Introduction
The sweetpotato whitefly Bemisia tabaci (Gennadius) (Homoptera: Aleyrodidae) is a cosmopolitan pest of field and horticultural crops (Oliveira et al., Reference Oliveira, Henneberry and Anderson2001) causing direct yield loss and economic damage in many economically important crops (Gerling et al., Reference Gerling, Motro and Horowitz1980; Osborne & Landa, Reference Osborne and Landa1992; Landa et al., Reference Landa, Osborne, Lopez and Eyal1994; Nomikou et al., Reference Nomikou, Janssen, Schraag and Sabelis2001). The life-cycle of Bemisia tabaci consists of three nymphal instars and a transitional form before adult emergence, lasting about 22 days in total depending on ambient temperatures (Salas & Mendoza, Reference Salas and Mendoza1995). The relatively immobile nymphal instars offer a possible opportunity for biological control agents to have an impact upon the population growth of this species (Gerling, Reference Gerling1990; Gerling & Mayer, Reference Gerling and Mayer1996). The first interceptions of this pest on plant material within the UK were in 1987, where it was subject to statutory controls aimed at eradication (Cheek & MacDonald, Reference Cheek and MacDonald1994; Cheek et al., Reference Cheek, Head and Malumphy1994). Within the UK, B. tabaci remains a notifiable pest subject to a policy of eradication if found on propagators' premises, plants moving in trade, and containment if outbreaks occur at nurseries (Cuthbertson, Reference Cuthbertson2005). However, due to populations developing high degrees of resistance to chemical pesticides (Prabhaker et al., Reference Prabhaker, Coudriet and Meyerdirk1985; Sharaf, Reference Sharaf1986; Cahill et al., Reference Cahill, Byrne, Denholm, Devonshire and Gorman1994, Reference Cahill, Gorman, Day, Denholm, Elbert and Nauen1996; Ahmad et al., Reference Ahmad, Arif, Ahmad and Denholm2002) control of outbreaks on protected crops has proved difficult and therefore other means of control are urgently required.
Although insecticides remain the primary tactic for managing B. tabaci (Palumbo et al., Reference Palumbo, Horowitz and Prabhaker2001), considerable research has focused on the development of alternative control methods (Gerling & Meyer, Reference Gerling and Mayer1996; Naranjo & Ellsworth, Reference Naranjo and Ellsworth2001; Cuthbertson et al., Reference Cuthbertson, Head, Walters and Gregory2003a, Reference Cuthbertson, Head, Walters and Murrayb; Naranjo et al., Reference Naranjo, Ellsworth and Hagler2004; Cuthbertson & Walters, Reference Cuthbertson and Walters2005a; Cuthbertson et al., Reference Cuthbertson, Walters and Northing2005a, Reference Cuthbertson, Walters and Deppeb). Entomopathogenic nematodes from the families Steinernematidae and Heterorhabditidae can penetrate and cause the death of many economically important pests (Poinar, Reference Poinar, Gaugler and Kaya1990; Stuart et al., Reference Stuart, Barbercheck, Grewal, Taylor and Hoy2006). Infective juveniles (IJ) of entomopathogenic nematodes are more mobile than the respective third-stage juveniles of the normal developmental cycle and therefore possibly more robust at host location and invasion (Ishibashi et al., Reference Ishibashi, Tabata and Kondo1987). As a result, they have the potential to form effective biological alternatives to chemical pesticides (Cross et al., Reference Cross, Solomon, Chandler, Jarrett, Richardson, Winstanley, Bathon, Huber, Keller, Langenbruch and Zimmerman1999). Foliar applications of Steinernema feltiae (Filipjev) (Nematoda: Steinernematidae) have successfully been used to control the quarantine leafminer species Liriomyza huidobrensis (Blanchard) (Diptera: Agromyzidae) on lettuce in commercial glasshouses (Williams & Walters, Reference Williams and Walters2000). In the laboratory, S. feltiae infective juveniles have been shown to have a high efficacy against second instar B. tabaci (Cuthbertson et al., Reference Cuthbertson, Head, Walters and Gregory2003a; Cuthbertson & Walters, Reference Cuthbertson and Walters2005b) and also that they can be successfully integrated with insecticides commonly used in the control of B. tabaci (Cuthbertson et al., Reference Cuthbertson, Head, Walters and Murray2003b).
The development of control strategies for non-indigenous insects is limited by UK legislation which precludes the intentional release of quarantine pests into ordinary experimental glasshouses (Williams & Walters, Reference Williams and Walters2000). In the current study, the efficacy of S. feltiae against second instar B. tabaci on five commercially important host plants within the laboratory environment was investigated. Information from the laboratory investigations was then used to select two host plants on which efficacy tests were conducted under glasshouse conditions using designated quarantine glasshouse cubicles. For commercial purposes, the efficacy of S. feltiae applied at different application rates was also investigated.
Materials and methods
Products and insect cultures
Bemisia tabaci was cultured under quarantine conditions in perspex cages (60 cm×60 cm×80 cm) on poinsettia (Euphorbia pulcherrima Wild. ex Klotzsch cv. Lilo Pink (Euphorbiaceae)) plants as hosts at 23±1°C with a 16:8 h light:dark (L:D) regime and an artificial dawn and dusk (Cuthbertson et al., Reference Cuthbertson, Head, Walters and Gregory2003a). A fresh batch of S. feltiae obtained as Nemasys® (Becker Underwood (UK) Inc.) was used in all experiments.
The efficacy of S. feltiae against B. tabaci infesting different host plants
Five species of commercial host plants were chosen for experimental purposes; tomato (Lycopersicon esculentum Mill. cv. Moneymaker (Solanaceae)); verbena (Verbena hybrida L. cv. Quartz Scarlet (Verbenaceae)); cucumber (Cucumis sativus L. cv. Telegraph Improved (Cucurbitaceae)); poinsettia (Euphorbia pulcherrima Wild. ex Klotzsch cv. Lilo Pink (Euphorbiaceae)); and chrysanthemum (Dendrathemum grandiflora L. cv. White Fresco (Asteraceae)). Plants were infested following the protocol of Cuthbertson et al. (Reference Cuthbertson, Head, Walters and Gregory2003a, Reference Cuthbertson, Head, Walters and Murrayb). Four clip cages, modelled on those described by MacGillivray & Anderson (Reference MacGillivray and Anderson1957), were positioned randomly on individual leaves (one cage per leaf) of each of 40 plants (eight of each plant species, four treated and four control for each species). Two male and five female B. tabaci were added to each cage and incubated for 48 h at 25±1°C, 65% relative humidity (r.h.) and 16:8 h L:D to allow egg-laying, after which adults were removed and infested leaves labelled. The plants were further incubated for another 12 days to allow eggs to hatch and larvae to reach the second larval instar (Butler et al., Reference Butler, Henneberry and Clayton1983; Bethke et al., Reference Bethke, Paine and Nuessly1991; Wang & Tsai, Reference Wang and Tsai1996), the stage most susceptible to nematode infection (Cuthbertson et al., Reference Cuthbertson, Head, Walters and Gregory2003a; Cuthbertson & Walters, Reference Cuthbertson and Walters2005b).
When a majority of second instar larvae were present the plants were sprayed with either one of two treatments: (i) water control+0.02% Agral solution (Syngenta Crop Protection Ltd., Cambridge, UK; active ingredient: alkyl phenol ethylene oxide); or (ii) a suspension of 10,000 S. feltiae IJ ml−1, with 0.02% Agral (Cuthbertson et al., Reference Cuthbertson, Head, Walters and Gregory2003a). Treatments resulted in approximately 160 S. feltiae IJ cm−2 of leaf area. Plants were sprayed to run-off using a Hozelock® Polyspray 2 hand held sprayer with a cone nozzle (Hozelock Ltd, Aylesbury, UK), returned to the environmental control cabinet while still wet and maintained at 20°C, 85% r.h., 12:12 L:D.
The efficacy of S. feltiae against B. tabaci under glasshouse conditions
Following the controlled environment trial, tomato and verbena (two important UK horticultural commodities) host plants were chosen for testing the efficacy of S. feltiae against B. tabaci under glasshouse conditions. Sixteen plants of each species (eight treated and eight control) were infested and incubated as outlined previously. After egg-laying had occurred and the adults had been removed, the plants were transferred in sealed cooler boxes to the designated glasshouse cubicle. The plants were arranged randomly throughout the cubicle. The conditions of the cubicle were set for 25°C for a further 12 days to allow the second instar larvae to develop. One hour before spraying occurred, the cubicle and plants were soaked with water in order to raise the humidity (>85%) and the temperature was lowered to 20°C (Cuthbertson & Walters, Reference Cuthbertson and Walters2005a). The plants were then sprayed as described previously during late evening on a dull, overcast day in mid-February. The plants remained within the cubicle for a further 3 days before being taken back to the quarantine laboratory in sealed cooler boxes for analysis of treatment effect.
The efficacy of S. feltiae against B. tabaci at various application rates
Twenty one of both tomato and verbena plants were each infested with five individual clip cages containing B. tabaci adults as outlined above and divided into seven groups. The larvae were again reared through to the second instar stage for experimental purposes. Suspensions of S. feltiae and 0.02% Agral solution were applied to equal numbers of plants at the following rates: 10,000; 5000; 2500; 1000; 1000 IJ ml−1 followed by a another application of 1000 IJ ml−1 24 h later. An equal number of control plants were sprayed with either water or a 0.02% Agral control spray. Plants were incubated under the same conditions as outlined above and larval mortality was assessed after 72 h.
Assessment
The numbers of live and dead individuals (determined by the presence of nematodes in the larvae) per infested leaf within the controlled environment and the glasshouse trials were recorded 72 h after the treatment had been applied (nematodes or water control). Data underwent analysis of variance and Tukey's honest significant difference (HSD) comparison test.
Results
Efficacy under laboratory conditions
The application of S. feltiae against B. tabaci on all five host plants resulted in significant mortality of the pest compared to the water control (F1,198=10,524; P<0.001) (fig. 1). Almost complete mortality of B. tabaci was obtained on tomato foliage (99.2%), while >90% mortality was recorded for the remaining four host plants (fig. 1). There was no significant difference in mortality of B. tabaci larvae obtained on the different plant species (F4,45=1.3266; P=0.275).
Efficacy of Steinernema feltiae against second instar Bemisia tabaci on a range of host plants applied within a controlled environment cabinet, 20°C, 85% relative humidity. Mortality recorded after 72 h. Bars are standard error (±SEM) of the mean. (
, Control;, S. feltiae.)
Efficacy under glasshouse conditions
Significantly higher mortality of B. tabaci on nematode treated plants than control plants was recorded in glasshouse experiments (F1,126=1243; P<0.001) (fig. 2). There was significantly higher larval mortality caused on tomato and verbena plants treated within the laboratory environment (fig. 1) compared to tomato and verbena plants treated within the glasshouse environment (fig. 2) (F1,102=30.4; P<0.001).
Efficacy of Steinernema feltiae against second instar Bemisia tabaci on tomato and verbena foliage under glasshouse conditions, 20°C, 85% relative humidity. Mortality recorded after 72 h. Bars are standard error (±SEM) of the mean. (, Control;, S. feltiae.)
Efficacy of S. feltiae at various application rates
On tomato plants
Using Tukey's HSD comparison test, there was no significant difference in mortality of B. tabaci larvae obtained by using half (5000 IJ ml−1) the normal recommended dose rate (10,000 IJ ml−1) (mean difference=−1.0; P=1.0) (fig. 3a). Also, there was no significant difference in mortality of B. tabaci larvae following application of either 2500 IJ ml−1 or 1000 IJ ml−1 followed 24 h later by another application of 1000 IJ ml−1 (1000×2) (mean difference=0.4; P=1.0).
Efficacy of Steinernema feltiae against second instar Bemisia tabaci on (a) tomato foliage and (b) verbena foliage applied at various application rates at 20°C, 85% relative humidity. Mortality recorded after 72 h. Bars are standard error (±SEM) of the mean.
On verbena plants
On verbena plants, full rate application (10,000 IJ ml−1) caused significantly higher B. tabaci larval mortality than the other dose rates (F6,43=44.3; P<0.001) (fig. 3b). Using Tukey's HSD comparison test, applying nematodes in sequential applications of 1000 IJ ml−1 caused as equally high mortality as did a single application of 2500 IJ ml−1 (mean difference=−0.4; P=1.0).
Discussion
Foliar application of the entomopathogenic nematode S. feltiae has been successfully used for the control of another UK quarantine pest, L. huidobrensis, in glasshouses (Williams & Walters, Reference Williams and Walters1994, Reference Williams and Walters2000). Here, they can offer high potential as a safe alternative to chemical pesticides for the treatment of invertebrate pests within glasshouse situations (Cross et al., Reference Cross, Solomon, Chandler, Jarrett, Richardson, Winstanley, Bathon, Huber, Keller, Langenbruch and Zimmerman1999). Their success in this situation has been the main driving force in researching their potential for use in integrated pest management of other quarantine species.
Careful targeting of the nematode application for effective pest control has been shown to be critical, as different instars often vary in their susceptibility to the control agent. In addition, the ambient temperature and humidity have been shown to be important factors determining nematode efficacy (Williams & Walters, Reference Williams and Walters1994; Candy, Reference Candy2003; Cuthbertson et al., Reference Cuthbertson, Head, Walters and Gregory2003a; Cuthbertson & Walters, Reference Cuthbertson and Walters2005b). Steinernema feltiae move through a thin film of water on the leaf surface to reach their insect hosts, before entering the body through natural openings. Maintenance of a water film relies in part on both temperature and relative humidity, which in glasshouses can be controlled to optimize efficacy of nematode treatments. However, commercial constraints result in the period in which environmental control can be maintained being limited. Therefore, for optimal use of S. feltiae, favourable conditions for nematode survival and efficacy must be maintained for up to 6–8 h following application to plant foliage (Williams & MacDonald, Reference Williams and MacDonald1995; Cuthbertson & Walters, Reference Cuthbertson and Walters2005b).
The present study provides evidence that S. feltiae under controlled laboratory conditions is able to exert an excellent degree of control over second instar B. tabaci on five commercial plant hosts. Within the glasshouse situation a lower level of B. tabaci mortality was obtained. However, the compatibility of S. feltiae and chemical insecticides (Cuthbertson et al., Reference Cuthbertson, Head, Walters and Murray2003b) gives potential for the incorporation of this nematode into integrated pest management programmes for the control of B. tabaci within UK glasshouses. Promising results of B. tabaci larval mortality in response to reducing the application rates of S. feltiae, especially on tomato foliage, offers the potential for more economically viable use of the agent, and encourages the incorporation of this bio-control agent into integrated pest management programmes. It would appear from the current findings that there is the potential for the recommended application rate to be halved without reduction in efficacy on tomato foliage. Similar rates of mortality have been achieved against leaf miners (Williams & MacDonald, Reference Williams and MacDonald1995) and, with modification of application rates and post-application conditions, economically viable control of leaf miners was obtained (Williams & Walters, Reference Williams and Walters2000). Further work is now required to optimize the application and post-application conditions required for the working of S. feltiae for B. tabaci control.
Acknowledgements
The authors thank Becker Underwood (UK) Inc. for supplying the nematodes used in this study and Mr Richard Natt (Central Science Laboratory Horticultural Manager) and his team for the maintenance of the quarantine glasshouse facilities and supply of plant material. The work was funded by Plant Health Division, Department for Environment, Food and Rural Affairs (Defra). Bemisia tabaci are held under quarantine licence number: PHL 251/4072 (01/2002).
