Captures in Methyl Eugenol and Cue-Lure Detection Traps with and Without Insecticides and with a Farma Tech Solid Lure and Insecticide Dispenser

ECOLOGY AND BEHAVIOR Captures in Methyl Eugenol and Cue-Lure Detection Traps With and Without Insecticides and With a Farma Tech Solid Lure and Insecticide Dispenser

ROGER I. VARGAS,1 R. E. BURNS,2 RONALD F. L. MAU,3 JOHN D. STARK,4 PETER COOK,5 3 AND JAIME C. PIN˜ ERO

U.S. PaciÞc Basin Agricultural Research Center, USDA-ARS, P.O. Box 4459, Hilo, HI 96720

J. Econ. Entomol. 102(2): 552Ð557 (2009) ABSTRACT Methyl eugenol (ME) and cue-lure (C-L) traps to detect tephritid ßies on the U.S. mainland were tested with and without insecticides under Hawaiian weather conditions against small populations of oriental fruit ßy, Bactrocera dorsalis (Hendel) and melon ßy, Bactrocera cucurbitae (Coquillett), respectively. In comparative tests, standard Jackson traps with naled and the Hawaii fruit ßy areawide pest management (AWPM) trap with 2,2-dichorovinyl dimethyl phosphate (DDVP) insecticidal strips outperformed traps without an insecticide. Addition of the reduced risk insecticide spinosad did not increase trap capture signiÞcantly compared with Jackson traps without an insecticide. Captures in AWPM traps with DDVP compared favorably with those for the Jackson trap with liquid naled (the Florida standard). In subsequent tests, captures with solid Farma Tech wafer dispensers with ME or C-L and DDVP placed inside Jackson and AWPM traps were equal to those for a Jackson trap with naled, currently used for detection of ME and C-L responding fruit ßies in Florida. Farma Tech ME and C-L wafers with DDVP would be more convenient and safer to handle than current liquid insecticide formulations (e.g., naled) used for detection programs in Florida.

KEY WORDS Tephritid ßies, Bactrocera dorsalis, B. cucurbitae, naled, spinosad

Within the dipteran family Tephritidae, the genus traps with ME or C-L are maintained in California and Bactrocera Macquart is composed of at least 440 spe- Florida for rapid detection of accidental introductions cies distributed primarily in tropical Asia, the South of fruit ßy species belonging to these two genera. PaciÞc, and Australia (White and Elson-Harris 1992). The oriental fruit ßy, Bactrocera dorsalis (Hendel), Males of numerous Bactrocera and Dacus F. species are and the melon ßy, Bactrocera cucurbitae (Coquillett), known to be highly attracted to either methyl eugenol are two major fruit ßy pest species in Hawaii with 173 (ME) (4-allyl-1, 2-dimethoxybenzene-carboxylate) and 125 host plant species recorded, respectively or cue-lure (C-L) [4-(p-acetoxyphenyl)-2-butanone] (Metcalf and Metcalf 1992). Monitoring is a primary (Metcalf and Metcalf 1992). In fact, at least 90% of the component of the Hawaii fruit ßy Area-Wide Pest Dacinae species (composed of the two major genera Management (AWPM) program, a community-based Bactrocera and Dacus) are strongly attracted to either statewide effort to transfer safe sustainable integrated ME or to C-L/raspberry ketone (Hardy 1979). For pest management (IPM) technologies to farmers and example, at least 176 species of the male Dacinae are residents (Mau et al. 2007, Vargas et al. 2008a). Simple attracted to C-L/raspberry ketone, and 58 species to bucket traps (termed AWPM traps) are often used for ME (Metcalf 1990). Of the 73 Bactrocera and Dacus detection and monitoring of fruit ßy populations. species that are agricultural pests, 41 respond to C-L/ Therefore, an AWPM trap also was included in these raspberry ketone, 22 to ME, and 10 to neither (White tests. The objective of the current study was to test the and Elson-Harris 1992). Large numbers of detection performance of C-L and ME detection traps without insecticides, with the reduced risk insecticide spi- This article reports the results of research only. Mention of a nosad, and with a novel solid lure and insecticide proprietary product does not constitute an endorsement or a recom- formulation to capture B. cucurbitae and B. dorsalis mendation by the USDA for its use. ßies at low population densities under Hawaiian cli- 1 Corresponding author, e-mail: [email protected]. 2 Division of Plant Industries, Florida Department of Agriculture matic conditions. These two fruit ßy species represent and Consumer Services, Gainesville, FL 32608. model systems for potential application of our Þndings 3 College of Tropical Agriculture and Human Resources, University to other Bactrocera and Dacus species. Different com- of Hawaii at Manoa, 3050 Maile Way, Honolulu, HI 96822. binations were compared against the standard Jackson 4 Washington State University, Puyallup Research and Extension Center, Puyallup, WA 98371. trap with ME and C-L used for detection in Florida 5 Farma Tech International, North Bend, WA 98045. and the AWPM trap used for monitoring in Hawaii. April 2009 VARGAS ET AL.: METHYL EUGENOL AND CUE-LURE DETECTION TRAPS 553

Table 1. Trap type, toxicant, and attractant (eiher methyl trap); and 4) ChamP trap (Seabright Laboratories, eugenol [ME] against B. dorsalis or cue-lure [C-L] against B. cu- Emeryville, CA) ϩ ME (6 ml). C-L treatments were curbitae) used in three ﬁeld experiments conducted at Waimea and Honokaa, HI placed on wicks inside traps as follows (Table 1): 1) Jackson trap ϩ 5% ([AI]) naled C-L solution (6 ml) Exp Trap Toxicant Attractant (standard Florida detection trap); 2) Jackson trap ϩ ϩ ϩ 1 Jackson 1% naled 6 ml ME C-L (6 ml); 3) AWPM trap C-L (6 ml) insecticidal Jackson None 6 ml ME strip; and 4) ChamP trap ϩ C-L (6 ml). AWPMa Hercon Vapor Tape (DDVP) 6 ml ME Studies were conducted in Waimea (for B. dorsalis) ChamP None 6 ml ME and Honokaa (for B. cucurbitae), Hawaii Island, HI, Jackson 1% naled 6 ml C-L Jackson None 6 ml C-L respectively, in areas where fruit ßy populations were AWPM Hercon Vapor Tape (DDVP) 6 ml C-L low. Trials of different ME traps were conducted from ChamP None 6 ml C-L 20 November 2003 to 12 February 2004 in a large stand 2 Jackson 1% naled 6 ml ME Psidium cattleianum Jackson None 6 ml ME of wild strawberry guava, L. near Jackson 5% spinosad 6 ml ME Waimea. Trials of different C-L traps were conducted AWPM Hercon Vapor Tape (DDVP) 6 ml ME in an abandoned vegetable Þeld near Honokaa from 20 ChamP None 6 ml ME November 2003 to 12 February 2004. Mean Ϯ SEM Jackson 1% naled 6 ml C-L Ϯ Њ Jackson None 6 ml C-L monthly temperature and rainfall were 16.95 0.14 C Jackson 5% spinosad 6 ml C-L and 0.42 Ϯ 0.09 cm for Waimea and 16.86 Ϯ 0.19ЊC and AWPM Hercon Vapor Tape (DDVP) 6 ml C-L 0.72 Ϯ 0.19 cm for Honokaa, during the evaluation ChamP None 6 ml C-L period. Treatments were replicated four times in a 3 Jackson 1% naled 6 ml ME AWPM Hercon Vapor Tape (DDVP) 6 ml ME randomized complete block design. To compensate Jackson Mallet-ME/DDVP wafer for position effects, traps were rotated one position AWPM Mallet-ME/DDVP wafer each week until all four positions in a block had been Jackson 1% naled 6 ml C-L AWPM Hercon Vapor Tape (DDVP) 6 ml C-L occupied by each trap. Wicks were replaced after 6 Jackson Mallet-C-L/DDVP wafer wk. ME traps were placed in guava trees in four AWPM Mallet-C-L/DDVP wafer patches of guava. C-L traps (20 m apart) were placed

a on Þberglass stakes 1 m above the ground and arranged One-liter plastic containers with four 3-cm-diameter holes, de- in four blocks around the periphery of an abandoned scribed in detail by Vargas et al. (2003). vegetable Þeld. Fruit ßies from different traps were emptied at weekly intervals into plastic bags and transported to the laboratory where counts were done. For We were speciÞcally interested in developing a more the statistical analysis, data for B. dorsalis or B. cucur- convenient and safer way to use lure and insecticide bitae captures (mean number of ßies per trap per d) combinations. were transformed to (x ϩ 1)1/2 to stabilize the variance, and then subjected to analysis of variance Materials and Methods (ANOVA) (Proc GLM). Means were separated by a Fisher least signiÞcant difference (LSD) test at the Experiment 1. Performance of Jackson Traps with P ϭ and without Naled, ChamP Traps, and AWPM Traps. 0.05 level (SAS Institute 1999). The standard treatment was the Jackson trap with ME Experiment 2. Comparisons of Jackson Traps with or C-L prepared in accordance with the Florida trap- Naled, Jackson Traps with Spinosad, Jackson Traps, ping manual used to maintain a detection grid of 12,518 ChamP Traps, and AWPM Traps with DDVP Strips. ME and 6,844 C-L traps throughout the state of Florida ME treatments were placed inside traps and on wicks ϩ (FDACS 2003). The ChamP trap is a noninsecticidal as follows (Table 1): 1) Jackson trap 1% ([AI]) naled ϩ fruit ßy monitoring trap used in California. Jackson ME solution (6 ml); 2) Jackson trap ME (6 ml); 3) ϩ and ChamP traps are illustrated in the International Jackson trap 5% ([AI]) spinosad ME (6 ml) solution; Atomic Energy Agency (IAEA 2003) trapping guide- 4) AWPM trap ϩ ME (6 ml) ϩ insecticidal strip; and 5) line for AWPM programs. The Hawaii AWPM trap is ChamP trap ϩ ME (6 ml). C-L treatments were placed constructed of plastic containers (Highland Plastics inside traps and on wicks as follows (Table 1): 1) Jackson Inc., Mira Loma, CA) (Highland Plastic no. 36, 1-liter trap ϩ 5% ([AI]) naled C-L solution (6 ml); 2) Jackson containers, 3.5-cm radius, 15 cm in height, with four trap ϩ C-L (6 ml); 3) Jackson trap ϩ 5% ([AI]) spinosad 3-cm-diameter holes) illustrated in Vargas et al. C-L solution; 4) AWPM trap ϩ C-L (6 ml) ϩ insecticidal (2003). ME and C-L liquid lures were obtained from strip; and 5) ChamP trap ϩ C-L (6 ml). The Þve dif- Farma Tech International (North Bend, WA). ME ferent ME traps were placed 20 m apart within a patch treatments were placed on wicks inside traps as fol- of strawberry guava trees (1.5 m above the ground) in lows (Table 1): 1) Jackson trap ϩ 1% ([AI]) naled four different locations (four replicates, total of 20 (Dibrom Concentrate, Valent USA Corp., Walnut traps) in a randomized complete block design. The Creek, CA) ME solution (6 ml) (standard Florida Þve C-L traps (20 m apart) were placed on Þberglass detection trap); 2) Jackson trap ϩ ME (6 ml); 3) stakes 1 m above the ground and arranged in four AWPM trap ϩ ME (6 ml) ϩ insecticidal strip (Hercon blocks around the periphery of an abandoned vege- Vaportape II (2,2-dichorovinyl dimethyl phosphate table Þeld. To minimize the effect of trap location, [DDVP], Emigsville, PA) (standard Hawaii AWPM traps were rotated one tree position each week until 554 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 102, no. 2

Table 2. For experiment 1, captures of B. dorsalis and B. cucurbitae in Jackson traps with naled, AWPM traps with DDVP, Champ trap, and Jackson trap baited with ME or C-L, respectively

Mean (Ϯ SEM) ßies/trap/day Lure Wk Jackson trap/naled AWPM trap/DDVP ChamP trap Jackson trap ME 1Ð4 12.64 Ϯ 2.33a 12.59 Ϯ 3.73a 10.23 Ϯ 2.99a 6.32 Ϯ 1.49a 5Ð8 19.98 Ϯ 3.61a 15.50 Ϯ 4.29a 11.48 Ϯ 2.47ab 5.10 Ϯ 0.62b 9Ð12 15.39 Ϯ 3.03a 10.17 Ϯ 1.91ab 8.82 Ϯ 2.10bc 4.16 Ϯ 0.89c C-L 1Ð4 0.20 Ϯ 0.06a 0.16 Ϯ 0.05a 0.12 Ϯ 0.03a 0.06 Ϯ 0.04a 5Ð8 0.13 Ϯ 0.03ab 0.25 Ϯ 0.10a 0.06 Ϯ 0.03b 0.02 Ϯ 0.01b 9Ð12 0.21 Ϯ 0.09a 0.21 Ϯ 0.08a 0.05 Ϯ 0.02ab 0.00 Ϯ 0.00b

Traps were surveyed weekly from 20 Nov 2003Ð12 Feb 2004 at Waimea (ME) and Honokaa (C-L), Hawaii Island, HI. a Values in each row followed by the same letters are not signiÞcantly different at the 0.05 level LSD, PROC GLM (SAS Insitute 1999) (ME, weeks 1Ð4: F ϭ 1.16; df ϭ 3, 60; P ϭ 0.33; weeks 5Ð8: F ϭ 4.21; df ϭ 3, 60; P ϭ 0.009; weeks 9Ð12: F ϭ 4.74; df ϭ 3, 60; P ϭ 0.0049; and C-L, weeks 1Ð4: F ϭ 1.84; df ϭ 3, 60; P ϭ 0.15; weeks 5Ð8: F ϭ 3.57; df ϭ 3, 60; P ϭ 0.02; weeks 9Ð12: F ϭ 3.08; df ϭ 3, 60; P ϭ 0.03). all Þve positions in a patch had been occupied by each were pooled for 24 wk by fruit ßy species, transformed trap. to (x ϩ 1)1/2 to stabilize the variance, and then sub- Evaluations were conducted in Waimea (for B. dor- jected to ANOVA (Proc GLM, SAS Institute 1999). salis) and Honokaa (for B. cucurbitae) for 50 wk from 26 February 2004 to 10 February 2005. Mean Ϯ SEM Results monthly temperature and rainfall were 18.12 Ϯ 0.41ЊC and 0.42 Ϯ 0.09 cm for Waimea and 18.24 Ϯ 0.41ЊC and Experiment 1. Performance of Jackson Traps with 0.80 Ϯ 0.28 cm for Honokaa, respectively. As in the Þrst and without Naled, ChamP Traps, and AWPM Traps experiment, traps were emptied at weekly intervals with DDVP. Captures of B. dorsalis for ME traps with into plastic bags and transported to the laboratory insecticides were not signiÞcantly different (P Ͼ 0.05) where counts were done. For the statistical analysis, from those without insecticides from week 1Ð4 but data for B. dorsalis or B. cucurbitae captures (mean differed signiÞcantly (P Ͻ 0.01) from weeks 5Ð8 and number of ßies per trap per d) were transformed to 9Ð12 (Table 2). For each of the two last periods, the (x ϩ 1)1/2 to stabilize the variance, and then subjected highest captures occurred in the Jackson traps con- to ANOVA (Proc GLM). Means were separated by a taining naled and in the AWPM trap containing Fisher LSD test at the P ϭ 0.05 level (SAS Institute DDVP, whereas Jackson traps without insecticide cap- 1999) at 5-wk intervals. tured signiÞcantly fewer ßies, followed by ChamP Experiment 3. Comparison of Jackson Traps with traps. Captures of B. cucurbitae in C-L traps with Either Naled or Farma Tech Wafers, and AWPM insecticides were not signiÞcantly different (P Ͼ 0.05) Traps with Either Vaportape or Farma Tech Wafers. than those without insecticides from weeks 1Ð4, but ME treatments were placed on wicks inside the fol- they were signiÞcantly different (P Ͻ 0.01) from lowing traps (Table 1): 1) Jackson trap ϩ 1% ([AI]) weeks 5Ð8 and 9Ð12. For the last two periods, the naled ME solution (6 ml); 2) AWPM trap ϩME (6 ml) ϩ AWPM trap with insecticidal (DDVP) strip per- Hercon Pest Strip; 3) Jackson trap ϩ MALLET-ME formed best, followed by Jackson traps with naled. (57.3%)/DDVP (3.4%) wafer (7.7 by 5 cm) (Farma Experiment 2. Comparisons of Jackson Traps with Tech, North Bend, WA); and 4) AWPM trap ϩ Naled, Jackson Traps with Spinosad, Jackson Traps, MALLET-ME/DDVP wafer. C-L treatments were ChamP Traps, and AWPM Traps with DDVP. For placed inside traps and on wicks as follows (Table 1): seven of the 10 time intervals signiÞcant differences in 1) Jackson trap ϩ 5% ([AI]) naled C-L solution (6 captures of B. dorsalis were detected among the Þve ml); 2) AWPM trap ϩ C-L (6 ml) ϩ vaportape; 3) ME traps evaluated (Table 3). Overall, the most ef- Jackson trap ϩ MALLET-C-L (52.0%)/DDVP (7%) fective traps were the Jackson trap containing naled wafer (5 by 2.5 cm) (Farma Tech, North Bend, WA); and the AWPM trap containing DDVP. The AWPM and 4) AWPM trap ϩ Farma Tech C-L/DDVP wafer. trap was outperformed by the Jackson trap with naled The experimental design and plot layout was the same in only three of the 10 time intervals. In no instance did as in the previous experiment. addition of spinosad to the Jackson trap signiÞcantly Trap capture data were collected in Waimea (for B. (P Ͼ 0.05) improve performance over the Jackson trap dorsalis) and Honokaa (for B. cucurbitae) for 24 wk alone. In seven of the 10 time intervals the ChamP trap from 24 April 2007 to 10 July 2007 and 2 August 2007 performed as well as the Jackson trap with naled. to 13 September 2007. Mean Ϯ SEM monthly tem- Captures of B. cucurbitae were signiÞcantly differ- perature and rainfall were 18.13 Ϯ 0.28ЊC and 0.09 Ϯ ent (P Ͻ 0.05) (Table 4) among the Þve traps evalu- 0.03 cm for Waimea and 18.63 Ϯ 0.32ЊC and 0.25 Ϯ 0.09 ated for eight of the 10 time intervals. Overall, the cm for Honokaa, respectively. All B. dorsalis and B. performance of the Jackson trap with naled was similar cucurbitae ßies captured were removed from traps at to the AWPM trap with an insecticidal strip. Addition weekly intervals and counted in the laboratory. For of spinosad to the Jackson trap improved performance the statistical analysis, data for B. dorsalis or B. cucur- signiÞcantly (P Ͻ 0.05) over the Jackson trap alone in bitae captures (mean number of ßies per trap per d) only one (week 26Ð30) of the 10 time intervals. Per- April 2009 VARGAS ET AL.: METHYL EUGENOL AND CUE-LURE DETECTION TRAPS 555

Table 3. For experiment 2, captures of B. dorsalis in traps baited with ME

Mean (Ϯ SEM) ßies/trap/day Wk Jackson trap/naled AWPM trap/DDVP Jackson trap/spinosad ChamP trap Jackson trap 1Ð5 4.13 Ϯ 1.30a 3.01 Ϯ 1.21ab 0.98 Ϯ 0.31bc 1.82 Ϯ 0.43abc 0.61 Ϯ 0.16c 6Ð10 1.73 Ϯ 0.67a 0.77 Ϯ 0.18b 0.41 Ϯ 0.14b 0.77 Ϯ 0.16b 0.20 Ϯ 0.07b 11Ð15 1.29 Ϯ 0.30a 0.49 Ϯ 0.12b 0.43 Ϯ 0.18b 1.19 Ϯ 0.35a 0.27 Ϯ 0.06b 16Ð20 0.29 Ϯ 0.10a 0.21 Ϯ 0.07a 0.12 Ϯ 0.06a 0.26 Ϯ 0.09a 0.05 Ϯ 0.02a 21Ð25 0.41 Ϯ 0.10a 0.28 Ϯ 0.09ab 0.13 Ϯ 0.05b 0.27 Ϯ 0.06ab 0.12 Ϯ 0.06b 26Ð30 5.54 Ϯ 1.32a 4.35 Ϯ 1.39a 2.90 Ϯ 0.85a 2.77 Ϯ 0.83a 3.15 Ϯ 0.92a 31Ð35 10.10 Ϯ 2.00a 6.20 Ϯ 1.84ab 5.11 Ϯ 1.37b 4.46 Ϯ 0.81b 4.19 Ϯ 1.12b 36Ð40 26.90 Ϯ 4.22a 10.73 Ϯ 1.78b 10.66 Ϯ 2.21b 8.79 Ϯ 1.41b 9.02 Ϯ 1.62b 41Ð45 12.94 Ϯ 3.17a 6.31 Ϯ 1.25a 8.80 Ϯ 3.09a 10.46 Ϯ 2.23a 3.88 Ϯ 0.96a 46Ð50 13.13 Ϯ 3.09ab 11.33 Ϯ 2.76abc 6.94 Ϯ 1.68bc 16.92 Ϯ 3.29a 4.75 Ϯ 0.84c

Traps were surveyed weekly from 26 February 2004Ð10 February 2005 at Waimea, Hawaii Island, HI. Values in each row followed by the same letters are not signiÞcantly different at the 0.05 level LSD, PROC GLM (SAS Institute 1999) (weeks 1Ð5: F ϭ 3.08; df ϭ 4, 95; P ϭ 0.02; weeks 6Ð10: F ϭ 3.28; df ϭ 4, 95; P ϭ 0.01; weeks 11Ð15: F ϭ 4.17; df ϭ 4, 95; P ϭ 0.004; weeks 16Ð20: F ϭ 1.79; df ϭ 4, 95; P ϭ 0.14; weeks 21Ð25: F ϭ 2.56; df ϭ 4, 95; P ϭ 0.04; weeks 26Ð30: F ϭ 1.19; df ϭ 4, 95; P ϭ 0.32; weeks 31Ð35: F ϭ 2.60; df ϭ 4, 95; P ϭ 0.04; weeks 36Ð40: F ϭ 9.71; df ϭ 4, 94; P Ͻ 0.0001; weeks 41Ð45: F ϭ 2.30; df ϭ 4, 95; P ϭ 0.06; weeks 46Ð50: F ϭ 3.75; df ϭ 4, 95; P ϭ 0.007). formance of the Jackson trap alone was similar to the Agriculture 2005). For detection of Bactrocera species, ChamP trap. generally larger than Ceratitis species, the addition of Experiment 3. Comparison of Jackson Traps with ME and C-L with an insecticide such as naled has been Either Naled or Farma Tech Wafers, and AWPM commonly used as is the case with Florida (FDACS Traps with Either Vaportape or Farma Tech Wafers. 2003). Almost 20,000 of these traps are maintained Over a 24-wk period, B. dorsalis captures for the dif- throughout the state for detection of fruit ßies. Larger ferent ME traps were not signiÞcantly different (P Ͼ numbers of similar traps are maintained throughout 0.05) (Table 5). Similarly, B. cucurbitae captures for California. Our results suggest that for detection of the the different C-L traps with insecticides were statis- two Bactrocera species evaluated in these tests, addi- tically similar (P Ͼ 0.05) over the entire 24-wk period. tion of naled, and potential use of Farma Tech wafers, is necessary for highest trap capture efÞciency. Overall, ChamP traps and Jackson traps alone did Discussion not perform as well as the Jackson trap with naled or The Jackson trap is inexpensive and easy to carry, the AWPM trap with DDVP. For traps without insec- handle, and service, and it also provides the opportu- ticides, captures by the ChamP trap were slightly bet- nity to service a greater number of traps per person- ter than those for the Jackson trap. Although not as hour than other commercial traps (IAEA 2003). This persistent as naled, spinosad, a reduced risk insecti- cardboard delta-shaped fruit ßy trap with a sticky cide, would be safer to handle and a more environ- cardboard insert is used worldwide with and without mentally friendly substitute for organophosphate in- insecticides for detection of fruit ßies. The Jackson secticides (Vargas et al. 2003). Previously, spinosad trap with trimedlure has become the standard tool for used in AWPM traps for male annihilation was statis- detection of Mediterranean fruit ßy, Ceratitis capitata tically comparable with organophosphate insecticides (Wiedemann) (California Department of Food and (Vargas et al. 2003). However, in the present tests and

Table 4. For experiment 2, captures of B. cucurbitae in traps baited with C-L

Mean (Ϯ SEM) ßies/trap/day Wk Jackson trap/naled AWPM trap/DDVP Jackson trap/spinosad ChamP trap Jackson trap 1Ð5 0.10 Ϯ 0.04ab 0.13 Ϯ 0.04a 0.10 Ϯ 0.03abc 0.04 Ϯ 0.02bc 0.02 Ϯ 0.01c 6Ð10 0.19 Ϯ 0.05a 0.13 Ϯ 0.03a 0.04 Ϯ 0.02b 0.02 Ϯ 0.01b 0.04 Ϯ 0.03b 11Ð15 0.10 Ϯ 0.04a 0.03 Ϯ 0.01b 0.04 Ϯ 0.01b 0.02 Ϯ 0.01b 0.00 Ϯ 0.0b 16Ð20 0.09 Ϯ 0.03a 0.10 Ϯ 0.03a 0.04 Ϯ 0.03a 0.05 Ϯ 0.04a 0.04 Ϯ 0.02a 21Ð25 0.44 Ϯ 0.15a 0.29 Ϯ 0.12a 0.37 Ϯ 0.15a 0.21 Ϯ 0.07a 0.14 Ϯ 0.05a 26Ð30 0.24 Ϯ 0.05a 0.16 Ϯ 0.04a 0.16 Ϯ 0.05a 0.04 Ϯ 0.01b 0.04 Ϯ 0.01b 31Ð35 0.18 Ϯ 0.05a 0.19 Ϯ 0.06a 0.10 Ϯ 0.03a 0.07 Ϯ 0.03a 0.04 Ϯ 0.02a 36Ð40 0.44 Ϯ 0.12a 0.34 Ϯ 0.13ab 0.13 Ϯ 0.04bc 0.07 Ϯ 0.04c 0.04 Ϯ 0.20c 41Ð45 0.22 Ϯ 0.10b 0.53 Ϯ 0.14a 0.16 Ϯ 0.05b 0.09 Ϯ 0.06b 0.04 Ϯ 0.01b 46Ð50 0.65 Ϯ 0.15a 0.45 Ϯ 0.1ab 0.18 Ϯ 0.05c 0.21 Ϯ 0.05cb 0.06 Ϯ 0.03c

Traps were surveyed weekly from 22 February 2004Ð3 February 2005 at Honokaa, Hawaii Island, HI. Values in each row followed by the same letters are not signiÞcantly different at the 0.05 level LSD, PROC GLM (SAS Institute 1999) (weeks 1Ð5: F ϭ 2.61; df ϭ 4, 95; P ϭ 0.04; weeks 6Ð10: F ϭ 6.18; df ϭ 4, 95; P ϭ 0.0002; weeks 11Ð15: F ϭ 3.64; df ϭ 4, 95; P ϭ 0.0084; weeks 16Ð20: F ϭ 1.04; df ϭ 4, 95; P ϭ 0.39; weeks 21Ð25: F ϭ 1.09; df ϭ 4, 95; P ϭ 0.36; weeks 26Ð30: F ϭ 5.73; df ϭ 4, 95; P ϭ 0.0004; weeks 31Ð35: F ϭ 2.24; df ϭ 4, 95; P ϭ 0.07; weeks 36Ð40: F ϭ 4.44; df ϭ 4, 95; P ϭ 0.0025; weeks 41Ð45: F ϭ 5.27; df ϭ 4,95; P ϭ 0.0007; weeks 46Ð50: F ϭ 7.36; df ϭ 4, 95; P Ͻ 0.0001). 556 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 102, no. 2

Table 5. For experiment 3, captures of B. dorsalis and B. and Jackson trap with Farma Tech wafers will be cucurbitae traps baited with methyl eugenol (ME) or cue-lure (C-L), tested as male annihilation devices. Farma Tech wa- respectively fers could be deployed as a naled alternative although Mean (Ϯ SEM) ßies/trap/d protection would still be required from insecticide Jackson/ AWPM/ fumes. Lure Wk Jackson/ AWPM/ Farma Tech Farma Tech In summary, our results suggest that the addition of naled DDVP strip Wafer Wafer spinosad would not improve trap captures and con- ME 24 2.0 Ϯ 0.73a 3.55 Ϯ 1.15a 2.22 Ϯ 0.61a 1.07 Ϯ 0.28a sequently naled cannot be excluded from ME/C-L C-L 24 1.26 Ϯ 0.22a 1.14 Ϯ 0.20a 1.24 Ϯ 0.20a 1.01 Ϯ 0.16a detection traps presently used in Florida. However, the Farma Tech wafer is a promising substitute for Traps were surveyed weekly from 24 April 2007 to 10 July 2007 and Jackson traps baited with both ME and C-L naled from 2 August to 13 September 2007 at Waimea (ME) and Honokaa solutions for detection of fruit ßies and should be (C-L), Hawaii Island, HI. Values in each row followed by the same letters are not signiÞcantly tested further not only for the AWPM program but different at the 0.05 level LSD, PROC GLM (SAS Institute 1999) (ME: also in Florida and California as the development of F ϭ 1.93; df ϭ 3, 284; P ϭ 0.1243; C-L: F ϭ 0.29; df ϭ 3, 284; P ϭ 0.8351). environmentally friendly area-wide IPM procedures would have important applications for early eradica- tion of accidental introductions of fruit ßies into the in the context of detection, addition of this reduced U.S. mainland. risk insecticide did not increase effectiveness compared with traps without an insecticide. Because it is Acknowledgments recognized that spinosad is a slow-acting insecticide and therefore some ßies may have escaped from traps, We thank Neil Miller, (PaciÞc Basin Agricultural Research but died later, then captures in Jackson traps with Center, USDA-ARS, Hilo, HI). Gregory Boyer, Charles Lee, spinosad may underestimate total kill. In general, un- Albert Kawabata, and Maria Derval Diaz (PaciÞc Basin Ag- der our approach the treatments that involved fast- ricultural Research Center, USDA-ARS, Kamuela, HI) for killing insecticides (AWPM traps with insecticidal assistance in collecting and analyzing Þeld data. We appre- strips and Jackson traps with naled) performed the ciate information provided by Abbie Fox (Division of Plant Industries, FDACS, Gainesville, FL). We thank Luc Leblanc best consistently. (University of Hawaii at Manoa, Department of Plant and Since 1999, a USDAÐARS AWPM program in Hawaii Environmental Protection Sciences, Honolulu, HI), Victoria has been funded for suppression of fruit ßies (Mau et Yokoyama (USDA-ARS-SJVASC, Parlier, CA), and Avi Ei- al. 2007; Vargas et al. 2008a). In the AWPM program, tam (APHIS-PPQ, Pickerington, OH) for comments on an we have consistently sought safer or reduced risk earlier draft of this manuscript. The Hawaii AWPM Program components. For example, spinosad-based GF-120 provided partial Þnancial support for this work. Fruit Fly Bait has been used effectively as a protein bait spray (Prokopy et al. 2003). Fipronil-based Am- References Cited ulet C-L stations (Vargas et al. 2005) were recently registered for male annihilation technique. A spray- [CDFA] California Department of Food and Agriculture. able SPLAT-ME is currently undergoing registration 2005. Insect trapping guide, 11th ed. CDFA, Sacra- with the United States Environmental Protection mento, CA. [FDACS] Florida Department of Agriculture and Con- Agency (Mau et al. 2007, Vargas et al. 2008b). The sumer Services. 2003. Florida fruit ßy detection manual. development of IPM approaches that include a solid Division of Plant Industry, Florida Department of Agri- lure/toxicant wafer trap for monitoring would have culture and Consumer Services, Gainesville, FL. important applications to detection and monitoring of Hardy, D. E. 1979. Review of economic fruit ßies of the fruit ßies not only in Hawaii but also throughout the South PaciÞc Region. Pac. Insects 20: 429Ð432. south and western PaciÞc, Australia, and tropical Asia [IAEA] International Atomic Energy Agency. 2003. Trap- where Bactrocera are serious economic pests. Imple- ping guidelines for area-wide fruit ßy programmes. In- mentation of DDVP (Vapor Tape) strips in place of ternational Atomic Energy Agency, Vienna, Austria. liquid naled represents an important improvement for Mau, R.F.L., E. B. Jang, and R. I. Vargas. 2007. The Hawaii fruit ßy area-wide fruit ßy pest management programme: worker safety. Likewise, there has been a move toward inßuence of partnership and a good education pro- replacement of liquid ME and C-L with solid formu- gramme, pp. 671Ð683. In M.J.B. Vreysen, A. S. Robinson, lations (i.e., Sentry ME or C-L plugs, Boseman, MT) and J. Hendrichs [eds.], Area-wide control of insect pests: without an insecticide. The prepackaged Farma Tech from research to Þeld implementation. Springer, Dor- wafer with a solid formulation of ME or C-L with drect, The Netherlands. DDVP, is a novel dispenser that performed equally as Metcalf, R. L. 1990. Chemical ecology of Dacinae fruit ßies well in AWPM and Jackson traps. These are the Þrst (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 83: 1017Ð published tests of these wafers. In the present test fruit 1030. ßy captures with Farma Tech wafers equaled or sur- Metcalf, R. L., and E. R. Metcalf. 1992. Fruit ßies of the family Tephritidae, pp. 109Ð152. In R. L. Metcalf and E. R. passed captures obtained with liquid formulations. Metcalf [eds.] Plant kairomones in insect ecology and The Farma Tech wafers will be further tested during control. Routledge, Chapman & Hall, Inc., London, the AWPM program in a variety of monitoring situa- United Kingdom. tions (high and low densities of ßies) and evaluated for Prokopy, R. J., N. W. Miller, J. C. Pin˜ ero, J. D. Barry, L. C. weathering properties. Furthermore, the bucket trap Tran, L. K. Oride, and R. I. Vargas. 2003. Effectiveness April 2009 VARGAS ET AL.: METHYL EUGENOL AND CUE-LURE DETECTION TRAPS 557

of GF-120 fruit ßy bait spray applied to border area plants Vargas, R. I., R.F.L. Mau, E.B. Jang, R. M. Faust, and L. for control of melon ßies (Diptera: Tephritidae). J. Econ. Wong. 2008a. The Hawaii Fruit Fly Area-Wide Pest Entomol. 96: 1485Ð1493. Management Program, pp. 300Ð325. In O. Koul, G. W. SAS Institute. 1999. SAS/STAT userÕs guide, version 6. In- Cuperus, and N. C. Elliott [eds.], Areawide IPM: the- stitute, Cary, NC. ory to implementation. CABI Books, London, United Vargas, R. I., N. W. Miller, and J. D. Stark. 2003. Field trials Kingdom. of spinosad as a replacement for naled DDVP, and mal- Vargas, R. I., J. D. Stark, M. Hertlein, A. Mafra-Neto, R. athion in methyl eugenol and cue-lure bucket traps to Coler, and J. C. Pin˜ ero. 2008b. Evaluation of SPLAT attract and kill male oriental fruit ßies and melon ßies with spinosad and methyl eugenol or cue-lure for “at- (Diptera: Tephritidae) in Hawaii. J. Econ. Entomol. 96: tract-and-kill” of oriental and melon fruit ßies (Diptera: 1780Ð1785. Tephritidae) in Hawaii. J. Econ. Entomol. 101: 759Ð768. Vargas, R. I., J. D. Stark, B. Mackey, and R. Bull. 2005. White, I. M., and M. M. Elson-Harris. 1992. Fruit ßies of Weathering trials of Amulet Cue-lure and methyl eugenol economic signiÞcance: their identiÞcation and bionomics. “attract and kill” stations with male melon ßies and ori- CAB International, Wallingford, United Kingdom. ental fruit ßies (Diptera: Tephritidae) in Hawaii. J. Econ. Entomol. 98: 1551Ð1559. Received 9 June 2008; accepted 30 October 2008.