Molecular, physiological, and growth responses to sodium stress in C4 grasses from a soil salinity gradient in the Serengeti ecosystem.

The concentration of soil sodium (Na) is an important factor that influences species distribution in the Serengeti short-grass plains, Tanzania. Experiments were conducted to characterize physiological (growth, photosynthetic, nutrients, and water relations) and molecular (heat shock proteins and organic solutes) responses to high soil sodium in four Serengeti C(4) grasses. The species tested were Andropogon greenwayi and three species of Sporobulus, S. ioclados, S. kentrophyllus and S. spicatus. Andropogon greenwayi occurs on locations with low soil Na concentrations, S. ioclados on low to moderate, S. kentrophyllus moderate to high, and S. spicatus on soils with high Na concentration.Among all four species, short-term physiological and molecular responses to Na treatments (0, 100, and 400 mmol/L Na) were correlated with their field soil Na concentrations. Sporobulus kentrophyllus and S. spicatus exhibited rapid molecular induction of heat shock proteins in response to experimental soil Na treatments within 24 h and had increased levels of proline within 96 h in contrast to A. greenwayi and S. ioclados. Photosynthetic rates and water relations were positively correlated with field soil Na concentrations and Hsp induction was clearly associated with photosynthetic tolerance. Long-term (6 wk) responses of the four species to Na treatment were consistent with the short-term responses to Na. Species that occur on low Na soils in the field did not survive past week 1 when treated with 400 mmol/L Na and exhibited significant reductions in biomass when treated with 100 mmol/L Na. Reduced biomass was associated with increased shoot tissue Na concentrations, and thus Na tolerance correlated with the Na concentrations of field leaf tissue. The results demonstrate that the community distribution of these species reflects their Na tolerance and that the observed physiological and molecular responses in tolerant species may have adaptive significance.

Influence of nutrient availability and tree wildling density on nutrient uptake by Oxalis acetosella and Acer saccharum.

Loss of nutrients following pulses of nutrient input in northern hardwood forests and the general effects of atmospheric deposition on forest communities are of concern. Uptake of nutrients by ground layer vegetation, including herbs and tree wildlings, may be important in both of these processes. We brought plants from the field (Catskill Mts, New York) and grew them under controlled environment conditions at two nutrient input levels to determine responses of Oxalis acetosella and Acer saccharum to increased nutrient input and tree wildling density. Oxalis nutrient concentration increased for many nutrients compared to field plants. Both species doubled their P concentration when P input doubled. Biomass of Oxalis was unaffected by both nutrient input level and Acer wildling density. Acer showed a similar response to increased nutrient input and its density did not affect the response of Oxalis to increased nutrient input. Results indicate that both species may be important to nutrient retention in northern hardwood forests.

Tradeoff between height and relative growth rate in a dominant grass from the Serengeti ecosystem.

We determined the relationship between plant height and whole-plant relative growth rate (g g-1 day-1) for ten genotypes of Sporobolus kentrophyllus collected from an intensively grazed site on the Serengeti Plains, Tanzania. Plants were grown for 7 weeks in a greenhouse in Syracuse, N.Y., and harvested weekly. Plants that received simulated bovine urine showed a negative relationship between plant height and growth rate, suggesting a genetic tradeoff between competitive ability if ungrazed (height) and ability to recover from grazing (growth rate). There was no height-growth rate relationship under nitrogen addition rates similar to field mineralization rates. In addition, faster-growing, shorter plants tended to have relatively higher above-ground growth rates than slower-growing, taller plants. These results suggest that natural selection has maintained a gradient of morphologies within this species ranging from short, rapidly growing genotypes adapted to intense grazing conditions to tall, slow-growing, grazer-susceptible genotypes that are superior light competitors in absence of herbivory.

Measuring Heterotroph-Induced Source-Sink Relationships in Panicum Coloratum with 11C Technology.

We report a synthesis from three series of experiments on source-sink relationships in Panicum coloratum L., a C"4 tropical grass obtained from the Serengeti grasslands of Africa. Studies on ^1^1C real-time analyses of P. coloratum to determine aboveground effects of grasshopper grazing and belowground effects of mycorrhizal inoculation and nematode feeding provided the database. A series of multi- and univariate statistical investigations of all available experimental data described responses of leaves, stems, and roots to these biological stresses. From a principal components analysis we have shown differences in distribution of C source-sink locations along three principal component axes, which accounted for 84% of the experimental variance. The first and second components (62% of variance) described C allocation to leaf, stem, and root sinks. The third component (22% of variance) showed a metabolic dichotomy between leaf starch sinks and labile carbon pools throughout the plant. We use the three principal components from a ^1^1C three-compartment model describing leaf, stem, and root C source and sink variables to present patterns, or fingerprints, of responses to the experiments. Time of day, treatment class, number of days since transplanting, and ecotype controlled a large amount of the overall variation in plant C fixation and reallocation. A comparison of ^1^2C leaf carbon exchange rates (CER) measured with an infrared gas analyzer and ^1^1C rates showed a high positive correlation. Slopes for grasshopper grazing, mycorrhizal inoculation experiments, and nematode feeding showed almost identical results; however, differences in the intercept developed as a function of ecotype. We noted a significantly lower intercept in morning studies, but no differences in the slope for morning compared to afternoon studies. CER and all ^1^1C variables for grasshopper and nematode experiments showed a lower coefficient of ^1^1C variables and higher for CER. We conclude that ^1^1C experiments provide the base for developing laboratory, field, modeling studies to incorporate aggregations of real-time C transfers within plants responding to biological stresses, including those of heterotrophs.

Grasses and Grazers, Science and Management.

The evidence that pure ecological science has influenced grazing management practices employed in livestock husbandry on the North American Great Plains is tenuous at best. Those practices arose from critical technological innovations in the 19th century that modified the essential properties of grazing ecosystems when applied to livestock husbandry. Ecological research has identified many significant differences as well as similarities between natural and human-defined grazing systems, and how husbandry can lead to overgrazing. There is considerable evidence that moderate grazing can increase primary productivity and unequivocal evidence that some plants benefit from the presence of grazing animals.

Effect of animal husbandry on herbivore-carrying capacity at a regional scale.

All significant properties of the herbivore trophic level, including biomass, consumption and productivity, are significantly correlated with primary productivity across a broad range of terrestrial ecosystems. Here we show that livestock biomass in South American agricultural ecosystems across a 25-fold gradient of primary productivity exhibited a relationship with a slope essentially identical to unmanaged ecosystems, but with a substantially greater y-intercept. Therefore the biomass of herbivores supported per unit of primary productivity is about an order of magnitude greater in agricultural than in natural ecosystems, for a given level of primary production. We also present evidence of an increase in livestock body size with primary productivity, a pattern previously characterized in natural ecosystems. To our knowledge this is the first quantitative documentation at a regional scale of the impact of animal husbandry practices, such as herding, stock selection and veterinary care, on the biomass and size-structure of livestock herds compared with native herbivores.

Effect of stress and time for recovery on the amount of compensatory growth after grazing.

We tested the hypothesis that the amount of compensatory growth after defoliation is affected by the level of stress at which plants grow when defoliated and by the length of time for recovery. Growth response to defoliation went from partial compensation when plants were growing at high relative growth rates (RGR) to overcompensation when plants were more stressed and growing at low RGR. Defoliation released plants from the limitation imposed by the accumulation of old and dead tissue and this release overrode the negative effect of biomass loss. Compensatory growth resulted from a higher RGR aboveground that was not associated with a reduction in RGR belowground. Time available for recovery had a major impact on the outcome of defoliation. With a short time for recovery, RGR was decreased by defoliation because an immediate increase in net assimilation rate was overridden by a reduction in the ratio of leaf area to plant weight. After defoliation, this ratio increased quickly due to a larger allocation to leaf growth and lower leaf specific weights, resulting in higher RGR. We conclude that the compensatory response to grazing depends on the type and level of stress limiting growth. Allocation and physiological responses to stress may positively or negatively affect the response to grazing and, simultaneously, grazing may alleviate or aggravate the effects of different types of stress.

Interactive effect of flooding and grazing on the growth of Serengeti grasses.

Grazing and flooding may potentially interact in particular habitats of many grassland regions around the world. We tested the hypothesis that grazing and flooding induce different and largely opposed allocation responses in individual plants. As a result, their combined effect on plant growth would be negative. We studied the response of three grass species from the Serengeti ecosystem (Tanzania) to the effects of flooding and clipping. Plants under the combined effect of flooding and clipping had lower growth rates than plants growing under the effect of either of the two factors individually. Plants under flooding grew taller and allocated more resources to stem growth than controls; for two of the three species, flooded plants also generated a new root system above soil level. All these morphological and physiological responses conflict with the ability of a plant to respond to defoliation with minimum reduction in growth rates. The three species showed a response to flooding reflecting their distribution ranges in the field: the species from the most flood-prone habitat showed a positive effect of flooding on growth, whereas the species from dry uplands showed a strong negative effect of flooding. Flood-tolerant species were taller and less tolerant of clipping than flooding sensitive species. Our results suggest that, in ecological time, individuals subjected to both flooding and grazing have their growth reduced to a greater extent than by either of the two factors acting individually, whereas in evolutionary time, species adapted to flooding are poor grazing tolerators and species adapted to grazing are poor flooding tolerators.

Ecosystem-level patterns of primary productivity and herbivory in terrestrial habitats.

Ecosystems are structurally organized as food webs within which energy is transmitted between trophic levels and dissipated into the environment. Energy flow between two trophic levels is given by the amount of production at the lower level and by the proportion of production that is consumed, assimilated and respired at the higher level. Considerable evidence indicates that food-web structure varies predictably in different habitats, but much less is known about quantitative relationships among food web fluxes. Many of the energetic properties of herbivores in African game parks are associated with rainfall and, by inference, with net primary productivity. Respiratory costs per unit production at the consumer trophic level are higher for homeotherms than for heterotherms. Plant secondary chemicals affect herbivore dietary choices and the allocation of plant resources to those chemicals varies with resource availability. How these phenomena are translated into ecosystem fluxes is unknown. We present evidence that herbivore biomass, consumption and productivity are closely correlated with plant productivity, suggesting that the latter is a principal integrator and indicator of functional processes in food webs.

Lack of compensatory growth under phosphorus deficiency in grazing-adapted grasses from the Serengeti Plains.

Two shortgrass species (Sporobolus ioclados and Eustachys paspaloides) and two midgrass species (E. paspaloides and Pennisetum mezianum) from the Serengeti grasslands of Tanzania were grown under conditions of extreme phosphorus (P) deficiency. Production of each of these species is maintained or enhanced by defoliation under adequate nutrient supply (McNaughton et al. 1983). However, under the P-deficient conditions of our experiment, defoliation caused a reduction in biomass of all plant parts of each species. Green leaf biomass was reduced most strongly by defoliation, and crowns were least affected. Yield of biomass and nutrients to grazers (green leaves+clipped material) was enhanced by weekly defoliation in the shortgrass grazing-adapted species, whereas yield to producers (live biomass and nutrients retained by the plant) and yield to decomposers (litter) were strongly reduced by defoliation in all species. Phosphate absorption capacity (V max) measured on excised roots was enhanced by defoliation in the grazing-adapted Sporobolus, but, due to low affinity (high K m) of roots of defoliated plants for phosphate, absorption rate was not greatly altered at low solution concentrations. Phosphate absorption capacity was reduced or unaffected by defoliation in other species. We conclude that under conditions of P deficiency, plants are unable to acquire the nutrients necessary to replenish large nutrient losses to grazers. In low-nutrient environments, compensatory growth (stimulation of production by grazing) is not a viable strategy. Therefore, in these environments plants respond evolutionarily to herbivores by developing chemical or morphological defenses.

Carbon partitioning patterns of mycorrhizal versus non-mycorrhizal plants: real-time dynamic measurements using 11 CO2.

Gas exchange and carbon allocation patterns were studied in two populations of Panicum coloratum, an Africa C-4 grass. The plants were grown in split-root pots, containing partially sterilized soil, with one side either inoculated (I) or not inoculated (NI) with a vesicular arbuscular (VA) mycorrhizal Fungus, Gigaspora margarita. Net carbon exchange rates (CER) and stomatal conductances were measured with conventional gas exchange apparatus, and carbon assimilation, translocation, and allocation were measured using photosynthetically-fixed 11 CO2 . Mycorrhizal infection on one half of the split-root system caused a 20%, increase in CER. The effect on CER was less in tillers on the opposite side of the plants from the infected half of the roots. The rate at which photosynthates were stored in the leaves was 45% higher. Sink activity (concentration of labelled photosynthates in stem phloem tissue) more than doubled in 1 versus NI plants. CER and stomatal conductances, along with most of the carbon allocation patterns, were nearly identical between the NI (control) high grazing and low grazing ecotypes. However, VA mycorrhizal fungi caused a greater storage of photosynthates in the low grazing ecotype.

Mineral nutrition and spatial concentrations of African ungulates.

Africa's abundant large herbivores are very heterogeneously distributed, both geographically and regionally. Within a region, some localities contain dense animal concentrations although areas nearby may be virtually unoccupied. Mixed-species herds are a conspicuous feature of areas where animals concentrate. The prevailing explanations of local distributional concentrations are (1) that different herbivore species facilitate each other's foraging, and (2) that animals are protected from predation by both intraspecific and interspecific association. If facilitation of grazing were an overriding factor, mixed species herds should move extensively with localized rain showers to obtain the greatest forage yield. If predation were the major factor influencing animal densities and distributions, rapid, unpredictable spatial movements would further reduce predation. But because resident, non-migratory species tend to occupy home ranges that are stable over time, neither of these hypotheses is totally compelling. Because tropical forages are of lower quality than temperate ones and are often chronically deficient in mineral elements, I tested the hypothesis that areas where animals concentrate are localities supporting forages of higher mineral content. I report here that the mineral content of foods is an important determinant of the spatial distributions of animals within the Serengeti National Park, Tanzania. Based on ecological critieria, magnesium, sodium and phosphorus appear particularly important.

Intraspecific variation in the response of Themeda triandra to defoliation: the effect of time of recovery and growth rates on compensatory growth.

The response to a single defoliation was studied on three clones of Themeda triandra collected in the short, mid, and tall grassland regions of the Serengeti National Park (Tanzania). These sites represent a gradient of decreasing grazing intensity. Growth, allocation pattern, and several morphometric traits were monitored during an 80-day period. Clipped plants of the short and medium clones fully compensated for the reduction of biomass, while plants of the tall clone showed overcompensation. During the first two weeks after clipping, clipped plants showed lower relative growth rates than unclipped ones, whereas the opposite was observed later on. Clipped plants compensated for the removal of leaf area by producing new leaves with lower specific weights and higher nitrogen content. They also produced more, smaller tillers. Although clipped plants mobilized nonstructural carbohydrates from roots and crowns, this did not account for a significant amount of growth. Relative growth rates of unclipped plants of the short clone were higher. The relative growth rate of the short clone diminished less after clipping, but also exhibited the lowest increase later. The tall clone was the most negatively affected early, but showed the highest compensation later. Compared to the other clones, the short ecotype showed many of the characteristics that defoliation induced in each individual of any clone: higher allocation to leaf area production, higher relative growth rate, higher number but smaller size of tillers, and lower leaf specific weights.

Ammonia volatilization and the effects of large grazing mammals on nutrient loss from East African grasslands.

Ammonia volatilization losses measured from soils at seven sites in the Serengeti National Park, Tanzania during the 1986 growing season ranged from 2.78±0.49% to 25.03±1.34% of nitrogen applied. Although peak ammonia losses ranged from 0.071±0.018 to 0.404±0.040 g N m-2 h-1, rates dropped to zero within four days, and calculations reveal that volatilization losses represent minor fluxes in the context of the system's nitrogen cycling. Volatilization losses were inversely correlated with grazing intensity experienced by a site, and it appears that large ungulates themselves contribute to nutrient conservation throught indirect interactive effects on system processes.

Responses of an African tall-grass (Hyparrhenia filipendula stapf.) to defoliation and limitations of water and nitrogen.

Hyparrhenia filipendula stapf., a tall (1-1.5 m) perennial grass common in dry-subhumid African savannas, was collected from Serengeti National Park in Tanzania, propagated vegetatively, and grown in controlled environments simulating conditions in nature. Plants were subjected to a factorial experiment with combinations of watering frequency, nitrogen supply, clipping height (10 and 15 cm) and clipping frequency (7 and 14 d). Biomass yield and allocation to various tissue types, morphometric traits, and growth processes were measured.Watering frequency affected leaf elongation rate while nitrogen affected tiller number. Clipped yield was strongly correlated with leaf elongation rate but not tiller number, therefore it was primarily controlled by the activities of intercalary rather than apical meristems. There was a negative exponential relationship between tillering and clipped yield per tiller. Plants that received both high nitrogen and high water closely followed a-3/2 power law in this tradeoff. The fraction of total net photosynthate allocated to roots was not significantly related to any environmental treatment. Root and crown growths were not affected by defoliation treatment; leaf blade and stem growths were inhibited; and sheaths were inhibited only under low water.Despite a tall stature, H. filipendula tolerated herbivory by increased photosynthetic rate (Wallace et al. 1984), through continued production of young tissues by intercalary meristems balanced against tiller number, and by a statistically constant proportional allocation to roots. Clipped yield increased only when both nitrogen and water were abundant, and then, proportional clipped yield did not surpass an upper asymptotic limit.

Responses of an African graminoid (Themeda triandra Forsk.) to frequent defoliation, nitrogen, and water: a limit of adaptation to herbivory.

Themeda triandra Forsk., a medium height perennial grass common in semi-arid Africa, was collected from Serengeti National Park in Tanzania, propagated vegetatively and grown in controlled environments simulating native conditions. The experiment demonstrated mechanisms that result in inability to withstand frequent defoliation, and thus how plant-herbivore interactions are affected by plant morpho-physiology. Clipping reduced most plant yield components. After two months leaf elongation rate was greater in clipped plants, but over the whole experiment aboveground yield was unaffected by clipping. Since clipping did not stimulate growth, residual leaf area was less in clipped plants. Tillering was not stimulated by clipping so there was no mechanism to increase leaf area below the clipping height. The resulting reduced leaf area accounted for reduced root and crown production by clipped plants.

Interactive regulation of grass yield and chemical properties by defoliation, a salivary chemical, and inorganic nutrition.

Clones of 2 C4 grass species, Sprobolus ioclados and S. pyramidalis, were obtained from more and less heavily grazed grasslands, respectively, in Tanzania's Serengeti National Park. Plants were grown in a factorial experiment to determine the effects of severe defoliation, nutrient limitation, and a salivary chemical (thiamine) on plant growth, nitrogen content, and non-structural carbohydrate content. The experimental design included: (1) species; (2) clipping, with plants either unclipped or clipped weekly to a height of 5 cm; (3) 0.2 ml of distilled water of 0.2 ml of 10 ppb aqueous thiamine sprayed on plants from an atomizer after clipping and identical treatments at the same time to unclipped plants; (4) phosphorus (P) at 0.2 or 1 mM; (5) nitrogen (N) at 3 or 15 mM. Clipping was the major variable affecting plant growth. Total and litter yields were reduced to half and residual plant yield was reduced to 30% of the values for unclipped plants. Clipping interacted strongly with other variables since they commonly had minor effects on clipped plants and major effects on unclipped plants. Exceptions to this generalization were generally due to better performance by S. ioclados under clipping. Compared to lower treatment levels, higher treatment levels promoted total yield of unclipped plants by 52% for N, 43% for thiamine, and 33% for P. In general, thiamine had greater effects than P but lesser effects than N. Thiamine promoted yield and modified the chemical balance of plants by promoting carbohydrate (CHO) concentrations and reducing N concentrations. N and P deficiencies promoted CHO accumulation. Clipping promoted the N of leaves and crowns and reduced the N levels in roots. Leaf blade water and N contents were positively correlated with very little scatter. The slope of the line was different for S. ioclados and S. pyramidalis. Leaf blade water and CHO contents were negatively related but there was more scatter and the species could not be distinguished. The species from more heavily grazed grasslands was conspicuously more sensitive to thiamine application. The results indicated that leaf treatment with thiamine, the only likely source of which in natural grasslands is saliva deposited by feeding herbivores, can have major effects on plant yield and metabolic balances at very low application levels. But under defoliation levels as severe as those imposed in this experiment, which reduced above ground plant biomass to a fourth of the level produced by unclipped plants, growth was so strongly limited by defoliation that neither thiamine nor inorganic nutrients affected plant yield residual from clipping. Therefore, whether chemicals such as thiamine that may be introduced onto grass foliage by grazing ungulates and other herbivores will influence the growth of grazed plants will depend upon the grazing intensity associated with the transfer.

Urea as a promotive coupler of plant-herbivore interactions.

Growth responses of Kyllinga nervosa Steud., a sedge from the Serengeti short-grass plains, were examined in a factorial experiment which included clipped and unclipped plants, and nitrogen supplied as either urea or ammonium nitrate. Results were expressed in relation to three transfer processes: flow to grazers, flow to producers and flow to reproduction. Clipping increased biomass and nitrogen flow to grazers by significantly increasing nitrogen uptake, aboveground nitrogen flow, and the weights of and proportional allocation to green leaf production. This was at the expense of flow to vegetative and sexual reproduction, since the weights and proportional investments in roots, crowns and reproductive structures were reduced. Urea nutrition increased flow to grazers and plant reproduction through increases in green leaf weight, flower weight, allocation to green leaves, flowers and stems, and aboveground: belowground biomass ratios. Stimulation of aboveground productivity by urea was a consequence of increased tillering rates.Interactive responses of clipping and nitrogen source regulated plant growth, thus controlling flow to each transfer process. Combined effects of clipping and urea resulted in compensatory production of both green leaves and flowers, and maximized biomass and nitrogen flow to grazers. Both urea and clipping tightened herbivore-producer recycling by significantly reducing litter nitrogen and carbon masses. In contrast, when plants were unclipped and grown on NH4NO3, biomass allocation and weights of roots and crowns were increased at the expense of aboveground tissues, thus increasing flow to primary producers. Plant growth responses to experimental treatment combinations simulating nutritional status of grazed and ungrazed field plants indicate that urea represents a potential importance beyond it nitrogen contribution by introducing a positive feedback to herbivores.

Grass leaf silicification: Natural selection for an inducible defense against herbivores.

Plants from four populations of three species of African grasses were collected from grasslands in Tanzania's Serengeti National Park that differ in the grazing intensity that they experience. Plants were grown in the laboratory in a factorial experiment in which variables were plant origin, species identification of plants, defoliation intensity, and supply of soluble silicate in the nutrient medium. All plants accumulated silica in leaf blades in the absence of soluble silicate from the nutrient medium. Plants native to the more heavily grazed grassland accumulated more silica in their leaf blades than did plants from the less heavily grazed site. Blade silica content was higher when plants were defoliated, indicating that silicification is an inducible defense against herbivores. The quantitative heterogeneity of this qualitatively homogeneous plant defense system may have contributed to the evolution of high species diversity in the grazing fauna.

The effects of clipping, nitrogen source and nitrogen concentration on the growth responses and nitrogen uptake of an east african sedge.

Kyllinga nervosa Steud., a sedge from the Serengeti short-grass plains, was subjected to a balanced factorial experiment which included unclipped plants and plants clipped weekly to a 5 cm height, nitrogen supplied as either nitrate or ammonium and three nitrogen concentrations. Tillering rates, green leaf nitrogen, and both green leaf weight and biomass investment in green leaf production increased with nitrogen concentration. Low nitrogen conserved investment in crown production and resulted in adjustments for nitrogen acquisition by increasing biomass allocation to root production. Nitrate nutrition stimulated green leaf weight, tillering rate, nitrogen redistribution and both crown and root nitrogen. Ammonium nutrition increased nitrogen uptake, total plant nitrogen accumulation, reproduction, litter weight and nitrogen loss to decomposers. Clipping increased investment in green leaf production at the expense of stem, root, crown and flower production. Compensatory green leaf production in response to clipping occurred only when plants were grown in ammonium. Clipping stimulated uptake rates of both ammonium and nitrate, and therefore total plant nitrogen accumulation. Results suggest a balanced utilization of both nitrate and ammonium may be necessary for optimal growth in this species.