BACKGROUND:A promising strategy to optimize biological control of insect pests is selecting crop varieties with indirect defense traits. Indirect plant defenses recruit natural enemies to kill pests and include chemical attractants like herbivore-induced plant volatiles. In prior laboratory assays, we found sorghum (Sorghum bicolor L.) cultivar ATx3409/RTx436 infested with sorghum aphid (Melanaphis sorghi Theobald) was attractive to natural enemies and emitted more chemical attractants than two other cultivars. In this field study, we manually infested 9-week-old sorghum plants with aphids and quantified differences in natural enemy and aphid densities among cultivars throughout the growing season. We also used field cages to control access of natural enemies to plants and estimate their effects on aphid suppression. RESULTS:We found strong evidence that indirect plant defenses confer economically relevant control of aphid pest populations and that laboratory assays can accurately predict natural enemy recruitment in the field. In 2022, there were three times more lady beetles (Coleoptera: Coccinellidae), lacewings (Neuroptera: Chrysopidae and Hemerobiidae), hover flies (Diptera: Syrphidae), and parasitoids (Hymenoptera: Braconidae and Aphelinidae) per aphid on ATx3409/RTx436 than on the other two cultivars. In the field cage experiment, natural enemies reduced aphid densities by up to 83% one week after aphid infestation. ATx3409/RTx436 was the only cultivar to remain below the economic threshold throughout the growing season, indicating that this cultivar would not require any pesticide applications to control aphids. In 2023, there were similar abundances of natural enemies and aphid densities across cultivars, the latter of which remained near zero throughout the growing season, likely due to extremely hot temperatures and drought that may have contributed to aphid mortality. CONCLUSION:Our findings demonstrate that indirect plant defenses enhance biological control and deliver economically important pest suppression. Cultivar screening and selection for indirect defense traits provides a promising avenue to improve crop protection and breeding for resistance. © 2025 The Author(s). Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry. This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.

The production of herbivore-induced plant volatiles (HIPVs) is a type of indirect defense used by plants to attract natural enemies and reduce herbivory by insect pests. In many crops little is known about genotypic variation in HIPV production or how this may affect natural enemy attraction. In this study, we identified and quantified HIPVs produced by 10 sorghum (Sorghum bicolor) cultivars infested with a prominent aphid pest, the sorghum aphid (Melanaphis sorghi Theobald). Volatiles were collected using dynamic headspace sampling techniques and identified and quantified using GC-MS. The total amounts of volatiles induced by the aphids did not differ among the 10 cultivars, but overall blends of volatiles differed significantly in composition. Most notably, aphid herbivory induced higher levels of methyl salicylate (MeSA) emission in two cultivars, whereas in four cultivars, the volatile emissions did not change in response to aphid infestation. Dual-choice olfactometer assays were used to determine preference of the aphid parasitoid, Aphelinus nigritus, and predator, Chrysoperla rufilabris, between plants of the same cultivar that were un-infested or infested with aphids. Two aphid-infested cultivars were preferred by natural enemies, while four other cultivars were more attractive to natural enemies when they were free of aphids. The remaining four cultivars elicited no response from parasitoids. Our work suggests that genetic variation in HIPV emissions greatly affects parasitoid and predator attraction to aphid-infested sorghum and that screening crop cultivars for specific predator and parasitoid attractants has the potential to improve the efficacy of biological control.

Interactions between plants and herbivores are central in most ecosystems, but their strength is highly variable. The amount of variability within a system is thought to influence most aspects of plant-herbivore biology, from ecological stability to plant defense evolution. Our understanding of what influences variability, however, is limited by sparse data. We collected standardized surveys of herbivory for 503 plant species at 790 sites across 116° of latitude. With these data, we show that within-population variability in herbivory increases with latitude, decreases with plant size, and is phylogenetically structured. Differences in the magnitude of variability are thus central to how plant-herbivore biology varies across macroscale gradients. We argue that increased focus on interaction variability will advance understanding of patterns of life on Earth.

Pollinators for the fiber crop cotton are underused despite evidence that cross-pollination can increase yields. In addition, existing research largely ignores the potential of insects other than bees (Hymenoptera: Apoidea: Anthophilia) to provide pollination services for cotton. We observed plant vaceae) and carrying cotton pollen grains on their bodies. We hypothesized that fleahoppers might contribute to cross-pollination of cotton as they forage among flowers. To test this hypothesis, we examined P. seriatus flower visitation frequency, cotton pollen load, pollen analog dispersal, and cross-pollination capacity. We found that cotton fleahoppers visited 21% of flowers observed in our field site and that they deposited a pollen analog on 12.5% of the flowers accessible in a field cage. However, individual cotton fleahoppers are likely too small to carry enough grains to fertilize selfsterile cotton flowers, because field collected cotton fleahoppers carried approximately 25 pollen grains per insect, which is less than what is needed for cotton flowers to set fruit. Overall, we found that cotton fleahoppers were unable to stimulate cotton fruit development in self-sterile flowers. Nevertheless, we predict that cotton fleahoppers may contribute to cross-pollination of cotton within a community of pollinators, and that they may pollinate their wild host plants which have smaller or clustered flowers. We encourage researchers to continue to investigate non-bee pollinators in wild and agroecosystems.

Abstract Human altered landscapes have caused declines in the diversity of wildlife where behaviorally plastic species (i.e., mesocarnivores and invasive species) tend to monopolize these areas and consume predictable and readily accessible food resources, such as human food waste and carrion. Increased consumption of carrion by vertebrates and invasive invertebrate species can alter population dynamics of native necrophagous insects relying on these resources. We tested the hypothesis that vertebrate scavengers and invasive species reduce blow fly (1) ability to use carrion and (2) reproduction in human‐impacted environments in central Texas, USA, with season, habitat (field and wooded landscapes), and carrion type (species of carrion and coat color) acting synergistically. Vertebrate scavengers in this habitat, of which 75% of the documented species were mesocarnivores and obligate scavengers, consumed 100% of carrion during the winter and 62% during summer despite having low species richness (2–5 species). Of the remaining carcasses available for arthropod activity during summer, the invasive red imported fire ant, Solenopsis invicta (Hymenoptera: Formicidae), monopolized 34%, and blow flies (e.g., Lucilia eximia and Chrysomya rufifacies [Diptera: Calliphoridae]) were only able to colonize 25%. Approximately 90% of carrion that was utilized by blow flies was co‐colonized by fire ants, and subsequent production of adult blow flies experienced up to a ninefold reduction in production compared with carcasses that were not scavenged by vertebrates or fire ants. Our results demonstrate oviposition resources used by blow flies in environments altered by human activity are reduced significantly by vertebrate scavengers and an invasive ant species. Future research should determine whether competitive interactions between vertebrate and invasive ant competitors for access to carrion resources have population‐level impacts to blow flies in human‐mediated ecosystems, or whether blow flies are able to shift to other resources to maintain sustainable populations and continue providing ecosystem services, such as pollination.

Ants have not been considered important in the process of vertebrate carrion decomposition, but a recent literature review reported over 150 carrion-visiting ant species. Though many ant species have been observed to remove carrion tissue and consume carrion-exuded liquids, the significance of ant recruitment to vertebrate carrion is poorly understood. We conducted a combination of field and laboratory experiments to quantify red imported fire ant recruitment to rodent carrion and determine whether consuming rodent carrion is beneficial to ant colony performance. In the field, 100% of rat carcasses were rapidly colonized by fire ants at high abundances. In our laboratory experiment, the performance of mice-fed fire ant colonies was poor when compared to colonies that were fed mice and insects or insects only. Our results suggest that there is a discrepancy between high levels of fire ant recruitment to vertebrate carrion and the poor colony performance when fed carrion. We hypothesize that fire ants are attracted to vertebrate carrion not because it is a high-quality food, but rather because it hosts large numbers of other invertebrates that can serve as prey for fire ants, potentially showcasing an interesting case of tritrophic interaction in carrion ecology.

Landscapes with more complex composition and configuration are generally expected to enhance natural enemy densities and pest suppression. To evaluate this hypothesis for an invasive aphid pest of sorghum, Melanaphis sorghi Theobald (Hemiptera: Aphididae), sampling in sorghum fields for aphids and natural enemies was conducted over two years in a southern U.S. coastal production region. Landscape composition and configuration of crop and noncrop elements were assessed using correlation and multivariate regression modeling to detect relationships with insects at different spatial scales. Significant models found more complex landscape configuration, particularly the amount of habitat edges, was associated with increased aphid and natural enemy abundance. Composition associated with noncrop habitats had the opposite effect. Numerical response of natural enemies was taxa dependent, with parasitism lower as landscape complexity increased, while predator numerical response was not affected by landscape complexity. These results indicate landscape complexity may increase both aphid and natural enemy abundance, but with decreasing parasitism and little association with predator numerical response. These relationships are likely contingent on overall environmental suitability to aphid population increase as results were less evident in the second year when average aphid abundance regularly exceeded the economic threshold. This study supports the importance of configuration, especially habitat borders, as a critical metric for determining pest-natural enemy dynamics within a large-scale cereal agroecosystem.

Whether increased natural enemy density or adding a second natural enemy species will provide superior pest suppression in greenhouse augmentative biological control is unknown for many commercially available natural enemy species. In this study, we use sweetpotato whiteflies, Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae), on poinsettias, Euphorbia pulcherrima Willd. ex Klotzsch (Malpighiales: Euphorbiaceae), to determine whether adding Amblyseius swirskii (Athias-Henriot) (Acari: Phytoseiidae) to Eretmocerus eremicus Rose and Zolnerowich (Hymenoptera: Aphelinidae) is better for B. tabaci suppression compared with either natural enemy alone, both with and without challenges with whitefly immigration or delayed natural enemy releases.The number of whiteflies on caged poinsettias treated with different natural enemy release rates (single or double rate), natural enemy species (one or two species), natural enemy delayed release (weeks 4 and 8), and whitefly immigration treatments (introduced at week 4 or week 8) was censused biweekly for 16 wk. Both species used in combination provided similar or better suppression of whiteflies compared with either natural enemy alone. Both species combined also provided superior suppression of whiteflies when challenged with whitefly immigration or delays in natural enemy releases compared with E. eremicus alone. Whitefly immigration or delays in E. eremicus releases did not increase whitefly populations, suggesting that suppression of whiteflies by E. eremicus alone is relatively robust.This study found no evidence for negative interactions between E. eremicus and A. swirskii for suppressing B. tabaci.

Stable isotope analysis is one of the most widely used techniques to estimate trophic position and provides fundamental insight into the structure and management of ecological communities. To account for the effects of geographic variation in isotope levels, trophic position is typically estimated relative to an isotope "baseline" (i.e., material representing geographic variation) using a methodology such as a formula or statistical analysis. There is, however, remarkable variation in the baselines and methodologies used to estimate trophic position from stable isotopes. The consequences of this lack of standardization are unknown but could result in biased or erroneous conclusions. We conducted a literature review to quantify the variation in baselines and methodologies used to estimate trophic position from stable isotopes. Next, we assessed the consequences of this variation on individual species estimates and food web structure by extracting published trophic positions and applying various baselines and methodologies to existing data sets. We identified 10 baselines and eight methodologies, the use of which varied by ecosystem studied. Moreover, we found that different baselines and methodologies yield significantly different trophic position estimates for individual species, as well as different conclusions about food web structure. Authors should avoid biological interpretations of absolute, stand-alone trophic positions (as these are prone to a number of biases). We recommend pairing stable isotope analysis with other techniques for more robust conclusions. Increased sample size may mitigate some of the variation caused by different baselines and methodologies; however, an alarmingly large proportion of studies collected only one sample in at least one trophic group (41% of all reviewed studies). Authors should collect a minimum of five samples per trophic group (but ten for best practices) from as many trophic groups as possible to increase statistical power and redundancy in comparisons. When sample size is unavoidably constrained, we recommend using compound-specific isotope analysis with a taxon-specific trophic discrimination factor, because it may be more accurate and require fewer samples to maintain appropriate statistical power. Implementing our recommendations will increase the robustness and accuracy of conclusions based on stable isotopes, resulting in better management decisions and a more accurate understanding of ecological communities.

Due to typesetting errors, Table 5 was not displayed correctly in the initial online publication. The original online article has been corrected.

We used high-throughput sequencing molecular gut content analysis (HTS-MGCA) to examine diets of the red imported fire ant (Solenopsis invicta; hereafter fire ant). Because adult fire ant workers rely on their larvae (brood) to digest prey and share it via trophallaxis with the rest of the colony, we conducted a study to determine if we could sequence the gut contents of final instar ant brood instead of adult workers to better capture colony diet. We sequenced all samples on the Illumina HiSeq 2500 using general primers to amplify and identify arthropod DNA within larval guts. We first verified our methods in a controlled-feeding experiment in the laboratory using crickets (Gryllodes sigillatus), beet armyworm (Spodoptera exigua), or a mixture of both species, and then we analyzed the diets of field-collected fire ants. The sequences from our controlled-feeding lab samples were very accurate, validating our analysis pipeline. We identified a total of 45 diet events from six field-collected fire ant samples. All fire ant field samples contained sequences from both detritus-based food webs (arthropods that consume decaying material) and plant-based food webs (arthropods that consume living plant material). HTS-MGCA of fire ant larvae was highly successful and provided novel insights into the community-level impacts of ant foraging. Our approach should be widely applicable to the study of resource use and trophic interactions in other eusocial insects.

The cotton agroecosystem is one of the most intensely managed, economically, and culturally important fiber crops worldwide including in the United States of America (U.S.), China, India, Pakistan, and Brazil. The composition and configuration of crop species and semi-natural habitat can have significant effects on ecosystem services such as pollination. Here we investigate the effect of crop and semi-natural habitat configuration in a large-scale cotton agroecosystem on the diversity and abundance of native bees. Interfaces sampled include cotton grown next to cotton, sorghum or semi-natural habitat. Collections of native bees across interface types revealed 32 species in 13 genera across 3 families. Average species richness ranged between 20.5 and 30.5 with the highest (30.5) at the interface of cotton and semi-natural habitat. The most abundant species was Melissodes tepaneca Cresson (> 4,000 individuals, ~75% of bees collected) with a higher number of individuals found in all cotton-crop interfaces compared to the cotton interface with semi-natural habitat or natural habitat alone. It was also found that interface type had a significant effect on the native bee communities. Communities of native bees in the cotton-crop interfaces tended to be more consistent in the abundance of species and number of species at each sampling site. While cotton grown next to semi-natural habitat had higher species richness, the number of bees collected varied. These data suggest that native bee communities persist in large-scale cotton agroecosystems and some species may thrive even when cotton-crop interfaces are dominant compared with semi-natural habitat. These data have native bee conservation implications that may improve potential pollination benefits to cotton production.

Evaluating the factors that promote invasive ant abundance is critical to assess their ecological impact and inform their management. Many invasive ant species show reduced nestmate recognition and an absence of boundaries between unrelated nests, which allow populations to achieve greater densities due to reduced intraspecific competition. We examined nestmate discrimination and colony boundaries in introduced populations of the red imported fire ant (Solenopsis invicta; hereafter, fire ant). Fire ants occur in two social forms: monogyne (colonies with a single egg-laying queen) and polygyne (colonies with multiple egg-laying queens). In contrast with monogyne nests, polygyne nests are thought to be interconnected due to the reduced antagonism between non-nestmate polygyne workers, perhaps because polygyne workers habituate the colony to an odour unique to Gp-9b -carrying adults. However, colony boundaries and nestmate discrimination are poorly documented, particularly for worker-brood interactions. To delimit boundaries between field colonies, we correlated the exchange of a 15 N-glycine tracer dissolved in a sucrose solution with social form. We also evaluated nestmate discrimination between polygyne workers and larvae in the laboratory. Counter to our expectations, polygyne colonies behaved identically to monogyne colonies, suggesting both social forms maintain strict colony boundaries. Polygyne workers also preferentially fed larval nestmates and may have selectively cannibalized non-nestmates. The levels of relatedness among workers in polygyne colonies was higher than those previously reported in North America (mean ± standard error: 0.269 ± 0.037). Our study highlights the importance of combining genetic analyses with direct quantification of resource exchange to better understand the factors influencing ant invasions.

In this case study, we investigate the efficacy and economics of using two natural enemies in an integrated pest management (IPM) program to manage sweetpotato whitefly, Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae), in commercial poinsettia (Euphorbia pulcherrima Willd. ex Klotzsch) production. Two similar greenhouses at each of three different grower locations were designated as either the IPM or the conventional insecticide greenhouses in southeastern United States. In the IPM greenhouses, we released Eretmocerus eremicus (Rose & Zolnerowich) (Hymenoptera: Aphelinidae) weekly and Amblyseius swirskii (Athias-Henriot) (Acari: Phytoseiidae) every 4 wk, and selective insecticides were used to treat high whitefly densities as needed. In the conventional greenhouses, growers were autonomous in their insecticide application decisions. All whitefly stages were counted weekly on a maximum of 20 leaves per 50 randomly sampled poinsettias and 50 flagged (i.e., revisited) poinsettias in every greenhouse. Whitefly densities were consistently similar or higher in the IPM greenhouses compared to their conventionally managed counterparts for the duration of the trial. The cost of inputs and labor for whitefly management in the IPM greenhouses was between 0.57- and 3.0-fold the cost of conventional management. Our study supports that releasing E. eremicus and A. swirskii can reduce insecticide applications by 25-78% and may be considered a feasible strategy to manage B. tabaci in commercial poinsettia production in place of conventional insecticidal control in southeastern United States.

Classical food web theory predicts that species at the base of food webs will be more abundant than those at the top. Likewise, it’s hypothesized that feeding at lower trophic positions (e.g., deriving more nitrogen from plant-based resources such as nectar or honeydew) plays an important role in the establishment, spread, and ecological dominance of invasive social insects. We tested the relationship between diet, abundance, and resource dominance using the invasive tawny crazy ant (Nylanderia fulva). We used stable isotope analysis, pitfall sampling, and attraction of ants to baits to investigate tawny crazy ant trophic position, abundance, foraging intensity, and competition with other ant species at sites across Texas, USA. Tawny crazy ant abundance varied from zero to > 1000 ants per pitfall trap depending on site and month. Counter to our predictions, however, there was no significant relationship between tawny crazy ant abundance and trophic position. Moreover, tawny crazy ant foraging activity and numerical dominance at baits were significantly higher when ants occupied a higher trophic position. Tawny crazy ants were much more predaceous than expected, as worker $${\updelta }$$ 15N values were often statistically indistinguishable from known arthropod predators (14/22 total observations). A review of the literature indicates that our results add to a growing body of evidence (15/24 studies) suggesting that ant abundance does not always correspond with a more herbivorous diet. Rather, other factors such as dietary flexibility and abiotic features may play a more important role in invasive ant abundance, particularly in the case of tawny crazy ants.

Climate change is predicted to increase the frequency of drought conditions and alter plant-insect interactions. Despite over 530 studies on the effects of water-deficit stress on plant-insect interactions, we still cannot accurately predict plant-insect interactions under drought conditions. Most studies have focused on how insect herbivores respond to water-deficit-stressed plants, with little attention on how stressed plants and changes in plant physiology may contribute to the variation in herbivore response. Variation in herbivore response to water-deficit-stressed plants may be due to stress-induced changes in plants and how these changes differ with stress severity. In this study, we determined the effects of water-deficit stress on cotton (Delta Pine 174RF) physiology in an agroecosystem using pulsed moderate and severe stress. Our goal was to determine how moderate and severe water-deficit stress affect cotton (Gossypium hirsutum L.) physiology differently in terms of photosynthesis, development, water use, and nutrient content. We found that moderate and severely stressed plants had many different physiological responses that may contribute to the variation we see in herbivore response to stressed plants. For instance, moderately stressed plants were more vigorous and more developed than severely stressed plants, whereas severely stressed plants had more amino acids than moderately stressed plants. We predict that insect herbivores feeding on moderately and severely stressed plants would encounter differences in C assimilation, water content (stomatal conductance, transpiration), chlorophyll content, and concentrations of amino acids and digestible carbohydrates. It is clear that field studies should consider how differences in stress severity can affect plant-insect interactions.

In this study, we surveyed the initial whitefly (Aleyrodidae) populations on rooted poinsettia (Euphorbia pulcherrima) cuttings at two commercial greenhouse facilities in both 2017 and 2018 to determine the initial whitefly population at the beginning of poinsettia production and surveyed finished poinsettias at multiple retailers in Tyler, TX, over 2 years to determine whitefly densities considered acceptable by retailers. The initial whitefly population (mean ± se) for all poinsettias was 0.02 ± 0.02 (2017) and 0.33 ± 0.13 (2018) nymphs per plant for grower facility A and 0.05 ± 0.05 (2017) and 0.02 ± 0.01 (2018) nymphs per plant for grower facility B. Of the total 2417 rooted poinsettia cuttings inspected at both locations over 2 years, 29 cuttings had whitefly nymphs (1.2%), 18 had pupae (0.7%), and 23 had exuviae (1.0%). On finished poinsettias sampled at retailers, 4.38 to 40.38 immatures (nymphs + pupae) per plant were found within 60 seconds for any given retailer over the 2 years. We found poinsettias with as many as 220 immatures and 32 adults on a single plant at retailers. This study is the first to quantify densities of whiteflies at retail stores over multiple years.

Unicoloniality, or the absence of behavioral boundaries between nests, is thought to promote ant abundance due to reduced intraspecific competition. Workers within unicolonial populations may increase their own inclusive fitness by preferentially caring for more related individuals (nepotism), but nepotism has only rarely been documented in ants. We tested for unicoloniality and nepotism in polygyne red imported fire ants (Solenopsis invicta; hereafter fire ants). Fire ants occur in two social forms: monogyne (i.e., colonies with a single egg-laying queen) and polygyne (i.e., colonies with multiple egg-laying queens). Introduced populations of polygyne fire ants are commonly referred to as unicolonial, but cooperation between and within colonies is poorly documented. To delimit boundaries between colonies in the field, we quantified the exchange of a 15N-glycine tracer dissolved in a sucrose solution and correlated this exchange with colony genetic structure. We also quantified within-colony conflict between workers and larvae using close siblings (i.e., from the same mother) and non-siblings (i.e., from a different mother). Counter to our expectations, polygyne colonies did not exchange resources or workers, indicating distinct colony boundaries. Polygyne workers also preferentially fed larval sibling and may have preferentially cannibalized non-siblings. Polygyne colony behavior was correlated with higher levels of within-mound relatedness between workers in the field than those previously reported in North America (mean ± SE: 0.269 ± 0.037). Our study challenges fundamental assumptions about introduced populations of polygyne fire ants and suggests that polygyne colonies are multicolonial and likely engage in high levels of intraspecific competition.

Above-ground plant tissues produce characteristic blends of volatile compounds in response to insect herbivory. These herbivore-induced plant volatiles (HIPVs) function in plant defence and mediate foraging decisions by herbivores and their natural enemies. The ecological roles of HIPVs as foraging cues for different trophic levels highlight an important conflict for herbivores that need to locate suitable host plants while avoiding competition and predation. Plant roots also emit HIPVs following herbivory, but our understanding of root-produced volatiles and their ecological functions in soil environments remains limited. Moreover, recent studies have documented the effects of temporal dynamics of plant volatile production on ecological interactions, but little is known about how root HIPVs change throughout herbivory or the resulting ecological implications from such changes. In this study, we examined the roles of HIPVs from roots of cucumber plantsCucumis sativusas foraging cues for a specialist herbivore, striped cucumber beetleAcalymma vittatumand its natural enemies, entomopathogenic nematodes (EPNs). We predicted HIPVs fromA. vittatum-damaged roots would attract EPNs, while repelling conspecific larvae that avoid competition, induced plant defences and increased risk of predation by EPNs. To capture the temporal dynamics of root HIPVs, we determined how HIPV-mediated interactions change over time with sustained herbivory. Initially (after 24 hr),A. vittatumherbivory onC. sativus, or mechanical wounding, induced greater production of root volatiles. These root HIPVs recruited EPNs and repelled foragingA. vittatumlarvae, although larval performance was not affected by prior damage. Sustained (7 days) herbivory by larvae reduced HIPVs to levels indistinguishable from undamaged control roots while mechanically damaged roots continued to produce higher levels of volatiles. Attenuation of HIPVs impaired indirect defence responses ofC. sativusby reducing recruitment of EPNs and deterrence ofA. vittatumlarvae. These results suggest that root HIPVs function as honest signals that indicate the presence of herbivores, induction of indirect plant defences and increased risk of predation by natural enemies. However, some herbivores may overcome this line of plant defence by attenuating production of HIPVs and thus altering the outcomes of subsequent interactions among plants, herbivores and natural enemies. A freePlain Language Summarycan be found within the Supporting Information of this article.

Indirect defence, the adaptive top-down control of herbivores by plant traits that enhance predation, is a central component of plant-herbivore interactions. However, the scope of interactions that comprise indirect defence and associated ecological and evolutionary processes has not been clearly defined. We argue that the range of plant traits that mediate indirect defence is much greater than previously thought, and we further organise major concepts surrounding their ecological functioning. Despite the wide range of plant traits and interacting organisms involved, indirect defences show commonalities when grouped. These categories are based on whether indirect defences boost natural enemy abundance via food or shelter resources, or, alternatively, increase natural enemy foraging efficiency via information or alteration of habitat complexity. The benefits of indirect defences to natural enemies should be further explored to establish the conditions in which indirect defence generates a plant-natural enemy mutualism. By considering the broader scope of plant-herbivore-natural enemy interactions that comprise indirect defence, we can better understand plant-based food webs, as well as the evolutionary processes that have shaped them.