A synthesis of subdisciplines: predator–prey interactions, and biodiversity and ecosystem functioning

The last 15 years has seen parallel surges of interest in two research areas that have rarely intersected: biodiversity and ecosystem functioning (BEF), and multispecies predator–prey interactions (PPI). Research addressing role of biodiversity in ecosystem functioning has focused primarily on single trophic‐level systems, emphasizing additive effects of diversity that manifest through resource partitioning and the sampling effect. Conversely, research addressing predator–prey interactions has focused on two trophic‐level systems, emphasizing indirect and non‐additive interactions among species. Here, we use a suite of consumer‐resource models to organize and synthesize the ways in which consumer species diversity affects the densities of both resources and consumer species. Specifically, we consider sampling effects, resource partitioning, indirect effects caused by intraguild interactions and non‐additive effects. We show that the relationship between consumer diversity and the density of resources and consumer species are broadly similar for systems with one vs. two trophic levels, and that indirect and non‐additive interactions generally do little more than modify the impacts of diversity established by the sampling effect and resource partitioning. The broad similarities between systems with one vs. two trophic levels argue for greater communication between researchers studying BEF, and researchers studying multispecies PPI.     

Complex phenotype-environment associations revealed in an East African cyprinid

Environmental factors influence phenotypes directly, as well as indirectly via trait correlations and interactions with other environmental variables. Using nine populations of the African cyprinid Barbus neumayeri, we employed path analysis to examine direct, indirect and total effects of two environmental variables, water flow (WF) and dissolved oxygen (DO), on several morphological traits. WF and DO directly influenced relative gill size, body shape and caudal fin shape in manners consistent with a priori predictions. Indirect effects also played an important role in the system: (1) strong, oppositely signed direct and indirect effects of WF on body shape resulted in a nonsignificant total effect; (2) DO had no direct effect on body shape, but a strong total effect via indirect effects on gill size; (3) WF indirectly influenced gill size via effects on DO. Only through examination of multiple environmental parameters and multiple traits can we hope to understand complex relationships between environment and phenotype.     

Temperature alters the relative abundance and population growth rates of species within the Dendroctonus frontalis (Coleoptera: Curculionidae) community

Temperature has strong effects on metabolic processes of individuals and demographics of populations, but effects on ecological communities are not well known. Many economically and ecologically important pest species have obligate associations with other organisms; therefore, effects of temperature on these species might be mediated by strong interactions. The southern pine beetle (Dendroctonus frontalis Zimmermann) harbors a rich community of phoretic mites and fungi that are linked by many strong direct and indirect interactions, providing multiple pathways for temperature to affect the system. We tested the effects of temperature on this community by manipulating communities within naturally infested sections of pine trees. Direct effects of temperature on component species were conspicuous and sometimes predictable based on single-species physiology, but there were also strong indirect effects of temperature via alteration of species interactions that could not have been predicted based on autecological temperature responses. Climatic variation, including directional warming, will likely influence ecological systems through direct physiological effects as well as indirect effects through species interactions.     

Richness–productivity relationships between trophic levels in a detritus-based system: significance of abundance and trophic linkage

Most theoretical and empirical studies of productivity–species richness relationships fail to consider linkages among trophic levels. We quantified productivity–richness relationships in detritus-based, water-filled tree-hole communities for two trophic levels: invertebrate consumers and the protozoans on which they feed. By analogy to theory for biomass partitioning among trophic levels, we predicted that consumer control would result in richness of protozoans in the lower trophic level being unaffected by increases in productivity, whereas richness of invertebrate consumers would increase with productivity. Our data were consistent with this prediction: consumer richness increased linearly, but protozoan richness was unrelated to changes in productivity. The productivity–richness relationships for all taxa combined were not necessarily consistent with relationships within each trophic level. We used path analysis to investigate the mechanisms that may produce the observed responses of trophic levels to changes in productivity. We tested the importance of the direct effect of productivity on richness and the indirect effect of productivity mediated by effects on total abundance. For protozoans, only direct effects of productivity on richness were important, but both direct and indirect effects of productivity on richness were important for invertebrates. Protozoan richness was strongly affected by top-down impacts of abundance of invertebrates. These results are consistent with theory on biomass partitioning among trophic levels and suggest a strong link between richness and abundance within and between trophic levels. Understanding how trophic level interactions determine productivity–richness relationships will likely be necessary in order for us to achieve a comprehensive understanding of the determinants of diversity. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Population dynamical consequences of reduced predator switching at low total prey densities

Several models of rapid switching by a predator in a two-prey environment are analyzed. The goal is to determine how the dynamics of the system and the potential indirect effects between prey are affected by the dependence of switching on total prey density. In exploring this question, the difference between the population-level consequences of switching in stable and cycling predator-prey systems is also examined. We concentrate on reduced switching at low densities, a feature that is likely because of the difficulty of distinguishing between two very low densities. The main findings are: (1) switching in unstable systems can produce positive indirect effects of one prey species on the other; and (2) reduced switching at low densities can greatly alter the dynamics of the system and the indirect effects between prey. Both of the possibilities are only evident in cycling systems. Reduced switching at low total prey densities leads to heavier predation on the slower-growing prey when both prey species are rare. As a consequence, there is a lag in the recovery of the slower-growing prey species after predator densities fall, and the dynamics of the two prey become desynchronized. The net result is increased indirect interactions between prey, and a greater likelihood of exclusion of the slower growing prey. The analysis of these models suggests a need for more empirical work to determine whether switching is reduced by very low total prey densities, and to study the long-term dynamics that occur in systems with switching predators.     

Indirect effects of Atrina zelandica on water column nitrogen and oxygen fluxes: The role of benthic macrofauna and microphytes

The effects of ecosystem engineers can be mediated by direct and indirect interactions. For suspension-feeding bivalves that contribute to benthic-pelagic coupling, indirect effects have been linked to hydrodynamics; however, the influence of these ecosystem engineers may also operate through interactions with surrounding sediments, microphytes and macrofauna that, in turn, affect nutrient and oxygen fluxes. This study investigated the indirect effects of an epibenthic suspension-feeding bivalve (Atrina zelandica) on ammonium and nitrate-nitrite effluxes from the sediment, and oxygen consumption in the overlying waters, under dark conditions, at two sites with different environmental characteristics. Location-dependent effects were observed in the relative strength of the effect of Atrina on microphyte and macrofaunal abundance. The difference between the strength of the effect of Atrina on macrofauna between sites was not driven by a single species or type of species; rather all the species decreasing in abundance away from Atrina were small species that utilised the sediment water interface. Location-dependent effects were also observed in the relative strength of the effect of microphyte and macrofaunal abundance on oxygen and nutrient fluxes. While microphytes were an important consumer of oxygen at one site, at the other site, small infaunal macrofauna were important. Similarly, the strength of the effect of surrounding macrofauna on ammonium efflux varied between sites. These findings demonstrate the importance of natural history and species functions for understanding complex responses. They suggest that indirect effects by key benthic macrofaunal species in marine systems can also be important to benthic-pelagic coupling. In particular, while key species are often large, excretion and respiration of smaller macrofauna can be important to the exchange of nutrients near the seafloor and oxygen consumption in the benthic boundary layer.     

Interspecific pairwise relationships among body size, clutch size and latitude: deconstructing a macroecological triangle in birds

Aim Ecogeographical ‘rules’, large‐scale patterns in ecological variables across geographical space, can provide important insights into the mechanisms of evolution and ecological assembly. However, interactions between rules could obscure both the observation of large‐scale patterns and their interpretation. Here, we examine a system of three variables interrelated by ecogeographical rules – the latitudinal increase in body size within closely related homeotherms (Bergmann’s rule), the negative allometry of clutch size (Calder’s rule) and the latitudinal increase in clutch size (Lack’s rule) – in a global dataset of birds. Location Global. Methods We used linear regressions and meta‐analysis techniques to quantify the three rules across clades and through the taxonomic hierarchy. Path analysis was used to quantify interactions between rules at multiple taxonomic levels, as a function of both phylogenetic inheritance of traits and indirect feedbacks between the three rules. Independent contrasts analyses were performed on four clades with available phylogenies, and the taxonomic partitioning of variation in each trait was quantified. Results Standardizing across all clades, Lack’s and Bergmann’s rules were supported at all taxonomic levels, with Calder’s rule being supported at the order level. Lack’s rule was consistently stronger and more often detected than the other two rules. Path analysis showed that the indirect effects often outweighed the direct effects of Calder’s rule at the genus level and Bergmann’s rule at the order level. Strong interactions between Calder’s and Bergmann’s rules led to a trade‐off between the rules depending on taxonomic resolution. Main conclusions We found strong interactions between Bergmann’s, Lack’s and Calder’s rules in birds, and these interactions varied in strength and direction over the taxonomic hierarchy and among avian clades. Ecogeographical rules may be masked by feedbacks from other, correlated variables, even when the underlying selective mechanism is operating. The apparently conflicting pairwise relationships among clutch size, body size and latitude illustrate the difficulty of interpreting individual pairwise correlations without recognition of interdependence with other variables.     

Direct and indirect interactions co‐determine species composition in nurse plant systems

Facilitation by nurse plants plays an important role in determining community composition in severe environments. Although the unidirectional effect of nurses on beneficiary species has received considerable research interest, nurse‐mediated interactions among beneficiary species (so‐called indirect interactions) are less known. Consequently, community composition in nurse plant systems is generally considered as a simple consequence of the facilitative effect of the nurse even though beneficiary species may significantly contribute to community assembly and modulate the direct nurse effects on the community. In an observational study we assessed nurse effects and nurse‐mediated beneficiary interactions in two contrasting nurse plant systems in dry environments using a newly developed framework. We quantified plant–plant interaction intensity using the relative interaction index (RII) at the community and species level for three Retama sphaerocarpa shrub size‐classes in a semiarid shrubland and four Arenaria tetraquetra cushion plant communities differing in aspect and elevation in dry alpine gravel habitats. The observed RII was split into nurse and beneficiary effects, and related to individual mass, species frequency and abundance using generalized linear mixed models. Results showed predominantly positive nurse effects and negative beneficiary interactions. The effect size of nurse plants, however, was significantly higher than the effect size of beneficiary species in both systems. Individual plant mass and abundance of species was dependent on the combined effects of nurse and beneficiary species whereas species occurrence was related to nurse effects only. Despite evident differences, the semiarid and alpine nurse plant systems showed strong functional parallelisms. We found interdependence between the effects of nurse and beneficiary species on beneficiary plant assemblages emphasizing their combined role on community assembly in both systems. Our results highlight the need to consider indirect interactions to understand fully plant community dynamics.     

Seasonal variation in prey preference,diet partitioning and niche breadth in a rich large carnivore guild

Large carnivore community structure is affected by direct and indirect interactions between intra-guild members. Co-existence between different species within a carnivore guild may occur through diet, habitat or temporal partitioning. Since carnivore species are highly dependent on availability and accessibility of prey, diet partitioning is potentially one of the most important mechanisms in allowing carnivores to co-exist. Intra-guild interactions may vary over time as carnivore prey preference and diet overlap can change due to seasonal changes in resource availability. We conducted scat analysis to compare the seasonal changes in prey preference, diet partitioning and niche breadth of four large carnivore species, namely leopard Panthera pardus, spotted hyena Crocuta crocuta, brown hyena Parahyaena brunnea and wild dog Lycaon pictus in central Tuli, Botswana. Large carnivores in central Tuli display a high dietary overlap, with spotted hyena and brown hyena displaying almost complete dietary overlap and the other carnivore species displaying slightly lower but still significant dietary overlap. Dietary niche breadth for both hyena species was high possibly due to their flexible foraging strategies, including scavenging, while leopard and wild dog showed a relatively low niche breadth, suggesting a more specialised diet. High dietary overlap in central Tuli is possibly explained by the high abundance of prey species in the area thereby reducing competition pressure between carnivore species. Our research highlights the need to assess the influence of diet partitioning in structuring large carnivore communities across multiple study sites, by demonstrating that in prey rich environments, the need for diet partitioning by carnivores to avoid competition may be limited.     

Habitat effects on the relative importance of trait- and density-mediated indirect interactions

Classical views of trophic cascades emphasize the primacy of consumptive predator effects on prey populations to the transmission of indirect effects [density-mediated indirect interactions (DMIIs)]. However, trophic cascades can also emerge without changes in the density of interacting species because of non-consumptive predator effects on prey traits such as foraging behaviour [trait-mediated indirect interactions (TMIIs)]. Although ecologists appreciate this point, measurements of the relative importance of each indirect predator effect are rare. Experiments with a three-level, rocky shore food chain containing an invasive predatory crab ( Carcinus maenas ), an intermediate consumer (the snail, Nucella lapillus ) and a basal resource (the barnacle, Semibalanus balanoides ) revealed that the strength of TMIIs is comparable with, or exceeds, that of DMIIs. Moreover, the sign and strength of each indirect predator effect depends on whether it is measured in risky or refuge habitats. Because habitat shifts are often responsible for the emergence of TMIIs, attention to the sign and strength of these interactions in both habitats will improve our understanding of the link between individual behaviour and community dynamics.     

Prey diversity as a driver of resource partitioning between river‐dwelling fish species

Although food resource partitioning among sympatric species has often been explored in riverine systems, the potential influence of prey diversity on resource partitioning is little known. Using empirical data, we modeled food resource partitioning (assessed as dietary overlap) of coexisting juvenile Atlantic salmon (Salmo salar) and alpine bullhead (Cottus poecilopus). Explanatory variables incorporated into the model were fish abundance, benthic prey diversity and abundance, and several dietary metrics to give a total of seventeen potential explanatory variables. First, a forward stepwise procedure based on the Akaike information criterion was used to select explanatory variables with significant effects on food resource partitioning. Then, linear mixed‐effect models were constructed using the selected explanatory variables and with sampling site as a random factor. Food resource partitioning between salmon and bullhead increased significantly with increasing prey diversity, and the variation in food resource partitioning was best described by the model that included prey diversity as the only explanatory variable. This study provides empirical support for the notion that prey diversity is a key driver of resource partitioning among competing species.     

Determining the mechanism by which fish diversity influences production

Understanding the ability of biodiversity to govern ecosystem function is essential with current pressures on natural communities from species invasions and extirpations. Changes in fish communities can be a major determinant of food web dynamics, and even small shifts in species composition or richness can translate into large effects on ecosystems. In addition, there is a large information gap in extrapolating results of small-scale biodiversity–ecosystem function experiments to natural systems with realistic environmental complexity. Thus, we tested the key mechanisms (resource complementarity and selection effect) for biodiversity to influence fish production in mesocosms and ponds. Fish diversity treatments were created by replicating species richness and species composition within each richness level. In mesocosms, increasing richness had a positive effect on fish biomass with an overyielding pattern indicating species mixtures were more productive than any individual species. Additive partitioning confirmed a positive net effect of biodiversity driven by a complementarity effect. Productivity was less affected by species diversity when species were more similar. Thus, the primary mechanism driving fish production in the mesocosms was resource complementarity. In the ponds, the mechanism driving fish production changed through time. The key mechanism was initially resource complementarity until production was influenced by the selection effect. Varying strength of intraspecific interactions resulting from differences in resource levels and heterogeneity likely caused differences in mechanisms between the mesocosm and pond experiments, as well as changes through time in the ponds. Understanding the mechanisms by which fish diversity governs ecosystem function and how environmental complexity and resource levels alter these relationships can be used to improve predictions for natural systems.     

Direct effects of warming increase woody plant abundance in a subarctic wetland

Both the direct effects of warming on a species’ vital rates and indirect effects of warming caused by interactions with neighboring species can influence plant populations. Furthermore, herbivory mediates the effects of warming on plant community composition in many systems. Thus, determining the importance of direct and indirect effects of warming, while considering the role of herbivory, can help predict long‐term plant community dynamics. We conducted a field experiment in the coastal wetlands of western Alaska to investigate how warming and herbivory influence the interactions and abundances of two common plant species, a sedge, Carex ramenskii, and a dwarf shrub, Salix ovalifolia. We used results from the experiment to model the equilibrium abundances of the species under different warming and grazing scenarios and to determine the contribution of direct and indirect effects to predict population changes. Consistent with the current composition of the landscape, model predictions suggest that Carex is more abundant than Salix under ambient temperatures with grazing (53% and 27% cover, respectively). However, with warming and grazing, Salix becomes more abundant than Carex (57% and 41% cover, respectively), reflecting both a negative response of Carex and a positive response of Salix to warming. While grazing reduced the cover of both species, herbivory did not prevent a shift in dominance from sedges to the dwarf shrub. Direct effects of climate change explained about 97% of the total predicted change in species cover, whereas indirect effects explained only 3% of the predicted change. Thus, indirect effects, mediated by interactions between Carex and Salix, were negligible, likely due to use of different niches and weak interspecific interactions. Results suggest that a 2°C increase could cause a shift in dominance from sedges to woody plants on the coast of western Alaska over decadal timescales, and this shift was largely a result of the direct effects of warming. Models predict this shift with or without goose herbivory. Our results are consistent with other studies showing an increase in woody plant abundance in the Arctic and suggest that shifts in plant–plant interactions are not driving this change.     

Direct and indirect effects of predation and predation risk in old-field interaction webs

ABSTRACT Indirect effects emerge when a change in the abundance of one species indirectly affects another by changing the abundances of intermediate species-called density-mediated indirect effects-or they arise when one species modifies how two other species interact-called trait-mediated indirect effects. I report on field experiments that evaluated how grass and herb biomass in old-field interaction webs was influenced indirectly by a spider carnivore through its interactions with a generalist and a grass-specialist grasshopper species. I manipulated interaction pathways between the spider and the plants using different combinations of the grasshopper species. I changed the modality of predator-prey interactions to isolate density-mediated from trait-mediated effects using natural spiders (predation spiders) or spiders that were prevented from subduing prey by mouthpart manipulation (risk spiders). I found that indirect effects were stronger in speciose, reticulate food webs than in linear food chains owing to a trait-mediated effect, a diet shift by herbivores in response to predation risk. Spiders alone did not have significant effects on grasshopper densities in the field experiments, removing any possibility of density-mediated indirect effects. The study illustrates that ecologists should not underestimate the importance of behavioral ecology in determining community-level interactions.     

Correlates of cell-mediated immunity in nestling house sparrows

Cell-mediated immunity is an important vertebrate defense against pathogens, but components of this response may vary in quality. Such variation could arise through the effects of ecology on optimal immunocompetence. We used injections of phytohaemagglutinin (PHA) to measure the factors influencing T-cell proliferation in nestling house sparrows (Passer domesticus). Bivariate analyses revealed positive associations with nestling mass and size, but no effect of ectoparasites. The response to PHA was, however, strongly affected by brood identity. A mixed model with brood identity as a random factor and nestling mass, size, number of ectoparasites, parental feeding rate, clutch size, brood size at hatching, and date uncovered significant positive correlations between PHA response and both nestling mass and the brood size at hatching. Because many of these variables are related hierarchically, we used path analysis to explore the relationships in more detail. We found that a nestling immune response was affected by several indirect paths. Brood size at hatch had both positive and negative paths, and date in the season had several indirect negative effects through its effect on brood size and nestling mass. The approach used and the results obtained offer some new ideas for incorporating immune responses into life history theory.     

Host‐plant genotypic diversity and community genetic interactions mediate aphid spatial distribution

Genetic variation in plants can influence the community structure of associated species, through both direct and indirect interactions. Herbivorous insects are known to feed on a restricted range of plants, and herbivore preference and performance can vary among host plants within a species due to genetically based traits of the plant (e.g., defensive compounds). In a natural system, we expect to find genetic variation within both plant and herbivore communities and we expect this variation to influence species interactions. Using a three‐species plant‐aphid model system, we investigated the effect of genetic diversity on genetic interactions among the community members. Our system involved a host plant (Hordeum vulgare) that was shared by an aphid (Sitobion avenae) and a hemi‐parasitic plant (Rhinanthus minor). We showed that aphids cluster more tightly in a genetically diverse host‐plant community than in a genetic monoculture, with host‐plant genetic diversity explaining up to 24% of the variation in aphid distribution. This is driven by differing preferences of the aphids to the different plant genotypes and their resulting performance on these plants. Within the two host‐plant diversity levels, aphid spatial distribution was influenced by an interaction among the aphid's own genotype, the genotype of a competing aphid, the origin of the parasitic plant population, and the host‐plant genotype. Thus, the overall outcome involves both direct (i.e., host plant to aphid) and indirect (i.e., parasitic plant to aphid) interactions across all these species. These results show that a complex genetic environment influences the distribution of herbivores among host plants. Thus, in genetically diverse systems, interspecific genetic interactions between the host plant and herbivore can influence the population dynamics of the system and could also structure local communities. We suggest that direct and indirect genotypic interactions among species can influence community structure and processes.     

Evolutionary indirect effects of biological invasions

Just as ecological indirect effects can have a wide range of consequences for community structure and ecosystem function, theory suggests that evolutionary indirect effects can also influence community dynamics and the outcome of species interactions. There is little empirical evidence documenting such effects, however. Here, I use a multi-generation selection experiment in the field to investigate: (1) how the exotic plant Medicago polymorpha and the exotic insect herbivore Hypera brunneipennis affect the evolution of anti-herbivore resistance traits in the native plant Lotus wrangelianus and (2) how observed Lotus evolutionary responses to Hypera alter interactions between Lotus and other members of the herbivore community. In one of two study populations, I document rapid evolutionary changes in Lotus resistance to Hypera in response to insecticide treatments that experimentally reduced Hypera abundance, and in response to Medicago-removal treatments that also reduced Hypera abundance. These evolutionary changes in response to Hypera result in reduced attack by aphids. Thus, an evolutionary change caused by one herbivore species alters interactions with other herbivore taxa, an example of an eco-evolutionary feedback. Given that many traits mediate interactions with multiple species, the effects of evolutionary changes in response to one key biotic selective agent may often cascade through interaction webs to influence additional community members.     

Multivariate control of root biomass in a semi-arid grassland on the Loess Plateau,China

Aim

Root biomass has long been under-represented in biodiversity–ecosystem functioning studies, despite its dominance in biomass in many arid and semi-arid ecosystems. We aimed to explore the multivariate control over root biomass by plant diversity, together with other biotic and abiotic factors and to evaluate the relative importance of these factors.

Methods

Above- and below-ground traits of 13 communities and soil properties were measured in semi-arid grasslands on the Loess Plateau, China. Structural equation modeling (SEM) was used to evaluate the relative importance of the community and soil characteristics, emphasizing the direct and indirect effects of plant diversity on root biomass.

Results

Significant indirect effects of plant species richness on root biomass were found, although no direct correlation was detected between them. In the indirect pathways, plant species richness showed a positive effect on soil total nitrogen, but a significant negative influence on soil total carbon. Soil total nitrogen and plant diversity had the largest and smallest total effect respectively on root biomass in the model.

Conclusions

Plant species richness was not the strongest determinant of root biomass but had a significant indirect effect, mediated through soil total carbon and nitrogen. This study suggests that greater plant species richness, through a positive influence on soil total nitrogen, may indirectly promote root carbon stock.     

Enemy-mediated apparent competition: empirical patterns and the evidence

Apparent competition arises when two victim species negatively affect each other (−,  −) by enhancing the equilibrium density or changing the foraging behaviour of a shared natural enemy. Shared enemies can also mediate non-reciprocal (−,  0) indirect effects, i.e. indirect amensalism, whenever one prey species is not affected by the presence of alternative prey. We review 34 studies on terrestrial and freshwater systems to evaluate the extent to which apparent competition has been perceived as a reciprocal (−,  −) or non-reciprocal (−,  0) interaction. We found only three studies showing reciprocal effects between apparent competitors. Indirect amensalism was documented in 10 studies and could be inferred for 16 other cases (76% in total). The remaining five studies provided insufficient data to determine the form of indirect interaction. The apparent prevalence of non-reciprocal enemy-mediated interactions resembles that observed for resource-based interspecific competition. Amensal indirect effects via shared predation may result from differences in population size, nutritional value, susceptibility to attack, or asynchronous dynamics of alternative prey, or the predator's feeding preferences. Moreover, experimental protocols may confound the actual form of apparent competition through short-term observations, incomplete designs, or biased consideration of conspicuous interactions, leading to reciprocal effects being overlooked. We conclude that, at present, it is still difficult to determine the relative role of apparent competition vs indirect amensalism in natural food webs because most published studies have failed to document in full interactions via shared enemies.     

How do belowground organisms influence plant-pollinator interactions?

Aims The majority of angiosperms are pollinated by animals, and this interaction is of enormous importance in both agricultural and natural systems. Pollinator behavior is influenced by plants' floral traits, and these traits may be modified by interactions with other community members. In recent years, knowledge of ecological linkages between above- and belowground organisms has grown tremendously. Soil communities are extremely diverse, and when their interactions with plants influence floral characteristics, they have the potential to alter pollinator attraction and visitation, but plant–pollinator interactions have been neglected in studies of the direct and indirect effects of soil organism–root interactions. Here, we review these belowground interactions, focusing on the effects of nitrogen-fixing bacteria, arbuscular mycorrhizal fungi and root-feeding herbivores, and their effects on floral traits and pollinators. Further, we identify gaps in our knowledge of these indirect effects and recommend promising directions and topics that should be addressed by future research.Important findings Belowground organisms can influence a wide variety of floral traits that are important mediators of pollinator attraction, including the number and size of flowers and nectar and pollen production. Other traits that are known to influence pollinators in some plant species, such as floral volatiles, color and nectar composition, have rarely or never been examined in the context of belowground plant interactions. Despite clear effects on flowers, relatively few studies have measured pollinator responses to belowground interactions. When these indirect effects have been studied, both arbuscular mycorrhizal fungi and root herbivores were found to shift pollinator visitation patterns. Depending on the interaction, these changes may either increase or decrease pollinator attraction. Finally, we discuss future directions for ecological studies that will more fully integrate belowground ecology with pollination biology. We advocate a multilevel approach to these questions to not only document indirect effect pathways between soil interactions and pollination but also identify the mechanisms driving changes in pollinator impacts and the resultant effects on plant fitness. A more thorough understanding of these indirect interactions will advance ecological theory and may inform management strategies in agriculture and conservation biology.