model for root:shoot ratios in vegetative plants. Ann. Bot. 36, 431-441. 

Tilman, D. (1988) Plant Strategies and the Dynamics and Structure of Plant Communities. Monographs in Population Biology 26. Princeton University Press, Princeton, NJ. 360 pp.

Vogt, K. A., Grier, C. C., Gower, S. T., Sprugel, D. G. and Vogt, Do J.(1986) Overestimation of net root production: a real or imaginary problem? Ecology 67, 577-579.

Willson, M. F. (1983) Plant Reproductive Ecology. John Wiley & Sons, New York.

Wilson, J. B. (1988) A review of evidence on the control of shoot:root ratio, in relation to models. Ann. Bot. 61, 433-449.