Methyl Salicylate Attracts Natural Enemies and Reduces Populations of Soybean Aphids (Hemiptera: Aphididae) in Soybean Agroecosystems

Transcription

1 FIELD AND FORAGE CROPS Methyl Salicylate Attracts Natural Enemies and Reduces Populations of Soybean Aphids (Hemiptera: Aphididae) in Soybean Agroecosystems RACHEL E. MALLINGER, 1 DAVID B. HOGG, AND CLAUDIO GRATTON Department of Entomology, University of WisconsinÐMadison, 1630 Linden Street, Madison, WI J. Econ. Entomol. 104(1): 115Ð124 (2011); DOI: /EC10253 ABSTRACT Methyl salicylate, an herbivore-induced plant volatile, has been shown to attract natural enemies and affect herbivore behavior. In this study, methyl salicylate was examined for its attractiveness to natural enemies of the soybean aphid, Aphis glycines Matsumura (Hemiptera: Aphididae), and for its direct effects on soybean aphid population growth rates. Methyl salicylate lures were deployed in plots within organic soybean [Glycine max (L.) Merr.] Þelds. Sticky card traps adjacent to and 1.5 m from the lure measured the relative abundance of natural enemies, and soybean aphid populations were monitored within treated and untreated plots. In addition, exclusion cage studies were conducted to determine methyl salicylateõs effect on soybean aphid population growth rates in the absence of natural enemies. SigniÞcantly greater numbers of syrphid ßies (Diptera: Syrphidae) and green lacewings (Neuroptera: Chrysopidae) were caught on traps adjacent to the methyl salicylate lure, but no differences in abundance were found at traps 1.5 m from the lure. Furthermore, abundance of soybean aphids was signiþcantly lower in methyl salicylate-treated plots. In exclusion cage studies, soybean aphid numbers were signiþcantly reduced on treated soybean plants when all plants were open to natural enemies. When plants were caged, however, soybean aphid numbers and population growth rates did not differ between treated and untreated plants suggesting no effect of methyl salicylate on soybean aphid reproduction and implicating the role of natural enemies in depressing aphid populations. Although aphid populations were reduced locally around methyl salicylate lures, larger scale studies are needed to assess the technology at the whole-þeld scale. KEY WORDS volatile methyl salicylate, soybean aphid, syrphid ßy, lacewing, herbivore-induced plant Herbivore-induced plant volatiles constitute one particular group of allelochemicals that are released when herbivore elicitors induce plant defense pathways (Reymond and Farmer 1998, Turlings and Ton 2006, Bruce and Pickett 2007, Felton and Tumlinson 2008, Schmelz et al. 2009). Herbivore-induced plant volatiles have been shown to mediate relationships between plants and insects through the attraction of natural enemies and the repulsion of herbivores (Turlings et al. 1990; Hildebrand et al. 1993; Takabayashi and Dicke 1996; Drukker et al. 2000; De Moraes et al. 2001; Kessler and Baldwin 2001; Cardoza et al. 2003; de Boer and Dicke 2004a,b, 2005; Ode 2006; Turlings and Ton 2006; Goggin 2007). It has long been anticipated that manipulation of herbivore-induced plant volatiles in synthetic or natural form can be used to concentrate and increase populations of natural enemies within an area of interest, such as a crop Þeld, or to repel pests from crop plants (Dicke et al. 1990, Turlings et al. 1990, Kessler and Baldwin 2002, Turlings and Ton 2006, Khan et al. 2008). 1 Corresponding author, remallinger@wisc.edu. Methyl salicylate is a phenolic compound that serves in plant defenses against pathogens and herbivores and is a common component of herbivore-induced plant volatile blends (Shulaev et al. 1997, Park et al. 2007, Vlot et al. 2008). Upon activation of the salicylic acid defense pathway, liquid methyl salicylate increases in concentration within a plant, and the volatile form is subsequently released in high concentrations (Kessler and Baldwin 2001, Kessler and Baldwin 2002, Park et al. 2007, Li et al. 2008, Vlot et al. 2008). It is speculated to be a good volatile for use in pest management because its presence does not vary among plant cultivars, as do other herbivore-induced plant volatile compounds (Takabayashi et al. 1991). SpeciÞcally, feeding by the soybean aphid, Aphis glycines Matsumura (Hemiptera: Aphididae), has been shown to activate the salicylic acid defense pathway within soybean, Glycine max (L.) Merr., plants (Li et al. 2008), resulting in the emission of high levels of methyl salicylate (Zhu and Park 2005). Furthermore, methyl salicylate can itself induce the salicylic acid defense pathway and subsequent release of methyl salicylate in healthy plants (Shulaev et al. 1997, Farmer 2001, Heidel and Baldwin 2004). The cascading release /11/0115Ð0124$04.00/ Entomological Society of America

2 116 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 104, no. 1 of this volatile can thereby extend the spatial range of methyl salicylateõs effects in an agroecosystem. Thus, through its role in plant defensive pathways and its role as a volatile signal, methyl salicylate has the potential to inßuence herbivoreðplant interactions. The capacity of methyl salicylate to attract natural enemies has been demonstrated in laboratories and agroecosystems, but the speciþc nature of responses is highly varied. When deployed in vineyards and hopyards, synthetic methyl salicylate attracted six groups of predatory insects and one genus of parasitoid wasp (James 2003a,b; James and Price 2004; James and Grasswitz 2005). It also has attracted predatory mites and an egg parasitoid in laboratory experiments (de Boer and Dicke 2004a,b, 2005; Williams et al. 2008) and anthocorid predators (Hemiptera: Anthocoridae) in both laboratories and cotton (Gossypium spp.) Þelds (Drukker et al. 2000, Yu et al. 2008). However, only the sevenspotted lady beetle, Coccinella septempunctata L., was attracted to methyl salicylate in soybean Þelds (Zhu and Park 2005). The system-speciþc responses of natural enemies to methyl salicylate suggest that factors such as concentration, release rate, deployment method, or deployment time during the season may strongly inßuence the volatileõs attractiveness to natural enemies (James 2003a,b; Zhu and Park 2005; Yu et al. 2008), making it difþcult to predict how effective it will be in pest management across agroecosystems. In addition to indirectly impacting herbivore populations through natural enemy attraction, methyl salicylate also can directly impact the behavior of herbivores. Methyl salicylate has been shown to function as a natural cue for aphids to alternate hosts in the spring, repelling them from their overwintering host and into their summer host (Pettersson et al. 1994, Glinwood and Petterson 2000). When deployed in barley (Hordeum vulgare L.), a summer host, synthetic methyl salicylate repelled aphid pests, delaying spring colonization and resulting in signiþcantly reduced aphid abundance (Ninkovic et al. 2003). Other studies have shown aphids to be attracted to low concentrations of methyl salicylate comparable with the constitutive levels emitted by their plant hosts (Han et al. 2005, Webster et al. 2008). Herbivore-induced plant volatiles also have the potential to affect herbivores through increasing plant antibiosis. The presence of synthetic volatiles as well as the natural release of volatiles reduced the number of eggs laid by moths on their host plants (Kessler and Baldwin 2001). Similarly, the artiþcial induction of salicylic acid defenses has been shown to reduce aphid population growth on tomato plants, Solanum lycopersicum L. (Cooper et al. 2004). Thus, herbivore-induced plant volatiles such as methyl salicylate may be capable of affecting changes within the host plant, including the induction of plant defenses that in turn reduce herbivore fecundity. The capacity to regulate herbivore reproduction together with the ability to attract natural enemies and repel herbivores suggests a signiþcant role for plant volatiles such as methyl salicylate in the management of pest populations. The objective of this study was to examine whether methyl salicylate formulations tested under Þeld conditions have the potential to reduce soybean aphid populations indirectly through attracting natural enemies or directly through effects on the aphids themselves. We hypothesized that in small Þeld plots containing a synthetic methyl salicylate emitter (hereafter lure), there would be an overall greater abundance of natural enemies such as lady beetles (Coleoptera: Coccinellidae), lacewings (Neuroptera), Orius spp. (Hemiptera: Anthocoridae), and braconid wasps (Hymenoptera: Braconidae) compared with untreated plots. We also predicted that natural enemy abundance would be greater nearest the methyl salicylate lure and would decline with distance from the lure. Finally, we hypothesized that the presence of high concentrations of synthetic methyl salicylate would result in reduced soybean aphid fecundity and population growth rates. The total effects of methyl salicylate would result in lower soybean aphid abundance within plots containing methyl salicylate lures compared with plots without lures. Materials and Methods Deployment of Methyl Salicylate. Methyl salicylate emitters were 5 g, 90 d, slow-release lures (Predalure, AgBio, Westminster, CO). In 2008, these lures were deployed in two organically managed soybean Þelds in south central Wisconsin in which no insect pest management treatments occurred. Field 1, located near Randolph, WI ( N, W), contained 45 acres of soybean (ÔBlue River 16A7 ) that was planted on 24 May Field 2, located near Columbus, WI ( N, W), was 28 acres of soybean (ÔViking ) planted on 3 June Both of these Þelds were surrounded on three sides by other Þeld crops (corn, Zea mays L; or soybean) and on one side by either a road or a backyard. The experimental design was a randomized complete block with 16 pairs of plots (2 by 2 m) dispersed throughout the two soybean Þelds. Each plot within a pair was randomly assigned as treated (receiving a methyl salicylate lure) or untreated (no lure). Treated and untreated plots within each pair were 150 m apart, and pairs were a minimum of 50 m from any Þeld edge and 150 m from other pairs of plots. In each treated plot, one methyl salicylate lure was hung from a plastic stake at the center of the plot at canopy height. Lures were put in the Þeld on 1 July 2008, 3 wk after the emergence of soybean, and replaced once at the end of July. Sampling for Natural Enemies and Aphids. One yellow sticky card trap (Pherocon AM, Trece Inc., Adair, OK) was attached to a plastic stake at canopy height, adjusted throughout the growing season, and placed in the center of each plot just below the methyl salicylate lure (hereafter adjacent to the lure) or alone in untreated plots. In addition, four yellow sticky card traps were hung at the corners of all treated plots (hereafter 1.5 m from the lure) and untreated plots. Traps were placed in the Þeld seven times throughout

3 February 2011 MALLINGER ET AL.: METHYL SALICYLATE AND SOYBEAN APHID 117 the growing season from 3 July 2008 when plants were at the V3 stage (three sets of unfolded trifoliate leaves) through 26 August 2008, plants at R5 stage (seed is 3 mm long in the pod at one of the four uppermost nodes on the main stem), and on each occasion were left in the Þeld for 4 d before collection. Once collected, traps were frozen and stored until analysis, at which point all natural enemies on traps were identiþed to family or species level. For corner traps, insect abundance is expressed as averages of the four traps. In addition, whole plant sampling was conducted within all plots to determine abundance of natural enemy immatures, natural enemy eggs and soybean aphids. On 12 dates throughout the 2008 growing season, 10Ð15 randomly selected plants in treated and untreated plots were examined (nondestructively) from top to bottom for natural enemies and soybean aphids. Natural enemy eggs and immatures were counted on a per plant basis and recorded to family level, and the total number of soybean aphids per plant also was recorded. Methyl Salicylate Effect on Aphid Population Growth: Exclusion Cage Experiment. To assess the potential direct effects of methyl salicylate on aphid population growth, an exclusion cage experiment was conducted. Soybean (ÔBSR 101 ) was planted in 15- cm-diameter pots in which multiple strands of nylon string, each 33 cm long, were taped to the inside to act as watering wicks. All pots contained two soybean plants and were grown in the greenhouse. When the plants reached the V2ÐV3 stage (two to three fully expanded sets of trifoliate leaves), two pots were placed in small plastic containers for deployment in the Þeld. Each container had two 13-cm-diameter holes cut in the lid to accommodate the two pots of soybean plants that sat 15 cm from the bottom of the container. Water was placed in the bottom of each container, and the nylon strings extending from the pots served as watering wicks. For both experimental pots in a container, 50 fourth-instar aphids, reared in the laboratory on soybean plants (ÔBSR 101 ) under a photoperiod of 16:8 (L:D) h, were distributed among the two soybean plants per experimental pot. One pot per container was then randomly assigned to be caged to exclude natural enemies. Nylon mesh bags (200- m mesh) were placed over the plants and closed with rubber bands around the top of the pot to prohibit movement of insects off of and onto the soybean plants. Plastic containers holding one caged and one open pot of soybean plants were deployed on 25 June 2009 in organically managed soybean Þelds at the University of Wisconsin MadisonÕs Agricultural Research Station in Arlington, WI. Containers were placed in the Þeld in pairs, with members of the pair separated by 50 m. One container of each pair was randomly assigned to be exposed to methyl salicylate: both the open and caged pots received one 5-g methyl salicylate lure (Predalure, AgBio) hung from a plastic stake inserted into the pots. The stake and lure were placed inside nylon mesh bags on caged pots. All untreated containers had identical stakes with no lures. The experimental design included a total of 15 pairs of containers (i.e., 15 treated and 15 untreated). These plants were left in the Þeld for 2 wk, after which period the mesh bags were removed from caged pots. Aphids, as well as any natural enemies, on caged and open pots were counted and expressed as aphids per pot. The low initial densities of aphids and the short time frame of the experiment were designed to reduce cage impacts on the natural emigration of winged aphids, which was found to be minimal when Þnal aphid densities were 4,000 aphids per plant (Donaldson et al. 2007). Methyl Salicylate Effect on Aphid Life Table Parameters: Clip Cage Experiment. In addition to the above-described exclusion cage study, a clip cage experiment was conducted to determine the effects of methyl salicylate on soybean aphid population growth parameters. Soybean plants at the R1 stage (plants having at least one ßower on any node) in organically managed soybean plots at the Agricultural Research Station in Arlington were used for this experiment. On 23 July 2009, 28 plants were randomly selected within these plots, with all plants at least 50 m away from any other selected plant. One fourth-instar aphid, reared in the laboratory on BSR 101 soybean plants under a photoperiod of 16:8 (L:D) h, was placed on the youngest, fully developed leaf of each selected plant and enclosed within a 5-cm-diameter clip cage as described in Myers and Gratton (2006). Half of the plants were randomly selected to be treated, and 5 g methyl salicylate lures (Predalures, AgBio) were hung from plastic stakes as close to each clip cage as possible. Untreated plants had stakes without a methyl salicylate lure placed close to the clip cage. A same-age cohort was achieved when all aphids in clip cages had produced at least two neonates on any 1 d (with previous offspring removed daily). The experiment started 24 h later when the initial parent and one of the neonates were removed to leave one 2-d-old aphid in each clip cage. All cages were subsequently checked every 24 h for 15 d, and after the aphids began reproducing (around the seventh day), the number of offspring per aphid was recorded daily. Offspring were left in the cages so as to avoid disturbance and the experiment was terminated before offspring reached maturity and had produced offspring themselves. Statistical Analyses. Natural enemy and aphid counts collected through sticky traps and whole plant samples were analyzed with repeated measures mixed model analyses of variance. Fixed effects included treatment, date, and the interaction between these two variables whereas random effects included plot nested within block, and date was treated as the repeated variable (nlme package, version , R Development Core Team 2009). All data were log or square root transformed where appropriate to meet the assumptions of normality and homogeneity of variance. Differences between Þnal aphid abundance on treated and untreated plants in the exclusion cage study were tested with mixed model analyses of variance. Analyses were done separately for aphids on

4 118 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 104, no. 1 caged plants (treated versus untreated) and open plants (treated versus untreated), and each analysis was conducted with a Þxed effect of treatment and a random effect of individual pot nested within block (nlme package, version , R Development Core Team 2009). In addition, a biocontrol service index was calculated to measure the functional activity of natural enemies on both the treated and untreated pots of plants (Gardiner et al. 2009). The biocontrol service index was determined by the relative changes in aphid abundance on caged and open pots of plants using the formula. 15 A c,n A o,n A c,n n 1 BSI n where A c,n is the Þnal number of aphids on caged plants for a given replicate, A o,n is the Þnal number of aphids on open plants for a given replicate, and n is the number of replicates, or pairs of caged and open plants. The biocontrol service index gives a value varying from 0 to 1, with the values increasing as the level of aphid predation increases. The biocontrol service indices of treated and untreated pots of plants were compared using mixed model analyses of variance with treatment as the Þxed effect and the container of pots nested within block as the random effect (nlme package version , R Development Core Team 2009). Data were transformed where appropriate to meet assumptions of normality and homogeneity of variance. Life table parameters including Þnite rate of increase, intrinsic rate of increase, net reproductive rate, and mean generation time were estimated for the aphid populations in clip cages by using PopTools 2.5 (Hood 2003) with standard errors of the estimates calculated by jackkniþng (Myers and Gratton 2006). Statistical differences in life table parameters between aphids in treated versus untreated clip cages were determined using randomization tests (Myers and Gratton 2006). The null hypothesis in such tests was that there was no difference in any of the life table parameters for aphids in treated and untreated clip cages. In the randomization tests, the individual replicates from treated and untreated clip cages were randomly shufßed to comprise two new groups containing aphids from both treatments. Life table parameters were calculated for each of the two randomly constructed groups, and this was repeated for 1,000 randomization iterations. A test statistic was computed as the average difference between life table parameters in the two randomly constructed groups over 1,000 iterations and this test statistic was then compared with the difference in life table parameters calculated from the original data. An overall P value for the statistical difference between treated and untreated clip cages was determined by the number of times that the difference between life table parameters in the two randomly constructed groups for each randomization iteration exceeded the difference between the life table parameters calculated from the original data. Results Abundance of Natural Enemy Adults. Sticky card traps placed at the center of treated plots, adjacent to the methyl salicylate lure, had signiþcantly more natural enemy adults compared with traps placed at the center of untreated plots (F 39.0; df 1, 7; P 0.001; Fig. 1A). There was also a signiþcant time treatment interaction (F 3.0; df 6, 84; P 0.01), with differences in natural enemy abundance being most pronounced during the month of July (Fig. 1A). Data from both soybean farms were analyzed together as there was no signiþcant interaction between treatment and farm (F 1.1; df 1, 6; P 0.34). Among individual natural enemy taxa, Syrphidae (Diptera), syrphid ßies, and Chrysopidae (Neuroptera), green lacewings, were both signiþcantly more abundant on sticky card traps placed in the center of treated plots compared with untreated plots with no signiþcant time treatment interactions for either taxon (Table 1). There was no signiþcant difference in abundance of other natural enemy taxa such as the lady beetles Coccinella septempunctata L. and Harmonia axyridis (Pallas); the insidious ßower bug, Orius insidious (Say); braconid wasps (Hymenoptera: Braconidae); and brown lacewings (Neuroptera: Hemerobiidae), with no signiþcant time treatment interactions for any taxon (Table 1). There was no signiþcant difference in the total number of natural enemy adults caught in sticky card traps placed at the corners of treated plots, 1.5 m from the methyl salicylate lure, compared with untreated plots (F 0.2; df 1, 7; P 0.67; Fig. 1B), with no signiþcant time treatment interaction (F 0.4; df 6, 84; P 0.88). Again, data from both farms were pooled because there was no signiþcant interaction between farm and treatment (F 0.4; df 1, 6; P 0.55). Furthermore, the abundance of speciþc natural enemy taxa found in these corner traps did not differ between treated and untreated plots, including C. septumpunctata, H. axyridis, O. insidiosus, Braconidae, Syrphidae, Chrysopidae, and Hemerobiidae (Table 2). There was a signiþcant time treatment interaction for H. axyridis but not for the other aforementioned taxa (Table 2). Abundance of Natural Enemy Immatures and Eggs. Abundance of natural enemy immatures, including larvae of Syrphidae, Coccinellidae, Chrysopidae, and Hemerobiidae, as well as nymphs of O. insidiosus, was not signiþcantly different in treated versus untreated plots (F 0.1; df 1, 7; P 0.72; Fig. 2A), and there was no signiþcant time treatment interaction (F 0.4; df 11, 154; P 0.95). In addition, there was no signiþcant difference in the abundance of natural enemy eggs, including those of Syrphidae, Chrysopidae, Hemerobiidae, and Coccinellidae, in treated versus untreated plots (F 0.04; df 1, 7; P 0.85; Fig. 2B), with no signiþcant time treatment interaction (F

5 February 2011 MALLINGER ET AL.: METHYL SALICYLATE AND SOYBEAN APHID 119 Fig. 1. Average abundance (per 4-d period), means SEM, of natural enemies captured on sticky card traps during the 2008 growing season in soybean Þelds in southern Wisconsin. (A) Sticky card traps placed in the center of untreated and treated plots (adjacent to the methyl salicylate lure). (B). Sticky card traps placed at the corners of untreated and treated plots (1.5 m from the methyl salicylate lure). 1.1; df 11, 154; P 0.40). Immature and egg abundances were pooled across both farms because there were no signiþcant interactions between farm and treatment for immatures (F 0.6; df 1, 6; P 0.47) or eggs (F 0.3; df 1, 6; P 0.59). Aphid Abundance. Aphids were signiþcantly more abundant in untreated plots compared with treated plots (F 8.0; df 1, 7; P 0.03; Fig. 3), with no signiþcant time treatment interaction (F 1.0; df 11, 154; P 0.31). Data of aphid abundance were Table 1. Average abundance of natural enemy taxa captured on sticky card traps placed in the center of untreated and treated plots (adjacent to the methyl salicylate lure) over seven 4-d periods during the 2008 growing season Avg per-trap ANOVA results Taxa Treatment abundance (mean SEM) Treatment Time Time treatment C. septempunctata Treated F 0.04, P 0.85 F 2.8, P 0.02* F 0.3, P 0.93 Untreated H. axyridis Treated F 1.2, P 0.31 F 0.8, P 0.61 F 1.8, P 0.12 Untreated O. insidiosus Treated F , P 1.00 F 27.0, P 0.001*** F 0.4, P 0.86 Untreated Braconidae Treated F 7.0, P 0.49 F 1.0, P 0.45 F 1.1, P 0.36 Untreated Syrphidae Treated F 45.0, P 0.001*** F 14.0, P 0.001*** F 1.0, P 0.26 Untreated Chrysopidae Treated F 5.3, P 0.05* F 2.0, P 0.08 F 1.5, P 0.20 Untreated Hemerobiidae Treated F 0.8, P 0.39 F 2.3, P 0.04* F 1.1, P 0.40 Untreated Treatment df 1, 7; time df 6, 84; and time treatment df 6, 84.

6 120 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 104, no. 1 Table 2. Average abundance of natural enemy taxa captured on sticky card traps placed in the corners of untreated and treated plots (1.5 from the methyl salicylate lure) over seven 4-d periods during the 2008 growing season Avg per-trap ANOVA results Taxa Treatment abundance (mean SEM) Treatment Time Time treatment C. septempunctata Treated F 0.1, P 0.77 F 5.2, P 0.001*** F 0.3, P 0.94 Untreated H. axyridis Treated F 0.3, P 0.60 F 1.9, P 0.09 F 2.4, P 0.04* Untreated O. insidiosus Treated F 0.2, P 0.66 F 40.0, P 0.001*** F 0.2, P 0.97 Untreated Braconidae Treated F 0.3, P 0.6 F 1.0, P 0.27 F 0.3, P 0.93 Untreated Syrphidae Treated F 0.1, P 0.73 F 20.0, P 0.001*** F 0.8, P 0.61 Untreated Chrysopidae Treated F 0.5, P 0.51 F 6.6, P 0.001*** F 0.9, P 0.53 Untreated Hemerobiidae Treated F 0.003, P 0.96 F 2.7, P 0.02* F 1.6, P 0.16 Untreated Treatment df 1, 7; time df 6, 84; and time treatment df 6, 84. pooled for both farms as there was no farm by treatment interaction (F 1.9; df 1, 6; P 0.22). Exclusion Cage Study: Aphid Population Growth. Aphid abundance at the end of the 2-wk exclusion cage study was not signiþcantly different between the treated and untreated caged plants (F 0.9; df 1, 14; P 0.36; Fig. 4) but was signiþcantly different between the treated and untreated open plants (F 6.1; df 1, 14; P 0.03; Fig. 4). The biocontrol service index, a measure of the functional activity level of Fig. 2. Average abundance, means SEM, of natural enemies sampled through whole plant sampling in methyl salicylate-treated and untreated plots over 12 sampling dates during the 2008 growing season. (A) Natural enemy immatures, including larvae of Coccinellidae, Chrysopidae, Hemerobiidae, and Syrphidae as well as nymphs of insidious ßower bug. (B) Natural enemy eggs, including those of Syrphidae, Chrysopidae, Hemerobiidae, and Coccinellidae.

7 February 2011 MALLINGER ET AL.: METHYL SALICYLATE AND SOYBEAN APHID 121 Clip Cage Study: Aphid Life Table Parameters. Aphids reared in clip cages on treated and untreated plants showed no signiþcant differences in any life table parameters including aphid Þnite rate of population increase ( ), intrinsic rate of population increase (r), net reproductive rate (R o ), and mean generation time (T) (Table 3). Fig. 3. Average abundance, means SEM, of soybean aphids per soybean plant sampled by whole-plant sampling in methyl salicylate-treated and untreated plots during the 2008 growing season. natural enemies (Gardiner et al. 2009), calculated for plants paired with methyl salicylate lures ( ), was marginally signiþcantly greater than that calculated for untreated plants ( ) (F 4.4; df 1, 14; P 0.06). Although few natural enemies were found on the open plants when aphids were counted, whole plant samples of natural enemies taken during the 2-wk experiment revealed the presence of two lady beetle species, Hippodamia convergens Guérin-Méneville, and Hippodamia variegata (Goeze), O. insidiosus, eggs of Syrphidae and Chrysopidae, harvestmen (Arachnida: Opiliones), and spiders in the study Þeld. Fig. 4. Final average aphid abundance, means SEM, on caged and uncaged pots of soybean plants exposed to methyl salicylate or untreated. All pots began with 50 aphids and were left in the Þeld for 14 d after which time aphid abundances were recorded. SigniÞcant differences are deþned at the P 0.05 level. Discussion Herbivore-induced plant volatiles have the potential to inßuence crop pests directly through decreased feeding performance and changes in host plant preference, or indirectly by affecting the third trophic level. Traps placed adjacent to methyl salicylate lures in treated soybean plots caught signiþcantly more natural enemies compared with traps in untreated plots where no lures were deployed. Moreover, ambient aphid populations were lower in methyl salicylate treated plots and predation intensity, as measured with predator-excluders, was higher on methyl salicylate treated plants, suggesting that methyl salicylate inßuences pest abundance through effects on natural enemies. However, this effect may be highly localized because at a distance of 1.5 m from the lure, there was no difference in natural enemy abundance between traps in treated and untreated plots. Although the results illustrate the efþcacy of methyl salicylate at a small spatial scale, they also raise questions on scaling this technology to the whole Þeld. Effect of Methyl Salicylate on Natural Enemies and Aphids. Syrphid ßies (Syrphidae) and green lacewings (Chrysopidae) were the only natural enemy taxa that were more abundant on traps adjacent to methyl salicylate lures compared with traps in untreated plots. This was consistent with previous studies that demonstrated the attraction of both of these groups to methyl salicylate in grapes and hops (James 2003a,b; James and Price 2004). In contrast, we found no attraction to methyl salicylate by the sevenspotted lady beetle, a predator of the soybean aphid (Rutledge et al. 2004), found previously to be attracted to methyl salicylate in soybean (Zhu and Park 2005). However, populations of all lady beetles were unusually low in our treated and untreated plots during the 2008 growing season (R.E.M., unpublished data), probably as a result of the late establishment of soybean aphids (Cullen 2008). Sticky card traps adjacent to methyl salicylate lures showed greater natural enemy abundance most noticeably in the Þrst half of the growing season, which suggests that methyl salicylate can attract natural enemies before herbivore outbreaks. Early attraction of natural enemies into an agroecosystem is believed to be an important attribute contributing to effective biological control of herbivores (Khan et al. 2008). Increased abundance of natural enemies on traps in methyl salicylate-treated plots was associated with lower soybean aphid abundance in treated plots. The relative impact of natural enemies on aphid abundance, as measured by an exclusion cage study, suggests that one mechanism through which methyl sa-

8 122 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 104, no. 1 Table 3. Life table parameters of soybean aphids raised in clip cages on methyl salicylate treated and untreated soybean plants Treatment n (95% CI) r (95% CI) R o (95% CI) T (95% CI) Untreated a (1.289, 1.297) 0.257a (0.254, 0.260) a (12.713, ) 9.217a (9.111, 9.341) Treated a (1.282, 1.288) 0.251a (0.249, 0.253) a (13.567, 14.1) a (9.976, ) n, number of female soybean aphids in the treatment group;, Þnite rate of population increase; r, intrinsic rate of population increase; R o, net reproductive rate; and T, mean generation time. The 95% CI were constructed using jackknife estimates of the parameters. Means within a column followed by the same letter are not signiþcantly different (P 0.05) as determined using Monte Carlo randomization tests. licylate affects aphid populations is via the attraction of natural enemies. In our study, the natural enemy group displaying the strongest attraction to methyl salicylate was syrphid ßies, which as adults do not feed on aphids but rather on pollen and nectar (Sommaggio 1999). Green lacewings also were attracted to methyl salicylate, but they have not been implicated as having strong effects on soybean aphid populations (Fox et al. 2004, Rutledge et al. 2004, Costamagna and Landis 2007, Costamagna et al. 2007) and also are predaceous primarily in the larval stage (Principi and Canard 1984, Costamagna and Landis 2007). Moreover, we did not Þnd greater numbers of green lacewing or syrphid larvae and eggs in treated plots. Thus, we remain uncertain as to which natural enemies are actually affecting aphid populations. However, our sampling methods may not have been adequate for capturing natural enemy immatures and eggs, because they are rarely found on sticky traps and can be difþcult to identify or locate in whole-plant sampling. Other taxa that may have been present in our system but were not adequately captured using our collection technique (sticky cards), such as spiders or ground beetles (Carabidae), also may have signiþcantly contributed to the predation of soybean aphids (Fox et al. 2005, Hajek et al. 2007, Birkhofer et al. 2008, Hannam et al. 2008, Schmidt et al. 2008, Oelbermann and Scheu 2009). We can reject in part the hypothesis that methyl salicylate-induced plant responses affect aphid populations directly. In the two studies where aphids were caged on plants in proximity of methyl salicylate (exclusion cage study and clip cage experiment), there was no difference in the Þnal number of aphids or in aphid life table parameters on treated versus untreated plants, offering no support for methyl salicylateõs role in reducing aphid fecundity. However, neither of these studies could determine whether methyl salicylate reduces soybean aphid abundance by repelling colonizing aphids or by increasing emigration of winged aphids via changes in host plant preference. The lower numbers of aphids on open, treated plants in the exclusion cage study as well as on plants within treated plots could be the result of multiple mechanisms such as methyl salicylate increasing the emigration of aphids and attracting natural enemies. Methyl Salicylate in Aphid Pest Management. The efþcacy of methyl salicylate in pest management is dependent on its ability to consistently and predictably attract natural enemies that can suppress pests at Þeld scales. The heterogeneous response of natural enemies to herbivore-induced plant volatiles found across multiple studies suggests that methyl salicylateõs attractiveness could be modulated by environmental factors as well as by the release rate and deployment mode of the volatile. For example, predatory mites were attracted to intermediate doses of methyl salicylate but showed no response to low doses and were repelled by high doses (de Boer and Dicke 2004b). In addition, the ability of an herbivore-induced plant volatile to attract a particular natural enemy taxon can depend on the natural enemyõs previous exposure to volatiles (de Boer and Dicke 2004a). Volatile blend, in turn, depends not only on the species of plant but also the plant cultivar and plant age (Takabayashi and Dicke 1996) as well as factors such as light intensity, temperature, soil moisture, and simultaneous attack by other pests or pathogens (Cardoza et al. 2002, Gouinguene and Turlings 2002). Therefore, natural enemies of aphids may display differential attraction to synthetic plant volatiles based on the species, cultivar, and plant age of their preyõs host and on environmental factors affecting plant volatile blend. Results of in-þeld studies are thus highly speciþc to the agroecosystem and may not translate to other crop Þelds. In addition, the range at which methyl salicylate attracts natural enemies will inßuence its role in pest management and the ability to scale this management approach to grower Þelds. In this study, we were unable to determine the spatial extent of methyl salicylateõs attractiveness. Natural enemies found in greater abundance on traps adjacent to the methyl salicylate lures may have originated from areas of soybean bordering the experimental plots or may have been attracted at longer distances from the off-crop habitat. Although we did not Þnd greater abundance of natural enemies at traps 1.5 m from the methyl salicylate lure, this result does not conclusively address the issue of range. Natural enemies may still have been attracted at a long distance, orienting directly to the methyl salicylate lures and resulting in no increased abundance at traps 1.5 m from the lure. Alternatively, our results could indicate that methyl salicylate has only a short-range attractiveness and operates to redistribute natural enemies within the crop Þeld, resulting in no overall increase in natural enemy abundance or reduction in pest numbers. For methyl salicylate to function at grower-relevant scales by increasing natural enemy abundance throughout the agroecosystem, natural enemies must respond to the volatile from off-crop habitats. The landscape context within which methyl salicylate is deployed could then affect the volatileõs role in pest manage-

9 February 2011 MALLINGER ET AL.: METHYL SALICYLATE AND SOYBEAN APHID 123 ment by supporting variable numbers of natural enemies. In this study, the organic Þelds used were bordered by impervious surface (roads) and other Þeld crops, with virtually no Þeld margins containing noncrop vegetation. It is possible that methyl salicylate would have had larger effects in an agroecosystem situated in a more complex landscape, bordered by woodlots and perennial vegetation that are known to support a higher abundance of natural enemies (Bianchi et al. 2006). In summary, the Þndings of this study illustrate methyl salicylateõs efþcacy in attracting natural enemies to a lure source and show its potential in reducing herbivore populations at small spatial scales. Research addressing the most appropriate method and timing of deployment, as well as how surrounding landscape and farm management decisions interact with methyl salicylate, will advance the potential for using this volatile lure in pest management. Ultimately, experiments performed at larger spatial scales, comparing treated and untreated crop Þelds, are needed to test the applicability of this technology for pest management. Acknowledgments We thank Katlyn Arnett, John Carlee, Dan Endow, Dale Jacques, Emily Houtler, Pat Moriarty, Merritt Singleton, Sarika Sharma, Matt Stangl, Brent Wittig, and Flora Zeng for assistance in the Þeld and laboratory. We also thank Camila Botero, Ting-Li Lin, and Seung Cheon Hong for help with statistics. Methyl salicylate lures were donated by AgBio. This manuscript was improved by comments from two anonymous reviewers. This work was funded through Hatch grant WIS04956 (awarded to D.B.H. and C.G.). References Cited Birkhofer, K., E. Gavish-Regev, K. Endlweber, Y. D. Lubin, K. von Berg, D. H. Wise, and S. Scheu Cursorial spiders retard initial aphid population growth at low densities in winter wheat. Bull. Entomol. Res. 98: 249Ð255. Bianchi, F.J.J.A., C.J.H. Booij, and T. Tscharntke Sustainable pest regulation in agricultural landscapes: a review on landscape composition, biodiversity and natural pest control. Proc. R. Soc. B 273: 1715Ð1727. Bruce, T.J.A., and J. A. Pickett Plant defense signaling induced by biotic attacks. Curr. Opin. Plant Biol. 10: 387Ð392. Cardoza, Y., H. T. Alborn, and J. H Tumlinson. In vivo volatile emissions from peanut plants induced by simultaneous fungal infection and insect damage. J. Chem. Ecol. 28: 161Ð174. Cardoza, Y., P.E.A. Teal, and J. H. Tumlinson Effect of peanut plant fungal infection on oviposition preference by Spodoptera exigua and on host-searching behavior by Cotesia marginiventris. Environ. Entomol. 32: 970Ð 976. Cooper, W. C., L. Jia, and F. L. Goggin Acquired and R-gene mediated resistance against the potato aphid in tomato. J. Chem. Ecol. 30: 2527Ð2542. Costamagna, A. C., and D. A. Landis Quantifying predation on soybean aphid through direct Þeld observations. Biol. Control 42: 16Ð24. Costamagna, A. C., D. A. Landis, and C. D. Difonzo Suppression of soybean aphid by generalist predators results in a trophic cascade in soybeans. Ecol. Appl. 17: 441Ð451. Cullen, E Soybean aphid update for late August Wisconsin Crop Manager. ( WCMNews/tabid/53/EntryId/624/Soybean-Aphid- Update-for-Late-August-2008.aspx). de Boer, J. G., and M. Dicke. 2004a. Experience with methyl salicylate affects behavioral responses of a predatory mite to blends of herbivore-induced plant volatiles. Entomol. Exp. Appl. 110: 181Ð189. de Boer, J. G., and M. Dicke. 2004b. The role of methyl salicylate in prey searching behavior of the predatory mite Phytoseiulus persimilis. J. Chem. Ecol. 30: 255Ð271. de Boer, J. G., and M. Dicke Information use by the predatory mite Phytoseiulus persimilis (Acari: Phytoseiidae), a specialized natural enemy of herbivorous spider mites. Appl. Entomol. Zool. 40: 1Ð12. De Moraes, C. M., M. Mescher, and J. H. Tumlinson Caterpillar-induced nocturnal plant volatiles repel conspeciþc females. Nature 410: 577Ð580. Dicke, M., M. W. Sabelis, J. Takabayashi, J. Bruin, and M. A. Posthumus Plant strategies of manipulating predator-prey interactions through allelochemicals Ð prospects for application in pest control. J. Chem. Ecol. 16: 3091Ð3118. Donaldson, J. R., S. W. Myers, and C. Gratton Densitydependent responses of soybean aphid (Aphid glycines Matsumura) populations to generalist predators in mid to late season soybean Þelds. Biol. Control 43: 111Ð118. Drukker, B., J. Briun, and M. Sabelis Anthocorid predators learn to associate herbivore-induced plant volatiles with presence or absence of prey. Physiol. Entomol. 25: 260Ð265. Farmer, E. E Surface to air signals. Nature 411: 854Ð 856. Felton, G. W., and J. H. Tumlinson Plant-insect dialogs: complex interactions at the plant-insect interface. Curr. Opin. Plant Biol. 11: 457Ð463. Fox, T. B., D. A. Landis, F. F. Cardoso, and C. D. Difonzo Predators suppress Aphis glycines Matsumura population growth in soybean. Environ. Entomol. 33: 608Ð618. Fox, T. B., D. A. Landis, F. F. Cardoso, and C. D. Difonzo Impact of predation on establishment of the soybean aphid, Aphis glycines, in soybean, Glycine max. Biocontrol 50: 545Ð563. Gardiner, M. M., D. A. Landis, C. Gratton, C. D. DiFonzo, M. O Neal, J. M. Chacon, M. T. Wayo, N. P. Schmidt, E. E. Mueller, and G. E. Heimpel Landscape diversity enhances biological control of an introduced crop pest in the north-central USA. Ecol. Appl. 19: 143Ð154. Glinwood, R. T., and J. Petterson Change in response of R. padi spring migrants to the repellent winter host component methyl salicylate. Entomol. Exp. Appl. 94: 325Ð330. Goggin, F Plant-aphid interactions: molecular and ecological perspectives. Curr. Opin. Plant Biol. 10: 399Ð408. Gouinguene, S. P., and T.C.J. Turlings The effects of abiotic factors on induced volatile emissions in corn plants. Plant Physiol. 129: 1296Ð1307. Han, B. Y., C. S. Zhou, and L. Cui Attraction of winged tea aphids to tea plant volatiles measured by olfactometry and electro-antennography. Trop. Agric. 82: 138Ð142. Hannam, J. J., J. K. Liebherr, and A. E. Hajek Climbing behavior and aphid predation by Agonum muelleri (Coleoptera: Carabidae). Can. Entomol. 140: 203Ð207.

10 124 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 104, no. 1 Hajek, A. E., J. J. Hannam, C. Neilsen, A. J. Bell, and J. K. Liebherr Distribution and abundance of Carabidae (Coleoptera) associated with soybean aphid (Hemiptera: Aphididae) populations in central New York. Ann. Entomol. Soc. Am. 100: 876Ð886. Heidel, A. J., and I. T. Baldwin Microarray analysis of salicylic acid and jasmonic acid signaling in responses of Nicotiana attenuata to attack by insects from multiple feeding guilds. Plant Cell Environ. 27: 1362Ð1373. Hildebrand, D. F., G. C. Brown, D. M. Jackson, and T. R. Hamiltonkemp Effects of some leaf-emitted volatile compounds on aphid population increase. J. Chem. Ecol. 19: 1875Ð1887. Hood, G. M PopTools version ( csiro.au/poptools). James, D. G. 2003a. Field evaluation of herbivore-induced plant volatiles as attractants for beneþcial insects; methyl salicylate and the green lacewing, Chrysopa nigricornis. J. Chem. Ecol. 29: 1601Ð1609. James, D. G. 2003b. Synthetic herbivore-induced plant volatiles as Þeld attractants for beneþcial insects. Environ. Entomol. 32: 977Ð982. James, D. G., and T. R. Grasswitz Synthetic herbivore-induced plant volatiles increase Þeld captures of parasitic wasps. Biocontrol 50: 871Ð880. James, D. G., and T. S. Price Field-testing of methyl salicylate for recruitment and retention of beneþcial insects in grapes and hops. J. Chem. Ecol. 30: 1613Ð1628. Kessler, A., and I. T. Baldwin Function of herbivoreinduced plant volatile emissions in nature. Science 291: 2141Ð2144. Kessler, A., and I. T. Baldwin Plant responses to insect herbivory: the emerging molecular analysis. Annu. Rev. Plant Biol. 53: 299Ð328. Khan, Z. R., D. G. James, C.A.O. Midega, and J. A. Pickett Chemical ecology and conservation biological control. Biol. Control 45: 210Ð224. Li, Y., J. Zou, M. Li, D. D. Bilgin, L. O. Vodkin, G. L. Hartman, and S. J. Clough Soybean defense responses to the soybean aphid. New Phytologist 179: 185Ð 195. Mye Myers, S. W., and C. Gratton Inßuence of potassium fertility on soybean aphid, Aphis glycines Matsumura (Hemiptera: Aphididae), population dynamics at a Þeld and regional scale. Environ. Entomol. 35: 219Ð227. Ninkovic, V., E. Ahmed, R. Glinwood, and J. Pettersson Effects of two types of semiochemical on population development of the bird cherry-oat aphid Rhopalosiphum padi in a barley crop. Agric. For. Entomol. 5: 27Ð33. Oelbermann, K., and S. Scheu Control of aphids on wheat by generalist predators: effects of predator density and the presence of alternative prey. Entomol. Exp. Appl. 132: 225Ð231. Ode, P. J Plant chemistry and natural enemy Þtness: effects on herbivore and natural enemy interactions. Annu. Rev. Entomol. 51: 163Ð185. Park, S-W., E. Kaimoyo, D. Kumar, S. Mosher, and D. F. Klessig Methyl salicylate is a critical mobile signal for plant systemic acquired resistance. Science 318: 113Ð 116. Pettersson, J., J. A. Pickett, B. J. Pye, A. Quiroz, L. E. Smart, L. J. Wadhams, and C. M. Woodcock Winter host component reduces colonization by bird-cherry-oat aphid, Rhopalosiphum padi (L.) (Homoptera, Aphididae), and other aphids in cereal Þelds. J. Chem. Ecol. 20: 2565Ð2574. Principi, M. M., and M. Canard Feeding habits, pp. 76Ð92. In M. Canard, Y. Séméria, and T. R. New (eds.), Biology of Chrysopidae. Dr W Junk, The Hague, The Netherlands. Reymond, P., and E. F. Farmer Jasmonate and salicylate as global signals for defense gene expression. Curr. Opin. Plant Biol. 1: 401Ð411. Rutledge, C. E., R. J. O Neil, T. B. Fox, and D. A. Landis Soybean aphid predators and their use in integrated pest management. Ann. Entomol. Soc. Am. 97: 240Ð248. Schmelz, E. A., J. E. Engleberth, H. T. Alborn, J. H. Tumlinson, III, and P.E.A. Teal Phytohormone-based activity mapping of insect herbivore-produced elicitors. Proc. Natl. Acad. Sci. U.S.A. 106: 653Ð657. Schmidt, N. P., M. E. O Neal, and P. M. Dixon Aphidophagous predators in Iowa soybean: a community comparison across multiple years and sampling methods. Ann. Entomol. Soc. Am. 101: 341Ð350. Shulaev, V., P. Silverman, and I. Raskin Airborne signaling by methyl salicylate in plant pathogen resistance. Nature 385: 718Ð721. Sommaggio, D Syrphidae: can they be used as environmental bioindicators? Agric. Ecosyst. Environ. 74: 343Ð356. Takabayashi, J., M. Dicke, and M. A. Posthumus Variation in composition of predator attracting allelochemicals emitted by herbivore-infested plants: relative inßuence of plant and herbivore. Chemoecology 2: 1Ð6. Takabayashi, J., and M. Dicke Plant-carnivore mutualism through herbivore-induced carnivore attractants. Trends Plant Sci. 1: 109Ð113. Turlings, T.J.C., and J. Ton Exploiting scents of distress: the prospect of manipulating herbivore-induced plant odours to enhance the control of agricultural pests. Curr. Opin. Plant Biol. 9: 421Ð427. Turlings, T. C., J. H. Tumlinson, and W. J. Lewis Exploitation of herbivore-induced plant odors by hostseeking parasitic wasps. Science 250: 1251Ð1253. Vlot, A. C., D. F. Klessig, and S-W. Park Systemic acquired resistance: the elusive signal(s). Curr. Opin. Plant Biol. 11: 436Ð442. Webster, B., T. Bruce, S. Dufour, C. Birkemeyer, M. Birkett, J. Hardie, and J. Pickett IdentiÞcation of volatile compounds used in host location by the black bean aphid, Aphis fabae. J. Chem. Ecol. 34: 1153Ð1161. Williams III, L., C. Rodriguez-Saona, S. C. Castle, and S. Zhu EAG-active herbivore-induced volatiles modify behavioral responses and host attack by an egg parasitoid. J. Chem. Ecol. 34: 1190Ð1201. Yu, H., Y. Zhong, K. Wu, X. W. Gao, and Y. Y. Guo Field-testing of synthetic herbivore-induced plant volatiles as attractants for beneþcial insect. Environ. Entomol. 37: 1410Ð1415. Zhu, J., and K.-C. Park Methyl salicylate, a soybean aphid induced plant volatile attractive to the predator Coccinella septempunctata. J. Chem. Ecol. 31: 1733Ð1746. Received 6 July 2010; accepted 8 September 2010.