General enquiries on this form should be made to: Defra, Procurements and Contracts Division (Science R&D Team) Telephone No. 0207 238 5734 E-mail: [email protected]

SID 5 Research Project Final Report

 Note In line with the Freedom of Information Act 2000, Defra aims to place the results Project identification of its completed research projects in the public domain wherever possible. The PS2120 SID 5 (Research Project Final Report) is 1. Defra Project code designed to capture the information on the results and outputs of Defra-funded 2. Project title research in a format that is easily Novel control of soil dwelling horticultural pests using the publishable through the Defra website. A staphylinid beetle Atheta coriaria SID 5 must be completed for all projects.  This form is in Word format and the boxes may be expanded or reduced, as 3. Contractor appropriate. organisation(s) ADAS

 ACCESS TO INFORMATION

The 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 54. Total Defra project costs £ 155,000 project. Defra may also disclose the (agreed fixed price) information to any outside organisation acting as an agent authorised by Defra to 01 April 2007 process final research reports on its 5. Project: start date ...... behalf. Defra intends to publish this form

on its website, unless there are strong end date ...... 31 March 2010 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.

SID 5 (Rev. 05/09) Page 1 of 20 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 Summary 7. 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. Control of the major pests cabbage root (CRF) and fly on conventionally grown horticultural brassicas and respectively is currently reliant on pesticides. For CRF control, young plants raised in modules in propagation glasshouses are drenched with chlorpyrifos before planting in the field but the approval for this use is being withdrawn. The alternative modular drench, spinosad is unpopular with growers due to its high cost. Pyrethroid sprays are widely used in carrot crops against adult carrot . This project aimed to evaluate the potential of the native predatory staphylinid beetle Atheta coriaria, for biological control of both CRF and carrot fly. A. coriaria is commercially available for control of sciarid and shore flies in protected crops but use of high release rates is expensive. HDC-funded research by ADAS developed a low-cost method for growers to rear their own A. coriaria using turkey-rearing feed as an artifical diet. This method offers opportunities for inundative release of the predators for control of various pests with ground-dwelling life stages, including those of field crops. This Defra-funded project aimed to encourage grower uptake of biological control methods on outdoor horticultural crops and to reduce pesticide usage, in line with the aims of CRD’s Alternative Crop Protection Technologies programme.

Objective 1: confirm that Atheta coriaria predate the eggs and young larvae of CRF and carrot fly Replicated 24-hr laboratory bioassays confirmed that individual A. coriaria adults ate means of up to 20 CRF eggs when offered 24 on damp filter paper in Petri dishes. Fewer eggs were eaten (a mean of 7 per day) when 20 eggs were offered on damp compost and even fewer were eaten (a mean of 3 per day) when offered on damp compost mixed with 5% turkey crumbs as an alternative food source. A. coriaria adults ate up to a mean of 12 young CRF larvae and A. coriaria larvae ate a mean of 10 CRF eggs when offered 16 on filter paper. Bioassays with carrot fly eggs were limited by available eggs. A. coriaria adults ate all eggs when offered 8 on filter paper. As carrot fly eggs are smaller than those of CRF, it is likely that predation rates on carrot fly eggs will be at least as high as on CRF eggs. These results were used to guide A. coriaria release rates in subsequent semi-field and field experiments done in Objective 2.

Objective 2: evaluate the efficacy and practicality of using the A. coriaria rearing-release system against CRF and carrot fly on crops in semi-field or commercial field conditions

Atheta coriaria development rates in brassica compost Glasshouse experiments were done to compare the development and multiplication rates of A. coriaria in brassica compost in replicated plastic boxes with turkey crumbs as food at a range of temperatures consistent with those in unheated glasshouses used for brassica propagation. During a typical 6-week

SID 5 (Rev. 05/09) Page 2 of 20 propagation period in summer, A. coriaria multiplied by around x14. However, in the cooler winter/early spring during a typical 9-week propagation period, the predators bred more slowly and multiplied by only x2. This indicated that to establish A. coriaria in the modules during propagation, many more would need to be added in the early spring period to achieve the required number of predators per module at planting out. Similar experiments with replicate rearing boxes maintained in incubators at constant low temperatures indicated that at 15°C, mean A. coriaria population increases were x5 and x10 after six and 12 weeks respectively, but at 10°C, adult predators survived for four weeks but no offspring developed.

Atheta coriaria flight behaviour A glasshouse experiment showed that A. coriaria adults will fly and disperse in unheated propagation glasshouses in February and March, when plants are raised for planting out in April, when first generation CRF egg laying usually starts. A. coriaria rearing-release boxes (with holes in the lids to allow predator escape) were placed in replicated mesh -proof cages in a research glasshouse. Assessment of A. coriaria on sticky traps and in 'bait pots' (pots of compost mixed with turkey crumbs) in the cages confirmed that the predators flew and dispersed when mean air temperatures were 11-14°C. The results also indicated that A. coriaria may fly and disperse in similar temperatures occurring in the field.

Atheta coriaria establishment in brassica plugs An experiment in research glasshouses during August and September 2007 tested methods for establishing A. coriaria in brassica modules during propagation so that predator-'seeded' modules could be planted in the field. Module trays were filled with plain compost (untreated control) or compost mixed with 1%, 5%, 10% or 25% turkey crumbs, to provide different amounts of food for the predators. After seeding with cauliflower seeds, A. coriaria were either broadcast over the modules or allowed to colonise them independently after leaving rearing-release boxes. After six weeks, the highest numbers of A. coriaria (10.4 per module) were recorded in those with 25% turkey crumbs, but this mix and the 10% mix inhibited seed germination and seedling survival. Very few predators established in the modules with plain compost or with 1% turkey crumbs. Compost with 5% turkey crumbs was the most suitable mix, allowing normal seed germination and leading to 75% of the modules containing A. coriaria, with a mean of 2.5 per module. Using rearing-release boxes led to higher numbers of predators per module than using the broadcasting method. These results were used to guide the subsequent semi-field efficacy experiment.

Semi-field efficacy experiment, CRF on brassicas A semi-field experiment during July and August 2008 tested whether A. coriaria, either raised in cauliflower modules during propagation or added after planting, could reduce the severity of CRF damage. Replicate young cauliflower plants were planted into pots for each of five treatments: 1) untreated control with no A. coriaria; 2) chlorpyrifos drench to the modules before planting; 3) A. coriaria raised with the modules with a mean of 2 adults and 2 larvae per pot at planting; 4) A. coriaria added after planting at 2 adults and 2 larvae per pot; 5) A. coriaria added after planting at 10 adults and 10 larvae per pot. The pots were placed outdoors in individual mesh insect-proof cages and 10 CRF eggs per pot were added to the soil immediately after planting and again one week later. After six weeks, root damage was assessed. A. coriaria added at the higher rate after planting was almost as effective as the chlorpyrifos treatment, reducing mean % Root Damage Index (RDI) to 5%, from 61% in untreated plants. A. coriaria established in the modules during propagation also reduced the RDI to a mean of 28%. At the end of the experiment, A. coriaria were found in soil in pots treated with all predator treatments, with significantly more (20 per pot) recorded in pots treated with the higher rate after planting. The results of this experiment were used to guide the treatments in the subsequent field experiment in a commercial cauliflower crop.

Semi-field efficacy experiment, carrots A similar semi-field cage experiment to that done with CRF was done during July to October 2009 to test whether A. coriaria could reduce carrot fly damage to carrots. Replicate pots of carrots were infested with five carrot fly eggs per pot and used to test two treatments: 1) untreated control and 2) A. coriaria at 10 adults and 10 larvae per pot. Results were inconclusive as no carrot fly damage was recorded in either treatment after 10 weeks. The eggs were assumed to be viable as all the extra eggs incubated in the laboratory at the time of carrot infestation successfully hatched.

CRF field experiment on a commercial farm A field experiment in a commercial cauliflower crop during May and June 2009 tested whether A. coriaria, either raised in cauliflower modules or added after planting, could reduce the severity of first generation CRF damage. Young cauliflower plants were raised in plain compost at a propagation nursery between 12 March and 6 May. To produce the A. coriaria in the modules during propagation, high numbers were broadcast over the modules together with a small amount of turkey crumbs. Rearing-release boxes were also used with the aim of establishing at least five adults and five larvae per module at planting. The experiment tested six replicated treatments: 1) untreated control; 2) chlorpyrifos drench to modules; 3)

SID 5 (Rev. 05/09) Page 3 of 20 spinosad drench to modules; 4) A. coriaria raised with modules; 5) A. coriaria added after planting at 10 adults and 10 larvae per plant; 6) A. coriaria added after planting as in treatment 5 and again one week later. The large plots were separated by similar sized 'buffer' plots with chlorpyrifos-treated plants to reduce the risk of A. coriaria moving to plots with other treatments. Although planting was done on 6 May when peak first generation egg laying had been forecast, CRF egg laying pressure during the experiment was relatively light and prolonged, peaking at only two eggs per plant per day in early June. There was no treatment effect on root damage to plants surviving the 8-week experiment, with mean RDI's in all treatments being below 40%. The main effect of CRF on untreated plants was the death of 12.5% of young plants checked by dry soil conditions at and during the first week after planting. A. coriaria added at planting out significantly reduced the mean percentage of dead plants (to 2.5%), as did modular drenches of chlorpyrifos and spinosad (to 0%). However, A. coriaria added twice after planting did not significantly reduce plant death (5%) and this result cannot be explained. A. coriaria raised with the modules did not reduce plant death (5%) but the cold spring had led to poor predator establishment during propagation, with means of only 0.6 adults and 0.4 larvae per module achieved at planting. Pitfall trapping during the experiment showed that numbers of A. coriaria were higher (up to a mean of 10.6 per trap) one week after release in plots where the beetles had been broadcast than in all other treatments. However, A. coriaria had dispersed equally to all plots within two weeks of planting. This could have been partly due to the dry soil conditions early in the experiment which would not have been favourable for the predators.

Objective 3: communicate the results and demonstrate the system to the industry Knowledge transfer activities included six presentations at industry conferences and meetings, three presentations at scientific conferences, two scientific papers and regular communications with key field vegetable growers and brassica propagators.

Options for new work The project has demonstrated the potential of A. coriaria for biological control of CRF. Further fundamental research is needed to quantify how many A. coriaria are needed to give sufficient and reliable control of CRF at a range of egg-laying pressures. Development work would then be needed on scaling up the A. coriaria rearing system for cost-effective use by propagators on a commercial scale and on practical and effective commercial release methods in the field.

Project Report to Defra 8. 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).

This project aimed to evaluate the potential for extending the use of the predatory staphylinid beetle Atheta coriaria, as a biological control agent for two key pests of field vegetable crops, cabbage root fly (CRF) and carrot fly. This predator is commercially available for use in protected crops for biological control of the ground-dwelling life stages of sciarid and shore flies. However, it is not widely used as high numbers need to be released for effective control of high pest densities and such releases are expensive. HDC-funded research (PC 239 and PC 239a), developed a ‘DIY’ rearing-release system, for growers to rear and maintain large numbers of the beetles in glasshouse crops, for low-cost biological control of sciarid and shore flies. This project built on the results of the HDC-funded research. The intended outcome of the project is to encourage increased grower uptake of biological control methods on outdoor horticultural crops and to reduce pesticide usage, in line with the aims of CRD’s Alternative Crop Protection Technologies programme.

SID 5 (Rev. 05/09) Page 4 of 20 The objectives of the project were:

1. Confirm that A. coriaria predate the eggs and young larvae of cabbage root fly and carrot fly. 2. Evaluate the efficacy and practicality of using an existing low cost/high output A. coriaria rearing-release system against the selected pests within realistic ICM programmes, in experiments on crops in near-commercial or commercial conditions. 3. Communicate the results and demonstrate the system to the industry.

Objective 1: confirm that Atheta coriaria predate the eggs and young larvae of cabbage root fly and carrot fly

Methods

Rearing method for Atheta coriaria Laboratory cultures of A. coriaria were established and maintained to provide adults and larvae for the predation studies and for other experiments in the project. The method was similar to that developed in Canada by Carney et al. (2002), but modified to use turkey-rearing crumbs (Dodson & Horrell Ltd., UK) as a food source rather than ground trout pellets (Bennison et al 2008). Research in HDC project PC 239 developed a ‘DIY’ rearing-release system for growers to rear and maintain large numbers of the beetles in glasshouses, for control of the ground- dwelling life stages of sciarid and shore flies (Bennison 2007, 2010). The original beetles for the ADAS culture were obtained from Syngenta Bioline Ltd. who are one of the commercial suppliers of A. coriaria and who were a contributor to HDC project PC 239. Beetles were reared in 3-litre plastic boxes containing 1.5 litres of dampened coir and vermiculite substrate. The boxes had snap-on lids, fitted with two ventilation holes covered with insect- proof mesh. Sixty A. coriaria adults were added to each box and every week, 5g turkey crumbs and water as required were added and incorporated into the substrate. The boxes were maintained at 25C with a 16:8 light:dark photoperiod for approximately 28 days. Under these conditions, the next generation of adult A. coriaria normally emerge after 19-23 days, and the mean population growth is x 20, i.e. 60 adults lead to a mean of around 1,200 mixed adults and larvae after 23 days (Bennison 2007). Adults taken from the cultures 23-28 days after the boxes had been set up (for use in predation studies and for setting up new boxes to maintain the culture) were assumed to be new generation adults as adult longevity in these conditions is no more than 21 days (Carney et al 2002).

Rearing method for CRF Laboratory cultures of CRF were established and maintained using methods similar to those described by Finch and Coaker (1969). CRF eggs were collected in soil from around the base of cauliflower stems in a commercial crop and were washed around the edges of whole swedes that had been placed in wet sand so that approximately one third of the swede was above the level of the sand. The infested swedes were placed into an insect rearing cage in a controlled temperature room (20°C ± 1°C, 16 hours light: 8 hours dark). Eggs were left to hatch and larvae to complete development feeding on the swede. After approximately 30 days pupae were collected by gently sieving the sand around the swede. Collected pupae were placed onto damp sand in a plastic plant pot and moved to a second insect rearing cage in the same controlled temperature room. Emerging CRF adults were provided with a food source consisting of cotton wool pads dampened with sugar solution. Additional cotton wool pads were dampened with water and then smeared with a little Marmite ® and sprinkled with a small amount of soya flour and ground brewers yeast. Petri dishes filled with damp sand onto which small cubes of swede were placed were introduced into the cage in order to provide an egg laying site for the CRF. Eggs were washed from the sand and swede cubes and collected by water floatation to use in the predation studies or to re- infest fresh swedes to maintain the CRF culture.

Rearing method for carrot fly Adult carrot flies were collected from carrot fields using a sweep net. The adults were maintained in an insect rearing cage in order to collect eggs, using a method adapted from McLeod et al, 1985. The cage was in a controlled temperature room (20°C ± 1°C, 16 hours light: 8 hours dark). Petri dishes filled with damp sand to which chopped fresh carrot greens were added were provided as egg laying sites. Cotton wool pads dampened with sugar solution and additional pads smeared with Marmite ® were provided as a source of water and food for the adults. Eggs were washed from the sand and collected by water floatation to use in the predation studies.

Predation studies Initial predation bioassays confined individual A. coriaria adults with CRF eggs on damp filter paper in Petri dishes in order to confirm that the predator will eat CRF eggs. Further replicated bioassays were completed, to determine the number of CRF eggs eaten over 24 hours by individual A. coriaria adults offered increasing

SID 5 (Rev. 05/09) Page 5 of 20 numbers of eggs on damp filter paper or on damp compost. Similar 24-hour bioassays on damp filter paper were completed to determine predation rates of first instar CRF larvae by A. coriaria adults, of CRF eggs by A. coriaria third instar larvae and of carrot fly eggs by A. coriaria adults. All bioassays were completed at 21°C with a 16:8 light:dark photoperiod, with a minimum of four replicates.

Results and Discussion

Predation studies The initial predation bioassays using A. coriaria adults confined with either CRF or carrot fly eggs on damp filter paper confirmed that the predator will eat the eggs of both pests. Often the entire egg was not eaten, the predator often consumed up to half the egg shell and the contents, leaving a chewed empty egg shell.

On damp filter paper, individual A. coriaria adults predated all CRF eggs when offered 2, 4 or 8 eggs, and predated means of up to 15 and 20 eggs when offered 16 and 24 eggs respectively (Figure 1). Fewer CRF eggs (a mean of 7) were predated by A. coriaria adults when 20 eggs were offered on damp compost rather than on damp filter paper. When offered 20 CRF eggs on damp compost mixed with 5% turkey crumbs, significantly fewer (P<0.05) CRF eggs were eaten (a mean of 2.9) than on plain damp compost, indicating that the predators eat fewer live prey when the artificial food source is present. A. coriaria adults predated up to a mean of 12 young CRF larvae when offered 16 on filter paper. A. coriaria larvae predated a mean of 10 CRF eggs over 24 hours when offered 16 on filter paper.

Bioassays with carrot fly eggs as prey were limited by the number of available eggs. A. coriaria adults ate all carrot fly eggs when offered 8 eggs over a 24-hr period. As carrot fly eggs are smaller than those of CRF, it is likely that A. coriaria predation rates on carrot fly eggs will be at least as high as on CRF eggs.

20

16 Atheta adult on filter paper

12 Atheta adult on compost

8 Atheta adult on compost + turkey crumbs

No. eggs No. consumed eggs 4 Atheta larva on filter paper

0 0 5 10 15 20 No. eggs offered

Figure 1. Mean numbers of CRF eggs eaten by individual A. coriaria adults or third instar larvae when offered increasing numbers of eggs on damp filter paper or compost with or without turkey crumbs as alternative food.

Conclusions

 A. coriaria predates on both CRF eggs and first instar larvae and on carrot fly eggs.  Individual A. coriaria adults ate up to 20 CRF eggs per day on damp filter paper, but predation rates were reduced to seven eggs per day when offered on the more natural, complex substrate of damp compost.  Predation rates were reduced still further to three eggs per day when offered on damp compost mixed with 5% turkey crumbs as an artificial food source.  These data were used to guide release rates of A. coriaria in the semi-field and field experiments done in Objective 2.

SID 5 (Rev. 05/09) Page 6 of 20 Objective 2: Evaluate the efficacy and practicality of using a low cost/high output Atheta coriaria rearing-release system against the selected pests within ICM programmes, in experiments on crops in near-commercial or commercial conditions

Atheta coriaria development rates in brassica compost

Work in HDC project PC 239 and PC 239a determined the development rate of A. coriaria and its multiplication rate at 25°C when reared in a substrate of coir and vermiculite and when fed on turkey-rearing crumbs as an artifical diet (Bennison 2007 and 2010). In this project, A. coriaria development and multiplication rates were investigated at a range of temperatures consistent with those in unheated glasshouses used for brassica propagation and in compost used for raising brassica plants.

Methods The A. coriaria rearing method developed in HDC-funded project PC 239 and described above was used to compare the development rate of the beetles when reared in ventilated plastic boxes containing either the 'standard' rearing substrate (coir and vermiculite) or a commercial compost used for raising brassica plants. Two experiments were done in unheated glasshouses, one during the summer (August to September 2007) and the second in late winter/early spring (February to April 2008). Three replicate boxes were used for each substrate and 60 adult A. coriaria were added to each box. Every week, 5g turkey-rearing crumbs and water as required were added to each box. The numbers of adults and larvae in 12 replicate 30-ml substrate sub-samples per box were recorded at selected intervals and the data were subjected to Analysis of Variance. A similar experiment was done to determine A. coriaria development and population increase rates in the coir/vermiculite substrate with turkey crumbs as food, when kept in incubators at constant temperatures of 15°C and 10°C.

Results and Discussion During the summer 2007 experiment, after 6 weeks, mean numbers of A. coriaria per box were statistically similar in both brassica and coir composts (857 and 546 respectively, representing x14 and x9 increases). Daily mean, maximum and minimum compost temperatures averaged across the 6-week period were 20°C, 26°C and 16°C respectively. These results indicated that during a typical 6-week cauliflower propagation period in summer, A. coriaria will develop at least as well in brassica compost as in the standard coir/vermiculite rearing medium, and could multiply by around x14 in brassica modules, if turkey-rearing crumbs are added to the brassica compost as food.

In the winter/early spring 2008 experiment, set up on 13 February, the first A. coriaria larvae were recorded 12 days after the experiment was set up, and on day 37, mean numbers of larvae per rearing box were 186 in coir and 154 in brassica compost, representing around x3 population increases. The experiment was continued for a total of nine weeks, as at this time of year, the brassica propagation period is longer than in the summer. In week 9, new generation adults had still not developed. Mean numbers of larvae per rearing box were significantly lower (113) in brassica compost than in coir (288), P<0.05. These numbers indicated only x2 and x5 beetle multiplication rates respectively. Daily mean, maximum and minimum compost temperatures averaged across the experimental period were 15°C, 26°C and 8°C respectively. The lower multiplication rates in peat-based brassica compost than the coir/vermiculite substrate was probably due to the peat-based compost remaining wetter and thus colder during this period. These results demonstrated that in a typical 9-week cauliflower propagation period during February to April, A. coriaria developed and bred much more slowly than during the summer. Thus higher numbers of A. coriaria would need to be added to modules in the early spring propagation period, to achieve the required number of predators per module at planting out.

In the experiment to determine A. coriaria development and population increase rates in coir/vermiculite substrate with turkey crumbs as food at constant temperatures of 15°C and 10°C, A. coriaria adults were still alive after four weeks at 10°C but no larvae had developed. It was concluded that a constant temperature of 10°C is too cool for either egg laying, egg hatch, or both. At a constant 15°C, first larvae were recorded after 12 days and new generation adults had developed after 42-81 days. Mean population increases after 42 and 81 days were x5 and x10 respectively. These increases are lower and slower than at 25°C when fed weekly with 5g turkey crumbs as food (mean x20 increase after 23 days, Bennison 2007).

SID 5 (Rev. 05/09) Page 7 of 20 Conclusions  During a typical 6-week cauliflower propagation period in the summer, A. coriaria can multiply by around x14 in brassica modules if turkey crumbs are provided as food. However, in cooler winter/early spring glasshouse temperatures, the predators may only multiply by x2 in peat-based brassica compost during a 9-week propagation period.  A. coriaria adults can survive for four weeks at a constant temperature of 10°C but this temperature is too low for egg laying, egg hatch, or both.  At a constant 15°C, mean A. coriaria population increases after six and 12 weeks were x5 and x10 respectively.

Atheta coriaria flight behaviour

Methods An experiment was done in February/March 2007 to determine whether A. coriaria adults will fly and disperse at temperatures experienced in unheated brassica propagation glasshouses early in the season. Three replicate insect-proof mesh cages (80 x 80 x 80 cm) were placed in an unheated research glasshouse with natural day length. One 'rearing-release box' containing a mean of 1200 A. coriaria mixed adults and larvae was placed in each cage. The box was a standard rearing box with the insect-proof mesh removed from the ventilation holes in the lid, to allow adult beetles to leave the boxes of their own accord. Numbers of A. coriaria adults flying from the boxes were recorded on a sticky trap hung from the roof of each cage, three times on each working day for two weeks. Two ‘bait’ pots containing damp compost mixed with turkey crumbs were placed in each cage, to offer the predators a ‘home’ once leaving the boxes. Numbers of A. coriaria adults and larvae were recorded in each bait pot three times per week during the 2-week experiment.

Results and Discussion The first adult A. coriaria on a sticky trap was recorded one day after set-up, when mean, maximum and minimum air temperatures were 11.2°C, 13.5°C and 9.5°C respectively. Adult beetles were found on sticky traps on each recording date during the experiment, with mean, maximum and minimum air temperatures over the 2-week period being 14.2°C, 18°C and 11.7°C respectively. Mean numbers of beetles per trap per day were 0.3 in week 1 and 1.4 in week 2. Peak numbers of beetles (a mean of three per trap per day) were recorded eight days after set-up, which was also the date on which maximum air temperature was recorded (21.6°C). Some A. coriaria colonised bait pots rather than flying to sticky traps, with mean numbers of beetles per pot per day being 0.2 in week 1 and 0.8 in week 2. It was not possible to determine the temperature ‘threshold’ for A. coriaria flight from this experiment, but the results demonstrated that adults flew and dispersed in unheated glasshouses during February and March, and indicated that more flight occurs on warmer days.

Conclusions  Adult A. coriaria flew in an unheated glasshouse during February and March when mean air temperatures were 11.2-14.2°C. Thus the beetles could disperse in brassica propagation glasshouses in early spring, when plants are raised for planting out in April, when first generation CRF eggs are usually laid in the field.  The observations also implied that A. coriaria may fly and disperse in similar temperatures experienced in the field.

Atheta coriaria establishment in brassica plugs

Methods An experiment was done in three research glasshouses during August and September 2007, to evaluate how well A. coriaria establish in brassica modules during the propagation period, with the aim of planting the predator- ‘seeded’ modules out in the field. Module trays (with 345 modules per tray) were filled with brassica propagation compost mixed with either no turkey crumbs (untreated control), or 1%, 5%, 10% or 25% (by volume) turkey crumbs, thus providing different amounts of food for the predators. The trays were sown with cauliflower seed at a commercial propagator’s nursery. In glasshouse 1, 4,500 A. coriaria mixed adults and larvae were broadcast over the filled module trays (a mean of 1.4 predators per module). In glasshouse 2, 4,500 A. coriaria were placed alongside the module trays in four rearing-release boxes (a mean of 1125 per box), thus allowing the predators to disperse from the boxes and enter the modules of their own accord. In glasshouse 3, the module trays were filled with plain brassica compost and 4,500 A. coriaria were broadcast over the module trays and plants at the end of the 6-week propagation period. This treatment tested an alternative strategy for adding the predators, i.e. just before planting out in the field, rather than allowing the population to build up in the modules during the propagation period.

SID 5 (Rev. 05/09) Page 8 of 20

Percentage germination was recorded on each tray by counting missing plants in each tray once germination had occurred. Height and vigour of the seedlings was assessed weekly on 10 randomly selected seedlings per tray. Ten randomly selected replicate modules were removed weekly from each of the trays in glasshouses 1 and 2 and numbers of A. coriaria adults and larvae per module were recorded after gently crumbling the compost onto a white plastic tray. Ten modules per tray were similarly assessed in glasshouse 3 two days after adding the predators (simulating the dispatch date to the grower for planting out in the field). The data were subjected to Analysis of Variance.

Results and Discussion

Germination and plant vigour Adding turkey crumbs to the compost at 10% or 25% inhibited cauliflower seed germination and seedling survival. At the end of the propagation period, only 55% and 0.2% of the modules contained plants, in those filled with compost with 10% and 25% turkey mash respectively, compared with 89% in control modules with plain compost. The compost with 1% and 5% turkey mash allowed normal seed germination and seedling survival, with 90% and 86% of the modules respectively containing plants at the end of the propagation period. Cauliflower plants were significantly shorter (P<0.05) in the compost with 5% turkey crumbs than those in the plain compost or in compost with 1% turkey crumbs for the first three weeks of the 6-week propagation period, but were of similar height to those in plain compost by week 4 and were significantly taller than those in plain compost in week 6 (P<0.05). This result implied that the turkey crumbs led to initial inhibition followed by stimulation of growth.

Establishment of A. coriaria in modules Very few A. coriaria (maximum mean of 0.3 per module) established in the modules filled with plain compost or with 1% turkey crumbs throughout the experiment, whether the predators were added directly or allowed to enter the modules independently. Numbers of A. coriaria in modules filled with compost and 5% turkey crumbs peaked in week 2, when significantly more (a mean of 2.5 per module, P<0.05) were recorded than in modules with plain compost or with 1% turkey crumbs (means of 0 and 0.1 per module respectively), but only in the glasshouse with A. coriaria rearing-release boxes (Figure 2). Numbers of A. coriaria established in modules filled with compost with 10% turkey crumbs were no higher than in those filled with compost and 5% turkey crumbs on any date. Numbers of A. coriaria in modules filled with compost and 25% turkey crumbs peaked in week 3, when significantly more (a mean of 10.4 per module in the glasshouse with rearing-release boxes and a mean of 4.8 per module in the glasshouse where the predators had been directly released) were found than in modules with all other compost mixes (P<0.05), Figure 2. Modules filled with compost and 25% turkey crumbs contained higher numbers of predators than those with all other compost mixes on all assessment dates.

The percentage of modules containing A. coriaria reached 100% in weeks 2 and 3 only, in those filled with compost and 25% turkey crumbs. In weeks 2 and 3, significantly more (75%) contained predators in the modules with 5% turkey crumbs than in those with 1% turkey crumbs (10% containing predators). During the final three weeks of the experiment, mean numbers of A. coriaria per module declined in all compost mixes, with a mean of 0.4 per module in those filled with the 25% turkey crumbs in week 6. The reason for this decline in numbers of predators per module towards the end of the experiment is unknown, but could possibly be associated with turkey crumb decay over time. In year 2 of the project, when raising plants for the semi-field efficacy experiment, A. coriaria were introduced three weeks before the planting out date to avoid a potential crash in predator numbers.

SID 5 (Rev. 05/09) Page 9 of 20 12

10 per per plug per plug 0% 8

1% Atheta

Atheta of of 6 5% 10% 4 25%

nnumber nnumber 2

Mean Mean 0 1 2 3 4 5 6

Weeks after set up

Figure 2. Mean numbers of A. coriaria per module in trays filled with compost mixed with 0%, 1%, 5%, 10% and 25% turkey crumbs, in the glasshouse where A. coriaria dispersed from rearing-release boxes and colonised the modules of their own accord.

Mean numbers of A. coriaria were 0.5 per module in the third glasshouse, where the predators were broadcast direct to the plants and modules two days before the end of the experiment. The mean percentage of plugs containing predators was 32%. This result indicates that much higher numbers of A.coriaria would need to be added to the module trays at the end of the propagation period, to result in 100% of the modules containing sufficient predators when planted out in the field. However, this method may be useful to supplement predator numbers if needed, before module despatch to growers.

Conclusions

 Brassica compost with 5% turkey crumbs was the most suitable mix for filling module trays, for establishment of A. coriaria in the modules for the semi-field efficacy experiment. This mix allowed normal seed germination, and although it inhibited plant growth for the first three weeks of the 6-week propagation period, the plants were of normal height by week 4.  Allowing A. coriaria to disperse from rearing-release boxes placed alongside the module trays throughout the propagation period led to higher numbers of predators per module, and percentage of modules with predators, than broadcasting the predators direct to the module trays after seed sowing. This method was used to establish the predators in modules raised for the semi-field efficacy experiment.  Results from predation studies in Objective 1 indicated that each adult predator ate around three CRF eggs per day when offered the eggs on damp compost containing 5% turkey crumbs. In most years, a mean of 4-5 CRF eggs are laid per day per plant during the peak egg-laying period which usually occurs in early May (ADAS unpublished data). During year 2 of the project, in the semi-field experiment, the aim was to manipulate the numbers and timing of A. coriaria establishment in modules during plant propagation, to result in 100% of the modules containing at least two predators when planted out.

Semi-field efficacy experiment, CRF on brassicas

Methods A pilot semi-field experiment was done at ADAS Boxworth during July and August 2008 to test whether A. coriaria, either raised in cauliflower modules or added after planting, could reduce the severity of CRF damage after planting. Young cauliflower plants were raised in research glasshouses. To establish A. coriaria in the modules, module trays were filled with either plain compost, to which turkey-rearing crumbs were added to the surface (5% of compost volume) or compost mixed with 5% turkey-rearing crumbs. Three A. coriaria rearing- release boxes, each with a mean of 1,500 per box were placed alongside the module trays three weeks before the planting out date, to allow the predators to leave the boxes and to colonise and breed in the modules. On the planting date, the plants raised in compost to which turkey crumbs had been added to the surface were selected for the experiment, as assessment of 10 random modules from each treatment showed that these modules had higher mean numbers of A. coriaria (2 adults and 2 larvae per module) than the modules filled with the mixture of compost and turkey crumbs (0.4 adults and 0.5 larvae per module). In addition, A. coriaria had established in 90% of the modules where turkey crumbs had been added to the surface compared with only 50% of the modules where turkey crumbs had been mixed in the compost.

SID 5 (Rev. 05/09) Page 10 of 20

Six replicate plants were planted into individual pots of soil-based compost, for each of the five treatments: 1) untreated control (plants raised with no A. coriaria); 2) chlorpyrifos drench (Dursban WG, label rate) applied to modules before planting; 3) A. coriaria raised with the modules with a mean of 2 adults and 2 larvae per pot; 4) A. coriaria added after planting at 2 adults and 2 larvae per pot; 5) A. coriaria added after planting at 10 adults and 10 larvae per pot. The pots were placed in individual insect-proof mesh cages to confine the A. coriaria and to prevent wild CRF from infesting the plants. The pots were placed outdoors in a randomised block design and were watered as needed to keep the soil damp in order to ensure successful CRF egg hatch and to provide suitable soil conditions for A. coriaria. Ten CRF eggs per pot were added to the soil around the base of the plants immediately after planting and again one week later, to simulate second generation CRF egg-laying. After six weeks the roots were examined for CRF damage using the ADAS Root Damage Index (RDI) scoring system. A. coriaria remaining in the soil in the pots used for the three A. coriaria treatments were recovered by floatation in concentrated magnesium sulphate and numbers of adults and larvae per pot were recorded. The data were subjected to Analysis of Variance.

Results and Discussion

CRF root damage Mean % root damage was significantly lower (P<0.05) in plants treated with A. coriaria at 10 adults and 10 larvae per plant (5% RDI) and in plants raised with A. coriaria in the modules (28% RDI) than in untreated controls (61% RDI), Figure 3. Mean % root damage was not significantly reduced in plants treated with A. coriaria at 2 adults and 2 larvae per plant (39% RDI). The mean numbers of A. coriaria per module where the predators were esablished during propagation were estimated at 2 adults and 2 larvae at planting. However, numbers of predators are likely to have been higher than this as the modules would also have contained eggs and pupae which were not visible during compost assessments. This could explain why this treatment was more effective than where 2 adults and 2 larvae were added by hand after planting. Chlorpyrifos was 100% effective (no root damage). As cauliflower plants are considered to tolerate up to a mean of 40% root damage without reduction of yield or quality in the absence of drought stress, the results with A. coriaria were encouraging.

Numbers of Atheta coriaria in soil at end of experiment At the end of the 6-week experiment, A. coriaria were found in the soil in pots treated with all predator treatments. Significantly more (P<0.05) adults and larvae (mean 20.3 per pot) were recovered from pots treated with 10 adults and 10 larvae after planting than in those treated with 2 adults and 2 larvae after planting (mean 10.8 per pot) or in those where the predators were established in the modules before planting (mean 6.3 per pot). However, this 'proof of principle' experiment was done during the summer when temperatures and soil moisture were conducive to A. coriaria survival and development, and the A. coriaria were confined to the pots using insect-proof mesh cages. Therefore the Atheta system was tested in a field experiment the following year, where plants were exposed to natural first generation CRF egg laying, earlier in the season in cooler temperatures with natural soil moisture levels and where the predators were free to disperse.

70 60 50 40 30 *

20 %CRF damage 10 * * 0 untreated chlorpyrifos Atheta 2+2 Atheta 10+10 Atheta in Prop. Treatment

Figure 3. Mean % CRF Root Damage Index (RDI) in cauliflower plants in the semi-field experiment comparing different strategies for using A. coriaria compared with the standard chlorpyrifos drench to modules and with untreated plants. * significantly lower RDI than in untreated control plants.

SID 5 (Rev. 05/09) Page 11 of 20 Conclusions  A. coriaria added to the soil around cauliflower plants after planting at 10 adults and 10 larvae per pot was almost as effective as the standard chlorpyrifos module drench treatment, with mean % RDI being only 5% six weeks after infestation with 20 CRF eggs per pot, compared with 61% RDI in untreated plants.  A. coriaria established in the modules during propagation also reduced root damage, to a mean RDI of 28%.  The results of the semi-field experiment indicated that in the field experiment on a commercial cauliflower crop, selected treatments should include A. coriaria added to the plants after planting at a rate of at least 10 adults and 10 larvae per plant, and A. coriaria established in the modules during propagation, with a target rate of more than 2 adults and 2 larvae per module.

Semi-field efficacy experiment, carrot fly

Methods

A similar semi-field experiment to that done with CRF was done at ADAS Boxworth between July and October 2009 to test whether A. coriaria could reduce carrot fly damage to carrots. Adult carrot flies were collected using a sweep net in a field of continuous carrots at Warwick HRI. Adults were placed in an egg-laying cage in the laboratory and eggs were collected using the same method as that described in Objective 1 for the predation studies with A. coriaria. Young carrots, 10-12 cm long were gently dug up from a field of organic carrots which had not been sprayed with pesticides. Individual carrot plants were potted into 3-litre pots with a soil/peat-based compost to represent field soil. Ten replicate pots for each of two treatments were used: 1) untreated control and 2) A. coriaria at 10 adults and 10 larvae per pot. An additional four untreated plants were used to check when the damage assessment should be done.

The pots were placed in individual insect-proof mesh cages to confine the A. coriaria and to prevent wild carrot flies from infesting the plants. The pots were placed outdoors in a randomised block design and were watered as needed to keep the soil damp in order to ensure successful carrot fly egg hatch and to provide suitable soil conditions for A. coriaria. On 31 July, immediately before adding the A. coriaria to pots used for treatment 2, five carrot fly eggs were added to the soil around the base of the plants in all pots, by making a shallow groove in the soil, washing on the eggs and gently covering the eggs with soil. Ten carrot fly eggs from the same batch used to infest the carrots were added to damp filter paper in Petri dishes and kept in a laboratory at 21°C, 16:8 hrs light: dark photoperiod and checked after 3, 4, 5 and 6 days to assess % egg hatch. The four extra infested carrot plants were removed from the soil six weeks after egg infestation to check for carrot fly damage and the 20 experimental plants were left for a further four weeks until 8 October for the final assessment of carrot fly damage.

Results and Discussion All ten of the carrot fly eggs incubated on damp filter paper hatched successfully after 5-6 days, indicating 100% viability. However, no carrot fly damage was visible on any of the carrots assessed six or ten weeks after egg infestation. This result was disappointing as the efficacy of A. coriaria in reducing carrot fly damage could not be assessed.

CRF field experiment on a commercial farm

Methods

An experiment was done in a commercial cauliflower crop in Lincolnshire during May and June 2009 to test whether A. coriaria, either raised in cauliflower modules or added after planting, could reduce the severity of first generation CRF damage in the field. Young cauliflower plants were raised in a commercial propagation nursery over an 8-week period between 12 March and 6 May. To produce the A. coriaria in the modules, 11 module trays, each with 345 modules were filled with plain compost onto which laboratory-reared A. coriaria were released, together with a small amount of turkey-rearing crumbs. As results in A. coriaria development rate experiments earlier in the project had shown that in an unheated glasshouse in early spring, A. coriaria had multiplied by only x2 over an 8-week period, large numbers of beetles were released to the modules to achieve a target mean of at least five adults and five larvae per module at planting. Approximately 14,000 (4 per module) and 18,500 (5 per module) A. coriaria were broadcast over the trays on 12 March and 2 April respectively. In addition, three rearing-release boxes, each containing 2,250 beetles were placed alongside the module trays on 7

SID 5 (Rev. 05/09) Page 12 of 20 April and these were replaced with 10 boxes, each containing 1,550 beetles on 23 April. 22 randomly selected modules were removed from the trays on 2, 23 and 30 April and on 6 May (planting date) and numbers of A. coriaria adults and larvae per module were recorded after destructive assessment.

A randomised block experiment design with four replicates per treatment was used to test six treatments: 1) untreated control; 2) chlorpyrifos drench applied to modules (Dursban WG at label rate of 30 g per 5,000 plants); 3) spinosad drench applied to modules (Tracer at label rate of 60 ml per 5,000 plants); 4) A. coriaria raised with modules (target of five adults and five larvae per module); 5) A. coriaria added at planting out; 6) A. coriaria added at planting out and again one week later. The two insecticide treatments were applied as pre-planting drenches to modules. For treatment 5, a mean of ten adult and 10 larval A. coriaria were added by hand to the soil at the base of each plant after planting out; this was repeated for treatment 6 after one week. A. coriaria were added by taking equal amounts of rearing substrate from prepared boxes containing known numbers of beetles.

Each plot was 8 x 7 metres containing 176 plants except for untreated and chlorpyrifos-treated plots, which were half this size containing 88 plants. All plots were separated by ‘buffer’ plots the same size as the full-sized plots, to reduce the risk of A. coriaria moving to plots with other treatments. The buffer plots were planted with chlorpyrifos-treated plants, as in the rest of the field.

Planting out was done on 6 May, when peak first generation CRF egg laying was forecast for this area of Lincolnshire (Warwick HRI / HDC Pest Bulletin). At planting, pitfall traps (8.5 cm diameter) were set up in each plot (two per full-sized plot and one per half-sized plot) and were collected and replaced weekly for four weeks. The traps were primarily used to detect any A. coriaria migration from plots where it had been released, but numbers and genera of other beetles were also recorded. Ten additional traps were used in the same field 50 m away from the trial area, to indicate whether A. coriaria occurred naturally in the field. No pesticides were applied by the grower to the experimental plots. Pirimicarb and lambda-cyhalothrin were applied to the rest of the field on 18 June.

Every 3-7 days throughout the experiment period, numbers of CRF eggs laid in the field were estimated by recording numbers of CRF eggs in soil taken from the base of five plants in a buffer plot in each of the four replicate blocks (20 plants in total). At the start of the experiment, ten assessment plants per plot were marked and every week the numbers of marked plants dying due to CRF damage were recorded. Eight weeks after planting on 30 June, the surviving assessment plants in each plot were examined in the laboratory and CRF root damage, root weight and numbers of CRF larvae and pupae in soil around the roots were recorded. An assessment of marketability was completed on an additional 10 plants per plot. Data were subjected to Analysis of Variance.

Results and Discussion

Production of A. coriaria in modules Despite the high numbers of A. coriaria that were released to the trays of plugs during propagation, only low numbers (a mean of 0.6 adults and 0.4 larvae) per plug had established by the planting date and beetles were present in only 55% of the plugs (Table 1). Temperatures in the propagation house were low during this period due to the cold spring, with overall mean compost temperatures of 12.6°C being recorded between 12 March and 6 May (Figure 4). Work earlier in the project on A. coriaria development in brassica compost at low temperatures showed that the beetles multiplied slowly at a constant temperature of 15°C but did not lay eggs at 10°C. Mean 24-hr compost temperatures reached or exceeded 15°C on only 12 dates (Figure 1) and although maximum temperatures reached over 30°C for a short time on some dates, temperatures are likely to have been too low for beetle multiplication for much of the propagation period. Another problem with establishing the A. coriaria in the modules with turkey crumbs as food was that the extra nitrogen released by the turkey crumbs led to the plants being taller than those raised for the other experiment treatments, making it more difficult to synchronise the ideal plant height for planting.

Table 1. Mean number of A. coriaria adults and larvae per plug and % plugs with A. coriaria during propagation

Date Mean no. adults per Mean no. larvae per plug % plugs with adults and plug /or larvae 2 April (3 wks after seeding) 0.1 0.1 18% 23 April 0.3 0.4 41% 30 April 0.8 0.1 50% 6 May (planting date) 0.6 0.4 55%

SID 5 (Rev. 05/09) Page 13 of 20 40.00

35.00 mean 30.00 max 25.00 min 20.00 15C 15.00 10.00 mean temperature °C mean temperature 5.00 0.00 12- 19- 26- 02- 09- 16- 23- 30- Mar Mar Mar Apr Apr Apr Apr Apr

Figure 4. Mean, minimum and maximum compost temperatures during propagation of cauliflower plants for the field experiment

CRF egg laying CRF eggs were laid throughout the experimental period (Figure 5). Although planting was done when peak first generation egg laying had been forecast (Warwick HRI/HDC Pest Bulletin), numbers of eggs laid remained relatively low through the experiment period, peaking at just over 2 eggs per day in early June, five weeks after planting. The egg counts on 11 May were particularly low and this may have been due to the sand placed round the base of the monitoring plants to aid egg extraction, blowing away in the strong wind during the first week after planting. To minimise this risk on further monitoring dates, field soil was collected from around the plants rather than using sand. First generation CRF egg laying usually peaks in early May and in some years up to 21 eggs per plant per day can be laid (ADAS unpublished data). At this site in 2009, first generation eggs were laid in low numbers throughout May and June, merging with second generation egg laying which started in late June. Thus the plants in this experiment were subjected to continuous but low CRF pressure.

2.5 2 1.5 1 0.5 0

Mean CRF eggs/plant/day CRFMean 11-May. 14-May 21-May 28-May 04-Jun 11-Jun 18-Jun 25-Jun Date

Figure 5. Mean numbers of CRF eggs/plant/day throughout the experiment period following planting on 6 May.

CRF damage Results indicated that the effects of A. coriaria treatments on CRF were generally comparable with those of the insecticide treatments but were more variable, both between replicate plants per treatment and between the two broadcast A. coriaria treatments. This may reflect the importance of environmental variables affecting A. coriaria biology and behaviour. For example, the cold spring in 2009 slowed the rate at which numbers of A. coriaria raised in the modules increased so that at planting out means of only 0.6 adults and 0.4 larvae were recorded per module rather than the target means of 5 adults and 5 larvae per module, and the distribution of A. coriaria between modules was irregular (Table 1). Similarly the very dry soil conditions at planting will have affected A. coriaria egg-laying and dispersal behaviour. Despite these factors, adding A. coriaria at planting out significantly reduced the percentage of dead plants recorded when compared with untreated plots (Table 2). It is not clear why A. coriaria added after planting and repeated one week later did not have a similar effect on the percentage of dead plants.

SID 5 (Rev. 05/09) Page 14 of 20 Mean root weights of plants treated with A. coriaria raised in the modules or added at planting were significantly heavier than root weights of untreated plants (Table 2). However, none of the treatments tested, including the insecticide drenches, significantly reduced mean root damage index (RDI) compared with that on untreated plants. This is likely to have been due to the low CRF pressure during the experimental period maintaining RDI scores at below 40%, which is considered as the level of damage tolerated by cauliflower plants without a reduction in yield or quality if the plants are not drought-stressed. In this experiment, the main effect of treatment was the reduction of early plant death due to attack of young plants at and shortly after planting. In addition to RDI's of surviving plants being unaffected by treatment, the final assessment of marketability found no significant difference between treatments, with mean percent marketable cauliflowers ranging between 50% in untreated plots to 67.5% in plots treated with spinosad drench. Mean numbers of CRF larvae, pupae and new adults collected from soil around each plant at harvest did not differ significantly between treatments.

Table 2. Summary of % plant death due to CRF damage, root damage, root weights and numbers of CRF larvae, pupae and new adults per plant. * indicates significant difference compared with untreated control (P<0.05).

Mean Assessment Scores Dead Root Root CRF adults, Treatment Plants Damage Weight (g) larvae & (%) Index (%) pupae/plant untreated 12.5 36.9 23.6 7.2 chlorpyrifos drench 0* 38.1 35.6* 2.8 spinosad drench 0* 30.9 43.2* 3.6 A. coriaria raised with modules 5 18.3 39.4* 4.2 A. coriaria added at planting out 2.5* 30.5 35.5* 5.5 A. coriaria added at planting out and 5 24.7 31.2 3.6 one week later

Atheta coriaria in pitfall traps Approximately 3,800 A. coriaria were released into each plot at planting, equivalent to 630,000 beetles per ha (over 1 million per ha when two releases were made). Pitfall traps in the experimental plots contained significantly higher (P<0.05) numbers of A. coriaria (up to a mean of 10.6 per trap) on 14 May, one week after the first release at planting, where the beetles had been broadcast in the field than traps in all other treatments (Figure 6). On this date, no A. coriaria were found in the traps 50 m away from the trial area, indicating that this species, although native to the UK, was not naturally occurring in the field and thus the A. coriaria found are likely to be those that were released. A. coriaria are usually found in highly organic substrates (R. Booth, personal communication) and so are unlikely to persist in the silt soils in this experimental field. More A. coriaria recorded in traps on 14 May than on any other date. On 21 May, one week after the second release, numbers of A. coriaria were no higher in pitfall traps in plots where two releases had been made than in traps in all other treatments. Adult A. coriaria are highly mobile and fly actively on warm days, thus in suitable conditions they can disperse rapidly. This was apparent in the field trial, as low numbers of A. coriaria were recorded in plots where they had not been released from 14 May onwards, despite the use of ‘buffer’ plots. In addition, very low numbers (up to a mean of 0.3 per trap) of A. coriaria were found in the traps 50 m away on 21 May. Adult dispersal could have been partly due to increasing temperatures and to the very dry soil conditions at planting and during the following week, as A. coriaria prefer damp substrates. In addition, only low numbers of CRF eggs were laid to provide a food source thus scarcity of food may have stimulated dispersal. A small amount of turkey crumbs were broadcast with the beetles to provide a local food source and to help reduce dispersal, and other insect eggs and larvae are likely to have been available as prey. Further work is needed on dispersal behaviour of A. coriaria and efficacy against CRF in different soil types, moisture levels and at different CRF egg pressures.

SID 5 (Rev. 05/09) Page 15 of 20 12 * 10 * 8 14 May 21 May 6 28 May 4 4 June 2

Mean no. Atheta per trap Mean no. Atheta 0 untreated chlorpyrifos spinosad drench A. coriaria raised A. coriaria added A. coriaria added drench with modules at planting out at planting out and one week later Treatment

Figure 6. Mean numbers of A. coriaria per pitfall trap after release on 6 and 14 May. * significantly more than in other treatments on 14 May (P<0.05).

Other beetles in pitfall traps Other staphylinid beetles recorded in the traps included Anotylus spp., Tachyporus spp. and ‘other Aleocharinae’ (including Aleochara spp. which were likely to include A. bilineata, a naturally-occurring CRF egg predator and pupal parasitoid). Of these, the ‘other Aleocharinae’ were the most numerous, reaching an overall mean of 15.2 per trap on 28 May (Figure 7). On this date traps amongst chlorpyrifos-treated plants, those raised with A. coriaria in modules or those treated with two field releases of the predator had significantly fewer ‘other Aleocharinae’ than those in untreated plots. The data on 28 May indicated that there were significant interactions between the A. coriaria and chlorpyrifos treatments and numbers of ‘other Aleocharinae’, such as reduced availability of CRF eggs in plots receiving these treatments and possible direct toxicity of chlorpyrifos. On 4 June significantly fewer ‘other Aleocharinae’ were recorded in all treated plots than in untreated plots. Again this may have been due to higher numbers of CRF eggs and larvae in untreated plots, but data on CRF numbers was not recorded until harvest on 30 June. Bembidion spp. were the most numerous of the carabid beetles recorded in pitfall traps and this is consistent with records of carabids on the same farm in 2005/06 (Eyre et al 2009). As with the Aleocharinae the highest numbers of Bembidion were recorded on 28 May when there was a mean of 4.4 per trap. Other carabids recorded included Agonum spp., Amara spp., Harpalus spp., Nebria spp., Pterostichus spp. and Trechus spp. There was no effect of any treatment on numbers of any of the carabid species recorded, which indicated that there were no significant interactions with A. coriaria during the trial period.

25

20 14 May 15 * 21 May * 10 28 May * 4 June ▲ ▲ 5 ▲ ▲ ▲

0 untreated chlorpyrifos spinosad drench A. coriaria raised A. coriaria added A. coriaria added

Mean Mean no. per trap of Aleocharinae drench with modules at planting out at planting out and one week later Treatment

Figure 7. Mean numbers of ‘other Aleocharinae’ per pitfall trap between 14 May and 4 June. * significantly fewer than in untreated plots on 28 May and ▲ significantly fewer than untreated plots on 4 June (P<0.05).

SID 5 (Rev. 05/09) Page 16 of 20 Potential commercial uptake of Atheta coriaria as a biological control agent of CRF These results provide the first data on the fate and pest control effect of A. coriaria following field release. Although further research is needed on the efficacy and behaviour of A. coriaria at different CRF egg pressures and in different soil types and moisture contents, these initial results indicate that A. coriaria is a good candidate for cost-effective inundative release for biological control of CRF, and its short persistence in certain soil types could present low risk to other resident and immigrant beetles.

The approval for use of chlorpyrifos for drenching brassica modules for CRF control will soon be withdrawn due to potential operator hazard. The alternative modular drench is spinosad. Use of spinosad is unpopular with propagators due to its expense (82 pence per tray of 345 plants) in comparison with using chlorpyrifos (3 pence per tray). Discussions with both conventional and organic brassica growers have confirmed commercial interest in further development of the A. coriaria technique. Release of A. coriaria in the field would be their preferred option rather than establishing the predators during propagation. This would avoid the problems with A. coriaria establishment in the modules in low early spring temperatures. Propagators have expressed an interest in rearing A. coriaria if the method developed in HDC-funded project PC 239 and PC 239a (Bennison 2007, 2010) can be scaled up for cost-effective use on a commercial scale. Current estimated costs of producing A. coriaria for field release at the rate of 10 adults and 10 larvae per plant would be similar to the cost of a modular drench of spinosad. However, bulk buying of rearing substrate and reduced labour time if the system was scaled up using fewer, larger rearing containers would reduce costs considerably. Results of previous research on using a staphylinid beetle, Aleochara bilineata (a CRF egg predator and pupal parasitoid) were not developed for commercial use against CRF, partly because mass-rearing is too expensive due to the need for host flies for the parasitic life stage (Finch, 1996, 2002; Hartfield & Finch, 2003).

Conclusions from field experiment

 CRF egg laying pressure was relatively light and unusually prolonged during the experiment period. Thus the main effect of CRF on untreated plants was the early death of young plants checked by dry soil conditions.  A. coriaria added at planting out at the rate of 10 adults and 10 larvae per plant significantly reduced the number of dead plants, as did the modular drenches of chlorpyrifos and spinosad.  A. coriaria, both added at planting and raised in the modules led to heavier root weights in surviving plants than in untreated plants, as did both pesticide drenches to modules. However, none of the treatments reduced root damage.  Pitfall trapping indicated that A. coriaria dispersed within two weeks after release. This could have been partly due to dry soil conditions.

Objective 3: Communicate the results and demonstrate the system to the industry

Industry presentations and publications

 Jude Bennison presented a summary of the project objectives and results to date at the UK vegetable industry conference, East of England Showground, Peterborough, 12 February 2008 (written summary also published in conference proceedings).

 Jude Bennison presented and discussed the results to date with brassica growers at the ADAS vegetable centre meeting, Warwick HRI, Boston, Lincs, 27 February 2008.

 Jude Bennison and Tom Pope presented and discussed the final results of the project with major conventional and organic UK vegetable growers and propagators at Marshall Bros (Butterwick) Ltd, Lincs (part of Produce World), on 15 December 2009. Marshall Bros and Westhorpe Flowers & Plants hosted the field experiment in 2009 and they, together with technical staff at Produce World guided the work in the field experiment so that the work fitted with commercial propagation and production practices.

 Jude Bennison presented an update on results at the UK vegetable industry conference, East of England Showground, Peterborough, 28 January 2009 (written summary also published in conference proceedings). A summary of the presentation was also published by a journalist in HDC News, March 2009.

SID 5 (Rev. 05/09) Page 17 of 20  Jude Bennison & Tom Pope presented a poster on the results of the project at the UK Brassica Growers Association Conference, 19 January 2010 and at the UK vegetable industry conference, East of England Showground, Peterborough, 27 January 2010. Tom Pope also demonstrated the Atheta coriaria rearing system at the latter conference. Presentations to the scientific community

Jude Bennison presented the results of the project at the AAB conference ‘Theoretical population ecology and practical biocontrol – bridging the gap’, Studley Castle, Warwicks, 6 December 2007.

Tom Pope presented a paper on the project results at the IOBC Working Group Meeting, Integrated control in field vegetables, Croatia, October 2009.

Jude Bennison & Tom Pope will present a paper on the project results at the IOBC Working Group Meeting, Landscape Management for Functional Biodiversity, Cambridge, June 2010.

Scientific publications

Bennison, Jude; Lole, Mike; Pope, Tom; Maulden, Kerry; Maher, Heather and Watling, Martyn. In press. Potential control of cabbage root fly (Delia radicum) with the predatory staphylinid beetle Atheta coriaria. IOBC/wprs Bulletin, Integrated Protection of Field Vegetables.

Bennison, Jude; Lole, Mike; Pope, Tom; Maher, Heather; Maulden, Kerry and Watling, Martyn. In press. What happens to the predator Atheta coriaria when inundatively released in the field for biological control of cabbage root fly? IOBC/wprs Bulletin, Landscape Management for Functional Biodiversity.

Possible future work

The project has demonstrated the potential of Atheta coriaria for biological control of cabbage root fly (CRF). The semi-field experiment was done during the summer months (at the time of second generation egg laying) with only one rate of CRF egg infestation and the field experiment was done in a year when first generation CRF laid low numbers of eggs over an unusually prolonged period. Further fundamental research is needed to quantify how many A. coriaria are needed to give sufficient and reliable control of both first and second generation CRF at a range of egg laying pressures. This would determine the most cost-effective predator release rates. A reliable semi-field pot test was developed in the project. This offers the opportunity for a further, similar experiment to be done at the time of first generation egg laying using a range of CRF egg and A. coriaria numbers, to determine the biological 'dose response' to increasing egg numbers. The results of the semi-field experiment should then be validated in replicated field trials.

Development work would then be needed on scaling up the A. coriaria rearing system for cost-effective use on a large scale by commercial propagators, and on practical and effective commercial release methods in the field, followed by knowledge transfer activities such as a factsheet and workshop for brassica propagators and growers.

Acknowledgements

Thanks to the following:  Richard GreatRex, Syngenta Bioline Ltd. for supplying the 'starter' culture of Atheta coriaria  Rosemary Collier, Warwick HRI for help in sourcing carrot flies  Roger Wood, Elsoms Seeds Ltd. for providing cailiflower seeds  John Overvoorde, Delfland Nurseries Ltd. for seeding brassica module trays  Roger White, Westhorpe Flowers & Plants for raising plants for the field experiment and for technical guidance and support  Phillip Effingham, Carolyn Coxe & colleagues, Marshall Bros. (Butterwick) Ltd and Emma Garrod, Produce World for hosting the field experiment and for technical guidance and support

SID 5 (Rev. 05/09) Page 18 of 20

References to published material 9. 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.

Bennison, J. 2007: Development of an on-nursery rearing system for Atheta coriaria for reduced cost biological control of sciarid and shore flies. First year report to HDC on project PC 239.

Bennison, J., Maulden, K., Maher, H. & Tomiczek, M. 2008: Development of a grower rearing-release system for Atheta coriaria, for low cost biological control of ground-dwelling pest life stages. IOBC/wprs Bulletin 32 (1): 21-24.

Bennison, J. 2010: Grower system for rearing the predatory beetle Atheta coriaria. HDC Factsheet 06/10.

Bennison, Jude; Lole, Mike; Pope, Tom; Maulden, Kerry; Maher, Heather and Watling, Martyn. In press: Potential control of cabbage root fly (Delia radicum) with the predatory staphylinid beetle Atheta coriaria. IOBC/wprs Bulletin, Integrated Control in field vegetables.

Bennison, Jude; Lole, Mike; Pope, Tom; Maher, Heather; Maulden, Kerry and Watling, Martyn. In press: What happens to the predator Atheta coriaria when inundatively released in the field for biological control of cabbage root fly? IOBC/wprs Bulletin, Landscape Management for Functional Biodiversity.

Carney, V.A., Diamond, J.C., Murphy, G.D. & Marshall, D. 2002: The potential of Atheta coriaria Kraatz (Coleoptera: Staphylinidae), as a biological control agent for use in greenhouse crops. IOBC/wprs Bulletin 25 (1): 37-40.

Eyre, M.D., Labanowska-Bury, D., Avayanos, J.G., White, R., Leifert, C. 2009: Ground beetles (Coleoptera, Carabidae) in an intensively managed vegetable crop landscape in eastern England. Agriculture, Ecosystems and Environment 131 (3-4), 340-346.

Finch, S. & Coaker, T. H. 1969: A method for the continuous rearing of the cabbage root fly Erioischia brassicae (Bch.) and some observations on its biology. Bulletin of Entomological Research 58: 619-627.

Finch, S. 1996: Problems associated with controlling the cabbage root fly by inundative releases of the rove beetle, Aleochara bilineata. IOBC/wprs Bulletin 19 (11): 152-155.

Finch, S. 2002: Biocontrol of the cabbage root fly by the release of predators. Defra (HH1830SFV) project final report.

Hartfield, C. & Finch, S. 2003: Releasing the rove beetle Aleochara bilineata in the field as a biological agent for controlling the immature stages of the cabbage root fly, Delia radicum. IOBC/wprs Bulletin 26 (3): 127-133.

McLeod, D.G.R., Whistlecraft, J.W. and Harris, C.R. 1985: An improved rearing procedure for the carrot rust fly (Diptera: ) with observations on life history and conditions controlling diapause induction and termination. Canadian Entomologist 117: 1017-1024.

SID 5 (Rev. 05/09) Page 19 of 20

SID 5 (Rev. 05/09) Page 20 of 20