Plant Health Division, International Centre of Insect Physiology and Ecology (icipe), PO Box GPO, Nairobi, Kenya

Transcription

1 Comparison of Food-Based Attractants for Bactrocera invadens (Diptera: Tephritidae) and Evaluation of Mazoferm Spinosad Bait Spray for Field Suppression in Mango Author(s): Sunday Ekesi, Samira Mohamed, and Chrysantus M. Tanga Source: Journal of Economic Entomology, 107(1): Published By: Entomological Society of America URL: BioOne ( 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 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.

2 HORTICULTURAL ENTOMOLOGY Comparison of Food-Based Attractants for Bactrocera invadens (Diptera: Tephritidae) and Evaluation of Mazoferm Spinosad Bait Spray for Field Suppression in Mango SUNDAY EKESI, 1 SAMIRA MOHAMED, AND CHRYSANTUS M. TANGA Plant Health Division, International Centre of Insect Physiology and Ecology (icipe), PO Box GPO, Nairobi, Kenya J. Econ. Entomol. 107(1): 299Ð309 (2014); DOI: ABSTRACT Catches of Bactrocera invadens Drew, Tsuruta, & White (Diptera: Tephritidae) in Multi-lure traps baited with six commercial food-based attractants, Mazoferm E802, Torula yeast, GF-120, Hymlure, Biolure, and Nulure were compared in mango orchards in Kenya. In 2007, Mazoferm E802 and Torula yeast were the most effective attractants and captured 2.4Ð2.6 times more females and 3.4Ð4.0 times more males than the standard Nulure. All attractants captured more females than males (ranging from 63 to 74%). In 2008, Mazoferm E802 was the most effective bait capturing 5.6 and 9.1 times more females and males, respectively, than the standard Nulure. Among all the attractants, in both years, Nulure captured the greatest proportion of females: 74% compared with 51Ð68% for the other attractants. In 2008, the use of Mazoferm E802 in combination with spinosad as a bait spray in mango orchards reduced B. invadens catches relative to the control by 87% within 4 wk and 90% within 8 wk. At harvest, the proportion of fruit infested was signiþcantly lower in the treated orchards (8%) compared with the control orchards (59%). Estimated mango yield was signiþcantly higher in orchards receiving the bait sprays (12,487 kg/ha) compared with control orchards (3,606 kg/ha). Based on bait spray costs, yield data, and monetary gains, a costðbeneþt ratio of 1:9.1 was realized, which is acceptable for growers. In 2009, the experiment was repeated with similar results. We have demonstrated that Mazoferm E802, used alone for monitoring of B. invadens or in conjunction with spinosad for population suppression, shows great promise in Kenya. KEY WORDS Bactrocera invadens, food bait, Kenya, population suppression, yield Bactrocera invadens Drew, Tsuruta, & White (Diptera: Tephritidae) is a recently described fruit ßy species from Asia that is of major quarantine concern in Africa and elsewhere (Drew et al. 2005, French 2005). It has been recorded from 30 host plant species, although mango, Mangifera indica L. (Sapinidales: Anacardiaceae), is clearly the preferred host (Mwatawala et al. 2006a, 2009; Rwomushana et al. 2008; Goergen et al. 2011). On mango, direct damage owing to B. invadens has been reported to range from 30 to 80% of the crop depending on the cultivar, locality, and season (Ekesi et al. 2006, Rwomushana 2008, Vayssières et al. 2009). In addition to direct losses, indirect losses attributed to quarantine restrictions on this pest have been enormous. For example, in the case of avocado, since 2007 Kenya has lost US$2 million annually owing to quarantine restrictions imposed by South Africa as a result of B. invadens (Otieno 2011). Export of other fruit species including mango and cucurbits from Kenya are already banned in the Seychelles, Mauritius, and South Africa. Trade in horticultural produce between Africa and the United States has also been severely 1 Corresponding author, sekesi@icipe.org. restricted by a recent U.S. Federal Order banning importation of particular cultivated fruits and vegetables from African countries where B. invadens has been reported (U.S. Department of AgricultureÐAnimal and Plant Health Inspection Service [USDAÐAPHIS] 2008). The European Union (EU) phytosanitary regulations in relation to non-european Tephritidae are tightening and interception and rejection of African mangoes in the EU, owing to fruit ßies, have been on the increase since the arrival of B. invadens (Guichard 2009). These direct and indirect costs have wide reaching socioeconomic implications for millions of people in rural and urban communities involved in the mango value chain across Africa. This has been further compounded by the introduction of uniform and strict quarantine regulations and maximum residue levels (MRL) by the EU that jeopardize the export of mangoes from Africa valued at US$42 million annually (Lux et al. 2003). Female tephritid fruit ßies require a posteclosion protein meal for ovarian development and egg maturation (Drew and Yuval 2000). Consequently, fruit ßy detection, monitoring, and suppression strategies have relied strongly on the use of food-based attractants /14/0299Ð0309$04.00/ Entomological Society of America

3 300 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 107, no. 1 made from proteins and fermenting sugars (Roessler 1989, International Atomic Energy Agency [IAEA] 2003). Although the main targets are the females, traps baited in this way capture both females and males of several fruit ßy species. An aqueous solution of corn hydrolysate commercialised as Nulure (Miller Chemical and Fertilizers, Hanover, PA) is considered as the standard for female-targeted detection, monitoring, and suppression in many countries (Gilbert et al. 1984, Moreno and Mangan 1995, IAEA 2003). Most recently, a dry synthetic attractant based on ammonium acetate, trimethylamine, and putrescine has been developed and marketed as Biolure. Reports suggest that this dry attractant has similar levels of efþcacy as the liquid product, Nulure (Epsky et al. 1999, Heath et al. 2007, Navarro-Llopis et al. 2013). Although B. invadens has been trapped in commercial crops using food-based attractants such as Nulure and Biolure (Ekesi et al. 2007b, Rwomushana 2008, Mwatawala et al. 2009) and Þeld suppression has been reported using GF-120, which is a combination of a food-based bait and the insecticide spinosad (Vayssières et al. 2009, Ekesi et al. 2011), no detailed studies to assess the relative attractiveness of various commercially available food-based attractants for both sexes of the pest have been conducted in Kenya. As Vargas et al. (2002) state, this is essential if they are to be used in effective detection, monitoring, and suppression strategies. In this study, we evaluated the relative efþcacy of six commercially available foodbased attractants in trapping male and female B. invadens in mango. Furthermore, we evaluated the efþcacy of the most attractive bait, sprayed in conjunction with spinosad for Þeld suppression of B. invadens in mango orchards in Kenya during two fruiting seasons. The study also estimated the costðbeneþt ratio of the bait-spinosad application based on monetary gains from mango yields and other associated treatment costs to guide future implementation and uptake by the growers. Materials and Methods Bait Evaluation Trials. Sites for Bait Evaluation Trials. The bait evaluation trials were conducted in mango orchards at Nguruman, Kajiado District, Rift Valley Province, Kenya. The orchard had a mixture of three mango cultivars, namely ÔKent,Õ ÔApple,Õ and ÔNgowe.Õ The mango season in Nguruman starts in August and ends in December. The Þrst trial was done from 5 October 2007 to 30 November 2007 and the second trial from 4 October 2008 to 29 November Nguruman is located at latitude S and longitude E, and at an altitude of 700 m above sea level (a.s.l). The mango-growing areas here are supplied, in addition to rains, with underground irrigation water from three rivers (Oloibortoto, Entasopia, and Sampu), which cut across the area Þnally discharging into Lake Natron in Tanzania. This area is also surrounded by a dry savannah belt making it an ecological island. In 2007, the average minimum and maximum temperatures during the experimental period were 22 and 35 C, respectively, and the minimum and maximum relative humidities (RH) were 33 and 76%, respectively. In 2008, the average minimum and maximum temperatures during the experimental period were 24 and 36 C, respectively, and the minimum and maximum RHs were 32 and 72%, respectively. Treatments and Experimental Design. Six foodbased bait treatments were evaluated each year: 1) Mazoferm E802 (Corn Products International, Eldoret, Kenya) at the on-the-label rate of 4%, 2) Torula yeast (ISCA Technologies, Riverside, CA) at the rate of 3 pellets (4.78 g/pellet) per 1,000 ml of water, 3) GF-120 (Dow AgroSciences, Indianapolis, IN) at the on-the-label rate of 18.2%, 4) Hymlure (Savoury Food Industries Ltd., Johannesburg, South Africa) at the on-the-label rate of 4%, 5) Biolure, a three-component synthetic lure containing ammonium acetate, trimethylamine, and putrescine (Suttera LLC, Bend, OR) as wet trap containing 0.01% Triton X-100, and 6) Nulure (Miller Chemical & Fertilizer Corporation, Hanover, PA) at 9%. Each bait was placed in a Multi-lure trap (Better World Manufacturing, Fresno, CA) for evaluation. These traps are a plastic version of the McPhail glass invaginated trap ( 17 cm in diameter) consisting of a yellow base ( 7 cm in height) and clear top ( 11 cm in height), and they were used in both years. In all the traps, 3% borax was added to preserve trapped ßies. The experiments were arranged in a randomized complete block design with Þve replicate mango orchards as blocks. The distance between blocks and traps within blocks were 30 and 15 m respectively. Traps were placed on randomly selected trees of the various cultivars in the orchard at 1.5Ð2 m above the ground and their position within each block was rotated sequentially every week, at the time they were checked for the presence of B. invadens, to ensure that the effect of treatment was not confounded by the effect of position. Apart from Biolure, all baits were replaced weekly at the time of trap checking. Biolure was replaced every 4 wk. At each weekly check, ßies were removed from the trap and the number and sex of B. invadens captured was recorded and a daily catch rate estimated. More than 96% of the fruit ßies caught were B. invadens, with only negligible numbers of the native Ceratitis cosyra (Walker). This native species has been largely displaced by the invasive species (Ekesi et al. 2009). Suppression Trials. Sites for the Suppression Trials. The suppression trials were conducted at Nthagaiya village, Runyenjes district, Eastern Province, Kenya (Fig. 1). Nthagaiya is located at latitude S and longitude and altitude of 1,365 m.a.s.l. This area has a single mango season from October to February. Four localities or blocks ( 3Ð4 km apart) to the north, south, east, and west were identiþed within the village environs (Fig. 1). Within each block two different mango orchards (1 ha in size) were selected each year (2008 and 2009) and allocated either to the treatment or to the control. Testing both the treatment and the control in the same orchard was avoided owing to the mobility of fruit ßies (see Peck and McQuate 2000, Piñero et al. 2009, Vayssières et al.

4 February 2014 EKESI ET AL.: B. invadens BAITS FOR MONITORING AND SUPPRESSION 301 Fig. 1. Map of the study location showing the treatment and control sites. 2009, Ekesi et al. 2011). To avoid movement of ßies between replicates and inßux of ßies from other orchards, all experimental mango orchards were at least 1.5 km apart from each other and relatively isolated from other unmanaged large-scale and smallholder mango orchards. None of the experimental orchards previously applied pest management measures, and the trials were done with the full cooperation of the growers. Three mango cultivars (Apple, ÔTommy Atkins,Õ and ÔVan DykeÕ) in rows 10 m apart (100 trees per ha) were present in all experimental orchards, and orchards were hand weeded as necessary. At full bloom, experimental orchards were treated with Mancozeb at the rate of 0.8 kg active ingredient per hectare against powdery mildew. Treatments and Application of Bait Spray. The foodbased attractant Mazoferm E802 was mixed with the insecticide spinosad for evaluation. At each of the four localities or blocks indicated in the Sites for the Suppression Trials section, a weekly application of the Mazoferm E802Ðspinosad bait spray was applied to one of the selected orchards (treatment) and the second orchard remained untreated (control). The rate applied wasñmazoferm E802 at 400 ml active ingredient per hectare with spinosad (Tracer 4 EC; Dow AgroSciences, Indianapolis, IN) at 100 g active ingredient per hectare at an output volume of 5 liters per hectare ( 50 milliliter per tree). The Mazoferm E802Ðspinosad bait spray was applied weekly as spot applications to 1 m 2 of the canopy of every tree within each of the four 1-ha treatment orchards using a CP 15 knapsack sprayer (Cooper Pegler, Sussex, United Kingdom). Spray application began on 29 November 2008 (2008 season) and 5 November 2009 (2009 season) just before the onset of fruit maturity and when monitoring data indicated that ßies had begun the usual seasonal population buildup. In total, eight spray applications were made to each treatment orchard each year. Population Monitoring With Traps. Fruit ßy populations were monitored in both years using Multi-lure traps baited with the standard protein hydrolysate, Nulure (2% active ingredient), for Þve consecutive weeks before the trials began and continued until mango harvest. Borax was added to the bait and water solution in each trap at the rate of 30 g/liter to preserve the captured ßies. At each mango orchard (four treatment and four control), Þve traps were positioned on randomly selected trees as described previously. Flies were collected weekly from the traps, preserved in 70% ethanol and returned to the laboratory where their identity was determined and a daily capture rate estimated. Traps were maintained, including the weekly replacement of bait, as described previously. Assessment of Fruit Infestation and Yield. At harvest, 100 mango fruits were sampled at random from each orchard and returned to the laboratory for processing. Each fruit was placed individually into a perforated polyethylene bag for transportation from the Þeld. Fruits were weighed and placed individually into 4-li-

5 302 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 107, no. 1 Table 1. Overall mean number of B. invadens adults captured per trap per day in mango orchards using different food-based attractants in 2007 and 2008 in Nguruman, Kenya Treatments Females Males Total % female 2007 Mazoferm E a a a a Torula yeast a a a a GF b b b a Hymlure d c cd a Biolure e d d a Nulure c c c a 2008 Mazoferm E a a a b Torula yeast b b b b GF c c c b Hymlure e d e b Biolure f e f b Nulure d d d a Means within a column followed by the same letter are not signiþcantly different to each other using TukeyÕs (HSD; P 0.05) test. ter rectangular transparent plastic containers (22 by 14.5 by 14 cm; Kenpoly, Kenya) on a thin layer (20 mm) of moist sterilized sand and incubated under ambient conditions (26Ð28 C) for 4Ð5 wk. The rims of these containers were covered with Þne netting held in place by the perforated container lid and capable of retaining adult tephritids. The sand served as a pupation substrate for larvae exiting the fruits and was checked daily for puparia, which were removed using forceps, counted, and placed in petri dishes (8.6 cm in diameter). After 4Ð5 wk, all fruit were dissected to retrieve any remaining larvae and puparia. Petri dishes containing puparia were placed individually into transparent Perspex cages (20 by 15 by 15 cm) until the emergence of adult ßies. Adult ßies were provided with water and fed on an artiþcial diet consisting of a 3:1 mixture, by volume, of sugar and enzymatic yeast hydrolysate ultrapure (USB Corporation, Cleveland, OH) for 4Ð5 d after emergence, until they had attained the full body coloration necessary for identiþcation. These ßies were then preserved in 70% ethanol for further identiþcation. A measure of infestation rate by fruit ßies was estimated based on the percentage of fruits infested after dissection. At harvest, marketable mango fruits were carefully picked using a hand-held pruner (Sauk City, WI) attached to a wooden pole (Copland 2007) from each replicated 1-ha plot and weighed on site. Marketable yields were determined and estimated as kilograms per hectare. Statistical Analysis. In all trials for each year, the numbers of fruit ßies captured and the Þgures for mango yield were transformed using Log (x 1) to normalize the variance while percentage fruit infestation and percentage reduction in fruit ßy population were arcsine-transformed before subjecting these data to analysis of variance. In the bait evaluation trials, the interaction between treatments and dates in trapping efþciency were tested using a factorial model with repeated measures analysis using PROC GLM. Mean female, mean male, total (male and female), and % female captured in each block were averaged over the sampling weeks and used for treatment comparison (Epsky et al. 1999, Barry et al. 2005) using PROC GLIM, and the means were compared using a TukeyÕs (HSD) test (P 0.05). In the suppression trials, we analyzed percentage reduction in fruit ßy population relative to the control for each year at two sampling dates (corresponding to 4 and 8 wk after the MazofermÐspinosad bait spray) using the formula of Henderson and Tilton (1955) as follows: {1 ([control before treatment after]/ [control after treatment before])} 100. Comparisons of traps catches, percentage fruit infestation, and mango yield between treated and control orchards were made using t-tests with PROC TTEST. For each year, costðbeneþt ratio was calculated based on monetary gains obtained at the farm gate price for marketable mango yields against the total cost of bait, insecticide, labor, and sprayer costs (Karel and Ashimogo 1991). All analyses were performed using the SAS (SAS Institute 2001) software package. Results Bait Evaluation. B. invadens Catches There was a signiþcant difference in the mean number of female B. invadens captured daily per trap among treatments (F 25.8; df 5,189; P ), dates (F 10.13; df 7,189; P ), and a signiþcant interaction between treatments and dates (F 2.86; df 35,189; P ; Table 1). For the mean number of male B. invadens captured daily per trap, there was a signiþcant difference among treatments (F 11.7; df 5,189; P ), dates (F 9.10; df 7,189; P ), and a signiþcant interaction between treatments and dates (F 3.13; df 35,189; P ; Table 1). For the mean total number of B. invadens captured daily per trap, there was a signiþcant difference among treatments (F ; df 5,189; P ), dates (F 13.19; df 7,189; P ), and a signiþcant interaction between treatments and dates (F 3.12; df 35,189; P ; Table 1; Fig. 2A). The percentage of adults captured that were female was not signiþcantly different among treatments (F 1.02; df 5,189; P ) and dates (F 1.01; df 7,189; P ),

6 February 2014 EKESI ET AL.: B. invadens BAITS FOR MONITORING AND SUPPRESSION 303 Fig. 2. Temporal trends in total number of B. invadens captured (male and female) using six food-based attractants in mango orchards in 2007 (A) and 2008 (B). Error bars denote SE. and there was no interaction between treatment and dates (F 0.87; df 35,189; P ; Table 1). Mazoferm E802 and Torula yeast were the most effective attractants and captured 2.4Ð2.6 times more females and 3.4Ð4.0 times more males, respectively, than the standard Nulure (Table 1). GF-120 was 1.5 and 2.0 times more effective in capturing females and males, respectively, than Nulure (Table 1). Biolure was the least effective of the attractants followed by Hymlure (Table 1). A comparison of the total (male and female) number of ßies captured among treatments showed that Mazoferm E802 and Torula yeast captured signiþcantly more ßies than the other attractants (Table 1). All attractants were female selective, with percentage female catches ranging from 63 to 74% across all attractants (Table 1). B. invadens Catches There was a signiþcant difference in the mean number of female B. invadens captured daily per trap among treatments (F 31.68; df 5,189; P ), dates (F 16.18; df 7,189; P ), and an interaction between treatments and dates (F 8.01; df 35,189; P ; Table 1). For the mean number of male B. invadens captured daily per trap, there was a signiþcant difference among treatments (F 15.2; df 5,189; P ), dates (F 18.25; df 7,189; P ), and an interaction between treatments and dates (F 2.18; df 35,189; P ; Table 1). For the mean total number of B. invadens captured daily per trap, there was a signiþcant difference among treatments (F 20.11; df 5,189; P ), dates (F 16.20; df 7,189; P ), and an interaction between treatment and dates (F 2.98; df 35,189; P ; Table 1; Fig. 2B). The percentage of adults captured that were female was not signiþcantly different among treatments (F 0.87; df 5,189; P ), dates (F 1.87; df 7,189; P ), and there was no interaction between treatments and dates (F 0.34; df 35,189; P ; Table 1). Mazoferm E802 was the most effective attractant and captured 5.6 and 9.1 times more females and males, respectively, than the standard Nulure (Table 1). The next most effective attractant was Torula yeast followed by GF-120 that captured 2.7 and 1.3 times more females and 5.5 and 2.7 times more males than the standard Nulure respectively (Table 1). Biolure and Hymlure were the least effective of all the attractants tested (Table 1). A comparison of the total number of B. invadens captured (male and female) showed that Mazoferm E802 captured signiþcantly more ßies than any other attractant (Table 1). Among all the attractants, the Nulure was the most female selective, with 74% female catches, while female catches in the other attractants ranged between 51 and 63% (Table 1). Temporal Trend of B. invadens Catches 2007 and During 2007, total B. invadens catches ranged from 0.4 ßies per trap per day (FTD) in traps baited with Biolure to 10.2 FTD in traps baited with Mazoferm E802 at the Þrst sampling date (12 October 2007; Fig. 2A). Peak populations were recorded on 23 November 2007 (21.6Ð85.8 FTD) and thereafter declined at the last day of sampling across all the treatments (11.8Ð69.3 FTD; Fig. 2A). In 2008, total B. invadens catches ranged from 1.1 to 34.2 FTD at the beginning of the trial (Fig. 2B). The populations continued to increase across all the treatments until the last date of sampling (29 November 2008), reaching 10.5Ð185.2 FTD (Fig. 2B). Suppression Trial Monitoring data taken before application of the bait spray showed that B. invadens catches ranged from 0.82 to 4.13 FTD and 0.52Ð8.7 FTD in the orchards assigned to the Mazoferm E802Ðspinosad bait spray and control, respectively, representing no signiþcant difference between treatments at time zero (t 0.56Ð1.82; df 6; P Ð0.6117; Fig. 3A). Two weeks after the Mazoferm E802Ðspinosad spray (13 December 2008), a signiþcant reduction in the number of B. invadens captured in the treated orchards (3.45 FTD) was recorded compared with the control orchards (7.18 FTD; t 8.18; df 6; P ; Fig. 3A). On the last sampling date (24 January 2009), B. invadens catches were 2.86 FTD in orchards treated with Mazoferm E802Ðspinosad and 7.18 FTD in the control orchards (t 21.15; df 6; P ; Fig. 3A). Percentage reduction in B. invadens catches in the treatment relative to the control was 87% after 4 wk and 90% after 8 wk. At harvest, percentage fruit infestation was significantly lower in the Mazoferm E802Ðspinosad-treated orchards (8%) compared with the control orchards (59%; t 26.15; df 6; P ; Fig. 4A). Marketable mango yield was signiþcantly larger (t 9.314; df 6; P ) in orchards receiving Mazoferm E802Ðspinosad bait spray (12,487 kg/ha) than in the

7 304 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 107, no. 1 Fig. 3. Mean number of B. invadens captured per trap per day following application of Mazoferm E802Ðspinosad bait sprays during 2008 (A) and 2009 (B). Arrow indicates dates of spray applications. Error bars denote SE. control (3,606 kg/ha), indicating a 71% yield gain owing to use of the bait spray (Table 2). The total monetary gain owing to baitðinsecticide applications was US$2,132/ha representing a costðbeneþt ratio of 1:9.1 (Table 2). Fig. 4. Mean percentage of mango fruits infested by B. invadens following application of Mazoferm E802Ðspinosad bait sprays during 2008 (A) and 2009 (B). Error bars denote SE. Suppression Trial Monitoring data taken before the bait spray was made showed that B. invadens catches ranged from 1.15 to 5.15 FTD and 1.12Ð5.03 FTD in the orchards assigned to the Mazoferm E802Ðspinosad bait spray and the control, respectively, representing no signiþcant difference between treatments at time zero (t 0.76Ð1.18; df 6; P Ð0.8762; Fig. 3B). Unlike the 2008 data, the impact of the Mazoferm E802Ðspinosad bait spray was not noticeable until 3 wk posttreatment (26 November 2009) when B. invadens catches were signiþcantly lower in the treated orchards (2.76 FTD) than in the control orchards (9.14 FTD; t 9.11; df 6; P ; Fig. 3B). On the last date of sampling (31 December 2009), B. invadens captures were 5.21 FTD in the orchards treated with Mazoferm E802Ðspinosad and FTD in the control orchards (t 22.87; df 6; P ; Fig. 3B). Percentage reduction in B. invadens catches in the treatment relative to the control was 67% after 4 wk and 81% after 8 wk. At harvest, percentage fruit infestation was significantly lower in the Mazoferm E802Ðspinosad-treated orchards (14%) compared with the control orchards (71%; t 31.41; df 6; P ; Fig. 4B). As with the previous year, marketable mango yield was signiþcantly larger (t 8.922; df 6; P ) in orchards receiving Mazoferm E802Ðspinosad bait spray (10,507 kg/ha) than in the control (2,602 kg/ ha), indicating a 75% yield gain owing to use of the bait spray (Table 2). The total monetary gain owing to baitðinsecticide applications was US$1,976/ha representing a costðbeneþt ratio of 1:8.2 (Table 2). Discussion Evaluation of Baits for Field Monitoring of B. invadens Populations. Vargas et al. (2003) reported that the type of protein in a food-based bait can inßuence the attractiveness of that bait to fruit ßies. The results of our experiments demonstrate that, although all the food attractants tested were able to attract both sexes of B. invadens, the level of attraction varied with the different protein baits assessed. We found that Mazoferm and Torula yeast outperformed all the other food attractants in terms of attractiveness to B. invadens while GF-120, the standard Nulure, and Hymlure were moderately effective and Biolure was the least effective. Earlier studies by Moreno and Mangan (1995) in Texas compared the attraction of Anastrepha ludens (Loew) to Mazoferm E802, Torula yeast, casein hydrolysate, yeast hydrolysate, and Nulure as a standard in citrus orchards. They reported that all the baits were signiþcantly more attractive than water or Nulure and indicated that any one of the attractants could be used to lure ßies in ultralow volume spray applications for the management of A. ludens. However, it was concluded that because Torula yeast, yeast hydrolysate, and casein hydrolysate were costly, the best candidate was Mazoferm E802. In Hawaii, Vargas and Prokopy (2006) compared attraction and feeding responses of Bactrocera dorsalis (Hendel) and Bactro-

8 February 2014 EKESI ET AL.: B. invadens BAITS FOR MONITORING AND SUPPRESSION 305 Table 2. Economics of B. invadens suppression using a Mazoferm E802 spinosad bait spray on mango in Kenya Treatment Marketable yield (kg/ha) a % yield gain owing to management Monetary gain (US$/ha) b Cost of Mazoferm E802Ðspinosad bait application (US$/ha) Cost of baitðinsecticide c Labor cost d Sprayer cost e Total cost 2008 MazofermÐspinosad 12,487.3a , :9.1 Control 3,605.5b Ð Ð Ð Ð Ð Ð Ð 2009 MazofermÐspinosad 10,507.3a , :8.2 Control 2,602.9b Ð Ð Ð Ð Ð Ð a Means within a column followed by the same letter are not signiþcantly different from each other by t-test (P 0.05). b Based on mango farm-gate prices of US$0.24/kg in 2008 and US$0.25/kg in c Based on 2008 Mazoferm E802 and spinosad prices of US$0.25/liter and US$160.00/liter, respectively. d Labor cost of bait-spray based on wages of US$4.00 per working day (9.5 h [ 2.4 per ha]). e Cost of sprayer based on depreciation at US$70.00 for a CP3 Knapsack sprayer over an estimated 5-yr life. CostÐbeneÞt ratio cera cucurbitae (Coquillett) to four different protein baits (Provesta 621, Mazoferm E802, GF-120, and Nulure) in Þeld cages. On the basis of attraction and feeding, only Mazoferm E802 rated higher than the standard Nulure among the four baits for both species. The results obtained from our studies with regard to attraction of B. invadens to Mazoferm E802 generally agree with previous studies and suggest that Mazoferm E802 holds great promise in the management of B. invadens in Africa. Torula yeast is currently recommended for detection and management of a wide range of fruit ßies from the genera Anastrepha, Bactrocera, Ceratitis, and Dacus and has been evaluated for detection of a number of different species (IAEA 2003) and also in fruit ßy mass rearing (Chang et al. 2006, Ekesi et al. 2007c). For example, Burrack et al. (2008) compared the effectiveness of Torula yeast in McPhail traps with ammonium bicarbonate as another food-based lure and spiroketal as a pheromone lure and reported that the Torula yeast was more attractive to Bactrocera oleae (Gmelin) than either of the other lures. Most recently, in a markðreleaseðrecapture study, Kendra et al. (2010) assessed the sampling range of Torula yeast and Biolure in a guava grove using both feral and sterile females of the Carribean fruit ßy, Anastrepha suspensa (Loew). The authors reported that recapture of both immature and mature females was signiþcantly greater with Torula yeast than with Biolure. To the best of our knowledge, the use of Torula yeast in fruit ßy management has largely been restricted to detection and monitoring and there are few reported cases of it being used in the Þeld for population suppression. Despite the relatively high cost (Moreno and Mangan 1995), Messing (1999) noted that Torula yeast was the preferred bait for detection and monitoring purpose over standard attractants such as Nulure because its ph remained stable over time at 9.2. However, while a solution of low or high optimal ph may be useful in trapping systems, phytotoxicity problems at such phs may prevent the use of Torula yeast as a bait spray applied directly to plants in Þeld suppression strategies. Research is increasingly moving toward the development of solid bait stations rather than liquid bait for fruit ßy suppression (Epsky et al. 1999, Mangan and Moreno 2007, IAEA 2009) and, given the broad spectrum of activity of Torula yeast for a range of different fruit ßy species, there is an urgent need to explore the development of Torula yeast baits station for Þeld suppression of fruit ßies including B. invadens. Currently, GF-120, a protein bait that also contains spinosad as the toxic element, is the baitðinsecticide mixture of choice in many fruit ßy suppression and eradication programs. For example, the Hawaiian area-wide fruit ßy suppression program uses this baitð insecticide mixture to target and suppress a number of different fruit ßy species in combination with other techniques such as male annihilation and sterile insect techniques (SIT; Vargas et al. 2003, Revis et al. 2004, McQuate et al. 2005, Mau et al. 2007). In Benin, Vayssières et al. (2009) reported signiþcant suppression of native and exotic fruit ßy species using GF-120 bait sprays, and in Kenya GF-120 contributed to a significant reduction of B. invadens when used in combination with the biopesticide Metarhizium anisopliae (Metschnikoff) Sorokin (Ekesi et al. 2011). In the current Þeld study, the attractiveness of GF-120 to B. invadens was signiþcantly lower than Mazoferm E802 and Torula yeast. This is perhaps surprising given that solbait, the active ingredient for attraction (proteinaceous component), in GF-120 is formulated as a dried and puriþed derivative of Mazoferm E802 called solulyls (Moreno and Mangan 2002). Pelz-Stelinski et al. (2006) demonstrated that attraction of Rhagoletis cingulata Loew to GF-120 could be signiþcantly increased by augmenting the concentration of ammonium acetate in the standard formulation, from 1 to 5 or 10% and so it would be worth exploring whether attraction and feeding by B. invadens to GF-120 can be increased using a similar approach. At the tested concentration of 4%, B. invadens were less attracted to odors of Hymlure when compared with Mazoferm E802, Torula yeast, GF-120, and Nulure. The reason for the poor performance of Hymlure is not very clear. However, Fabre et al. (2003) demonstrated that increasing the concentration of Hymlure from 1 to 10% increases the attraction of the bait to B. cucurbitae. It would be interesting to test the attraction of B. invadens to Hymlure at slightly higher concentrations than tested in the current study. The

9 306 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 107, no. 1 possible phytotoxicity effect of higher concentrations to the target host plant would need to be evaluated if used for Þeld suppression. Although originally developed for Ceratitis capitata (Wiedemann) (Heath et al. 2007), Biolure has shown increased attraction to a broad range of fruit ßy species including B. oleae, Bactrocera latifrons (Hendel), and several species of Anastrepha (Canal et al. 2007, Ros et al. 2007, Mziray et al. 2010). Ekesi et al. (2007b) also demonstrated that B. invadens was attracted to odors of the components of the Biolure. However, when compared with other attractants in the current study, Biolure was by far the least effective. This is in contrast to other studies where it has been used to monitor and estimate populations of B. invadens (Mwatawala et al. 2006b). Based on the results of our current study, we would suggest that the population size of B. invadens could be underestimated if Biolure was used for monitoring, and that other attractants, such as Mazoferm E802 and Torula yeast, would be better monitoring tools than Biolure for B. invadens. Apart from Hymlure which was inconsistent, all protein baits tested caught more females than males in both years. This is perhaps not surprising, given that protein baits act as food attractants and their effectiveness relies on the fact that immature females need protein meals to develop mature eggs (Drew and Yuval 2000). In Israel, Multi-lure traps baited with Biolure captured twice as many females of C. capitata than males in trials conducted in citrus orchards (Gazit et al. 1998). Similarly, in Greece, Katsoyannos et al. (1999) reported that Biolure captured two to Þve times more females of C. capitata than males in citrus orchards. The female selectivity for Biolure observed in our study is therefore similar to previous reports. The bait evaluation trials all began at the period of mango fruit maturity when highest numbers of B. invadens were predicted to be present as the fruits ripened toward harvesting. Rwomushana (2008) used the same orchards as we used in our study to assess the seasonal dynamics of B. invadens and reported peak catches in November and December when mangoes were at the ripening stage. The high number of B. invadens captured during November in our study can, therefore, be attributed to the abundance of ripe mangoes during the experimental period. Suppression of B. invadens Populations in the Field. In the Þeld suppression trials, up to 90% (2008) and 81% (2009) reductions in B. invadens catches were recorded in the Mazoferm E802Ðspinosad bait sprayed orchards relative to the untreated control orchards 8 wk after spraying was initiated. This represents a signiþcant level of suppression of B. invadens populations in treated orchards compared with the control orchards. Fruit infestation by B. invadens ranged between 5 and 10% in the Mazoferm E802Ðspinosad bait sprayed orchards compared with 56Ð62% infestation in the control orchards. This is in line with the work of Peck and McQuate (2000) who reported signiþcant levels of suppression of C. capitata on coffee (average daily catch of 1 FTD) compared with unsprayed control Þelds (average daily catches of 28 FTD) following application of Mazoferm E802 mixed with spinosad. Most recently, Mangan and Moreno (2007) showed that Mazoferm E802 bait formulation mixed with phloxine B as the toxic component within a tent baiting station reduced populations of sterile-released A. ludens adults by 70Ð90% in 4 d in a citrus orchard compared with the control orchards treated with the bait but without the Phoxine B toxin. In trials conducted over two seasons in 2005 and 2006 on cherry, Yee (2006) also demonstrated that there were 86Ð92% fewer larvae of Rhagoletis indifferens Curran in fruits following application of a Mazoferm E802 and spinosad bait sprays. There are no comparable studies with regard to use of Mazoferm E802 in conjunction with spinosad for the control of fruit ßies in Africa. However, in Benin, Vayssières et al. (2009) observed up to 8% mango fruit infestation by B. invadens and C. cosyra 10 wk after application of GF-120 (which is also a combination of spinosad and a bait derived from Mazoferm E802) compared with 48% infestation in the control. Ekesi et al. (2011) also reported mango fruit infestation ranging from 28 to 30% by B. invadens in orchards receiving six treatments of GF-120 bait spray compared with 52Ð60% fruit infestation in an untreated mango orchard. GF-120 is, however, still not registered in Kenya unlike Mazoferm E802, which is locally available as an animal feed. Until registration is accomplished, growers now have the option of using Mazoferm E802 for monitoring and suppression of B. invadens. Mango yields were 3.5Ð4 times higher in Mazoferm E802Ðspinosad-treated orchards compared with the untreated control with a costðbeneþt ratio of 1:8.2Ð9.1 in both years. Ordish (1963) showed that a costðbeneþt ratio of 1:1.5 is satisfactory in sophisticated agriculture with crops of a high value per hectare. However, Bernet (1974) argued that higher costðbeneþt ratios of 1:3 and above were needed in small-scale farming before growers would invest in pest management measures. CostÐbeneÞt ratios obtained during the 2 yr of the bait spray trials are satisfactory for both small-scale and large-scale farming (Ordish 1963, Bernet 1974) and proþtable for the management of B. invadens. One concern with our trials was that we did not measure ßy migration and the degree to which ßy movement occurred within and between treatments and blocks. However, members of the genus Bactrocera Macquart have relatively low movement in the presence of host plants (Fletcher 1987). In markð releaseðrecapture studies, Peck and McQuate (2004) showed that movement of B. latifrons did not exceed 200 m during the 6-wk study period. Fletcher and Economopoulos (1976) showed that migration of B. oleae did not exceed 190 m from the release point after 14 d. In other tephritid species, Kovaleski et al. (1999) found that 94.7% of marked Anastrepha fraterculus (Wiedemann) remained within 200 m of the release point for 20 d while Plant and Cunningham (1991) showed that 70% of released and recaptured C. capitata stayed within 150 m during the 8 d of the experiment. One exception to the above studies is the re-

10 February 2014 EKESI ET AL.: B. invadens BAITS FOR MONITORING AND SUPPRESSION 307 capture of A. ludens at 9 km from the release point (Thomas and Loera-Gallardo, 1998). Although caution must be taken in generalizing migration in markð releaseðrecapture studies, the distances between treatments within blocks ( 1.5 km) and experimental blocks ( 3Ð4 km) used in our study, which are also within the range used by other authors working on fruit ßy Þeld suppression (Peck and McQuate 2000, Piñero et al. 2009, Vayssières et al. 2009, Ekesi et al. 2011), greatly exceed migration range of majority of the fruit ßy species listed above. It is therefore unlikely that ßy migration affected the results of our experiments. In conclusion, B. invadens is rated as a devastating quarantine pest (French 2005, USDAÐAPHIS 2008). Although a fruit infestation range of only 5Ð10% was observed in orchards following treatment with Mazoferm E802Ðspinosad and an appreciable costðbeneþt ratio may be satisfactory for growers targeting domestic local and urban markets in Kenya, this level of control is still not sufþcient for lucrative quarantinesensitive markets abroad. Previous experience with other fruit ßy species in similar agroecologies in Latin America and the South PaciÞc has shown that fruit ßy suppression is unlikely to be successful if based on a single management technique (Aluja et al. 1996, Allwood and Drew 1997). An integrated pest management approach offers the best method to improve the economies of the production system by reducing yield losses and enabling growers to comply with stringent quality on the export market (Lux et al. 2003). In this regard, we recommend that the baiting technique used in this study should be integrated with other methods such as male annihilation technique (MAT) using methyl eugenol, application of biopesticides, use of parasitoids, and orchard sanitation (Ekesi et al. 2007a, Mohamed et al. 2010, Ekesi et al. 2011). Furthermore, without postharvest treatment to provide quarantine security, export of fruits and vegetables attacked by B. invadens will be limited by quarantine restrictions. To complement the preharvest management measures described here, cold disinfestation treatments for citrus and avocado have been developed (Grout et al. 2011, Ware et al. 2012) and hot water treatment parameters for mangoes are being generated (S. E. et al., unpublished data). Once completed and implemented, this should increase the export market potential of mango fruits from Africa. Acknowledgments We thank the mango growers at Nguruman and Nthagaiya and especially Bainito Atonya and Henry Ngare for providing the experimental orchards for the trials. We also greatly appreciate the efforts by P. W. Nderitu and P. Agola for their various contributions to the Þeld collection of data and processing. We thank E. Muchugu for the map of the experimental site. Comments from S. Subramanian greatly improved the quality of the article. The German Ministry for Economic Cooperation and Development (BMZ) provided funding for this research. References Cited Allwood, A. J., and R.A.I. Drew Control strategies for fruit ßies (family Tephritidae) in the South PaciÞc, pp. 171Ð178. In A. J. Allwood and R.A.I. Drew (eds.), Management of Fruit Flies in the PaciÞc. ACIAR Proceedings No. 76. Sun Photoset Pty Ltd., Brisbane, Australia. Aluja, M., H. Celedonio-Hurtado, P. Liedo, M. Cabrera, F. Castillo, J. Guillen, and E. Rios Seasonal population ßuctuations and ecological implications for management of Anastrepha fruit ßies (Diptera: Tephritidae) in commercial mango orchards in southern Mexico. J. Econ. Entomol. 89: 654Ð667. Barry, J. D., W. J. Sciarappa, L.A.F. Teixeira, and S. Polavarapu Comparative effectiveness of different insecticides for organic management of blueberry maggot (Diptera: Tephritidae). J. Econ. Entomol. 98: 1236Ð1241. Bernet, E. E Economic aspects of pesticide use with special reference to African conditions, p. 36. In Proceedings of seminar on the efþcient and safe use of pesticides in agriculture and public health in Africa, OctoberÐNovember 1974, Amsen, Nairobi, Kenya. Burrack, H. J., J. Connell, and F. Zalom Comparison of olive fruit ßy (Bactrocera oleae (Gmelin)) (Diptera: Tephritidae) captures in several commercial traps in California. Int. J. Pest Manage. 54: 227Ð234. Canal, N. A., M. C. De Chacon, P. E. Galeano, S. G. Gomez, M. J. Gomez, J. H. Ramirez, A. Osorio, and M. R. Castaneda Response of Anastrepha species (Diptera: Tephritidae) to synthetic attractants in Colombia, pp. 129Ð141. In Development of improved attractants and their integration into fruit ßy SIT management programmes. Proceedings of a Final Research Coordination Meeting, IAEA-TECDOC-1574, 5Ð7 May International Atomic Energy Agency, Vienna, Austria. Chang, C. L., R. I. Vargas, C. Caceres, E. Jang, and I. K. Cho Development and assessment of a liquid diet for Bactrocera dorsalis (Diptera: Tephritidae). Ann. Entomol. Soc. Am. 99: 1191Ð1198. Copland, R. S Host fruit processingñpurpose, tools and methodology, pp. F1ÐF11. In S. Ekesi, and M. K. Billah (eds.), A Field Guide to the Management of Economically Important Tephritid Fruit Flies in Africa. ICIPE Science Press, Nairobi, Kenya. Drew, R.A.I., K. Tsuruta, and I. M. White A new species of pest fruit ßy (Diptera: Tephritidae: Dacinae) from Sri Lanka and Africa. Afr. Entomol. 13: 149Ð154. Drew, R.A.I., and B. Yuval The evolution of fruit ßy feeding behaviour. Fruit ßies, pp. 731Ð749. CRC, Boca Raton, FL. Ekesi, S., S. Dimbi, and N. K. Maniania. 2007a. The role of entomopathogenic fungi in the integrated management of Tephritid fruit ßies (Diptera: Tephritidae) with emphasis on species occurring in Africa, pp. 239Ð274. In S. Ekesi and N. K. Maniania (eds.), Use of Entomopathogenic Fungi in Biological Pest Management. Research SignPost, Kerala. Ekesi, S., P. W. Nderitu, and I. Rwomushana Field infestation, life history and demographic parameters of Bactrocera invadens Drew, Tsuruta & White, a new invasive fruit ßy species in Africa. Bull. Entomol. Res. 96: 379Ð386. Ekesi, S., S. A. Lux, and M. K. Billah. 2007b. Field comparison of food-based synthetic attractants and traps for African Tephritid fruit ßies, pp. 205Ð222. In Development of Improved Attractants and Their Integration Into Fruit Fly SIT Management Programmes. Proceedings of a Final Research Coordination Meeting, IAEA-TECDOC-1574,

11 308 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 107, no. 1 5Ð7 May 2005, International Atomic Energy Agency, Vienna, Austria. Ekesi, S., P. W. Nderitu, and C. L. Chang. 2007c. Adaptation to and small-scale rearing of the invasive fruit ßy Bactrocera invadens (Diptera: Tephritidae) on artiþcial diet. Ann. Entomol. Soc. Am. 100: 562Ð567. Ekesi, S., M. K. Billah, P. W. Nderitu, S. A. Lux, and I. Rwomushana Evidence for competitive displacement of the mango fruit ßy, Ceratitis cosyra by the invasive fruit ßy, Bactrocera invadens (Diptera: Tephritidae) on mango and mechanisms contributing to the displacement. J. Econ. Entomol. 102: 981Ð991. Ekesi, S., N. K. Maniania, and S. A. Mohamed EfÞcacy of soil application of Metarhizium anisopliae and the use of GF-120 spinosad bait spray for suppression of Bactrocera invadens (Diptera: Tephritidae) in mango orchards. Biocontrol Sci. Technol. 21: 299Ð316. Epsky, N. D., J. Hendrichs, B. I. Katsoyannos, L. A. Vasquez, J. P. Ros, A. Zumreoglu, R. Pereira, A. Bakri, S. I. Seewooruthun, and R. R. Heath Field evaluation of female-targeted trapping systems for Ceratitis capitata (Diptera: Tephritidae) in seven countries. J. Econ. Entomol. 92: 156Ð164. Fabre, F., P. Ryckewaert, P. F. Duyck, F. Chiroleu, and S. Quilici Comparison of the efþcacy of different food attractants and their concentration for melon ßy (Diptera: Tephritidae). J. Econ. Entomol. 96: 231Ð238. Fletcher, B. S The biology of Dacine fruit ßies. Annu. Rev. Entomol. 32: 115Ð144. Fletcher, B. S., and A. P. Economopoulos Dispersal of normal and irradiated laboratory strains and wild strains of the olive ßy Dacus oleae in an olive grove. Entomol. Exp. Appl. 20: 183Ð194. French, C The new invasive Bactrocera species. Insect pest control newsletter, No. 65, pp. 19Ð20. International Atomic Energy Agency, Vienna, Austria. Gazit, Y., Y. Rossler, N. D. Epsky, and R. R. Heath Trapping females of the Mediterranean fruit ßy (Diptera: Tephritidae) in Israel: comparison of lures and trap types. J. Econ. Entomol. 91: 1355Ð1359. Goergen, G., J.-F. Vayssières, D. Gnanvossou, and M. Tindo Bactrocera invadens (Diptera: Tephritidae), a new invasive Fruit ßy pest for the afrotropical region: host plant range and distribution in West and Central Africa. Environ. Entomol. 40: 844Ð854. Gilbert, A. J., R. Bingham, M. A. Nicolas, and R. A. Clark Insect trapping guide. Pest Detection/Emergency Projects, State of California Department of Food and Agriculture, Sacramento, CA. Grout, T. G., J. H. Daneel, S. A. Mohamed, S. Ekesi, P. W. Nderitu, P. R. Stephen, and V. Hattingh Cold susceptibility and disinfestation of Bactrocera invadens (Diptera: Tephritidae) in Oranges. J. Econ. Entomol. 104: 1180Ð1188. Guichard, C EU Interceptions rising in In Fighting fruit ßies regionally in sub-saharan Africa, p. 4. COLEACP/CIRAD, Information Letter 4. COLEACP, Cedex, France. Heath, R. R., R. E. Burns, P. E. Kendra, and R. Mangan Fruit ßy trapping and controlñpast, present and future. In Development of Improved Attractants and Their Integration Into Fruit Fly SIT Management Programmes. Proceedings of a Final Research Coordination Meeting, IAEA-TECDOC-1574, 5Ð7 May 2005, pp. 7Ð10. International Atomic Energy Agency, Vienna, Austria. Henderson, D. F., and E. W. Tilton Tests with acaricides against brown wheat mite. J. Econ. Entomol. 48: 157Ð161. (IAEA) International Atomic Energy Agency Trapping guidelines for Area-wide fruit ßy programmes. International Atomic Energy Agency, Vienna, Austria. (IAEA) International Atomic Energy Agency Development of bait stations for fruit ßy suppression in support of SIT. Report and recommendations of the consultants group meeting organized by the Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, Mazatlán, Mexico, 30 OctoberÐ1 November Karel, A. K., and G. C. Ashimogo Economics of insect control on common beans and soybeans in Tanzania. J. Econ. Entomol. 84: 996Ð1000. Katsoyannos, B. I., R. R. Heath, N. T. Papadopoulos, N. D. Epsky, and J. Hendrichs Field evaluation of Mediterranean fruit ßy (Diptera: Tephritidae) female selective attractants for use in monitoring programs. J. Econ. Entomol. 92: 583Ð589. Kendra, P. E., N. D. Epsky, and R. R. Heath Effective sampling range of food-based attractants for female Anastrepha suspensa (Diptera: Tephritidae). J. Econ. Entomol. 103: 533Ð540. Kovaleski, A., R. Sugayama, and A. Malavasi Movement of Anastrepha fraterculus from native breeding sites in to apple orchards in Southern Brazil. Entomol. Exp. Appl. 91: 457Ð463. Lux, S. A., S. Ekesi., S. Dimbi, S. A. Mohamed, and M. K. Billah Mango infesting fruit ßies in Africa - perspectives and limitations of biological approaches to their management, pp. 277Ð293. In P. Neuenschwander, C. Borgemeister, and J. Langewald (eds.), Biological Control in Integrated Pest Management Systems in Africa. CAB International, Wallingford, United Kingdom. Mangan, R. L., and D. Moreno Development of bait stations for fruit ßy population suppression. J. Econ. Entomol. 100: 440Ð450. Mau, R., E. Jang, and R. Vargas The Hawaii Area-wide fruit ßy pest management programme: inßuence of partnerships and a good education programme, pp. 671Ð683. In M.J.B. Vreysen, A. S. Robinson, and J. Hendrich (eds.), Area-Wide Control of Insect Pests: From Research to Field Implementation. Springer, Amsterdam, The Netherlands. McQuate, G. T., C. D. Sylva, and E. B. Jang Mediterranean fruit ßy (Diptera: Tephritidae) suppression in persimmon through bait sprays in adjacent coffee plantings. J. Appl. Entomol. 129: 110Ð117. Messing, R Managing fruit ßies on farms in Hawaii, University of Hawaii. CTAHR Brief. No. 4. Mohamed, S. A., S. Ekesi, and R. Hanna Old and new host-parasitoid associations: parasitization of the invasive fruit ßy Bactrocera invadens (Diptera: Tephritidae) and Þve other African fruit ßy species by Fopius arisanus, an Asian opiine parasitoid. Biocontrol Sci. Technol. 10: 183Ð 196. Moreno, D. S., and R. L. Mangan Responses of the Mexican fruit ßy (Diptera: Tephritidae) to two hydrlosed proteins and incorporation of phloxine B to kill adults, pp. 257Ð279. In R. J. Heitz and K. Downum (eds.), Light Activated Pest Control. Supplement. ACS Symposium Series 616. American Chemical Society, Washington, DC. Moreno, D. S., and R. L. Mangan Bait matrix for novel toxicants for use in control of fruit ßies (Diptera: Tephritidae), pp. 333Ð362. In G. Hallman, G. and C. P. Schwalbe (eds.), Invasive Arthropods in Agriculture. Science Publishers Inc., EnÞeld, NH. Mwatawala, M. W., M. De Meyer, R. H. Makundi, and A. P. Maerere. 2006a. Seasonality and host utilization of the