Insecticidal Activity of Plant Powders against the Parasitoid, Pteromalus venustus, and Its Host, the Alfalfa Leafcutting Bee

Developing a bee-friendly alternative to traditional insecticides used within commercial environments can contribute to reductions in pesticide exposure experienced by managed bees. We performed acute contact toxicity studies using fifteen plant powders from seven plant families against a parasitoid pest, Pteromalus venustus, and its host, the Alfalfa leafcutting bee (ALB). Ajwain, cinnamon, clove, cumin, fennel, ginger, nutmeg, oregano and turmeric applied at low contact concentrations had sufficient fumigant properties to cause equivalent or higher parasitoid mortality as that obtained with the traditional insecticide. Nutmeg adversely affected adult ALBs at both low and high contact concentrations, thus eliminating it as a candidate. Increasing the contact concentrations did not consistently increase parasitoid control but did increase adverse effects on the ALBs. In addition, the efficacious plant powders significantly reduced the sexual function and fertility of the female parasitoids, a feature not associated with the traditional insecticide. The dual nature of the mechanisms underlying the effects of the plant powders may translate into effective control of the parasitoid populations in the commercial environment. The results reported here support further evaluations of Ajwain, cinnamon, clove, cumin, fennel, ginger, oregano and turmeric as potential botanical insecticides for control of P. venustus.


Introduction
Alfalfa leafcutting bees (ALBs, Megachile rotundata Fab.) are gregarious, solitary bees that are important pollinators of alfalfa and canola. Each year, about 4 billion ALB adults are released into alfalfa and canola fields in western Canada and the northwestern US to pollinate the flowers, maximizing seed production [1,2]. Shelters containing nest blocks are provided within the fields for the mated female bees to construct and provision their nests. To do this, they gather leaves, nectar and pollen for individual nest cells, laying one egg in each cell after depositing the provisions and finally enclosing and capping off the developing larvae within the leaf-based cocoon. At the end of the summer, the nest blocks containing the cocoons are collected from the fields, dried for a period, and then harvested before being stored until the following spring. At that time, the cocoons are added to trays and transferred to an incubator maintained under standard ALB rearing conditions. The trayed cocoons are incubated for an average of 21 days, where male bees emerge between day 18 and 28, while the female bees emerge between day 21 and 28 [3]. The newly emerged bees may be held in the cold to synchronize emergence with crop phenology or transported directly to the field and released. Trays remain within the field for 7 to 10 days to ensure that female emergence is complete and all bees have left the trays. dead bees within cocoons) were placed in plastic boxes (35 × 19 cm) with a metal screened lid for ventilation and Sefar ® e'mesh (Basic Mesh-123 Mesh 70 Micron PW; Davis International, Fairport, New York, NY, USA) to contain any emerging parasitoids. The boxes were maintained at 28 • C, 16:8 photoperiod and 50% relative humidity for 30 days, the standard incubation period. At day 15, two water-soaked pieces of paper towel placed within a weigh boat were added to the box to increase the relative humidity. Peak male adult emergence fell within 17 to 24 days of the 30-day incubation period. The adult male bees were transferred from the boxes using feather light forceps into 50 mL Fisherbrand™ conical tubes (Fisher Scientific Company, Ottawa, Ontario, Canada) just prior to use in the bioassays.

Rearing of Parasitoids
Stored cocoon matrix (~156 g) was transferred to 1000 mL transparent polypropylene containers. The central part of the lid had metal screening to allow for ventilation, and Sefar ® e'mesh was glued over the screening to contain any emerging parasitoids. A range of 30 to 50 female parasitoids were added to each container from an initial starter colony set up in a similar manner, the lid was replaced and then it was maintained at 28 • C, 16:8 photoperiod and 50% relative humidity for 15 days. Adult P. venustus were collected using an aspirator system (Tetra ® Whisper 60 Air Pump; PetSmart, Lethbridge, AB, Canada) with tubing connectors attached to a Fisherbrand™ specimen container (Fisher Scientific Company, Ottawa, ON, Canada) between day 12 and 15, just prior to use in the bioassays.

Acute Contact Toxicity-Adult P. venustus
A preliminary examination of the distribution of 1% plant powder within the cocoon matrix (w/w) in commercial trays and bioassay containers suggested that the adult parasitoids could come into contact with either an electrostatic amount of particles (low contact) from emergence and walking within the cocoon matrix, or a thick coating of particles (high contact) from settled plant powders within the cocoon matrix. All plant powders exhibited this uneven distribution within the cocoon matrix. To mimic those conditions, we physically coated adult parasitoids with low or high contact concentrations of each plant powder. This would provide us with insight into the dose range necessary to achieve control in the commercial tray environment. We also calculated the powder concentrations relative to the container volume, a method used to compare the released EO vapor potency among plant powders.

Low Contact with Plant Powders
A mixture of ten male and female parasitoids were collected as described in Section 2.2. The lid was removed from a 90 mm Phoenix Biomedical Petri dish lined with Whatman ® (Fisher Scientific Company, Ottawa, Ontario, Canada) 85 mm filter paper that was previously dusted with a plant material using a stainless-steel sieve (8 cm). The parasitoids were added to the dish. The exposure concentration for each plant material in the bioassay container was 26 mg/cm 2 of Ajwain seed, 22 mg/cm 2 ground basil leaves, 28 mg/cm 2 cinnamon powder, 19 mg/cm 2 clove powder, 33 mg/cm 2 coriander powder, 62 mg/cm 2 cumin powder, 47 mg/cm 2 ground Fenugreek seed, 21 mg/cm 2 ground fennel seed, 79 mg/cm 2 ginger powder, 47 mg/cm 2 nutmeg powder, 22 mg/cm 2 ground oregano leaves, 16 mg/cm 2 ground rosemary, 20 mg/cm 2 sage powder, 22 mg/cm 2 ground thyme leaves and 29 mg/cm 2 turmeric powder. The variation in the amounts applied reflects the differences in particle size and how it affects the distribution on the filter paper. Honey water (1:1) was supplied ab libitum via a cotton pad, with no visible contamination from the plant powders transferred to the feeding surface. The lid was then replaced and taped to the bottom of the dish to prevent the adult parasitoids from escaping. Untreated parasitoids were handled in the same manner, but in the absence of a plant powder. Normally, the untreated parasitoids will pick up an electrostatic concentration of leaf powder as they emerge from the cocoons. This served as the negative control. The bioassay containers were transported to an incubator and maintained at 28 • C, 16:8 photoperiod and 50% relative humidity. Mortality was assessed at 24, 48, 72 and 96 h. Ten adult P. venustus were weighed before and after exposure to the plant powders. No weight changes were observed per individual adult, but sparse plant particles were visible on the parasitoids when examined with a dissecting microscope. The experiment was replicated six times using cocoons obtained from different regions within Alberta.

High Contact with Plant Powders
A mixture of ten male and female parasitoids were collected as described in Section 2.2 and transferred to a 1.5 mL Axygen ® (Fisher Scientific Company, Ottawa, ON, Canada) tube containing 0.1 g of a plant powder. The exposure concentration for each plant material in the bioassay container was equivalent to 589 mg/cm 2 , or about 20 times the concentration encountered for the low contact with plant powders in Section 2.4.1. The tube was closed and gently rolled by hand to distribute the plant material onto the parasitoids. The tube contents were poured into a 90 mm Phoenix Biomedical Petri dish lined with Whatman ® 85 mm filter paper. Honey water (1:1) was supplied ab libitum via a cotton pad, with no visible contamination from the plant powders transferred to the feeding surface. The lid was then replaced and taped to the bottom of the dish to prevent the parasitoids from escaping. Untreated parasitoids were handled in the same manner, but in the absence of plant powder. The bioassay containers were transported to an incubator and maintained at 28 • C, 16:8 photoperiod and 50% relative humidity. Mortality was assessed at 24, 48, 72 and 96 h. Ten adult P. venustus were weighed before and after exposure to the tube-dusting. No weight changes were observed, but adults were visibly coated with plant particles. The experiment was replicated six times using cocoons obtained from different regions within Alberta.

Reproductive Toxicity-P. venustus
Each efficacious plant powder, as determined in Section 2.4, was applied to the filter paper as described in Section 2.4.1, but at half to a quarter of the concentration to prevent female parasitoid mortality, or about 4.75 to 40 mg/cm 2 . Clove and turmeric were applied at the lowest concentration to prevent mortality within the study period. Ten female and two male parasitoids were collected as described in Section 2.2 and added to a Petri dish lined with the treated filter paper. Honey water (1:1) was supplied ab libitum via a gauze-covered Eppendorf tube that was taped to the filter paper. The lid was then placed on top and taped to the bottom to prevent the adult parasitoids from escaping. Untreated parasitoids were handled in the same manner, but in the absence of plant powder. The Petri dishes were transported to an incubator and maintained at 28 • C, 16:8 photoperiod and 50% relative humidity for 24 h. Then, each female parasitoid was transferred to the lid of a 112 mL disposable polypropylene dish fitted with four caps (VWR disposable/sterile 8-strip caps for library tubes; VWR International, Edmonton, AB, Canada) glued to the lid, each containing one ALB fifth-instar prepupa that had been removed from the cocoon. Honey water was supplied ab libitum by adding a 1 cm piece of cotton swab. The bottom of the dish was then snapped onto the lid. Untreated parasitoids were handled in the same manner, but in the absence of plant powder. The bioassay containers were transported to an incubator and maintained at 28 • C, 16:8 photoperiod and 50% relative humidity Insects 2020, 11, 359 5 of 20 for 7 days. Each cup was then opened and assessed for the number of parasitized prepupa or the number of parasitoid larvae per prepupa. The experiment was replicated three times using parasitoids obtained from different regions within Alberta.
2.6. Acute Contact Toxicity-Adult ALBs 2.6.1. Low Contact with Plant Powders Ten adult male ALBs were collected, as described in Section 2.1, and then transferred to a 90 mm Phoenix Biomedical Petri dish lined with 85 mm Whatman ® filter paper that was previously dusted with a plant powder applied as described in Section 2.4.1. The Petri dishes were gently shaken, and the ALBs interacted with the plant powders for an additional 5 min. The adult bees were then cooled for 3 min at 4 • C and finally, transferred to a 112 mL polypropylene dish with two 2.5 × 2.5 cm coarse floor vent filters that served as perches. The lid of the plastic container was customized with a metal screened lid to provide ventilation and a port to hold an inverted 5 mL Eppendorf ® tube (Fisher Scientific Company, Ottawa, ON, Canada). This tube contained 1:1 honey water and was covered with layered cotton gauze held in place with an elastic. Untreated bees were handled in the same manner except no plant powder was applied to the filter paper. The bioassay containers were transported to an incubator and maintained at 28 • C, 16:8 photoperiod and 50% relative humidity. Mortality was assessed at 24, 48, 72 and 96 h. Individual adult bees were weighed before and after exposure. The mean weight of the plant material picked up by an adult bee ranged between 0.5 and 1 mg per bee. Since 10 bees were added to each container, this represents a range between 5 and 10 mg per container, or 32 to 65 µg/cm 2 . The experiment was replicated six times using cocoons obtained from different regions within Alberta.

High Contact with Plant Powders
Ten adult male ALBs were collected as described in Section 2.1 and then added to a 50 mL Fisherbrand™ conical tube containing 2 g of plant material. The lid was placed onto the tube and gently rolled four to six times. The visibly coated bees were then transferred to a 112 mL plastic container with two 2.5 × 2.5 cm coarse floor vent filters that served as a perch. The lid of the plastic container was customized with a metal screened lid to provide ventilation and a port to hold an inverted 5 mL Eppendorf ® tube. The 5 mL tube contained 1:1 honey water and was covered with layered cotton gauze held in place with an elastic. Untreated bees were handled in the same manner except no plant powder was applied to the filter paper. The bioassay containers were transported to an incubator and maintained at 28 • C, 16:8 photoperiod and 50% relative humidity. Mortality was assessed at 24, 48, 72 and 96 h. Individual adult bees were weighed before and after exposure to a tube-dusting of each plant material. The mean weight of the plant material picked up by an adult bee ranged between 3 and 5 mg per bee. Since 10 bees were added to each container, this represents a plant powder concentration between 30 and 50 mg per container, which is 0.20 to 0.33 mg/cm 2 or about 6 times higher than the amount applied in the low contact bioassays, Section 2.6.1. The experiment was replicated six times using cocoons obtained from different regions within Alberta.

Statistical Analysis
The count data were analyzed using the Loglinear component within SYSTAT 13.0 (Systat Software, Inc., San Jose, CA, USA), with two discrete variables, treatment and time-period. We examined the test for Model terms panel comparing the likelihood-ratio chi-square for the full model to the same value for the smaller model. To determine whether the removal of a term results in a significant decrease in the fit, we then looked at the difference in these statistics. If the fit was worsened with the removal of the term, it remained in the model. With each change in the model, we checked Raftery's BIC (Bayesian Information Criterion), which when negative, concludes that the model is preferable to the saturated model. For each factor in the model, z-scores were provided within the analysis, and the probabilities associated with the z-scores were obtained using z-tables. The results are expressed as a treatment mean or the effect of the treatments on a measured count variable, and this is compared with the effects of individual treatments.
Insects 2020, 11, x 9 of 21 member of Fabaceae, Fenugreek (3.2, −2.484, p = 0.006), significantly decreased parasitoid mortality compared with the treatment mean, indicating that it has no effect on the adult parasitoids. The hierarchy for relative potency of the plant powders is clove, fennel, nutmeg > Ajwain, cinnamon, turmeric > cumin, ginger, oregano. The change in the hierarchy from the low contact exposure suggests that clove, fennel and nutmeg have a higher essential oil content that facilitates the recorded higher parasitoid mortality. Only four plant powders act quickly (24 hr) under high contact conditions to elicit parasitoid mortality, including Ajwain, clove, cumin and fennel.      to the treatment mean, parasitizing 82% lower ALB prepupae than with the untreated parasitoids. Although fennel (0.55, 1.623, p = 0.0526) significantly increased the number of parasitized ALB prepupae compared with the treatment mean, it was 72% lower than with the untreated parasitoids. This suggests that the plant powders change the sexual function of the female parasitoids, resulting in fewer attacks.

Low Contact with Plant Powders (32 to 65 µg/cm 2 )
There was a significant effect of time (χ 2 = 220.019, df = 3, p < 0.005) and treatment (χ 2 = 48.48, df = 15, p < 0.005) on adult male ALB mortality. There was a significantly lower adult ALB mortality recorded at 24 h of exposure (1, −6.094, p < 0.0002), followed by a significant increase in adult ALB mortality between 72 (1.  (Figure 8). Interestingly, basil elicited highly variable mortality (0-9) resulting in a high mean mortality at 96 h, suggesting that there may be a genetic or management component to the regional differences in ALB sensitivity. In contrast, rosemary (4.2, 3.606, p < 0.0002) and sage (1.9, 2.019, p = 0.0217) significantly increased ALB mortality compared with the treatment mean. Since Vapona ® applications elicit about a 16% adult ALB loss [8], only rosemary poses a greater risk for ALB losses than the current method for controlling the parasitoid. For the family Apiaceae, several plant powders either significantly decreased adult ALB mortality or were equivalent to the treatment mean, including coriander (2.0, −1.863, p = 0.0314), cumin (1.8, −0.634, p = 0.2643) and fennel (1.4, −1.128, p = 0.1292; Figure 9). These powders, therefore, pose no risk for adult ALB losses. In contrast, Ajwain (2.5, 2.019, p = 0.0217) significantly increased the number of dead ALBs compared with the treatment mean, but was within the 16% losses reported for  Figure 10). These plant powders also did not pose a risk for adult ALB losses. Clove elicited high ALB losses, but the effect was highly variable (0-9 dead) and was dependent upon the origin of the cocoons. In contrast, one member of the Myristicaceae family (nutmeg, 3.1, 3.385, p = 0.0003) significantly increased adult ALB losses compared with the treatment mean, eliciting about a 72% greater loss than that recorded for untreated ALBs. The hierarchy of the relative potency for the plant powders to harm ALBs is nutmeg, rosemary > clove. Within the ALB commercial industry, 16% ALB losses [8] are acceptable. Based upon our results, nutmeg and rosemary cause higher ALB losses than the acceptable commercial level, eliminating them as candidates for parasitoid control.  (Figure 8). Interestingly, basil elicited highly variable mortality (0-9) resulting in a high mean mortality at 96 hrs, suggesting that there may be a genetic or management component to the regional differences in ALB sensitivity. In contrast, rosemary (4.2, 3.606, p < 0.0002) and sage (1.9, 2.019, p = 0.0217) significantly increased ALB mortality compared with the treatment mean. Since Vapona ® applications elicit about a 16% adult ALB loss [8], only rosemary poses a greater risk for ALB losses than the current method for controlling the parasitoid. For the family Apiaceae, several plant powders either significantly decreased adult ALB mortality or were equivalent to the treatment mean, including coriander (2.0, -1.863, p = 0.0314), cumin (1.8, −0.634, p = 0.2643) and fennel (1.4, −1.128, p = 0.1292; Figure 9). These powders, therefore, pose no risk for adult ALB losses. In contrast, Ajwain ( Figure 10). These plant powders also did not pose a risk for adult ALB losses. Clove elicited high ALB losses, but the effect was highly variable (0-9 dead) and was dependent upon the origin of the cocoons. In contrast, one member of the Myristicaceae family (nutmeg, 3.1, 3.385, p = 0.0003) significantly increased adult ALB losses compared with the treatment mean, eliciting about a 72% greater loss than that recorded for untreated ALBs. The hierarchy of the relative potency for the plant powders to harm ALBs is nutmeg, rosemary > clove. Within the ALB commercial industry, 16% ALB losses [8] are acceptable. Based upon our results, nutmeg and rosemary cause higher ALB losses than the acceptable commercial level, eliminating them as candidates for parasitoid control.      There was a significant effect of time (χ 2 = 97.508, df = 3, p < 0.005) and treatment (χ 2 = 72.636, df = 15, p < 0.005) on adult male ALB mortality. There was a significant lower mean ALB mortality recorded at 24 h of exposure (1.6, −6.623, p < 0.0002), followed by a gradual significant increase in adult ALB mortality at 72 (2.9, 3.318, p = 0.0005) and 96 h (3.6, 8.243, p < 0.0002; Figures 11-13). For untreated ALBs (1.0, −2.332, p = 0.0099), the mean number of dead adult ALBs was significantly lower (64%) than the treatment mean of 1.2. For the family Lamiaceae, the mean number of dead adult ALBs for basil (1.3, −1.379, p = 0.0838), oregano (1.7, −0.610, p = 0.2709), thyme (0.2, −1.762, p = 0.0392) and sage (1.8, −0.418, p = 0.3372) either significantly decreased ALB mortality or was equivalent to the treatment mean of 1.2 ( Figure 11). In contrast, rosemary (2.5, 2.028, p = 0.0212) significantly increased adult ALB mortality compared with the treatment mean, and these losses were above those reported with Vapona ® applications [8].  (Figure 13), but were within the 16% bee loss attributed to Vapona ® applications [8]. In contrast, one member of the family Myristicaceae (nutmeg, 4.5, 6.711, p < 0.0002) and one member of the family Myrtaceae (clove, 2.8, 2.028, p = 0.0212) significantly increased adult ALB mortality compared with the treatment mean, and these losses are above those reported with Vapona ® applications [8]. As with the lower contact results, clove elicited high ALB losses at 96 h, but was highly variable, suggesting that there may be a genetic or management component to the regional differences in bee sensitivity. The hierarchy for the relative potency of the negative effects of the plant powders on adult ALBs is nutmeg > cumin, fennel, rosemary > clove. Only nutmeg poses a higher risk for ALB losses than the currently applied insecticide and, therefore, has been eliminated as a candidate for parasitoid control.
One of the factors affecting the outcome for contact toxicity with high concentrations or coatings of plant powders was related to bee behavior. The heavily coated bees tended to readily "buzz" the powder off the body within minutes of release into the bioassay container, and followed this with intensive grooming. The different particle sizes contributed to the times required for the "buzz" behavior to remove the particles from the bee's body. Fine powders such as turmeric and cinnamon stayed on the body for the longest period and required intensive grooming, while course powders such as clove or rosemary were removed immediately followed by intensive grooming. Despite this, increasing the particle distribution on the bee body for oregano, sage, cumin, fennel, ginger, cinnamon and nutmeg increased either the mortality or the time required to elicit ALB mortality. This suggests that it is the EO content rather than the amount of powder deposited on the insect that is eliciting the effects on the adult ALBs. Since adult ALBs will come in contact with the plant powders as they emerge from and interact with the cocoon matrix during the spring incubation, we would suggest applications equivalent to the low concentration to ensure that the adult ALBs are not adversely affected.
Overall, the results suggest that clove, nutmeg and rosemary pose the highest risk to adult male ALBs, while Ajwain, cinnamon, fennel, turmeric, cumin, ginger and oregano are good candidates to apply to cocoons for reducing P. venustus populations.
2.028, p = 0.0212) significantly increased adult ALB mortality compared with the treatment mean, and these losses are above those reported with Vapona ® applications [8]. As with the lower contact results, clove elicited high ALB losses at 96 hrs, but was highly variable, suggesting that there may be a genetic or management component to the regional differences in bee sensitivity. The hierarchy for the relative potency of the negative effects of the plant powders on adult ALBs is nutmeg > cumin, fennel, rosemary > clove. Only nutmeg poses a higher risk for ALB losses than the currently applied insecticide and, therefore, has been eliminated as a candidate for parasitoid control.

Discussion
Plant powders and their corresponding EOs have been investigated extensively, with their biggest successes in controlling stored product pests [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29]. In the current study, the stored commodity is not a grain but ALBs, and the pest insect is a parasitoid (P. venustus). This lends an added question as to the safety or risk for the commodity, developing ALBs. Current control strategies for P. venustus involve applying Vapona ® (active ingredient, Dichlorvos) at day 7 of the 30-day spring incubation period to overlap with the emergence of the first generation of P. venustus [3,7]. The Vapona ® is removed on day 13 and the incubators are vented to release any remaining vapor. This approach suppresses the rate of parasitism by the first generation of parasitoids by about 67% [8]. Concerns over the safety of the Vapona ® vapor for farmers in the confined incubators, together with the risk of the residues for the developing ALBs, provided the impetus to investigate alternative control measures for this parasitoid. Previous studies suggest that several EOs are harmful to beneficial parasitoids (Braconidae [21], Platygastridae [33], Pteromalidae [22][23][24][25]31,32,34], Trichogrammatidae [23,30,34,35]) applied in IPM programs for stored product pests. We investigated the contact toxicity of fifteen plant powders from seven plant families against adult P. venustus, but also examined whether exposure to the plant materials was safe for adult male ALBs. We identified nine plant powders that were toxic to adult P. venustus, including Ajwain, cinnamon, clove, cumin, fennel, ginger, nutmeg, oregano and turmeric. Of these, nutmeg and rosemary pose a high risk for adult male ALB toxicity, while clove poses a high risk for adult male ALB toxicity if the ALBs are from specific regions within Alberta. Therefore, nutmeg and rosemary should be eliminated from further studies, while clove should be investigated with caution.
Typically, the next phase of investigation for the efficacious plant powders is to extract the EOs and determine the toxicity against P. venustus and ALBs [12,19,20,29]. However, plant powders have several advantages over EOs within commercial ALB production; they are easy to obtain, cost effective, easy to incorporate at harvesting from the nest blocks, safe for the farmer, less toxic than EOs [11][12][13][14] and are safer for the developing ALBs. In addition, the EO vapors are more unstable than plant powders, undergoing oxidative damage, chemical transformations or polymerization reactions when exposed to heat, water, light or air [37,38]. Under the conditions employed during the spring incubation of the ALB cocoons, EOs would degrade rapidly (30 • C, 50% relative humidity), not penetrate the cocoon layer and potentially concentrate within the cocoons at the surface of the trays, thus harming the developing ALBs. The greater stability of the plant powders is a benefit and must be balanced by its lower potency when compared with EOs. A factor that must be addressed for commercial applications is the EO content of plant powders [12], which vary depending upon the plant source [13,23] and the manufacturing process [12,20]. This can be overcome at the commercial level by adding an additional step, testing the plant powder, prior to the large-scale spring incubation to ensure efficacy against the pest without harming the ALBs. Although this would be a challenge for some agricultural sectors, this is feasible for managers of ALBs.
To be effective fumigants, the EOs from the plant powders must vaporize under ALB spring incubation conditions. In this scenario, the released vapor can travel throughout the cocoon matrix, poisoning the parasitoid larvae or emerging adult parasitoids. In one earlier study, plant powders from the family Lamiaceae reduced the rate of parasitism or the reproductive activity of a beneficial Pteromalid wasp (24%), and reduced larval parasitoid development (28%), which is suggestive of reproductive toxicity and fumigant properties [31,32]. Many studies have monitored the fumigant properties of EOs against beneficial parasitoids. However, these studies used EOs from plant genera and/or species that are not available in Canada, except for basil [22], ginger [23,35], oregano [33,35] and thyme [33,35]. We demonstrated significant adult P. venustus mortality after exposure to oregano and ginger powders, but not with thyme or basil powders. EOs from several other plant families (Apiaceae, Lauraceae, Mytraceae, Zingiberaceae, Lamiaceae) are effective fumigants for beneficial parasitoids or Coleopteran pests of stored products, such as grains .
Exposure of Trichogramma wasps to the LC01 of an EO from a member of the Apiaceae family reduced female longevity to 4.5 days post-exposure compared with 10 days for untreated parasitoids [30]. In the current study, members of this family (Ajwain, cumin and fennel) elicited high toxicity for adult P. venustus, peaking between 3 and 4 days post-exposure and indicating that the plant powders are releasing sufficient EOs to act as effective fumigants. These powders had equivalent or higher toxicity against P. venustus compared with Vapona ® (67%), supporting their potential for commercial applications.
The EOs from several members of the Lamiaceae or Myrtaceae families cause high adult mortality for Braconid [21], Pteromalid [22,31,32], Platygastrid [33] and Trichogramma wasps [35]. We examined the impact of two other members of the Lamiaceae family, rosemary and sage plant powders, which have strong fumigant properties for pest insects [26]. These powders did not achieve equivalent mortality to Vapona ® applications, regardless of contact concentration, eliminating them as candidates for commercial applications. In contrast, a member of the Myrtaceae family, clove, was highly toxic to adult P. venustus, but it also harmed the adult ALBs, making this a less favorable candidate than other efficacious plant powders with no effect on the ALBs.
There is one study regarding the Zingiberaceae family and EO activity against beneficial wasps. Increasing concentrations of ginger oil (Zingiberaceae) cause significant larval and adult mortality of Anisopteromalus wasps, but had no effect on larval stages of a Trichogramma wasp [23]. We confirmed that ginger powder caused a level of toxicity for adult P. venustus equivalent to Vapona ® , supporting its potential use in commercial applications. Interestingly, ginger oil extracted from a different plant species than that used in the current or an earlier study caused significant mortality of immature Trichogramma wasps [35], suggesting that the source of the plant powder or EO can affect outcomes.
There is no information regarding the fumigant properties of turmeric (Zingerberaceae) and cinnamon (Lauraceae) powder against beneficial parasitoids. However, the toxicity of turmeric and cinnamon powders was evaluated for several stored product pests [13,16,18,27] with highly variable results depending upon the stored product pest examined. A 1% cinnamon and turmeric powder application was ineffective against the rice weevil, but did cause significant mortality at 5% concentration, 90% and 32%, respectively [18]. In contrast, 2.5% cinnamon or turmeric powder applications elicit significant mortality for the cowpea weevil [18]. The unpredictable nature of the response is further illustrated by a very low concentration of 0.5% cinnamon powder, causing 100% Khapra and red flour beetle mortality [13]. Regardless, both cinnamon and turmeric powders elicit significant adult P. venustus mortality, about 20% higher than that achieved with Vapona ® applications. The variation in the reported responses to plant powders is suspected to be a function of whether a matrix, such as grain, is present or absent in the bioassay containers, with higher mortalities achieved in grain-free containers [12]. If we assume that the toxicity will decline within the mixed environment of the cocoon matrix, then only those plant powders with higher toxicity than Vapona ® applications should be considered for further investigation. Ajwain, cinnamon, clove, fennel, ginger and turmeric meet this criterion, with caution as previously noted regarding clove because of its adverse effects on ALBs.
Plant powders with fumigant and repellant properties are more effective against stored product pests than plant powders displaying only one mode of action [13,17,18,27]. Ajwain is a strong repellant for Trichogramma wasps [30]. For pest insects, cinnamon bark is a repellant for coleopteran beetles, while ginger rhizomes and stems are repellent for aphids [29]. Repellency is also measured by the effects of plant powders on the size of the F1 populations related to the rejection of the oviposition site and reproductive toxicity. A 5% concentration of cinnamon, clove, cumin, fennel, ginger and turmeric significantly reduces the F1 generation of rice weevil between 35 and 100% [16]. Similarly, 0.85% turmeric and cinnamon powder effectively reduces the F1 generation for the lesser grain borer [27]. Other pests, such as red flour beetle and Khapra beetle, require higher concentrations of cinnamon powder (1-5%) to significantly reduce the F1 generation [13]. In the current study, Ajwain, cinnamon, clove, cumin, fennel, ginger, oregano and turmeric are repellant as reflected in their reproductive toxicity for P. venustus.
Overall, the dual mechanisms displayed by Ajwain, cinnamon, cumin, fennel, ginger, oregano and turmeric against P. venustus, together with their safety for ALBs, lend further support for their potential use in protecting stored developing ALBs from parasitoid attacks during the spring incubation.

Conclusions
For over 20 years, an organophosphate insecticide, Vapona ® , has been used to control adult P. venustus attacking developing ALBs during spring incubation in commercial operations. This study identified several candidate plant powders with dual mechanisms for targeting and reducing P. venustus populations that are safe for the adult male ALBs. Further studies are required to ensure their safety for female bees and to understand the mechanisms underlying the toxicity recorded for the efficacious plant powders.