New Phytol. (1999), 142, 373-417

Tansley Review No. 104Calcium physiology and terrestrialecosystem processes

S. B . McLAUGHLINl* A N D R . WIMMER2

‘Environmental Sciences Division, Oak Ridge, National Laboratory, Oak Ridge, TN,U S A

‘) 2 Austrian Agricultural University, Vienna, Austria

Received 6 July 1998 ; accepted 8 March 1999

C O N T E N T S

SummaryI. INTRODUCTION - A HYPOTHESIS

11. EFFECTS OF CALClUhl ON PHYSIOLOGICAL

PROCESSES

1. The chemical uniqueness of calcium( a ) Cytotoxicit)(b) Binding properties(c) Stimulation/displacement

potential2. Cal&n signaling and plant responses

to environmental stress(a) Control principles(b) Carbohydrate metabolism(c) Synthesis and function of

membranes and cell walls(d) Disease resistance and wound repair(e) Cold tolerance(f) Stomata1 regulation

111. CALCIURI UPTAKE AND DISTRIBUTION AT THE

WHOLE-PLANT LEVEL

1. Uptake at the root-soil interface2. Transport and exchange in stems3. Exchange of calcium by foliage

Ii’:, tiCOSYSTE&l PROCESSES AND CALClUhl SUPPLY

1. Plant succession and soil acidification

374

376376376376

376

377377378

379380381382

382382384385387387

2. Plant adaptations to nutrient deficiency(a) Morphological adaptations(b) Physiological adaptations

v. PLANT AND ECOSYSTEM RESPONSES TO HUbIAN

ALTER.+TlONS I N CALClU:Xl SUPPLY

1. Increased atmospheric inputs of acidity(a) Reductions in soil cation pools(b) Inhibition of calcium uptake and

effects on root function(c) Increased leaching of calcium from

foliage(d) Physiological indicators of altered

forest function(e) Wood chemistry, structure and

function2. Forest management

(a) Harvesting effects on nutrient supply(b) Managing forest nutrient supply

VI. CONCLUSION

1. Whole-tree perspectives2. Ecosystem perspectives

VII. EVALUATION OF THE HYPOTHESIS

Acknowledgements,References

Calcium occupies a unique position among plant nutrients both chemically and functionally. Its chemicalproperties allow it to exist in a wide range of binding states and to serve in both structural and messenger roles.Despite its &portance in many plant processes, Ca mobility is low, making Ca uptake and distribution rate alimiting process for many key plant functions. Ca plays an essential role in regulating many physiological processesthat influence both growth and responses to environmental stre’sses. Included among these are: water and solutemovement, influenced through effects on membrane structure and stomata1 function; cell dikision and cell wallsynthesis; direct or signaling roles in systems involved in plant defense and repair of damage from biotic andabiotic stress; rates of respiratory metabolism and translocation; and structural chemistry and function of woodysupport tissues. Forest trees, because of their size and age capacity, have been examined for evidence of limitationsimposed by the timing and level of Ca supply. Examination of Ca physiology and biogeochemical cycling forforested systems reveals many indications that Ca supply places important limitations on forest structure and

389389390

391391392

393

395

397

402404404405407407408410411411

*Author for correspondence (fax + 1 423 576 9939; e-mail urv@ornl.gov)

,

374 B. McLaughlin and R. Wimmer

function. These limitations are likely to be most significant with older trees, later successional stages, high levelsof soil acidity and/or high canopy Ca leaching losses, or under conditions where plant competition is high ortranspiration is limited by high humidity or low soil moisture. Evidence of structural and physiological adaptationsof forests to limited Ca supply; indicators of system dysfunction at many levels under reduced Ca supply; and thepositive responses of diverse indicators of forest vitality in liming experiments indicate that Ca is more importantto forest function and structure than has generally been recognized. Lack of recognition of Ca limitations is duein part to that fact some important plant functions are controlled by changes in very small physiologically activepools within the cytoplasm, and whole-leaf Ca levels may not reflect these limitations. An additional aspect is thefact that Ca availability has declined significantly for many forests in just the past fen decades. Additional researchon the role of Ca supply in resistance of forests to disease, changes in structural integrity of woody tissues,restrictions on rooting patterns and function-and uptake of other nutrients, notably X;, is needed. Increasedunderstanding of the physiological ecology of Ca supply can be anticipated to protide important insights that willaid in future protection and management of both natural and commercial forest systems.

Key words: calcium, forests, ecosystem, physiology, structure, function.

I. INTRODUCTION - A HYPOTHESIS

The diverse functions of Ca in regulating physio-logical and structural processes in both animals andplants are now widely recognized after decades ofresearch. Ca is unique among the elements com-prising living systems because of its role as amessenger for many different types of physiologicalprocesses and its unique chemical binding properties(Hepler & Wayne, 1985). For these reasons Ca playsan important role in adding stability and structuralintegrity to biological tissues at scales ranging fromintercellular membranes to the cell walls of woodystems. Reviews over the past 15 years have docu-mented steady progress in the definition of anincreasingly broad array of responses and mech-anisms by which Ca functions to influence plantprocesses (Bnngerth, 1979; Clarkson & Hanson,1980; Ferguson, 1984; Kirkby & Pilbeam, 1984;Rorison & Robinson, 1984; Hepler 8L Wayne, 1985;Kauss, 1987; Pooviah, 1988; Roberts & Harmon?1992; Bush, 1995; McAinsh et al., 1996; Webb etal. , 1996; Trewavas & Rlalho, 1997; McAinsh &Hetherington, 1998).

The number of processes in which Ca plays animportant regulatory role is impressive, includingphosphorylation of nuclear proteins; cell division;cell ~wall and membrane synthesis and function;intra- and intercellular signaling; protein synthesis;responses to environmental stimuli, including lowtemperatures, gravity, insects and disease; stomata1regulation; and carbohydrate metabolism. The di-versity of these processes is even more remarkablewith respect to higher plant function when it isrecognized”that, among the primary plant macro-nutrients, Ca is the least mobile, making the issue ofCa supply and conser\‘ation critical to growth anddevelopment in resource-limited environments.

The needito place Ca metabolism in the broaderperspective of factors that determine its availabilityin the external environment (Bangerth, 1979) andaffect its distribution within the plant (Kirkby &Pilbeam, 1984) has become increasingly important inrecent years. There are two reasons for this: first, a

wealth of recent research on Ca metabolism andsignaling in plants has made additional understand-ing of the nature and potential connectivity of Ca-regulated processes at the whole-plant levelincreasingly possible; and second, because there areincreasing indications that the activities of humansin recent decades have reduced the availability of Cain terrestrial ecosystems.

Environmental influences that produce significantspatial and temporal gradients in Ca supply are ofinterest from the perspectives of both plant speciesand plant community development. These influencesinclude changes in regional scale atmospheric de-position of strong acids, which have long beenrecognized as a factor that increases Ca loss fromforests (Overrein, 1972); large-scale reductions inatmospheric deposition of Ca to terrestrial eco-sys tems (Hedin et al. , 1994); and whole-treeharvesting, which can remove important stores of Cain foliage and branches from both temperate(Johnson et al., 1988; Federer et al., 1989) andtropical forests (Jordan 6; Herrera, 1981). Bio-geochemical responses of interest include changes inpatterns of Ca uptake, retention, and cycling atorganizational scales ranging from cell membranes(DeHayes e t a l . , 1997) to forested watersheds(Johnson & Lindberg, 1992; Federer et al:, 1989;Ulrich & Matzner, 1986).

Studies of nutrient cycling in terrestrial eco-sys tems a t scales ranging from multispecies lab-oratory microcosms (Van Voris et al., 1980) towatersheds (Federer et al., 1989) indicate that loss ofCa is an important early indicator of disturbance ofnutrient cycling in terrestrial ecosystems. Recentdocumentation of significant long-term losses of Cain soils and vegetation within the Hubbard Brook‘Il’atershed in the northeastern US.4 (Likens et al.,1996, 1998) indicates that issues of Ca supply andcycling might provide increasingly importantinsights into future patterns of forest growth andecosystem function.

20 years ago Bangerth (1979) emphasized thatmany factors in the environment ultimately affect Caconcentration in plant organs, and he called for

I ransplrarv..,,,,t,-r- ion

FUNCTIONMembrane integrityStomata1 regulationEnzyme activationCarbohydrate metabolismCold hardinessDef ensekhemical-ph’isical

GROWTHCell division -Cell wall synthesisStress tolerance

STRUCTURECanopy integrityLeaf formWood quality

,Deposition

txcnange

Rotit’distribution

Leaching

Physiological processes pa s”pp’y rate

Fig. 1. Diagram sho\ving indicators of forest physiological function, grolvth and structure that are linked tobiogeochemical cycles through processes that control rates of Ca supply.

i _‘.

extending the field of experimentation in Ca metab-olism into more complex environmental settings. Hewanted field experiments ’ . . . to define which facto&are decisive for Ca uptake and distribution.’ Hefurther noted that ‘Such experiments are far morelaborious and expensive than investigations in con-trolled environments, but they are necessary for amore efficient utilization of the already accumulatedinformation on the basic mechanisms of Ca uptake,distribution, and function.’ Many types of suchexperiments have now been performed, sometimesby design, and sometimes by chance, as research hasincreasingly focused on understanding and pre-dicting plant and ecosystem responses to disturbanceand to the impacts of current and projected futurechanges in chemical and physical climate. These

experiments have provided a format for examining awide array of processes potentially influenced by Casupply levels along both spatial and temporalgradients.

*Here, we propose a hypothesis that stems fromtwo previously noted aspects of the metabolicfunctions of Ca: regulation of many critical meta-bolic processes; and limited mobility in plants. Ourhypothesis is that Ca supply ‘exerts a significantcontrol on both the structure and the function offorest ecosystems. We reason that for a nutrientelement having these properties, adaptive strategiesto limitations in Ca supply will have had animportant influence on the physiology, morphology,growth and evolutionary development of individualplant species as well as the differentiation of plant

376

communities. We propose that mechanisms for Caconservation will be increasingly evident on bothphysiological and structural adaptations in forests atthe later successional stages as availability becomesincreasingly limited. Our primary emphasis is onforest systems because their longevity, diversity andstructural complexity make them excellent testinggrounds for evaluating this hypothesis. Within forestsystems we will focus on the interplay of Ca supplyand Ca flux between compartments as they in-fluence physiology, growth and nutrient cycling atorganizational scales ranging from the cell to foreststructure and function (Fig. 1).

II. EFFECTS OF CALCIUM ON PHYSIOLOGICAL

PROCESSES

1. The chemical uniquelzess of calcium

Fundamental to an appreciation for the varied rolesthat Ca plays in plant systems is an understanding ofthe unique chemical properties that allow it to existin a wide variety of binding states, and thereby tofulfill diverse biochemical functions. Hepler &Wayne (1985) have provided an excellent overview ofthe chemical attributes that define the ‘fitness of Ca’to fill these diverse roles. Here, Ca chemistry isconsidered in relation to three principal attributes:cytotoxicity; binding properties; and stimulation/displacement potential.

(a) Cytotoxicity. Ironically, this highly functionalregulator of cellular processes is also highly toxic tocytoplasmic processes at the millimolar concen-trations that exist outside the cytoplasm. Thistoxicity is a consequence of the precipitation ofinorganic phosphate, which impairs phosphorousmetabolism within the cell. Maintenance of themicromolar concentrations necessary for cellfunctions in the cytoplasm is accomplished byenergy-requiring Ca transporters, which functionwithin the plasma membrane, as well as within othermembrane systems that delimit other subcellularorganelles (Kauss, 1987). These pumps maintain abalance betqeen Ca export from the cytoplasm’ byactive transport through antiports and influx throughspecific channels (Bush, 1995). As a result, cyto-plasmic concentrations are maintained in the micro-molar range (lo+- 10-j M), but Ca levels in theother subcellular organelles, notably the vacuole, arein the millifiolar range, a gradient of three to fourorders of magnitude.

The end result of the steep concentration gradientsthat exist within the cell is that the potential for Cagradients to be promoted by external stimuli is veryhigh, and that the magnitude of change required inmembrane sjrstems of the subcellular organellesto affect significantly cytoplasmic Ca levels is verylow (Bush, 1995). The active transport of Ca withinthe cell, resulting in the maintenance of steep

intracellular gradients, is in striking contrast to themore passive processes by which Ca moves from thesoil into the plant through the xylem and the generallack of internal retranslocation at the organ level. Forthis reason, plant growth is very sensitive to changesin Ca supply rates, and estimates of Ca sufficiencyfor small seedlings under steady-state supply innutrient culture (Goransson & Eldhuset, 1995) mayunderestimate the amounts required to maintainoptimal growth under variable supply and demandsequences typically experienced in the field.

(b) Binding properties. The diversity of bindingconditions that Ca can occupy is a consequence of itslarge nuclear size, charge density, and the availabilityof outer-shell electrons. These electrons allow Ca toparticipate in multiple configurations, including bothcovalent and ionic bonding. Because of its size andlow charge density, Ca is specifically favored inbonds with neutral oxygen sites in the biologicalligands, such as polysaccharides and lipids, andforms relatively weak bonds at multiple bondingsites. In comparison to its close competitor, thesmaller, more charge dense, Mg ion, Ca has arelatively low affinity for water and N atoms, and isbound more readily to oxygen at the multiple bindingpoints in cell membranes and cell walls. A lowaffinity for binding to water provides two advantagesfor Ca in biological systems : an enhanced capacity tobind to less highly charged ligands, because removalof water reduces the energy available for bonding;and fast reaction times, because Ca can quickly‘shed’ water and become actively involved inalternate binding states. The release of water by Caions is approximately 100 times faster than for Mg,for example. The capacity to occupy multiplebinding sites also allows Ca to be bound in muchmore complex configurations than Mg, perhapsexplaining its capacity to be bound to proteins, andas a site-specific enzyme cofactor (Roberts &Harmon, 1992).

(c) Stimulation/displacement potential. Bush (1995)refers to 12 types of stimuli that cause changes incytosolic Ca concentrations in various plant systems.These include a wide diversity of stimuli, such as lowtemperature, touch, gravity and the plant growthregulators auxin, cytokinin, ABA and GA. Most ofthese stimuli are accompanied by changes in cyto-solic pH, indicating that eschange of hydrogen forCa in cell membranes is involved in the signalingprocess (Bush, 1995). As an example, ion displace-ment of Ca in root membranes has been shown tooccur in the order Al > H > Fe > Mn (Stienen &Bauch, 1988); thus, the fluxes of these acidic cationsthrough cell walls of fine roots influence’ thedynamics of retention and release of Ca, and thefunctional integrity of associated cell membranes(Zhao et al., 1987). This is an important concept inaddressing changes in Ca availability in developingplant communities, because normal patterns in forest

. I

Ca signaling

Envirr;;$ntal

Physical -

Chemical -

Electrical -

z

c----.

1 Indirect fCa ,- ---A 1 Genetic code

1 TranscriotionI

c----\

1 Direct ID e v e l o p m e n ,

Ca k----d

>Ca

Ca ,

377

receptor receptors\-e--dBiologica lresponse

Plasma membrane __t Cytoplasm - Cell -Organ *Organism

Fig. 2. Conceptual diagram of the types and sequences of signals involved in the Ca-mediated activation ofreceptors, which control diverse physiological functions in plants. Both direct and indirect control, involvingprimary and secondary receptors operating in both local (intracellular) and remote control functions areenvisioned (see also Trewavai & M~lho, 1937).

succession, as well as many human influences on In Fig. 2, we have detailed the principal functionalthose patterns, lead to diminishing availability of the interrelationships among types and sequences ofbasic cations, such as Ca and Mg, and increasing signals and responses involved in Ca signaling at theavailability of acidic cations, such as H, Al, Fe and cellular level (Trewavas & Malho, 1997). CriticalMn in the soil environment at later successional features of those responses can be summarized asstages (Ulrich & Matzner, 1986). follows :

2. Calcium signaling.ond plant responses toevwironmentol stress

(a) Control principles. Three keys to Ca regulation ofplant processes are the maintenance of a homeostaticbackground of regulated compartments of Ca be-tween the cytosol, the plasmalemma and the in-tracellular organelles; sensitivity of those compart-ments to estetnal chemical and physical stimuli thatcreate Ca impulses within the system; and astimulus-response system by which Ca signals aretranslated into the plant response. The first twoequip ,tbe plant system to respond to environmentalstimuli’;’ the last provides the signal by which aresponse occurs.

Significant advances in understanding of thediversity, timing and functional complexity of Casignaling in plants have been made in the past decadeas new Ca-specific dyes have allowed temporal andspatial sequences at the intracellular level to bedetailed (\Vebb et al. , 1996). This has led toimportant new conceptual advances related to howCa-induced signals are generated and transmitted inplants and hoiv such responses are translated into thephysiological responses that govern form and func-t i o n (Trelvavas & M a l h o , 1 9 9 7 ; hIcAinsh &Hetherington, 1998; Snedden & Fromm, 1998).Here, we focus on conceptual highlights that relateto the timing and specificity of signals and theirrelationship to lvhole-plant physiology.

l Plasma membrane. The plasma membrane plays acentral role in transmitting Ca signals and inmaintaining homeostasis. A series of ports and gateshelp regulate Ca flow and maintains the three order-of-magnitude gradients between cytoplasmic andmembrane Ca contents, Other organelles have theirown signaling systems such that Ca serves as anintegrator of many types of information flow amongorganelles.l Signals and noise. Ca signals can be produced bya variety of external stimuli and recognized as specificstimuli by internal receptors or combinations ofreceptors that elicit responses. The recognition thatsome hypes of signals occur as oscillations, with botha specific frequency and amplitude, provides an easyanalogy to encoding in radio waves, which can beAM (amplitude- modulated) or FM (frequency-modulated), or perhaps both, in plant systems. Suchsignals can also have filters to remove noise andestablish thresholds, as well as secondary amplifiersand transmitters to select and enhance relevantportions of the induced signals. A simple model ofthe translation of Ca oscillation,s to signal enzymeactivation in plants is shown in Fig. 3. A centralfeature of this scheme is Ca regulation of proteinphosphorylation, and alteration of the balance be-tween phosphorylation and dephosphorylation ofproteins that controls enzyme-regulated reactionkinetics.l Receptors. Ca-activated proteins represent animportant class of receptors for Ca-moderated

378 a?ld R. Wimmer

Stimulus

Protein Ca2+-independenthosphatase protein kinase

I IProtein - P r o t e i n - P

1Response

Fig. 3. A model for decoding the information contained inCa spikes and oscillations that comprise signals in cytosolicfree Ca levels. Such signals are envisioned to controlrelative rates of phosphorylation and dephosphorylationand hence activation levels of enzymes (WAinsh &Hetherington, 1998).

responses in p lants (IJIcAinsh & Hetherington,1998). The identification of Ca-modulated proteinsand their modes of action has proceeded rapidly inrecent years, and Roberts & Harmon (1992) notedthat over 150 such proteins have been identified. Themost well characterized group of messenger proteinsfor Ca-induced signals are members of a Ca-bindingfamily known as calmodulin-binding proteins(Snedden & Fromm, 1998). Calmodulin is ideallysuited to serve as a Ca-activated signal because of itshigh solubility, low molecular weight, and the factthat it changes in charge (and conformation) uponbinding with Ca (Kauss, 1987). Although calmodulinhas no activity on its own, the role of messenger iseser ted through i ts capaci ty to act ivate otherenzymes. Thus, Ca-activated calmodulin can serveas part of an information cascade; directing activitiesof secondary systems at locations remote from theprimary site of action. The identification of multi-gene families of calmodulin and calmodulin-relatedproteins, and the growing list of known targets ofcalmodulin, suggest that a complex Ca-based regu-latory network controls a wide variety of responses tothe envirosment (Snedden & Fromm, 1998).l ROM and RAM. Trewavas & Malho (1997)coined a useful computer-based analogy to relateprogrammable translation systems for the variableCa-specific signals produced by external environ-mental stim,uli (‘ random-access memory ‘, RAM) tothe ‘read-only memory’ (ROM) by which plants aregenetically shaped and biochemically constrained.Thus, the functionally more-complex responses thatlead to plant growth and development can be

envisioned to involve increasing levels of coor-dination between genetic ‘ROM ’ and environment-stimulated ‘RAM’.l Ca movement and storage as signals. The im-portance of Ca concentration to cell signals and cellgrowth has apparently led to built-in controls thathelp dampen the influences of ‘ . . . the vagaries of thexylem source’ (Trewavas & Malho, 1997). Theseinclude reservoirs in the membranes themselves, aswell as accumulations as Ca-oxalate in specializedcells, the ideoblasts , from which subsequentsolubilization can occur in response to localized Cadepletion. Thus, the size and distribution of thesepools can provide some ‘memory’ of previousexposure, as well as providing buffering againstsubsequent changes in Ca supply. Additionally, it isnow recognized that Ca signals can be transmitted aschemical waves within the plant. The geometry ofconducting pathways and anatomical structures cancreate variable hydraulic pressure fields thatinfluence these waves. Because the primary move-ment of Ca in plants is along longitudinal axes ofwater flow, such waves provide some interestingpossibilities for integrating the influence of flow andstructure to help coordinate phyllotaxis in plants.

In the following sections we provide examples ofthe importance and diversity of Ca signals operatingin various direct and indirect ways to influenceselected processes that are important to whole-plantfunction and associated responses to stress.

(b) Carbohydrate metabolism. Through its acti-vation of a wide variety of protein kinases (Fig. 3),Ca has the capacity to influence many aspects ofcarbohydrate metabolism. However, the effects ofCa on carbohydrate metabolism have been morefrequently related to its influence on respiratorymetabolism and growth than to energy capture inphotosynthesis. Indirect Ca-related control of res-piration has been attributed to the loss of cellularcompartmentalization and reduced membrane in-tegrity associated with Ca deficiency (Bangerth,1979), although the specific mechanisms of thiscontrol are not well characterized.

At supraoptimal levels of Ca supply, plants have awell-developed Ca immobilization system involvingthe formation of Ca-oxalate, which precipitates, andhence detoxifies, excess Ca. The formation of Ca-oxalate deposits can be shown to occur in bothfoliage (Fink, 1991) and woody stems of trees,particularly those growing in Ca-rich environments(Gourlay, 1995). Fink (1991) found insoluble Ca-oxalate crystals in Norway spruce (Picea abik)needles outside the mesophyll cell walls in whichsoluble Ca was bound as Ca pectate in the middlelamella.’ The diversion of photosynthetically fixedcarbon to produce osalic acid represents an energ]sink that will divert some fraction of carbon andenergy to this protective role. Oxalic acid is a majorconstituent of the TCA respiratory cycle in plants

and occurs in crystalline deposits at levels comprisingup to 50yb of some plant tissues (Ranson, 1965).Thus, it should be available in amounts adequate toimmobilize excess Ca when the need arises.

The more widely documented example of thefunction of Ca in carbon metabolism is its capacity torepress respiratory activity of fruits. Respiratoryactivity is typically enhanced in fruits that sufferfrom poor Ca supply (Bangerth, 1979). Suppressionof respiratory metabolism and delayed ripening offruits by Ca have led to the use of Ca sprays to retardsenescence and diseases of fruits in horticulture(Ferguson, 1984). Though less widely studied, Cadeficiency can also increase the respiratory rates ofleafy tissue of herbaceous crops (Pal et al., 1973) andforest trees (McLaughlin et al., 1991, 1993). Rathermodest reductions in foliar Ca have also been shownto reduce translocation of photosynthate from foliageof soybean plants (Gossett et al., 1977).

The deterioration of a broad spectrum of essentialphysiological processes in Ca-deficient plants has ledto recognition of the role of Ca supply in delayingplant senescence (Pooviah, 1988). Symptoms of Cadeficiency that are commonly associated with sen-escence include loss of protein, loss of chlorophyll,and reduced integrity of cell membrane and cell wallsystems. Thus, the maintenance of adequate suppliesof Ca will be particularly important with perennialspecies such as forest trees, where partitioning ofcarbon between growth, respiration and defensefunctions becomes increasingly important with ageand the associated increasing respiratory demandson whole-tree energy budgets (McLaughlin &Shriner, 1980; Lechowicz, 1987 ; \Varing, 1987).

(c) Sy,lthesis atzd faction of membranes and cellwalls. Ca is essential for the structural integrity ofmembranes and cell walls because of both itsactivation of enzymes involved in structural bio-synthesis (Brummel & MacLachlan, 1989) and itsstabilizing influence on structural and functionalattributes of these structures (Jones & Lunt, 1967).Membrane architecture and function is dependenton structural and spatial interrelationships betweenlipid and protein membrane components that regu-late transmembrane flow of both nonpolar and polarmaterials (Palta & Li, 1978). Ca helps stabilizemembrane structure through the formation of Ca-dependent hydrated crosslinks within the more polarregions (MihcJrsky, 1985). Leaching or displacementof Ca by exposure to increased atmospheric or soilacidity has been shown to lead to altered permeabilityof root membranes (Zhao et a/., 1987), while foliarleaching by acid mists can affect membranes re-sponsible for regulation of water and electrolytes infoliage of conifers (Eamus et al., 1989). Alteration ofmembrane structure by Ca deficiency leads toaccumulation of Ca in vesicles, which is a typicalsymptom appearing in senescent cells (Hecht-Bucholz, 1979; Poovaiah, 1988).

379

Ca influences the composition of cell walls inhigher plants through its effect on the activities ofthe synthesizing enzymes. Activation is effected bymovement of terminal complexes on the surface ofthe plasma membrane to form microfibrils. Newfibril layers are sequentially deposited between theplasmalemma and the already-formed microfibrils.Modifications of cell wall composition are related tothe relative activities of enzymes involved in the cellwall formation processes, including the gycosyl-transferases, which are activated by Ca (Brummell &MacLachlan, 1989). Differentiation between cellu-lose and callose formation in the cell wall-formingprocess is also controlled by Ca, with greater relativeamounts of callose, a wound-repair agent, beingpreferentially formed at the higher Ca levels (Delmer& Amor, 1995).

Wood formation in forest trees is influenced by Casupply rates, both as a function of changes in cellwall plasticity of initiating cambial cells, as well asthrough stabilization of newly initiated cell wallsduring the growth process (Demarty et al., 1981).Cambial cell walls change structurally andchemically as they go through alternating active andresting stages during the year. The shift of cambialcells from active to resting is accompanied by wallthickening and stiffening processes (Catesson, 1990).By contrast , the resumption of activity in thecambium of ash (Fraxinus excelsior) was accom-panied by localized lysis of the radial walls (Catesson,1990; Funada & Catesson, 1991). Because cross-linking of acidic pectins by Ca bridges is animportant factor of wall rigidification, the breakdownof Ca-linked pectin molecules in the cambial zoneappears to be a necessary prerequisite to increasingcell wall plasticity and the resumption of radialexpansion of cambial cells (Funada & Catesson,1991).

The initiation of cell division at the cambiumbegins with fusion of Golgi vesicles to form a cellplate consisting mainly of pectin. The principalcomponent of pectin is a partially methylated a-1,4-polygalacturonic acid, and some of its carboxylgroups form Ca salts to produce a hard gel by cross-linking between the molecules. On both sides of thispectin layer, primary and secondary walls are formedsuccessively. Therefore, pectic substances constituteone of the major carbohydrates of the intercellularportion ic developing wood cells and are a majorcomponent of the primary wall of dicots (O’Neil etal., 1990). Pectin is later degraded or removed priorto the Iignification, apparently releasing Ca ions thatrvere utilized during lignification (Westermark et al.,1986).

Diinisch et al. (1998) found K to be the drivingforce in cell enlargement of differentiating spruce *tracheids (primary wall phase), while Ca is especiall)important for the secondary wall synthesis andlienification. This orocess of lienification starts in

380 Wimmer

the ceil corner regions and proceeds in the middlelamella region after the primary wall has formed, andjust before the formation of the secondary wall hasstarted. Pectin has an important role duringlignification in the cambial region and Ca plays amajor role in this process. Westermark (1982)suggested that lignin in wood cells is polymerizedwithin the cell wall by a Ca-dependent mechanism.Ca-mediated peroxidase activity is involved in ligninsynthesis, gradually converting the hydrophilic gelof the primary wall into a hydrophobic gel, com-prised of lignin and hemicellulose, and releasing Ca.

The strength and flexibility of cell walls isdetermined by the biochemical composition, wallthickness, and stabilizing cross linkages. Mechanicalflexibility of wheat coleoptile tissue was found to bereduced at high Ca supply resulting in a more rigidcell wall with reduced plasticity (Tagawa & Bonner,1957). The assumption is that high Ca concen-trations link and stiffen the polysaccharides of thewall, thereby preventing the turgor-driven extensionfrom occurring. Cleland & Rayle (1977) argue thatthe inhibition of growth by high-tissue Ca contentresults more from an inhibition of biochemical wall-loosening processes than a direct stiffening action ofCa through formation of Ca-bridges,

(d) Disease resistance and wound repair. Theassociation of Ca nutrition with plant resistance to awide variety of diseases and physiological disordersis a well-documented result of the important struc-tural and biochemical roles that Ca plays in defenseand repair (Bangerth, 1979; Kirkby & Pilbeam,1984; Pooviah, 1988). In addition, Ca functions asan early-warning signal aiding in the recognition ofpathogen invasion (Dixon et al., 199-C). In evaluatingthe role of Ca in plant defense, it is important toconsider both how Ca deficiencies develop spatiallyand temporally nithin the plant, as well as where andhow Ca is specifically involved in the defense process.

Shear (1975) has listed over 30 Ca-related dis-orders that affect fruit and vegetable crops, includ,ingboth deficiency and toxicity responses. Most wereassociated nith foliage and fruits and represent awide variety of rots and decays, and abnormalpatterns of tissue growth, that become more abun-dant at low Ca concentrations. Diseases of fruits andvegetables have been of obvious interest from aneconomic perspective, but are also significant physio-logically because they often represent the problemsassociated ‘&th poor Ca distribution rather thanoverall low Ca levels in the plant (Kirkaby &Philbeam, 1984). The apparent cause of the highsensitivity of fruits and vegetables to Ca-relateddiseases is that they are non-transpiring organs,without, direct linkages to supply by transpiration(Clarkson & Hanson, 1980).

Increasing Ca supply to fruit trees with liming ordirect application \vith various Ca-containing spraysis no\? well established as a means of remediating

diseases of fruits in the horticultural industry (Zocchi& NIignani, 1995; Raese & Drake, 1996). Stark(1964) cites numerous studies demonstrating that Cafertilization can increase tree resistance to forestinsects or adversely affect insect survival. AlthoughBangerth (1979) noted that Ca deficiencies cansometimes occur on plants growing in Ca-rich soils,and that liming is not always the answer, it appearsevident from many studies that increasing Ca supplyfrom the soil generally improves Ca distribution andthe alleviation of many types of diseases. Twoimportant components of plant defense, recognitionand containment, which are dependent on Ca signals,are also influenced by Ca supply.

Rapid recognition and response to stress fromwounding or disease can be an important measure ofthe likelihood that a plant will survive that damage.Davies (1987a) lists three types of wound signalsproduced at the site of injury to plant tissues. Theseinclude induction of chemical signals, such ashormones, oligosaccharides, or phermones; elec-trical signals, which result in an electrochemicalgradient or action potential that can alter polarity ofplant membranes; and physical signals, such asplugging of plasmodesmata, changes in cytoplasmicstreaming’and changes in cell turgor that shift thebalance of apoplast-symplast relationships.

Plant responses to these wound signals includeboth chemical and physical defenses that isolate thewound area and/or retard development of thepathogen (Dixon et nl., 1994). Rapid death of cells atthe point of injury or attack-the hypersensitiver e s p o n s e -can provide fast, effective protection byisolating the damaged area. Ca has been shown to berequired for the hypersensitive response in soybeansand to be release2 in response to a wound-inducedburst in H,O, (Levine et al., 1996). Plant release ofantimicrobial agents, such as phytoalexins andhydrolytic enzymes, contr ibute by at tacking orretarding the pathogen chemically. In addition, theformation of protective physical barriers with lignin,cell wall proteins, or callose (Dixon et al., 1994;Davies, 1987a) may essentially wall-off the damagedor infected tissue. The formation of protectivebarriers is particularly important as a defensemechanism with forest trees (Shigo, 1984).

Although it is clear that plant defenses againstinjury and disease are multicomponent systems,which are both influenced by many aspects of hostand pathogen physiology and modified by theenvironment, it is also clear that Ca plays a majorrole in several important aspects of this process. Thisincludes serving as a primary intracellular signal inresponse to ion leakage produced by injury (Dixon etal., 1994), as well as acting as a secondary in-tracellular sign.al induced by intercellular actionpotentials (Davies, 1987b). In addition, Ca has beenshown to be required for synthesis of phytoalexin, anantimicrobial agent (Dison et al., 1994). Its signaling

381

L J I ,

1994 1995Sample date

Fig. 4. Seasonal changes in membrane-associated Ca in current-year red spruce foliage show significant shiftsassociated with phenological changes and winter hardening (DeHayes ef al., 1997). These were not apparentin whole-needle Ca levels nor in one-year-old foliage. Initial freezing temperatures were experienced in mid-November.

role in cascade systems equips Ca to activate a hvidevariety of enzyme systems. Ca-activated enzymesynthesis is aiso knolvn to play an important role inthe formation of two protective compounds that helpthe plant seal off wound areas: callose and lignin(Dixon et ol., 199+).

The signaling and response system by whichcallose is formed is now tvell known (Kaus, 1987). Itis one of a wide variety of enzymatic systems inwhich calmodulin plays a central role as a Ca-mediated signal (Roberts & Harmon, 1992). Plantcells respond rapidly to cellular injury to formcallose, a carbohydrate polymer of a-1-3 glucans thatserves to plug damage and/or strengthen stressedcell \valls and membranes. The Ca-activated enzyme\vhich stimulates callose deposition is 2-l-3 glucansynthase, and the signal for call&e formation, whichmay start as rapidly as 10 min after a wound signal,can be produced by cell wall fragments generated b,injury.‘or by significant disruption in membranepermeability (Kaus, 1987). A key component inactivation of callose depositiqn on cell walls inresponse to these fragments is a shift in membranepermeability, and a release of Ca. Removal of boundCa from membranes also apparently reduces theircapacity to form callose. Loss of membrane integrityis in itself k”potentially important factor in thevirulence of many pathogens for which containmentof nutrient flow from the host to the pathogen is anessential part of defense (Hancock & Huisman,1981). A demonstrable decrease in cellulose (~-1-kglucan synthetase) activity and a rapid increase incallase (r-1-3 synthetase) in response to wounding(Delmer e t nf., 1985) suggests that plants areequipped rapidly to shift their biosynthetic ma-chinery from growth to structural defense whenchallenged.

Lignin is another structural defense agent, whichcan be activated in response to wounding andinfection (Dison et nl., 1994). Increased ligninsynthesis has been shown to have both local andsystemic, control elements in cucumber (Czrcumissatizm) plants (Dean & Kuc, 1987). Thus, the initialwound signal can produce enhanced lignin synthesispotential in leaves not initially infected by thepathogen. Increased lignin synthesis in tree stems isa prominent feature of the response of both balsamfir (Abies bnlsnwen) and Fraser fir to attack by theBalsam wooley adelgid (Aclelgis balsanzea), an aphid-like sucking insect that has caused significantmortality to these species in eastern Xorth Americain the past three decades (Timmel, 1986).

Ca appears to play a role in lignin synthesis in bothnormal tree-growth processes (Eklund c9: Eliasson,1990) and lignin increases in response to wounding.Increases in perosidase, which stimulates ligninsynthesis, have been associated with the systemiceffects of infection in cucumber leaves (Dean & Kuc,1987). Rapid increases in perosidase activity havealso been suggested as part of a Ca-moderatedresponse to the generation of action potentials inwounded plants (Panel et nl., 1985).

In summary, Ca plays an essential role in all of thebasic categories of plant defense against disease:maintenance of membrane integrity; signaling mul-tiple pathways of defense through enzyme activation;release of phytoalesin, an antimicrobial agent; repairand reinforcement of damaged membranes and cellwalls; and synthesis of structural barriers.

(e) Cold tolerance. The importance of cytoplasmicCa levels in maintenance of plant membrane per-meability led hIinorsky (1985) to postulate that Caplayed an important role in protecting plants fromchilling injury. nIaintenance of membrane integrity

382 S. B. McL.mlghlin and R. Wimmer

in promoting cold tolerance is, in some respects,analogous to the protective role of Ca in defenseagainst plant disease (Davies, 1987a). nlembraneregulation of water extrusion from the cytoplasmduring cold hardening,as well as resistance tocellular dehydration during the formation of estra-cellular ice crystals, is involved in freezing resistance(Levitt, 1978; Guy, 1990).

Interest in the physiology of winter damage to redspruce (Picea rtrbens) in the north-eastern USA hasbeen an important aspect of research examiningdeclines in growth of this species at high-elevationsites during the past three decades (Johnson et al.,1958; Cook ei Johnson, 1989; DeHayes, 1992). It isnow apparent from analyses of seasonal changes inmembrane-associated Ca in red spruce foliage(Dehayes et al., 1997) that increases in such Ca arerelated to increased cold hardening. Temperaturefluctuations were found to regulate the membrane-associated Ca and the subsequent dehardening-hardening cycle in red spruce as shown in Fig. 4.This study documented an important aspect of Caphysiology of plant organs - that a significant changein a very small, but physiologically active com-partment can be sufficient to alter physiologicalperformance of that organ in the absence of de-monstrable changes in overal’l tissue levels.

The capacity of plants’ to maintain adequatetranspiration rates in highly Ca-enriched environ-ments would be challenged ‘by such a strategy. *However, calcicoles can sequester incoming Ca asCa-osalate in epiderma! trichomes, and thus avoidhigh Ca concentrations in guard cells and stomata!closure (DeSilva et nl., 1996). Accumulation of Ca-osalate crystals in the woody tissues of trees in aridenvironments may be indicative of a similar seques-tration strategy (Gourlay, 1995) that helps themadapt to the excess supply potential of their growingenvironment.

111. CALCIUbI U P T A K E ASD DISTRIBCTIOS A T

T H E WHOLE-PLAXT L E V E L

(f) Stomntal regfflntion. Ca plays a role in control ofleaf gas exchange through its regulation of stomata1guard cell tclrgor relations and stomata1 opening andclosing (McAinsh et al., 1996). Stomata1 closure ispromoted by high concentrations of cytosolic free Cain guard cells, which can in turn be stimulated byseveral types of environmental stimuli, includingabscisic acid, increased CO, and osidative stresses(McAinsh e t nl., 1997). Although the specificmechanisms of Ca-induced closure are not known,oscillations in cjtoplasmic Ca concentrations arethought to be related to apoplastic supply rates andto provide stimulus-specific’signatures that enableplants to differentiate among stimuli (hlcAinsh etal., 1996).

Bangerth (1979) reviewed a wide range of studiesthat indicate that Ca supply patterns in plants mightbe analogous to the source-sink concept by whichcarbohydrates are distributed along gradients frompoints of production to points of greatest use (1Varingg L Patrick, 1975). These Ca gradients appear to beprimarily driven by relative rates of water movementto transpiring organs and relative rates of Ca usealong the transport pathway. Thus, rapidly growingstems, twigs, or foliage act as competing sinks fordirectional Ca flow and use (Clarkson, 1984).Together with Ca uptake rate from the soil, therelative activities and distribution of these utilizationcenters are important in determining relative avail-ability of Ca to storage organs and their resistance tomany types of disease. Reduced availability of soilwater can also significantly increase the severity ofCa-deficiency disorders in storage organs (Bangerth,1979; Ferguson, 198-l), an apparent consequence oflinkages of Ca supply with the transpiration stream.

1. Uptake at the root-soil irlterface

Viewed from a whole-plant perspective, linkagesbetween fluctuations in stomata1 guard cell functionand Ca concentrations in the transpiration streamcould potentially provide a sensitive indicator of thebalance between Ca (and water) uptake and use alongthe t ransport pathsvay f rom roots to shoots .Operating in conjunction ivith plant water potential,Ca concent&ions in the transpiration stream wouldprovide a means of influencing water-use effi&ncy.Rapid growth, and associated depletion of Ca, maythereby provide a signal of a physiological ‘need’ foropen stomates driven by Ca depletion at growthcenters along the transpiration pathway; slowedgrowth ivould lead to increased Ca concentrations inthe transpiration stream. Such a strategy wouldfacilitate stomata1 closure, and improved nutrientand water-use eficiency at slower groivth rates.

Ca uptake into roots occurs principally by passivemovement in the miss flow of soil water driven bqthe transpiration stream (Bangerth, 1979). Thus,root ‘activi’ty’ in obtaining Ca is indirectly keyed tocarbon assimilation in the canopy through con-current regulation of fluses of water and carbonthrough stomates. The concentrations of Ca in thesoil solution are controlled by soil chemistry anddiffusidn along concentration gradients. Criticalfeatures of Ca supply capacity include the soil cationexchange capacity, the proportional representationof Ca in the eschangeable base cation pool, the rateat Fvhich mineralization of soil organic matter andprimary minerals release soluble Ca, and the pH ofthe soil solution (Ulrich S; hlatzner, 1986):Totalcation concentration is also balanced and controlledby levels of mineral acid anions in the soil solution(Robarge & Johnson, 1992).

It is important to recognize the nonlinearity of theeffects of soil pH on Ca availability in soil solutions(Reuss PC Johnson, 1986). Xt pH levels Z 5.0,

383

Cation mobilization-

Baset saturation

-Increasing solutions anions 4

Fig. 5. Ca availability in soil solutions is stronglyinfluenced by soil weathering, cation-exchange capacity ofsoils, base saturation and inputs of strong acid anions,which mobilize exchangeable AI mire rapidly than Ca. Atlower base-saturation levels, large changes in relativeavailability of Al and Ca are produced by relatively smallchanges in ionic strength. At Ca:Al ratios < 1.0 Caavailability to plant roots is further reduced by chemicalinterference of Al with Ca uptake.

increasing acidity is associated with both increasedmineral weathering and increased availability of Cawithin the soil solution. This: is effected by re-placement of Ca on soil cation-exchange sites by H’or other acidic cations and release of Ca to the soilsolution, from which it may be absorbed by roots orlost by leaching. Both natural and human influencesthat acidify soils will make a relatively largerproportion of the soil-exchangeable Ca available inthe soil solution (Ulrich & Matzner, 1986). As soilpH drops below 5.0, A13+ is mobilized more rapidlythan Cap+, because release of cations is proportionalto charge, resulting in a greater relative represen-tation of Al in the resulting soil solution (Reuss &Johnson, 1986). In addition, as the pH drops below4.5, significant stores of Ca-oxalate, which in somesoils can represent half of the exchangeable pools inthe forest f loor, can become rapidly dissolved(Cromack et al., 1979). The mobilization of Ca byanion loading is not tied to the natural annual cyclesof temperature and associated processes of decompo-sition and root growth that help match nutrientuptake with nutrient availability. For this reason,stochastic deposition patterns release ‘excess’ cationsthat may be lost from the system during much of theannual cycle. Over time this will result in loss of basesaturation’df exchange sites, unless weathering anddecomposition re-supply the system at rates equi-valent to losses to uptake and leaching.

At very low soil-solution pH, Ca mobilization isreduced, Al becomes the dominant cation, andresulting Ca: Al ratios decrease. Mineral-acid anionsare particularly important in determining cationconcentrations at this stag6 because, as anion concen-trations increase, a greater proportion of exchange-able cations, including Ca, but especially Al, arereleased from exchange sites within the soil complex.

Al (pmol I-‘)

Fig. 6. Al concentrations affecting Ca uptake and rootelongation for Norway spruce seedlings in solution cultureindicate a relatively greater effect of Al on Ca uptake thanon growth at lower AI concentrations (Godbold & Kettner,1991; Godbold et al., 1988).

The interplay of soil weathering, percentage basesaturation, and anion-induced mobilization ininfluencing Ca: Al ratios in soils is depicted in Fig.5. Ca availability for root uptake decreases evenmore rapidly below soil pH 4.0, because Al, a strongcompetitor with Ca with a large soil reservoir, i’sreleased even more rapidly than Ca and interfereswith Ca uptake and root growth. Studies with treeseedlings in nutrient culture indicate that at soilsolution Ca: Al ratios below 1 .O, Al interferes withCa uptake by roots (Rost-Siebert, 1983; Schroederet al., 1988). Patterns of Ca uptake and growth innutrient solutions (Fig. 6) indicate that interferencewith Ca uptake occurs at lower Al concentrationsthan are required for direct inhibition of root growth(Godbold et al., 1988; Godbold & Kaettner, 1991).

Bangerth (1979) described typical Ca concen-trations at the root-soil interface as 0.1-l mMcompared to soil-solution values in the range of3.4-14 mM. However, in forest systems, particularlythose in later successional stages or in industrializedregions, soil-solution Ca values may be much lowerand associated with low Ca: Al ratios and in-terference from Al. Johnson et al. (1991), forexample, found typical A-horizon soil solution Cavalues to range between 10 and 150 pM and Ca : Alratios to average approximately 0.4 for a highelevation southern Appalachian (USA) red spruceforest. Values recorded at a comparable depth in abeech (Fagus sy[oatica) forest in Germany were inthe range lo-600 pILI Ca (iLIatzner & Prenzel, 1992).

The typical pathway for uptake of mineralelements involves initial movement into the freespace of the root apoplast, and subsequent basipetalmovement through apoplastic pathways. This pat-tern of movement through the cell walls avoidstrans-membrane passage and the toxicity problemsthat contact with the cytoplasm would entail(Haynes, 1980). Thus, Ca uptake is confined to theunsuberized tine root tips that have no casparian

384

strip, which would force a symplastic detour. Inaddition, because xylem movement of Ca is uni-direct ional and passive, f ine roots cannot beresupplied with Ca from basipetal tissues and areparticularly susceptible to low Ca supply from thesoil (Bangerth, 1979).

’ The status and the cation-exchange capacity of theapoplast is particularly important for Ca uptake,because Ca must move entirely through the apoplastto the conducting xylem. In the apoplast, the cation-exchange capacity of the cell wall is influenced by therelative concentrations of cellulose, hemicelluloseand lignin, constituents that can vary significantlybetween plant species and can have an importantinfluence on adsorption of specific ions. Below pH4.0; uptake of both Ca and Mg by fine roots ofNorway spruce seedlings is blocked by competingions in hydroponic cultures in the order of Al >H> Fe >Mn (Stienen & Bauch, 1988). Interferencewith and/or replacement of the less tightly bound Cais a part of this effect. Schroeder et al., (1988) foundthat the combination of pH 4.5 and increased Al : Caratios reduced Ca binding in root cell walls by 83 %.The low Ca binding (-80 y’) found in cell wallsof trees from a field site acidified by atmosphericdeposition (soil pH 3.0) was partially reversible bysubsequently adding pH 4.7 solution, but entailed aloss of 40 y0 of the original exchange capacity. Thus,residual effects on root uptake of Ca can be producedby prior exposure to high Al concentrations.

Although coping with low Ca supply is only one ofseveral stresses that plants growing in very acidicenvironments must face, interspecific differences inthe cation-exchange capacity of root cell walls mayprovide some adaptive advantage in Ca uptake. Thecation-exchange capacity of root cell walls has beenfound to be reduced in acidofilous herbs (Blarney etal., 1990; Koedam et al., 1992). The suggested gainsfrom such a strategy were attributed to reducing Aldisplacement of Ca from cell walls (Koedam et al.,1992). This adaptation would appear to represent again if cell-wall Ca involved in tissue structuralintegrity were protected at non-exchangeable sites,or if low retention of Ca in the root walls facilitatedrapid movement to shoots.

2. Transport and exchange in stems

Ca is found in’ greater quantities than any otherinorganic element in plants, and, as previously noted,its influx to the xylem is a function of transpirationflow (illarschner, 1995). Thus, like water supply, theflow of Ca is closely linked to the functional leaf areamaintained through the conducting sapwood cross-sectional arei’(Rogers & Hinckley, 1979; Whitehead& Jarvis, 1981; Ryan, 1989). Whereas transpirationdetermines the total amount of Ca that enters thetree, annual growth rates regulate the Ca that isincorporated into woody tissues.

Although Ca moves as a part of the transpirationstream, the pathway is no longer considered to be adirect one. Rather Ca is now considered to move asa part of an ion-exchange complex, with adsorptionand desorption occurring from active exchange siteswithin the cell walls of the xylem sapwood along theway (Ferguson & Bollard, 1976; Hanger, 1979). Theion-exchanging stem wood functions as a columnin equilibrium with the transpiration stream(Momoshima & Bondietti, 1990). On sites withadequate Ca supply, growth rate appears to beinversely related to potential Ca exchange activity inthe xylem. This occurs because increased growth isassociated with wider, but reduced numbers of,annual rings in the functional sapwood area. Inslower growing trees, with smaller annual growthrings, more rings and a larger surface area of cell wallsurfaces are included in the conducting cross-section,thereby increasing the potential influences of cellwall exchange on Ca supply rates. Therefore, aslower growing tree will retain sap flow within aparticular annual ring longer than a faster-growingtree with sapwood that has less rings involved.Increased surface-area contact within the sapwoodincreases the opportunity for exchange reactionsbetween Ca in the sap and the cell walls (Guyette &Cutter, 1995). Species differ in this respect becauseof differences in xylem structure that result insystematic variations in efficiency of water transportthrough the stem. The role of reduced Ca supply inthe production of smaller rings is an interestingissue, and whether it represents a consequence ofslow growth or an adaptation to nutrient supplylimitations that limit growth is unknown.

The cation exchange capacity of red spruce xylemhas been examined under laboratory conditions andfound to be very sensitive to both sap pH and radialdistance from the pith (Momoshima & Bondietti,1990). The Ca content of the xylem was estimated tooccupy 60-80 “/, of the cat ion binding si tes .Increased sap acidity resulted in displacement of Cafrom cell wall exchange sites. In addition, age/diameter-related decreases in binding capacity resultin a natural pattern of decreasing Ca retention withtree age (Momoshima & Bondietti, 1990; Shortle etal. 1995). This is apparently attributable to adecrease of the cellulose: lignin ratio accompaniedby increased crystallinity of cellulose in older wood(Erickson & Arima, 1974; \Vellwood et al., 1974).The relationship of adsorbed Ca in stemwood to sappH and Ca concentration is thought to represent anequilibrium reaction involving feedbacks betweensap pH, sapwood biGding status, and concentrationsof Ca and competing cations. For example, either anincrease in sap acidity by.0.5 pH units at constant Caconcentrations or an 80% drop in sap Ca concen-trations at constant pH was found to reduce cell wallbinding of Ca by about 25 y. (Momoshima S:Bondietti, 1990).

The cell wall surface contains a high number ofnegative binding si tes provided by l ignin andcellulose. Total Ca exchange capacity of individualtracheids depends on the inner surface area, as wellas the ultrastructural and chemical composition of itscell walls. The cation exchange capacity per cm2vessel surface has been calculated at the rather highvalue of lo-’ mol equivalent which led Van de Geijn& Petit (1979) to hypothesize that cations mustpenetrate into the cell wall. For a 2 mm-thick cellwall, Kuhn et al. (1997) calculated cation exchangecapacity to be 300-500 mol md3. These authors werealso able to demonstrate in a LAMMA/SI1CISmicroprobe study that Ca is evenly distributedwithin the cell wall, which means that Ca may accessthe entire donan free space of the apoplast.

These studies raise a series of questions regardingthe significance of changes in cell wall Ca measuredat near-ground level :

What fraction of the total available Ca in cell wallsof both roots and stems is actually exchangeableunder field conditions ?How do these changes affect the level and timingof Ca supplied to the growing centers further upthe stem?What are the consequences of changes in cell wallCa on the structural integrity o’f cell wallsthemselves ?

Isotopic analysis of cation exchange by Norwayspruce roots (Kuhn et al., 1995) suggests that theexchangeable fraction in cell walls of fine mycorrhizalroots, at least, is quite high, approaching 100%. 1~addition, occupancy of binding sites in the cell wallby alternate cations, such as Al, was found actually tofacilitate movement of Ca through the xylem toapical growth centers. This resulted from morebasipetal anionic binding sites in the stem beingblocked. Thus, although net uptake was reduced bylower pH and the presence of Al in solution, a higherfraction of Ca absorbed by roots moved up the stempast occupied binding sites along the path from rootsto shoots. This raises an interesting question relativeto the potential role of the exchangeable Ca in thecell wall as a reservoir for Ca supply to acropetalgrowing centers. Displacement of ‘bound’ Ca by Al

under intermittent acidic pulses related to soilwarming events, or more gradual increases of eithernatural or anthropogenic origin, might serve toincrease stem Ca supplies to the upper canopy inforest trees. This would result from displacing boundCa in xylem wood and by reduced adsorption of Cain the sap stream at sites now occupied by Al or otheracidic cations.

3. Ca exchange by foliage

Plant uptake of Ca from the soil is partially offset byrelatively high Ca losses from foliar leaching b>

385

Table 1. Net canopy fluxes of K; Ca, and Mg andrelationships to total base cation exchangefrom forestedsites in the IFS study (Ragsdale et al. 1992)

Flux (Sd)

‘& of total baseEq ha-’ yr-’ cat ions

I( 199 (97) 51 (11)C a 113 (48) 21 (9)a 68 (33) 13 (6)

Data are means (f 1 SD) from 12 sites.

incident rainfall. Among the cations exchanged by adeciduous canopy in east Tennessee (USA), Ca-leaching rates exceeded those for Mg by approsi-mately 50% and comprised from 40 y. (yellowpoplar, Liriodendron tulipifera) to 75 y. (oak, Quercusspp.) of the total cations leached (Johnson et al.,1985). In general, total Ca losses are dependent ontotal leaf area and are higher for mature forests, latesuccessional species ‘and evergreen canopy types. Acomparison of net canopy exchange of Ca from 12forests examined in the Integrated Forest Study(IFS; Johnson & Lindberg, 1992), as shown inTable 1, indicates that annual leaching rates rangedfrom > 50 to > 200 eq Ca ha-’ yr-’ (Ragsdale et al.,1992). The significance of foliar Ca exchange at theselevels is potentially high, both in terms of the totalannual Ca accumulation by forests and in terms ofstanding pools in foliage. Compared to net annualincrement, average foliar Ca losses of 113 kg ha-’(Table 1) exceeded net annual forest increment forsome Ca-depleted forest systems and amounted toover 50% of net annual uptake for approximately30 91; of the forests compared in IFS (Johnson &Lindberg, 1992; Fig. 7). Compared to standingfoliar Ca pools, leaching rates for a mature forestcanopy in Tennessee equated to approximately 20 9,of the annual foliar Ca pool for yellow poplar butonly 12 “/b of that pool for oak (Johnson et al., 1992).These losses compare favorably to the 4-13 ?‘bleaching losses for an early successional forest innearby North Carolina (Potter et al., 1991).

It may be argued that Ca leached from foliage isnot truly lost from the system because subsequentroot uptake and cycling occur from the soil; however,it is important to consider the origin of leached Ca.Detailed chemical Ca budgets show that a significantfraction of leached Ca is derived from internalsources. Microscopic analysis of internal Ca dis-tribution of conifer needles leached by acidic rainfallindicates that depletion of Ca occurs from apoplasticreserves (Fink, 1991). nfeasurements of changes indroplet chemistry on prewashed dogwood (Cornusjorida) leaf surfaces indicate that Ca efflux from thedogwood leaf interior was rapid, occurring withinminutes of contact, and 32-98O, of the droplet H+

386 S. B. McLaughlin and R. Wimmer

(al

SS2SSl ST2 ST1 NSlNS2 FL MS WF SB2 HF SBI TL FS CH DL GL CP LPI RA DF ‘LP2

S i t e

3 0 0 0

*No uptake data

-lOOO! . . . . . . . . ~ . . . . . . . . . . 1

, ‘.

SSPSSl ST2 ST1 NSlNS2 FL MS WF SB2 HF SBl TL FS CH DL GL CP LPl RA DF LP2

S i t e

Fig. 7. Comparison of Ca pools and fluxes for diverse forests types from the Integrated Forest Study (Johnson,1992). (a) Pools in vegetation and in the forest floor dominate soil-available Ca for many forest types.(b) Leaching losses vary significantly in magnitude and importance relative to Ca increment across these IFSsites (Johnson, 1992). Forest-site types discussed in this paper include high-elevation red spruce in theSouthern Appalachian mountains (SS and ST) and at Whiteface bIountain, New York (WF). Other forestsrepresented include Norway spruce (NS) in Norway, southern hardwoods (CH at Coweeta, North Carolina),northern hardwoods (HF in New York) and southern loblolly pine (DL, GL and LP).

was exchanged for Ca*+ (Willey & Hackney, 1991). Thus, although Ca loss from individual rain eventsLoss of leaf Ca in a single l-cm rain event at pH 4.0 may be relative small relative to total leaf Cawas sufficient produce a Ca concentration of 18-24 (Mitterhuber et ol., 1989), the cumulative effect ofmmol in foliar leachate and a loss of 1 “/o of total these losses over a typical growing season may befoliar Ca. In cell systems in which cytoplasmic large if considered from the perspectives of either theconcentrations are in the range lO-‘j- 10-s M (Bush, cumulative loss over an entire season or the episodic1995), shifts of 1.8 to 2.4 x 10m5 M Ca would be loss of Ca concentrations that are large relative to theanticipated to alter metabolic rates and induce shifts micromolar quantities involved in intracellular Cain the homeostasis of cells. signaling jus t discussed. This is indicared by nu-

387

merous studies that now demonstrate the adverse 20physiological effects of increased foliar leaching in

I a I 1 I 0 I c

rainfall acidified with strong acids (Schulze, 1989; 0 0McLaughlin & Kohut, 1992).

Iv. E C O S Y S T E M P R O C E S S E S A N D C A L C I U M

S U P P L Y

Processes that regulate Ca availability and influenceits effects on plant production over longer timeperiods and larger spatial scales are of interest fromthe perspectives of ecosystem function and regu-lation. Our interest here is both in the changes thatoccur naturally over space and time, but also inevidence that these changes significantly influenceplant and community structure and function.

- 0 Stages used in regression ’ c

0 I , I , I I I I0 50 100 150 200 250

Stand age (vr)

Fig. 8. Loss and/or organic sequestration of Ca from soils

1. Plant swcession and soil acidification is an important component of natural succession. In thisexample, which began with Ca-rich glacially derived

Plant succession depends on well recognized inter- substrate soi l , Ca levels both affected and were affected byrelationships between the growth habits and resource changing community composition and nutrient-use

requirements of competing vegetation and changespatterns (Marion et al., 1993a).

in the capacity of the site to supply required nutrientsand water, and to distribute solar radiation amongcomponents of the plant community over time.Modification of forest soils by the chronosequence ofvegetation types that leads from pioneering speciesto a climax forest community is an important part offorest successional processes. Over long time frames,it will lead to major changes in availability of all ofthe plant nutrients, but particularly Ca and N.

Ulrich (1983, 1984) describes a typical chrono-sequence of post-glacial primary succession thatleads from a base-rich, N-poor soil condition to aforest community in which increasing amounts ofcations and N are incorporated into vegetation, andinto the soil organic matter that helps buffer thesystem against nutrient deficiency. This wholeprocess, which may take many hundreds of yearsfully to develop, is centered around the chemicalrequirement to dalance the reducing power ofphotosynthesis and organic acids it produces withneutralizing influences of cations removed from thesoil and incorporated into phytomass. It leadsultimately to a more acidic forest soil, reducedconcentrations of base cations, particularly soil Ca,and accumulation of these cations in woody biomass,particularly foliage and branches, and in the organichumus layer (Ulrich & Matzner, 1986).

The acids ptoduced during the early aggradingstage of forest development are initially weakcarbonic acids from organic matter decompositionand root respirdtion. However, as organic matter andassociated N accumulate over time, soil pH levels arereduced by nitr ic and organic acid productionassociated with accumulation of humus in the uppersoil horizon. Thus, forest growth itself is anacidifying process that leads to three importantchemical processes that reduce the availability of Ca

for plant uptake over time. These include: in-corporation of Ca into the woody biomass, where itis temporarily unavailable to the system; loss ofavailable Ca from the soil thiough increased leachingas increasing amounts of exchangeable Ca from soilreserves are replaced by H+ in the soil-exchangecomplex; and reduced availability of Ca to the rootsattributable to competitive interference from Al andother acidic cations that increase in the soil solutionas pH levels drop to values < 5.0. Soil acidificationis not a process restricted to industrial regions withhigh atmospheric deposition of strong acids, but cansignificantly influence Ca availability in remote forestsystems as well.

The decreasing trend of soil Ca that occurs duringthe earlier stages of the developmental chrono-sequence described by Ulrich (1983) is well illus-trated by soil development occurring with sequentiallife stages during a 250-yr primary successionalsequence along the Tanana River in Alaska (Marion,1993 ; Fig. 8). These soils lost approximately 75 % ofthe original Ca content over 250 yr from internal acidproduction, which was calculated to have added 10keq ha-’ to the system each year. Although stillalkaline at the time of the study, the soil profile hadacidified by 2-3 pH units (pH 8.4 to pH 5.4-5.6) inthe upper 10 cm at the oldest (white spruce, Piceaglauca) successional stage (nlarion et al., 1993).

With time, natural successional processes, corn-bined with climate, lead to the redistribution oforganic matter and associated base cations fromdeeper profiles to an N-enriched organic surfacehorizon. Strong mineral and organic acids generatedin this layer leach the mineral soil below, andassociated cation losses and uptake by vegetationmay greatly exceed buffering and resupply by

389

type. In this study, root growth into implantedvermiculite cylinders, both with and without nu-trient addition, was several times higher on a Ca-deficient Amazonian forest soil than in forests whereN, P, or K were primary limiting nutrients.

2. Plant adaptations to nutrient d&iency

The high effectiveness of even relatively smallincrements in Ca on overall root-growth responsesraises an important point about evaluation offertilizer responses in Ca-limiting environments -the addition of N with an associated strong anion willincrease Ca availability if soils have adequate basesaiuration to permit Ca mobilization into the soilsolution (Fig. 5). Improved root growth and overallplant nutrition by these interactions could makeseparation of primary and secondary nutrient effectsof N addition difficult under these conditions.

One measure of the significance of nutrientlimitations to plant development and function is thediversity of plant morphological and physiologicaladaptations to conditions that limit nutrient supply.We examine here first the spectrum of adaptationsthat influences plant nutrient acquisition and con-servation, and subsequently, the potential import-ance of Ca deficiency as a driver of such adaptations.We consider the adaptations and their linkages to Casupply as one measure of the strength of our originalhypothesis - that Ca limitations can significantlyinfluence forest structure.

Esperiments involving addition of more Ca withseedlings and saplings confirm the potential signifi-cance of increased Ca supply for tropical forestspecies. These include experiments with acid-leached soils from a tropical rain forest in Singapore(Burlsem et al., 1995) and physiological studies with13 woody species from an oligotrophic Amazonianrain forest (Reich et al., 1995). These experiments,together with the documentation of Ca deficiency ofmature Terra Firme Amazonian forests by Cuevasand Medina (1988), indicate that Ca can be theprimary nutrient limiting tree growth in cation-depleted tropical settings. Particularly relevant arephysiological s tudies that have explored bothchanges in relative availability of Ca, N and P andtheir physiological effects for species spanning earlyto late successional stages in the Amazonian rainforest (Reich et al., 1995). Foliar Ca, P and N werefound to be highest in early successional species,although only Ca continued to decline through thelater successional stages. Although photosyntheticrates were correlated with N and, P in earlysuccessional species, at the lower levels of Ca, N andP occurring with later successional species, leafphotosynthetic rates were strongly and linearlycorrelated with leaf Ca concentrations (R* = 0.80)and not with leaf concentrations of N or P (Reich etal., 1995).

(a) Morphological adaptations. Chapin (1980) hasreviewed a wide range of native plant ‘strategies’ forreducing nutrient stress, and these are summarizedbriefly here as a basis for further hypothesis testing.Examples cited by Chapin were developed mainlyfrom studies involving N and P limitations on nativeherbaceous plants, but many of the same principlesapply with cations as well, and where they do not, thecontrasts can be useful. Plants in general shiftresources between roots and shoots depending onwhere the resources in most growth-limiting supplyare acquired. As a consequence, growth underconstraints in resource availability is typically shiftedtowards the organs involved in acquisition of theresource most limiting overall plant growth (i.e. thesite of maximum availability of the most limitingresource).

Other evidence that Ca is an important componentof the nutrient-deficiency complex that alters growthpatterns of tropical rain forests includes:

The influences of nutrient stress and low light onshifting root:shoot ratios are classical examples ofadaptations that result in a shift in carbon allocationto overcome low nutrients (higher R: S) or lowerlight (higher S:R). Although these adjustments arepossible to varying degrees within the physiologicalplasticity of individual plants and species, there arealso interspecific differences that appear in thesuccessional sequence that leads from pioneeringspecies, where light and nutrients are generally moreabundant, to late successional species where theseresources become more limiting. Thus, both nutrientand shade stress develop in parallel over time duringforest succession.

Chapin (1980) emphasized the importance of slow

growth rate as an adaptive s trategy with bothphysiological and phenological consequences .Slower growth rates, and lower nutrient demand,equip late successional species to survive at lownutrient supply rates. This strategy may also fa-cilitate resumption of growth following episodicnutrient stress periods, which can terminate seasonalgrowth of faster-growing species. Prominent amongthe nutrient-conserving strategies discussed by.Chapin (1980) are adaptations involving leaf lon-gevity and structure. The genera1 pattern with latersuccessional species is toward increased leaf lon-gevity and sclerophylly, both strategies that reduce

The absence of typical structural modificationsindicative o’f‘nutrient limitations on N-poor sitesthat have moderate Ca levels (Scott et al., 1992).The presence of stunting patterns at other sitesthat have adequate N levels (Bruijnzeel et al.,1993).The nature of nutrient-conserving adaptationsthat occur in these forests.

Clearly, additional field experimentation is neededto evaluate both functional and structural responsesof tropical forests to increases in Ca supply.

Calcium in terrestrial ecosystem processes 391

Table 2. Percentages of forest stands with NUE 2 25 y0 above theminimum value at >50% of the observed range of litter content levels fo;each forest type

Forest type

Element

CaNP

Tropical Deciduousy0 (sample No.) o/0 (sample No.)

71 (31) 42 (24)33 (28) 58 (26)31(31) 39 (23)

Conifero/0 (sample No.)

69 (16)50 (32)44 (16)

Data were derived from Vitousek (1982). Increased responsiveness of NUE toreduced Ca relative to N and P suggests that Ca becomes limiting at higherrelative nutrient contents than N or P for some forest types.

nutrients. For. example, the ranges in ratios ofmaximum to minimum nutrient concentrations infoliage of Amazonian forest species were 1: 65, 1: 12,1: 10, 1: 5 and 1:4 for Ca, Mg, K, P and N,respectively (Reich et al., 1995). These data suggestthat either Ca content is not very important, andhence that plants can exist in a wide variety ofconditions with little regard for Ca concentration or,alternatively, that Ca is very important and thatplants have adapted differential Ca uptake andutilization strategies to address these variations.

Vitousek (1982) developed the concept of nutrient-use efficiency (NUE) of leaf litter as a measure of thecapacity of diverse forest types to increase foliarbiomass production per unit of nutrient mass underconditions where total nutrient uptake and return inlitter is reduced. He concluded from increases inNUE at low nutrient-supply rates that increasedphysiological efficiency of utilization of N, P and Caoccurred’ as a physiological adjustment to lownutrient supply. These concepts are difficult to applyto Ca in a way comparable to N or P, because bothN and P are retranslocated before leaf senescence.However, our analysis of the changes in apparentNUE among nutrients for three forest types, shownin Table 2, suggest that, for some forest types,calculated NUE responds more readily to reducedCa supply, even when expressed at the whole-leaflevel, than to either N or P. This is supported by thedata of Cuevas and Medina (1986), who noted thatAmazonian forests of Tierra Ferme, which arerelatively high in N and very low in Ca, have leaflitter with high Ca NUE values.

In the Vjttousek (1982) analysis, two data pointsstood apart as outliers, both having unusually lowNUE and low total Ca return in litterfall. Both werefrom the heavily polluted temperate coniferous forestin Solling, Germany (Cole & Rapp, 1981) for whichelement fluxes, particularly N, S, Ca and Al, havebeen significantly altered by combinations of naturaland anthropogenic acidity (Ulrich & Matzner, 1986).The very low Ca uptake and return in litter in thisacidified forest ecosystem suggests that both ‘uptakeand use of Ca were reduced at these sites. This is

supported by substantial mechanistic evidence of the.role of anthropogenic inputs of actdity in amplifyingnatural acidification processes and biogeochemicalcycles that regulate availability of Ca and other basecations for plant uptake (Ulrich & Matzner, 1986;Robarge & Johnson, 1992). The causes and. impli-cations of these limitations for forest processesprovide further insights into Ca regulation ofecosystem processes at both whole-tree and com-munity scales.

v. PLANT AND ECOSYSTEM RESPONSES TO

HUMAN ALTERATIONS IN CALCIUM SUPPLY

The appearance in recent decades of widespread anddiverse symptoms of physiological stress on severalforest tree species in industrialized regions of theUSA and Western Europe has been extensivelyreviewed and debated with respect to both primaryand secondary causes and long-term significance toforest health (Rehfuess, 1981; Schutt & Cowling,1985; McLaughlin, 1985; Prinz, 1987; Pitelka &Raynal, 1989; and Rehfuess, 1991). In the case ofEuropean forests, increased forest growth observedin many areas in the past two decades raisesinteresting questions about the causes and signifi-cance of widespread foliar symptoms of stress clearlyapparent in many areas of Central Europe in the late1970s and early 1980s (Sterba, 1996). Evaluation ofthis apparent anomaly necessitates considering threefactors that are important in evaluating plantresponses to any stress: mechanisms of effect,changes in stress level over time and interactingenvironmental. variables. Here, we address theseissues in the context of the role of base cation supply

in areas where responses have been documented, andthe effects of time, location and atmospheric Ndeposition on forest physiology and forest nutrientcycles. *

1. Increased atmosphek inputs of acidity

Acidification of forests by natural processes of cationsequestration and internal acid production has beenaugmented in recent decades over widespread areas

. .

Calcium in terrestrial ecosystem processes 393

high-elevation sites in the Great Smoky Mountains,for example, very high Ca deposition levels, coupledwith high leaching rates and low Ca uptake byvegetation, are apparently leading to net accumu-lation of Ca even in the presence of high leachingrates (Johnson, 1992). Here apparent low uptake ofCa .by vegetation may constitute a form of self-regulation that has potentially adverse effects onforest physiological function.

(b) Inhibition of calcium uptake and effects on rootfunction. The elucidation of the reaction kinetics forAl and Ca mabilization in soils (Reuss & Johnson,1986) has led to the realization that large and rapidchanges in soil-solution chemistry are possible as aresult of strong anion inputs from the atmosphere(Robarge & Johnson, 1992). The non-linearity ofchanges in Ca and Al in soil solutions at lower soilpH values and low base saturation means that rathersmall changes in soil acidity or, alternatively, smallchanges in ionic strength of the soil solution cansignificantly shift the Ca:Al ratio (Reuss & Johnson,1986; Fig. 5). It is important to note that thisresponse requires only inputs of strong anions suchas SO, and NO,, not changes in soil pH per se.

, Al toxicity has been an important consideration inthe efforts to evaluate the potential of acidicdeposition to affect forest ecosystems for almost twodecades (Ulrich et al., 1980). Effects of Al on rootstructure and function can be direct at higherconcentrations or indirect as a consequence ofinterference with Ca uptake or retention. Oleksyn.etal. (1996) reported changes in organic chemistry,growth, structure and function of roots of Scats pine(P. sylvestris) seedlings at Al concentrations (0.5mM) that can be found in soil at European sites withhigh levels of acidic deposition. Controlled studies ofCa uptake, already discussed, indicate the import-ance of Al and other acidic cations in blocking Cauptake and/or replacing Ca from the apoplast of rootcell walls. Significant disruption of Ca flow to shootsand effects on both shoot and root growth inhydroponic studies with Norway spruce have beendemonstrated at a soil solution Ca:Al ratio of 0.67,well above the seasonal volume-weighted meanCa:Al ratio (0.36) measured at a high-elevationSouthern Appalachian site in the USA (Johnsonet al., 1991).

. Reduced Ca: Al ratios in roots of red spruce in theUSA (Shortb & Smith, 1988) and Norway spruce inGermany (Bauch, 1983) have been considered animportant indicator of declining tree vigor and rootfunction of those species. Low Ca:Al in fine roottips, in addition to inhibiting root growth, inducesincreased levels of a stress-related polyamine,putrescene, which can be detected in the foliage ofeven apparently healthy red spruce trees (Shortle etal., 1997). A broader evaluation of the implicationsof changes in the Ca : Al ratio in both soil solutionsand tree t issues (Cronan & Grigal, 1995) has

Probability of tree stress or toxicity

Fig. 10. A survey of forest responses to variations in theCa: Al ratio in soil solutions has been used to developgeneralized response surfaces that describe relative risks ofimpairment of forest processes by low Ca and high Allevels. Contrasts between mean, sensitive and moreresistant species are noted (Cronan & Grigal, 1995).

indicated that ‘this ratio may be a valuable indicatorof changes in forest function. From a review of awide range of studies, Cronan & Grigal (1995)classified the effects of decreasing Ca : Al ratios in thesoil solution in terms of increasing relative risks toforest ‘function (Fig.. 10). The > 50% risk level inthe mean response surface developed in Fig. 10occurs at a Ca : dl ratio of 1 .O, with sensitive speciesexpected to respond negatively at about twice theCa: Al concentration levels as resistant species.

Foliage typically experiences reduced function athigher Ca: Al ratios than roots. For example, theestimated 5Ogb risk level associated with foliarCa:Al values was > 12.5, 60 times higher than thethreshold for 50 yb’ risk to fine roots (0.2) (Cronan &Grigal, 1995). This is more likely a result of reducedmovement of the more-strongly adsorbed Al ion toshoots rather than higher sensitivity of foliage to Ca.Reduced Ca transport to foliage from roots can becaused by a combination of factors leading to eitherchemical interference with Ca uptake by Al, physio-logical effects on root integrity or function (Zhao,1987), or reductions in root mass.

Root distribution within the soil profile is animportant component of forest structure that alsoappears to be influenced by the combined effects ofCa and Al supply. The Ca:Al values of the mineralsoil at acidic forest sites are typically quite lowcompared to the more base-rich organic soils justbelow the forest floor (Johnson et al., 1991; Fig. 9),and this condition can limit rooting ixi the deepermineral soils. Evidence that acidic depositionreduces the effective depth of root production withinprofiles of poorly buffered acidic soils comes frommany different studies including: contrasts in fine-root distribution between healthy and decliningNorway spruce stands in the field (Meyer et al.,1988); contrasts in mature red spruce at field sitesdiffering in cloud exposure and soil-solution Ca: Allevels (Joslin &, Wolfe, 1992); examination of N-

395

this species (Robarge, 1989). These concentrationsare substantially lower than the levels reported byMcNulty et al. (1996) in response to soil-h’ additionsof up to 31.4 kg N ha-’ yr-‘, which raised foliar Nconcentrations from appros. 8 mg g-l to 15 mg g-lX. The study site in this case was a high-elevationsite with relatively low previous N deposition levels(5.4 kg ha-’ yr-‘). The absence of high foliar N levels(Friedland et al., 1988) in the presence of high Klevels in soil solutions, combined with a relativelygreater input of N from atmospheric sources,indicates that N uptake efficiency by roots, regardlessof the form of N, may be rather low at these sites.Total foliar N uptake is also reduced by loss of treebasal area and reduced growth of remaining trees atmany of these high-elevation spruce-fir sites(McLaughlin et al., 1987). This raises the possibilitythat the efficiency of root uptake of N undet theseconditions may be reduced by N accumulation, Almobilization, and associated effects on Ca uptakeand growth of new absorptive roots. Rennenberg etal. (1998) found that NO, uptake in both spruce andbeech forests could be completely inhibited by highN loads and attributed the effect to an accumulationof organic amino compounds translocated fromshoots to roots. This accumulation would probablyalso be enhanced by processes that limit root growthand metabolite use.

Episodic fluctuations in soil solution’ Al levels inN-saturated forests are strongly related to SO,, andX0, concentration in the upper soil horizons(Johnson et al. , 1991). As a consequence, Alconcentrations and particularly Ca:AI ratios in thesoil solutions of such forests can frequently reachlevels (.\I at >, 100 mm 1-i and Ca : Al < 1 .O) that canaffect root function (Fig. 11). Calculated Ca:Alratios in the soil solution at this high elevation sprucesite dropped to approx. 0.25 during high Al intervalsand averaged 0.36 over the three-year study interval.Because periods of high Al and NO, concentration insoil solutions often occur together and at times whenCa:.-\l,, ratios are lowest, root uptake potential inthese N-saturated soils would be out of phase with Navailability. Thus, for N uptake from soils with lowbase saturation, the mobilization of Al by stronganion inputs means that biological availability of Nwould be lowest at the time that soil solution N levelsare highest, thereby reducing N uptake and con-tributing to N accumulation. An additional, po-tentially important, component of reduced N uptakepotential is reduced fine-root production and theshift in root distribution to shallower zones in thesoil profile.

Yet, for many forests, current atmospheric sourcesof N may provide a fertilizer response. For esample,under relatively low inputs of afmospheric K innorthern Sweden, relatively young forests on morerecently glaciated soils have a long history of positivegrowth responses to N fertilization and appear, in

general, to be far from N saturation (Binkley &Hogberg, 1997). Foliar Ca levels do not indicateobvious Ca deficiency. By contrast, forests along thesouthern coast, which receive the highest levels of Nand S, are showing increasing signs of N saturation(Binkley & Hogberg, 1997). Fertilization esperi-ments in southern Sweden provide evidence ofpositive growth responses to added Ca, which hasbeen shown to increase the effectiveness of added Nin stimulating growth (Nohrstedt et al. , 1993).Under these conditions, observed responses to Nfertilization can still be positive even at base:Alratios < 1 .O in the soil solution (Binkley & Hogberg,1997). Foliar analyses from these studies indicatethat Ca concentrations can be maintained at mod-erate levels (m 3.0 mg g-‘) under Swedish growingconditions. Foliar Ca concentrations have also beenshown to be reduced by N fertilization and signifi-cantly increased (38 %) by irrigation with N-freeliquid fertilizer containing modest Ca concentrations(218 kg ha-‘) (Nilsson & Wiklund, 1995).

The importance of adequate Ca supply for Nuptake has also been demonstrated in a variety ofEuropean liming studies in the field (Huettl, 1989).In addition, reduced N levels in xylem sap ofdeclining Norway spruce trees in Germany havebeen observed in mature trees - a response that waswell correlated with periods of low xylem Caconcentrations and low water potential during thegrowing season (Osonubi et al. , 1988). In thenortheastern USA, Ellsworth and Liu (1994)reported a strong linear relationship between foliarN and foliar Ca (R’ = 0.64) in limed sugar maplestands. In southern Sweden, addition of even smallamounts of Ca (69 kg ha-‘) in combination with150 kg of N ha” and 25 kg ha-’ of P increasedtree growth by 35 yb above the N-only treatment.Foliar N concentrations were increased above con-trols by three times the amount observed in theN-only treatment (Nohrstedt et al., 1993). Suchresponses are logically linked to two physiologicalshifts induced by improved Ca nutrition: increasedfine-root production, and improved root function asCa:AI ratios are increased.. (c) Increased leaching of calcium from foliage.

Leaching of foliar cations by rainfall is a significantcomponent of nutrient cycles of widely divergentforest types. However, i t becomes even moreimportant when rainfall exposures are augmentedwith frequent exposure to clouds, particularly thosein industrial regions that hat.e been chemicallyenriched by strong acids (Lovett et al., 1982). Forestsat mountain sites in the eastern US.4 are exposed tohighly acidic mists for from lo*, of the time in thenorth-east (Miller et al., 1993) to 30 o/0 in the south-east (Saxena & Lin, 1990). The ionic content of theseclouds is typically lo-20 times that of rainfall at thesame sites and, as a result, cloud interception bycanopies accounts for 20-SO*,, of the total SO,

396 S. B. McLaughlin and R. Wimmer

A horizon250 , , , , , , , , , , , , , , , , , , , , , I I I I I I I I I I I

(a)

0.8 ,1,,1,,11,111111111,11,,,,11,111,1-

Ib)

1985 1986 1987 1988

Fig. 11, (a) Temporal variations in concentrations of Al and Ca in A-horizon soil solutions from a high-elevation spruce forest show similar dynamics and chemical dominance by Al, Ivhich can interfere with Cauptake by plant roots. (b) Ca:AI molar ratios remained below the 1.0 level indicative of potential interferenceof Al with Ca uptake by roots. (original data with coefficients of variation of around 33 O0 from Johnson et at.,1991).

depoSited. Enrichment with both SO, and NO,results in the occurrence of occasional mist pHvalues < 3.0 (Lindberg, 1992) and minimum valuesaround pH 2.5 (Saxena & Lin, 1990). The degree ofcation leaching by individual mist etients isinfluenced by mist pH and has been shown to resultin losses of up to 36”/b of the total amount of foliarCa for red Spruce (Joslin et al., 1988).

The loss of Ca from the canopy during rain or mistesposure events represent an intensity change, butone must ask how significant such events can berelative to the larger pools of Ca present in foliage.Experiments with detached twigs indicate that thequantities leached are probably derived from apo-plastic supplies and may sum 15?0 of the totalapoplast ic cat ion-eschange capaci ty at pH 2.1(Turner & Van Broekheusen, 1992). In Norway

spruce, cation depletion by pH 2.5 mist in epidermalcells was not uniform, but localized in the outer walls(i.e. in the direction of foliar contact (Fink, 1991)).Based on the high concentrations of apoplastic Carelative to the small physiologically active com-ponent of leaf Ca, i t has been suggested thatapoplastic Ca reserves in conifers are sufficient tocompensate for low Ca supply from the soil (Gulpenet al., 1995) thereby reducing the likelihood of Cadeficiency directly limiting Ca supply for foresttrees.

Several considerations suggest that maintenanceof Ca supply rate at the canopy and whole tree levelsare important. First, the concentrations of Ca lost infoliar leaching can be several times larger than thechanges in cytoplasmic Ca concentrations associatedwith physiological and biochemical signaling. Thus

Calcium in terrestrial ecosystem processes 397

the potential for these events adversely to affectcellular metabolism is high. Second, there is nowevidence that important changes in physiologicalfunctions are linked to foliar leaching of Ca. Finally,liming studies have conclusively demonstratedimproved physiological performance of decliningtrees, including improved canopy vigor, withimproved Ca supply.

(d) Physiological indicators of alteredforestfunction.When interdisciplinary teams have investigated theunderlying mechanisms of forest responses in areaswith high levels of acidic deposition (Hutterman,1985 ; Schulze, 1989 ; Eagar & Adams, 1992 ; Johnson& Lindberg, 1992), changes in forest nutrient statusinduced by increased atmospheric deposition ofstrong mineral acids have been consistentlyrecognized as an, important contributing factor to theobserved alteration of soil processes and/or physio-logical responses. From these and many other relatedstudies, increased understanding has been gained ofthe factors controlling nutrient cycling, and inparticular, Ca cycling, in forests. The limitedmobility of Ca, its physiological importance, andaccelerated loss/immobiliz,ation rates with increasesin acidic deposition have led to the suggestion thatreduced Ca availability could play a primary role inmany of the observed physiological responses offorests to acidic deposition (McLaughlin & Kohut,1992).

In evaluating the effects of acidic deposition onforest ecosystem processes, i t has becomeincreasingly apparent that interactions with naturalbiotic and abiotic stresses are a fundamental com-ponent of the expression of forest responses topollution stress at al l levels (Manion, 1981;McLaughlin, 1985 ; Fuhrer, 1990; Rehfuess, 1991).Here, we examine four measures of response thathave typically been associated with the appearance ofsymptoms of forest decline in recent decades. Allhave either explicit or potential mechanistic linkagesto the known effects of reduced Ca supply, and, inour opinion, all represent areas that are fertile forfurther research. Each represents a different di-mension of Ca control of whole-plant function that islinked to various aspects of Ca metabolism.

An increase in occurrence of winter damage tocurrent-year foliage has been an early and significantvisual indicator of stress to red spruce at highelevations in the north-eastern USA (Friedland etal., 1984), and has also been an important theme ofdiverse research efforts aimed at determining themechanisms responsible (DeHayes; 1992). In theearly stages of recognition of red spruce decline,damage and loss of current-year foliage was found tobe closely associated temporally and spatially withthe incidence of widespread radial growth declines athigh-elevation sites (Johnson et al., 1988). Increasedfrequency of terminal leader death was also closelyassociated with radial growth decline (LeBlanc &

Raynal, 1990). The associated growth decline wasnot attributable to uniquely low temperatures, butrather resulted from an apparent increase in sen-sitivity of red spruce to winter damage at normalwinter temperature (Cook & Johnson, 1989).

A subsequent series of laboratory experimentsindicated that foliar exposure to acidic rain and mistwas a key component of increased sensi t ivi ty.Simulated acid mist-exposure of foliage increasedion leakage from needles and increased the sensitivityof foliage to cold damage (Fowler et al., 1989;Cape et al., 1991). Field studies with mist-exclusionchambers were also used to evaluate the effects ofambient mist/rain on both cold damage and ionleakage of both branches of mature trees (Vann et al.,1992) in the north-east and seedlings in the south-east (DeHayes et al., 1991; Thornton et al., 1994).The results of laboratory and field studies wereessentially the same - acid mist-exposure increasedsensitivity of red spruce foliage to cold damage.Insights into mechanisms for this predispositionhave become increasingly apparent as new tech-niques for detecting changes in membrane-asso-ciated Ca have been developed. In initial experi-ments with this technique, DeHayes et al. (1997)found that changes in membrane-associated Ca wereassociated with red spruce hardening-dehardeningcycles during successive low-high temperature fluctu-ations, as previously discussed (Fig. 3). Similarly,changes in cytoplasmic free Ca have been found toplay a role in cold-shock effects on Arabidopsis, asignal that provides the plant with a protective‘memory’ against damage from future cold events(Knight et al., 1996).

.

The l inkages of changes in the membrane-associated Ca signal to acidic deposition were furtherclarified in subsequent experiments by D. H.DeHayes (pers. comm.) that documented a reductionin membrane-associated Ca and a loss in coldhardiness for red spruce exposed to mists acidifiedwith H,SO,. The 4-10°C elevation in the temper-ature threshold for cold damage by pH 3.0 mistexposure, was associated with loss of membrane-associated Ca and membrane destabilization. Thephysiological significance of these changes led theauthors to project increases in susceptibility ofmontane red spruce to ‘low temperature and otherstresses that compromise overall forest health’. Thesignificance of a shift in internal status of such asmall component of total leaf Ca suggests that totalleaf Ca values could have little relevance to thephysiological effects of Ca. However, it is apparentthat both intensity factors and capacity factors thatinfluence longer-term supply rates are important.Empirical data on Ca influences on carbohydratemetabolism, as well as liming experiments in thefield, support the importance of both modes ofcontrol.

Foliar cation deficiencies have been an important

Calcium in terrestrial ecosystem processes 399

into the deeper mineral soil, were produced by acidicmist/rain treatments, and this response was onlypartially reversed by fertilization with Ca, Mg, or aCa-Rig combination. Increased needle weight inresponse to Ca fertilization and evidence of com-petit ive interactions between Mg and Ca wereobserved in these esperiments as well as infertilization studies with red spruce saplings (VannIiegroet et al., 1993) and mature trees (Joslin &\!‘olfe, 1994) in the field.

Although physiological responses of red spruce tocontrolled acid mist/ rain’ esposures closely par-alleled physiological changes observed in the field atsites with high frequency of exposure to acidicdeposition, the foliar Ca concentrations at whichresponses were produced in controlled greenhousestudies were approsimately twice the foliar Caconcentrations at which dark respiration wasenhanced in foliage of saplings in prior field studies@IcLaughlin et nl., 1993). This led the authors toconclude that observed physiological responses weresignificantly influenced both by the smaller changesin a physiologically active compartment of foliar Cathat is reduced by acid mist/rain exposure, as well asby total foliar Ca and exchangeable soil supply levels(McLaughlin et nl., 1993). This conclusion iscompatible with that reached by DeHayes et of.(1997) based on direct empirical evidence of theimportance of membrane-associated Ca pools.

The importance of esposure-related changes inphysiologically active Ca may esplain the apparentlack of a strong relationship of Ca levels in soil orfoliage with severity of crokvn deterioration ormortality at a relatively high Ca site, \VhitefaceRiountain, in the north-eastern USA (Johnson et al.,1994~). The low soil Al concentration, and relativelyhigh soil Ca concentration, which increases withelevation at this site, result in Ca:Al ratios in soilsolutions that are only occasionally < 1 .O. Ca uptakeat this site is relatively high (Johnson, 1992) and nodecrease in total foliar Ca (mean concentrations areabout. 2.2 mg g-l) occurs with increasing elevation inspite “of the fact that crown deterioration andmortality increase strongly with elevation (Johnsonet al., 1994c). The LVhiteface site, because of theadvanced state of deterioration of its high-elevationspruce forests, is somewhat difficult to evaluate interms of responses tb nutrient supply. The per-centage of dead and declining red spruce trees at thislocation \vas quite high, with 100 “/b spruce mortalityin 15 no of the plots and > 60:*; foliar loss at 303, ofthe plots (Johnson et al., 1994~). Such opening up ofthe canopy, iccompanying reduction of foliar area ofremaining trees, and the increased return of litter tothe forest floor would be espected to have increasedCa supply to the remaining foliage at this site. Theseconditions, coupled with the high foliar exposure toacidic deposition at this location (bIiller el nl., 1993),may esplain the high variability in foliar Ca

concentrations (ranging from 1-3 mg g-l; Johnson etal., 1994c) at a site with generally higher Ca supplypotential (Fig. 7).

An important component of the decline in healthof red spruce at lyhiteface blountain has been thefrequent occurrence of winter injury and associatedloss of foliage, as already discussed. The repeatedloss of a portion of current foliage from ninter-damaged xhoots can be an important drain on thecarbohydrate-production capacity of affected trees.Physiological studies indicate that reduced photo-synthetic production associated with low foliar Caconcentrations w o u l d esacerbate carbohydratereductions by winter damage (Amundson et. cd.,1992).

Sugar maple decline has also been an importantissue in regional analysis of forest health in KorthAmerica (Bauce & Allen, 1991, RIcLaughlin 8i Percy,1999). Although crown mortality and dieback ofsugar maple had been reported over extensive areasin Canada, particularly Quebec, by 1982 (Roy et al.,1985), a substantial improvement in health of thisspecies across much of Canada has been reported inmost recent surveys (Hall et al., 1997). More detailedspatial and soil chemical analysis of the currentcondition of sugar maple has revealed that crowndieback within Canada is most apparent on soils withlow base saturation and high levels of acidicdeposition (Hall et al., 1997).

hIechanistic evidence linking the severity of mapledecline to acidic soils with low cation-exchangecapacity has become increasingly apparent in recentyears (Adams & Hutchinson, 1992). In the weakenedupper crowns of declining sugar maple, reduced netphoto synthesis (Pn) has been related to reducedfoliar Ca and reduced N (Ellsworth & Liu, 1994). Inaddition, annual basal areagrowth trends (1978-1987vs. 1958-1967) of overstory maple in New York andPennsylvania (USA) were found to be significantlyand positively correlated (P < 0.01) with soil Caconcentrat ions and negatively correlated (P <O.Oj)with soil Al, I-I’ and Fe across a wide range ofgrowth rates (Heisey, 1995). Among foliar nutrients,only foliar Ca (r = 0.76; P < 0.01) and Mg (r =0.64; P ,< 0.05) were significantly related to growthof canopy trees. Interestingly, growth of subcanopytrees declined by an average of 56O/b across all sitesin this study regardless of growth trends of canop}trees, which varied widely (-80 y0 to +98 O/b).h’egative correlations of subcanopy tree growth rates

‘*wth soil chemistry were strongest with the acidiccations Al and Fe. Correlations of all variables wereweaker overall for subcanopy trees, presumablybecause of the strong influences of overstory com-petition on growth of these trees. Under theseconditions, correlations of understory growth withsoil iLIg (r = 0.59; P = 0.05) and Ca (r = 0.54, ns)were positive and stronger that those of N, P or K(the masimum response among these was for soil 1\:

401

0.85 million hectares) by a systematic network ofplots in seven south-eastern states. The very highmortality (within 3 yr of infection), lack of re-production and rapid spread of the disease of thisvaluable wildlife forage species led to a series ofexploratory studies that we will summarize herebecause they collectively provide strong inferentialevidence suggesting that reduced Ca supply todogwood would be an important contributing factorto the development of this disease.

First, it should be noted that dogwood is an earlysuccessional, Ca-accumulating species, with foliarCa concentrations approx. six times those of redspruce within the same region. Ca accumulates indogwood berries, and its nutritionally rich twigsmake its fruit a valuable source of Ca for birds and itstwigs important forage for browsers (Britton et al.,1997). The occurrence of dogwood anthracnose hasbeen closely linked to those conditions under whichCa uptake from soils would be limited by reducedtranspiration (shaded locations, wet coves and north-ern slopes where temperatures are lower). Dramaticincreases in dogwood anthracnose occurred during1989, which was an exceptionally wet, cool summerin much of the south-eastern USA.. Plants in sunnylocations showed increased resistance, while thosefrom higher elevations were more sensitive.

Controlled studies suggest that acid deposition canplay a role in increasing the susceptibility of dogwoodto drought stress as well as increasing its inherentsusceptibility to anthracnose (Britton et al., 1997).Exposure to increasingly acidic rainfall within theambient pH range (5.5, 4.5, 3.5, 2.5) increased post-exposure success of Discda infection in experimentsrepeated on two consecutive years. Rapid equi-libration of internal cation status with acidic dropletshas been demonstrated as well (Willey & Hackney,1991), indicating the potential significance ofaccelerated foliar leaching of Ca. Improved pro-tection of dogwood was provided by fungicidalsprays, and fertilization improved growth, butapparently did not affect resistance. Addition ofphosphorous increased sensitivity. Addition of limereportedly increased resistance (USDA, 1991).

In summary, examination of a wide variety ofindicators that characterize variable influences onplant demand relative to Ca supply reveals a patternof response that is consistent with a significant rolefor Ca deficiency in the development and/or ameli-oration of dogwood anthracnose. Based on the

collective symptomology, we suggest that reducedCa supply to dogwood leaves at low soil Caconcentration, low transpirat ion rates, and/oraccelerated leaching of foliar reserves from base-depleted soils reduces the natural resistance of thisCa-requiring species to the infecting Disczrlus fungus.Accelerated foliar leaching of Ca in regions of highatmospheric deposition, particularly in combinationwith conditions that limit foliar uptake of Ca, ma}

also be an important predisposing factor that con-tributes to synchronizing the appearance of foliarstress across diverse sites. We will return later to theissues of justification and research tools to test suchspeculation at a mechanistic level.

In Poland, Witkowski et al. (1987) attributed therapid development of outbreaks of phytophagousand secondary cambio- and xylophagous insectsduring the 1980s to increased susceptibility of spruceand fir trees weakened by regional air pollution.These attacks were considered unusual both in therapidity with which they developed and the fact thattrees in the National forests that were affected hadbeen notably resistant to both harsh environmentalconditions and diseases in the past.

There are interesting parallels between thisEuropean scenario and the development of theBa!sam W’ooley Adelgid (BWA) on firs in the easternUSA (Timmel, 1986). BWA, a cambial feedinginsect with five life stages, can rapidly build lethalpopulations on infected trees. It was introduced intothe northeastern USA in 1907 and BWA developedto epidemic proportions on both balsam fir (Abiesbalasamea) in the northeastern USA and Canada andFraser fir (Abies balasamea) in the southeastern USAin the 1960s (Eagar, 1984). This timing correspondedclosely with the onset of reduced growth of mature,high-elevation red spruce at widespread locations inthe eastern USA (McLaughlin et al., 1987). Con-current declines in radial growth of Fraser fir andbalsam fir were also noted in the Southeastern USA(Adams et al., 1985). Fraser fir has been particularlysusceptible to BWA, and a 1988 survey reportedmortality of 44-91 loO’ for trees > 12.5 cm dbh overthe approx. 26000 ha range of this species in theAppalachian Mountains in the southeastern USA(Dull et al., 1988).

Although the potential role of air pollution inpredisposing Fraser fir to BWA was recognizedapprox. 15 yr ago (Hain & Arthur, 1985), there hasbeen little work on specific mechanisms of resistancesince BW.4 resistance of adult trees was apparentonly in limited populations, notably on Mt Rogers atthe northern extremity of the range of Fraser fir(Eagar, 1984). Viewed from the perspectives of therole of Ca in disease resistance, evidence of changesin Ca availability to Fraser fir throughout the region,and Ca-related physiological changes to fir docu-mented at field test sites, we believe that several linesof evidence support a more comprehensive evalu-ation of the potential role of nutrient deficiency inaccelerating BWA development in this low-Caenvironment.

First, the increased formation of lignin-reinforcedcompression wood along the bole is now recognizedas a defense mechanism for balsam fir (Timmel,1986) as well as Fraser fir following BWA attack(Hollingsworth i% Hain, 1991). Increased hgninformation is a well-recognized mechanism for in-

403

of increased soil acidification followed by diminishedcation availability in response to continued strong,anion inputs.

The downward shift in wood Ca at some forestsites in industrialized regions in the past 2-3 decadeshas often been associated with reduced growth offorest t rees (Bauch, 198.3; Baes & lVlcLaughlin,

1984; Bondietti et al. , 1989; Watmaugh, 1997);however, there are additional implications of suchchanges. They relate to potential alteration in thestructural characteristics of wood formed duringperiods of reduced Ca supply. Because Ca is sostrongly involved in the wood-formation process,and in crosslinking within wood structure,reductions in wood Ca have the potential to affectstructural properties as well as the rate and totalquantity of wood formed. Studies with red spruce(Bondietti et al., 1989), loblolly pine (Pinzrs taeda,Jordan et al., 1990) and yellow poplar (McClenahen& Vimmersted,t, 1993) have shown that temporalvariations in Ca:Al ratios in wood are statisticallyrelated to annual variations in radial increment ofmature trees. Whether this is directly related toreduced Ca availability at the cambium during woodformation, or to effects on carbohydrate availabilityfor growth, is not known.

Wood structural properties depend to a largeextent on the sizes and lengths of xylem. cellsproduced. Controlled studies with smaller trees havedemonstrated a wide range of effects on woodformation and anatomy in response to variations inCa supply rates. In an early study on Ca effects andxylem anatomy of one-year-old loblolly pines(Davis, 1949), cross-sectional areas of Ca-deficienttrees were considerably reduced due to reducedproportions of primary tissue of the cortex and thepith. These differences were explained by adecreased number of cells, and to a lesser extent bydecreased cell size, under reduced Ca supply.Cambial divisions were obviously most affected byCa-deficiency. For deciduous trees, Lamb &Murphey (1968) found that Ca deficiency of silvermaple (Acer saccharinum) seedlings reduced radialgrowth, which was accompanied by increasinglengths of fibers and vessels. Eastern cottonwood(Populus deltoides) responded to increasing Ca withincreases in the percentage of fibers and vessels(Foulger et al. , 1971), while fiber and vesseldimensions wgre not sensitive to a change. Para-doxically, Western larch (Larix occidentalis) seed-lings produced under low soil Ca levels produced8y0 more tracheids per unit area, resin canals thatwere + 10 y0 larger and, most significantly, tracheidlengths that increased by + 16 y0 (Brady, 1969).

Wood density as an indicator for the total amountof the wall material may increase or decrease inresponse to altered nutrient supply, and effects of Caon wood density have been inconsistent. Brady(1969) reported a slight decrease of larch wood

density under low Ca supply while Murphey andMcAdoo (1969) reported increased density inRobinia pseudacacia. The effectiveness of wood inresisting any particular form of applied force is anadditional, more-relevant measure that considerstissue proportions and integrity of lignified cellsrather than pure wall material percentages. Onemeasure of the capacity of a material to resist stressis given by the specific stress index (Panshin &DeZeeuw, 1980). Ca deficiency can cause substantialchanges in the mechanical properties of wood (Table3). These changes include reduced strength per unitmass under tension, reduced resistance to rupture,and reduced elasticity, all of which effectively weakenthe wood in response to mechanical stress.

Ca can also influence the properties of supporttissues, including causing loss of structural integrityof stems of soybean (Glycine max; Albrecht & Davis,1929) and increased brittleness and hardness of pinestems (Davis, 1949). In an exploratory test onindividual spruce tracheids, Wimmer & McLaughlin(1996) showed that lignin, Ca and wood mechanicalproperties (Wimmer et al., 1997) are related to eachanother. Where low Ca in the xylem tissue was foundto produce lower lignin proportions, the hardnessand elasticity of secondary wall layers and cellcorners were also altered (Wimmer & Lucas, 1997).

Inhibition of lignification of wood is a potentiallyimportant consequence of low Ca availability in veryacidic soils (Eklund & Eliasson, 1990; Eklund, 1991).Reduced lignification would be more likely in thecanopies of larger trees because of both increasedtransport distances (resistances) and competition forCa within the sap stream. Reduction in lignin withinthe crown is of interest because there lignin is widelyinvolved with branch support and in maintaining themechanical stability of the tree. Maximum lignincontents are typically found slightly below the baseof the crown (Jansons, 1966). From the perspectiveof wood structure, reduced lignin content in the cellwalls could lead to a more fragile and brittle xylemwood, thereby increasing the susceptibility of treesto breakage during high mechanical stress periodsinduced by ice or strong wind. Reduced foliar lignincontents have, in fact, been reported in the uppercanopies of oak-maple and pine forests in theNortheastern USA (Wessman et al., 1989). LOWl ignin contents, which could be detected fromspectral data by remote imagery, were well related tolow foliar N concentrations and high Nmineralization rates in associated soils. In a relatedstudy with red spruce examined at multiple sitesacross the same region, reduced canopy lignin wasalso found to be significantly related to increasedannual N deposition, increased soil N, and reducedCa and >Ig in foliage and in the forest floor (McNultyet al., 1991). Foliar Ca was strongly related to bothforest floor Ca (r = 0.73; P < O.Ol), and to foliarlignin (I = 0.66) suggesting that base cation de-

Calcium in terrestrial ecosystem processes 4 0 5

( + 2 12 kg ha-’ vs. 56 kg ha-’ for SAW) 15 yr after theharvests, inclusion of the Ca removed in residue inwhole-tree harvest (approx. 690 kg ha-‘) results in acalculated net loss of 478 kg ha-’ of Ca from thewhole-tree harvest site, while the SAW site shows again of 56 kg ha-’ of Ca, presumably associated withprimary mineral weathering.

An interesting feature of this study was the trendtowards increasing Ca levels in the upper 15 cm ofsoils of all treatments (unharvested (+ 175 SO), SA\V(196 ‘$b) and whole-tree harvest (48 yb)) at the end ofthe 15-yr interval. Increasing C and N levels inupper so i l s of all treatments suggests thatmineralization, uptake and rooting were activeresupply mechanisms to the upper soil profile duringthe 15 yr interval. Despite the increases in apparentCa availability in the shallow soils following whole-tree harvest, relative availability of exchangeable Cawas lower in both A (-56 “/b, P < 0.01) and E(-54%) P < 0.10) soil horizons for whole-treeharvest vs. S4W. The reduced relative availability ofCa with the whole-tree harvest treatment wassupported by foliar Ca concentrations that werelower in whole-tree harvest by an average of 27 9/b forsis species compared, but Ca levels were abovewhole-leaf deficiency levels. Thus, while whole-treeharvest resulted in increased Ca losses from this site,rates of resupply of exchangeable Ca by minerali-zation and uptake from the dolomitic parentmaterial were apparently high enough to resupplysurface soils during the initial 15 yr following thisharvest cycle.

Over longer-term cycles, repeated harvesting canalso be evaluated as a soil-acidifying processassociated with the net removal of cation bases fromthe soil (Ulrich R: Matzner, 1986). The cationremoval rate for conventional bole harvest in aGerman Norway spruce forest was estimated to beequivalent to addition of 0.5-1.0 keq ha-’ yr-’ ofacidity, and approximately to balance naturalmineralization rates: For spruce and beech forests,branch removal increased base cation losses by 100 “/hand 20 9,; over bole-only removal and represented asignificant increase in soil acidification, particularlywhen coupled with high rates of acidic deposition(Ulrich & Matzner, 1986). Knowledge of the relativeimportance of weathering, leaching and uptake ratesby vegetation coupled with estimates of the pools infoliage and forest floor can provide forest managersimportant insiihts into long-term sustainability ofnutrient cycles with available harvesting options.Those choices appear to be particularly importantfor Ca because relatively rapid cycling times andpotential system losses following harvest are im-portant processes in uptake and retention of Ca.

(b) Mana&ngforest nutrient sllppl~. Understandingof the physiological implications of base cationdeficiency has been strongly reinforced by experiencewith liming and fertilization experiments in Central

Europe. Review of the role of natural and inducednutrient deficiencies and the effects of liming andfertilization as remediation tools (Zoettl & Huettl,1986; Huettl, 1989; Kreutzer, 1995) has providedmany insights into the physiological basis of theeffects of Ca on forest growth as well as importantinsights into effects on soil chemistry and litterdecomposition processes. Such studies have alsobeen important in evaluating interactions among Caand other nutrients, most notably N. Huettl (1989)summarized the results of numerous fertilization andliming experiments designed to remedy nutrientimbalances due to ‘accelerated soil acidification andincreased nutrient leaching from the canopy.. . thatdemonstrated a fast and sustained revitalization ofdeclining forest ecosystems’. Among the evidence ofrecovery following liming was increased nutrientuptake, including N, increased vigor of forestcanopies, improved growth, resistance to frostdamage, and enhanced production of fine roots.Changes in rooting patterns associated with dis-tribution of Ca to deeper soils have been reported asroots follow liming effects into deeper soil profiles(Huettl, 1989). Such changes are of particularinterest in trying to understand the effects of limingon both forest nutrient uptake and growth.

Perhaps most interesting are the effects of limingon N uptake and metabolism and litter decompo-sition rates (Kreutzer, 1995). Heavy liming (30 tha-‘) has been shown to reduce N,O emissions to theatmosphere by 75 “/b from N-saturated soil in thehigh deposition (N and S) Solling area in Germany(Brumme & Beese, 1992). Five years after limeapplication, the pH of the humus at this site wasraised from 4.5 and 6.5, and CO, emissions wereincreased by 31 ?A, suggesting that microbial de-composition rates and, perhaps, root growth, hadincreased. However, soil solution nitrate concen-trations were not increased by liming in this study,and the authors suggest that a shift to greater N,:N,O in the emission gases may have occurred,indicating liming-induced alteration of N metab-olism. High N,O emissions from N-saturated forestsoils are considered a consequence of reducedefficiency of N utilization associated with chronichigh inputs of atmospheric N (Goulding et al.,1998). With liming at modest levels, increased Nuptake and retention are also likely as the zone ofrooting activity changes and total root mass isincreased.

In one of the few long-term liming studiesdesigned to ameliorate the effects of acidic depositionin the USA, liming was found to increase signi-ficantly basal area growth and improve the crowncondition of overstory sugar maple in Pennsylvania(Long et al., 1997). Foliar Ca concentrationsdoubled, Mg quadrupled, and seed yields wereincreased in these tests with dolomitic limestoneaddition at 22 Mg ha-‘. Interestingly, growth

Calcium in terrestrial ecosystem processes

V I . C0NCLUS10N

1. Whole-tree perspectives

In this article we have sequentially reviewed somefundamental aspects of Ca regulation of basicphysiological processes at cellular , organ andorganismic levels. In addition, we have examined thebiogeochemical basis of Ca effects on some ecosystemprocesses. Wherever possible we have sought torelate mechanisms of action to biological control ofprocesses at the whole-plant and whole-system level.As one moves up through successive organizationalscales, the diversity of influences of multiplenutrients and multiple chemical and biologicals t resses on plant and ecosystem funct ion areincreasingly apparent and conclusions regardingspecific influences of any single influence are lessdirect. However, Ca occupies a unique physiologicalniche in the diversity of its influences on plantfunction and, for this reason, inferences regardingCa regulation of forest processes can be examinedfrom many dimensions.

At the cellular level, the interconnected apoplasticnetwork of cell walls and cell membranes sur-rounding cytoplasm that functions at three orders ofmagnitude lower Ca levels has some interestinganalogies to the neurological system of animals.Electrochemical signals sent through this system inmultiple directions appear to play a role in local andmore remote defense against pathogens and physicaland chemical environmental stresses, such as leafcation exchange, low temperatures and cell wall

407

damage. Environmental signals, that change thetiming and distribution of Ca fluxes and pool sizes inthe plant play an important role in biologicalresponses at multiple levels and some of these aresummarized in Table 4. It can be seen from this tablethat the internal signals that provide biochemicalcontrol, typically in the order of a few mg l-i, arerelatively small compared to chemical signals pro-vided by foliar leaching from the atmosphere and thetemporal variations in Ca concentrations suppliedfrom the soil solution through xylem transport.

I t is not now clear how plants integrate theinfluences of variable Ca supply imposed by themore stressful and more variable soil chemicalenvironment encountered under increasingly acidicsoil conditions into the control of physiologicalprocesses. I t is now apparent, however, thatincreased foliar leaching of Ca can have adversephysiological effects that are not readily reversed orcompensated by local (membrane) resupply. Manyliming experiments now verify the positive effects ofincreased soil Ca supply rates on improved canopyfunction and structure, indicating that increasing Casupply rates can reduce or eliminate symptoms ofphysiological stress in the upper canopy.

Important roles for Ca have now been identified inthe following areas: the structure and function ofmembrane systems that regulate nutrient flowthrough roots and both water and solute fluxesthrough leaf membranes; stomata1 control of waterflux from foliage; activation of enzyme systems thatare involved in cell wall synthesis and in providingphysical and chemical defense against injury and

Table 4. Comparative strengths of jluxes and signals involved in calcium supply and regulation of somephysiological and ecological processes in terrestrial vegetation

Compartment Content Signal* Species/condi t ions Reference

Rainfal l 440 pg 1-t 440 pg 1-l Deciduous, USA/TN Johnson et al. (1985)Cytoplasm 0.04400 pg 1-r General range Bush (1995)

20 pg I-’ Stomata1 guard cel l nIcAinsh et nf. (1996)Knight et al. (1996)

Cell walls’;.‘membranespg I-’ Arnbidopsis cold shock

40 mg 1-l 38 mg 1-t Red spruce seasonal DeHayes et al. (1997)Foliage 1.5-2.5 mg g-’ Red spruce field Robarge et al. (1989)

> 3.0 mg g-r Fraser fir fieldGeneral forests Cape et nl. (1990)

Lechate net flux 800 pg I-’ Red spruce pH 3.0 DeHayes et al. (1998)100 pg I-’ Red spruce pH 5.0 DeHayes et al. (1998)

Xylem sap 48 mg g-l Healthy Norway spruce Osonubi et nl. (1988)36 mg I-’ Declining Norway spruce Osonubi et 01.‘(1988)

Xylem cell walli ’ 0.5-l .O mg g-’ Red spruce Bondetti et nl. (1990)Root cortex 0.5 mg g-r Norway spruce Godbold et al. (1988)

0.1-l .6 mg g-’ 0.01-0.06 mg g-’ Beech cross section Huh et nl. (1992)Soil solution 2.4 mg 1-l 0.4-24 mg 1-r Beech (Soiling, Germany) hlatzner 8; Prenzel

(1992)10 cm depth

4 mg I-’ 3-15 mg 1-l Spruce (Soiling; Germany) llIatzner 8r Prenzel(1992)

0.40-2.6 mg I-’ 0.4-6.0 mg 1-l Spruce USA/SA Johnson et ul. (1991)

*Signal is defined as a change in content (an estimate of typical levels from cited sources) produced by changingenvironmental and/or physiological conditions.

Calcium in terrestrial ecosystem processes 4 0 9

would expect such modifications to be more evidentin later successional stages of forest developmentwhen increasing amounts of system Ca are tied up inwoody stems.

Studies of the effects of acidic deposition onforested watersheds have added significantly tocurrent understanding of the importance of reducedCa supply to forest system structure and function.Among the observed responses are reduced size anddistribution of root systems; increased mortalityrates of mature trees from amplifying naturalstresses, such as winter injury; and loss of upper-canopy integrity, Additional studies in Europeindicate that, with continued Ca loss, reduced flow ofCa through food chains can adversely affect thereproductive success and skeletal integrity of somebird species (Graveland, 1996). Interestingly, birds,like plants, have no capacity to store Ca and thus Casupply at the time of nesting is critical to eggshellintegrity and nesting success.

The results of liming studies indicate that in-creasing Ca supply in Ca-depleted systems can havevery positive effects on the vitality of both soil andvegetation. Included among the positive effects areimproved fine-root production and root distribution,an important factor in responses of forests todrought, nutrient uptake and belowground com-petition. However, it is apparent that additionalresearch is needed on liming application rates andparticle sizes to optimize Ca supply with time andsoil depth. Over-liming and use of strong anioncarriers can have negative effects on decompositionand root distribution that must be recognized andavoided. In many types of forests, a significantfraction of total available Ca is contained in above-ground biomass, making management of forest

, residues a particularly critical issue to maintaininglonger-term Ca supply.

Interrelationships between Ca and N can beespected to become increasingly important in futureforest growth and nutrient cycling. In Europe,patterns of increasing basal area growth in manyareas have followed the widespread foliar sympto-mology and evidence of declining forest vigorapparent in the late 1970s and early 1980s(Schadauer, 1996; Sterba, 1996; Pretzsch, 1996). Incontrast to improving growth at lower elevations,reduced height and volume growth trends havecontinued in the mountains of Central Europe(Pretzsch, 1996). Several factors must be consideredin relating growth rates in the 1980s to previous andpotential future growth. These include improvingclimatic conditions in the 1980s following the verydry 197Os, implementation of improved silviculturaltechniques over the past 30 yr, and increasing Nsupply from atmospheric deposition, which has hadits most positive effects on stands with lower initialN levels (Schadauer, 1996). Another biological factoris the improved genetics of managed-stand and

natural selection for stress resistance within un-managed stands. Superimposed on this has been thesignificant reduction in pollution stress from re-ducing SO,, deposition in the last two decades. Sulfuremission decreased between 198 1 and 199 1, amount-ing to 67% in western Germany and 35 y& ineastern Germany, which began its reduction pro-gram later in the decade (Meesenburg et al., 1995).European reductions in, SO, and base cations havebeen 2976 and 39%, respectively, for The Nether-lands, and 63 y. and 27 y. in Sweden (Hedin et al.,1994). In north-west Germany, where deposition ofSO,, has decreased by SO-70%, N deposition hasremained at about the same levels, and Ca depositionhas decreased. The net effects of changes in these.multiple factors on soils at Solling, a site heavilyimpacted by deposition in the past, has been a5 S-60 :J, reduction in exchangeable base cationsduring the past two decades (Wesselink et al., 1995).

For many reasons, it is difficult to make confidentprojections of future forest growth patterns inEurope (Sterba, 1996). However, from the mech-anistic approach we have presented here, it is usefulto consider how whole-tree nutrient supply patternsfor Ca might be expected to respond to the shifts ingrowth and deposition patterns discussed. In brief,canopy processes, both Ca absorption from de-position and foliar Ca leaching from acidic de-position, should be significantly reduced in im-portance due to reduced emissions of S. Reduced Sdeposition would also be expected to reduce .A1concentrations in soil solutions, a response that hasbeen observed at the Solling Forest (Meesenburg etal., 1995). As a consequence, competitive inter-ference of Al with Ca uptake would be expected to bereduced. However, reduced atmospheric inputs ofCa, which in some areas are comparable to reducedSO, deposition, can be ‘expected to diminish theeffectiveness of reduced S inputs on Ca cycling(Likens et al., 1996). By contrast, the continued highN deposition rates, which in north-west Germanysignificantly exceed forest uptake, can be expected tocontinue to mobilize and deplete Ca and Mg byleaching and accelerated sequestration in growth. N-fertilization experiments in Sweden suggest thatincreased forest growth in response to high N loadingoccurs at the expense of reducing carbohydrateallocation to root systems (Persson & Majdi, 1995).At some point , reduced Ca supply would beanticipated to limit growth and N uptake in soilswith lower base saturation and to accelerate processesthat lead to N saturation.

In Sweden, results of mapping base cation: Alratios in soils and critical loads of acid depositionsuggest that over 800 o of forests currently receiveacidic deposition in excess of present critical loads,and ‘growth reductions have been predicted(Sverdrup kt al., 1992). However, Binkley andHogberg (1997) have cautioned against applying

.

an improved capacity to use this knowledge toprotect and enhance forest growth and forest eco-system integrity.

ACKNOWLEDGEbIENTS

The authors wish to thank the US Forest Service and theUS Environmental Protection Agency for research sup-port. S. B. M. thanks S. G. Hildebrand for grantingsabbatical leave enabling the preparation of this review; D.J. Johnson, T. J. Tschaplinski and R. V. O’Neill for theirearly technical assessment; and Karen Wilson for literatureretrieval. R. W. received support through the AustrianProgramme for Advanced Research and Technology(APART) of the Austrian Academy of Sciences. Thismanuscript is Publication No. 4886 of the EnvironmentalSciences Division, Oak Ridge National Laboratory.ORNL is operated by Lockheed Martin Energy Researchfor the US Department of Energy under contract No. DE-AC05960R22464.

R E F E R E N C E S

Aber JD, Nadelhoffer KJ, Steudler P, Melillo JM. 1989.Nitrogen saturation in northern forest ecosystems. BioScience39: 378-386.

Adams CM, Hutchinson TC. 1992. Fine-root growth andchemical composition in declining Central Ontario sugar maplestands. Catradian Journal of Forest Research 22: 1489-1503.

Adams HS, Stevenson SL, Blasing TJ, Duvick DN. 1985.Growth trend declines of spruce and fir in mid-Appalachiansubalpine forest. Environmental and Experimental Botany 25:315-325.

Alban, DH, Perala DA, Schlaegel BE. 1978. Biomass andnutrient distribution in aspen, pine, and spruce stands on thesame soil type in Minnesota. Canadian ~oournal of ForestResearch 8: 290-299.

Albrecht WA, Davis DE. 1929. Physiological importance ofcalcium in legume inoculation. Botanical Gazette 88: 310-321.

Amundson RG, Hadley JL, Fincher JF, Fellows S, AlscherRG. 1992. Comparisons of seasonal changes in photosyntheticcapacity, pigments, and carbohydrates of healthy sapling andmature red spruce and of declining and healthy red spruce.Canadian Jorrrnal of Forest Research 22: 1605-l 616.

Andersen CP, McLaughlin SB, Roy WK. 1991. A comparisonof seasonal patterns of photosynthate production and use inbranches of red spruce saplings at two elevations. CanadiatrJournal of Forest Research 21: 455-461.

Axelsson B. 1985. Biomass dynamics in the nutrition experimentat St&an. &rr,ml Royal Swedish Academy of Agriculture andForestry 17: 30-39.

Baes CF III, McLaughlin SB. 1984. Trace elements in treerings: Evidence of recent and historical air pollution. Science224: 49-t-497.

Blarney FPC, Edmeades DC, Wheeler DM. 1990. Role of rootcation-eschange capacity in differential aluminum tolerance of

Baes CF III, McLaughlin SB 1986. Multielemental analysis oftree rings: A survey of coniferous trees in the Great SmokyMountains National Park. Technical Report ORNL-6155. OakRidge National Laboratory, Oak Ridge, TN.

Bangerth F. 1979. Calcium-related physiological disorders ofplants. Anmml Review of Phytopathology 17: 97-122.

Bauce E, Allen DC. 1991. Etiology of a sugar maple decline.Canadian Journal of Forest Research 21: 686-693.

Bauch J. 1983. Biological alterations in the stem and root of firand spruce due to pollution influence. In: Ulrich B, Pankrath J,eds. Accctmulation of Air Polhctants in Forest Ecosystems.Dordrecht, The Netherlands: D Reidel, 145-160.

Binkley D, Hogberg P. 1997. Does atmospheric deposition ofnitrogen threaten Swedish forests? Forest Ecology and Man-agement 92: 119-152.

411

Bockheim JG, Langley-Turnbaugh S. 1997. Biogeochemicalcycling in coniferous ecosystems on different aged marineterraces in coastal Oregon. Journal of Environmental Quality 26:292-301,

Bondietti EA, Baes CF III, McLaughlin SB. 1989. Radialtrends in cation ratios in tree rings as indicators of the impact ofatmospheric deposition on forests. Canadian Journal of ForestResearch 19: 586-594.

Bondietti EA, McLaughlin SB. 1992. Evidence of historicalinfluence of acidic deposition on wood and soil chemistry. In :Johnson DfV:, Lindberg SE, eds. Atmospheric depositiotr audforest mrtrient cycling. New York: Springer-Verlag, 358-377.

Bondietti EA, Momoshima N, Shortle WC, Smith KT. 1990.A historical perspective on divalent cation trends in red sprucestemwood and the hypothetical relationship to acidic de-position. Canadiau Journal of Forest Research 20: 18X-1858.

Boxman AW, Cobben PLW, Roelofs JGM. 1994. Does(K+iXlg t Ca+ P) fertilization lead to recovery of tree health ina nitrogen stressed Quercus rtrbra L. stand? EnvironmentalPollution 85: 297-303.

Brady SL. 1969. The effect of mineral nutrients on the anatomy ofLarix occideutalis (Nutt.) seedlings. Thesis, Pennsylvania StateUniversity.

Britton KO, Berrang PC, Mavity E. 1997. Soil effects mediateinteraction of dogwood anthracnose and acidic precipitation.In: Mickler RA, Fox S, eds. The procilrctivity and sustaitrabilit4of southertl forest ecosystems in a changing environment. SewYork: Springer-Verlag, 635-646

Bruck RI, Robarge WP. 1988. Change in forest structure in theboreal montane ecosystem of Mount i\Iitchell, North Carolina.European Journal of Forest Patho1og.v 18: 357-366.

Bruijnzeel LA, Waterloo MJ, Proctor J, Kuiters AT,Kotterink B. 1993. Hydrological observations in montane rainforests on Gunung Silam, Sabah, Malaysia, with specialreference to the ‘Massenerhebung’ effect. Journal of Ecolog>81: 145-167.

Brumme R, Beese F. 1992. Effects of liming and nitrogenfertilization on emissions of CO, and N,O from a temporateforest. Journal of Geophysical Research 97: 12,851-l 2,858.

Brummel D, MacLachlan .1989. Calcium antagonist interfereswith ausin-regulated xyloglucan glycosyltransferase levels inpea membranes. Biochemistry and Biophysics 1014: 298-30-l.

Bucciarelli B, Ostry ME, Anderson NA, Vance CP. 1993.Activity of phenylalanine ammonia-lyase, O-methyl trans-ferase, and cinnamyl alcohol dehydrogenase in Poptrl~utremuloides resistant and susceptible to Hypoxylon mammatwn.Phytopatholog-v 83 : 1415.

Burslem DFRP, Grubb PJ, Turner, IM. 1995. Responses tonutrient addit ion among shade-tolerant tree seedlings oflowland tropical rain forest in Singapore.Jotwnnl ojEco1og.v 83:113-122.

Bush DS. 1995. Calcium regulation in plant cells and its role insignaling. Annual Review of Plant Physiology aud PlantMolecular Biology 46: 95-l 22.

Cape JN, Freer-Smith PH, Paterson, IS, Parkinson JA,Wolfenden J. 1990. The nutritional status of Picea abies (L.)Karst. across Europe, and implications for ‘forest decline’.Trees 4: 21 l-224.

Cape JN, Leith ID, Fowler D, Murray MB, Sheppard LJ,Eamus D, Wilson, RHF..1991. Sulphate and ammonium inmist impair the frost hardening of red spruce seedlings. .VeccPhytologist 118: 119-126.

Catesson AM. 1990. Cambial cytology and biochemistry. In:lqubal M, ed. The vascular cambium. England: ResearchStudies Press, 63-l 12.

Chapin FS III. 1980. The mineral nutrition of nild plants.Awrunl Review of Ecologv aud Systematics 11: 233-260.

Christianson MN, Foy CD. 1979. Fate and function of calciumin tissue. Commwrications in Soil Science and Plant Analysis 10:427442.

Clarkson DT. 1984. Calcium transport between tissues and itsdistribution in the plant. Plant, Cell and Environment 7:449-456.

Choong MF, Lucas PW, Ong JSY, Pereira B, Tan HTW,Turner IM. 1992. Leaf fracture toughness and sclerophyllv:their correlations and ecological implications. New I’/~~folog~sf121: 597-610.

Lotus species. Journal of Plant Nutrition 13: 729-744.

Calcium in terrestrial ecosystem processes 413

and forest decline. Proceeding National Academy of Sciences 85:3888-3892.

Godbold DL, Kettner C. 1991. Use of root elongation studies todetermine aluminum and lead toxicity in Picea abies seedlings.Journal of Plant Physiology 138: 23 l-235.

Goransson A, Eldhuset TD. 1995. Effects of aluminium ions onuptake of calcium, magnesium and nitrogen in Betcda pendalaseedlings growing at high and low nutrient supply rates. Water,Air, and Soil Pollntion 83: 351-361.

Gossett DR, Egli DB, Leggett JE. 1977. The influence of calciumdeficiency on the translocation of photosynthetically fixed “ Cin soybeans. Plant and Soil 48: 243-25 1.

Goulding KWT, Bailey NJ, Bradbury NJ, Hargreaves P,Howe M, Murphy DV, Poulton PR, Willison TW. 1998.Nitrogen deposition and its contribution to nitrogen cyclingand associated soil processes. New Phytologist 139: 49-58.

Gourlay I. 1995. Calcium oxalate crystals in African Acaciaspecies and their analysis by scanning proton microprobe. UfYJownal of Tropical Ecology 11: 12 l-l 40.

Graveland J. 1996. Asian eggshell formation in calcium-rich andcalcium-poor habitats: Importance of snail shells and anthro-pogenic calcium sources. Canadian Jownal of Zoologv 74:1035-lo&.

Grubb PJ. 1977. Control of forest growth and distribution on wettropical mountains: with special reference to mineral nutrition.Annctal Re&ec of Ecology and Systetnatics 8: 83-107.

Gulpen M, Turk S, Fink S. 1995. Ca nutrition of conifers.Zeltschrift &rr PJ?anzenermahrang and Bodenkande 158 :519-527.

Guy CL. 1990. Cold acclimitation and freezing tolerance: role ofprotein metabolism. Annaal RezGezo of Plant Physiology 41:187-223.

Guyette RP, Cutter BE. 1995. Water-use efficiency as monitoredby dendrochemistry. In: Lewis TE, ed. Tree rings as indicatorsof ecosystetn health. Boca Raton: CRC Press, Inc., 95-122.

Hain FP, Arthur FH. 1985. The role of atmospheric depositionin the latitudinal variation of Fraser fir mortality caused by thebalsam woolly adelgid, Adelges piceae (Raw) (Hemipt.:Adelgidae): A hypothesis. Zeitschrift Angezcandl Entotnologia99: l-C5-152.

Hall JP, Bowers W, Hirvonen H, Hogan G, Foster N,Morrison I, Percy K, Cox R, Arp P. 1997.1997 Canadian acidrain assessment, vol. 4, The Effects on Canada’s Forests,Saturn1 Resources Canada, 47 pp.

Hancock JG, Huisman OC. 1981. Nutrient movement in host-pathogen systems. Attnt~al Recietc of Phytopatholog~. 19 :309-331.

Hanger BC. 1979. The movement of calci,um in p lan ts .Comttrwicatinns in Soil Science and Plant Analysis 10: 171-193.

Haynes RJ. 1980. Ion exchange properties of roots and ionicinteractions within the root apoplasm: Their role in ionaccumulation by plants. The Botanical Reziecc 46: 7t99

Hecht-Buchholz C. 1979. Calcium deficiency and plant ultra-structure. Cotntnwricotiotrs in Soil and Plant Science Analysis10:‘67-81.

Hedin LO, Granat L, Likens GE, Bulshand TA, GallowayJN, Butler TJ, Rodhe H. 1994. Steep declines in atmosphericbase cations in regions of Europe and North America. Nature367: 351-354.

Heisey RM. 1995. Growth trends and nutritional status of sugarmaple stands on the Appalachian plateau of Pennsylvania, U. S.A. Water, Air, and Soil Polltrtion 82: 675-693.

Hepler PK, Wayne RO. 1985. Calcium and plant development.Annaal Rerr’eic of Plant Physiology 36: 397439.

Hollingsworth RG, Hain FP. 1991. Balsam woolly adelgid(homoptera: adelgidae) and spruce-fir decline in the southernAppalachians: Assessing pest relevance in a damaged eco-system. Florida Entomologist 74: 179-187.

Huettl RF. 1989. Liming and fertilization as mitigation tools indeclining forest ecosystems. Water, Air, and Soil Polltction 44:93-l 18.

Huh M, Bengtsson B, Larsson NP, Yang C. 1992. Particleinduced X-ray emission microanalysis of root samples frombeech (Fcguu sylratica). Scanning of Microscopy 6: 58 I-590.

Huttermann A. 1985. The effects of acid deposition on thephysiology of the forest ecosystem. Experietrtia 41: 58t-590.

Jansons N. 1966. Lignin content in various cardinal directions of

spruce annual increments. The University of British Columbia,Faculty of Forestry, Canada, Translation No. 39.

Jenny H. 1980. The soil resowce: origin and beha&w. Springer-Verlag, New York.

Johnson AH, Andersen SB, Siccama TG. 1994b. Acid rain andsoils of the Adirondacks. I. Changes in pH and availablecalcium, 1930-t 984. Canadian Jownal of Forest Research 24:39-45.

Johnson AH, Cook ER, Siccama TG. 1988. Climate and redspruce growth and decline in the northern Appalachians.Proceedings National Academy of Sciences, USA 85 : 5369-5373.

Johnson AH, Friedland AJ, Miller EK, Siccama TG. 1994a.Acid rain and soils of the .4dirondacks. III. Rates of soilacidification in a montane spruce-fir forest at 1VhitefacehIountain, New York. Canadian~oarnal of Forest Research 24:663-669.

Johnson AH, Schwartzman TN, Battles JJ, Miller R, MillerEK, Friedland AJ, Vann DR. 1994~. Acid rain and soils of theAdirondacks. II. Evaluation of calcium and aluminum as causesof red spruce decline at Whiteface 11Iountain, New York.Canadian Jownal of Forest Research 24: 654-662.

Johnson DW. 1992. Base cation distribution and cycling. In:Johnson DLV:, Lindberg SE. eds. Atmospheric Deposition andForest 1Vatrient Cyclittg. New York, US.4: Springer-Verlag.275-332.

Johnson DW, Fernandez IJ. 1992. Soil-mediated effects ofatmospheric deposition on eastern U. S. spruce-fir forests. In:Eagar C, Adams MB, eds. Ecology and decline of red sprrrce in theEastern United States. New York. US.4: Springer-Verlag.

Johnson DW, Kelly JM, Swank WT, Cole DW, Van MiegroetH, Hornbeck JW, Pierce RS, Van Lear D. 1988. The effectsof leaching and whole-tree harvesting on cation budgets ofseveral forests. Joarnal of Enrirontnental Qaality 17: 418424.

Johnson DW, Lindberg SE. 1992. Atmospheric deposition andforest mrtrient cycling. New Tork: Springer-Verlag.

Johnson DW, Richter DD, Lovett GM, Lindberg SE. 1985.The effects of atmospheric deposition on potassium, calcium,and magnesium cycling in two deciduous forests. CanadianJownal of Forest Research 15: 773-782.

Johnson DW, Todd DE. 1987. Nutrient export by leaching andwhole-tree harvesting in a loblolly pine and mixed oak forest.Plant and Soil 102: 99-109.

Johnson DW, Todd DE. 1998. Harvesting effects on long-termchanges in nutrient pools of mised oak forest.Journal of the SoilScience Society of America 62: 1725-1735.

Johnson DW, Van Miegroet H, Lindberg SE, Todd DE,Harrison RB. 1991. Nutrient cycling in red spruce forests ofthe Great Smoky ilIountains. Canadian Jaarnal of ForestResearch 21: 769-787.

Jones RGW, Lunt OR. 1967. The function of calcium in plants.Botanical Reriezc 33: 407-426.

Jordan CF, Herrera R. 1981. Tropical rain forests: Are nutrientsreally critical ? American h’atttralis 117: 167-180.

Jordan DN, Wright LM, Lockaby BG. 1990. Relationshipbetween xylem trace metals and radial growth of loblolly pine inrural Alabama. Journal of Ewirotrtnental Qaality 19: SOtSO8.

Joslin JD, Kelly JM, Van Miegroet H. 1992. Soil chemistry andnutrition of Xorth American spruce-fir stands: Evidence forrecent change. Jownnl of Etr~irottttrental Qnalit? 21: 12-30.

Joslin JD, McDuffie CM, Brewer PF. 1988. Acidic cloudwaterand cation loss from red spruce foliage. Water, Air, and SoilPollation 39: 355-363.

Joslin JD, Wolfe MH. 1992. Red spruce soil solution chemistqand root distribution across a cloud water deposition gradient.Canadian Jorwttal of Forest Research 22: 893-904.

Joslin JD, Wolfe MH. 1994. Foliar deficiencies of maturesouthern Appalachian red spruce determined from ferrilizationtrials. Soil Science Society of Atnerica Journal 58: 1572-l 579.

Kauss H. 1987. Some aspects of calcium-dependent regulation inplant metabolism. Antnral Review of Plant Physiology 38:47-72.

Kelly JM, Mays PA. 1989. Root zone physical and chemicalcharacteristics in southeastern spruce-fir stands. Soil ScienceSociety J7oarnal 53: 1248-1255.

Kirkby EA, Pilbeam DJ. 1984. Calcium as a plant nutrient.Plant, Cell and Etwirotrtnent 7: 397-405.

Knight H, Trewavas AJ, Knight MR. 1996. Cold calcium

415

t rends in atmospheric deposi t ion and seepage output innorthwest German forest ecosystems. Water, Air, and SoilPollution 85: 611-616.

Meyer J, Schneider BU, Werk K, Oren R, Schulze ED. 1988.Performance of two Picea abies stands at different stages ofdecline. V. Root tip and ectomycorrhizae development andtheir relations to above ground and soil nutrients. Oecologia 77:7-13.

Miller EK, Panek JA, Friedland A], Kadlecek J, MohnenVA. 1993. Atmospheric deposition to a high-elevation forest atCvhiteface hIountain, New York, USA. Tellus 45B: 209-227.

Miller HG. 1995. The influence of stand development on nutrientdemand, growth and allocation. Plant and Soil 168-169:225-232.

Minorsky PV. 1985. An heuristic hypothesis of chilling injury inplants: A role for calcium as the primary physiologicaltransducer of injury. Plant, Cell and Environment 8: 75-94.

Mitterhuber E, Pfanz H, Kaiser WM. 1989. Leaching ofsolutes by the action of acidic rain: A comparison of efflux fromtwigs and single needles of Picea abies (L. Karst.). Plant, Celland Environment 12: 93-100.

.Momoshima N, Bondietti EA. 1990. Cation binding in wood:

Applications to understanding historical changes in divalentcation availability to red spruce. Canadian Journal of ForestResearch 20: 18X)-18+9.

Mueller-Dombois D. 1983. Canopy dieback and successionalprocesses in Pacific forests. Pacific Science 37: 317.

Murphey WK, McAdoo JG. 1969. Effects of nutrient deficiencyon tensile properties of the xylem of Robinia pseudacaria L.seedlings. Forest Science 15: 251-253.

National Park Service. 1995. Summary of Forest Insect DiseaseImpacts, Great Smoky iXIountains National Park. (hnonymousreport).

Nilsson LW, Wiklund K. 1995. Nutrient balance and P, K, Ca,hIg. S and B accumulation in a Norway spruce stand followingammonium sulphate application, fertigation, irrigation droughtand X-free-fertilisation. Plant and Soil 168: 437-446.

Nodvin SC, Van Miegroet H, Lindberg SE, Nicholas NS,Johnson DW. 1995. Acidic deposition, ecosystem processes,and ni trogen saturat ion in a high elevation southernAppalachian watershed. Water, Air, and Soil Pollution 85:1647-1652.

Nohrstedt HO, Sikstriim U, Ring E. 1993. Experiments withvitality fertilisation in Norway spruce stands in southernSweden. Vista, Sweden: SkogForsk Report 2/93,

Odum HT. 1970. Rain forest structure and mineral-cyclinghomeostasis. In: Odum MT, Piegeon RF, eds. A tropical rain

a study of irradiation aud ecology at El Verde, Puerto Rico.Oak Ridge, TN, USA: U. S. Atomic Energy Commission,Division of Technical Information.

Oleksyn J, Karolewski P, Giertych MJ, Werner A, TjoelkerMC, Reich RB. 1996. Altered root growth and plant chemistryof Piers syle.estris seedlings subjected to alumintim in nutrientsolution. Treds 10: 135-144.

O’Neil,,,M, Albersheim P, Darvill A. 1990. The pecticpolysaccharides of primary walls. Methods of Plant Biochemistry2: 415-l-11.

Oren, R, Schulze ED, Werk KS, Meyer J, Schneider BU,Heilmeier H. 1988. Performance of two Picea abies (L.) Karst.stands at different stages of decline. I. Carbon relations andstand growth. Oecologia 75: 2.5-37.

Osonubi 0, Oren R, Werk KS, Schulze ED, Heilmer H. 1988.Performance of two Picea abies stands at different stages ofdecline. IV. $ylem sap concentrations of magnesium, calcium,potassium, and nitrogen. Oecologia 77: 1-6.

Ostry ME. 1985. Susceptibility of Populus species and hybrids todisease in the North Central United States. Plant Disease 69:755-757.

Ostry M. 1994. Poplar disease research: Host resistance andpathogen variability. Plant Science 18: 89-94.

Overrein LN, 1972. Sulphur pol lut ion pat terns observed;leaching of calcium in forest soil determined. Ambio 1: 145-147.

Oyonarte C, Perez-Pujalte A, Delgado G, Delgado R,Almendros G. 1994. Factors affecting soil organic matterturnover in a iXIediterranean ecosystems from Sierra de Gador(Spain): An analytical approach. Commanicatio~rs in Soil Scienceaad Aualysis 25: 1929-1945.

Pal RN, Rai VK, Laloraya MM. 1973. Calcium in relation to

respiratory activity of peanut and linsied plants. Biochemie andPhysiologic der Pflanzen (BPP) 164: 258-265.

Palta JP, Li PH. 1978. Cell membrane properties in relationshipto freezing injury. In: Li PH, Sakai .4, eds. Plant cold hardinessandfreering stress: mechanisms and crop implications. New York,USA: Academic Press, 93-107.

Panel C, Castillo FJ, Kiefer S, Greppin H. 1985. The regulationof plant peroxidases by calcium. In: Trewavas AJ, ed. Molecularand cellular aspects of calcium in plant development. N.4TOScientific Affairs Division, 365-366.

Panshin AJ, DeZeeuw C. 1980. Textbook of wood technology, 4thedn. New York, USA: iXIcGraw-Hill.

Parchomchuk P, Neilsen GH, Hogue EJ. 1993. Effect of dripfert i l ization of ammonium nitrate and ammonium poly-phosphate on soil pH and cation leaching. Canadian $xrrnal ofSoil Science 73 : 157-l 64.

Persson H, Majdi H. 1995. Effects of acid deposition on treeroots in Swedish forest stands. Tl’ater, Air, and Soil Pollution85: 1287-1292.

Pitelka LF, Raynal DJ. 1989. Forest decline and acidicdeposition. Ecology 70: 2-10.

Poovaiah BW. 1988. Calcium and senescence. Chapter 11. In:Nooden LD, Leopold AC, eds. Senescence and agirzg in plauts.New York, US.4: Academic Press.

Pop VV, Postolache T, Vasu A, Craciun C. 1992. Calcophilousearthworm activity in soil; an experimental approach. SoilBiology and Biochemistry 24:1+83-1490.

Potter CS, Ragsdale HL, Swank WT. 1991. Atmosphericdeposition and foliar leaching in a regenerating southernAppalachian forest canopy. JorrrNal of Ecology 79: 97-l 15.

Pretzsch H. 1996. Growth trends of forests in southern Germany.Berlin, Germany: Springer-Verlag.

Prinz B. 1987. Causes of forest damage in Europe. i%Iajorhypotheses and factors. Environment 29: 10-37.

Raese JT, Drake SR. 1996. Yield increased and fruit disordersdecreased with repeated annual calcium sprays on ‘Anjour’pears. Journal Tree Fruit Production 1: 51-59.

RagsdaIe HL, Lindberg SE, Lovett GM, Schaefer DA. 1992.Atmospheric deposition and throughfall fluxes of base cations.In: Johnson DW, Lindberg SE, eds. Atmospheric deposition artdforest nutrient cycling. New York, US.4: Springer-Verlag,235-252.

Raich JW, Rilery RH, Vitousek PM. 1994. Use of root ingrowthcores to assess nutrient limitations in forest ecosystems.Cauadian j%tcr~~al of Forest Research 24: 2135-2138.

Ranson SL. 1965. The plant acids. Chapter 20. In: Bonner J,Varner JE, eds. Plant biochemistry. Kew York, USA: AcademicPress.

Rehfuess KE. 1981. On the impact of acid precipitation inforested ecosystems. ForstzcissenschaftlLhes Centralblatt 100:363.

Rehfuess KE. 1991. Review of forest decline research activitiesand results in the Federal Republic of Germany. Journal ofEwirownental Science and Health A26: 415-445.

Reich PB, Oleksyn J, Tjoelker MC. 1994. Relationship ofaluminum and calcium to net CO,exchange among diverseScats pine provenances under pollution stress in Poland.Oecologia 97: 82-92.

Reich PB, Ellsworth DS, Uhl C. 1995. Leaf carbon and nutrientassimilation and conservation in species of differing successionalstatus in an oligotrophic Amazonian forest. Functional Ecology9: 65-76.

Reiners WA. 1992. Twenty years of ecosystem reorganizationfollowing experimental deforestation and regrowth suppression.Ecological Monographs 62’: 503-523.

Rennenberg H, Kreutzer K, Papen H, Weber P. 1998.Consequences of high loads of nitrogen for spruce (Picea abies)and beech (Fagus sylvatica) forests. iyew Phytologist 139: 71-86.

Reuss JO, Johnson DW. 1986. Acid deposi t ion and theacidification of streams and waters. Xew York, US.4: Springer-Vertag.

Richter DD, Markewitz D, Wells CC, Allen HL, April R,Heine PR, Urrego B. 1994. Soil chemical change during threedecades in an old-field loblolly pine (Pinus taeda L.) ecosystem.Ecology 75: 1+63-l-+73.

Robarge, WP, Pye JM, Bruck RI. 1989. Foliar elementalcomposition of spruce-fir in the southern blue ridge province.Plant attd Soil 114: 19-34.

4 1 7

Vitousek PM, Gosz JR, Crier CC, Melilo JM, Reiners WA,Todd RL. 1979. Kitrate losses from disturbed ecosystems.Science 204: 469-474.

Vitousek PM, Fahey T, Johnson DW, Swift MJ. 1988. Elementinteractions in forest ecosystems: Succession, allometry andinput-output budgets. Biogeochemistry 5: 7-34.

Waide RB, Zimmerman JK, Scatena FN. 1998. Controls ofprimary productivity: Lessons from the Luquillo hIountains inPuerto Rico. Ecology 79: 3 l-37.

Waring RH. 1987. Characteristics of trees predisposed to die.BioScience 37: 569-57-l.

Waring PE, Patrick J. 1975. Source-sink relations and thepartitioning of assimilates in the plant. In: Cooper JJP, ed.Photosynthesis and productivity in dtifferent environments. NewYork. USA: Cambridge Universitv Press. 481499.

Watmaugh SA. 1997. Ai evaluation bf the u$e of dendrochemicalanalyses in environmental monitoring. Environmental Reviews.3: 181-201.

Webb AR, McAinsh MR, Taylor JE, Hetherington AM. 1996.Calcium ions as intracellular second messengers in higherplants. Adz*ances in Botanical Research 22:

Wellwood RW, Sastry CBR, Micko MM, Paszner L. 1974. Onsome possible specific gravity, holo- and (a)-cellulose, tracheidweight/length and cellulose crystallinity relationship in a SOO-year-old Douglas-fir treg. Ho/zforschwzg 28: 91-94.

Wesselink LG, Meiwes KJ, Matzner E, Stein A. 1995. Long-term changes in water and soil chemistry in spruce and beechforests, Soiling, Germany. Environmental Scieme and Tech-nology 29: 51-58.

Wessman CA, Aber JD, Peterson DL. 1989. An evaluation ofimaging spectrometry for estimating forest canopy chemistry.International Journal of Remote Sensing 10: 1293-1316.

Westermark U. 1982. Calcium promoted phenolic coupling bysuperoxide radical - a possible reaction in wood. Wood Scienceand Technology 16: 71-78.

Westermark U, Hardell H-L, Iversen T. 1986. The content ofprotein and pectin in the lignitied middle lamella/primary wallfrom spruce fibers. Holzforschtrng 40: 65-68.

Whitehead D, Jarvis PG. 1981. Cohducting sapwood area,foliage area and permeability in mature trees of Picea sitchensisand Pinus contorta. Canadian -7ournal of Forest Research 15:940-947.

Willey JD, Hackney JH. 1991. Chemical interactions betweenacid rain and dogwood leaves. TheJournal of the Elisha MitchellScientific Society 107: 83-88.

Wimmer R, McLaughlin SB. 1996. Possible relationshipsbetween chemistry and mechanical properties in the micro-structure of red spruce xylem. Radiocarbon 659-668.

Wimmer R, Lucas BN. 1997. Comparing mechanical propertiesof secondary wall and cell corner middle lamella in sprucewood. IAWA~octrnal 18: 77-78.

Wimmer R, Lucas Bn, Tsui TY, Oliver WC. 1997. Longi-tudinal hardness and Young’s modulus of spruce tracheidsecondary walls using indentation technique. Wood Science andTechnology 31: 131-141.

Witkowski Z, Madziara-Borusiewicz K, Plonka P, Zurek Z.1987. Insect outbreaks in mountain Kational Parks in Poland -their causes, course and effects. Ekologia Polska 35: 463-492.

Yoder BJ, Ryan MC, Waring RH, Schoettle AW, KaufmannMR. 1994. Evidence of reduced photosynthetic rates in oldtrees. Forest Science 40: 513-527.

Zech W, Schneider BU, Rohle H. 1991. Element compositionof leaves and wood of beech (Fagus sylvatica L.) on SO,-polluted sites of the SE-Bavarian hlountain. Water, Air, andSoil Pollution 54: 97-106.

Zhao XJ, Sucoff E, Stadelmann EJ. 1987. AIs+ and Ca”alteration of membrane permeability of Qttercus rttbra rootcortex cells. Plant Physi&gy 83:1591162.

Zimmerman R. Oren R. Schulze ED, Werk KS. 1988.Performance oi two Picea abies stands ai different stages ofdecline. II. Photosynthesis and leaf conductance. Oecologia 76:513-518.

Zocchi G, Mignani I. 1995. Calcium physiology and metabolismin fruit trees. Acta Hortictrlturae 383: 15-23.

Zoettl HW, Huettl RF. 1986. Nutrient supply and forest declinein southwest Germany. Water, Air, and Soil Pollution 31:449462.