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 34

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 HH3115TSF

2. Project title

IPM methods for blackcurrant gall mite and leaf midge: synthesis, validation and implementation in UK commercial blackcurrant production

3. Contractororganisation(s)

East Malling ResearchNew RoadEast MallingKent ME19 6BJ     

54. Total Defra project costs £ 219,525

5. Project: start date................ 01 August 2003

end date................. 31 December 2006

SID 5 (2/05) Page 2 of 34

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.

Blackcurrant gall mite and leaf midge are the most important pests of blackcurrant in the UK. Before this project commenced, they were controlled by foliar sprays of the broad-spectrum synthetic pyrethroid acaricide fenpropathrin (Meothrin). However, fenpropathrin is being withdrawn in 2008 due to the EU pesticide review. The gall mite is the vector of reversion virus disease, which causes sterility in blackcurrant bushes, and is the principal factor limiting the life of blackcurrant plantations. The mite colonises developing blackcurrant buds in spring and its feeding initiates prolific growth of cells within the bud resulting in the characteristic ‘big bud’ disorder. Control with acaricides is aimed at preventing successful migration and infestation of the new axillary buds and the resultant spread of virus infection. Control of mites within the galls is not possible with the acaricides currently available, which are not able to penetrate the galls. Traditionally, three sprays of fenpropathrin were applied per season for control of the mite, the first just before flowering, the second at the end of flowering and the third 10 days later. Fenpropathrin is harmful to a very wide range of arthropods, including the natural enemies of gall mite and leaf midge. The purpose of this project was to synthesize and validate IPM methods for blackcurrant gall mite and leaf midge that are compatible with IPM practices for other pests, are safer to humans and the environment and that reduce the occurrence of pesticide residues and to implement them in UK commercial blackcurrant production.

A six-year field experiment at East Malling Research demonstrated the long-term effectiveness of integrated use of gall mite or gall mite plus reversion virus resistant varieties with acaricide sprays. Combining host plant resistance to gall mite (in the variety Ben Hope) with use of early sulphur acaricide sprays gave 97.6% control of gall mite and 88 % control of reversion virus. Combining host plant resistance to gall mite plus reversion virus disease (in the variety Ben Gairn) with use of early sulphur acaricide sprays gave > 99.5% control of gall mite and 98-100 % control of reversion virus. The relative susceptibility of the two varieties to gall mite was found to be as follows: Ben Alder : Ben Gairn : Ben Hope = 4.8-8.7 : 1.7-2.2 : 1. It was shown that gall mite gall numbers increase exponentially from year to year, 80% of the variation in numbers on the log scale being accounted for by the relationship n=B*exp(k*year) where n is the average number of gall mites per bush, year is the interval in years from the start of the experiment (2000=1, 2001=2, 2002=2 etc), the coefficient B taking differing values for different varieties and k different values for different treatments. On the susceptible variety Ben Alder, reversion virus infection was first seen on a small number of bushes at flowering in year 3. The

SID 5 (2/05) Page 3 of 34

distribution was not related to treatment. The proportion of bushes infected increased steadily from year to year, until practically all bushes were infected at the spring of year 7. Ben Gairn appeared to be fully resistant to reversion virus until the end of the experiment when 5 bushes were found with possible symptoms of infection. However, the presence of infection in these bushes was not confirmed in DNA tests at SCRI. Further tests are planned in 2007. On Ben Hope, the first appearance of infection by reversion virus occurred in year 4, one year later than on the susceptible variety Ben Alder. The subsequent increase in infection occurred at a slightly lower rate on Ben Hope than Ben Alder. By the end of the experiment, approximately half the untreated bushes were infected. The incidences of infection on bushes that received acaricide treatments were markedly lower. An important conclusion is that varietal resistance needs to be protected by sprays of acaricides to lessen the likelihood of the development of resistance breaking strains of gall mite and/or reversion virus disease.

Sulphur was shown to be the most effective acaricide for control of the gall mite, two early season sprays giving superior control to the standard Meothrin spray programme. However, it was shown that sulphur can cause substantial phytotoxicity to blackcurrant with the severity of visual symptoms, yield loss and growth reduction, depending greatly on variety, temperature conditions and growth stage at application. The phytotoxicity of sulphur sprays to five widely grown blackcurrant varieties was quantified. Baldwin and Ben Gairn were most susceptible, Ben Hope and Ben Lomond intermediate and Ben Tirran only slightly susceptible to sulphur phytotoxicity. Importantly, a sulphur spray applied at the late dormant/bud-burst growth stage did not cause phytotoxicity to any variety. However, a spray applied at the first grape visible growth stage caused significant phytotoxicity to the susceptible varieties Baldwin and Ben Gairn. It was concluded that the use of sulphur on these varieties should be avoided where possible. Phytotoxic effects of applications at this time to Ben Lomond, Ben Hope and Ben Tirran were at most small and insignificant. Later sprays, either pre- or post-blossom were phytotoxic, the severity of losses depending on variety and temperature conditions. On the susceptible varieties (Baldwin and Ben Gairn), severe leaf blackening, chlorosis and drop and yield losses of up to 27% may occur, depending on temperature conditions at the time of application. On moderately susceptible varieties (Ben Lomond, Ben Hope), phytotoxicity symptoms are likely to be less severe and losses more moderate (~10%). Applications of sulphur when temperatures are high (> 25 ˚C) exacerbate phytotoxic effects. Application in these conditions should be avoided. A sensible, cautious approach would be to avoid making applications temperatures are > 20˚C.

Optimum timing and use of sulphur to gain maximum efficacy with minimal risk of phytotoxicity was established. The preferred gall mite treatment developed in this project of a spray of sulphur 800 g/l at 10 kg/ha at bud burst, followed by a second spray just before grape visible (except on Baldwin and Gairn where this spray should be omitted).

Masai was identified to be the most effective alternative acaricide to be used for supplementary growing season sprays when sulphur cannot be used because of the risk of phytotoxicity. Masai is comparatively crop safe and is unlikely to cause significant phytotoxicity. However, it was found only to be partially effective and complete control of gall mite could not be achieved even with two supplementary sprays. One disadvantage of Masai apart from its relatively high cost is that it has a high risk to bees and cannot be used during flowering. It was also shown that the novel Bayer product, UKA378b, had some promise as a gall mite acaricide and requires further evaluation. Indeed, it may prove superior to Masai. The EMR gall mite migration forecasting model provided accurate predictions of the start, 5% and 50% migration. This model may be useful in timing sprays but is not suitable for timing sulphur sprays (because of phytotoxicity) except on Ben Tirran which was shown not be susceptible to sulphur.

A further important objective was to determine whether the entomopathogenic fungus Lecanicillium longisporum (formerly known as Verticillium lecanii) could be exploited as a biocontrol agent of gall mite and identify the optimum time of year when it should be applied. Unfortunately, the fungus showed little promise as a biocontrol agent for gall mite. Pathogenicity of the fungus to the mite could not be clearly demonstrated even with direct exposure and in ideal conditions in the laboratory.

The gall mite IPM methods were validated in commercial practice in a large scale experiment in a commercial plantation. The 3 x fenpropathrin and IPM programmes performed similarly for gall mite control, but were inadequate because the plantation was too heavily infested. This, suggests that gall mite control is increasingly difficult as populations increase and priority should be given to keeping plantations as clean as possible for as long as possible.

SID 5 (2/05) Page 4 of 34

For blackcurrant leaf midge, the key objective was to test the hypothesis that the parasitoid Platygaster demades, anthocorids and other natural enemies can establish and naturally regulate leaf midge populations to below damaging levels in commercial plantations. However, no evidence was found that Platygaster demades is a key natural enemy of blackcurrant leaf midge. Observations also indicated that generalist predators appear to be of only secondary importance in regulating leaf midge numbers, populations only developing in response to high pest populations and that they are not able to reduce leaf midge numbers to the low levels needed to prevent considerable attack. Observations suggested that anthocorids are the dominant leaf midge generalist predator. Selective insecticidal controls for leaf midge are only likely to be effective if they are applied early in the attack, before midge larvae are inside galls. Further work to complete the identification of the midge’s sex pheromone so that sex pheromone traps can be used for timing of sprays is the highest priority for future research.

The findings of this work have major implications for gall mite and leaf midge IPM in blackcurrants. They have been comprehensively and universally implemented in commercial blackcurrant production in the UK.

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).

OBJECTIVES AND EXTENT TO WHICH THEY HAVE BEEN MET

Objective 1. Synthesize and validate IPM methods for gall mite

Objective 1.1. Determine the long-term effectiveness of selective acaricide sprays at the start and peak of migration, timed by the East Malling Research forecasting model, for gall mite and reversion virus prevention on resistant versus susceptible cultivarsThis objective has been met in full. A six year experiment demonstrated the long-term effectiveness of integrated use of gall mite or gall mite plus reversion virus resistant varieties with acaricide sprays. Combining host plant resistance to gall mite (Ben Hope) with use of early sulphur acaricide sprays gave 97.6% control of gall mite and 88 % control of reversion virus. Combining host plant resistance to gall mite plus reversion virus disease (Ben Gairn) with use of early sulphur acaricide sprays gave > 99.5% control of gall mite and 98- 100 % control of reversion virus. Sulphur was found to be the most effective acaricide but because it was found to be phytotoxic to many blackcurrant varieties (see objective 1.2 below), use had to be confined to two sprays in the early season, the first of these at bud break before mite migration is likely to commence on most varieties, and the second at the first grape visible growth stage, approximately at the time of the start of migration. Thus the mite migration model could not be used to time these sprays. However, its use to time supplementary acaricide sprays was demonstrated. An important overall conclusion is that host plant resistance should be protected with sprays of acaricides to lessen the likelihood of the development of resistance breaking strains of gall mite and/or reversion virus disease.

Objective 1.2. Determine whether sulphur can be used on blackcurrant before and/or after flowering by quantifying the phytotoxic effects of sprays of sulphur including effects on growth and yield

SID 5 (2/05) Page 5 of 34

This objective was met in full. The phytotoxic effects of sulphur to blackcurrant were quantified in 3 comprehensive field experiments. The experiments demonstrated that sulphur sprays can cause substantial phytotoxicity to blackcurrant but the severity of visual symptoms, yield loss and growth reduction depend greatly on variety, temperature conditions and growth stage at application. The susceptibility of the five of the most widely grown blackcurrant varieties was quantified. It was shown that early season sprays of sulphur are not or are only very slightly phytotoxic even to the most susceptible varieties. On susceptible varieties, yield losses of up to 27% from later sprays were demonstrated. On moderately susceptible varieties, yield losses were more moderate (~10%). It was also demonstrated that application of sulphur when temperatures are high (> 25 ˚C) exacerbate phytotoxic effects leading to the conclusion that application in these conditions should be avoided.

Objective 1.3. Identify an alternative selective acaricide to fenpropathrin for control of gall mite at the peak of migration that is not phytotoxic to blackcurrantThis objective was met in full. Four comprehensive field experiments at East Malling research showed that sulphur was the most effective acaricide for gall mite control of a wide range of products tested. Optimum timing and use of sulphur to gain maximum efficacy with minimal risk of phytotoxicity was established. Masai was identified to be the most effective alternative acaricide too be used for supplementary growing season sprays when sulphur cannot be used because of the risk of phytotoxicity. However, Masai was found only to be partially effective and complete control of gall mite could not be achieved even with two supplementary sprays of Masai. One disadvantage of Masai apart from its relatively high cost is that it has a high risk to bees and cannot be used during flowering. It was also shown in the final experiment that the novel Bayer product, UKA378b, shows some promise as a gall mite acaricide and requires further evaluation and may prove superior to Masai. It was also shown that the EMR gall mite migration forecasting model provides fairly accurate predictions of the start, 5% and 50% migration. This model may be useful in timing of sprays but is not suitable for timing sulphur sprays (because of phytotoxicity) except on Ben Tirran which was shown not be susceptible to sulphur phytotoxicity.

Objective 1.4. Determine whether foliar sprays of the commercially available strain of the entomopathogenic fungus Lecanicillium longisporum (formerly known as Verticillium lecanii) in the product Vertalec can be exploited as a biocontrol agent of gall mite and identify the optimum time of year when it should be appliedThis objective was met in full. The entomopathogenic fungus Lecanicillium longisporum (Vertalec) showed little promise as a biocontrol agent for gall mite. Programmes of multiple sprays of the fungus in the field neither caused infection inside galls, nor reductions in gall formation compared to untreated controls. Pathogenicity of the fungus to the mite could not be clearly demonstrated even with direct exposure and in ideal conditions in the laboratory.

Objective 1.5. Validate the IPM methods for gall mite in commercial practiceThis objective was met in full. In a large scale IPM experiment in a commercial plantation the conventional and IPM programmes performed similarly for gall mite control, but were inadequate because the plantation was too heavily infested. This, together with data from the long-term gall mite IPM experiment (objective 1.1) and the acaricides trials (objective 1.3), suggests that gall mite control is increasingly difficult as populations increase and priority should be given to keeping plantations as clean as possible for as long as possible. The increase in gall mite population from year to year is exponential until saturation is reached.

Objective 2. Synthesize and validate IPM methods for blackcurrant leaf midge

Objective 2.1. To test the hypothesis that the parasitoid Platygaster demades, anthocorids and other natural enemies can establish and naturally regulate leaf midge populations to below damaging levels in commercial plantations over a three year period if broad-spectrum insecticides are not used and other practices to enhance the natural enemies are implemented.This objective was met. However, no evidence was found that Platygaster demades is a key natural enemy of blackcurrant leaf midge able to regulate populations to below damaging levels. It was also shown that generalist predators appear to be of only secondary importance in regulating leaf midge numbers, populations only developing in response to high pest populations and that they are not able

SID 5 (2/05) Page 6 of 34

to reduce leaf midge numbers to the low levels needed to prevent considerable attack. Observations suggested that anthocorid bugs are the dominant leaf midge predator.

Objective 2.2. Determine whether or not sprays of sulphur or of the selective acaricide identified in 1.3 above have an adverse affect on the key natural enemies of leaf midgeThis objective was not met. Populations of Platygaster demades that developed were too small to determine the effects of treatments on the parasitoid.

Objective 2.3. To identify a selective insecticide treatment for control of blackcurrant leaf midgeTwo field experiments were conducted but the second of these failed because of weather conditions shortly after spraying and failure of populations of the pest to develop subsequently. The first experiment indicated that a wide range of products tested have at best only very limited curative activity against existing infestations of semi-mature and mature larvae in galls. None of the treatments had a worthwhile degree of efficacy when applied at this late stage in the larval attack and none of the products tested were more effective than Meothrin. The results highlight the importance of treatment application at the early stages of oviposition and larval attack. In separate work by EMR and NRI, good progress in identification of the sex pheromone of the blackcurrant leaf midge has been made. The pheromone will almost certainly prove crucial in timing of sprays and when this is done it may be possible to identify products that have selective activity against the midge without harm to its important natural enemies.

Objective 3. Deliver and promote the IPM methods to UK blackcurrant growers

This objective has been comprehensively met. The findings of this project have been fully and almost universally implemented in UK blackcurrant production.

METHODS USED AND RESULTS OBTAINED

Objective 1. Synthesize and validate IPM methods for gall mite

Objective 1.1. Determine the long-term effectiveness of selective acaricide sprays at the start and peak of migration, timed by the East Malling Research forecasting model, for gall mite and reversion virus prevention on resistant versus susceptible cultivarsIn the late 1990s, two new blackcurrant varieties, Ben Hope resistant to gall mite and the Ben Gairn resistant to reversion virus, were released by SCRI and they have now been widely planted commercially in the UK. Work in this objective of the project was to investigate whether the use of host plant resistance combined with sulphur sprays can adequately prevent gall mite and reversion virus infection under conditions of severe pest pressure in the long term, continuing work in the previous three-year DEFRA-funded project HH1942SSF. In the previous project, an experimental plantation consisting of a 5x5 Latin square of plots split by the three varieties Ben Hope, Ben Gairn and Ben Alder (standard susceptible variety) was established. Three sulphur based spray programmes were compared with a standard programme of three sprays of fenpropathrin (Meothrin) at fixed growth stages and untreated control. However, reversion virus infection was only just starting to develop in the susceptible variety so this experiment was continued from 2003 - 2005 with a final reversion virus assessment in spring 2006. The treatments were modified so that only IPM of gall mite and reversion virus were investigated in order that the long-term effects of the treatments could be determined. Here, the increases in gall mite populations and the increase and spread of the virus over the 6 year period of the experiment are reported. This work is reported fully in Cross and Harris (2004d, 2004f, 2005e, 2006c) and Cross (2006).

Materials and methodsThe IPM experiment was conducted in a purpose planted blackcurrant plantation at East Malling Research. It comprised a 5x5 Latin square of 25 plots. Each plot consisted of 6 rows of 6 blackcurrant bushes; two rows of each of three varieties in a three-way split plot design. The varieties were Ben Gairn (reversion resistant but gall mite susceptible), Ben Hope (gall mite resistant but reversion susceptible) and Ben Alder (susceptible to both gall mite and reversion). Five different spray programmes were applied annually as treatments in a Latin square experiment design with five

SID 5 (2/05) Page 7 of 34

replicate plots of each treatment. The treatments applied in the first three years and the new treatments in 2003-2005 are shown in Table 1. Products, active ingredients and formulations and dose rates are given in Table 2.

Table 1. Treatments applied in the experimental plantation used for the IPM experiment at East Malling in 2000-2002 and modified treatments applied in 2003-2005

Treatment Insecticide/acaricide sprays2000-2002

Insecticide/acaricide sprays2003-2005

A 2 sprays sulphur (bud-burst, grape visible)

1 spray sulphur

B 2 sprays of sulphur (bud-burst, grape visible), 1 Dursban just pre-flower

2 sprays sulphur

C 2 sprays sulphur (bud-burst, grape visible), 1 Aphox pre-flower

1 spray sulphur, 1 spray Masai

D 3 sprays Meothrin, pre flower, post flower and + 10-14 days

3 sprays Meothrin, pre flower, post flower and + 10-14 days

E Untreated Untreated

Table 2. Products and their dose rates of application

Product Active ingredient and formulation

Product dose rate/ha

Aphox Pirimicarb 50% w/w WG 560 gDursban Chlorpyrifos 480 g/l EC 1 lUnited Phosphorus Flowable Sulphur Sulphur 800 g/l SC 12.5 lMasai Tebufenpyrad 20% w/v WB 0.5 kgMeothrin Fenpropathrin 100 g/l EC 0.5 l

Sprays were applied with a Holder N5S air assisted sprayer at a spray volume of 500 l/ha. The whole experimental area also received a programme of sprays of fungicides for mildew and leaf spot control. Each year during early flower, each bush was inspected for symptoms of infection by reversion virus disease. The presence or absence of symptoms on each bush was recorded. In the dormant period following each growing season, the blackcurrant gall mite galls on each bush were counted and recorded.

For the end of season counts of gall mite galls, the large differences between varieties and years meant that one over all analyses of variance of the whole data set was inappropriate. Therefore, separate analyses of variance were done on the numbers of gall mite galls per bush, after square root transformation to stabilise variances, for each assessment date on each variety so that comparisons could be made between the spray treatments for each separate data set. Analyses were not done where data contained numerous zero values (Ben Gairn in years 1 and 2, Ben Hope in years 1, 2 and 3). Because of the change in treatments at the start of year 4, the year 3 end of season counts were used as a covariate for adjusting the subsequent year 4, year 5 and year 6 counts. The transformation used was less satisfactory in some analyses than others because the SED values were greater than some of the individual means.

The yearly increase in gall density for each spray treatment on each variety was plotted against time from start of the experiment in years (2000 = 1, 2001 = 2, 2002 = 3 etc.) on a log versus linear scale. These graphs were approximately linear indicating that the increase in gall density was roughly exponential with different rates of increase for the different variety*treatment combinations. After exploration of the data in a variety of ways, it was found that the relationship count = B.exp(k*year) could be fitted to the data on a logarithmic scale, thus fitting the linear relationship log(count) = log(B) + k*year (zero counts were adjusted to 0.005 to allow an overall model to be fitted to the means per bush). Fitting a reduced model where the intercepts log(B) varied only for variety and the slopes k varied only for treatment was as effective as fitting 15 separate lines. Thus, the relative effectiveness of the treatments could be conveniently expressed in terms of the rates k, and the overall varietal differences by the intercepts B. No statistical analyses of the proportions of bushes with reversion virus were done

ResultsAt the end of year 1, small numbers of gall mite galls (0.22-0.5/bush) were present for all treatments on Ben Alder. Only an occasional gall was found on Ben Gairn, but no galls were present on Ben Hope

SID 5 (2/05) Page 8 of 34

(Table 3). Statistical analyses was not appropriate but even at the end of year 1, the Meothrin and untreated treatments on Ben Alder had the highest numbers of galls.

At the end of year 2, a 17.7 fold increase in gall numbers from the previous year had occurred on the untreated Ben Alder plots and statistically significant effects of the spray treatments were apparent. Treatments A, B, and C (the treatments including 2 early sulphur sprays) had significantly smaller numbers of galls than the untreated control, treatment D (3 x Meothrin) having intermediate numbers which did not differ significantly from the untreated control or the 3 sulphur treatments. Small numbers of galls were present on the untreated and Meothrin treated plots on Ben Gairn but not on the plots that received the sulphur treatments. No galls were present on Ben Hope.

At the end of year 3, gall numbers from the previous year on the untreated Ben Alder plots had deceased by 2.3 fold. This decrease was probably due to prolonged periods of heavy rainfall during the migration period in 2002. However, the same treatment effects as in year 2 were apparent in the data.

At the end of years 4, 5 and 6, gall numbers on the untreated Ben Alder plots had increased by 3.1, 8.8 and 3.6 fold respectively (total = 98 fold) from the previous year. Increase factors for the 3 x Meothrin treatment on Ben Alder were 6, 6.2 and 3.9 fold (total = 145 fold) and for the three sulphur treatments (A, B and C) which performed similarly, increase factors were 4.6, 13.8 and 3.4 fold respectively (total = 145 fold). The start of the increase in gall numbers was delayed by a year on Ben Gairn and by two years on Ben Hope. Rates of increase over the previous year in years 4, 5 and 6 on the untreated controls were 68 fold on Ben Gairn and 125 fold on Ben Hope. Throughout the data, the highest numbers of galls were found on the untreated controls, followed by Ben Alder, with fewer on Ben Gairn and the least on Ben Hope. The smallest numbers of galls were found on the three sulphur treatments A, B and C, which performed similarly. Treatment D (3 x Meothrin) had intermediate numbers of galls.

The yearly increase in gall density for each spray treatment on each variety was plotted against time from start of the experiment in years (2000 = 1, 2001 = 2, 2002 =3 etc.) on a log versus linear scale. These graphs were approximately linear indicating that the increase in gall density was roughly exponential with different rates of increase for the different variety*treatment combinations.

The values of the rate parameter k in the relationship count=B*exp(k*year) conveniently indicate the relative efficacy of the treatments (Table 4). It can be seen that the rate of increase for treatments A, B and C are very similar, but those for D and E substantially greater (both are significantly higher than those for A, B and C at p < 0.001). The ratio of the constant B for the varieties Ben Alder : Ben Gairn : Ben Hope is approximately 5 : 2 : 1. (4.8:2.2:1).

Table 3. Mean (n) and mean square root transformed total (√n) numbers of gall mite galls per bush at the end of each year. Means followed by the same letter do not differ significantly in a Duncan’s multiple range test (p = 0.05)

TreatmentBen Alder

(susceptible)Ben Gairn

(reversion resistant)Ben Hope

(gall mite resistant)n √n n √n n √n

Year 1 (2000)A. 2 Sulphur 0.22 0.20 0.03 0.03 0.00 0.00B. 2 Sulphur, Dursban 0.20 0.17 0.02 0.02 0.00 0.00C. 2 Sulphur, Aphox 0.23 0.20 0.02 0.02 0.00 0.00D. 3 Meothrin 0.50 0.41 0.00 0.00 0.00 0.00E. Untreated 0.43 0.36 0.02 0.02 0.00 0.00

SED (12 d.f.)Year 2 (2001)A. 2 Sulphur 2.32 1.37 bc 0.00 0.00 0.00 0.00B. 2 Sulphur, Dursban 2.07 1.14 c 0.00 0.00 0.00 0.00C. 2 Sulphur, Aphox 3.45 1.55 bc 0.00 0.00 0.00 0.00D. 3 Meothrin 5.98 1.94 ab 0.15 0.12 0.00 0.00E. Untreated 7.60 2.56 a 0.35 0.27 0.00 0.00

SED (12 d.f.) 0.251Year 3 (2002)A. 2 Sulphur 0.2 0.12 b 0.0 0.00 b 0.0 0.00B. 2 Sulphur, Dursban 0.1 0.09 b 0.0 0.03 b 0.0 0.02C. 2 Sulphur, Aphox 0.5 0.39 b 0.0 0.02 b 0.0 0.00D. 3 Meothrin 0.8 0.52 b 0.1 0.06 b 0.0 0.03E. Untreated 3.3 1.59 a 1.0 0.68 a 0.3 0.20

SED (12 d.f.) 0.206 0.135

SID 5 (2/05) Page 9 of 34

Year 4 (2003)A. 1 Sulphur 1.2 0.45 b 0.2 0.04 b 0.2 0.11 bB. 2 Sulphur 1.3 0.48 b 0.4 0.17 b 0.1 0.04 bC. 1 Sulphur, 1 Masai 1.2 0.43 b 0.2 0.05 b 0.1 0.05 bD. 3 Meothrin 4.8 1.85 a 1.2 0.59 b 0.9 0.56 aE. Untreated 10.3 2.95 a 5.8 2.15 a 1.4 0.70 a

SED (12 d.f.) 0.415 0.371 0.133Year 5 (2004)A. 1 Sulphur 17.1 2.61 b 1.3 0.14 b 0.6 0.13 bB. 2 Sulphur 15.5 2.14 b 1.1 0.15 b 0.4 0.08 bC. 1 Sulphur, 1 Masai 18.4 2.87 b 1.2 0.16 b 0.6 0.09 bD. 3 Meothrin 29.8 4.76 b 5.8 1.85 b 4.7 1.74 aE. Untreated 91.1 8.59 a 18.6 3.85 a 6.1 2.20 a

SED (12 d.f.) 1.363 0.695 0.256Year 6 (2005)A. 1 Sulphur 52.9 5.88 b 1.0 0.36 c 4.8 1.05 bB. 2 Sulphur 61.5 5.93 b 1.2 0.53 c 1.8 0.41 bC. 1 Sulphur, 1 Masai 61.1 5.59 b 2.9 0.98 bc 2.6 0.59 bD. 3 Meothrin 115.9 9.99 b 22.5 4.02 ab 33.0 5.02 aE. Untreated 329.2 16.69 a 62.8 6.87 a 37.6 5.65 a

SED (12 d.f.) 1.966 1.180 0.527

Table 4. Estimates of the rate parameter k and intercepts B in the equation n=B*exp(k*year) where n = total number of mites present per bush and year is the time in years from the start of the experiment (2000=1, 2001 = 2, etc.)

Variety BBen Alder 0.072Ben Gairn 0.033Ben Hope 0.015Treatment kA 0.741B 0.770C 0.754D 1.415E 1.687

No reversion virus was found in years 1 or 2 (Table 5). The first reversion virus infection was recorded at flowering in year 3 in a total of 5 out of 300 bushes of Ben Alder, when 4 bushes which received treatment C and 1 bush which received treatment D were found to be infected. The incidence of virus infection on Ben Alder increased steadily from year to year thereafter. The proportion of bushes infected was consistently greatest on the untreated control but numbers of infected bushes were similar on the other treatments with no obvious treatment effects. By the end of the experiment in May 2006 practically every bush of Ben Alder was infected with no obvious treatment differences.

Ben Gairn remained completely free of virus throughout the first 6 years of the experiment. However, at the end of the experiment in May 2006, 5 bushes had possible symptoms of reversion virus infection. Samples of infected shoots were sent to SCRI for confirmation of infection, but these tests proved negative. The bushes have been maintained for further tests by SCRI in 2007.

On Ben Hope, the first signs of infection were found in year 4. The distribution was erratic with no obvious effects of spray treatments. The proportion of infected bushes steadily increased in subsequent years but it remained lowest on treatment B and at the end of the experiment was highest on the untreated control.

SID 5 (2/05) Page 10 of 34

Table 5. Total number of bushes (n = 60) showing symptoms of reversion virus at the early flowering growth stage each year

Treatment Ben Alder(susceptible)

Ben Gairn(reversion resistant)

Ben Hope(gall mite resistant)

Year 1 (2000)A. 2 sulphur 0 0 0B. 2 sulphur+chlorpyrifos 0 0 0C. 2 sulphur+pirimicarb 0 0 0D. 3 fenpropathrin 0 0 0E. untreated 0 0 0Year 2 (2001)A. 2 sulphur 0 0 0B. 2 sulphur+chlorpyrifos 0 0 0C. 2 sulphur+pirimicarb 0 0 0D. 3 fenpropathrin 0 0 0E. untreated 0 0 0Year 3 (2002)A. 2 sulphur 0 0 0B. 2 sulphur+chlorpyrifos 0 0 0C. 2 sulphur+pirimicarb 4 0 0D. 3 fenpropathrin 1 0 0E. untreated 0 0 0Year 4 (2003)A. 1 sulphur 2 0 6B. 2 sulphur 5 0 0C. 1 sulphur+ 1 tebufenpyrad 3 0 0D. 3 fenpropathrin 2 0 2E. untreated 6 0 0Year 5 (2004)A. 1 sulphur 11 0 2B. 2 sulphur 18 0 0C. 1 sulphur+ 1 tebufenpyrad 14 0 2D. 3 fenpropathrin 17 0 2E. untreated 22 0 1

SID 5 (2/05) Page 11 of 34

Year 6 (2005)A. 1 sulphur 21 0 12B. 2 sulphur 24 0 0C. 1 sulphur+ 1 tebufenpyrad 22 0 7D. 3 fenpropathrin 32 0 9E. untreated 31 0 7Year 7 (2006)A. 1 sulphur 56 1? 10B. 2 sulphur 56 0 2C. 1 sulphur+ 1 tebufenpyrad 58 0 9D. 3 fenpropathrin 59 1? 10E. untreated 59 3? 28

ConclusionsThis experiment provided a rigorous comparison of the comparative efficacy of the spray treatment*variety combinations in preventing the gall mite infestation and reversion infection and their increases. The plots were small (compared to commercial blackcurrant plantations and the high levels of infestation and infection on the Ben Alder bushes, particularly in the latter years of the experiment, placed adjacent plots under heavy pressure. Differences between treatments and varieties would probably have been greater if plots had been much larger.

Gall numbers increased from year to year, by different factors depending on the variety and the treatment, except in year 3 (2002) when gall numbers decreased. This was probably due to heavy rainfall on several days in late April, May and early June 2002 during the migration period of the gall mite.

The results clearly show that the varieties Ben Gairn and Ben Hope are highly though not completely resistant to gall mite. The ratio of susceptibility in terms of the numbers of galls that developed by the end of year 6 for Ben Alder : Ben Gairn : Ben Hope was 8.7 : 1.7 1. These ratios are somewhat higher than the relative values of the constant B in the relationship count=B*exp(k*year) fitted to the whole data where the ratios were approximately 5 : 2 : 1, but this is likely to be due to the fitting of this overall relationship being carried out on the log scale and thus down-weighting the particularly high values. However, both these results indicate Ben Gairn and Ben Hope to be fairly, though not fully, resistant to gall mite in comparison with the susceptible Ben Alder standard, with Ben Hope being more resistant then Ben Gairn.

The conclusions of this work are as follows: The relative susceptibility of the varieties to gall mite, as quantified by the number of galls

present on them by the end of year 6 or the relative values of the rate of increase parameter B in the relationship n=B*exp(k*year), is as follows: Ben Alder : Ben Gairn : Ben Hope = 4.8-8.7 : 1.7-2.2 : 1

Thus Ben Gairn and Ben Hope are fairly, though not fully, resistant to gall mite in comparison with the susceptible Ben Alder standard, with Ben Hope being more resistant then Ben Gairn.

There was no difference in the performance of the 1 sulphur, 2 sulphur or 1 sulphur then 1 Masai treatments.

Sulphur or sulphur then Masai spray programmes gave significantly better control of gall mite than the standard 3 Meothrin spray treatment.

Gall mite numbers increased exponentially from year to year, 80% of the variation in numbers on the log scale being accounted for by the relationship n=B*exp(k*year) where n is the average number of gall mites per bush, year is the interval in years from the start of the experiment (2000=1, 2001=2, 2002=2 etc), the coefficient B taking differing values for different varieties and k different values for different treatments.

It appeared that heavy rainfall during gall mite migration in 2002 caused numbers of galls to decrease in that year. Otherwise, numbers increased.

On the susceptible variety Ben Alder, reversion virus infection was first seen on a small number of bushes at flowering in year 3. The proportion of bushes infected increased steadily from year to year subsequently until practically all bushes were infected in the spring of year 7. High levels of infection occurred at a similar rate of increase on all plots, despite the treatments with acaricides which did not significantly delay the spread of infection.

Ben Gairn appeared to be fully resistant to reversion virus until the end of the experiment when 5 bushes were found with possible symptoms of infection. However, the presence of infection in these bushes was not confirmed in DNA tests at SCRI.

SID 5 (2/05) Page 12 of 34

On Ben Hope, first appearance of infection by reversion virus infection occurred in year 4, one year later than on the susceptible variety Ben Alder. The subsequent increase in infection occurred at a slightly lower rate on Ben Hope than Ben Alder. By the end of the experiment approximately half the untreated bushes were infected with markedly lower incidences of infection on bushes that received acaricide treatments.

Combining host plant resistance with use of early sulphur acaricide sprays gave a very high degree of gall mite and reversion virus as follows:

Treatment % control of gall mite over 6

years

% control of reversion virusover 6 years

Ben Hope + early sulphur 97.6 88Ben Gairn + early sulphur 99.5 98-100

Varietal resistance needs to be protected by sprays of acaricides to lessen the likelihood of the development of resistance breaking strains of gall mite and/or reversion virus disease.

Objective 1.2. Determine whether sulphur can be used on blackcurrant before and/or after flowering by quantifying the phytotoxic effects of sprays of sulphur including effects on growth and yieldThree replicated field experiments were done to investigate possible phytotoxic effects of foliar sprays of sulphur to blackcurrant in purpose-planted blackcurrant plantations at EMR in 2003, 2004 and 2005. Experimental evidence (see objective 1.3 below) showed that, of a very wide range of materials tested, sulphur was the most effective gall mite acaricide. However, the crop safety of sulphur sprays, particularly when applied during flower or fruit development, was in doubt. The effects of sulphur sprays on growth and yield of different blackcurrant varieties needed to be quantified so that any limits on timings of applications could be determined. This work is reported in full in Cross and Harris (2004a, 2005c, 2005f).

Methods and materialsThe first experiment in 2003 was a factorial comparison of the effect of single sprays of sulphur pre- or post-flower (GS F1-I3) or a programme of 3 sprays (pre-flower, post flower and 14 days later) on the growth and yield of five blackcurrant varieties; Baldwin, Ben Gairn, Ben Hope, Ben Lomond, Ben Tirran. The experimental plantation consisted of ten 4x4 Latin squares, 2 of each of the five blackcurrant varieties. Each Latin square consisted of 16 plots each of 6 blackcurrant bushes in a row. The plantation was planted in spring 2002 with 1 year old rooted bushes that were in their 3rd year of growth at the time the experiment was carried out.

The second experiment in 2004 compared single applications just before grape (= flower bud) emergence (GS C3-D) or at the end of flowering, or of two sprays one at each of these timings, on the yield and growth of the blackcurrant varieties Baldwin, Ben Gairn, Ben Hope Ben Lomond or Ben Tirran. The same experimental plantation was used as in 2003 but the 2004 treatments were re-allocated orthogonally to the 2003 treatments.

The third experiment in 2005 determined effects of two early season foliar sprays of sulphur at the late dormant and first grape visible growth stages and the effect of a range of additional sprays of sulphur and or Masai (tebufenpyrad) on the blackcurrant varieties Ben Hope and Ben Tirran (Table 6). These same treatments were also evaluated in the acaricides trial done in 2005 (See objective 1.3 below). The experiment consisted of six 8x4 arrays of plots, three arrays of each variety, each array containing four replicates of each of the eight treatments. Each plot consisted of six bushes in a row. The plantation was planted in spring 2004 with 1 year old rooted bushes so the bushes were in their 3rd year of growth at the time the experiment was done.

Table 6. Treatments tested in the third sulphur phytotoxicity experiment in 2005

Treatment name Late dormant First grape visible Additional sprays†A SL,SL Sulphur SC Sulphur SC -B SL,SL,M Sulphur SC Sulphur SC MasaiC SL,SL,M+1/3SL Sulphur SC Sulphur SC Masai + Sulphur SC 1/3 rateD SP,SP,M Sulphur DF Sulphur DF MasaiE SL,SL,1/3SL Sulphur SC Sulphur SC Sulphur SC 1/3 rate

SID 5 (2/05) Page 13 of 34

F SL,SL,1/3SLx3 Sulphur SC Sulphur SC 3 x Sulphur SC 1/3 rateG SL,SL,1/10SLx3 Sulphur SC Sulphur SC 3 x Sulphur SC 1/10 rateH Untreated Untreated Untreated Untreated† single sprays were applied at the end of flower, the programmes of sprays were applied at approximately 2 week intervals following on from the first grape visible spray

In the first two experiments, an 800 g/l sprayable concentrate (SC) sulphur formulation (Headland Sulphur) was applied at a dose of 12.5 l product/ha throughout. In the third experiment, the full dose rates for application of the sprayable concentrate and dry flowable (DF) formulations of sulphur were 10 litres and 10 kg of 80% product respectively. In all three experiments, sprays were applied at 500 l/ha with a hand lance, which gave complete cover.

Temperature conditions at the time of application of sulphur sprays were potentially critical as grower observations indicated that phytotoxicity was most severe in hot conditions. Generally, air temperatures at the time of spraying were < 20 ˚C except for the last sprays of sulphur in each experiment when air temperatures were 22-22.5, 25 and 26-30 ˚C at the time of spraying in the first, second and third experiments respectively.

The effects of the treatments were assessed by regularly examining the bushes for visual symptoms of phytotoxicity, measuring yields of hand picked and dropped ripe fruit and the length of the season’s extension growth in the dormant period after the sprays had been applied. Analysis of variance was done on the data, with appropriate square root or log transformation to stabilise variances as necessary. Covariance adjustment for the total length of extension growth before the treatments were applied was also done.

ResultsIn the first experiment, slight phytotoxicity symptoms were apparent in early June 2003. The symptoms consisted of slight yellowing of the foliage which was most pronounced on the plots where the pre-blossom spray had been applied and on the plots where both the pre- and post-blossom spray had been applied. Symptoms were very slight on the plots that had received the post blossom spray only. There were large statistical (P = 0.002) differences in the average yields of the different varieties. Ben Hope had by far the greatest mean yield (7.49 t/ha), Ben Gairn the smallest (2.89 t/ha). The ranking of the varieties in increasing order of yield was Ben Gairn < Baldwin < Ben Lomond < Ben Tirran < Ben Hope. The sulphur spray treatment factor also had highly significant effects (P = 0.003). The programme of three sprays caused the greatest reduction in overall mean yield compared to the untreated control by 14.6% (Table 7). The single pre-blossom spray caused the second greatest reduction in the overall mean yield (by 11.4%) but the single post blossom spray treatment did not reduce the overall mean yield compared to the control. The greatest percentage reductions in yield occurred on Baldwin suggesting that this variety may be more susceptible to sulphur than the others, though the differences were not statistically significant. There were statistically significant differences between the shoot growth of the varieties but the effects of the spray treatments and their interaction with variety were not significant.

Table 7. Percentage increase (positive values) or percentage decrease (negative values) in mean yield (upper table) compared to the untreated control in the first sulphur phytotoxicity experiment in 2003

Variety Sulphur spray treatmentPre-flower Post-flower 3 spray programme

Ben Gairn -5.6 +3.0 -14.2**Ben Lomond -0.8 +3.9 -5.3*Ben Hope -16.0** -7.0* -20.1**Baldwin -19.5** -18.8** -34.1**Ben Tirran -10.8 +10.9 + 0.3Mean -11.4** -0.2 -14.6**Significantly less than the control for that variety * P ≤ 0.05, ** P ≤ 0.01

In the second experiment in 2004, phytotoxicity symptoms were clearly visible on all the varieties except Ben Tirran in mid-June. The pre-grape emergence + end of flower sulphur treatments caused the most severe phytotoxicity followed by the end of flower treatment with the least phytotoxicity being caused by the pre-grape emergence treatment (Table 8). However, the severity of the symptoms differed markedly between varieties being most severe on Ben Gairn where the lower leaves were severely blackened followed by Baldwin. Ben Hope and Ben Lomond showed only slight phytotoxicity

SID 5 (2/05) Page 14 of 34

symptoms and the effects of the treatments were barely perceptible on the Ben Tirran. The phytotoxicity caused leaf drop and appeared to slightly stunt the growth of the plants.

There were strong treatment effects due to both variety and sulphur treatment on yield. The general interaction between these treatments was not quite significant (P = 0.087) even with the more rigorous analysis on the logarithmic scale, but there did appear to be minor differences in the effects of treatments on the different varieties. The yields of Ben Lomond and Ben Tirran did not appear to be significantly reduced by any of the sulphur treatments (Table 9). The pre-grape + after flower sulphur treatment (treatment 3) was the most phytotoxic, reducing yield by 19% averaged across all varieties, but by 27% on the most susceptible variety, Baldwin. The end of flowering treatment reduced yield by 13.5% on average with the strongest treatment effects on Baldwin. The pre-grape emergence treatments to Baldwin significantly (P < 0.05) reduced yield by 17% compared to the untreated control for that variety, but did not significantly reduce yields of the other varieties.

There were overall differences between varieties in the mean length of extension shoot, the numbers of shoots per bush. However, the spray treatment factor was not significant for these variates. There were significant (P = 0.036) differences between varieties in total extension growth per bush and also significant effects of the sulphur treatment (P = 0.030). In respect of the sulphur treatments, there was slight evidence of greater overall growth for the untreated controls than for the two single spray treatments, but the logical pattern was somewhat upset by the combined treatment having the second highest total growth (although this was not statistically significant from any other treatment).

Table 8. Relative severity of phytotoxicity symptoms on the 14 June 2004

VarietySulphur treatment

Pre-grape emerged End flower Pre-grape + after flower

Baldwin Very slight Slight ModerateBen Gairn Moderate Severe Very severeBen Hope None Very slight SlightBen Lomond None Very slight SlightBen Tirran None None None

Table 9. Percentage increase (positive values) or percentage decrease (negative values) in mean yield (upper table) compared to the untreated control in the second sulphur phytotoxicity experiment in 2004

Variety Sulphur spray treatmentPre-grape emerged End flower Pre-grape + end flower

Ben Gairn -16.9 -32.8 -27.2Ben Lomond -9.0 -17.2 -34.3Ben Hope -5.1 -10.7 -17.1Baldwin 9.5 -1.8 -12.3Ben Tirran -7.9 -7.9 -9.6Mean -5.9 -14.1 -20.1Significantly less than the control for that variety * P ≤ 0.05, ** P ≤ 0.01

In the third experiment in 2005, no visual phytotoxicity symptoms were observed. The analyses of variance showed strong, statistically significant (p = 0.032) treatment effects for Ben Hope (Table 10). LSD testing (P = 0.05) indicated that treatments C, D, E and G all significantly reduced yield (P<0.05) compared to the untreated control. On average, these treatments reduced yield by 20% compared to the untreated control. Reduction by treatment F was not quite significant at the 5% level, but very nearly so. The fact that treatment D reduced yield significantly whereas treatment B did not suggests that the two early season sprays of the DF formulation of sulphur caused phytotoxicity whereas the SC formulation did not. All the treatments with additional sulphur SC sprays reduced yield significantly, indicating that the application of sulphur SC at these later timings causes phytotoxicity to Ben Hope, even when only a single spray at 1/3 dose is applied. Treatment effects were not significant for Ben Tirran (p = 0.885).

SID 5 (2/05) Page 15 of 34

Table 10. Mean yield (t/ha) in the third blackcurrant phytotoxicity experiment in 2005

Treatment and timing Ben Hope Ben TirranA SL,SL 6.54 2.88B SL,SL,M 7.10 3.57C SL,SL,M+1/3SL 5.54* 3.58D SP,SP,M 5.56* 3.25E SL,SL,1/3SL 5.53* 3.21F SL,SL,1/3SLx3 5.80† 3.10G SL,SL,1/10SLx3 5.69* 3.01H Untreated 7.00 3.26Fprob 0.032 0.885SED (77 d.f.) 0.623 0.536LSD (P=0.05) 1.241 1.067*Significantly less than the control P < 0.05†Significantly less than control P < 0.06

ConclusionsIn the first experiment, a single foliar spray of sulphur applied just before flowering was phytotoxic to 2 year old bushes of the blackcurrant varieties Ben Gairn, Ben Hope, Ben Lomond, Baldwin and Ben Tirran causing leaf discoloration, an 11.4% reduction in yield and possible slight reductions in growth. A single spray just of sulphur just after flower did not significantly reduce yield or growth. A programme of 3 sprays, one just pre-flowering, one post flowering and a third approximately 14 days later, caused greater phytotoxicity than the single pre-flowering spray, reducing yield on average by 14.6%. The data suggests that Baldwin may be more sensitive to sulphur than the other varieties, but this could not be proven by detailed statistical analyses. In this experiment, phytotoxicity from post flowering sprays was less pronounced than in the 2004 experiment. This was probably because temperatures at the time of application were lower, 14 ºC and 21 °C for the sprays at the end of flower and 2 weeks later respectively.

In the second experiment, the treatments caused clear visible symptoms of phytotoxicity on all the varieties except Ben Tirran. The two spray (pre-grape emergence + end of flower) treatment caused the most severe phytotoxicity, followed by the end of flower treatment with the least phytotoxicity being caused by the pre-grape emergence treatment. The severity of the visual phytotoxicity symptoms differed markedly between varieties. Symptoms were most severe on Ben Gairn where the lower leaves were blackened, followed by Baldwin. Ben Hope and Ben Lomond showed only slight symptoms. The effects of the treatments were barely perceptible on the Ben Tirran. The yields of Ben Lomond and Ben Tirran did not appear to be reduced significantly by any of the sulphur treatments. The pre-grape + end of flowering sulphur treatment reduced yield by 19% averaged across all varieties, but by 27% on Baldwin. The single end of flowering treatment reduced yield by 13.5% on average with strongest treatment effects on Baldwin. The pre-grape emergence spray reduced the yield of Baldwin by 17%, but did not significantly reduce the yields of the other varieties. The sulphur treatments did not affect mean length of extension shoots or the numbers of shoots per bush. There was some evidence of greater overall growth for the untreated controls than for the two single spray treatments, but the logical pattern was somewhat upset by the combined treatment having the second highest total growth (although this was not significantly different from any other treatment). The severity of the phytotoxicity caused by the end of flower sprays may have been exacerbated by the high temperatures (20.5-25.0 ºC) when treatments were applied.

In the third experiment, none of the treatments tested caused any visual symptoms of phytotoxicity on either Ben Hope or Ben Tirran. None of the treatments had any significant effects on the yield of Ben Tirran, though there was some evidence that the two treatments that included the programmes of 3 additional sprays of Sulphur SC at 1/3 or 1/10 rate may have slightly decreased total extension growth of Ben Tirran. Two early season sprays of sulphur SC, at the bud burst and first grape visible growth stages did not cause significant phytotoxic yield or growth reduction to Ben Hope. Sprays of Masai applied in hot conditions (air temp 26-30 ºC) at end of flowering were not phytotoxic to Ben Hope. A single additional spray of sulphur SC at the end of flowering, or the programmes of 3 sprays of sulphur SC at 1/3 or 1/10 rate at fortnightly intervals, were phytotoxic to Ben Hope, reducing yield by 20% and significantly reducing extension growth. Application of the end of flowering sprays in hot conditions (air temp 26-30 ºC) is likely to have caused or exacerbated the phytotoxic effects. The overall conclusions of this work are:

SID 5 (2/05) Page 16 of 34

Sulphur 800 g/l SC applied at 10 or 12.5 l product in 500 l water/ha can cause substantial phytotoxicity to blackcurrant but the severity of visual symptoms, yield loss and growth reduction depend greatly on variety, temperature conditions and growth stage of application.

Of the five varieties investigated, Baldwin and Ben Gairn were most susceptible, Ben Hope and Ben Lomond intermediate and Ben Tirran only slightly susceptible to sulphur phytotoxicity.

A sulphur spray applied at the late dormant/bud-burst growth stage did not cause phytotoxicity to any variety.

A sulphur spray applied at the first grape visible growth stage caused significant phytotoxicity to the susceptible varieties Baldwin and Ben Gairn. Use of sulphur on these varieties should be avoided if possible. Phytotoxic effects of applications at this time to Ben Lomond, Ben Hope and Ben Tirran were at most small and insignificant.

Later sprays, either pre- or post blossom were phytotoxic, the severity of losses depending on variety and temperature conditions. On the susceptible varieties (Baldwin and Ben Gairn), severe leaf blackening, chlorosis and drop and yield losses of up to 27% may occur, depending on temperature conditions at the time of application. On moderately susceptible varieties (Ben Lomond, Ben Hope) phytotoxicity symptoms are likely to be less severe and losses more moderate (~10%).

Applications of sulphur when temperatures are high (> 25 ˚C) exacerbated phytotoxic effects. Application in these conditions should be avoided. A sensible, cautious approach would be to avoid making applications if temperatures are > 20 ˚C.

The preferred gall mite treatment developed in this project of a spray of sulphur 800 g/l at 10 kg /ha at bud burst, followed by a second spray just before grape visible (except on Baldwin and Ben Gairn where this spray should be omitted) in at risk plantations supplemented with an application of Masai just before or just after flowering is comparatively crop safe and is unlikely to cause significant phytotoxicity.

Objective 1.3. Identify an alternative selective acaricide to fenpropathrin for control of gall mite at the peak of migration that is not phytotoxic to blackcurrantThe objective was to test a range of acaricide treatments for control of blackcurrant gall mite to find alternatives to Meothrin, which is being withdrawn due to the EU pesticides reviews and to optimise the use of sulphur for gall mite control. Previous experiments at East Malling have repeatedly shown sulphur to be a highly effective gall mite acaricide. However, sulphur has been shown to be phytotoxic to blackcurrant, causing leaf scorch and significant reductions in yield, especially when used during flower or fruit development (see objective 1.2 above). This work is reported in full in Cross and Harris (2004b, 2005b, 2005d, 2006a).

Methods and materialsReplicated field experiments were done in purpose planted experimental plantations at EMR in 2003, 2004, 2005 and 2006 to evaluate the effect of a wide range of programmes of acaricide spray programmes on the emergence of mites from galls and on the number of new galls formed by the end of the growing season.

In the 2003 experiment, thirteen treatments were compared (Table 11). Seven of the treatments were two foliar sprays of different acaricide products, the first spray at the predicted start of the migration of the gall mite shortly after the predicted peak of the migration (Timing S). The EMR gall mite migration forecasting model (Cross & Ridout, 2001) was used to predict the date of the start and of the peak of the migration. A single spray of sulphur in admixture with the latex adjuvant Bond at the start of the migration (Timing P) was included as a comparison with the two sulphur + Bond spray treatment. Two further treatments were a programme of 4 sprays of the Botrytis fungicide Elvaron Multi at approximately equal intervals through the flowering period (Timing B) and a programme of 5 sprays of the Vertalec formulation of the entomopathogenic fungus Lecanicillium longisporum at approximately equal intervals throughout the migration period (Timing M). A full standard traditional commercial treatment comprising two early season sprays of sulphur followed by three sprays of Meothrin at the standard traditional timings (just pre flower, just post flower and 10-14 days later) (Timing F) was included as the positive control. Water only (Timing S) and untreated treatments were provided as negative controls.

Two experiments were done in 2004. Experiment 1 evaluated 12 treatments including an untreated control (Table 12). Eight of the treatments consisted of a late dormant spray of sulphur SC, 4 with an additional sulphur and 4 with an additional Masai spray at first grape visible. Each group of 4 treatments had an additional Masai spray at peak mite emergence (in mid-flower) or 14 days later. In

SID 5 (2/05) Page 17 of 34

each group of 4 treatments, 2 treatments had an additional sulphur SC or Masai 7 days later. The remaining treatments were sulphur SC at the late dormant growth stage with and without the adjuvant Bond, and sulphur SC at the late dormant growth stage plus a programme of five 1/3 rate sulphur SC sprays at approximately 2 week intervals throughout the migration period. Experiment 2 compared the efficacy of the 2 kg/ha label recommended dose of the sulphur 80% w/w DF formulation Kumulus with the higher dose (10 kg/ha) normally used commercially (Table 13).

The experiment in 2005 evaluated two early season sulphur SC sprays supplemented with additional sulphur SC or Masai sprays. A treatment with a dry flowable (DF) formulation and a programme of 3 sprays of the novel acaricide acrinathrin were included (Table 14). The 2006 experiment aimed to validate the findings of previous trials and evaluate the novel acaricides Oberon, Floramite, Kanemite, UKA378b and Vertalec (±Codacide oil) (Table 15).

Standard rates of application of the products tested are given in Appendix Table 1. All sprays were applied with a hand lance, at a volume rate of 1000 l/ha in the 2003 experiment and 500 l/ha in the other experiments.

The emergence pattern of gall mite was monitored using miniature sticky traps (Cross & Ridout, 2001). A total of 32 traps were set in early March, each above a gall on the variety Ben Lomond in the untreated plots. The numbers of gall mites caught in each trap were counted at 2-4 day intervals until early June when the migration had ceased. The EMR gall mite migration forecasting model (Cross & Ridout, 2001) was used to predict first, 5% and peak emergence, based on daily maximum and minimum air temperatures recorded by weather stations in Kent.

The effects of the treatments on blackcurrant gall mite were assessed in two ways, (1) by counting the numbers of mites emerging from galls that successfully migrated a distance of 5 cm to miniature sticky traps (Cross & Ridout, 2001); (2) by counting the number of galls present per plot in the dormant period before the experiment began and again in the dormant period after the experiment was completed. For practical reasons, two separate experiments were done for each of these assessments. For the miniature sticky trap assessment, an experimental plantation of mature Ben Lomond bushes heavily infested with gall mite with a randomised complete block experimental design with four replicates was used. Emerging mites were captured using four miniature sticky traps in each plot, each set 5 cm above a gall on a separate shoot. Plots were single bushes. Thus, there were 16 miniature sticky traps per treatment, with a double replicated untreated control. The mites captured in each trap were counted and the traps refreshed at 7 day intervals throughout the migration period. For the pre and end of season gall counts, a separate experimental plantation of young, lightly infested bushes with a randomised complete block experimental; design with 16 replicated was used. Plots were single bushes. The total numbers of galls present on each plot were counted in the dormant season before the experiment commenced and again in the dormant period after the season when the experiment was completed. Pre- and end of season gall counts were also done in the separate experimental plantation used for the miniature sticky trapping.

Total catches of gall mites in the miniature sticky traps and the end of season gall counts were subjected to analysis of variance, in the latter case with covariance adjustment for the pre experiment gall counts. Square root or log transformation of the data was used to stabilise variances where required.

ResultsThe dates of first, 5% and 50 % emergence predicted by the gall mite emergence model of Cross and Ridout (2001) based on daily maximum and minimum air temperatures were in reasonably close agreement with observed dates (Table 11).

Table 11. Observed and predicted dates of first, 5% and 50% gall mite emergence in 2003-2006First emergence 5% emergence 50% emergence

Observed Predicted Observed Predicted Observed Predicted2003 28 Mar 23 Mar 12 Apr 11 Apr 25 Apr 26 Apr2004 5 Apr 1 Apr 14 Apr 15 Apr 27 Apr 29 Apr2005 28 Mar – 4 Apr 30 Mar 16 Apr 13 Apr 29 Apr 2 May2006 17-19 Apr 8 Apr 24 Apr 20 Apr 5 May 4 May

Analysis of variance of the square root transformed total numbers of mites captured per trap (Table 12) showed that the water, Vertalec, Elvaron Multi, 2 Mavrik (treatment F) and 2 Meothrin (treatment H) treatments did not significantly reduce the total numbers of mites captured per gall compared to the untreated control. All the other treatments significantly reduced the numbers captured but the

SID 5 (2/05) Page 18 of 34

differences between these treatments were not significant. The effective treatments appeared to reduce the numbers of mites captured throughout the migration period and not simply daily emergence. The numbers of galls present at the end of the season (adjusted for the pre-experiment count) (Table 5) showed almost the same treatment effects, though for this variate both the 2 Mavrik and 2 Meothrin treatments significantly reduced gall numbers compared to the untreated control but the reduction by the 2 Masai treatment was not quite statistically significant. The 2 sulphur then 3 Meothrin and the 1 sulphur then 1 Masai treatments (treatments A and B) ranked as being the most effective in both data sets.

SID 5 (2/05) Page 19 of 34

Table 12. Mean (n) and mean square root (√n) total numbers of mites caught per trap and mean numbers of galls per bush at the end of the season on Ben Lomond and Ben Tirran and mean square root (√n) number in the 2003 gall mite acaricide experiment. End of season gall counts are adjusted for the pre-trial count

Treatment No. of mites/trap End of season galls/bushn √n Lomond Tirran Mean Mean √n

A 2 Sulphur, 3 Meothrin F 2.1 1.31 10.2 6.0 8.1 2.56B 1 Sulphur, 1 Masai S 3.7 1.69 10.2 5.8 8.0 2.72C 1 Sulphur + bond P 10.1 2.79 9.6 13.2 11.4 3.02D 2 Sulphur + bond S 11.9 3.02 13.3 10.9 12.1 3.12E 2 Sulphur S 7.1 2.38 10.9 16.5 13.7 3.28F 2 Mavrik S 23.6 4.08 13.6 16.0 14.8 3.44G 2 Masai S 11.8 3.02 18.0 21.6 19.8 4.07H 2 Meothrin S 22.3 4.46 11.2 22.2 16.7 3.58I 2 Sequel S 8.9 2.65 13.9 19.4 16.6 3.71J 4 Elvaron Multi B 118.9 7.97 16.1 26.6 21.3 4.13K 5 Vertalec M 43.4 6.14 26.4 31.1 28.7 4.87L 2 Water S 70.3 7.04 22.2 34.7 28.5 4.92M Untreated - 48.1 6.03 22.5 30.0 26.3 4.77

Fprob <0.001 <0.001Sed (>177 d.f.) 1.054 1.054

F = 2 sprays of sulphur at bud break and first grape emerged then 3 sprays of Meothrin just before flowering, just after flowering and 10-14 days later;S = standard at the start and peak of migrationB = during the flowering period as for BotrytisM = 5 sprays at equal intervals throughout the migration period

In experiment 1 in 2004, all the spray treatments greatly reduced numbers of mites captured compared to the untreated control (p < 0.001) (Table 13). However, most of the effect appears to have been caused by the first sulphur spray applied in the late dormant period. Comparing treatment I (dormant sulphur alone) with treatments A-H (dormant sulphur + additional sprays) suggests that the additional sprays did have some benefit but the differences for any individual treatment were not statistically significant. Treatment K (full dose sulphur on 26 March followed by the programme of five 1/3 dose sulphur sprays performed well. There was no evidence of benefit of addition of Bond to the 26 March sulphur spray. All the chemical treatments significantly (p < 0.01) reduced the end of season gall count compared to the untreated control (Table 13). The programme of sulphur sprays stood out as being the most effective treatment. The early sulphur followed by varying numbers of sulphur or Masai treatments (treatments A-H) all performed similarly. Masai performed similarly to sulphur for the latter sprays though the mean of the four treatments A-D which had sulphur as the second spray (mean = 3.33) was smaller than the mean of the 4 treatments E-H where Masai was the second spray (mean = 4.15). Though the values for all these treatments were lower than the mean for the single sulphur spray, the differences were not statistically significant at the p = 0.01 level indicating that the first sulphur spray gave the most benefit. There was no evidence of any significant benefit from the addition of Bond.

In experiment 2 in 2004, both the high and the low (recommended) dose Kumulus treatments reduced the numbers of migrating gall mites captured by > 90%, but there was no significant difference between the doses (Table 14). Neither the high nor the low dose Kumulus treatments significantly affected the end of season gall count. The mite migration data suggests that the treatments were applied too late and that the early season part of the migration is of greatest importance. This finding is corroborated by the results of experiment 1.

SID 5 (2/05) Page 20 of 34

Table 13. Mean total (n) and mean Log10(n+1) total numbers of mites captured per gall from 7 April – 9 June 2004 and end of season gall counts per bush (n) and mean square root transformed count per bush (√n) on 30 November 2004 in experiment 1. Values have been covariance adjusted for the pre-season gall count

Treatment† and timing‡ Mites captured/gall Ends of season galls/bush1 2 3 4 5 n Log10(n+1)* n √n *S S M S 2.0 0.313 a 3.37 1.685 abcS S M M 5.4 0.523 a 3.62 1.708 abcS S M 3.7 0.368 a 3.53 1.800 abcS S M 3.0 0.323 a 2.79 1.307 abS M M S 4.1 0.498 a 4.09 1.735 abcS M M M 14.9 0.523 a 3.32 1.700 abcS M M 4.9 0.663 ab 3.31 1.658 abcS M M 6.9 0.611 ab 5.88 2.366 cS 7.8 0.584 ab 6.82 2.362 c

S+B 13.8 0.955 b 4.60 2.005 bcProg of S sprays 4.1 0.516 a 1.68 1.082 aUntreated 157.5 1.972 c 20.18 4.170 d

Fprob <0.001 <0.001SED (181 df) – Comparisons with control 0.154 0.2985

Other comparisons 0.178 0.3447† S = sulphur SC, M = Masai, S+B = sulphur SC + Bond‡ 1 = late dormant, 2 = 1st grape visible, 3 = peak emergence (mid flower), 4 = peak emergence + 14 days (end flower); 5= peak emergence + 21 days*Means followed by the same letter do not differ significantly (Duncan’s multiple range test, P = 0.05)

Table 14. Mean total (n) and mean Log10(n+1) total numbers of mites captured per gall from 7 April – 9 June 2004 and end of season gall counts per bush (n) and mean square root transformed count per bush (√n) on 30 November 2004 in experiment 2. End of season gall counts have been covariance adjusted for the pre-season gall count

Treatment† Mites captured/gall Ends of season galls/bushn Log10(n+1) n √n †

3 x sulphur DF low 17.5 0.914 288 15.763 x sulphur DF 14.3 0.922 286 15.58Untreated 176.5 2.059 214 14.04

Fprob <0.001 0.582SED (>41 df) - Comparisons with control 0.167 1.805

Other comparisons 0.192 1.805† Sprays applied at first grape visible, peak mite emergence (mid flower) and peak mite emergence + 14 daysSulphur DF low = at 2 kg/ha + 0.01% Agral

In the 2005 experiment, all the spray treatments significantly reduced total numbers of mites captured compared to the untreated control (P<0.001) (Table 15). The reduction was by >92% for all treatments except the Acrinathrinx4 treatment which reduced total numbers captured by 77%. The 2 early sulphur then programme of three 1/3 rate sulphur sprays stood out as being the best treatment, though the reduction in gall mite numbers did not differ significantly from the two early sulphurs then one 1/3 rate sulphur at the end of flowering treatment. The two early sprays of sulphur plus three 1/10 rate sulphur sprays after flowering was marginally, though significantly, less effective than the where the 1/3 dose sulphur spray programme was used. No benefit from the Masai was apparent nor benefit from using the dry flowable (DF) formulation of sulphur versus the sprayable concentrate (SC) formulation. Indeed, no statistically significant benefit was achieved from any of the post flowering treatments except treatment the three additional 1/3 rate sulphur sprays. The analysis of variance of the log10(n+1) transformed end of season gall counts, covariance adjusted for the pre-season count, had highly significant treatment effects (P<0.001) and showed a similar pattern of effects to the total catches in the miniature sticky traps. All the treatments except acrinathrin reduced end of season gall numbers compared to the untreated control. The two early season sulphur sprays then a programme of 3 x 1/3 rate sulphur sprays stood out as the most effective treatment, though it did not differ significantly from the other effective treatments.

SID 5 (2/05) Page 21 of 34

Table 15. Mean total (n)‡ and mean Log10(n+1) total numbers of mites captured per gall from 7 April – 6 June 2005 and end of season numbers of galls per bush (covariance adjusted for pre-season count)

Treatment† and timing‡ Mites / gall in sticky traps End of season galls/bush1 2 3 N* Log10(n+1)# N* Log10(n+1)#S S 21.7 1.356 c 18.0 1.278 bcS S M 19.8 1.318 c 15.8 1.224 bcS S M+S 1/10 20.6 1.335 c 22.0 1.361 bc

SD SD M 17.8 1.275 c 11.1 1.082 cdS S S 1/3 15.4 1.216 cd 11.5 1.096 cdS S 3x S 1/3 8.8 0.989 d 6.9 0.900 dS S 3xS 1/10 19.3 1.308 c 10.7 1.067 cd

A 3xA 62.9 1.806 b 28.3 1.467 abUntreated control 277.3 2.444 a 45.7 1.669 a

Fprob <0.001 < 0.001SED (>100 df) – Comp with control 0.0982 0.1243

Other comparisons 0.1134 0.2465†S = sulphur SC, M = Masai, SD =sulphur DF, A = Acrinathrin‡ 1= at late dormant growth stage; 2= at first grape visible; 3 = for single sprays = at end of flower, = otherwise programme of 3 sprays at 2 week intervals with last spray at end of flower*Back-transformed values#Means followed by the same letter do not differ significantly (Duncan’s multiple range test, P = 0.05)

In the 2006 trial, no mites were recorded in any of the traps on 12 April but small numbers were apparent on 19 April. Emergence progressed rapidly with steep increases on the untreated control in late April and early May but with the rate of emergence declining thereafter. The Codacide, Oberon, Floramite and Kanemite treatments clearly had no or, at best, only modest affects on the numbers and rates of emergence of mites. Vertalec appeared to have some affect at first but this subsequently broke down. Sulphur, sulphur + (early) Masai and sulphur + (late) Masai all greatly suppressed the rate and numbers of mite emergence, but small numbers of mites were recorded on all dates after the start of emergence. UKA378b and Vertalec + Codacide were partially effective. The analysis of variance of the log10(n+1) transformed total catches of mites showed strong treatment affects (P<0.001) (Table 16). However, only the sulphur, sulphur + early Masai and sulphur + late Masai treatments reduced total numbers significantly compared to the untreated control. In the experiment on the young lightly infested bushes in KF306, the analysis of variance of the log10(n+1) transformed numbers of gall per bush covariance adjusted for the pre-season count showed highly significant treatment affects (Table 16). Sulphur, sulphur + Masai and UKA378b were clearly the most effective reducing end of season gall counts by > 90%. Codacide and Vertalec + Codacide also reduced end of season gall numbers significantly compared to the untreated but only by 27% and 45% respectively. Oberon, Floramite, Kanemite, Vertalec were ineffective. In the experiment on the heavily infested mature bushes in plot KF288, treatment affects were not quite significant at the P = 0.05 level (Table 16). However, the treatment means showed similar trends to those in the experiment in the lightly infested plantation. Sulphur, sulphur + Masai, UKA378b had the lowest end of season gall counts, though numbers of galls were only reduced by 32-55% by these treatments.

SID 5 (2/05) Page 22 of 34

Table 16. Mean (n) and mean log10(n+1) transformed total numbers of gall mites captured per miniature sticky trap and end of season gall counts in the experiments in the lightly and heavily gall infested plantations in 2006. End of season gall counts have been covariance adjusted for the pre-season count

Treatment† and timing‡ Mites/sticky trap Galls / bushLight infestation Heavy infestation

1 2 3 4 n Log10(n+1)* n Log10(n+1)* n Log10(n+1)S + + 11 0.769 cd 0.47 0.106 c 107 1.919S+M + + + 14 0.501 d 0.57 0.161 c 120 1.988S+ M + + + 5 0.504 d 0.51 0.099 c 162 1.904Coda + + + + 221 1.800 a 5.21 0.688 b 181 2.265Oberon + + 448 1.994 a 8.36 0.835 a 504 2.517Floramite + + 203 1.857 a 6.54 0.794 a 637 2.606Kanemite + + 263 1.721 ab 6.98 0.807 a 409 2.552UKA378b + + 76 1.604 ab 0.63 0.165 c 127 1.867Vertalec + + + + 194 1.346 abc 6.36 0.742 ab 359 2.194Vertalec+Coda + + + + 46 1.116 bcd 3.91 0.544 b 465 2.258Untreated 313 1.440 ab 7.09 0.802 a 239 2.286

Fprob 0.001 <0.001 0.056SED (>29 df) – comparisons with control 0.2391 0.0987 0.2615

SED (>29 df) – other comparisons 0.2761 0.0987 0.2615LSD (P = 0.05) - comparisons with control 0.4718 0.1950 0.5248

LSD (P = 0.05) – other comparisons 0.5448 0.1950 0.5248† S = sulphur SC, M = Masai, Coda = Codacide oil‡ 1 = at bud burst; 2 = at 1st grape visible; 3 = pre flower; 4 = post flower* means followed by the same letter do not differ significantly in a Duncan’s multiple range test

ConclusionsIn the first experiment in 2003, the water, 5 x Vertalec, 4 x Elvaron Multi, 2 x Mavrik and 2 x Meothrin treatments did not significantly reduce the total numbers of mites captured per gall compared to the untreated control. All of the other treatments significantly reduced the numbers captured but the differences between these treatments were not statistically significant. The numbers of galls present at the end of the season showed almost the same treatment effects, though in this instance both the 2 x Mavrik and 2 x Meothrin treatments significantly reduced gall numbers compared to the untreated control, but the reduction by the 2 x Masai treatment was not quite statistically significant. The 2 x sulphur then 3 x Meothrin and the ‘1 sulphur then 1 Masai’ treatments ranked as being the most effective treatments both for reducing the numbers of migrating mites and the numbers of galls. The sulphur then Masai treatment is a promising replacement for the standard 2 x sulphur then 3 x Meothrin treatment. Use of sulphur in the early and later parts of the season, supplemented with Masai at times during the migration when sulphur is likely to be phytotoxic was identified as the most promising strategy for commercial gall mite control.

In the 2004 experiments, treatment with sulphur (12.5 l of 800 g/l SC in 1000 l water/ha) at the late dormant growth stage of Ben Lomond gave approximately 66% control of gall mite galls. A supplementary spray of sulphur (at the same rate) or Masai (0.5 kg in 1000 l water/ha) at the first grape visible growth stage plus one or two additional sprays of Masai at the peak of mite emergence on 6-7 May or 14 or 21 days later, or with an additional spray of sulphur at this latter timing improved the reduction in the number of galls formed to 71-86%, but did not give complete control. The best control was achieved by the late dormant spray of sulphur SC followed by five 1/3 rate sprays of sulphur SC spanning the mite emergence period which reduced the number of galls formed by 92%. Programmes of 3 sprays of sulphur 80 % w/w DF (Kumulus DF) at a dose of 2.0 or 12.5 kg in 1000 l water/ha at first grape visible (15-16 April), at peak mite emergence (6-7 May) and 14 days later failed to reduce the number of galls formed by the end of the season. Catches of mites on miniature sticky traps indicated that, although these treatments reduced the total numbers of migrating mites by over 90%, they did not control the early part of the migration which occurred before the first spray was applied. Poor reduction in the numbers of galls can thus be attributed to the failure to control the mites in the early part of the migration period. Although Masai is probably slightly less effective than sulphur, use of Masai after the first grape visible growth stage may be preferable because of possible phytotoxic effects of sulphur. No obvious visual symptoms of phytotoxicity were observed in this experiment. However, the phytotoxicity of programmes of reduced rate sulphur sprays after the first grape visible growth stage needed to be investigated.

SID 5 (2/05) Page 23 of 34

In the 2005 experiment, all the treatments tested controlled blackcurrant gall mite though the programme of 4 sprays of acrinathrin was less effective. The acrinathrin programme reduced end of season gall numbers by 38% and numbers of migrating mites by 77%. Two early sprays of flowable sulphur (at bud burst and first grape visible) reduced numbers of migrating mites by 92% and end of season gall numbers by 61%. Control of mites and galls by the dry flowable (DF) formulation of sulphur did not differ significantly from the degree of control with the suspension concentrate (SC) formulation. Addition at petal fall of a spray of Masai, of a single spray of sulphur at 1/3 rate, or of both did not significantly improve control over that achieved with the two early sprays of sulphur alone. Significantly improved control (85% and 97% for end of season gall numbers and total mites captured respectively) was obtained by addition to the bud burst and first grape visible sprays of a programme 3 x 1/3 rate sprays of sulphur at approximately 2 week intervals starting at the end of flowering. No improvement in control was obtained from a the additional programme of 1/10 rate sulphur sprays. No visual phytotoxicity symptoms were observed in this trial (variety Ben Lomond). Phytotoxicity of these treatments was investigated on Ben Hope and Ben Tirran in a separate trial in 2005.

In the 2006 trial, treatments which comprised two early season sprays of sulphur plus an additional spray of Masai pre-flower or post-flower, were the most effective reducing numbers of migrating mites by >95%. These treatments reduced end of season gall numbers by >92% on young lightly infested bushes but only by <50% on heavily infested bushes. The additional spray of Masai did not significantly improve control. Two sprays of a novel product from Bayer, UKA378b, applied just pre and post-flowering, showed acaricidal activity, significantly reducing end of season gall counts though the reduction in migrating mites was not statistically significant. Four sprays of Codacide oil, alone or in admixture with Vertalec, also showed limited activity. End of season gall counts were reduced by < 45%. Vertalec alone was ineffective as were Oberon, Kanemite and Floramite. None of the treatments caused any symptoms of phytotoxicity.

In summary, the conclusions form work under this objective are: Sulphur is the most effective gall mite acaricide currently available of a wide range of materials

tested. A spray of sulphur 800 g/l SC should be applied at a dose of 10 l/ha bud burst to all

blackcurrant plantations. Application at this time is not phytotoxic and can give up to 80% control of gall mite.

A second spray of sulphur 800 g/l SC should be applied at a dose of 10 l product/ha just before the first grape visible growth stage to all plantations, except to Ben Gairn and Baldwin and other varieties that are highly susceptible to sulphur phytotoxicity where such treatment may cause limited yield reductions. Where Baldwin plantations are at high risk, the second sulphur spray may still be beneficial at this time but there may be a modest yield penalty due to phytotoxicity.

An additional programme of sulphur sprays at 3.3 l/ha should be applied at 10-14 day intervals throughout the gall mite migration period to Ben Tirran and varieties that are not susceptible to sulphur phytotoxicity. Applications in hot conditions (> 25 ˚C ) should be avoided.

On varieties that are moderately or highly susceptible to phytotoxicity due to sulphur (Ben Lomond, Ben Hope, Ben Gairn, Baldwin) one or more supplementary sprays of Masai should be applied either just before or just after flowering. Masai has a high risk to bees and cannot be used during flowering.

The novel Bayer product, UKA378b, shows some promise as a gall mite acaricide and requires further evaluation. It may prove superior to Masai.

The EMR gall mite migration forecasting model provides fairly accurate predictions of the start, 5% and 50% migration. This model may be useful in timing of sprays but early season sprays of sulphur have to be applied at fixed growth stages to avoid phytotoxicity. Masai cannot be used at the peak of mite migration, which normally occurs during flower, because of its high risk to bees. It may prove useful for application of other effective acaricide materials if they are identified in future.

Objective 1.4. Determine whether foliar sprays of the commercially available strain of the entomopathogenic fungus Lecanicillium longisporum (Vertalec) can be exploited as a biocontrol agent of gall mite and identify the optimum time of year when it should be appliedKanagaratnam et al. (1981) working at the Glasshouse Crops Research Institute, Littlehampton, conducted a survey in England in 1977-78 on fungal pathogens of the blackcurrant gall mite. Infectivity tests were carried out in the laboratory on 5 fungal pathogens of arthropods to evaluate their potential

SID 5 (2/05) Page 24 of 34

as biological control agents. The survey revealed that many dead mites were associated with Lecanicillium longisporum, the only fungal pathogen found. The laboratory tests confirmed that L. longisporum was an effective fungal pathogen of the mite. The pathogenic status of L. longisporum, however, remained obscure, as the fungus appeared unable to penetrate live mites but grew profusely and sporulated on dead mites. It was thought possible that the fungus produced a toxin following germination or growth of spores on the mite cuticle. The mode of pathogenicity appeared unusual and was considered to warrant further investigation. It was thought that the best results for L. longisporum as a biological control agent would be obtained when sprays containing spores were applied in moist weather in spring as the 'big buds' were opening to prevent the spread of mites to uninfested plants; the ability of the fungus to grow as a saprophyte might possibly prolong its active period during mite spread. In 1999, V. lecanii was found to be a common natural entomopathogen of blackcurrant gall mite in New Zealand (G. Langford, pers. comm.). However, work to further investigate the pathogen and explore its use as a biocontrol agent was not done because an application from the New Zealand government to fund the work was unsuccessful. The work reported here had two objectives. They were to evaluate the efficacy of programmes of foliar sprays of a formulated product of L. longisporum (Vertalec), applied during the gall mite migration period in spring or during gall formation in the summer and autumn or both, (1) in causing infection in blackcurrant gall mite galls (2) preventing formation of new blackcurrant gall mite galls. Two experiments were conducted in 2003, one to address each of the two objectives. The experiment to address the first objective (experiment 1) was done in a plantation heavily infested with gall mite with large numbers of galls. The experiment to address the second objective (experiment 2) was done in a plantation with an initially low density of gall mite galls. This work was reported in full in Cross and Harris (2004c).

Subsequent to the field trials a laboratory experiment in controlled environment conditions was conducted in spring 2005 to establish whether blackcurrant gall mite was susceptible to parasitisation by L. longisporum (formerly named Verticillium lecanii), in particular whether infection of mites inside galls or emerging from galls can be caused by dipping galls in a aqueous suspension of spores. This laboratory experiment was reported in full in Van Wezel et al. (2006).

Materials and methodsThe two replicated field experiments were done in 2003 in two adjacent blackcurrant plantations at EMR that had been heavily infested with gall mite for several years and were badly infected with reversion virus disease. One plantation had not been cut down to the ground in spring 2002 and was very heavily infested with gall mite galls on the old and young wood. The other had been cut down to the ground in spring 2002 and had made vigorous re-growth in 2002 that was only lightly infested with gall mite galls. Each planting consisted of alternate rows of Ben Tirran, Ben Lomond. Treatments (Table 17) were programmes of foliar sprays of Vertalec, a commercially available formulation of the entomopathogenic fungus Lecanicillium longisporum, normally used for controlling aphids on glasshouse crops. A programme of five sprays of Vertalec during the gall mite migration period in April and May was compared with a programme of five sprays applied during the gall formation period from June to August and a programme of 10 sprays combining both these two treatment. Water only and untreated controls were also included.

Table 17. Treatments applied in the field experiments in 2003 to evaluate programmes of sprays of Vertalec against blackcurrant gall mite

Treatment Product Timing of sprays No. sprays Dates of application (2003)A Vertalec Migration period 5 2, 16, 30 Apr, 15, 28 May B Vertalec Gall formation period 5 11, 27 Jun, 9, 24 Jul, 7 AugC Vertalec All 10 All dates for A and BD Water All 10 All dates for A and BE Untreated None 0 -

The Vertalec sprays were all applied at a concentration of 6 g/l at a spray volume of 1000 l/ha (dose rate = 6 kg product applied/ha). Vertalec was pre-soaked in water (1g in 5 ml of water) for approximately 24 hours before the spray solutions were prepared. Sprays were applied with a Cooper Pegler CP 2000 knapsack sprayer at 1000 l/ha (6 l of spray to be made up per treatment per experiment). Randomised complete block experimental designs with six replicate two bush plots were used. The same experimental design and layout were used for the two experiments but with different randomisations of the treatments within the blocks. Each plot consisted of two bushes, one of Ben Lomond and one of Ben Tirran.

SID 5 (2/05) Page 25 of 34

In experiment 1 (causing infection in galls) the total number of galls on each bush in each plot were counted in the dormant period on before the first sprays were applied and again in the dormant period after all the sprays had been applied. In experiment 2 (reducing numbers of galls formed) in the December 2003, a sample of 10 galls was collected from each untreated plot of the variety Ben Lomond and from each plot of the variety Ben Lomond that had received the full programme of Vertalec. Thus 120 Ben Lomond galls were samples, 60 from the untreated control and 60 from the 10 Vertalec sprayed treatment. Each individual gall was then examined and tested for internal infections of L. longisporum using the following procedure: In the laboratory at East Malling, each gall was surface sterilised with 20% bleach for 10 minutes then washed in distilled water and left to dry. Each gall was then cut into quarters ensuring no cross infection between galls could occur. One quarter of each gall was carefully examined under a stereo microscope to see whether any obvious signs of internal infection were visible. The four quarters of each gall were then placed on the surface of an agar (20 g technical agar no. 3 + 10g mycological peptone + 40 g glucose/litre) plate. The quarters were equi-spaced on the surface of the agar with one of the cut surfaces in contact with the agar medium. For a positive control, 10 plates were streaked with a 1g/litre solution of Vertalec. The plates were incubated at 20 C for 5 days. Each plate was then examined for growth of fungal colonies on the surface of the agar arising from each gall quarter. A second assessment was made 7 days later on 15 December. On 19 December 2003, colonies that were suspected to be L. longisporum were sub cultured then sent to a specialist for identification. Statistical analysis was not appropriate for experiment 1. For experiment 2, after appropriate square root transformation, analysis of variance of the end of season gall counts was done with covariate adjustment for the pre-experiment counts.

The laboratory experiment (Phytotron unit, 20 oC, 95%RH) was conducted in spring 2005 to establish whether blackcurrant gall mite was susceptible to parasitisation by L. longisporum, in particular whether infection of mites inside galls or emerging from galls can be caused by dipping galls in a aqueous suspension of spores. Treatments tested were Vertalec (0.2%) (A), Vertalec plus the adjuvant Codacide oil (0.3%) (B), adjuvant only (C) and untreated (D). Mites emerging from galls were captured in miniature sticky traps and examined for mycosis. At the end of the experiment, after 2 weeks, the galls were cut open and the mites inside examined and tested for infection.

ResultsIn experiment 1 (causing internal infections in galls), no L. longisporum infections were found in any of the galls from the Ben Lomond plots that had received the 10 sprays of Vertalec (treatment C) or from the untreated plots. A fungus which was isolated from the galls which superficially appeared to be similar to L. longisporum was found to be a non-entomopathogenic species (possibly a Fusarium sp.) on close examination by D. Chandler. In experiment 2 (effects on numbers of galls formed), none of the treatments significantly reduced the number of galls formed per bush compared with the untreated or water controls which themselves did not differ significantly (Table 18).

Table 18. Mean (n) and mean square root (√n) end of season numbers of gall per bush adjusted for per-season in the field experiments in 2003 to evaluate programmes of sprays of Vertalec against blackcurrant gall mite

Treatment Timing of sprays n √n

Lomond Tirran Mean Lomond Tirran MeanA. Vertalec Migration 135 339 237 11.4 17.8 14.6B. Vertalec Gall formation 166 332 249 12.0 17.2 14.6C. Vertalec All 94 295 195 9.3 16.2 12.8D. Water All 91 410 251 8.8 19.7 14.3E. Untreated None 158 207 183 12.0 14.0 13.0

Fprob SEDVariety < 0.001 0.83

Treatment 0.444 1.33Variety.Treatment 0.036 1.87

In the laboratory experiment in 2005, the mite trapping data showed that 9 days after the start of the experiment, the treatments with the adjuvant Codacide oil (B and C) had reduced numbers of mites emerging compared to the Vertalec alone (A) or the untreated control (D). No fungus resembling L. longisporum was found on the emerged mites at the outset, but putative positive samples were found 3 days after the start of the experiments in treatments A, B and C and in the untreated material after 9 days. Inside the galls, the Vertalec (A) and untreated control (D) were mostly fungus free, but the

SID 5 (2/05) Page 26 of 34

treatments containing Codacide oil (B and C) showed considerable amounts of fungus parasitizing the mites. Examination of these mites showed mycelia nearly always formed around mites, even when the body of the mite had been degraded. Fungal penetration via the mouthparts occurred, but in very low numbers, and was not found to be L. longisporum. Putative L. longisporum mycelium was discovered within the buds in treatments B and C only, but only in low quantities.

ConclusionsNo natural or artificially induced infection by L. longisporum nor reduction in gall formation was detected in the field experiments. Several possible explanations for this are evident. 1) The methods used to detect infections were not effective, 2) the L. longisporum was unable to enter the galls to cause infection, 3) the fungus, or at least the strain of it tested, is not pathogenic. The results of the laboratory experiment suggested that the Vertalec strain of L. longisporum was not virulent enough to be used as a control agent against blackcurrant gall mite.

Objective 1.5. Validate the IPM methods for gall mite in commercial practice

Objective 2. Synthesize and validate IPM methods for blackcurrant leaf midge

Objective 2.1. To test the hypothesis that the parasitoid Platygaster demades, anthocorids and other natural enemies can establish and naturally regulate leaf midge populations to below damaging levels in commercial plantations over a three year period if broad-spectrum insecticides are not used and other practices to enhance the natural enemies are implemented.Work under objectives 1.5 and 2.1 are considered together because these objectives were addressed in the same large scale field experiment. The work was reported in full in Cross and Harris (2004d, 2005e, 2006b).

A new blackcurrant gall mite and leaf midge IPM experiment was established in an existing mature commercial blackcurrant plantation at Upper Horton Farm, Canterbury, Kent. The plantation cv Ben Lomond, approximately 2.3 ha in area, was divided into 12 plots of approximately equal size in a 4 x 3 array. The IPM experiment compared three treatments (Table 19), an IPM treatment using selective insecticides and acaricides only, a conventional treatment using broad spectrum products and an untreated control. The treatments were applied in the 2003 and 2004 growing seasons. Thereafter, the trial was abandoned (see below).

Table 19. Treatments in the gall mite and leaf midge IPM experiment at Upper Horton farm

Conventional treatment IPM treatment Untreated Sulphur at bud break Sulphur at first grape

emerged Dursban pre flower Meothrin just pre flower Meothrin end of flower

Sulphur at predicted first emergence of gall mite

Aphox pre-flower Sulphur or Masai at peak

emergence

No insecticides or acaricides

Sprays were applied at a volume of 500 l/ha with the growers Commandair orchard airblast sprayer. The whole experimental area also received a programme of sprays of fungicides for mildew and leaf spot control. In the dormant period before the experiment commenced and each year subsequently, a pre-experiment assessment of the density of blackcurrant gall mite galls was made in each plot. This was done each time on the same marked individual bushes. The numbers of bushes infected with reversion virus was assessed in each of 3 rows in the centre of each plot at early flower. Numbers of first generation leaf midge galls on 5 shoots at the base and 5 shoots in the terminals of each of 10 bushes per plot were counted. A sample of galls from each plot was collected. These were transferred the following day to plastic boxes containing paper towel, one box for each plot. Larvae completed their development and exited the galls to seek cocooning sites for pupal development. A sample of 100 larvae from each plot was collected and transferred individually to multi-well dishes containing moist filter paper. They were held in a incubator at 18˚C till they emerged. The number of adult blackcurrant leaf midge and parasitoids emerging were recorded. Attempts were made to repeat the leaf midge damage and parasitism assessments for the second leaf midge generation in both 2003 and

SID 5 (2/05) Page 27 of 34

2004, but this proved impossible as the bushes had stopped growing and no growing shoots were provided for the midge attack.

Because the IPM trial at Upper Horton Farm had to be abandoned, in 2006, further work was done to ascertain the incidence of parasitism of blackcurrant leaf midge larvae by P. demades in commercial blackcurrant plantations. Additionally, in late June, four blackcurrant plantations in western England subject to different pesticide management were beat sampled to determine the relative abundance on generalist predators and pests to gain insights into the effects of pesticide management on generalist predator communities. Samples of first generation blackcurrant leaf midge galls containing mature blackcurrant leaf midge larvae were collected from 6 commercial plantations in Norfolk, Kent and Herefordshire between 12 and 23 May 2006. Similar samples of second generation galls were collected from 11 plantations between 20 June and 3 July 2006. Up to 100 larvae from each sample were removed from the galls and transferred to a watch glass containing a small quantity of tap water. The anterior region of each larva was opened with dissecting needles and the contents examined for the presence of proto-larvae of Platygaster demades. The number of larvae examined and the number found to be parasitized was recorded.

Additionally, four blackcurrant plantations in western England that were subject to different pesticide management programmes were beat-sampled on 28 June 2008 (Table 2). Twenty five individual bushes widely spread throughout the area of each plantation were beat sampled using the standard beating method (2 beats per bush over 0.25 m2 beating tray). Additionally, 2 blackcurrant leaf midge galls from each of the 25 bushes were unfurled and the numbers of leaf midge larvae and anthocorid predators contained in the galls counted.

ResultsPre-trial mean gall mite gall numbers were similar for the different treatments in March 2003 but there were large significant differences between treatments in both March 2004 and March 2005, with much higher numbers of galls on the untreated controls than in the other two treatments, which were similar (Table 20). However, the treatments could only be considered to be partially effective as they did not prevent a ~3 fold increase in gall numbers in 2003 and a > 20 fold increase in 2004. However, increases were much greater on the untreated control in 2003, though the increase was limited by saturation in 2004. This shows that the conventional and IPM programmes performed similarly for gall mite control, but were inadequate in a mature heavily infested plantation where the gall mite infestation was running away. The very high populations meant that the experiment had to be abandoned.

Table 20. Pre-trial mean number of gall mite galls per bush in March 2003 and mean (n) and mean square root (√n) numbers of gall mite gall per bush in March 2004 and 2005, in the dormant season after the first and second years of treatment applications

Treatment March 2003 March 2004 March 2005n n √n n √n

IPM 4.8 14.2 3.41 73.1 7.29Conventional 4.9 15.4 3.46 84.0 8.02Untreated 5.4 187.9 12.57 278.4 16.05

Fprob 0.6 <0.001 0.006SED (5 df) 0.58 1.279 1.899

In May in both 2003 and 2004, after sprays of Meothrin had been applied to the Conventional plots, the numbers of leaf midge galls were significantly smaller on the conventionally treated plots than the IPM or untreated plots that had not received the Meothrin treatment (Table 21). Basal shoots were more heavily attacked than terminal shoots. The interaction between treatment and sampling positions was significant for both the untransformed and square root transformed data. This is because a slightly greater proportional reduction in numbers occurred in the shoots in the base of the bush than in the terminals comparing the IPM and conventional treatments. This result clearly shows Meothrin to be an effective control treatment for leaf midge but the IPM treatment did not reduce leaf midge gall numbers compared to the untreated control, indicating that natural enemies were not having a regulatory effect on populations so reducing leaf midge damage.

Rearing a sample of 100 mature larvae per plot to adult showed that a very small percentage (< 1%) of larvae were parasitised by P. demades in 2003 (Table 22). This number rose slightly in 2004 but remained low (< 3%). This work showed that the parasite did occur in the experimental site, though parasitism levels were small and unlikely to regulate populations sufficiently to reduce leaf midge damage.

SID 5 (2/05) Page 28 of 34

Table 21. Mean (n) and mean square root (√n) number of leaf midge galls per five shoots in May 2003 and 2004, after sprays of Meothrin had been applied to the conventional plots

Treatment n √nBase Terminal Mean Base Terminal Mean

2003IPM 12.6 6.6 9.6 3.32 2.35 2.83Conventional 2.9 2.1 2.5 1.40 1.17 1.29Untreated 12.7 8.8 10.7 3.43 2.87 3.15Mean 9.4 5.8 7.6 2.72 2.13 2.42

Treat Position Treat.PosFprob Fprob 0.002 <0.001 0.028SED SED 0.278 0.140 0.327

df df 5 224 >72004IPM 7.70 7.10 7.40 2.751 2.617 2.684Conventional 2.65 3.35 3.00 1.432 1.704 1.568Untreated 7.90 8.50 8.20 2.780 2.898 2.839Mean 6.08 6.32 6.20 2.321 2.406 2.364

Treat Position Treat.PosFprob 0.002 0.248 0.080SED 0.2113 0.0737 0.2298

df 6 105 8

Table 22. Parasitism of first generation leaf midge larvae (n=400) by the parasitoid Platygaster demades

TreatmentTotal number of leaf

midge adults emerged

Total number of Platygaster demades

adults emerged% parasitism

2003 2004 2003 2004 2003 2004IPM 227 292 2 5 0.88 1.71Conventional 250 307 1 9 0.40 2.93Untreated 205 286 1 6 0.49 2.10

In the grower plantations in 2006, a total of 1345 larvae were examined from 17 samples but none were found to be parasitised by P. demades.

Leaf midge galls were abundant in all the four plantations beat sampled in June 2006, practically every growing shoot being attacked by blackcurrant leaf midge. However, on the date of sampling, live leaf midge larvae were only found in galls in the conventionally sprayed Ben Lomond plantation. No larvae were found in the galls at the other sites. Generalist arthropod predators were abundant in beat samples from the IPM, unsprayed and organic plantations (totals of 115, 158 and 94 individuals/25 bushes respectively) but much less numerous in the conventionally treated plantation (7 indiviuals/25 bushes) (Table 23). In the IPM and unsprayed plantations, anthocorid predators and the predatory mite Anystis sp. were particularly abundant but no nabid bugs were found. In contrast in the organic plantation, there were comparatively few anthocorids or Anystis sp., but nabid bugs were abundant. Few anthocorids and no nabids were found in the conventional plantation. The predatory mirid, Heterotoma planicornis, was abundant in the IPM and organic plantations but infrequent in the unsprayed and conventional plantations. Capsid bugs were particularly abundant in the IPM plantation and were causing extensive damage to shoots. Small numbers of adult vine weevils were also found in this plantation. Anthocorids were found inside leaf midge galls, and nymphs were most numerous in galls containing leaf midge larvae (Table 23). Very high populations of leaf hoppers occurred in the unsprayed plantation and these were causing severe leaf discolouration. Very high populations of capsid bugs (mainly Lygocoris pabulinus) occurred in the IPM plantation causing damage to shoots. Selective insecticides for controlling these pests need to be identified for future development of IPM in blackcurrants.

SID 5 (2/05) Page 29 of 34

Table 23. Numbers of predatory arthropods found in four blackcurrant plantations in western England subject to different pesticide management, by beat sampling 25 bushes on each plot. Numbers of anthocorid predators and leaf midge larvae found in 50 leaf midge gall on 28 June 2006 were also counted

Pesticide manage-

mentVariety

No. in beat samples from 25 bushes No. in 50 midge galls

Anthocorid Other predators Pest Anthocorid

adul

ts

nym

phs

Het

erot

oma

plan

icor

nis

Nab

ids

Any

stis

sp.

Earw

igs

Cap

sid

Vine

wee

vil

adul

ts

nym

phs

IPM B. Gairn 28 28 32 0 25 2 36 2 2 2 0Unsprayed B. Gairn 27 43 7 0 69 12 6 0 5 5 0Organic B. Gairn 7 4 22 40 6 15 3 0 2 1 0Conventional B. Lomond 2 3 2 0 0 0 1 0 1 8 26

ConclusionsIn the IPM experiment at Upper Horton Farm, the conventional and IPM programmes performed similarly for gall mite control, but were inadequate because the plantation was too heavily infested. This, together with data from the long-term gall mite IPM experiment (objective 1.1) and the acaricides trials (objective 1.3), suggests that gall mite control is increasingly difficult as populations increase and priority should be given to keeping plantations as clean as possible for as long as possible. The increase in the gall mite population from year to year was exponential until a high proportion of buds was galled.

No evidence to suggest that the egg-larval parasitoid Platygaster demades is an effective parasitoid that can naturally regulate leaf midge populations was found either in the IPM experiment at Upper Horton Farm in 2003-2004 or in the commercial plantations in 2006 when both first and second generation leaf midge larvae were sampled. This was surprising because the parasite has been found at significant levels in samples from various farms in previous years. A likely reason for the failure of the parasitoid to establish in 2006 is poor synchronisation with its host. Monitoring of the flight of apple leaf midge using sex pheromone traps and adult P. demades using yellow sticky traps at East Malling indicated that the first flight of P. demades occurs in June - July with a second generation flight in August-September. Thus, the first generation of the midge occurred well before the midge flight. The first flight of the parasitoid seems to coincide better with the second generation of the midge. Further work is needed to investigate the incidence of the parasitoid.

The beat sampling for generalist arthropod predators in the four plantations subject to different management suggests that large populations of insect predators occur in IPM, organic and untreated plantations with few predators in conventionally sprayed plantations. However, despite the occurrence of high populations of predators, leaf midge was still a significant pest in all plantations, attacking practically every available shoot. This suggests that such predators are of only secondary importance in regulating leaf midge numbers, populations only developing in response to high pest populations and that they are not able to reduce leaf midge numbers to the low levels needed to prevent considerable attack. Observations suggest that anthocorid bugs are the dominant leaf midge predator. The results also suggests that nabid bugs establish at high levels in long established organic plantations and these might out-compete anthocorids and possibly prey on the predatory mite Anystis sp.. The results suggest that conventional pesticide management appears to more or less eliminate the bulk of predatory arthropods including the predatory mirid bug Heterotoma planicornis, nabid bugs, earwigs and Anystis sp.

In summary, the key findings are: The sulphur based gall mite control programme is as effective as the 3 x Meothrin programme but

cannot prevent a breakdown in control when populations are high. Gall mite control is increasingly difficult as populations increase and priority should be given to

keeping plantations as clean as possible for as long as possible. The increase in gall mite population from year to year is exponential until saturation is reached.

No evidence was gained to demonstrate that the parasitoid Platygaster demades is a key natural enemy of blackcurrant leaf midge able to regulate attacks to low, non-damaging levels.

SID 5 (2/05) Page 30 of 34

Generalist predators appear to be of only secondary importance in regulating leaf midge numbers, populations only developing in response to high pest populations and that they are not able to reduce leaf midge numbers to the low levels needed to prevent considerable attack. Observations suggest that anthocorid bugs are the dominant leaf midge predator.

Objective 2.2. Determine whether or not sprays of sulphur or of the selective acaricide identified in 1.3 above have an adverse affect on the key natural enemies of leaf midge.Regrettably, populations of Platygaster demades that occurred in the IPM experiment at Upper Horton Farm were too low to determine the effects of sulphur or other acaricides on the parasitoid.

Objective 2.3. To identify a selective insecticide treatment for control of blackcurrant leaf midgeIn a previous field trial in 2002, foliar sprays deltamethrin, fenpropathrin (Meothrin) and bifenthrin (Talstar) were the most effective products reducing numbers by 98, 99 and 97%, respectively, compared with the untreated control. Chlorpyrifos, nicotine, triazamate, thiacloprid, rotenone and spinosad were partially effective reducing numbers of live larvae per bush by 86, 57, 56, 84, 65 and 72%, respectively, compared with the untreated control. Diflubenzuron had little effect. Although deltamethrin and bifenthrin are potential replacements for Meothrin, they are equally harmful to natural enemies and disruptive to IPM. The aim was to identify alternative, more selective treatments for leaf midge control.

In this project, a field trial at Upper Horton Farm, Bridge, Kent, in 2005, evaluated single foliar sprays (500 l ha-1) of chlorpyrifos (Lorsban WG), fenpropathrin (Meothrin), thiacloprid (Calypso), spinosad (Tracer), acetamiprid (Gazelle), acrinathrin (Acrinathrin), pymetrozine (Plenum WG) and formetanate (Dicarzol) for control of existing larval infestations of the blackcurrant leaf midge on blackcurrant variety Ben Lomond. Sprays were applied on 10 May 2005 in the middle stages of the first generation larval attack when semi-mature and some mature larvae were present in galls. The effects of the treatments were assessed by counting the number of galls and the number of live larvae contained in galls 7 and 14 days after treatment.

The results indicated that all the products tested have at best only very limited curative activity against existing infestations of semi-mature and mature larvae in galls. None of the treatments had a worthwhile degree of efficacy when applied at this late stage in the larval attack and none of the products tested were more effective than Meothrin. It was not possible to discern treatments which had limited efficacy from those that were completely ineffective. The standard product Meothrin had the lowest mean numbers of larvae at the second assessment but this did not differ significantly from the untreated control. The results highlight the importance of treatment application at the early stages of oviposition and larval attack. None of the products produced any visual symptoms of phytotoxicity. This experiment was reported in full in Cross & Harris (2006).

An attempt to repeat the experiment at an earlier stage of attack when eggs are first being laid and larvae just starting to hatch was made on the 31 July 2006 in a heavily infested plantation at Burr Farm, Brenchley. Heavy rain immediately after treatment application washed off the spray treatments and no treatment differences developed. The site was monitored weekly throughout the rest of the season but levels of leaf midge did not increase sufficiently for the trial to be repeated.

Objective 3. Deliver and promote the IPM methods to UK blackcurrant growers

Full oral reports of this work were given by the project leader to the GlaxoSmithKline blackcurrant conferences in 2003, 2004, 2005 and 2006. All the work has been comprehensively reported in 15 reports (see references) plus in two papers in international conference proceedings.

The results have been fully assimilated into best practice which has been communicated to growers through periodic agronomic bulletins and visits by GlaxoSmithKline fieldsmen. The key findings have been imported into normal grower practice and are universally implemented in UK blackcurrant production.

POSSIBLE FUTURE WORK

SID 5 (2/05) Page 31 of 34

Whilst the integrated use of resistant varieties and early season sprays of sulphur supplemented with additional sprays of Masai or sulphur probably provides the highest standard of control of blackcurrant gall mite with the materials available, currently the treatment is not fully effective. It cannot prevent the exponential increase in gall mite when high populations occur. A programme of screening novel acaricides for gall mite activity needs to be continued. Further investigation of the novel Bayer product UKA378b is needed. Priority should also be given to ensuring that new blackcurrant varieties have gall mite and reversion virus resistance.

Further investigation is also needed by SCRI to determine whether or not the Ben Gairn bushes with symptoms of reversion virus disease in the long-term gall mite IPM experiment at EMR are positively infected. Samples will be sent for testing in spring 2007. If the disease is confirmed, it will be the first record in the variety.

The recent partial identification of the blackcurrant leaf midge sex pheromone is an important breakthrough. The identification needs to be completed and work done to exploit the pheromone for pest monitoring and control. It may then be possible to get better results from more accurate timing of selective insecticides to control the leaf midge.

ACTION RESULTING FROM THE RESEARCH (IP, KNOWLEDGE TRANSFER)

No IP has arisen from this work. Knowledge transfer is covered in the report on objective 3 above.

Appendix

Appendix Table 1. Products evaluated, their active ingredients and formulations and dose rates of application except in 2006

Product a.i. Dose product(l or kg /ha)

Acrinathrin acrinathrin 75 g/l EW 0.6 lAgral non-ionic wetter 0.01%Bond latex adjuvant (0.14% conc) 700 mlElvaron Multi tolylfluanid 50% w/w WG 1.6 kgHeadland Sulphur sulphur 800 g/l SC 12.5 lMasai tebufenpyrad 20% w/w WB 0.5 kgMavrik taufluvalinate 240 g/l EW 0.2 lMeothrin fenpropathrin 100 g/l E 1.0 lSequel fenpyroximate 51.3 g/l SC 2 lSulphur DF sulphur 80% DF 12.5 kgSulphur SC sulphur 800 g/l SC 12.5 lVertalec Lecanicillium longisporum 20% w/w WP 6.0 kg

SID 5 (2/05) Page 32 of 34

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.

Cross, J. V. 2006. Further investigation of IPM methods for blackcurrant gall mite and leaf midge. IOBC Bulletin in press.

Cross, J. V. & Harris, A. L. 2004a. Tests of the phytotoxicity of sulphur to blackcurrants 2003. Confidential report to GlaxoSmithKline/HDC growers fund issued 26 January 2004, 11pp.

Cross, J. V. & Harris, A. L. 2004b. Evaluation of acaricides for control of blackcurrant gall mite 2003 Confidential report to GlaxoSmithKline/HDC growers fund issued 2 February 2004, 12pp.

Cross, J. V. & Harris, A. L. 2004c. Tests of Verticillium lecanii for control of blackcurrant gall mite, 2003. Confidential report to GlaxoSmithKline/HDC growers fund issued 16 March 2004, 7pp.

Cross, J. V. & Harris, A. L. 2004d. IPM methods for blackcurrant gall mite and leaf midge, 2003. Confidential report to GlaxoSmithKline/HDC growers fund issued 5 July 2004, 14pp

Cross, J. V. & Harris, A. L. 2004e. Exploiting natural enemies in Integrated Pest Management in blackcurrant crops. Proceedings of 4th IOBC workshop on Integrated Soft Fruit Production, Conthey, Switzerland, October 2003. IOBC/WPRS bulletin 27(4), 9-16.

Cross, J. V. & Harris, A. L. 2004f. Exploiting natural enemies in Integrated Pest Management in blackcurrant crops. Proceedings of 4th IOBC workshop on Integrated Soft Fruit Production, Conthey, Switzerland, October 2003. IOBC/WPRS bulletin 27(4), 9-16.

Cross J V. & Harris, A. L. 2005a. IPM in blackcurrants. Plant it. Issue 8 August 2005, 4-5. HH3115TSF

Cross, J. V.& Harris, A. L. 2005b. Evaluation of acaricides for control of blackcurrant gall mite 2005 Confidential report to GlaxoSmithKline/HDC growers fund issued 13 Nov 2005, 13pp.

Cross, J. V. & Harris, A. L. 2005c. Tests of the phytotoxicity of sulphur to blackcurrants 2004. Confidential report to GlaxoSmithKline/HDC growers fund issued 24 February 2005, 19 pp.

Cross, J. V. & Harris, A. L. 2005d. Evaluation of acaricides for control of blackcurrant gall mite 2004 Confidential report to GlaxoSmithKline/HDC growers fund issued 21 February 2005, 13pp.

Cross, J. V. & Harris, A. L. 2005e. IPM methods for blackcurrant gall mite and leaf midge, 2004. Confidential report to GlaxoSmithKline/HDC growers fund issued 27 April 2005, 16pp.

Cross, J. V. & Harris, A. L. 2005f. Tests of the phytotoxicity of sulphur to blackcurrants 2005. Confidential report to GlaxoSmithKline/HDC growers fund issued 28 Nov 2005, 13 pp.

Cross, J. V. & Harris, A. L. 2006a. Evaluation of acaricides for control of blackcurrant gall mite 2006. Report to GlaxoSmithKline Blackcurrant Growers research Committee issued 1 December 2006, 15 pp.

Cross, J. V. & Harris, A. L. 2006b. Monitoring commercial blackcurrant crops for natural enemies of blackcurrant leaf midge 2006. Report to GlaxoSmithKline Blackcurrant Growers research Committee issued 1 December 2006, 15 pp.

Cross, J. V. & Harris, A. L. 2006c. IPM methods for blackcurrant gall mite: Final report of long-term experiment at EMR. Report to GlaxoSmithKline Blackcurrant Growers research Committee issued 6 December 2006, 20 pp.

Cross, J. V. & Harris A. L. 2006d. Evaluation of insecticides for control of blackcurrant leaf midge 2005. Report to GlaxoSmithKline Blackcurrant Growers research Committee issued 1 December 2006, 15 pp.

Cross, J. V. & Harris A. L. 2006. Evaluation of insecticides for control of blackcurrant leaf midge 2005. Report to GlaxoSmithKline Blackcurrant Growers research Committee issued 1 December 2006, 15 pp.

Cross, J. V. & Ridout, M. S. (2001). Emergence of the blackcurrant gall mite (Cecidophyopsis ribis) from galls in spring. Journal of Horticultural Science and Biotechnology 76, 311-319.

Kanagaratnam, P. Hall, R. A. & Burges, H. D. 1981. Effect of fungi on the black currant gall mite, Cecidophyopsis ribis. Plant Pathology. 1981. 30: 2, 117-118.

Van Wezel, R., Fountain, M. T. & Cross, J. V. 2006. Laboratory investigation of biocontrol of blackcurrant gall mite with Lecanicillium longisporum. Report to GlaxoSmithKline Blackcurrant Growers research Committee issued 3 December 2006, 16 pp.

SID 5 (2/05) Page 33 of 34

SID 5 (2/05) Page 34 of 34