Tree diversity enhances predation by birds but not by arthropods across climate gradients.

Tree diversity can promote both predator abundance and diversity. However, whether this translates into increased predation and top-down control of herbivores across predator taxonomic groups and contrasting environmental conditions remains unresolved. We used a global network of tree diversity experiments (TreeDivNet) spread across three continents and three biomes to test the effects of tree species richness on predation across varying climatic conditions of temperature and precipitation. We recorded bird and arthropod predation attempts on plasticine caterpillars in monocultures and tree species mixtures. Both tree species richness and temperature increased predation by birds but not by arthropods. Furthermore, the effects of tree species richness on predation were consistent across the studied climatic gradient. Our findings provide evidence that tree diversity strengthens top-down control of insect herbivores by birds, underscoring the need to implement conservation strategies that safeguard tree diversity to sustain ecosystem services provided by natural enemies in forests.

Temporal variation in tree diversity effects on birds and its implications for top-down control of insect herbivores in a tropical system.

Tree diversity promotes predator abundance and diversity, but evidence linking these effects to increased predation pressure on herbivores remains limited. In addition, tree diversity effects on predators can vary temporally as a function of environmental variation, or due to contrasting responses by different predator types. In a multi-year study, we assessed temporal variation in tree diversity effects on bird community abundance, diversity, and predation rates as a whole and by functional group based on feeding guild (omnivores vs. insectivores) and migratory status (migrant vs. resident). To this end, we conducted bird point counts in tree monocultures and polycultures and assessed attacks on clay caterpillars four times over a 2-year period in a tree diversity experiment in Yucatan, Mexico. Tree diversity effects on the bird community varied across surveys, with positive effects on bird abundance and diversity in most but not all surveys. Tree diversity had stronger and more consistent effects on omnivorous and resident birds than on insectivorous and migratory species. Tree diversity effects on attack rates also varied temporally but patterns did not align with variation in bird abundance or diversity. Thus, while we found support for predicted increases in bird abundance, diversity, and predation pressure with tree diversity, these responses exhibited substantial variation over time and the former two were uncoupled from patterns of predation pressure, as well as contingent on bird functional traits. These results underscore the need for long-term studies measuring responses by different predator functional groups to better understand tree diversity effects on top-down control.

Test of Specificity in Signalling between Potato Plants in Response to Infection by Fusarium Solani and Phytophthora Infestans.

Plant-plant signalling via volatile organic compounds (VOCs) in response to insect herbivory has been widely studied, but its occurrence and specificity in response to pathogen attack has received much less attention. To fill this gap, we carried out a greenhouse experiment using two fungal pathogens (Fusarium solani and Phytophthora infestans) to test for specificity in VOC induction and signalling between potato plants (Solanum tuberosum). We paired potato plants in plastic cages, one acting as VOC emitter and the other as receiver, and subjected emitters to one of the following treatments: no infection (control), infected by F. solani, or infected by P. infestans. We measured total emission and composition of VOCs released by emitter plants to test for pathogen-specificity in VOC induction, and then conducted a pathogen infection bioassay to assess resistance levels on receiver plants by subjecting half of the receivers of each emitter treatment to F. solani infection and the other half to P. infestans infection. This allowed us to test for specificity in plant VOC signalling by comparing its effects on conspecific and heterospecific sequential infections. Results showed that infection by neither F. solani or P. infestans produced quantitative (total emissions) or qualitative (compositional) changes in VOC emissions. Mirroring these patterns, emitter infection treatment (control vs. pathogen infection) did not produce a significant change in pathogen infection levels on receiver plants in any case (i.e., either for conspecific or heterospecific sequential infections), indicating a lack of signalling effects which precluded pathogen-based specificity in signalling. We discuss possible mechanisms for lack of pathogen effects on VOC emissions and call for future work testing for pathogen specificity in plant-plant signalling and its implications for plant-pathogen interactions under ecologically relevant scenarios involving infections by multiple pathogens.

Effect of herbivore load on VOC-mediated plant communication in potato.

MAIN CONCLUSION: VOC emissions increased with herbivore load, but this did not result in concomitant increases in resistance in neighbouring plants, suggesting that communication occurred independently of herbivore load in emitter plants. Herbivore-damaged plants emit volatile organic compounds (VOCs) that can alert neighbours and boost their resistance. While VOC-mediated plant communication has been shown to be herbivore-specific, we know little about its contingency on variation in herbivore load. To address this knowledge gap, we tested herbivore load effects on VOC-mediated communication between potato plants (Solanum tuberosum) using the generalist herbivore Spodoptera exigua. First, we tested whether herbivore load (three levels: undamaged control, low, and high load) affected total VOC emissions and composition. Second, we matched emitter and receiver plants and subjected emitters to the same herbivore load treatments. Finally, we performed a bioassay with S. exigua on receivers to test for induced resistance due to VOC-mediated communication. We found that herbivory significantly increased total VOC emissions relative to control plants, and that such increase was greater under high herbivore load. In contrast, we found no detectable effect of herbivory, regardless of the load, on VOC composition. The communication experiment showed that VOCs released by herbivore-induced emitters boosted resistance in receivers (i.e., lower leaf damage than receivers exposed to VOCs released by control emitters), but the magnitude of such effect was similar for both levels of emitter herbivore load. These findings suggest that changes in VOCs due to variation in herbivore load do not modify the outcomes of plant communication.

Aphid and caterpillar feeding drive similar patterns of induced defences and resistance to subsequent herbivory in wild cotton.

MAIN CONCLUSION: Our results indicate caterpillars and aphids cause similar levels of induced defences and resistance against caterpillars in wild cotton plants. These symmetrical effects are not consistent with patterns predicted by plant defensive signaling crosstalk and call for further work addressing the biochemical mechanisms underpinning these results. Plant-induced responses to attack often mediate interactions between different species of insect herbivores. These effects are predicted to be contingent on the herbivore's feeding guild, whereby prior feeding by insects should negatively impact subsequent feeding by insects of the same guild (induced resistance) but may positively influence insects of a different guild (induced susceptibility) due to interfering crosstalk between plant biochemical pathways specific to each feeding guild. We compared the effects of prior feeding by leaf-chewing caterpillars (Spodoptera frugiperda) vs. sap-sucking aphids (Aphis gossypii) on induced defences in wild cotton (Gossypium hirsutum) and the consequences of these attacks on subsequently feeding caterpillars (S. frugiperda). To this end, we conducted a greenhouse experiment where cotton plants were either left undamaged or first exposed to caterpillar or aphid feeding, and we subsequently placed caterpillars on the plants to assess their performance. We also collected leaves to assess the induction of chemical defences in response to herbivory. We found that prior feeding by both aphids and caterpillars resulted in reductions in consumed leaf area, caterpillar mass gain, and caterpillar survival compared with control plants. Concomitantly, prior aphid and caterpillar herbivory caused similar increases in phenolic compounds (flavonoids and hydroxycinnamic acids) and defensive terpenoids (hemigossypolone) compared with control plants. Overall, these findings indicate that these insects confer a similar mode and level of induced resistance in wild cotton plants, calling for further work addressing the biochemical mechanisms underpinning these effects.

Soil salinization disrupts plant-plant signaling effects on extra-floral nectar induction in wild cotton.

Plant-plant interactions via volatile organic compounds (VOCs) have received much attention, but how abiotic stresses affect these interactions is poorly understood. We tested the effect of VOCs exposure from damaged conspecifics on the production of extra-floral nectar (EFN) in wild cotton plants (Gossypium hirsutum), a coastal species in northern Yucatan (Mexico), and whether soil salinization affected these responses. We placed plants in mesh cages, and within each cage assigned plants as emitters or receivers. We exposed emitters to either ambient or augmented soil salinity to simulate a salinity shock, and within each group subjected half of the emitters to no damage or artificial leaf damage with caterpillar regurgitant. Damage increased the emission of sesquiterpenes and aromatic compounds under ambient but not under augmented salinity. Correspondingly, exposure to VOCs from damaged emitters had effect on receiver EFN induction, but this effect was contingent on salinization. Receivers produced more EFN in response to damage after being exposed to VOCs from damaged emitters when the latter were grown under ambient salinity, but not when they were subjected to salinization. These results suggest complex effects of abiotic factors on VOC-mediated plant interactions.

Volatile-Mediated Signalling Between Potato Plants in Response to Insect Herbivory is not Contingent on Soil Nutrients.

Plant-plant signalling via volatile organic compounds (VOCs) has been studied intensively, but its contingency on abiotic conditions (e.g., soil nutrients, drought, warming) is poorly understood. To address this gap, we carried out a greenhouse experiment testing whether soil nutrients influenced signalling between potato (Solanum tuberosum) plants in response to insect leaf herbivory by the generalist caterpillar Spodoptera exigua. We placed pairs of plants in plastic cages, where one plant acted as a VOC emitter and the other as a receiver. We factorially manipulated soil nutrients for both emitter and receiver plants, namely: unfertilized (baseline soil nutrients) vs. fertilized (augmented nutrients). Then, to test for signalling effects, half of the emitters within each fertilization level were damaged by S. exigua larvae and the other half remained undamaged. Three days after placing larvae, we collected VOCs from emitter plants to test for herbivory and fertilization effects on VOC emissions and placed S. exigua larvae on receivers to test for signalling effects on leaf consumption and larval mass gain as proxies of induced resistance. We found that herbivory increased total VOC emissions and altered VOC composition by emitter plants, but these effects were not contingent on fertilization. In addition, bioassay results showed that receivers exposed to VOCs from herbivore-damaged emitters had lower levels of herbivory compared to receivers exposed to undamaged emitters. However, and consistent with VOC results, fertilization did not influence herbivore-induced signalling effects on receiver resistance to herbivory. In sum, we found evidence of S. exigua-induced signalling effects on resistance to herbivory in potato plants but such effects were not affected by increased soil nutrients. These results call for further work testing signalling effects under broader range of nutrient concentration levels (including nutrient limitation), teasing apart the effects of specific nutrients, and incorporating other abiotic factors likely to interact or covary with soil nutrients.

Variation in the Outcome of Plant-Mediated Pathogen Interactions in Potato: Effects of Initial Infections on Conspecific vs. Heterospecific Subsequent Infections.

Plants are often attacked sequentially by multiple enemies. Pathogen sequential co-infections can lead to indirect interactions mediated by plant induced responses whose outcome is contingent on differences in the magnitude and type of plant induced defences elicited by different species or guilds. To date, however, most studies have tested unidirectional effects of one pathogen on another, not discerning between conspecific vs. heterospecific infections, and often not measuring plant induced responses underlying such outcomes. To address this, we conducted a greenhouse experiment testing for the impact of initial infection by two leaf pathogens (Alternaria solani and Phytophthora infestans) on subsequent infection by each of these pathogens on potato (Solanum tuberosum) plants, and also measured induced plant defences (phenolic compounds) to inform on interaction outcomes. We found contrasting results depending on the identity of the initially infecting pathogen. Specifically, initial infection by A. solani drove induced resistance (lower necrosis) by subsequently infecting A. solani (conspecific induced resistance) but had no effect on subsequent infection by P. infestans. In contrast, initial infection by P. infestans drove induced resistance to subsequent infection by both conspecifics and A. solani. Patterns of plant induced defences correlated with (and potentially explained) induced resistance to conspecific but not heterospecific (e.g., in the case of P. infestans) subsequent infection. Overall, these results further our understanding of plant-mediated pathogen interactions by showing that plant-mediated interactions between pathogen species can be asymmetrical and in some cases not reciprocal, that pathogen species can vary in the importance of conspecific vs. heterospecific effects, and shed mechanistic insight into the role of plant induced responses driving such interactions.

Experimental Growth Conditions affect Direct and Indirect Defences in two Cotton Species.

Cotton has been used as a model plant to study direct and indirect plant defence against herbivorous insects. However, the plant growing conditions could have an important effect on the outcome of such plant defence studies. We examined how common experimental growth conditions influence constitutive and inducible defences in two species of cotton, Gossypium hirsutum and G. herbaceum. We induced plants by applying caterpillar regurgitant to mechanical wounds to compare the induction levels between plants of both species grown in greenhouse or phytotron conditions. For this we measured defence metabolites (gossypol and heliocides) and performance of Spodoptera frugiperda caterpillars on different leaves, the emission of plant volatiles, and their attractiveness to parasitic wasps. Induction increased the levels of defence metabolites, which in turn decreased the performance of S. frugiperda larvae. Constitutive and induced defence levels were the highest in plants grown in the phytotron (compared to greenhouse plants), G. hirsutum and young leaves. Defence induction was more pronounced in plants grown in the phytotron and in young leaves. Also, the differences between growing conditions were more evident for metabolites in the youngest leaves, indicating an interaction with plant ontogeny. The composition of emitted volatiles was different between plants from the two growth conditions, with greenhouse-grown plants showing more variation than phytotron-grown plants. Also, G. hirsutum released higher amounts of volatiles and attracted more parasitic wasps than G. herbaceum. Overall, these results highlight the importance of experimental abiotic factors in plant defence induction and ontogeny of defences. We therefore suggest careful consideration in selecting the appropriate experimental growing conditions for studies on plant defences.

Apparent inhibition of induced plant volatiles by a fungal pathogen prevents airborne communication between potato plants.

Plant communication in response to insect herbivory has been increasingly studied, whereas that involving pathogen attack has received much less attention. We tested for communication between potato (Solanum tuberosum) plants in response to leaf infection by the fungal pathogen Sclerotinia sclerotiorum. To this end, we measured the total amount and composition of volatile organic compounds (VOCs) produced by control and infected emitter plants, as well as tested for induced resistance of receiver plants exposed to VOCs from emitters. We further tested for changes in the expression of defensive genes due to pathogen infection. Fungal infection did not significantly affect the total amount or composition of VOCs produced by emitter plants. Correspondingly, we found no evidence of higher resistance to the pathogen in receiver plants exposed to VOCs from infected emitters relative to control emitters. Molecular analyses indicated that pathogen infection drove a down-regulation of genes coding for VOC precursors, potentially explaining the absence of pathogen effects on VOC emissions and thus of communication. Overall, these results indicate no evidence of airborne communication between potato plants in response to fungal infection and point at pathogen inhibition of VOC emissions as a likely explanation for this result.

Effects of latitude and conspecific plant density on insect leaf herbivory in oak saplings and seedlings.

PREMISE: Abiotic factors and plant species traits have been shown to drive latitudinal gradients in herbivory, and yet, population-level factors have been largely overlooked within this context. One such factor is plant density, which may influence the strength of herbivory and may vary with latitude. METHODS: We measured insect herbivory and conspecific plant density (CPD) of oak (Quercus robur) seedlings and saplings along a 17° latitudinal gradient (2700 km) to test whether herbivory exhibited a latitudinal gradient, whether herbivory was associated with CPD, and whether such an association changed with latitude. RESULTS: We found a positive but saturating association between latitude and leaf herbivory. Furthermore, we found no significant relationship between CPD and herbivory, and such lack of density effects remained consistent throughout the sampled latitudinal gradient. CONCLUSIONS: Despite the apparently negligible influence of plant density on herbivory for Q. robur, further research with other plant taxa and in different types of plant communities are needed to investigate density-dependent processes shaping geographical variation in plant-herbivore interactions.

Latitudinal variation in seed predation correlates with latitudinal variation in seed defensive and nutritional traits in a widespread oak species.

BACKGROUND AND AIMS: Classic theory on geographical gradients in plant-herbivore interactions assumes that herbivore pressure and plant defences increase towards warmer and more stable climates found at lower latitudes. However, the generality of these expectations has been recently called into question by conflicting empirical evidence. One possible explanation for this ambiguity is that most studies have reported on patterns of either herbivory or plant defences whereas few have measured both, thus preventing a full understanding of the implications of observed patterns for plant-herbivore interactions. In addition, studies have typically not measured climatic factors affecting plant-herbivore interactions, despite their expected influence on plant and herbivore traits. METHODS: Here we tested for latitudinal variation in insect seed predation and seed traits putatively associated with insect attack across 36 Quercus robur populations distributed along a 20° latitudinal gradient. We then further investigated the associations between climatic factors, seed traits and seed predation to test for climate-based mechanisms of latitudinal variation in seed predation. KEY RESULTS: We found strong but contrasting latitudinal clines in seed predation and seed traits, whereby seed predation increased whereas seed phenolics and phosphorus decreased towards lower latitudes. We also found a strong direct association between temperature and seed predation, with the latter increasing towards warmer climates. In addition, temperature was negatively associated with seed traits, with populations at warmer sites having lower levels of total phenolics and phosphorus. In turn, these negative associations between temperature and seed traits led to a positive indirect association between temperature and seed predation. CONCLUSIONS: These results help unravel how plant-herbivore interactions play out along latitudinal gradients and expose the role of climate in driving these outcomes through its dual effects on plant defences and herbivores. Accordingly, this emphasizes the need to account for abiotic variation while testing concurrently for latitudinal variation in plant traits and herbivore pressure.

Tree diversity effects through a temporal lens: Implications for the abundance, diversity and stability of foraging birds.

Tree diversity exerts a strong influence on consumer communities, but most work has involved single time point measurements over short time periods. Describing temporal variation associated with diversity effects over longer time periods is necessary to fully understand the effects of tree diversity on ecological function. We conducted a year-long study in an experimental system in southern Mexico assessing the effects of tree diversity on the abundance and diversity of foraging birds. To this end, we recorded bird visitation patterns in 32 tree plots (21 × 21 m; 12 tree species monocultures, 20 four-species polycultures) every 45 days (n = 8 surveys) and for each plot estimated bird abundance, richness, functional diversity (FD) and phylogenetic diversity (PD). In each case, we reported temporal (intra-annual) variation in the magnitude of tree diversity effects, and calculated the temporal stability of these bird responses. Across surveys, tree diversity noticeably affected bird responses, demonstrated by significantly higher abundance (43%), richness (32%), PD (25%) and FD (25%) of birds visiting polyculture plots compared to monoculture plots, as well as a distinct species composition between plot types. We also found intra-annual variation in tree diversity effects on these response variables, ranging from surveys for which the diversity effect was not significant to surveys where a significant 80% increase (e.g. for bird FD and PD) was observed in polyculture relative to monoculture plots. Notably, tree diversity increased the stability of all bird responses, with polycultures having a greater stability abundance (18%), richness (38%), PD (32%), and FD (35%) of birds visiting tree species polycultures compared to monocultures. These results show that tree diversity not only increases bird visitation to plots, but also stabilizes bird habitat usage over time in ways that could implicate insurance-related mechanisms. Such findings are highly relevant for understanding the long-term effects of plant diversity on vertebrates and the persistence of bird-related ecosystem functions. More work is needed to unveil the ecological mechanisms behind temporal variation in vertebrate responses to tree diversity and their consequences for community structure and function.

Tri-trophic interactions: bridging species, communities and ecosystems.

A vast body of research demonstrates that many ecological and evolutionary processes can only be understood from a tri-trophic viewpoint, that is, one that moves beyond the pairwise interactions of neighbouring trophic levels to consider the emergent features of interactions among multiple trophic levels. Despite its unifying potential, tri-trophic research has been fragmented, following two distinct paths. One has focused on the population biology and evolutionary ecology of simple food chains of interacting species. The other has focused on bottom-up and top-down controls over the distribution of biomass across trophic levels and other ecosystem-level variables. Here, we propose pathways to bridge these two long-standing perspectives. We argue that an expanded theory of tri-trophic interactions (TTIs) can unify our understanding of biological processes across scales and levels of organisation, ranging from species evolution and pairwise interactions to community structure and ecosystem function. To do so requires addressing how community structure and ecosystem function arise as emergent properties of component TTIs, and, in turn, how species traits and TTIs are shaped by the ecosystem processes and the abiotic environment in which they are embedded. We conclude that novel insights will come from applying tri-trophic theory systematically across all levels of biological organisation.

Parallel increases in insect herbivory and defenses with increasing elevation for both saplings and adult trees of oak (Quercus) species.

PREMISE: Herbivory is predicted to increase toward warmer and more stable climates found at lower elevations, and this increase should select for higher plant defenses. Still, a number of recent studies have reported either no evidence of such gradients or reverse patterns. One source of inconsistency may be that plant ontogenetic variation is usually not accounted for and may influence levels of plant defenses and herbivory. METHODS: We tested for elevational gradients in insect leaf herbivory and leaf traits putatively associated with herbivore resistance across eight oak (Quercus, Fagaceae) species and compared these patterns for saplings and adult trees. To this end, we surveyed insect leaf herbivory and leaf traits (phenolic compounds, toughness and nutrients) in naturally occurring populations of each oak species at low-, mid- or high-elevation sites throughout the Iberian Peninsula. RESULTS: Leaf herbivory and chemical defenses (lignins) were unexpectedly higher at mid- and high-elevation sites than at low-elevation sites. In addition, leaf chemical defenses (lignins and condensed tannins) were higher for saplings than adult trees, whereas herbivory did not significantly differ between ontogenetic stages. Overall, elevational variation in herbivory and plant chemical defenses were consistent across ontogenetic stages (i.e., elevational gradients were not contingent upon tree ontogeny), and herbivory and leaf traits were not associated across elevations. CONCLUSIONS: These findings suggest disassociated patterns of elevational variation in herbivory and leaf traits, which, in turn, are independent of plant ontogenetic stage.

Greater phylogenetic distance from native oaks predicts escape from insect leaf herbivores by non-native oak saplings.

PREMISE: Non-native plant species have been hypothesized to experience lower herbivory in novel environments as a function of their phylogenetic distance from native plant species. Although recent work has found support for this prediction, the plant traits responsible for such patterns have been largely overlooked. METHODS: In a common garden experiment in northwestern Spain, we tested whether oak species (Quercus spp.) not native to this region that are phylogenetically more distantly related to native species exhibit less insect leaf herbivory. In addition, we also investigated plant traits potentially correlated with any such effect of phylogenetic distance. RESULTS: As expected, phylogenetic distance from native species negatively predicted insect leaf herbivory on non-native oaks. In addition, we found that the leaf traits, namely phosphorus and condensed tannins, were significantly associated with herbivory, suggesting that they are associated with the effect of phylogenetic distance on leaf herbivory on non-native oak species. CONCLUSIONS: This study contributes to a better understanding of how evolutionary relationships (relatedness) between native and non-native plant species determine the latter's success in novel environments via locally shared enemies, and encourages more work investigating the plant traits that mediate the effects of phylogenetic distance on enemy escape.

Weather cues associated with masting behavior dampen the negative autocorrelation between past and current reproduction in oaks.

PREMISE OF THE STUDY: The influence of weather conditions on masting and the ecological advantages of this reproductive behavior have been the subject of much interest. Weather conditions act as cues influencing reproduction of individual plants, and similar responses expressed across many individuals lead to population-level synchrony in reproductive output. In turn, synchrony leads to benefits from economies of scale such as enhanced pollination success and seed predator satiation. However, there may also be individual-level benefits from reproductive responses to weather cues, which may explain the origin of masting in the absence of economies of scale. In a previous study, we found support for a mechanism whereby individual responses to weather cues attenuate the negative autocorrelation between past and current annual seed production-a pattern typically attributed to resource limitation and reproductive tradeoffs among years. METHODS: Here we provide a follow-up and more robust evaluation of this hypothesis in 12 species of oaks (Quercus spp.), testing for a negative autocorrelation (tradeoff) between past and current reproduction and whether responses to weather cues associated with masting reduce the strength of this negative autocorrelation. KEY RESULTS: Our results showed a strong negative autocorrelation for 11 of the species, and that species-specific reproductive responses to weather cues dampened this negative autocorrelation in 10 of them. CONCLUSIONS: This dampening effect presumably reflects a reduction in resource limitation or increased resource use associated with weather conditions, and suggests that responses to weather cues conferring these advantages should be selected for based on individual benefits.

Bottom-up control of geographic variation in insect herbivory on wild cotton (Gossypium hirsutum) by plant defenses and climate.

PREMISE: The occurrence and amount of herbivory are shaped by bottom-up forces, primarily plant traits (e.g., defenses), and by abiotic factors. Addressing these concurrent effects in a spatial context has been useful in efforts to understand the mechanisms governing variation in plant-herbivore interactions. Still, few studies have evaluated the simultaneous influence of multiple sources of bottom-up variation on spatial variation in herbivory. METHODS: We tested to what extent chemical (phenolics, production of terpenoid glands) and physical (pubescence) defensive plant traits and climatic factors are associated with variation in herbivory by leaf-chewing insects across populations of wild cotton (Gossypium hirsutum). RESULTS: We found substantial population variation in cotton leaf defenses and insect leaf herbivory. Leaf pubescence, but not gossypol gland density or phenolic content, was significantly negatively associated with herbivory by leaf-chewing insects. In addition, there were direct effects of climate on defenses and herbivory, with leaf pubescence increasing toward drier conditions and leaf damage increasing toward wetter and cooler conditions. There was no evidence, however, of indirect effects (via plant defenses) of climate on herbivory. CONCLUSIONS: These results suggest that spatial variation in insect herbivory on wild G. hirsutum is predominantly driven by concurrent and independent influences of population variation in leaf pubescence and climatic factors.

Tropical tree diversity mediates foraging and predatory effects of insectivorous birds.

Biodiversity affects the structure of ecological communities, but little is known about the interactive effects of diversity across multiple trophic levels. We used a large-scale forest diversity experiment to investigate the effects of tropical tree species richness on insectivorous birds, and the subsequent indirect effect on predation rates by birds. Diverse plots (four tree species) had higher bird abundance (61%), phylogenetic diversity (61%), and functional diversity (55%) than predicted based on single-species monocultures, which corresponded to higher attack rates on artificial caterpillars (65%). Tree diversity effects on attack rate were driven by complementarity among tree species, with increases in attack rate observed on all tree species in polycultures. Attack rates on artificial caterpillars were higher in plots with higher bird abundance and diversity, but the indirect effect of tree species richness was mediated by bird diversity, providing evidence that diversity can interact across trophic levels with consequences tied to ecosystem services and function.