Methodology for Assessing Progeny Production and Grain Damage on Commodities Treated with Insecticides

: In evaluating insecticides, progeny production on grain commodities can be evaluated by either exposing adults on a commodity for a given time period, then removing them and assessing mortality and progeny production, or by leaving the adults on the commodity continuously, and then assessing progeny production. Little research directly compares these methodologies. Thus, our aims were to: 1) determine residual efficacy of Diacon IGR+ (methoprene+deltamethrin) and Gravista (methoprene+deltamethrin+piperonyl butoxide) on wheat, corn, and brown rice over the course of a year, using bioassays with select stored product insects at different time intervals, and 2) directly compare the two different methods of parental adult exposure on progeny production. Adults were either exposed for 7 d, then removed and assessed for survival, and the commodities were held for 6–7 weeks to assess progeny production, or adults were continuously exposed on the commodities for 6–7 weeks. Commodities were aged and sampled every 3 months for 12 months. Afterwards, samples were examined for progeny, sample weight loss, and insect feeding damage. Each insecticide killed exposed adults and prevented progeny of Rhyzopertha dominica on wheat and brown rice, and Tribolium castaneum on corn. There was extensive survival of Sitophilus spp. on all commodities, though Gravista did initially suppress S. oryzae on wheat and S. zeamais on corn compared to Diacon IGR+. Progeny, weight loss, and insect feeding damage were positively correlated in the 7 d exposure compared with continuous parental exposure. Both insecticides will control R. dominica and externally ‐ feeding insects, but may exhibit reduced effectiveness for Sitophilus spp., especially S. oryzae . Food managers can utilize these data to more effectively plan management programs.


Introduction
The insect growth regulator (IGR) methoprene has a long history of use as an insecticide in many agricultural systems throughout the world [1]. It has been evaluated and used alone and in combination with contact toxicants for many years to control stored product insects in Australia [2][3][4]. It was originally registered in the United States (US) as a grain protectant in the mid-1980s under the trade name Altosid ® , but that particular product was later withdrawn from the market [5]. Methoprene was reintroduced under the trade name Diacon II ® , at application rates of 1, 2,5, and 5 ppm for stored grains. This product was in turn replaced by Diacon IGR ® , at application rates of 1.25 and 2.5 ppm.
IGRs are only effective on immature insects and will not kill adults, a fact that must be acknowledged when conducting research with or promoting the use of IGRs. However, the from the incubator and held for about a week in a freezer at about −18 °C. Subsequently, the vials were removed from the freezer, and allowed to warm on a laboratory counter. The wheat in each vial was first weighed on a Mettler balance. The wheat was then sifted to remove adult progeny, along with the frass and feeding damage. The wheat was weighed again, the frass was weighed, and number of adults was recorded. Progeny production for the group of R. dominica and S. oryzae that were continually exposed on the wheat was determined by subtracting the 10 original parental adults. After this process was completed, the wheat was discarded. This entire procedure was repeated by sampling the wheat, collecting the wheat for each replicated treatment in the vials, introducing the groups of R. dominica and S. oryzae held for 7 days or continuously exposed on the wheat. Samples were processed at 0, 3, 6, 9, and 12 months after initial treatment of grain.

Insecticide Treatment and Bioassay Procedures for Corn
The corn was treated on 13 August 2018. The same treatments, untreated controls, the low and high rates of Diacon IGR+, and the one rate of Gravista were also used for corn with five replicates for each treatment. The label rate for treatment of the low rate of Diacon IGR+ on corn was 266 mL in 18.75 L for 25,454 kg, which was equivalent to 0.052 mL for 5 kg of corn in a volumetric ratio of 3.7 mL in proportion to the label. Due to difficulties in measuring this small amount, 0.35 mL of formulation was mixed with water in a 25 mL flask, and for each replicate, 3.7 mL of the formulation was dispensed onto the corn using the airbrush system described for wheat. The amount of formulation was doubled for the high rate of Diacon IGR+. Thus, the amount of Gravista specified for treatment of corn was 1062 mL per 25,454 kg, 0.21 mL in 3.7 mL water, or 0.56 of the formulation in 10 mL. The same volume of 3.7 mL described for the previous treatments was dispensed onto the corn. The treatment procedures were the same as described above for wheat regarding the dispensing of the insecticides (or untreated controls) onto the corn, placing each replicate in separate 18.75 L plastic buckets, and storing them on the floor of a different empty grain bin at the CGAHR.
The corn was first sampled after one day, as described for wheat. About 80 g of corn was collected into each of four 120-mL vials for each replicate and treatment (80 total). Ten mixed-sex adult 1-2-week-old T. castaneum or S. zeamais were placed in one vial for each of the four replicated treatments (20 total) and held for 7 d or put into the final set of vials for continuous exposure. The process of holding the corn in the chamber, and the subsequent analyses at 3, 6, 9, and 12 months, were the same as described for wheat.

Insecticide Treatment and Bioassay Procedures for Brown Rice
The brown rice was treated on 17 September 2018 with the same four treatments with five replicates as described above. Label specifications for treatment of brown rice with the low rate of Diacon IGR+ state application of 212 mL of formulation into 18.75 mL to cover 20,454 kg, which is a volumetric spray rate of 4.6 mL per 5 kg of brown rice. The amount of 0.052 mL of Diacon IGR+ needed was again too small to be formulated into 4.6 mL, so 0.28 mL was formulated into a 25 mL flask, and the airbrush system was used to dispense the 4.6 mL needed for each replicate and to treat the brown rice as described above. The amount of formulation was doubled for the high rate. The specifications for the Gravista were 845 mL of the formulation per 20,454 kg, 0.21 mL/4.6 mL, or 0.45 mL in 10 mL, dispensed as 4.6 mL for each replicate. The five untreated control replicates were sprayed with 4.6 mL of distilled water. The process of treating the rice, transferring the rice to buckets and holding in the grain bin, and the subsequent sampling assessment procedures was as described for wheat and corn. The insect species were R. dominica and S. oryzae.

Statistical Analysis
Data for the biological efficacy studies on each of the commodities were analyzed using Mixed-Model Procedures (e.g., proc mixed) in the Statistical Analysis System (SAS Version 9.4, SAS Institute, Cary, NC, USA). Data for wheat and brown rice were separated by species, with main effects for wheat and brown rice treatment and residual bioassay month, with analysis variables, progeny production in the 7 day exposures versus the continuous exposures, feeding damage, and frass weight. Interactions and random effects were also included in the Analysis of Variance (ANOVA). Data were also separated by species for the tests with corn and analyzed as described above for wheat and brown rice. The Means Procedure in SAS was used to generate mean values for comparisons, and these treatment means were separated using Tukey's Honest Significant Different Test, at  = 0.05, as an option under the Proc Mixed Procedure in SAS. When two treatments were compared, the t-test Procedure in SAS was used. Since post-treatment months were an ordered sequence, the data were not analyzed by mean separation tests, but instead were analyzed using month as a main effect, and separating data by month. A series of analyses using Proc Corr in SAS was also done for each commodity to correlate the initial parental mortality of each insect species with progeny production, correlate progeny production, sample weight loss, and frass weight of samples when parental adults were exposed for 7 days with the samples when adults were continuously exposed, and combining all samples and correlating progeny production with sample weight loss and progeny production, and also correlating frass weight with sample weight loss. Data were combined for all treatments for these correlation analyses.

Wheat
The ANOVA analyses for R. dominica and S. oryzae after the 7 day exposures are shown in Table  1. Main effects for month, treatment, and their interactions, were generally significant for the percentage of live adults, adults that were knocked down, or adults that were dead for R. dominica, except for the main effect of month for live adults. All main effects and interactions were significant for S. oryzae. The percentage of live R. dominica in untreated controls after the 7 day exposure ranged from 92% to 100% depending on month, with few knocked down or dead (Table 2). There were only a few live adults in any of the treatments at bioassay months 0-9, and mortality was generally over 90%. At month 12, there was an increase in percentage knockdown with a corresponding decrease in mortality in all of the treatments. There were no instances where there were significant differences in live adults, adults knocked down, and dead adults between the three insecticide treatments. At month 0 (one day after treatment), the percentage of live S. oryzae in the Gravista treatment was less than in controls or the two rates of Diacon IGR+, but after that time, there were live adults in all treatments, and in some instances, the percentage of live adults was greater in the Gravista treatment than in the highest rate of Diacon IGR+ (Table 2). At months 9 and 12, most of the adults exposed on the treated wheat were alive or knocked down, with few dead adults in any of the treatments. There were no F1 progeny for R. dominica when parental adults were either exposed continuously on the treated wheat or removed after the 7 day exposures, so two-way t-tests were done by month to determine differences in the two exposure regimes on number of F1 progeny, percentage of sample weight loss caused by adults and/or progeny, and feeding damage in untreated controls (Table 3). Progeny production ranged from 47.6 to 33.8 individual adults, depending on month and exposure, and was greater in samples where adults were continually exposed compared with removal after 7 days. Percentage weight loss ranged from 1.65% to 6.27% and was greater in samples where parental adults were continuously exposed in three of the five possible comparisons. Feeding damage ranged from 0.80 g to 4.51 g and was greater in samples where parental adults were continuously exposed in four of five comparisons. Table 3. Number of F1 adult progeny, % of sample weight loss due to insect feeding, and feeding damage (frass) (mean ± SE for all) from exposure of ten mixed-sex parental adult R. dominica exposed continuously for about 8 weeks (Cont.) on ~80 g of untreated wheat (UTC), or parental adults were removed after 1 week and samples were held for an additional 8 weeks (7 day). Data are separated by month. No progeny development on wheat treated with low and high rates of Diacon IGR+ or the Gravista formulation; no overall ANOVA done, only comparison between the two parental exposure regimes. Data are separated by month, bioassays were done at 1 day (month 0), and 3, 6, 9, and 12 months after treatment 1 . A complete ANOVA analysis was done for S. oryzae, with main effects for month, treatment, and exposure regime, and all their interactions, for variables F1 progeny, sample weight loss, and feeding damage ( Table 4). All main effects were significant for all three variables, while interactions were significant only for month by treatment and month by exposure for F1 progeny, and only month by treatment for weight loss and feeding damage. A series of analyses was then done by separating the two exposure regimes and analyzing for differences between treatments (including controls) for F1 progeny, sample weight loss, and feeding damage (Table 5). At months 0-6, there were more progeny in untreated controls and the low rate of Diacon IGR+ compared to the high rate and Gravista treatments, but there were few differences at 9 and 12 months. After month 0, average progeny ranged from a low of 22.8 in the high rate of Diacon IGR+ to a high of 401.7 in the low rate, with month to month variation. At 12 months, progeny were so numerous in untreated controls that the samples could not be processed due to excessive fungal development, along with the high progeny numbers. Values for sample weight loss and feeding damage followed the same general pattern as shown for F1 progeny development, with little difference between untreated controls and treatments at month 9, but with slightly less weight loss and feeding damage at month 12 in the Gravista treatment for the 7 day parental exposures. Overall, the F1 progeny numbers were generally comparable to or greater than F1 progeny of R. dominica, but the sample weight loss and feeding damage caused by S. oryzae was less than that caused by R. dominica. In contrast to R. dominica, frass weight never exceeded one gram in the S. oryzae samples for any treatment including controls. Table 4. ANOVA Table for main effects month (MO), Treatment (untreated controls, two rates of Diacon IGR+, and the Gravista formulation product), and count (CT, parental insects removed after 7 days or left continuously on the wheat for 8 weeks), and all associated interactions, for S. oryzae F1 progeny, % of sample weight loss, and weight in grams of feeding damage.  Table 5. Number of F1 adult progeny, % of sample weight loss due to insect feeding, and feeding damage (frass) (mean ± SE for all) from exposure of ten mixed-sex parental adult S. oryzae exposed on ~80 g of untreated wheat (UTC) wheat or wheat treated with the low rate of Diacon IGR+ (LDiacon IGR+, 0.5 ppm deltamethrin + 1.25 ppm methoprene), the high rate of Diacon IGR+ (HDiacon IGR+, 1.0 ppm+2.5 ppm methoprene), or the Gravista. Parental adults were either continuously exposed for about 8 weeks (Cont.) or were removed after 1 week and samples were held for an additional 8 weeks (7 day). Data are separated by month, bioassays were done at 1 day (month 0), and 3, 6, 9, and 12 months after treatment 1,2 . Survival of parental R. dominica was strongly correlated with progeny production, but there was considerable variation in F1 progeny even when survival was 100% ( Figure 1A). Survival of parental S. oryzae was also correlated with progeny production but there was more variation at levels where parental survival was less than 100% compared to R. dominica, and also variation when survival was 100% ( Figure 1B). Progeny production of both species in the 7 day exposures was correlated with production in the continuous exposures, but with more variation, and hence, a lower r value in S. oryzae (Figure 2A,B). Similarly, sample weight loss and frass weight was correlated for R. dominica and S. oryzae ( Figure 2C-F), with stronger correlation and greater r values for R. dominica compared to S. oryzae. For the final set of correlations, progeny production of both species was strongly correlated with sample weight loss ( Figure 3A,B) and frass weight ( Figure 3C,D), and frass weight was correlated with sample weight loss ( Figure 3D

Corn
The ANOVA for the 7 day exposures of T. castaneum and S. oryzae is shown in Table 6. Except for the main effect for month and variable live adults of T. castaneum, all main effects and interactions were significant for percentage live, knocked down, or dead adults after exposure. A series of oneway ANOVAs was done next by month for the three variables (Table 7). All exposed T. castaneum adults on untreated controls were alive after 7 days at all bioassay months except for month 12. The percentage of live, knocked down, and dead adults in the treatments varied considerably with month; for example, the average of dead adults on the low rate of Diacon IGR+ was 92% at month 6, and 0% and 56.9% at months 3 and 12, respectively. The percentage of live adults in the Gravista treatment was always 0 but rarely lower than the percentages of live adults in the two Diacon IGR+ treatments. All S. zeamais in the untreated corn were alive after the 7 day exposures ( Table 7). The percentage of live adults was lower in the Gravista treatment compared to the two Diacon IGR+ treatments at months 0, 3, and 12, but there were no differences between treatments at month 6. Knockdown in the two Diacon IGR+ treatments ranged from 0 to 69.1 at months 0 and 3 but at 6 and 12 months did not exceed 18.2%. Similarly, the percentages of dead S. zeamais seemed to fluctuate with month in all three insecticide treatments but was generally greatest in the Gravista treatment.  The ANOVA for main effects of month, treatment, and exposure condition for variables F1 progeny, sample weight loss, and weight of feeding damage were all significant at P < 0.001, except with weight loss by month due to T. castaneum (Table 8). Only about half of the interactions were significant for both species. The series of one-way ANOVAs was then done by month and exposure conditions for both species. Tribolium castaneum did not complete develop on corn treated with any of the three insecticides, and sample weight loss and feeding damage was negligible, thus data are reported only for untreated controls; however, progeny production to the adult stage was low even on the untreated controls, except for month 12, likely because whole grains are not a preferred food source for T. castaneum ( Table 9). The only significant difference for progeny production occurred at month 3, when progeny production was unexpectedly greater when the parental adults were removed after 7 d compared to the continuous exposure of parental adults. Weight loss and feeding damage was greatest for the continual exposures of parental adults at months 6-12, and all months except month 9, respectively. Average progeny production of S. zeamais in the three insecticide treatments ranged from 0 to 67.2, fluctuated with month, and was generally lower than progeny in untreated controls; however, progeny production was rarely different among the three insecticide treatments in either the continuous parental exposures or samples whereby the parental adults were removed after 7 days (Table 10). Similarly, the percentage weight loss and amount of feeding damage in the insecticide treatments was usually less than in the controls, but rarely different from each other except for scattered samples during the 12 month testing period. Table 8. ANOVA Table for main effects month (MO), treatment (untreated controls, two rates of Diacon IGR+, and the Gravista formulation), and count (CT, parental insects removed after 7 days or left continuously on the corn for 8 weeks), and all associated interactions, for T. castaneum and S. zeamais F1 progeny, % of sample weight loss, and weight in grams of feeding damage.  Table 9. Number of F1 adult progeny, % of sample weight loss due to insect feeding, and feeding damage (frass) (mean ± SE for all) from exposure of ten mixed-sex parental adult T. castaneum exposed continuously for about 8 weeks (Cont.) on ~80 g of untreated corn (UTC), or parental adults were removed after 1 week and samples were held for an additional 8 weeks (7 day). No progeny development on corn treated with either rate of Diacon IGR+ or the Gravista formulation, no overall ANOVA done, only comparison between the two parental exposure regimes. Data are separated by month, bioassays were done at 1 day (month 0), and 3, 6, 9, and 12 months after treatment 1 .  Table 10. Table 5. Number of F1 adult progeny, % of sample weight loss due to insect feeding, and feeding damage (frass) (mean ± SE for all) from exposure of ten mixed-sex parental adult S. zeamais exposed on ~80 g of untreated corn (UTC) wheat or corn treated with the low rate of Diacon IGR+ (Ldiacon IGR+, 0.5 ppm deltamethrin + 1.25 ppm methoprene), the high rate of Diacon IGR+ (Hdiacon IGR+, 1.0 ppm+2.5 ppm methoprene), or the Gravista formulation (methoprene + deltamethrin + piperonyl butoxide synergist). Parental adults were either continuously exposed for about 8 weeks (Cont.) or were removed after 1 week and samples were held for an additional 8 weeks (7 day). Data are separated by month, bioassays were done at 1 day (month 0), and 3, 6, 9, and 12 months after treatment 1,2 . Survival of parental T. castaneum was strongly correlated with progeny production, but even when there was parental survival, there was no progeny production, and also variation when survival was 100% ( Figure 4A). The relatively low progeny production on corn perhaps contributed to the variation, which was not surprising given the low reproductive capacity of this species on whole grain kernels. Survival of parental S. zeamais was also correlated with progeny production, but similar to S. oryzae on wheat, there was considerable variation, where parental survival ranged from 20% to 80%, and also variation when survival was 100% ( Figure 4B). Progeny production of both species in the 7 day exposures was correlated with production in the continuous exposures, with more variation in S. zeamais, as reflected by the low r values ( Figure 5A,B). Sample weight loss and frass weight was correlated between the two parental exposure regimes for both species, with stronger correlation and greater r values for T. castaneum compared to S. zeamais ( Figure 5C-F). Progeny production of both species was strongly correlated with sample weight loss ( Figure 6A,B) and frass weight ( Figure 6C,D), and frass weight was correlated with sample weight loss ( Figure  6D,E), but with lower r values for T. castaneum compared to S. zeamais. The r values for all correlations in Figure 6 Figure 6. Correlation of corn (C) sample weight loss and frass weight with F1 progeny of T. castaneum (A,B), sample weight loss with frass weight (C), sample weight loss and frass weight with F1 progeny of S. zeamais (D,E), and sample weight loss with frass weight (F). Data for the two parental exposure regimes combined, data for all treatments and all bioassay months combined.

Brown Rice
The ANOVA analyses for R. dominica and S. oryzae after the 7 day exposures on untreated and treated brown rice are shown in Table 11. Main effects month, treatment, and the interactions were generally significant for the percentage of live adults, adults that were knocked down, or adults that were dead for R. dominica, except for main effect month for knockdown of exposed adults. Main effects month and treatment were significant for S. oryzae for live and dead adults, but treatment was not significant for knockdown. Some but not all interactions were significant for S. oryzae. Survival of R. dominica on untreated brown rice was 100% except for month 9. There were generally no live adults in any of the chemical treatments except for the low rate of Diacon IGR+, and only one instance where live adults were present after 7 d of exposure on the brown rice treated with the Gravista product. There were adult R. dominica that were knocked down or dead in the treated brown rice at all bioassay points, and in most comparisons, the treatment values were different from controls, but there were no differences among the three insecticide treatments for percentage knocked down or dead. Except for month 3, the percentage of live S. oryzae after the 7 day exposures exceeded 90% in all three insecticide treatments, and were usually not significantly different from the untreated controls or from each other (Table 12). Thus, the percentage of knocked down or dead adults was also low in untreated controls and in the three insecticide treatments. The ANOVA Table for F1 progeny production, sample weight loss, and frass weight for R. dominica and S. oryzae are presented in Table 13. Main effects treatment and exposure were significant for progeny production of R. dominica, but not month, while all main effects were significant for sample weight loss and frass rate. About half of the interactions were significant. Main effect treatment was not significant for S. oryzae for any of the three analysis variables. Table 13. ANOVA Table for main effects month (MO), treatment (untreated controls, two rates of Diacon IGR+, and the Gravista formulation), and count (CT, parental insects removed after 7 days or left continuously on the corn for 8 weeks), and all associated interactions, for R. dominica and S. oryzae F1 progeny, % of sample weight loss, and weight in grams of feeding damage 1 . There was no progeny production of R. dominica in any of the three treatments, similar to the results for the exposures on wheat, thus, data were analyzed only by exposure method for untreated controls (Table 14). There were no differences in progeny production, and only one occasion each when weight loss and frass was greater in the continual parental exposures versus the 7 day exposures. Since the overall ANOVA showed no treatment effects for S. oryzae on brown rice, data for treatment were combined and analyzed by month (Table 15). Samples for month 12 could not be processed because the progeny numbers were so great that the rice was clumped with heavy fungal development. Progeny production ranged from 94.4 to 432.0, and was greater in the continual parental exposures compared to the 7 day parental exposures at months 0 and 3 but not at months 6 and 9. Weight loss was greater in the continual exposures at month 0, but greater in the 7 day exposures at month 9. Although frass weight was significant in the overall ANOVA, none of the comparisons between continual parental exposure and 7 day parental exposures were different. Table 14. Number of F1 adult progeny, % of sample weight loss due to insect feeding, and feeding damage (frass) (mean ± SE for all) from exposure of ten mixed-sex parental adult R. dominica exposed continuously for about 8 weeks (Cont.) on ~80 g of untreated brown rice (UTC), or parental adults were removed after 1 week and samples were held for an additional 8 weeks (7 day). No progeny development on brown rice treated with low and high rates of Diacon IGR+ or the Gravista formulation, no overall ANOVA done, only comparison between the two parental exposure regimes. Data are separated by month, bioassays were done at 1 day (month 0), and 3, 6, 9, and 12 months after treatment 1 .  Table 15. Data for S. oryzae combined for treatments since overall ANOVA showed no treatment effect. Number of F1 adult progeny, % of sample weight loss due to insect feeding, and feeding damage (frass) (mean ± SE for all) from exposure of ten mixed-sex parental adult S. oryzae exposed continuously for about 8 weeks (Cont.) on ~80 g of untreated brown rice, or adults removed after 7 days of exposure (7 day). Only comparison between the two parental exposure regimes were done. Data are separated by month, bioassays were done at 1 day (month 0), and 3, 6, 9 months after treatment 1 . Survival of parental R. dominica on brown rice was strongly correlated with progeny production, similar to results with this species exposed on wheat, but even when survival was 100%, there was still variation in progeny production ( Figure 7A). Survival of parental S. oryzae was also correlated with progeny production, with considerable survival on all treatments, very high progeny production, and variation in progeny production with 100% parental survival ( Figure 7B). Progeny production of both R. dominica and S. oryzae in the 7 day exposures was correlated with production in the continuous exposures, with more variation in S. oryzae as reflected by the low r values ( Figure  8A,B). Sample weight loss and frass weight was correlated between the two parental exposure regimes for both species, with stronger correlation and greater r values for R. dominica compared to S. oryzae ( Figure 8C-F). Progeny production of both species was strongly correlated with sample weight loss ( Figure 9A,B) and frass weight ( Figure 9C,D), and frass weight was correlated with sample weight loss ( Figure 9D

Discussion
There have been several recent studies with grain protectants whereby authors have exposed parental adults, obtained mortality values, and then, held the grains to obtain progeny assessments [16,17]. While this is a valid methodology, one of the objectives of this study was to see if continual exposure of parental adults altered the progeny production values, using methodology that has been used for assessments of IGRs alone where parental mortality is not expected [5]. Results from this study showed that exposing parental adults for one week to assess mortality, and then waiting to do progeny counts from those exposed parental adults, did not appreciably alter the relative differences between the treatments compared to the continual exposures. In fact, the opposite occurred because in some instances, the samples were so overloaded with progeny that damage assessments could not be performed. Progeny production, sample weight loss, and insect feeding damage were all correlated with the two exposure regimes for all target species on all three commodities. Thus, it would seem that providing some measure of adult mortality after exposure should be a valid component of experimental protocols for evaluating grain protectants.
Parental mortality was clearly correlated with progeny production of all target species. While this may be expected, there is evidence from previous studies with diatomaceous earth (DE) products that while adult mortality occurred after exposure, the time interval required for death allowed for parental oviposition, and subsequent progeny production [18,19]. This was especially prominent for Sitophilus weevils. Historical evidence and recent research have shown that pyrethroids, including deltamethrin, can be effective on Sitophilus weevils, but either high concentrations or long exposure times at current label rates may be necessary for complete kill [20]. Physical grain damage was also correlated with progeny production, emphasizing that the purpose of an insecticidal grain protectant should be to kill adults before they are able to oviposit.
The new Gravista formulation did provide some measure of increased control of S. oryzae on wheat and S. zeamais on corn, but comparisons between the two commodities can be misleading because insecticidal efficacy can vary depending on the specific grain. Kavallieratos et al. [21] exposed adults and larvae of Trogoderma granarium (Everts), the khapra beetle, on wheat, barley, maize, and rough rice treated with six different insecticides. Mortality of larvae and adults was generally greater for all insecticides on wheat and barley compared to maize and rough rice. In a similar study [22], greater mortality of adults and larvae of Tenebrio molitor (L.), the lesser mealworm, exposed on wheat and barley treated with pirimiphos-methyl, deltamethrin, spinosad and SilicoSec compared to the same insecticides applied to maize. Plus, a corn kernel is much larger than a wheat kernel, with about 30 wheat kernels per gram compared to 3 corn kernels per gram, depending on the variety. Thus, reproduction of Sitophilus spp., an internal feeder, on corn will be much less compared to what can be produced on wheat; there may also be inherent characteristics of the two species that affect reproduction.
The husk of rough rice will offer some protection from stored product insects [23]. In the current study, none of the treatments controlled S. oryzae on brown rice, which has lost the protective husk during the milling process. Brown rice is very susceptible to S. oryzae, and in previous studies with brown rice, progeny of S. oryzae quickly depleted the food resources in untreated controls used in insecticide experiments, often creating so much mold that the brown rice was turned into a gooey mass, similar to results for the 12 month data in this study [5,24,25]. Tribolium castaneum (Herbst), the red flour beetle, can also grow and develop on brown rice [24,25]. Since brown rice is most likely to be stored in bags rather than in bulk, as it is a product sold directly to consumers, alternative treatments for brown rice, including fumigation or cold storage, may be necessary to prevent economic loss. Storing brown rice in bags that are comprised of packaging in which deltamethrin is incorporated into the laminate layers of the packaging may also be an alternative to grain protectants or used in combination with a grain protectant [26,27]. Field studies in Ghana have shown the utility of these bags for protecting stored bagged maize in a tropical environment [28]. A new deltamethrinincorporated netting material is also effective on stored product insect adults [29] and immatures [30], and has recently been registered with the US Environmental Protection Agency.
The results of this current study also show that it is difficult to evaluate grain protectants on rough rice because of the protection provided by the husk leads to low parental survival and low progeny production of both R. dominica and S. oryzae [5]. In this cited study, untreated rough rice and brown rice, and rough rice and brown rice treated with methoprene alone were stored at ambient conditions for 24 months on the floor of a grain bin at the CGAHR. The rice, along with untreated wheat and corn, was bioassayed at bimonthly intervals. There was little progeny production of R. dominica on untreated rough rice until 18 months, which was most likely due to ageing of the rough rice and weakness developing in the husk. Both R. dominica and S. oryzae may require a crack or split in the rough rice husk for optimum development [23,31,32]. Rhyzopertha dominica can be controlled on brown rice with either methoprene or the combination methoprene-deltamethrin grain protectants [5,6]. However, control of S. oryzae on brown rice may require additional insecticide inputs.

Conclusions
Evaluation of insecticides can be done by exposing adults on treated grains and then removing them after a given time period, or continually exposing parental adults and examining progeny. Removing parental adults after exposure seemed to be the best method for evaluating the insecticides tested in our study. The combination treatments of the pyrethrin deltamethrin and the IGR deltamethrin, with and without piperonyl butoxide synergist, gave residual control of Rhyzopertha dominia and T. castaneum on different grains. There was some indication of increased control of S. zeamais on corn with the Gravista product compared to the IGR+. However, brown rice was extremely susceptible to S. oryzae, and none of the insecticides gave complete control of this species for the residual time period tested in this study. Brown rice is usually stored in bags inside warehouses in most developed countries, and additional insecticide treatments prior to bagging or fumigation during the time the bags are stored may be necessary to control Sitophilus spp. weevils.