Fungal grass endophytes and arthropod communities: lessons from plant defence theory and multitrophic interactions

Alkaloids produced by systemic fungal endophytes of grasses are thought to act as defensive agents against herbivores. Endophytic alkaloids may reduce arthropod herbivore abundances and diversity in agronomic grasses. Yet, accumulating evidence, particularly from native grasses, shows that herbivore preference, abundances and species richness are sometimes greater on endophyte-infected plants, even those with high alkaloids, contrary to the notion of defensive mutualism. We argue that these conflicting results are entirely consistent with well-developed concepts of plant defence theory and tri-trophic interactions. Plant secondary chemicals and endophytic alkaloids often fail to protect plants because: (1) specialist herbivores evolve to detoxify and use defensive chemicals for growth and survival; and (2) natural enemies of herbivores may be more negatively affected by alkaloids than are herbivores. Endophytes and their alkaloids may have profound, but often highly variable, effects on communities, which are also consistent with existing theories of plant defence and community genetics.     

Gene flow in the endophyte Neotyphodium and implications for coevolution with Festuca arizonica

Arizona fescue (Festuca arizonica) often harbours asymptomatic, asexual endophytic fungi from the genus Neotyphodium. In agronomic grasses, Neotyphodium endophytes are often credited with a wide range of mutualistic benefits to its host many of which are related to fungal production of alkaloids for herbivore deterrence. Neotyphodium in the native grass Arizona fescue, however, usually produces alkaloids at levels too low to deter herbivores, and in general, does not behave mutualistically. This study uses microsatellite markers to examine rates of gene flow among four Arizona populations of Neotyphodium. Haplotypic diversity was generally low; only one population contained more than two haplotypes. Haplotypes carrying multiple loci for some or all of the microsatellite loci were also found, indicating a vegetative hybridization event between Neotyphodium and the grass choke pathogen from the genus Epichloë. Gene flow between Neotyphodium populations is very low, and likely much lower than the pollen mediated gene flow of its host. These differing rates of gene flow are predicted to create trait mismatching between endophyte and host and may explain the low, or lack of, alkaloid production by Neotyphodium in Arizona fescue and other native grass species.     

Forest habitat characteristics affect balance between sexual reproduction and clonal propagation of the ectomycorrhizal mushroom Hebeloma cylindrosporum

Endophytic fungi, especially asexual, systemic endophytes in grasses, are generally viewed as plant mutualists, mainly through the action of mycotoxins, such as alkaloids in infected grasses, which protect the host plant from herbivores. Most of the evidence for the defensive mutualism concept is derived from studies of agronomic grass cultivars, which may be atypical of many endophyte-host interactions. I argue that endophytes in native plants, even asexual, seed-borne ones, rarely act as defensive mutualists. In contrast to domesticated grasses where infection frequencies of highly toxic plants often approach 100%, natural grass populations are usually mosaics of uninfected and infected plants. The latter, however, usually vary enormously in alkaloid levels, from none to levels that may affect herbivores. This variation may result from diverse endophyte and host genotypic combinations that are maintained by changing selective pressures, such as competition, herbivory and abiotic factors. Other processes, such as spatial structuring of host populations and endophytes that act as reproductive parasites of their hosts, may maintain infection levels of seed-borne endophytes in natural populations, without the endophyte acting as a mutualist.     

Fungal endophytes of native grasses decrease insect herbivore preference and performance

Endophytic fungal symbionts of grasses are well known for their protective benefit of herbivory reduction. However, the majority of studies on endophyte–grass symbioses have been conducted on economically important, agricultural species—particularly tall fescue (Lolium arundinaceum) and perennial ryegrass (Lolium perenne)—raising the hypothesis that strong benefits are the product of artificial selection. We examined whether fungal endophytes found in natural populations of native grass species deterred insect herbivores. By testing several native grass–endophyte symbiota, we examined phylogenetic signals in the effects of endophytes on insects and compared the relative importance of herbivore and symbiotum identity in the outcome of the interactions. Preference was assessed using three herbivore species [Spodoptera frugiperda (Lepidoptera), Schistocerca americana (Orthoptera), Rhopalosiphum padi (Hemiptera)] and ten native symbiota, which spanned seven grass genera. We also assessed herbivore performance in a no choice experiment for five native symbiota against S. frugiperda. We compared greenhouse and laboratory trials with natural levels of herbivory measured in experimental field populations. In all cases, we included the agronomic grass species, L. arundinaceum, to compare with results from the native grasses. Both in the field and in experimental trials, herbivores showed a significant preference for endophyte-free plant material for the majority of native grasses, with up to three times lower herbivory for endophyte-symbiotic plants; however, the degree of response depended on the identity of the herbivore species. Endophyte presence also significantly reduced performance of S. frugiperda for the majority of grass species. In contrast, the endophyte in L. arundinaceum had few significant anti-herbivore effects, except for a reduction in herbivory at one of two field sites. Our results demonstrate that the mechanisms by which native symbionts deter herbivores are at least as potent as those in model agricultural systems, despite the absence of artificial selection.     

Comparisons of arthropod assemblages on an invasive and native trees: abundance,diversity and damage

The success of exotic plants may be due to lower herbivore loads than those on native plants (Enemies Release Hypothesis). Predictions of this hypothesis include lower herbivore abundances, diversity, and damage on introduced plant species compared to native ones. Greater density or diversity of predators and parasitoids on exotic versus native plants may also reduce regulation of exotic plants by herbivores. To test these predictions, we measured arthropod abundance, arthropod diversity, and foliar damage on invasive Chinese tallow tree (Triadica sebifera) and three native tree species: silver maple (Acer saccharinum), sycamore (Platanus occidentalis), and sweetgum (Liquidambar styraciflua). Arthropod samples were collected with canopy sweep nets from six 20 year old monoculture plots of each species at a southeast Texas site. A total of 2,700 individuals and 285 species of arthropods were caught. Overall, the species richness and abundance of arthropods on tallow tree were similar to the natives. But, ordination (NMS) showed community composition differed on tallow tree compared to all three native trees. It supported an arthropod community that had relatively lower herbivore abundance but relatively more predator species compared to the native species examined. Leaves were collected to determine damage. Tallow tree experienced less mining damage than native trees. The results of this study supported the Enemies Release Hypothesis predictions that tallow tree would have low herbivore loads which may contribute to its invasive success. Moreover, a shift in the arthropod community to fewer herbivores without a reduction in predators may further limit regulation of this exotic species by herbivores in its introduced range.     

Experimental Manipulation of Grassland Plant Diversity Induces Complex Shifts in Aboveground Arthropod Diversity

Changes in producer diversity cause multiple changes in consumer communities through various mechanisms. However, past analyses investigating the relationship between plant diversity and arthropod consumers focused only on few aspects of arthropod diversity, e.g. species richness and abundance. Yet, shifts in understudied facets of arthropod diversity like relative abundances or species dominance may have strong effects on arthropod-mediated ecosystem functions. Here we analyze the relationship between plant species richness and arthropod diversity using four complementary diversity indices, namely: abundance, species richness, evenness (equitability of the abundance distribution) and dominance (relative abundance of the dominant species). Along an experimental gradient of plant species richness (1, 2, 4, 8, 16 and 60 plant species), we sampled herbivorous and carnivorous arthropods using pitfall traps and suction sampling during a whole vegetation period. We tested whether plant species richness affects consumer diversity directly (i), or indirectly through increased productivity (ii). Further, we tested the impact of plant community composition on arthropod diversity by testing for the effects of plant functional groups (iii). Abundance and species richness of both herbivores and carnivores increased with increasing plant species richness, but the underlying mechanisms differed between the two trophic groups. While higher species richness in herbivores was caused by an increase in resource diversity, carnivore richness was driven by plant productivity. Evenness of herbivore communities did not change along the gradient in plant species richness, whereas evenness of carnivores declined. The abundance of dominant herbivore species showed no response to changes in plant species richness, but the dominant carnivores were more abundant in species-rich plant communities. The functional composition of plant communities had small impacts on herbivore communities, whereas carnivore communities were affected by forbs of small stature, grasses and legumes. Contrasting patterns in the abundance of dominant species imply different levels of resource specialization for dominant herbivores (narrow food spectrum) and carnivores (broad food spectrum). That in turn could heavily affect ecosystem functions mediated by herbivorous and carnivorous arthropods, such as herbivory or biological pest control.     

Tree diversity drives diversity of arthropod herbivores,but successional stage mediates detritivores

The high tree diversity of subtropical forests is linked to the biodiversity of other trophic levels. Disentangling the effects of tree species richness and composition, forest age, and stand structure on higher trophic levels in a forest landscape is important for understanding the factors that promote biodiversity and ecosystem functioning. Using a plot network spanning gradients of tree diversity and secondary succession in subtropical forest, we tested the effects of tree community characteristics (species richness and composition) and forest succession (stand age) on arthropod community characteristics (morphotype diversity, abundance and composition) of four arthropod functional groups. We posit that these gradients differentially affect the arthropod functional groups, which mediates the diversity, composition, and abundance of arthropods in subtropical forests. We found that herbivore richness was positively related to tree species richness. Furthermore, the composition of herbivore communities was associated with tree species composition. In contrast, detritivore richness and composition was associated with stand age instead of tree diversity. Predator and pollinator richness and abundance were not strongly related to either gradient, although positive trends with tree species richness were found for predators. The weaker effect of tree diversity on predators suggests a cascading diversity effect from trees to herbivores to predators. Our results suggest that arthropod diversity in a subtropical forest reflects the net outcome of complex interactions among variables associated with tree diversity and stand age. Despite this complexity, there are clear linkages between the overall richness and composition of tree and arthropod communities, in particular herbivores, demonstrating that these trophic levels directly impact each other.     

Grass-herbivore interactions altered by strains of a native endophyte

Many plants support symbiotic microbes, such as endophytic fungi, that can alter interactions with herbivores. Most endophyte research has focused on agronomically important species, with less known about the ecological roles of native endophytes in native plants. In particular, whether genetic variation among endophyte symbionts affects herbivores of plant hosts remains unresolved for most native endophytes. Here, we investigate the importance of native isolates of the endophyte Epichlo? elymi in affecting herbivory of the native grass host, Elymus hystrix. Experimental fungal isolate-plant genotype combinations and endophyte-free control plants were grown in a common garden and exposed to natural arthropod herbivory. Fungal isolates differed in their effects on two types of herbivory, chewing and scraping. Isolates exhibiting greater sexual reproduction were associated with greater herbivore damage than primarily asexual isolates. Endophyte infection also altered patterns of herbivory within plants, with stroma-bearing tillers experiencing up to 30% greater damage than nonstroma-bearing tillers. Results suggest that intraspecific genetic variation in endophytes, like plant genetic variation, can have important 'bottom-up' effects on herbivores in native systems.     

Dynamics of plant and arthropod diversity during old field succession

The successional dynamics of arthropod diversity in 18 abandoned agricultural fields (age 15-54 yr) at Cedar Creek. MN. USA were determined using sweep net sampling (44833 individuals of 618 species). Total arthropod species richness and equitability (J), but not abundance, increased significantly with field successional age. Herbivore and parasite species richness, but not detritivore and predator species richness, also increased significantly with field age. All of these arthropod variables were significantly positively correlated with plant species richness in the fields. When plant species richness was included as a covariate in regressions, there were no longer any significant effects of field age. These results supported the hypothesis that increases in arthropod diversity with field age are influenced by increases in plant diversity. The additional significant positive dependence of herbivore species richness on predator species richness suggests that predator-prey interactions may also influence the successional dynamics of arthropod diversity. Nine of the ten most common arthropod species decreased in abundance with field age, two of them significantly. The abundances of these two generalist forb-feeding species, Melanoplus femurrubrum (Orthoptera: Acrididae) and Scaphytopius acutus (Homoptera: Cicadellidae). each depended significantly on amount of forbs. The average body size of arthropod species (total and herbivores) decreased significantly with field age. An efficiency vs specialization hypothesis predicts such a decrease. Because plants in later secondary succession are generally less palatable, a diversity of smaller, potentially more specialized herbivores may have an advantage over larger and more efficient herbivores in later succession.     

Plant species loss decreases arthropod diversity and shifts trophic structure

Plant diversity is predicted to be positively linked to the diversity of herbivores and predators in a foodweb. Yet, the relationship between plant and animal diversity is explained by a variety of competing hypotheses, with mixed empirical results for each hypothesis. We sampled arthropods for over a decade in an experiment that manipulated the number of grassland plant species. We found that herbivore and predator species richness were strongly, positively related to plant species richness, and that these relationships were caused by different mechanisms at herbivore and predator trophic levels. Even more dramatic was the threefold increase, from low- to high-plant species richness, in abundances of predatory and parasitoid arthropods relative to their herbivorous prey. Our results demonstrate that, over the long term, the loss of plant species propagates through food webs, greatly decreasing arthropod species richness, shifting a predator-dominated trophic structure to being herbivore dominated, and likely impacting ecosystem functioning and services.     

Interactions among fungal endophytes,grasses and herbivores

The interaction between two species often depends on the presence or absence of a third species. One widespread three-species interaction involves fungal endophytes infecting grasses and the herbivores that feed upon them. The endophytes are allied with the fungal family Clavicipitaceae and grow systemically in intercellular spaces in above-ground plant tissues including seeds. Like relatedClaviceps species, the endophytes produce a variety of alkaloids that make the host plants toxic or distasteful to herbivores. A large number of grass species are infected, especially cool-season grasses in temperate areas. Field and laboratory studies have shown that herbivores avoid infected plants in choice trials and suffer increased mortality and decreased growth on infected grasses in feeding experiments. Resistance to herbivores may provide a selective advantage to infected plants in competitive interactions with noninfected plants. Recent studies have shown that differential herbivory can reverse competitive hierarchies among plant species. Both endophyte-infected and noninfected tall fescue grass (Festuca arundinacea) are outcompeted by orchardgrass (Dactylis glomerata) in the absence of insect herbivory. However, when herbivores are present infected tall fescue outcompetes orchardgrass. These results suggest that the frequency of infection in grass species and grassland communities will increase over time. Several studies are reviewed illustrating increases in infection frequency within grass populations subject to herbivore pressure. Endophytic fungi may be important regulators of plant-herbivore interactions and so indirectly affect the structure and dynamics of plant communities.     

Effects of endophyte infection on the competitive ability of Achnatherum sibiricum depend on endophyte species and nitrogen availability

Aims The leaves of almost all terrestrial plant species are colonized by endophytic fungi. Compared to agronomic grasses, which usually harbor few endophytes, native grasses generally possess greater endophyte species diversity. Existing studies examining endophyte effects on natural grasses under competition normally considered the infection status (infected or uninfected), and rarely considered endophyte species.     

Biogeographical comparison of the arthropod herbivore communities associated with Lepidium draba in its native, expanded and introduced ranges

Aim To examine the composition and structure of the arthropod community on the invasive weed Lepidium draba in its native, expanded and introduced ranges, in order to elucidate the lack of a biotic constraint that may facilitate invasion. Location Europe and western North America. Methods Identical sampling protocols were used to collect data from a total of 35 populations of L. draba in its native (Eastern European), expanded (Western European) and introduced (western US) ranges. A bootstrapping analysis was used to compare herbivore richness, diversity and evenness among the regions. Core species groups (monophages, oligophages and polyphages) on the plant were defined and their abundances and host utilization patterns described. Results Species richness was greatest in the native range, while species diversity and evenness were similar in the native and expanded range, but significantly greater than in the introduced range of L. draba. Specialist herbivore abundance was greater in the native and expanded compared with the introduced range. Oligophagous Brassicaceae‐feeders were equally abundant in all three ranges, and polyphagous herbivore abundance was significantly greater in the introduced range. Overall herbivore abundance was greater in the introduced range. Host utilization was more complete in the two European ranges due to monophagous herbivores that do not exist in the introduced range. Root feeders and gall formers were completely absent from the introduced range, which was dominated by generalist sap‐sucking herbivores. However, one indigenous stem‐mining weevil, Ceutorhynchus americanus, occurred on L. draba in the introduced range. Main conclusions This is, to our knowledge, the first study documenting greater herbivore abundance on an invasive weed in its introduced, compared with its native, range. However, greater abundance does not necessarily translate to greater impact. We argue that, despite the greater total herbivore abundance in the introduced range, differences in the herbivore community structure (specialist vs. generalist herbivory) may contribute to the invasion success of L. draba in the western USA.     

Plant diversity induces shifts in the functional structure and diversity across trophic levels

Changes to primary producer diversity can cascade up to consumers and affect ecosystem processes. Although the effect of producer diversity on higher trophic groups have been studied, these studies often quantify taxonomy‐based measures of biodiversity, like species richness, which do not necessarily reflect the functioning of these communities. In this study, we assess how plant species richness affects the functional composition and diversity of higher trophic levels and discuss how this might affect ecosystem processes, such as herbivory, predation and decomposition. Based on six different consumer traits, we examined the functional composition of arthropod communities sampled in experimental plots that differed in plant species richness. The two components we focused on were functional variation in the consumer community structure (functional structure) and functional diversity, expressed as functional richness, evenness and divergence. We found a consistent positive effect of plant species richness on the functional richness of herbivores, carnivores, and omnivores, but not decomposers, and contrasting patterns for functional evenness and divergence. Increasing plant species richness shifted the omnivore community to more predatory and less mobile species, and the herbivore community to more specialized and smaller species. This was accompanied by a shift towards more species occurring in the vegetation than in the ground layer. Our study shows that plant species richness strongly affects the functional structure and diversity of aboveground arthropod communities. The observed shifts in body size (herbivores), specialization (herbivores), and feeding mode (omnivores) together with changes in the functional diversity may underlie previously observed increases in herbivory and predation in plant communities of higher diversity.     

Asexual Endophytes in a Native Grass: Tradeoffs in Mortality, Growth, Reproduction, and Alkaloid Production

Neotyphodium endophytes are asexual, seed-borne fungal symbionts that are thought to interact mutualistically with their grass hosts. Benefits include increased growth, reproduction, and resistance to herbivores via endophytic alkaloids. Although these benefits are well established in infected introduced, agronomic grasses, little is known about the cost and benefits of endophyte infection in native grass populations. These populations exist as mosaics of uninfected and infected plants, with the latter often comprised of plants that vary widely in alkaloid content. We tested the costs and benefits of endophyte infections with varying alkaloids in the native grass Achnatherum robustum (sleepygrass). We conducted a 4-year field experiment, where herbivory and water availability were controlled and survival, growth, and reproduction of three maternal plant genotypes [uninfected plants (E−), infected plants with high levels of ergot alkaloids (E+A+), and infected plants with no alkaloids (E+A−)] were monitored over three growing seasons. Generally, E+A+ plants had reduced growth over the three growing seasons and lower seed production than E− or E+A− plants, suggesting a cost of alkaloid production. The reduction in vegetative biomass in E+A+ plants was most pronounced under supplemented water, contrary to the prediction that additional resources would offset the cost of alkaloid production. Also, E+A+ plants showed no advantage in growth, seed production, or reproductive effort under full herbivory relative to E− or E+A− grasses, contrary to the predictions of the defensive mutualism hypothesis. However, E+A+ plants had higher overwintering survival than E+A− plants in early plant ontogeny, suggesting that alkaloids associated with infection may protect against below ground herbivory or harsh winter conditions. Our results suggest that the mosaic of E−, E+A+, and E+A− plants observed in nature may result from varying biotic and abiotic selective factors that maintain the presence of uninfected plants and infected plants that vary in alkaloid production.     

Effects of exotic annual grass litter and local environmental gradients on annual plant community structure

Exotic annual grasses have been introduced into many semi-arid ecosystems worldwide, often to the detriment of native plant communities. The accumulation of litter from these grasses (i.e. residual dry biomass) has been demonstrated to negatively impact native plant communities and promote positive feedbacks to exotic grass persistence. More targeted experiments are needed, however, to determine the relative impact of exotic grass litter on plant community structure across local environmental gradients. We experimentally added exotic grass litter to annual forb-dominated open woodland communities positioned along natural canopy cover gradients in southwest Western Australia. These communities are an important component of this region’s plant biodiversity hotspot and are documented to be under threat from exotic annual grasses. After a one-year treatment period, we measured the effects of exotic grass litter, soil properties, and canopy cover on native and exotic species richness and abundance, as well as common species’ biomass and abundances. Plant community structure was more strongly influenced by soil properties and canopy cover than by grass litter. Total plant abundances per plot, however, were significantly lower in litter addition plots than control plots, a trend driven by native species. Exotic grass litter was also associated with lower abundances of one very common native species: Waitzia acuminata. Our results suggest that exotic grass litter limits the establishment of some native species in this system. Over multiple years, these subtle impacts may contribute substantially to the successful advancement of exotic species into this system, particularly in certain microenvironments.     

Response of fungal endophyte communities within Andropogon gerardii (Big bluestem) to nutrient addition and herbivore exclusion

Fungal endophytes may alter plant responses to the environment, but how does the environment affect the communities of fungal symbionts within plants? We examined the impact of nutrient addition and herbivore exclusion on endophyte communities of the prairie grass Andropogon gerardii in a full factorial field experiment. Fungi were cultured from stems, young leaves, and mature leaves, ITS sequences obtained, and endophyte incidence, community richness, and composition analyzed. Results indicate that in plots where nutrient addition and herbivore exclusion treatments had been applied separately, fungal endophyte incidence, community composition or evenness did not differ, but that greater species richness was observed in plots with nutrient addition and herbivore exclusion treatments applied in combination, compared to other treatments. Further, although fungal community composition was significantly different in stem and leaf tissues, OTU richness was greater in all endophyte communities in nutrient addition plus herbivore exclusion treatments, regardless of tissue type. Our results indicate the distinct fungal endophyte communities found in different plant tissues respond similarly to environmental factors.     

Proportion of exotics and relatedness of host species mediate the positive effect of plant richness on the species richness of fruit flies

1. All else being equal, the greater the local species richness of plants, the greater the number of associated herbivore species. Because most herbivore insects feed on a subset of closely related plant species, plant phylogenetic diversity is expected to play a key role in determining the number of herbivore species. What is not well known, however, is how an increase in the species richness of exotic plants affects the species richness of herbivores. 2. In this study, we used plant–fruit fly interactions to investigate the influence of the proportion and species richness of exotic host plants on the species richness of herbivorous insects. We also tested whether the phylogenetic diversity of host plants increases when the number of exotic plant species increases. 3. We found that the species richness of fruit flies is more accurately predicted by the richness of native host plants than by total plant species richness (including both native and exotic species). The proportion of exotic host species and the phylogenetic diversity of host plants had negative and positive effects, respectively, on the species richness of fruit flies. 4. Our findings suggest that a positive effect of plant richness on herbivore richness occurs only when an increase in plant diversity involves plant species with which native herbivores share some evolutionary history.     

Fungal endophytes: common host plant symbionts but uncommon mutualists

Fungal endophytes are extremely common and highly diverse microorganismsthat live within plant tissues, but usually remain asymptomatic.Endophytes traditionally have been considered plant mutualists,mainly by reducing herbivory via production of mycotoxins, suchas alkaloids. However, the vast majority of endophytes, especiallyhorizontally-transmitted ones commonly found in woody plants,apparently have little or no effect on herbivores. For the systemic,vertically-transmitted endophytes of grasses, mutualistic interactionsvia increased resistance to herbivores and pathogens are morecommon, as predicted by evolutionary theory. However, even inthese obligate symbioses, endophytes are often neutral or evenpathogenic to the host grass, depending on endophyte and plantgenotype and environmental conditions. We present a graphical model based upon variation in nitrogenflux in the host plant. Nitrogen is a common currency in endophyte/hostand plant/herbivore interactions in terms of limitations tohost plant growth, enhanced uptake by endophytes, demand forsynthesis of nitrogen-rich alkaloids, and herbivore preferenceand performance. Our graphical model predicts that low alkaloid-producingendophytes should persist in populations when soil nutrientsand herbivory are low. Alternatively, high alkaloid endophytesare favored under increasing herbivory and increasing soil nitrogen,at least to some point. At very high soil nitrogen levels, uninfectedplants may be favored over either type of infected plants. Thesepredictions are supported by patterns of infection and alkaloidproduction in nature, as well by a manipulative field experiment.However, plant genotype and other environmental factors, suchas available water, interact with the presence of the endophyteto influence host plant performance.     

Effects of genetic diversity of grass on insect species diversity at higher trophic levels are not due to cascading diversity effects

Genetic diversity in plant populations has been shown to affect the species diversity of insects. In grasses, infection with fungal endophytes can also have strong effects on insects, potentially modifying the effects of plant genetic diversity. We manipulated the genetic diversity and endophyte infection of a grass in a field experiment. We show that diversity of primary parasitoids (3rd trophic level) and, especially, secondary parasitoids (4th trophic level) increases with grass genetic diversity while there was no effect of endophyte infection. The increase in insect diversity appeared to be due to a complementarity effect rather than a sampling effect. The higher parasitoid diversity could not be explained by a cascading diversity effect because herbivore diversity was not affected and the same herbivore species were present in all treatments. The effects on the higher trophic levels must therefore be due to a direct response to plant traits or mediated by effects on traits at intermediate trophic levels.