The use of sniffing dogs for the detection of Rhynchophorus ferrugineus

Transcription

1 Phytoparasitica (2014) 42: DOI /s The use of sniffing dogs for the detection of Rhynchophorus ferrugineus Pompeo Suma & Alessandra La Pergola & Santi Longo & Victoria Soroker Received: 2 April 2013 /Accepted: 14 July 2013 /Published online: 14 August 2013 # Springer Science+Business Media Dordrecht 2013 Abstract The red palm weevil (RPW) Rhynchophorus ferrugineus (Olivier) (Coleoptera: Dryophthoridae) is the most serious pest of Canary Island date palms (Phoenix canariensis Hort. ex Chabaud) and date palms (Phoenix dactylifera L.). Early infestations of this invasive weevil are difficult to detect under field conditions. The adults are very elusive and not visible inside the dense frond of the palms. Efficient methods for detecting this pest are needed for its successful management under field conditions and in particular at ports of entry to avoid its' spreading via infested imported palms. The present study evaluated the ability of Rottweiler and Golden Retriever dogs (Canis familiaris L.) to detect Canary Island palms that were baited with RPW adults or larvae under field conditions. During these experiments, the dogs successfully detected 78% of the baits hidden in the crown of the palms. These encouraging results need to be confirmed by evaluation of the performance of sniffing dogs in detection of naturally infested palms of different species. Keywords Early detection. Phoenix canariensis. Phoenix dactylifera. Red palm weevil P. Suma (*) : A. La Pergola : S. Longo Department of Agri-food and Environmental Systems Management, Section of Applied Entomology, University of Catania, via S. Sofia 100, Catania, Italy suma@unict.it V. Soroker Department of Entomology, Agricultural Research Organization, The Volcani Center, P.O. Box 6, Bet Dagan 50250, Israel Introduction During the last two decades, infestations of the red palm weevil (RPW) Rhynchophorus ferrugineus (Oliv.) (Coleoptera: Dryophthoridae) have spread in the Middle East, North Africa and Mediterranean Europe (Barranco et al. 1996; Buyukozturk et al. 2011; Dembilio et al. 2009). The weevil also has been found in the Caribbean basin (Aruba, in 2008) (EPPO RS 2009/002) and recently in California (USDA APHIS-PPQ, 2010). The devastating damage induced by this insect to Canary Island date palms (Phoenix canariensis Hort. ex Chabaud) and other ornamental palms in Italy has been well documented (Conti et al. 2008; Longo & Tamburino 2005; Sacchetti et al. 2005). More recently, in Sicily, the weevil infestation was also recorded on other ornamental palms including Dwarf palmetto (Sabal minor), Guadalupe Palm (Brahea edulis), and Pigmy Date Palm (Phoenix roebelenii O Brien) (Giovino et al. 2012; Longo et al. 2011; P. Suma, unpublished data). Field trials have shown that early infestations of this weevil can be controlled by using chemical (e.g. injections of systemic insecticides into the palm stem) and non-chemical (e.g. frond pruning) approaches (Ferry & Gómez 2008; Hernàndez- Marante et al. 2003; La Mantia et al. 2008). However, early infestations usually do not cause obvious visual symptoms and escape detection. At these early stages of the infestation, which may last several months, the oviposition sites are invisible while adult weevils are elusive and hidden by the dense vegetation of the fronds (Rahalker et al.1985; Wattanapongsiri 1966). Freshly hatched larvae bore into the petioles of the fronds without visible production of frass or other specific symptoms. Crown asymmetry is visible in the

2 270 Phytoparasitica (2014) 42: case of an extended infestation level when the palm is seriously damaged. Only at advanced stages of the infestation the injury symptoms become obvious because of the collapse of the fronds induced by the feeding and tunneling of the larvae and also because of fermentation odors emitted by the infested palm tissues. These advanced infestation stages are rarely successfully managed and often result in the palm s death (Blumberg et al. 2001; Ferry&Gomez 1998; Murphy &Briscoe1999). The detection of the RPW early infestation is a prerequisite for successful control/management of the pest. Various methods and approaches were evaluated over the years for early detection of RPW infestations, including visual and acoustic methods. Visual inspections using basket cranes are not only very expensive, but not feasible in impervious sites and often ineffective, due to the concealed nature of the pest, as previously stated. Acoustic detection of RPW feeding and tunneling larvae is effective only with active larvae, but fails to detect inactive stages such as eggs and pupae (Gutiérrez et al. 2010; Mankin 2011; Pinhas et al. 2008; Soroker et al. 2004). Other detection practices rely on the capability of dogs to sense volatile chemical compounds. These may be derived directly from the palms, damaged plant tissue, weevils or weevil frass. It is known that RPW males produce an aggregation pheromone that attracts both sexes (Hallett et al. 1993). Because dogs are reliable and efficient scent-detectors with specialized olfactory organs (Albone 1984), they can be trained to locate biological and non-biological scents (Browne et al. 2006) and are able to locate small quantities of the target odor (Oxley &Waggoner 2009). Dogs are, in fact, used for biosecurity in locations such as airports, ports and borders, to prevent introduction of unwanted items including many pests (Adams & Johnson 1994; Lorenzo et al. 2003; Rouhi1997). The use of dogs with associated handlers was also reported to detect chemical contaminant, illegal drugs, and search for missing people, avalanche victims and survivors at disaster sites (Arner et al. 1986; Fenton 1992; Hebard 1993). Dogs are considered to be the most reliable and cost efficient explosive detectors (Furton & Myers, 2001; Lorenzo et al. 2003, cited by Browne et al. 2006); they can detect a target compound in mixtures, but they learn to respond to only a few as it was shown for the diagnosis of some types of cancer (Pickel et al. 2004). There are many accounts of dogs trained to locate insect pests such as the Gypsy moth Lymantria dispar L. (Wallner & Ellis 1976), the Eastern subterranean termites Reticulitermes flavipes (Koehlar) (Brooks et al. 2003), the bed bug Cimex lectularius (L.) (Pfiester et al. 2008), and the red imported fire ant Solenopsis invicta Buren (Lin et al. 2011). Nakash et al. (2000) conducted a preliminary test on the possibility of using dogs for detection of tissue from RPW-infested date palms (Phoenix dactylifera L.) in date plantations, with encouraging results. However, the efficiency of detection has not yet been determined. This study was conducted in Sicily, where the damage inflicted by the weevil is severe. The objective was to evaluate the accuracy and detection ability of dogs to search for and locate Canary Island date palms baited with live life stages of the red palm weevil under field conditions. Materials and methods Rhynchophorus ferrugineus Adult weevils, collected from traps baited with the aggregation pheromone Ferrugineol [4S,5S]-4-methyl-5-nonanol (Rhyfer 220, Intrachem Bio Italia), ethyl acetate and a water solution of sugar beet molasses as co-attractants, were used to start the stock colonies. The weevils were sorted into groups of 25 pairs in PVC cages ( cm depth) perforated on the upper side, covered by mesh, and containing slices of apple or palm trunk as food. The cages were kept at 25 ± 1 C, 75 ± 5% r.h., and a 16-h light photoperiod in an insectary. The colonies were periodically supplemented with specimens of different developmental stages originating from infested Canary Island date palms from the urban and suburban area of Catania province (Italy). Larvae were separated by size and reared in groups of 30 in plastic cages provided with apple slices and moved to a new vial containing parts of fresh palm tissue used to build the cocoon at pupation time. Cocoons were kept in individual covered plastic boxes until adult emergence. Dogs and their training At the beginning of this study four dogs two Golden Retrievers (GR1 and GR2) and two Rottweilers (RW1 and RW2) were utilized: GR1 was a 7-yr-old female, GR2 a 6-yr-old female, RW2 a 9- yr-old female previously trained for utility work (i.e. pet therapy), and RW1 a 6-month-old male. None had any previous training in sniffing.

3 Phytoparasitica (2014) 42: The training procedure used was in accordance with the USCS method, allowing the dog to associate a retrieve toy (tennis ball) with the target odor (USCS 1979). For this purpose, the dogs were goaded into playing with a tennis ball drilled and filled with RPW adults or with a mixture of 30 g of RPW-infested palm tissues and five late instar larvae in order to familiarize the dog with this kind of scent. In the second part of training, the tennis ball was hidden underground. The handler verbally urged the dog to find the ball. The dog that found and retrieved the ball was rewarded. The last training steps were carried out in a small experimental plot (ca 20 m 2 ), using potted Canary Island date palms (P. canariensis) artificially baited with PVC tubes (Falcon 50 ml conical tubes) with perforated caps to allow scent release and filled with the same RPW material used to bait the balls. The baited tubes were arranged randomly in the fronds of the palms. The dogs were then goaded into searching the baited palms. At the end of the search, the Golden Retrievers sat down near and the Rottweilers barked at the baited trees, each waiting for their reward. Thrice-weekly training sessions were carried out for 3 months with few differences among dogs. At the end of these training steps, three dogs were selected: the two Golden Retrievers (GR1 and GR2) and one Rottweiler (RW2). Although the young RW1 was classified at the beginning of the training activity as adequate for the field evaluation, it was decided to exclude it from the final evaluation because it became easily distracted, losing concentration during the search tests. All the dog trials were conducted according to the standards of the Council for International Organizations of Medical Sciences (CIOMS 1985). Field experiment The tests were conducted outdoors in in a randomized block design. A total of 12 uninfested, potted P. canariensis (4 5 years old and cm stem diam) palms were arranged in a grid pattern of three rows of four, 1 m distant from each other. Four different odors (=baits) were simultaneously used in the trial to test the detection ability of the dogs. The baits used were placed in PVC tubes (Falcon 50 ml conical tubes) with perforated caps to allow scent release and filled with: (i) five RPW male adults, (ii) five RPW female adults, (iii) five pairs of adult males and females, (iv) five late instar larvae (4 5 cm in length) of R. ferrugineus plus about 30 g of infested palm material, and (v) an empty tube used as control. The tubes containing the RPW life-stages and empty controls were used as a dog target in three trials conducted in 3 days. In each trial four tubes baited with RPW and eight empty control tubes were randomly hidden in the crown of the 12 palm trees. The arrangement of the tubes in the crown of the palm trees differed in each trial. Tubes were hidden in the palm trees 3 h before the search by the dogs. The dog handlers were not aware of target locations, to prevent any bias. Each dog was then guided on a leash by the handler to sniff each single palm, allowing three passes per palm tree. Palms targeted by the dogs were checked for the presence of the bait. Dogs were rewarded when they detected the baited plant. Each dog was exposed to the 12 (4 RPW + 8 empty) tubes in each trial. The dog s response during three passes per single palm tree was recorded. There were, in total, 36 chances for baits to be recognized by each dog in each trial. The dogs' responses to the four different baited tubes in the three trials were categorized as true/positive and false/positive. Successful location of the tubes containing the RPW instars was recorded as a true positive response, whereas incorrect location of the empty tubes was recorded as a false positive response. Statistical analysis Data on the responses of dogs from the 3 days of palm inspection were analyzed by using one-way ANOVA. As the effect of days was not found to be significant, the data were pooled to calculate the percentage of true positive and false positive responses for each dog. True positive responses were calculated out of the total baited palm trees examined, multiplied by the number of examinations (total 36) and false positives out of the total examinations of palm trees with the empty tubes (total 72). Data were then transformed (arcsine square root) and analyzed by two-way analysis of variance (ANOVA), with main effects as the dogs and the baits. Means were separated by the Student- Newman-Keuls test (P=0.05; STATSOFT Italia, 1997). Results The reactions of the dogs were separated into two parts: true/positive detection reaction to tubes containing RPW baits, and false/positive detection reaction to empty tubes. The presence of the RPW live instars inside the tubes significantly affected the responses (F 1,313 =

4 272 Phytoparasitica (2014) 42: Table 1 Comparison between different dogs in Rhynchophorus ferrugineus (red palm weevil, RPW) detection ability. The data are mean percentages of true/positive and false/positive responses by dogs trained in searching for RPW when allowed to inspect unbaited or baited palm trees with different live weevil life stages Dog z True/positive responses y (mean% ± SE) RW 78 ± 7.0 a 11 ± 3.9 a GR1 81 ± 6.7 a 17 ± 4.5 a GR2 75 ± 7.3 a 10 ± 3.5 a Mean w 78 ± ± 3.9 False/positive responses x (mean% ± SE) z RW = Rottweiler; GR1 and GR2 = Golden Retriever; y Response to baited palm trees with RPW live instars x Response to unbaited palm trees. w Within columns, means do not differ significantly (P=0.05; Student-Newman-Keuls on arcsin square root transformed data; STATSOFT Italia, 1997). a No statistical differences are present between the evaluated parameters ; P<0.05). Trained dogs were able to provide true/positive indications in 78% of the cases (Table 1) for all the baits used (males, females, pairs and larvae plus infested tissues). There was no significant difference among each dog s ability positively to indicate the presence of the RPW in the inspected palm trees (F 2,105 =0.15; P=0.8; Table 1). Although detection seemed to be better with female and pair baits tested, no significant difference was found among the dogs in their capability to discriminate the different baits (F 6,96 =0.29; P=0.94) (Fig. 1). On the other hand, dogs falsely indicated the presence of the RPW instars in 12% (Table 1) of the palms. Responses by the dogs to palms baited with empty tubes were not significantly different (F 2,204 =0.94; P=0.39) (Table 1). No significant differences were observed in the capability of the dogs to smell specific weevil life stages for both the true/positive and false/positive responses (F 3,104 =1.88; P=0.13 and F 3,203 =0.97; P=0.41, respectively) (Table 2). Discussion Detection of a wood-boring pest is a challenging task. In the case of the red palm weevil infestation there is no information of induced emission of unique volatile organic compounds by the palm tree in response to insect attack. On the other hand, red palm weevil infestation causes fermentation of affected palm tissue that is probably associated with a release of some volatiles. The infested palms may contain different developmental stages of the pest. Indeed, all three trained dogs demonstrated the ability correctly to identify RPW-baited palms in 78% of cases regardless of the different baits tested, with a false/positive response rate of 12%. This level of sensitivity is similar to previous studies on detector dogs, e.g. dogs trained to detect brown tree snakes hidden in cargo from Guam had an overall accuracy of 70% (Engeman et al. 1998) and three dogs trained to detect catfish off-flavor compound showed an overall accuracy of 77% (Shelby et al. 2004). However, in many studies on insect detection by dogs, the minimum acceptable standard for positive response is 90% and a false/positive accordingly: 10%. In some cases, detection accuracy above 90% was also reported: a Fig. 1 Percentage of true positive indications (mean% ± SE) by dogs of palm trees baited with live red palm weevil life stages enclosed in PVC tubes. No significant difference was observed among the dogs in their ability to detect the different baits (P=0.05; Student-Newman-Keuls on arcsin square root transformed data; STATSOFT Italia, 1997). RW, Rottweiler; 1GR and 2 GR, Golden Retrievers

5 Phytoparasitica (2014) 42: Table 2 The differences in detectability of different baits. The data are mean percentages of true/positive and false/positive responses by three dogs when allowed to inspect unbaited or baited palm trees with males, females, larvae or pairs of Rhynchophorus ferrugineus (red palm weevil, RPW) Baits True/positive responses z (mean% ± SE) x males 67 ± 9.2 a 17 ± 5.3 a larvae 70 ± 8.9 a 15 ± 5.0 a females 89 ± 6.2 a 9 ± 4.2 a pairs 85 ± 7.0 a 8 ± 3.7 a False/positive responses y (mean% ± SE) x z Response to palm trees baited with RPW live life stages dogs is a new approach that deserves to be taken into consideration and should be added to the conventional RPW inspection methods for the detection of this pest. Acknowledgments The authors would like to thank Dr. Renato Inserra of the Division of Plant Industry, Florida Department of Agriculture and Consumer Services, for having read and commented on an early version of this paper, and the European Community s Seventh Framework Programme under grant agreements: No. FP7 KBBE Grant "Palm Protect" that is allowing us to continue the research aimed at evaluating the ability of the dogs in early detection of the palm trees infested by the red palm weevil. y Response to unbaited palm trees x Within columns, means do not differ significantly (P=0.05; Student-Newman-Keuls on arcsin square root transformed data; STATSOFT Italia, 1997). References German pointer trained to locate screwworms showed a detection rate of 99.7% (Welch 1990); German shepherds trained to detect gypsy moth egg masses showed an overall detection accuracy of 95% (Wallner & Ellis 1976); six dogs trained to detect the Eastern subterranean termites showed an overall accuracy of 96% (Brooks et al. 2003) and recently, Lin et al.(2011)were able to train three beagles to detect the red imported fire antwithanaccuracyof98%. The relatively high rates of false/positive responses in our study are in agreement with other works reported in the literature (Engeman et al. 1998; Shelby et al. 2004) and could be caused by factors related to the difficulty of the task, inadequate training (e.g. length of the searching time) and/or to intrinsic characteristics of the tested dogs (e.g. the age, the breed, their personality) and eventually to the lack of previous experience in sniffing. In conclusion, our study indicated that trained dogs can locate Canary Island date palms containing an odor emission of the weevil under field conditions. However, the dog s ability to detect the early infestation stages remains to be determined. The use of sniffing dogs seems to fit well for palm inspection at nurseries, ports of entry and/or quarantine facilities where the fronds of palms are accessible and environmental conditions favor dog sniffing activity. Studies are in progress to evaluate the detection limits and minimal detectable infestation levels by the dogs. The effect of palm size, species and infestation locus are important factors that are evaluated in these studies. The RPW detection using Adams, G. J., & Johnson, K. G. (1994). Sleep, work, and the effects of shift work in drug detector dogs Canis familiaris. Applied Animal Behaviour Science, 41, Albone, E. S. (1984). Mammalian semiochemistry: the investigation of chemical signals between mammals. Chichester, UK: Wiley. Arner, L. D., Johnson, G. R., & Shovronek, H. S. (1986). Delineating toxic areas by canine olfaction. Journal of Hazardous Materials, 13, Barranco, P., de la Peña, J., & Cabello, T. (1996). El picudo rojo de las palmeras, Rhynchophorus ferrugineus (Olivier), nueva plaga en Europa (Coleoptera, Curculionidae). Phytoma España, 67, Blumberg, D., Navon, A., Kehat, E., & Levski, S. (2001). Date palm pests in Israel early second millennium. Alon Hanotea, 55, Brooks, S. E., Oi, F. M., & Koehler, P. G. (2003). Ability of canine termite detectors to locate live termites and discriminate them from non termite material. Journal of Economic Entomology, 96, Browne, C., Stafford, K., & Fordham, R. (2006). The use of scent detection-dog. Irish Veterinary Journal, 59, Buyukozturk, H. D., Kutuk, H., & Birisik, N. (2011). Current status of red palm weevil in Turkey. EPPO Bulletin, 41, CIOMS. (1985). Council for International Organizations of Medical Sciences. Principios éticos en la investigación: International guiding principles for biomedical research involving animals [accessed June, 2013]. Available: ch/publications/guidelines/1985_texts_of_guidelines.htm. Conti, F., Tamburino, V., Raciti, E., & Sesto, F. (2008). Il punteruolo rosso delle palme Rhynchophorus ferrugineus in Sicilia orientale: monitoraggio e strategie di contenimento. In conference paper: Giornate Fitopatologiche 2008, Cervia (RA), marzo 2008, (1) pp Dembilio, O., Jacas, J. A., & Llácer, E. (2009). Are the palms Washingtonia filifera and Chamaerops humilis suitable hosts for the red palm weevil, Rhynchophorus ferrugineus (Col. Curculionidae)? Journal of Applied Entomology, 133,

6 274 Phytoparasitica (2014) 42: Engeman, R. M., Vice, D. S., Rodriguez, D. V., Gruver, K. S., Santos, W. S., & Pitzler, M. E. (1998). Effectiveness of the detector dogs used for deterring the dispersal of brown tree snakes. Pacific Conservation Biology, 4, EPPO Reporting Service (2009). No. 1, /002 First record of Rhynchophorus ferrugineus in Curaçao, Netherlands Antilles. Fenton, V. (1992). The use of dogs in search, rescue and recovery. Journal of Wilderness Medicine, 3, Ferry, M., & Gomez, S. (1998). The red palm weevil in the Mediterranean area. Journal of the International Palm Society, 46, Ferry, M., & Gómez, S. (2008). Une nouvelle stratégie contre le charançon rouge des palmiers. Phytoma La Défense des Végétaux, 620, Furton, K. G., & Myers, L. J. (2001). The scientific foundation and efficacy of the use of canines as chemical detectors for explosives. Talanta, 54, Giovino, A., Scibetta, S., Gugliuzza, G., Longo, S., Suma, P., & La Mantia, T. (2012). Attacks of Rhynchophorus ferrugineus (Olivier) (Coleoptera Curculionidae) on natural specimens of Dwarf Fan Palm Chamaerops humilis L. (Arecaceae) in Sicily. Naturalista Siciliano, S. IV, Xxxvi, Gutiérrez, A., Ruiz, V., Moltó, E., Tapia, G., & Téllez, M. (2010). Development of a bioacoustic sensor for the early detection of Red Palm Weevil (Rhynchophorus ferrugineus Olivier). Crop Protection, 29, Hallett, R. H., Gries, G., Gries, R., Borden, J. H., Czyzewska, E., Oehlschlager, A. C., et al. (1993). Aggregation pheromones of two Asian palm weevils. Rhynchophorus ferrugineus and R. vulneratus. Naturwissenschaften, 80, Hebard, C. (1993). Use of search and rescue dogs. Journal of the American Veterinary Medical Association, 203, Hernàndez-Marante, D., Folk, F., Sanchez, A., & Fernandez- Escobar, R. (2003). Control de curculionide ferruginoso de las palmeras (Rhynchophorus ferrugineus Olivier) mediante inyecciones al tronco y pulverizacòn foliar. Boletin de Sanidad Vegetal Plagas, 29, La Mantia, G., Lo Verde, G., & Ferry, M. (2008). Le palme colpite dal Punteruolo rinate con la dendrochirurgia. L Informatore Agrario, 35, Lin, H., Chi, W., Lin, C., Tseng, Y., Chen, W., Kung, Y., et al. (2011). Fire ant-detecting canines: a complementary method in detecting Red Imported Fire Ants. Household and Structural Insects, 104, Longo, S., Anderson, P. J., Smith, T. R., Stanley, J. D., & Inserra, R. N. (2011). New palm hosts for the red palm weevil, Rhynchophorus ferrugineus, in Sicily. Palms, 55, Longo, S., & Tamburino, V. (2005). Gravi infestazioni di punteruolo rosso dellapalma. L Informatore Agrario, 50, Lorenzo,N.,Wan,T.I.,Harper,R.J.,Hsu,Y.L.,Chow,M., Rose, S., et al. (2003). Laboratory and field experiments used to identify Canis lupis var. familiaris active odor signature chemicals from drugs, explosives, and humans. Analytical and Bioanalytical Chemistry, 376, Mankin, R. W. (2011). Recent developments in the use of acoustic sensors and signal processing tools to target early infestations of red palm weevil in agricultural environments. Florida Entomologist, 94, Murphy, S. T., & Briscoe, B. R. (1999). The red palm weevil as an alien invasive: biology and the prospects for biological control as a component of IPM. Biocontrol, 20, Nakash, J., Osem, Y., & Kehat, M. (2000). A suggestion to use dogs for detecting red palm weevil (Rhynchophorus ferrugineus) infestation in date palms in Israel. Phytoparasitica, 28, Oxley, J. C., & Waggoner, L. P. (2009). Detection of explosives by dogs. In M. Marshall & J. C. Oxley (Eds.), Aspects of explosives detection (pp ). Dorsdorf, the Netherlands: Springer. Pfiester, M., Koehler, P. G., & Pereira, R. M. (2008). Ability of bed bug-detecting canine to locate live bed bugs and viable bed bug eggs. Journal of Economic Entomology, 101, Pickel, D., Manucy, G. P., Walker, D. B., Hall, S. B., & Walker, J. C. (2004). Evidence for canine olfactory detection of melanoma. Applied Animal Behaviour Science, 89, Pinhas, J., Soroker, V., Hetzroni, A., Mizrach, A., Teicher, M., & Goldberger, J. (2008). Automatic acoustic detection of the red palm weevil. Computers and Electronics in Agriculture, 63, Rahalkar, G. W., Awalkar, M. R., Rananavare, H. D., Tamhankar, A., & Shanthram, S. (1985). Rhynchophorus ferrugineus (pp ) In: P. Singh & R. F. Moore (Eds.), Handbook of insect rearing. New York, NY: Elsevier. Rouhi, A. M. (1997). Detecting illegal substances. Chemical and Engineering News, 75, Sacchetti, P., Camera, A., Granchietti, A., Rosi, M. C., & Marzialetti, P. (2005). Prima segnalazione in Italia del curculionide delle palme, Rhynchophorus ferrugineus. Notiziario del Centro Sperimentale per il Vivaismo di Pistoia, 144, 6 9. Shelby, R. A., Schrader, K. K., Tucker, A., Klesius, P. H., & Myers, L. J. (2004). Detection of catfish off-flavour compounds by trained dogs. Aquaculture Research, 35, Soroker, V., Nakache, Y., Landau, U., Mizrach, A., Hetzroni, A., & Gerling, D. (2004). Utilization of sounding methodology to detect infestation by Rhynchophorus ferrugineus on palm offshoots. Phytoparasitica, 32, 6 8. StatSoft Italia (1997). STATISTICA per Windows [Manuale programma per computer]. StatSoft Italia S.r.l., via Parenzo, Vigonza (Padova). U.S. Customs Service. (1979). U.S. Customs narcotics detector dog training. U.S. Customs Service (USCS), Washington, DC. USDA APHIS-PPQ (2010) First U.S. Detection of Red Palm Weevil, Rhynchophorus ferrugineus, in California. Letter to State Plant Regulatory Officials. Wallner, W. E., & Ellis, T. L. (1976). Olfactory detection of gypsy moth pheromone and eggs masses by domestic canines. Environmental Entomology, 5, Wattanapongsiri, A. (1966). A revision of the genera Rhynchophorus and Dynamis (Coleoptera: Curculionidae). Bangkok, Thailand: Department of Agriculture Science Bulletin. 1. Welch, J. B. (1990). A detector dog for screwworms (Diptera: Calliphoridae). Journal of Econonimc Entomology, 83,