Optimizing the releasing strategy used for the biological control of the sugarcane borer Diatraea saccharalis by Trichogramma galloi with computer modeling and simulation.

One of the challenges in augmentative biological control programs is the definition of releasing strategy for natural enemies, especially when macro-organisms are involved. Important information about the density of insects to be released and frequency of releases usually requires a great number of experiments, which implies time and space that are not always readily available. In order to provide science-based responses for these questions, computational models offer an in silico option to simulate different biocontrol agent releasing scenarios. This allows decision-makers to focus their efforts to more feasible options. The major insect pest in sugarcane crops is the sugarcane borer Diatraea saccharalis, which can be managed using the egg parasitoid Trichogramma galloi. The current strategy consists in releasing 50,000 insects per hectare for each release, in three weekly releases. Here, we present a simulation model to check whether this releasing strategy is optimal against the sugarcane borer. A sensitive analysis revealed that the population of the pest is more affected by the number of releases rather than by the density of parasitoids released. Only the number of releases demonstrated an ability to drive the population curve of the pest towards a negative growth. For example, releasing a total of 600,000 insects per hectare in three releases led to a lower pest control efficacy that releasing only 250,000 insects per hectare in five releases. A higher number of releases covers a wider range of time, increasing the likelihood of releasing parasitoids at the correct time given that the egg stage is short. Based on these results, it is suggested that, if modifications to the releasing strategy are desired, increasing the number of releases from 3 to 5 at weekly intervals is most likely preferable.

Why do predators attack parasitized prey? Insights from a probabilistic model and a literature survey.

Predators and parasitoids often encounter parasitized prey or hosts during foraging. While the outcomes of such encounters have been extensively studied for insect parasitoids, the consequences of a predator encountering parasitized prey have received less attention. One extreme example involves the potter wasp Delta dimidiatipenne that frequently provision their nest with parasitized caterpillars, despite the low suitability of this prey for consumption by their offspring. This raises two main questions: (1) why do female potter wasps continue collecting parasitized caterpillars? and (2) is this an exceptional example, or do predatory insects often suffer from fitness costs due to encounters with parasitized prey? We addressed the first question using a probabilistic mathematical model predicting the value of discrimination between parasitized and unparasitized prey for the potter wasp, and the second question by surveying the literature for examples in which the parasitism status of prey affected prey susceptibility, suitability, or prey choice by a predator. The model demonstrates that only under certain conditions is discrimination against parasitized prey beneficial in terms of the potter wasp's lifetime reproductive success. The literature survey suggests that the occurrence of encounters and consumption of parasitized prey is common, but the overall consequences of such interactions have rarely been quantified. We conclude that the profitability and ability of a predator to discriminate against parasitized prey under natural conditions may be limited and call for additional studies quantifying the outcome of such interactions.

Trait-based approaches to predicting biological control success: challenges and prospects.

Identifying traits that are associated with success of introduced natural enemies in establishing and controlling pest insects has occupied researchers and biological control practitioners for decades. Unfortunately, consistent general relationships have been difficult to detect, preventing a priori ranking of candidate biological control agents based on their traits. We summarise previous efforts and propose a series of potential explanations for the lack of clear patterns. We argue that the quality of current datasets is insufficient to detect complex trait-efficacy relationships and suggest several measures by which current limitations may be overcome. We conclude that efforts to address this elusive issue have not yet been exhausted and that further explorations are likely to be worthwhile.

Dynamic Economic Thresholds for Insecticide Applications Against Agricultural Pests: Importance of Pest and Natural Enemy Migration.

In Integrated Pest Management programs, insecticides are applied to agricultural crops when pest densities exceed a predetermined economic threshold. Under conditions of high natural enemy density, however, the economic threshold can be increased, allowing for fewer insecticide applications. These adjustments, called 'dynamic thresholds', allow farmers to exploit existing biological control interactions without economic loss. Further, the ability of natural enemies to disperse from, and subsequently immigrate into, insecticide-sprayed areas can affect their biological control potential. We develop a theoretical approach to incorporate both pest and natural enemy movement across field borders into dynamic thresholds and explore how these affect insecticide applications and farmer incomes. Our model follows a pest and its specialist natural enemy over one growing season. An insecticide that targets the pest also induces mortality of the natural enemy, both via direct toxicity and reduced resource pest densities. Pest and natural enemy populations recover after spraying through within-field reproduction and by immigration from neighboring unsprayed areas. The number of insecticide applications and per-season farmer revenues are calculated for economic thresholds that are either fixed (ignoring natural enemy densities) or dynamic (incorporating them). The model predicts that using dynamic thresholds always leads to reduced insecticide application. The benefit of dynamic thresholds in reducing insecticide use is highest when natural enemies rapidly recolonize sprayed areas, and when insecticide efficacy is low. We discuss real-life situations in which monitoring of natural enemies would substantially reduce insecticide use and other scenarios where the presence of beneficial organisms may lead to threshold modifications.

Can sociality facilitate learning of complex tasks? Lessons from bees and flowers.

The emergence of animal societies is a major evolutionary transition, but its implications for learning-dependent innovations are insufficiently understood. Bees, with lifestyles ranging from solitary to eusocial, are ideal models for exploring social evolution. Here, we ask how and why bees may acquire a new 'technology', foraging on morphologically complex flowers, and whether eusociality facilitates this technological shift. We consider 'complex' flowers that produce high food rewards but are difficult to access, versus 'simple' flowers offering easily accessible yet lower rewards. Complex flowers are less profitable than simple flowers to naive bees but become more rewarding after a learning period. We model how social bees optimally choose between simple and complex flowers over time, to maximize their colony's food balance. The model predicts no effect of colony size on the bees' flower choices. More foraging on complex flowers is predicted as colony longevity, its proportion of foragers, individual longevity and learning ability increase. Of these traits, only long-lived colonies and abundant foragers characterize eusocial bees. Thus, we predict that eusociality supports, but is not mandatory for, learning to exploit complex flowers. A re-analysis of a large published dataset of bee-flower interactions supports these conclusions. We discuss parallels between the evolution of insect sociality and other major transitions that provide scaffolds for learning innovations. This article is part of the theme issue 'Human socio-cultural evolution in light of evolutionary transitions'.

The advantage of sex: Reinserting fluctuating selection in the pluralist approach.

The advantage of sex, and its fixation in some clades and species all over the eukaryote tree of life, is considered an evolutionary enigma, especially regarding its assumed two-fold cost. Several likely hypotheses have been proposed such as (1) a better response to the negative frequency-dependent selection imposed by the "Red Queen" hypothesis; (2) the competition between siblings induced by the Tangled Bank hypothesis; (3) the existence of genetic and of (4) ecological factors that can diminish the cost of sex to less than the standard assumed two-fold; and (5) a better maintenance of genetic diversity and its resulting phenotypic variation, providing a selective advantage in randomly fluctuating environments. While these hypotheses have mostly been studied separately, they can also act simultaneously. This was advocated by several studies which presented a pluralist point of view. Only three among the five causes cited above were considered yet in such a framework: the Red Queen hypothesis, the Tangled Bank and the genetic factors lowering the cost of sex. We thus simulated the evolution of a finite mutating population undergoing negative frequency-dependent selection on phenotypes and a two-fold (or less) cost of sexuality, experiencing randomly fluctuating selection along generations. The individuals inherited their reproductive modes, either clonal or sexual. We found that exclusive sexuality begins to fix in populations exposed to environmental variation that exceeds the width of one ecological niche (twice the standard deviation of a Gaussian response to environment). This threshold was lowered by increasing negative frequency-dependent selection and when reducing the two-fold cost of sex. It contributes advocating that the different processes involved in a short-term advantage of sex and recombination can act in combination to favor the fixation of sexual reproduction in populations.

The combined effect of host and food availability on optimized parasitoid life-history traits based on a three-dimensional trade-off surface.

The reproductive success of many insects is considered to be limited by two main factors: the availability of mature eggs to lay (termed egg limitation) and the time to locate suitable hosts (termed time limitation). High host density in the environment is likely to enhance oviposition opportunities, thereby selecting for higher investment in egg supply. In contrast, a shortage of food (e.g. sugar sources) is likely to increase the risk of time limitation, thereby selecting for higher allocation to initial energy reserves. To our knowledge, the combined effect of host and food availability on these optimal life-history allocations has never been investigated. We thus modelled their simultaneous effects on a three-dimensional trade-off between initial investment in energy reserves, egg number and egg size, while focusing on insect parasitoids. The model was based on Monte Carlo simulations coupled with genetic algorithms, in order to identify the optimal life-history traits of a single simulated parasitoid female in an environment in which both hosts and food are present in varying densities. Our results reproduced the simple predictions described above. However, some novel predictions were also obtained, especially when specific interactions between the different factors were examined and their effects on the three-dimensional life-history surface were considered. The work sheds light on long-lasting debates regarding the relative importance of time versus egg limitation in determining insect life-history traits and highlights the complexity of life-history evolution, where several environmental factors act simultaneously on multiple traits.

Pollinator Behavior Drives Sexual Specializations in the Hermaphrodite Flowers of a Heterodichogamous Tree.

Dioecy, the specialization of individuals into either male-only or female-only sexual function, has multiple evolutionary origins in plants. One proposed ancestral mating system is heterodichogamy, two morphs of cross-fertilizing hermaphrodite flowers that differ in their timing of flowering. Previous research suggested that small specializations in these morphs' functional genders could facilitate their evolution into separate sexes. We tested the possible role of pollinators in driving such specializations. Ziziphus spina-christi is an insect-pollinated heterodichogamous tree with self-incompatible flowers and two sympatric flowering morphs. We compared the flower development patterns, floral food rewards, pollinator visits, and fruit production between the two morphs. Male-phase flowers of Z. spina-christi's "Early" and "Late" morphs open before dawn and around noon, respectively, and transition into female-phase 7-8 h later. Flowers of both morphs contain similar nectar and pollen rewards, and receive visits by flies (their ancestral pollinators) at similar rates, mostly during the morning. Consequently, the Early morph functions largely as pollen donor. The Late morph, functioning as female in the morning, produces more fruit. We developed an evolutionary probabilistic model, inspired by Z. spina-christi's reproductive system, to test whether pollinator visit patterns could potentially play a role in an evolutionary transition from heterodichogamy towards dioecy. The model predicts that reproductive incompatibility within flowering morphs promotes their evolution into different sexes. Furthermore, the pollinators' morning activity drives the Early and Late morphs' specialization into male and female functions, respectively. Thus, while not required for transitioning from heterodichogamy to dioecy, pollinator-mediated selection is expected to influence which sexual specialization evolves in each of the flowering morphs.

Breeding success of a marine central place forager in the context of climate change: A modeling approach.

In response to climate warming, a southward shift in productive frontal systems serving as the main foraging sites for many top predator species is likely to occur in Subantarctic areas. Central place foragers, such as seabirds and pinnipeds, are thus likely to cope with an increase in the distance between foraging locations and their land-based breeding colonies. Understanding how central place foragers should modify their foraging behavior in response to changes in prey accessibility appears crucial. A spatially explicit individual-based simulation model (Marine Central Place Forager Simulator (MarCPFS)), including bio-energetic components, was built to evaluate effects of possible changes in prey resources accessibility on individual performances and breeding success. The study was calibrated on a particular example: the Antarctic fur seal (Arctocephalus gazella), which alternates between oceanic areas in which females feed and the land-based colony in which they suckle their young over a 120 days rearing period. Our model shows the importance of the distance covered to feed and prey aggregation which appeared to be key factors to which animals are highly sensitive. Memorization and learning abilities also appear to be essential breeding success traits. Females were found to be most successful for intermediate levels of prey aggregation and short distance to the resource, resulting in optimal female body length. Increased distance to resources due to climate warming should hinder pups' growth and survival while female body length should increase.

Does Plant Cultivar Difference Modify the Bottom-Up Effects of Resource Limitation on Plant-Insect Herbivore Interactions?

Variation in resource input to plants triggers bottom-up effects on plant-insect herbivore interactions. However, variation in plant intrinsic traits in response to resource availability may modify the bottom-up effects. Furthermore, the consequences also may depend on the feeding strategy of insect herbivores belonging to different feeding guilds. We evaluated the performance of two insect herbivores from distinct feeding guilds, the leaf miner Tuta absoluta and the phloem feeder Bemisia tabaci. We offered the insects two tomato cultivars growing under optimal nitrogen input vs. nitrogen limitation, or under optimal water input vs. water limitation. We found that: (i) the two cultivars differed in their responses to nitrogen and water limitation by regulating primary (leaf-gas exchange related parameters, leaf nitrogen content, and leaf C/N ratio) and secondary metabolism (main defensive compounds: glycoalkaloids); (ii) for both plant cultivars, nitrogen or water limitation significantly affected T. absoluta survival and development, while B. tabaci survival was affected only by nitrogen limitation; and surprisingly (iii) plant cultivar differences did not modify the negative bottom-up effects of resource limitation on the two insect herbivores. In conclusion, the negative effects of resource limitation cascaded up to insect herbivores even though plant cultivars exhibited various adaptive traits to resource limitation.

Side effects of Bacillus thuringiensis var. kurstaki on the hymenopterous parasitic wasp Trichogramma chilonis.

Most of the detrimental effects of using conventional insecticides to control crop pests are now well identified and are nowadays major arguments for replacing such compounds by the use of biological control agents. In this respect, the bacterium Bacillus thuringiensis var. kurstaki and Trichogramma (Hymenoptera: Trichogrammatidae) parasitic wasp species are both effective against lepidopterous pests and can actually be used concomitantly. In this work, we studied the potential side effects of B. thuringiensis var. kurstaki on Trichogramma chilonis females. We first evidenced an acute toxicity of B. thuringiensis on T. chilonis. Then, after ingestion of B. thuringiensis at sublethal doses, we focused on life history traits of T. chilonis such as longevity, reproductive success and the time spent on host eggs patches. The reproductive success of T. chilonis was not modified by B. thuringiensis while a significant effect was observed on longevity and the time spent on host eggs patches. The physiological and ecological meanings of the results obtained are discussed.

The effect of direct interspecific competition on patch exploitation strategies in parasitoid wasps.

The presence of competitors may affect the pay-off of individuals' foraging strategies. They should therefore modify their resource exploitation decisions accordingly. In such a direct competition situation, theory predicts that individuals should stay longer on a resource patch than when foraging alone. However, models predicting patch residence time focus on intraspecific competition without agonistic interactions. Here, we investigate the patch use strategies of females of two parasitoid species, Eupelmus vuilleti and Dinarmus basalis, attacking the same host, Callosobruchus maculatus, knowing that D. basalis is more aggressive and can exclude E. vuilleti during pairwise contests for single hosts. Our results showed that time allocation and oviposition strategies differed in relation to the species and type of competition (i.e., presence/absence of competitor, simultaneous/sequential female introduction or resident/intruder female). Eupelmus vuilleti females tended to wait in the patch surroundings for D. basalis females' departure to return and exploit hosts parasitized by the opponent (after destruction of her eggs). In contrast, D. basalis females tended to self-superparasitize and stay motionless near the hosts. After detecting an E. vuilleti female entering the patch, they attacked and chased her permanently from the patch. Females of both species spent less time in the patch when faced with a competitor than when alone. This study is the first to test the influence of direct interspecific competition and arrival order on patch exploitation strategies in parasitoid species, and highlights the necessity to include agonistic behaviors in theoretical models predicting optimal patch residence time in competitive situations.

Effect of plant nitrogen and water status on the foraging behavior and fitness of an omnivorous arthropod.

Omnivorous arthropods make dietary choices according to the environment in which they forage, mainly availability/quality of plant and/or prey resources. Such decisions and their subsequent impacts on life-history traits may be affected by the availability of nutrients and water to plants, that is, through bottom-up forces. By setting up arenas for feeding behavior observation as well as glasshouse cages for plant preference assessment, we studied effects of the presence of prey (Lepidoptera eggs) and nitrogen/water availability to host tomato plants on the foraging behavior and life-history traits in the omnivorous predator Macrolophus pygmaeus (Heteroptera: Miridae). In the absence of prey, the predator fed equally on the plants treated with various levels of nitrogen and water. In the presence of prey, however, the feeding rate on plants decreased when the plant received low water input. The feeding rate on prey was positively correlated with feeding rate on plants; that is, prey feeding increased with plant feeding when the plants received high water input. Moreover, plants receiving high water input attracted more M. pygmaeus adults compared with those receiving low water input. For M. pygmaeus fitness, the presence of prey enhanced its fertility and longevity, but the longevity decreased when plants received low compared with high water input. In conclusion, the omnivorous predator may be obliged to feed on plants to obtain water, and plant water status may be a limiting factor for the foraging behavior and fitness of the omnivorous predator.

The functional response predicts the effect of resource distribution on the optimal movement rate of consumers.

Understanding how often individuals should move when foraging over patchy habitats is a central question in ecology. By combining optimality and functional response theories, we show analytically how the optimal movement rate varies with the average resource level (enrichment) and resource distribution (patch heterogeneity). We find that the type of functional response predicts the effect of enrichment in homogeneous habitats: enrichment should decrease movement for decelerating functional responses, but increase movement for accelerating responses. An intermediate resource level thus maximises movement for type-III responses. Counterintuitively, greater movement costs favour an increase in movement. In heterogeneous habitats predictions further depend on how enrichment alters the variance of resource distribution. Greater patch variance always increases the optimal rate of movement, except for type-IV functional responses. While the functional response is well established as a fundamental determinant of consumer-resource dynamics, our results indicate its importance extends to the understanding of individual movement strategies.

New parasitoid-predator associations: female parasitoids do not avoid competition with generalist predators when sharing invasive prey.

Optimal habitat selection is essential for species survival in ecosystems, and interspecific competition is a key ecological mechanism for many observed species association patterns. Specialized animal species are commonly affected by resource and interference competition with generalist and/or omnivorous competitors, so avoidance behavior could be expected. We hypothesize that specialist species may exploit broad range cues from such potential resource competitors (i.e., cues possibly common to various generalist and/or omnivorous predators) to avoid costly competition regarding food or reproduction, even in new species associations. We tested this hypothesis by studying short-term interactions between a native larval parasitoid and a native generalist omnivorous predator recently sharing the same invasive host/prey, the leaf miner Tuta absoluta. We observed a strong negative effect of kleptoparasitism (food resource stealing) instead of classical intraguild predation on immature parasitoids. There was no evidence that parasitoid females avoided the omnivorous predator when searching for oviposition sites, although we studied both long- and short-range known detection mechanisms. Therefore, we conclude that broad range cue avoidance may not exist in our biological system, probably because it would lead to too much oviposition site avoidance which would not be an efficient and, thus, beneficial strategy. If confirmed in other parasitoids or specialist predators, our findings may have implications for population dynamics, especially in the current context of increasing invasive species and the resulting creation of many new species associations.

Extensive transcription analysis of the Hyposoter didymator Ichnovirus genome in permissive and non-permissive lepidopteran host species.

Ichnoviruses are large dsDNA viruses that belong to the Polydnaviridae family. They are specifically associated with endoparasitic wasps of the family Ichneumonidae and essential for host parasitization by these wasps. We sequenced the Hyposoter didymator Ichnovirus (HdIV) encapsidated genome for further analysis of the transcription pattern of the entire set of HdIV genes following the parasitization of four different lepidopteran host species. The HdIV genome was found to consist of at least 50 circular dsDNA molecules, carrying 135 genes, 98 of which formed 18 gene families. The HdIV genome had general features typical of Ichnovirus (IV) genomes and closely resembled that of the IV carried by Hyposoter fugitivus. Subsequent transcriptomic analysis with Illumina technology during the course of Spodoptera frugiperda parasitization led to the identification of a small subset of less than 30 genes with high RPKM values in permissive hosts, consisting with these genes encoding crucial virulence proteins. Comparisons of HdIV expression profiles between host species revealed differences in transcript levels for given HdIV genes between two permissive hosts, S. frugiperda and Pseudoplusia includens. However, we found no evident intrafamily gene-specific transcription pattern consistent with the presence of multigenic families within IV genomes reflecting an ability of the wasps concerned to exploit different host species. Interestingly, in two non-permissive hosts, Mamestra brassiccae and Anticarsia gemmatalis (most of the parasitoid eggs were eliminated by the host cellular immune response), HdIV genes were generally less strongly transcribed than in permissive hosts. This suggests that successful parasitism is dependent on the expression of given HdIV genes exceeding a particular threshold value. These results raise questions about the mecanisms involved in regulating IV gene expression according to the nature of the lepidopteran host species encountered.

Intraguild interactions between two egg parasitoids of a true bug in semi-field and field conditions.

Research on interspecific competitive interactions among insect parasitoids has often been characterized by laboratory studies in which host insects are exposed to female parasitoids of different species in various sequences and combinations. In the last years, an increasing number of studies have investigated interspecific interactions under field and semi-field conditions although just a few number of works focused on egg parasitoids. In this work, we undertook a two-year study to investigate interspecific interactions between Trissolcus basalis (Wollaston) (Hymenoptera: Platygastridae) and Ooencyrtus telenomicida (Vassiliev) (Hymenoptera: Encyrtidae), two egg parasitoids of the pest Nezara viridula (L.) (Heteroptera: Pentatomidae) that co-occur in cultivated crops. Under semi-field (in out-door mesh cages) and field conditions, we investigated: 1) the seasonal occurrence of competing parasitoid species on sentinel egg masses; 2) the impact achieved by competing species on the shared host on naturally laid egg masses; 3) the outcome of intraguild interactions under controlled conditions. Results from sentinel egg masses showed that T. basalis occurs in May and successfully parasitizes hosts until the end of September/beginning of October, whereas O. telenomicida is mainly occurring in July-August. In both years, it was found that T. basalis is predominant. From naturally laid egg masses, results indicated that T. basalis achieves higher impact on the hosts, even in those egg masses which are parasitized by more than one female of different species ( =  multiparasitism). Results from manipulating intraguild interactions showed that T. basalis achieves higher impact on N. viridula when released alone, but it suffers from competition with O. telenomicida. The ecological factors that play a role in intraguild interactions in the context of biological control perspective are discussed.

Avoiding pitfalls in estimating heritability with the common options approach.

In many circumstances, heritability estimates are subject to two potentially interacting pitfalls: the spatial and the regression to the mean (RTM) fallacies. The spatial fallacy occurs when the set of potential movement options differs among individuals according to where individuals depart. The RTM fallacy occurs when extreme measurements are followed by measurements that are closer to the mean. We simulated data from the largest published heritability study of a behavioural trait, colony size choice, to examine the operation of the two fallacies. We found that spurious heritabilities are generated under a wide range of conditions both in experimental and correlative estimates of heritability. Classically designed cross-foster experiments can actually increase the frequency of spurious heritabilities. Simulations showed that experiments providing all individuals with the identical set of options, such as by fostering all offspring in the same breeding location, are immune to the two pitfalls.

How optimal foragers should respond to habitat changes: a reanalysis of the Marginal Value Theorem.

The Marginal Value Theorem (MVT) is a cornerstone of biological theory. It connects the quality and distribution of patches in a fragmented habitat to the optimal time an individual should spend exploiting them, and thus its optimal rate of movement. However, predictions regarding how habitat alterations should impact optimal strategies have remained elusive, with heavy reliance on graphical arguments. Here we derive the sensitivity of realized fitness and optimal residence times to general habitat attributes, for homogeneous and heterogeneous habitats, retaining the level of generality of the MVT. We provide new predictions on how altering travel times, patch qualities and/or relative abundances should affect optimal strategies, and study the consequences of habitat heterogeneity. We show that knowledge of average characteristics is in general not sufficient to predict the change in the average rate of movement. We apply our results to examine the conditions under which the optimal strategies are invariant to scaling. We prove a previously conjectured form of invariance in homogeneous habitats, but show that invariances to scaling are not generic in heterogeneous habitats. We also consider the relative exploitation of patches that differ in quality, clarifying the conditions under which it is adaptive to stay longer on poorer patches.