Search

Plant litter is the major source of energy and nutrients in stream ecosystems and its decomposition is vital for ecosystem nutrient cycling and functioning. Invertebrates are key contributors to instream litter decomposition, yet quantification of their effects and drivers at the global scale remains lacking. Here, we systematically synthesized data comprising 2707 observations from 141 studies of stream litter decomposition to assess the contribution and drivers of invertebrates to the decomposition process across the globe. We found that (1) the presence of invertebrates enhanced instream litter decomposition globally by an average of 74%; (2) initial litter quality and stream water physicochemical properties were equal drivers of invertebrate effects on litter decomposition, while invertebrate effects on litter decomposition were not affected by climatic region, mesh size of coarse-mesh bags or mycorrhizal association of plants providing leaf litter; and (3) the contribution of invertebrates to litter decomposition was greatest during the early stages of litter mass loss (0-20%). Our results, besides quantitatively synthesizing the global pattern of invertebrate contribution to instream litter decomposition, highlight the most significant effects of invertebrates on litter decomposition at early rather than middle or late decomposition stages, providing support for the inclusion of invertebrates in global dynamic models of litter decomposition in streams to explore mechanisms and impacts of terrestrial, aquatic, and atmospheric carbon fluxes. ispartof: BIOLOGICAL REVIEWS vol:97 issue:6 pages:2023-2038 ispartof: location:England status: published

Aims Biomass allocation to different organs is a fundamental plant ecophysiological process to better respond to changing environments; yet, it remains poorly understood how patterns of biomass allocation respond to nitrogen (N) additions across terrestrial ecosystems worldwide. Methods We conducted a meta-analysis using 5474 pairwise observations from 333 articles to assess how N addition affected plant biomass and biomass allocation among different organs. We also tested the ‘ratio-based optimal partitioning’ vs. the ‘isometric allocation’ hypotheses to explain potential N addition effects on biomass allocation. Important Findings We found that (i) N addition significantly increased whole plant biomass and the biomass of different organs, but decreased root:shoot ratio (RS) and root mass fraction (RMF) while no effects of N addition on leaf mass fraction and stem mass fraction at the global scale; (ii) the effects of N addition on ratio-based biomass allocation were mediated by individual or interactive effects of moderator variables such as experimental conditions, plant functional types, latitudes and rates of N addition and (iii) N addition did not affect allometric relationships among different organs, suggesting that decreases in RS and RMF may result from isometric allocation patterns following increases in whole plant biomass. Despite alteration of ratio-based biomass allocation between root and shoot by N addition, the unaffected allometric scaling relationships among different organs (including root vs. shoot) suggest that plant biomass allocation patterns are more appropriately explained by the isometric allocation hypothesis rather than the optimal partitioning hypothesis. Our findings contribute to better understand N-induced effects on allometric relationships of terrestrial plants, and suggest that these ecophysiological responses should be incorporated into models that aim to predict how terrestrial ecosystems may respond to enhanced N deposition under future global change scenarios.

Long-term nitrogen (N) fertilization imposes strong selection on nitrifying communities in agricultural soil, but how a progressively changing niche affects potentially active nitrifiers in the field remains poorly understood. Using a 44-year grassland fertilization experiment, we investigated community shifts of active nitrifiers by DNA-based stable isotope probing (SIP) of field soils that received no fertilization (CK), high levels of organic cattle manure (HC), and chemical N fertilization (CF). Incubation of DNA-SIP microcosms showed significant nitrification activities in CF and HC soils, whereas no activity occurred in CK soils. The 44 years of inorganic N fertilization selected only 13C-ammonia-oxidizing bacteria (AOB), whereas cattle slurry applications created a niche in which both ammonia-oxidizing archaea (AOA) and AOB could be actively 13C-labeled. Phylogenetic analysis indicated that Nitrosospira sp. 62-like AOB dominated inorganically fertilized CF soils, while Nitrosospira sp. 41-like AOB were abundant in organically fertilized HC soils. The 13C-AOA in HC soils were affiliated with the 29i4 lineage. The 13C-nitrite-oxidizing bacteria (NOB) were dominated by both Nitrospira- and Nitrobacter-like communities in CF soils, and the latter was overwhelmingly abundant in HC soils. The 13C-labeled nitrifying communities in SIP microcosms of CF and HC soils were largely similar to those predominant under field conditions. These results provide direct evidence for a strong selection of distinctly active nitrifiers after 44 years of different fertilization regimes in the field. Our findings imply that niche differentiation of nitrifying communities could be assessed as a net result of microbial adaption over 44 years to inorganic and organic N fertilization in the field, where distinct nitrifiers have been shaped by intensified anthropogenic N input.

In this study, we show that composition shifts of ammonia oxidizer communities imposed by a 22-year field fertilization regime could be well retained in both fresh and air-dried soils. The abundance and composition of ammonia-oxidizing bacteria (AOB) and archaea (AOA) were measured in fresh soils, which received no fertilization (CK), chemical fertilization (NPK), and chemical plus organic matter fertilization (NPK/OM) for 22 years. The air-drying treatment of fresh soil was also conducted for pairwise analysis. We found that in fresh soils DGGE fingerprints of AOB showed significant changes under both NPK and NPK/OM treatments when compared with control (CK) and that microbial shift was almost identical in air-dried soils. Long-term nutrient fertilization did not affect AOA communities in either air-dried or fresh soils. Compared to CK treatment, real-time PCR indicated that AOB amoA genes increased significantly in fresh soils of NPK (59-fold) and NPK/OM (48-fold) plots and in air-dried NPK and NPK/OM soils by 22-fold and 19-fold respectively. Our results demonstrate that community shifts of AOB in fresh soils under chronic N fertilization could be well preserved in air-dried soils, despite the apparent decline in absolute abundance of ammonia oxidizers. These results suggest that air-dried soil could be a useful resource for deciphering the adaptive strategy of ammonia oxidizers under N enrichment when the significant changes of community composition occurred in fresh soils.

Plant roots can significantly alter soil pH and the chemical concentration and distribution of different elements in the rhizosphere environment. Here we ask whether cadmium (Cd) bioavailability in the rhizosphere of Cd-hyperaccumulator Sedum plumbizincicola can be influenced by root-induced effects on soil pH. The Cd-hyperaccumulator S. plumbizincicola and the Cd non-hyperaccumulator ecotype Sedum alfredii were both grown in four different Cd-contaminated soils. We used the planar optode imaging technique to produce two-dimensional and high-resolution measurements of soil pH. Shoot excess cation concentration, root architecture and Cd concentrations ([Cd]) in soil pore water were also measured. Spatial analyses based on kernel density estimate of roots (KDE) and a Moran’s I correlogram were performed to assess spatial patterns and potential relationships among root distribution, soil pH and [Cd]. Both Sedum species showed root-induced increases in soil acidification (i.e. soil pH decreases of 0.1 to 0.62 units), which were clearly associated with greater root density of these plants. Remarkable excess cation uptakes by both Sedum species were detected and likely a driving factor for the root-induced acidification. The presence of the roots of S. plumbizincicola were then related to higher [Cd] in the rhizosphere than in bulk soil in Orthic Acrisol (+342%) and in Hydragric Antrosol soils (+296%). The hyperaccumulator S. plumbizincicola had larger root systems, higher acidification ability, and was associated with greater soil [Cd] than S. alfredii. Spatial patterns of root distribution and soil pH were similar between Sedum plants, however, spatial patterns of [Cd] differed across polluted soils. Rhizosphere acidification induced by S. plumbizincicola plants can play an important role on soil Cd mobilization, but overall effects on soil Cd bioavailability will depend on intrinsic soil biogeochemical properties.

Human-managed grasslands receive significant inputs of fertilizing materials, which can greatly influence soil biological processes associated with the cycling of nitrogen (N), including microbial nitrification. Here we specifically address how soil ammonia oxidizing archaea (AOA) and ammonia oxidizing bacteria (AOB) respond to 23 years of different management practices in a permanent grassland experiment. We found that AOB amoA gene copy numbers were significantly higher in limed soils (associated with greater pH values) whereas AOA amoA gene copy numbers were higher in grazed grasslands. The chronic addition of inorganic N fertilizer in the form of ammonium nitrate (either applied alone or in combination with other macro-nutrients) greatly contributed to increase AOB abundance. Our study brings evidence of how soil-nitrifying communities can differently respond to the long-term effect of animal (i.e. rabbit) grazing and to repeated applications of nutrient (e.g. NH4NO3) and non-nutrient (i.e. CaCO3) fertilizing materials.

Long-term nutrient fertilization of grassland soils greatly increases plant yields but also profoundly alters ecosystem phosphorus (P) dynamics. Here, we addressed how long-term P fertilization may affect ecosystem P budget, P use efficiency (PUE) and the abundance of arbuscular mycorrhizal fungi (AMF), which play a key role in the acquisition of P by plants. We found that 47 years of organic P applications increased soil P availability and total soil P stocks up to 1600% and 400%, respectively, compared to unfertilized-control soils. Grassland soils could retain up to 62% and 48% of P applied since 1970 in organic and inorganic forms, respectively. Nutrient treatments significantly affected rates of AMF root colonization (%), which were higher in control and NPK-fertilized plots when compared to soils receiving increasing applications of organic P. Plant PUE increased with greater AMF root colonization, which remained high (i.e. 50-to-75%) even after ~50 years of continuous ‘normal’ rates of agronomic P inputs (~30 kg P ha−1 year−1). AMF abundance, however, decreased under higher P applications and we found a negative relationship between soil P availability or soil P stocks and rates of AMF root colonization. Our study demonstrates that (1) AMF root colonization is still high in soils, which have received consistent but moderate P inputs for over four decades, and (2) moderate rates of P fertilization are related to a more conservative P ecosystem budget whereby the amount of P retained in soils and up-taken by plants on an annual basis is higher than the amount of P added through fertilization. This is possible only if extra P is ‘mined’ from the soil P ‘bank’ and made available to plant uptake. We suggest that AMF could play a significant role in intensively-managed grasslands contributing to increase P sustainability by reducing the need for extra P fertilizer.

Summary The productivity of species-diverse plant assemblages strongly depends on the temporal dynamics of nutrient uptake by competing neighbouring plants. Our understanding, however, of how rates of nitrogen (N), phosphorous (P) and potassium (K) uptake might change through time between neighbouring plant species under field conditions is still very limited. Here, we specifically measure the temporal trajectories of N, P and K uptake by staple food plants such as wheat (Triticum aestivum L.), barley (Hordeum vulgare L.) and maize (Zea mays L.) when growing either in monocultures or in intercropping systems. We ask how (i) plant species combinations, (ii) N fertilization and (iii) film mulching might affect key indexes of N, P and K uptake over time. We fit logistic models to characterize the nutrient uptake trajectories. Maximum cumulative N, P and K uptake (kg ha−1) by wheat and barley were significantly greater in wheat–maize or barley–maize intercropping systems than in wheat or barley monocultures. Cumulative nutrient uptake by intercropped maize (either with wheat or with barley) was reduced by interspecific competition at early growth stages, but it increased rapidly after wheat and barley were harvested. Maximum cumulative N and P (but not K) uptake by intercropped maize were significantly higher than the uptake by monoculture maize, particularly when N fertilizer or film mulching was applied. Intercropping induced a significant temporal niche differentiation in maximum daily nutrient uptake rates (kg ha−1 day−1) between intercropped species. Fertilization had much stronger effects on maximum cumulative nutrient uptake of maize than that of wheat and barley. Mulching significantly increased the maximum cumulative nutrient uptake of maize and advanced the time to reach its maximum daily P and K uptake rates. Our study provides evidence of an important temporal niche differentiation mechanism (‘temporal complementarity’) in nutrient uptake rates between neighbouring plant species. A better understanding of temporal trajectories of interspecific nutrient uptake rates remains crucial if we want to maximize the nutrient-use efficiency and sustain overyielding (i.e. high food production) in plant species-diverse systems such as intercropping.

Important biochemical reactions in soils are catalyzed by extracellular enzymes, which are synthesized by microbes and plant roots. Although enzyme activities can significantly affect the decomposition of soil organic matter and thus influence the storage and cycling of carbon (C) and nitrogen (N), it is not clear how enzyme activities relate to changes in the C and N content of different grassland soils. Here we address whether the activity of C-acquiring (b-1,4-glucosidase, BG) and N-acquiring (L-leucine amino-peptidase (LAP) and b-1,4-N-acetyl-glucosaminidase (NAG)) enzymes is linked to changes in the C and N content of a variety of human-managed grassland soils. We selected soils which have a well-documented management history going back at least 19 years in relation to changes in land use (grazing, mowing, ploughing), nutrient fertilization and lime (CaCO 3) applications. Overall we found a positive relationship between BG activity and soil C content as well as between LAP þ NAG activity and soil N. These positive relationships occurred across grasslands with very different soil pH and management history but not in intensively managed grasslands where increases in soil bulk density (i.e. high soil compaction) negatively affected enzyme activity. We also found evidence that chronic nutrient fertilization contributed to increases in soil C content and this was associated with a significant increase in BG activity when compared to unfertilized soils. Our study suggests that while the activities of C-and N-acquiring soil enzymes are positively related to soil C and N content, these activities respond significantly to changes in management (i.e. soil compaction and nutrient fertilization). In particular, the link between BG activity and the C content of long-term fertilized soils deserves further investigation if we wish to improve our understanding of the C sequestration potential of human-managed grassland soils.

Plant elemental composition can indicate resource limitation, and changes in key elemental ratios (e.g. plant C:N ratios) can influence rates including herbivory, nutrient recycling, and pathogen infection. Although plant stoichiometry can influence ecosystem-level processes, very few studies have addressed whether and how plant C:N stoichiometry changes with plant diversity and composition. Here, using two long-term experimental manipulations of plant diversity (Jena and Cedar Creek), we test whether plant richness (species and functional groups) or composition (functional group proportions) affects temporal trends and variability of community-wide C:N stoichiometry. Site fertility determined the initial community-scale C:N ratio. Communities growing on N-poor soil (Cedar Creek) began with higher C:N ratios than communities growing on N-rich soil (Jena). However, site-level plant C:N ratios converged through time, most rapidly in high diversity plots. In Jena, plant community C:N ratios increased. This temporal trend was stronger with increasing richness. However, temporal variability of C:N decreased as plant richness increased. In contrast, C:N decreased over time at Cedar Creek, most strongly at high species and functional richness, whereas the temporal variability of C:N increased with both measures of diversity at this site. Thus, temporal trends in the mean and variability of C:N were underlain by concordant changes among sites in functional group proportions. In particular, the convergence of community-scale C:N over time at these very different sites was mainly due to increasing proportions of forbs at both sites, replacing high mean C:N (C4 grasses, Cedar Creek) or low C:N (legumes, Jena) species. Diversity amplified this convergence; although temporal trends differed in sign between the sites, these trends increased in magnitude with increasing species richness. Our results suggest a predictive mechanistic link between trends in plant diversity and functional group composition and trends in the many ecosystem rates that depend on aboveground community C:N. Synthesis We compared the effect of plant diversity on the temporal dynamics of community stoichiometry in two long-term grassland diversity experiments: the Cedar Creek and Jena Experiments. Changes in community C:N ratios were accelerated by increasing diversity at both sites, but in opposite directions depending on soil fertility. Stoichiometry changes were driven by shifts of functional group composition differing in their elemental compositions, the identity of the functional groups depending on the site. Thus, we highlighted that community turnover constrained the effect of diversity on plant stoichiometry at both sites.

Chronic nutrient addition to grassland soils can strongly influence the composition and abundance of nitrifying microbial communities. Despite the fact that nitrifying microbes play a crucial role in regulating ecosystem nitrogen (N) cycling, our understanding of how long-term N fertilization might influence nitrifying microbial groups is still limited. Here we used soil from a 44-year-old grassland fertilization experiment and performed high-throughput pyrosequencing analyses (and real-time quantitative PCR) to determine whether and how the identity and abundance of nitrifying microbes has changed in response to chronic inorganic (chemical fertilizer) and organic (cattle slurry) N additions. We found that the amoA genes of ammonia-oxidizing archaea (AOA) significantly increased under organic N additions, whereas ammonia-oxidizing bacteria (AOB) increased with the addition of inorganic N. Proportional changes of AOA, AOB and nitrite-oxidizing bacteria (NOB) demonstrate that nitrifying phylotypes are influenced by chronic N additions. We also found that AOA/AOB ratios increased with higher application rates of cattle slurry suggesting that AOA may affect N cycling more in soils receiving animal manures, whereas AOB are functionally more important in chemically fertilized soils. Phylogenetic analysis shows that shifts in AOA and AOB community structure occurred through time across N fertilization treatments. For example, (a) fosmid 29i4-like AOA, (b) Nitrosospira cluster 3-like AOB, and (c) Nitrospira-like NOB dominated nitrifying communities in fertilized soils. Finally, high-throughput pyrosequencing of 16S rRNA genes show that N fertilization (either inorganic or organic) increased the abundance of Actinobacteria in soils while it decreased the abundance of Proteobacteria. Our study is one of the first to show that long-term N additions to soils can greatly affect nitrifying communities, and that phylogenetically and functionally distinct nitrifiers have developed through time in response to chronic N fertilization.

Intercropping is commonly practiced worldwide because of its benefits to plant productivity and resource-use efficiency. Belowground interactions in these species-diverse agro-ecosystems can greatly contribute to enhancing crop yields; however, our understanding remains quite limited of how plant roots might interact to influence crop biomass, photosynthetic rates, and the regulation of different proteins involved in CO2 fixation and photosynthesis. We address this research gap by using a pot experiment that included three root-barrier treatments with full, partial and no root interactions between foxtail millet (Setaria italica (L.) P.Beauv.) and peanut (Arachis hypogaea L.) across two growing seasons. Biomass of millet and peanut plants in the treatment with full root interaction was 3.4 and 3.0 times higher, respectively, than in the treatment with no root interaction. Net photosynthetic rates also significantly increased by 112–127% and 275–306% in millet and peanut, respectively, with full root interaction compared with no root interaction. Root interactions (without barriers) contributed to the upregulation of key proteins in millet plants (i.e. ribulose 1,5-biphosphate carboxylase; chloroplast β-carbonic anhydrase; phosphoglucomutase, cytoplasmic 2; and phosphoenolpyruvate carboxylase) and in peanut plants (i.e. ribulose 1,5-biphosphate carboxylase; glyceraldehyde-3-phosphate dehydrogenase; and phosphoglycerate kinase). Our results provide experimental evidence of a molecular basis that interspecific facilitation driven by positive root interactions can contribute to enhancing plant productivity and photosynthesis.

The relationship between plant diversity and productivity in grasslands could depend, partly, on how diversity affects vertical distributions of root biomass in soil; yet, no prior study has evaluated the links among diversity, root depth distributions, and productivity in a long-term experiment. We use data from a 12-year experiment to ask how plant species richness and composition influenced both observed and expected root depth distributions of plant communities. Expected root depth distributions were based on the abundance of species in each community and two traits of species that were measured in monocultures: root depth distributions and root to shoot ratios. The observed proportion of deep root biomass increased more than expected with species richness and was positively correlated with aboveground productivity. Indeed, the proportion of deep root biomass explained variation in productivity even after accounting for legume presence/abundance, and greater nitrogen availability in diverse plots. Diverse plots had root depth distributions that were twice as deep as expected from their species composition and corresponding monoculture traits, partly due to interactions between C4 grasses and legumes. These results suggest the productivity of diverse plant communities was partly dependent on belowground plant interactions that caused roots to be distributed more deeply in soil.

Two primary approaches to perennial biofuel crop production studied so far are fertilized grassmonocultures and low-input high-diversity grasslands. While high-yielding perennial grass varieties arebeing developed in fertilized monocultures, breeding for yield in low-input high-diversity systemswould be difficult. Before initiating breeding for low-input systems, it is therefore important to know theminimum number of functional groups and species required for maximum biomass harvest from lowinputgrasslands. We controlled the number of perennial grassland species in 168 plots in Minnesota,USA. Species were selected at random from a pool of 18, and 1, 2, 4, 8, or 16 were planted in each plot.Aboveground biomass was measured annually, and the plots were burned each spring. We found astrongly positive log-linear relationship between average annual aboveground biomass and plantedspecies number, but a large proportion of plot-to-plot variability remained unexplained. Weperformed aconditional analysis of the aboveground biomass data to determine whether considering species identitywould reduce the minimum number of species necessary in order to achieve yields similar to the highestdiversity treatments. A model that accounted for the presence of legumes in general, and for the presenceof the legume species Lupinus perennis in particular, showed no increase in biomass yield with increasedspecies number. Over 11 years, average yields of L. perennis/C4 grass bicultures were similar to those of16-species (maximum diversity) plots, and both were >200% greater than the average of monocultures.Thus, under low-input conditions, the choice of the appropriate few perennial plant species for eachlocation might result in systems with biomass yields similar to those from high-diversity systems.Because breeding biofuel crops in diverse mixtures would introduce complexity that is unwarranted interms of maximum biomass yield, the first biofuel crop breeding programs for low-input systems arelikely to accelerate progress by focusing on grass–legume bicultures.

In a 13-year grassland biodiversity experiment in Minnesota, USA, we addressed two main questions: What set of ecological mechanisms caused aboveground productivity to become approximately 340% greater in highly diverse plant mixtures than in the average monoculture? Why did the effect of diversity on productivity become so much stronger through time? Because our grassland system is N limited, we simultaneously measured critical variables associated with the storage and cycling of this element, such as plant and soil N pools, soil N availability, soil N mineralization rates, and plant N-use efficiency, as well as the initial soil N concentration of each diversity plot when the experiment was established in 1994. We used linear and multiple regression analyses to test for potential effects of these variables on aboveground productivity and to address whether and how such variables were in turn affected by plant species diversity and functional composition across years and also at different time intervals within the same year. We found that seven variables simultaneously controlled productivity: (1) initial total soil nitrogen (N) of each plot, (2) diversity-dependent increases in total soil N through time, (3) soil N mineralization rates, (4) soil nitrate (NO3-) utilization, (5) increases in plant N-use efficiency at greater plant diversity, (6) legume presence, and (7) higher species numbers. The surprising continued significance of higher plant diversity may occur because of its effects on seasonal capture of soil NO3- and moisture and on the accumulation of root-N pools, all of which may have also increased productivity through time at higher species numbers.

Summary 1. Plant functional composition may indirectly affect fine root processes both qualitatively (e.g. by influencing root chemistry) and quantitatively (e.g. by influencing root biomass and thus soil carbon (C) inputs and the soil environment). Despite the potential implications for ecosystem nitrogen (N) cycling, few studies have addressed the linkages between plant functional composition, root decay, root detritus N dynamics and soil N mineralization rates. 2. Here, using data from a large grassland biodiversity experiment, we first show that plant functional composition affected fine root mass loss, root detritus N dynamics and net soil N mineralization rates through its effects on root chemistry rather than on the environment of decomposition. In particular, the presence of legumes and non-leguminous forbs contributed to greater fine root decomposition which in turn enhanced root N release and net soil N mineralization rates compared with C3 and C4 grasses. 3. Second, we show that all fine roots released N immediately during decomposition and showed very little N immobilization regardless of plant composition. As a consequence, there was no evidence of increased root or soil N immobilization rates with increased below-ground plant biomass (i.e. increased soil C inputs) even though root biomass negatively affected root decay. 4. Our results suggest that fine roots represent an active soil N pool that may sustain plant uptake while other soil N forms are being immobilized in microbial biomass and/or sequestered into soil organic matter. However, fine roots may also represent a source of recalcitrant plant detritus that is returned to the soil (i.e. fine roots of C4 and C3 grasses) and that can contribute to an increase in the soil organic matter pool. 5. Synthesis . An important implication of our study is that the simultaneous presence of different plant functional groups (in plant mixtures) with opposite effects on root mass loss, root N release and soil N mineralization rates may be crucial for sustaining multiple ecosystem services such as productivity and soil C and N sequestration in many N-limited grassland systems.

Most explanations for the positive effect of plant species diversity on productivity have focused on the efficiency of resource use, implicitly assuming that resource supply is constant. To test this assumption, we grew seedlings of Echinacea purpurea in soil collected beneath 10-year-old, experimental plant communities containing one, two, four, eight, or 16 native grassland species. The results of this greenhouse bioassay challenge the assumption of constant resource supply; we found that bioassay seedlings grown in soil collected from experimental communities containing 16 plant species produced 70% more biomass than seedlings grown in soil collected beneath monocultures. This increase was likely attributable to greater soil N availability, which had increased in higher diversity communities over the 10-year-duration of the experiment. In a distinction akin to the selection/complementarity partition commonly made in studies of diversity and productivity, we further determined whether the additive effects of functional groups or the interactive effects of functional groups explained the increase in fertility with diversity. The increase in bioassay seedling biomass with diversity was largely explained by a concomitant increase in N-fixer, C4 grass, forb, and C3 grass biomass with diversity, suggesting that the additive effects of these four functional groups at higher diversity contributed to enhance N availability and retention. Nevertheless, diversity still explained a significant amount of the residual variation in bioassay seedling biomass after functional group biomass was included in a multiple regression, suggesting that interactions also increased fertility in diverse communities. Our results suggest a mechanism, the fertility effect, by which increased plant species diversity may increase community productivity over time by increasing the supply of nutrients via both greater inputs and greater retention.

We studied the composition of a savanna woody plant community across a natural herbivory gradient maintained by both browsing and grazing ungulates in an arid part of the Kruger National Park, South Africa. We focused on (1) short-term browsing effects on reproductive and morphological traits of a dominant-palatable woody species, Acacia nigrescens, Miller, (2) the relationship between browsing–grazing intensity and soil parameters, (3) the effects of herbivore–soil interactions on woody species richness and composition, and (4) browser-induced effects on the representation of woody plant functional traits. We show that the number of pods carried by A. nigrescens trees as well as the internode length of external tree branches both decreased significantly at high browsing intensity. Moreover, we found that total soil nitrogen (N) and soil cation (Ca, Na, Mg, and K) concentrations varied significantly according to grazing rather than browsing intensity with soil nutrients decreasing at heavily grazed sites. Although herbivory and soil

Suppressed growth forms of woody species are common where fire and herbivory are major ecosystem drivers, such as in African savannas. Nevertheless, despite their importance in maintaining plant population viability, woody plants in suppressed growth forms have received little attention in ecological studies. We measured a set of morpho-functional traits and investigated plant density variation in suppressed growth forms (hereafter we refer to them as saplings) of two common species, Acacia nigrescens and Acacia tortilis, across sites that had undergone very different histories of attack from large herbivores while fire had been absent for at least 13 y. We show that heavily browsed saplings have higher regrowth abilities, twice the number of stems produced by the main root crown and twice the root diameter at 5 cm soil depth, than lightly browsed saplings. This suggests that Acacia saplings are resilient to chronic herbivory and show high morphological plasticity. However, we show that mammalian herbivores can strongly limit sapling recruitment to mature size classes and possibly affect variation in sapling density between heavily and lightly browsed sites. Further studies should investigate whether the persistence of the sapling bank can be ascribed to the ``storage effect'' by which a plant's reproductive potential is ``stored'' in a suppressed growth form until a window of opportunity allows rapid maturity.

Summary 1. The mechanisms controlling soil carbon (C) and nitrogen (N) accumulation are crucial for explaining why soils are major terrestrial C sinks. Such mechanisms have been mainly addressed by imposing short-term, step-changes in CO 2 , temperature and N fertilization rates on either monocultures or low-diversity plant assemblages. No studies have addressed the long-term effects of plant functional diversity (i.e. plant functional composition) on rates of soil C accumulation in N-limited grasslands where fixation is the main source of N for plants. 2. Here we measure net soil C and N accumulation to 1 m soil-depth during a 12-year-long grassland biodiversity experiment established on agriculturally degraded soils at Cedar Creek, Minnesota, USA. 3. We show that high-diversity mixtures of perennial grassland plant species stored 500% and 600% more soil C and N than, on average, did monoculture plots of the same species. Moreover, the presence of C4 grasses and legumes increased soil C accumulation by 193% and 522%, respectively. Higher soil C and N accrual resulted both from increased C and N inputs via (i) higher root biomass, and (ii) from greater root biomass accumulation to 60 cm soil depth resulting from the presence of highly complementary functional groups (i.e. C4 grasses and legumes). 4. Our results suggest that the joint presence of C4 grass and legume species is a key cause of greater soil C and N accumulation in both higher and lower diversity plant assemblages. This is because legumes have unique access to N, and C4 grasses take up and use N efficiently, increasing below-ground biomass and thus soil C and N inputs. 5. Synthesis. We demonstrate that plant functional complementarity is a key reason why higher plant diversity leads to greater soil C and N accumulation on agriculturally degraded soils. We suggest the combination of key C4 grass‐legume species may greatly increase ecosystem services such as soil C accumulation and biomass (biofuel) production in both high- and low-diversity N-limited grassland systems.

We investigated the linkages between leaf litter quality and decomposability in a savanna plant community dominated by palatable-spinescent tree species. We measured: (1) leaf litter decomposability across five woody species that differ in leaf chemistry; (2) mass decomposition, nitrogen (N); and carbon (C) dynamics in leaf litter of a staple browse species (Acacia nigrescens) as well as (3) variation in litter composition across six sites that experienced very different histories of attack from large herbivores. All decomposition trials included litter bags filled with chopped straw to control for variation in site effects. We found a positive relationship between litter quality and decomposability, but we also found that Acacia and straw litter mass remaining did not significantly vary between heavily and lightly browsed sites. This is despite the fact that both the quality and composition of litter returned to the soil were significantly different across sites. We observed greater N resorption from senescing Acacia leaves at heavily browsed sites, which in turn contributed to increase the C:N ratio of leaf litter and caused greater litter N immobilization over time. This, together with the significantly lower tree- and herb-leaf litter mass beneath heavily browsed trees, should negatively affect decomposition rates. However, estimated dung and urine N deposition from both browsers and grazers was significantly greater at high- than at low-herbivory sites. We hypothesize that N inputs from dung and urine boost litter N mineralization and decomposition (especially following seasonal rainfall events), and thereby offset the effects of poor leaf litter quality at chronically browsed sites.

We measured browsing-induced responses of Acacia trees to investigate "browsing lawns" as an analogy to grazing lawns in a semiarid eutrophic African savanna. During the two-year field study, we measured plant tolerance, resistance, and phenological traits, while comparing variation in leaf nitrogen and specific leaf area (SLA) across stands of Acacia nigrescens, Miller, that had experienced markedly different histories of attack from large herbivores. Trees in heavily browsed stands developed (1) tolerance traits such as high regrowth abilities in shoots and leaves, high annual branch growth rates, extensive tree branching and evidence of internal N translocation, and (2) resistance traits such as close thorn spacing. However, phenological "escape" responses were weak even in heavily browsed stands. Overall, browsing strongly affected plant morpho-functional traits and decreased both the number of trees carrying pods and the number of pods per tree in heavily browsed stands. Hence, there is experimental evidence that tolerance and resistance traits may occur simultaneously at heavily browsed sites, but this comes at the expense of reproductive success. Such tolerance and resistance traits may coexist if browsers trigger and maintain a positive feedback loop in which trees are continually investing in regrowth (tolerance), and if the plant's physical defenses (resistance) are not nutritionally costly and are long-lived. Our results confirm that chronic browsing by ungulates can maintain A. nigrescens trees in a hedged state that is analogous to a grazing lawn. Further research is needed to fully understand the long-term effects of chronic browsing on reproduction within such tree populations, as well as the overall effects on nutrient cycling at the ecosystem level.

Many tropical pioneer species depend on the presence of high seed densities in the soil for successful recruitment following canopy disturbance (Cheke et al. 1979, Dalling & Hubbell 2002, Guevara Sada & Gómez Pompa 1972, Whitmore 1983). However determinants of variation in the composition and abundance of soil seed banks remain poorly understood. Seed bank densities can be affected by rates of seed predation and pathogen infection on the surface and in the soil, by intrinsic rates of loss in viability following dispersal, and by variation in the timing and duration of fruit production (Dalling et al. 1997, Garwood 1983, Murray & Garcia 2002). Here we compare seasonal fluctuations in seed bank density in five Panamanian forests varying in elevation and seasonality of precipitation (Table 1). We predict that lowland forests should show stronger intra-annual fluctuation in seed bank densities than montane forests because seed production and loss rates should be higher under conditions of greater resource availability, and where consistent high temperatures support greater abundance or activity of seed predators and pathogens (Brühl et al. 1999). Secondly, among lowland sites, we predict greater fluctuations in seed bank densities at drier, more seasonal sites where seasonally favourable conditions for seedling recruitment may select for interspecific synchrony in fruit production (Daubenmire 1972, Garwood 1983).

Variation among forests in environmental and biotic conditions may strongly influence seed fate with important consequences for the abundance and distribution of plant species. Here we examine the post-dispersal seed removal rates of six pioneer species (Cecropia peltata, Miconia argentea, Luehea seemannii, Trema micrantha, Apeiba aspera and Jacaranda copaia) from the soil surface at five sites in Panama varying in elevation (0–1100 m) and seasonality (0–4 mo dry season). We compared removal rates of washed seeds placed in vertebrate exclosures, invertebrate exclosures, and unprotected controls in January and June. Overall, removal rates of unprotected seeds were similar among sites. Almost all seed removal could be attributed to litter ants in two subfamilies (Myrmicinae and Ponerinae). Little or no removal was recorded from invertebrate exclosures while vertebrate exclosures had no effect on removal either in lowland and montane forests. Seed removal rates were high for four animal-dispersed species (mean 45% removed over 2 d), whereas two wind-dispersed species were largely untouched (mean 2% removed). These results indicate that seed dispersal characteristics, rather than site characteristics, may be the strongest determinant of the post-dispersal seed fate of pioneers.

Biomass allocation to plant aboveground and belowground compartments may change in response to changes in soil nutrient fertility (e.g., nitrogen (N) and phosphorous (P) availability). It is not clear however whether and how changes in soil fertility could influence carbon(C):nutrient stoichiometry of wild plants distributed along environmental gradients. Here we use ecological stoichiometric theory to test two hypotheses: (1) C:N (or C:P) ratios of different plant compartments (i.e., roots, stems and leaves) increase when soil N (or P) availability decreases, (2) the relative availability of N compared to P in soils predictably influences C:nutrient ratios (e.g., high soil N:P ratios are associated with high plant C:P ratios). Data from 72 wild plant species were collected along a gradient of soil development determined by the temporal progression of four primary ecological successions across Europe. We found significant changes in soil N and P availability and content along the four vegetation successio...