Developmental Morphology and Physiology of Grasses
D. D. Briske
- Architectural Organization
- Root Systems
- Tiller Demography
- Tiller Recruitment
- Tiller Longevity
- Leaf Demography
- Benefits of Vegetative Growth
- Grazing Avoidance
- Mechanical Mechanisms
- Biochemical Mechanisms
- Grazing Tolerance
- Morphological Mechanisms
- Physiological Mechanisms
- Compensatory photosynthesis
- Carbon allocation
- Carbohydrate reserves
- Plant water status
- Root growth and function
- Apical dominance
- Compensatory growth
- Cost of Grazing Resistance
- Competitive Interactions
- Population Structure
List of Figures
The vegetation managed in livestock and wildlife production systems is produced by a series of developmental and physiological processes within individual plants. The series of structural changes displayed by organisms from inception to maturity, including cellular division, differentiation and growth, are collectively referenced as developmental morphology (Esau 1960). The developmental morphology of plants defines their architectural organization, influences their palatability and accessibility to herbivores, and affects their ability to grow following defoliation. Physiological processes establish the capacity for solar energy capture and product synthesis necessary to sustain structural development.
The predominant impact of grazing on plant growth is a reduction in photosynthetic capacity associated with a decrease in leaf area. Species cope with grazing by minimizing the probability of being grazed and/or rapidly replacing leaf area removed by herbivores. Morphological attributes and biochemical compounds influence the probability and severity of grazing by affecting tissue accessibility and palatability. The capacity for rapid leaf replacement is conferred by physiological processes and meristem availability.
The inherent morphological and physiological attributes of individual species influence the structure and function of populations and communities by determining the extent of competitive interactions among plant species. Grazing alters competitive interactions among species by removing various amounts of leaf area and establishing the potential for differential growth rates following similar defoliation severities. Species composition is altered when a particular intensity, frequency and/or seasonality of grazing shifts the competitive advantage from one group of species to another. Species composition changes subsequently influence livestock production and managerial strategies by affecting the quantity, quality and seasonality of plant production. Consequently, the design and evaluation of grazing management strategies must be based in part on the developmental morphology and physiological function of the dominant plant species to conserve rangeland resources and maintain production stability.
The developmental morphology of grasses is remarkably similar among species with only minor morphological variations separating growth-forms. Individual phytomers, which consist of a blade, sheath, node, internode and axillary bud, form the basic unit of growth (Hyder 1972, Briske 1986, Fig. 4.1). The size, number and spatial arrangement of phytomers determines the architectural organization of individual tillers. A tiller consists of a series of phytomers successively differentiated from an apical meristem with the initial phytomer located nearest the soil surface. Individual grass plants are composed of an assemblage of tillers originating from the axillary buds of previous tiller generations.
Initially, the entire leaf primordium is meristematic, but cellular division is quickly restricted to intercalary meristems (meristematic tissue separated from the apical meristem by a region of non-meristematic tissue) (Dahl and Hyder 1977). Intercalary meristems are located in narrow zones at the base of the blade, sheath and internode (Sharman 1945, Etter 1951, Langer 1972). Intercalary meristem activity ceases within the blade when the ligule is formed and within the sheath when the ligule becomes exposed. Consequently, the blade ceases elongation prior to the sheath, while internode elongation is dependent upon species and phenology. The basal location of the intercalary meristem within the blade and sheath explain why leaf elongation can occur following defoliation without replacement of the leaf tip (Hyder 1974).
Floral induction marks the transition of the apical meristem from a vegetative to a reproductive status (Sharman 1945, Etter 1951, Langer 1972). Floral induction occurs in response to a photoperiodic stimulus (i.e., day length) following a sufficient juvenile growth period. At the time of floral induction, both leaf primordia and axillary buds are rapidly differentiated producing a double ridge appearance on the apical meristem. Spiklet primordia differentiate from the axillary buds while further development of leaf primordia is suppressed halting additional vegetative development. Vegetative growth can only occur from immature intercalary meristems of existing phytomers or from previously differentiated axillary buds in reproductive tillers.
Morphological variation of individual tillers is largely a consequence of the number and length of phytomers comprising the tiller. Variation in tiller architecture among tall, mid- and short grasses does not originate from a major deviation in the pattern of developmental morphology, but rather results from a variable number and/or size of phytomers determining cumulative tiller height. Internode elongation increases phytomer size and is most frequently associated with reproductive tiller development, but may also occur in nonreproductive tillers in a small number of species. In culmed vegetative tillers, the apical meristem is elevated above the soil surface by internode elongation while in a vegetative condition. Culm elongation originates from the activity of intercalary meristems located at the base of the several uppermost internodes. In reproductive tillers, the inflorescence and several uppermost leaves are elevated above the soil surface presumably to facilitate wind pollination. The developmental morphology of reproductive tillers is similar to vegetative tillers prior to floral induction (Hyder 1974).
The mechanisms determining whether a tiller, rhizome or stolon develop from an axillary bud are not clearly understood. These shoot-types are initiated from a finite number of axillary buds which may partially explain the observed seasonality of tiller and rhizome recruitment (i.e., rapid tiller recruitment may potentially limit rhizome recruitment). High nitrogen availability, high temperatures and short photoperiods promote tiller development to a greater extent than rhizome development in quackgrass (McIntyre 1967). Bermudagrass stolons exposed to a high ratio of red:far-red radiation displayed an upward curvature, increased leaf and internode elongation and lower carbohydrate concentrations than stolons grown in darkness or in radiation with a low red:far-red ratio (Willemoes et al. 1987). These data suggest that phytochrome (a proteinaceous pigment sensitive to specific wavelengths) may regulate the differentiation of stolons, rhizomes or tillers by affecting the distribution of photosynthetic products within the plant. Radiation quality has also been implicated in the regulation of tiller recruitment in several grasses (Casal et al. 1986, Kasperbauer and Karlen 1986).
Adventitious roots differ from seminal roots in both diameter and mass per unit length. The large diameter of adventitious roots is associated with a greater cross sectional area of xylem to enhance water transport to the shoot system (Wilson et al. 1976). The large root mass per unit surface area may explain why adventitious roots are not initiated until seedlings produce sufficient leaf area and photosynthetic capacity to support their development 1 to 3 weeks following seedling emergence (Wilson and Briske 1979). In the case of vegetative tiller development a seminal root system is not initiated. Juvenile tillers are supported by parental tillers until the adventitious root system develops (Welker et al. 1987). Adventitious root development is initiated at approximately the third to fourth leaf stage in little bluestem (Carman and Briske 1982).
Tillering. The perenniality and sustained productivity of grasses are conferred by the successive production of relatively short-lived tillers (Langer 1963, Tomlinson 1974). Successive tiller recruitment produces a series of connected tiller generations referred to as tiller hierarchies or families (Langer 1963). The number of generations within a hierarchy is determined by the rate of tiller recruitment and tiller longevity as influenced by genetic and environmental constraints. The number of tillers per hierarchy and number of hierarchies per plant define the size and architectural configuration of the plant. With increasing plant size and age, these tiller hierarchies become separated as the initial tiller generations die and decompose (Gatsuk et al. 1980). Each plant fragment is capable of survival and may continue tiller development as previously described. The hollow crown phenomenon characteristic of many perennial grasses is very likely a natural consequence of the developmental morphology of grasses and not a symptom of plant stress (Gatsuk et al. 1980). The disproportionate initiation of tillers on the plant periphery eventually reduces the density of tillers and axillary buds necessary for continuation of tiller recruitment within the plant interior (Butler and Briske 1988, Olson and Richards 1988a).
The number of live tillers per plant or per unit area is determined by the rate and seasonality of tiller recruitment in relation to tiller longevity. Changes in tiller density occur when recruitment lags behind or exceeds mortality (Fig. 4.5). The density of live tillers defines the potential for biomass production, within the constraints of resource availability, by determining the number of intercalary meristems, apical meristems and axillary buds available for growth (Olson and Richards 1986b). The continued existence of perennial grasses is also dependent upon the successive development of axillary buds capable of initiating subsequent tiller generations and perpetuating the plant. If tiller recruitment was suspended for an interval equivalent to the maximum longevity of the most recently developed tillers, the plant would lose all meristematic potential and cease to exist. Growth could potentially resume from axillary buds, but their longevity in relation to that of parental tiller is not known.
Reproductive tiller development terminates the development of leaf primordia and is followed by tiller death as the existing phytomers senesce (Noble et al. 1979, Fig. 4.6). Vegetative tiller mortality, coincident with flowering of parental tillers within the plant, is assumed to be a consequence of the shading of smaller vegetative tillers and a reduction in resource import as culm and inflorescence development increase resource demand within reproductive tillers (Ong 1978). It is the smallest, but not necessarily the youngest, tillers which experience mortality when the plant is stressed. The relationship between recruitment and mortality is generally described as a density-dependent process with either mortality increasing resource availability for the recruitment of new tillers or recruitment contributing to the mortality of existing tillers by resource depletion (Robson 1968).
Leaf longevity also displays seasonal patterns of recruitment and mortality (Vine 1983, Chapman et al. 1983, 1984). Leaves initiated when growing conditions are most favorable, spring and early summer in temperate environments, have shorter longevities than those initiated during periods of less favorable environmental conditions. Leaves of perennial ryegrass and browntop exhibited mean longevities of 60 - 70 days when initiated in the spring and summer compared to 70 - 105 days when initiated in the fall and winter (Chapman et al. 1984). Leaf longevities of less than 90 days during favorable growing conditions indicate that grazing must closely follow leaf initiation to optimize the harvest of live leaves. The environmental conditions experienced by emerging leaves may program their subsequent development to a larger extent than the conditions during growth or maturity.
Synchronous leaf initiation and senescence maintains a relatively constant leaf number per tiller throughout much of their developmental history. Generally, a tiller possesses an emerging leaf, immature leaf, mature leaf and senescing leaf (Anslow 1966, Chapman et al. 1984). The net difference between leaf initiation and senescence represents the number of live leaves per tiller. Leaf demography, by determining the amount of live leaf area per tiller, influences both the potential photosynthetic capacity of the tiller and the amount of leaf biomass available for consumption by herbivores. Leaf and tiller demography collectively determine the rate of biomass turnover (i.e., production versus senescence) within the community.
Grasses exhibit vegetative growth or reproduction by the successive recruitment of tillers from previous generations. Each tiller establishes a shoot and root system (adventitious) to acquire resources from the environment. Vegetative growth may confer several ecological advantages originating from resource allocation between and among tillers within individual plants (Pitelka and Ashman 1985). The capacity for continuous tiller replacement and site occupation based upon parental support of juvenile tillers is perhaps the most significant ecological benefit. Resource allocation from parental to juvenile tillers confers a greater likelihood of establishment and survival in comparison with seedlings which must become established from energy and nutrient reserves available within the endosperm (Tripathi and Harper 1973).
Survival and growth of stressed tillers are also enhanced by resource import from associated non-stressed tillers within a plant (Gifford and Marshall 1973, Ong and Marshall 1979, Welker et al. 1987). Root growth, as estimated by both total length and numbers of roots, of recently recruited juvenile tillers progressively decreased as the juvenile tiller, the parental tiller or the parental tiller and all remaining tillers within little bluestem plants were defoliated (Carman and Briske 1982). Greater growth rates were observed in tall fescue plants following defoliation as the percentage of undefoliated tillers within the plant increased (Matches 1966). Both observations support the conclusion that defoliated tillers were deriving support from associated, nondefoliated tillers.
Vegetative growth theoretically confers plants with potential immortality. Individual plants of hard fescue have been estimated to be greater than 1000 years old (Harberd 1962). However, the few age estimates available for North American perennial grasses indicate that their life spans are relatively short. Estimates of individual plant longevities on the Jornada Experimental Range in New Mexico, including tobosa grass, black grama and red threeawn, indicate that maximum plant longevity does not exceed 30 years (Wright and Van Dyne 1976). These estimates of plant longevity are corroborated by the work of West et. al.,(1979) on the U.S. Sheep Station in Idaho and Canfield (1957) on the Santa Rita Experimental Range in Arizona.
Grazing resistance is an ambiguous term used to describe the relative ability of plants to survive grazing. However, strategies to cope with grazing vary greatly in form and expression among plant species. Additional insight can be gained by organizing grazing resistance into a tolerance and avoidance component (Stuart-Hill and Mentis 1982, Briske 1986, Fig. 4.7). Avoidance mechanisms reduce the probability and severity of plant defoliation (i.e., escape mechanisms), while tolerance mechanisms facilitate growth following defoliation (i.e., mechanisms of rapid leaf replacement). The ability of a species to survive grazing undoubtedly results from a combination of these two components, but in certain species and under specific environmental conditions, one component may predominate over another.
Grazing resistance within ecological plants groups may be generally ranked as follows: herbaceous monocots > herbaceous dicots > deciduous shrubs and trees > evergreen shrubs and trees (Archer and Tieszen 1986). This ranking is based upon both morphological and physiological considerations. Apical and intercalary meristems within monocots are less vulnerable to large herbivores because of their basal location within the plant. Meristems are located at terminal and lateral positions of shoots in dicots increasing their susceptibility to large herbivores. Many woody plants, evergreen species most notably, possess slow growth rates and low rates of resource acquisition which require that individual leaves be retained for long periods (Chapin 1980). Consequently, these species are less efficient in replacing photosynthetic surfaces removed by herbivores and frequently rely on avoidance mechanisms rather than tolerance mechanisms to cope with grazing.
Mechanical Mechanisms. Avoidance mechanisms primarily influence plant accessibility and palatability to specific herbivores. At the individual tiller level of organization, the probability and severity of defoliation may be reduced by a number of avoidance mechanisms originating from a variety of morphological parameters (Fig 4.8). Tissue accessibility is primarily a function of tissue proximity to the soil surface as influenced by the length and angle of leaves and tillers. Species possessing culmed vegetative shoots are especially susceptible to defoliation because apical meristems are elevated above the soil surface and readily accessible to herbivores (Branson 1953, Rechenthin 1956, Booysen et al. 1963). Mechanical deterrents including spines, awns and epidermal characteristics (e.g., silica bodies, pubescence and cuticular waxes) directly influence palatability (McNaughton et al. 1985, Cooper and Owen-Smith 1986, Young 1987). Leaf anatomy, primarily the presence of vascular bundles associated with the C4 photosynthetic pathway (i.e., Kranz leaf anatomy), has been demonstrated to influence species selection in herbivorous insects (Caswell et al. 1973). Greater cell wall thickening in the bundle sheath cells (specific mesophyll cells) of warm-season grasses limits digestibility by impeding microbial access to cellular contents within the digestive tract (Akin and Burdick 1977, see Chapter 2).
Avoidance mechanisms may also originate at the plant level of organization from reduced tissue accessibility or by interspecific association. At the level of the individual plant, tissue accessibility is primarily a function of tiller height, number of culmed tillers and the amount of dead material which has accumulated within the plant. Grazing intensity was inversely related to the basal area of individual crested wheatgrass plants (Norton and Johnson 1983). Grazing intensity in large plants (basal area > 200 cm2) was reduced, relative to their canopy volume, by the accumulation of dead culms and litter. Grazing intensity became proportional to canopy volume when this material was removed. Shrub species may develop a "hedged" appearance with frequent grazing through the initiation of numerous shoots which form a mechanical barrier protecting leaves and meristems within the canopy.
The association of palatable species with less palatable species may also influence the frequency and intensity of plant defoliation (McNaughton 1978). The protection afforded to grasses growing within the canopy of low growing shrubs serves as a frequently observed example (Davis and Bonham 1979). Although, not well documented, there is no reason to suspect that this phenomena does not occur among herbaceous plant species as well. This implies that not only the relative amount, but also the spatial distribution of herbaceous species may be an important parameter regulating plant utilization by herbivores.
Optimal defense theory indicates that the most apparent plants within a community rely on quantitative defenses because they are easily located by herbivores (Rhodes 1979, Provenza and Malechek 1983). Less apparent plants frequently possess the less costly qualitative compounds since they have a lower probability of being grazed. The observation that approximately 80% of woody perennials contain tannins in comparison with only 15% of the herbaceous dicots serves to substantiate this point (Rhodes and Cates 1976).
Grazing avoidance mechanisms do not necessarily remain constant, but may increase with an increase in grazing intensity. The qualitative biochemical compounds which increase in response to grazing, referred to as inducible defenses, have previously been referenced (Rhodes 1985, Young 1987). Similarly, grazing can modify morphological parameters influencing avoidance mechanisms. Long-term grazing has been observed to function as a selection pressure against the tall upright growth-form in several perennial grasses (Detling and Painter 1983, Carman and Briske 1985). The remaining growth-forms, characterized by a large number of smaller statured tillers with reduced leaf numbers and blade areas, are better able to avoid grazing because less biomass is removed by herbivores and a greater number of meristems may escape grazing to facilitate growth following defoliation. Contrastingly, plants characterized by a small number of large tillers with large leaf areas are more competitive in environments with dense canopies. It is currently uncertain whether this herbivore-induced shift in growth-form is a result of genotypic selection or phenotypic plasticity.
Morphological Mechanisms. Leaf replacement potential, defined as the rate at which leaf area is reestablished following defoliation, is largely a function of the number, source and location of meristems within a plant following defoliation (Fig 4.9). Growth will occur most rapidly from intercalary meristems, followed by newly developed leaf primordia, and least rapidly from newly initiated axillary buds (Cook and Stoddart 1953, Hyder 1972, Briske 1986). Growth from intercalary meristems results from the expansion of previously differentiated cells, whereas growth from axillary buds is delayed by the time required for differentiation and growth of leaf primordia. However, axillary buds ensure perenniality by providing a meristematic source for the production of subsequent tillers in contrast to the limited activity of intercalary meristems (i.e., phytomer growth only). Consequently, species possessing a high ratio of reproductive or culmed vegetative tillers are best suited to intermittent defoliation rather than continuous grazing (Hyder 1974). When the apical meristem assumes reproductive status or is removed by grazing, leaf replacement must originate from axillary buds which require the greatest time interval following defoliation.
Compensatory photosynthesis. Grazing alters the age structure of leaves within plant canopies in addition to reducing total leaf area. This has direct consequences for the photosynthetic capacity of plants because leaves generally exhibit maximum photosynthetic rates at about the time of full expansion and decline thereafter (Caldwell 1984). Consequently, leaves of defoliated plants may display greater rates of photosynthesis than nondefoliated plants because many of the leaves are chronologically younger and more efficient photosynthetically. However, the carbon gain capacity of plants is a function of both total leaf area and photosynthetic rate.
Increased rates of photosynthesis following partial defoliation relative to similar aged leaves of nondefoliated plants is referred to as compensatory photosynthesis (Nowak and Caldwell 1984). Compensatory photosynthesis has been observed in a number of species with leaf photosynthetic rates increasing within the range of 15 - 50% in comparison with nondefoliated leaves (Gifford and Marshall 1973, Dyer et al. 1982, Wallace et al. 1984, Nowak and Caldwell 1984). Maximum photosynthetic rates occur several days following defoliation, a portion of which is attributable to decreasing photosynthetic rates of nondefoliated leaves as they age and senesce. Photosynthetic response is influenced by leaf position relative to the location of plant defoliation. Photosynthetic rates of the remaining portions of defoliated leaves generally decrease (Detling et al. 1979, Dyer et al. 1982), while photosynthetic rates of nondefoliated leaves on tillers from which associated leaves have been removed or nondefoliated tillers within plants from which associated tillers have been removed, generally increase (Gifford and Marshall 1973, Dyer et al. 1982). Potential mechanisms contributing to compensatory photosynthesis include increased carboxylase activity, increased leaf conductance to carbon dioxide and a decrease in feedback inhibition resulting from a greater demand for carbon following defoliation (i.e., greater sink strength) (Hodgkinson et al. 1972, Gifford and Marshall 1973, Wallace et al. 1984).
Although compensatory photosynthesis does occur, its significance to grazing tolerance appears limited. Compensatory photosynthesis occurs in both crested and bluebunch wheatgrass, but is most evident in the oldest leaves which comprise a small percentage of the total photosynthetic area (Nowak and Caldwell 1984). The total amount of compensatory photosynthesis resulting from this portion of the canopy is insufficient to explain the difference in grazing tolerance between these two species.
Carbohydrate reserves are utilized for plant growth and maintenance when photosynthetic capacity is limited, as evidenced by the reduction in reserves following defoliation (Deregibus et al. 1982). However, a substantial amount of information has developed suggesting that the contribution of carbohydrate reserves to the leaf replacement potential of perennial grasses may not be as large as previously assumed (May 1960, Ryle and Powell 1975, Atkinson and Farrar 1983, Caldwell 1984). Investigations utilizing labelled carbon indicate that carbohydrates allocated to the root system are not capable of being remobilized for subsequent use above-ground following defoliation (Davidson and Milthorpe 1966).
The amount of reserve carbon is not directly related to leaf replacement potential in crested and bluebunch wheatgrass and does not account for the wide variation in grazing tolerance between them (Richards and Caldwell 1985). Carbohydrate pools (tissue mass x carbohydrate concentration) within the crowns of the two wheatgrass species only contain an amount of carbohydrate equivalent to that produced in approximately 3 days of photosynthesis. Consequently, plant growth is more dependent upon current photosynthesis than stored carbon within 3 days of defoliation. In addition, rates of leaf elongation or tillering are not dependent upon carbohydrate concentrations in leaves, crowns or roots of tall fescue, orchardgrass or canarygrass (Sambo 1983, Zarrough et al. 1984, Volenec and Nelson 1984). These findings support the view of May (1960), who indicated that use of the term reserve evokes a false conception of their contribution to growth following defoliation (Deregibus et al. 1982).
The magnitude of carbohydrate reserves necessary to ensure plant survival and maintain maximum leaf replacement potential has not been established for individual species or species groups. Carbohydrate concentrations of 1 - 6% have been suggested as minimum reserve levels in grasses, but these estimates are far from conclusive (Caldwell 1984). In addition, total carbohydrate pools must be quantified by determining both carbohydrate concentrations and weight of the associated plant organ(s) (Richards and Caldwell 1985). Defoliation could potentially reduce total carbohydrate availability within a plant by reducing the weight of crown tissue without necessarily altering carbohydrate concentration.
It has been suggested that plant growth may be limited to a greater extent by the availability or activation of axillary buds than the amount of reserve carbon (Watson and Casper 1984, Richards and Caldwell 1985). A meristematic growth limitation would promote the accumulation of carbon reserves because photosynthesis would exceed growth and respiration requirements (White 1973, Caldwell 1984). Although insufficient carbon undoubtedly limits plant growth, a direct relationship between the amount of carbohydrate reserves and plant growth has not been established.
Inconsistencies associated with analytical techniques for carbohydrate extraction have also curtailed research progress. Estimates of nonstructural carbohydrates derived by acid extraction may be two to three times greater than for boiling water extraction because a greater amount of structural carbohydrates are apparently digested (Richards and Caldwell 1985). Compounds other than nonstructural carbohydrates (e.g., proteins, hemicellulose and organic acids), which are not currently evaluated, may also contribute energy and organic intermediates to plant growth following defoliation. Additional investigation of the analytical techniques and reserve compounds within the context of whole plant carbon balance is required to define the relationship of carbohydrate reserves to grazing tolerance.
Root growth cessation affects both lateral and vertical development of root systems (Schuster 1964, Smoliak et al. 1972) as well as detrimentally influencing root initiation, diameter, branching and total production (Biswell and Weaver 1933, Jameson 1963, Evans 1973, Carman and Briske 1982, Richards 1984). Root mortality has also been observed following defoliation (Weaver and Zink 1946, Hodgkinson and Baas Becking 1977, Troughton 1981). These detrimental responses collectively serve to reduce the total absorptive surfaces and soil volume explored for water and nutrients.
The capacity for nutrient absorption per unit length in temperate, perennial grasses parallels the growth responses following defoliation. Phosphorus absorption, root elongation rate and respiration rate remained suppressed for an 8-day observation period following defoliation of orchardgrass to a height of 2.5 cm (Davidson and Milthorpe 1966). Similar responses of root growth and function to defoliation originate from the dependence of both processes on energy produced in plant respiration (Caldwell et al. 1987). Respiration, in turn, functions upon substrate produced in photosynthesis.
It would appear unduly restrictive to presume that only the removal of apical meristems by grazing could serve as an environmental cue to induce tillering. How would tillering occur in plant populations subjected to limited grazing? Tiller recruitment has been observed to occur in response to grazing even though apical meristems were insufficiently elevated to be removed by livestock (Butler and Briske 1988). Conversely, removal of apical meristems from tillers of crested and bluebunch wheatgrass did not always result in accelerated tiller recruitment (Olson and Richards 1988b, Richards et al. 1988). In contrast to the temperate species, lateral bud growth was stimulated by both apical meristem and canopy removal in three tropical grasses. Expanding leaves and either the inflorescence or elongating culm were observed to be the source of apical dominance in vegetative and reproductive ryegrass tillers, respectively (Laidlaw and Berrie 1974). These conflicting observations attest to the complexity associated with the regulation of tiller recruitment in perennial grasses (Youngner 1972).
Light quality has been implicated in the control of axillary bud expansion in several grass species (Deregibus et al. 1985, Casal et al. 1985, 1986, Kasperbauer and Karlen 1986). This photomorphogenetic response is presumably mediated by phytochrome, as is flowering and branching in many dicotyledonous species. A decrease in the ratio of red:far-red radiation associated with increasing canopy development may signal the diminishing availability of resources and suppress additional tiller recruitment (Deregibus et al. 1985, Simon and Lemaire 1987). Partial removal of the plant canopy by grazing may increase the ratio of red:far-red radiation and promote tillering without disturbing the apical meristem. The versatility of phytochrome as a mechanism regulating tiller recruitment is yet to be proven, but the direct inhibition of axillary buds as presented by Leopold (1949) is overly restrictive.
Grazing does influence both the seasonality and total number of tillers recruited by affecting axillary bud development. Tiller recruitment in ungrazed populations of little bluestem in central Texas is restricted to spring and fall coincident with the bimodal precipitation pattern of the region (Butler and Briske 1988, Briske and Butler 1989). An intermediate severity of grazing extended the period of tiller recruitment throughout the spring and summer. Grazing did not significantly increase total tiller recruitment, however, when the number of recruited tillers were summed over the entire growing season. Similarly, grazing promoted tiller recruitment in the spring in addition to the normal pattern of fall recruitment in crested wheatgrass (Olson and Richards 1988b). Grazing at the time of culm elongation or thereafter reduced subsequent tiller recruitment while grazing prior to culm elongation produced little affect in comparison with ungrazed plants (Olson and Richards 1988c, Busso et al. 1989). The absence of a positive tillering response in a species with the demonstrated grazing tolerance of crested wheatgrass confirms the conclusion of Ellison (1960), that grazing generally inhibits tillering over the long-term. Removal of a large portion of the photosynthetic surfaces from a plant apparently reduces the amount of resources available for tiller growth irrespective of the mechanisms regulating the tillering process (e.g., apical dominance, light quality, etc., Youngner 1972).
Ill-defined terminology has undoubtedly contributed to the conflicting viewpoints on this topic. Compensatory growth may be generally defined as any positive plant response to injury (Belsky 1986). Overcompensation, more specifically, describes an increase in the cumulative total dry weight of grazed plants, including the biomass removed by defoliation, in excess of that produced by ungrazed plants (Belsky 1986). In other words, an increase in the rate of plant growth following defoliation does not in itself constitute greater production. It must also be established if the growth increase is maintained for a sufficient period to offset the reduction in biomass resulting from defoliation.
An additional source of confusion is associated with the potential mechanisms contributing to compensatory growth. Although evidence for the involvement of inherent physiological mechanisms does exist (e.g., compensatory photosynthesis), the modification of competitive interactions among plants following defoliation is probably of far greater consequence. In greenhouse experiments, Belsky (1986) noted that in the few cases where compensatory growth was observed, it occurred when shortgrasses were growing in competition with tallgrass species. Defoliation to a uniform height may have placed the tallgrasses at a competitive disadvantage rather than directly inducing a physiological response which enhanced growth of the shortgrass species. The benefit a plant derives from defoliation of its neighbors coincident with its own defoliation has been termed competitive fitness (Belsky 1986).
The "costs" associated with grazing resistance are most clearly defined in relation to the production of secondary compounds which deter herbivores (i.e., avoidance mechanisms) (Coley 1986). Tree seedlings with high concentrations of tannin were grazed to a lesser extent, but displayed lower rates of leaf production than seedlings with low tannin concentrations. The growth reduction can be interpreted as the cost associated with the production of tannins. A trade-off may also exist between competitive ability and grazing resistance for similar reasons. Plants of white clover possessing cyanogenic compounds were less effective competitors than acyanogenic plants in the absence of grazing because of the energy diverted to grazing avoidance (i.e., cyanogenic compounds) (Dirzo and Harper 1982). In the presence of grazing however, the cyanogenic plants would presumably be better competitors because cyanogenic compounds would reduce the intensity of grazing.
Morphological and physiological resistance mechanisms do not represent such clear costs to the plant. What costs are incurred by the decumbent growth-form of western wheatgrass resulting from long-term grazing (Detling and Painter 1983) or the short-term increase in resource allocation to the shoots of crested wheatgrass following defoliation (Richards 1984)? These mechanisms may not represent a cost in terms of the diversion of previously assimilated energy, but rather a reduction in the potential for subsequent production and resource acquisition. Hyder (1972) has cautioned that placing excessive emphasis on grazing resistant species may decrease productivity. Wise strategies of grazing management must be coupled with inherent grazing resistance mechanisms of existing species to optimize plant and animal productivity in grazed systems.
Grazing-induced modifications in species composition have been documented in numerous grasslands throughout the world (Voigt and Weaver 1951, Branson and Weaver 1953, Ellison 1960, Williams 1969, Noy-Meir et al. 1989). Compositional changes frequently involve the replacement of higher successional species by lower successional species (Canfield 1957). The lower successional species are frequently mid- or shortgrass species held in a subordinate position by competitive interaction with species possessing greater stature (Arnold 1955, Belsky 1986). Grazing reduces the competitive ability of the mid- and tallgrass thereby increasing the relative abundance of lower successional grasses and forbs and establishing the potential for shrub invasion. This scenario of species replacement in response to grazing has frequently been inferred from field observation, but has not been experimentally verified. A scenario incorporating elements of developmental morphology, grazing resistance, competitive interactions and population structure is presented as a mechanistic explanation for species replacement in grasslands. Herbaceous retrogression is emphasized in this section while its implications to woody plant encroachment are discussed in Chapter 5.
Plant species do not grow or respond to grazing as isolated individuals, but rather as members of a population and community. It has previously been demonstrated that individual grass plants consist of an assemblage of phytomers and tillers (Fig. 4.1). Similarly, grass populations reflect the number of plants per unit area and the number of tillers per plant (Fig. 4.10). Grassland communities are further composed of an aggregation of populations variously arranged in terms of abundance and space.
The first, and most direct, mechanism by which grazing alters competitive interactions involves the differential utilization of populations within the community in response to the relative display of avoidance mechanisms. Species grazed severely are placed at a disadvantage when competing with associated species grazed less severely. Production of bluebunch wheatgrass plants subjected to 50% canopy removal just prior to culm elongation was equivalent to the production of nondefoliated plants growing with full competition when associated vegetation within a 90 cm radius of the defoliated plants was clipped at ground level (Fig 4.11). When competition from associated vegetation was removed by tilling within a 90 cm radius, defoliated plants produced three times the biomass of nondefoliated plants growing with full competition. These data clearly demonstrate that the ability of plants to respond to defoliation is not only determined by an inherent suite of morphological and physiological characteristics, but also by competitive pressure from associated species (Caldwell 1984).
The second mechanism by which competitive interactions among plant species may be altered in response to grazing involves the differential ability of species to grow following a similar intensity of defoliation based upon the possession of various tolerance mechanisms. Species rapidly replacing photosynthetic surfaces gain a competitive advantage over associated species that grow more slowly following defoliation. Crested wheatgrass exhibits greater leaf replacement potential than bluebunch wheatgrass following an equivalent intensity of defoliation. This response is due in part to the ability of crested wheatgrass to rapidly initiate a greater number of tillers and to allocate carbon to reestablish photosynthetic surfaces while temporarily decreasing allocation below-ground (Caldwell et al. 1981, Richards 1984). Inequitable responses to defoliation between these two species can be attributed to the differential expression of tolerance mechanisms because similarity in plant architecture minimizes the influence of avoidance mechanisms (Caldwell 1984).
Species grazed less severely (i.e., avoidance mechanisms), capable of growing more rapidly following defoliation (i.e., tolerance mechanisms), or possessing a combination of these two resistance components realize a competitive advantage within the community. These species, through the possession of a greater canopy area, are able to intercept greater amounts of solar energy and assimilate greater amounts of carbon further enhancing their competitive ability. By allocating a greater amount of carbon below-ground, grazing resistant species may more effectively explore the soil profile for water and nutrients and preempt resources that may have been utilized by associated species prior to defoliation (Mueggler 1972, Eissenstat and Caldwell 1988).
Grazing management partially governs the intensity of competitive interactions by regulating the relative frequency and intensity of defoliation among plant species within grassland communities. Lenient grazing may not alter species composition appreciably, even though species may be grazed non-uniformly or respond differentially, because the intensity is insufficient to alter plant growth and subsequent competitive interactions. However, as stocking rate and defoliation intensity increase, differential utilization and growth among species becomes intensified altering competitive interactions and ultimately contributing to species replacement. In addition, the species or combination of herbivore species affect the relative frequency, intensity and seasonality of grazing within communities based on preference and behavioral differences (see Chapter 2 and Chapter 3). Sheep grazing sagebrush steppe, for example, can shift community composition toward grass dominance more rapidly than the total exclusion of grazing (Laycock 1967).
Season of grazing in relation to the progression of phenological development among species plays a major role in determining the outcome of competitive interactions. Species grazed throughout their entire growth period are placed at a competitive disadvantage in the presence of species possessing growth periods which do not coincide entirely with the grazing season. This is very likely the reason Texas Wintergrass, a cool season species, increases in relative abundance in grasslands of central Texas which are grazed most intensively during the spring and summer (Launchbaugh 1955). Similarly, fall grazing has been observed to favor dominance of warm season grasses in relation to shrub species in the cold desert of the western U.S. (West et al. 1979).
Given these considerations, it is to be expected that competitive interactions among species are modified on a relative rather than an absolute basis. Consequently, it is possible for an individual species to decrease in relative abundance in one community, but increase within another in response to grazing. The inherent mechanisms of grazing resistance probably remain constant over the distributional range of a species, but the expression of relative resistance mechanisms change in relation to the resistance mechanisms of associated species. The response of individual species has been observed to vary depending upon the intensity of grazing and topographic position within a mixed-grass prairie community. Relative abundance of western wheatgrass decreased regardless of grazing intensity or topographic position suggesting that it possessed limited grazing resistance relative to the associated species (see Chapter 5, Fig. 5.4). By contrast, the relative grazing resistance of buffalograss varied in relation to topographic position. Relative abundance decreased in the swale, but increased on the ridge. Livestock behavior and environmental variables may further influence competitive interactions among plant species to produce an array of plant responses (see Chapter 3).
Grazing-induced modifications of competitive interactions are eventually expressed at the population level through the modification of basal area and tiller demography of individual plants. A reduction in basal area of individual plants may be the initial and predominate response contributing to the decline of bunchgrass populations in response to grazing (Butler and Briske 1988). Grazed populations of several perennial grasses have been observed to consist of individuals with smaller basal areas in comparison with ungrazed populations (Pond 1960, Hickey 1961). This decrease in individual plant basal area is very likely a consequence of the fragmentation of individual large plants into smaller units (Fig. 4.12). Consequently, plant density may remain constant or even increase while basal area per plant decreases. Further, an increase in tiller number per unit of remaining basal area may initially offset the decrease in total basal area thereby maintaining a constant tiller density. However, with continued severe grazing the decrease in individual plant basal area may become so great that tiller density declines within the population.
This decline in population structure is very likely paralleled by a decrease in resource acquisition within the community. Consequently, the more grazing resistant species within the community utilize a greater proportion of the available resources (Caldwell et al. 1987). The continued existence of populations composed entirely of plants with reduced basal areas may be jeopardized by the inability to effectively compete with populations of less severely grazed species and increased susceptibility to extreme abiotic conditions (e.g., drought or temperature extremes). Although bunchgrass populations composed of numerous, small plants appear very persistent, a large reduction in basal area may predispose the population to elimination from the community.
Species replacement influences the quantity, quality and variability of biomass production by altering the initial harvest and subsequent flow of energy through the ecosystem (see Chapter 1). In many rangeland systems, the ratio of unpalatable to palatable species increases with increasing grazing severity (Noy-Meir and Walker 1986). Although this may not decrease total productivity of the system, it reduces the proportion of energy transferred through the grazing food chain (i.e., plants to herbivores). Consequently, as the proportion of unpalatable species increases a greater proportion of the solar energy captured in photosynthesis is transferred into the decomposer compartment following plant senescence or is incorporated into woody biomass.
Compositional changes involving an increase in the abundance of annual and short-lived perennial species limits the amount of solar energy captured by reducing the proportion of the growing season during which a canopy is present. Additional difficulty is encountered in efficiently harvesting biomass in annual systems based on the large variability and limited duration of the growth period. Finally, it is possible for severe grazing to modify the production potential of a site through soil erosion and alteration of hydrological properties (see Chapter 6). In this case, fewer resources are available to sustain biomass production.
Plant growth originates from the incorporation of photosynthetic products into cells and tissues differentiated from meristems. Tillers represent the sum of all tissues differentiated from a single apical meristem to form the basic unit of growth. The grass plant is composed of an assemblage of tillers initiated from axillary buds of previous tiller generations. Variation in the size and number of phytomers comprising the tiller and the pattern of tiller emergence contribute to the architectural distinction among grass growth-forms (e.g., bunchgrass versus sodgrass).
Grazing resistance can be organized into avoidance and tolerance components. Avoidance mechanisms consist of morphological characteristics or biochemical compounds which reduce the probability and severity of plant defoliation at several levels of vegetation organization. Tolerance mechanisms are conferred by physiological processes at the tiller and plant levels to enhance growth following grazing. Both components contribute to grazing resistance, but the relative magnitude and associated cost of each component are poorly understood.
Grazing management modifies competitive interactions by influencing the frequency, intensity and seasonality of plant defoliation (Fig. 4.13). Plant species grazed less frequently and intensively, or with a greater capacity to grow following defoliation, display greater leaf areas for photosynthesis and attain a competitive advantage. Grazing-induced modifications in competitive interactions are eventually expressed as modifications in plant and population structure. A decrease in total basal area, plant density or tiller density of a given species is ultimately manifested in a relative reduction in resource acquisition within the community. Shifts in species composition subsequently alter the quantity, quality and variability of plant production by modifying the amount and pattern of energy flow through the ecosystem (see Chapter 1).
Vegetation response to grazing may be investigated at one or more levels within the hierarchical organization of grasslands (e.g., tiller, plant, population or community). For example, plant productivity may be reduced by a decrease in individual tiller weight, tiller number per plant, or plant density in response to grazing. Consequently, insight into mechanisms occurring at higher hierarchical levels (e.g., community) requires that processes at lower hierarchical levels (e.g., population, plant and tiller) be investigated (Archer and Tieszen 1986, Brown and Allen 1989, see Chapter 5). Hierarchical levels of vegetation organization may respond in a comparable manner to affect vegetation dynamics, but frequently additional complexity is encountered by the occurrence of opposing responses between or among levels. Grazing has been observed to increase tiller density, but concomitantly decrease individual tiller weight (Jones et al. 1982) or increase plant density while reducing basal area per plant (Butler and Briske 1988). It is essential that several hierarchical levels be considered when evaluating vegetation responses to grazing to avoid incomplete or erroneous conclusions.
Research oriented at the population level of vegetation organization possesses the potential for integrating the divergent sources of information available from individual plant and community studies. These two research perspectives have not been effectively unified into an information base for vegetation management in grazed systems. Community level investigations describe species composition shifts and biomass dynamics, but do not yield insights into mechanistic cause-effect relationships. Conversely, reductionist investigations at the individual plant level yield mechanistic insights, but are frequently too narrow in scope to identify interactions and properties of systems at levels of organization suitable for vegetation management. A major limitation to the extrapolation of plant level studies is the minimal amount of information concerning competitive interactions and population ecology of dominant plant species.
Figure 4.1 The devlopmental morphology of grasses originates from the successive differentiation of phytomers from the apical meristem of individual tillers.
Figure 4.2 Leaf primordia (blade and sheath of individual phytomers) are different from the upper portion of the apical meristem to form a series of primordia at succesive stages of development.
Figure 4.3 Tiller initiation from auxillary buds in the crown of a grass plant.
Figure 4.4 Variation within the grass growth form originates from the pattern of tiller emergence expressed by various species groups.
Figure 4.5 Live tiller density as a consequence of tiller recruitment and mortality within a population.
Figure 4.6 Seasonality of tiller recruitment, flowering, and mortality within a population of tall fescue.
Figure 4.7 Organization of grazing resistance into avoidance and tolerance components.
Figure 4.8 Categories of avoidance mechanism and associated morphological characteristics at the tiller and plant levels of organization that may potentially reduce the probability of being grazed.
Figure 4.9 Sources of meristematic activity in a grass plant.
Figure 4.10 Hierarchical organization of grassland communities is dependent upon the density and spatial arrangements of tillers in individual plants and within and amoung species.
Figure 4.11 Response of bluebunch wheatgrass to three defoliation intensities in the presence of full, partial, or no competition from associated vegetation.
Figure 4.12 Relative response of plants and tillers of increaser and decreaser species to increasing grazing intensity.
Figure 4.13 Proposed cause-effect relationship between grazing-induced competitive interactions and modified plant and population structure as it may regulate species composition at the community level.