COMMODITY TREATMENT AND QUARANTINE ENTOMOLOGY

Weathering and Chemical Degradation of Methyl Eugenol and Raspberry Ketone Solid Dispensers for Detection, Monitoring, and Male Annihilation of Bactrocera dorsalis and Bactrocera cucurbitae (Diptera: Tephritidae) in Hawaii Author(s): Roger I. Vargas, Steven K. Souder, Eddie Nkomo, Peter J. Cook, Bruce Mackey, and John D. Stark Source: Journal of Economic Entomology, 108(4):1612-1623. Published By: Entomological Society of America URL: http://www.bioone.org/doi/full/10.1093/jee/tov137 BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne s Terms of Use, available at www.bioone.org/page/ terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder. BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

COMMODITY TREATMENT AND QUARANTINE ENTOMOLOGY Weathering and Chemical Degradation of Methyl Eugenol and Raspberry Ketone Solid Dispensers for Detection, Monitoring, and Male Annihilation of Bactrocera dorsalis and Bactrocera cucurbitae (Diptera: Tephritidae) in Hawaii ROGER I. VARGAS, 1,2 STEVEN K. SOUDER, 1 EDDIE NKOMO, 3 PETER J. COOK, 3 BRUCE MACKEY, 4 AND JOHN D. STARK 5 J. Econ. Entomol. 108(4): 1612 1623 (2015); DOI: 10.1093/jee/tov137 ABSTRACT Solid male lure dispensers containing methyl eugenol (ME) and raspberry ketone (RK), or mixtures of the lures (ME þ RK), and dimethyl dichloro-vinyl phosphate (DDVP) were evaluated in area-wide pest management bucket or Jackson traps in commercial papaya (Carica papaya L.) orchards where both oriental fruit fly, Bactrocera dorsalis (Hendel), and melon fly, Bactrocera cucurbitae (Coquillett), are pests. Captures of B. dorsalis with fresh wafers in Jackson and bucket traps were significantly higher on the basis of ME concentration (Mallet ME [56%] > Mallet MR [31.2%] > Mallet MC [23.1%]). Captures of B. cucurbitae with fresh wafers in Jackson and bucket traps were not different regardless of concentration of RK (Mallet BR [20.1%] ¼ Mallet MR [18.3%] ¼ Mallet MC [15.9%]). Captures of B. dorsalis with fresh wafers, compared with weathered wafers, were significantly different after week 12; captures of B. cucurbitae were not significantly different after 16 wk. Chemical analyses revealed presence of RK in dispensers in constant amounts throughout the 16-wk trial. Degradation of both ME and DDVP over time was predicted with a high level of confidence by nonlinear asymptotic exponential decay curves. Results provide supportive data to deploy solid ME and RK wafers (with DDVP) in fruit fly traps for detection programs, as is the current practice with solid TML dispensers placed in Jackson traps. Wafers with ME and RK might be used in place of two separate traps for detection of both ME and RK responding fruit flies and could potentially reduce cost of materials and labor by 50%. KEY WORDS tephritid fly, oriental fruit fly, melon fly, male lure Spread of alien fruit flies (Diptera: Tephritidae) throughout the world has increased the need for improved detection, monitoring, and control systems (Vargas et al. 2010a, 2012). Early detection methods for accidental introductions of fruit flies in the United States (e.g., California and Florida) involves deployment of large numbers of Jackson traps baited with highly the attractive male-specific lures methyl eugenol (ME; 4-allyl-1, 2-dimethoxybenzene-carboxylate) and cue-lure [C-L; 4-(p-acetoxyphenyl)-2-butanone] to detect such invasive flies as oriental fruit fly, Bactrocera dorsalis (Hendel), and melon fly, Bactrocera cucurbitae (Coquillett), respectively (International Atomic Energy Agency [IAEA] 2003, Florida Department of Agriculture and Consumer Services [FDACS] 2003, California Department of Food and Agriculture [CDFA] 2010). C-L has never been isolated as a natural product but 1 Daniel K. Inouye, U.S. Pacific Basin Agricultural Research Center, USDA-ARS, 64 Nowelo St. Hilo, HI 96720. 2 Corresponding author, e-mail: roger.vargas@ars.usda.gov. 3 Farma Tech International, 1546 Boalch Avenue NW, Suite 30, North Bend, WA 98045. 4 USDA-ARS-PWA, Albany, CA 94710. 5 Washington State University, Puyallup Research and Extension Center, Puyallup, WA 98371. quickly hydrolyzes to form raspberry ketone (RK; 4-(p-hydroxyphenyl)-2-butanone; Vargas et al. 2010b). RK is thought to be the attractive molecule and is more persistent because it volatilizes at a lower rate than C-L (Metcalf and Metcalf 1992). The total number of sites in operation is often in the tens of thousands, for example >30,000 sites are monitored in the state of California (Vargas et al. 2010a). These trapping systems are used in association with liquid insecticide formulations such as naled (dimethyl 1, 2-dibromo-2, 2-dichloroethyl phosphate), which may pose shipping, environmental, and worker safety concerns (Vargas et al. 2009). Similarly, where established, monitoring systems are often important in determining timing of treatments for control (Mau et al. 2007). Finally, one important technique used in the suppression of fruit flies in areas of establishment is the use of the male annihilation technique (MAT) through mass trapping (Vargas et al. 2010a). Replacement of liquid lures and environmentally unfriendly organophosphate pesticides with safer alternative insecticides has been a primary goal of the Hawaii Fruit Fly Area-Wide Pest Management (AWPM) program (Vargas et al. 2008, 2010a, 2012). The Hawaii AWPM program promoted adoption of novel solid detection and MAT dispensers for safer and more effective methods of detection, monitoring, and Published by Oxford University Press on behalf of Entomological Society of America 2015. This work is written by US Government employees and is in the public domain in the US.

August 2015 VARGAS ET AL.: LONGEVITY OF SOLID ME/C-L/RK DISPENSERS 1613 control (Vargas et al. 2010a). In studies in Tahiti with B. dorsalis and Bactrocera tryoni (Froggatt), Queensland fruit fly, a C-L RK responding fly, traps baited with ME and C-L combined in a single Mallet MC wafer with DDVP (Farma Tech International, North Bend, WA) captured as many B. tryoni and B. dorsalis as individual traps baited with standard liquid ME and C-L with malathion formulations (Leblanc et al. 2011). In addition, results from an evaluation in Hawaii with solid triple lure dispensers (FT Mallet TMR) containing trimedlure (TML), ME, and RK (Shelly et al. 2012, Vargas et al. 2012) provide even more evidence for eliminating the use of hard-to-handle liquid lures and hazardous organophosphate liquids in standard trapping systems. Incorporating a multilure dispenser would reduce the number of traps. Solid TML dispensers have been used for detection and monitoring of Mediterranean fruit fly, Ceratitis capitata (Wiedemann), for many years. A single or multilure dispenser that attracts ME and C-L/RK responding flies to the same trap could potentially reduce the cost of trap and labor requirements by >50%. This could have a profound effect on detection or suppression programs in California, Florida, Australia, New Zealand, Pacific Island nations, Africa, South America, and Hawaii. The current study evaluates the performance of solid lure wafers with DDVP insecticide in individual and combined formulations in monitoring, MAT traps, or both under Hawaiian climatic conditions, where both B. dorsalis and B. cucurbitae were present. These lure wafers were used extensively in the Hawaii AWPM program and we did chemical analyses of many of the prototypes (Vargas et al. 2009). The overall goal of the present study was to provide efficacy and chemical formulation data for producing a more convenient, effective, and safer means to use male lures and insecticides for improved detection or suppression of invasive fruit flies that respond to either MEorC-L RK.Furthermore,thesesolidlurewafers could potentially be used effectively as male annihilation devicesaspartofanintegratedpestmanagement(ipm) program (Piñero et al. 2009, Vargas et al. 2010a) in areas of fruit fly establishment (e.g., Hawaii). These studies are unique in that they summarize: 1) chemical degradation data for the compounds present in the dispensers with time, 2) bioassays of the same weathered solid dispensers, and 3) weathered dispensers compared to fresh dispensers with time. Materials and Methods Field Bioassays. Evaluations were conducted by researchers at the U.S. Department of Agriculture (USDA), Agricultural Research Service (ARS), Daniel K. Inouye, U.S. Pacific Basin Agricultural Research Center, Hilo, HI, on Hawaii Island during the Hawaii AWPM program. The standard monitoring or detection traps used were the Hawaii AWPM bucket trap, which consists of a plastic 1-liter container (3.5 cm in radius and 15 cm in height; Highland Plastic no. 36; Highland Plastics Inc., Mira Loma, CA) with four lateral 3-cmdiameter holes used in the AWPM program and the Jackson trap with a sticky insert (both traps are illustrated in Vargas et al. 2010a) used in fruit fly programs on the U.S. mainland. The major emphasis of this study was on the Jackson trap for use in detection programs on the U.S. mainland. The AWPM data were collected for application in male annihilation against multiple fruit fly species in Hawaii IPM programs (Vargas et al. 2010a). Bioassays of four different solid lure treatments (against B. dorsalis and B. cucurbitae) were conducted from 10 May through 30 August 2010 in a commercial papaya orchard (100 ha) located near Keaau, HI (latitude 19 25 0 N; longitude 154 54 0 W) in the district of Puna of Hawaii Island, the major papaya growning region in the state. Very large populations of both B. dorsalis and B. cucurbitae were present in these orchards. Test materials were obtained from Farma Tech International (North Bend, WA) and each wafer was 7.7 by 5 cm and solid lure wafers (active ingredients [AI] by weight) were as follows: 1) FT Mallet ME (56% ME, 3.4% DDVP, one lure contained 8.0 g AI) wafer; 2) FT Mallet MR (31.2% ME, 18.3% RK, 3.7% DDVP, one lure contained 7.1 g AI) wafer; 3) FT Mallet MC (23.1% ME, 15.9% Benzyl Acetate, 15.9% RK, 3.6% DDVP, one lure contained 7.0 g AI) wafer; 4) FT Mallet BR (29.5% Benzyl Acetate, 20.1% RK, 3.7% DDVP, one lure contained 7.1 g AI). Benzyl acetate was used to increase the solubility of RK. Previous trials with C-L dispensers had indicated chemical reactions occurring inside the Mylar packages. Fresh and weathered solid lure dispensers were evaluated over a 16-wk period. Weathering. Solid lure dispensers were weathered simultaneously at the University of Hawaii Waiakea Agricultural Research Station, Panaewa, HI (latitude 19 38 0 N, longitude 155 5 0 W). This experiment station is 80 ha in size, has an elevation of 241 m with an annual rainfall of 4,130 mm and an average yearly temperature of 21.7 C (minimum 17.2 C, maximum 26.1 C). The station grounds contain various ornamental plant shade houses and tropical fruit tree orchards including: macadamia (Macadamia integrifolia Maiden & Betche), guava (Psidium guajava L.), lychee (Litchi chinensis Sonnerat), longan (Dimocarpus longan Table 1. Farma Tech insecticidal lures were weathered under tarps at Waiakea Agricultural Research Station, Panaewa, HI, from 10 May to 30 August 2010 Month Rainfall total (mm) Rainfall per day (mm) Temperature ( C) % RH Wind speed (m/s) May 160.37 5.17 6 1.0 21.27 6 0.05 85.61 6 0.19 0.54 6 0.01 June 154.29 5.19 6 1.33 21.55 6 0.05 85.88 6 0.18 0.30 6 0.01 July 170.90 5.51 6 1.12 21.97 6 0.06 85.22 6 0.20 0.30 6 0.01 Aug. 167.56 5.59 6 1.21 22.27 6 0.05 84.82 6 0.19 0.49 6 0.01 Summarized weather data are presented (mean 6 SEM).

1614 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 108, no. 4 Loureiro), papaya (Carica papaya L.), citrus (Citrus L.), plumeria (Plumeria L.), anthurium [Anthurium palmatum (L.) G. Don], and palm (Bactris gasipaes Kunth). Four separate weathering stations were made for each of the four treatments to minimize crosscontamination from responding wild fruit flies. Weathering stations were constructed 100 m apart within a palm orchard (0.4 ha) in a completely randomized block design. Four solid lure treatments were weathered separately under a tarpaulin (Foremost Tarps, Mdl# 110128, 10 by 12, Auburn, WA). Each weathering site had eight weathering lines, each attached between two poles (2 m in height; four Jackson trap lines and four AWPM trap lines) that contained 16 traps each. Each solid lure wafer was weathered inside Jackson or AWPM traps. In total, 128 solid lure dispensers (64 in Jackson traps and 64 in AWPM traps) were weathered for each of the four solid lure treatments. Weather data were collected with a Vantage Pro 2 (Davis Instruments, Hayward, CA; Table 1). Weekly bioassays for Jackson and AWPM traps were started on the same day each week between at 7:45 and 8:30 am. Traps were arranged in a randomized complete block design. To compensate for position effects, Jackson and AWPM trap blocks were alternated and traps within a block were rotated clockwise. Weathered treatments (four dispensers from four different lines) were placed in a new Jackson or AWPM trap for field evaluations. Traps were placed 20 m apart and hung from papaya trees. Jackson traps were serviced and removed 30 min after deployment. AWPM traps were serviced and removed 60 min after deployment. Jackson trap inserts were removed and AWPM traps were emptied into individually identified paper bags and returned to the laboratory for fruit fly counts. Immediately after each weekly bioassay, weathered solid lure wafers from Jackson traps were sent overnight for chemical analysis. Chemical Analysis. Only samples from Jackson traps were sent to Farmatech International Corporation (North Bend, WA) and used for chemical analyses. Each week, four samples from the four different dispensers were analyzed (16 wafers per week). Amounts of ME, RK, benzyl acetate, and DDVP remaining in weathered dispensers were determined using a Shimadzu GC 2010 gas chromatograph (GC) and Shimadzu GC Solution software. Analytical residue methods used in this study included weighing the lure, followed by placing a measured portion of the lure in 100 ml of chloroform for 48 h. A 3-ml sample of this was then taken and diluted in 10 ml of chloroform. The same was done separately with 3 ml of a calibration Fig. 1. Chemical analysis (A) of Mallet ME solid dispensers placed inside Jackson traps with corresponding (B) B. dorsalis (fresh vs. weathered) captures (mean 6 SEM) at Island Princess Papaya Orchards, Keaau, Hawaii Island, HI, from 10 May to 30 August 2010. Error bars in the same category with different letters indicate a significant difference (P 0.05), PROC TTest (SAS Institute 2013).

August 2015 VARGAS ET AL.: LONGEVITY OF SOLID ME/C-L/RK DISPENSERS 1615 standard, and then again with 3 ml of a check standard. The three samples were then analyzed in the following order: calibration standard, check standard, and sample, with the Shimadzu GC 2010 GC equipped with a split or splitless injection port, operated in the splitless mode at a temperature of 225 C. The column used was a nonpolar crossbond (100% dimethyl polysiloxane) Rtx-1 capillary column, 30 m by 0.53 mm by 1.5 mm and the carrier gas was helium (He) at a pressure of 4.7 psi using the Pressure Flow Control Mode. The oven temperature was programmed from 110 C(hold time 1 min) to 220 C at 8 C/min (hold time 9 min) with the flame ionization detector held at 275 C. The amounts of each active component in the check standard and samples were determined by an external calibration method with the active component of each check standard within 6 3% of the nominal value. Statistical Analysis. PROC GLM and PROC TTEST were used to compare data (treatment and control; SAS Institute 2013, Cary, NC). Data were transformed to (x þ 0.5) 1/2 or log (x) to stabilize variances when necessary, but untransformed data are presented. For PROC GLM, a post hoc test was performed using a Tukey s honestly significant difference (HSD) test. To calculate captures based on proportions, captures for each dispenser were divided by the total captures with ME or RK each week for B. dorsalis and B. cucurbitae, respectively. Proportions were pooled and analyzed by one-way ANOVA and Tukey s HSD test. ME and DDVP losses were expressed as the percentage loss over time (days). These percentages were then fitted to the asymptotic regression model: Y ¼ a þ b exp ð cdayþ In which, y ¼ % ME or % DDVP. Curves were fitted using PROC NLIN (SAS Institute 2013). A P 0.05 criterion was used as a significance level for all statistical tests. Results Captures (mean 6 SEM) of B. dorsalis with Mallet ME dispensers (fresh vs. weathered) were compared over 16 wk in Jackson traps with corresponding chemical analysis of Mallet ME (Fig. 1A and B). After 12 wk, there was a significant difference between fresh and weathered Mallet ME dispensers. Captures (mean 6 SEM) of B. cucurbitae with Mallet BR dispensers (fresh vs. weathered) in Jackson traps were analyzed over 16 wk and compared with corresponding chemical analysis of Mallet BR (Fig. 2A and B). There were no significant differences with week and amount of RK remained constant with time. Captures (mean 6 SEM) of B. dorsalis and B. cucurbitae with Mallet MR dispensers (fresh vs. weathered) in Jackson Fig. 2. Chemical analysis (A) of Mallet BR solid dispensers placed inside Jackson traps with corresponding B. cucurbitae (fresh vs. weathered) captures (mean 6 SEM) at Island Princess papaya orchards, Keaau, Hawaii Island, HI, from 10 May to 30 August 2010. Paired comparisons over the 16-wk period indicated no significant differences between fresh and weathered dispensers (P > 0.05), PROC TTest (SAS Institute 2013).

1616 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 108, no. 4 traps were analyzed over 16 wk along with the corresponding chemical analysis of Mallet MR (Fig. 3A, B, and C). There were no significant differences over 16 wk for B. cucurbitae captures. However, there were significant differences for B. dorsalis captures after 12 wk. The RK remained relatively constant over time, while the amount of ME and DDVP decreased rapidly. Captures (mean 6 SEM) of (B) B. dorsalis and (C) B. cucurbitae with Mallet MC solid lure dispensers (fresh vs. weathered) in Jackson traps with the corresponding (A) chemical analysis of FT Mallet MC were compared (Fig. 4 A, B, and C). There were no significant differences over 16 wk for B. cucurbitae captures except at 7 wk, where the weathered dispenser was significantly more attractive than the fresh dispenser. However, there were significant differences for B. dorsalis captures after 12 wk. The RK remained relatively constant over time, while ME and DDVP decreased more Fig. 3. Chemical analysis (A) of Mallet MR solid dispensers placed inside Jackson traps with corresponding B. dorsalis and B. cucurbitae (fresh vs. weathered) captures (mean 6 SEM) evaluated at Island Princess papaya orchards, Keaau, Hawaii Island, HI, from 10 May to 30 August 2010. (B) Error bars in the same category with different letters indicate a significant difference (P 0.05), PROC TTest (SAS Institute 2013). (C) Paired comparisons over the 16-wk period indicated no significant differences between fresh and weathered dispensers (P > 0.05), PROC TTest (SAS Institute 2013).

August 2015 VARGAS ET AL.: LONGEVITY OF SOLID ME/C-L/RK DISPENSERS 1617 Fig. 4. Chemical analysis of (A) Mallet MC solid dispensers placed inside Jackson traps with corresponding B. dorsalis (B) and B. cucurbitae (C) (fresh vs. weathered) captures (mean 6 SEM) evaluated at Island Princess papaya orchards, Keaau, Hawaii Island, HI, from 10 May to 30 August 2010. (B) Error bars in the same category with different letters indicate a significant difference (P 0.05), PROC TTest (SAS Institute 2013). (C) Error bars in the same category with different letters indicate a significant difference (P 0.05) PROC TTest (SAS Institute 2013). rapidly. Three different weathered solid lure dispensers were compared in (B) Jackson traps and (C) for B. dorsalis captures (mean 6 SEM) with the corresponding (A) degradation of ME from Jackson traps (Fig. 5 A, B, and C). B. dorsalis captures were equal or higher with the dispenser containing the highest amount of ME. Three different weathered solid lure dispensers were compared in (B) Jackson traps and (C) for B. cucurbitae captures (mean 6 SEM) with the corresponding (A) degradation of RK sampled from Jackson traps (Fig. 6 A, B, and C). B. cucurbitae captures were not significantly different for the three dispensers, regardless of the amount of RK contained in the dispenser. Total captures of B. dorsalis over 16 wk were significantly different on the basis of ME concentration (Mallet ME [56%] > Mallet MR [31.2%] > Mallet MC [23.1%]; Fig. 7A). Total captures of B. cucurbitae over 16 wk were not significantly different on the basis of RK concentration (Mallet BR [20.1%] ¼ Mallet MR [18.3%] ¼ Mallet MC [15.9%];

1618 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 108, no. 4 Fig. 5. Chemical degradation (A) of ME from three different weathered solid lure dispensers with corresponding captures (mean 6 SEM) of B. dorsalis inside (B) Jackson traps and (C) AWPM traps. Traps were hung in papaya trees at Island Princess papaya orchards, Keaau, Hawaii Island, HI, from 10 May to 30 August 2010. (B) Error bars in the same category with different letters are significantly different (P 0.05) PROC GLM, Tukey s HSD test (SAS Institute 2013). (C) Error bars in the same category with different letters are significantly different (P 0.05) PROC GLM, Tukey s HSD test (SAS Institute 2013). Fig. 7B). Degradation of ME over time was predicted with a high level of confidence by nonlinear asymptotic exponential decay curves with R 2 values ranging from 0.936 to 0.985 for ME (Fig. 8). Based on slope parameter estimates (c) and 95% CIs, ME in Mallet ME dispensers degraded significantly faster than ME in Mallet MR and Mallet MC. Degradation of DDVP over time was also predicted with a high level of confidence by nonlinear asymptotic exponential decay curves with R 2 values ranging from 0.960 to 0.993. Based on slope parameter estimates (c) and 95% CIs DDVP in Mallet BR degraded significantly faster than DDVP in Mallet MR, Mallet MC, and Mallet ME dispensers (Fig. 9). Discussion In tests with solid lure insecticide wafers, Vargas et al. (2010a) demonstrated that standard Jackson traps or AWPM bucket traps with Mallet-ME wafers impregnated with DDVP and the AWPM trap with Scentry ME cones (Scentry Biologicals, Inc., Billings, MT) and

August 2015 VARGAS ET AL.: LONGEVITY OF SOLID ME/C-L/RK DISPENSERS 1619 Fig. 6. Chemical degradation of RK from (A) three different weathered solid lure dispensers with corresponding captures (mean 6 SEM) of B. cucurbitae inside (B) Jackson traps and (C) AWPM traps. Traps were hung in papaya trees at Island Princess papaya orchards, Keaau, Hawaii Island, HI, from 10 May to 30 August 2010. (B) Error bars in the same category with different letters are significantly different (P 0.05) PROC GLM, Tukey s HSD test (SAS Institute 2013). (C) Error bars in the same category with different letters are significantly different (P 0.05) PROC GLM, Tukey s HSD test (SAS Institute 2013). DDVP vapor tape performed as well as the standard Jackson trap with liquid ME and naled against B. dorsalis. Similarly, Jackson traps or AWPM traps with Mallet-C-L wafers impregnated with DDVP or the AWPM trap with Scentry C-L plugs with vapor tape performed as well as a standard Jackson trap with liquid C-L and naled against B. cucurbitae (Vargas et al. 2010a). During the Hawaii AWPM fruit fly program, solid lure dispensers were tested extensively as both monitoring and male annihilation dispensers (Vargas et al. 2008, 2010a; Mau et al. 2007; Leblanc 2011) over a 10-yr period. Prepackaged solid lure dispensers were found to be a more convenient and safer alternative to traps with cotton wicks with liquid lures (ME and C-L) and insecticide mixtures. Few differences have generally been found between captures of different species of Bactrocera with solid and liquid forms of ME and C-L (Shelly 2010; Shelly et al. 2011a,b, 2012). In the present studies of weathering of solid lure dispensers, captures of B. dorsalis with fresh wafers, compared with weathered wafers, were significantly different after week 12; captures of B. cucurbitae were

1620 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 108, no. 4 Fig. 7. Captures (mean proportions % 6 SEM) of (A) B. dorsalis (F (2, 48) ¼ 107.51; P 0.001) with three different ME dispensers and (B) B. cucurbitae (F (2, 48) ¼ 1.593; P ¼ 0.214) with three different RK dispensers placed inside Jackson traps and hung in papaya trees at Island Princess papaya orchards, Keaau, Hawaii Island, HI, from 10 May to 30 August 2010. Error bars in the same category with different letters are significantly different (P 0.05) PROC GLM, Tukey s HSD test (SAS Institute 2013). Fig. 8. Nonlinear asymptotic regression of % ME in three solid lure dispensers weathered under Hawaii climatic conditions from 10 May to 30 August 2010. For Mallet ME, significance test for the model was F (2,14) ¼ 202.09, P < 0.0001; estimate (95% CL) a ¼ 0.17 ( 0.49, 0.15), b ¼ 0.79 (0.50, 1.09), and c ¼ 0.0034 (0.0032, 0.0177); for Mallet MR, significance test for the model was F (2,14) ¼ 454.02, P < 0.0001; estimate (95% CL) a ¼ 0.089 (0.070, 0.109), b ¼ 0.24 (0.22, 0.26), and c ¼ 0.024 (0.018, 0.030); and for Mallet MC, significance test for the model was F (2,14) ¼ 103.07, P < 0.0001; estimate (95% CL) a ¼ 0.023 ( 0.020, 0.065), b ¼ 0.25 (0.21, 0.29), and c ¼ 0.024 (0.012, 0.035) (SAS Institute 2013). not significantly different after 16 wk. These results suggest that RK dispenser attractiveness is longer-lived than ME and are useful in determining when lures inside traps should be replaced. Furthermore, bioassay results in the present studies generally supported chemical analysis data for solid lures. Vargas et al. (2009) found no difference between captures of B. dorsalis and B. cucurbitae with DDVP vapor tape and liquid naled, as used in detection programs in California and Florida. Subsequent studies by Shelly (2013) confirmed these results. Lloyd et al. (1998) found that ME efficacy with fiberboard ( Caneite ) blocks followed a negative exponential curve over 52 wk. After 8 wk of exposure to weather, efficacy of fiberboard blocks was reduced by 50% in comparison with a new block and the ME content was reduced by 73%. Malathion ( maldison ) content of blocks did not change over 28 wk, with a small loss over 52 wk. In the present study, degradation of ME and DDVP were both described effectively by asymptotic nonlinear regression curves, whereas degradation of RK was more gradual and constant over the 16-wk trial and did not fit asymptotic nonlinear curves as well as those for ME and DDVP. Addition of benzyl acetate to mixtures appeared to have little effect on captures. In studies by Vargas et al. (2005), Amulet C-L dispensers with fipronil and 0.5 ml of C-L were effective for at least 77 d, while Amulet ME dispensers with fipronil and 4 ml of ME were effective for only 21 d. Addition of ME to weathered stations restored their attractiveness and kill properties. Leblanc et al. (2013) suggested the rapid degradation under tropical climates

August 2015 VARGAS ET AL.: LONGEVITY OF SOLID ME/C-L/RK DISPENSERS 1621 Fig. 9. Nonlinear asymptotic regression curves of % DDVP in four solid lure dispensers weathered under Hawaii climatic conditions from 10 May to 30 August 2010. For Mallet BR, significance test for the model was F (2,14) ¼ 166.66, P < 0.0001; estimate (95% CL) a ¼ 0.011 ( 0.033, 0.010), b ¼ 0.051 (0.031, 0.070), and c ¼ 0.011 (0.003, 0.019); for Mallet MR, significance test for the model was F (2,14) ¼ 983.29, P < 0.0001; estimate (95% CL) a ¼ 0.0052 (0.0041, 0.0063), b ¼ 0.034 (0.033, 0.036), and c ¼ 0.036 (0.031, 0.040; for Mallet MC, significance test for the model was F (2,14) ¼ 150.63, P < 0.0001; estimate (95% CL) a ¼ 0.0009 ( 0.0026, 0.0045), b ¼ 0.042 (0.036, 0.047), and c ¼ 0.035 (0.023, 0.046); and for Mallet ME, significance test for the model was F (2,14) ¼ 1027.58, P < 0.0001; estimate (95% CL) a ¼ 0.0004 ( 0.0013, 0.0005), b ¼ 0.035 (0.034, 0.037), and c ¼ 0.044 (0.039, 0.048) (SAS Institute 2013). of these ME stations made of pressed paper was one of the reasons for the failure of the eradication program for B. dorsalis in French Polynesia. Results reported here provide additional data for improved formulation of solid ME and RK solid wafers, as is the current practice with solid TML dispensers placed in Jackson traps with sticky inserts. However, because of the larger size of the Bactrocera flies, either dispensers with DDVP added or dispensers with DDVP vapor tape will have to be used in traps (Vargas et al. 2009). Captures with mixtures of liquid ME and C-L on wicks in bucket traps were previously examined in Hawaii by Vargas et al. (2000). They found little effect of using different mixtures on B. cucurbitae captures; however, over time captures of B. dorsalis were reduced with the amount of ME initially incorporated on the wick. As chemical analyses did not accompany those studies, the following two factors were not recognized: 1) the high volatility of ME and low volatility of C-L/RK; and 2) when using mixtures of C-L/RK and ME, this volatility difference could be compensated for by using small amounts of C-L/RK and large amounts of ME. The present study clarifies these two issues. Captures of B. dorsalis with fresh wafers in Jackson and AWPM traps were significantly different on the basis of ME concentration (Mallet ME [56%] > Mallet MR [31.2%] > Mallet MC [23.1%]). Captures of B. cucurbitae with fresh wafers in Jackson and AWPM traps were not significantly different, regardless of concentration of RK (Mallet BR [20.1%] ¼ Mallet MR [18.3%] ¼ Mallet MC [15.9%]). The most recent results of field experiments conducted in Hawaii coffee fields (Shelly 2012, Vargas et al. 2012) confirm these conclusions. Both studies conducted independently suggest that traps baited with a solid formulation containing three lures (TML, ME, and RK plus the toxicant DDVP ¼ Mallet-TMR), but with a high concentration of ME and a large dispenser surface area, captured similar numbers of males of C. capitata, B. cucurbitae, and B. dorsalis, as three traps, each baited with a single liquid lure (plus the toxicant naled). Three possible applications for solid lure detection traps have been proposed for large survey programs utilizing ME, C-L/RK, and TML traps: 1) three individual traps with three separate solid wafers (TML, Mallet ME, and Mallet RK or C-L), 2) two individual traps with two solid wafers (TML and MR or MC), or 3) one trap with Mallet TMR (TML, ME, and RK). Results supporting option 3 have been discussed previously by Vargas et al. (2012) and Shelly et al. (2012). Results presented here provide both bioassay and chemical degradation data collected in Hawaii to support options 1 or 2 whereby solid lure dispensers could be deployed in a similar fashion to TML traps for detection of C. capitata (Jang et al. 2003; 2005) either as separate ME and RK lures or mixtures of the two. In Hawaii, based on data presented here and field experience with the AWPM program, solid lure dispensers in monitoring traps needed to be replaced after 3 mo and vapor tape after 6 to 8 wk. Further tests are currently underway in California to test weathering of solid lures under California climatic conditions. With respect to male annihilation programs in Hawaii where B. dorsalis and B. cucurbitae distributions often overlap MAT traps containing ME and C-L would provide a single dispenser that can be effectively used against flies responding to either lure. Enclosing dispensers inside bucket traps would not only provide protection from the weather but also make the device visible, retrievable, and reusable with limited environmental contamination and exposure to humans and pets (Vargas et al. 2003). However, inclusion of an

1622 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 108, no. 4 insecticide in wafers would most likely require that the dispenser would have to be registered with US Environmental Protection Agency. However, a registration exists for DDVP vapor tape whereby they could be used with solid dispensers containing an approved lure inside a trap for either survey work or male annihilation in an IPM program, without the further expense of registering a new lure insecticide combination (that is a lure impregnated with DDVP). Comparison of the vapor tape lure combination with lure DDVP combinations in the same dispenser have shown few differences (Vargas et al. 2009). Future tests will examine escape-proof traps (Hiramoto et al. 2006) with combination wafers without an insecticide for applications as environmentally friendly area-wide IPM procedures on farms, home gardens, and residential areas in Hawaii. Acknowledgments We thank Russell Ijima (Pacific Basin Agricultural Research Center, USDA-ARS, Hilo, HI) and Amy E. Morice (UC Davis) for assistance in collecting field data. We also thank Yu Nakane for assisting in fruit fly counts. We thank W.H. Shipman Limited, in particular Randall S. Akao, and many Big Island papaya growers for allowing us access to field sites for these evaluations. The USDA-ARS Hawaii Fruit Fly Area-Wide Pest Management program and IR-4 provided partial financial support for this work. We thank Nic Manoukis and Jaime Piñero for comments on an earlier draft of this manuscript. This article reports the results of research only. Mention of a proprietary product does not constitute an endorsement or a recommendation by the USDA for its use. USDA is an equal opportunity employer. References Cited (CDFA) California Department of Food and Agriculture. 2010. Insect trapping guide, 12th ed. CDFA, Sacramento, CA. (FDACS) Florida Department of Agriculture and Consumer Services. 2004. Floridafruitflydetectionmanual. Division of Plant Industry, FDACS, Gainesville, FL. Hiramoto,M.K.,L.Arita-Tsutsumi,andE.Jang.2006.Test of effectiveness of newly formulated plastic matrix with methyl eugenol for monitoring Bactrocera dorsalis (Hendel) populations. Proc. Hawaii. Entomol. Soc. 38: 103 110. (IAEA) International Atomic Energy Agency. 2003. Trapping guidelines for area-wide fruit fly programmes. IAEA, Vienna, Austria. Jang,E.B.,T.Holler,M.Cristofaro,S.Lux,A.S.Raw,A.L. Moses, and L. A. Carvalho. 2003. Improved attractants for areawide detection and control of Mediterranean fruit fly, Ceratitis capitata (Wiedemann): Response of sterile and wild flies to (-) enantiomer of ceralure B1. J. Econ. Entomol. 96: 1719 1723. Jang,E.B.,A.Khrimian,T.Holler,V.Casana-Giner,S. Lux, and L. A. Carvalho. 2005. Field response of Mediterranean fruit fly (Diptera: Tephritidae) to ceralure B1: Evaluations of enantiomer B1 ratios on fly captures. J. Econ. Entomol. 98: 1139 1143. Leblanc,L.,R.I.Vargas,B.Mackey,R.Putoa,andJ.C. Piñero. 2011. Evaluation of cue-lure and methyl eugenol solid lure and insecticide dispensers for fruit fly (Diptera: Tephritidae) monitoring and control in Tahiti. Fla. Entomol. 94: 510 516. Leblanc,L.,R.I.Vargas,andR.Putoa.2013.From eradication to containment: Invasion of French Polynesia by Bactrocera dorsalis (Hendel) (Diptera: Tephritidae) and releases of two natural enemies: A 17 year case study. Proc. Hawaii. Entomol. Soc. 45: 31 43. Mau,R.F.L.,E.B.Jang,andR.I.Vargas.2007.The Hawaii fruit fly area-wide fruit fly pest management programme: Influence of partnership and a good education programme, pp. 671 683. In M. J. B. Vreysen, A. S. Robinson, and J. Hendrichs (eds.), Area-wide control of insect pests: From research to field implementation. Springer, Dordrect, The Netherlands. Metcalf, R. L., and E. R. Metcalf. 1992. Fruit flies of the family Tephritidae, pp.139 152 In R. L. Metcalf and E. R. Metcalf (eds.), Plant kairomones in insect ecology and control. Routledge, Chapman & Hall Inc., New York, NY. Piñero,J.C.,R.F.L.MauandR.I.Vargas.2009.Managing oriental fruit fly (Diptera: Tephritidae), with spinosad-based protein bait sprays and sanitation in papaya orchards in Hawaii. J. Econ. Entomol. 102: 1123 1132. SAS Institute. 2013. SAS 9.2 online documentation. SAS Institute Inc., Cary, NC. Shelly,T.E.2010.Capture of Bactrocera males (Diptera: Tephritidae) in parapheromone-baited traps: A comparison of liquid versus solid formulations. Proc. Hawaii. Entomol. Soc. 42: 1 8. Shelly,T.E.2013.Male lures and the detection of Bactrocera fruit flies (Diptera: Tephritidae): Performance of solid dispensers with separate insecticidal strips relative to standard liquid lures. Proc. Hawaii. Entomol. Soc. 45: 119 128. Shelly, T. E., R. Kurashima, J. Nishimoto, A. Diaz, J. Leathers, M. War, and D. Joseph. 2011a. Capture of Bactrocera fruit flies (Diptera: Tephritidae) in traps baited with liquid versus solid formulations of male lures. J. Asia-Pac. Entomol. 14: 463 467. Shelly,T.E.,R.Kurashima,andJ.Nishimoto.2011b.Capture of Bactrocera males (Diptera: Tephritidae) in parapheromone-baited traps: Performance of solid dispensers with different loadings of attractants and toxicant. Proc. Hawaii. Entomol. Soc. 43: 33 47. Shelly,T.E.,J.Nishimoto,andR.Kurashima.2012.Captures of three economically important fruit fly species (Diptera: Tephritidae) in traps baited with liquid versus solid formulations of male lures in a Hawaiian coffee field. J. Econ. Entomol. 105: 1186 1193. Vargas,R.I.,J.D.Stark,M.H.Kido,H.M.Ketter,andL. C. Whitehand. 2000. Methyl eugenol and cue-lure traps for suppression of male oriental fruit flies and melon flies (Diptera: Tephritidae) in Hawaii: Effects of lure mixtures and weathering. J. Econ. Entomol. 93: 81 87. Vargas,R.I.,N.W.Miller,andJ.D.Stark.2003.Field trials of spinosad as a replacement for naled ddvp, and malathion in methyl eugenol and cue-lure bucket traps to attract and kill male oriental fruit flies and melon flies (Diptera: Tephritidae) in Hawaii. J. Econ. Entomol. 96: 1780 1785. Vargas, R. I., J. D. Stark, B. Mackey, and R. Bull. 2005. Weathering trials of Amulet cue-lure and methyl eugenol attract and kill stations with male melon flies and oriental fruit flies (Diptera: Tephritidae) in Hawaii. J. Econ. Entomol. 98: 1551 1559. Vargas,R.I.,R.F.L.Mau,E.B.Jang,R.M.Faust,and L. Wong. 2008. The Hawaii fruit fly area-wide pest management program, pp. 300 325. In O.Koul,G.W. Cuperus, and N. C. Elliott (eds). Areawide IPM: Theory to Implementation. CABI Books, London, United Kingdom. Vargas, R. I., R. E. Burns, R. F. L. Mau, J. D. Stark, P. Cook, and J. C. Piñero. 2009. Captures in methyl eugenol and cue-lure detection traps with and without insecticides andwithafarmatechsolidlureandinsecticidedispenser. J. Econ. Entomol. 102: 552 557. Vargas,R.I.,R.F.L.Mau,J.D.Stark,J.C.Piñero,L.Leblanc, and S. K. Souder. 2010a. Evaluation of methyl

August 2015 VARGAS ET AL.: LONGEVITY OF SOLID ME/C-L/RK DISPENSERS 1623 eugenol and cue-lure traps with solid lure and insecticide dispensers for fruit fly monitoring and male annihilation in the Hawaii areawide pest management program. J. Econ. Entomol. 103: 409 415. Vargas,R.I.,T.E.Shelly,L.Leblanc,andJ.C.Piñero. 2010b. Recent advances in methyl eugenol and cuelure technologies for fruit fly detection, monitoring and control, pp. 575 596, In G. Litwack (ed.), Vitamins and hormones, section: Pheromones, vol 83. Academic Press, Burlington. Vargas, R. I., S. K. Souder, B. Mackey, P. J. Cook, J. G. Morse,andJ.D.Stark.2012.Field trials of solid triple lure (trimedlure, methyl eugenol, raspberry ketone, and DDVP) dispensers for detection and male annihilation of Ceratitis capitata (Wiedemann), Bactrocera dorsalis (Hendel) and Bactrocera cucurbitae (Coquillett) (Diptera: Tephritidae) in Hawaii. J. Econ. Entomol. 105: 1557 1565. Received 3 February 2015; accepted 6 May 2015.