Toxicity of Insecticides on Various Life Stages of Two Tortricid Pests of Cranberries and on a Non-Target Predator

Laboratory and extended laboratory bioassays were conducted to determine the residual toxicities of various insecticides against two key pests of cranberries, Sparganothis sulfureana and Choristoneura parallela (Lepidoptera: Tortricidae), and their non-target effects on the predatory Orius insidiosus (Hemiptera: Anthocoridae). The effects of nine insecticides with different modes of action on S. sulfureana and Ch. parallela eggs, larvae, and adults were tested in the laboratory, while the efficacy of a post-bloom application on larval mortality and mass of these pests and on adult O. insidiosus was evaluated in extended laboratory experiments. The organophosphate chlorpyrifos and the spinosyn spinetoram provided long-lasting (seven-day) control against all stages of both pests. The growth regulator methoxyfenozide and the diamides chlorantraniliprole and cyantraniliprole had strong (1–7 days) larvicidal, particularly on young larvae, and growth inhibitory activity, but only the diamides were adulticidal. Among neonicotinoids, acetamiprid had stronger ovicidal and adulticidal activity than thiamethoxam, showing within-insecticide class differences in toxicities; however, both were weak on larvae. Lethality of novaluron and indoxacarb was inconsistent, varying depending on species and stage. Chlorpyrifos was most toxic to O. insidiosus. These results show species- and stage-specific toxicities, and greater compatibility with biological control, of the newer reduced-risk classes of insecticides than older chemistries.


Introduction
The use of insecticides in agroecosystems is constantly changing in response to the needs for products with greater public safety and reduced environmental risks [1]. This has led to the development of new classes of chemicals with novel modes of action [2,3]. Moreover, the implementation of the Food Quality Protection Act (FQPA) by the Environmental Protection Agency (EPA) in the United States of America (USA) [4] and the European Directive 2009/128/EC on sustainable use of pesticides and regulation in Europe (REGULATION (EC) No. 1107 [5] has resulted in increased restrictions on the use of broad-spectrum insecticides, and their replacement larvae, and adults) of these pests can overlap, whereas in the spring (May) only the larval stages are present [17]; therefore, these stages are the primary targets of insecticide applications. Orius insidiosus is also abundant in July in cranberries [25]. Thus, these studies have important implications for managing S. sulfureana and Ch. parallela and the conservation of biological control agents during the cranberry growing season when insecticide applications are most critical.

Insects
Eggs, larvae, and adults of S. sulfureana and Ch. parallela were obtained from laboratory colonies at the Rutgers P.E. Marucci Center (Chatsworth, NJ, USA). The colonies have been kept for over 10 years; wild larvae have been added to the colonies every year to maintain genetic integrity. Adults were kept inside air-filled plastic bags, that were also used as an oviposition substrate, and provided with cotton balls saturated with 10% m/v sugar-water solution. Egg masses laid on the bag's interior surface were collected and placed in 5 cm Petri dishes (Thermo Fisher Scientific Inc., Pittsburgh, PA, USA), on a slightly moistened (with distillate water) piece of filter paper (Whatman No. 4; Maidstone, Kent, UK). The egg masses were kept in an incubator at 25˘5˝C, 50%-60% RH, and 10:14 (L:D), and monitored daily until they hatched. Subsequently, neonate larvae were placed in 28.3 g plastic shot glasses (four larvae per glass) (Frontier Agricultural Sciences, Newark, DE, USA), containing commercial "stonefly Heliothis" diet (Ward's Natural Science, Rochester, NY, USA). Once larval development was completed, pupae were removed from the diet and placed in 453.6 g plastic deli containers (Placon, Madison, WI, USA), lined with Kimwipe paper (11.4 cmˆ21.3 cm; Fischer Scientific), until the adults emerged. The same rearing protocol was used for both species.

Treatments
Insecticide treatments were prepared at a concentration equivalent to the labeled field application rate (Table 1). Insecticides were selected to represent a wide array of modes of action and chemical classes. These insecticides can be broadly classified into: broad-spectrum organophosphates (chlorpyrifos), organophosphate-replacements (novaluron, acetamiprid, and thiamethoxam), and reduced-risk (chlorantraniliprole, spinetoram, cyantraniliprole, methoxyfenozide, and indoxacarb) [28]. All insecticides evaluated, except for cyantraniliprole (which is under the process of registration), are currently registered for use in cranberries in the USA [29]. Chlorpyrifos was used as the grower standard. All experiments were conducted at the Rutgers P.E. Marucci Center for Blueberry and Cranberry Research and Extension in Chatsworth, New Jersey, NJ, USA.

Laboratory Assays
A series of experiments was conducted in 2011 and 2013 to evaluate the effect of various insecticides (Table 1) on S. sulfureana and Ch. parallela eggs, larvae (neonates and 3rd instars), and adults (males and females). Treatments were applied using a Potter precision spray tower (Burkard Scientific, Uxbridge, UK), with an output equivalent to 280 L/hectare (30 gallons/acre), diluted in deionized water (spray volume = 28 mL/m 2 ). In addition to the insecticides in Table 1, experiments included an untreated control (deionized water alone) for a total of 10 treatments. The spray tower was calibrated between treatments.

Egg Toxicity
The ovicidal activity of insecticides (Table 1) was measured by placing 7 egg masses of either S. sulfureana or Ch. parallela in a Petri dish (5 cm diameter; Fisher Scientific) lined with a slightly moistened filter paper (Whatman No. 4). The number of eggs per mass was counted before treatment. Treatments were sprayed directly on top of the egg masses and allowed to dry for 30 min. Egg masses were placed in individual Petri dishes that were sealed with Parafilm M (Pechiney Plastic Packaging, Chicago, IL, USA), and then placed in an incubator at 25˘5˝C; 50%-60% RH and 10:14 (L:D). Eggs were checked daily 10-14 days after treatment to monitor for their viability (number of developed eggs) and egg hatch (number of emerged larvae); the eggs of both pest species take about 10-12 days to hatch [17,30,31]. Viability of eggs was determined under a microscope by recording the number of eggs that changed coloration (for S. sulfureana from light green-yellow to dark green-black color; and for Ch. parallela from yellow-orange to black), which is indicative of egg development; non-viable eggs were those that did not change in color. Each Petri dish was considered as a replicate, and each treatment was replicated 10 times; S. sulfureana and Ch. parallela were tested separately.

Larval Toxicity
The toxicity of insecticides to S. sulfureana and Ch. parallela larvae was evaluated in Petri dishes (4 cm diameter; Fisher Scientific) half-filled with artificial diet (see above, Section 2.1). Insecticides were applied directly to the surface of the diet, and residues were allowed to dry for 2 h before placing the larvae in the Petri dishes. Either two 1st instars or one 3rd instar larvae were then placed on the surface of the dried diet in each Petri dish using a fine paint brush. Larvae were exposed to treated diet at 0 (i.e., 2 h after treatment) and 3 days after treatment (3 DAT); with a new set of insects and diet for each date. To prevent larvae from escaping, treated Petri dishes were sealed with Parafilm M (Pechiney Plastic Packaging), and then placed in an incubator at the conditions described above. Larval mortality was recorded 7 days after exposure to treated diet. Percent mortality was calculated by adding the number of dead and missing larvae (assumed to have died and broken down). Each Petri dish was considered as a replicate, and each treatment and residue age was replicated 10 times; each moth species was tested separately.

Adult Toxicity
To evaluate the toxicity of insecticides on S. sulfureana and Ch. parallela adults, the inner surfaces of Petri dishes (5 cm diameter; Fisher Scientific) were treated with insecticides and allowed to dry for 2 h. After drying, one adult was placed inside each Petri dish together with a small piece of cotton ball saturated with a 10% m/v sugar-water solution to serve as food. Petri dishes were sealed with Parafilm M (Pechiney Plastic Packaging), and then placed in an incubator (as described above, Section 2.1.). Adults were exposed to treated surfaces at 0, 3, and 7 DAT; with a new set of insects and dishes for each date. Adult mortality was recorded 48 h after insecticide exposure. Insects that were moving but failed to remain upright were considered moribund (i.e., knocked down), whereas insects that exhibited no or very little movement when touched with a probe were considered dead. Each treatment was replicated 10 times for each species and residue age (5 replicates for each gender).

Extended Laboratory Experiments
Because the substrate on which an insecticide is deposited can affect its activity (e.g., [32]), as can the conditions during residue aging, extended laboratory trials were conducted to evaluate the mortality and growth effects of various insecticides (Table 1) on S. sulfureana and Ch. parallela 1st and 3rd instar larvae exposed to residues on cranberry foliage from 2012 to 2015. In addition to the insecticides listed in Table 1, an untreated control was included for a total of 10 treatments. Although most treatments were tested in 2012 ( Table 1), because of limited availability of larvae for bioassays, experiments were expanded across 4 years. In all years, however, chlorpyrifos (used as grower standard) and untreated control treatments were included as reference for comparison. The experiments were done in a mature cranberry bog cv. "Early Black" located at the Rutgers P.E. Marucci Center. The bog was managed under standard commercial conditions (e.g., fertilizer, fungicides, etc.); however, no insecticides were applied. Treatments were applied to 60 cmˆ60 cm plots, separated by a 15 cm buffer zone, and were replicated 4 times. Treatment applications were made from mid-July through early-August (post-bloom) with a CO 2 backpack sprayer (R&D Sprayers, Opelousas, LA, USA), using a 1 L plastic bottle, and done early in the morning at wind speeds of <4 km/h to avoid drift; during the spray the sprayer's nozzle was kept within 30 cm from the top of the canopy to further avoid drift. No precipitation was recorded during the experiments. The sprayer was calibrated to deliver 467 L per hectare (50 gallons/acre) at 30 psi, using a single TeeJet VS nozzle (size 110015; Spraying Systems Co, Wheaton, IL, USA), yielding 47 mL per cm 2 . This application volume is within the range recommended for optimal canopy deposition with insecticides under commercial conditions in cranberries. After treatment, foliage (uprights) from field plots was collected, placed in plastic bags, and brought to the laboratory 1, 3, and 7 DAT. Foliage from all four plots of each treatment was combined before selecting uprights used in bioassays. In the laboratory, 3 or 4 uprights (each~15 cm in length) from a given treatment were inserted into water-filled florists' water picks that were secured on Styrofoam trays. Uprights were randomly selected from all foliage collected for that treatment. The top of uprights (containing foliage and fruits) was enclosed in a ventilated 40 dram plastic snap cap vial (3 cmˆ5 cm) to prevent larvae from escaping, while the lower parts (bare stems only) passed through a hole in the vial bottoms and were placed inside the water picks. Three 1st instars or one 3rd instar larvae were placed inside each vial. Vials were placed on a light bench in the laboratory (25˘5˝C, 50%-60% RH and 15:9 L:D). Larval mortality (% of dead and missing larvae) and mass of surviving 3rd instar larvae were assessed after 7 days. Larval mass was recorded for 3 and 7 DAT using a Mettler-Toledo AE50 analytical balance of 0.1 mg resolution (Mettler Instrument Corp., Hightstown, NJ, USA). There were 10 vials per treatment for each species and residue age, and each vial was considered a replicate.

Non-Target Effects
The toxicity of various insecticides on O. insidiosus adults was tested in an extended laboratory experiment. All insecticides listed in Table 1 were tested in this experiment, except for thiamethoxam, and an untreated control for a total of 9 treatments. The experiment was conducted in July 2012 and followed the same protocol as described above (see Section 2.4). Residual activity was evaluated 3 and 7 DAT. Groups of 5 O. insidiosus adults (mix of males and females) were placed in each vial, mortality was recorded 24 h after exposure, and there were six vials per treatment and residue age (replicates).

Statistical Analysis
Mean percent mortality of eggs, larvae, and adults of S. sulfureana and Ch. parallela for each treatment at each time check (residual activity) was arcsine square-root transformed. Similarly, mean percent mortality of O. insidiosus adults for each treatment and time check was arcsine square-root transformed. Mass data were natural log (ln)-transformed. Transformed mortality and mass data were analyzed using analysis of variance (ANOVA), with treatment, instar, time check (date), or gender, and their interactions, as factors [33]. To simplify our analysis, extended laboratory data for each species and instar were combined across years because larval mortality in untreated control and chlorpyrifos (grower standard) treatments did not change among years (p > 0.05) [34]. To avoid pseudo-replication, if more than 1 larva were placed per vial (replicate), larval mortality and mass data for each vial were averaged to obtain a mean value for each vial prior to analysis. Untransformed percent data are shown in the Figures. Mean differences were compared using Tukey's test (α = 0.05) after a significant F-test.

Egg Toxicity
For both pest species, the insecticides evaluated here had no effect on percent egg viability ( Table 2); however, some insecticides had a negative impact on egg hatch (Table 2; Figure 1). For both species, acetamiprid, spinetoram, novaluron, and chlorpyrifos had the highest ovicidal activity; while chlorantraniliprole and indoxacarb were the least toxic to S. sulfureana and Ch. parallela eggs ( Figure 1). The insect growth regulator methoxyfenozide had intermediate ovicidal toxicity against both species. There were, however, some discrepancies between species: cyantraniliprole and thiamethoxam were relatively harmless to S. sulfureana eggs but toxic to Ch. parallela eggs ( Figure 1).

Larval Toxicity
Insecticide treatment, larval instar, and DAT affected S. sulfureana and Ch. parallela larval mortality (Table 2). However, these factors interacted in different ways to affect both species, indicating some species-specific effects. For instance, all factors interacted to affect S. sulfureana larval mortality, while only treatment and instar interacted to affect Ch. parallela larval mortality (Table 2).
As expected, 3rd instar larvae of both species were more resistant to insecticides than 1st instar larvae, but this effect depended on the insecticide (significant Treatment × Instar interaction for both species; Table 2). In fact, only spinetoram, methoxyfenozide, and chlorpyrifos were highly toxic to 3rd instar S. sulfureana and Ch. parallela larvae at 0 and 3 DAT (Figures 2 and 3). The neonicotinoids acetamiprid and thiamethoxam were weak against 3rd instars. Contrasting both species, indoxacarb was only toxic to S. sulfureana 3rd instar larvae, while the diamides chlorantraniliprole and cyantraniliprole and the chitin inhibitor novaluron were more toxic to Ch. parallela 3rd instar larvae (Figures 2 and 3).

Larval Toxicity
Insecticide treatment, larval instar, and DAT affected S. sulfureana and Ch. parallela larval mortality (Table 2). However, these factors interacted in different ways to affect both species, indicating some species-specific effects. For instance, all factors interacted to affect S. sulfureana larval mortality, while only treatment and instar interacted to affect Ch. parallela larval mortality (Table 2).
As expected, 3rd instar larvae of both species were more resistant to insecticides than 1st instar larvae, but this effect depended on the insecticide (significant TreatmentˆInstar interaction for both species; Table 2). In fact, only spinetoram, methoxyfenozide, and chlorpyrifos were highly toxic to 3rd instar S. sulfureana and Ch. parallela larvae at 0 and 3 DAT (Figures 2 and 3). The neonicotinoids acetamiprid and thiamethoxam were weak against 3rd instars. Contrasting both species, indoxacarb was only toxic to S. sulfureana 3rd instar larvae, while the diamides chlorantraniliprole and cyantraniliprole and the chitin inhibitor novaluron were more toxic to Ch. parallela 3rd instar larvae (Figures 2 and 3).

Extended Laboratory Experiments
Insecticide class, instar, and residual age, and their two-way interaction, had an effect on S. sulfureana and Ch. parallela larval mortality (Table 3). For S. sulfureana, the effect of insecticide on larval mortality was influenced by instar and residual age (significant three-way interaction; Table 3).

Extended Laboratory Experiments
Insecticide class, instar, and residual age, and their two-way interaction, had an effect on S. sulfureana and Ch. parallela larval mortality (Table 3). For S. sulfureana, the effect of insecticide on larval mortality was influenced by instar and residual age (significant three-way interaction; Table 3).

Discussion
In the past two decades, new classes of reduced-risk insecticides have been registered for insect pest control in many crops including cranberries; yet, few studies have compared the toxicity and residual activities of these insecticides on different life stages of pests. In this study, we found a high degree of specificity on the toxicities of reduced-risk insecticides on two key pests of cranberries, S. sulfureana and Ch. parallela, depending on insect stage (egg, larvae, and adult) and gender, and insecticide residual age. In general, only the spinosyn spinetoram and the organophosphate chlorpyrifos had broad-spectrum activities against all stages of the two lepidopteran pests tested here, while the remaining insecticides, that included organophosphate-replacement and reduced-risk chemistries, were more target specific in their toxicities. Moreover, as compared with the organophosphate chlorpyrifos, the newer insecticides had less negative effects on the predator O. insidiosus, indicating that these insecticides may be more compatible with biological control and can be implemented into IPM programs for the control of both pests.
Our experiments demonstrate high insecticidal and good residual (1-7 days) activities of the organophosphate chlorpyrifos and the spinosyn spinetoram against eggs, larvae, and adults of S. sulfureana and Ch. parallela. Insecticides affected egg hatch, but not their viability, indicating an effect at later stages of egg development. Both moth species protect their eggs with scales, which may reduce exposure of eggs to some insecticides; however, there were some species-specific differences in their susceptibility of eggs to insecticides (see below), indicating that other factors may also be involved. Charlet and Busaca [35] reported that chlorpyrifos applied at the proper time of the year prevents sunflower yield losses from the banded sunflower moth, Cochylis hospes Walsingham (Lepidoptera: Cochylidae). Chlorpyrifos is an insecticide intensively used in cranberries in New Jersey because of its broad-spectrum activity not just on chewing but also sucking insect pests; this insecticide is, however, under consideration for elimination by EPA in USA crops. Therefore, finding alternatives to chlorpyrifos is critical. The spinosyn spinosad was also toxic to all life stages (eggs, larvae, and adults) of the cranberry fruitworm Acrobasis vaccinii Riley (Lepidoptera: Pyralidae) [36]. Moreover, spinetoram had a lower LC 50 value against the fall armyworm, Spodoptera frugiperda (J.E. Smith) (Lepidoptera: Noctuidae), larvae than the LC 50 s of seven other insecticides tested [37]. These studies together with ours demonstrate the broad-spectrum efficacy of spinosyns against different life stages of various lepidopteran pests; spinosyns can thus be used as alternatives to conventional broad-spectrum insecticides to manage S. sulfureana and Ch. parallela.
Two neonicotinoid insecticides were tested here: acetamiprid and thiamethoxam, but their efficacy depended on insect species and life stage. Acetamiprid had ovicidal activity on both S. sulfureana and Ch. parallela, while thiamethoxam was effective only on Ch. parallela eggs. Acetamiprid also had stronger adulticidal activity than thiamethoxam; thiamethoxam being toxic only to S. sulfureana males but not females, and moderately toxic to Ch. parallela adults. Both were weak against larvae, with acetamiprid tending to be more toxic than thiamethoxam and younger larvae being more susceptible to both insecticides than older instars. All together, these result indicate differential toxicities of insecticides that belong to the same class, with distinct contact toxicities on different insect species and stages. These differential toxic effects are likely due to the systemic nature of these insecticides [38], which might be influenced by their levels of penetration and persistence in insect and plant tissues [39]. For instance, Wise et al. [36] reported strong ovicidal and larvicidal, but weaker adulticidal, activities of the neonicotinoids acetamiprid and thiacloprid on A. vaccinii than broad-spectrum insecticides. They indicate that these neonicotinoids have relatively short-lived adult contact toxicity due to residue profiles on vegetable substrate; with declining toxicities as residues penetrate plant tissues. Yue et al. [40] also showed differential toxicities of neonicotinods depending on larval age, with higher concentrations and longer exposure periods of thiamethoxam required to provide 100% larval mortality as larvae of the Indianmeal moth, Plodia interpunctella (Hübner) (Lepidoptera: Pyralidae), age. Because of their systemic activity, neonicotinoids have mainly been used to target sucking insects [41]; our results show variable toxicities of this class of insecticides on two key lepidopteran pests of cranberries, which will likely limit their adoption.
The insect growth regulators (IGRs) methoxyfenozide (an ecdysone mimic [42]) and novaluron (a chitin inhibitor [43]) affected various S. sulfureana and Ch. parallela life stages differently. Methoxyfenozide was one of the most stage-specific insecticides tested, having weak ovicidal and adulticidal activity but excellent residual activity on 1st and 3rd instar larvae of both cranberry pests, indicating that this insecticide likely operated mainly via ingestion [42]. Stage-specific toxicity of methoxyfenozide was previously reported against A. vaccinii [36] and grape berry moth, Paralobesia viteana (Clemens) [44]. Wise et al. [36] also demonstrated that, although it does not have direct toxicity on eggs, methoxyfenozide showed ovi-larvicidal activity on A. vaccinii when applied after eggs were laid. Because methoxyfenozide is safe on bees, it is the only insecticide tested that can be used during bloom in cranberries, when S. sulfureana and Ch. parallela eggs are laid, so precise timing of spray applications is required for optimal control. The benzoylurea novaluron, on the other hand, had strong activity on S. sulfureana and Ch. parallela eggs, but weak adulticidal activity and moderate larvicidal, particularly on 1st instars, activity. Ovicidal activity of novaluron has been shown for lepidopteran (e.g., [36,45]), and coleopteran (e.g., [46,47]), pests. The use of the IGR novaluron as an ovicidal-larvicidal insecticide, as well as neonicotinoids, in cranberries may be limited by their potential toxicity to bees (e.g., [48][49][50]).
The newer insecticides, the anthranilic diamides chlorantraniliprole and cyantraniliprole affect muscle contraction [51]. These insecticides had strong activity on young S. sulfureana and Ch. parallela larvae, but were weak on 3rd instars in both laboratory and extended laboratory experiments, in particular on those of S. sulfureana. Stage-specific mortality effects could be due to metabolic differences in detoxification enzymes, as well as their feeding habits as older instars of both pest species produce silk webs [30,31], which may reduce their exposure to insecticides; however, these possibilities remain to be explored. Furthermore, the effects of insecticides on larval mortality and mass were assessed after seven days of exposure to treated diet or foliage; thus, it is possibly that our bioassay did not reflect fully the effects of these insecticides on larvae and stronger negative effects on 3rd instars could be seen after a longer exposure. Slow larval death might not be problematic to growers if exposed larvae cease feeding as shown by the reduced mass gained by Ch. parallela fed diamide-treated foliage; however, this was not the case for S. sulfureana, where diamide-exposed larvae gained as much mass as the controls. Both diamide insecticides also showed adulticidal activity and, in the case of chlorantraniliprole, ovicidal activity on Ch. parallela, suggesting exposure through contact in addition to via ingestion [52,53]. Most adults exposed to diamides were found alive but in a moribund state, indicating a slow-acting mode of action of this class of insecticides on both older instar larvae and adults, which is consistent with their negative effects on insect movement.
Interestingly, the oxadiazine insecticide indoxacarb had differential larval toxicity depending on pest species. It was toxic to S. sulfureana larvae but not to Ch. parallela larvae in both laboratory and extended laboratory experiments. This difference in toxicity between species is not likely due to resistance to indoxacarb in Ch. parallela because this insecticide has rarely been used to control this pest in cranberries in New Jersey; instead, physiological differences in levels of detoxification enzymes and/or behavioral differences in feeding habits are more possible explanations and worth exploring further. Indoxacarb had no ovicidal and only weak adulticidal activity. Liu et al. [54] also reported no toxic effects of indoxacarb on eggs of the diamondback moth, Plutella xylostella L., but a high efficacy, comparable to spinosad, against larvae. Unfortunately, the level of species-specificity of indoxacarb will likely limit its adoption for the control of these pests in cranberries, as has occurred thus far.
Reduced-risk insecticides are expected to be more compatible with biological control as compared with conventional broad-spectrum insecticides (e.g., [55]); yet, only a few studies have examined their non-target effects on natural enemies of agricultural pests to date [56]. In this study, we showed that organophosphate-replacement and reduced-risk insecticides were less toxic than the organophosphate chlorpyrifos on the insect predator O. insidiosus; however, all these newer insecticides, except for methoxyfenozide, had residual (three days) toxicities higher than the control. Similarly, Roubos et al. [57] showed high levels of toxicity of organophosphates on O. insidiosus; however, they also reported high mortality to acetamiprid and to exposure to fresh-residues of spinetoram, as well as moderate toxicities of indoxacarb and low toxicities of methoxyfenozide, novaluron, and chlorantraniliprole. The study by Roubos et al. [57] used insecticide-treated Petri dishes, whereas our bioassays were done with cranberry-treated foliage; O. insidiosus exposure to insecticides may thus vary depending on the type of treated substrate [32].
In summary, we have identified several insecticides that can be used as replacements to organophosphates (i.e., chlorpyrifos) for S. sulfureana and Ch. parallela control in cranberries. Except for spinetoram, these newer insecticides had higher stage-specificity. Results from laboratory and extended laboratory experiments were consistent, indicating that substrate (diet versus foliage) did not influence larval exposure to the insecticides tested or their residual activity, which could be the case if insecticides are absorbed or broken down by plant tissues [58]. Both pest species overwinter as young larvae; thus, applications of insecticides such as chlorantraniliprole, methoxyfenozide, spinetoram, and cyantraniliprole, that are highly effective against this stage, should be applied early in the season if numbers exceed economic thresholds [17,30,31]. Fewer options are available later in the spring because older instars are more tolerant to some of these insecticides such as the diamides, so scouting and timing of insecticides are important. In the summer, 2nd generation larvae need to be controlled before they enter the fruit. Sex pheromones are available to monitor adult flight of both species [59][60][61], and for S. sulfureana a degree-day model is available to predict time of egg laying and hatch [62]. Among the insecticides tested here, methoxyfenozide is the only option during bloom in cranberries to prevent newly-hatched larvae from entering the fruit. After bloom, chlorantraniliprole, spinetoram, and cyantraniliprole could also be used as alternatives to chlorpyrifos (after cyantraniliprole is registered in cranberries). Although larvae are the main targets of insecticide applications due to their damage to fruit, spinetoram has the added benefit of having ovicidal activity, and chlorantraniliprole, spinetoram, and cyantraniliprole having adulticidal activity; these insecticides were also more compatible with biological control than organophosphates.

Conclusions
Our laboratory and extended laboratory experiments demonstrate that organophosphatealternative and reduced-risk insecticides can be used for effective control of S. sulfureana and Ch. parallela in cranberries. These insecticides were, in general, more selective, i.e., stage-specific, than conventional broad-spectrum insecticides, and will thus require more intensive scouting and precise timing of application. Implementation of these newer insecticides into current IPM programs will help reduce the reliance of growers on broad-spectrum insecticides (e.g., chlorpyrifos), while conserving biological control agents in cranberries. Moreover, these various classes of insecticides should be rotated pre-and post-bloom within and between seasons to prevent onset of resistant populations. In fact, organophosphate-resistant populations of S. sulfureana have been reported in cranberries in Massachusetts, USA [17].