Nitrogen fixation does not axiomatically lead to phosphorus limitation in aquatic ecosystems

A long‐standing debate in ecology deals with the role of nitrogen and phosphorus in management and restoration of aquatic ecosystems. It has been argued that nutrient reduction strategies to combat blooms of phytoplankton or floating plants should solely focus on phosphorus (P). The underlying argument is that reducing nitrogen (N) inputs is ineffective because N₂‐fixing species will compensate for N deficits, thus perpetuating P limitation of primary production. A mechanistic understanding of this principle is, however, incomplete. Here, we use resource competition theory, a complex dynamic ecosystem model and a 32‐year field data set on eutrophic, floating‐plant dominated ecosystems to show that the growth of non‐N₂‐fixing species can become N limited under high P and low N inputs, even in the presence of N₂ fixing species. N₂‐fixers typically require higher P concentrations than non‐N₂‐fixers to persist. Hence, the N₂ fixers cannot deplete the P concentration enough for the non‐N₂‐fixing community to become P limited because they would be outcompeted. These findings provide a testable mechanistic basis for the need to consider the reduction of both N and P inputs to most effectively restore nutrient over‐enriched aquatic ecosystems.     

Nitrogen fixation in the High Arctic: a source of ‘new’ nitrogen?

Biological nitrogen (N₂) fixation performed by diazotrophs (N₂ fixing bacteria) is thought to be one of the main sources of plant available N in pristine ecosystems like arctic tundra. However, direct evidence of a transfer of fixed N₂ to non-diazotroph associated plants is lacking to date. Here, we present results from an in situ ¹⁵N–N₂ labelling study in the High Arctic. Three dominant vegetation types (organic crust composed of free-living cyanobacteria, mosses, cotton grass) were subjected to acetylene reduction assays (ARA) performed regularly throughout the growing season, as well as ¹⁵N–N₂ incubations. The ¹⁵N-label was followed into the dominant N₂ fixer associations, soil, soil microbial biomass and non-diazotroph associated plants three days and three weeks after labelling. Mosses contributed most to habitat N₂ fixation throughout the measuring campaigns, and N₂ fixation activity was highest at the beginning of the growing season in all plots. Fixed ¹⁵N–N₂ became quickly (within 3 days) available to non-diazotroph associated plants in all investigated vegetation types, proving that N₂ fixation is an actual source of available N in pristine ecosystems.     

Effects of modified nutrient concentrations and ratios on the structure and function of the native phytoplankton community in the Neuse River Estuary, North Carolina, USA

A variety of analyses were used to assess the structure (community composition) and function (assimilation number, nitrogen fixation) of phytoplankton in the Neuse River Estuary (NRE), NC under ambient and modified nutrient concentrations. Dilution bioassays were employed to reduce the concentration of nitrogen (N) or both N and phosphorus (P) and thus compare varied DIN:DIP ratios. Experimental manipulations created conditions that may result from mandated N load reductions to the estuary. We hypothesized that unilateral reduction of N loading to the NRE would increase the activity, abundance and diversity of N₂ fixing cyanobacteria. Changes in phytoplankton primary productivity, N₂ fixation (nitrogenase activity), genetic potential for N₂ fixation (presence of nifH), phytoplankton taxonomic composition (diagnostic photopigment concentration) and abundances of N₂ fixing cyanobacteria (microscopy) were determined. Decreasing ambient DIN:DIP ratios in NRE samples resulted in increased rates of N₂ fixation when seed populations were present and environmental conditions were amenable. Decreasing the DIN:DIP ratio did not lead to an increase in the abundance or diversity of N₂ fixing cyanobacteria. Because N₂ fixing cyanobacteria were only actively fixing nitrogen during periods of low riverine N discharge (summer and early autumn), lowering nutrient ratios may not have a major impact on the NRE. However, the maximum potential amount of N from N₂ fixation was calculated using rates from this study and was found to be approximately 3% of total riverine loading of N to the NRE. Because N₂ fixation occurs farther downstream and later in the year than riverine N loading to the NRE, there is potential for N₂ fixation to modify N dynamics. Analyses of the phytoplankton community as a whole in these relatively short term experiments indicated that reduced DIN:DIP may not have a major impact on their structure and function.     

Nitrogen fixation in legumes and actinorhizal plants in natural ecosystems: values obtained using 15N natural abundance

Background: Nitrogen fixation has been quantified for a range of crop legumes and actinorhizal plants under different agricultural/agroforestry conditions, but much less is known of legume and actinorhizal plant N₂ fixation in natural ecosystems.

Aims: To assess the proportion of total plant N derived from the atmosphere via the process of N₂ fixation (%Ndfa) by actinorhizal and legume plants in natural ecosystems and their N input into these ecosystems as indicated by their ¹⁵N natural abundance.

Methods: A comprehensive collation of published values of %Ndfa for legumes and actinorhizal plants in natural ecosystems and their N input into these ecosystems as estimated by their ¹⁵N natural abundance was carried out by searching the ISI Web of Science database using relevant key words.

Results: The %Ndfa was consistently large for actinorhizal plants but very variable for legumes in natural ecosystems, and the average value for %Ndfa was substantially greater for actinorhizal plants. High soil N, in particular, but also low soil P and water content were correlated with low legume N₂ fixation. N input into ecosystems from N₂ fixation was very variable for actinorhizal and legume plants and greatly dependent on their biomass within the system.

Conclusions: Measurement of ¹⁵N natural abundance has given greater understanding of where legume and actinorhizal plant N₂ fixation is important in natural ecosystems. Across studies, the average value for %Ndfa was substantially greater for actinorhizal plants than for legumes, and the relative abilities of the two groups of plants to utilise mineral N requires further study.     

The regulation of symbiotic N2 fixation: a conceptual model of N feedback from the ecosystem to the gene expression level

Symbiotic nitrogen (N₂) fixation in legumes may give the host plant a distinct competitive advantage; at the same time it is mainly responsible for introducing N into terrestrial ecosystems which may ultimately benefit all organisms. Depending on environmental conditions, symbiotic N₂ fixation may be tuned to the plant's N demand or specifically inhibited (a disadvantage for plants which depend mainly on symbiotic N₂ fixation), or even prevented. Thus, the ecological range for symbiotic N₂ fixation can be narrower than that of the host plants. A shortage of mineral N is the only case in which adverse environmental conditions clearly favour symbiotic N₂ fixation. Variations in number or mass of nodules or nodule morphology are persistent features, that may represent one kind of regulation of N₂ fixation. In addition, varying O₂ permeability of nodules functions as a rapid and reversible control of N₂ fixation which may compensate partially or fully for poor nodulation. The plant's demand for symbiotically fixed N is thought to play a central role in modulating both nodulation and N₂ fixation activity; an N feedback mechanism is assumed. The control of symbiotic N₂ fixation operates through a series of ecophysiological triggers which are also influenced by complex interactions between legume plants and other organisms in the ecosystem. The proportion of legume biomass and the performance of symbiotic N₂ fixation in each individual legume are the main parameters which determine the amount of symbiotically fixed N introduced into a terrestrial ecosystem. The various triggers and N feedback mechanisms from the whole ecosystem to the gene expression level which regulate symbiotic N₂ fixation in terrestrial ecosystems are reviewed and discussed in terms of a conceptual model. Although the presented model is based primarily on our knowledge about the physiology of a few leguminous crop species and of ecosystem processes in managed, perennial grassland in temperate climatic conditions, it may stimulate thinking about functional relationships between symbiotic N₂ fixation and terrestrial ecosystems at various system levels.     

Direct evidence that symbiotic N2 fixation in fertile grassland is an important trait for a strong response of plants to elevated atmospheric CO2

Although legumes showed a clearly superior yield response to elevated atmospheric pCO₂ compared to nonlegumes in a variety of field experiments, the extent to which this is due to symbiotic N₂ fixation per se has yet to be determined. Thus, effectively and ineffectively nodulating lucerne (Medicago sativa L.) plants with a very similar genetic background were grown in competition with each other on fertile soil in the Swiss FACE experiment in order to monitor their CO₂ response. Under elevated atmospheric pCO₂, effectively nodulating lucerne, thus capable of symbiotically fixing N₂, strongly increased the harvestable biomass and the N yield, independent of N fertilization. In contrast, the harvestable biomass and N yield of ineffectively nodulating plants were affected negatively by elevated atmospheric pCO₂ when N fertilization was low. Large amounts of N fertilizer enabled the plants to respond more favourably to elevated atmospheric pCO₂, although not as strongly as effectively nodulating plants. The CO₂‐induced increase in N yield of the effectively nodulating plants was attributed solely to an increase in symbiotic N₂ fixation of 50–175%, depending on the N fertilization treatment. N yield derived from the uptake of mineral N from the soil was, however, not affected by elevated pCO₂. This result demonstrates that, in fertile soil and under temperate climatic conditions, symbiotic N₂ fixation per se is responsible for the considerably greater amount of above‐ground biomass and the higher N yield under elevated atmospheric pCO₂. This supports the assumption that symbiotic N₂ fixation plays a key role in maintaining the C/N balance in terrestrial ecosystems in a CO₂‐rich world.     

Biological nitrogen fixation by alternative nitrogenases in terrestrial ecosystems: a review

Biological nitrogen fixation (BNF), a key reaction of the nitrogen cycle, is catalyzed by the enzyme nitrogenase. The best studied isoform of this metalloenzyme requires molybdenum (Mo) at its active center to reduce atmospheric dinitrogen (N₂) into bioavailable ammonium. The Mo-dependent nitrogenase is found in all diazotrophs and is the only nitrogenase reported in diazotrophs that form N₂-fixing symbioses with higher plants. In addition to the canonical Mo nitrogenase, two alternative nitrogenases, which use either vanadium (V) or iron (Fe) instead of Mo are known to fix nitrogen. They have been identified in ecologically important groups including free-living bacteria in soils and freshwaters and as symbionts of certain cryptogamic covers. Despite the discovery of these alternative isoforms more than 40 years ago, BNF is still believed to primarily rely on Mo. Here, we review existing studies on alternative nitrogenases in terrestrial settings, spanning inland forests to coastal ecosystems. These studies show frequent Mo limitation of BNF, ubiquitous distribution of alternative nitrogenase genes and significant contributions of alternative nitrogenases to N₂ fixation in ecosystems ranging from the tropics to the subarctic. The effect of temperature on nitrogenase isoform activity and regulation is also discussed. We present recently developed methods for measuring alternative nitrogenase activity in the field and discuss the associated analytical challenges. Finally, we discuss how the enzymatic diversity of nitrogenase forces a re-examination of existing knowledge gaps and our understanding of BNF in nature.

     

Adaptation of methods for evaluating N2 fixation in food legumes and legume cover crops

Methods for partitioning the nitrogen assimilated by nodulated legumes, between nitrogen derived from soil sources and from N₂ fixation, are described as applied in peninsular Malaysia. The analysis of nitrogenous components translocated from the roots to the shoots of nodulated plants in the xylem sap is outlined, with some precautions to be observed for applications in the tropics. Some examples of the use of the technique in surverying apparent N₂ fixation by tropical legumes, in studying interrow cropping in plantation systems and in assessing effects of experimental treatments on N₂ fixation by food legumes, are described. Techniques for assesing N₂ fixation by means of¹⁵N abundance have been used to show that applications of nitrogenous fertilizers commonly used in Malaysia for soybeans depress N₂ fixation, that similar results are obtained with natural abundance and¹⁵N-enrichment methods and that, in at least two locations in Malaysia, differences between the natural abundance of¹⁵N in plant-available soil nitrogen and in atmospheric N₂ are great enough to permit application to measurement of N₂ fixation by leguminous crops.     

Stratification,nitrogen fixation,and cyanobacterial bloom stage regulate the planktonic food web structure

Changes in the complexity of planktonic food webs may be expected in future aquatic systems due to increases in sea surface temperature and an enhanced stratification of the water column. Under these conditions, the growth of unpalatable, filamentous, N₂‐fixing cyanobacterial blooms, and their effect on planktonic food webs will become increasingly important. The planktonic food web structure in aquatic ecosystems at times of filamentous cyanobacterial blooms is currently unresolved, with discordant lines of evidence suggesting that herbivores dominate the mesozooplankton or that mesozooplankton organisms are mainly carnivorous. Here, we use a set of proxies derived from amino acid nitrogen stable isotopes from two mesozooplankton size fractions to identify changes in the nitrogen source and the planktonic food web structure across different microplankton communities. A transition from herbivory to carnivory in mesozooplankton between more eutrophic, near‐coastal sites and more oligotrophic, offshore sites was accompanied by an increasing diversity of microplankton communities with aging filamentous cyanobacterial blooms. Our analyses of 124 biotic and abiotic variables using multivariate statistics confirmed salinity as a major driver for the biomass distribution of non‐N₂‐fixing microplankton species such as dinoflagellates. However, we provide strong evidence that stratification, N₂ fixation, and the stage of the cyanobacterial blooms regulated much of the microplankton diversity and the mean trophic position and size of the metabolic nitrogen pool in mesozooplankton. Our empirical, macroscale data set consistently unifies contrasting results of the dominant feeding mode in mesozooplankton during blooms of unpalatable, filamentous, N₂‐fixing cyanobacteria by identifying the at times important role of heterotrophic microbial food webs. Thus, carnivory, rather than herbivory, dominates in mesozooplankton during aging and decaying cyanobacterial blooms with hitherto uncharacterized consequences for the biogeochemical functions of mesozooplankton.     

Genetic specificity of nitrogen nutrition in leguminous plants

Summary Mineral nitrogen did not increase grain yield and seed protein levels ofVicia faba L. andLupinus luteus L. in field trials and pot experiments. Fixed N₂ was substituted by mineral nitrogen in these cases because of inhibition of N₂ fixation by mineral nitrogen. Contrary to these results mineral nitrogen increased grain yields and seed protein amounts ofLupinus albus L.,Pisum sativum L., andGlycine max. (L.) Merr. The nitrogen effect was caused at an early stage by saving energy due to inhibition of N₂ fixation (measurement of gas exchange by means of IRGA). In case of the N application after flowering grain, yields and seed protein levels increased because the mineral N was an additional nitrogen source for plants. At this stage the plants had ceased fixing atmospheric nitrogen. The high sink activity of growing fruits induced a lack of assimilates in nodules (determined by means of¹⁴CO₂ application). The N effect was therefore the consequence of the lower assimilate pool for supplying root nodules in these plants in comparison withVicia faba L. andLupinus luteus L. Hence it follows that response to mineral nitrogen can be a criterion for discovering more effective Rhizobium-host combinations.     

Nitrogen Transfer Within and Between Plants Through Common Mycorrhizal Networks (CMNs)

Mycorrhizae play a critical role in nutrient capture from soils. Arbuscular mycorrhizae (AM) and ectomycorrhizae (EM) are the most important mycorrhizae in agricultural and natural ecosystems. AM and EM fungi use inorganic NH₄ ⁺ and NO₃ ^?, and most EM fungi are capable of using organic nitrogen. The heavier stable isotope ¹⁵N is discriminated against during biogeochemical and biochemical processes. Differences in ¹⁵N (atom%) or δ¹⁵N (‰) provide nitrogen movement information in an experimental system. A range of 20 to 50% of one-way N-transfer has been observed from legumes to nonlegumes. Mycorrhizal fungal mycelia can extend from one plant's roots to another plant's roots to form common mycorrhizal networks (CMNs). Individual species, genera, even families of plants can be interconnected by CMNs. They are capable of facilitating nutrient uptake and flux. Nutrients such as carbon, nitrogen and phosphorus and other elements may then move via either AM or EM networks from plant to plant. Both ¹⁵N labeling and ¹⁵N natural abundance techniques have been employed to trace N movement between plants interconnected by AM or EM networks. Fine mesh (25～45 μm) has been used to separate root systems and allow only hyphal penetration and linkages but no root contact between plants. In many studies, nitrogen from N₂-fixing mycorrhizal plants transferred to non-N₂–fixing mycorrhizal plants (one-way N-transfer). In a few studies, N is also transferred from non-N₂–fixing mycorrhizal plants to N₂-fixing mycorrhizal plants (two-way N-transfer). There is controversy about whether N-transfer is direct through CMNs, or indirect through the soil. The lack of convincing data underlines the need for creative, careful experimental manipulations. Nitrogen is crucial to productivity in most terrestrial ecosystems, and there are potential benefits of management in soil-plant systems to enhance N-transfer. Thus, two-way N-transfer warrants further investigation with many species and under field conditions.     

Bryophyte-Cyanobacteria Associations during Primary Succession in Recently Deglaciated Areas of Tierra del Fuego (Chile)

Bryophyte establishment represents a positive feedback process that enhances soil development in newly exposed terrain. Further, biological nitrogen (N) fixation by cyanobacteria in association with mosses can be an important supply of N to terrestrial ecosystems, however the role of these associations during post-glacial primary succession is not yet fully understood. Here, we analyzed chronosequences in front of two receding glaciers with contrasting climatic conditions (wetter vs drier) at Cordillera Darwin (Tierra del Fuego) and found that most mosses had the capacity to support an epiphytic flora of cyanobacteria and exhibited high rates of N₂ fixation. Pioneer moss-cyanobacteria associations showed the highest N₂ fixation rates (4.60 and 4.96 µg N g⁻¹ bryo. d⁻¹) very early after glacier retreat (4 and 7 years) which may help accelerate soil development under wetter conditions. In drier climate, N₂ fixation on bryophyte-cyanobacteria associations was also high (0.94 and 1.42 µg N g⁻¹ bryo. d⁻¹) but peaked at intermediate-aged sites (26 and 66 years). N₂ fixation capacity on bryophytes was primarily driven by epiphytic cyanobacteria abundance rather than community composition. Most liverworts showed low colonization and N₂ fixation rates, and mosses did not exhibit consistent differences across life forms and habitat (saxicolous vs terricolous). We also found a clear relationship between cyanobacteria genera and the stages of ecological succession, but no relationship was found with host species identity. Glacier forelands in Tierra del Fuego show fast rates of soil transformation which imply large quantities of N inputs. Our results highlight the potential contribution of bryophyte-cyanobacteria associations to N accumulation during post-glacial primary succession and further describe the factors that drive N₂-fixation rates in post-glacial areas with very low N deposition.     

Natural15N abundance as a method of estimating the contribution of biologically fixed nitrogen to N2-fixing systems: Potential for non-legumes

The¹⁵N abundance of plants usually closely reflects the¹⁵N abundance of their major immediate N source(s); plant-available soil N in the case of non-N₂-fixing plants and atmospheric N₂ in the case of N₂ fixing plants. The¹⁵N abundance values of these sources are usually sufficiently different from each other that a significant and systematic difference in the¹⁵N abundance between the two kinds of plants can be detected. This difference provides the basis for the natural¹⁵N abundance method of estimating the relative contribution of atmospheric N₂ to N₂-fixing plants growing in natural and agricultural settings. The natural¹⁵N abundance method has certain advantages over more conventional methods, particularly in natural ecosystems, since disturbance of the system is not required and the measurements may be made on samples dried in the field. This method has been tested mainly with legumes in agricultural settings. The tests have demonstrated the validity of this method of arriving at semi-quantitative estimates of biological N₂-fixation in these settings. More limited tests and applications have been made for legumes in natural ecosystems. An understanding of the limits and utility of this method in these systems is beginning to emerge. Examples of systematic measurements of differences in¹⁵N abundance between non-legume N₂-fixing systems and neighbouring non-fixing systems are more unusual. In principle, application of the method to estimate N₂-fixation by nodulated non-legumes, using the natural¹⁵N abundance method, is as feasible as estimating N₂-fixation by legumes. Most of the studies involving N₂-fixing non-legumes are with this type of system (e.g., Ceanothus, Chamabatia, Eleagnus, Alnus, Myrica, and so forth). Resuls of these studies are described. Applicability for associative N₂-fixation is an empirical question, the answer to which probably depends upon the degree to which fixed N goes predominantly to the plant rather than to the soil N pool. The natural¹⁵N abundance method is probably not well suited to assessing the contribution of N₂-fixation by free-living microorganisms in their natural habitat, particularly soil microorganisms.This work was supported in part by subcontracts under grants from the US National Science Foundation (DEB79-21971 and BSR821618)     

Fixation of Dinitrogen-15 Associated with Rice Plants

Rice plants (IR26 and Latisail) obtained at near heading stage from a wetland field were transferred to water culture and exposed to ¹⁵N₂ in a gas-tight growth chamber for 7 days to measure N₂-fixing activities associated with the rice. The activities measured varied from 6.5 to 11.6 μmol of N₂ fixed per hill per day. The outer leaf sheath had about 2.5 times higher N₂-fixing activities per unit weight than the root. Slight activities were also found in the basal node and inner leaf sheath. Wrapping basal parts of the stem with aluminum foil did not decrease the activities of N₂ fixation in these parts. Thus, the outer leaf sheath as well as the root are N₂-fixing sites in rice plants. N₂ fixation found in above-ground parts is not due to photoautotrophic organisms. Less than 10% of the fixed nitrogen was translocated from the fixing sites to the leaf blades and the young panicles.     

Symbiotic performance of grain and wild herbaceous legumes in the Okavango Delta and Tswapong region of Botswana

The low inherent soil fertility, especially nitrogen (N) constrains arable agriculture in Botswana. Nitrogen is usually added to soil through inorganic fertilizer application. In this study, biological nitrogen fixation by legumes is explored as an alternative source of N. The objectives of this study were to measure levels of N₂ fixation by grain legumes such as cowpea, Bambara groundnut and groundnut in farmers’ fields as well as to estimated N₂ fixation by indigenous herbaceous legumes growing in the Okavango Delta. Four flowering plants per species were sampled from the panhandle part of the Okavango Delta and Tswapong area. Nitrogen fixation was measured using the ¹⁵N stable isotope natural abundance technique. The δ¹⁵N values of indigenous herbaceous legumes indicated that they fixed N₂ (?1.88 to +1.35 ‰) with the lowest value measured in Chamaecrista absus growing in Ngarange (Okavango Delta). The δ¹⁵N values of grain legumes growing on farmers’ fields ranging from ?1.2 ‰ to +3.3 ‰ indicated that they were fixing N₂. For grain legumes growing at most farms, %Ndfa were above 50% indicating that they largely depended on symbiotic fixation for their N nutrition. With optimal planting density, Bambara groundnuts on farmers’ fields could potentially fix over 90 kg N/ha in some parts of Tswapong area and about 60 kg N/ha in areas around the Okavango Delta. Results from this study have shown that herbaceous indigenous legumes and cultivated legumes play an important role in the cycling of N in the soil. It has also been shown that biological N₂ on farmer’s field could potentially supply the much needed N for the legumes and the subsequent cereal crops if plant densities are optimized with the potential to increase food security and mitigate climate change.     

Benthic N2 fixation in coral reefs and the potential effects of human‐induced environmental change

Tropical coral reefs are among the most productive and diverse ecosystems, despite being surrounded by ocean waters where nutrients are in short supply. Benthic dinitrogen (N₂) fixation is a significant internal source of “new” nitrogen (N) in reef ecosystems, but related information appears to be sparse. Here, we review the current state (and gaps) of knowledge on N₂ fixation associated with coral reef organisms and their ecosystems. By summarizing the existing literature, we show that benthic N₂ fixation is an omnipresent process in tropical reef environments. Highest N₂ fixation rates are detected in reef‐associated cyanobacterial mats and sea grass meadows, clearly showing the significance of these functional groups, if present, to the input of new N in reef ecosystems. Nonetheless, key benthic organisms such as hard corals also importantly contribute to benthic N₂ fixation in the reef. Given the usually high coral coverage of healthy reef systems, these results indicate that benthic symbiotic associations may be more important than previously thought. In fact, mutualisms between carbon (C) and N₂ fixers have likely evolved that may enable reef communities to mitigate N limitation. We then explore the potential effects of the increasing human interferences on the process of benthic reef N₂ fixation via changes in diazotrophic populations, enzymatic activities, or availability of benthic substrates favorable to these microorganisms. Current knowledge indicates positive effects of ocean acidification, warming, and deoxygenation and negative effects of increased ultraviolet radiation on the amount of N fixed in coral reefs. Eutrophication may either boost or suppress N₂ fixation, depending on the nutrient becoming limiting. As N₂ fixation appears to play a fundamental role in nutrient‐limited reef ecosystems, these assumptions need to be expanded and confirmed by future research efforts addressing the knowledge gaps identified in this review.     

Top-down effects of insect herbivores during early succession: influence on biomass and plant dominance

We tested the hypothesis that phytophagous insects would have a strong top-down effect on early successional plant communities and would thus alter the course of succession. To test this hypothesis, we suppressed above-ground insects at regular intervals with a broad-spectrum insecticide through the first 3 years of old-field succession at three widely scattered locations in central New York State. Insect herbivory substantially reduced total plant biomass to a similar degree at all three sites by reducing the abundance of meadow goldenrod, Solidago altissima. As a result, Euthamia graminifolia dominated control plots whereas S. altissima dominated insecticide-treated plots by the third year of succession. S. altissima is the dominant old-field herbaceous species in this region but typically requires at least 5 years to become dominant. Past explanations for this delay have implicated colonization limitation whereas our data demonstrate that insect herbivory is a likely alternative explanation. A widespread, highly polyphagous insect, the xylem-tapping spittlebug, Philaenus spumarius, appeared to be the herbivore responsible for the reduction in standing crop biomass at all three sites. Insect herbivory typically caused little direct leaf tissue loss for the ten plant species we examined, including S. altissima. Consequently, the amount of leaf area removed was not a reliable indicator of the influence of insect herbivory on standing crop biomass or on early succession. Overall, we found a strong top-down effect of insect herbivores on biomass at several sites, so our results may be broadly applicable. These findings run counter to generalizations that top-down effects of herbivores, particularly insects, are weak in terrestrial systems. These generalizations may not apply to insects, such as spittlebugs, that can potentially mount an effective defense (i.e., spittle) against predators and subsequently reach relatively high abundance on common plant species. Our results suggest that insect herbivory may play an important but often overlooked role during early old-field succession. Received: 26 December 1998 / Accepted: 3 April 1999     

Relationship between growth,nitrogen fixation and assimilation in a legume (Medicago sativa L.)

Summary Symbiotic N₂ fixation, NO ₃ ⁻ assimilation and protein accumulation in the shoots were measured simultaneously in alfalfa (Medicago sativa L.) grown in the field or in pots, in order to study how the balance between the two modes of nitrogen nutrition could be influenced by agronomic factors, such as harvest, mineral nitrogen supply and drought stress. During periods of rapid growth, fixation and assimilation may function simultaneously; they are antagonistic at the beginning and at the end of the growth cycle, when the nitrogen requirement of the plant is lower. When nitrogen nutrition does not limit growth, mineral nitrogen supply favours assimilation at the expense of fixation, but does not modify the amount of nitrogen accumulated, which is adjusted to the growth capacity of the plant. After cutting, nitrate assimilation compensated for the decrease in fixation and supplied the plant with the nitrogen required by the regrowth, the proliferation of which determined the fixation recovery. Drought stress decreased N₂ fixation much more than NO ₃ ⁻ assimilation. The latter made growth recovery possible when water supply conditions became normal again. These results suggested the existence of an optimum level of nitrate assimilation, which differed depending on the age of the plants and allowed both maximum growth and fixing activity.     

The relationship between N isotopic fractionation within soybean and N2 fixation during soybean development

The contribution of N₂ fixation to overall soybean N uptake has most commonly been quantified by N isotope‐based methods, which rely on isotopic differences in plant N between legumes and non‐fixing reference plants. The choice of non‐fixing reference plants is critical for the accuracy of isotope‐based methods, and mismatched reference plants remain a potential source of error. Accurate estimates of soybean N₂ fixation also require information on N isotopic fractionation within soybean. On the basis of a previous observation of a close correlation between an expression of N fractionation within soybean and the proportion of plant N derived from atmosphere (%Ndfa) determined by ¹⁵N natural abundance, this field study aimed at assessing the relationship between various expressions describing intraplant ¹⁵N or N partitioning and %Ndfa during soybean development. Starting from a late vegetative stage until beginning senescence, the N content and N isotopic composition of shoots, roots and nodules of nodulated and non‐nodulated soybeans was determined at eight different developmental stages. Regression analysis showed that %Ndfa most closely correlated with the difference in the N isotopic composition of shoot N minus that of root including nodule N, and that this relationship was similar to that obtained in a previous multi‐site field study. We therefore consider this expression to hold promise as a means of quantifying %Ndfa independent of a reference plant, which would avoid some of the external sources of error introduced by the use of reference plants in determining %Ndfa.     

Community structures of N2‐fixing bacteria associated with the phyllosphere of a Holm oak forest and their response to drought

Biological nitrogen (N) fixation is a key pathway in terrestrial ecosystems and is therefore critical for understanding the responses of ecosystems to global environmental changes. The free‐living diazotrophic community is distributed along the canopy‐to‐soil profile, but the ecological significance of epiphyllic N₂ fixers, despite their functional relevance, on plant foliar surfaces remains very poorly understood compared with the N₂‐fixing community in forest litter and soils. We assessed the community structure of N₂ fixers and overall bacteria by genetic fingerprinting (t‐RFLP) to explore the seasonal successional patterns of the microbial community in the natural phyllosphere of a Holm oak (Quercus ilex) forest submitted to 12‐year field experiment of rain exclusion mimicking the conditions of drought projected for the coming decades. Leaves of Holm oak were analysed in different seasons over a period of 1.5 years. The bacterial community of the phyllosphere did not correspond to the surrounding soil biome in the same area. These analyses provided field evidence for the presence of free‐living diazotrophs associated with the tissues of leaves of Holm oak, the dominant tree species of many Mediterranean forests. The results also revealed that the community composition is affected seasonally and inter‐annually by the environment, and that the composition shifts in response to climate change. Drought treatment increased the richness of the epiphyllic microbial community, especially during the summer. These changes were associated with higher C:N ratios of leaves observed in response to drought in semiarid areas. This epiphyllic microbiota that can potentially fix N₂ extends the capacity of plants to adapt to the environment.