General enquiries on this form should be made to:Defra, Science Directorate, Management Support and Finance Team,Telephone No. 020 7238 1612E-mail: research.competitions@defra.gsi.gov.uk

SID 5 Research Project Final Report

SID 5 (2/05) Page 1 of 58

NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.

A SID 5A form must be completed where a project is paid on a monthly basis or against quarterly invoices. No SID 5A is required where payments are made at milestone points. When a SID 5A is required, no SID 5 form will be accepted without the accompanying SID 5A.

This form is in Word format and the boxes may be expanded or reduced, as appropriate.

ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code HH3102 TPC

2. Project title

Exploiting knowledge of western flower thrips behaviour to improve efficacy of biological control measures

3. Contractororganisation(s)

ADAS BoxworthBoxworthCambridgeCB3 8NA                         

54. Total Defra project costs £ 499,724

5. Project: start date................ 01 December 2002

end date................. 28 February 2006

SID 5 (2/05) Page 2 of 58

6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...................................................................................YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They

should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domain

Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the

intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.In accordance with Defra policy objectives, this project aimed to reduce the use of chemical pesticides against western flower thrips (WFT) on protected horticultural crops, by exploiting new knowledge of WFT behaviour to develop more effective strategies for using biopesticides. The biopesticides used in the research were the insect-pathogenic nematodes, Steinernema feltiae and the insect-pathogenic fungus, Beauveria bassiana. S. feltiae (Nemasys F) is already available in the UK for control of thrips but B. bassiana is not yet approved in the UK. The representative host crops used in the research were pot chrysanthemum, cv. Swingtime and cucumber.

Objective 1: To determine the spatial and temporal distribution and activity patterns of WFT on two contrasting crops, in order to identify the most effective hours and placement for applications of microbiological control agents to reach the target life stages. The spatial distribution of WFT life stages was determined on heavily infested glasshouse crops of pot chrysanthemum and cucumber at sequential crop growth stages. On both crops, more WFT were found on older, lower leaves than on younger, upper leaves and more larvae were found on the lower than the upper leaf surfaces. On chrysanthemum, once buds and flowers had developed, there were more WFT in buds and flowers than on leaves. For effective control of WFT life stages on the plants, applications of any effective biopesticides would need to achieve good coverage of both leaf surfaces during the vegetative stage and of buds and flowers on chrysanthemums. On both crops, 95-98% of the WFT larvae dropped to the ground to pupate and this behaviour offers the opportunity to use biopesticides against the ground-dwelling stages (late second stage larvae, prepupae and pupae), either instead of, or in addition to applying them to the plants. On cucumber, significant numbers of prepupae or pupae were only found on the undersides of older lower leaves, which were partly lying on the polythene floor. Removal of lower cucumber leaves touching the floor may be a useful cultural control strategy. During the same glasshouse experiments, the location of WFT adults was determined on both crops and in the air over 72-hour periods. A suction trap was used to monitor flight activity and a WFT flight model was produced. WFT did not fly by night or when the air temperature was less than 20°C or above 40°C. Peak flight activity was typically in the late afternoon when the temperature was around 30°C. Time-lapse video was used in both crops over several days to record diel (24-hr period) activity patterns. Diel movement of WFT adults on the plants seemed to be related to avoiding extremes of microclimate, with more present on lower cucumber leaves in the afternoon than the morning and more on lower leaf undersides by day than by night. It is unlikely that biopesticides applied to plants would be affected by diel patterns of WFT activity, as WFT were generally as active on plants by day as by night and the proportion of adults in flight by day was low. However, on both crops, most larvae dropped to the ground to find a pupation site in the early evening and novel control measures might be designed to exploit this behaviour.

SID 5 (2/05) Page 3 of 58

Objective 2: To quantify the efficacy and persistence of Steinernema feltiae and Beauveria bassiana against WFT life stages on selected host plant material and growing media A laboratory bioassay was developed to test the effect of controlled doses of either S. feltiae or B. bassiana on mortality and speed of kill of WFT adult females and second stage larvae (L2s) when applied to chrysanthemum leaf discs and of WFT ground-dwelling life stages when applied to growing substrates. The S. feltiae product used was Nemasys F and the B. bassiana products tested were Naturalis and Botanigard, and isolates cultured from these products, code numbers 432.99 and 433.99 respectively. Nemasys F typically killed a mean of approximately 50% WFT L2s at day 5, but this was significantly different to the mean 30% kill given by water controls in only one set of bioassays. Only high rates of Nemasys F (x32 recommended rate) killed significant numbers of WFT adult females, (a mean of 80% kill at day 5, compared with a mean of 25% kill in water controls). Both WFT adults and larvae showed a marked avoidance response to S. feltiae and this defence behaviour could reduce nematode infection. Recent research in Canada showed that Nemasys F killed WFT adults only when the thrips were immobilised first. Nemasys F at recommended rate killed a mean of 54% WFT pupae in compost bioassays but this was not significantly more than the 26% killed in controls and no dose response was given with higher rates of nematodes. B. bassiana (Botanigard) killed significant numbers of WFT L2s but mean % kill was low (35% at day 7) and Naturalis had no significant effect. WFT adults were more susceptible than larvae to B. bassiana, with approximately 60% kill at day 7 by both Botanigard and isolate 432.99. B. bassiana did not kill WFT ground-dwelling stages in compost. Glasshouse experiments were done to determine the persistence of S. feltiae and B. bassiana on chrysanthemum and cucumber leaves and on chrysanthemum buds and flowers. Most S. feltiae were inactive within three hours of Nemasys F application, giving a very short ‘window’ for infection of WFT before plant surfaces dry. B. bassiana conidia persisted for at least 350 hrs on cucumber leaves and for 70-120 hrs on chrysanthemum leaves and buds. Botanigard persisted longer than Naturalis. A glasshouse experiment was done to determine the persistence of S. feltiae in compost over an 8-week period following application to pots of chrysanthemum cuttings. Numbers of viable S. feltiae in the compost declined by 67% over the first five weeks but then stabilised, so that a proportion were still active eight weeks after application, when the plants were ready for marketing. This result may allow nematode application rates and timings to be planned for persistent control of WFT in the compost or substrate. A glasshouse experiment on cucumber showed that weekly Nemasys F applications to either the plants or the growing substrate reduced WFT numbers, but numbers were too low to compare efficacy of the two nematode placements. Two consecutive glasshouse experiments on pot chrysanthemums in replicate thrips-proof cages demonstrated the comparative efficacy of S. feltiae when applied to the plants only, the compost only or to both plants and compost. In both experiments, WFT numbers were significantly reduced where Nemasys F had been applied to the compost only or to both plants and compost, but not where applied to plants only. This result supports those given in the leaf bioassays and indicates that nematode control of WFT in the compost or substrate has a more important role in reducing WFT populations than control of WFT on the plants. In Experiment 2, Nemasys F applied to both plants and compost gave greater reductions in WFT numbers than when applied to compost only. It is possible that in addition to killing WFT ground-dwelling stages, Nemasys F applications may cause some WFT adults and larvae to drop from the plants to escape infection, but on reaching the compost they may become infected. Nemasys F is currently recommended as a spray to plants to just before run-off. Further work would be needed to determine whether adapted application methods could optimise WFT control.

Objective 3: To determine whether the efficacy of S. feltiae and B. bassiana can be improved by manipulating WFT behaviour with semiochemicals or by adapting use of glasshouse lighting Laboratory bioassays were done to test the effects of semiochemicals (behaviour-modifying plant extracts) on the activity behaviour of WFT adults and L2s on cucumber and chrysanthemum leaves. The antifeedants tested were extracts from Tasmannia stipitata and T. lanceolata, plants which contain polygodial (a known insect antifeedant) and synthetic polygodial as a comparison. The volatile semiochemical tested was (E)-β-farnesene (EBF), shown in previous Defra-funded research to attract WFT. Both Tasmannia spp. extracts increased WFT movement in bioassays and the polygodial content was shown to be largely responsible. EBF did not affect WFT behaviour at the concentration tested. Laboratory tests showed that the T. lanceolata extract was incompatible with B. bassiana; it inhibited Botanigard germination, prevented the formation of fungal colonies and reduced infection of WFT adults. Laboratory tests showed that mixing either Tasmannia spp. extract with S. feltiae did not reduce nematode viability. However, in glasshouse experiments on pot chrysanthemums, spraying a mix of the T. lanceolata antifeedant and Nemasys F did not improve control of WFT, although there was a short-term antifeedant effect. The original hypothesis that increasing WFT movement to enhance contact and thus infection with S. feltiae was shown to be invalid as it is now thought that stationary thrips (e.g. pupae) are more susceptible to nematode infection than moving ones (adults and larvae). Initial laboratory work had indicated that a short light burst during the night could induce prolonged walking activity in WFT adult females. However, a glasshouse experiment on cucumber plants showed that a 30 minute light burst during the night did not increase WFT adult walking activity because, contrary

SID 5 (2/05) Page 4 of 58

to expectation, walking activity continued at night in glasshouses.

Objective 4: To quantify the effects of S. feltiae, B. bassiana and semiochemicals on other selected biological control agents used in Integrated Pest Management ( IPM) on protected crops Laboratory bioassays were done to test the effect of S. feltiae (Nemasys F) and B. bassiana (Botanigard) on the predatory midge larvae, Aphidoletes aphidimyza, which are widely used for aphid control within IPM programmes. Botanigard did not reduce survival of larvae at day 7 and should be compatible with A. aphidimyza. Nemasys F killed 50% of the larvae at day 5 and nematodes were found inside dissected bodies but this mortality was not significantly different from the 27% killed in water controls. Further work would be needed to test the effect of weekly applications of Nemasys F on A. aphidimyza larvae on plants and on pupal cocoons in growing media. Laboratory bioassays were done to test the effect of the T. lanceolata extract and of synthetic polygodial on the behaviour of the aphid parasitoid, Aphidius ervi and on the settlement of winged aphids, Aphis gossypii on cucumber plants. Neither treatment had any significant effect on the behaviour of A. ervi but both treatments reduced aphid settlement on cucumber. The antifeedants have potential to reduce aphid colonisation of plants with unlikely harmful effects on aphid parasitoids.

Knowledge transfer activities included 15 grower presentations, five articles in the horticultural press/newsletters, 12 scientific conference/meeting presentations, two scientific papers and regular communications with growers and the industry during ADAS consultancy and technology transfer work. The project has identified key new information on WFT behaviour, e.g. synchronised larval dropping from host plants and pupation in growing media and substrates. This information, together with new information gained on the persistence and efficacy of S. feltiae and B. bassiana against WFT life stages on plants and in substrates, offers the opportunity to develop novel biopesticide application methods for optimum control of WFT and of other pests, as substitutes for chemical pesticides within IPM programmes. An expression of interest has been submitted to Defra PSD and the potential for Horticulture LINK funding is being discussed.

Project Report to Defra8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with

details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).

tronObjective 1. To determine the spatial and temporal distribution and activity patterns of WFT on two contrasting crops, in order to identify the most effective hours and placement for applications of microbiological control agents to reach the target life stages (ADAS and Keele)

1 (a) Spatial distribution of WFT life stages on pot chrysanthemum and cucumber (ADAS)

MethodsSpatial distribution of plant-dwelling WFT life stages, pot chrysanthemumsDuring May and June 2003, pot chrysanthemums, cv. ‘Swingtime’ were grown in a research glasshouse at ADAS Boxworth. There were 90 pots of chrysanthemums (three benches covered with capillary matting, each bench with 30 pots). Glasshouse environmental conditions and plant husbandry was as in commercial practice. Once the pot chrysanthemums were rooted, the plants were grown in ‘short days’ (11hL:13hD) to initiate flowering, and at a minimum temperature of 18C, venting at 21C and with an automatic aluminium shade screen used at 30C / 40kl light. To ensure a heavy infestation of WFT, adults

SID 5 (2/05) Page 5 of 58

were released on five occasions (a total of 40 per plant, in a ratio of 12 females:1 male on each release date). Numbers and location of WFT females, males, first and second instar larvae, prepupae and pupae were recorded on 18 plants randomly selected on each of six dates from weekly samples of upper, middle and lower leaves and from samples of the different bud and flower stages. Leaf assessments were done in-situ using a head magnifier and numbers of WFT on both upper and lower surfaces of each leaf were recorded. Buds and flowers were sampled into alcohol and dissected under a binocular microscope to ensure all WFT were recorded. All sampling and assessments were done between 2 and 4pm.

Spatial distribution of plant-dwelling WFT life stages, cucumbersDuring May and June 2003, cucumbers, cv. ‘Aviance’ were grown in rockwool slabs on a white polythene floor in a research glasshouse at ADAS Boxworth. There were 18 plants (three rows of six plants). Glasshouse environmental conditions and plant husbandry were as in commercial practice. Minimum temperatures were 18C, venting at 24C. WFT adults were released on two occasions (a total of 50 per plant, 12 females:1 male on each release date). Numbers and location of all WFT life stages were recorded on the same six plants (two per row) on six dates from weekly samples of upper, middle and lower leaves and from an apical leaf just behind the growing point on the main stem and the first lateral stem. Assessments were done in-situ using a head magnifier between 2 and 4pm and numbers of WFT on both upper and lower leaf surfaces were recorded.

Proportions of WFT pupating on plants or in growing media / substrate, pot chrysanthemumsImmediately after the final assessment, when the plants were at the flowering stage, six plants were used to assess the proportions of WFT pupating on the plants, in the compost or in the capillary matting on which the pots were stood. Black polythene coated with insect-trapping glue was used under six plants and the numbers of WFT larvae dropping from each plant to pupate were recorded after a 24-hr period. The leaves and stems of each plant were then washed in alcohol to collect all WFT and the buds and flowers were placed into alcohol and dissected under a binocular microscope. The total numbers of larval and pupal stages on each plant were recorded. The proportion of WFT dropping from the plants to pupate in either the compost or the capillary matting substrate was calculated using the numbers of larvae dropping off the plant, the numbers of larvae and pupae remaining on the plant and the published mean development times of the larval and pupal stages at the glasshouse temperature range over the 24-hr period.

Proportions of WFT pupating on plants or in growing media / substrate, cucumbersImmediately after the final assessment, four plants were used to assess the proportions of WFT pupating on the plants or in/on the rockwool blocks, slabs or polythene substrate. White sticky polythene sheets were used under the plants over a 24-hr period to trap larvae falling off the plants and then the four plants were destructively sampled and all leaves, stems, flowers and fruit were washed in alcohol, which was then sieved to retain all thrips, which were counted under a binocular microscope. Using the same method as used for the pot chrysanthemums, the proportions of WFT pupating on/off the plants were calculated.

Results and Discussion

Spatial distribution of plant-dwelling WFT life stages, pot chrysanthemumsVery few WFT adults (less than 0.5 per leaf sample) were found on the leaves on all assessment dates, so no pattern of spatial distribution of adults could be determined. There were more WFT larvae on leaves from the bottom of the canopy than on those from the middle or upper canopy, peaking on assessment 4 (means of 9.2 per bottom leaf leaf, 5.3 per middle leaf and 1.1 per top leaf), when the plants were just before the bud-break stage (Fig.1). Numbers of larvae on the upper and lower surfaces of leaves were very similar, although there were consistently more on the lower than on the upper surfaces (Fig.2). At assessments 5 and 6, when the plants had developed opening buds and flowers, numbers of adults and larvae increased in the bud and flower stages and numbers of larvae correspondingly dropped on the leaf samples to less than 1 per leaf at all canopy levels Figs 1 and 3). It is known that adult WFT are attracted to chrysanthemum opening buds and flowers to feed and lay eggs, and the data indicates that there may also have been some movement of larvae from leaves to buds and flowers. The results show that if S. feltiae and B. bassiana are effective against WFT adults and larvae on pot chrysanthemum foliage, buds or flowers, it would be important to achieve good coverage of both sides of the leaves during the vegetative stage and of the buds and flowers from the bud stage onwards.

Spatial distribution of plant-dwelling WFT life stages, cucumbersVery few thrips were found on the small, young apical leaves. More WFT (totals of all life stages) were found on the older leaves at the bottom of the canopy than on leaves from the middle or top of the canopy, reaching means of 203 per bottom leaf, 104 per middle leaf and 59 per top leaf (Fig. 4). More WFT were found on the lower sides of the leaves than the upper sides (Fig. 5) and this tendency was more obvious for larvae and pupae than for adults. Very few prepupae or pupae were found on the leaves until the last

SID 5 (2/05) Page 6 of 58

three assessments, when up to 25 pupal stages per leaf were found, on the undersides of leaves at the bottom of the canopy which were partly lying on the polythene floor. These results indicate that if S. feltiae and B. bassiana are effective against WFT adults and larvae it would be important to achieve good coverage of both leaf surfaces, particularly at the bottom of the canopy. As a cultural strategy to reduce the numbers of WFT pupating on the bottom leaves, any leaves touching the floor should be removed.

Proportions of WFT pupating on plants or in growing media / substrate, pot chrysanthemums and cucumberOn pot chrysanthemum, cv. ‘Swingtime’, 95-96% of the WFT larvae dropped to the ground to pupate (Tables 1 and 2). Of the 4-5% which pupated on the plants most pupae were in the open flowers including inside the disc florets, but some were on stems, leaves and in opening buds. This result differed from those of Canadian workers on pot chrysanthemum cv. ‘Miramar’ which indicated that only 47% WFT larvae dropped to the ground to pupate (Broadbent et al, 2003). However, in the Canadian work, the experimental methods were different to those used in this project and 43% of their thrips were unaccounted for. The two sets of results were compared and discussed in a joint presentation with the Canadian workers at the International Congress of Entomology in Brisbane (Bennison et al, 2004). On cucumber, very similar results were given to those on pot chrysanthemum, with 98% WFT larvae dropping to the ground to pupate (Tables 1 and 2). The few WFT which pupated on the plants were found on leaves, stems and occasionally on fruit. This is new data for cucumber; prior to this research, it was assumed that most WFT pupated on cucumber plants (Jacobson, 1997). Our results are similar to those published on French bean, which showed that 98% WFT dropped from the plants to pupate in the soil (Berndt et al, 2004). The results indicate that there are opportunities for using integrated control methods, including either S. feltiae or B. bassiana against the ground-dwelling WFT stages, either as an alternative to, or in addition to applying control measures to the plants.

1 (b) Temporal and spatial distribution of WFT adults on crops and in the air (ADAS and Keele)The aim of this work was to identify where and when WFT were active over the diel period on pot chrysanthemum and cucumber.

MethodsDuring the same experiments as in 1 (a), the temporal and spatial distribution of WFT adults was determined by counting the numbers visible on the surface of pot chrysanthemum and cucumber plants. WFT adults on pot chrysanthemums were counted on upper and lower surfaces of leaves, at three leaf heights, on three development stages of buds/flowers sampled into alcohol and at eight times over the diel period. The assessments were repeated on three days and at two stages of the crop’s development in May and June 2003. Within the cucumber crop, WFT adults were counted at seven foliage heights, on both upper and lower leaf surfaces at seven times over the diel period and this was repeated on three days in June 2003. A Johnson-Taylor suction trap simultaneously sampled the aerial density of WFT adults on an hourly basis. In addition, the suction trap was used to gather further data in a pot chrysanthemum crop in the ADAS Boxworth glasshouses in June 2003 and in January 2004. Data were analysed with paired t-tests and two-way ANOVA.

Results and DiscussionWhen the pot chrysanthemum crop was at the bud-break (20-22 May 2003), insufficient adult WFT were observed on the leaves for useful analysis (total=38). At the open flower stage (11-14 June 2003), there was no significant difference between the number of WFT adults on the upper and lower surfaces of leaves (P=0.36) or at the three different heights of leaf within the crop (P=0.13) and so the counts for numbers of adult WFT on both surfaces and at the three heights were combined. As many adult WFT were present on the leaves by day as by night (P=0.70). Analysis of numbers of WFT in buds/flowers at three stages of development showed no significant difference in the numbers of adult male, adult female, combined adult WFT or larval WFT between night and day (P>0.1), except that more larvae were present in the buds at bud-break by day than by night. This result is on the margin of significance (P=0.05) and is out of line with the other data. However, over the three days of the experiment there was a significant decline in the number of larvae in the buds at bud-break (P=0.02), possibly because larvae were moving to the increasing number of open flowers on the plants, and this could have artificially distorted the difference in numbers between night and day.

In the cucumber crop, tests to validate the visual assessments showed that male and larval WFT were not adequately distinguished by eye over the diel period, but counts of adult females were shown to be accurate. This meant that night-time adult counts, when males and females were counted together, and day-time counts of males were not reliable and so were excluded from the analysis. In the cucumber crop, approximately seven times more adult female WFT were found on the lower leaves than on the middle or top leaves. Unlike the spatial distribution of larvae and pupae on cucumber, reported in 1 (a), there was no significant difference in the numbers of adults on lower and upper surfaces of lower canopy leaves (P=0.43) and so the data from these counts were combined. About 37% more adult female WFT were

SID 5 (2/05) Page 7 of 58

found on the lower leaves in the afternoon, when temperatures were higher, than in the morning (P<0.001).

Suction trap catches from both pot chrysanthemum and cucumber crops showed that significantly more adult WFT flew by day than by night (P=0.01). Virtually no WFT flew by night. Flight typically peaked in the afternoon (Fig. 6). Male and female flight patterns were similar.

1 (c) Laboratory walking bioassays (Keele)

MethodsInfrared recording equipment was used to assess the numbers of adult WFT visible and the proportion moving on flowers of pot chrysanthemum plants, cv. Swingtime kept in a constant temperature (CT) room (25±1°C, 16L:8D, n=8 days). A camera was focused on the flowers and WFT activity was recorded using a time-lapse VCR. The recording was reviewed for one minute at 15 min intervals and the maximum number of adults present during each minute in the field of view and the proportion moving were recorded. Data were analysed with paired t-tests.

Bioassays were designed to identify when thrips were walking over the diel period. EthoVision ver. 3.1 software (Noldus, Wageningen) and infrared video recording equipment, which is invisible to WFT, were used to track walking of adult male and female WFT within a plastic arena (diameter 15 mm, height 3 mm) at 25±1°C, 16L:8D. Optic-fibre cold-light sources provided illumination during the photoperiod (10.6 Wm-2) and low-level lighting in the scotophase for one experiment. WFT were provided with a diet of 10% sucrose solution that they could feed on through a membrane. Data were analysed with paired t-tests.

Results and DiscussionAlthough adult WFT on chrysanthemum flowers in the CT room walked as much by day as by night (P=0.99) (Fig. 7), significantly more adult WFT were present on the surface of the flowers by day than by night (P=0.04). However, the effect was small, with only about 24% more present by day than by night. Such a small difference may also have been present in the glasshouse (reported in 1 (d), but could well have been masked by the environmental variability, which was much greater than in a CT room. Walking activity in small arenas showed a distinctly bimodal curve (Fig. 8) with an increase in walking at the beginning and end of the light period and with little movement at night or in the middle of the day. Thrips were almost stationary at night within the small arena, but, in contrast, they walked at night in a glasshouse (Fig. 7). To ascertain if ambient light at night in the glasshouse, from the moon and street lighting, stimulated movement of WFT and caused the difference in diel patterns, a low-level light (0.0125 Wm-2) was applied to the small arena during the scotophase. It was hypothesised that the absence of any ambient light at night in the EthoVision arena suppressed walking. However, females showed much less walking activity during the night than the day (P<0.005), regardless of whether or not there was a low light present in the scotophase (Fig. 8), so the difference in light level at night did not explain the difference in nocturnal activity. Another possible explanation of the difference in nocturnal activity between the small arena and plants in the glasshouse is that plant volatiles stimulate walking activity. Males showed a similar diel pattern to females (Fig. 8), but the difference between day and night was not significant (P=0.21). This could be because males tended to be more active during the first night of the experiment, probably as a result of a more prolonged disturbance effect following the setting up of the experiment.

1 (d) Glasshouse walking activity (Keele)

MethodsUsing infrared recording equipment, the numbers present and proportion moving of WFT were assessed on pot chrysanthemum cv. Swingtime and cucumber plants cv. Tanja in glasshouses. The camera was focused on either one chrysanthemum flower head (n=6 days) or the lower surface of a cucumber leaf (n=4 days for larvae and n=8 days for adults) and recorded by time-lapse VCR. Assessment of the video was made in the same way as in 1 (c). Data were analysed with paired t-tests.

Results and DiscussionThe same proportion of WFT adults walked in the light as in the dark on flowering chrysanthemums (P=0.54) and on the underside of cucumber leaves (P=0.22) (Fig. 7). Larvae also moved as much by day as by night on the underside of cucumber leaves (P=0.99) (Fig. 7). There was no significant difference in the number of adults on the surface of chrysanthemum flowers by day compared with by night (P=0.11) or for larvae on the underside of cucumber leaves by day compared with by night (P=0.74). However about twice as many adults were present during the day than during the night on the underside of cucumber leaves (P<0.001). Unusually hot weather and high glasshouse temperatures (>40°C) occurred during these experiments and adults may have retreated to the shaded side of lower leaves by day to escape microclimatic extremes. This movement would reduce their exposure to any foliar sprays of either S. feltiae or B. bassiana. However, the results showed that they were still as active by day as by night.

SID 5 (2/05) Page 8 of 58

Larvae, however, tended to stay on the underside of leaves throughout the day and this result is consistent with the spatial distribution of larvae on cucumber leaves reported in 1 (a), where although more larvae were found on leaf undersides than upper sides, numbers on upper sides increased as thrips densities increased. In the 1 (a) experiment the WFT density was higher than on the plants used in 1 (d). This result is consistent with those of Gaum et al (1994) who reported that WFT larvae are usually on cucumber undersides, but in high densities they are also found on the upper sides.

Additional experiments with water traps under the plants showed that larvae dropped to the ground to pupate predominantly over a few hours in the evening in both pot chrysanthemum (P<0.001) and cucumber crops (P<0.001) (data not shown). This suggests a time of day when larvae are exposed on the ground and novel measures could be designed to exploit this phase.

1 (e) Flight activity model (Keele)

MethodsHourly flight activity was recorded on 20-22 May 2003, 11-14 June 2003 and 11-14 January 2004 in a pot chrysanthemum crop and 17-19 June 2003 in a cucumber crop. The recording on both crops during May and June 2004 was done during the same experiment as in 1 (a) and (b). Light, temperature and relative humidity data were collected every 5 min with a data logger. This information was used to develop a flight activity model to identify the conditions when adult WFT fly during the diel period. Data were analysed with regression analysis and ANOVA

Results and DiscussionTen days of suction trap data was used to produce a model. A graphical method was used to estimate a threshold light level of 12 Wm-2 below which male and female WFT do not fly, but above which they fly if the correct conditions are met. Analysis from data with a wide range of temperatures (17-19 June 2003 in a cucumber crop) showed that there was a significant quadratic relationship with temperature (P=0.015), which agreed with other evidence in the literature (Pearsall, 2002; O’Leary, 2005). Flight increased between 20°C and 30°C and decreased from 30°C to 40°C. A simple mathematical model was produced with no flight at light <12 Wm-2, temperature <20°C or temperature ≥40°C , and with flight increasing linearly from zero at 20°C to a maximum at 30°C and then decreasing linearly to zero at 40°C. This generated a flight index from 0-1, with no flight at 0 and maximum flight at 1. The model applied to both males and females, so it could be applied to total adult flight activity. An example of the good fit of the model is illustrated in Fig. 9. Flight activity would therefore be predicted to be highest by day when temperatures are approaching 30°C.

Conclusions from work in Objective 1 On both pot chrysanthemum and cucumber, more WFT were found on older, lower leaves than on

younger, upper leaves and more larvae were found on the lower than the upper leaf surfaces. On pot chrysanthemum, once the plants had developed buds and flowers, the proportion of WFT

in buds and flowers increased and the proportion on leaves correspondingly decreased. For effective control of WFT adults and larvae on the plants, applications of any effective

microbiological control agents would need to achieve good coverage of both sides of the leaves during the vegetative stage and of buds and flowers on chrysanthemums from the bud stage onwards.

On both pot chrysanthemum and cucumber, 95-98% of WFT larvae dropped to the ground to pupate, thus there is the opportunity to use any effective microbiological control agents against WFT ground-dwelling stages (late second stage larvae, prepupae and pupae).

On cucumber, very few pupal stages were found on leaves until the final three weeks of the experiment, when pupae were found on the undersides of older lower leaves which were partly lying on the floor. Removal of such leaves may be a useful cultural control strategy on cucumber.

WFT adults and larvae showed similar high levels of walking activity on the surface of pot chrysanthemum and cucumber crops both by day and night.

Diel movement of WFT adults seemed to be related to avoiding extremes of microclimate, with more adults present on lower cucumber leaves in the afternoon than the morning and more adults on the lower sides of lower leaves by day than by night.

Adult male and female WFT do not fly by night (light < 12 Wm-2) or when the air temperature is too cold (<20°C) or too hot (40°C).

Adult male and female WFT exhibit peak flight activity by day, typically in the late afternoon when the temperature is in the region of 30°C.

It is unlikely that biological control measures would be affected by diel patterns of WFT activity as WFT were generally as active on the plants by day as by night and the proportion of adults in flight by day was low.

WFT larvae drop to the ground to pupate in the early evening in chrysanthemum and cucumber crops, suggesting a time of day when larvae are exposed on the ground. Novel control measures

SID 5 (2/05) Page 9 of 58

could be designed to exploit this behaviour.

Objective 2. To quantify the efficacy and persistence of Steinernema feltiae and Beauveria bassiana against all WFT life stages on selected host plant material and growing media (ADAS and WHRI)

2(a) Develop laboratory bioassays to measure WFT life stage susceptibility to S. feltiae (ADAS) and B. bassiana (WHRI)

The aim was to develop a reliable bioassay to test the effect of a controlled dose of either nematodes or fungal conidia on mortality and speed of kill of WFT adult females and second instar larvae, when applied to a leaf surface or growing substrate in a reproducible manner, using a bioassay arena and controlled conditions that are optimal for each of the biopesticides and that minimise control thrips mortality.

Methods and ResultsBioassay for leaf-dwelling WFT stagesMethods using chrysanthemum or bean leaves or leaf discs in Petri dishes or Tashiro cages (Tashiro, 1967) were investigated using replicated experiments but were discarded due to problems with thrips escape, high control mortality, leaf deterioration or failure to allow daily non-destructive assessment of thrips mortality. A novel bioassay was developed using synchronised aged WFT adult females or second instar larvae on chrysanthemum (cv. ‘Swingtime’) leaf discs maintained on damp filter paper in an adapted 2 cm diameter Petri dish with sub-irrigation and lid ventilation. WFT were confined to the dishes using a 1.5 cm diameter plastic ring secured by perforated Clingfilm and Parafilm. The leaf discs were sprayed with suspensions of either S. feltiae or conidia of B. bassiana using a Potter tower prior to adding the thrips. The dishes were maintained at 21C and at a 16 h photoperiod. This method successfully confined the small, very active thrips to the experimental arenas, maintained leaf vigour for at least a week, avoided thrips drowning in condensation and allowed daily non-destructive assessment of thrips mortality.

Bioassay for ground-dwelling WFT stagesA bioassay was developed using 6cm x 3cm ventilated plastic pots containing 15-20 g of a peat-based compost (Sinclair pot chrysanthemum compost) to a depth of 1.5-2 cm. WFT late second instar larvae (four days old) were added to a chrysanthemum leaf disc (1.5 cm diameter) and allowed to pupate in the compost. After three days the leaf disc was removed together with any remaining larvae, the numbers of which were recorded. Survival was measured by numbers of adults emerging and getting trapped on insect barrier glue spread over the inside of the pot lid. Survival of the ground-dwelling stages was best when the moisture content of the compost was >75%, determined by a drying and weighing method. For bioassays with B. bassiana the conidial suspensions were applied to the compost just before the leaf disc and thrips were added, but for bioassays with S. feltiae, nematodes were applied to the compost just after the leaf disc was removed, as if applied earlier, thrips larvae were observed to walk away from treated compost and get trapped in the glue on the lid, as an escape response to the nematodes. Both conidia and nematodes were applied with a Potter tower and the pots were maintained at 21C and at a 16 h photoperiod, inside a sealed plastic box containing a Petri dish of watery agar to maintain humidity.

2(b) Quantify WFT life stage susceptibility to S. feltiae (ADAS) and B. bassiana (WHRI)

MethodsLeaf-dwelling WFT stages and S. feltiaeUsing the method developed in 2(a), replicated bioassays were done to test the susceptibility of WFT second stage larvae (L2s) and adult females to controlled doses of S. feltiae on chrysanthemum leaves over a 5-day period. The S. feltiae product used was ‘Nemasys F’ which is formulated in a gel carrier to avoid leaving a visible residue on foliage. In the initial set of bioassays, water was compared with water plus Agral at a 0.05% concentration and with S. feltiae at the recommended rate (250,000 per 1000m2 in 100 l water) plus Agral at a 0.05% concentration. In subsequent bioassays, water plus Agral was used as the control and Agral was also used in all S.feltiae treatments, as this is currently the most commonly used wetter applied with foliar sprays of the nematodes in commercial practice. S. feltiae (‘Nemasys F’) was applied at recommended (x1), double (x2) and eight times (x8) the recommended rate (recommended rate being equivalent to 25 per cm2). Five WFT L2s or five adult females were used per dish and there were five replicate dishes per treatment. Numbers of live and dead thrips were assessed after one, two and five days. Data were analysed using Chi-squared tests for the initial set of bioassays and thereafter using a generalised linear model (GLM) using a binomial distribution with a logit link function, and using confidence limits to predict mean % WFT mortality.

Leaf-dwelling WFT stages and B. bassiana

SID 5 (2/05) Page 10 of 58

Using the method developed in 2(a), replicated bioassays were done to test the susceptibility of WFT second stage larvae (L2s) and adult females to controlled doses of B. bassiana on chrysanthemum leaves over a 7-day period. B. bassiana was applied as the products Naturalis and Botanigard, as well as isolates cultured from the products, code numbers 432.99 and 433.99 respectively. Applications of Naturalis were not tested against adult WFT. The products were applied at the manufacturers’ recommended rates (Naturalis = 1.0 x 105 ml-1; BotaniGard = 5.5 x 107ml-1) while cultured isolates were applied at 1 x 108 conidia ml-1. All treatments were applied with 0.01% Agral and the control was water with Agral. Numbers of live and dead WFT were assessed daily for seven days and any cadavers were removed, incubated and assessed for mycosis after a further seven days. For larvae, the bioassay was repeated on three separate occasions with 10-15 dishes per treatment and 15-20 dishes per control and with five thrips per dish. For adults, the bioassay was repeated on three separate occasions with 14-15 dishes per treatment and 20 dishes per control, with five thrips per dish. Data were analysed with a generalised linear model (GLM) using a binomial distribution with a logit link function.

Ground-dwelling WFT stages and S. feltiaeUsing the method developed in 2 (a), replicated bioassays were done to evaluate the efficacy of S. feltiae (Nemasys F) against WFT prepupae and pupae in compost with 78% moisture content. The bioassay was done on three occasions. Control pots were treated with water and nematodes were applied at either recommended rate (250 million in 100l water per 1000m2, equivalent to 25 per cm2) or at twice this rate (50 per cm2) In the first set of bioassays, nematodes were also applied at x8 recommended rate (200 per cm2). Six WFT late L2s were used per pot and there were six pots per treatment. Emerged adults per pot were recorded seven and 14 days after treatment and percentage assumed pupal mortality was calculated. An additional set of bioassays were done using damp capillary matting as the substrate instead of compost. For these bioassays, identical pots were used as for compost, but they had a 1cm layer of set Plaster of Paris and charcoal in the base, dampened with water to maintain the dampness of the capillary matting. Eight late WFT L2s were used in each pot for this additional bioassay and there were seven replicates per treatment. Data were analysed using GLM as for the leaf bioassays. Ground-dwelling stages and B. bassianaUsing the method developed in 2(a), replicated bioassays were done to evaluate the virulence of B. bassiana against WFT prepupae and pupae. B. bassiana was applied as BotaniGard ES at concentrations of 8.8 x 107ml-1. Groups of 15-20 WFT late L2s were used per bioassay pot. The numbers of emerged adults per pot were recorded seven days after treatment. The bioassay was done on three occasions with five pots per treatment on the first occasion and with ten pots per treatment on the second and third occasions. Data were analysed using GLM.

Results and DiscussionLeaf-dwelling WFT stages and S. feltiaeDuring the last year of the project, after these bioassays had been completed, it was found that the Potter Tower was delivering approximately four times as many nematodes as it had been calibrated to apply. Thus Nemasys F applied at recommended, x2 and x8 rates was actually applied at x3.5, x8 and x32 rates. In the initial set of bioassays on WFT L2s, where water alone was compared with water plus Agral as a control treatment, water plus Agral led to higher thrips mortality (32%) than water alone (14%) five days after treatment, but this difference was not statistically different (Fig. 10). Nemasys F applied at x3.5 rate with Agral killed significantly more WFT L2s (52%) five days after treatment than either of the water treatments (Fig. 10). Subsequent bioassays comparing different rates of Nemasys F used water plus Agral as the control, and all nematode treatments included Agral, as this is the most commonly used wetter applied with Nemasys F by commercial growers.

In subsequent bioassays comparing different rates of Nemasys F against WFT L2s, none of the rates (x3.5, x8 or x32) killed significantly more thrips than in the controls (Fig. 11). These bioassays were repeated on several occasions but the significant effect of Nemasys F on WFT L2s given in the initial bioassays was not repeated, due to large variation between % thrips mortality given in the replicate dishes. It was concluded that although Nemasys F can kill WFT L2s at the dose rates tested, a statistically significant effect at the 95% level was rarely given.

In bioassays comparing different rates of Nemasys F against WFT adult females, only x32 rate killed significantly more adult thrips five days after treatment (80% mortality) than the 25% mortality given in the controls (Fig.12). Canadian work using filter paper bioassays also showed that Nemasys F at up to x8 rates failed to give significant kill of adult WFT except when they were immobilised on sticky tape (Buitenhuis, 2005). In our bioassays, both WFT adults and larvae were observed to have a marked avoidance response to S. feltiae, flicking their abdomens and attempting to clean nematodes off their bodies using their legs. This defence behaviour may affect the success rate of S. feltiae gaining entry into the thrips’ bodies. It is concluded that consistent kill of WFT adults by Nemasys F was only given in our

SID 5 (2/05) Page 11 of 58

bioassays at very high rates of application, and that immobile WFT (e.g. pupae) are likely to be more susceptible to infection by S. feltiae than moving ones, due to the avoidance response of the mobile life stages.

Leaf-dwelling WFT stages and B. bassianaNaturalis had no significant effect on L2 mortality compared with the control, but significant larval mortality levels of approximately 35% at day 7 were observed with Botanigard and the cultured isolates 432.99 (BotaniGard) and 433.99 (Naturalis) (Fig. 13). The proportion of larval cadavers supporting fungal mycelium were low (Table 3), which probably reflects the generally low levels of infection, although in addition the small size of the insect may provide insufficient biomass for high levels of fungal outgrowth and sporulation. The lack of significant larval mortality with Naturalis is probably due to the low concentration of inoculum used in this product compared to BotaniGard. Adult WFT were more susceptible to fungal infection than the larvae, with approximately 60% mortality obtained with both BotaniGard and 432.99 at 7 days post inoculation (Fig. 14). It is likely that moulting of the larvae reduced their susceptibility to fungal infection, as entomopathogenic fungi have contact activity and infectious conidia can be lost when the integument is shed at ecdysis. The overall levels of mortality obtained with the fungi were unexpectedly low, considering that about 70% control of WFT populations were observed previously in glasshouse experiments on cucumber crops in HDC PC 129 (Jacobson et al., 2001). It is possible that chrysanthemum secondary metabolites are inhibitory to fungal infection and further investigation in this area is warranted. There are also opportunities to look for fungal isolates with higher efficacy, since there is often significant variation in virulence to a host species between and within species of entomopathogenic fungi. Research to identify fungal isolates with improved virulence to WFT is ongoing in the USA and Australia, however identification of virulent European fungal isolates for use within the EU is also warranted

Ground-dwelling WFT stages and S. feltiaeThere were no significant differences between assumed percentage pupal mortality in the controls and in any of the nematode treatments (Fig. 15). In Experiment 3, the 54.1% mean assumed mortality in the recommended rate (25 per cm2) nematode treatment was almost significantly greater than the 26.3% mean mortality in the controls (P=0.06). This result is similar to that reported by Helyer et al. (1995) where compost drenches of S. feltiae (Nemasys) led to 59.8% mortality of WFT pupae when applied at 36 per cm2. We found no significant differences in the efficacy of S. feltiae applied at 25, 50 or 200 per cm2, whereas Helyer et al reported a significant increase in mortality rate (96.8%) where Nemasys was used at 107 per cm2, which is approximately equivalent to x4 current recommended rates for control of WFT. Ebssa et al (2001) reported that compost drenches of a laboratory strain of S. feltiae at 100 per cm2 gave approximately 50% WFT pupal mortality which was similar to our result (42%) at this rate (Fig. 15). Ebssa et al found that increase in dose rate up to 400 per cm2 led to increasing mortality of up to 85%, but dose rates above 400 per cm2 did not improve efficacy. In our bioassays using Nemasys F drenches to capillary matting, where nematodes at recommended and x2 rates led to 34 and 46% mortality respectively, but these mortality rates were not significantly different from the control mortality of 24%. These results were similar to those on compost in Experiment 1 (Fig. 15). Buitenhuis & Shipp (2005) reported that S. feltiae (Nemasys F) drenched onto filter paper led to significant kill of WFT prepupae and pupae, with a dose rate approximating to 25 per cm2 leading to 56-59% mortality of pupal stages, which is similar to our result on compost at this rate in Experiment 3. Buitenhuis & Shipp reported no significant increase in mortality of pupae with increasing dose rate, which is also in agreement with our results.

Ground-dwelling stages and B. bassianaThere was generally a low emergence rate of adults in the bioassays, with 95-96% assumed mortalities of the ground-dwelling stages. There was no evidence that the fungal treatments increased pupal mortality. A mean of 4.18% of thrips larvae were recovered as adults in the control and a mean of 3.38% in the Botanigard treatment which was not significantly different (p = 0.05) from the control. It is possible that the pupae had low intrinsic susceptibility to B. bassiana infection and the data suggest that the fungus would be more effective when applied to foliage. However, other fungus species or isolates could be active against pupae. Helyer et al. (1995) reported 80% control of WFT pupae using an isolate of Metarhizium anisopliae, albeit at very high doses. In addition, conidia of B. bassiana may have been adsorbed onto organic matter within the compost, reducing contact with pupae. In previous Defra funded research, we showed that conidia of M. anisopliae are rapidly absorbed by compost when applied to this substrate as a drench (Chandler & Davidson, 2006).

2 (c) Quantify the persistence of nematodes and fungal conidia on foliage and in flowers (ADAS and WHRI)

MethodsReplicated experiments were done in glasshouses to quantify the persistence of S. feltiae and B. bassiana

SID 5 (2/05) Page 12 of 58

on cucumber leaves (WHRI) and on pot chrysanthemum foliage, buds and flowers (ADAS), Plants were treated with either S. feltiae (Nemasys F) at the recommended rate of 250 million nematodes per 10002 in 100l water with Agral at a 0.04% concentration, or with B. bassiana. B. bassiana was applied as Naturalis and Botanigard at recommended rates and as isolates cultured from both products, code numbers 432.99 and 433.99 respectively, at a rate of 1 x 108 conidia ml-1. Treatments were applied to the plants to just before run off. Treatments were applied to the cucumbers in the late afternoon. Treatments were applied to the pot chrysanthemums in either the morning or the late afternoon, depending on sampling times following application, followed by immediate black-out. Treatments were applied on three occasions to the cucumbers and on four occasions to the pot chrysanthemums. The pot chrysanthemums were treated at the vegetative, bud initiation, bud-break and open flower stages. An additional treatment was applied at the flowering stage, to assess whether an application of water 0.5h after nematode application increased nematode persistence. After each treatment, leaf, bud or flower samples were taken from four replicate plants per treatment (pot chrysanthemums) and from four leaf canopy positions (cucumber). The samples were taken at selected time intervals (up to 24h post-application for nematodes and up to 350h post-application for B. bassiana) for counts of viable nematodes or conidia. Nematodes were washed from whole leaves, buds or flowers (pot chrysanthemum) or from 2 x 2 cm leaf samples (cucumber) and numbers of active, inactive or dead nematodes recorded using a binocular microscope. Fungal conidia were washed from leaf discs (1cm) or from whole chrysanthemum buds and plated onto a selective medium. Conidia viability was assessed in terms of the numbers of colony forming units recovered. The data were analysed by t-test, ANOVA or by regression analysis using a best fit model as appropriate.

Results and DiscussionOn cucumber leaves from all parts of the crop canopy, most of the nematodes were inactive within two hours of application and all were inactive three hours after application (Fig. 16). There was no significant effect of leaf position within the canopy on nematode persistence. On pot chrysanthemum plants, most of the nematodes were inactive within two or three hours of application on leaves, buds and flowers (Figs 17, 18 and 19). Leaves, buds and flowers were observed to dry quickly after nematode application. On only one spraying occasion, when the plants were at the bud-break stage, 49% nematodes were still active on lower leaves 16 hours after application (Fig. 17) and this was probably due to the lower leaves being shaded by leaves higher in the canopy. When water was applied to plants half an hour after nematode application, significantly more (55%) nematodes were still active on fully open flowers one hour after nematode application then without the water spray (11% active) but within two hours after nematode application, all nematodes were inactive with or without the water spray (Fig. 20). The additional water spray did not improve nematode persistence in opening buds (Fig. 20) or on top leaves.B. bassiana conidia persisted for much longer than the nematodes: on cucumber leaves for >350 h (data not shown) and on chrysanthemum leaves and opening buds for 70-120 h (Figs 21 and 22). On chrysanthemum leaves and buds, the numbers of colony forming units declined over time but on cucumber leaves there was relatively little decline, e.g. those of the cultured isolate 432.99 declined by only 15% over 350 h. There was no evidence that the formulated products persisted longer than their cultured isolates on either crop. On chrysanthemum, very few Naturalis conidia were recovered from either leaves or buds, whereas the unformulated isolate 43.99 were found in similar numbers to those of both the commercial and unformulated Botanigard products (Figs 21 and 22). On cucumber leaves, Botanigard was more persistent than Naturalis. The long persistence of Botanigard and the two unformulated isolates may have been partly due to secondary cycling, i.e. vegetative growth followed by sporulation, although there was no evidence of increased fungal populations on leaf surfaces, which might be expected should secondary cycling occur. The data suggest that B. bassiana is well adapted to conditions on leaf and bud surfaces, and should it be effective against WFT plant-dwelling life stages, its long persistence is likely to benefit control. 2 (d) Quantify the persistence of nematodes against ground-dwelling WFT stages in compost (ADAS)

MethodsRooted pot chrysanthemum cuttings were sprayed with Nemasys F at x8 recommended rate (200 per cm2) on two weekly occasions. Over the following eight weeks, a compost sample was taken from each of five replicate pots on six occasions at weekly or fortnightly intervals using a #13 (2 cm diameter) cork borer. S. feltiae were extracted from each sample using a modified Baermann funnel technique and mean numbers per cm2 compost were recorded.

Results and DiscussionBetween two and five weeks after application, estimated numbers of nematodes fell from 257 to 84 per cm2 (a decline of 67%) and then remained at approximately 100 per cm2 (equivalent to x4 recommended rate) for the remaining three weeks of the experiment (Fig. 23). This decline curve in nematode numbers in compost over seven weeks is similar to the 75% decline reported with S. feltiae in the compost of potted lilies over 60 days when used for control of sciarid flies (Gouge & Hague, 1994). Both our data and the published work showed that a proportion of the S. feltiae reaching the compost can persist for at least

SID 5 (2/05) Page 13 of 58

eight weeks after application, i.e. the whole production time for pot chrysanthemums following rooting. The results in 2 (c) and (d) demonstrated that nematode persistence in compost is much greater than on foliage, buds or flowers. The role of nematodes in the compost or substrate e.g. capillary matting in control of WFT populations was further investigated in 2 (e) below.

2 (e) Glasshouse experiments to test the efficacy of applications of S. feltiae to either plants or to growing substrate (ADAS and WHRI)

This work was done to meet milestone 03/04 but is more appropriate to report under Objective 2.

Pot chrysanthemums (ADAS)MethodsTwo consecutive experiments were done in a glasshouse compartment at ADAS Boxworth, to test the comparative efficacy of Nemasys F applied to the plants, to the compost, or to both plants and compost, on chrysanthemum plants under high WFT pressure. In each experiment, 12 thrips-proof mesh cages were stood on capillary matting. Each cage contained four pots, each with a single rooted pot chrysanthemum cutting, cv. Swingtime. There were three replicate cages for each of four treatments: (1) Water plus Agral applied to both the plants and the compost (control), (2) S. feltiae (Nemasys F) applied to just before run-off at x2 rate (500 million per 1000m2 in 100 l water), applied to plants only (for each pot, a circular piece of absorbent Tex-R ground-cover matting was stapled to thick cardboard and used to cover the compost and overhang the edge of each pot), (3) Nemasys F applied to compost only (each plant was covered with a plastic bag secured with a twisted tie around the base of the stem) and (4) Nemasys F applied to both plants and compost. Agral was used in all treatments at a 0.04% concentration. Treatments were applied to the pots each week for eight weeks, i.e. from rooting until flowering. In Experiment 2, the treatments were also applied to the cuttings from sticking i.e. for the full 10 weeks of the crop’s life. Treatments were applied in the late afternoon, during partial blackout, followed by full blackout, as per commercial practice. 20 WFT adults (18 females and two males) were released into each cage on two occasions, when the plants were four and five weeks old (at the vegetative and bud initiation stages). Compost samples were taken from the entire compost depth from one previously unsampled pot in each nematode-treated cage with a #13 (2 cm diameter) cork borer, prior to nematode application on three occasions. S. feltiae were extracted from each sample using a modified Baermann funnel technique and mean numbers per cm2 compost were recorded. Immediately after taking each compost sample, an empty plastic tube was placed in each hole in the compost, to collect nematode suspension during application. After spraying, these tubes were collected and numbers of live S. feltiae recorded. When the plants were 10 weeks old and at the marketing (flowering) stage, % damaged leaves and flowers and marketability were assessed on each plant and numbers of WFT per cage were recorded after destructive assessments. Each plant was cut off at compost level, leaves were washed in 70% alcohol, buds and flowers were dissected in alcohol under a binocular microscope, and one blue and one yellow sticky trap were left in each cage for two weeks after plant removal, to catch any remaining adults or those emerging from pupae in the compost. Mean % leaves damaged, mean % leaf and petal areas damaged and log10 transformed mean numbers of WFT per cage were analysed by ANOVA. Mean % damaged flowers were analysed by GLIM regression in order to take into account the varying numbers of flowers per plant.

Results and DiscussionIn both Experiments 1 and 2, approximately 50 per cm2 S. feltiae were retrieved from collection tubes placed in the compost and sampled immediately after nematode application on three sampling dates, demonstrating that the numbers of nematodes reaching the compost were consistent with the numbers applied (Figs 25 and 28). In both experiments, compost sampling immediately before three nematode application occasions (on each date this would be two weeks after the previous nematode application) indicated that numbers of S. feltiae in compost remained at approximately 50 per cm2 or above during the sampling period (Figs 24 and 27). In Experiment 1, where nematodes had been applied to compost alone, the results indicated that there had been a cumulative effect of sequential nematodes applications, with approximately 200 per cm2 extracted from compost samples taken just before the 6th and 8th applications (Fig. 24). This cumulative effect was only indicated on the final sampling date in Experiment 2 (Fig. 27). It is likely that numbers of nematodes in compost fluctuated between application dates, as results from work in 2 (d) showed that numbers decline over time, but as weekly applications were made in Experiments 1 and 2, at least 50 per cm2 remained in the compost throughout the experiments.

In both experiments, numbers of WFT per cage were significantly reduced where nematodes had been applied to compost only or to both plants and compost, but not where nematodes had been applied to plants only (Tables 4 and 5, Figs 26 and 29). These results are consistent with those of the leaf bioassays reported in 2 (b). In Experiment 2 only, back-transformed mean numbers of WFT per cage were significantly lower where nematodes had been applied to both plants and compost (14.6 per cage) than

SID 5 (2/05) Page 14 of 58

where nematodes had been applied to compost only (60.9 per cage, Table 5). In Experiment 1, mean % flowers damaged were significantly reduced where nematodes had been applied to compost only or to both plants and compost (Table 4). In Experiment 2, % flowers damaged and % petal area damaged were only significantly reduced where nematodes had been applied to both plants and compost (Table 5). The results of both cage experiments indicate that when Nemasys F is applied as weekly sprays to pot chrysanthemums, control of WFT life stages in the compost or substrate has an important role in reducing WFT populations. The greater reductions in WFT numbers and flower damage given by Nemasys F applied to both plants and compost than when applied to compost only in Experiment 2 cannot be explained by any corresponding improved control of WFT and flower damage given by Nemasys F applied to plants only than in water plus Agral controls. As a strong avoidance response was shown by WFT on nematode-treated leaves in laboratory bioassays, it is possible that on plants sprayed with Nemasys F, some WFT dropped from treated leaves, buds and flowers to the compost in an attempt to escape infection, but subsequently became infected in pots where nematodes had been allowed to reach the compost and where active, viable nematodes were present throughout the life of the experiment. The level of flower damage and % plants marketable in these experiments reflected the high numbers of WFT which were released to the cages in order to achieve the objective and do not represent WFT densities occurring on commercial nurseries where Nemasys F is used for control.

The failure of Nemasys F to reduce numbers of WFT when applied to plants only could also explain why growers find that Nemasys F does not give a quick ‘knockdown’ of WFT as an effective insecticide would and that several weekly applications are needed before a reduction of WFT on sticky traps is seen on commercial nurseries. The results could also explain why growers who have delayed the start of weekly Nemasys F programmes until the bud stage on pot chrysanthemums to reduce costs, have found that this has given poorer control of WFT than when the nematodes have been applied from the rooting stage onwards. This poorer WFT control could be due to fewer nematodes reaching the compost from the bud stage onwards than at earlier growth stages, due to the plant canopy closing over. Nemasys F is currently recommended to be applied to just before run-off, as it was thought that the nematodes mainly kill thrips on the plants rather than those in the compost or substrate. Our results indicate that even when applied to just before run-off, significant numbers of nematodes reach the growing medium or substrate. The key new information gained on the fate of S. feltiae after application and its persistence and efficacy against WFT life stages on plants and in substrates offers the opportunity for further development of application methods for optimum WFT control.

Cucumbers (WHRI)MethodsAn experiment was done in three glasshouses at Warwick HRI. In each glasshouse, 12 cucumber plants were grown in two rows of six, in 20 cm diameter pots of compost. WFT adults were released to each plant at 35 per plant and left for 14 days. One of the following treatments was applied in one glasshouse each: (1) S. feltiae (Nemasys F) applied to just before run-off to the foliage at x2 rate (500 million per 1000m2 in 100 l water), (2) Nemasys F applied to compost and capillary matting on floor and (3) untreated control. Treatments were applied in the late afternoon, weekly from day zero to day 21. Numbers of WFT (all life stages) were counted on leaves at the top, middle and bottom of the plants on days 6, 13, 20 and 27. Samples from leaves (1 cm diam discs), compost (4cm cores taken with a 1 cm diam cork borer) and matting (2 cm by 2 cm) were taken immediately after treatment and six days after each treatment. S. feltiae were extracted from these samples by washing the leaf discs and using an adapted Baermann funnel technique for the compost and matting samples and numbers of active nematodes were recorded.Results and DiscussionWFT numbers on the cucumber leaves were low, with a mean of only 2.7 WFT per untreated leaf on day 27 (Fig. 30). Nemasys F applied to either foliage or to compost and matting significantly reduced WFT numbers compared with those on untreated leaves (Fig. 30). The overall low numbers of WFT did not allow the comparative efficacy of the different placements of Nemasys F to be distinguished during the experimental time available. Numbers of active nematodes per cm2 leaf immediately after each application ranged from 35-59 (Fig. 31) which was consistent with the numbers applied (50 per cm2). Numbers of active nematodes per cm2 matting and compost were higher than expected immediately after each application, even on day 0, when no accumulation of nematodes could have occurred (Fig. 31). It is possible that the compost and matting could have been sprayed inaccurately. No active nematodes were found on leaf samples six days after each application date (Fig. 32). The short survival of S. feltiae on leaves is consistent with results in 2 (c) which showed that nematodes are inactive within three hours after application to cucumber foliage. Numbers of active nematodes in matting and compost samples six days after application ranged from 25-113 and 106-299 per cm2 respectively (Fig. 32). These numbers are similar to those surviving in compost samples between application dates in the pot chrysanthemum cage experiments reported above, where Nemasys F applied to only the compost gave significant reduction in WFT numbers.

SID 5 (2/05) Page 15 of 58

Conclusions from work in Objective 2 S. feltiae (Nemasys F) plus Agral can kill WFT L2s on chrysanthemum leaves in laboratory

bioassays, typically giving approximately 50% kill after five days. However, this % kill was not always significantly different than the 30% kill given by water plus Agral controls.

Only high rates of Nemasys F (x32 recommended rate) killed significant numbers of WFT adults in laboratory bioassays (80% kill, compared with 25% kill in controls).

Both WFT adults and larvae show a marked avoidance response to S. feltiae and this defence behaviour could affect the success of nematode infection.

Recent research in Canada has shown that Nemasys F killed WFT adults only when they were immobilised first.

Nemasys F at recommended rate killed 54% WFT pupae in compost bioassays although this was not significantly more than the 26% kill in controls. There was no evidence of a dose response with Nemasys F against ground-dwelling WFT life stages.

B. bassiana (Botanigard) killed significant numbers of WFT L2s on chrysanthemum leaves in bioassys, but % kill was low (35% at day 7) and Naturalis had no significant effect.

WFT adults were more susceptible than larvae to B. bassiana, with approximately 60% kill at day 7 on chrysanthemum leaves by both Botanigard and isolate 432.99 cultured from Botanigard.

It is possible that chrysanthemum secondary metabolites inhibit B. bassiana, as previous HDC-funded research indicated that B. bassiana was effective against WFT on cucumber.

There was no evidence that B. bassiana killed ground-dwelling stages of WFT in compost. S. feltiae have short persistence on plant surfaces after application in glasshouses. Most

were inactive within three hours of Nemasys F application to cucumber leaves or to pot chrysanthemum leaves, buds and flowers. Thus there is a very short ‘window’ for S. feltiae to infect WFT on plants before plant surfaces dry.

S. feltiae have long persistence in compost, with a proportion surviving for at least eight weeks after application, giving a long ‘window’ for infecting ground-dwelling WFT stages.

B. bassiana conidia persisted for at least 350 hours after glasshouse application to cucumber leaves and for 70-120 hours on pot chrysanthemum leaves and buds. Botanigard was more persistent than Naturalis.

In a glasshouse experiment on cucumber, weekly Nemasys F applications to either the leaves or the growing substrate significantly reduced numbers of WFT, but WFT numbers were too low to allow the efficacy of the different nematode placements to be compared.

In glasshouse cage experiments with pot chrysanthemums, numbers of WFT were significantly reduced where Nemasys F had been applied weekly to the compost only or to both plants and compost, but not where the nematodes had been applied to plants only. This result is consistent with those in the leaf bioassays and indicates that nematode control of WFT life stages in the compost or substrate has a more important role than control of WFT on the plants in reducing WFT populations.

In one of the cage experiments, Nemasys F applied to both chrysanthemum plants and compost gave greater reductions in numbers of WFT and in flower damage than when applied to compost only. It is possible that in addition to killing WFT ground-dwelling stages, Nemasys F applications may cause some WFT adults and larvae to drop from the plants to escape infection, but on reaching the compost they become infected.

The current recommendation for WFT control with Nemasys F is to make weekly applications to just before run-off. Although this gives significant control of WFT on pot chrysanthemums, it is possible that adapted application methods could optimise WFT control. Such methods would need to be tested in further development work.

Objective 3. To determine whether the efficacy of S. feltiae and B. bassiana can be improved by manipulating WFT behaviour with semiochemicals or by adapting use of glasshouse lighting (ADAS, RR, Keele and WHRI)

3 (a) Laboratory behavioural studies with semiochemicals (RR)MethodsBioassays, using whole leaves of chrysanthemum, cv. Swingtime, and cucumber, cv. bush cucumber, were developed to investigate the effects of antifeedants and volatile semiochemicals on the behaviour of WFT adults and 2nd instar larvae (L2s). Bioassay on cucumber leaf: A cucumber leaf was placed between a modified Petri dish (9cm diameter) and a flat plate that when clamped together formed a ventilated, thrips-proof arena. An individual WFT was placed in the centre of the arena through a hole in the top. The WFT was observed for 20 minutes under an illuminated magnifying glass at 20°C +/- 2°C. Different behaviours including walking, being still,

SID 5 (2/05) Page 16 of 58

and flight were recorded and analysed using the Observer software (Noldus©). Location on the arena cage or leaf was also recorded. Where possible, WFT were tested with all semiochemicals on the same day to take account of day to day behavioural changes. At least 20 replicates were done for adult males and females and 10 replicates for L2s for the first antifeedant Tasmannia stipitata. Fewer replicates were done with the other antifeedants due to the time taken to acquire these alternatives when the T. stipitata became unavailable from Australia part-way through the project. Data were analysed by t test at P=0.05. The ‘% time spent data’ were arcsine transformed prior to analysis.

Bioassay on chrysanthemum leaf: Bioassays with chrysanthemum leaves could not be done using the method developed for cucumber due to the lack of contrast between the WFT and the dark leaf surface. A modified system was developed in which the leaf was enclosed in a small ventilated Perspex arena (22 x 22mm square by 30mm high). The arena was lit by infra red light emitting diodes, under which the WFT was easily recorded by an infra red sensitive camera with a zoom lens, onto a TV monitor and to video tape. Behavioural observations were recorded with Observer, from the TV screen. Ten replicates were done for each bioassay and the data analysed by t-test.

Antifeedants: These were extracts of Tasmannia stipitata and T. lanceolata, plants containing polygodial (37% and 32% respectively) an antifeedant active against aphids and WFT, and synthetic polygodial as a comparison. For the antifeedant bioassays, whole cucumber plants were sprayed to run-off and excised leaves were used either the same day or left on the plant for 24 or 48h before testing. For the antifeedant sprays, a 3g/litre, (i.e. 3000ppm, approximately equivalent to 1000ppm of the active ingredient) solution of extract was used in an ethanol/Agral formulation. An ethanol/Agral solution was used as a control. To extend our understanding of the action of the antifeedant extract, pure polygodial was synthesised using a newly developed route, which provided multi-gram samples of product. As polygodial is insoluble in water and is in relatively short supply, an ethanol formulation was used and applied to whole cucumber plants by low volume spinning disc sprayer at a rate equivalent to that found in the antifeedant extract (1000ppm). An ethanol formulation of T. lanceolata was applied at 3000ppm by spinning disc as a direct comparison. Control leaves were untreated or sprayed with aqueous ethanol. These tests were performed with adult WFT only, with a minimum of seven replicates per treatment.

Volatile semiochemicals: To test the effects of volatile semiochemicals, a filter paper disc, treated with either 1µl of hexane (control) or 1µl of a 1mg/ml solution of (E)-β-farnesene (EBF), was placed in the middle of an untreated cucumber leaf within the arena. Each bioassay was performed using an individual WFT for a period of 20 min and the behaviours were recorded as in the other bioassays.

Results and DiscussionAntifeedants T. stipitata on cucumberPreliminary bioassays on untreated cucumber leaves showed differences in adult male and female behaviour, with males spending significantly more time being still than females (Fig 33). In the presence of the antifeedant, both males and females spent significantly more time walking and less time on the treated leaf surface throughout the majority of the test periods (Figs. 34-36). The number of flights by both sexes was also significantly increased by the antifeedant (Fig.37). T. stipitata significantly increased walking behaviour of L2s for up to 48h after treatment (Fig 38), and significantly reduced time spent on the treated leaf surface for up to 24h after treatment (Fig 39), as observed for adults. These results show that both adult and L2 movement is significantly increased by the antifeedant and could result in greater contact with microbiological control agents on plants. However, the increase in flight behaviour indicates that adults will avoid contact with antifeedant and will probably fly from the plant. Thrips larvae cannot avoid the antifeedant by flying away, but will seek out untreated areas. This could provide a short-term reduction in pest pressure on the crop, provided that good coverage could be achieved from this treatment.

T. stipitata on chrysanthemumResults of behavioural observations of WFT adults and L2s on chrysanthemum leaves, either untreated or sprayed to run off with T. stipitata, were similar to those on cucumber. Adult walking behaviour was increased, although this was only significant for females (Figs 40 & 41). Time spent on the leaf surface and number of flights by adults was also significantly increased (Figs 42 & 43). Unlike on cucumber, there was no significant increase in walking by L2s on the treated chrysanthemum leaves (Fig 44). However, L2s did spend significantly less time on the antifeedant-treated leaf compared with the control (Fig 45). As similar results for WFT behaviour were given on cucumber and on chrysanthemum leaves, all subsequent assays were carried out on cucumber as they could be done in real time. When the supply of the T. stipitata antifeedant from Australia became unavailable, to allow the work to continue, pure polygodial was synthesised. This enabled a comparison of the activity of the active ingredient against the whole plant extract. Ultimately, we sourced another polygodial-rich antifeedant extract, from a related plant, T. lanceolata, grown and produced by a UK company. GC analysis showed it to contain approximately 32%

SID 5 (2/05) Page 17 of 58

polygodial, a similar level to the T. stipitata extract (Fig 46).

Polygodial and T. lanceolata on cucumberThese treatments had been applied in ethanol by spinning disc. There was a significant reduction in time spent and in time being still for female WFT on polygodial-treated leaves immediately after treatment, but no other treatment effects were significantly different (Figs 47-49). Thrips made very few flights from leaves with these treatments. This application method does not give complete coverage and may not have been as effective as the spray to run-off. However, there were no significant differences between the polygodial and the T. lanceolata treatments, indicating that the effects of the antifeedant extract on WFT behaviour are due to the polygodial.

Tasmannia lanceolata on cucumber sprayed to run-offAs the spinning disc application results were less clear than expected, a limited series of experiments with T. lanceolata sprayed to run-off were done (Figs 50-52) to determine WFT adult response to treatment prior to the glasshouse experiments. As for T. stipitata, WFT spent less time on the treated leaf and walking behaviour was increased, although, due to variability in thrips behaviour and low replication, the results for T. lanceolata were not as conclusive. Interestingly, there were fewer flights from the treated leaf with this antifeedant extract, which was a more refined extract than the T. stipitata.

Volatile semiochemicals:In the presence of EBF no significant behavioural differences, compared to the untreated leaf, were observed for either males or females at the concentration tested.

3 (b) Laboratory bioassays with semiochemicals and S. feltiae (RR)

MethodsEffects of antifeedants on S. feltiaeLaboratory bioassays: The extract of T. stipitata had been shown to be non-toxic to S. feltiae nematodes in a previous MAFF-funded project (HH1838SPC). A suspension of S. feltiae (Nemasys F) was made up in water plus Agral at 0.5 ml per 10 l water to give 10,000 nematodes/ml (equivalent to x4 recommended rate for WFT control). A sub-sample was taken and after an appropriate dilution with water, numbers of dead, live and moribund nematodes were counted in four replicate 1ml samples. The extract of T. lanceolata was made up to x2 and x4 the selected rate of 3000ppm in an ethanol/Agral solution. Equal volumes (12.5 ml) of the nematode suspension and each of the antifeedant solutions were mixed together in individual Petri dishes to give full (3000ppm) and x2 (6000ppm) the selected antifeedant rate. Nematode suspensions in water plus Agral and 6% ethanol were used as controls. There were five replicates per treatment. The dishes were left uncovered and incubated at 22ºC in the dark for 6h before assessment, when the contents were emptied into a beaker, made up to 50ml with water and aerated. Numbers of dead, moribund and live nematodes were assessed after an appropriate dilution. A pure polygodial formulation would be incompatible as a tank mix with S. feltiae, as it requires at least 15% ethanol to dissolve the compound and this rate of alcohol was toxic to S. feltiae in the Petri dish test.

Results and DiscussionA mean of 97.5% S. feltiae were still alive after incubation of nematode suspensions with T. lanceolata at both rates of the antifeedant for six hours (Fig 53). This result indicated that the antifeedant did not affect nematode survival, even at double the rate found to be effective against WFT.

3 (c) Laboratory bioassays with semiochemicals and B. bassiana (WHRI)

MethodsThe effect of the antifeedant extracted from T. lanceolata on B. bassiana was investigated in a series of laboratory experiments:Germination of conidiaThe effect of the antifeedant on the germination of conidia was tested using two Botanigard products : Botanigard ES (emulsifiable solution) and Botanigard WP (wettable powder). The antifeedant was applied using two different methods. In the first method, antifeedant solution (prepared in ethanol and 0.1% Agral) was incorporated into Sabouraud’s dextrose agar (SDA) at 3000 ppm. Serially diluted aliquots (20µl) of Botanigard WP and SE, prepared originally at the manufacturers recommended rate, were then pipetted onto the SDA + antifeedant. In the second method, serially diluted aliquots of Botanigard WP and SE were mixed with antifeedant solution to simulate a tank mix, and then plated onto SDA. Controls consisted of blanks prepared without antifeedant. Replicate plates were incubated at 23C, in darkness, for 24h before the numbers of germinated and ungerminated conidia were counted.

Numbers of colony forming unitsA similar test was done to measure the numbers of colony forming units developing when conidia, treated

SID 5 (2/05) Page 18 of 58

as above, were plated onto SDA.

Virulence of B. bassianaA laboratory bioassay was done to examine the effect of the antifeedant on the virulence of B. bassiana against adult female WFT. The bioassay method was as described previously in 2 (b). B. bassiana was used as Botanigard ES at the manufacturers recommended rate, and was either tank mixed with the antifeedant as described above (3000 ppm) and sprayed onto chrysanthemum leaf surfaces, or applied to the leaf surface following a spray application of the antifeedant which was then allowed to air dry. The numbers of live and dead adults were assessed daily for seven days and any dead larvae were removed, incubated and assessed for mycosis after a further seven days. The bioassay used 7 - 9 dishes per treatment and 11 dishes per control with five thrips per dish.

Results and DiscussionGermination of conidiaThe antifeedant inhibited germination of both Botanigard WP and ES applied as a tank mix or onto antifeedant incorporated into SDA (Table 6).

Numbers of colony forming unitsThe antifeedant prevented the formation of fungal colonies of both Botanigard WP and ES when applied as a tank mix or directly incorporated into SDA (data not shown).

Virulence of B. bassianaThere were indications that the antifeedant reduced the infectivity of the fungus and in particular inhibited outgrowth and sporulation on the surface of cadavers (Table 7); however the negative effect of the antifeedant was less than that observed with the germination and colony viability experiments. A small number of comparable studies have been done on the effects of fungicides on entomopathogenic fungi and they indicate that compounds which are inhibitory to fungal growth and germination in agar tend to have less of an effect on the plant scale performance of the fungus (Jaros-Su et al. 1999; Chandler & Davidson, 2006).

3 (d) Small-scale glasshouse experiments with semiochemicals and S. feltiae (RR)

MethodsDue to the incompatibility of the antifeedant and B. bassiana and the loss of supply of T. stipitata, the combination tested was the extract of T. lanceolata with S. feltiae. Eight glasshouse experiments were done using pot chrysanthemums, cv. ‘Swingtime’, at the opening bud/flower stage. For each experiment, there were four replicate pots, each with five chrysanthemum plants per treatment. Each pot was infested with approximately 40 WFT (adults and larvae in equal ratios) and the treatments were sprayed to just before run-off in the early evening. The treatments were: a) Water, b) An Agral/ethanol formulation control, c) S. feltiae (Nemasys F) at 10,000 per ml (x4 recommended rate) in Agral solution, d) T. lanceolata extract in Agral and ethanol solution, and e) A Nemasys F plus T. lanceolata mixed formulation. The treatments were re-randomised for each experiment. Sticky paper was placed beneath each pot to catch any WFT falling from the plants after treatment. The numbers of WFT per pot were assessed after 48h. Six buds and six leaves were taken randomly from each pot and put into 95% ethanol. These samples were dissected and assessed for numbers of WFT adults, L1s, L2s and pupae. After sampling, the plants were cut at compost level and beaten onto a white tray. Numbers of each WFT life stage were counted per pot. The data from the eight experiments were combined and analysed by ANOVA and significant differences between treatments were then assessed by t test at P=0.05.

Results and DiscussionS. feltiae alone had no significant effect on the numbers of WFT 48 h after treatment, compared with the water and blank formulation controls. This result is consistent with the bioassay results reported in 2 (b). WFT caught on the sticky traps under the plants were very few and were not analysed separately, but they were added into the overall totals. There were no significant treatment effects for WFT on the leaf samples (Fig 54). In the buds, where WFT numbers were higher, the antifeedant treatments significantly reduced numbers of L2s and total WFT (Fig 55). Similarly, when thrips were beaten from the plants, the antifeedant treatments led to significantly fewer WFT than the other treatments (Fig 56). When the results from all the assessments in all experiments were combined, the antifeedant treatments led to significantly fewer WFT than the other treatments (Fig 57). There was no statistical evidence that the combination of antifeedant and nematodes was more effective than the antifeedant alone.

3 (e) and (f) Laboratory bioassays and glasshouse studies with adapted use of glasshouse lighting (Keele)This work aimed to identify whether a short burst of light during the night would increase WFT walking activity. The work was limited to glasshouse studies which were more realistic than laboratory bioassays.

SID 5 (2/05) Page 19 of 58

MethodsCucumber plants (cv. Tanja) grown in a glasshouse were infested with approximately 600 WFT throughout June and July 2005. Glasshouse sodium lighting illuminated the cucumbers at 01.00 h BST for 30 min and thus provided a burst of light (10 Wm-2). Background light levels were 0 Wm-2 at night. Infrared video recording equipment was used to assess the numbers of adult WFT present and the proportion moving on the cucumber leaves. The recording was reviewed for one minute at 15 min intervals for two hours prior to light burst and for two hours after. Data were analysed with a paired t-test.

Results and DiscussionA burst of light at night did not increase walking activity in adult WFT (P=0.10) or increase the numbers present (P=0.93) when the recording before the light burst was compared to that after the light burst. This result seemed to be because, contrary to expectation, thrips walking activity continued at night (Fig. 7), so a light burst would make little difference.

Conclusions from work in Objective 3 Extract of Tasmannia stipitata increased movement of WFT adults and larvae and adult flight and

could provide a short-term antifeedant effect. Extract of Tasmannia lanceolata could provide a short-term reduction in WFT numbers on

chrysanthemums due to its antifeedant effect. The polygodial within the antifeedant extracts is largely responsible for the effects on WFT

behaviour. S. feltiae and the two antifeedant Tasmannia spp. plant extracts can be applied as a tank mix with

no toxic effects on the nematodes. Pure polygodial is incompatible. A combination of the T. lanceolata antifeedant and S. feltiae did not give significantly greater

control of WFT than the antifeedant alone. Although the Tasmannia spp. extracts increased WFT movement and thus could also increase

contact with microbiological control agents, they do not have potential for enhancing control by either S. feltiae or B. bassiana. S. feltiae is now thought to be more effective against stationary WFT than moving ones and T. lanceolata extracts adversely affected B. bassiana germination, viability and infectivity.

The volatile semiochemical (E)-β-farnesene failed to influence WFT behaviour in the bioassay arena used, possibly due to a saturation effect in the enclosed chamber.

A light burst at night did not increase walking activity of adult WFT on cucumbers in the glasshouse, since they were already very active.

The efficacy of S. feltiae and B. bassiana cannot be improved by manipulating WFT behaviour with the semiochemicals tested or by adapting use of glasshouse lighting

Objective 4. To quantify the effects of S. feltiae, B. bassiana and semiochemicals on other selected biological control agents used in IPM on protected crops (ADAS, WHRI and RR)

4 (a) Laboratory bioassays on effects of S. feltiae (ADAS) and B. bassiana (WHRI) on other biological control agentsKnowledge reviewThe scientific literature was examined, and other scientists and producers of biological control agents contacted for known information on the compatibility of S. feltiae and of entomopathogenic fungi with potential to control WFT, with arthropod biological control agents used within IPM on protected crops.There were four reports of incompatibility between isolates of entomopathogenic fungi and arthropod biocontrol agents, including B. bassiana infecting Encarsia formosa and Phytoseiulus spp., as well as Metarhizium anisopliae infecting P. persimilis. However, there were also reports of isolates of B. bassiana, Beauveria brongniartii and Verticillium lecanii having no effect on populations of E. formosa, P. persimilis, Aphidius colemani, Trichogramma spp. and Amblyseius cucumeris. These conflicting reports probably arise as a result of researchers using different fungal isolates and different methods in their bioassays. In commercial practice, V. lecanii is used for aphid and whitefly control within IPM programmes with no observed adverse effects on other biological control agents. Steinernema spp. infective juveniles enter the host body through the mouth, anus or spiracles. There is no reported evidence that S. feltiae can infect predatory mites used as biological control agents e.g. Phytoseiulus persimilis, Amblyseius cucumeris and Hypoaspis spp. and it is thought that the body orifices in mites are too small to allow nematode entry. Research in a Horticulture LINK project showed that S. feltiae will infect leaf miner larvae but only low numbers of the larvae and pupae of the leaf miner parasitoid, Diglyphus isaea, and using both the parasitoids and nematodes improved leaf miner control (Jacobson, 1999). S. feltiae were shown in previous Defra-funded research to be safe to nymphs of the

SID 5 (2/05) Page 20 of 58

predatory bug Orius laevigatus (Bennison, 2002). Research in Canada has shown that third instar larvae of the predatory beetles, Atheta coriaria ( now available in the UK for control of sciarid and shore flies) can be killed by S. feltiae, although the adult beetles are not susceptible (Jandricic et al, 2005). As S. feltiae are used for biological control of dipteran (fly) pests e.g. sciarid flies, dipterous biological control agents are likely to be vulnerable. A significant gap in knowledge was identified as the compatibility between S. feltiae and B. bassiana with the predatory midge Aphidoletes aphidimyza, which is used widely for aphid control within IPM programmes. A. aphidimyza larvae feed on aphids on plant foliage and then drop to the ground to pupate, thus both leaf-dwelling and ground-dwelling stages could be susceptible to S. feltiae when used for WFT control.

Methods Bioassays with S. feltiaeA. aphidimyza eggs were reared to 3-day old larvae on Aulacorthum solani. Three A. aphidimyza larvae and 20 A. solani were placed in each of ten replicate Petri dishes (5 cm diameter) with a moist filter paper disc. One of two treatments (0.24 ml) was added to each dish using a pipette: (1) Water plus 0.04% Agral (control) and (2) S. feltiae (Nemasys F) at the recommended rate (250 million in 100 l water per 1000m2, equivalent to 25 per cm2) plus 0.04% Agral. The dishes were incubated at 21 2 C with a 16 h photoperiod. Live, dead or missing larvae were recorded three days after treatment. Dead larvae were dissected in a drop of water and the presence or absence of nematodes in the body were recorded. Data was analysed using a GLM to predict % A. aphidimyza mortality. Bioassays with B. bassianaA. aphidimyza eggs were reared to 1-day old larvae on Myzus persicae on brassica leaves. Bioassay units consisted of a Petri dish (5cm diameter) with a filter paper disc sprayed with 2ml of B. bassiana conidial suspensions or a wetting agent control using a Potter tower. B. bassiana was applied as Botanigard ES at a concentration of 1.0 x 108 conidia.ml-1). Ten 1-day old A. aphidimyza larvae and M. persicae ad libitum were added to each unit and incubated at 21C with a 16 h photoperiod. The numbers of live A. aphidimyza were recorded after seven days. Ten replicate bioassay arenas were used per treatment and the experiment was repeated on two occasions.

Results and DiscussionBioassays with S. feltiaeMean predicted % mortalities of A. aphidimyza were 26.7% in the control and 50% in the Nemasys F treatment and nematodes were found inside dead A. aphidimyza larval bodies, but these differences in mortality were not significantly different. The International Organisation of Biological Control (IOBC) testing scheme for side effects of pesticides on biological control agents categorises pesticides to be ‘slightly harmful’ if 25-50% are killed and ‘moderately harmful’ if 50-75% are killed. Further work would be needed to confirm the effect of weekly applications of Nemasys F on A. aphidimyza larvae on plants and on pupal cocoons in the growing medium. Bioassays with B. bassianaMean % survival of A. aphidimyza larvae seven days after treatment was 37.4% in the control and 34.2% in the Botanigard treatment. Treatment with Botanigard had no significant effect ( p = 0.05) on survival, indicating that the fungus would be compatible with A. aphidimyza.

4 (b) Laboratory bioassays on behavioural effects of semiochemicals on non-target biological control agents (RR)MethodsAntifeedant effects on aphid parasitoidsThe effects of applications of pure polygodial and of the T. lanceolata extract on the behaviour of the aphid parasitoid, Aphidius ervi, were tested immediately after application and at 24 and 48h after treatment. Small cucumber plants, cv. bush cucumber, with two true leaves were treated as follows: a) Untreated, b) ethanol, c) T. lanceolata extract at 3000ppm in ethanol, d) polygodial at 1000ppm in ethanol. Treatments were applied by low volume spinning disc sprayer. The test plant was placed on a turntable within an open sided cage (50x 50x50cm), allowing the observer to rotate the plant if the parasitoid moved to the back. For each observation a single mated female A. ervi was allowed to walk from a small glass vial onto the base of the stem of the test plant. Behavioural observations (walking, being still and cleaning) were made using Observer software. The test ended when the parasitoid left the plant. There were at least 17 replicates for each treatment. Where possible, all treatments were tested during each observation period to take account of day to day changes in parasitoid behaviour.Antifeedant effects against aphidsThe effect of applications of pure polygodial and of extracts of T. stipitata and T. lanceolata on the settlement of alate Aphis gossypii was assessed. Each trial used 40 cucumber plants, cv. bush cucumber, with two true leaves. There were four treatments each with 10 replicates. In trials 1 & 2 ethanolic treatments were applied by low volume spinning disc sprayer: a) untreated, b) ethanol, c) T. lanceolata

SID 5 (2/05) Page 21 of 58

extract at 3000ppm in ethanol, d) polygodial at 1000ppm in ethanol. In trial 3 aqueous treatments were applied to run-off: a) water, b) ethanol/Agral formulation, c) T. stipitata extract at 3000ppm in ethanol/Agral formulation, d) T. lanceolata extract at 3000ppm in ethanol/Agral formulation. The plants were left to dry for 0.5h after application and then transferred to individual, ventilated, cylindrical cages (24cm high x 12cm dia) in a CE room (16:8 L:D; 20°C +/- 2°C). Ten alate A. gossypii, which had been reared on cucumber, were introduced into each cage and the number of aphids settled was recorded after 2, 5 and 24h.

Results and DiscussionAntifeedant effects on aphid parasitoidsParasitoid behaviour varied greatly between individuals and there were no significant treatment effects for either time spent or different behaviours on plants immediately after application, except for time spent cleaning on T. lanceolata-treated plants compared to the ethanol controls (Figs 58 & 59). There were no significant differences at 24 and 48h after application. The antifeedants are unlikely to be a problem for A. ervi, however further tests with a spray to run-off should be considered.Antifeedant effects against aphidsIn trials 1 and 2, there was no effect on alate A. gossypii settling behaviour on cucumber plants after application of polygodial and T. lanceolata extract by spinning disc sprayer (Figs 60 & 61), except for polygodial after 5h in trial 1 (Fig 60). However, in trial 3, both antifeedant extracts significantly reduced aphid settlement when applied to run-off (Fig 62). These antifeedant extracts have potential to reduce aphid colonisation. However, as with the WFT tests good coverage of the plant with a high volume spray is needed for the antifeedants to have significant effects against aphids.

Conclusions from work in Objective 4 S. feltiae (Nemasys F) can infect and kill Aphidoletes aphidimyza larvae but in laboratory

bioassays, did not significantly increase % mortality compared with a water plus Agral control. B. bassiana (Botanigard ES) did not affect % survival of A. aphidimyza larvae in laboratory

bioassays and the two biological control agents should be compatible. T. lanceolata extract and the synthetic polygodial applied by a low-volume spinning disc sprayer

had no significant effect on time spent or behaviour of Aphidius ervi on the plants. Extracts of T. stipitata and T. lanceolata have the potential to reduce aphid colonisation of

cucumber by the aphid A. gossypii, when applied to run-off to achieve good plant coverage.

Knowledge Transfer

Presentations on the research results given by the research collaborators: Jude Bennison: Eight Defra-funded IPM workshops for growers of protected ornamentals during

2003 (five in January and three in October). Jude Bennison: IPM workshop for growers of protected ornamentals in Australia, February 2003. Jude Bennison: Two IPM workshops for growers of protected ornamentals in New Zealand,

February 2003. Jude Bennison: HDC WFT Review Meeting for funders, researchers and the industry, at HRI

Wellesbourne, 1 May 2003. Neil Holmes, Keele University: First Symposium on Palaearctic Thysanoptera, Gödöllő, Hungary,

August 2003. Jude Bennison: EU COST meeting on entomopathogenic nematodes against thrips, Helsinki, July

2004. Jude Bennison, gave training course on identification and control of thrips to growers and

biological control staff in Kenya, July 2004. Guthrie, F., Smart, L., Wadhams, L and Bennison, J. The behavioural effects of a plant-derived

antifeedant on western flower thrips. Poster presentation at the Meeting of the Royal Entomological Society, York University, 21-23 July, 2004.

Jude Bennison gave 2 presentations on the results of the project at the International Congress of Entomologists, Brisbane, Australia, August 2004.

Jude Bennison: ‘Integrated control of thrips in the UK’, Canadian Greenhouse Conference, Toronto, October 2004.

Jude Bennison and Neil Holmes: VIII International Symposium on Thysanoptera and Tospoviruses, California, September 2005.

Monique Tomiczek, IOBC working group meeting, Integrated Control in Protected crops, Finland, April 2005.

Jude Bennison and Monique Tomiczek, EU COST meeting on entomopathogenic nematodes, Hannover, Germany, April 2005.

Jude Bennison, grower seminar, Finland, April 2005.

SID 5 (2/05) Page 22 of 58

Monique Tomiczek and Kerry Maulden, EU COST meeting on practical methods for research on entomopathogenic nematodes and thrips, The Netherlands, January 2006.

Publications of the research results given by the research collaborators: Jude Bennison: ‘Using nematodes for pest control’, ADAS Bedding and Pot plant notes for

growers, June 2003. Bennison, J.A., Broadbent, A.B., Kirk, W.D.J. and Maulden, K.A. (2004). WFT pupation behaviour

on greenhouse chrysanthemums and implications for biological control. Abstracts of International Congress of Entomologists, Brisbane, August 2004.

Bennison, J.A. and Steiner, M. (2004). Biological control of thrips in greenhouses: the European and Australian experience. Abstracts of International Congress of Entomologists, Brisbane, August 2004.

Jude Bennison and the project consortium (2005). Controlling WFT and thrips-transmitted viruses in protected crops. Defra ‘Plant-IT’ Issue 7, p.2.

Jude Bennison, Monique Tomiczek and Kerry Maulden (2005). WFT behaviour helps improve bio-control. Grower 1 September 2005, 13-15.

Monique Tomiczek: Biological control of WFT, ADAS Bedding and Pot plant notes for growers, September 2005.

Options for new workThe project has identified key new information on WFT behaviour, e.g. synchronised larval dropping from host plants and pupation in growing media and substrates. This information, together with new information gained on the fate of S. feltiae and B. bassiana after application and their persistence and efficacy against WFT life stages on plants and in substrates, offers the opportunity to develop novel biopesticide application methods for optimum control of WFT and of other pests, as substitutes for chemical pesticides within IPM programmes. An expression of interest has been submitted to Defra PSD and the potential for Horticulture LINK funding is being discussed.

AcknowledgementsThanks to the following:

Becker Underwood for supplying S. feltiae and exchange of technical information on entomopathogenic nematodes.

Cleangro for supplying pot chrysanthemum cuttings. Fargro for supplying the Botanigard ES Koppert BV for supplying A. aphidimyza eggs. Australian Bioactive Compounds for supplying the T. stipitata extracts. Botanix for sourcing and extracting the T. lanceolata. Bayer CropScience for the deltamethrin used in the suction trap

References to published materialBennison, J. (2002). Final report to Defra on project HH1838SPC, Improving biological control of western flower thrips on chrysanthemums.

Bennison, J.A., Broadbent, A.B., Kirk, W.D.J., Maulden, K.A. and Shipp, J.L. (2004). Western flower thrips behaviour on greenhouse chrysanthemums and implications for integrated control. Abstract proceedings of the XXII International Congress of Entomology, 15-21 August 2004, Brisbane, Australia.

Broadbent, A.B., Rhainds, M., Shipp, L., Murphy, G. and Wainman, L. (2003). Pupation behaviour of western flower thrips (Thysanoptera: Thripidae) on potted chrysanthemum. The Canadian Entomologist 135, 741-744.

Buitenhuis, R. and Shipp, J.L. (2005). Efficacy of entomopathogenic nematode Steinernema feltiae (Rhabditita: Steinernematidae) as influenced by Frankliniella occidentalis (Thysanoptera: Thripidae) developmental stage and host plant stage. Journal of Economic Entomology 98, 1480-1485.

Chandler, D. & Davidson, G. (2006). Evaluation of the entomopathogenic fungus Metarhizium anisopliae against soil dwelling stages of the cabbage maggot, Delia radicum (L.) (Diptera: Anthomyiidae), in glasshouse and field experiments, and the effect of fungicides on fungal activity. Journal of Economic Entomology, 95, in press

Ebssa, L., Borgemeister, C., Berndt, O. & Poehling, H.-M. (2001). Effiacy of entomopathogenic nematodes against soil-dwelling life stages of western flower thrips, Frankliniella occidentalis

SID 5 (2/05) Page 23 of 58

(Thysanoptera: Thripidae). Journal of Invertegrate Pathology 78, 119-127.

Gaum, W.G., Giliomee, J.H. and Pringle, K.L. (1994). Life history and life tables of western flower thrips, Frankliniella occidentalis (Thysanoptera: Thripidae), on English cucumbers. Bulletin of Entomological Research 84, 219-224.

Gouge, D.H. and Hague, N,G.M. (1994). Control of sciarids in glass and propagation houses with Steinernema feltiae. Brighton Crop Protection Conference – Pests and Diseases – 1994, 1073-1078.

Helyer, N.L., Brobyn, P.J., Richardson, P.N. and Edmondson, R. (1995). Control of western flower thrips (Frankliniella occidentalis Pergande) pupae in compost. Annals of Applied Biology 127, 405-412.

Jacobson, R. (1997) Integrated Pest Management in glasshouses. In Thrips as Crop Pests (LEWIS, T., Ed.) CAB International, Wallingford, UK, p 644.

Jandricic, S., Murphy, G., Broadbent, B., Scott-Dupree, C. and Harris, R. (2005). Compatibility of Atheta coriaria with other biological control agents used in greenhouse production. IOBC/wprs Bulletin 28 (1), 135-138.

O’Leary, A.T. (2005). Flight and oviposition in the western flower thrips. PhD thesis, Keele University.

Robb, K.L. (1989). Analysis of Frankliniella occidentalis (Pergande) as a pest of floricultural crops in California greenhouses. PhD Thesis, Univ. California, Riverside, USA. XIV + 135pp.

Pearsall, I.A. (2002). Daily flight activity of the western flower thrips (Thysan., Thripidae) in nectarine orchards in British Colombia, Canada. Journal of Applied Entomology 126, 293-302.

Tashiro, H. (1967). Selfwatering acrylic cages for confining insects and mites on detached leaves. Journal of Economic Entomology 60, 254-356.

Appendix

SID 5 (2/05) Page 24 of 58

0123456789

10

1 2 3 4 5 6

Assessment

Mea

n no

. WFT

larv

ae p

er le

af TopMiddle

Bottom

Fig. 1 Mean numbers of WFT larvae per pot chrysanthemum leaf from the top, middle or bottom of the canopy on six weekly assessment dates, from the vegetative stage (assessments 1 and 2) to the bud stages (weeks 4 and 5) and the flowering stages (week 6).

0123456789

10

1 2 3 4 5 6

Assessment

Mea

n no

. WFT

larv

ae

Upper

Lower

Fig. 2 Mean numbers of WFT larvae on upper and lower surfaces of pot chrysanthemum leaves sampled from the top, middle and lower canopy at the vegetative stage (assessments 1 and 2), bud stages (assessments 3 and 4) and opening bud and flower stages (assessments 5 and 6).

0

1

2

3

4

5

6

1 2 3 4 5 6

Assessment

Mea

n no

. WFT

per

bud

/flow

er

AdultsLarave

SID 5 (2/05) Page 25 of 58

Fig. 3 Mean numbers of WFT adults and larvae per pot chrysanthemum leaf terminal (assessments 1 and 2), per bud (assessments 3 and 4) and per opening bud and flower (assessments 5 and 6).

0

50

100

150

200

250

1 2 3 4 5 6

Assessment number

Mea

n no

. WFT

per

leaf

TopMiddle

Bottom

Fig. 4 Mean numbers of WFT (all life stages) per cucumber leaf from the top, middle or bottom canopy on six weekly assessment dates.

050

100150200

250300350

400450

1 2 3 4 5 6

Assessment number

Mea

n nu

mbe

r WFT

upper

lower

Fig. 5 Mean numbers of WFT (all life stages) on upper and lower surfaces of cucumber leaves sampled from the top, middle and lower canopy on six weekly assessment dates.

Table 1. Mean numbers of WFT larvae per pot chrysanthemum and cucumber plant, mean numbers dropping to sticky traps on the ground over 24 hrs, and development rates of WFT larval and pupal stages on pot chrysanthemum and cucumber at the prevailing temperatures in the glasshouses over the 24-hr period

Pot chrysanthemum CucumberMean no. L2 dropping to ground to sticky traps over 24 hrs

61.5 3,424

Mean no. L1 and L2 per plant over 24 hrs

230 19,379

Mean no. prepupae and pupae per plant over 24 hrs

10.3 196

Mean duration of L1 and L2 stages at prevailing temperature

4.5 days 6.6 days

Mean duration of prepupal and pupal stages at prevailing temperature

3.7 days1 4 days2

SID 5 (2/05) Page 26 of 58

1 Mean temperatures in glasshouse with pot chryanthemums over the 24-hr period was 25.8C (range 22-34.6C). Nearest published data on WFT development rates on chrysanthemum is at fluctuating temperatures of 18.5-36C (Robb, 1989).

2 Mean temperatures in glasshouse with cucumbers over the 24-hr period was 22.3C (range 17.8-29.8). Nearest published data on WFT development rates on cucumber is at constant temperature of 23C (Gaum et al, 1994); no data on cucumber at fluctuating temperatures available.

Table 2. Calculations to quantify the percentage of WFT pupating on pot chrysanthemums and cucumber or on the ground beneath the two host plants.

Pot chrysanthemum CucumberL1 plus L2 on plant divided by duration time of L1 plus L2 on host plant at prevailing temperature = mean number of L2 which develop into pupae over the 24 hr period (L)

230 = 51.14.5

19379 = 29366.6

Prepupae plus pupae on plant divided by duration time of pupal stages on host plant at prevailing temperature = mean number of pupae which start development on the plant over the 24 hr period (P)

10.3 = 2.83.7

196 = 494

Mean % pupating on plants over the 24-hr period (P divided by L)

2.8 = 0.054 (5%) 51.1

49 = 0.017 (1.7%)2936

Mean % pupating on ground over the 24-hr period

95% 98.3%

Verfication: P divided by numbers of larvae dropping to sticky traps over the 24 hr period plus P = mean % pupating on plant

2.8 = 0.04 (4%) 61.5 + 2.8

= 96% pupating on ground

49 = 0.014 (1.4%)3424 + 49

= 98.6% pupating on ground

0 4 8 12 16 20 24

0

10

20

30

40

50

Mea

n tra

p ca

tch

(thrip

s h-1

)

Time (h BST/GMT)

Chr May 2003 (n=3) Chr June 2003a (n=4) Chr June 2003b (n=2) Chr Jan 2004 (n=3) Cu June 2003 (n=3)

Fig. 6 Mean diel flight curves each obtained over three days from Johnson-Taylor suction trap sampling in a pot chrysanthemum crop (Chr) and a cucumber crop (Cu). In the chrysanthemum crop, screens maintained a scotophase from 18.00 - 07.00 h.

SID 5 (2/05) Page 27 of 58

0 4 8 12 16 20 24

0.0

0.2

0.4

0.6

0.8

1.0

Time (h GMT)

Mea

n pr

opor

tion

of th

rips

mov

ing

Chr CT room adults Chr glasshouse adults Cu glasshouse larvae Cu glasshouse adults

Fig. 7. The proportion of WFT moving over a diel period in crops. Key: Chr = pot chrysanthemum, Cu = cucumber.

0 4 8 12 16 20 24

0

5

10

15

20

25

Males, no night light Females, no night light Females, low night light

Mea

n pe

rcen

tage

of t

ime

mov

ing

Time (h GMT)

Fig. 8. Mean diel movement curves for adult WFT males and females. The vertical lines indicate the transitions between photophase and scotophase (05.00 h, 21.00 h). The scotophase was either no light (0.0 Wm-2) or a low night-light (0.0125 Wm-2).

SID 5 (2/05) Page 28 of 58

6 12 18 24 6 12 18 24 6 12 18 24

0

1

2

3

4

5

6

7

8

Obs

erve

d tra

p ca

tch

(thrip

s h-1

)

Time (h GMT)

Observed

0.0

0.1

0.2

0.3

0.4

0.5

lightlightlight dark darkdarkdark

Expected

Exp

ecte

d tra

p ca

tch

inde

x

Fig. 9. The diel flight curves in a pot chrysanthemum crop from 11-14 January 2004 compared with the predicted flight curves from the flight activity model based on temperature and light. The vertical lines indicate the transitions between photophase and scotophase, maintained by automated screens.

0

10

20

30

40

50

60

Water Water + Agral Nemasys F + Agral

Treatment

% m

orta

lity

*

Fig.10. Mean % mortality of WFT L2s five days after treatment with water, water plus Agral, or Nemasys F at x3.5 recommended rate plus Agral.* Significantly different from other treatments, P<0.05

SID 5 (2/05) Page 29 of 58

0

20

40

60

80

1 2 5

Days after application

% W

FT L

2 m

orta

lity

Control

X 3.5

X 8

X 32

Fig. 11. Mean % mortality of WFT L2s one, two and five days after application with water plus Agral (control), Nemasys F at x3.5, x8 or x32 recommmended dose rate, all nematode treatments applied with Agral.No treatments significantly different from controls.

0

20

40

60

80

100

1 2 5

Days after application

% m

orta

lity

Control

x 3.5

x 32*

*

*

Fig. 12. Mean % mortality of WFT adult females one, two and five days after treatment with water plus Agral (control) and ‘Nemasys F’ applied at x3.5 and x32 recommended rates plus Agral.* significantly greater than in controls, P<0.05.

SID 5 (2/05) Page 30 of 58

0

5

10

15

20

25

30

35

40

45

Control BotaniGard 432.99 Naturalis 433.99

Treatment

% m

orta

lity

*****

***

Fig. 13. Mean % mortality of WFT L2s seven days after treatment with 0.01% Agral (control), Botanigard, isolate 432.99 cultured from Botanigard, Naturalis and isolate 433.99 cultured from Naturalis. All Beauveria treatments applied with 0.01% Agral.** significantly greater than in controls, p<0.01*** significantly greater than in controls, p<0.001

0

10

20

30

40

50

60

70

Control BotaniGard 432.99 Naturalis 433.99

Treatment

% m

orta

lity

*** ******

Fig. 14. Mean % mortality of WFT adult females seven days after treatment with 0.01% Agral (control), Botanigard, isolate 432.99 cultured from Botanigard and isolate 433.99 cultured from Naturalis. All Beauveria treatments applied with 0.01% Agral.

SID 5 (2/05) Page 31 of 58

Table 3. Proportion of WFT cadavers supporting fungal mycelium (mycosis) in laboratory bioassays with Beauveria treatments against WFT second instar larvae (L2s) and adults.

Treatment WFT L2s WFT adults% mycosis s.e. % mycosis s.e.

control 0 0 0 0Botanigard 0 0 61.3 5.1432.99 4.8 3.7 39.6 5.1Naturalis 0 0433,99 1.2 1.2 26.5 4.9

010

20

30

40

50

6070

80

90

1 2 3

Experiment

Mea

n %

ass

umed

mor

talit

y +

SE

Water only

Recommended ratenematodes2 x recommendedrate nematodes8 x recommendedrate nematodes

Fig. 15. Mean assumed % mortality of WFT ground-dwelling stages in three compost bioassays, 14 days after treatment with Nemasys F at recommended, x2 and x8 rates. None of the treatments gave significantly greater mortality than in water controls.

SID 5 (2/05) Page 32 of 58

0

50

100

0 1 2 3 17 24

Hours after nematode application

% A

ctiv

e ne

mat

odes

per

leaf

sam

ple

top

upper

lower

bottom

Fig. 16. % active nematodes on leaf samples taken from the top, upper middle, lower middle and bottom canopy of cucumber plants, at different time intervals after Nemasys F application.

0

25

50

75

100

125

0 1 2 16 24

Hours after nematode application

% A

ctiv

e ne

mat

odes

per

leaf

/bud top

middle

bottomgreenbud

Fig. 17. % active nematodes on leaf samples taken on leaf samples from top, middle and bottom canopy and on buds at the ‘green bud’ stage of pot chrysanthemum plants

SID 5 (2/05) Page 33 of 58

0

25

50

75

100

125

0 1 3 5

Hours after nematode application

% A

ctiv

e ne

mat

odes

per

leaf

/bud

top

middlebottom

bud

Fig. 18. % active nematodes on leaf samples taken on leaf samples from top, middle and bottom canopy and on buds at the ‘bud break’ stage of pot chrysanthemum plants

0

50

100

0 1 2 3Hours after nematode application

% a

ctiv

e ne

mat

odes

Full flower

Opening Bud

Fig. 19. % active nematodes taken from opening bud and flower samples of pot chrysanthemum plants

SID 5 (2/05) Page 34 of 58

0

50

100

0 1 2

Hours after nematode application

% a

ctiv

e ne

mat

odes

per

flow

erWithout water

With water

**

Fig, 20. % active nematodes taken from opening bud and flower samples of pot chrysanthemum plants, with or without a spray of water applied half an hour after nematode application* significantly more active nematodes than without water, p<0.05

0

200

400

600

800

0 16 20 24 40 48 72

Hours after application

No. c

olon

ies

NaturalisBotanigard

432.99433.99

Fig. 21. Mean number of Beauveria bassiana colonies per plate from pot chrysanthemum leaf discs

SID 5 (2/05) Page 35 of 58

Fig. 22. Mean number of Beauveria bassiana colonies per plate from pot chrysanthemum opening buds

Fig. 23. Mean numbers of active S. feltiae per cm2 compost over the 8-week period following two weekly applications of Nemasys F at the rate of 200 per cm2

SID 5 (2/05) Page 36 of 58

0

50

100

150

200

250

6 8 10

Nematode application times (weeks after cuttings stuck)

Estim

ated

num

ber n

emat

odes

per

cm

2 com

post

Plant

Compost

Plant & compost

Fig. 24. Pot chrysanthemums cage Experiment 1: Mean numbers of active S. feltiae per cm2

compost when sampled from compost immediately before Nemasys F applications (7 days after previous application) at the rate of 50 per cm2, 6, 8 and 10 weeks after pot chrysanthemum cuttings were stuck.

0

25

50

75

6 8 10

Nematode application times (weeks after cuttings stuck)

Num

ber o

f nem

atod

es re

achi

ng a

cm

2 of c

ompo

st

Plant

Compost

Plant & compost

Fig. 25. Pot chrysanthemums cage Experiment 1: Mean numbers of S. feltiae reaching cm2

compost when collected from tubes inserted into compost immediately after Nemasys F applications at the rate of 50 per cm2, 6, 8 and 10 weeks after pot chrysanthemum cuttings were stuck.

Table 4. Pot chrysanthemums cage Experiment 1: mean % leaves and flowers damaged, mean numbers of WFT per cage and mean % marketable plants treated with Nemasys F to plants only, to

SID 5 (2/05) Page 37 of 58

compost only or to plants and compost, compared with the water control. *** significantly less than in controls, (P<0.001). * significantly less than in the other treatment, (P<0.05). Treatments with the same letter are statistically similar, treatments with different letters are significantly differerent.

Treatment Mean %leaves damaged

Mean % flowers damaged

Mean %marketable plants

Mean no. WFT per cage (all life stages)

Back-transformed (log10) mean no. WFT per cage

1. Control (water to plant & compost)

31.9 79 a 0 42.233.8

(mean of treatments 1 &

2) 2. Nematodes to plant only

12.9 74 a 0 37.7

3. Nematodes to compost only

15.4 30*** b 8 18 10.7* (mean of

treatments 3 & 4)4. Nematodes

to plant & compost

6.2 7*** c 33 10.3

0

10

20

30

40

Mea

n ba

ck-tr

ansf

orm

ed n

umbe

r of W

FT p

er

cage

Nemasys F or water toplants

Nemasys F to compost

*

Fig. 26. Pot chrysanthemum cage Experiment 1. Mean back-transformed numbers of WFT per cage, with plants treated with either Nemasys F or water, or where Nemasys F was allowed to reach the compost. * significantly fewer WFT than in the other treatment.

SID 5 (2/05) Page 38 of 58

Fig. 27. Pot chrysanthemums cage Experiment 2: Mean numbers of active S. feltiae per cm2

compost when sampled from compost immediately before Nemasys F applications (7 days after previous application) at the rate of 50 per cm2, 6, 9 and 11 weeks after pot chrysanthemum cuttings were stuck.

0

20

40

60

80

6 9

Nematode application (weeks after cuttings stuck)

Num

ber n

emat

odes

reac

hing

1c

m2

com

post

Plant

CompostPlant & Compost

Fig. 28. Pot chrysanthemums cage Experiment 2: Mean numbers of S. feltiae reaching cm2

compost when collected from tubes inserted into compost immediately after Nemasys F applications at the rate of 50 per cm2, 6 and 9 weeks after pot chrysanthemum cuttings were stuck.

SID 5 (2/05) Page 39 of 58

Table 5. Pot chrysanthemums cage Experiment 2: mean % leaves and flowers damaged, mean % petal area damaged, mean numbers of WFT per cage and mean % marketable plants treated with Nemasys F to plants only, to compost only or to plants and compost, compared with the water control. *, ** and ***significantly less than in controls, (P<0.05, P<0.01 and P<0.001 respectively). Treatments with the same letter are statistically similar, treatments with different letters are significantly different.

Treatment Mean % leaves damaged

Mean % flowersdamaged

Mean % petal area damaged

Mean %Marketable plants

Back-transformed mean no. WFT per cage (all life stages)

1. Control (water to plant & compost)

76.7 a 97% a 43.8 a 0 250.2 a

2. Nematodes to plant only

20.4*** c 96.4% a 42.9 a 0 135.8 ab

3. Nematodes to compost only

37.9** b 79.2% a 38.8 a 0 60.9* b

4. Nematodes to plant & compost

2.6*** d 25%** b 2.1*** b 42 14.6** c

0

50

100

150

200

250

300

Control (waterplus Agral to

plant andcompost)

Nematodesplus Agral to

plant only

Nematodesplus Agral tocompost only

Nematodesplus Agral to

plant andcompost

Back

-tra

nsfo

rmed

mea

n nu

mbe

r WFT

per

ca

ge (a

ll lif

e st

ages

)

a

ab

* b

** c

Fig. 29. Pot chrysanthemum cage Experiment 2. Mean back-transformed numbers of WFT per

SID 5 (2/05) Page 40 of 58

cage, with plants treated with water, Nemasys F to plants only, to compost only or to both plants and compost. * and ** significantly less than in controls, (P<0.05 and P<0.01 respectively). Treatments with the same letter are statistically similar, treatments with different letters are significantly different.

0

1

2

3

4

6 13 20 27

Days after application

Mea

n W

FT p

er le

af Control

Nemasys F (compostand matting)

Nemasys F (leaves)*

* *

*

*

Fig. 30. Mean WFT per cucumber leaf, on untreated control plants and where Nemasys F was applied to either the leaves or the compost and capillary matting on the glasshouse floor.* significantly fewer WFT than in untreated controls (P<0.05).

0

50

100

150

200

250

300

0 6 14 21

Application day

Num

ber n

emat

odes

per

cm

2

leaf

matting

compost / 10

Fig. 31. Mean numbers of active S. feltiae per cm2 on cucumber leaf, capillary matting and compost samples immediately after each application date. Numbers in compost samples were x10 higher than shown on scale, thus numbers/10 are presented.

SID 5 (2/05) Page 41 of 58

0

50

100

150

200

250

300

0+6 6+7 14+6 21+6

Post-application sample day

Num

ber o

f nem

atod

es p

er c

m2

leaf

matting

compost

Fig. 32. Mean numbers of active S. feltiae per cm2 on cucumber leaf, capillary matting and compost samples 6 or 7 days after each application date. Actual numbers in compost samples presented.

Comparison of adult WFT behaviour on untreated cucumber leaf

0

10

20

30

40

50

60

70

80

w alking still

P<0.05 for time spent walking and still

% ti

me

spen

t

control females

control males

Fig. 33. % time spent walking or being still by WFT adult males and females on untreated cucumber leaves.

SID 5 (2/05) Page 42 of 58

Adult male WFT on antifeedant treated cucumber leaf

0

10

20

3040

50

60

70

80

still w alking

P<0.05 for antifeedant v control

% ti

me

spen

tcontrol T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 34. % time spent walking or being still by WFT adult males on untreated or antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment.

Adult female WFT on antifeedant treated cucumber leaf

0

10

20

30

40

50

60

70

still w alking

P<0.05 for antifeedant except after 48h

% ti

me

spen

t

control T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 35. % time spent walking or being still by WFT adult females on untreated or antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment.

SID 5 (2/05) Page 43 of 58

Time spent by adult WFT on antifeedant treated cucumber leaf

0102030405060708090

100

females males

P<0.05 for antifeedant except for females at 48h

% ti

me

spen

t/20m

incontrol T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 36. % time spent by adult WFT on untreated or antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment.

Flight behaviour of adult WFT on antifeedant treated cucumber leaf

00.5

11.5

22.5

33.5

44.5

females males

P<0.05 for antifeedant except at 48h

mea

n no

. flig

hts/

20m

in

control T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 37. Mean number of flights per 20 min by adult WFT on untreated or antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment.

SID 5 (2/05) Page 44 of 58

Second instar larvae (L2) of WFT on antifeedant treated cucumber leaf

0

10

20

30

40

50

60

70

80

still w alking

P<0.05 for antifeedant

% ti

me

spen

tcontrol T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 38. % time spent walking or being still by WFT L2s on untreated or antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment.

Time spent by L2 of WFT on antifeedant treated cucumber leaf

0

20

40

60

80

100

120

on leaf

P<0.05 for antifeedant except at 48h

% ti

me

spen

t

control T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 39. % time spent by WFT L2s on untreated or antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment.

SID 5 (2/05) Page 45 of 58

Adult male WFT on antifeedant treated chrysanthemum leaf

0

10

20

30

40

50

60

70

still w alking

Not Significant

% ti

me

spen

tcontrol T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 40. % time spent walking or being still by WFT adult males on untreated or antifeedant-treated chrysanthemum leaves 0, 24 and 48 h after treatment.

Adult female WFT on antifeedant treated chrysanthemum leaf

0

10

20

30

40

50

60

70

80

still w alking

P<0.05 for antifeedant except at 48h

% ti

me

spen

t

control T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 41. % time spent walking or being still by WFT adult females on untreated or antifeedant-treated chrysanthemum leaves 0, 24 and 48 h after treatment.

SID 5 (2/05) Page 46 of 58

Time spent by adult WFT on antifeedant treated chrysanthemum leaf

0

10

20

30

40

50

60

70

80

females males

P<0.05 for antifeedant except at 48h for females

% ti

me

spen

tcontrol T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 42. % time spent by WFT adults on untreated or antifeedant-treated chrysanthemum leaves 0, 24 and 48 h after treatment.

Flight behaviour of adult WFT on antifeedant treated chrysanthemum leaf

0

1

2

3

4

5

6

7

females males

P<0.05 forantifeedant except at 48h for females and 24 and 48h for males

mea

n no

. flig

hts/

20 m

in

control T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 43. Mean number of flights per 20 mins by WFT adults on untreated and antifeedant-treated chrysanthemum leaves 0, 24 and 48 h after treatment.

SID 5 (2/05) Page 47 of 58

L2 on antifeedant treated chrysanthemum leaf

0

10

20

30

40

50

60

70

still w alking

Not significant

% ti

me

spen

tcontrol T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 44. Mean % time spent walking or being still by WFT L2s on untreated and antifeedant-treated chrysanthemum leaves 0, 24 and 48 h after treatment.

Time spent by L2 on antifeedant treated chrysanthemum leaf

0102030405060708090

on leaf

P<0.05 for antifeedant

% ti

me

spen

t

control T. stipitata T. stipitata 24h T. stipitata 48h

Fig. 45. Mean % time spent by WFT L2s on untreated and antifeedant-treated chrysanthemum leaves 0, 24 and 48 h after treatment.

SID 5 (2/05) Page 48 of 58

Fig. 46. GC trace of T. lanceolata extract showing polygodial content

Adult female WFT on antifeedant treated cucumber leaf

0

10

20

30

40

50

60

70

80

still w alking

application by spinning disc; still P<0.05 for polygodial v controls and for T. lanceolata v ethanol

% ti

me

spen

t

control

aqueous ethanol

polygodial

polygodial 24h

polygodial 48h

T. lanceolata

T. lanceolata 24h

T. lanceolata 48h

Fig. 47. Mean % time spent walking or being still by WFT females on untreated and antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment application by spinning disc.

SID 5 (2/05) Page 49 of 58

Adult male WFT on antifeedant treated cucumber leaf

0

10

20

30

40

50

60

70

80

still w alking

application by spinning disc; No significant differences

% ti

me

spen

tcontrol

aqueous ethanol

polygodial

polygodial 24h

polygodial 48h

T. lanceolata

T. lanceolata 24h

T. lanceolata 48h

Fig. 48. Mean % time spent walking or being still by WFT males on untreated and antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment application by spinning disc.

Time spent by adult WFT on antifeedant treated cucumber leaf

0

20

40

60

80

100

120

females males

application by spinning disc; females P<0.05 for ethanol v polygodial

% ti

me

spen

t/20

min

control

aqueous ethanol

polygodial

polygodial 24h

polygodial 48h

T. lanceolata

T. lanceolata 24h

T. lanceolata 48h

Fig. 49. Mean % time spent by WFT adults on untreated and antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment application by spinning disc.

SID 5 (2/05) Page 50 of 58

Adult female WFT on antifeedant treated cucumber leaf

0

10

20

30

40

50

60

70

still w alking

spray to run off; still P<0.05 immediately after treatment

% ti

me

spen

tcontrol T. lanceolata T. lanceolata 24h T. lanceolata 48h

Fig. 50. Mean % time spent walking or being still by WFT females on untreated and antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment application to run-off.

Adult male WFT on antifeedant treated cucumber leaf

0

10

20

30

40

50

60

70

80

still w alking

spray to run off; still P<0.05 except at 24h

% ti

me

spen

t control

T. lanceolata

T. lanceolata 24h

T. lanceolata 48h

Fig. 51. Mean % time spent walking or being still by WFT males on untreated and antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment application to run-off.

SID 5 (2/05) Page 51 of 58

Time spent by adult WFT on antifeedant treated cucumber leaf

0

20

40

60

80

100

120

females males

Spray to run off; males P<0.05 for T. lanceolata

% ti

me

spen

tcontrol T. lanceolata T. lanceolata 24h T. lanceolata 48h

Fig. 52. Mean % time spent by WFT adults on untreated and antifeedant-treated cucumber leaves 0, 24 and 48 h after treatment application to run-off.

Petri dish assay of the effects of Tasmannia lanceolata extract on Steinernema feltiae nematodes after 6h

0

20

40

60

80

100

120

Sub-sample ofnematode mix

Nematodesalone

Nematodes with3000ppm T.lanceolata

Nematodes with6000ppm T.lanceolata

no significant differences

mea

n no

. nem

atod

es/d

ish

Dead

Morbid

Live

Fig. 53. Mean numbers of live, dead and moribund S. feltiae 6h after mixing with the T. lanceolata antifeedant at two concentrations or with a water plus ethanol control.

Table 6. Effect of antifeedant extracted from Tasmania lanceolata on the germination of conidia of B. bassiana, used as BotaniGard WP and ES

Treatment Mean % germination (s.e.)Botanigard ES 82 4 (3.23)Botanigard WP 80.1 (1.16)Botanigard ES: tank mix with antifeedant 0Botanigard WP: tank mix with antifeedant 0Botanigard ES: antifeedant in agar 0Botanigard WP: antifeedant in agar 0

SID 5 (2/05) Page 52 of 58

Table 7. Effect of antifeedant extracted from Tasmania lanceolata on the virulence of Botanigard ES to adult female WFT in a laboratory bioassay

Treatment Mean % WFT mortality at 7 dpost-inoculation (s.e.)

Mean % mycosis (s.e.)

Control 15.3 (4.68) 0Antifeedant 22.7 (7.10) 0Botanigard ES 56.2 (13.64) 30.7 (11.65)Antifeedant + Botanigard (tank mix)

30.4 (5.81) 0

Antifeedant then Botanigard 37.1 (5.86) 0

WFT on chrysanthemum leaf samples from glass house trials

0

1

2

3

4

5

6

7

8

Water blank formulationethanol+agral

Nematodes+agral Nematodes & T.lanceolata

T. lanceolata+ethanol+agral

mea

n no

. wft/

24 le

aves

L2 L1 ADULT total

Fig. 54. Mean number of WFT per 24 leaves on leaf samples from pot chrysanthemums treated with water or ethanol + Agral controls, Nemasys F with Agral, Nemasys F with the T. lanceolata antifeedant or with the antifeedant alone (no significant treatment effect).

SID 5 (2/05) Page 53 of 58

WFT on chrysanthemum bud samples from glass house trials

0

5

10

15

20

25

30

35

40

45

Water blank formulationethanol+agral

Nematodes+agral Nematodes & T.lanceolata

T. lanceolata+ethanol+agral

T. lanceolata treatments P<0.05 for L2 and for total

mea

n no

. wft/

24 b

uds

L2 L1 ADULT total

Fig. 55. Mean number of WFT per 24 buds on bud samples from pot chrysanthemums treated with water or ethanol + Agral controls, Nemasys F with Agral, Nemasys F with the T. lanceolata antifeedant or with the antifeedant alone.

WFT beaten off chrysanthemum plants in glass house trials

0

10

20

30

40

50

60

Water blank formulationethanol+agral

Nematodes+agral Nematodes & T.lanceolata

T. lanceolata+ethanol+agral

P<0.05 for T. lanceolata and T. lanceolata + nematodes

mea

n no

./4 p

ots

Fig. 56. Mean number of WFT per four pots of chrysanthemums treated with water or ethanol + Agral controls, Nemasys F with Agral, Nemasys F with the T. lanceolata antifeedant or with the antifeedant alone.

SID 5 (2/05) Page 54 of 58

Total WFT on chrysanthemums in glass house trials

0

20

40

60

80

100

120

Water blank formulationethanol+agral

Nematodes+agral Nematodes & T.lanceolata

T. lanceolata+ethanol+agral

P<0.05 for T. lanceolata and T. lanceolata + nematodes

mea

n no

./4 p

ots

Fig. 57. Mean number of WFT per four pots of chrysanthemums from all glasshouse experiments where plants were treated with water or ethanol + Agral controls, Nemasys F with Agral, Nemasys F with the T. lanceolata antifeedant or with the antifeedant alone.

Mean time spent by Aphidius ervi on treated cucmber plants

0

1

2

3

4

5

6

7

8

untreated ethanol polygodial T.lanceolata

Not significant

min

utes

Fig. 58. Mean time spent by Aphidius ervi on untreated cucumber plants or those treated with ethanol, polygodial or the T. lanceolata antifeedant.

SID 5 (2/05) Page 55 of 58

Behaviour of Aphidius ervi on treated cucumber plants

0

10

20

30

40

50

60

70

80

untreated ethanol polygodial T.lanceolata

Not significant except cleaning P<0.05 for ethanol v T. lanceolata

% ti

me

spen

t

w alking

still

cleaning

Fig. 59. Mean % time spent walking, cleaning or being still by Aphidius ervi on untreated cucumber plants or those treated with ethanol, polygodial or the T. lanceolata antifeedant.

Aphis gossypii settling test: spinning disc spray on cucumber Trial 1

0123456789

10

control ethanol T. lanceolata polygodial

n=10; polygodial significantly different from control and ethanol at 5h P<0.05

mea

n no

. of a

phid

s se

ttled

2 hr

5 hr

24 hr

Fig. 60. Mean number of Aphis gossypii settled on untreated or ethanol-treated cucumber plants and on those treated with T. lanceolata or polygodial antifeedants by spnning disc sprayer, Trial 1.

SID 5 (2/05) Page 56 of 58

Aphis gossypii settling test: spinning disc spray on cucumber Trial 2

0123456789

10

control ethanol T. lanceolata polygodial

n=10; Not Significant

mea

n no

. of a

phid

s se

ttled

2 hr

5 hr24 hr

Fig. 61. Mean number of Aphis gossypii settled on untreated or ethanol-treated cucumber plants and on those treated with T. lanceolata or polygodial antifeedants by spnning disc sprayer, Trial 2.

Aphis gossypii settling test: spray to run off on cucumber Trial 3

0123456789

10

control ethanol T. lanceolata T. stipitata

n=10; Treatments significantly different from control and ethanol P<0.05

mea

n no

. of a

phid

s se

ttled

2 hr

5 hr

24 hr

Fig. 62. Mean number of Aphis gossypii settled on untreated or ethanol-treated cucumber plants and on those treated with T. lanceolata or T. stipitata antifeedants by high volume spray to run-off.

References to published material9. This section should be used to record links (hypertext links where possible) or references to other

published material generated by, or relating to this project.

SID 5 (2/05) Page 57 of 58

See end of report, before Appendices.

SID 5 (2/05) Page 58 of 58