Herbivore-induced plant volatiles, not natural enemies, mediate a positive indirect interaction between insect herbivores.

Many insect herbivores engage in apparent competition whereby two species interact through shared natural enemies. Upon insect attack, plants release volatile blends that attract natural enemies, but whether these volatiles mediate apparent competition between herbivores is not yet known. We investigate the role of volatiles that are emitted by bean plants upon infestation by Acyrthosiphon pisum aphids on the population dynamics and fitness of Sitobion avenae aphids, and on wheat phloem sap metabolites. In a field experiment, the dynamics of S. avenae aphids on wheat were studied by crossing two treatments: exposure of aphid colonies to A. pisum-induced bean volatiles and exclusion of natural enemies. Glasshouse experiments and analyses of primary metabolites in wheat phloem exudates were performed to better understand the results from the field experiment. In the field, bean volatiles did not affect S. avenae dynamics or survival when aphids were exposed to natural enemies. When protected from them, however, volatiles led to larger aphid colonies. In agreement with this observation, in glasshouse experiments, aphid-induced bean volatiles increased the survival of S. avenae aphids on wheat plants, but not on an artificial diet. This suggests that volatiles may benefit S. avenae colonies via metabolic changes in wheat plants, although we did not find any effect on wheat phloem exudate composition. We report a potential case of associational susceptibility whereby plant volatiles weaken the defences of receiving plants, thus leading to increased herbivore performance.

LEDs Make It Resilient: Effects on Plant Growth and Defense.

Light spectral composition influences plant growth and metabolism, and has important consequences for interactions with plant-feeding arthropods and their natural enemies. In greenhouse horticulture, light spectral composition can be precisely manipulated by light-emitting diodes (LEDs), and LEDs are already used to optimize crop production and quality. However, because light quality also modulates plant secondary metabolism and defense, it is important to understand the underlying mechanisms in the context of the growth-defense trade-off. We review the effects of the spectral composition of supplemental light currently used, or potentially used, in greenhouse horticulture on the mechanisms underlying plant growth and defense. This information is important for exploring opportunities to optimize crop performance and pest management, and thus for developing resilient crop-production systems.

Plant-mediated interactions between two herbivores differentially affect a subsequently arriving third herbivore in populations of wild cabbage.

Plants are part of biodiverse communities and frequently suffer from attack by multiple herbivorous insects. Plant responses to these herbivores are specific for insect feeding guilds: aphids and caterpillars induce different plant phenotypes. Moreover, plants respond differentially to single or dual herbivory, which may cascade into a chain of interactions in terms of resistance to other community members. Whether differential responses to single or dual herbivory have consequences for plant resistance to yet a third herbivore is unknown. We assessed the effects of single or dual herbivory by Brevicoryne brassicae aphids and/or Plutella xylostella caterpillars on resistance of plants from three natural populations of wild cabbage to feeding by caterpillars of Mamestra brassicae. We measured plant gene expression and phytohormone concentrations to illustrate mechanisms involved in induced responses. Performance of both B. brassicae and P. xylostella was reduced when feeding simultaneously with the other herbivore, compared to feeding alone. Gene expression and phytohormone concentrations in plants exposed to dual herbivory were different from those found in plants exposed to herbivory by either insect alone. Plants previously induced by both P. xylostella and B. brassicae negatively affected growth of the subsequently arriving M. brassicae. Furthermore, induced responses varied between wild cabbage populations. Feeding by multiple herbivores differentially activates plant defences, which has plant-mediated negative consequences for a subsequently arriving herbivore. Plant population-specific responses suggest that plant populations adapt to the specific communities of insect herbivores. Our study contributes to the understanding of plant defence plasticity in response to multiple insect attacks.

Variation in plant-mediated interactions between rhizobacteria and caterpillars: potential role of soil composition.

Selected strains of non-pathogenic rhizobacteria can trigger induced systemic resistance (ISR) in plants against aboveground insect herbivores. However, the underlying mechanisms of plant-mediated interactions between rhizobacteria and herbivorous insects are still poorly understood. Using Arabidopsis thaliana Col-0-Pseudomonas fluorescens WCS417r as a model system, we investigated the performance and the molecular mechanisms underlying plant-mediated effects of rhizobacteria on the generalist caterpillar Mamestra brassicae and the specialist Pieris brassicae. Rhizobacteria colonisation of Arabidopsis roots resulted in decreased larval weight of M. brassicae, whereas no effect was observed on larval weight of P. brassicae. Using a jasmonic acid (JA)-impaired mutant (dde2-2), we confirmed the importance of JA in rhizobacteria-mediated ISR against M. brassicae. Interestingly, in some experiments we also observed rhizobacteria-induced systemic susceptibility to M. brassicae. The role of soil composition in the variable outcomes of microbe-plant-insect interactions was then assessed by comparing M. brassicae performance and gene transcription in plants grown in potting soil or a mixture of potting soil and sand in a 1:1 ratio. In a mixture of potting soil and sand, rhizobacteria treatment had a consistent negative effect on M. brassicae, whereas the effect was more variable in potting soil. Interestingly, at 24 h post-infestation (hpi) rhizobacteria treatment primed plants grown in a mixture of potting soil and sand for stronger expression of the JA- and ethylene-regulated genes PDF1.2 and HEL, respectively. Our study shows that soil composition can modulate rhizobacteria-plant-insect interactions, and is a factor that should be considered when studying these belowground-aboveground interactions.

Rhizobacteria modify plant-aphid interactions: a case of induced systemic susceptibility.

Beneficial microbes, such as plant growth-promoting rhizobacteria and mycorrhizal fungi, may have a plant-mediated effect on insects aboveground. The plant growth-promoting rhizobacterium Pseudomonas fluorescens can induce systemic resistance in Arabidopsis thaliana against several microbial pathogens and chewing insects. However, the plant-mediated effect of these beneficial microbes on phloem-feeding insects is not well understood. Using Arabidopsis as a model, we here report that P. fluorescens has a positive effect on the performance (weight gain and intrinsic rate of increase) of the generalist aphid Myzus persicae, while no effect was recorded on the crucifer specialist aphid Brevicoryne brassicae. Additionally, transcriptional analyses of selected marker genes revealed that in the plant-microbe interaction with M. persicae, rhizobacteria (i) prime the plant for enhanced expression of LOX2, a gene involved in the jasmonic acid (JA)-regulated defence pathway, and (ii) suppress the expression of ABA1, a gene involved in the abscisic acid (ABA) signalling pathway, at several time points. In contrast, almost no effect of the plant-microbe interaction with B. brassicae was found at the transcriptional level. This study presents the first data on rhizobacteria-induced systemic susceptibility to an herbivorous insect, supporting the pattern proposed for other belowground beneficial microbes and aboveground phloem feeders. Moreover, we provide further evidence that at the transcript level, soil-borne microbes modify plant-aphid interactions.

Herbivore-induced volatiles of cabbage (Brassica oleracea) prime defence responses in neighbouring intact plants.

When attacked by herbivores, plants release herbivore-induced plant volatiles (HIPV) that may function in direct defence by repelling herbivores or reducing their growth. Emission of HIPV may also contribute to indirect defence by attracting natural enemies of the herbivore. Here, cabbage (Brassica oleracea L.) plants (receiver plants) previously exposed to HIPV and subsequently induced through feeding by five Pieris brassicae L. caterpillars attracted more Cotesia glomerata L. parasitoids than control plants. HIPVs to which receiver plants had been exposed were emitted by B. oleracea infested with 50 P. brassicae caterpillars. Control plants had been exposed to volatiles from undamaged plants. In contrast, there were no differences in the attraction of wasps to receiver plants induced through feeding of one or ten larvae of P. brassicae compared to control plants. In addition, RT-PCR demonstrated higher levels of LIPOXYGENASE (BoLOX) transcripts in HIPV-exposed receiver plants. Exposure to HIPV from emitter plants significantly inhibited the growth rate of both P. brassicae and Mamestra brassicae caterpillars compared to growth rates of caterpillars feeding on control receiver plants. Our results demonstrate plant-plant signalling leading to priming of both indirect and direct defence in HIPV-exposed B. oleracea plants.

CREB expression in the brains of two closely related parasitic wasp species that differ in long-term memory formation.

The cAMP/PKA signalling pathway and transcription factor cAMP response element-binding protein (CREB) play key roles in long-term memory (LTM) formation. We used two closely related parasitic wasp species, Cotesia glomerata and Cotesia rubecula, which were previously shown to be different in LTM formation, and sequenced at least nine different CREB transcripts in both wasp species. The splicing patterns, functional domains and amino acid sequences were similar to those found in the CREB genes of other organisms. The predicted amino acid sequences of the CREB isoforms were identical in both wasp species. Using real-time quantitative PCR we found that two low abundant CREB transcripts are differentially expressed in the two wasps, whereas the expression levels of high abundant transcripts are similar.

The role of volatiles in aggregation and host-seeking of the haematophagous poultry red mite Dermanyssus gallinae (Acari: Dermanyssidae).

Infestations with ectoparasitic poultry red mites (Dermanyssus gallinae) pose an increasing threat to poultry health and welfare. Because of resistance to acaricides and higher scrutiny of poultry products, alternative and environmentally safe management strategies are warranted. Therefore, we investigated how volatile cues shape the behavior of D. gallinae and how this knowledge may be exploited in the development of an attract-and-kill method to control mite populations. A Y-tube olfactometer bio-assay was used to evaluate choices of mites in response to cues related to conspecific mites as well as related to their chicken host. Both recently fed and starved mites showed a strong preference (84 and 85%, respectively) for volatiles from conspecific, fed mites as compared to a control stream of clean air. Mites were also significantly attracted to 'aged feathers' (that had remained in the litter for 3-4 days), but not to 'fresh feathers'. Interestingly, an air stream containing 2.5% CO(2), which mimics the natural concentration in air exhaled by chickens, did attract fed mites, but inhibited the attraction of unfed mites towards volatiles from aged feathers. We conclude that both mite-related cues (aggregation pheromones) and host-related cues (kairomones) mediate the behavior of the poultry mite. We discuss the options to exploit this knowledge as the 'attract' component of attract-and-kill strategies for the control of D. gallinae.

Flower vs. leaf feeding by Pieris brassicae: glucosinolate-rich flower tissues are preferred and sustain higher growth rate.

Interactions between butterflies and caterpillars in the genus Pieris and plants in the family Brassicaceae are among the best explored in the field of insect-plant biology. However, we report here for the first time that Pieris brassicae, commonly assumed to be a typical folivore, actually prefers to feed on flowers of three Brassica nigra genotypes rather than on their leaves. First- and second-instar caterpillars were observed to feed primarily on leaves, whereas late second and early third instars migrated via the small leaves of the flower branches to the flower buds and flowers. Once flower feeding began, no further leaf feeding was observed. We investigated growth rates of caterpillars having access exclusively to either leaves of flowering plants or flowers. In addition, we analyzed glucosinolate concentrations in leaves and flowers. Late-second- and early-third-instar P. brassicae caterpillars moved upward into the inflorescences of B. nigra and fed on buds and flowers until the end of the final (fifth) instar, after which they entered into the wandering stage, leaving the plant in search of a pupation site. Flower feeding sustained a significantly higher growth rate than leaf feeding. Flowers contained levels of glucosinolates up to five times higher than those of leaves. Five glucosinolates were identified: the aliphatic sinigrin, the aromatic phenylethylglucosinolate, and three indole glucosinolates: glucobrassicin, 4-methoxyglucobrassicin, and 4-hydroxyglucobrassicin. Tissue type and genotype were the most important factors affecting levels of identified glucosinolates. Sinigrin was by far the most abundant compound in all three genotypes. Sinigrin, 4-hydroxyglucobrassicin, and phenylethylglucosinolate were present at significantly higher levels in flowers than in leaves. In response to caterpillar feeding, sinigrin levels in both leaves and flowers were significantly higher than in undamaged plants, whereas 4-hydroxyglucobrassicin leaf levels were lower. Our results show that feeding on flower tissues, containing higher concentrations of glucosinolates, provides P. brassicae with a nutritional benefit in terms of higher growth rate. This preference appears to be in contrast to published negative effects of volatile glucosinolate breakdown products on the closely related Pieris rapae.

Olfactory responses of banana weevil predators to volatiles from banana pseudostem tissue and synthetic pheromone.

As a response to attack by herbivores, plants can emit a variety of volatile substances that attract natural enemies of these insect pests. Predators of the banana weevil, Cosmopolites sordidus (Germar) (Coleoptera: Curculionidae) such as Dactylosternum abdominale (Coleoptera: Hydrophilidae) and Pheidole megacephala (Hymenoptera: Formicidae), are normally found in association with weevil-infested rotten pseudostems and harvested stumps. We investigated whether these predators are attracted to such environments in response to volatiles produced by the host plant, by the weevil, or by the weevil plant complex. We evaluated predator responses towards volatiles from banana pseudostem tissue (synomones) and the synthetic banana weevil aggregation pheromone Cosmolure+ in a two-choice olfactometer. The beetle D. abdominale was attracted to fermenting banana pseudostem tissue and Cosmolure+, whereas the ant P. megacephala was attracted only to fermented pseudostem tissue. Both predators were attracted to banana pseudostem tissue that had been damaged by weevil larvae irrespective of weevil presence. Adding pheromone did not enhance predator response to volatiles from pseudostem tissue fed on by weevils. The numbers of both predators recovered with pseudostem traps in the field from banana mats with a pheromone trap were similar to those in pseudostem traps at different distance ranges from the pheromone. Our study shows that the generalist predators D. abdominale and P. megacephala use volatiles from fermented banana pseudostem tissue as the major chemical cue when searching for prey.

Behavioural responses of diamondback moth Plutella xylostella (Lepidoptera: Plutellidae) to extracts derived from Melia azedarach and Azadirachta indica.

The impact of three different doses of botanical insecticide derived from the syringa tree, Melia azedarach and the neem tree, Azadirachta indica was tested on the behaviour of the diamondback moth, Plutella xylostella (Linnaeus). Both botanical insecticides had a significant impact on larval behaviour. At higher doses the extracts showed feeding deterrent activity, with larvae preferring the untreated sides of cabbage leaves and consuming less of the treated half of cabbage leaves. The botanical insecticides had less of an effect on the oviposition behaviour of P. xylostella moths. In laboratory and glasshouse trials, significantly fewer eggs were oviposited on the plants that had been treated with syringa extracts. Therefore, the syringa extracts appear to have a repellent effect. In contrast, when exposed to the neem extracts the moths did not discriminate between control plants and treated plants. Behavioural observation indicated that, despite the lower number of eggs oviposited on cabbage treated with syringa extracts, the moths chose cabbage treated with the highest dose of syringa more often than they chose control cabbage plants. Similar observations were found in cabbage plants treated with neem, moths chose the medium dose more often than they chose the control. Oviposition and feeding deterrent properties are important factors in pest control, and results from this study indicate that botanical insecticides have the potential to be incorporated into control programmes for P. xylostella in South Africa.

Indirect defence of plants against herbivores: using Arabidopsis thaliana as a model plant.

In their defence against pathogens, herbivorous insects, and mites, plants employ many induced responses. One of these responses is the induced emission of volatiles upon herbivory. These volatiles can guide predators or parasitoids to their herbivorous prey, and thus benefit both plant and carnivore. This use of carnivores by plants is termed indirect defence and has been reported for many plant species, including elm, pine, maize, Lima bean, cotton, cucumber, tobacco, tomato, cabbage, and Arabidopsis thaliana. Herbivory activates an intricate signalling web and finally results in defence responses such as increased production of volatiles. Although several components of this signalling web are known (for example the plant hormones jasmonic acid, salicylic acid, and ethylene), our understanding of how these components interact and how other components are involved is still limited. Here we review the knowledge on elicitation and signal transduction of herbivory-induced volatile production. Additionally, we discuss how use of the model plant Arabidopsis thaliana can enhance our understanding of signal transduction in indirect defence and how cross-talk and trade-offs with signal transduction in direct defence against herbivores and pathogens influences plant responses.

Attraction of Phytoseiulus persimilis (Acari: Phytoseiidae) towards volatiles from various Tetranychus urticae-infested plant species.

Plants infested with the spider mite Tetranychus urticae Koch, may indirectly defend themselves by releasing volatiles that attract the predatory mite Phytoseiulus persimilis Athias-Henriot. Several plants from different plant families that varied in the level of spider mite acceptance were tested in an olfactometer. The predatory mites were significantly attracted to the spider mite-infested leaves of all test plant species. No differences in attractiveness of the infested plant leaves were found for predatory mites reared on spider mites on the different test plants or on lima bean. Thus, experience with the spider mite-induced plant volatiles did not affect the predatory mites. Jasmonic acid was applied to ginkgo leaves to induce a mimic of a spider mite-induced volatile blend, because the spider mites did not survive when incubated on ginkgo. The volatile blend induced in ginkgo by jasmonic acid was slightly attractive to predatory mites. Plants with a high degree of direct defence were thought to invest less in indirect defence than plants with a low degree of direct defence. However, plants that had a strong direct defence such as ginkgo and sweet pepper, did emit induced volatiles that attracted the predatory mite. This indicates that a combination of direct and indirect defence is to some extent compatible in plant species.

Herbivore-induced volatile production by Arabidopsis thaliana leads to attraction of the parasitoid Cotesia rubecula: chemical, behavioral, and gene-expression analysis.

Many plant species defend themselves against herbivorous insects indirectly by producing volatiles in response to herbivory. These volatiles attract carnivorous enemies of the herbivores. Research on the model plant Arabidopsis thaliana (L.) Heynh. has contributed considerably to the unraveling of signal transduction pathways involved in direct plant defense mechanisms against pathogens. Here, we demonstrate that Arabidopsis is also a good candidate for studying signal transduction pathways involved in indirect defense mechanisms by showing that: (1) Adult females of Cotesia rubecula, a specialist parasitic wasp of Pieris rapae caterpillars, are attracted to P. rapae-infested Arabidopsis plants. (2) Arabidopsis infested by P. rapae emits volatiles from several major biosynthetic pathways, including terpenoids and green leaf volatiles. The blends from herbivore-infested and artificially damaged plants are similar. However, differences can be found with respect to a few components of the blend, such as two nitriles and the monoterpene myrcene, that were produced exclusively by caterpillar-infested plants, and methyl salicylate, that was produced in larger amounts by caterpillar-infested plants. (3) Genes from major biosynthetic pathways involved in volatile production are induced by caterpillar feeding. These include AtTPS10, encoding a terpene synthase involved in myrcene production, AtPAL1, encoding phenylalanine ammonia-lyase involved in methyl salicylate production, and AtLOX2 and AtHPL, encoding lipoxygenase and hydroperoxide lyase, respectively, both involved in the production of green leaf volatiles. AtAOS, encoding allene oxide synthase, involved in the production of jasmonic acid, also was induced by herbivory.

Comparison of cultivars of ornamental crop Gerbera jamesonii on production of spider mite-induced volatiles, and their attractiveness to the predator Phytoseiulus persimilis.

We investigated whether volatiles produced by spider mite-damaged plants of four gerbera cultivars differ in attractiveness to Phytoseiulus persimilis, a specialist predator of spider mites, and how the mite-induced odor blends differ in chemical composition. The gerbera cultivars differed in resistance, as expressed in terms of spider mite intrinsic rate of population increase (rm). In order of increasing resistance these were Sirtaki, Rondena, Fame, and Bianca. To correct for differences in damage inflicted on the cultivars, we developed a method to compare the attractiveness of the blends, based on the assumption that a larger amount of spider mite damage leads to higher attraction of P persimilis. Spider mite-induced volatiles of cultivars Rondena and Bianca were preferred over those of cultivar Sirtaki. Spider mite-induced volatiles of cultivars Sirtaki and Fame did not differ in attractiveness to P. persimilis. Sirtaki plants had a lower relative production of terpenes than the other three cultivars. This was attributed to a low production of cis-alpha-bergamotene, trans-alpha-bergamotene, trans-beta-bergamotene, and (E)-beta-farnesene. The emission of (E)-beta-ocimene and linalool was lower in Sirtaki and Fame leaves than in Bianca and Rondena. The importance of these chemical differences in the differential attraction of predatory mites is discussed.

Leaf pubescence and two-spotted spider mite webbing influence phytoseiid behavior and population density.

Phytoseiid mites, both in agricultural and natural systems, can play an important role in the regulation of herbivorous mites. Host plant traits, such as leaf pubescence, may influence the dynamics between predator and prey. In this study, we examined the influence of leaf surface characteristics (leaf pubescence and two-spotted spider mite webbing) on the behavior of two species of predatory mites, the generalist Typhlodromus pyri and the spider mite specialist Phytoseiulus persimilis. In laboratory trials, T. pyri females consistently spent more time and deposited more eggs on leaf discs from trichome-rich apple varieties compared to relatively trichome-poor varieties. A similar result was found when the choice involved trichome-rich and trichome-poor apple varieties planted into the same pot where leaves were allowed to touch so that the mites could freely move from leaf to leaf. To further explore the effect of structure created by pubescence and to remove possible confounding effects of chemical cues, we added cotton fibers to trichome-free bean leaves. T. pyri females consistently spent more time and deposited more eggs on the side of a glabrous bean leaf with artificial cotton fibers versus the side without added fibers. When given a choice between two densities of cotton fibers, T. pyri females consistently selected the highest density of available fibers in which to to reside and oviposit. T. pyri also preferred cotton fiber configurations in which it could move underneath and access the plant surface. The artificial pubescent leaf was also used to test the effect of leaf hairs and two-spotted spider mite webbing on the behavior of P. persimilis. P. persimilis females preferred residing and ovipositing on surfaces with cotton fibers or two-spotted spider mite webbing than on bean leaf areas without these structures. When presented a choice between cotton fibers or webbing, the behavior of P. persimilis females depended on the cotton fiber density. In a mixed-variety apple orchard, we investigated the relationship between leaf pubescence and phytoseiid density under field conditions. We found a highly significant, positive relationship between density of trichomes on leaves and abundance of T. pyri, whereas spider mite prey numbers were uniformly low and unrelated to trichome density. These field results suggest that the behavioral responses found in our laboratory experiments have population consequences.

Chemical ecology of host-plant selection by herbivorous arthropods: a multitrophic perspective.

Most herbivorous arthropods are specialists that feed on one or a few related plant species. To understand why this is so, both mechanistic and functional studies have been carried out, predominantly restricted to bitrophic aspects. Host-selection behaviour of herbivorous arthropods has been intensively studied and this has provided ample evidence for the role of secondary plant chemicals as source of information in behavioural decisions of herbivores. Many evolutionary studies have regarded co-evolution between plants and herbivores to explain the diversity of secondary plant chemicals and host specialisation of herbivores. However, many cases remain unexplained where herbivores select host plants that are suboptimal in terms of fitness returns. A stimulating paper by Bernays and Graham [(1988) Ecology 69, 886-892)] has initiated a discussion on the need of a multitrophic perspective to understand the evolution of host-plant specialisation by herbivorous arthropods. However, this has hardly resulted in ecological studies on host-selection behaviour that take a multitrophic perspective. Yet, evidence is accumulating that constitutive and induced infochemicals from natural enemies and competitors can affect herbivore behaviour. These cues may constitute important information on fitness prospects, just as plant cues can do. In this paper I selectively review how information from organisms at different trophic levels varies in space and time and how herbivores can integratively exploit this information during host selection. In doing so, research areas are identified that are likely to provide important new insights to explain several of the questions in herbivore host selection that remain unanswered so far. These research areas are at the interface of evolutionary ecology, behavioural ecology and chemical ecology.

Induced response of tomato plants to injury by green and red strains of Tetranychus urticae.

We studied the induced response of tomato plants to the green strain and the red strain of the spider mite Tetranychus urticae. We focused on the olfactory response of the predatory mite Phytoseiulus persimilis to volatiles from T. urticae-infested tomato leaves in a Y-tube olfactometer. Tomato leaves attracted the predatory mites when slightly infested with the red strain, or moderately or heavily infested with the green strain. In contrast, neither leaves that were slightly infested with green-strain mites, nor leaves that were moderately or heavily infested with the red strain attracted the predators. We discuss the specific defensive responses of tomato plants to each of the two strains.

Trichomes and spider-mite webbing protect predatory mite eggs from intraguild predation.

Predaceous arthropods are frequently more abundant on plants with leaves that are pubescent or bear domatia than on plants with glabrous leaves. We explored the hypothesis that for some predatory mites this is because pubescence affords protection from intraguild predation. In laboratory experiments, we tested whether apple leaf pubescence protected Typhlodromus pyri eggs from predation by western flower thrips, Frankliniella occidentalis. To investigate the effect of pubescence further, we added cotton fibers to trichome-free leaves. We also determined whether webbing produced by Tetranychus urticae protected Phytoseiulus persimilis eggs from predation by F. occidentalis. Predation by thrips on T. pyri eggs oviposited on field-collected pubescent "Erwin Bauer" apple leaves was significantly less than on glabrous "Crittenden" apple leaves. Phytoseiid eggs oviposited in the cotton fibers were preyed upon significantly less than those on the trichome-free bean disk. Increasing the cotton fiber density from 5 to 20 fibers only slightly further reduced predation by thrips on T. pyri eggs. Thrips fed upon significantly fewer P. persimilis eggs oviposited in Te. urticae webbing than eggs oviposited on a surface that differed only in the absence of Te. urticae web. We conclude that a complex leaf topography reduces intensity of intraguild predation in this system.

Specificity of herbivore-induced plant defences.

In addition to induced direct defence, plants can defend themselves indirectly by improving the effectiveness of enemies of herbivores. Plants can respond to arthropod herbivory with the induction of a blend of volatiles that attracts predators and/or parasitoids of herbivores. Carnivorous arthropods can discriminate between infested plants and mechanically wounded plants, and between plants infested by different herbivore species. The volatile blends emitted by different plant species infested by the same herbivore species show large qualitative differences, whereas blends emitted by plants of the same species, but infested by different herbivore species are mostly qualitatively similar with quantitative variation. Carnivores can discriminate between blends that differ qualitatively and/or quantitatively. However, it remains unknown what differences in blends are used by carnivorous arthropods in this discrimination. Signal transduction pathways involved in the induction of direct and indirect defence seem to overlap. Direct and indirect defence may interfere with each other's effectiveness. For application of direct and indirect defence in agriculture, it is important to compare the relative importance of these two defence types in the same plant species.