Increased chemical resistance explains low herbivore colonization of introduced seaweed

The success of introduced species is often attributed to release from co-evolved enemies in the new range and a subsequent decreased allocation to defense (EICA), but these hypotheses have rarely been evaluated for systems with low host-specificity of enemies. Here, we compare herbivore utilization of the brown seaweed, Fucus evanescens, and its coexisting competitors both in its native and new ranges, to test certain predictions derived from these hypotheses in a system dominated by generalist herbivores. While F. evanescens was shown to be a preferred host in its native range, invading populations supported a less diverse herbivore fauna and it was less preferred in laboratory choice experiments with important herbivores, when compared to co-occurring seaweeds. These results are consistent with the enemy release hypothesis, despite the fact that the herbivore communities in both regions were mainly composed of generalist species. However, in contrast to the prediction of EICA, analysis of anti-grazing compounds indicated a higher allocation to defense in introduced compared to native F. evanescens. The results suggest that the invader is subjected to less intense enemy control in the new range, but that this is due to an increased allocation to defense rather than release from specialized herbivores. This indicates that increased resistance to herbivory might be an important strategy for invasion success in systems dominated by generalist herbivores.     

Increased resistance to generalist herbivores in invasive populations of the California poppy (Eschscholzia californica)

Escape from enemies in the native range is often assumed to contribute to the successful invasion of exotic species. Following optimal defence theory, which assumes a trade‐off between herbivore resistance and plant growth, some have predicted that the success of invasive species could be the result of the evolution of lower resistance to herbivores and increased allocation of resources to growth and reproduction. Lack of evidence for ubiquitous costs of producing plant toxins, and the recognition that invasive species may escape specialist, but not generalist enemies, has led to a new prediction: invasive species may escape ecological trade‐offs associated with specialist herbivores, and evolve increased, rather than decreased, production of defensive compounds that are effective at deterring generalist herbivores in the introduced range. We tested the performance of two generalist lepidopteran herbivores, Trichoplusia ni and Orgyia vetusta, when raised on diets of native and invasive populations of the California poppy, Eschscholzia californica. Pupae of T. ni were significantly larger when reared on native populations. Similarly, caterpillars of O. vetusta performed significantly better when raised on native populations, indicating that invasive populations of the California poppy are more resistant to herbivores than native populations. The chance of successful establishment of some non‐indigenous plant species may be increased by retaining resistance to generalist herbivores, and in some cases, invasive species may be able to escape ecological trade‐offs in their new range and evolve, as we observed, even greater resistance to generalist herbivores than native plants.     

A meta-analysis of genetic correlations between plant resistances to multiple enemies

Genetic correlations between plant resistances to multiple natural enemies are important because they have the potential to determine the mode of selection that natural enemies impose on a host plant, the structure of herbivore and pathogen communities, and the success of plant breeding for resistance to multiple diseases and pests. We conducted a meta-analysis of 29 published studies of 16 different plant species reporting a total of 467 genetic correlations between resistances to multiple herbivores or pathogens. In general, genetic associations between resistances to multiple natural enemies tended to be positive regardless of the breeding design, type of attacker, and type of host plant. Positive genetic correlations between resistances were stronger when both attackers were pathogens or generalist herbivores and when resistance to different enemies was tested independently, suggesting that generalists may be affected by the same plant resistance traits and that interactions among natural enemies are common. Although the mean associations between resistances were positive, indicating the prevalence of diffuse selection and generalized defenses against multiple enemies, the large variation in both the strength and the direction of the associations suggests a continuum between pairwise and diffuse selection.     

Induced responses in Ipomoea hederacea: simulated mammalian herbivory induces resistance and susceptibility to insect herbivores

Multispecies interactions between plants and natural enemies are ubiquitous, and often lead to diffuse interactions between plants and their herbivores. Non-specific induced responses, where responses induced by one species affect other species, are one potential mechanism generating diffuse interactions. Using 57 inbred lines of the Ivyleaf morning glory, Ipomoea hederacea, in a greenhouse experiment, we examined whether simulated mammalian herbivory induced responses that could affect plant resistance to the generalist insect herbivore, Spodoptera exigua. Inbred lines were highly variable for induced responses, ranging from induced resistance to induced susceptibility, with the rank-order for resistance in inbred lines changing between clipping and control treatments. We failed to detect significant genetic correlations between induced responses and trichome density, or that clipping modified the negative relationship between trichome density and Spodoptera exigua consumption and biomass. Our results suggest that non-specific induced responses can mediate the diffuse evolutionary relationship between I. hederacea and its herbivores, and that genetic variation in induced responses are an important component of this interaction. Handling Error: Heikki Hokkanen     

Responses of insect herbivores and herbivory to habitat fragmentation: a hierarchical meta‐analysis

Loss and fragmentation of natural habitats can lead to alterations of plant–animal interactions and ecosystems functioning. Insect herbivory, an important antagonistic interaction is expected to be influenced by habitat fragmentation through direct negative effects on herbivore community richness and indirect positive effects due to losses of natural enemies. Plant community changes with habitat fragmentation added to the indirect effects but with little predictable impact. Here, we evaluated habitat fragmentation effects on both herbivory and herbivore diversity, using novel hierarchical meta‐analyses. Across 89 studies, we found a negative effect of habitat fragmentation on abundance and species richness of herbivores, but only a non‐significant trend on herbivory. Reduced area and increased isolation of remaining fragments yielded the strongest effect on abundance and species richness, while specialist herbivores were the most vulnerable to habitat fragmentation. These fragmentation effects were more pronounced in studies with large spatial extent. The strong reduction in herbivore diversity, but not herbivory, indicates how important common generalist species can be in maintaining herbivory as a major ecosystem process.     

Effects of invasive seaweeds on feeding preference and performance of a keystone Mediterranean herbivore

The consequences of invasive species on ecosystem processes and ecological interactions remain poorly understood. Predator–prey interactions are fundamental in shaping species evolution and community structure and can be strongly modified by species introductions. To fully understand the ecological effects of invasive species on trophic linkages it is important to characterize novel interactions between native predators and exotic prey and to identify the impacts of invasive species on the performance of native predators. Although seaweed invasions are a growing global concern, our understanding of invasive algae—herbivore interactions is still very limited. We used a series of feeding experiments between a native herbivore and four invasive algae in the Mediterranean Sea to examine the potential of native sea urchins to consume invasive seaweeds and the impacts of invasive seaweed on herbivore performance. We found that three of the four invasive species examined are avoided by native herbivores, and that feeding behaviour in sea urchins is not driven by plant nutritional quality. On the other hand, Caulerpa racemosa is readily consumed by sea urchins, but may escape enemy control by reducing their performance. Recognizing the negative impacts of C. racemosa on herbivore performance has highlighted an enemy escape mechanism that contributes to explaining how this widespread invasive alga, which is preferred and consumed by herbivores, is not eradicated by grazing in the field. Furthermore, given the ecological and economic importance of sea urchins, negative impacts of invasive seaweeds on their performance could have dramatic effects on ecosystem function and services, and should be accounted for in sea urchin population management strategies.     

Copper Contamination Impairs Herbivore Initiation of Seaweed Inducible Defenses and Decreases Their Effectiveness

Seaweed-herbivore interactions are often mediated by environmental conditions, yet the roles of emerging anthropogenic stressors on these interactions are poorly understood. For example, chemical contaminants have unknown consequences on seaweed inducible resistance and herbivore response to these defenses despite known deleterious effects of contaminants on animal inducible defenses. Here, we investigated the effect of copper contamination on the interactions between a snail herbivore and a brown seaweed that displays inducible resistance to grazing. We examined seaweed inducible resistance and its effectiveness for organisms exposed to copper at two time points, either during induction or after herbivores had already induced seaweed defenses. Under ambient conditions, non-grazed tissues were more palatable than grazed tissues. However, copper additions negated the preference for non-grazed tissues regardless of the timing of copper exposure, suggesting that copper decreased both how herbivores initiated these inducible defenses and their subsequent effectiveness. Copper decreased stimulation of defenses, at least in part, by suppressing snail grazing pressure—the cue that turns inducible defenses on. Copper decreased effectiveness of defenses by preventing snails from preferentially consuming non-grazed seaweed. Thus, contaminants can potentially stress communities by changing seaweed-herbivore interactions mediated via inducible defenses. Given the ubiquity of seaweed inducible resistance and their potential influence on herbivores, we hypothesize that copper contamination may change the impact of these resistant traits on herbivores.     

Loss of specificity: native but not invasive populations of Triadica sebifera vary in tolerance to different herbivores

During introduction, invasive plants can be released from specialist herbivores, but may retain generalist herbivores and encounter novel enemies. For fast-growing invasive plants, tolerance of herbivory via compensatory regrowth may be an important defense against generalist herbivory, but it is unclear whether tolerance responses are specifically induced by different herbivores and whether specificity differs among native and invasive plant populations. We conducted a greenhouse experiment to examine the variation among native and invasive populations of Chinese tallow tree, Triadica sebifera, in their specificity of tolerance responses to herbivores by exposing plants to herbivory from either one of two generalist caterpillars occurring in the introduced range of Triadica. Simultaneously, we measured the specificity of another defensive trait, extrafloral nectar (EFN) production, to detect potential tradeoffs between resistance and tolerance of herbivores. Invasive populations had higher aboveground biomass tolerance than native populations, and responded non-specifically to either herbivore, while native populations had significantly different and specific aboveground biomass responses to the two herbivores. Both caterpillar species similarly induced EFN in native and invasive populations. Plant tolerance and EFN were positively correlated or had no relationship and biomass in control and herbivore-damaged plants was positively correlated, suggesting little costs of tolerance. Relationships among these vegetative traits depended on herbivore type, suggesting that some defense traits may have positive associations with growth-related processes that are differently induced by herbivores. Importantly, loss of specificity in invasive populations indicates subtle evolutionary changes in defenses in invasive plants that may relate to and enhance their invasive success.     

Native plant/herbivore interactions as determinants of the ecological and evolutionary effects of invasive mammalian herbivores: the case of the common brushtail possum

In their native range, mammalian herbivores exist in a suite of direct and indirect ecological and evolutionary relationships with plant populations and communities. Outside their native range these herbivores become embedded in a multitude of new ecological and evolutionary interactions with native plant species in the new range. Sound knowledge of the plant/herbivore interactions in the herbivores’ native range provides an ideal framework to better understand their effects in the introduced range. The example of the common brushtail possum (Trichosurus vulpecula) and its introduction to New Zealand from Australia provides an excellent case study. In Australia, the common brushtail possum is a widespread generalist herbivore and it is thought that this generalist lifestyle has equipped the species well for successful colonisation of New Zealand. In Australia the brushtail possum has co-existed with highly chemically defended foliage since the Oligocene and recent papers have supported the role of possums as agents of selection on eucalypt defences. While the chemical profile of New Zealand foliage is comparatively poorly understood, possums do show clear selectivity between and within populations and some of these interactions may be mediated by the animals ability to ‘cope’ with PSMs, coupled with maintaining its generalist diet. While possums have had less time to effect evolutionary change in New Zealand species, their impacts on plant fitness have been well documented. However, further knowledge on variation and heritability of foliage traits driving possum preferences is needed to elucidate the ecological and evolutionary plant/possum interactions in the invasive range.     

Facing herbivory as you grow up: the ontogeny of resistance in plants

As plants develop from seeds to seedlings, juveniles and mature stages, their ontogeny can constrain the expression of resistance to herbivore damage. Nevertheless, ecological and evolutionary theories regarding interactions between plants, herbivores and their natural enemies are largely based on observations and experiments conducted at a single ontogenetic stage. Owing to resource allocation and architectural constraints in plants, and the influence of herbivore foraging behavior, resistance to herbivores is likely to change during plant development. We propose that such changes are likely to occur in a non-linear fashion and suggest that the role of ontogeny should be incorporated as an important factor in new syntheses of plant defense theory.     

Differentiation and adaptation in <Emphasis Type="Italic">Brassica nigra</Emphasis> populations: interactions with related herbivores

Local adaptation and population differentiation of plants are well documented, but studies on interactions with natural enemies are rare. In particular, evidence for plant adaptation to the local biotic environment, such as herbivores remains poor. We used the black mustard Brassica nigra, an annual species of river valley and coastal habitats to (1) analyse population differentiation in plant traits and herbivory in a common garden experiment, (2) examine home versus away differences in a reciprocal transplant experiment and (3) test whether plants are adapted to local herbivores or vice versa under standard greenhouse conditions. In the common garden experiment, we found significant differentiation in plant traits, leaf damage and herbivore number among seven populations of B. nigra from France and Germany (distance 15–1,000 km). Differences were particularly strong among coastal and river valley populations and did not necessarily increase with geographical distance. A herbivore removal treatment did not change population differentiation when compared with the control allowing natural colonisation. The reciprocal transplant experiment at a scale of 15–30 km did not reveal local plant adaptation, whilst one dominant herbivore species (Meligethes aeneus) occurred in significantly higher numbers on local plants. A greenhouse experiment combining three aphid (Brevicoryne brassicae) and plant populations of the same provenance indicated herbivore adaptation to their local plants rather than plant adaptation, but overall contrasts between local and non-local combinations were not significant. The results suggest that herbivores may counteract local plant adaptation to other environmental factors. Our study has important implications for plant translocations in ecological restoration projects.     

Second and third trophic level effects of differences in plant species reflect dietary specialisation of herbivores and their endoparasitoids

In natural populations, plants demonstrate an array of indirect and direct defence strategies that help to protect them from their herbivores and pathogens. Indirect defences include the release of odours that attract the natural enemies of herbivores, whereas direct defences may include the production of secondary compounds, allelochemicals that impair herbivore development or repel herbivore attack. Although both strategies have been well studied independently, comparatively little attention has been paid to examining the conflict that may arise between indirect and direct defences, such as when the performance of ‘recruited’ parasitoids or predators is negatively affected by plant allelochemicals. Here, we examine the growth and development of polyphagous and oligophagous folivores and their respective endoparasitoids on three crucifer species. One of the species, Brassica oleracea, was recently cultivated, whereas populations of B. nigra and Barbarea vulgaris occur naturally. Additionally, these species possess contrasting life‐history patterns and are also known to exhibit differences in secondary chemistry. The development of the generalist herbivore–parasitoid system was much more variable over the three crucifers than that exhibited by the specialists. Moreover, generalist herbivore and/or parasitoid fitness‐related traits (survival, development time, pupal, or adult size) were much more negatively affected on the wild crucifers than in the specialist association. Our results suggest that the relative importance of direct and indirect defences in plants may rest on the degree of dietary specialisation exhibited by herbivores and their natural enemies, and on the level of toxicity in the plant species under investigation.     

Leaf traits and herbivore selection in the field and in cafeteria experiments

Abstract Despite the vast diversity and complexity of herbivores, plants and their interactions, most authors agree that a small number of components of leaf quality affect preference by generalist herbivores in a predictable way. However, herbivore preference is determined not only by intrinsic plant attributes and herbivore biology but also by the environmental context. Within this framework, we aimed to analyse general interspecific trends in the association between herbivory and leaf traits over a wide range of angiosperms from central Argentina. We (i) tested for consistent associations between leaf traits, consumption in the field, and preference of generalist invertebrate herbivores in cafeteria experiments; (ii) assessed how well herbivore preferences in cafeterias matched leaf consumption in the field; and (iii) developed a simple conceptual model linking leaf traits, herbivore preference in cafeterias and consumption in the field. In general, we found that tender leaves with higher nutritional quality were preferred by herbivores, both in the field and in cafeteria experiments. According to our model, this relationship between field and cafeteria consumption and leaf quality is observed when generalist herbivores and plants of high accessibility are considered. However, differences between leaf consumption in the field and in cafeteria experiments can also be found. At least two reasons can account for this: (i) specialized plant–herbivore relationships often occur in the field, whereas cafeteria experiments tend to consider only one or a few generalist herbivores; (ii) different plant species growing in the field often differ in their degree of accessibility to herbivores, whereas in cafeteria experiments all species are equally accessible. Our results add new evidence to a growing consensus that, although herbivory in the field is determined by many factors, consistent patterns of differential susceptibility to foliar feeders can be found in leaves differing in nutritional quality and thus in resource‐use strategy.     

Predicting the effects of climate change on natural enemies of agricultural pests

Climate change can have diverse effects on natural enemies of pest species. Here we review these effects and their likely impacts on pest control. The fitness of natural enemies can be altered in response to changes in herbivore quality and size induced by temperature and CO₂ effects on plants. The susceptibility of herbivores to predation and parasitism could be decreased through the production of additional plant foliage or altered timing of herbivore life cycles in response to plant phenological changes. The effectiveness of natural enemies in controlling pests will decrease if pest distributions shift into regions outside the distribution of their natural enemies, although a new community of enemies might then provide some level of control. As well as being affected by climate through host plants and associated herbivores, the abundance and activity of natural enemies will be altered through adaptive management strategies adopted by farmers to cope with climate change. These strategies may lead to a mismatch between pests and enemies in space and time, decreasing their effectiveness for biocontrol. Because of the diverse and often indirect effects of climate change on natural enemies, predictions will be difficult unless there is a good understanding of the way environmental effects impact on tritrophic interactions. In addition, evolutionary changes in both hosts and natural enemies might have unexpected consequences on levels of biocontrol exerted by enemies. We consider interactions between the pest light brown apple moth and its natural enemies to illustrate the type of data that needs to be collected to make useful predictions.     

When do herbivores affect plant invasion? Evidence for the natural enemies and biotic resistance hypotheses

Two venerable hypotheses, widely cited as explanations for either the success or failure of introduced species in recipient communities, are the natural enemies hypothesis and the biotic resistance hypothesis. The natural enemies hypothesis posits that introduced organisms spread rapidly because they are liberated from their co‐evolved predators, pathogens and herbivores. The biotic resistance hypothesis asserts that introduced species often fail to invade communities because strong biotic interactions with native species hinder their establishment and spread. We reviewed the evidence for both of these hypotheses as they relate to the importance of non‐domesticated herbivores in affecting the success or failure of plant invasion.
To evaluate the natural enemies hypothesis, one must determine how commonly native herbivores have population‐level impacts on native plants. If native herbivores seldom limit native plant abundance, then there is little reason to think that introduced plants benefit from escape from these enemies. Studies of native herbivore‐native plant interactions reveal that plant life‐history greatly mediates the strength with which specialist herbivores suppress plant abundance. Relatively short‐lived plants that rely on current seed production for regeneration are most vulnerable to herbivory that reduces seed production. As such, these plants may gain the greatest advantage from escaping their specialist enemies in recipient communities. In contrast, native plants that are long lived or that possess long‐lived seedbanks may not be kept “in check” by native herbivores. For these species, escape from native enemies may have little to do with their success as exotics; they are abundant both where they are native and introduced.
Evidence for native herbivores providing biotic resistance to invasion by exotics is conflicting. Our review reveals that: 1) introduced plants can attract a diverse assemblage of native herbivores and that 2) native herbivores can reduce introduced plant growth, seed set and survival. However, the generality of these impacts is unclear, and evidence that herbivory actually limits or reduces introduced plant spread is scarce. The degree to which native herbivores provide biotic resistance to either exotic plant establishment or spread may be greatly determined by their functional and numerical responses to exotic plants, which we know little about. Generalist herbivores, through their direct effects on seed dispersal and their indirect effects in altering the outcome of native–non‐native plant competitive interactions, may have more of a facilitative than negative effect on exotic plant abundance.     

Direct and ecological costs of resistance and tolerance in the stinging nettle

Plant resistance and tolerance to herbivores, parasites, pathogens, and abiotic factors may involve two types of costs. First, resistance and tolerance may be costly in terms of plant fitness. Second, resistance and tolerance to multiple enemies may involve ecological trade-offs. Our study species, the stinging nettle ( Urtica dioica L.) has significant variation among seed families in resistance and tolerance as well as costs of resistance and tolerance to the holoparasitic plant Cuscuta europaea L. Here we report on variation among seed families (i.e. genetic) in tolerance to nutrient limitation and in resistance to both mammalian herbivores (i.e. number of stinging trichomes) and an invertebrate herbivore (i.e. inverse of the performance of a generalist snail, Arianta arbustorum). Our results indicate direct fitness costs of snail resistance in terms of host reproduction whereas we did not detect fitness costs of mammalian resistance or tolerance to nutrient limitation. We further tested for ecological trade-offs among tolerance or resistance to the parasitic plant, herbivore resistance, and tolerance to nutrient limitation in the stinging nettle. Tolerance of nettles to nutrient limitation and resistance to mammalian herbivores tended to correlate negatively. However, there were no significant correlations among resistance and tolerance to the different natural enemies (i.e. parasitic plants, snails, and mammals). The results of this greenhouse study thus suggest that resistance and tolerance of nettles to diverse enemies are free to evolve independently of each other but not completely without direct costs in terms of plant fitness.     

Nutrient-limited food webs with up to three trophic levels: feasibility, stability, assembly rules, and effects of nutrient enrichment

Community structure is controlled, among multiple factors, by competition and predation. Using the R* rule and graphical analysis, we analyse here the feasibility, stability and assembly rules of resource-based food webs with up to three trophic levels. In particular, we show that (1) the stability of a food web with two plants and two generalist herbivores does not require that plants' resource exploitation abilities trade-off with resistance to the two herbivores, and (2) food webs with two plants and either one generalist herbivore and a carnivore or two generalist herbivores and two generalist carnivores are not feasible because of cascade competition between top consumers. The relative strength of species interactions and the relative impacts of plants and herbivores on factors which control their growth also play a critical role. We discuss how community structure constrains assembly rules and yields cascades of extinctions in food webs.     

Seasonality of herbivory and communication between individuals of sagebrush

Seasonal changes in herbivore numbers and in plant defenses are well known to influence plant–herbivore interactions. Some plant defenses are induced in response to herbivore attack or cues correlated with risk of attack although seasonal variation in these defenses is relatively poorly known. We previously reported that sagebrush becomes more resistant to its herbivores when neighboring plants have been experimentally clipped with scissors. In this study we asked whether herbivory to leaves of sagebrush varied seasonally and whether there was seasonal variation in natural levels of damage when neighbors were clipped. We found that sagebrush accumulated most chewing damage early in the season, soon after the spring flush of new leaves. This damage was caused by generalist grasshoppers, deer, specialist caterpillars, beetles, gall makers, and other less common herbivores. Sagebrush showed no evidence of preferentially abscising leaves that had been experimentally clipped. Experimental clipping by Trirhabda pilosa beetle larvae caused neighbors to accumulate less herbivore damage later that season, similar to results in which clipping was done with scissors. Induced resistance caused by experimentally clipping a neighbor was affected by season; plants with neighbors clipped in May accumulated less damage throughout the season relative to plants with unclipped neighbors or neighbors clipped later in the summer. We found a correlation between seasonal herbivore pressure, damage accumulated by plants, and induced responses to experimentally clipping neighbors. The causal mechanisms responsible for this correlation are unknown although a strong seasonal effect was clear.     

Genetic architecture of susceptibility to herbivores in hybrid willows

We performed a common garden experiment using parental, F1, F2, and backcross willow hybrids to test the hypothesis that hybrid willows experience breakdown of resistance to herbivores. After exposing plants to herbivores in the field, we measured the densities/damage caused by 13 insect herbivores and one herbivorous mite. Using joint-scaling tests, we determined the contribution of additive, dominance, and epistasis to variation in susceptibility to herbivores (measured either as density or damage level) among the six genetic classes. We found the genetic architecture of susceptibility/resistance in the parental species to be complex, involving additive, dominance, and epistasis for each herbivore species. Although genic interactions altered plant susceptibility for each of the 14 herbivores, three distinct patterns of response of herbivores to hybrids were expressed. One pattern, observed in four herbivore species, supported the hypothesis of breakdown of resistance genes in recombinant hybrids. A second pattern, shown by six other herbivore species, supported the hypothesis of hybrid breakdown of host recognition genes. In other words, epistatic interactions for host recognition traits (probably oviposition/feeding stimulants or attractants) appeared to be important in determining herbivore abundance for those six species. The final patterns supported a structure of dominance, either for host recognition traits (in the case of three herbivore species) or for host resistance traits (for one herbivore species). The combination of differing responses of herbivore species, including members of the same genus and tribe, and the ubiquitous importance of epistasis suggests that many genes affect herbivore resistance in this hybrid willow system.     

Drought changes plant chemistry and causes contrasting responses in lepidopteran herbivores

Drought events are predicted to increase due to climate change, yet consequences for plant–insect interactions are only partially understood. Drought‐mediated interactions between herbivores and their host plants are affected by a combination of factors, including characteristics of the affected plant, its associated herbivore and of the prevailing drought. Studying the effect of these factors in combination may provide important insight into plant and herbivore responses to drought. We studied drought effects on plant resistance to two leaf‐chewing herbivores by considering differing growth conditions, plant chemistry and insect responses in concert. We exposed Alliaria petiolata plants from several wild populations to different intensities of intermittent drought stress and quantified drought‐mediated changes in plant chemistry. Simultaneously, we assessed behavior (feeding preference) and performance of two lepidopteran herbivores: Pieris brassicae, a specialist, and Spodoptera littoralis, a generalist. Drought led to lowest concentrations of secondary defense compounds in severely stressed plants, without affecting total nitrogen content. Additionally, drought evoked opposite patterns in feeding preferences (plant palatability) between the herbivore species. Pieris brassicae consumed most of well‐watered plants, while S. littoralis preferred severely drought‐stressed plants. Hence, feeding preferences of S. littoralis reflected changes in plant secondary chemistry. Contrary to their feeding preference, P. brassicae performed better on drought‐stressed than on well‐watered plants, with faster development and higher attained pupal mass (plant suitability). Spodoptera littoralis showed retarded development in all treatments. In conclusion, drought caused plant secondary defense compounds to decrease consistently across all studied plant populations, which evoked contrasting feeding preferences of two herbivore species of the same feeding guild. These results suggest herbivore specificity as a possible explanation for herbivore responses to drought and emphasize the importance of herbivore characteristics such as feeding specialization in understanding and predicting consequences of future drought events.