Imidacloprid As a Contact Arrestant for Larvae of the European Chafer, Amphimallon Majale

See discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/230748527

Imidacloprid as a contact arrestant for larvae of the European chafer, Amphimallon majale

ARTICLE in PEST MANAGEMENT SCIENCE · APRIL 2013 Impact Factor: 2.74 · DOI: 10.1002/ps.3394 · Source: PubMed

DOWNLOADS VIEWS 247 230

4 AUTHORS, INCLUDING:

Gengping Zhu Matthew J Petersen Dolphin (PG) College of Life Sciences Roanoke College

13 PUBLICATIONS 47 CITATIONS 22 PUBLICATIONS 70 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Matthew J Petersen Retrieved on: 16 August 2015 Research Article

Received: 29 September 2011 Revised: 30 June 2012 Accepted article published: 3 August 2012 Published online in Wiley Online Library:

(wileyonlinelibrary.com) DOI 10.1002/ps.3394 Imidacloprid as a contact arrestant for larvae of the European chafer, Amphimallon majale a,b b c b,d Gengping Zhu, ∗ Matthew J Petersen, Guoqing Liu and Daniel C Peck

Abstract

BACKGROUND: Manipulative studies of the behavioral response of soil-dwelling insects to insecticides and other antagonists are stymied by the difﬁculties of observing and interpreting interactions played out below ground. Six experiments were carried out using X-ray radiography to quantify the movement of the European chafer, Amphimallon majale (Razoumowsky), larvae in response to imidacloprid and how this was affected by host plant cues and cold temperature.

RESULTS: The movement of third instars was arrested in imidacloprid-treated soil at 0.6 ppm concentration. At 0.8 ppm, the arrestant effect of imidacloprid was stronger than the attraction cue posed by germinating≥ grass seed. There≥ was a less disruptive effect on dispersal distance in vertical versus lateral panels. In vertical panels, there was a less disruptive effect on downward movement under a cold temperature treatment that simulated overwintering conditions.

CONCLUSION: Larvae of A. majale do not remotely detect imidacloprid in the soil; they neither evade contact, nor are repelled after contact. Imidacloprid thereby acts as a contact arrestant to disrupt grub movement. This ﬁnding might help to explain the synergistic effect of imidacloprid in combination with other biological agents for white grub control, and its effects on grub overwintering behavior. c 2012 Society of Chemical Industry ! Keywords: white grub; European chafer; Amphimallon majale; movement arrestant; imidacloprid; sublethal effect; X-ray radiography

1 INTRODUCTION the outcome of insecticidal interventions against a suite of The interrelated factors that influence insecticidal activity in the root-feeding and soil-inhabiting insect pest complexes that are of soil include chemistry and formulation of the active ingredient, global agricultural and economic concern. A better understanding soil and climatic conditions, susceptibility and behavior of the of how insecticides influence the sublethal behavior of soil insects target species and insect developmental stage.1 Additionally, the will lead to improved soil insect management via better targeting, presence of both synthetic insecticides and biological control greater efficacy and novel application approaches. agents can significantly influence the behavior of taxa from Imidacloprid is a versatile insecticide that is used for systemic fo- major soil insect pest complexes. Understanding how insecticides liar and soil drench applications and as seed dressing, film coating, and neurotoxicants modify insect behaviors, including mating, pelleting and multilayer coating, allowing for relatively safe and 11,12 host-finding and feeding, may offer new avenues for effective efficacious protection of young plants against insect attack. pest management.2 Among soil-inhabiting arthropod pests, It has a relatively broad spectrum of activity and a long-lasting repellence behavior has been widely tested and is highly relevant effect against early insect developmental stages. Sublethal ef- to insect susceptibility and control product efficacy.3–10 Larvae fects after exposure to imidacloprid have been documented for of the wireworms (Coleoptera: Elateridae) Agriotes obscurus L. several insect pests. Disruption of host evaluation behavior was 13 and Limonius canus (LeConte) were repelled by tefluthrin-treated observed in Bemisia tabaci Gennadius (Hemiptera: Aleyrodidae). seeds after feeding for 15 min in soil microcosms.4,5 Larvae In choice tests, adult B. tabaci and alate morphs of Myzus per- ≤ of the weevil (Coleoptera: Curculionidae) Otiorhynchus sulcatus sicae (Sulzer) (Hemiptera: Aphididae) preferred untreated leaves (F.) were significantly deterred by bifenthrin absence of the host plant Picea abies L. Karst. (Pinales: Pinaceae).6 The mole crickets (Orthoptera: Gryllotalpidae) Scapteriscus borellii Giglio-Tos ∗ Correspondence to: Gengping Zhu, College of Environmental Science and and Scapteriscus vicinus Scudder modified their movement Engineering, Nankai University, Tianjin 300071, China. E-mail: [email protected] in response to Metarhizium anisopliae (Metchnikoff) Sorokin (Hypocreales: Clavacipitaceae) and Beauveria bassiana (Balsamo) a College of Environmental Science and Engineering, Nankai University, Tianjin, Vuillemin (Hypocreales: Clavacipitaceae), ultimately reducing China 7,8 their exposure to the entomopathogenic fungi. Among white b Department of Entomology, New York State Agricultural Experiment Station, grubs (Coleoptera: Scarabaeidae), the larvae of Popillia japonica Cornell University, Geneva, NY, USA Newman (Japanese beetle) avoid soil that contains mycelial c College of Life Sciences, Nankai University, Tianjin, China particles of M. anisopliae.9,10 This suggests that detection of and orientation to soil-borne antagonists are likely to influence d EntomoTech Fundamentals, Geneva, NY, USA

Pest Manag Sci (2012) www.soci.org c 2012 Society of Chemical Industry ! www.soci.org G Zhu et al. over systemically treated leaves.14,15 Antifeedant effects were documented in M. persicae.15–17 Reduced fecundity was documented in Diaphorina citri Kuwayama (Hemiptera: Psyllidae),18 Ceutorhynchus obstrictus (Marsham) (Coleoptera: Curculionidae)19 and Nilaparvata lugens (Stål) (Hemiptera: Delphacidae).20 With regard to white grubs, recent studies have shown an impact of imidacloprid on oviposition behavior,21 increased susceptibility to infection by entomopathogens22–25 and greater overwintering mortality of larvae.26 While the alteration of insect behavior after exposure to imidacloprid has been demonstrated for white grubs, repellence behavior has yet to be thoroughly examined under controlled conditions. Because of imidacloprid’s widespread use in turf environments, understanding and exploiting potential insect behavioral modifications, such as repellence behavior, could be important for management of soil insects, particularly those displaying low susceptibility to contact insecticides. Since its first North American detection in 1940,27 Amphimallon majale (Razoumowsky) (Coleoptera: Scarabaeidae) has become a dominant white grub pest of turfgrass in many areas of Figure 1. Diagrams of the PVC pipe and Plexiglas panel microcosms. New York State. Elsewhere it has caused problems in other (A) PVC pipe with treated soil (106 g, left) and untreated soil (144 g, right); 28 29 evasion before contact with the treated zone would be evidence of remote nursery crops and in winter wheat. The species occurs from repellence; evasion upon contact would be evidence of contact repellence. the northeast United States west to Michigan, with a disjunct (B) Plexiglas panel in a horizontal (left) and vertical (right) position; the five population now established in British Columbia.30 Compared circles indicate grub placement sites. with the three other major species of turf-infesting white grubs in the northeast United States, Asiatic garden beetle [Maladera castanea (Arrow)], Japanese beetle and oriental beetle [Anomala 2 EXPERIMENTAL METHODS 2.1 Source of insects orientalis (Waterhouse)], control of A. majale has proven to be difficult as it is less susceptible to insecticidal controls31 and to Third-instar A. majale collected from three field sites in New York State were used to conduct all laboratory experiments. Grubs environmental stressors such as drought.32 Behavior modifications were collected on 19 October 2009 in Fulton (Battle Island Golf due to insecticide exposure may help to explain the decreased Course, Oswego County), on 27 October 2009 in Victor (Parkview ability for control of A. majale. Fairways Golf Course, Ontario County) and on 9 November 2009 in To gain a better understanding of the nature of imidacloprid’s Rochester (Park Point at RIT, Monroe County). Larvae were held in sublethal effects on A. majale larval behavior, a series of soil, with a piece of sod from the collection site included as a food experiments was conducted with different combinations of source, at 4.4 ◦C until 24 h before the start of experiments when antagonists and agonists, namely soil-incorporated imidacloprid they were transferred to 23 ◦C. and germinating grass seed. Third-instar A. majale were chosen for the following studies not only because of the species’ pest status but also because the developmental stage can be 2.2 Radiographic imaging maintained in the laboratory for several months and is the Non-destructiveradiographicimagesofinsectswithinmicrocosms target of curative control in the field. The first objective of were made with a Faxitron X-ray cabinet (Model 43855B; Hewlett- this study was to determine (1) whether larvae could detect Packard, Palo Alto, CA). This system has been widely used in soil and orient away from imidacloprid, with two kinds of repellence insect behavior studies.7,9,10,32 The voltage was set at 74 kVp. being considered: evading imidacloprid before contact with the A digital X-ray scanner system (Model EZ400; NTB Eletronische treated zone, or evasion upon contact with the insecticides. The Geraete GmbH, Dickel, Germany) was used to capture, read and former would be evidence of remote repellence, while the latter manipulate the images. would be evidence of contact repellence. Subsequent objectives were to determine (2) what concentrations cause changes in 2.3 Experimental design larval behavior response, (3) how the response is modified Experiments were conducted in soil microcosms designed to by the presence of host plants and (4) whether imidacloprid provide an experimental chamber for studying grub movement causes disruption to grub overwintering behavior. The study in which the moisture, temperature and homogeneity of the and management of root-feeding insects is stymied by how soil could be controlled. Regardless of treatment and design, difficult it is to access, interpret and manipulate interactions all microcosms contained sterilized and screened (0.5 0.5 cm 33 × played out below ground. Monitoring soil insect behavior is mesh) sandy loam soil (89.4% sand, 8.8% silt, 1.8% clay; pH 7.5). difficult and usually requires significant alteration of the insect’s Two kinds of soil microcosm were used: polyvinyl chloride (PVC) habitat.34 Non-destructive techniques useful for studying insects pipes (Fig. 1A) to measure single grub response and Plexiglas withinsoilincluderadioactivetagging,35 acousticmethods,36 X-ray panels (Fig. 1B) to measure multiple grub response. Radiographs computedtomography(CT)37 andX-rayradiography.32 Herein,use were taken of all PVC pipes and Plexiglas panels at regular, is made of a non-destructive radiographic technique to determine predetermined time intervals. The location of each grub within whether imidacloprid influences the movement of A. majale larvae the microcosm was marked on the radiograph and recorded through soil microcosms. as distance relative to its original release point. A commercial wileyonlinelibrary.com/journal/ps c 2012 Society of Chemical Industry Pest Manag Sci (2012) ! Imidacloprid as a contact arrestant of A. majale www.soci.org formulation of imidacloprid (Merit 2F; Bayer Environmental of the 11 treatments of experiment 1 (176 pipes) were used. Science, Research Triangle Park, NC) was used to prepare different Grass seed (Kentucky-31 tall fescue, Festuca arundinacea Schreb) concentrations of soil microcosms. The label rate of Merit 2F for (0.2 g) was added in the end of the pipe as a food source and A. majale control is 17 mL 1000 ft 2 ( 443 g imidacloprid ha 1), positive attractant. The seed was added by settling it onto the − = − which approximates a concentration of 0.3 ppm for typical sandy pipe bottom before adding soil, or by adding it to the soil surface loam soil. before inserting the rubber stopper. Pipes were radiographed after infestation (2, 6, 26, 50, 74, 98, 144 and 266 h) and then 2.3.1 Single-grub tests with PVC pipe microcosms destructively sampled to measure grub mortality. It was predicted The design of the PVC pipe microcosms was based on Villani that the results would vary from experiment 1 insofar as insect et al.9 Each pipe (diameter 2.5 cm, length 33.0 cm) (Fig. 1A) held movement would be arrested after contact with the germinating a uniform amount of soil (250 g). To introduce grubs, each pipe seed. had a hole (0.8 cm diameter) on one edge, midway between the For experiment 3, protocols were designed to measure how ends. The other edge of the pipe had ten smaller-diameter holes imidaclopridreducedthegrub’sresponsetograssseedunderano- (0.4 cm) evenly distributed along its length to serve as markers to choicesituation.Atotalof18replicatesoftwotreatments(36pipes) measure grub location relative to the release point. Holes were were used. Both treatments had grass seed (0.5 g) on one side, sealed on the outside with duct tape. Each end of the pipe was with or without imidacloprid (0.8 ppm). The rate of imidacloprid sealed with a rubber stopper that filled an equal amount of the was chosen on the basis of the results of experiments 1 and 2 as pipe’s length (2 cm). Dry soil was added a certain amount of being just above the threshold of grub response. The opposite side imidacloprid solution (for treated soil) or tap water (for untreated was untreated and without seed. Pipes were radiographed after soil) to raise the soil moisture to 12.5% before adding to the pipe. infestation(2,6,26,50,74,98,144and266 h)andthendestructively Using a plastic funnel, the pipe was filled through two steps. Firstly, sampled to measure grub mortality. It was predicted that, without treated soil (106 g) was placed into one side and compressed to imidacloprid, grubs would displace towards the germinating seed, 12 cm with a wooden dowel. The treated soil column thereby while with imidacloprid the displacement towards the seed would extended from the rubber stopper to 2.0 cm distance from the be reduced. grub insertion hole, or midline. Secondly, untreated soil (144 g) was added and compressed against the treated soil, leaving 2 cm 2.3.2 Multiple-grub tests with Plexiglas panel microcosms to accommodate the rubber stopper. As a result, the grub was The design of the Plexiglas panel microcosms was based on Villani inserted into untreated soil and 2 cm away from treated soil. To and Wright32 and Villani and Nyrop.38 Two Plexiglas sheets were control for compaction bias, the order in which each side of the separated and walled by a wooden frame (3.5 25 30 cm) × × pipe was filled with soil treatments was alternated. Once prepared, (Fig. 1B) to hold a uniform amount of soil (3400 g). One pane of pipes were stored horizontally under controlled conditions (28 ◦C, the box was removable for grub placement. Soil moisture was 80% RH, photoperiod 12 : 12 light : dark). After 24 h, an individual raised to 4% for both treated and untreated soil. The panel was put grub was introduced at the midline of each pipe by placing its in a vertical position for filling with soil according to the specific head in contact with untreated soil through the insertion hole. details of experiments 4, 5 and 6. Once prepared, panels were After release, each grub was observed and replaced if it could not maintained horizontally under controlled conditions (28 ◦C, 80% descend into soil within 10 min. To prevent water loss, the insertion RH). After 24 h, one pane was removed to make five holes for hole was covered with masking tape before and after infestation. grub infestation according to each experiment. After the grub was The pipes were then assembled in sets of six (experiments 1 and placed in each hole, the Plexiglas was reaffixed to the panel. 3) or eight (experiment 2), alternating the left–right orientation For experiment 4, protocols were designed to study movement of the treated and untreated sides, placed on a Plexiglas plate, of multiple grubs in response to imidacloprid over time. Treated and secured with duct tape for radiography. Plates were stored soil(1700 g)wasaddedtothebottomofthepanelandcompressed horizontally and at random compass directions in the growth to a height of 15 cm. Untreated soil (1700 g) was then added and chamber throughout the course of experiment. Pipes were only compressed to the top of the panel for an additional layer of 15 cm removed from controlled conditions to be radiographed. (Fig. 1B, horizontal position). Grass seed (4 g) was mixed evenly For experiment 1, protocols were designed to ascertain the throughout the volume of treated and untreated soil to provide threshold concentration of soil-incorporated imidacloprid for the grub with a food source over the course of the experiment. grub response, and to examine changes in that response over The order of filling (treated versus untreated) was alternated to time. A total of 12 replicates of 11 treatments (132 pipes) reduce the soil compaction bias. A total of four replicates of were used in the experiment. One-half of each pipe was filled three treatments [0 (untreated control), 0.4 and 2.0 ppm] (12 total with untreated soil, while the other half was filled with one of panels) were used. The concentrations were chosen to represent the 11 imidacloprid-incorporated soil treatments [0.0 (untreated a below- and above-threshold concentration respectively. Panels control), 0.1, 0.2, 0.4, 0.5, 0.6, 0.8, 1.0, 2.0, 5.0 and 10.0 ppm]. were maintained in a horizontal position in the growth chamber These concentrations represented a realistic range of imidacloprid (28 ◦C, 80% RH) throughout the course of the experiment. Five concentrations in the environment. No grass seed was included. larvae were introduced along the midline in each panel (Fig. 1B, Pipes were radiographed after infestation (2, 6, 26, 50, 74 and horizontal position). All panels were radiographed at uniform time 98 h). It was predicted that, below some threshold concentration, intervals after infestation (24, 72 and 144 h). It was predicted that grubs would displace evenly to the untreated and treated sides, grub movement would be reduced at 2.0 ppm, but not at 0.4 ppm. and, above some threshold concentration, grubs would displace For experiment 5, protocols were designed to study the effect of preferentially to the untreated side. imidacloprid on vertical dispersal under overwintering conditions For experiment 2, protocols were designed to build on the simulated by cold temperature. Untreated soil (2720 g) was added results of experiment 1 by including grass seed and making to the bottom of the panel and compressed to a height of 24 cm. observations over a longer time series. A total of 16 replicates Treated soil (680 g) was then added and compressed to the top of

Pest Manag Sci (2012) c 2012 Society of Chemical Industry wileyonlinelibrary.com/journal/ps ! www.soci.org G Zhu et al.

Figure 2. Percentage of A. majale larvae displacing to the untreated side away from a range of imidacloprid concentrations at 6, 26, 50, 74 and 98 h after introduction. An asterisk (*) indicates significant displacement to the treated side with respect to an even distribution (50%–50%). the panel for an additional layer of 6 cm (Fig. 1B, vertical position). assigned as the treated side. An ANOVA was also used to determine Given the short duration of the experiment (48 h), no grass seed whether insecticide treatment affected the mean location of grubs wasused.Atotaloffourreplicatesofthreetreatments[0(untreated relative to the release point in experiment 3. Orthogonal contrasts control), 0.4, and 2.0 ppm] (12 total panels) were used. Five grubs were used to make a separate comparison for each time period. were introduced along a horizon 3 cm from the top in each panel This analytical approach is similar to the profile analysis described (Fig. 1B, vertical position). After infestation, panels were kept in in Villani et al.9 Owing to the small number of replicates, ANOVA a vertical position in a growth chamber under cold temperature was not used for experiments 1 and 2. conditions (4.4 ◦C, 70% RH) during the course of the experiment. In panel experiments, the distribution of grubs at each time All panels were radiographed at uniform time intervals after grub interval was displayed visually as a box plot constructed using placement (6, 24 and 48 h). It was predicted that downward grub SigmaPlot 11.2 software.39 This approach is similar to the protocol movement would be reduced at low (0.4 ppm) but not high described in Villani and Wright32 and Villani and Nyrop38 and (2.0 ppm) concentrations. is useful for revealing the rate of movement and the change For experiment 6, protocols were designed to validate results in population distribution over time. ANOVA of the mean grub of the previous experiment and further elaborate the role of position relative to the release line was used to test the effect of temperature on vertical dispersal. Panels were prepared as in imidacloprid on grub dispersal length in the panel in experiments experiment 5, but all had an untreated bottom zone and a top 4, 5 and 6. In experiments 4 and 5, ANOVA comparisons were zone of imidacloprid-incorporated soil (2.0 ppm). A total of four made between the untreated and each two imidacloprid-treated replicates of three environmental conditions (held vertically at groups. In experiment 6, the comparisons were made between the 4.4 ◦C, vertically at 28 ◦C and horizontally at 28 ◦C) (12 total panels) two vertical treatments and between the two 28 ◦C treatments. An were used. All panels were radiographed at uniform time intervals effect of treatment on the distribution of grubs from the center axis ( 2 cm) and beyond was also compared using a chi-square test after grub placement (6, 24 and 48 h). Vertical treatments were ≤ compared in order to test the effect of cold temperature on in experiment 4 (Fig. 1B, horizontal position). All the chi-square 40 the disruption of imidacloprid on downward movement. It was analyses were conducted in Microsoft Excel 2007, while the predicted that downward dispersal length would be longer under ANOVA analyses were conducted using JMP Statistical Discovery 41 the cold temperature conditions. The two 28 ◦C treatments were Software. compared in order to test the effect of pane orientation on grub dispersal. Third-instar grubs used in the experiment were collected before winter; onset of cold temperature during this 3 RESULTS developmental stage elicits a positive geotropic response where 3.1 Single-grub tests with PVC pipe microcosms larvae tend to move downwards in the soil column.38 It was 3.1.1 Experiment 1 predicted that dispersal from the release point would be longer in Grubs in the untreated control pipes showed no significant the vertical position than in the lateral position. difference in displacement between sides at all five time intervals, confirming the lack of a directional bias inherent to the experimental protocols (Fig. 2). In the treated pipes, the 2.4 Statistical analysis proportion of grubs displacing to the untreated side was <50% Chi-square tests were used to test for differences in the proportion for all concentrations. There was significant displacement to the of grubs distributed with respect to treatment and time. In treated side for all five time intervals at 0.2 and 10.0 ppm, for four tube experiments, the proportion of grubs in the untreated side time intervals at 5.0 ppm, for two time intervals at 0.5 ppm and was compared against a null expectation of even distribution for one time interval at 1.0 ppm (chi-square, df 1, P 0.05). = ≤ between sides (50%–50%) for each treatment. This analytical The disruption of displacement to the end of the treated approach is similar to the profile analysis described in Fry et al.10 side increased with imidacloprid concentration (Fig. 3). At lower A significantly higher proportion of grubs first displacing to the concentrations, in particular 0, 0.1, 0.4 and 0.5 ppm, grubs untreated side was considered evidence for remote repellence displaced along the entire length of the pipe, which was consistent (Fig. 1A). Alternatively, a significantly higher proportion of grubs with an uninterrupted search for food sources (see experiment 2). first displacing to the treated side, and then displacing to the At concentrations of 0.6 ppm, displacement was interrupted ≥ untreated side, was considered evidence for contact repellence. after contact with treated soil, and no grubs displaced to the For the untreated control, the side filled second (106 g) was end ( 6 position) of the treated side. Changes in the pattern of − wileyonlinelibrary.com/journal/ps c 2012 Society of Chemical Industry Pest Manag Sci (2012) ! Imidacloprid as a contact arrestant of A. majale www.soci.org

Figure 3. Response of A. majale larvae to a range of imidacloprid concentrations and untreated control. Graphs indicate shifts in grub proportion (y-axis) in microcosms (x-axis) over time (z-axis) (x-axis: left side treated side, right side untreated; the 6 to 6 axis represents the distance grub moved from the treated to the untreated side). = = − grub distribution from 0.5, 0.6 to 0.8 ppm highlighted the effect of proportion of grubs displacing to the untreated side was <62% increasing imidacloprid concentration on grub distribution, from for all concentrations and time intervals. There was significant no apparent effect to defined peaks at the positions of 4 and 2 displacement to the treated side for six time intervals at 2.0 ppm, − − respectively (Fig. 3). The relatively low proportion of grubs that for four time intervals at 0.5 ppm, for two time intervals at 1.0 displaced to the untreated side might be due to the unexpected and 10.0 ppm and for one time interval at 0.6 and 5.0 ppm (chi- arrestant effect of imidacloprid-treated soil (Fig. 2). square, df 1, P 0.05). The proportion of grubs displaced to the = ≤ untreated side tended to decline with increasing imidacloprid 3.1.2 Experiment 2 concentration: 25–60% over the seven time intervals at low Grub mortality was 25% in the untreated pipes, and ranged from concentrations (0–0.4 ppm), 10–60% at middle concentrations 19 to 50% across different concentrations in the treated pipes. In (0.5–1.0 ppm) and 5–40% at high concentrations (2.0–10.0 ppm). the untreated pipes there were no significant differences in grub The displacement of grubs was significantly inhibited by the displacement between sides at all seven time intervals (Fig. 4). The presence of grass seed. In the untreated control pipes, grubs

Pest Manag Sci (2012) c 2012 Society of Chemical Industry wileyonlinelibrary.com/journal/ps ! www.soci.org G Zhu et al.

Figure 4. Percentage of A. majale larvae displacing to the untreated side away from a range of imidacloprid concentrations at 6, 26, 50, 98, 146, 194 and 266 h after introduction. An asterisk (*) indicates significant displacement to the treated side compared with an even distribution (50%–50%). displaced to pipe ends within 50 h and stayed there until the At 0.4 ppm, grubs were arrested at the release point at 24 h, end of the experiment (Fig. 5), in contrast to the untreated pipes followed by some displacement into the untreated zone at 72 h, without food sources in experiment 1 (Fig. 3). At 98 h, for instance, and displaced in both directions at 144 h. At 2.0 ppm, grubs did 94% of grubs were at the 6 and 6 positions when food was not displace beyond the release points, and most were arrested − present (Fig. 5), but only 17% were at those positions when on the axis at 72 and 144 h. Only three grubs escaped the axis at food was absent (Fig. 3). Significant shifts in the pattern of grub 24 h and displaced into the untreated zone. displacement were found with respect to both concentration and time. At 0–0.5 ppm, two crests appeared at the 6 and 6 positions. − 3.2.2 Experiment 5 At 50 h, all grubs had displaced to the end of the pipe in untreated pipes, but 45% had displaced in the pipes at 0.1–0.5 ppm. At In the untreated panels, 85% of grubs dispersed downwards ≤ low concentrations, displacement to the treated end was delayed from the release point. In the treated panels, the grubs also and arrested by the presence of imidacloprid in the soil. There tended to move downwards (Fig. 7B). The box plot column were also two crests at 0.6–10.0 ppm, but these appeared at representing majority grub population was shorter in the 0.4 ppm the 2 and 6 positions after 98 h. After 266 h, no grubs had and 2.0 ppm treated panels than in the untreated control at − displaced through the imidacloprid-treated soil to the pipe end, each time interval. The shorter column and its position relative and none had turned back to displace to the untreated side. to the release point indicates the decrease in grub dispersal Displacement to the untreated side was reduced after contacting length. As concentration increased from 0 (untreated control) to the imidacloprid-treated soil. 0.4 and 2.0 ppm, mean dispersal length shortened. There was a significant effect of imidacloprid on mean position relative to the release line at 2.0 ppm (ANOVA, F 4.49, df 1, 95, = = 3.1.3 Experiment 3 P < 0.05), but not at 0.4 ppm. Mean dispersal length in the Grub mortality was 17 and 25% after 266 h in the untreated untreated panels was 13.7 7.29, 14.5 7.88 and 15.1 8.21 cm ± ± ± (Fig. 6A) and treated (Fig. 6B) pipes respectively. In the untreated at 6, 24 and 48 h respectively, while, in the treated panel, mean pipes there was no significant effect of seed on grub displacement dispersal length decreased to 12.4 5.08, 14.1 6.64 and ± ± between sides at 2, 6, 26, 50, 74 and 144 h, but there was at 98 14.8 6.86 cm in the 0.4 ppm treatment and to 11.4 3.91, ± ± and 266 h (chi-square, df 1, P 0.05). The proportion of grubs 12.6 4.93 and 13.5 5.46 cm in the 2.0 ppm treatment. = ≤ ± ± displacing towards seed was 66.7–83.3% over the last five time Under low-temperature conditions, downward dispersal was still intervals. In the treated pipes there was a significant difference disrupted by imidacloprid, but to a lesser degree than lateral in distribution between sides at 98, 144 and 266 h (chi-square, dispersal (experiment 4). df 1, P 0.05), but not at earlier time intervals. Unlike the = ≤ untreated pipes where grubs displaced to seed, displacement in 3.2.3 Experiment 6 the treated pipes was arrested at the imidacloprid-treated soil (Fig. 6). Imidacloprid treatments had a significant effect on grub For the two vertical treatments, at 28 ◦C, by 48 h only 15% of distribution:significancewasobservedat50,98and266 h(ANOVA, grubs escaped the imidacloprid-treated zone, while, at 4.4 ◦C, 45% 50 h: F 12.09, df 1, 255, P 0.0005; 98 h: F 11.25, df 1, escaped to the untreated zone. The mean downward dispersal = = = = = length at 4.4 ◦C was 6.1 6.64, 6.4 6.96 and 6.6 7.02 cm at 6, 255, P 0.0009; 266 h: F 8.52, df 1, 255, P 0.0038). ± ± ± = = = = 24 and 48 h respectively. In contrast, at 28 ◦C the corresponding values decreased to 4.4 6.19, 4.8 7.19 and 4.2 6.42 cm. ± ± ± 3.2 Multiple-grub tests with Plexiglas panel microcosms There was a marginal significance effect of imidacloprid on the 3.2.1 Experiment 4 grub mean position relative to the release line between the two In the untreated panels, grubs displaced in both directions at treatments (ANOVA, F 3.27, df 1, 95, P 0.073). Therefore, = = = 24 h, and continued to displace at 72 and 144 h (Fig. 7A), with no imidacloprid had a less disruptive effect on downward dispersal significant difference in displacement between sides for all three under the cold temperature treatment. time intervals. There was a significant effect of treatments on grub For the two 28 ◦C treatments, the difference between the distribution at 72 and 144 h (chi-square, df 1, P 0.05). The horizontal and vertical treatment of grub distribution was small = ≤ mean location of grub position relative to the release line between (Fig. 8): 20 and 15% of grubs escaped the imidacloprid-treated the two treatments and control was not significantly different. zone respectively. However, compared with the grubs distributed wileyonlinelibrary.com/journal/ps c 2012 Society of Chemical Industry Pest Manag Sci (2012) ! Imidacloprid as a contact arrestant of A. majale www.soci.org

Figure 5. Response of A. majale larvae to a range of imidacloprid concentrations and untreated control. Grass seeds added in both ends of the pipe serve as food source for the grubs. Graphs indicate shifts in grub proportion (y-axis) in microcosms (x-axis) over time (z-axis) (x-axis: left side treated side, right side untreated; the 6 to 6 axis represents the distance grub moved from the treated to the untreated side). The control and the= 0.1–0.5 ppm concentration= treatments show− a similar pattern, with two crests at positions 6 and 6 after 98 h. The 0.6–10.0 ppm concentration treatments show a similar pattern, with crests at positions 2 and 6 after 98 h. − − just around the release point in the horizontal treatment, several 4 DISCUSSION grubs escaped the release point and moved downwards in the The behavioral response of insect pests to antagonists can act vertical treatment. The mean dispersal distance in the horizontal either to enhance or to diminish the efficacy of insecticidal treatment was 2.4 1.86, 2.4 1.90 and 2.5 2.08 cm at 6, treatments by increasing or decreasing, respectively, their contact ± ± ± 24 and 48 h respectively. In contrast, in the vertical treatment with the treatment. The effect of chemicals on insect behavior the corresponding values increased to 4.4 6.19, 4.8 7.19 and can be classified as arrestant, stimulant (locomotor, feeding, ± ± 4.2 6.42 cm.Themeanlocationofgrubsrelativetothereleaseline ovipositional), attractant, repellent or deterrent.42,43 The observed ± between the two treatments was statistically significant (ANOVA, grub morality was constant among treatments in experiments 2 F 8.66, df 1, 95, P < 0.05). Therefore, imidacloprid had a less and 3, suggesting that the movement response was not caused = = disruptive effect on dispersal length in the vertical position than by the toxicity of imidacloprid. The present results indicate that in the lateral position. A. majale larvae cannot remotely detect imidacloprid in the soil at

Pest Manag Sci (2012) c 2012 Society of Chemical Industry wileyonlinelibrary.com/journal/ps ! www.soci.org G Zhu et al.

Figure 6. Effect of imidacloprid on A. majale larvae searching for food sources in untreated soil (A) and imidacloprid-treated (B) soil at 0.8 ppm. Graphs indicate shifts in grub proportion (y-axis) in microcosms (x-axis) over time (z-axis) (x-axis: left side with seed only, or seed and imidacloprid, right side untreated; the 6 to 6 axis represents the distance grub moved from the left to the right side). = = −

Figure 7. Box plot of grub population distribution in Plexiglas panels in the horizontal (A) and the vertical (B) position. (A) Grub population in the treated (upper) and untreated (down) side at 24, 72 and 144 h after grub placement. (B) Grub population in vertical microcosms at 6, 24 and 48 h after grub placement in the untreated control and the surface application of imidacloprid-treated soil at 0.4 and 2.0 ppm concentration.

prolonged contact with the chemical indicates that imidacloprid was acting as an arrestant. The impact of imidacloprid on insect behavior has been little studied; however, it has been shown to elicit a behavioral response in select taxa. Female B. tabaci did not detect imidacloprid in choice experiments using surface-treated leaf discs.14 Host plant evaluation behavior of B. tabaci was also affected significantly by imidacloprid when ingested, although it was not similarly affected through direct contact.13 The lack of significant differences in directional movement indicates that larvae of A. majale show no preference for soil not treated or treated with imidacloprid. Although it is not possible to define a threshold concentration from this work, it is apparent that concentrations of 0.6 ppm ≥ arrest movement as early as 6 h after contact. In applying these results to field settings, the efficacy of imidacloprid for curative Figure 8. Box plot of grub population distribution in microcosms at 6, 24 control of third-instar white grub should be enhanced by insect and 48 h after grub placement with the microcosms held in three different behavior, and any apparent lack of control cannot be attributed to conditions (horizontally at 28 ◦C, vertically at 28 ◦C and vertically at 4.4 ◦C). repellency. Furthermore, field application of imidacloprid would not expel A. majale grubs from treated areas and into surrounding any of the tested concentrations, nor do larvae move away from untreated areas. imidacloprid after contacting the insecticide, even when present The attraction of A. majale to available food sources could act at high concentrations. Furthermore, lack of grub movement after to draw larvae into the presence of imidacloprid-treated soils, wileyonlinelibrary.com/journal/ps c 2012 Society of Chemical Industry Pest Manag Sci (2012) ! Imidacloprid as a contact arrestant of A. majale www.soci.org thus facilitating greater utility of the product. Miller and Strickler44 by sublethal rates of neonicotinoids.22–25 Koppenhofer¨ et al.24 suggested that host plant location and acceptance by insects is the suggest that the major factor responsible for synergistic interac- result of a shift in balance between negative and positive internal tions between imidacloprid and entomopathogenic nematodes and external inputs. The shifting balance analogy is additionally may be the general disruption of normal nerve function that applicable to published patterns of P. japonica grubs searching for causes drastically reduced activity of the grubs, thus increasing germinatingseedasafoodsource38 andthebehaviorpatternsofA. their exposure to the infective juvenile nematodes. The present majale displayed here. The internal grub status interacting with the result is in line with this theory and provides empirical evidence external factors determines their response to the various external that, when exposed to imidacloprid, even at low rates, A. ma- stimuli present in the environment, or as shown here within soil jale movement is greatly reduced. This pattern may explain the microcosms. Within soil microcosms, larvae were clearly shown to synergistic interaction between Heterorhabditis bacteriophera and be attracted to available food sources, searching the soil column imidacloprid for control of A. majale.25 until coming into contact with grass seed. During experiments, some grubs chose the ‘right’ direction into the untreated side, resulting in their successfully reaching the food source. Other 5 CONCLUSIONS grubs chose the ‘wrong’ direction into the imidacloprid-treated Amphimallon majale larvae do not remotely detect imidacloprid side, resulting in a reduction in movement and the cessation of in the soil. They neither evade contact, nor are repelled after search. By incorporating imidacloprid, it was revealed that the contact. Imidacloprid functions as a contact arrestant to disrupt elicitors arresting movement dominated those that promoted the grub movement. Movement is arrested after contact with search for food. Further understanding the behavioral response of concentrations of 0.6 ppm. This finding might help to explain ≥ white grubs to agonists and antagonists might open opportunities the synergistic effect of imidacloprid in combination with other to use their properties to retain grubs in, or attract grubs to, treated biological agents for the control of late-instar white grubs, and the zones in some field conditions. disruption of grub overwintering behavior by imidacloprid. The capacity of imidacloprid to impact upon the overwintering behavior of A. majale has been reported previously, but direct emprical evidence is lacking. Grewal et al.26 proposed that ACKNOWLEDGEMENTS imidacloprid reduces the survival of overwintering P. japonica The authors thank Dan Olmstead for his valuable assistance with larvae (third instar) by altering their normal overwintering grub collections, soil preparations and statistical analyses, Akiko behavior. The result was attained, however, by comparing the Seto for technical support, Anuar Morales for developing early mortality rate of third-instar grubs 15 days after treatment in protocols, Masanori Seto for helpful comments and suggestions the fall and 210 days after treatment in the next spring. Under and Jim Opela, Rich Belowski and David Damaske for access to controlled laboratory conditions and through the use of non- grub collection sites. Major support was provided by the China destructive X-ray technology, direct observations on vertical Scholarship Council through a fellowship awarded to GZ and GL, displacement were made. Under the conditions of the present and by the Federal Formula Funds projects NYG-621404 and NYG- study, the application of imidacloprid did not significantly deter 621819 awarded to DCP. This paper is dedicated to the late Mike the downward movement of A. majale larvae after short-term Villani (1953-2001) and Haruo Tashiro (1917-2009). exposure, although within the timeframe of these experiments it was apparent that movement after treatment was somewhat impaired. The third instars used in this experiment have an innate REFERENCES positive geotropic response and tend to move downwards under 1 Harris CR, Factors influencing the effectiveness of soil insecticides. cold temperature conditions.38 As such, the sluggish movement Annu Rev Entomol 17:177–198 (1972). already stimulated by low temperature might be magnified by 2 Haynes KF, Sublethal effects of neurotoxic insecticides on insect behavior. Annu Rev Entomol 33:149–168 (1988). the presence of imidacloprid, and lead to a longer residence time 3 van Herk WG, Vernon RS, Moffat C and Harding C, Response of the and potential exposure to adverse environmental conditions. As Pacific Coast wireworm, Limonius canus, and the dusky wireworm, these microcosm experiments recorded response after short-term Agriotes obscures (Coleoptera: Elateridae), to insecticide-treated exposure,notmorethan30 minbeforeholdingtheminthegrowth wheat seeds in a soil bioassay. Environ Entomol 89:7–19 (2008). chamber, the extended exposure times under natural conditions 4 van Herk WG and Vernon RS, Soil bioassay for studying behavioral responses of wireworms (Coleoptera: Elateridae) to insecticide- might elicit astronger responseeffect.Additionally,whether larvae treated wheat seed. Environ Entomol 36:1441–1449 (2007). still reserve the capacity to move after a more prolonged time of 5 van Herk WG, Vernon RS and Roitberg BD, Repellency of a wireworm, exposure to imidacloprid, and could still be released by the stimuli Agriotes obscurus (Coleoptera: Elateridae), on exposure to synthetic of increasing temperature, is a subject of needed future study. insecticides in a soil-less bioassay. Environ Entomol 37:534–545 (2008). Like other neonicotinoids, imidacloprid acts selectively on the 6 Kepler RM and Bruck DJ, Examination of the interaction between insect central nervous system as an agonist of the post-synaptic the black vine weevil (Coleoptera: Curculionidae) and an nicotinic acetylcholine receptors.45 The resulting processes may entomopathogenic fungus reveals a new tritrophic interaction. include insect behavior modification. Such changes in behavior Environ Entomol 35:1021–1029 (2006). may lead to a reduction in ecological fitness and increased expo- 7 Villani MG, Allee LL, Preston-Wilsey LM, Consolie N, Xia Y and Brandenburg RL, Use of radiography and tunnel castings for sure to biotic and abiotic mortality factors. Modifications of white observing mole cricket (Orthoptera: Gryllotalpidae) behavior in grub behavior, such as those shown here, could make them more soil. Am Entomol 48:42–50 (2002). susceptible to biological control agents. Coupling control mea- 8 Thompson SR and Brandenburg RL, Tunneling responses of mole sures in this way could provide a new approach for reduced-risk crickets (Orthoptera: Gryllotalpidae) to the entomopathogenic fungus, Beauveria bassiana. Environ Entomol 34:140–147 (2005). white grub management in turf. One example of the potential 9 Villani MG, Krueger SR, Schroeder PC, Consolie F, Consolie NH, synergy between control measures is increased susceptibility of Preston-Wilsey LM, et al, Soil application effects of Metarhizium third instars to entomopathogenic nematodes when challenged anisopliae on Japanese beetle (Coleoptera: Scarabaeidae) behavior

Pest Manag Sci (2012) c 2012 Society of Chemical Industry wileyonlinelibrary.com/journal/ps ! www.soci.org G Zhu et al.

and survival in turfgrass microcosms. Environ Entomol 23:502–513 26 Grewal PS, Power KT and Shetlar DJ, Neonicotinoid insecticides alter (1994). diapause behavior and survival of overwintering white grubs 10 Fry RC, Fergusson-Kolmes LA, Kolmes SA and Villani MG, Radiographic (Coleoptera: Scarabaeidae). Pest Manag Sci 57:852–857 (2001). study of the response of Japanese beetle larvae (Coleoptera: 27 Tashiro H, The European chafer, a continuing lawn problem in New Scarabaeidae) to soil-incorporated mycelial particles of Metarhizium York. New York’s Food and Life Sci Bull 16:1–6 (1972). anisopliae (Deuteromycetes). J New York Entomol Soc 105:113–120 28 Smitley DR, Incidence of Popillia japonica (Coleoptera: Scarabaeidae) (1997). and other scarab larvae in nursery fields. J Econ Entomol 11 Elbert A, Becker B, Harwtig J and Erdelen C, Imidacloprid – a new 89:1262–1266 (1996). systemic insecticide. Pflanzenschutz-Nachr Bayer (German edn) 29 Vittum P, Villani MG and Tashiro H, Turfgrass Insects of the United States 44:113–136 (1991). and Canada. Cornell University Press, Ithaca, NY (1999). 12 Elbert A, Haas M, Springer B, Thielert W and Nauen R, Applied 30 Jackson TA and Klein MG, Scarabs as pests: a continuing problem. aspects of neonicotinoid uses in crop protection. Pest Manag Sci Coleopterists Soc Monograph 5:102–119 (2006). 64:1099–1105 (2008). 31 Morales-Rodriguez A, Ospina A and Peck DC, Variation in the 13 Isaacs R, Cahill M and Byrne D, Host plant evaluation behaviour of laboratory susceptibility of turf-infesting white grubs (Coleoptera: Bemisia tabaci and its modification by external or internal uptake of Scarabaeidae) to biological, biorational and chemical control imidacloprid. Physiol Entomol 24:101–108 (1999). products. Pest Manag Sci 66:90–99 (2010). 14 Nauen R, Koob B and Elbert A, Antifeedant effects of sublethal dosages 32 Villani MG and Wright RJ, Use of radiography in behavioral studies of of imidacloprid on Bemisia tabaci. Entomol Exp Appl 88:287–293 turfgrass infesting scarab grub species (Coleoptera: Scarabaeidae). (1998). Bull Entomol Soc Am 34:132–144 (1988). 15 Nauen R, Behaviour modifying effects of low systemic concentrations 33 Hunter MD, Out of sight, out of mind: the impacts of root-feeding of imidacloprid on Myzus persicae with special reference to an insects in natural and managed systems. Agric For Entomol 3:3–9 antifeeding response. Pestic Sci 44:145–153 (1995). (2001). 16 Devine GJ, Harling ZK, Scarr AW and Devonshire AL, Lethal and 34 Villani MG and Wright RJ, Environmental influences on soil sublethal effects of imidacloprid on nicotine-tolerant Myzus macroarthropod behavior in agricultural systems. Ann Rev Entomol nicotianae and Myzus persicae. Pestic Sci 48:57–62 (1996). 35:249–269 (1990). 17 Mowry TM and Ophus JD, Effects of sub-lethal imidacloprid levels on 35 Fredericksen CF and Lilly JH, Measuring wireworm reactions to soil potato leafroll virus transmission by Myzus persicae. Entomol Exp insecticides by tagging with radioactive cobalt. J Econ Entomol Appl 103:249–255 (2002). 48:438–442 (1955). 18 Boina DR, Onagbola EO, Salyani M and Stelinski LL, Antifeedant and 36 Mankin RW, Brandhorst-Hubbard J, Flanders KL, Zhang M, Crocker RL, sublethal effects of imidacloprid on Asian citrus psyllid, Diaphorina Lapointe SL, et al, Eavesdropping on insects hidden in soil and citri. Pest Manag Sci 65:870–877 (2009). interior structures of plants. J Econ Entomol 93:1173–1182 (2000). 19 Dosdall LM, Responses of the cabbage seedpod weevil, Ceutorhynchus 37 Harrison RD, Gardner WA, Tollner WE and Kinard DJ, X-ray computed obstrictus (Marsham) (Coleoptera: Curculionidae), to seed tomography studies of the burrowing behavior of fourth- treatments of canola (Brassica napus L.) with the neonicotinoid instar Pecan weevil (Coleoptera: Curculionidae). J Econ Entomol compounds clothianidin and imidacloprid. Pest Manag Sci 86:1714–1719 (1993). 65:1329–1336 (2009). 38 Villani MG and Nyrop JP, Age dependent movement patterns of 20 Bao H, Liu S, Gu J, Wang X, Liang X and Liu Z, Sublethal effects of Japanese beetle and European chafer grubs in soil–turfgrass four insecticides on the reproduction and wing formation of brown microcosms. Environ Entomol 20:241–251 (1991). planthopper, Nilaparvatalugens. PestManagSci 65:170–174 (2009). 39 SigmaPlot for Windows. Version 11.2. SPSS, Inc., Chicago, IL (2009). 21 George J, Redmond CT, Royalty RN and Potter DA, Residual effects 40 Microsoft Office Excel 2007. Microsoft Corporation, San Francisco, CA of imidacloprid on Japanese beetle (Coleoptera: Scarabaeidae) (2007). oviposition, egg hatch, and larval viability in turfgrass. J Econ 41 JMP: The Statistical Discovery Software. Version 7. SAS Institute, Cary, Entomol 100:431–439 (2007). NC (2003). 22 Koppenhofer¨ AM and Kaya HK, Synergism of imidacloprid and an 42 Dethier VG, Barton BL and Smith CN, The designation of chemicals entomopathogenic nematode: a novel approach to white grub in terms of the responses they elicit from insects. J Econ Entomol (Coleoptera: Scarabaeidae) control in turfgrass. J Econ Entomol 53:134–136 (1960). 91:618–623 (1998). 43 Miller JR, Siegert PY, Amimo FA and Walker ED, Designation of 23 Koppenhofer¨ AM, Brown I, Gaugler R, Grewal PS, Kaya HK and chemicals in terms of the locomotor responses they elicit from Klein MG, Synergism of entomopathogenic nematodes and insects: an update of Dethier, et al. (1960). J Econ Entomol imidacloprid against white grubs: greenhouse and field evaluation. 102:2056–2060 (2009). Biol Cont 19:245–251 (2000). 44 Miller JR and Strickler KL, Finding and accepting host plants, in 24 Koppenhofer¨ AM, Grewal PS and Kaya HK, Synergism of imidacloprid Chemical Ecology of Insects, ed. by Bell WJ and Carde RT. Chapman and entomopathogenic nematodes against white grubs: the and Hall, New York, NY, pp. 127–157 (1984). mechanism. Entomol Exp Appl 94:283–293 (2000). 45 Jeschke P and Nauen R, Neonicotinoids – from zero to hero in 25 Morales-Rodriguez A and Peck DC, Synergies between biological and insecticide chemistry. Pest Manag Sci 64:1084–1098 (2008). neonicotinoidinsecticidesforthecurativecontrolofthewhitegrubs Amphimallon majale and Popillia japonica. Biol Cont 51:169–180 (2009).

wileyonlinelibrary.com/journal/ps c 2012 Society of Chemical Industry Pest Manag Sci (2012) !