IOBC/WPRS Working Group “Integrated Plant Protection in Fruit

Home , Acleris comariana, Anthonomus rubi, Armine (chemical), Byturidae, Campylomma, Campylomma verbasci

IOBC/WPRS

Working Group “Integrated Plant Protection in Fruit Crops”

Subgroup “Soft Fruits”

Proceedings of

Workshop on Integrated Soft Fruit Production

East Malling (United Kingdom) 24-27 September 2007

Editors Ch. Linder & J.V. Cross

IOBC/WPRS Bulletin Bulletin OILB/SROP Vol. 39, 2008

The content of the contributions is in the responsibility of the authors

The IOBC/WPRS Bulletin is published by the International Organization for Biological and Integrated Control of Noxious Animals and Plants, West Palearctic Regional Section (IOBC/WPRS)

Le Bulletin OILB/SROP est publié par l‘Organisation Internationale de Lutte Biologique et Intégrée contre les Animaux et les Plantes Nuisibles, section Regionale Ouest Paléarctique (OILB/SROP)

The Publication Commission of the IOBC/WPRS:

Horst Bathon Luc Tirry Julius Kuehn Institute (JKI), Federal University of Gent Research Centre for Cultivated Plants Laboratory of Agrozoology Institute for Biological Control Department of Crop Protection Heinrichstr. 243 Coupure Links 653 D-64287 Darmstadt (Germany) B-9000 Gent (Belgium) Tel +49 6151 407-225, Fax +49 6151 407-290 Tel +32-9-2646152, Fax +32-9-2646239 e-mail: [email protected] e-mail: [email protected]

Address General Secretariat:

Dr. Philippe C. Nicot INRA – Unité de Pathologie Végétale Domaine St Maurice - B.P. 94 F-84143 Montfavet Cedex (France)

ISBN 978-92-9067-213-5 http://www.iobc-wprs.org

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008

Contents

Development of semiochemical attractants, lures and traps for raspberry beetle, Byturus tomentosus at SCRI; from fundamental chemical ecology to testing IPM tools with growers. N. Birch, S. Gordon, T. Shepherd, W. Griffiths, G. Robertson, T. Woodford, R. Brennan...... 1-3

Mass trapping of raspberry beetle as a possible control method - pilot trials in Norway. N. Trandem, S. Gordon, N. Birch, M. Ekeland, N. Heiberg...... 5-10

Monitoring raspberry cane midge, Resseliella theobaldi, with sex pheromone traps: results from 2006. J. Cross, C. Baroffio, A. Grassi, D. Hall, B. Łabanowska, S. Milenković, T. Nilsson, M. Shternshis, C. Tornéus, N. Trandem, G. Vétek...... 11-17

Raspberry cane midge – flight dynamics, egg laying and the efficacy of the neonicotinoid insecticide acetamiprid on primocane fruiting raspberry. B. Łabanowska, J. Cross ...... 19-25

Interference between raspberry cane midge (Resseliella theobaldi) sex pheromone traps – A one season trial in a Swedish raspberry plantation. T. Nilsson, C. Tornéus...... 27-31

Some preliminary investigations into the sex pheromones of mirid soft fruit pests. M. Fountain, J. Cross, G. Jaastad, D. Hall ...... 33-40

Identification of black currant leaf midge Dasineura tetensi (Rübsaamen) female sex pheromone. L. Amarawardana, D. Hall, J. Cross, C. Nagy ...... 41-46

Biological control of the currant clearwing moth Synanthedon tipuliformis by mating disruption. C. A. Baroffio, Ch. Carlen...... 47-49

Notes on the parasitoids of the raspberry cane midge, Resseliella theobaldi (Barnes, 1927) (Diptera: Cecidomyiidae) and the rose stem girdler, Agrilus cuprescens (Ménétriés, 1832) (Coleoptera: Buprestidae). G. Vétek, C. Thuróczy, B. Pénzes ...... 51-64

Open field and laboratory surveys to evaluate the susceptibility of red raspberry genotypes to Tetranychus urticae Koch and Resseliella theobaldi (Barnes). A. Grassi, R. Maines, M. Grisenti, M. Eccher, A. Saviane, L. Giongo ...... 65-70

Harmfulness of raspberry gall midge, Lasioptera rubi Schrank (Diptera, Cecidomyi- idae), to some raspberry cultivars. S. Milenković, S. Tanasković...... 71-75

Patterns in the within-cane distribution of the gall-like swellings caused by Agrilus cuprescens (Coleoptera: Buprestidae) and the rate of raspberry infestation. M. Váňová, P. Tóth, J. Lukáš ...... 77-84

i ii

Post harvest control of the eriophyoid mite Phyllocoptes gracilis on raspberries. Ch. Linder, C. Baroffio, Ch. Mittaz...... 85-87

Biological control of two-spotted spider mite with Phytoseiulus persimilis in everbearer strawberry. C.A. Baroffio, Ch. Linder ...... 89-92

Does the presence of multiple phytoseiid species affect biocontrol of Tetranychus urticae on strawberry? J. Fitzgerald ...... 93-96

Field control of strawberry mite Phytonemus pallidus. B. Gobin, E. Bangels ...... 97-100

New generation acaricides for control of two important strawberry pests: The Two- spotted spider mite and the strawberry mite. B. Łabanowska ...... 101-106

Efficacy of 3 neonicotinoid insecticides for the control of the green leafhopper Asymmetrasca (Empoasca) decedens Paoli, a new pest on cultivated red raspberry in Trentino, Italy. A. Grassi, R. Maines, A. Saviane...... 107-113

Integrated Pest Management in protected strawberry crops: Increased returns, fewer pests and reduced pesiticide use. C. Sampson...... 115-120

Integrating biological control measures against strawberry pests – Preliminary results with strawberry tortrix in Denmark. L. Sigsgaard ...... 121-124

Biological control of aphids with Chrysoperla carnea on strawberry M. Turquet, J.-J. Pommier, M. Piron, E. Lascaux, G. Lorin ...... 125-129

Severing damage by Anthonomus rubi populations in the UK. C. Jay, J. Cross, C. Burgess...... 131-136

Studies on control of the vine weevil, Otiorhynchus sulcatus using entomopathogenic nematodes. S. Haukeland ...... 137-138

Use of entomopathogenic fungi for vine weevil and thrips control T.M. Butt, F.A. Shah, M.A. Ansari...... 139

Breeding for durable resistance to the large raspberry aphid, Amphorophora ideai, in field and protected raspberry plantations: Co-evolution and IPDM. N. Birch, S. Gordon, R. Brennan, N. Jennings, C. Mitchell...... 140

Aphid biology and the development of a programme to manage the spread of Blue- berry scorch virus in south western British Columbia, Canada. D.A. Raworth, S. Mathur, M. Sweeney, V. Brookes ...... 141-147

Advances in IPM for black currant A. Harris, J. Cross...... 149-154

iii

Sub-lethal exposure of honey bees to crop-protection – Feeding behaviour and flower visits. B. Gobin, K. Heylen, J. Billen, R. Huybrechts, L. Arckens ...... 155-159

Raspberry certification: How it benefits the raspberry sector? C. Eckert, C. Calvin ...... 161-163

Raspberry root rot control in the Scottish raspberry certification scheme. A. Schlenzig ...... 165-167

Biofumigation to control Verticillium wilt of strawberry: Potency and pitfalls. V.V. Michel, S. Dahal-Tscherrig, H. Ahmed, A. Dutheil ...... 169-176

The influence of weed covering on short-day strawberries in the autumn. R. Faby ...... 177-179

Evaluation of alternative chemicals for control of botrytis in raspberry. A. Berrie, T. O’Neill, E. Wedgwood, B. Ellerker ...... 181-187

Efficacy of Metschnikowia fructicola (Shemer®) against post-harvest soft fruit (berries) rots in northern Italy (Trentino). D. Prodorutti, A. Ferrari, A. Pellegrini, I. Pertot...... 189-192

Low doses of copper control leaf spot diseases caused by Mycosphaerella ribis and Drepanopeziza ribis in black currants. A. Stensvand, A. Dobson, S. Mogan ...... 193-196

Effectiveness of a tryfloxystrobin and tolyfluanid mixture for control of blackcurrant diseases. A. Broniarek-Niemiec, A. Bielenin...... 197-201

Interactions between isolates of powdery mildew (Podosphaera aphanis) and cultivars of strawberry, Fragaria x ananassa. X. Xu, J. Robinson, D. Simpson ...... 203-209

Potential role of cleistothecia in strawberry powdery mildew. X. Xu, J. Robinson, A. Berrie ...... 211-215

Ontogenic resistance against powdery mildew (Podosphaera macularis) in leaf tissue of strawberry. D.M. Gadoury, A. Stensvand, R.C. Seem, M.C. Heidenreich ...... 217

Colletotrichum acutatum: Survival in plant debris and infection of some weeds and cultivated plants. P. Parikka, A. Lemmetty ...... 219-222

Wild and cultivated Potentilla spp. may serve as alternate hosts and possible reservoirs of strawberry viruses. D. Yohalem, K. Lower ...... 223-224 iv v

List of Participants

AMARAWARDANA Lakmali BOULLENGER Amelie Natural Resources Institute BCP University of Greenwich at Medway Occupation Road Central Avenue Wye, Ashford, Kent Chatham Maritime TN25 5EN Kent, ME4 4TB UK UK [email protected] [email protected] BOUVEROUX Jurgen ALLEN Janet Biobest NV ADAS UK LTD Ilse Velden 18 Pibworth Cottage 2260 Westerlo Aldworth, Reading, Belgium Berks RG89RU [email protected] UK [email protected] BRONIAREK-NIEMIEC Agata Research Institute of Pomology and BAROFFIO Catherine Aryelle Floriculture Agroscope ACW 96-100 Skierniewice Centre des Fougères, Poland 1964 Conthey [email protected] Switzerland [email protected] CANO Javier Eurosemillas SA BELDA SUAREZ Jose Paseo de la Victoria 31 Koppert Espana 14004 Cordoba Parcela P.14 Nave 3, Pol. Ciudad del Spain Transporte del Poniente [email protected] 04745 La Mojonera Almeria CHRISTENSEN Dan Haunstrup Spain The Danish Fruit and Vegetable Advisory [email protected] Service Udkaersvej 15 BERRIE Angela 8200 Aarhus N East Malling Research Denmark New Road, East Malling, Kent [email protected] ME19 6BJ UK CIPRIANI Stéphanie [email protected] Syngenta Bioline 4 bis les Hillaires BIRCH Nick 33230 Abzac SCRI, Invergowrie, Dundee France DD2 5DA [email protected] UK [email protected] vi

COOK Rob FOUNTAIN Michelle FAST Ltd East Malling Research North Street New Road, East Malling, Kent Sheldwich, Faversham, Kent ME19 6BJ ME13 0LN UK UK [email protected] [email protected] GOBIN Bruno CROSS Jerry PC Fruit East Malling Research De Brede Akker 13 New Road, East Malling, Kent 3800 Sint Truiden ME19 6BJ Belgium UK [email protected] [email protected] GRIFFITHS Dave ECKERT Cathy Angus Soft Fruits CTIFL, Lanxade East Seaton, Arbroath 24130 Prigonrieux DD115SD France UK [email protected] [email protected]

ENKEGAARD Annie GULLIFORD Eleanor Faculty of Agriculture Sciences Angus Soft Fruits Aarhus University East Seaton, Arbroath Department of Integrated Pest DD115SD Management UK Research Centre Flakkebjerg [email protected] Forsøgsvej 1 4200 Slagelse HARRIS Adrian Denmark East Malling Research [email protected] New Road, East Malling, Kent ME19 6BJ FABY Rudolf UK VBOG Langförden, [email protected] Spredaer Straße 2 49377 Vechta-Langförden HAUKELAND Solveig Germany Bioforsk [email protected] Plant Health and Plant Protection Høgskolveien 7 FITZGERALD Jean 1432 Ås East Malling Research Norway New Road, East Malling, Kent [email protected] ME19 6BJ UK JAASTAD Gunnhild [email protected] East Malling Research New Road, East Malling, Kent ME19 6BJ UK [email protected] vii

JAY Chantelle LINDER Christian East Malling Research Agroscope Station de recherches New Road, East Malling, Kent Agroscope Changins-Wädenswil ACW, ME19 6BJ Route de Duillier UK 1260 Nyon [email protected] Switzerland [email protected] JOANG Malin LRF Konsult AB LOPEZ-MEDINA Jose Box 565 University de Huelva 201 25 Malmö Ctra. Huelva - Palos de la Frontera, S/N. Sweden 21819 La Rábida - Palos de la Frontera [email protected] Spain [email protected] JONES Simon BCP LORIN Guillaume Occupation Road Hortis Aquitaine Wye, Ashford, Kent Maison Jeannette TN25 5EN 24140 Douville UK France [email protected] [email protected]

JONGEDIJK Gerard MALAVOLTA Carlo Naktuinbouw Regione Emilia Romagna Caro van Eyckstraat 17 Viale Silvani 6 6708 NH Wageningen 40122 Bologna Netherland Italy [email protected] [email protected]

ŁABANOWSKA Barbara Helena McHUGH Rosalind Research Institute of Pomology and Scottish Agricultural Science Agency Floriculture Roddinglaw Road 96-100 Skierniewice Edinburgh Poland EH12 9FJ [email protected] UK [email protected] LASCAUX Emilie Koppert France MELIN Cédric 14 rue de la Communauté GFW-ASBL 44860 Pont St Martin 234 Chaussée de Charleroi France 5030 Gembloux [email protected] Belgium [email protected]

viii

MICHEL Vincent Valentin PARIKKA Paivi Agroscope ACW MTT Centre des Fougères Agrifood Research Finland 1964 Conthey 31600 Jokioinen Switzerland Finland [email protected] [email protected]

MILENKOVIĆ Slobodan PETERSEN Bodil Damgaard Fruit Research Institute Fruit and Vegetable Advisory Service Kralja Petre I 9, Industrivej 31 C 3200 Cacak 4230 Skaelskoer Serbia Denmark [email protected] [email protected]

MOORE Graham PIRON Mireille FAST Ltd Koppert France North Street 14 rue de la Communauté Sheldwich, Faversham, Kent 44860 Pont St Martin ME13 0LN France UK [email protected] [email protected] PRODORUTTI Daniele MORGAN Karen IASMA Research Center-Plant Protection FAST Ltd Departement North Street Via Edmondo Mach 1 Sheldwich, Faversham, Kent 38010 San Michele all’Adige (TN) ME13 0LN Italy UK [email protected] [email protected] RAWORTH David Arnold NILSSON Thilda Agriculture and Agri-Food Canada Swedish Board of Agriculture PO Box 1000 – 6947 #7 Highway Box 12 Agassiz 23053 Alnarp BC V0M 1A0 Sweden Canada [email protected] [email protected]

OLSZAK Remigiusz REID Caroline Research Institute of Pomology and Syngenta Bioline Floriculture 140 Tilehouse 96-100 Skierniewice Green Lane, Knowle, West Midlands Poland B93 9ER [email protected] UK [email protected]

ROBBE Alain STEFANANOVA Miryana GFW-ASBL Angus Soft Fruits, 234 Chaussée de Charleroi East Seaton, Arbroath 5030 Gembloux DD11 5SD Belgium UK [email protected] [email protected]

SAMPSON Clare STENSVAND Arne BCP Bioforsk, Norwegian Institute for Occupation Road Agricultural and Environmental Research Wye, Ashford, Kent Høgskoleveien 7 UK 1432 Ås [email protected] Norway [email protected] SCHLENZIG Alexandra Scottish Agricultural Science Agency STERK Guido Roddinglaw Road Biobest NV Edinburgh Ilse Velden 18 EH12 9FJ 2260 Westerlo UK Belgium [email protected] [email protected]

SHAH Farooq SURIS Santiago Department of Biological Sciences Eurosemillas SA University of Wales Paseo de la Victoria 31 Swansea 14004 Cordoba SA2 8PP Spain UK [email protected] [email protected] TORNÉUS Christer SIGSGAARD Lene Swedish Board of Agriculture Department of Ecology Box 12 University of Copenhagen 23053 Alnarp Thorvaldsenvej 40 Sweden 1872 Frederisksberg C [email protected] Denmark [email protected] TÓTH Peter Slovak Agricultural University SPAK Joseph Department of Plant Protection Biology Centre ASCR, v.v.i. Tr. A. Hlinku 2 Institute of Plant Molecular Biology 949 76 Nitra Branišovská 31/1160 Slovak Republic 370 05 České Budějovice [email protected] Czech Republic [email protected]

TRANDEM Nina WALTER Philip Bioforsk, Norwegian Institute for IPM Consultant Agricultural and Environmental Research Acorn Cottage Høgskoleveien 7 Chapel Lane, West Wittering, 1432 Ås Chichester, West Sussex Norway PO20 8QG [email protected] UK [email protected] TURQUET Marion Hortis Aquitaine XU Xiangming Maison Jeanette East Malling Research 24140 Douville New Road, East Malling, Kent France ME19 6BJ [email protected] UK [email protected] VÁŇOVÁ Martina Slovak Academy of Sciences YOHALEM David Simon Branch for Woody Plants Biology East Malling Research Department Protection and Creation of New Road, East Malling, Kent Greenness ME19 6BJ Academická 2 UK 949 01 Nitra [email protected] Slovak Republic [email protected]

VÉTEK Gábor Corvinus University of Budapest Faculty of Horticultural Science Department of Entomology Villanyi ut 29-43 1118 Budapest Hungary [email protected]

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 1-3

Development of semiochemical attractants, lures and traps for raspberry beetle, Byturus tomentosus at SCRI; from fundamental chemical ecology to testing IPM tools with growers

Nicholas Birch, Stuart Gordon, Tom Shepherd, Wynne Griffiths, Graham Robertson, Trefor Woodford, Rex Brennan SCRI, Invergowrie, Dundee, DD2 5DA

Abstract: Raspberry beetle adults are attracted to flowers of their hosts primarily by colour and odour (floral volatiles). SCRI scientists have investigated this chemical ecology interaction for several years, using a multi-disciplinary approach involving phytochemistry, insect behaviour, and GC-EAG electrophysiology. We will present a historical overview, explaining how these techniques have allowed us to identify the key flower attractants from a complex mixture of volatiles emitted by raspberry flowers. We will then go on to explain how recent (EU-CRAFT, Horticulture Development Council) and current (Defra HortLINK) work has progressed the optimization of raspberry beetle traps for U.K. growers needing IPM solutions due to demands for zero pesticide residue levels on fruit. We will explain how we are developing and testing slow release lures and different trap designs, together with collaborators at East Malling Research, Natural Resources Institute, AgriSense Ltd and also with Norwegian scientists, testing prototype traps on organic soft fruit farms.

Key words: raspberry beetle, trapping, host attractant, semiochemicals, IPM

Introduction

The raspberry beetle, Byturus tomentosus (Degeer) causes severe damage to raspberries by both the adults (which feed on buds and flowers) and by larvae (which feed inside the developing fruit). The threshold for damage is very low in fresh raspberries because the presence of only a few larvae can mean that the whole consignment is rejected by the fruit marketing agents (Gordon et al. 1997). Due to ongoing reductions in allowable pesticide residues on fresh raspberries to a point that is effectively zero in the U.K. and a large increase in protected raspberry production (under plastic tunnels) there is a strong demand for alternative means to control this pest using semiochemical-enhanced trapping methodologies. At SCRI, we have developed a novel trap, based on the key visual and olfactory characteristics of the raspberry flower. This trap is now being tested on commercial farms in Scotland and England (conventional/IPM).and on organic smallholdings in Norway (see Trandem et al. this volume). Previous research at SCRI successfully identified the most active floral attractant volatiles from raspberry flowers using a combination of GC-EAG (gas chromatogram linked to electro-antennogram) and behavioural bioassays using linear track olfactometers and a wind tunnel (Woodford et al. 2003; Birch et al. 2004; Mitchell et al. 2004).

1 2

Materials and methods

Sites and duration of on-farm experiments in Scotland The trials were conducted at two farms in Eastern Scotland from 1 May – 11 July 2007 (Table 1). Two commercial Scottish plantations were chosen. Each were growing cv. Glen Ample under Spanish type tunnels covered with polyethylene sheeting from flowering to after harvest and each plantation was approximately 1 hectare in size. These sites were chosen as they were thought to have low to moderate populations of raspberry beetle due to previous use of insecticides.

Table 1: Site details.

Site 1 details Wester Essendy, Blairgowrie, Perthshire, Scotland Field name NGR Cultivar Age (years) E1 NO 135 435 Glen Ample 2 E2 NO 135 435 Glen Ample 5 E3 NO 135 435 Glen Ample 3 Site 2 details Blairgowrie, Perthshire, Scotland Field name NGR Cultivar Age (years) J1 NO 216 462 Glen Ample 8 J2 NO 216 462 Glen Ample 8 J3 NO 216 462 Glen Ample 8

Treatments Devices were modified AgriSense bucket traps with white Correx cross-vanes and with a polythene vial dispenser containing initially 2.5 ml of attractant coded ‘B’. The funnel traps contained 3 cm of 25% antifreeze (ethylene glycol) in water with a drop of detergent (Teepol®) to reduce surface tension. The trap treatments were applied at least two weeks before flowering commenced in each plantation. Three trap deployment treatments were used: perimeter, lattice and control. Perimeter trapping consisted of the traps being suspended from the top wire at the outer most position in the plantation at 8 m spacing around the entire perimeter of the plantation (50 traps / hectare around the edge of each 1 ha plantation). The traps in the lattice trapping treatment were positioned regularly through out the plantation, suspended from the top wire, at a density of 50 traps per hectare. The control treatment contained no buckets traps but did contain a small number of sticky traps for monitoring local raspberry beetle numbers in untreated areas.

Assessments Eight bucket traps per treatment were checked weekly and the number of beetles recorded. Sampling of all the traps for adult raspberry beetles was undertaken at flowering and at the end of the trial. The number of non-target insects was also recorded. Standard white sticky traps were positioned in all three treatments to allow weekly monitoring of the raspberry beetle populations in control areas without traps.

Results and discussion

Up to the green fruit stage the lattice treatment (within crop) was x 6 more effective for trapping raspberry beetles than the perimeter layout at one site but not different at the other (due to a lower raspberry beetle population). After the green fruit stage both spatial layouts worked equally well in terms of catching adult raspberry beetles, but catch numbers were lower than in the pre- green fruit stage. In control plots (without enhanced bucket traps) numbers of raspberry beetles were variable between sites but more were caught around the edges than in the middle. The numbers of raspberry beetle eggs found on sampled flowers was very low in all trap treatments, which was also reflected in the very low incidence of larvae and damage to berries. Although some non-target insects were caught (mainly bees, wasps, flies and other beetle species), the numbers were low and had no obvious effect on pollination or biocontrol within the crop. These first year results and previous studies indicate that the enhanced bucket traps are effective in trapping and controlling damage caused by raspberry beetle in protected raspberry plantations with low-moderate populations of raspberry beetles (due to management history, climate and local habitat). The most effective time to use them is pre-flowering of the raspberry crop, when there is no competition from raspberry flowers. Further trials will be undertaken in Scotland and England in 2008 to assess variation in efficacy caused by site and season and to fine tune the spatial deployment under different environmental conditions. We are also setting up collaborative trials in France and Switzerland, to complement the ongoing trials in Norway, providing a wide range of pest pressures and growing conditions.

Acknowledgements

We thank Defra HortLINK and RERAD for providing funding and our project collaborators, particularly at East Malling Research, Agrisense Ltd, Natural Resources Institute and ADAS, for help with these experiments.

References

Gordon, S.C., Woodford, J.A.T. & Birch, A.N.E. 1997: Arthropod pests of Rubus in Europe: pest status, current and future control strategies. – Journal of Horticultural Science 72: 831-862. Woodford, J.A.T., Birch, A.N.E., Gordon, S.C., Griffiths, D.W., McNicol, J.W. & Robertson, G.W. 2003: Controlling raspberry beetle without insecticides. – IOBC/wprs Bulletin 26(2): 87. Birch, A.N.E., Gordon, S.C., Fenton, B., Malloch, G., Mitchell, C., Jones, A.T., Griffiths, D.W., Brennan, R.M., Graham, J. & Woodford, J.A.T. 2004: Developing a sustainable IPM system for high value Rubus crops (raspberry, blackberry) for Europe. – Acta Horticulturae 649: 289-292. Mitchell, C., Gordon, S.C., Birch, A.N.E. & Hubbard, S.F. 2004: Developing a 'lure and kill' system for raspberry beetle, Byturus tomentosus, in Rubus production. – IOBC/wprs Bulletin 27(4): 113-118.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 5-10

Mass trapping of raspberry beetle as a possible control method – pilot trials in Norway

Nina Trandem1, Stuart C. Gordon2, A. Nicholas E. Birch2, Mari Ekeland3, Nina Heiberg4 1 Bioforsk, Plant Health and Plant Protection Division, Høgskoleveien 7, 1432 Ås, Norway; 2 SCRI, Invergowrie, Dundee DD2 5DA, Scotland, U.K.; 3 Myrerskogveien 28, 0495 Oslo, Norway; 4 Njøs Frukt- og bærsenter, Njøsavegen 5, 6863 Leikanger, Norway

Abstract: Berry damage by the larva of raspberry beetle (Byturus tomentosus) is a major risk for growers of organic raspberry. The identification of two volatile compounds in raspberry flowers that are attractive to the beetle by scientists at SCRI has facilitated the idea of mass trapping as an alternative control method to using conventional insecticides. The challenges are to design efficient and user-friendly traps, and to document if and how such traps can reduce the number of ovipositing beetles sufficiently to get a low level of berry damage. These questions are being investigated as part of a large UK project (Defra HortLINK) which is developing IPDM (Integrated Pest and Disease Management) for protected raspberries. We here report the results from a cooperating Norwegian project in which the volatile ‘compound B’ was used in pilot trials, 2003-2006. A combined collision- funnel trap from SCRI and AgriSense with the compound in a slow release lure attracted and killed a high number of raspberry beetles in the weeks before flowering, but more studies are needed to find a trap strategy that consistently leads to less berry damage. Norwegian organic fields are small, with large populations of raspberry beetle, and usually with wild raspberry growing nearby. A successful mass trapping strategy must therefore pay equal attention to immigrating and resident beetles.

Key words: Byturus tomentosus, organic growing, pest, raspberry, semiochemicals

Introduction

The larva of raspberry beetle (Byturus tomentosus) is the well known “worm” found in raspberry fruit. The beetle can relatively easily be controlled by synthetic pesticides like pyrethroids in the spring, but for growers of organic raspberry the beetle is a severe problem as no documented alternative control methods exist. Raspberry beetle emerges from its over-wintering site in the ground some weeks before flowering, and can in this period be found on the raspberry plants and nearby pollen sources like bird cherry (Prunus padus) and other early-flowering rosaceous plants. The beetle requires a body temperature above 15ºC to be able to fly, and flight activity is highest in warm and dry weather (Willmer et al. 1996). The beetle also feeds on raspberry leaves and flower buds. When the raspberry starts flowering, the beetle mates and oviposits in the open flowers. Each female can lay up to 120 eggs, usually one per flower (Antonin 1984; Gordon et al. 1997). The larvae feed on the developing fruit (husks and druplets; Taylor & Gordon 1975). The fully fed larvae leave the berries and enter the ground beneath the plants to overwinter. In Norway a major part of the population requires two years to complete the lifecycle, spending almost one and half years in the ground (Stenseth 1974). Further south the normal lifecycle is univoltine, with only one winter spent in the ground. Raspberry fields in Norway, and organic ones in particular, are typically small in size (<<1ha), and wild raspberry, a source of raspberry beetle, grows nearby most fields.

5 6

Two volatile compounds occurring in raspberry flowers, denoted ‘compound A’ and ‘compound B’, have been identified as strong attractants for raspberry beetles of both sexes by scientists at the SCRI (Birch et al 1996; 2004; Woodford et al. 2003). The attractants can be used for enhancing the efficiency of white sticky traps used for monitoring, but have also made it possible to envision an attract-and-kill (mass trapping) control method for raspberry grown without pesticides (Gordon et al. 1997; 2006; Woodford et al. 2003). Since 2003, The Norwegian Agricultural and Environmental Research Institute (Bioforsk, formerly The Norwegian Crop Research Institute) has done trials with ‘compound B’ under a collaboration agreement with SCRI/ MRS Ltd. The Norwegian trials have been a part of a national pilot project on development of organic fruit growing and have been of a preliminary nature. We here report the main results and experiences 2003-2006.

Material and methods

Trap density gradient in one field, 2003 In a 16 x 36 m unsprayed experimental field with mixed cultivars at Njøs, W-Norway, 20 white sticky traps (single vanes of Rebell® bianco, Andermatt Biocontrol) were distributed to form a gradient of trap density, increasing from zero in one end to 1 trap per 3 m plant row in the other. The traps were suspended from the plant-supporting wire and enhanced with olfactory lures consisting of 2 ml chromacol vials containing ‘compund B’. A cotton wick dispensed the attractant through a hole in the polyethylene cap. The trap period was 20 May to 27 June, with flowering starting in mid-June. Berries were assessed for larval damage during harvest by dividing the field into 24 plots of 6 m row, sampling 100 berries weekly from each plot for 3 weeks.

Mass trapping in private gardens, 2004 White sticky traps with lures as in 2003 were deployed in a density of 1 every 3 row m (cv mainly Veten and Asker) in 10 private gardens at Ås, SE-Norway, about 3 weeks before start of flowering. Sticky traps were changed 1-4 times per month, before becoming saturated with target and non-target insects (Schmid et al. 2001), and the trapping continued about 4 weeks into the flowering. The number of traps in each garden varied from 3 to 22, due to varying raspberry planting size. Twenty other gardens served as controls: 10 of them had no traps at all, and 10 had white sticky traps without lures. During the picking season, 100 berries were randomly sampled in each garden once a week (3 weeks in total) and assessed for larval damage. More details are found in Ekeland (2005).

Test of new trap type, 2005 Twenty new combination traps were distributed to 6 organic fields across Norway. The combination traps consisted of a green funnel trap (AgriSense) modified by SCRI by inserting white non-sticky cross-vanes between the top and the funnel + bucket part. Lures (slow release plastic dispensers from AgriSense with ‘compound B’) were placed in the cross-vanes. Water with a drop of detergent trapped the beetles falling through the funnel into the bucket. Most localities received two combination traps, to be positioned 1.2 m above ground in the field interior 15-20 m from each other, the two traps having 1 and 2 lures, respectively. In addition, unmodified funnel traps from AgriSense and Pherobank, also enhanced with ‘compound B’, were tried. Raspberry beetle catch was recorded and soapy water renewed weekly. Berry damage was to be assessed 2-3 times during harvest at three sites in each field: within 1 m of each combination trap and midway between them.

Combination traps along field perimeter, 2006 Ten growers from several parts of Norway each received 30 combination traps with plastic lures. The traps were to be put every 6-8 m (i.e., 2x row distance) along the whole perimeter of a ~0.1 ha organic field or part of field. Growers received instructions to mount the traps ~1.5 m above ground on poles put up ~2 m away from the raspberry rows, to record the catch in 12 of the traps weekly (as well as renewing soapy water in all traps) during the pre-flowering period, and to register raspberry beetle damage in the berries at 7-9 spots at 2 times during harvest.

Results and discussion

Trap density gradient in one field, 2003 The field had a very small raspberry beetle population, probably because it had been sprayed all previous years: 37 beetles were caught in total (31 of them before flowering). As expected from this the larval damage was low, only 0.4% of the 2700 fruits sampled. Beetles were not removed when counted (2 times weekly), and it was observed that some of them disappeared. This may have been due to beetles freeing themselves from the glue, or predation by birds. There was a positive correlation between total beetle catch and trap density in the plots (r2=0.37; p=0.01), but no correlation between beetle catch and berry damage, nor any unequivocal gradient in berry damage from one end of the field to the other.

Mass trapping in private gardens, 2004 The beetle populations in the 30 gardens were many times larger than in the field used the previous year, but the pattern that traps with lures were most attractive before flowering was the same. Traps without lures had similar efficiency throughout the trapping period (Table 1).

Table 1. Median beetle catch per trap day in white sticky traps with and without an olfactory lure during 3 successive periods in 2004. Data from private gardens, 1 trap/ 3 row m.

Period (related to Sticky traps w/ lure Sticky trap w/o lure Enhancement flowering) (n=10 gardens) (n=10 gardens) factor with lure 3-7 weeks before 7.4 0.7 10.6 0-3 weeks before 3.4 0.9 3.8 4 weeks during 1.5 0.9 1.7

Average berry damage was similar in the three treatments (Figure 1), thus trapping did not influence larval population size. A probable explanation is that at such high beetle densities there is strong competition for oviposition sites (flowers), and the number of beetles trapped was not enough to counter the compensating effect of remaining females being able to lay more eggs.

Test of new trap type, 2005 The new combination traps worked very well. They did not become saturated or got entangled in foliage, and users were not polluted with glue. Although not tested directly, the combination traps also seemed more efficient; in the only field (NW-Norway) receiving more than two combination traps, an average of 300 beetles were caught per trap (n=6) during pre- blossom (~4 weeks), while the year before sticky traps caught 77 beetles per trap (n=8) in the same field and period (in a trial not further described here). Combination traps have also been more efficient than sticky traps in UK trials (N. Birch & S. Gordon, unpublished data).

% berries with larval damage 100 Sticky traps+lure 80 Sticky traps only 60 No traps

0 123 Week of harvest

Figure 1. Berry damage in the 2004 mass trapping trial (data from 3 x 10 private gardens).

The cross-vane part of the combination traps was very important; the catches in unmodified funnel traps were less than 5% of that in combination traps. The white cross- vanes acted both as visual lures (mimicking the colour of raspberry flowers) and collision devices. Providing combination traps with two lures in stead of one did not increase the catch (median pre-blossom catch 371 beetles per trap with one lure; 246 with two; n= 5 growers providing data). Berry damage varied greatly from field to field (13-83%), and there was no significant difference between the three sites investigated in each field (mean damage 42 % close to traps w/one lure, 43 % w/two lures, and 40 % midway between traps; ANOVA, p=0.98, n=5 fields). Thus, berries close to traps did not seem to have an increased risk of damage.

Combination traps along field perimeter, 2006 Eight growers returned data on weekly beetle catch for 9 fields in total, and on berry damage for 8 fields. Like in previous years, many raspberry beetles trapped usually meant many berries damaged and vice versa (Figure 2). If the perimeter trapping worked, one would have expected a low level of damage in all fields, also in the ones with many beetles. With one notable exception, this was not the case. Neither was there any consistent pattern that damage in 2006 was lower than in 2005, when trapping was less intensive (5 fields were also in the 2005 trap test, Table 2).

Table 2. Total number of beetles trapped during pre-blossom and berry damage level reported for five organic fields where data was collected in both 2005 and 2006. Number of combination traps present in the field in parentheses. Most fields had a mix of cultivars.

Year 2005 2006 Site (planting year) Beetle catch Damage Beetle catch* Damage Ryfylke (2003/04) 30 (2) 19 % 323 (30) reg. missing Årnes (1992-2005) 1086 (2) 13 % 2060 (30) 4 % Nordmøre (2001) 1526 (6) 51 % 1346 (27) 46 % Helgeland (1999) 1250 (2) 43 % 1055 (8**) >50 % Bodø (1999) 1930 (2) 83 % 9135 (30) ca 80 % * Extrapolated from the traps (12 in most cases) where the catch was counted. **Field was only 0.05 ha.

Trap rate vs berry damage, perimeter trial 2006

100

% berry damage berry % 20

0 0246810 Beetles caught per trap day before flowering

Figure 2. There was a positive correlation between trap rate and berry damage in the perimeter trial (n=8 organic fields). The outlier is Årnes (cf. Table 2).

Conclusions

The trapping intensity used in the trials was insufficient to reduce the larval population in Norwegian raspberry fields. Trials with higher trapping intensity, putting traps in the field perimeter and nearby wild raspberry as well as in the field interior, should be carried out in the Norwegian fields. In a Swiss trial the berry damage was 21 % in plots with sticky traps and 34 % in control plots without sticky traps (Schmid et al. 2006). In this trial the trap density was ~3 times the one in our 2004 trial, and the beetle population (judged from trap rates) about one tenth. Tests of mass trapping in the UK on experimental plots and commercial farms look promising (Birch et al. this volume). Further, observations of beetle behaviour around and on traps may give clues to improvement of trap design. A major constraint to mass trapping in Norway is weather conditions during pre-flowering period; during cold or windy weather the beetles do not fly and cannot be trapped. Catching resident beetles in tunnels is probably less weather dependent. These and other questions are being addressed in the Defra HortLINK project in UK and a new Norwegian project on tunnel growing of organic raspberry.

Acknowledgements

We thank the growers for counting trapped beetles and damaged berries in their fields, Roald Lunde and Marianne Bøthun for field work in W-Norway, Øystein Kjos and Karin Westrum for trap assemblage and distribution, Eline Hågvar for co-supervising the Master project of ME, and AgriSense for providing plastic dispensers and other materials. Bioforsk received financial support from the Norwegian Agricultural Authority (SLF) 2004-2006. SCRI receives funding from Defra HortLINK and from RERAD on chemical ecology and development of sustainable agriculture using IPM.

References

Antonin, Ph. 1984: Le ver des framboises, Byturus tomentosus F. – Revue Suisse Vitic. Arboric. Hortic. 16: 103-5.

Birch, A.N.E., Gordon, S.C., Griffiths, D.W., Harrison, R.E., McNicol, R.J., Robertson, G.W., Spencer, B., Wishart, J. & Woodford, J.A.T. 1996: The role of flower volatiles in host attraction and recognition by raspberry beetle, Byturus tomentosus. – IOBC/wprs Bulletin 19 (5): 117-122. Birch, A.N.E., Gordon, S.C., Fenton, B., Malloch, G., Mitchell, C., Jones, A.T., Griffiths, D.W., Brennan, R.M., Graham, J. & Woodford, J.A.T. 2004: Developing a sustainable IPM system for high value Rubus crops (raspberry, blackberry) for Europe. – Acta Horticulturae 649: 289-292. Ekeland, M. 2005: Limfeller forsterket med duftstoff fra bringebær (Rubus idaeus) som alternativ bekjempelse av bringebærbille (Byturus tomentosus). – MSc thesis, Norwegian University of Life Sciences, 116 pp. [In Norwegian with English summary]. Gordon, S.C., Birch, A.N.E. & Mitchell, C. 2006: Integrated pest management of pests of raspberry (Rubus idaeus) – possible developments in Europe by 2015. – IOBC/wprs Bulletin 29 (9): 123-130. Gordon, S.C., Woodford, J.A.T. & Birch A.N.E. 1997: Arthropod pests of Rubus in Europe: Pest status, current and future control strategies. – J. Hort. Sci 72: 831-862. Schmid, A., Höhn, H., Schmid, K., Weibel, F. & Daniel, C. 2006: Effectiveness and side effects of glue-traps to decrease damages caused by Byturus tomentosus in raspberry. – J. Pest Sci. 79: 137-142. Schmid, K., Graf, B., Höhn, H. & Höpli H.U. 2001: Der Himbeerkäfer geht auf den Leim – Befallsprognose mit Weissfallen, Schädlingsbekämpfung. – Schweiz. Z. Obst -Weinbau 8: 198-201. Stenseth, C. 1974. Livssyklus og fenologi hos bringebærbille, Byturus tomentosus (Col., Byturidae). – Forskning og Forsøk i Landbruket 25: 191-99. [In Norwegian, English summary and legends]. Taylor, C.E. & Gordon, S.C. 1975: Further observations on the biology and control of the raspberry beetle (Byturus tomentosus (Deg.)) in eastern Scotland. – J. Hort. Sci. 50: 105- 112. Willmer, P.G., Hughes, J. P., Woodford, J.A.T. & Gordon, S.C. 1996: The effects of crop microclimate and associated physiological constraints on the seasonal and diurnal distribution patterns of raspberry beetle (Byturus tomentosus) on the host plant Rubus idaeus. – Ecol. Entomol. 21: 87-97. Woodford, J.A.T., Birch, A.N.E., Gordon, S.C., Griffiths, D.W., McNicol, J.W. & Robertson, G.W. 2003: Controlling raspberry beetle without insecticides. – IOBC/wprs Bulletin 26 (2): 87-92.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 11-17

Monitoring raspberry cane midge, Resseliella theobaldi, with sex pheromone traps: results from 2006

Jerry Cross1, Catherine Baroffio2, Alberto Grassi3, David Hall4, Barbara Łabanowska5, Slobodan Milenković6, Thilda Nilsson7, Margarita Shternshis8, Christer Tornéus7, Nina Trandem9, Gábor Vétek10 1 East Malling Research, New Road, East Malling Kent ME19 6BJ UK; 2 Agroscope Changins-Wädenswil ACW, Centre des Fougères, 1964 Conthey, Switzerland; 3 IASMA Research Center – Plant Protection Department, Via E. Mach, 1 – 38010 San Michele all’Adige (TN), Italy; 4 Natural Resources Institute, Central Avenue, Chatham Maritime, Kent ME4 4TB UK; 5 Research Institute of Pomology, Pomologiczna 18, 96-100 Skierniewice, Poland; 6 Fruit Research Institute Čačak, Kralja Petra I/9, Serbia; 7 Swedish Board of Agriculture, Plant Protection Centre, P.O. Box 12, S-230 53 Alnarp, Sweden; 8 Novosibirsk State Agrarian University, Novosibirsk-39, 630039, Russia; 9 Bioforsk, Plant Health and Plant Protection Division, Høgskoleveien 7, N-1432 Ås, Norway; 10 Corvinus University of Budapest, Faculty of Horticultural Science, Department of Entomology, Villányi út 29–43., H-1118 Budapest, Hungary

Abstract: The sex pheromone of the raspberry cane midge has been identified and synthesised by East Malling Research (EMR) and Natural Resources Institute (NRI) and has proved to be highly attractive and useful for pest monitoring. East Malling Research coordinated a collaborative ring test of standard raspberry cane midge sex pheromone traps in 2006. The aims were to investigate the seasonal temporal pattern of the midge flight in different raspberry production regions of Europe and the relationship between the magnitude of catches and the numbers of eggs and larvae which developed subsequently in artificial splits in the primocane of untreated raspberry plantations. The standard raspberry cane midge sex pheromone trap used for the ring test consisted of a white delta trap containing a 20 x 20 cm sticky base and with a rubber septum lure impregnated with 10 µg of the raspberry cane midge sex pheromone racemate. Pairs of traps, separated by >20 m, were deployed in the centre of raspberry plantations at a height of 0.5 m in Italy, Hungary, Norway, Poland, Russia, Serbia, Sweden, Switzerland, Norway and the UK. The traps proved effective and easy to use for monitoring the flight of adult male raspberry cane midge. There were very large variations (~ 30 fold) between plantations in total numbers of midges caught over the season indicating plantations which are at comparatively low and high risk from the pest. Three generations of adult flight were apparent in Norway, Russia and Sweden and four generations in the central European countries with possibly 5 generations in Italy, though later generations were often difficult to distinguish. In the northern countries, the 1st generation first and peak flight occurred on Julian days 150 and 165, respectively, whereas in Italy the 1st generation first and peak flight occurred approximately on Julian days 110 and 130 respectively. The 1st generation flights occurred much earlier in polytunnel protected crops than in open field crops. Data obtained on the occurrence of larvae were variable in quality but a linear relationship between the peak numbers of males captured in the pheromone traps per week for a given generation (M) and the peak numbers of eggs and larvae per cm in artificial splits in the primocanes for that generation subsequently (L) was apparent (L = 0.025 M; R2=0.61). A nominal threshold of 30 midges per trap per week had been proposed but the linear relationship derived indicates that this threshold, which would result in ~ 0.75 eggs + larvae/cm, is too high. In reality, the degree of larval infestation that occurs and the resultant severity of crop damage will depend on the numbers of natural splits in the crop. The ring test is being continued in several countries in 2007.

Key words: Resseliella theobaldi, crop damage assessment, damage threshold, midge blight, pest monitoring, pheromone trap

11 12

Introduction

The sex pheromone of the raspberry cane midge Resseliella theobaldi (Barnes) (Diptera: Cecidomyiidae) has been identified and synthesised as (S)-2-acetoxy-5-undecanone by EMR and NRI and has proved to be highly attractive and useful for pest monitoring (Cross & Hall 2006). Standard large white delta traps (20 x 20 cm base) with a rubber septum lure loaded with 10 µg of the raspberry cane midge sex pheromone racemate have been adopted in the UK as a standard for monitoring the pest in commercial crops and are now available commercially from Agralan Ltd, The Old Brickyard, Ashton Keynes, Swindon, Wiltshire SN6 6QR UK. The racemate is also highly attractive, the opposite enantiomer being non- inhibitory. The lures are attractive for several months but it is recommended that they are replaced every four weeks. The number of midges captured in these traps have been shown to be greatly affected by the height of trap deployment, the greatest numbers being captured when the traps are placed on the ground. Therefore, a standard height for trap deployment of 0.5 m has been adopted for practical purposes. However prior to the work reported here, no information was available on thresholds and interpretation of trap catches to detemine the need and timing of sprays. In 2006, East Malling Research coordinated a collaborative ring test of standard raspberry cane midge sex pheromone traps in 8 European countries plus at one site in Siberia, Russia. The objective was to establish an economic damage threshold for raspberry cane midge catches in sex pheromone traps. It is recognised that thresholds may need to differ for different varieties and in different cropping systems. This was done by monitoring the flight activity of male raspberry cane midge in raspberry plantations with different midge populations throughout the season using sex pheromone traps and determining how flight activity relates to larval populations and crop damage. An overview of the results of this work, which is being continued in 2007, is reported here.

Materials and methods

The ring test was done in 8 European countries plus at one site in Siberia, Russia (Table 1). A number of well separated, reasonably large raspberry plantations of one or two (the dominant) cultivars were selected with different populations of raspberry cane midge in each country/production region. In part of the plantation, preferably in the centre, an area not to be treated with insecticides for raspberry cane midge was demarked. Ideally, this area should be sufficiently wide for insecticide spray drift into the central 2-3 rows to be minimal. At each site, two (white) delta pheromone traps were deployed in the centre of each plantation, separated by at least 20-30m. They were suspended so that the base was at a height of 0.5 m above the ground. One trap was oriented parallel to the rows, one at right angles, to avoid bias (the catches from the two traps were averaged). Each trap contained a 20 x 20 cm white sticky base. Traps were baited with a standard lure loaded with 10 µg of the raspberry cane midge sex pheromone major component racemate, which were replaced at 1 month intervals. Traps, lures and bases were supplied by East Malling Research to those participating.

At most sites, traps were set out before the first flight of the midges in spring and maintained and recorded weekly until the midge flight ceases at the end of the growing season. The sticky bases were refreshed weekly. The numbers of midges in each trap were counted as follows: 0 – 200 = exact count; >201-1000 = estimated to nearest 10; > 1000 = estimated to nearest 100 except in Norway the numbers were estimated to nearest 50 above 150. Sticky bases were renewed on each recording occasion (unless they are empty or virtually empty and could be cleaned). The species and sex of the midges were checked. Key recognition features of raspberry cane midge (Resseliella theobaldi) and other similar midges in the genus are 1) Diptera, two winged fly; 2) body about 2 mm long, abdomen red/orange in colour; 3) wings sparsely clothed with short dark hairs, very few veins (only 3 are easily visible, the two anterior ones reaching the wing margin); 4) the first wing vein reaches the wing tip at its apex more or less exactly; 5) long legs, often broken on sticky bases; 6) males have a pair of claspers on rear of abdomen (females, which are not attracted by the pheromone, have a long protractible ovipositor); 7) male antennae long, filiform, characteristically curved back over head in a graceful arch and beaded (beads each separated by a short narrow stem), with conspicuous whorls of hairs on each segment (Note that female antennae are not markedly beaded and are much shorter). Other midges caught in significant numbers were to be recorded but none occurred. In order to provide oviposition sites for females, weekly, throughout the growing season, a 10 cm long, artificial split was made in each of 20 primocanes in the central, untreated area of each plot. This was done by drawing a hooked needle vertically along the cane, making a slit through the periderm. Care was taken not to injure the cambium below. The needle tip was angled sideways (tangentially to the circumference of the cane) so that the periderm was separated from the cambium tissue and making a flap under which ovipositing cane midge females could lay their eggs. Making this flap was critical for otherwise the wound healed and oviposition was be impossible. The canes in which artificial splits were made were marked with coloured tape so that they could easily be re-located. It was intended that each week, 10 primocanes with one week old and 10 primocanes with 2-week-old artificial splits, would be collected from each plantation and the number of eggs and larvae in each split counted under a binocular microscope in the laboratory. However, at some sites the sampling was done fortnightly and at the sites in Norway it was done only once. L1 and L2 larvae, which are small and translucent, turning slightly pinkish, were counted separately from L3 and L4 larvae which are larger, opaque and salmon pink or orange/yellow. The length of each split was recorded so that the number of larvae per unit length of split could be calculated. The growth stage of the raspberries was recorded on each sampling occasion.

Results and discussion

Variations in total trap catches The traps proved effective and easy to use for monitoring the flight of adult male raspberry cane midge. There were very large variations (~ 30 fold) between plantations in total numbers of midges caught over the season indicating plantations which are at comparatively low and high risk from the pest. The smallest total number (445) was captured at Otham, Kent, UK, a plantation treated with chlorpyrifos each year to control cane midge. Chlorpyrifos is highly effective against the midge giving > 90% control. The greatest total number was captured in the untreated plantation at Berkenye, Hungary (13845). Variations in totals numbers captured probably reflect a number of interacting factors including the susceptibility of the variety, the age and size of the plantation and the history of insecticide treatments against cane midge. It

is assumed that plantations with high numbers are at greater risk of damage if control measures are not applied. However, no information is available on the likely level of damage which would result if plantations with low numbers of midges were not treated with insecticide.

Number of generations The numbers of generations that occurred at the different sites were often difficult to distinguish, either because later generations were overlapping, because of strong fluctuations in numbers of midges caught or in some cases because too few midges were caught. Three generations of adult flight were apparent in Norway (Figure 1), Russia and Sweden. The first generation was distinct with numbers of midges caught falling to low levels before the second generation flight commenced. The later two generations overlapped. Four generations of adult flight were apparent in central Europe with possibly even 5 generations in the protected crops in Italy (Figure 2), though later generations were difficult to distinguish.

1000 900 Gloppestad 800 Vereide 700 600 500 400 Trap catch Trap 300 200 100 0 100 120 140 160 180 200 220 240 260 280 300 Julian day

Figure 1. Pheromone trap catches of midges in the two fields in Norway in 2006.

1000 900 800 Levico Novalado 700 600 500 400 Trap catch 300 200 100 0 100 120 140 160 180 200 220 240 260 280 300 Julian day

Figure 2. Pheromone trap catches of midges in the protected raspberry crops at Levico and Novaledo, Italy in 2006.

Time of first generation flight In the northern countries, the 1st generation first and peak flight occurred on Julian days ~150 and ~165, respectively, whereas in the protecteed crops in Italy, the 1st generation first and peak flight occurred approximately on Julian days 110 and 130 respectively (Table 2). The 1st generation flights occurred much earlier in polytunnel protected crops than in open field crops.

Table 2. Total number of midges captured per trap, number of midge generations and Julian dates of 5% and peak catch of first generation.

Total number Julian Julian date of No of Country Site midges date of peak catch of generations captured/trap 5% catch 1st generation Italy Pergine 8241 5? 112 130 Roncogno 2097 5? 124 137 Levico 6664 5? 114 145 Novaledo 4500 5? 126 137 Hungary Berkenye A 5853 4 118 135 Berkenye B 13845 4 116 129 Norway Gloppestad 5175 3 148 160 Vereide 6779 3 148 167 Poland Dabrowice 3503 4 † 138 Michrowek 5147 4 † 131 Russia Novosibirsk 4930 3 153 160 Serbia Zdravljak 656 4 132 142 Arilje 959 4 115 129 Djeradj 2438 4 † 124 Sweden Bodarpsgården 1110 3 156 171 Switzerland Ardon 3608 4 122 130 Bruson 5176 3 143 152 Salins 2299 4 128 146 UK Otham 445 4 126 138 Penshurst 4817 4 106 124 † Monitoring was started too late

Relationship between trap catches and numbers of larvae in splits Data obtained on the occurrence of larvae were variable in quality but a linear relationship between the peak numbers of males captured in the pheromone traps per week for a given generation (M) and the peak numbers of eggs and larvae per cm in splits in the primocanes for that generation subsequently (L) was apparent (L = 0.025 M; R2=0.61) (Figure 3). A nominal threshold of 30 midges per trap per week had been proposed but the linear relationship derived indicates that this threshold, which would result in ~ 0.75 eggs + larvae/cm, is too high. In reality, the degree of larval infestation that occurs and the resultant severity of crop damage will depend on the numbers of natural splits in the crop. The ring test is being continued in several countries in 2007.

50 y = 0.0253x 2 40 R = 0.6085

10 Peakeggs+ larvae / cm 0 0 200 400 600 800 1000 1200 1400 Peak catch

Figure 3. Relationship between the peak numbers of males captured in the pheromone traps per week for a given generation (generations 1-3 only) and the peak numbers of eggs and larvae per cm in splits in the primocanes for that generation subsequently.

Recommendations for use of raspberrry cane midge pheromone traps

• Raspberry cane midge sex pheromone traps available commercially from Agralan, UK since spring 2007. They are the standard white delta design with 20 x 20 cm bases. • The traps should be deployed from 1 April – 30 Sept and monitored weekly. Earlier initial deployment may be necessary in crops which are protected very early. • Sticky bases should be renewed each week unless none or very few midges that can be scraped off are present. • The standard height of trap deployment is 0.5 m. • At least 1 trap should be deployed per crop system per farm, in the centre of the crop. • Lures should be renewed monthly. • Store lures in freezer ( -20 ˚C). • A nominal threshold 30 midges/trap is proposed for timing sprays. • Concentrate on control of first generation but control of late season (post-harvest) generations should be considered if numbers are high. • Sprays should be applied a few days after threshold is reached. • OP insecticides (chlorpyrifos, diazinon) are the most effective. • Peak catches of 500 midges/generation led to an average of 10 eggs + larvae/cm of split. • Total catch of 1000 midges/generation led to an average of 20 eggs + larvae per split. Tolerance threshold for larvae is unknown but probably very low, e.g. 0.1 larvae/cm. • Damage threshold is low and significant larval infestation could occur below the nominal threshold of 30 midges/trap in some circumstances.

Acknowledgements

We acknowledge Rune Vereide for assistance with the field work in Norway.

Reference

Cross, J.V. & Hall, D.R. 2006: Sex pheromone of raspberry cane midge. – IOBC/wprs Bulletin 29(9): 105-109.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 19-25

Raspberry cane midge – flight dynamics, egg laying and the efficacy of the neonicotinoid insecticide acetamiprid on primocane fruiting raspberry

Barbara H. Łabanowska, Jerry V. Cross Research Institute of Pomology and Floriculture, Pomologiczna 18, 96-100 Skierniewice, Poland; East Malling Research, New Road, East Malling, Kent ME19 6BJ UK

Abstract: The raspberry cane midge, Resseliella theobaldi (Barnes), is one of the main pests of raspberry plantations in Poland. Larvae damage the primocanes of raspberry. The pest is very dangerous to cultivars fruiting in June-July as well as on primocane fruiting cultivars (August - October harvest). The experiment reported here was as part of the ‘ring test’ coordinated by East Malling Research (UK). The flight of the midges was monitored weekly using standard pheromone traps in untreated commercial raspberry plantations in Poland. The standard white delta traps (20 x 20 cm base) were baited with a rubber septum lure loaded with 10 µg of the raspberry cane midge sex pheromone ‘racemate’ and were placed at a height of 0.5 m above soil. Numbers of eggs and larvae were monitored in artificial splits on primocanes. The midge developed at least three generation a year in Poland. The flight started at the end of April or at the beginning of May and lasted until beginning of October. In an unprotected raspberry plantation, on cultivar Polka, the first generation of the midge’s intensive flight was noted in the end of April or in the first decade of May with the second generation in the second half of July. The most intensive flight and the highest number of males caught in pheromone sticky traps was noted at the end of August and at the beginning of September. This was thought to be the third generation of the raspberry cane midge. The neonicotinoid insecticide, Mospilan 20 SP (acetamiprid) showed promising results in the control of the raspberry cane midge, similar to the organophosphate malathion.

Key words: Resseliella theobaldi, Rubus, primocane, acetamiprid, control

Introduction

The raspberry cane midge, Resseliella theobaldi (Barnes), is one of the main pests of raspberry plantations in Poland. Larvae damage the primocanes of raspberry by direct feeding under the skin. This pest is very dangerous on cultivars fruiting in June-July as well as on primocane fruiting cultivars (August - October harvest). Usually the raspberry cane midge appears together with various raspberry stem diseases including Didimella applanata and Botrytis cinerea. This insect is also a damaging pest in England, Scotland, Switzerland, Hungary and other countries (Birch et al. 2004; Gordon et al. 2002a,b; Vétek et al. 2006). It is important to know the term of the pest’s flight and to find safe insecticides to control it (Gordon et al. 2002a), which will be useful in IPM programmes. According to Antonin et al. (1998) control in spring, at the beginning of the first generation’s flight is recommended. The best method to fix the date of chemical control of some pests, i.e. Synanthedon tipuliformis on black currant, is using pheromone traps to catch moths (Łabanowska 2001). To fix the optimal time of chemical control of black currant stem midge (Resseliella ribis), a similar pest to the raspberry cane midge, the dynamic of egg laying in artificial injuries on one year shoots were observed (Łabanowska 1997, 2001). The aim of this

19 20

work was monitoring the population of the raspberry cane midge using pheromone traps (Cross & Hall 2005) and checking the efficacy of a neonicotinoid insecticide to control it.

Material and methods

The research study was a part of the ‘ring test’ coordinated by East Malling Research.

Monitoring of the raspberry cane midge flight The experiments were carried out during 2006 – 2007 on a raspberry plantation cv Polka fruiting in August-September on primocanes. The experiment was carried out on the part of 4- 5 year old commercial plantation in Michrówek, near Warsaw (central Poland). The standard pheromone white delta traps (20 x 20 cm base) were baited with a rubber septum lure containing 10 µg of the raspberry cane midge sex pheromone ‘racemate’ and were placed at a height of 0.5 m above the soil. The flight of midges was monitored weekly. The pheromone lures were changed monthly during the season, but the trap sticky bases were changed every week.

Monitoring the raspberry cane midge egg laying Numbers of eggs and larvae were monitored in artificial splits (injuries) on primocanes, which were made once a week throughout the whole vegetation season. Every week the artificial splits were made in 20 primocanes, on untreated and on treated raspberry rows (it was small part of a large (~ 1.5 ha) production plantation). One week later 10 primocanes were collected and taken to laboratory. Next 10 primocanes were collected two weeks after the splits had been made. This was repeated during whole vegetation season. Numbers of eggs and larvae were counted in artificial splits on primocanes in the laboratory under a stereoscopic microscope.

Chemical control Experiments to evaluate the efficacy of insecticides for the control of the raspberry cane midge were carried out during 2006-2007. The experiments were localized on a part of the same commercial raspberry plantation Polka in Michrówek. The experiments were carried out in a randomized block design with four replicates of 10.5 m2 –12 m2 each (1 row 3.5-4 m long). Before treatment the 5 artificial injuries, 2-5 cm long, in the periderm of the 4 primocanes were made. It was a way to provoke the females to lay eggs there, instead of in natural splits on young canes. After that, the raspberry plots were sprayed. A knapsack motor sprayer ‘Stihl’ was used to apply the spray solution in a volume of 750 l/ha. The efficacy of the insecticides was evaluated 8-10 days after spray-treatment. The primocanes with artificial injuries were collected for laboratory examination. The length of injuries was measured in cm. The eggs and larvae in splits were counted under a stereoscopic microscope. Results are shown as a number of eggs and larvae per 10 cm of injury. The dates of collection of canes and treatments in each experiment are shown under the Tables 1a and 1b. In 2006, in the plantation where the 4 treatment programme was applied, in the autumn, 10 canes from each plot were collected and taken to the laboratory. The number of natural splits was counted and measured and the number of larvae under skin was recorded. The first and the second treatment were applied during the flight of pest’s first generation, when the primocanes were 40 and 50 cm high. The next treatment was applied when first flower buds were observed, and the last one just before blossom. The dates are shown in the Table 2. 21

The results were analysed statistically on data transformed according to functions y = log (x + 1), where x – number of eggs and larvae per 10 cm of split. Significance of differences between means was tested using Newman-Keuls test at level α = 0.05 The efficacy of product in % was calculated on base of averages according to Abbott’s formula, expressed as: C% = a-b/a x 100, where: a – number of individuals on untreated, check plot; b – number of individuals after treatment on treated plot.

Results

Monitoring the raspberry cane midge flight In 2006, the flight of the raspberry cane midge started in the first half of May and lasted until the beginning of October. In an unprotected raspberry plantation (Polka), the first generation intensive flight and the highest number of males caught on pheromone sticky traps was noted in the first decade of May, and the second generation in the second half of July (Figure 1A). The most intensive flight and the highest number of males caught in traps were noted at the end of August and the beginning of September. This was thought to be the third generation of the raspberry cane midge. In 2007, the flight started at the end of April and lasted until the beginning of October (Figure 1B). The most intensive flight of the first generation of midges was observed in the second decade of May, the second generation – at the end of the second and beginning of the third decade of June and the last generation during second half of July and beginning of August. In general, two years of observation confirmed that the raspberry cane midge has at least 3 generation a year in Poland. The first generation appears at the end of April and in May, before blossom of all raspberry cultivars. The second generation appears just before and during blossom of primocane fruiting cultivars, or during harvest of cultivars fruiting on second year canes. The last generation is flying during harvest of primocanes fruiting cultivars but after harvest of cultivars fruiting on second year canes.

900 900 A B beginning of harvest 800 beginning of harvest 800

700 700

600 Untreated 600 Treated beginning of blossom beginning of blossom 500 500

400 400 Untreated

300 300 Treated No of males in trap No of males in trap 200 200

100 100

0 0 26 10 24 7 21 5 19 2 16 30 13 27 11 112582272031731142812 April May June July August September May June July August September

Figure 1. Dynamic of the raspberry cane midge (R. theobaldi) flight in A) Michrówek 2006; B) Michrowek 2007. Vertical arrows indicate treatments.

Monitoring the raspberry cane midge egg laying In 2006 in the unprotected raspberry plantation the highest number of eggs and larvae of the raspberry cane midge was noted in the second half of May (in one week old splits) and at the beginning of June (in two weeks splits) (Figure 2). There were over 20 eggs and larvae per cm 22

of split. Later on in season the number of eggs and larvae was much lower. In 2007 the most intensive flight of the first generation was noted a bit earlier – in the first decade of May, but the second generation - in the second half of July, and the third generation at the end of August and of the beginning of September (Figure 3). In May there were over 30 eggs and larvae per cm of split. In the next months the number of eggs was lower. Generally, during the whole vegetation season eggs were laid to artificial splits on primocanes and larvae were feeding under the skin. These observations allowed us to establish the time of massive flight and egg laying and also to fix the optimal date for chemical control. Monitoring of egg laying was used with good effect to fix the term of the black currant stem midge control (Łabanowska 1997), but the sex pheromone traps were better to monitor the black currant clearwing moth (Łabanowska 2001).

30 8 A beginning of harvest B beginning of harvest

6 20 eggs eggs larvae beginning of blossom beginning of blossom larvae 4

No of eggs andNo larvae of eggs andNo larvae of 2

0 0 11258 227 203 1731142812 11 25 8 22 7 20 3 17 31 14 28 12 April May June July August September May June July August September

Figure 2. Number of eggs and larvae of the raspberry cane midge (R. theobaldi) per 1 cm of split in unprotected raspberries - Michrówek 2006 –A) one week old split; B) two weeks old split.

50 50 A beginning of harvest B beginning of harvest

40 40

eggs eggs larvae 30 beginning of blossom 30 beginning of blossom larvae

20 20 No of eggs andNo larvae of eggs andNo larvae of 10 10

0 0 3173114281225923620418 3 1731142812259 236 204 18 May June July August September May June July August September

Figure 3. Number of eggs and larvae of the raspberry cane midge (R. theobaldi) per 1 cm of split in unprotected raspberries - Michrówek 2007 –A) one week old split; B) two weeks old split.

Chemical control The chemical control of the pest is not easy because the adults, eggs and larvae were present in the plantation during the whole vegetation season. The best time to control the raspberry cane midge is during the flight of first generation of the pest, in all kinds of plantation. The possibility to control the second and the third generation depends on the time of fruiting of grown cultivar. The neonicotinoid insecticide Mospilan 20 SP (acetamiprid) showed promising results in the control of the raspberry cane midge. The insecticide used in the plantation with artificially injured canes reduced the number of eggs and larvae by 77.9 - 85.3%, depending on the experiment (Table 1a,b). The results were similar to the standard insecticide malathion (Fyfanon 500 EC).

Table 1. Efficacy of acetamiprid (Mospilan 20 SP) in the control of raspberry cane midge on raspberry (artificial injuries on canes) a. Michrówek, 2006 Number injuries Number eggs and Dose rate Insecticide with larvae per 10 larvae per 10 cm Reduction in % l/kg/ha injuries * injury Untreated – check - 2.74 c** 3.40 b - Mospilan 20 SP 0.125 0.4 ab 0.50 a 85.3 Fyfanon 500 EC 2.5 0.0 a 0.0 a 100.0 Artificial injuries has been made on 8 of July 2006, before treatment * Assessment of the eggs and larvae numbers in injuries - 18.07.2006 b. Kruszew I, 2006 Number injuries Number eggs and Dose rate Insecticide with larvae per 10 larvae per 10 cm Reduction in % l/kg/ha injuries * injury Untreated – check - 4.76 a ** 9.18 a - Mospilan 20 SP 0.125 2.33 a 2.03 a 77.9 Fyfanon 500 EC 2.5 1.35 a 1.37 a 81.5 * Artificial injuries has been made on 22 of June 2006, before treatment * Assessment of the eggs and larvae numbers in injuries – 4.07.2006 ** Means followed by the same letter do not differ significantly at P=0.05 according to Newman- Keuls test

In one experiment, on plantation where a programme of 4 treatments of Mospilan 20 SP was applied, over 72% reduction of the pest was obtained (Table 2). However, the treatment was applied in spring and early summer only. The results were similar to the standard organophosphate compound Fyfanon 500 EC (malathion). The method will be useful for controlling the pest on all raspberry cultivars, but acetamiprid might also be used on plantations where an Integrated Pest Control Programme is applied. 24

Table 2. Efficacy of acetamiprid (Mospilan 20 SP) used in a programme for the control of raspberry cane midge on raspberry plantation - Michrówek, 2006

No of No of natural Dose rate injuries with No of larvae Reduction in Insecticide injuries per l/kg/ha larvae per 10 per 10 shoots % 10 shoots shoots Untreated – check - 55.3 a ** 19.8 b 99.2 b - Mospilan 20 SP 0.125 35.8 a 10.6 ab 27.5 a 72.3 Fyfanon 500 EC 2.5 32.3 a 8.2 a 25.4 a 74.4 • Dates of treatment: 12 of May – primocanes 40 cm in high • 25 of May - primocanes 50 cm in high • 22 of June - first flower buds appears • 8.07 of July - just before blooming: Assessment - 23 of October 2006 **Explanations see Table 1

Conclusions

The raspberry cane midge has at least three generations per year, and adults, eggs and larvae are present in the plantation during the whole vegetation season. On primocane fruiting cultivars chemical control of the raspberry cane midge is possible during the flight of the first and second and also a part of the third generation. On cultivars fruiting on second years canes, chemical control is possible during the flight of the first and third generation. The neonicotinoid insecticid Mospilan 20 SP (acetamiprid) showed promising results for the control of the raspberry cane midge.

Acknowledgments

The authors would like to thank Elżbieta Paradowska and Stanisław Lesiak for their technical help in conducting the experiments.

References

Antonin, P., Mittaz, C., Terrattaz, R. & Linder, Ch. 1998: Raspberry cane midge Resseliella theobaldi (Barnes) II. Towards a control strategy. – Rev. Suisse vitic. arboric. hortic. 30: 389-393. Abstract. Birch, A.N.E., Gordon, S.C., Fenton, B., Malloch, G., Mitchell, C., Jones, A.T., Griffiths, W., Brennan, R., Graham, J. & Woodford, J.A.T. 2004: Developing a sustainable IPM System for high value Rubus crops (Raspberry, Blackberry) for Europe. – Acta Hort. 649: 289-292. Cross, J. & Hall, D. 2005: Sex Pheromone of Raspberry Cane Midge. – Presentation at Soft Fruit Conference, Ashford, 23-24 November 2005. Gordon, S.C., Woodford, J.A.T., Barrie, I.A., Grassi, A., Zini, M., Tuovinen, T., Lindquist, I., Höhn, H., Schmid, K., Breniaux, D. & Brazier C. 2002a: Development of a Pan- European monitoring system to predict emergence of first-generation raspberry cane midge in raspberry. – Acta Hort. 585: 1: 343-348. Gordon, S.C., Woodford, J.A.T., Grassi, A., Höhn, H. & Tuovinen T. 2002b: The ‘Racer’ Project – a blueprint for Rubus IPM research. – Acta Hort. 585: 1:303-307. 25

Łabanowska, B.H. 1997: The egg laying and effectiveness of control of the black currant stem midge – Resseliella ribis (Marik.) [Diptera, Cecidomyidae]. – Zesz. Nauk. Inst. Sadow. Kwiac. 4: 135-147. (in Polish with english summary). Łabanowska, B.H. 2001: Monitoring and control of currant clearwing moth (Synanthedon tipuliformis (CI) and black currant stem midge (Resseliella ribis Marik.) on black currant plantations. – IOBC wprs. Bulletin 24 (5): 253-258. Vétek, G., Fail, J. & Penzes, B. 2006: Susceptibility of raspberry cultivars to the raspberry cane midge (Resseliella theobaldi Barnes). – J. Fruit Ornam. Plant Res. 14 (Suppl. 3): 61-66. 26 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 27-31

Interference between raspberry cane midge (Resseliella theobaldi) sex pheromone traps – A one season trial in a Swedish raspberry plantation

Thilda Nilsson1, 2, Christer Tornéus1 1 Swedish Board of Agriculture, Plant Protection Centre, P.O. Box 12, S-230 53 Alnarp, Sweden; 2 Swedish University of Agriculture, Alnarp, Sweden

Abstract: Raspberry cane midge is a serious pest in raspberries. Larvae feed under the epidermis of first year canes and in these wounds several fungi can infect. The complex of larval feeding and fungal infection is termed midge blight. Efficient control is difficult to achieve due to the sheltered place of the larval feeding and the short lifespan of adults. Monitoring of the male flight period is possible with sex pheromone traps. A field trial was conducted to see if pheromone traps interfere when hung at a set distance from each other. The results indicate that traps placed in close vicinity (7 and 20 metres) show a decrease in number of trapped midges, thus monitoring traps should be placed more than 20 metres apart to give representative data of the male flight pattern.

Keywords: Resseliella theobaldi, raspberry cane midge, sex pheromone traps, raspberry

Introduction

Raspberry cane midge, Resseliella theobaldi (Barnes), is a serious pest of raspberry. The midge belongs to the gall midge family, Cecidomyiidae. The raspberry cane midge causes damage to first-year canes of raspberries by larval feeding under the epidermis of canes. In these wounds several fungal infections can occur (Gordon & Williamson 1991; Pitcher 1952). This complex of larval feeding and fungal infection is called midge blight (Pitcher & Webb 1952). Symptoms of midge blight are dark, sunken lesions on the canes and the following year heavily infested canes can die back or fail to produce lateral shoots (Pitcher 1952). The protected feeding place for larvae under the epidermis of first-year canes is a contributing factor to the difficulties with control. The timing of insecticidal applications in spring is another problem which can be aided by monitoring with pheromone traps (Cross & Hall 2005). In this work the trap catches of sex pheromone traps placed in a square pattern at a set distance are evaluated. The hypothesis stated that there will be a proven interference between pheromone traps located in a near vicinity of each other, resulting in fewer midges trapped in these due to confusion.

Material and methods

Pheromone traps of 20 x 20 cm white delta design with a rubber septum lure loaded with 10 µg of the sex pheromone racemate were used in the trial. Interference between pheromone traps was studied by mounting of five delta traps in a square with a middle trap, i.e. making a dice pattern. Two different sizes of squares were used, one with the outer traps 20 metres apart and one with the outer traps seven metres apart (Figure 1). Between the two squares there was a control distance of approximately 60 metres. In addition a single trap was

27 28

deployed as a control an additional 60 metres from the closest trap. Two sprays of an organic phosphorous insecticide, Gusathion, were applied twice before flowering. The interference experiment was started on the 7th of May in connection with the sightings of the first generation. The sticky bases were changed if needed in connection to the observations conducted twice a week. Insects were counted by using hand lens or under a stereomicroscope. A repetition of the interference experiment was conducted when repression of catches were seen in the small square compared to the large square and the control trap. The repetition experiment was set-up on the 17th of July. The first interference trial was ended on the 27th of July when some of the rows were eradicated. The second trial continued until the 20th of September. In total 24 readings of trap catches from the first repetition were made and 17 readings were made from the second repetition.

Small square: 7 metres Large square: 20 metres

Figure 1. Schematic picture of the set-up of pheromone traps. Two sizes of squares were used, outer traps 7 metres apart and outer traps 20 metres apart.

Results and discussion

The results from the interference trial show statistical differences in number of caught midges from the two middle traps and the control trap. The control trap caught most midges, the middle trap of the large square caught fewer midges but more than the middle trap in the small square. In both squares significant differences in catch amount were found between the outer traps, indicating external variables affecting the result. Thus the results indicate that monitoring traps placed too close to each other give unrepresentative data. Interference between traps in the square pattern was analysed by the non-parametric Friedman’s test (Figure 2). The test results indicate that there were statistical differences between the number of caught midges in the middle traps of the two square sizes and the single control traps in both replicates (rep.1: p=0.002, rep.2: p=0.018). The differences between trap catches of the four outer traps in each square were also analysed with Friedman’s test to see if the trap catches differed significantly (Table 1). The results from the test showed that trap catches all differed significantly except in the large square in the second repetition. In the rows where the first repetition of the interference experiment was conducted it could be observed that traps with low trap catches were hung in plants that showed symptoms of either virus or were of poor vigour. When these effects 29

became clear the square of traps were moved a few meters to healthier plants. This was the case for both the small and the large square in the first repetition, as these traps were placed before symptoms occurred. It seems like several factors affect trap catches and that the poor vigour of some plants lead to the differences in trap catches between the outer traps.

mean Interference between traps caught midges Rep. 2 140 Rep. 1 120 100 80 60 40 20 0 Middle trap in small Middle trap in large square Single control trap square

Figure 2. The mean number of trapped midges in the control trap and the two middle traps.

Table 1. The results from the statistical test where the differences in trap catches of the outer traps in the squares are analysed. All gave significant differences in trap catches except for large square in rep. 2.

Treatment Small square Large square Small square Large square (rep. 1) (rep. 1) (rep. 2) (rep. 2) - 13,6 - 8,4 - 42,9 - 55 Mean trapped midges - 19,3 - 10,5 - 53,5 - 63,8 from the four outer - 8,8 - 30,1 - 72,1 - 73,8 traps - 25,1 - 37 - 59,3 - 55,3 Statistical difference p= 0.001* p= 0.000* p= 0.011* p= 0.508 ns

The trap catches from the traps in the experiment were recorded twice a week and the recordings show three periods that could correspond to three generations: the first from late May to mid June, the second from mid July to the end of July, the third occurred in August (Figure 3-4). Two sprays of an insecticide were made on the 16th and 28th of May with the result of a fast decrease in the amount of caught male midges after the second application . Pheromone traps situated in the near vicinity of each other affect the result of the trap catch. Middle traps in the small squares caught less midges than middle traps in large squares and the highest number of midges were trapped in the single control traps. Thus the results indicate that monitoring traps placed too close to each other give unrepresentative data. 30

Rep. 1 Middle trap in small square caught midges / Single control trap trap Middle trap in large square 160 140 120 100 80 60 40 20 0

j j j j a a un un jul jul m m -jun -j -jun -j -jul 9-ma 6- 3-ma 0- 6 3 0 7 8 0 1 2 3 0 1 2 2 04-jul 11- 1 25-

Figure 3. Trap catches from the experiment repetition 1 that was carried out from May until end of July. The two arrows indicate the two applications of the insecticide.

Rep. 2 Middle trap in small square caught midges / Single control trap Middle trap in large square trap 240 200 160 120 80 40 0

ul g g g j aug au u u aug 4- 3-aug 1-aug 9- 20-jul 2 28-jul 01- 05-aug 09- 1 17-a 2 25-a 2 02-sep

Figure 4. Trap catches from the second repetition of the experiment, which was carried out from July until September.

Acknowledgments

We would like to send our gratitude towards the East Malling Research and Natural Resources Institute that provided with pheromone lures and traps. A special thanks to Mr Jerry Cross for his hospitality and appreciated input. Swedish Board of Agriculture provided with the financial support. We also send our gratitude to Kerstin and Johan Biärsjö, owners of the plantation.

References

Gordon, S.C. & Williamson, B. 1991: Raspberry cane midge. – In: Compendium of raspberry and blackberry diseases and insects. Ellis, Converse, Williams, Williamson (eds.): 75-76. 31

Pitcher, R.S. & Webb, P.C.R. 1952: Observations on the raspberry cane midge (Thomasini- ana theobaldi (Barnes)) II. “Midge Blight”, a fungal invasion of the raspberry cane following injury by T. theobaldi. – J. Hortic. Sci. 27: 95-100. Pitcher, R.S. 1952: Observations on the Raspberry cane midge (Thomasiniana theobaldi (Barnes)) I. Biology. – J. Hortic.Sci. 27: 71-94. Cross, J. & Hall, D. 2006: Sex pheromone of Raspberry cane midge. – IOBC wprs Bulletin: 29 (9): 105-109. 32 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 33-40

Some preliminary investigations into the sex pheromones of mirid soft fruit pests

Michelle Fountain1, Jerry Cross1, Gunnhild Jaastad1, David Hall2 1 East Malling Research, New Road, East Malling Kent ME19 6BJ UK; 2 Natural Resources Institute, University of Greenwich, Central Avenue, Chatham Maritime, Kent ME4 4TB UK

Abstract: The European tarnished plant bug, Lygus rugulipennis, is an important pest of soft fruit crops throughout Europe causing fruit deformities in strawberry and raspberry crops. Lygocoris pabulinus is often responsible for killing fruiting laterals in raspberry and damaging blackcurrent shoots. Routine sprays of broad-spectrum insecticides, used to control other pests, have kept mirid pests at low levels. However, a reduction and replacement of broad-spectrum pesticides with more selective ones is causing resurgence. As part of a larger programme to develop practical, long-lived and economic monitoring devices for these pests for use by growers, we report some initial findings and challenges involved in understanding the mechanisms and requirements for monitoring mirids. We determined that female L. rugulipennis attract males between 0500-1030 hours and that green sticky base delta traps were effective. Field tests in 2007 using micro-capillary dispensers with a reservoir, baited with various combinations of the 3 compounds produced by female L. rugulipennis and L. pabulinus (hexyl butyrate, (E)-2-hexenyl butyrate, and (E)-4-oxo-2-hexenal) failed to attract males.

Key words: capsid, European tarnished plant bug, Lygus rugulipennis, mirid, common green capsid, Lygocoris pabulinus, pest monitoring, pheromone lure

Introduction

Mirid pests, commonly known as capsids (Heteroptera, Miridae, Mirinae), are common and important pests of horticultural and agricultural crops world-wide. In the UK, the European tarnished plant bug, Lygus rugulipennis, the common green capsid, Lygocoris pabulinus, and the nettle capsid, Liocoris tripustulatus, are common pests of various horticultural crops. In conventional systems they are controlled by sprays of broad-spectrum insecticides (Garthwaite & Thomas 2001), organophosphorus insecticides (OPs) being the most effective and frequently used. However, there is an increased removal of OPs from use in pest control and neonicotinoids and other modern insecticide groups have only a limited efficacy against capsids. Insect growth regulators are ineffective against these pests and organic crops are particulary susceptable. These capsids are common and often abundant on a wide range of wild plants including common weeds. They have comparatively few natural enemies and none that seem to reduce populations to low levels in the wild. Effective biocontrol methods have not been developed for them in Europe, restricting the development of IPM programmes. Crop invasion by capsids is sporadic, unpredictable, and can cause severe economic losses. They cause damage at low population densities (<1 capsid/40 plants) and are difficult to detect in crop inspections (Jay et al. 2004). Insect sampling methods (sweep-net or beating- tray sampling) are time consuming and unsuitable for use by growers. Late season strawberry crops require almost routine treatment with insecticides in mid and late summer again L. rugulipennis. Feeding in flowers and on green fruits can cause up to

33 34

80% crop loss (Cross 2004). L. rugulipennis causes similar damage in raspberry and L. pabulinus terminates fruiting laterals in this crop (Cross 2004). L. rugulipennis has also become a significant problem on cucumber in protected crops where the use of broad- spectrum insecticides in IPM programmes has been reduced (Jacobson & Hargreaves 1996; Jacobson 1999; Jacobson 2002). L. pabulinus is a damaging pest of blackcurrant, attacking the shoots and a sporadic pest on apple and pear (Bus et al. 1995; Umpelby et al. 1995; Garthwaite et al. 2000; Anon. 2004). Liocoris tripustulatus causes surface scars and misshapes in peppers and aubergines rendering the fruit unmarketable (Jacobson 2001). Several species of mirid have been shown to produce sex pheromones. Traps baited with virgin females of both L. rugulipennis (Innocenzi et al. 1998; Glinwood et al. 2003) and L. pabulinus (Blommers et al. 1988) have been shown to attract conspecific males. During 30 years of work in the USA and Canada on the sex pheromones of Lygus lineolaris (Gueldner & Parrot 1978; Aldrich 1988; Wardle et al. 2003) and Lygus hesperus (Ho & Millar 2002), several potential pheromone components were identified, however, attraction to a synthetic lure was never demonstrated. Until recently, attraction to synthetic pheromone lures has been demonstrated in only eight mirid species belonging to different genera (Smith et al. 1991; Millar et al. 1997; Millar & Rice 1998; Zhang et al. 2003; Zhang et al. 2003; Kakizaki & Sugie 2001; Downham et al. 2002; Okutani-Akamatsu et al., 2007). Sex pheromone identification in mirids has been hampered by the abundant defensive secretions, present in the metathoracic scent gland, that are released upon disturbance (Aldrich 1988). Furthermore, it is probable that certain compounds can function both as components of the pheromone and as defensive secretions (cf. Blum 1981, 1996). Previous research on L. rugulipennis (Innocenzi et al. 2004; Cross & Hall 2003) identified three female-specific pheromone components, hexyl butyrate, (E)-2-hexenyl butyrate and (E)-4-oxo-2-hexenal. However, in initial field trials, traps baited with synthesised blends of these chemicals dispensed from standard pheromone dispensers failed to catch significant numbers of males (Innocenzi et al. 2004). Between 1996 and 2000, the same three compounds were shown to be produced by both male and females L. pabulinus (Drijfhout 2001; Groot 2000; Drijfhout et al. 2000), but these compounds only caused electroantennogram responses in the antennae of males (Drijfhout et al. 2002). No attraction of males could be demonstrated in laboratory or field when they were dispensed in various blends and release rates from standard pheromone dispensers (Groot 2001). Groot et al. (2001) reported that hexyl butyrate inhibited pheromone release by the females. An important breakthrough occurred in subsequent field trials by researchers at East Malling Research and the Natural Resources Institute in the UK, in which the chemicals were released from glass microcapillary tubes, as reported by Kakizaki & Sugie (2001) for another insect pheromone. Different blends of the components were found to attract L. rugulipennis and its congener Lygus pratensis. This was the first time a Lygus bug pheromone had been identified and attraction in the field demonstrated (Innocenzi et al. 2005). This paper outlines some preliminary work to estimate what time of day females are ‘calling’ (i.e. most probably releasing pheromone) and to determine the efficacy of delta traps for capturing L. rugulipennis males. Further tests measured attration of male capsids to micro- capillary reservoir lures baited with synthetic pheromone compounds in the field.

Materials and methods

In order to obtain individual unmated female L. rugulipennis, laboratory cultures were maintained. 35

Time of day males attracted to females A small scale field trial using caged females in traps was set up in a triangular shaped area (170x150x75m) of weeds (‘Ditton Field’ at EMR) with herbs including fathen, scentless mayweed and sowthistle. Green delta traps (20x20 cm) with brown sticky bases (with extra Ecotac added) were suspended at weed height on two canes taped together at the top. All traps, with the exception of the control traps, contained either a laboratory reared unmated female (matured to adult 16/08/07-28/08/07) or a newly captured field collected female L. rugulipennis. Females were contained in a cage which consisted of a hair roller with gauze around the outside and a lid at either end, holding the gauze in place. The cage also contained a piece of damp cotton wool to maintain the humidity and a section of bean as food. The cage was placed vertically through a hole made in the top of the delta trap, so that the cage did not touch the sticky base. The traps were arranged in 4 rows of 7, 10 m apart in a grid (numbered 1-28) and 10 m in from the edge of the plot. Trap numbers 5, 12, 16 and 24 were control traps (no female inside). Counts of the number, sex and species of mirids caught in the traps was made at approximately 05:00, 08:00, noon, 18:00 and 21:00 hours each day between 30 August – 7 September 2007.

Micro-capillary reservoir lure tested in field A large scale randomised block experiment comparing catches of capsids in green delta traps bated with different lure blends was done in fields of 1) apple, 2) blackcurrant, 3) strawberry, 4) cucumber and 5) weeds with capsid infestation the previous year. At sites 1-4, four different treatments (A-D) comprising micro-capillary lures containing different blends of hexyl butyrate (HB), (E)-2-hexenyl butyrate (E2HB) and (E)-4-oxo-2-hexenal (KA) and an untreated (treatment E) were used (Table 1). At site 5, an additional treatment comprising a caged unmated L. rugulipennis female was included (see above). Dispensers were sample vials fitted with a 5µl glass micro-capillary. The total release rate was approximately 1µg/hr. On 17 July, 10 mg of the blend was dispensed into each lure (approximately 400 days life). A Latin Square design was used at each site. Plots were individual delta traps each containing a single lure (no lure in untreated) and more than 10 m apart. The lures were suspended vertically inside the traps from a wire twist tie. Sticky bases had additional Ecotac applied. The traps were deployed at sites 2, 3 and 5 on 23 July, at site 4 on 1 August, and at site 1 on 31 July. Traps were suspended at mid crop height except at site 4 (cucumber) where each trap was placed above the crop. Weekly counts of the numbers of males and females of each species of capsid caught in each trap were made. Capsid identification was confirmed using a microscope in the laboratory. The trials were ended on 10 August 2007.

Results and discussion

Time of day males attracted to females Culturing of Lygus rugulipennis was successful through the summer months and provided unmated mirids for testing. Unmated L. rugulipennis females attracted more males than newly captured field females (data not shown). This may have been because newly caught field females had been recently mated. There was a high variation between the number, day and consecutive days of males captured in female baited traps, indicating that the variation between ‘calling’ behaviour by females was large. More males were captured in the delta traps in the morning compared to the night-time or afternoon (Figure 1). Closer examination of the data revealed that more males were captured between 0500-0900/1030 compared to 36

0830-1200/1300 (Figure 2). This is in contrast to the North American species L. lineolaris and L. Hesperus, where higher numbers of males were captured from late afternoon to dusk (Blackmer, et al., in press).

Table 1. Treatments used were different blends of hexyl butyrate (HB), (E)-2-hexenyl butyrate (E2HB), (E)-4-oxo-2-hexenal (KA), unmated female L. rugulipennis and an untreated control (E).

Ratios of components in blends Virgin female Treatment HB (mg) E2HB (mg) KA (mg) L. rugulipennis A 300 200 0 0 B 470 0 30 0 C 0 450 50 0 D 290 190 20 0 E 0 0 0 0 F (site 5 only) 0 0 0 1 Notes: All except E and F + 10% BHT as antioxidant. Ratio from single L. rugulipennis female 1.5:1.0:0.08, hexyl butyrate, (E)-2-hexenyl butyrate, and (E)-4-oxo-2-hexenal.

The green delta traps in this study were effective at capturing male Lygus bugs. Green, blue, clear and translucent yellow traps (sticky cylinders) were also found to be effective at capturing Lygus males in trials by Blackmer et al. (in press), compared to red and black traps. These researchers advised using green, blue or clear traps to minimize the capture of beneficial predators.

Micro-capillary reservoir lure tested in field No male L. rugulipennis males were captured at site 1 (apple) or site 3 (strawberry). At site 2 (blackcurrant) only one male L. pabulinus was found in one trap of treatment D. Although L. rugulipennis was abundant at site 5 (weedy vine, demonstrated by sweep netting), only one male L. rugulipennis was trapped in treatment E (no lure) and two male L. rugulipennis in treatment D. The caged females trapped 4 male L. rugulipennis. Innocenzi et al. (2004) identified three components, hexyl butyrate, (E)-2-hexenyl butyrate and (E)-4-oxo-2-hexenal released by single L. rugulipennis females in a ratio of 1.5:1.0:0.08. In contrast, females of the western tarnished plant bug, Lygus hesperus, released the same three components when attacked by ants in a ratio of 1.0:0.44:0.04 (Byers 2006). In addition, in Y-track olfactometer tests, attraction of L. lineolaris males to virgin females was significantly reduced when a dispenser loaded with hexyl butyrate was placed with the virgin females. Hexyl butyrate was found to significantly repel males (Zhang et al., 2007). A group containing similar compounds ((E)-2-hexenyl (E)-2-hexenoate, (E)-2-hexenyl (Z)-3- hexenoate, tetradecyl isobutyrate and octadecyl isobutyrate in a blend), have also been demonstrated to be aggregation pheromones in the male bean bug, Riptortus clavatus (Yasuda et al. 2007).This indicates that the same components may be used in different ratios (and quantities) to attract males, as a repellent, or as an aggregation or defensive mechanism.

70 60 60 50 50 40 40 30 30 20

number of males of number 20 males of number 10 10 0 0 dawn-noon noon-evening evening-dawn 0500-0900/1030 0830-1200/1300

Figure 1. Number of male L. rugulipennis Figure 2. Number of male L. rugulipennis captured in sticky base green delta traps captured in sticky base green delta traps between 8 August – 9 September at between 8 August – 9 September at two different times of the day. time periods during the morning

In this study, field tests with various combinations of the 3 compounds shown to be produced by female L. rugulipennis were tested using micro-capillary dispensers with a reservoir. The ratios and blends tested did not attract significant numbers of male mirids compared to unmated females. Future work will concentrate on using a Piezo-Electric Release System (Girling & Cardé 2007) to optimise the amount and ratio of the three components in order to maximise the attraction of male mirids to traps.

Acknowledgements

We would like to thank Defra for funding this project (HortLINK project HL0184) together with the Industrial Partners, Horticultural Development Council, GlaxoSmithKline Blackcurrant Grower’s Research Fund, GlaxoSmithKline, East Malling Trust, East Malling Ltd., Agrisense, Cucumber Growers Association, KG Growers Ltd. and Donald J Moor. We owe gratitude to the Project Co-ordinator, Mr Tom Maynard (GSK Blackcurrant Growers Association) and other researchers who have been instrumental in delivering the objectives of the project, so far, including, Mr D. Farman (NRI), Ms L. Amarawardana (NRI) and Mr A.L. Harris, (EMR).

References

Aldrich, J.R. 1988: Chemical ecology of the Heteroptera. – Annual Reviews of Entomology 33: 211-238. Anonymous 2004: Pesticides Residues Committee: Pesticides Residues Monitoring Report October-December 2004: 132 pp. Blackmer, J.L., Byers, J.A. & Rodriguez-Saona, C. In press: Evaluation of color traps for monitoring Lygus spp.: Design, placement, height, time of day, and non-target effects. Crop Protection. Blum, M.S. 1981: Chemical defences of arthropods. – Academic Press Inc., New York. Blum, M.S. 1996: Semiochemical parsimony in the arthropoda. – Annual Reviews of Entomology 41: 353-374. 38

Blommers, L., Bus, V., de Jongh, E. & Lentjes, G. 1988: Attraction of males by virgin females of the green capsid bug, Lygocoris pabulinus (Heteroptera: Miridae). – Entomologische Berichten Amsterdam 48: 175-179. Bus, V.G.M., Mols, P.J.M. & Blommers, L.H.M. 1985: Monitoring of green capsid bug Lygocoris pabulinus (Hemiptera: Miridae) in apple orchards. – Mededelingen van de Faculteit Landbouwwentenschappen Rijksuniversiteit Gent 50(2b): 505-510. Byers, J.A. 2006: Production and predator-induced release of volatile chemicals by the plant bug Lygus hesperus. – Journal of Chemical Ecology 32: 2205-2218. Conti, E., Frati, F., Salerno, G., Bin, F., Birkett, M.A., Chamberlain, K., Pickett, J.A. & Woodcock, C.M. 2006: Role of the host plant, Vicia faba, and induced volatiles in host location and sexual communication of Lygus rugulipennis. – 22nd Annual Meeting of the International Society of Chemical Ecology, Barcelona, Spain, July 2006. Abstract S5-O2. Cross, J.V. 2004: European tarnished plant bug on strawberries and other soft fruits. – Horticultural Development Council Fact Sheet 19/04: 6 pp. Cross, J.V. & Hall, D.R. 2003: Pheromones of strawberry blossom weevil and European tarnished plant bug for monitoring and control in strawberry crops. – Final report of Defra project HH1939SSF issued August 2003: 30pp. Downham, M.C.A., Cork, A., Farman, D., Hall, D., Innocenzi, P., Phythian, S., Padi, B., Lowor, S. & Sarfo, J. 2002: Sex pheromone components of the cocoa mirids, Distantiella theobroma and Sahlbergella singularis (Heteroptera: Miridae). – In: Abstract Book, 19th Annual Meeting of International Society of Chemical Ecology, Hamburg, Germany, 3-7 August 2002: 167. Drijfhout, F.P. 2001: Studies towards the sex pheromone of the green capsid bug. – PhD thesis, Wageningen University, The Netherlands: 152 pp. Drijfhout, F.P, van Beek, T.A., Visser, J.H. & de Groot, A. 2000: On-line thermal desorption- gas chromatography of intact insects for pheromone analysis. – Journal of Chemical Ecology 26: 1383-1392. Drijfhout, F.P, Groot, A.T, Posthumus, M.A., van Beek, T.A, & de Groot, A. 2002: Coupled gas chromatographic-electroantennographic responses of Lygocoris pabulinus (L.) to female and male produced volatiles. – Chemoecology 12: 113-118. Drijfhout, F.P, Groot, A.T, van Beek, T.A. & Visser, J.H. 2003: Mate location in the green capsid bug, Lygocoris pabulinus. – Entomologia experimentalis et applicata 106: 73-77. Drijfhout, F.P. & Groot, A.T. 2001: Close-range attraction in Lygocoris pabulinus. – Journal of Chemical Ecology 27: 1133-1149. Garthwaite, D.G. & Thomas, M.R. 2001: Pesticide useage survey report 181: Soft fruit in Great Britain: 42 pp. Garthwaite, D.G., Thomas, M.R. & Dean, S.M. 2000: Pesticide useage survey report 172: Orchards and fruit stores in Great Britain: 39 pp. Girling, R.D. & Cardé, R.T. 2007: Analysis and manipulation of the structure of odor plumes from a piezo-electric release system and measurements of upwind flight of male almond moths, Cadra cautella, to pheromone plumes. – Journal of Chemical Ecology 33:1927- 1945. Glinwood, R., Pettersson, J., Kularatne, S., Ahmed, E. & Kumar, V. 2003: Female European tarnished plant bugs, Lygus rugulipennis (Heteroptera: Miridae), are attracted to odours from conspecific females. – Acta Agric. Scand. Sect. A 53: 29032. Groot, A.T. 2000: Sexual behaviour of the green capsid bug. – PhD thesis, Wageningen University, The Netherlands: 156 pp. 39

Groot, A.T., Timmer, R., Gort, G., Lelyveld, G.P., Drijfhout, F.P., van Beek, T.A. & Visser, J.H. 1999: Sex-related perception of insect and plant volatiles in Lygocoris pabulinus. – Journal of Chemical Ecology 25: 2357-2371. Groot, A.T., Drijfhout, F.P., Heijboer, A., van Beek, T.A. & Visser, J.H. 2001: Disruption of sexual communication in the mirid bug Lygocoris pabulinus by hexyl butyrate. – Agriculture and Forest Entomology 3: 49-55. Gueldner, R.C. & Parrot, W.L. 1978: Volatile constituents of the tarnished plant bug. – Insect Biochemistry 8: 389-391. Ho, H.Y. & Millar, J.G. 2002: Identification, electroantennogram screening, and field bioassays of volatile chemicals from Lygus hesperus Knight (Heteroptera: Miridae). – Zoological Studies 41: 311-320. Innocenzi, P.J., Hall, D.R., Sumathi, C., Cross, J.V. & Jacobson, R.J. 1998: Studies of the sex pheromone of the European tarnished plant bug, Lygus rugulipennis (Het. Miridae). – Brighton Crop Protection Conference – Pests and Diseases 8: 829-832. Innocenzi, P.J., Hall, D.R., Cross, J.V., Masuh, H., Phythian, S.J., Chittamuru, S. & Guarino, S. 2004: Investigation of long-range female sex pheromone of the European tarnished plant bug Lygus rugulipennis: chemical, electrophysiological and field studies. – Journal of Chemical Ecology 30: 1509-1529. Innocenzi, P.I., Hall, D.R., Cross, J.V. & Hesketh, H. 2005: Attraction of male European tarnished plant bug, Lygus rugulipennis, to components of the female sex pheromone in the field. – Journal of Chemical Ecology 31: 1401-1413. Jacobson, R.J. 1999: Capsids (Het. Miridae): A new challenge to IPM in protected salad crops in the UK. – Med. Fac. Landbouww. Univ. Gent 64/3a: 67-72. Jacobson, R.J. 2001: Control of capsid bugs within IPM programmes in protected crops. – Report of contract work undertaken for HDC October 2001: 38 pp. Jacobson, R.J. 2002: Lygus rugulipennis Poppius (Het. Miridae): Options for integrated control in glasshouse-grown cucumbers. – IOBC/WPRS Bulletin 25(1): 111-114. Jacobson, R.J. & Hargreaves, D. 1996: Capsid bugs in protected crops. A guide to the recognition of capsids and the damage they cause. – Horticultural Development Council Fact Sheet 37/96, December 1996, 4 pp. Jay, C.N., Cross, J.V. & Burgess, C. 2004: The relationship between populations of European tarnished plant bug (Lygus rugulipennis) and crop losses due to fruit malformation in everbearer strawberries. – Crop Protection 23: 825-834. Kakazaki, M. & Sugie, H. 2001: Identification of female sex pheromone of the rice leaf bug, Trigonotylus caelestialium. – Journal of Chemical Ecology 27: 2447-2458. Millar, J.G. & Rice, R.E. 1998: Sex pheromone of the plant bug Phytocoris californicus. – Journal of Economic Entomology 23: 1743-1755. Millar, J.G., Rice, R.E. & Wang, Q. 1997: Sex pheromone of the mirid bug Phytocoris relativus. – Journal of Chemical Ecology 23: 1743-1755. Okutani-Akamatsu, K., Watanabe, T. & Azuma, M. 2007: Mating attraction by Stenotus rubrovittatus (Heteroptera: Miridae) females and its relationship to ovarian development. – Journal of Economic Entomology 100(4): 1276-1281. Smith, R.F., Pierce, H.D. & Borden, J.H. 1991: Sex pheromone of the Mullein bug, Campylomma verbasci (Meyer) (Heteroptera: Miridae). – Journal of Chemical Ecology 17: 1437-1447. Umpelby, R.A., Solomon, M.G. & Cross, J.V. 1995: Review of pests of apple. – Report of Ministry of Agriculture, Fisheries and Food, Horticulture and Potatoes Division project HH1722STF issued 10 August 1995: 51 pp. 40

Wardle, A.R., Borden, J.H., Pierce, H.D. Jr. & Gries, R. 2003: Volatile compounds released by disturbed and calm adults of the tarnished plant bug, Lygus lineolaris. Journal of Chemical Ecology 29: 931-944. Yasudu, T., Mitutani, N., Endo, N., Fukuda, T., Matsuyama, T., Ito, K., Moriya, S. & Sasaki, R. 2007: A new component of attractive aggregation pheromone in the bean bug, Riptortus clavatus (Thunberg) (Heteroptera: Alydidae). Applied Entomology and Zoology 42(1): 1–7. Zhang, Q.H. & Aldrich, J.R. 2003: Pheromones of milkweed bugs (Heteroptera: Lygaeidae) attract wayward plant bugs: sex pheromones of two Phytocoris Mirids. Journal of Chemical Ecology 29: 1835-1851. Zhang, O-H., Chauhan, K.R., Zhang, A., Snodgrass, L.G., Dickens, J.C. & Aldrich, J.R. 2007: Antennal and behavioral responses of Lygus lineolaris (Palisot de Beauvois) (Heteroptera: Miridae) to metathoracic scent gland compounds. Journal of Entomological Science 42 (l): 92-104.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 41-46

Identification of black currant leaf midge, Dasineura tetensi (Rübsaamen) female sex pheromone

Lakmali Amarawardana1, David Hall1, Jerry Cross2, Csaba Nagy2 1 Natural Resources Institute, University of Greenwich, Chatham Maritime, Kent, ME4 4TB, UK; 2 East Malling Research, New Road, East Malling , Kent ME19 6BJ UK

Abstract: The blackcurrant leaf midge, Dasineura tetensi (Rübsaamen) (Diptera: Cecidomyiidae), is an important pest of blackcurrant that damages the young leaves and reduces plant growth.The presence of a female-produced sex pheromone was demonstrated previously and work is in progress to identify this to provide growers with new tools for monitoring and control of the pest. Larvae collected from infested plant materials were reared individually in plastic tubes to adulthood. The volatiles from unmated males and females were collected separately by air entrainment. Volatile samples were analysed by gas chromatography (GC) coupled to electroantennographic (EAG) recording from the antenna of a male midge, and by GC coupled to mass spectrometry (MS). In GC-EAG analyses, two consistent EAG responses were observed from male midge preparations. GC-MS analyses and GC retention data of the major component indicated that this is a 17-carbon diacetate with one double bond. Hydrogenation gave a product with mass spectrum consistent with that of 2,12-diacetoxyheptadecane, but no product could be detected after reaction with dimethyl disulphide which would indicate the position of the double bond. No mass spectrum could be obtained for the minor component because of the small amount present. Work is in progress to complete identification of both components.

Key words: Dasineura tetensi, blackcurrant leaf midge, sex pheromone, electroantennography

Introduction

The black currant leaf midge Dasineura tetensi (Rübsaamen) (Diptera: Cecidomyiidae) is a serious pest of blackcurrant. Females lay eggs on the new terminal growth and larvae feed on leaves causing them to distort and discolour which leads to reduction in growth of the canes and eventually reduces fruit production. Its importance in the UK has declined in recent years due to widespread use of Meothrin against blackcurrant gall mite. However, the midge is likely to become more important with introduction of varieties which are resistant to gall mite and reversion virus and which will be treated with less costly acaricides which are ineffective against the midge. D. tetensi can be controlled by spraying broad-spectrum insecticides, but pest monitoring is important in order to prevent calender date-based application of insecticides. Female D. tetensi were shown to produce a sex pheromone which attracts males by Garthwaite et al. (1986), and here we report progress in identification of the female sex pheromone of D. tetensi which could provide new tools for both monitoring and control of this pest.

41 42

Materials and methods

Insect collection Shoots infested with D. tetensi were collected from a field in Horsmonden, UK, in May 2007. Shoots were stored in plastic boxes (19 cm x 10.6 cm x 7.5 cm) at room temperature with wet paper towel to keep the plant material in fresh condition. The fully-grown larvae crawled out from the curled edges of damaged leaves and transferred individually into small clear plastic tubes (1.5 cm i.d. x 2.3 cm) containing a piece of wet filter paper. Larvae were reared to adulthood at 23-18°C and 16L: 8D light cycle.

Pheromone collection Pheromone was collected by means of air entrainment. Female and male insects were sexed on the basis of antennal morphology and introduced into separate glass chambers (5.3 cm i.d. x 13 cm) of which one end was connected to a charcoal filter (20 cm x 2 cm; 10-18 mesh, Fisher Chemicals, UK) ) and the other to a Pasteur pipette (4 mm i.d.) containing the adsorbent Porapak Q (200 mg; 80-100 µm, Waters Associates Inc., USA). Air was drawn through the chambers with a diaphragm pump (300 ml/ min; M 361C, Charles Austen Pump Ltd, UK). Entrainment was carried out for three to seven days continuously and insects were introduced at 24 h intervals. Filters were removed and eluted with dichloromethane (1.5 ml; Pesticide Grade) and the samples were concentrated to approximately one fifth of the original volume and refrigerated until analysis.

Gas chromatography coupled to electroantennography (GC-EAG) GC-EAG analyses were carried out as described by Cork et al. (1990). A HP6890 (Aglilent Technologies) gas chromatograph was used equipped with a flame ionisation detector (FID) and fused silica capillary columns (30 m x 0.32 mm x 0.25 µm film thickness) coated with polar (Supelcowax-10, Supelco) and non polar (SPB-1, Supelco) phases. The temperature of the oven was programmed at 50°C for 2 min, then at 10°C/min to 250°C and held for 5 min. Injection was splitless at 220°C and helium was used as carrier gas (2.4 ml/min). The column effluent was split to the FID and a T-piece, the contents of which were carried by a stream of nitrogen (200 ml/min) as a puff over the antennal preparation for 3 s every 17 s (Cork et al. 1990). The antennal responses were recorded using a portable EAG recording unit (INR-2, Syntech, The Netherlands). The male midges were anaesthetised with carbon dioxide and the legs and the wings were removed with a scalpel blade before mounting. The glass electrodes were pulled and filled with electrolyte (0.1 M solution of KCl with 1% polyvinylpyrrolidine). The tips of the electrodes were broken off so that both antennae could be inserted into the recording electrode and the abdomen inserted into the reference electrode. Signals were amplified and analysed with EZChrom software (Elite v3.0).

Gas chromatography linked to mass spectrometry (GC-MS) GC-MS analysis was carried out with Varian CP 3800 gas chromatograph linked to Varian Saturn 2200 ion trap mass spectrometer. Samples were analysed on fused silica capillary columns (30 m x 0.25 mm i.d.) coated with polar (Supelcowax-10, Supelco) and non-polar (VF5, Varian) phases. Oven temperature was programmed at 50°C for 2 min, then at 6°C/min to 250°C and held for 5 min. Helium was used as the carrier gas (1.0 ml/min) and the injection was splitless. GC retention times were converted to Retention Indices (RI) relative to the retention times of straight-chain acetates.

Microchemical reactions Hydrogenation of an aliquot (20 µl) of collections of volatiles from females was carried out by removing the dichloromethane, dissolving the residue in hexane with a small amount of 10% palladium on charcoal as catalyst and bubbling in hydrogen through a length of fused silica capillary for 1 min. The reaction mixture was analysed by GC-MS. For derivatisation with dimethyl disulphide (DMDS) an aliquot (100 µl) of the combined volatile collections was mixed with of DMDS (20 µl) and iodine in ether (6 mg/ml; 10 µl) and heated for 4 hr at 63˚C (Buser et al. 1983). The reaction was quenched with equal volumes (100 µl) of an aqueous solution of sodium thiosulphate (5%) and hexane. The reaction was analysed by GC-MS on the non-polar column.

Results and discussion

0.020 2500 FID EAG Retention Time Name 0.018 2000

0.016 1500

1000 0.014

500 0.012

FIDVolts 0

0.010 * EAG millivolts

-500 0.008 * -1000 0.006

-1500 0.004

-2000 17.0 17.2 17.4 17.6 17.8 18.0 18.2 18.4 18.6 18.8 19.0 19.2 19.4 19.6 19.8 20.0 Minutes

0.020 3000 FID EAG Retention Time Name

0.018 2000

0.016 1000

0.014 * 0

0.012 FIDVolts -1000 EAG millivolts 0.010 * -2000 0.008

-3000 0.006

0.004 -4000 18.00 18.25 18.50 18.75 19.00 19.25 19.50 19.75 20.00 20.25 20.50 20.75 21.00 21.25 21.50 21.75 22.00 Minutes Figure 1. GC-EAG analysis of volatile collections from female D. tetensi with male EAG preparation (upper: polar column; lower: non-polar column; EAG responses to pheromone components are marked with ). 44

Approximately 12,300 larvae were collected and reared in individual plastic tubes under laboratory conditions, but only 36% reached adulthood. In GC-EAG analyses male midges consistently responded to two components in the volatile collections from D. tetensi females on both polar and non polar columns (Figure 1). The major response of higher intensity appeared at RI 2042 and 1835 on polar and non polar columns respectively and the minor response at RI 1999 and 1720 respectively. In GC-MS analyses a small peak was observed at the RI’s of the major component on both polar and non-polar columns. The mass spectrum (Figure 2) showed ions at m/z 43 and 61 characteristic of acetates. The ion at m/z 234 was consistent with loss of two acetate groups from a mono-unsaturated, 17-carbon diacetate of molecular weight 354. The presence of the tiny fragment at m/z 87 suggested that one acetate group was located on the second carbon atom, as in other midge pheromones (e.g. Gries et al. 2005; Hillbur et al. 1999). Furthermore, the GC retention indices of the major pheromone component were very similar to those of (Z)-2,13-diacetoxy-8-heptadecene, a component of the female sex pheromone of the pear leaf midge, D. pyri, identified previously (Amarawardana et al. 2006).

Figure 2. Mass spectrum of major component of female sex pheromone of D. tetensi.

After hydrogenation of the collection of volatiles from females of D. tetensi, GC-MS analysis showed a peak considered to be the hydrogenated major EAG active component at RI’s of 2042 and 1871 on polar and non-polar columns respectively. This shift is consistent with saturation of a single double bond. In the mass spectrum (Figure 3), the values of certain fragments such as m/z 234, 150, 251 and 83 in the original spectrum had increased by m/z 2 units, further confirming the presence of a single double bond. This spectrum closely resembled that of 2,12-diacetoxy-heptadecane, one of the components of the female sex pheromone of the red cedar cone midge, Mayetiola thujae, reported by Gries et al. (2005). In particular the significant ion at m/z 165 corresponded to loss of the terminal C5H11 fragment after loss of the two acetate groups (236-71). In order to determine the position of the double bond, reaction with DMDS was attempted. This gives an adduct with two methylthio groups on the carbons of the double bond which strongly directs fragmentation between these carbons in the mass spectrum (Buser et al. 1983). Reaction of DMDS with nanogram quantities of (Z)-2,13-diacetoxy-8- heptadecene gave a product with strong ions at m/z 171 and 157 in the mass spectrum. However, after similar reaction of a sample of the collection of volatiles from female 45

D. tetensi, no product with these or ions corresponding to other positions of the double bond could be observed around the same retention time in GC-MS analyses.

Figure 3. Mass spectrum of major component of D. tetensi pheromone after hydrogenation.

Thus the major component of the pheromone is thought to be a monounsaturated 2,12- diacetoxyheptadecane, but further work is required to determine the position and configuration of the double bond. Furthermore the diacetate can exist as four diastereomers and it is likely that only one of these is produced naturally by the female and is attractive to male D. tetensi. Because of the small quantity present, it has so far proved impossible to obtain a mass spectrum of the compound causing the minor EAG response from male D. tetensi. It is likely that this has a chemical structure related to that of the major component, but further work is required to identify this. Once identification of the major pheromone component has been completed, this will be made available to growers for testing in field trapping tests.

Acknowledgements

We thank UK Horticultural Development Council and the Worshipful Company of Fruiterers for funding this project.

References

Amarawardana, L., Hall, D. & Cross, J. 2006: Investigations on the sex pheromone of pear leaf midge, Dasineura pyri (Bouché), and other gall midge pests of fruit crops. – Conference proceedings of Workshop on Arthropod Pest Problems in Pome Fruit Production. Lleida (Spain), September 2006: 45. Barnes, H.F. 1948: Gall Midges of Economic Importance. Vol.III Gall midges of Fruit. – Crosby Lockwood, London. Buser, H.-R., Arn, H., Guerin, P. & Rauscher, S. 1983: Determination of double bond position in mono-unsaturated acetates by mass spectrometry of dimethyl disulfide adducts. – Analytical Chemistry 55: 818-822. 46

Cork, A., Beevor, P.S., Gough, J.E. & Hall, D.R. 1990: Gas chromatography liked to electroantennography: a versatile technique for identifying insect semiochemicals. – In: Chromatography and isolation of insect Hormones and Pheromones. A.R. McCaffery and I.D. Wilson (eds.), Plenum Press, London: 271-279. Garthwaite, D.G., Wall, C. & Wardlow, L.R. 1986: Further evidence for a female sex pheromone in the blackcurrant leaf midge Dasineura tetensi. –Proceedings of the British Crop Protection Conference – Pests and Diseases 1: 355-357. Gries, R., Khaskin, G., Bennett, R.G., Miroshnychenko, A., Burden, K. & Gries, G. 2005: (S,S)-2,12-, (S,S)-2,13- and (S,S)-2,14-diacetoxyheptadecanes: sex pheromone components of red cedar cone midge, Mayetiola thujae. – Journal of Chemical Ecology. 31: 2933-2946. Hillbur, Y., Anderson, P., Arn, H., Bengtsson, M. & Löfqvist, J. 1999: Identification of sex pheromone components of the pea midge, Contarinia pisi (Diptera: Cecidomyiidae). Naturwissenschaften 86: 292-294.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 47-49

Biological control of the currant clearwing moth Synanthedon tipuliformis by mating disruption

Catherine A. Baroffio, Christoph Carlen Agroscope Changins – Wädenswil ACW, Centre de recherche de Conthey, 1964 Conthey, Switzerland

Abstract: The currant clearwing moth Synanthedon tipuliformis Clerck is a major pest in Ribes ssp. crops, mostly red- and blackcurrants. Two mating disruption trials were conducted between 2002 and 2005 on redcurrant plots in Geneva and Thurgau (Switzerland). In both trials, the number of adults decreased significantly. The density of 600 dispensers per ha showed good efficacy against the clearwing moth.

Key words: currant clearwing moth, mating disruption, pheromone, Ribes, Synanthedon tipuliformis

Introduction

The currant clearwing moth Synanthedon tipuliformis Clerck is an important pest of various currant crops. The larvae feed and develop on the food reserves contained within the pith of Ribes canes. This weakens the stems, reduces yield and causes penetrating lesions for fungal diseases. There is only one generation per year and the emergence peak coincides with the harvest period (May-June) (Baggiolini & Dupperex 1963). In some Swiss regions (Thurgau and Geneva) sometimes more than 90% of redcurrant canes show symptoms. Chemical treatments are prohibited because of the risk of residues in the harvested fruits. Mating disruption (MD) has been successfully used in Italy (Grassi et al. 2002) and in New Zealand (Suckling et al. 2005). This paper summarizes efficacy trials carried out with this technique in Thurgau and Geneva between 2001 and 2005.

Material and methods

In Thurgau, MD was studied between 2001 and 2005. The MD plot and the control field consisted of 1 ha of covered redcurrants and 0.3 ha of uncovered blackcurrants respectively. The two plots were separated by more than 300 meters. In Geneva, the study was conducted between 2003 and 2005 in uncovered redcurrants plots. The MD plot was 1.2 ha and the control plot 0.2 ha. The distance between the plots was greater than 200 meters. The dispensers used are commercialized by the Japanese company Shin-Etsu and sold in Switzerland under the commercial name Isonet Z. They contain 2 pheromones: 52 mg of E2, Z13-18Ac and 2 mg of E3, Z13-18Ac. In mid-May, 600 dispensers/ha were deployed in the upper third part of the bushes. More details are given in Carlen et al. (2006). Effectiveness was assessed in December by sampling 25 canes per plot and counting the number of larvae/meter of wood.

Results and discussion

In Thurgau, S. tipuliformis populations decreased by 97% in the MD plot between 2001 and 2005 while they doubled in the control (Figure 1). In Geneva, the reduction was less

47 48

spectacular with a decrease of 58% in the MD plot while the control populations remained stable. These latter results were similar to those obtained by Grassi et al. (2002) with the same dispenser density but on much smaller uncovered surfaces. The better results obtained in Thurgau seem to indicate that covering the crop increases the efficacy of MD. None-the-less, in both regions, the population levels in the MD plots were reduced to around the threshold values of 0.1 to 0.15 larvae per cane as defined by Gottwald & Künzel (1994). MD had very good efficacy at 600 dispensers/ha. For bigger plots (> 3 ha) a reduction to 300 dispensers/ha is conceivable.

4.5 MD Thurgau 4 Control (Thurgau) MD (Geneva) 3.5 Control (Geneva)

2.5

/ linear m of main shoots 2

1.5

1 S. tipuliformis

0.5 NC 0 2001 2002 2003 2004 2005

Figure 1. Evolution of S. tipuliformis larvae per linear meter of main shoots in Thurgau and Geneva. NC: not checked.

Acknowledgements

We thank Pierre-Joseph Charmillot for his advices and Samuel Stüssi and Daniel Zingg from Andermatt Biocontrol for the providing of the dispensers.

References

Baggiolini, M. & Duperrex, H. 1963: Observations sur la biologie et la nuisibilité de la sésie du groseillier et du cassis Synanthedon tipuliformis Clerck. – La recherche agronomique en Suisse 2: 13-32. Carlen, C., Baroffio, C., Mittaz, C. & Auderset, C. 2006: Succès de la lutte par confusion sexuelle contre la sésie du groseillier. – Revue suisse Vitic. Arboric. Hortic. 38 (3): 183- 187. Gottwald, R. & Künzel, K. 1994: Neue Erkenntnisse zur Populationsökologie des Johannis- beerglasflüglers (Synanthedon tipuliformis Clerck). Gesunde Pflanzen 46 (4): 131-136. 49

Grassi, A., Zini, M. & Forno, F. 2002: Mating disruption field trials to control the currant clearwing moth, Synanthedon tipuliformis Clerck: a three-year study. – Bull. IOBC wprs Bulletin 25 (9): 69-76. Suckling, D.M., Gibb, A.R., Burnip, G.M., Snelling, C., De Ruiter, J., Langford, G. & El- Sayed, A.M. 2005: Optimization of pheromone lure and trap characteristics for currant clearwing, Synanthedon tipuliformis. – J. Chem. Ecol. 31 (2): 393-406. 50 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 51-64

Gábor Vétek1, Csaba Thuróczy2, Béla Pénzes1 1 Corvinus University of Budapest, Faculty of Horticultural Science, Department of Entomology, Villányi út 29–43, H-1118 Budapest, Hungary; 2 Malomárok u. 27, H-9730 Kőszeg, Hungary

Abstract: A wide range of beneficial insects controlling raspberry cane pest populations has been identified since the first surveys on the arthropod fauna of the plant were carried out. As far as the parasitoids of the raspberry cane midge (Resseliella theobaldi) and the rose stem girdler (Agrilus cuprescens) are concerned, a large number of references on different parasitoid species can be found in the literature dealing with the biology and control of the pests. In some cases, the specimens of the reared or collected parasitoids were identified by former or present-day specialists, but the citation of species names from old publications without checking them also seems to be quite widespread. As it has turned out that in many cases invalid, ambiguous or incorrect names are mentioned in publications, the aim of the authors of this paper was to give a review of the cited parasitoids of the two, above-mentioned raspberry cane pests together with making their taxonomic status clear so as to help researchers and plant protection engineers use the correct, presently accepted names of these beneficial insects. This work has been carried out within the scope of an international cooperation in which curators and specialists of national insect collections and researchers of plant protection institutes from Europe helped the investigations. On the basis of this study, the most frequently cited natural enemy of raspberry cane midge is Tetrastichus inunctus, while those of rose stem girdler are Tetrastichus heeringi and T. agrilorum. However, the type of T. inunctus is lost so this species name is now ambiguous, and the specimens earlier reared from the larvae of rose stem girdler and identified as T. agrilorum need to be re-examined. According to the authors’ rearing experiments, Aprostocetus epicharmus proved to be the primary parasitoid of raspberry cane midge, while it was Tetrastichus heeringi which parasitized the larvae of rose stem girdler significantly. Further studies on this topic are in progress.

Key words: Resseliella theobaldi, Agrilus cuprescens, parasitoids

Introduction

Red raspberry (Rubus idaeus) is one of the most important soft fruits grown in many parts of Europe. On the other hand it can be regarded as a minor crop compared with pome or stone fruits. Due to this fact, apart from some basic studies, relatively little research work has been carried out concerning integrated plant protection methods in raspberry, although some quite promising results have been achieved in the past few years. One of the most important problems in putting IPM in raspberry into practice is that growers tend to apply broad- spectrum insecticides against pests without exact timing. The reason for this is that the special fauna, the exact biology of insects and the interactions between the host plant and pests and pests and their natural enemies are still not understood adequately in case of this plant species. Due to the lack of this information, scheduled sprayings are widely practiced in order to avoid

51 52

all possible crop damage. However, there is an increasing demand by growers to reduce the application of chemicals so as to moderate the costs of raspberry production, and also by consumers because of their rightful demand for pesticide-free fruit products. It is the researchers’ task to develop novel methods in raspberry protection to make pest control more effective. The basis of carrying out fundamental research work in this area is the extensive knowledge of the literature already published on the subject. However, taking over all data from earlier studies without any review may give rise to some inaccurate or even incorrect statements and conclusions, the multiplication of which can’t be avoided by citations in later scientific studies. As far as the species composition of beneficial arthropods controlling raspberry cane pest populations are concerned, the number of publications dealing with species names and the interrelationship between hosts and their natural enemies is quite limited in spite of the fact that IPM is unimaginable without making the status of parasitoids clear. As a first step in this topic, the aim of this work was to help researchers and plant protection engineers become acquainted with those parasitoid species which may indeed play a role in the control of two widespread raspberry cane pests, the raspberry cane midge (Resseliella theobaldi) and the rose stem girdler (Agrilus cuprescens). The authors would like to emphasise the importance of using the correct, presently accepted names of these beneficial arthropods in order to avoid inaccurate or false citations of species in future publications. The clarification of the status of raspberry cane pest parasitoids can also contribute to the avoidance of misunderstandings concerning their biology.

Revision of literature on the parasitoids of two dangerous raspberry cane pests I. Raspberry cane midge – Resseliella theobaldi (Barnes, 1927) It was in the UK, in 1925 that a hymenopterous parasitoid species was first reared from the larvae of the raspberry cane midge by Barnes (Barnes 1926, 1927a), the author who gave the description of the pest (Barnes 1927b). The parasitoid was found to cause 100% mortality in midge larvae overwintering on raspberry canes. Besides this observation, it was supposed that the parasitized larvae remain on the canes while the healthy ones descend to the soil, and this fact was regarded as an explanation of next year’s midge attacks in spite of the established rate of parasitism (Barnes 1927a). The way of overwintering was later confirmed by Fox Wilson and Green (1944) and further studies of Barnes (1944). The first mention of identified raspberry cane midge parasitoids was given by Bachmann (1950) from Switzerland. The species are namely Tetrastichus flavovarius Nees, Tetrastichus roesellae (De G.) Nees and only one specimen of a Leptacis sp. According to the author, the rate of parasitism can be considerable, but it is of minor importance because these natural enemies of the midge are unable to prevent raspberry canes from being damaged by larvae of the pest. In the studies of Pitcher (1952, 1955), three hymenopterous species are described on the basis of several years’ observations carried out in the UK between 1946 and 1950. Tetrastichus inunctus (Nees), Piestopleura catillus (Wlk.) and a Leptacis sp. were reared by the author. The latter two were found to emerge later than their host (June–July) and attack principally the second generation of midge larvae. A second generation of them was also established to appear in late August and parasitize the third midge generation. According to the author’s investigations, the eggs of these parasitoids were laid in midge larvae in their final instar while they were still on the cane. The parasitized larvae fell to the ground in the normal way but, after forming a cocoon, their skin became swollen and hard as the parasitoid larva grew and pupated inside it. The rate of parasitism usually remained small. T. inunctus is considered to be of major importance in reducing the second and third raspberry cane midge 53

generations. Adults of this species were found to emerge earlier than those of the other two parasitoids and attack mainly the first larval generation of the midge. The females of T. inunctus were observed to locate the host larvae by using their antennae and drive the ovipositor through the plant tissue to reach them. The midge larvae were soon immobilized and metamorphosis of the parasitoid took place within their bloated and blackened skins. Emergence began again in August, leading to parasitism of the late second and the third midge generations. The resulting larvae overwintered on the canes within the host skins. It is also indicated that the parasitoid species noted by Barnes (1944) was identified as T. inunctus (Pitcher 1952). T. inunctus, and another species not mentioned earlier, Tetrastichus brevicornis (= flavovarius auct. nec. Nees) Nees, were reported from Germany in 1955–1956. The parasitoid attacked the larvae of the midge from the beginning of August, and the degree of parasitism reached 45% (Fritzsche 1958). Parasitized raspberry cane midge larvae were also found in Slovenian raspberry plantations (Masten 1958). The parasitoid wasp was identified as T. inunctus. It is interesting that, according to the author, attacked midge larvae do not die soon after they have been parasitized but they continue feeding even for a longer time than healthy midge larvae, and pupate under the bark of the cane. Hence it is the pupae of the pest which will be killed by the parasitoid, and adults of T. inunctus will emerge from dead midge pupae. These observations are partly in contrast with those of Pitcher’s (1952). The rate of parasitism can be high in the second and third generations of the midge, but it is of low significance because by that time of the year raspberry canes have lignified to such an extent that they show good resistance to midge damage (Masten 1958). Investigations on the biology of raspberry cane midge parasitoids were carried out by Stoyanov (1963) in Bulgaria, between 1957 and 1958. The hymenopterous parasitoids found were supposed to belong to the same species as those described by Pitcher (1952). Stoyanov (1963) established a similar relative importance of the three parasitoids, T. inunctus, P. catillus and the Leptacis sp., as noted by Pitcher (1952). According to Stoyanov (1963), the peak of emergence of the first generation of the most significant species, T. inunctus, is in late May (in the region of Sofia), and adults of the wasp parasitize the well-developed larvae of the first midge generation. The following emergences of the wasp coincide with the presence of late instar midge larvae on canes. Compared with the first larval generation of the midge, the rate of parasitism is said to be higher during the second and third midge generations. In the catalogue of Domenichini (1966) on Palaearctic Tetrastichinae, the raspberry cane midge is listed as a host of Tetrastichus vincius (Walker). Ambrus (1973) gives some information on the biology of T. inunctus, P. catillus and the Leptacis sp. In a concise work on horticultural pests edited by Balás & Sáringer (1982), Tetrastichus tymber Walk. is mentioned as the parasitoid of the raspberry cane midge. A short mention of the natural enemies of raspberry cane midge was made by Gordon & Williamson (1984). The authors write that several parasitoids of the pest were identified in Great Britain, but only two are common, namely Synopeas craterus (Wlk.) and a Tetrastichus sp. The first was found in Scotland infesting second-generation larvae of the midge. It is indicated that the parasitized larvae complete their development on the cane, fall to the ground and spin cocoons as usual. However, the parasitoid continues to grow within the cocoon and an adult wasp emerges in the following July. Tetrastichus spp. were recorded from southeastern England. Larvae parasitized by one of these species remain on the cane while the parasitoid completes its develpoment within them. These natural enemies can not adequately control midge populations in Great Britain (Gordon & Williamson 1984). 54

In the first volume of the Manual of Palaearctic Diptera, which, including other information, gives a general overview of flies of economic importance in the Palaearctic region, Darvas & al. (2000) mention Tetrastichus lycidas Walker and Tetrastichus vincius Walker as known parasitoids of the pest, but information is not given concerning either their biology or significance. Shternshis & al. (2002) tested some plant protection products containing biological control agents against midge blight in the Siberian region of Russia. Evaluating the results of the sparying experiments, they did not find any negative influence of the BCAs on a parasitoid species, identified as Aphanogmus sp., the larvae of which developed in raspberry cane midge larvae. In northern Italy, Aprostocetus epicharmus Walker was identified as a parasitoid of the pest, which infested larvae of the midge to a considerable extent from the beginning of June in the case of moderate application of pesticides (Grassi 2003). A. epicharmus was also found to parasitize the larvae of the pest in Hungarian raspberry plantations. The population dynamics of the two species, the host-parasitoid interrelationship and the possible effect of raspberry cultivars on the biology of these species were studied in detail between 2002–2005 (Vétek & al. 2006).

II. Rose stem girdler – Agrilus cuprescens cuprescens (Ménétriés, 1832) The taxonomy of A. cuprescens has been confused due to its extensive Euro-siberian distribution and obvious morphological and chromatic variability. Numerous names were proposed for the species, subspecies or infrasubspecies rank, respectively. Thus, on the basis of Jendek’s revision of A. cuprescens and related species (Jendek 2003), the authors of this present paper tried to find references to parasitoids of the rose stem girdler. The first and most important results of this research are described below but the work is still in progress as, according to Jendek (2003), A. cuprescens cuprescens has 16 available synonyms. The first reference to a parasitoid species, namely Tetrastichus rugglesi Roh., infesting larvae of the rose stem girdler to a considerable extent (called the bronze cane-borer in the article concerned) was found in an American journal (Mundinger 1941). Another hymenopterous species, Cubocephalus annulatus Cresson, is mentioned by Townes & Gupta (1962) as reared from raspberry canes together with rose stem girdler at the end of June, 1938, by Mundinger. It was in 1964 that Tetrastichus agrili Clwt. was found to cause 57.6% mortality of larvae of the rose stem girdler in a Hungarian raspberry plantation (Reichart 1968). According to the author, several parasitoid larvae attack a single larva of the pest, the parasitoid overwinters in the larval stage, and its adults emerge later than those of the rose stem girdler do. In Bulgaria, the rose stem girdler is a major pest of oil-bearing rose, so its biology was studied by several authors. As part of these investigations, some natural enemies of the pest were identified. Nikolova (1969) mentions two species: Tetrastichus heeringi Del., which parasitized larvae of the pest significantly (43–92%), and a Tetrastichus sp., which parasitized the eggs of the rose stem girdler. In a later work of the author, one more species is listed besides the above-mentioned two parasitoids, namely a Xylophrurus sp. (Nikolova 1972). In the former Yugoslavia, it was Dobrivojević (1970) who studied the entomophagous Hymenoptera and Diptera species of raspberry pests. Unfortunately, the list of parasitoid species is a bit confusing because of several misspellings and contradictions in the paper. It seems that at least five beneficial hymenopterous insects can be named which, according to the author, may take part in the control of rose stem girdler populations: T. heeringi, Tetrastichus agrilorum Ratz., Eupelmella vesicularis Ratz., an Eurytoma sp. belonging to the 55

rosae group and Heterospilus rubicola n. sp. However, in some parts of the article further species are also mentioned such as Platygaster cottei Kieff. and Eurytoma rosae group, E. curculionum Mayr. Based on more than two decades of studies on the natural enemies of the rose stem girdler, Tsalbukov (1983) mentions that T. heeringi is the most important parasitoid of the pest. It controlled the populations of the rose stem girdler very effectively in 1978–79. In a concise work on Coleopteran pests occurring in Hungary, T. agrilorum is named as a natural enemy of the rose stem girdler (Reichart 1990). According to Graham (1991), the rose stem girdler is a host of T. heeringi, the latter being a gregarious, endophagous parasite of the host larvae. In Austria, Idinger (1991) found that T. heeringi parasitized the rose stem girdler considerably. Molnár & al. (2001) mention T. agrilorum as an important parasitoid infesting rose stem girdler larvae in Hungarian raspberry plantations. All the above-mentioned data found in literature dealing with the parasitoids of the raspberry cane midge and the rose stem girdler are summarized below (Table 1 and 2). The species and auctor names are shown in the same way as indicated in the original papers. The aim of showing this table is to give an overview of the beneficial hymenopterous species known from earlier studies, and of researchers who referred to the parasitoids.

Table 1. List of parasitoids of Resseliella theobaldi (Barnes) on the basis of data found in literature in chronological order.

Species name References Tetrastichus flavovarius Nees, Tetrastichus roesellae (De G.) Nees, Bachmann (1950) Leptacis sp. Tetrastichus inunctus (Nees), Piestopleura catillus (Wlk.), Pitcher (1952, 1955) Leptacis sp. Tetrastichus inunctus Nees, Tetrastichus brevicornis (= flavovarius auct. nec. Nees) Fritzsche (1958) Nees Tetrastichus inunctus Nees Masten (1958) Tetrastichus inunctus Nees, Piestopleura catillus Walk., Stoyanov (1963) Leptacis sp. Tetrastichus vincius (Walker) Domenichini (1966) Tetrastichus inunctus (Nees.), Piestopleura catillus (Wlk.), Ambrus (1973) Leptacis sp. Tetrastichus tymber Walk. Balás & Sáringer (1982) Synopeas craterus (Wlk.), Gordon & Williamson (1984) Tetrastichus sp. Tetrastichus lycidas Walker, Darvas et al. (2000) Tetrastichus vincius Walker Aphanogmus sp. Shternshis et al. (2002) Aprostocetus epicharmus Walker Grassi (2003) Aprostocetus epicharmus Walker Vétek et al. (2006)

Table 2. List of parasitoids of Agrilus cuprescens cuprescens (Ménétriés) on the basis of data found in literature in chronological order.

Species name References Tetrastichus rugglesi Roh. Mundinger (1941) Cubocephalus annulatus Cresson Townes & Gupta (1962) Tetrastichus agrili Clwt. Reichart (1968) Tetrastichus heeringi Del., Nikolova (1969) Tetrastichus sp. Tetrastichus heeringi Delucchi, Tetrastichus agrilorum Ratz., Eupelmella vesicularis Ratz., Eurytoma rosae group, Dobrivojević (1970) Heterospilus rubicola n. sp., Platygaster cottei Kieff., Eurytoma rosae group, E. curculionum Mayr Tetrastichus heeringi Del., Tetrastichus sp., Nikolova (1972) Xylophrurus sp. Tetrastichus heeringi Tsalbukov (1983) Tetrastichus agrilorum Ratzeburg Reichart (1990) Tetrastichus heeringi Delucchi Graham (1991) Tetrastichus heeringi Delucchi Idinger (1991) Tetrastichus agrilorum Molnár et al. (2001) Tetrastichus heeringi Delucchi Vétek et al. (present paper)

Materials and methods

This work has been carried out within the scope of an international cooperation in which curators and specialists of national insect collections and researchers of plant protection institutes from Europe helped the investigations. As the first step, the relevant publications were checked if the name of the specialist who identified the parasitoid species is given. Based on this information, we tried to get in touch with present-day curators of national insect collections, where the specimens of the earlier identified parasitoid species were supposed to have been placed by the specialist. At the same time, the former workplace of the author who reared parasitoid specimens was also attempted to be found in order to reveal whether some identified specimens had been sent back to that place by the specialist. When the specimens considered were found, we asked for a loan of them so as to make their taxonomic status clear on the basis of the most modern and currently accepted works of literature on hymenopterous parasitoids. Otherwise, if the name of the specialist was not mentioned in the article, we tried to look for the author for further information. In this case, the aim was to find out whether the indicated species had been identified or the name of the species is only a citation from another paper not shown in the references. Nevertheless, since 2002, the authors of this paper have also been carrying out rearing experiments together with biological investigations of parasitoids of both pest species to make their status clear.

Results and discussion

In this part of the paper, the present status of the earlier indicated parasitoids of the two raspberry cane pests is discussed on the basis of our research completed with the results of the authors’ rearing experiments. The parasitoid species are grouped according to families. The referred name is indicated first, followed by the valid one if the original name has been changed.

I. Parasitoids related to the raspberry cane midge

Chalcidoidea: Eulophidae

Tetrastichus brevicornis (Panzer) – Sigmophora brevicornis (Panzer, 1804) The species is indicated by Fritzsche (1958). It was determined by Ch. Ferrière. The major part of the former specialist’s collection can be found in the Natural History Museum of the City of Geneva, Switzerland, where Ferrière was once the curator of the Hymenoptera collection. Although some specimens of T. brevicornis are placed there, none of them was reared from the raspberry cane midge. S. brevicornis is a very common parasitoid of the Tetrastichinae subfamily infesting several gall midge species, but as the raspberry cane midge is not mentioned as its host either by Domenichini (1966) or Graham (1987), the information given by Fritzsche (1958) needs confirmation by rearing experiments. No specimens of this parasitoid have been reared by us yet.

Aprostocetus epicharmus (Walker, 1839) Grassi (2003) and Vétek & al. (2006) found this parasitoid species of the raspberry cane midge. It was G. Viggiani who identified the parasitoids for Grassi, and Cs. Thuróczy who identified the reared specimens for Vétek & al. The results of these rearing experiments confirm the status of A. epicharmus as an important natural enemy of Resseliella theobaldi.

Tetrastichus flavovarius (Nees, 1834) – species inquirendae This species, mentioned by Bachmann (1950), was determined by Ch. Ferrière. The workplace of Bachmann was the Eidgenössische Versuchsanstalt für Obst-, Wein- und Gartenbau, Wädenswil. Specimens of the mentioned species and any other one referred to by Bachmann (1950) were not found there at all. Moreover, the species name, T. flavovarius is now ambiguous as the type is considered to be lost. Graham (1985) indicates T. flavovarius as a possible synonym of Aprostocetus forsteri (Walker). As the host of this species lives on the flower-heads of centaurea (Centaurea spp.), the raspberry cane midge is not probable to be a host of this wasp. Nevertheless, it should be noted that A. epicharmus adults were also collected from the dry anthodium of rough star-thistle (Centaurea aspera) by Graham (1987).

Tetrastichus inunctus (Nees, 1834) – species inquirendae Five references to this species were found in literature in relation to the raspberry cane midge (Pitcher 1952, 1955, Fritzsche 1958, Masten 1958, Ambrus 1973). A sixth one, that of Stoyanov’s (1963), should also be mentioned. In spite of the fact that this latter author did not ask any specialists to identify the wasps he reared, based on the biological description of one of the identified species given by Barnes (1944) (indeed Pitcher (1952) – false citation), Stoyanov (1963) supposes the major part of his reared specimens to be T. inunctus, and describes the host-parasitoid interrelationship in details. In each publication Ferrière, Nixon or both of them are named as specialists who helped the work. The paper of Ambrus (1973) is an 58

exception, and the author is supposed to refer to Pitcher’s (1952) data. T. inunctus specimens could not be found in the specialists’ collections. We tried to look for some colleagues of the institutes where the authors were working, but, unfortunately, we got negative answers to our question whether parasitic wasps reared by these authors have been preserved. The species name, T. inunctus is now ambiguous, as the type is considered to be lost. However, a range of similarities with Aprostocetus epicharmus Walker concerning its biology has been revealed due to our rearing experiments, so the species earlier referred to as T. inunctus in the literature dealing with the raspberry cane midge is likely to agree with A. epicharmus (Vétek & al. 2006).

Tetrastichus lycidas (Walker) – Aprostocetus lycidas (Walker, 1839) The mention of this name by Darvas & al. (2000) is obviously a mistake in the publication, because the species parasitizes only Hartigiola annulipes (Hartig) and Mikiola fagi (Hartig), both of which are monophagous pests on beech (Fagus sylvatica) (Graham 1987).

Tetrastichus roesellae (Nees) – Aprostocetus roesellae (Nees, 1834) The species was reared by Bachmann (1950) and identified by Ferrière. This is an obvious misidentification because the host of this species is possibly a moth, which occured in large numbers on dill (Anethum graveolens) (Nees 1834). T. roesellae specimens can not be found in Ferrière’s collection, hence we could not check any specimens.

Tetrastichus sp. On the basis of the description given by Gordon & Williamson (1984) with respect to the biology of this not exactly determined species, it is likely that the species considered agrees with Aprostocetus epicharmus Walker. This can be supposed by the study of Vétek & al. (2006).

Tetrastichus tymber (Walker) – Aprostocetus tymber (Walker, 1839) In their book, Balás & Sáringer (1982) refer to Masten’s (1958) and Stoyanov’s (1963) studies in connection with the aforesaid species when indicating the natural enemy of the raspberry cane midge, but, on the basis of data published in these papers and shown above, this can be regarded as a false citation. Nevertheless, T. tymber is a parasitoid of Helicomyia saliciperda (Dufour), which lives on willow (Salix spp.) (Graham 1987).

Tetrastichus vincius (Walker) – syn. of Aprostocetus epicharmus (Walker, 1839) Domenichini (1966) mentions T. vincius as a parasitoid of the raspberry cane midge. However, the species was later synonymized by Graham (1987) as Aprostocetus epicharmus (Walker), but in this work there is no reference to the raspberry cane midge as a host of A. epicharmus. On the basis of our rearing experiments, A. epicharmus is the principal parasitoid of the raspberry cane midge (Vétek & al. 2006). T. vincius is also noted by Darvas & al. (2000) in connection with the pest. We were informed by the authors that the species name is a citation. The source was not mentioned, but it was probably the work of Domenichini (1966).

Ceraphronoidea: Ceraphronidae

Aphanogmus sp. This species was found to parasitize raspberry cane midge larvae in the Siberian region of Russia (Shternshis & al. 2002). Unfortunately, the reared insects could not be located and at 59

present no information is available on the specialist who identified the specimens either. Most of the European Aphanogmus spp. infest gall midges (Alekseev 1978), hence the mentioned species can be regarded as a possible natural enemy of the raspberry cane midge.

Platygastroidea: Platygastridae

Leptacis sp. Specimens belonging to this genus were reared by Bachmann (1950), Pitcher (1952, 1955) and maybe Stoyanov (1963). Ambrus (1973) also mentions this species, but on the basis of his description it can be supposed that he refers to Pitcher’s (1952) results in his study. In connection with Bachmann’s (1950) work, it can be said that the specimen of the mentioned Leptacis sp. reared from the raspberry cane midge is not available in either Ferrière’s collection or Bachmann’s former workplace. Pitcher (1952, 1955) gives thanks to G. E. J. Nixon besides Ferrière, for helping him identify the species which are indicated in his publications concerned. Nixon was once working for the Commonwealth Institute of Entomology. As the CIE no longer exists, Nixon’s collection was supposed to be found at the Natural History Museum, London, UK. Based on the information of the curator of the Hymenoptera collection, two specimens of the Leptacis sp. reared by Pitcher from the raspberry cane midge are placed there. No further material could be found at East Malling Research, Pitcher’s former workplace. Many Leptacis spp. are known from Europe infesting gall midges (Cecidomyiidae) (Vlug 1995), thus the indicated Leptacis sp. may parasitize the raspberry cane midge. This could be confirmed by further rearing experiments. We have not managed to find specimens belonging to this genus yet.

Piestopleura catilla (Walker, 1835) It is Pitcher (1952, 1955) and maybe Stoyanov (1963), who succeeded in rearing specimens of a species named P. catillus. It is probable that Ambrus (1973) only refers to Pitcher’s (1952) data in his work. The specimens reared by Pitcher could not be located in either Ferrière’s or Nixon’s collection or at EMR. We found only one author, Vlug (1995), who refers to this species as Piestopleura catilla (Walker) and mentions the raspberry cane midge as its host citing Pitcher’s above-mentioned two papers. Rearing experiments are required to confirm that P. catilla is indeed a parasitoid of the pest.

Synopeas craterus (Walker, 1835) This species is mentioned as a common parasitoid of Resseliella theobaldi in Great Britain (Gordon & Williamson 1984). We were informed that the species was found in Scotland but not reared or identified by the authors. According to Kozlov (1978 cit. Vlug 1995), S. craterus infests Resseliella (= Thomasiniana) ribis (Marikovskij). The biological description of S. craterus given by Gordon & Williamson (1984) is very similar to that of Pitcher’s (1952) in connection with the Leptacis sp. and P. catillus. Each of the three species belongs to the same family (Platygastridae). If specimens of the reared S. craterus were found, the relationship between the three species and that of between S. craterus and the raspberry cane midge could be clarified.

II. Parasitoids related to the rose stem girdler

Chalcidoidea: Eulophidae

Tetrastichus agrilorum (Ratzeburg) - Baryscapus agrilorum (Ratzeburg, 1844) T. agrilorum is noted by several authors as a parasitoid of the pest (Dobrivojević 1970; Reichart 1990; Molnár & al. 2001). We started to look for the specimens reared by Dobrivojević (1970) and determined by G. Domenichini as T. agrilorum in order to re- examine the material. Four specimens (three females and one male) were found in the collection of the NHM of the City of Geneva, Switzerland. We re-examined the material, and the specimens proved to be different from T. agrilorum. Therefore, the determination given by Domenichini should be regarded as a misidentification. On the other hand, these specimens are very similar to the highly variable Baryscapus endemus, which is widely distributed in Europe. Graham (1991) mentions that it was reared from numerous hosts, though, it is probably not a primary parasitoid but a hyperparasitoid in most cases. At present, each of the re-examined specimens can be regarded as a Baryscapus sp., nr. endemus (Walker, 1839). A Baryscapus sp. was reared by us, too, in small numbers from damaged raspberry canes collected from raspberry plantations in different growing regions of Hungary, in 2003–2004 and 2007. These specimens belong to the same species as those reared by Dobrivojević. Further investigation is required to clarify the taxonomic status of Dobrivojević’s and our specimens. Reichart (1990) refers to one of his previous works (Reichart 1968) when giving some information on the biology of T. agrilorum, determined by G. Szelényi. However, in his earlier work (Reichart 1968) Tetrastichus agrili Crawford is mentioned. On the basis of these sources in which all details about the two parasitoids are the same with the exception of the name of the hymenopterous species, the following can be presumed. In his former study, Reichart (1968), perhaps based on data found in some foreign literature, mentioned only a supposed name of the natural enemy of the pest, the specimens of which were reared but not determined exactly then. These specimens were identified later by Szelényi, and due to this process it turned out that the previously reared wasps actually do not belong to the species T. agrili but T. agrilorum, and this was indicated in the author’s later work (Reichart 1990). We deemed it necessary to re-examine the material of Reichart, but T. agrilorum specimens were not found in either Szelényi’s collection located at the Hungarian Natural History Museum or at the author’s former workplace (Plant Protection Institute, Hungary). In connection with the mention of T. agrilorum by Molnár & al. (2001), we were informed that the species name is only a citation. The source was the work of Reichart (1990). We have not managed to rear B. agrilorum yet.

Tetrastichus agrili Crawford, 1915 Reichart (1968) mentions that T. agrili caused a considerable larval mortality of the rose stem girdler in a Hungarian raspberry plantation, in 1964. However, the name of the specialist who identified the specimens is not indicated in the author’s work, and the reared insects can not be found at his former workplace either, so the specimens could not be checked. As T. agrili is a Nerctic species (Graham 1991), it does not occur in Hungary. Further details on the mention of this species have been discussed above in connection with the publications of Reichart (1968, 1990).

Tetrastichus heeringi Delucchi, 1954 T. heeringi is the most fequently mentioned parasitoid of the rose stem girdler. Its impotance in the control of populations of the pest is emphasized both in case of raspberry and rose by 61

many authors (Nikolova 1969, 1972; Dobrivojević 1970; Tsalbukov 1983; Graham 1991; Idinger 1991). We also found this beneficial insect to be a significant parasitoid species infesting larvae of Agrilus cuprescens. T. heeringi was reared from damaged raspberry canes collected from raspberry plantations in different growing regions of Hungary, in 2003–2004 and 2007. These data confirm the status of T. heeringi as a significant natural enemy of Agrilus cuprescens.

Tetrastichus rugglesi (Rohwer) - Baryscapus rugglesi (Rohwer, 1919) This species is mentioned by Mundinger (1941). It was a bit surprising to find this old reference in an American journal because the host (rose stem girdler) is not native to North America but was introduced into the continent (Jendek 2003; Larson 2003). However, there is a native cane borer in North America which causes similar damage to commercial and wild Rubus spp. to that of the rose stem girdler. This pest is called the rednecked cane borer (Agrilus ruficollis (Fabricius)) (Johnson 1991). Therefore, it might be supposed that the parasitoid, which is a Nearctic species and can not be found in the Palaearctic region (Noyes 2002), accepted the introduced raspberry pest as an alternative host.

Tetrastichus sp. Nikolova (1969) indicated this species as an egg parasitoid of the rose stem girdler. She also refers to a Tetrastichus sp. in one of her later works (Nikolova 1972). These seem to be interesting data, therefore we would have liked to locate the specimens reared by the author. We were informed that at the Institute of Zoology, Bulgaria, no insect material was left by Nikolova. The specialist who determined the reared specimens was Z. Boucek. Boucek’s collection is deposited in the National Museum of Prague, but it is not properly arranged systematically. The recovery of the species concerned is not easy, and should be carried out in the future. We have not found any egg parasitoids of the pest yet.

Chalcidoidea: Eupelmidae

Eupelmella vesicularis (Retzius) - Eupelmus vesicularis (Retzius, 1783) It was Ferrière who identified this species for Dobrivojević (1970). These specimens can not be found in the NHM of the City of Geneva, Switzerland. E. vesicularis is one of the most common species from the Chalcidoidea superfamily. It can be a primary or a secondary parasitoid as well. Therefore, its status as a natural enemy of the rose stem girdler can not be excluded but needs to be confirmed by well-arranged rearing experiments.

Chalcidoidea: Eurytomidae

Eurytoma rosae group and Eurytoma curculionum Mayr, 1878 The Eurytoma spp. are very difficult to identify. It was M. F. Claridge who determined the group for Dobrivojević (1970) and supposed some species to be E. curculionum. The latter species is presumably the result of a not properly arranged rearing experiment, as the known hosts of this parasitoid belong to the Curculionidae family. However, in case of Dobrivojević’s (1970) paper, mistakes in the text can not be excluded either. Further investigation is required to confirm the data of the author.

Ichneumonoidea: Braconidae

Heterospilus rubicola Fischer, 1968 H. rubicola was described by Fischer as a new species on the basis of some specimens from Serbia reared by Dobrivojević (1970). They are deposited in Fischer’s collection, in the Natural History Museum of Vienna. On the basis of this material the rose stem girdler is a host of H. rubicola.

Ichneumonoidea: Ichneumonidae

Cubocephalus annulatus (Cresson, 1864) The Nearctic species (Dicky & al. 2005), which belongs to the family Ichneumonidae, was indicated by Townes & Gupta (1962). We can only suppose that the situation is similar to that of B. rugglesi. Further investigation is needed to solve this problem. Nevertheless, neither of these two parasitoid species has been found to infest the rose stem girdler in Europe yet.

Xylophrurus sp. This member of the Ichneumonidae family was reared from the rose stem girdler by Nikolova (1972) in one case. At present, several species belonging to this genus can be found in the Palaearctic and Nearctic regions, too (Dicky & al. 2005). The recovery of the indicated specimen or further rearing experiments are needed to confirm its status. We have not reared Xylophrurus spp. yet.

Platygastroidea: Platygastridae

Platygaster cottei Kieffer, 1913 The mention of this species by Dobrivojević (1970) in connection with the rose stem girdler is obviously a mistake in his paper. P. cottei parasitizes a gall midge, Craneiobia corni (Giraud), which infests common dogwood (Cornus sanguinea L.) (Vlug 1995).

Acknowledgements

The authors would like to thank the following specialists, curators of national insect collections and researchers of plant protection institutes for their cooperation in this study: Bernhard Merz (NHM of the City of Geneva, Switzerland), Catherine Baroffio, Heinrich Höhn, Charly Mittaz, Christian Linder (Agroscope Changins-Wädenswil ACW), David Notton, John Noyes (NHM, London, UK), Jerry Cross (East Malling Research, UK), Špela Modic (Agricultural Institute of Slovenia), Marcela Skuhravá (Czech Republic), Klára Balázs, Béla Darvas (Plant Protection Institute, Hungary), Stuart Gordon (Scottish Crop Research Institute, UK), Margarita Shternshis (Novosibirsk State Agrarian University, Russia), Alberto Grassi (Instituto Agrario San Michele all’Adige, Italy), Zoltán László (NHM, Hungary), Mária Szántóné Veszelka (Directorate for Plant Protection and Soil Conservation, Nógrád county, Hungary), Toshko Ljubomirov (Institute of Zoology, Bulgaria), Peter Boyadzhiev (University of Plovdiv, Bulgaria), Vladimir Sakalian (Bulgaria), Eduard Jendek (Institute of Zoology, Slovakia), Jan Macek (Department of Entomology, National Museum, Czech Republic), Maximilian Fischer and Dominique Zimmermann (NHM, Vienna, Austria). This reserach was supported by the “GAK 2005 FKUT1kol” project and the Deák Ferenc Fellowship.

References

Alekseev, V.N. 1978: Ceraphronoidea. – In: Opredelitel nasekomykh Evropeyskoy chasti SSSR, Tom III, Pereponchatokrylye, ed: Trjapitzin: 664-691. Ambrus, B. 1973: A szeder és málna gubacslegyei (Diptera: Cecidomyidae). – Állattaniközlemények 60 (1-4): 25-37. Bachmann, F. 1950: Untersuchungen über die Gallmücke Thomasiniana theobaldi Barnes an Himbeerruten. – Schweizerische Zeitschrift für Obst- und Weinbau 59: 386-392. Balás, G. & Sáringer, Gy. (eds.) 1982: Kertészeti kártevők. – Akadémiai Kiadó, Budapest, 1069 pp. Barnes, H.F. 1926: The gall midges of blackberries and raspberries. – Journal of Pomology 5: 137-140. Barnes, H.F. 1927a: Material for a Monograph of the British Cecidomyidae or Gall midges. III. Gall midges attacking fruit. – Journal of the South-Eastern Agricultural College, Wye, Kent 24: 101-114. Barnes, H.F. 1927b: British gall midges. I. – Thomasiniana theobaldi sp. n. (The Raspberry Cane Midge). – Entomologist’s Monthly Magazine 63: 219-221. Barnes, H.F. 1944: Investigations on the raspberry cane midge, 1943–44. – Journal of the Royal Horticultural Society 69: 370-375. Darvas, B., Skuhravá, M. & Andersen, A. 2000: Agricultural dipteran pests of the Palaearctic region. – In: Contributions to a Manual of Palaearctic Diptera 1: General and Applied Dipterology. Papp and Darvas (eds.): 565-649. Dicky, S.Y., Achterberg, K. van & Horstmann, K. 2005: World Ichneumonoidea 2004, Taxonomy, biology, morphology and distribution. – CDrom. Taxapad. Dobrivojević, K. 1970: Prilog poznavanju parazita nekih štetnih insekata maline. – Jugoslovensko voćarstvo 4 (11-12): 199-204. Domenichini, G. 1966: Hym. Eulophidae, Palaearctic Tetrastichinae. – In: Index of Entomo- phagous Insects 1. Delucchi and Remaudière (eds.): 13-101. Fox Wilson, G. & Green, D.E. 1944: Observations on two raspberry troubles. – Journal of the Royal Horticultural Society 69: 79-86. Fritzsche, R. 1958: Beiträge zur Ätiologie des Himbeerrutensterbens. – Archiv für Gartenbau 6 (3-4): 171-216. Gordon, S.C. & Williamson, B. 1984: Raspberry cane blight and midge blight. – Leaflet 905. Ministry of Agriculture, Fisheries and Food, Alnwick, UK: 9 pp. Graham, M.W.R. de V. 1985: On some Rondani types of Chalcidoidea (Hym.) in theHaliday collection, Dublin. – Entomologist’s Monthly Magazine 121: 160. Graham, M.W.R. de V. 1987: A reclassification of the European Tetrastichinae (Hymeno- ptera: Eulophidae), with a revision of certain genera. – Bulletin of the British Museum (Natural History), Entomology series 55 (1): 1-392. Graham, M.W.R. de V. 1991: A reclassification of the European Tetrastichinae(Hymenoptera: Eulophidae): revision of the remaining genera. – Memoirs of the American Entomolo- gical Institute 49. The American Entomological Institute, Gainesville, Florida, USA: 322 pp. Grassi, A. 2003: Cecidomia del pollone (Resseliella theobaldi). – In: Guida al riconoscimento e alla gestione integrata e biologica delle principali avversità dei piccoli frutti in Trentino. Grassi and Pertot (eds.): 6-15. Idinger, J. 1991: Untersuchungen zur Verbreitung, Biologie, Ökologie sowie Prognose- und Bekämpfungsmöglichkeiten von Agrilus aurichalceus Redt. (Himbeerprachtkäfer). – Pflanzenschutz Wien 2: 3-4. 64

Jendek, E. 2003: Revision of Agrilus cuprescens (Ménétriés, 1832) and related species (Coleoptera: Buprestidae). – Zootaxa 317: 1–22. Johnson, D.T. 1991: Rednecked Cane Borer. – In: Compendium of Raspberry and Blackberry Diseases and Insects. Ellis, Converse, Williams and Williamson (eds.): 75. Larson, D.J. 2003: The rose stem girdler (Agrilus aurichalceus Redtenbacher) (Insecta: Coleoptera: Buprestidae), a new threat to prairie roses. – Blue Jay 61 (3): 176-178. Masten, V. 1958: Thomasiniana theobaldi Barnes – opasan štetnik malina (Prethodno saopštenje). Заштита Биља (Београд) 49-50: 201-207. Molnár, J., Szántóné Veszelka, M., Szeőke, K. & Vörös, G. 2001: Integrált védekezés termesztett növényeink kártevői ellen – Gyümölcsfélék. – In: Kártevők elleni védekezés II, ed. Seprős: 237-309. Mundinger, F.G. 1941: Two Buprestid Cane-borers of brambles with experiments on control. – Journal of Economic Entomology 34 (4): 532-537. Nees von Esenbeck, C.G. 1834: Hymenopterorum Ichneumonibus affinium. Monographiae, genera Europaea et species illustrantes 2. – Stuttgartiae et Tubingae: 448 pp. [Nikolova, V.] Николова, В. 1969: Ценологични проучвания в насажденията с маслодайна роза. Академия на Селскостопанските Науки, Институт за Защита на Яастенията – Гара Костинброд, 171 pp. [Nikolova, V.] Николова, В. 1972: Ентомоценологични и биологични проучвания в насаждения с маслодайна роза. Част IV Hymenoptera и Diptera. – Известия на Зоологическия Институт с Музей 35: 107-138. Noyes, J.S. 2002: Interactive Catalogue of World Chalcidoidea (2001 – second edition). CD- ROM. – Taxapad and The Natural History Museum. Pitcher, R.S. 1952: Observations on the raspberry cane midge (Thomasiniana theobaldi Barnes). I. Biology. – Journal of Horticultural Science 27 (2): 71-94. Pitcher, R.S. 1955: A comparison of a group of gall midges of the genus Thomasiniana Strand, E. 1916 and a description of a new species from wild blackberry. – Bulletin of Entomological Research 46: 27-38. Reichart, G. 1968: Málna kártevői. – In: Növényvédelmi Enciklopédia 2, ed. Ubrizsy: 267-279. Reichart, G. 1990: Málna-karcsúdíszbogár (Agrilus aurichalceus Redtenbacher). – In: A növényvédelmi állattan kézikönyve 3/A. Jermy and Balázs (eds.): 82-84. Shternshis, M.V., Beljaev, A.A., Shpatova, T.V., Bokova, J.V. & Duzhak, A.B. 2002: Field testing of BACTICIDE®, PHYTOVERM® and CHITINASE for control of the raspberry midge blight in Siberia. – BioControl 47 (6): 697-706. [Stoyanov, D.] Cтoянoв, Д. 1963: Проучвания върху малиновото комарче - Thomasiniana theobaldi Barnes в България. Известия на Института за Защита на Растенията – Гара Коcтинброд 4: 41-66. Townes, H. & Gupta, V.K. 1962: Ichneumon-flies of America north of Mexico: 4. Subfamily Gelinae, tribe Hemigastrini. – Memoirs of the American Entomological Institute 2. The American Entomological Institute, Gainesville, Florida, USA: 302 pp. [Tsalbukov, P.] Цалбуков, П. 1983: Взаимоотношения между агрилуса по маслодайната роза и Теtrastichus heeringi. РасmumeΛнa защuma 31 (7): 2-3. Vétek, G., Thuróczy, Cs., Pénzes, B. 2006: Interrelationship between the raspberry cane midge, Resseliella theobaldi (Diptera: Cecidomyiidae) and its parasitoid, Aprostocetus epicharmus (Hymenoptera: Eulophidae). – Bulletin of Entomological Research 96: 367-372. Vlug, H.J. 1995: Catalogue of the Platygastridae (Platygastroidea) of the world (Insecta: Hymenoptera). – In: Hymenopterorum Catalogus (nova editio). Achterberg (ed.): 168 pp.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 65-70

Open field and laboratory surveys to evaluate the susceptibility of red raspberry genotypes to Tetranychus urticae Koch and Resseliella theobaldi Barnes

Alberto Grassi1, Romano Maines1, Marcella Grisenti2, Marco Eccher2, Alessio Saviane2, Lara Giongo2 IASMA Research Center – 1Plant Protection Dept.- 2Agrifood Quality Dept. – Via E.Mach, 1 38010 San Michele all’Adige (TN) - Italy

Abstract: Since 2005, a screening aimed to test the susceptibility of floricane and primocane raspberry genotypes to the key pests two-spotted spider mite (TSSM) Tetranychus urticae and the raspberry cane midge (RCM) Resseliella theobaldi was started at Vigolo Vattaro (Trento, Northern Italy). Observations on T. urticae susceptibility were carried out by means of leaf inspections (5 terminal leaflets/accession) at 15-30 day intervals from May to August. The burden of TSSM that developed on each variety at the end of each year was expressed as cumulative mites/day. A classification of tolerance to TSSM was proposed using the total of the cumulative mites/day of the two seasons. In 2006, a laboratory test was carried out on 27 genotypes that had in 2005 differently expressed tolerance degrees to TSSM. The tolerance to R. theobaldi was evaluated using the technique for scoring midge blight proposed by Williamson & Hargreaves (1979). At the autumn pruning stage in 2005 and 2006, 7-10 canes/accessions were selected. The basal portion (30 cm) was scraped to remove the periderm. A 1-10 disease score was attributed to each cane on the basis of the extension of the vascular lesions beneath feeding wounds of R. theobaldi.. The average class of damage of the two seasons for each variety was used to develop a classification of tolerance.

Key words: red raspberry genotypes, susceptibility, Tetranychus urticae, Resseliella theobaldi.

Introduction

Evaluation and selection of new varieties suitable for the fresh market is a major objective of the research programme for the development of the soft fruits sector at IASMA Research Centre. Since 2003, 225 floricane and primocane genotypes were grown and monitored in an experimental field in Vigolo Vattaro (Trento, Northern Italy), located at 700 m above the sea level. The Rubus plants under evaluation were chosen from commercial varieties and advanced selections from several breeding programs. Five to ten plants of each accession were planted in pots, drip-fertigated and evaluated in single plots. Besides the parameters related to quality and quantity, since 2005 a screening aimed to test the susceptibility of these genotypes to the key pests two-spotted spider mite (TSSM) Tetranychus urticae Koch and the raspberry cane midge (RCM) Resseliella theobaldi Barnes was started. Here, as an example of the results obtained, we report the results for the autumnal fruiting genotypes.

Material and methods

Susceptibility to TSSM; open field surveys The population dynamic of the mite was assessed on 104 genotypes by means of visual inspections (with the aid of a 10 x hand lens) of 1 terminal leaflet/plant/accession at 15-30 day intervals from May to August. The number of TSSM mobile stages on the underside of the

65 66

leaf was estimated by means of a ranking system (Guignard 1968 unpub.). The mobile forms of predators, Stethorus spp. and phytoseiids in particular, were exactly counted. The burden of TSSM developed on each variety at the end of the season was expressed as cumulative mite- days per leaf. A classification of tolerance to TSSM was proposed using the sum of the cumulative mite-days per leaf of the two years.

Susceptibility to TSSM; laboratory test A test was carried out in 2006, on 14 autumnal fruiting genotypes that in open field surveys in 2005 expressed different degrees of tolerance to TSSM. Ten TSSM females were placed on the underside of each of 5 terminal leaflets for every selected accession. Then, the leaves were inserted in a polystyrene platform floating on water. A sticky barrier to prevent the females escape was created at the insertion point. The units were kept at room temperature (min.16.5- max.18.6 °C/min.70.9% - max 76% RH). One week later, the number of TSSM eggs and larvae on both sides of each leaflet was counted. The method used is a modification of the original one proposed by Sonneveld et al. (1997).

Susceptibility to RCM The tolerance to R. theobaldi was evaluated on 98 genotypes using the technique for scoring midge blight proposed by Williamson & Hargreaves (1979). At the autumn pruning stage in 2005 and 2006, 7-10 canes/accession were selected. The basal portion (30 cm) was scraped to remove the periderm. Using the reference map, a 0-5 disease score was visually attributed to each cane on the basis of the extension of the vascular lesions beneath raspberry cane midge feeding wounds. The average class of damage of the two seasons for each variety was used to develop a classification of tolerance. Also the number of cuts on the bark of each cane portion was estimated in 2006 according to 3 classes of presence (between 0 and 5 cuts = 2.5; between 5 and 10 cuts = 7.5; > 10 cuts = 10).

Results and discussion

Susceptibility to TSSM; open field screening. The screening in the field revealed different levels of susceptibility to TSSM among the different genotypes examined; the most important results are reported in the Figure 1. Varieties tolerant to different extents may be considered those below the arbitrary limit value of 150 cumulated mite-days per leaf. When the presence of predators was high, it is likely that the infestation of TSSM was superior to the one recorded. Therefore, among the varieties that resulted more tolerant, or less susceptible – the most interesting for breeding purposes – those with the least number of predators have much more probabilities to show this desired character. So, regarding the susceptibility to TSSM the most interesting accessions seem to be VV 01, VV 01 683, VV 01 520, VV 02 301, VV 01 455 and VV 02 19, that are advanced selections from the local breeding program. Among the varieties, Josephine, Ljiulin and Caroline showed a promising performance on the two years, though a rather large number of predators was recorded on them. Within the advanced selections several Autumn Bliss derivates from different breeding programs showed a good tolerance too. As in our results, also in Bylemans et al. (2003), Autumn Bliss is reported to be extremely susceptible to T. urticae, while first generation selections markedly seem to show improved tolerance. Josephine and Caroline are important commercial varieties, that develop little grey mould too, have a similar aromatic profile for some compound classes 67

and cluster very differently when compared to Autumn Bliss, Heritage, Himbotop and other top primocane varieties (Giongo et al. 2007). In addition to the TSSM predators, other environmental factors could have influenced the infestation development and, consequently, the correctness of the information we obtained. In fact, the number of plants (5) per accession was fairly low and they were non- randomly located. Anyway, the results can be considered significant; all the accessions in the different seasons had the same opportunity to be infested by TSSM, since plant to plant contacts let the mite horizontally move between them. An higher natural demographic TSSM development in 2006 compared to 2005 (average of 70 mite-days per leaf in 2005, 190 in 2006), allowed for more stringent screening conditions.

800 5 750 700 mite-days per leaf total n° of predators 650 4 600 f

550 total n° ofpredators/lea 500 3 450 400 350 2

300 f

cumulative mite-days per250 lea 200 150 1 100 50 0 0 ANNE VV 01VV CITRIA POLKA MBT-F1 LJIULIN JACLYN VV 02 19 VV VV 01VV 683 01VV 520 02VV 301 01VV 455 01VV 720 VV 01VV 481 VV 03VV 443 01VV 395 02VV 419 02VV 659 02VV 208 01VV 505 ALLGOLD HERITAGE CAROLINE JOSEPHINE HIMBOTOP VV 04 A 488 04 A VV BABIE LETA BABIE VV 01 331-531 VV JOAN SQUIRE AUTUMN BLISS AUTUMN INDIAN SUMNER INDIAN Figure 1. Results of the open field screening for the assessment of susceptibility to TSSM (2005-2006).

Susceptibility to TSSM; laboratory screening. In order to overcome the inconvenience described above related to the field trial and to acquire a better knowledge, an additional experiment was set up in 2006 under controlled conditions. The test was conducted on 14 accessions that previously showed different tolerance levels to TSSM at the end of the first open field screening in 2005. In contrast to the field experiment, in controlled conditions all the tested accessions showed a nearly identical tolerance to the mite, producing similar values of eggs and larvae developed by each female (Table 1). These results suggest that the tolerance/susceptibility of the raspberry plant to TSSM is the result of several environmental factors combination (relationships between plant, environment, climate, phytophagous, predators, etc.), that are not present in controlled settings. Another hypothesis for mite resistance of raspberry plants is related to the involvement of plant compounds, e.g. terpenoids, particularly expressed in the wax of the surface. They are known as anti-herbivore defense mechanism in the plants and documented in raspberry as involved in aphid tolerance mechanisms (Shepherd et al. 2000). In some cases, herbivore-induced volatiles may also act as direct defense, repelling (De Moraes et al. 2001; Kessler & Baldwin 2001) or intoxicating herbivores and pathogens (Andersen et al. 1994) and 68

the possibility that these substances also act in plant-plant communication has been discussed (Arimura et al. 2000). The concentration and diversity in our study of these compounds may be different in the various tested raspberry genotypes in open field growing conditions, determining the different levels of susceptibility to TSSM we recorded. In laboratory screening conditions, using detached leaves, these compounds and their effects against TSSM may have been inactivated, giving rise to similar egg and larvae production on the investigated genotypes. It is interesting to detect that Jaclyn, on which the least number of eggs and larvae per female was recorded, also showed the lowest percentage of females, eggs and larvae on the underside of the leaf. This information could suggest a possible resistance of this variety to TSSM. The underside of the leaf could be inhospitable for the mite, because of its chemical/ physical characteristics. Unfortunately, this resistance was not confirmed in the open field screening.

Table 1. Results of the laboratory test for the assessment of susceptibility to TSSM.

Autumnal fruiting raspberry genotypes - susceptibility to TSSM - 2006 Laboratory test N° of N° of TSSM Cumulative mite- % of TSSM % of TSSM Average n° of TSSM females N° of TSSM days per leaf females found eggs+larvae TSSM Raspberry genotype females (alive+dead) eggs+larvae 2005+2006 on the found on the eggs+larvae/female inserted found after 1 counted (open field underside underside found after 1 week (5/6/2006) week (12/6/2006) surveys) JACLYN 50 49 41 78 1300 26.53 215.0 AUTUMN BLISS 50 50 86 94 1382 27.64 761.0 POLKA 50 51 80 96 1475 28.92 199.8 HIMBOTOP 50 52 92 100 1519 29.21 246.8 CAROLINE 50 48 98 100 1494 31.12 48.0 HERITAGE 50 46 91 99 1449 31.50 118.6 LJIULIN 50 47 85 94 1488 31.66 135.9 ALLGOLD 50 48 96 99 1530 31.87 322.0 BABIE LETA 50 46 59 91 1517 32.98 440.8 CITRIA 50 49 86 94 1623 33.12 630.2 ANNE 50 49 90 95 1655 33.77 205.7 JOAN SQUIRE 50 47 100 100 1605 34.15 262.6 MBT-F1 50 49 98 100 1777 36.26 216.4 JOSEPHINE 50 49 94 100 1778 36.28 80.3

Susceptibility to RCM Similarly to the analysis conducted on TSSM, the genotypes evaluated in open field showed different levels of tolerance to RCM (most important results are reported in the Figure 2). The most interesting accessions, probably endowed with an higher tolerance to RCM, are those with a lower average class of damage in the two seasons, reasonably with values ≤ 1. VV 02 659, VV 02 09, VV 02 672, VV 01 331 531 and VV 02 301 are advanced selections with Tulameen in the ancestors. Very interesting for the fresh market, they showed the lowest damage to RCM. Among the varieties, Allgold, a yellow Autumn Bliss seedling, Babie Leta from Eastern Europe and Jaclyn from US showed the greatest tolerance. Among the most susceptible, Polka, from Poland, and Joan Squire from the UK were recorded. As for TSSM, the best conditions in order to evaluate the sensitivity to RCM occurred in 2006, because of an higher infestation compared to the previous year that produced an average class of damage of 1.1 in 2005, while it was valued 2 in 2006. 69

As it can be seen from table 2, the most tolerant varieties showed also on average a low number of natural splits. This is in contrast with the highly susceptible genotypes, for which 10 splits/cane were registered. Genotypes that more easily tend to split on the bark are also more susceptible to R. theobaldi infestations, since the females of this midge lay their eggs exclusively in the cuts. It is also interesting to note that among the most tolerant genotypes, some of them showed an high average number of splits/cane: looking in detail at these splits it was however possible to verify that many of them were internally healed and not fully opened around the circumference of the cane, as was the case for most of the susceptible varieties. This kind of cuts are less susceptible to infestation by RCM larvae. Therefore, the results show that in the overall resistance process the capacity of the genotype to self-repair from inside the splits is of great importance.

4,50 average class of RCM damage (2005+2006) 4,00 3,50

3,00

2,50 2,00

1,50

1,00 0,50

0,00 ANNE CITRIA POLKA MBT-F1 T X 92 5 T X LJIULIN JACLYN VV 02 09 VV 02 19 AB 89 X 4 VV 02VV 659 02VV 672 02VV 301 01VV 505 01VV 395 01VV 683 02VV 208 02VV 419 VV 01VV 720 VV 01VV 520 VV 03VV 443 01VV 481 ALLGOLD HERITAGE BC 86-6-15 CAROLINE JOSEPHINE HIMBOTOP BABIE LETA BABIE JOAN SQUIRE VV 01 331-531 VV AUTUMN BLISS INDIAN SUMNER Figure 2. Results of the screening for the assessment of susceptibility to RCM (2005 -2006).

Table 2. Average number of splits/cane.

VV 02 659 02 VV 09 02 VV 672 02 VV 331 531 01 VV 301 02 VV ALLGOLD 505 01 VV 395 01 VV JACLYN 683 01 VV 208 02 VV 419 02 VV HERITAGE POLKA SQUIRE JOAN LJIULIN JOSEPHINE 720 01 VV AB X 89 4 5 X 92 T BC 86-6-15

average n° of splits/cane in 2.52.58 3.52.55 9 1010105 10101010101010101010 2006

average class of 0.1 0.1 0.2 0.3 0.4 0.6 0.7 0.7 0.7 0.9 0.95 1 2.6 2.7 2.8 2.88 2.95 3.3 3.6 4 4 RCM damage (2005+2006) MORE TOLERANT GENOTYPES HIGHLY SUSCEPTIBLE GENOTYPES

Conclusions

Tetranychus urticae and Resseliella theobaldi are two of the key pests of cultivated raspberry in Trentino. 70

For their control, every year the growers invest much effort, including chemical treatments. The selection of improved varieties not only for their qualitative aspects, but also for resistance/tolerance to these pests, is therefore of primary importance to help the growers to safeguard the agro-ecosystems and to produce healthy fruits for the consumers. The inspection methods we adopted permitted to highlight important differences among the investigated genotypes. However, it is necessary to acquire further historical information, to repeat for more years the open field investigations and to improve the experimental conditions (a larger number of plants randomly positioned in the field, reduction of the predator interferences on TSSM infestation development, artificial inoculation of the plants in field with TSSM females, etc.). Moreover, the results of the laboratory screening confirm that a multidisciplinary approach to the study of raspberry plant resistance against the pests is necessary, with the involvement of specific biochemical and physiology studies.

References

Andersen, R.A., Hamilton-Kemp, T.R., Hildebrand, D.F., McCracken, C.T. Jr, Collins, R.W. & Fleming, P.D. 1994: Structure-antifungal activity relationships among volatile C6 and C9 aliphatic aldehydes, ketones and alcohols. – J. Agric. Food. Chem. 42: 1563-1568. Arimura, G., Ozawa, R. & Shimoda, T. 2000: Herbivore-induced volatiles elicit defence genes in lima bean leaves. – Nature 406: 512-515. Bylemans, D., Janssen, C., Latet, G., Meesters, P., Peusens, G., Pitsioudis, F. & Wagelmans, G. 2003: Pest control by means of natural enemies in raspberry and red currants under plastic tunnel. – IOBC wprs Bulletin 26 (2): 37-44. De Moraes, C.M, Mescheer, M.C. & Tumlinson, J.H. 2001: Caterpillar-induced nocturnal plant volatiles repel non specific females. – Nature 410: 577-580. Giongo, L., Aprea, E., Carlin, S., Palmieri, L., Saviane, A., Grassi, A. & Gasperi, F. 2007: Multidisciplinary characterization of primocane raspberry genotypes for fruit quality compared to floricane fruiting cultivars. – Acta Horticulturae - Eucarpia Fruit breeding and genetics. In press. Kessler, A. & Baldwin, I.T. 2001: Defensive function of herbivore-induced plant volatiles emissions in nature. – Science 291: 2141-2144. Sheperd, T., Robertson, G.W., Griffiths, D.W. & Birch, A.N.E 2000: Plants and Aphids: The chemical ecology of infestation. – Annual Report of the Scottish Crop Research Institute 1999/2000. Scottish Crop research Institute, Dundee: 122-124. Sonneveld, T., Wainwright, H. & Labuschagne, L. 1997: Two methods for determining the resistance of strawberry cultivars to Two-spotted Mite (Acari:Tetranychidae). –Acta Horticulturae (ISHS) 439: 199-204. Williamson, B. & Hargreaves, J. 1979: A technique for scoring midge blight of red raspberry, a disease complex caused by Resseliella theobaldi and associated fungi. – Ann. Appl. Biol. 91: 297-301. Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 71-75

Harmfulness of raspberry gall midge, Lasioptera rubi Schrank (Diptera, Cecidomyiidae), to some raspberry cultivars

Slobodan Milenković1, Snežana Tanasković2 1 Fruit Research Institute Čačak, Kralja Petra I/9; 2 Faculty of Agronomy Čačak, Cara Dušana 34, Serbia

Abstract: The raspberry gall midge, Lasioptera rubi Schrank (Diptera, Cecidomyiidae) is an economically important pest, especially raspberry plantings where canes are not removed after harvest. The objective of this study was determination of the harmfulness of L. rubi in an untreated raspberry cultivar plantation at the Fruit Research Institute, Serbia. For the assessment, canes were collected from a planting of raspberry cultivars established in 2002. The trial involved five raspberry cultivars planted in a random arrangement with 100 plants in a replicated design. The intra-row and inter-row planting distances were 0.33 m and 2.5 m respectively. The sampling was performed on Willamette, Meeker, Latham, and hybrid K 81-6, and the rows were selected at random. Five meters of each of the chosen rows were marked. Canes displaying morphological changes in gall form were collected from the marked row sections. The samples were taken to the laboratory where cane diameter, length and width were measured, and the numbers of cane galls and larvae were checked. In the collected canes, galls occurred in different forms i.e. elongated, roundish or canker wound forms. Galls were observed along the entire cane length, at different cane sections (lower, medium and upper parts) as well as on petioles. Different levels of harmfulness of raspberry gall midge were recorded over the two years of study on the different raspberry cultivars.

Key words: Lasioptera rubi, crop damage, raspberry, cultivars

Introduction

In Serbian raspberry growing regions, large raspberry aphid (Amphorophora idaei Born.), small raspberry aphid (Aphis idaei van der Goot), raspberry beetle (Byturus tomentosus F.), strawberry blossom weevil (Antonomus rubi Hbst.) and raspberry stem girdler (Agrilus aurichalceus Retd.) are the economically the most serious insects pests. The pressure of midges, i.e. raspberry cane midge Resseliella theobaldi Barnes and raspberry gall midge Lasioptera rubi Schrank. (Heeger.) has been on the rise lately (Milenković & Ranković 2001; Gordon et. al. 2003; Milenković 2005). R. theobaldi and L. rubi from the Cecidomyiidae (Diptera) family are economically serious pests distributed throughout Europe. Within the certification schemes OEEP/EPPO (1993) for the Rubus genus and hybrids PM 4/10(1), these two pest insects are deemed damaging organisms which require preventive measures of monitoring (compulsory visual monitoring) aimed at the elimination thereof from the growing field (for all categories of reproductive planting material utilized for the propagation of certified planting material). Compulsory chemical control measures are envisaged within good agricultural practice (GAP 2/26(1)) of OEEP/EPPO (2002) schemes. Earlier investigations in Serbia have shown that raspberry gall midge develops one generation per year. In May or June, females lay eggs in groups on the surface of primocanes,

71 72

very close to the developing flower buds (Simova-Tošić 1970), i.e. averaging 15 eggs laterally at the base of the flowering buds (Anonymous 2007). The development of the embryo takes from eight to ten days. Immediately upon hatching, white larvae tunnel into the canes, forming globules underneath the epidermis. The gall is visible three to six weeks later, in July or August. Orange yellow to red larvae are housed inside the galls where they overwinter. Most commonly, an orange cocoon is formed during April in the following year and development of the imago takes two to three weeks. The adult insect is dark in colour, 1.5-3 mm long, with a small head, black eyes, and glassy, closely-overlapping wings. The thorax is brown, with sliver stripes at the front. The abdomen is dark brownish, with silvery gray stripes of transverse volumes. Damage is made visible by the incidence of galls. Asymmetrical globules, 3 - 4 cm long and 1.5 - 2 cm wide (Anonymous 2007) are observed on canes, lateral canes, petioles and stalks. Initially, the galls are green but take on a light brownish colour and become lignified later in the season. The gall surface cracks, and a chamber inside houses a variable number of orange red larvae. Growth of the infested canes and fruit formation is inhibited. It is a common sight that canes break just beneath the galls. The most efficient control measure is removal of the infested canes subsequently upon harvesting and all through the following April, prior to the emergence of the new generation imago. The pest pressure is increased if mechanical injuries have been inflicted on canes (by tools or hail). In raspberry plantings of Western Serbia, in the region of Čačak, about 30% of plants with galls of raspberry gall midge were evidenced upon the occurrence of hail in 1999 (Ivanović et al. 1999).

Material and methods

The trial was set up in the raspberry planting of the Fruit Research Institute Čačak, at the 'Zdravljak' site, the GPS coordinates being USR 7444859 and 4855139, the altitude 649 m, southern orientation. The planting was established in 2002. It included five genotypes planted in a random design in 10 north-east facing rows with at least four replications per genotype and 100 plants per replication. The intra-row and inter-row planting distance was 0.33m and 2.5m respectively. During the growing period 2006, four years after planting, six floricanes per metre were recorded on average. Sampling was performed on Willamette, Meeker, Latham and the hybrid K 81-6 on 2 m per row in four replications. Rows and row sections intended for sampling were determined by the random choice method. All canes with symptoms of morphological changes in the form of galls were collected from the designated sections in the rows. Sampling was performed on August 2nd, 2006 and July 6th, 2007. At the same time we made artificial splits on first primocanes in the the row, after collection in July 2007. During this vegetation season we did a second control, on October 12th, by visual inspection of all replicates. Collected canes were brought to laboratory where the diameter of canes was measured, and diameter, length and width of galls and the number of larvae in galls were recorded. The features were measured using an ˝Inox˝calliper, precision ± 0.05 mm. Average values of temperature and humidity over the assessment period were precisely recorded.

Results and discussion

Throughout the research period, over both growing seasons, galls made by raspberry gall midge were found in Latham, whereas in Willamette and hybrid K 81-6 the globules were found only in the 2006 growing season. No galls have been evidenced in Meeker over the research period (Table 1-2). During the first research year, the galls were covering the entire 73

length of primocanes, being sporadically observed on fruiting canes and petioles (photo 1), whereas in the following vegetation period these were found only on primocanes at a height not exceeding 70 cm. The most vigorous canes were observed in hybrid K 81-6. The incidence of galls in 2007 was observed about three weeks earlier than the previous growing season, probably because of high air temperatures and low humidities.

Table 1. Mean values by cultivars/year, summer sampling.

% of mean No of mean size of galls (cm) mean Ø of mean No of infested galls / Cultivar infested canes larvae/galls canes infested length width Ø / year (cm) (min.-max.) canes 06 07 06 07 06 07 06 07 06 07 06 07 06 07 Meeker 0 0 0 0 ------17.2 5.18 Latham 10 12.5 1.2 1.83 0.68 1 1.8 1.64 1.11 0.95 1.25 1.15 (5-51) (1-10) 9.4 Willamette 8 0 1.25 0 0.74 - 1.97 - 1.05 - 1.24 - - (2-29) 10.9 K 81-6 16 0 1.5 0 1.05 - 1.52 - 1.17 - 1.36 - - (0-37)

The largest number of infested canes was found on hybrid K 81-6 which had 16% of canes infested, followed by Latham and Willamette which had 10% and 8% of canes infested, respectively. The highest larval pressure was recorded in a gall in Latham (51), while in hybrid K 81-6 some parasited larvae were found. The calculated values for 2006 show that the highest larval pressure was found in Latham (17.2), which further implies that this cultivar is the most susceptible to raspberry gall midge under domestic conditions of growing. The pest pressure decreased to 10.9 in hybrid K 81-6 and 9.4 in Willamette. As regards the average number of galls per cane, the number showed a downward tendency, i.e. K 81-6 - 1.5 galls, Willamette - 1.25, Latham - 1.2. As for the cane thickness, the highest values are found in Willamette, with lower values in K 81-6 and Latham. Galls were longest in Willamette and Latham followed by hybrid K 81-6 . In respect of the width and diameter of galls, the highest values were recorded in K 81-6, Latham and Willamette. The results infer that the galls were largest in hybrid K 81-6, whereas Latham and Willamette recorded similar values. The growing season 2007 was characterized by higher temperatures and low humidities as compared to the previous season. Such conditions were best suited to Latham in which primocane diameter was 1 cm during this growing period and which averagely recorded 12.5% of infested canes (Table 1). As compared to the previous growing season galls were less elongated, the average number of larvae being almost three times lower (Table 1). The incidence of galls in October were recorded on Latham, Willamette and hybrid K 81-6. If we compare the two samplings made in 2007 and mean measured values on Latham, we can conclude that canes were thinner in second sampling, as expected. But, galls had a greater diameter, width and length (Table 2). The percentage of infestation was the same over both controls. However, in the 2nd collecting 50% of the galls were empty. In two galls we recorded young and older larvae in the same gall cabin. On Willamete we registered 8.3% of infested canes but 40% of the galls were empty. On hybrid K 81-6, 6.2% of the canes were infested and 25% of them were empty. In one gall we observered the presence of young and old larvae as well as one pupa (photo 3). 74

Table 2. Comparison of the mean values between summer and autumn controls in 2007.

% of mean No of mean size of galls (cm) mean Ø of mean No of infested galls / Cultivar infested canes larvae/galls canes infested length width Ø / sampling (cm) (min.-max.) canes 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd Meeker 0 0 0 0 ------0.8 5.18 3.25 Latham 12.5 12.5 1.83 1.33 1 1.64 1.86 0.95 1.14 1.15 1.38 1 (1-10) (0-10) 0.8 3.2 Willamette 0 8.3 0 1.25 - - 1.87 - 0.94 - 1.13 - 2 (0-11) 1.1 4.75 K 81-6 0 6.25 0 1.33 - - 1.72 - 1.12 - 1.75 - 3 (0-8)

Our results demonstrate that in conditions of high temperatures and lower humidity raspberry gall midge has a very extended flight period or maybe two generations per year. We cannot exclude or define first or second generations, but new assessment should be done to prove our hypothesis.

Photo 1. Galls on raspberry petioles. Note colour change

Photo 2. Galls – diameter, empty and with larvae.

Photo 3. Different development stages found in a gall cabin, October 2007 sampling.

References

Anonymous, 2007: Himbeerrutengallmücke. http://www.insektenbox.de/zweifl/himgal.htm Dobrivojević, K. 1968: Economically important raspberry pests in production regions of Valjevo and Čačak. – Plant protection, 19 (100-101): 253-271. [in Serbian] Gordon, S.C., Milenkovic, S., Cross, J.V. & Stanisavljevic, M. 2003: Raspberry pests in Serbia. – IOBC wprs Bulletin 26(2): 23-27. Ivanović, M., Petanović, R. & Milenković, S. 1999: Current problems in raspberry and blackberry protection. – IV Plant protection Yugoslav symposium. Proceeding of abstracts. 32-33. [in Serbian] Masten, V. 1958: Thomasiniana theobaldi Barnes. Dangerous raspberry pest. – Plant protection: 48-50. [in Serbian] Milenković, S. 2005: Raspberry pest and diseases in Serbia. – COST: WG3 First Meeting. Report. Wageningen 7/ 9 July 2005. Meeting code: COST-863-070705-01479. Milenković, S. & Ranković, M. 2001: Integral raspberry protection. – Proceedings of winter school for agronomist. Faculty of agronomy, Čačak 5(5):59-65. [in Serbian] Milenković, S., Tanasković, S. & Sretenović, D. 2006: Follow up of flight raspberry cane midge Resseliella theobaldi (Diptera, Cecidomyiidae) by pheromone trap. – VIII Plant protection Yogoslav symposium. Proceeding of abstracts: 117-118. [in Serbian] OEPP/EPPO 1993: Recommendations made by EPPO Council in 1992: scheme for the production of classified vegetatively propagated ornamental plants to satisfy health standards. – Bulletin OEPP/EPPO Bulletin 23: 735-736. OEPP/EPPO 2002: Good plant protection practice Ribes and Rubus crops. – Bulletin OEPP/EPPO Bulletin 32: 367-369. Simova-Tošić, D. 1970: Little known raspberry pests from family Cecidomyidae (Diptera). – Jugoslav horticulture: 193-197. Stamenković, S., Pešić, M. & Milenković, S. 1996: Raspberry and blackberry pests. – Plant doctor 2: 136-150. [in Serbian]

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 77-84

Patterns in the within-cane distribution of the gall-like swellings caused by Agrilus cuprescens (Coleoptera: Buprestidae) and the rate of raspberry infestation

Martina Váňová1, Peter Tóth2, Jozef Lukáš3 1 Slovak Academy of Sciences, Institute of Forest Ecology, Branch for Wood Plants Biology, Akademická 2, SK-94901 Nitra, Slovak Republic; 2 Slovak Agricultural University, Department of Plant Protection, Tr. A. Hlinku 2, SK-94901 Nitra, Slovak Republic; 3 Comenius University, Department of Ecology, Mlynská dolina B-2, SK-84215, Bratislava, Slovak Republic

Abstract: During 2004-2006, a neglected raspberry (Rubus idaeus L.) plantation (locality Sazdice) and a private back garden (locality Demandice) were investigated to determine the rate and intensity of raspberry infestation by Agrilus cuprescens (Ménétriés, 1832) and within-cane distribution of gall- like swellings caused by the pest larvae in Slovakia. The mean diameter of gall-like swellings ranged from 0.90±0.14 to 1.12±0.22 cm. Although the within cane distribution of gall-like swellings varied between years, the highest densities were observed from ground level up to 60 cm height-above- ground at the both localities. The rate of raspberry infestation fluctuated from 8.5% to 23.9% at Demandice and from 66.5% to 81.5% at Sazdice. The mean of gall-like swellings per cane ranged from 1.25±0.58 to 1.45±0.74 at Demandice, whereas from 1.58±0.81 to 2.7±1.58 per cane at Sazdice. The results show that the neglected plantations are heavily attacked by A. cuprescens, while the treated localities, where infested canes are annually removed, are damaged only slightly. Destruction of infested canes seems to be the simplest and a sufficient remedy against A. cuprescens.

Key words: control, intensity of infestation, pest, rose stem girdler, Rubus idaeus, symptom parameters

Introduction

Quality of red raspberry (Rubus idaeus L.) yield is the most important feature of saleable production. Although many arthropods are found in red raspberry plantations within Europe, only a few cause damage resulting in yield loss or reduction in fruit quality. The rose stem girdler (RSG), Agrilus cuprescens (Ménétriés, 1832) (Coleoptera: Buprestidae), is one of the most important pests causing the damage of canes and subsequently fruits. The species belongs to the group of relatively new pests emerging as a result of changes in pesticide usage, the cultivation and climate. The RSG is a member of the Buprestidae, adults in the family are commonly known as jewel beetles but the larvae are known as flat-headed borers. Adult buprestids are very active fliers, and the group is one of the most destructive insect families in the forest (Bright 1987). They kill their hosts by feeding in the soft phloem layer under the outer bark, cutting off the connections between the roots and the shoots and preventing the transport of nutrients and water (Haack & Benjamin 1982). The common hosts of the species are wild and cultivated raspberry and blackberry (Balás 1966; Bílý 1982), but wild, ornamental and perfumery roses can also be infested (Brussino & Scaramozzino 1982).

77 78

The RSG is widespread in Europe, the Near East, Asia Minor, the territory of North Caucasus (Reichart 1990) and also in North America (Stamenkovič et al. 1996). The brownish or greenish-golden, metallic bronzed elongated buprestid beetles are 4.5-8 mm long (Schaefer 1949). The egg is half-elliptical, 1-1.5 mm long, and is cemented to the cane by a gelatinous substance of yellowish colour which hardens in air (Reichart 1968). The milky white legless larva has a small dark brown flattened head. Fully grown it reaches a length of 7-20 mm. Length of the pupa which is the same colour is 7-8 mm (Reichart 1990). The species has only one generation a year (Lekic 1966; Brussino & Scaramozzino 1982; Reichart 1990; Kuroli 1999). The overwintering larvae, still in the previous season's girdled canes, pupate in April and most of the adults emerge from late May until end of June. The emergence period lasts 24-60 days depending on temperature (Kuroli 1999). Adults feed on the young foliage of raspberries and blackberries and mate repeatedly. Oviposition begins 5-7 days after feeding and starts mostly in early June (Lekic 1966). Eggs are laid singly on the exterior of the new canes, mainly on the side exposed to sunlight (Lekic 1966; Fischer–Colbrie et al. 1987), usually near a bud towards the base of the cane (Brussino & Scaramozzino 1982). Upon hatching, the young larva bores into the sap wood and under the surface of the bark, forming a spiral upward tunnel encircling the cane five or six times. The cane girdling causes the formation of a gall-like swelling (GLS), which appears in July-August. After girdling the stem, the larva bores into the pith of the stem, where it lives and bores in an upward direction. As it grows, the tunnel spreads and is filled with excrement and detritus. In late August the larva excavates an enlarge cavity in the cane and bites out through the cane wall making a D- shaped exit hole which is plugged up with detritus (Kuroli 1999). Adults feed on foliage, causing only minor leaf loss. The major damage is done by the larvae as they girdle the canes (Kuroli 1996). Spirals interrupt sap circulation (Brussino & Scaramozzino 1982) and this in turn slows growth, favoring the insect. The girdling weakens the canes and they can not support a full crop the following season. Fruit production from infested canes is limited. Average fruit weight being lowered from 0.5 to 22%, the losses in yield ranged from 21 to 37%. In fruits have a decreased content of protein, glucose and total sugars. The infestation also negatively affects vegetative behaviour of the raspberry plants, lowering their density, height and diameter (Tsolova 2002). Girdled canes are more susceptible to winter injury. In extreme cases, the cane may be killed outright (Reichart 1990). Such impact is very important especially for raspberry growers. One of the greatest challenges facing the successful management of the pest is the ability to accurately detect and suppose the presence of GLSs caused by the pest. Up to now, little or no relevant information exists about within-cane distribution of swollen areas caused by A. cuprescens larvae and about the rate of raspberry stands infestation under field conditions. The main objective of the work reported here was to obtain information that would provide a basis for effective management of the RSG in raspberry. Therefore the partial aims of the present study were to determine the following: (1) characteristics of symptoms; (2) within-cane distribution of symptoms; (3) rate of raspberry stands infestation; and (4) intensity of raspberry cane infestation. Due to its narrow host range, A. cuprescens was mentioned as a biocontrol agent that show potential for biological control of Rosa rubiginosa in New Zealand (Zwolfer 1965) and Rosa multiflora Thunberg ex. Murray in North America (Amrine 2002). There are also possibilities to control weedy Rubus spp. plants spreading within natural ecosystems outside Europe. As to any biological control program, the basic need is to study all aspects of the life cycle and impact of selected species. Thus, the research presented could contribute also in this way. 79

Material and methods

Two localities in southwest of Slovakia were investigated during 2004-2006 (Table 1). Both localities represented the warm and dry climate region with a mild winter (Zaťko 2002). Locality Demandice was a red raspberry plantation in a private back garden, where the crop was grown according to local farmers’ practices (biennial growing method). No chemical treatment was applied at the site. The only action realized was annual winter pruning and besides all infested canes where removed. The locality was situated at the bottom of the valley with exposure to the northwest, partially shaded by acacia forest. The raspberry cultivars grown in the back garden were Bulgarian Ruby and Ljulin. Locality Sazdice was a large (c 4 ha), more than 10 years old neglected red raspberry plantation with high stage of succession without chemical and agricultural treatments. The locality was southeast exposed and was situated in more open hilly country. The raspberry cultivars grown at the plantation were Canby and Heritage. Both localities were checked regularly each year after leaf fall, when the symptoms of the RSG infestation were clearly visible.

Table 1. Details of localities in southwest of Slovakia, where investigations of A. cuprescens were carried out (2004-2006). Coord. = coordinates; Alt. = altitude in meters; Temp. = average temperature in °C; Rad. = global radiation; Crop = covering crop species.

Locality Alt. a Coord. a Relief Temp. b Rad. b Crop 48˚07'N Hilly 1250- Demandice 140 8-9 Raspberry 18˚47'E country 1300 48˚05'N Hilly Sazdice 140 9-10 1300< Raspberry 18˚47'E country a Magellan GPS 315; b According to Zaťko (2002); Global radiation is expressed in kcal/cm2

To determine the characteristics of GLSs, the infested canes were cut down close to the ground. While 60 randomly selected infested canes were taken at Sazdice, all infested canes were evaluated at Demandice each year (there were 55, 89 and 29 infested canes in 2004, 2005 and 2006, respectively). The position of GLSs within canes was measured from the base of the cane to the midpoint of the swellings using a tape measure. The diameter of swellings was measured with a caliper rule at widest point of the swellings. The total number of GLSs was counted and recorded for each of the infested canes. The intensity and rate of infestation were counted for each locality and year separately. The intensity of raspberry infestation was expressed as an average number of GLSs per infested cane. The calculations were based on the data about total numbers of GLSs per each evaluated cane. The rate of raspberry infestation reflected the percentage of infested canes at the locality. At Demandice all raspberry canes were evaluated (297, 373 and 340 canes in 2004, 2005 and 2006, respectively). Because of the large acreage at Sazdice it was impracticable to evaluate all raspberry canes. Thus 200 canes were chosen at random within the site. Four smaller, equally sized subplots were selected within the plantation. In each subplot, 50 canes were randomly selected and evaluated. At both localities the infested canes were expressed as a percentage of the total number of canes evaluated at the site to give the rate of infestation. 80

Results

During 2004-2006, a total of 1790 red raspberry canes were sampled at the two checked localities. From these, 791 canes (44.2%) were infested by A. cuprescens (having one or more GLS). The within-cane distribution of GLSs varied from 2 to 192 cm (Figure 1 and 2) height- above-ground. The swellings were never found higher than 124 cm above ground at the neglected red raspberry plantation at Sazdice and were found at very low densities, or not at all, in smaller diameter canes at the both localities. The mean diameter of GLSs at Demandice and Sazdice was 0.90±0.14 and 0.94±0.19 cm in the year 2004, 0.97±0.18 and 1.02±0.16 cm in 2005, and 1.12±0.22 and 0.97±0.14 in 2006, respectively. The maximal diameter of GLS was 2.25 cm and minimal 0.54 cm; both diameters were recorded at locality Sazdice in 2004. Although the within cane distribution of GLSs varied between years, the highest densities of GLSs were observed beginning at ground level up to 60 cm height-above-ground at both localities during the study (Figure 1 and 2). 65.5% of the GLS at Demandice and 84.6% at Sazdice were recorded within the section 0-60 cm. The maximal number of GLS per cane revealed during the survey was 8 at Sazdice (in 2004) and 4 at Demandice (each year).

2004 20 2005 2006

10 Number of gall-like swellings gall-like of Number 5

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 Height-above-ground section of the raspberry cane (cm)

Figure 1. Within-raspberry cane distribution of the gall-like swellings caused by the A. cuprescens at the locality Demandice during 2004-2006.

The rate of raspberry infestation in the cultivated private back garden at Demandice was much lower in all investigated years than it was at neglected plantation in Sazdice. The rate of infestation fluctuated from 8.5% to 23.9% infested raspberry plants at Demandice and from 66.5% to 81.5% at Sazdice (Table 2). The highest intensity of raspberry infestation was recorded at the Sazdice in each observed year also. The mean of GLS per cane ranged from 1.25±0.58 to 1.45±0.74 what meant usually 1-2 GLS per infested cane at Demandice, whereas from 1.58±0.81 to 2.7±1.58 i.e. 1-4 GLS per cane at Sazdice (Table 2).

35 2004

30 2005 2006 25

Number of gall-like swellings gall-like of Number 10

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 Height-above-ground section of the raspberry cane (cm)

Figure 2. Within-raspberry cane distribution of the gall-like swellings caused by the A. cuprescens at the locality Sazdice during 2004-2006.

Table 2. The rate and intensity of raspberry plantation infestation by A. cuprescens at Demandice and Sazdice during 2004-2006.

Year 2004 2005 2006 Locality Demandice Sazdice Demandice Sazdice Demandice Sazdice Intensity of 1.25±0.58 2.7±1.58 1.9±0.49 2.3±1.08 1.45±0.74 1.58±0.81 infestation1 Rate of 18.5% 72% 23.9% 81.5% 8.5% 66.5% infestation2 1 Mean number of gall-like swellings caused by A. cuprescens per infested raspberry canes 2 Percentage of raspberry plants infested by A. cuprescens per site

Discussion

Common patterns of GLS distribution were observed among the raspberry canes investigated during the study (Figure 1 and 2). Although GLS were found more-less across the whole length of canes (2-192 cm high-above-ground), the highest numbers were clustered within the cane section 0-60 cm high-above-ground. The mean diameter of GLS ranged from 0.90±0.14 to 1.12±0.22 cm at both localities during the three years. Kuroli (1999) also recorded similar data (0.9-1.5 cm). The careful examination of data obtained shows wider range of GLS placement across the canes at Demandice as at Sazdice, even if both localities were situated close to each other (5 km distance) and have similar weather conditions (Table 1). From an ecological perspective, it appears reasonable that cane diameter would be the most important variable affecting the within cane distribution of GLS. The general condition, nutritional quality and protective value of canes are probably different within cultivated (Demandice) and neglected (Sazdice) crop. Wood-boring insects require bark thick enough to provide 82

adequate nutrition (Hanks et al. 1993), protection from desiccation, extreme temperatures (Acker & Nielsen 1986) and natural enemies (Wermelinger 2002). Another explanation for different within-cane distributions of GLS as well as rate and intensity of raspberry infestation could be related to the microclimate of the localities. Demandice was partially shaded by acacia forest, while Sazdice was a sunny site situated in more open hilly country. Although A. cuprescens was observed at both localities, a distinct preference for the sunny site was observed. This idea confirms previous descriptions of Agrilus species distributions, e.g. A. liragus, A. anxius (Barter 1965) and A. planipennis prefer to be in sunlit areas and are more active on sunny days (Yu 1992). Besides, Reichart (1990) stated that a later starting and longer lasting period of RSG adult emergence and afterwards also egg lying in the case of colder weather or microclimate. Suitable larval feeding positions are in the higher section of canes in these situations. Microclimate could play as a very important feature of pest’s behaviour and GLS distribution. Height-above-ground within-cane distribution of the GLS plays an important role in fruit losses and its value. High densities of GLS at lower part of canes can highly increase fruit losses and in the case of a very low position of attack the whole cane dies off. On the contrary, a higher distribution results in more moderate losses (Kuroli 1999). The majority of GLS were found in the lower sections of canes between 0 and 60 cm height-above-ground at both investigated localities (Figure 1 and 2). Reichart (1990) recorded a similar distribution. The intensity of raspberry cane infestation recorded ranged from 1-2 GLS per infested cane at Demandice to 1-4 GLS at Sazdice (Table 2). It was evident that raspberry plants suffered from heavier RSG attack at neglected locality Sazdice, which was documented also by higher maximal number of GLS per cane (8) despite Demandice (4). Kuroli (1999) reported generally one GLS within infested raspberry cane, and only rarely 2-3, very seldom 4 GLS were found. The occurrence of one GLS per cane is rare on blackberries, where usually 2-9 GLS were noted (Kuroli 1999) and 2-3 larvae of RSG per infested cane are typical within Rosa spp. canes (Armine & Stasny 1993). These species, growing often as wild plants around raspberry plantations, could be very important sources of annual infestation even within well- kept plantations. It is evident that recorded values of the rate of raspberry plantation infestation were similar within single localities annually, but differed greatly between localities (Table 2). The rate of infestation ranged from 8.5% in 2006 to 23.9% in 2005 at cultivated locality Demandice. Reichart (1990) reported similar (5-24%) infestation. At the neglected plantation the rate of infestation could reach 90% (Stamenkovič et al. 1996). This result was confirmed also by our data, when the rate of raspberry infestation reached 81.5% at neglected plantations at Sazdice in 2005 (Table 2). Our results show that the neglected plantations could be heavily attacked by RSG. On the other hand, the treated localities, where infested canes are annually removed, are only slightly damaged. Destruction of infested canes is the simplest remedy but, to be effective, wild host plants in the vicinity should be eliminated too. Chemical treatment against RSG is useless in this way. To reach low cane infestation, it is essential to remove and burn or compost infested canes because A. cuprescens is able to successfully complete its development also within cut canes (Vétek & Pénzes 2005). Contrary to the raspberry protection against the pest, higher rate of infestation is needed for biological control of invasive Rosa and Rubus species by A. cuprescens in the areas where these species are spreading. Zwolfer (1965) reported that relatively small number of larvae were able to severely damage or to kill the vegetative system of Rosa rubiginosa in New Zealand and Armine & Stasny (1993) stated that from 12 to 20% of R. multiflora canes were infested by the RSG at several sites of Ohio, Indiana and West Virginia (USA). The RSG is a non-indigenous species in North America and has been established throughout its eastern part. 83

Because of relatively low incidence and high parasitism Armine (2002) noted only the minor importance of the RSG for biological control agent of multiflora rose. Due to this affirmation our survey shows that despite the high parasitism, the rate of infestation could reach high values in raspberry plantations, so our study could provide an interesting starting point for future studies relating to biological control by the RSG.

Acknowledgements

Part of this work was supported by the Grant Agency VEGA, project No 2/7026/27 and No 1/2424/05

References

Akers, R.C. & Nielsen, D.G. 1986: Influence of post-felling treatment of birch logs on emergence success of bronze birch borer, Agrilus anxius, adults (Coleoptera: Buprestidae). – J. Entomol. Sci. 21: 63-66. Amrine, Jr., J.W. 2002: Multiflora Rose. – In: Biological Control of Invasive Plants in the Eastern United States, eds. Van Driesche, Blossey, Hoddle, Lyon and Reardon: 265-292. Amrine, Jr., J.W. & Stasny, T.A. 1993: Biological control of multiflora rose. In: Biological Pollution, ed. McKnight: 9-21. Balás, G. 1966: Kertészeti növények állati kártevői. – Mezőgazdasági Kiadó, Budapest, Hungary. Barter, G.W. 1965: Survival and development of the bronze poplar borer Agrilus liragus Barter & Brown (Coleoptera: Buprestidae). – Can. Entomol. 97: 1063-1068. Bílý, S. 1982: The Buprestidae (Coleoptera) of Fennoscandia and Denmark. – Fauna Ent. Scandinavica 10: 109. Bright, D.E. 1987: The Metallic Wood-Boring Beetles of Canada and Alaska. Coleoptera: Buprestidae. – Research Branch, Agriculture Canada, Canada. Brussino, G. & Scaramozzino, P.L. 1982: The presence of Agrilus aurichalceus Redt. (Coleoptera: Buprestidae) on raspberry in Piedmont. – Informatore Fitopatologico 32: 55-58. Fischer-Colbrie, P., Blumel, S. & Hausdorf, H. 1987: The raspberry buprestid beetle - a new pest of raspberry crops in Austria? – Pflanzenschutz 3: 8-9. Haack, R.A. & Benjamin, D.A. 1982: The biology and ecology of the twolined chestnut borer, Agrilus bilineatus (Coleoptera: Buprestidae), on oaks, Quercus spp., in Wisconsin. – Can. Entomol. 114: 385-396. Hanks, L.M., Paine, T.D. & Millar, J.G. 1993: Host species preference and larval performance in the wood-boring beetle, Phoracantha semipunctata F. – Oecologia 95: 22-29. Kuroli, G. 1996: Károsít a málna karcsúdíszbogár. – Növényvédelmi Tanácsok 5: 4-5. Kuroli, G. 1999: A málna karcsúdíszbogár (Agrilus aurichalceus Redtenbacher). – Növény- védelem 35: 257-263. Lekic, M. 1966: A contribution to the study of the ecology of the raspberry cane borer A. aurichalceus. – Arh. Poljopr. Nauke 19: 135-145. Reichart, G. 1968: Málna karcsúdíszbogár. – In: Bogyósgyümölcsűek növényvédelme, ed. Csorba: 15-16. Reichart, G. 1990: Málna karcsúdíszbogár (Agrilus aurichalceus Redt.). – In: A növényvédelmi állattan kézikönyve 3/A, eds. Jermy and Balázs: 82-84. Schaefer, L. 1949: Les Buprestides de France. – Editions Scientifiques E. Le Moult, Paris, 511 pp. 84

Stamenkovič, S., Pestič M. & Milenkovič, S. 1996: Stetocine maline i kupine. – Bilj. lekar 24: 136-150. Tsolova, E. 2002: Damage caused by Agrilus aurichalceus Redt. (Coleoptera: Buprestidae) in raspberry. – Rasteniev' dni Nauki 39: 231-234. Vétek, G. & Pénzes, B. 2005: New data about the damage and life cycle of Rose stem girdler (Agrilus cuprescens Ménetriés). – In: 5th International Conference of PhD Students. University of Miskolc, Hungary, 14-20 August 2005 Wermelinger, B. 2002: Development and distribution of predators and parasitoids during two consecutive years of an Ips typographus (Col., Scolytidae) infestation. – J. Appl. Entomol. 126: 521-527. Yu, C. 1992: Agrilus marcopoli Obenberger (Coleoptera: Buprestidae). – In: Forest Insects of China: 400-401. Zaťko, M. 2002: Primary landscape structure. – In: Atlas krajiny Slovenskej republiky: 344 pp. Zwölfer, H. 1965: Coleoptera attacking Rosa in Europe. – In: Rep. Commonw. Inst. Biol. Control, Rosa rubiginosa Project, 6 pp.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 85-87

Post harvest control of the eriophyoid mite Phyllocoptes gracilis on raspberries

Christian Linder1, Catherine Baroffio2, Charly Mittaz2 1 Agroscope Changins-Wädenswil ACW, CP 1012, 1260 Nyon 1, Switzerland; 2 Agroscope Changins-Wädenswil ACW Centre Les Fougères, 1964 Conthey, Switzerland

Abstract: The eriophyoid mite Phyllocoptes gracilis (Nalepa) is an important secondary pest species in Swiss raspberry production. The pending withdrawal of the only registered miticide bromopropylate has created the need to develop alternative control strategies against this mite. In autumn 2006, first efficacy trials were carried out in a heavily infested plot of the variety Glen Ample in Nendaz (Switzerland). Single, post harvest applications of wettable sulphur and of the miticide spirodiclofen were both very effective and provided excellent pest control. Visual inspection the following spring confirmed the value of the control strategies tested. The costs of post harvest treatments with wettable sulphur are lower than bromopropylate applications and the timing of application does not pose a risk of pesticide residues. This study supports a quick registration of the tested products against P. gracilis.

Key words: Phyllocoptes gracilis, raspberries, chemical control, post harvest treatments, sulphur, spirodiclofen

Introduction

The leaf and bud mite, Phyllocoptes gracilis (Nalepa) is a common and important secondary pest species in Swiss raspberry production (Linder et al. 1998; Baroffio 2007). Damages on susceptible varieties (e.g. Glen Ample) are often spectacular and plant growth and fruit quality may be reduced after several years of mite attack (De Lillo & Duso 1996). Although phytoseiid mites feed on P. gracilis, they are not able to prevent pest outbreaks. Currently, Swiss growers control the pest by the use of bromopropylate, the only miticide registered. Even though its application is effective, it is limited to the pre-blossom period to avoid pesticide residues on fruits. However, its pending withdrawal from the Swiss pesticide market has created the need for a quick development of alternative control strategies against the mite. An interesting alternative might be the post harvest use of the miticide spiridoclofen or wettable sulphur. A great advantage of post harvest pest control is the prevention of pesticide residues on fruits. It has been successfully applied against the pear leaf blister mite (Epitrimerus pyri) (Daniel et al. 2004). A field trial was launched to determine the suitability of post harvest mite control in raspberry.

Material and methods

In September 2006, the impact of post-harvest application of spiridoclofen and wettable sulphur on P. gracilis population was tested in a raspberry field (Glen Ample) in Nendaz (Valais, Switzerland). The four variants of the trial are listed in table 1. Each treatment was applied to two rows (± 20 meters) of heavily infested plants on the 6th of September. An “Echo SHR” knapsack sprayer was used to apply a water volume of 1000 l/ha.

85 86

Table 1. Variants tested in the Nendaz post harvest trial 2006.

Products Active ingredients Concentrations Control - - Envidor spirodiclofen (22.3%) 0.04% Thiovit wettable sulphur (80%) 1% Thiovit wettable sulphur (80%) 2%

Leaf samples to assess mite density were taken 2 days before as well as 15 and 170 days after application. Mite density was examined on 5 x 10 foliar disks of 4.5 cm2 randomly removed from 5 x 10 terminal leaflets per variant. Foliar disks were agitated for 20 minutes in a solution of water + 0.1% wetting agent (Teepol). After removal of the leaves, the solution was vacuum-filtered on 90 mm ∅ black paper filters (Schleicher & Schuell). Individuals were then counted under a dissecting microscope. Densities are expressed in mites/cm2. The efficacy of miticides was calculated by the Henderson-Tilton formula.

Results and discussion

The three variants showed all an excellent efficacy against P. gracilis 15 days after treatment (Figure 1). Wettable sulphur showed a slight dose-effect, but the gain of efficacy by the higher sulphur concentration was minor.

14 T-2 days T+15 days 12 T+170 days 2 10 / cm 8

6 P. gracilis 4

2 97.9 92.3 99 98 97.4 100 0 control spirodiclofen sulphur 1% sulphur 2% Variants & Henderson-Tilton efficacy

Figure 1. Mean densities of P. gracilis ± SD and Henderson-Tilton efficacy in the Nendaz post harvest trial 2006.

In early Spring 2007 (T+170 days), mite densities on treated plants were still lower than in the untreated control. Unfortunately a pre-blossom miticide treatment against Tetranychus urticae made by the grower prevented the made of additional surveys. Nonetheless these first results are promising in many regards: 87

- Major reduction of P. gracilis' overwintering population, which probably leads to a reduced need of spring applications; - No risk of pesticide residues due to the timing of the application; - More favourable weather conditions in late summer than in spring increase the efficacy of miticides; - Despite the high doses of sulphur applied, no phytotoxicity could be observed on Glen Ample. This observation has to be confirmed on other raspberry varieties; - An economically very interesting strategy for growers; the costs of a 1% wettable sulphur treatment are 25% lower than a bromopropylate and 52% lower than a spirodiclofen application, respectively; - The post harvest application of spirodiclofen may also reduce Tetranychus urticae and consequently its pre-harvest control attempts (Mariéthoz et al. 1994). Long-term efficacy trials should complete this study to support a quick registration of the tested products against P. gracilis.

References

Baroffio, C. 2007: Culture des framboises. Maladies et ravageurs. – In: Guide des petits fruits 2007. Ed. Fruit Union Suisse, 6302 Zoug. 125 pp. Daniel, C., Wyss, E. & Linder, Ch. 2004: Application de soufre en automne: une nouvelle manière de lutter contre l’ériophyide à galles du poirier. – Revue suisse de Vitic. Arboric. Hortic. 36 (4): 199-203. De Lillo, E. & Duso, C. 1996: Damages and Control of Eriophyoid Mites in Crops : Currants and Berries. – In: Eriophyoid mites. Their Biology, Natural Enemies and Control. Lindquist E.E., Sabelis M.W., Bruin J., eds., Elsevier, Amsterdam. 790 pp. Linder, Ch., Antonin, P., Mittaz, Ch. & Terrettaz, R. 1998: Ravageurs du framboisier. – Revue suisse de Vitic. Arboric. Hortic. 30 (2): 127-129. Mariéthoz, J., Baillod, M., Linder, Ch., Antonin, Ph. & Mittaz, Ch. 1994: Distribution, méthodes de contrôle et stratégies de lutte chimique et biologique contre l’acarien jaune, Tetranychus urticae Koch, dans les cultures de framboisiers. – Revue suisse de Vitic. Arboric. Hortic. 26 (5): 315-321. 88 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 89-92

Biological control of two-spotted spider mite with Phytoseiulus persimilis in everbearer strawberry

Catherine A. Baroffio, Christian Linder Agroscope Changins – Wädenswil ACW, Centre de recherche de Conthey, 1964 Conthey, Switzerland; Agroscope Changins – Wädenswil ACW, Centre de recherche de Nyon, 1260 Nyon, Switzerland

Abstract: Biological control of two-spotted spider mite Tetranychus urticae is an interesting technique in everbearer strawberry production where long harvesting periods prevent a classical acaricide treatment schedule. During a three year study on soilless strawberries, Phytoseiulus persimilis gave satisfactory control of T. urticae under certain conditions. Predators had to be released as soon as the leaf occupation rate by the pest reached 20% to 30%. Above this threshold, it was necessary to apply a soap treatment before the P. persimilis releases. With 10 P. persimilis /m2, it took two weeks until the pest populations started to decrease. T. urticae was usually controlled after four weeks. This strategy did not affect fruit quality and the costs remained affordable for the growers.

Key words: biological control, strawberry, two-spotted spidermite, Tetranychus urticae, Phytoseiulus persimilis

Introduction

In Switzerland, the two-spotted spider mite (TSSM) Tetranychus urticae Koch is an important pest in strawberry production (Antonin et al. 1997; Collectif 2007). In early season crops, Linder et al. (2003) demonstrated that a relatively high attack level (up to 120 TSSM/leaf) had only a low impact on yield, even with noticeable leaf damage. This ability to tolerate some damage allows the use of biological control. Although widely used, this method is not always successful because many factors can affect its achievement: choice and density of predatory mites, time of release, climate, side-effects of chemical treatments, costs. Never-the-less, in everbearer strawberries biological control is the only alternative to chemical treatment which cannot be used during the long harvest period. Because of its rapidity of action, its specialisation on TSSM and its cost, the predatory mite Phytoseiulus persimilis (PP) Athias- Henriot is often used. With the increasing development of soilless everbearer strawberries cultures in Switzerland (Ançay & Carlen 2006) it was of particular interest to test biological control under practical conditions. Results of a three years study are briefly described.

Material and methods

All the trials have been conducted between 2003 and 2005 at the ACW Center of Conthey (Valais - Switzerland) and in a nearby commercial plot (Table 1). TSSM and PP populations were regularly estimated by examining 25 terminal leaflets per tunnel. The densities, the percentage of occupied leaves and the TSSM load expressed in TSSM-day/leaf were calculated according to the method described by Baroffio et al. (2007).

89 90

Table 1. Description of the trials made in different strawberry crops in Valais between 2003 and 2005.

Phytoseiulus Acaricides Trial Year Place Shelter Variety persimilis Release Density Date Treatments Mara des Tunnel 07.05 10/m2 A 2003 Conthey Bois, - - (200 m2) 16.05 10/m2 Elsinore Greenhouse 25.05 10 m2 B 2004 Conthey Elsegarde - - (70 m2) 26.08 10/m2 Tunnel various C 2004 Conthey 30.07 10/m2 28.07 soap: 20 l/ha (200 m2) varieties Greenhouse D 2005 Conthey Elsinore 14.06 10/m2 - - (70 m2) Tunnel Charlotte, E 2005 Conthey 10.06 10/m2 - - (200 m2) Elsinore 12.07 3/m2 Tunnels 20.05 spirodiclofen:0,4l/ha F 2005 Ardon Elsinore 20.07 6/m2 (1500 m2) 15.07 soap: 20 l/ha 26.07 6/m2

Results and discussion

PP alone or combined with acaricide treatments gave good control of TSSM (Figure 1 and table 2). In trial A, the predatory mite was released twice with a one-week interval on high TSSM populations (90% of occupied leaves). Trial B showed that with the same pest occupation rate, a single PP-release was not sufficient to cover the whole season. Later in the season, a second release was not able to control the TSSM population. Trial C showed that with high pest occupation rates, a soap treatment combined with a PP release was effective and as cheap as the former strategy (B). The best results were obtained in trials D and E, with single PP-releases made in mid-June and when the TSSM occupation rates were close to 30% (or 3 mobile forms/leaf). At 10 PP/m2 (2 PP/plant), the costs reached 1850 sFr/ha (-50% compared to A and B and –14% compared to C). In trial F, a classical acaricide treatment before blossom did not enable good subsequent biocontrol of the pest. Later in the season a combined soap treatment and PP release strategy was more successful but at a slightly higher cost. The cumulated number of TSSM varied between 100 and 4900 (TSSM-day/leaf) at the end of the season (Table 2). During the main yield periods all the trials showed a cumulated TSSM number lower than 2500 TSSM-day/leaf. The quality and quantity of the yield was satisfactory in all the trials. These results obtained in soilless cultures and with different varieties confirm the results of Oatman et al. (1982) and Linder et al. (2003) who did not notice any difference in strawberry yield with TSSM loads varying between 168 and 3800 TSSM-day/leaf. In our conditions no additional treatment against other pests was applied. This played a very important part in the achieved success.

100 100 A D 80 80 TSSM

60 PP 60

40 40

20 20

0 0

9 2 5 0 6 100 16 1 2 2 28 31 34 37 4 1001 19 22 25 28 31 34 37 40 B E 80 80

60 60

40 40

20 20

0 0

100 8 1 100 9 2 5 % occupied leaves by TSSM and PP 16 19 22 25 2 3 34 37 40 16 1 2 2 28 31 34 37 40 C F 80 80

60 60

40 40

20 20

0 0 may june july aug. sept. oct. may june july aug. sept. oct. 1 4 7 0 8 1 16 19 22 25 28 3 3 3 4 16 19 22 25 2 3 34 37 40

Figure 1. Percentage of leaves occupied by T. urticae (TSSM) and P. persimilis (PP) in the various trials (A to F). P. persimilis release; Acaricide treament.

Conclusions

P. persimilis is an efficient control agent of TSSM in soilless everbearer strawberries grown in tunnel or glasshouse. The predators must be ordered as soon as the TSSM occupation rate reaches 10% and released when the occupation rates do not exceed 20 to 30%. P. persimilis must be released immediately after reception (1x10 or 2x5 PP/m2 within a one-week interval). After the release, the TSSM populations still increase for two or three weeks. Regulation is reached after 4 weeks. If the T. urticae populations do not decrease after 3 weeks, it is recommended to make a new release. When TSSM populations exceed a threshold of 30% of occupied leaves, it is recommended that soap treatment is applied first and then P. persimilis should be released a week later at a the density of 10 PP/m2. Such a strategy does not affect yield or fruit quality and its cost is affordable for growers. 92

Table 2. Results of biological control of TSSM with PP on soil-less strawberry under glasshouse or tunnel.

% TSSM TSSM/leaf Cumulated Trial occupation Cost (sFr/ha) when PP-release TSSM Load when PP-release

92 4.2 A 1134 3700.- 92 27.9 88 14.6 B 4919 3700.- 40 3.7 C 100 105.2 2414 2144.- D 30 3 100 1850.- E 32 3.2 549 1850.- 96 38.4 F 98 14.1 1396 2255.- 100 43.5

Acknowledgements

The authors would like to thank E. Daures, P. Pierre, C. Mittaz, A. Ancay, Ch. Auderset and B. Sauthier for their precious help in the different trials.

References

Ançay, A. & Carlen, Ch. 2006: Fraisiers remontants sur substrat: comparaison de nouvelles variétés et de deux densités de plantation. – Revue suisse Vitic. Arboric. Hortic. 38 (2), 129-134. Antonin, Ph., Baillod, M., Linder, Ch. & Mittaz, Ch. 1997: Problématique de la lutte chimique contre l’acarien jaune commun, Tetranychus urticae Koch, en cultures de fraisiers. – Revue suisse Vitic. Arboric. Hortic. 29 (3): 179-187. Baroffio, C., Carlen, C., Mittaz, C. & Linder, Ch. 2007: Succès de la lutte biologique contre les acariens jaunes dans les fraisiers remontants à l'aide de Phytoseiukus persimilis. – Revue suisse Vitic. Arboric. Hortic. 39 (2): 117-121. Collectif 2007: Guide des petits fruits. – FUS, 6302 Zoug, 125 pp. Linder, Ch., Carlen, Ch. & Mittaz, Ch. 2003: Nuisibilité de l’acarien jaune Tetranychus urticae Koch et stratégies de lutte dans les cultures de fraisiers précoces. – Revue suisse Vitic. Arboric. Hortic. 35 (4): 235-240. Oatman, E.R., Sances, F.V.,Lapré, L.F., Toscano, N.C. & Voth, V. 1982: Effects of different infestation levels of the twospotted spider mite on strawberry yield in winter plantings in Southern California. – J. Econ. Ent. 75: 94-96. Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 93-96

Does the presence of multiple phytoseiid species affect biocontrol of Tetranychus urticae on strawberry?

Jean Fitzgerald East Malling Research, East Malling, Kent ME19 6BJ, UK

Abstract: Experiments were undertaken on strawberry leaf arenas and on potted strawberry plants held at 20ºC, and in field plantings of strawberries to determine if the presence of multiple species of phytoseiid mites affected biocontrol of the target pest species, Tetranychus urticae. In leaf arena and potted plant experiments conducted over 10 and 5 days respectively, no interactions between predators were seen. However, in field experiments conducted over a longer period there was a reduction in control of T. urticae when both Phytoseiulus persimilis and Neoseiulus cucumeris were present on the plants.

Key words: Neoseiulus californicus, Neoseiulus cucumeris, Phytoseiulus persimilis, Typhlodromus pyri, interactions, biocontrol

Introduction

The spider mite Tetranychus urticae is a serious pest on strawberry. Phytoseiid mites are the main predators of this pest. The predatory mites that colonise naturally in SE England are Neoseiulus cucumeris, N. californicus and Typhlodromus pyri; N. californicus is not a native UK species, but has been released in the past and has been shown to overwinter successfully (Jolly 2001; Fitzgerald unpublished results). Typhlodromus pyri colonises slowly and is unlikely to me an important biocontrol agent for T. urticae in short rotation strawberry plantations. Predatory mite species that can be artificially released in open field plantations in the UK are Phytoseiulus persimilis and N. cucumeris. Earlier work has shown that P. persimilis, N. cucumeris and N. californicus can reduce T. urticae numbers on strawberry (Easterbrook 1992; Easterbrook et al. 2001; Fitzgerald & Easterbrook, 2003). The overall aim of this research was to assess possible interactions among the species present and to provide the biological information needed to develop optimised mite pest management strategies on strawberry.

Methods

Leaf arena experiments Pairwise combinations of the predator species (P. persimilis, N. californicus, N. cucumeris and T. pyri) and a no-predator control were used to assess interactions when T. urticae was present on leaf disc arenas in a constant temperature (CT) room at 20ºC with a 16L:8D photoperiod. Ten adult female T. urticae were placed on each leaf disc and left for 48 hours to lay eggs, after which time predators were released. The adult T. urticae were left on the discs and continued to lay eggs throughout the experiment. Numbers of T. urticae were recorded at 24, 48 and then every 48 hours after predator release until 10 days after release (DAR). Data were log transformed and compared by repeated measures ANOVA. A regression analysis was also performed using indicator variables, with the presence/absence of each predator given by 1/0 and a 2 indicating a pair of the same species.

93 94

Potted plant experiments Potted strawberry plants were reduced to two leaflets and two flower/fruit clusters plus a developing unexpanded trifoliate leaf, and were infested with 8 adult female T. urticae. After one week, numbers of all stages of T. urticae on the leaflets were recorded under a stereomicroscope, and all the adults were removed. Phytoseiids were added to the plants in all pairwise combinations of species as above with four predators released on each plant. Experiments were done in a CT room as described above. Numbers of T. urticae were recorded 5 days after predator release on all plant parts. Data were compared by ANOVA and by regression analysis as described above.

Field experiments Tetranychus urticae were released and allowed to establish on field grown strawberry plants. Predator treatments were releases of 20 P. persimilis, 20 N. cucumeris, or 10 P. persimilis plus 10 N. cucumeris per plant with a no predator release control. The experiment was a randomised complete blocks design. Leaf samples were taken 11, 46 and 64 DAR and numbers of mites on the leaves counted. Data were compared by repeated measures ANOVA and by regression analysis as above.

Results

Leaf arena experiments By 6 DAR, T. urticae active stages were significantly lower than the control in all treatments except those including T. pyri (Table 1), and by 10 DAR all treatments, except where T. pyri alone had been released, showed a significant reduction compared to the control. A similar result was obtained for T. urticae eggs. Using the linear regression analysis for T. urticae active stages and for eggs 10 DAR, the main effects of each predator, after adjusting for the other three, were significant for P. persimilis, N. californicus and N. cucumeris (P<0.05), but not for T. pyri (P = 0.195 for active stages). None of the 2-way interactions were statistically significant, thus there was no indication of any interactions between predator species in these arena experiments.

Potted plant experiments The main effects for P. persimilis and N. californicus were statistically significant for both T. urticae eggs and active stages (P<0.01 for both species for both variables) (Table 1), but not for the other phytoseiid species. However, releases of P. persimilis and N. californicus individually were at least as effective as releases of both species together, particularly for T. urticae active stages.

Field experiments Repeated measures ANOVA showed there was a significant reduction of T. urticae eggs where P. persimilis were released (P<0.01), but not in the N. cucumeris release treatment (P = 0.073). There was a significant reduction in numbers of T. urticae active stages in both the P. persimilis and N. cucumeris treatments (P<0.01; P<0.05 respectively), and also a significant interaction between predators (P<0.05); no predator interaction was apparent when numbers of T. urticae eggs were analysed (P = 0.131).

Table 1. Mean numbers of T. urticae active stages in leaf arena and potted plant experiments: Data are log10(n+1) transformed with back-transformed counts in parentheses. Control = no predator release. For the potted plant experiment pre-release numbers were used as a co- variate.

Treatment Leaf arena experiment Potted plant experiment 6 DAR 5 DAR Control 2.046 (110.2) 1.990 (96.7) Tp/Tp 1.738 (53.7) 1.753 (55.6) Tp/Ncal 1.725 (52.1) 1.354 (21.6) Tp/Ncuc 1.110 (11.9) 1.484 (29.5) Tp/Pp 1.461 (27.9) 1.449 (23.1) Pp/Pp 0.690 (3.9) 1.155 (13.3) Pp/Ncal 1.279 (18.0) 1.184 (14.3) Pp/Ncuc 1.013 (9.3) 1.667 (45.5) Ncal/Ncal 1.164 (13.6) 1.267 (17.5) Ncuc/Ncuc 1.082 (11.1) 1.836 (67.5) Ncal/Ncuc 1.390 (23.5) 1.432 (26.0) SED (~119 df) 0.3364 SED (59 df) 0.214 P=0.05 P<0.001

Discussion

Work at EMR has shown that P. persimilis, a specialist predator of tetranychid mites, is an effective biocontrol agent for T. urticae on strawberry (Easterbrook 1992). The other phytoseiids tested in our experiments are more generalist predators (e.g. McMurtry & Croft 1997), but Easterbrook et al. (2001) showed that N. cucumeris was able to reduce populations of T. urticae in strawberry. Research has shown that pest control may not be enhanced by releasing additional predator species to existing predator-prey systems (e.g. Rosenheim 1998; Sih et al. 1985). This reduction in biocontrol of pests when multiple species of predators are present may be a result of intraguild predation (Rosenheim et al. 1995), where a predator consumes a developmental stage of another predator species. Our experiments were undertaken to assess the effects of the presence of multiple predators on biocontrol of T. urticae in strawberry. In leaf disc arena experiments P. persimilis was the most effective species in reducing T. urticae numbers. Neoseiulus californicus and N. cucumeris were more effective than T. pyri in reducing numbers of T. urticae active stages, and P. persimilis and N. cucumeris gave the greatest reductions in T. urticae egg numbers. On potted plants, P. persimilis and N. californicus were again the most effective predators tested at reducing numbers of T. urticae. No interactions among predatory mite species were detected in our laboratory experiments; no detectable interaction among phytoseiid predators was seen in similar potted plant experiments where two pest species were present (T. urticae and the tarsonemid mite Phytonemus pallidus ssp. fragariae) (Fitzgerald et al. 2007). In the field experiments numbers of T. urticae were significantly reduced where releases of either P. persimilis or N. cucumeris had been made. However, there was also an interaction between these predator species such that there was a reduced level of biocontrol of T. urticae where both predators had been released. This negative interaction between N. cucumeris and P. persimilis demonstrated in these field experiments may in part be due to intraguild 96

predation (e.g. Yao & Chant 1989; Zhang & Croft 1995). Our laboratory experiments may have been conducted over too short a period to detect these effects. These results show the importance of conducting field tests to confirm the biocontrol efficacy of predator species and to uncover any possible interactions among potential predators. In the field experiments, but not in laboratory experiments, an interaction was seen between P. persimilis and N. cucumeris such that control of T. urticae was less effective when both predator species were released together.

Acknowledgements

This work was funded by the UK Department for Environment, Food and Rural Affairs (Defra). I thank Gillian Arnold for statistical analysis, and Nicola Pepper, Mike Easterbrook and Tom Pope for technical assistance.

References

Easterbrook, M.A. 1992: The possibilities for control of two-spotted spider mite Tetranychus urticae on field-grown strawberries in the UK by predatory mites. – Biocontrol Sci. Techn. 2: 235-245. Easterbrook, M.A., Fitzgerald, J.D. & Solomon, M.G. 2001: Biological control of strawberry tarsonemid mite Phytonemus pallidus and two-spotted spider mite Tetranychus urticae on strawberry in the UK using species of Neoseiulus (Amblyseius) (Acari: Phytoseiidae). – Exp. Appl. Acarol. 25: 25-36. Fitzgerald, J.D. & Easterbrook, M.A. 2003: Phytoseiids for control of spider mite, Tetranychus urticae and tarsonemid mite, Phytonemus pallidus, on strawberry in UK. – IOBC/WPRS Bull. 26 (2): 107-111. Fitzgerald, J.D., Pepper, N., Easterbrook, M.A., Pope, T. & Solomon, M.G. 2007: Interactions among phytophagous mites, and introduced and naturally occurring predatory mites, on strawberry in the UK. – Exp. Appl. Acarol. 43: 33-47. Jolly, R.L. 2001: The status of the predatory mite Neoseiulus californicus (McGregor) (Acari: Phytoseiidae) in the UK, and its potential as a biocontrol agent of Panonychus ulmi (Koch) (Acari:Tetranychidae). – PhD thesis. University of Birmingham. 173 pp. McMurtry, J.A. & Croft, B.A. 1997: Life-styles of phytoseiid mites and their role in biological control. – Ann. Rev. Entomol. 42: 291-321. Rosenheim, J.A. 1998: Higher-order predators and the regulation of insect herbivore populations. – Ann. Rev. Entomol. 43: 421-447. Rosenheim, J.A., Kaya, H.K., Ehler, L.E., Marois, J.J. & Jaffee, B.A. 1995: Intraguild predation among biological control agents - theory and evidence. – Biol. Contr. 5: 303- 335. Sih, A., Crowley, P., McPeek, M., Petranka, J. & Strohmeier, K. 1985: Predation, competition, and prey communities: a review of field experiments. – Ann. Rev. Ecol. Syst. 16: 269-311. Yao, D.S. & Chant, D.A. 1989: Population growth and predation interference between two species of predatory phytoseiid mites (Acarina:Phytoseiidae) in interactive systems. – Oecologia 80: 443-455. Zhang, Z.-G. & Croft, B.A. 1995: Interspecific competition and predation between immature Amblyseius fallacis, Amblyseius andersoni, Typhlodromus occidentalis and Typhlo- dromus pyri (Acari: Phytoseiidae). – Exp. Appl. Acarol. 19: 247-257.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 97-100

Field control of strawberry mite Phytonemus pallidus

Bruno Gobin, Eva Bangels pcfruit Zoology department; Fruittuinweg 1, B-3800 Sint-Truiden, Belgium

Abstract: Belgian soft fruit growing is moving towards full integrated pest management, although it has not yet reached the advanced level of IPM as hard fruit. Only few acaricides are registered for use in strawberry in Belgium. Although alternate use of these products is generally sufficient for spider mite control, achieving tarsonemid mite control is far more difficult. This paper presents the results of a recent trial on the biological efficacy of various active ingredients and their formulations to tarsonemid mites. Only partial control can be achieved with currently available acaricides, although at low infestation this might be sufficient to limit commercial losses. Growers must use strict hygienic measures in their fields and use clean plant material to minimize the danger of spreading this pest and selective crop protection to allow build-up of predatory mite populations to assist control of Phytonemus pallidus.

Key words: Tarsonemus, biological efficacy, acaricide, Fragaria

Introduction

Strawberry mite Phytonemus (Steneotarsonemus) pallidus ssp. fragariae (Zimmerman) is a persistent pest of both open air and covered strawberry crops. Adult and juvenile stages feed on plant juices in the youngest leaves, making them less efficient for photosynthesis once unfolded. Affected plants are weakened and produce less fruit that is of poor quality due to deformations. This can result in considerable yield losses (Alford 1972). Strawberry mite control is difficult for two reasons. Firstly, the number of acaricides approved for use against this pest is small in most European countries. The severity of Phytonemus problems increased progressively in those countries abandoning the use of endosulfan following an European level withdrawal decision. Secondly, the mite resides mainly in the folded heart leaves of strawberry plants, making it difficult to reach with pesticide applications. Furthermore, the pest symptoms are often only noticed just before harvest, when both bee activity and essential pre-harvest intervals limit the choice of compounds. For this reason, recent research focused on (1) the potential for biocontrol of this pest by means of predatory phytoseiid mites (Easterbrook et al. 2001) and (2) keeping fields Phytonemus-free by means of pre-treating plant material (Klinger 1969). Both strategies proved to be successful at reducing tarsonemid populations (Croft et al. 1998; Hellqvist 2002; Easterbrook et al. 2003), but often failed short of eliminating the pest completely. Phytoseiid mites indeed feed only on mobile Phytonemus, not on eggs, and the larger predatory mite stages cannot reach more cryptic microhabitats in folded strawberry leaves (Easterbrook et al. 2003). Temperatures used in hot water treatments have to balance effectiveness on tarsonemids to sensitivity of plant material; two parameters with seasonal variability depending on both pest and plant phenology (MacLachlan & Duggan 1979; Hellqvist 2002). Thus, though growers should maximally implement these biological control systems to minimize pest pressure, chemical applications will remain necessary to correct outbreaks. To successfully implement an IPM strategy it is imperative that growers use chemicals that are selective for predatory mites throughout the year. This includes acaricide correction

97 98

treatments aimed against tarsonemid mites. On the other hand, as tarsonemids are difficult to control, growers are likely to look for a good trade-off between efficacy and selectivity. In this manuscript we present data on the biological efficacy of a number of acaracides against tarsonemids.

Material and methods

We compared the biological efficacy of eight acaricides in comparison to endosulfan and water treated controls in a field trial in 2005 (Table 1). We chose a strawberry field planted with Darselect in 2004. Already in summer 2004 an important infection with Phytonemus pallidus was detected in this field. The trial was set up as a standard EPPO protocol as a randomized block design with 4 replicates per treatment. Replicates were 6.25m² surface area covering 40 strawberry plants. Spray solutions were prepared by adding test products to 2.5 l water (corresponding to 1000 l/ha) and divided in four equal parts to obtain homogeneous application of active ingredient on each replicate. Each replicate was thus treated separately with 0.625 l of solution. All test products were sprayed at the 4th of May (A), for 1 treatment, Kanemite, a 2nd application was performed two weeks later (18th of May; B). We assessed the number of Phytonemus mites just before treatment (precount) and at 5 DAT A, 13 DAT A, 30 DAT A (16 DAT B). At these dates, we collected 10 heartleaves per plot and counted the number of mobile mite stages on each leaf. For this purpose, heartleaves were unfolded under a binocular microscope and examined at 40 x magnification. We performed overall statistical analysis (ANOVA) with a Duncan posthoc analysis on the counts of mobile stages per plot. Biological efficacy of the tested acaricides was determined using the Henderson-Tilton formula, that calculates % reduction compared to water-treated control while taking into account initial differences between replicates assessed at the precount.

Table 1. The dose rate and application dates of the different test objects in a field experiment on Darselect strawberry.

Dose Commercial Dose rate Active Ingredient formulation rate application name (l/ha) (g ai/ha) water control 4 May endosulfan Endosulfan 350EC 525 1.5 4 May abamectin 1 Vertimec 18EC 22.5 1.25 4 May abamectin 2 Vertimec 18EC 45 2.5 4 May flufenoxuron Cascade 100DC 26 0.26 4 May fenpyroxymate 1 Naja 50SC 40 0.8 4 May fenpyroxymate 2 Naja 50SC 50 1 4 May tebufenpyrad Masai 20WP 100 0.5 4 May tolylfluanide Euparen M 50WG 1250 2.5 4 May spirodiclofen Envidor 240SC 96 0.4 4 May pyridaben Sanmite 20WP 100 0.5 4 May acequinocyl Kanemite 150SC 180 1.2 4 May + 18 May

Results and discussion

Strawberries were infested with a moderate to high number of tarsonemid mites. At precount (3rd of May), we counted 55.3±15.2 (mean ±SD) mobile stages per 10 heartleaves in the water treated check, and these numbers rose steadily throughout the trial: 96.8± 63.2 at 5 DATA, 89.5±34.2 at 13 DATA and 193.0±62.2 at 30 DATA. The standard deviations indicate a high variability between replicates. The increase in population is likely due to an optimal egg- laying period during the warm temperatures of the last days of April and beginning of May (maxima reaching 30°C). The first half of May was cooler (maxima around 14°C) but the latter half of May saw maxima rising above 25°C again. This is well above development thresholds for Phytonemus pallidus (Easterbrook et al. 2003).The different test products showed different patterns of biological efficacy in this trial (Figure 1).

100,0 5 DATA 13 DATA 30 DATA 90,0 80,0 70,0 cd 60,0 d abc bcd 50,0 bcd bcd bcd bcd 40,0 abcd abc c a bc 30,0 bc

% Henderson-Tilton % abc 20,0 abc abc abc ab abc 10,0 a a 0,0 18EC 23g 18EC 45g Abamectin Abamectin Pyridaben Endosulfan 20WP 100g 240SC 96g Acequinocyl 100DC 26g 100DC 50SC 40g 50SC 50g 150SC 180g 350EC 525g 20WP 100g Tolylfluanide Flufenoxuron Spirodiclofen 50WG 1250g Tebufenpyrad Fenpyroximate Fenpyroximate Figure 1. The biological efficacy of test products at different dates after treatment, represented as mean % reduction according to the Henderson-Tilton formula. Error bars show standard deviations. Significant differences were found at 13 DAT A (ANOVA SS=42216; df=11; F=2.35; p=0.028; italic lettering; control=abc) and 30 DAT A (ANOVA SS= 77826; df=11; p=0.012; standard lettering,; control=ab).

Overall control levels for a single application were relatively low (below 70%), even though we used quite a large spray volume. Water quantity should be high in tarsonemid control as it improves penetration of the sprayed product to the heart leaves. Endosulfan, the old standard before it was withdrawn by EU regulators, gave just over 50% control. This is lower than the efficacy obtained in three field trials in Poland (Łabanowska 2006). Four products (abamectin, fenpyroxymate, tebufenpyrad and pyridaben) neared 70% biological efficacy, but differed in persistence of activity. The highest dose rate of abamectin appeared less efficient, but was not significantly different from the lower dose rate. Fenpyroximate performed similar to a field trial in Poland (Łabanowska 2006). This might indicate some variability in biological activity, but the high variability in this trial calls for caution in interpreting small differences. In this trial, abamectin at 23g ai/ha gave the longest lasting efficacy, followed by tebufenpyrad. Other acaricides show lower efficacies, insufficient to control the pest alone, but might act as supportive measures, when chosen in control of other 100

pests, such as fungal diseases (tolylfluanid) or spider mites (spirodiclofen). Belgium regulatory restrictions allow only a single application per mode-of-action per year. Therefore our trial setup includes single applications only. Given the relatively low maximal control achieved this way, growers will need at least two control applications at short interval to prevent population recovery. The absence of complete control by chemical treatments underlines the importance of preventive methods and biological control measures alike. Hot water treatments of plant material are essential to minimize tarsonemid mite risks during the growth season. The use of Phytoseiid-selective crop protection would allow for the establishment of predatory mites to prevent outbreaks and help eliminate any Phytonemus that survives following a corrective treatment. Prey-predator ratio’s are important for tarsonemid control (Easterbrook et al. 2001), and targeted release of predatory mites should be considered when predatory mite numbers in the crop are too low. Acknowledgements

This research was funded through IWT Agronomical Research Project nr. 04/669.

References

Alford, D.V. 1972: The effect of Tarsonemus fragariae Zimmermann (Acarina: Tarsonemidae) on strawberry yields. – Ann. Appl. Biol. 70: 13-18. Croft, B.A., Pratt, P.D., Kosela, G. & Kaufman, D. 1998: Predation, reproduction and impact of phytoseid mites (Acari: Phytoseiidae) on cyclamen mite (Acari: Tarsonemidae) on strawberry. – J. Econ. Ent. 91: 1307-1314. Easterbrook, M.A., Fitzgerald, B.C. & Solomon, M. 2001: Biological control of strawberry tarsonemid mite Phytonemus pallidus and two-spotted spider mite Tetranychus urticae on strawberry in the UK using species of Neoseiulus (Amblyseus) (Acari:Phytoseiidae). – Exp. Appl. Acarol. 25: 25-36. Easterbrook, M.A., Fitzgerald, J.D., Pinch, C., Tooley, J. & Xu, X.M. 2003: Development times and fecundity of three important arthropod pests of strawberry in the United Kingdom. – Ann. Appl. Biol. 143: 325-331. Hellqvist, S. 2002: Heat tolerance of strawberry tarsonemid mite Phytonemus pallidus. – Ann. Appl. Biol. 141: 67-71. Klinger, J. 1969: Die Wirkung der Warmwasserbehandlung gegen Erdbeerblattälchen und Erdbeermilben. – Schweiz. Z. Obst-Weinbau 105: 605-609. Łabanowska, B.H. 2006: Potential control agents for controlling the strawberry mite (Phytonemus pallidus ssp. fragariae Zimm.) after the withdrawal of endosulfan. – J. Fruit Ornam. Plant Res. 14: 67-72. MacLachlan, J.B. & Duggan, J.J. 1979: An effective method for hot-water treatment and propagation of strawberry runners. – Irish J. Agric. Res. 18: 301-304.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 101-106

New generation acaricides for control of two important strawberry pests: The two-spotted spider mite and the strawberry mite

Barbara H. Łabanowska Research Institute of Pomology and Floriculture, Pomologiczna 18, 96-100 Skierniewice, Poland

Abstract: The strawberry mite, Phytonemus pallidus Zimm. and the two-spotted spider mite, Tetranychus urticae Koch, are key pests in strawberry plantations in Poland. Some new acaricides including Floramite 240 SC (bifenazate) and Envidor 240 SC (spirodiclofen), as well as Omite 570 EW (propargite) and Magus 200 SC (fenazaquin) were tested for control of two major pests of strawberry: the two-spotted spider mite and the strawberry mite. These acaricides when used after harvest against the strawberry mite (two treatments at a seven day interval) gave slightly poorer results than endosulfan (also two treatments), but endosulfan has been withdrawn from use. All the above acaricides were also tested against the two-spotted spider mite (as a single treatment) and they gave good control, similar in efficiency to the standard acaricides Omite 30 WP (propargite) and Nissorun 050 EC (hexythiazox). The results showed that the control of both pests is achievable.

Key words: Phytonemus pallidus Tetranychus urticae, chemical control, strawberry, bifenazate, spirodiclofen, propargite, fenazaquin, strawberry mite, two-spotted spider mite

Introduction

The strawberry mite Phytonemus pallidus Zimm. and the two-spotted spider mite Tetranychus urticae Koch are key pests on strawberry plantations in Poland (Łabanowska 2000; Łabanowska 2003). All strawberry cultivars tested in Poland are infested with these two pests (Łabanowska & Bielenin 2002; Łabanowska 2004) and there is a persistent need to control them. The strawberry mite feeds inside youngest, still unopened leaves and also between smallest furled strawberry leaves from the early spring until September. There are usually 3-5 generations per year. Severely infested plants become stunted and the yield is useless (Alford 1972). Recently two of the best acaricides to control the strawberry mite, endosulfan and amitraz, were withdrawn and cannot be used anymore in the strawberry protection program (Łabanowska 2006). This is the reason why new acaricides useful in controlling these pests need to be found. Some researches are looking for biological control of Phytonemus pallidus (Easterbrook et al. 2001; Tuovinen 2000, 2002) The two-spotted spider mite feeds on older leaves, on the lower surface throughout the growing season (usually 3-5 generations per year). As a result of mite feeding on leaves they become yellow and dry out. The plant growth and yield are lower than on healthy plants. The control of the two-spotted spider mite is still less of a problem than controlling the strawberry mite. Besides chemical methods, also experiments with biological control, with predatory mites belonging to the family Phytoseiidae, of this pest on strawberry were carried out in different countries (Cross 1984; Easterbrook 1992; Cross et al. 1996; Easterbrook et al. 2001; Rhodes & Liburd 2006).

101 102

The aim of this work was to evaluate the efficacy of few new acaricides such as Floramite 240 SC (bifenazate) a Chemtura product, Envidor 240 SC (spirodiclofen) a Bayer product (Nauen et al. 2000), the new formulation of propargite – Omite 570 EW against both strawberry pests and Magus 200 SC (fenazaquin) as well as Ortus 05 SC (fenpiroxymate) only against the strawberry mite.

Materials and methods

The experiments were conducted in 2002-2006 at the Research Institute of Pomology and Floriculture in Skierniewice, on 2-3 years old strawberry plantations, cultivar Senga Sengana. Each experiment was designed in 4 replications. A knapsack motor sprayer ‘Stihl’ was used to apply the acaricide (equal to 750 l of spraying liquid per ha). Against the strawberry mite two sprays, at a one week interval, were applied in the summer, after the harvest of strawberry. Against the two spotted spider mite one spray was applied just before blossom of strawberry. The strawberry mite population density was estimated just before the first treatment and 3-4 times, at one week intervals, after the second spraying. The motile forms of mites and eggs were counted separately on leaves, under a stereoscopic microscope. The samples of 10 leaves were taken randomly from each replication. The two-spotted spider mite population density was estimated just before spraying and 4-5 times after treatment, every 1 or 2 weeks. The active stages of mites and eggs were counted separately using Henderson and McBurnie’s (1943) technique. The samples of 30 leaves were taken randomly from each plot of the four replications. The numerical data were analysed statistically. Duncan’s multiple range test was used for mean separation at the 5% significance level. For two-spotted spider mite the Cumulative Indices of Infestation (CII) was calculated according to the formula of Wratten et al. (1979), and then recalculated into percentages, assuming the CII for the check = 100.

Results and discussion

Control of the strawberry mite (Phytonemus pallidus ssp. fragariae Zimm.) In the 2003 trial, the new acaricide Envidor 240 SC at the rate 0.6 l/ha + Silwett L-77 applied twice, at an interval of one week, gave promising results, in some data similar to standard Mitac 200 EC (Table 1). Also Envidor 240 SC at the rate 0.6 l/ha reduced the population of the mites. In next experiment, in 2004, new acaricides were tested: Envidor 240 SC (0.8 l/ha), Envidor 240 SC + Silwett L-77, a new formulation of propragite as Omite 570 EW + Silwett L-77 and Floramite 240 SC (0.8 l/ha), reduced the number of the strawberry mite, and the results were similar or a little poorer than those with standard Sanmite 20 WP (Table 2). Also in 2005 Envidor 240 SC, Omite 570 EW, Magus 200 SC and Ortus 05 SC reduced the strawberry mite population compared with untreated plants, but the results were poorer than after using Sanmite 20 WP (Table 3). The egg populations were correlated with the numbers of active stages. On all treated plants, the number of the strawberry mite was much lower than on untreated plants. Generally, Envidor 240 SC, Floramite 240 SC, Omite 570 EW, Magus 200 SC and Ortus 05 SC used to control the strawberry mite on strawberry reduced the numbers of the pest, results were similar or a little poorer compared to the standard acaricides Mitac 200 EC and Sanmite 20 WP. The results obtained with Envidor 240 SC and Floramite 240 SC confirmed those from the first experiment (Łabanowska 2003). Also according to Raudonis (2006) spirodiclofen was toxic or moderately toxic, depending on the rate, for Tarsonemus pallidus. Thiodan 350 EC was the best acaricide, as is usually found (Łabanowska 1992, 103

2003). But Thiodan 350 EC and Mitac 200 EC has been withdrawn from use on strawberry. In this situation new acaricides: Envidor 240 SC and Floramite 240 SC after registration could be used to control Phytonemus pallidus on strawberry. Omite 570 EW is just registered for use, but only after harvest, because in the spring phytotoxicity on leaves was noted.

Table 1. Average number of Phytonemus pallidus per leaf- Góra Puławska 2003.

Weeks after 2-nd treatment* Acaricide l/ha 1 2 3 4 Envidor 240 SC + Silwett L-77 0.4/0.375 10.1 c** 5.1 c 5.4 c 19.8 d Envidor 240 SC 0.6 4.4 b 4.0 c 6.1 c 9.1 c Envidor 240 SC + Silwett L-77 0.6/0.375 3.8 b 3.6 c 6.8 c 3.3 b Mitac 200 EC 3.0 4.2 b 1.2 b 1.2 b 2.2 b Thiodan 350 EC 2.5 0.02 a 0.4 a 0.5 a 0.8 a Check (untreated) - 25.1 d 23.2 d 22.1 d 22.1 d *Treatments after harvest: 15.07 and 22.07.2003. Before treatments - 22-29 mites/ leaf and 57-69 eggs/ leaf were present. Silwett L-77 – wetting agent **Means followed by the same letter are not signiificantly different at the 5 % level (Duncan’s multiple range t - test)

Table 2. Average number of Phytonemus pallidus per leaf - Dębowa Góra 2004.

Weeks after 2-nd treatment* Acaricide l or kg/ha 1 2 3 Sanmite 20 WP 2.25 9.1 b** 4.5 a 7.1 a Envidor 240 SC 0.8 6.1 ab 13.2 b 11.6 ab Envidor 240 SC + Silwett L-77 0.6/0.375 6.5 ab 15.3 b 14.7 bcd Omite 570 EW + Silwett L-77 2.0/0.375 6.9 ab 10.4 b 13.1 abc Floramite 240 SC 0.8 3.9 a 10.7 b 21.7 cd Check (untreated) - 29.6 c 29.8 c 26.5 d *Treatments after harvest, on August 03 and 09, 2004. Before treatments, 22 to 58 mites and 20 to 57 eggs per leaf were present **Explanations see under Table 1

Table 3. Average number of Phytonemus pallidus per leaf - Aleksandrów 2005.

Weeks after 2-nd treatment* Acaricide l/ha 1 2 3 Magus 200 SC 0.9 8.6 c** 12.3 d 11.8 d Magus 200 SC + Silwett L-77 0.9 /0.375 3.2 a 5.9 bc 10.6 d Magus 200 SC 1.2 13.1 d 7.7 c 7.7 c Envidor 240 SC 0.6 18.2 efg 7.1 c 5.0 b Envidor 240 SC 0.8 15.7 def 13.2 d 7.4 c Envidor 240 SC + Silwett L-77 0.6/0.375 14.4 de 4.3 b 5.0 b Omite 570 EW 2.0 19.7 fg 7.8 c 7.6 c Ortus 05 SC 1.2 17.2 efg 6.5 c 9.0 cd Sanmite 20 WP 2.25 5.4 b 1.9 a 2.9 a Check (untreated) - 22.5 g 28.4 e 19.8 e *Treatments after harvest, on July 18 and 26, 2005. Before treatments: 15 to 40 mites and 5 to 29 eggs per leaf were present **Explanations see under Table 1 104

Control of the two-spotted spider mite (Tetranychus urticae Koch) The new acaricides showed very high efficacy in control of the two-spotted spider mite: Envidor 240 SC (0.3 and 0.4 l/ha), Envidor 240 SC – 0.3 l/ha + Silwett L-77, as well as Floramite 240 SC (0.6 l/ha) and Floramite 240 SC – 0.6 l/ha + Silwett L-77 (Table 4 and 5). The new formulation of propargite – Omite 570 EW – 1.5 l/ha + Silwett L-77 also provided very good results. All the new acaricides gave similar results as the standards: Nissorun 050 EC (0.9 l/ha) and Omite 30 WP (2.25 kg/ha). On all treated plants the mite population was much lower than on the untreated, check plants.Generally Envidor 240 SC, Floramite 240 SC, Omite 570 EW alone and with Silwett L-77 gave good control of the two-spotted spider mite on strawberry. These results confirmed some earlier ones obtained on black currant (Łabanowska 2002). Also on strawberry promising results with bifenazate were obtained in Florida (Rhodes & Liburd, 2006) and spirodiclofen in Lithuania (Raudonis, 2006).

Table 4. Average number of Tetranychus urticae per leaf - Miedniewice 2002.

l or Weeks after treatment* Acaricide kg/ha 1 2 3 4 5 CII Floramite 240 SC 0.6 0.8 a** 0.4 a 0.1 a 0.1 ab 0.7 bcd 4.9 a Floramite 240 SC + Silwet L-77 0.6/0.375 0.9 ab 0.5 a 0.2 a 0.4 bcd 0.4 bc 5.9 ab Omite 570 EW 1.5 1.0 ab 0.3 a 0.3 ab 0.7 c 0.5 bc 6.6 abc Omite 570 EW + Silwet L-77 1.5/0.375 0.9 ab 0.3 a 0.3 ab 0.1 ab 0.4 bc 4.4 a Omite 570 EW 2.0 0.7 a 0.3 a 0.3 ab 0.9 d 1.0 cd 7.5abcd Omite 30 WP 2.25 2.2 c 0.8 ab 0.4 b 0.4 bcd 0.4 bc 9.1 bcd Nissorun 050 EC 0.9 1.6 bc 0.7 ab 0.2 a 0.02 a 0.3 b 6.1 ab Envidor 240 SC 0.3 2.1 c 1.3 b 0.9 c 0.04 a 0.05 a 10.3 d Envidor 240 SC 0.4 1.5 bc 1.3 b 0.6 b 0.3 b 0.1 a 9.8 cd Check – untreated - 13.3 d 12.0 c 9.2 d 4.3 e 1.5 d 100 e *Date of treatment: June 26. 2002; Before treatment there were 10-12 mites and 24-26 eggs/leaf **Explanations see under Table 1

Table 5. Average number of Tetranychus urticae per leaf - Złota 2004.

l or Weeks after treatment and dates* CII Acaricide kg/ha 1 2 3 4 Envidor 240 SC 0.3 0.1 bc* 0.1 a 0.2 bc 0.03 a 5.4 a Envidor 240 SC + Silwett L-77 0.3/0.375 0.2 c 0.05 a 0.2 bc 0.2 abc 4.6 a Envidor 240 SC 0.4 0.2 c 0.2 a 0.3 c 0.3 bc 7.0 a Omite 570 EW 1.5 0.04 ab 0.02 a 0.04 a 0.1 ab 1.8 a Floramite 240 SC 0.6 0.0 a 0.05 a 0.1 a 0.1 ab 2.9 a Nissorun 050 EC 0.9 0.08 bc 0.1 a 0.1 bc 0.1 ab 3.4 a Omite 30 WP 2.25 0.0 a 0.1 a 0.3 c 0.8cd 8.2 a Check – untreated - 7.6 d 2.9 b 2.9 d 1.8 d 100 b *Date of treatment: July 14, 2004; Before treatment there were 5-18 mites and 2-12 eggs/leaf **Explanations see under Table 1

105

Conclusions

The acaricides Envidor 240 SC (spirodiclofen), Floramite 240 SC (bifenazate), Omite 570 EW (propargite) applied twice (with one week interval) after harvest of strawberry reduced strawberry mite, Phytonemus pallidus. Envidor 240 SC, Floramite 240 SC, Omite 570 EW used once, provided good control of the two-spotted spider mite on strawberry, similar to the control achieved with the standards Nissorun 050 EC (hexythiazox) and Omite 30 WP (propargite). To control the strawberry mite the higher rate of acaricides were needed. Acaricides at a lower rate, but with the wetting agent Silwett L-77 gave usually similar results to the acaricides at a higher rate. Magus 200 SC (fenazaquin), Ortus 05 SC (fenpyroximate) registered to control the two-spotted spider mite, used in higher rate and twice, reduced the strawberry mite number.

Acknowledgements I would like to thank Elżbieta Paradowska for technical help in conducting the experiments and Dorota Łabanowska-Bury for English corrections.

References

Alford, D.V. 1972: The effect of Tarsonemus fragariae Zimmermann (Acarina: Tarsonemidae) on strawberry yields. – Ann. App. Biol., 70: 13-18. Cross, J.V. 1984: Biological control of two-spotted spider mite (Tetranychus urticae) by Phytoseiulus persimilis on strawberries grown in ‘walk-in’ plastic tunnels, and a simplified method of spider mite population assessment. – Plant Pathol. 33: 417-423. Cross, J.V., Burgess, C.M. & Hanks, G.R. 1996: Integrating insecticide use with biological control of two-spotted spider mite (Tetranychus urticae) by Phytoseiulus persimilis on strawberry in the UK. Proceedings. Brighton Crop Protection Conference – Pests and Diseases, 18-21 November 1996, Brighton, United Kingdom: 899-906. Easterbrook, M.A. 1992: The possibilities for control of two-spotted spider mite Tetranychus urticae on field-grown strawberries in the UK by predatory mites. – Biocontrol Sci. Technol. 2: 235-245. Easterbrook, M.A., Fitzgerald, J.D. & Solomon, M.G. 2001: Biological control of strawberry tarsonemid mite Phytonemus pallidus and two-spotted spider mite Tetranychus urticae on strawberry in the UK using species of Neoseiulus (Amblyseius) (Acari: Phytoseidae). – Exp. Appl. Acarol. 25: 25-36. Henderson, C.F. & McBurnie, H.F. 1943: Sampling technique for determining populations of the citrus red mite and its predator. – USDA Circ. 671: 1-11. Łabanowska, B.H. 1992: Effectiveness of acaricides in the control of strawberry mite (Phytonemus pallidus ssp. fragariae Zimm.). – Fruit Sc. Rep. 19 (3): 137-146. Łabanowska, B.H. 1995: Effectiveness of new generation acaricides in the control of two- spotted spider mite (Tetranychus urticae Koch.) on strawberry. – IOBC/wprs Bulletin 19 (4): 415-416. Łabanowska, B.H. 2000: Harmfulness of the strawberry mite (Phytonemus pallidus ssp. fragariae Zimm.) and the two-spotted spider mite on the strawberry and susceptibility of some cultivars for infestation. – Materiały XXXVI Symp. Akarol. “Akarologia Polska u progu nowego tysiaclecia”. Kazimierz Dolny. 1999: 358-361, SGGW, Warszawa. Łabanowska, B.H. 2002: Efficacy of Envidor 240 SC in the control of the two-spotted spider mite (Tetranychus urticae Koch.) on the black currant plantations in Poland. – Acta Hort. (ISHS) 585 (1): 363-367. 106

Łabanowska, B.H. 2003: Control of strawberry mite (Phytonemus pallidus ssp. fragariae Zimm.) with new acaricides (in Polish with English summary). – Progress in Plant Protection / Postępy w Ochronie Roślin 43 (2): 781-783. Łabanowska, B.H. 2004: Spread of the strawberry mite (Phytonemus pallidus ssp. fragariae Zimm.) on thirteen strawberry cultivars. – J. Fruit Ornam. Plant Res. 12: 105-111. Łabanowska, B.H. 2006: Potential agents for controlling the strawberry mite (Phytonemus pallidus ssp. fragariae Zimm.) after the withdrawal of endosulfan. – J. Fruit Ornam. Plant Res. 14 (suppl. 3): 67-72. Łabanowska B.H. & Bielenin, A. 2002: Infestation of strawberry cultivars with some pests and diseases in Poland. – Acta Hort. (ISHS) 567 (2): 705-708. Nauen, R., Stumpf, N. & Elbert, A. 2000: Efficacy of Bay 2740, a new acaricidal tetronic acid derivative, against tetranychid spider mite species resistant to conventional acaricides. – The BCPC Conference – Pests &Diseases 2000, 4 D-9: 453-458. Raudonis, L. 2006: Comparative toxicity of spirodiclofen and lambdacihalotrin to Tetra- nychus urticae, Tarsonemus pallidus and predatory mite Amblyseius andersoni in strawberry site under field conditions. – Agronomy research 4 (Special issue): 317-322. Rhodes, E.M. & Liburd, O.E. 2006: Evaluation of predatory mites and acramite for control of two-spotted spider mites in strawberries in north central Florida. – J. Econ. Entomol. 99 (4): 1291-1298. Tuovinen, T. 2000: Integrated control of the strawberry mite (Phytonemus pallidus) in the nordic multi-year growing system. – Acta Horticulturae 525: 389-392. Tuovinen, T. 2002: Biological control of strawberry mite: A case study. – Acta Horticulturae 567: 671-674. Wratten, S.D., Lee, G. & Stevens, D.J. 1979: Duration of cereal aphid populations and the effects on wheat yield and quality. – Proc. British Crop Protec. Conf. 1: 1-8. Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 107-113

Efficacy of 3 neonicotinoid insecticides for the control of the green leafhopper Asymmetrasca (Empoasca) decedens Paoli, a new pest on cultivated red raspberry in Trentino, Italy

Alberto Grassi1, Romano Maines1, Alessio Saviane2 IASMA Research Center – 1Plant Protection Dept.- 2Agrifood Quality Dept. – Via E.Mach, 1 38010 San Michele all’Adige (TN) - ITALY

Abstract: In 2006, two trials were carried out in Trentino, North Italy, in order to evaluate the efficacy against Asymmetrasca decedens of three neonicotinoids, thiametoxam (Actara), thiacloprid (Calypso) and acetamiprid (Epik). The three compounds were compared with an untreated plot and a malathion (Smart EW) treated plot (standard chemical). A single spray was applied before blossom, on the peak of the first generation nymphs population. The effectiveness was assessed 7, 14, 21, 28 and 35 days after treatment, by counting the number of live nymphs on the underside of 60-120 complete leaves/plot. Leaves inspections were carried out in order to evaluate the side effects of the insecticides on Tetranychus urticae, Neotetranychus rubi and on the indigenous predatory mites Amblyseius andersoni and Euseius finlandicus. The characteristics of the three tested neonicotinoids in terms of efficacy and side effects, better fulfil the requirements of the A. decedens integrated control than the standard chemicals used so far, and deserve our efforts to extend their registration on raspberry in Italy.

Key words: Asymmetrasca (Empoasca) decedens, chemical control, neonicotinoids insecticides, side effects

Introduction

The extremely limited availability of insecticides, mostly of old conception, is getting more and more an unsustainable aspect in the modern integrated raspberry production in Italy. Due to inadequate features of action, these chemicals show a poor effectiveness, particularly against new pests, as for example the green leafhopper Asymmetrasca decedens Paoli. Moreover, their repeated use may alter the ecological balances and cause outbreaks of other pests (as often occurs for Tetranychus urticae Koch after treatments with pyrethroids or similar compounds). The aim of these trials was to evaluate the efficacy against A. decedens of three new generation insecticides, potentially more compatible with the integrated raspberry production local programmes, the neonicotinoids thiametoxam (Actara), thiacloprid (Calypso) and acetamiprid (Epik). Preliminary agreements with the parent agrochemical companies were signed in order to get assurance they would sustain the registration costs. This approach has been already successful in other recent situations, allowing for the first time in Italy the registration of two acaricides on raspberry.

Material and methods

Experimental sites and plot description The trial was organised in two raspberry plantations. Site 1, of about 700 m2, was located in Pergine Valsugana (530 m a.s.l). Two year old plants of cv. Polka are arranged in 12 rows, 28

107 108

m long, on a 2.5 x 0.25 m planting spacing (2 rows/tunnel). Each treatment was applied in 4 times replicated and randomized plots 12 m long, with about 48 canes. Site 2, of about 550 m2 was located in Frassilongo, Mocheni’s Valley (1100 m a.s.l). Plants of cv. Heritage, 4 years old, are arranged in 15 rows, 22 m long, on a 1.65 x 0.25 m planting spacing (3 rows/tunnel). Each treatment was applied to 4 times replicated and randomized plots 10 m long (about 40 canes).

Treatments A motorised knapsack sprayer (mod. SEH06/Italdifra – 7.5 bar working pressure) was used to apply the insecticides at volume rates of 1700 (site 1) and 2400 (site 2) litres of water per hectare. The insecticides were sprayed before flowering, at the phenological growth stage BBCH 51 (21/6 site 1 and 26/6 site 2), when first generation nymphs with wings boxes were recorded. Other details are reported in the table 1.

Table 1. Details of the treatments applied in the experimental sites.

Treatment Active ingredient (a.i) Insecticide product application rate (insecticide product) and concentration site 1 site 2 Smart EW malathion 40% 3332 ml 4800 ml (standard chemical) Actara 25 WG thiamethoxam 25% 400 g 400 g Calypso thiacloprid 40.4% 250 ml 250 ml Epik acetamiprid 20% 375 g 375 g Check - untreated - - -

Inspection methods To assess the effectiveness of the insecticides against A. decedens, live nymphs were counted directly in the field on the underside of 15-30 complete leaves randomly collected from basal, median and apical portions of canes on both sides of each replication. Inspections were carried out 1 or 4 days before the date of spraying and 7, 14, 21, 28, 35 days after. Side effects of the insecticides on naturally occurring populations of spider mites (T. urticae at site 1 and Neotetranychus rubi Trägårdh at site 2) and predatory mites (Amblyseius andersoni Chant at site 1 and Euseius finlandicus Oudemans at site 2) were evaluated also by means of visual inspections, under a binocular microscope, of 5 terminal leaflets/replication, randomly sampled from the median portion of the canes before the date of spray and 7, 14, 28, 35 days after. The density of T. urticae and N. rubi populations was estimated using a ranking system (Guignard 1968 unpub.), while phytoseiids eggs and mobile stages were exactly counted. The % of leaves occupied by each of the mite species was also recorded.

Statistical analysis Data were analyzed by Kruskal-Wallis ANOVA (P ≤ 0.05) and median test analysis, using the software Statistica 7 (Stat. Soft. Inc.). The output was used subsequently by a program that performed a nonparametric multiple comparison test between treatments. This test compared the absolute value of the differences in mean ranks for all pairs with a critical value which was determined using a normal approximation with suitable adjustment of alpha to take the multiple comparisons into account.

109

Results and discussion

Efficacy against A. decedens At both the sites, the tested neonicotinoid insecticides, particularly Actara and Calypso, showed an higher efficacy against A. decedens than that obtained with the standard chemical, malathion (Figure 1).

0,9 site 1 0,8 check 0,7 Smart Actara 0,6 Epik Calypso 0,5 nymph/leaf 0,4

0,3

A. decedens a a 0,2 a a

0,1 b a a b bb b c c c b 0 T-1 T+7 T+14 T+21 T+28 T+35 p-level 0.827 0.048 0.010 0.008 0.112 0.122

0,9

0,8 site 2

0,7

0,6

0,5

nymph/leaf a a 0,4 a 0,3 b a A. decedens 0,2 b a 0,1 a a bbb c c c ccc b b 0 T-4 T+7 T+14 T+21 T+28 T+35 p-level 0.508 0.017 0.003 0.007 0.047 0.094

Figure 1. Population dynamic of A.decedens in the different treated blocks of the site 1 and 2. For each inspection date, the means of the different treatments followed by the same letter were not significantly different (P<0.05, Kruskal-Wallis test). 110

This was also confirmed by the statistical analysis; significant differences between the blocks sprayed with neonicotinoids and the malathion treated block were observed in most of the post-application inspections. They proved to have a very fast and strong control action on the leafhopper population; just 7 days after the spray, their efficacy was 100% compared to the untreated block. Another important feature was the long lasting residual activity, especially for Actara and Calypso. High levels of effectiveness were maintained for 3-4 weeks after application, making possible an almost complete control of the first generation of the pest and the development of a weak second generation, below the damage threshold. Marked differences in terms of persistence and overall efficacy were recorded between the two sites for all the tested chemicals. From 28 days after the application, in the blocks of site 1 treated with neonicotinoids, we observed an evident decrease of efficacy. As it can be seen in the figure 1, in the check block of the site 1 most of the first generation nymphs developed on the plants after the treatment, overlapping with the second generation nymphs. This was due to a frost at the end of May/beginning of June, with temperatures below 0°C in this site. The flight of the overwintering colonising adults temporarily stopped, so delaying the development of the nymphs on the plants. Many of them were exposed to the residual effect of the compounds only. The good systemic properties of Calypso and especially Actara however guaranteed a control, although not optimal, of the nymphs in these blocks at the end of the trial, keeping their number lower than that recorded on malathion and check blocks. At site 2, temperatures at the end of May/beginning of June did not fall below 0°C. The infestation developed normally and the spray hit the peak of the first generation nymph population; in fact, the untreated block curve in the figure 1 decrease after the spray. This permitted a better result of A. decedens control in this site. Malathion showed in this case a good efficacy also, although less than that recorded with the neonicotinoids. Results obtained with Actara at this site are particularly interesting considering that the infestation before the treatment was higher than in the untreated block.

Side effects on spider mites and phytoseiids Figure 2, concerning phytophagous mites, clearly indicate that after the treatment, an higher population developed in the blocks sprayed with neonicotinoids than in the untreated and malathion blocks. The statistical analysis showed highly significant differences between the treatments in some of the post-application inspections. Malathion exhibited an important secondary miticidal activity against T. urticae and N. rubi, limiting their population development better than the natural control in the untreated block. The adverse effects of the neonicotinoids on naturally occurring phytoseiids (mortality, reduction of the predatory abilities, interferences with the numerical response to increasing prey population) are probably one of the reasons of T. urticae and N. rubi population outbreaks after the treatment. E. finlandicus (site 2, figure 3) was more negatively affected than A. andersoni (site 1, figure 3). Among the tested neonicotinoids, Actara showed the most important initial impact on the phytoseiids population development after the spray. The progressive decaying of residues and their side effects on predators, and a great availability of prey, allowed the resumption of the phytoseiids population and their control on phytophagous mites at the end of the trial in neonicotinoids treated blocks of both the sites. The population 35 days after the treatment was even higher than that developed in the check block. The neonicotinoids (particularly Actara and Calypso) may have had also a reproductive stimulating effect (hormoligosis) on the mites (both pests and predators). This phenomenon has been already supposed by other authors (James & Price 2002) for similar chloronicotinoid compounds (e.g. imidacloprid). 111

12 site 1 a check 10 Smart Actara a 8 Epik b Calypso a c 6 b a mobile stages/leaf mobile c 4 d T. urticae T. urticae b 2 dd c b b

0 T-1 T+7 T+14 T+28 T+35 p-level 0.893 0.852 0.001 <0.001 0.017

a 600 b site 2 a

500

b 400

300 mobile stages/leaf mobile

200 c c N. rubi N.

100 d d

cbaba dd 0 T-4 T+7 T+14 T+28 T+35 p-level 0.048 0.099 0.109 <0.001 <0.001

Figure 2. Population dynamic of T.urticae and N.rubi in the different treated blocks of the site 1 and 2. For each inspection date, the means of the different treatments followed by the same letter were not significantly different (P<0.05, Kruskal-Wallis test). 112

4 site 1 a 3,5 check Smart 3 Actara Epik 2,5 Calypso b

2 b

mobile stages/leaf b 1,5

A.andersoni A.andersoni c

0,5

0 T-1 T+7 T+14 T+28 T+35 p-level 0.136 0.278 0.557 0.728 0.002

4 site 2 3,5

2,5

mobile stages/leaf a

1,5 a a 1 b a E.finlandicus b b b 0,5 c b c b bc c c d cbc d cd 0 T-4 T+7 T+14 T+28 T+35 p-level 0.105 0.002 0.013 <0.00 0.003

Figure 3. Population dynamic of A.andersoni and E.finlandicus in the different treated blocks of the site 1 and 2. For each inspection date, the means of the different treatments followed by the same letter were not significantly different (P<0.05, Kruskal-Wallis test). 113

Conclusions

The three tested neonicotinoids, particularly Actara 25 WG and Calypso, showed a more satisfactory efficacy than that obtained with the standard chemical malathion in the control of the green leafhopper A. decedens on raspberry. To get the maximum persistence of activity and a good control of the second more dangerous generation of this pest, a correct timing of the treatment on the peak of first generation nymphal population is very important. Therefore, it will be necessary to improve our knowledge about the biology and behaviour of this pest in our specific environmental conditions and to supply the growers with a practical and efficacy monitoring method. The stimulation of the phytophagous mite population development observed in the blocks treated with neonicotinoids is the most worrying effect we recorded in this trial. Further surveys are necessary to better understand the possible reasons (side effects on predatory mites and insects, hormoligosis). Awaiting a better knowledge of the effects of these new compounds on the agro-ecosystems, a careful and responsible utilization is necessary in case of their registration on raspberry in Italy (max 1 treatment/year, constant monitoring of the reactions of the biocenosis).

References

James, D.G. & Price, T.S. 2002: Fecundity in Twospotted spider mite (Acari: Tetranychidae) is increased by direct and systemic exposure to imidacloprid. – J. Econ. Ent. (95) 4: 729- 732. 114 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 115-120

Integrated Pest Management in protected strawberry crops: Increased returns, fewer pests and reduced pesticide use

Clare Sampson Biological Crop Protection Ltd, Occupation Road, Wye, Kent, TN25 5EN, UK

Abstract: Biological Crop Protection Ltd. has developed an Integrated Pest Management (IPM) programme for the main pest species attacking protected strawberry crops. Control of thrips, Frankliniella occidentalis and spider mites, Tetranychus urticae can be achieved with minimal pesticide input using the programme. The glasshouse whitefly, Trialeurodes vaporariorum was an increasing problem in UK protected strawberries. The physical product Eradicoat™, a starch polymer, was tested in a commercial crop. Two treatments of Eradicoat resulted in 99% control when the plants were small and good leaf cover could be achieved, and 55% control when the canopy was dense. Weekly releases of the parasitic wasp Encarsia formosa 1/m2/ week maintained low whitefly numbers and prevented crop damage. The use of an IPM programme in a commercial crop, resulted in higher yields and more Class 1 fruit compared to a chemical programme, raising grower returns by 18%.

Key words: Biological Control, protected strawberries, whiteflies, thrips, spider mites, Encarsia formosa, Eradicoat, Integrated Pest Management, cost benefit analysis

Introduction

Biological Crop Protection Ltd (BCP Ltd.), produce and supply natural enemies for the control of pests. Biological control has been used in UK strawberries since the 1960s (Garthwaite 2000). Until the 1980s, Phytoseiulus persimilis was the main predator used, for the control of spider mites, Tetranychus urticae. Since then, the biological programme has expanded to control thrips, aphids, tarsonemid mites, whiteflies, caterpillars and vine weevils. The area of protected strawberry crops in the UK has increased to over 2900 ha in 2007 (DEFRA statistics). Demand for biological control has risen, partly driven by resistance to available insecticides, especially in thrips and spider mite populations (Price et al. 2002). Increasingly, consumers demand pesticide free produce and workers are reluctant to spray chemicals. Following a decade of experience and trials in protected strawberry crops around the country, BCP Ltd has developed an Integrated Pest Management (IPM) programme for the main pest species (Table 1). In recent years the glasshouse whitefly, Trialeurodes vaporariorum, has increased as a pest in protected strawberries and control methods are required that do not disrupt the biological control of other pests. The parasitic wasp Encarsia formosa (ENCSURE), controls glasshouse whitefly effectively in protected salad crops (Garthwaite 2000), but the use of sulphur burners for powdery mildew control and low winter temperatures, compromise its establishment in protected strawberries. Encarsia formosa has very little activity below 15°C (Madueke & Coaker 1984). BCP Ltd showed that the physical product Eradicoat™ (starch polymer) kills all stages of whitefly in laboratory trials and that E. formosa established in protected strawberry crops when released once glasshouse temperatures average >15°C with < 2 hours sulphur burning per night (Sampson 2005). This paper reports on grower trials to determine the efficacy of Eradicoat™ (starch polymer) in crops; on the use of E. formosa to control T. vaporariorum and on the economic benefits of using biological pest control in protected strawberry crops.

115 116

Table 1. BCP Biological control programme for pests of protected strawberry crops.

Natural Enemy Species and Release Rates Pest Species Comments Preventative Low High SPIDER MITE AMBSURE (cal) PHYTOSURE (p) FELSURE Use P. persimilis or F. acarisuga for out- Tetranychus Amblyseius Phytoseiulus Feltiella breaks. A. californicus urticae californicus persimilis acarisuga permitted only in 2x3 / m2 2.5 to 10 / m2 1 tub / hot-spot permanent structures. fortnightly for up to 4 weeks THRIPS AMBSURE ABS AMBSURE (c) ORISURE Establish Amblyseius before flowering and Frankliniella Amblyseius Amblyseius Orius laevigatus before thrips occur. occidentalis cucumeris cucumeris Up to 2 / m2 1 sachet / 2 m 50-100 / m2 Orius can be used before flowering at flowering. ERADICOAT or March to September on repeat if thrips MAJESTIK everbearer crops. HYPOSURE build up. Use over the tops of flowers to re- Hypoaspis miles duce adult thrips 150-250 / m2 APHIDS APHIDOSURE Aphidoletes sp. Aphidoletes sp. Aphidoletes go into or Aphidius spp. or Aphidius spp. diapause in short Aphis gossypii, Aphidoletes 0.5 to 1 / m2 1 to 2 / m2 daylengths. Use from Myzus ascaloni- aphidimyza weekly weekly, March – October cus, Macrosi- 5 to 10 / m2 in phum sp., Aula- APHISURE hot spots Use A. colemani for corthum sp. Myzus and Aphis spp. Aphidius and A. ervi for Macro- colemani or ervi siphum and Aulacor 0.25 / m2 weekly thum spp. WHITEFLIES ENCSURE ENCSURE ERADICOAT or Start releases when MAJESTIK block temperatures Trialeurodes Encarsia formosa Encarsia formosa average > 15°C vaporariorum 1 / m2 weekly 1 / m2 weekly VINE WEEVIL NEMASYS L Drench larvae in late autumn >5°C. Re-treat Otiorhynchus Steinernema kraussei in spring for high sulcatus 25,000 nematodes / plant if present populations CATERPILLAR Monitor with phero- Spray DIPEL DF Bacillus thur- Good spray-cover is mone traps for specific ingiensis when moths are seen at essential. Tortrix spp. species 100g/100 litres of water. Repeat sprays at 3 week intervals

Materials and methods

Eradicoat, grower use Assessments were done at Polehouse Nurseries in Lawford, Essex (7 glasshouse blocks, 30,000 sq m). The crops (Elsanta) were grown in raised peat troughs and had an existing whitefly population. In mid-February 2005, the grower applied two treatments of Eradicoat (25 ml/litre of water) at a seven day interval after the outer leaves had been trimmed and the 117

blocks heated to initiate new growth. The plants were sprayed to run-off with a boom sprayer, using flat fan nozzles. On the day before the first treatment and the day after the second treatment, 10 plants from each block were chosen at random and the numbers of adult whiteflies counted per plant (n=70). This was repeated in mid-May, when the leaf canopy was large.

Encarsia formosa, grower trial The trial was done in commercial strawberry crops (Elsanta) grown in raised peat troughs at Hempnall, Norfolk. All blocks had an existing whitefly infestation and were treated with Oberon (spiromesifen) on 27th February 2006 to reduce over-wintering whitefly and spider mite populations. Encarsia formosa pupae were released at 1/m2 per week into three glasshouse blocks (4092, 3720 & 2880 sq m) from 13th April to 25th May 2006. The loose pupae were placed in ENCSURE release cups using one cup per 100 m2. Average block temperatures were c. 16°C and maximum daytime temperatures c. 25°C. Three blocks (1160, 5292 & 11,328 sq m) were left untreated as a control. Sulphur burning for the control of powdery mildew was restricted to two hours per night. Yellow sticky traps (10cm x 25cm) were placed 10 cm above the crop foliage, with one per 250 m2. The numbers of adult whiteflies per trap were counted weekly and the traps replaced. On the final assessment, one mature leaf was collected from each of 30 plants selected at random from each block. On each sample leaf the numbers of live whitefly scales were counted. Statistical analysis of the data was done using analysis of variance (ANOVA).

IPM, cost : benefit analysis The trial was done by Polehouse Nurseries (Marcar, pers. Comm.) in high value commercial spring crops (DarSelect) grown in raised peat troughs at Martham, Norfolk. Two crops in adjacent blocks (1,500 m2) were grown from February to July 2005. Pests were controlled using chemical pesticides in one block and an IPM programme in an adjacent block. For each block, the cost of each pest control was recorded, including labour and application costs. During cropping, the yields and class of harvested fruit were recorded and the total returns calculated.

Results

Eradicoat, grower use In 2005, two Eradicoat treatments at a 7 day interval reduced whitefly adult numbers by 99% in February when there was a small leaf canopy and by 55% in May when there was a large leaf canopy (Table 2).

Table 2. Effect of Eradicoat on adult whitefly numbers in a commercial crop.

Timing Mean numbers of adult whiteflies per plant (n=70) Pre-treatment Post treatment % Control February 30.3 0.1 99 May 5.6 2.6 55

Encarsia formosa, grower trial Releases of E. formosa significantly reduced the numbers of adult whiteflies on traps (p<0.01, Figure 1) and whitefly scales on leaves (p<0.01, Table 3). 118

45 40 35 30 25 Encarsia 20 No Treatment 15 10 Mean Nos. adults / trap 5 0 6.4 13.4 21.4 29.4 16.5 23.5

Figure 1. Effect of Encarsia formosa releases on numbers of adult whiteflies.

On the final assessment, live whitefly scales were either absent from leaves or at a very low level in all the glasshouse blocks treated with Encarsia formosa. In all untreated glasshouses there were small areas of plants with sooty mould with sticky fruit and whitefly numbers were significantly higher (Table 3).

Table 3. Mean numbers of live whitefly scales per leaf in blocks with and without E. formosa on 23rd May 2006.

Treatment Block a Block b Block c Totals all blocks (n=30) (n=30) (n=30) (n=90) Encarsia, 1/m2 0 0.4 0 0.1 Untreated 16.5 1.6 2.6 6.9

Table 4. Costs and returns of integrated and chemical pest control programmes.

Chemical Block IPM Block No. of insecticide treatments 7 2 Yield 1st class fruit 1.97 kg/m2 2.33 kg/m2 Yield 2nd class fruit 0.67 kg/m2 0.47 kg/m2 % Class 1 Fruit 75% 83% Cost of pest control 25p/m2 36p/m2 Returns per m2 £17.73 £20.93

IPM, cost : benefit analysis In the block where chemical pest control was used, seven insecticide treatments were used: abamectin (x3), fenbutatin oxide (x1), clofentezine (x1) and pirimicarb (x2). In the block where IPM was used, 5 species of natural enemy were released to control thrips, spider mites and sciarid flies: Amblyseius californicus (6/m2), A. cucumeris ABS sachets (1 sachet/2 m length of crop), Hypoaspis miles (130/m2), Feltiella acarisuga (1 pot/500m2) and Phyto- seiulus persimilis (4/m2). Pirimicarb was used for aphid control. Other pests (whiteflies and 119

caterpillars) did not exceed the economic damage threshold and were not treated. Using biological control was more expensive than chemical control, but resulted in higher yields, better fruit quality and 18% greater returns, which more than compensated for the higher costs (Table 4).

Discussion

Biological pest control is now being used in the majority of glasshouse strawberry crops in the U.K. Adoption of the BCP IPM programme in protected strawberry crops at Polehouse Nurseries has led to continued reduction in insecticide use and more stable pest control with fewer pest outbreaks. The programme shown is a guide only. Actual recommendations will vary between nurseries, cropping system and pest risk. The main pests are well controlled using biological control agents with the back up of an experienced advisor. Western Flower Thrips (WFT), Frankliniella occidentalis, is potentially the most damaging pest in protected strawberries. When chemical control was used at Polehouse Nurseries in 2002, between six and ten insecticide applications per crop failed to prevent crop damage because WFT was resistant to the range of pesticides available. Since then, spinosad has been registered for use in the UK. Spinosad has been effective against WFT and is compatible with biocontrol. However, WFT resistance to spinosad is widespread in many parts of the world and growers are now reporting resistance in the UK. Therefore growers are advised against routine spinosad applications in order to delay the development of resistance. Since the IPM programme was adopted at Polehouse Nurseries, Amblyseius cucumeris was used as the main control of WFT (Table 1) and thrips control improved annually. In the past two years, no routine insecticide treatments were used against thrips and in 2007 no corrective insecticide treatments were needed against thrips in any of the 30 separate blocks. The programme provided robust and reliable thrips control. Spider mites, Tetranychus urticae, are well controlled biologically as long as the timing and distribution of the predators is right. Feltiella acarisuga works well when released into “hot-spots” and established early in the year. Phytoseiulus persimilis is then used for outbreaks and blanket applications. Several species of aphids attack strawberries. The best control is achieved using the generalist aphid predator, Aphidoletes aphidimyza. However, A. aphidimyza does not tolerate any sulphur burning, which is why it was not used in the IPM trial at Polehouse Nurseries where sulphur burning is used. Growers that use sulphur burning to control powdery mildew, Spaerotheca macularis f. sp. fragariae, can use Aphidius ervi and A. colemani to control the complex of aphid species that attacks strawberry but this may be uneconomic where there are high aphid populations. In this situation the integration of pirimicarb and pymetrozine is advised in combination with Aphidius spp.. There are a range of effective IPM methods of controlling whiteflies, Trialeurodes vaporariorum. BCP trials demonstrated cost effective control using Encarsia formosa at a rate of 1/m2 per week (Figure 1, Table 3). Encarsia formosa maintained low whitefly numbers throughout the trial period and prevented crop damage. This method can only be used where glasshouse temperatures average above 15°C and where sulphur burning is restricted. Sulphur burning is harmful to many natural enemies including predatory mites, A. aphidimyza and E. formosa (Sterk & Put 2004). There are a number of IPM compatible methods available if control is required through the winter period, when average temperatures are below 15°C. Eradicoat showed excellent efficacy in February (99% control, Table 2), when the numbers and size of the leaves were relatively small (c. 5 leaves), but less control in May (55% control, 120

Table 2), when the canopy was dense (>10 leaves) and difficult to penetrate. The difference in control is explained by the difficulty in achieving good spray coverage in a dense canopy. Eradicoat acts entirely physically, causing suffocation by blocking the spiracles in inverte- brates (Cordon, unpublished data). Eradicoat is therefore a good option early in the season or to knock back numbers during harvesting. Both spiromesifen and pymetrozine can also be integrated into a biological programme if required. The implementation of this whitefly control programme at Polehouse Nurseries reduced whitefly numbers from over 500 adult whiteflies per plant in 2004 to <1 per 10 plants in 2007. The comparison between an IPM and a chemical pest control programme shows a clear economic benefit in adopting IPM even though the initial costs of IPM were slightly higher when comparing adjacent blocks in the same year. The main reason for this was the decrease in fruit quality and sale price in the chemical control block due to thrips damage. Over 5 years, the benefits of adopting IPM at Polehouse Nurseries improved annually. Each year there were fewer pest outbreaks and pesticide applications reduced by over 60% (nursery records). Because pest levels declined the crop protection costs declined by 11% over the same period, even though the commodity prices had increased. With a small number of effective pesticides registered for use in protected strawberries, integrating their use with biological control reduces the numbers of applications needed, delays the development of resistance and minimises the amount of pesticide residues in food. Integrating biological control agents to control pests in protected strawberry crops is therefore more cost effective and more sustainable than using a chemical pest control programme and is recommended to all growers.

Acknowledgements

A special thanks to Peter Wensak, Jon Marcar and staff at Polehouse Nurseries. Polehouse nurseries hosted all the presented trial work. Jon Marcar set up and collated the information comparing the costs and benefits of IPM versus chemical control methods.

References

Garthwaite, D. 2000: Changes in biological control usage in Great Britain between 1968 and 1995 with particular reference to biological control on tomato crops. – Biocontrol Science and Technology 10: 451-457. Madueke, E.D.N. & Coaker, T.H. 1984: Temperature requirements of the white fly Trialeurodes vaporariorum (Homoptera: Aleyrodidae) and its parasitoid Encarsia formosa (Hymenoptera: Aphelinidae). – Entomologia Generalis 9: 149-154. Price, J.F., Legard, D.E. & Chandler, C.K. 2002: Two-spotted spider mite resistance to abamectin miticide on strawberry and strategies for resistance management. – Acta Horticulturae 567: 683-685. Sampson, C. 2005: Reducing pesticide use in protected strawberry crops. Commercial experience and grower trials to improve control of glasshouse whitefly, Trialeurodes vaporariorum. – The BCPC International Congress. 8B-4: 283-288. Sterk, G. & Put, K. 2004: Side effects manual. – Biobest Biological Systems. 26-27.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 121-124

Integrating biological control measures against strawberry pests - Preliminary results with strawberry tortrix in Denmark

Lene Sigsgaard Department of Ecology, Faculty of Life Sciences, University of Copenhagen, Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark

Abstract: A three-year study on biological control of strawberry pests was initiated in 2007. Preliminary results from the first year’s study of strawberry tortrix Acleris comariana (Lienig & Zeller) are reported. The strawberry tortrix is a common and widespread pest in Danish strawberry production. The practice of spraying with pyrethroids may have negative consequences for natural enemies. Bacillus thuringiensis may provide an alternative control option, though temperature requirements for a good effect limit its use in the field in spring. The strawberry tortrix study aims to contribute to the knowledge of the natural enemy complex of this pest and how functional biodiversity may be affected by field margins and farming practices.

Key words: strawberry, strawberry tortrix, predation, field margin, Acleris comariana

Introduction

In 2007 we initiated a collaborative project between our university, the University of Aarhus, EWH Bioproduction, four strawberry growers and the advisory service. The objective is to develop biological control of pests in strawberries with a focus on identification of beneficials and development and integration of biological control methods and strategies towards several of the most serious pests in strawberry, with a long term goal of reducing pesticide use. The project consist of three work packages a) spider mites and tarsonemid mites control with predatory mites, b) Strawberry tortrix Acleris comariana (Lienig & Zeller) (Lepidoptera) conservation biological control and control with Bacillus thuringiensis var kurstaki and c) Strawberry weevil and Otiorhynchus spp. control with entomopathogenic fungi and a preliminary early warning system for the strawberry weevil. As the project progresses we will test the integration of promising methods. This paper deals only with point b). In IP and conventional strawberry production there are 0-3 annual treatments with pyrethroids to control strawberry tortrix. Pyrethroids are also used against strawberry weevils. It is desirable to find alternative control methods against the strawberry tortrix since pyrethroids can negatively impact beneficials hereby hindering or complicating the use of biological control against other pests in strawberry. Functional biodiversity can be a powerful tool to control insect pests (Conservation biological control) (Gurr et al. 2003). It builds on the assumption that the structure and composition of adjacent habitats can be used to manipulate arthropod populations, both of pests and beneficials, in farmland. To assess the potential for conservation biological control of different field boundaries surrounding strawberry fields we did a field study comparing the egg predation and predator activity in field margins bordering ponds or streams, field margins along hedgerows and margins to another crop. In addition we did a first assessment of the parasitoids of strawberry tortrix. Microbiological control may also be an option, and many organic growers already use B. thuringiensis against Lepidopteran pests in Brassica crops. Temperature requirements for a good effect limit its use in the open field in spring where the first generation of tortricids are

121 122

active, but an autumn treatment with B. thuringiensis at the time when the second generation is active was assessed.

Materials and methods

Four strawberry farms were included in the study, all in Eastern Denmark. The sites were Kildebrønde (20 km south of Copenhagen), Lille Skensved and Klippinge on Sealand and Holeby on the island of Lolland. All sites are surrounded by diverse traditional farmland, but with respect to Kildebrønde in a fairly urbanised area. In Kildebrønde and Klippinge strawberries are produced following IP principles. The remaining two producers follow organic production principles. Three first year strawberry fields and one second year strawberry field (Holeby) (all field with Honeoye) were sampled for strawberry tortrix from April till August. On three occasions in April, June and August pitfall sampling and predation on egg cards was assessed as a measure predation activity. In three strawberry fields a field margin along a hedgerow was sampled and in three a field margin bordering a pond or stream was sampled. In June we also assessed predation in the margin of the field bordering another crop (cereal and Brassica). In each habitat five pitfalls and five “egg cards” - sandpaper squares (10 cm2) each with 20 Ephestia kuehniella Zeller (Lepidoptera: Pyralidae) eggs serving as model prey were placed in a line with 1 m between each pitfall and egg card and 5 m distance between replicates. Both pitfalls and egg cards were protected from rain by transparent plastic roofs. Pitfall data remain to be analyzed and are not included here. Egg cards were pinned to the ground with toothpicks and collected after 24 h. Predation activity, the number of eggs removed from cards, was analyzed for the effects of type of field margin, type of cropping system and time of season on predation using Proc mixed (SAS Inst. 1999). In June and August by the end of the first and second larval generations, larvae and pupae of A. comariana were collected and reared in the laboratory to get a first assessment of parasitoids active on A. comariana and possible differences among localities. On the two organic farms we tested the effect of B. thuringiensis (Dipel ES) in a simple plot experiment. Plots (replicates) were 15 m long. The plot width was 11 m in Lille Skensved (Symphony) and 6 m in Klippinge (Honeoye). Each treatment (B. thuringiensis or control) was replicated three times. The first treatment was done when a high number of young larvae were observed in the field, the next treatment was 7-10 days later. One week later the number of larvae per m row was assessed. The effect was analysed using Proc mixed (SAS Inst 1999).

Results

Two annual generations of the strawberry tortrix were observed, as in the UK (Turner 1968). Eggs overwinter. First larvae were observed 2 May 2007. The second generation of larvae appeared after mid-June. Spring larvae cause the most important damage as they feed at a time when flowers and fruit are developing. Already in mid April predation activity on the egg cards could be registered. Predation was highest along field boundaries, and least within the field. Among the boundaries tested, the highest predation activity was found in field margins bordering a pond or stream. For example in Lille Skensved 52 ± 12% (SE) of the eggs in the hedgerow margin were eaten after 24h, 32 ± 11% in the field and 68 ± 15% by the margin near the pond. The exception was the strawberry tunnels at Kildebrønde where spring temperatures due to the tunnels were higher. Here predation activity in the field was already as high as by the stream. In early June and mid August the difference between field and margins had largely disappeared, though on 123

the last sampling date predation activity had fallen below that by the stream in the Kildebrønde tunnel crop where the plastic roof had recently been removed. The crossed effect of date x margin x farm was highly significant (df = 32, F = 7.5, P < 0.0001). All the emerging tortricids from field collected and laboratory reared larvae and pupae were identified to be A. comariana (in June a total of 82 adults). The most common parasitoid found in the study was Copidosoma aretas (Walker) (Encyrtidae: Hymenoptera), a poly- embryonic egg-larval parasitoid. Of the larvae collected in June 4% were parasitized by C. aretas. Of the total collected number of immature individuals in June (larvae, pupae, dead parasitized larvae) 26% were parasitized by C. aretas. The genus has recently been totally revised (Guerrieri & Noyes 2005). In much lower numbers were found Ichneumonidae, Braconidae (Hymenoptera) and Tachinidae (Diptera). Specimens of these three families have not been identified further yet. The B. thuringiensis treatment halved the tortricid density compared to controls in both farms (control: 1.7 ± 0.2 larva/m row (mean ± SE); treatment 3.2 ± 0.3 larva/m row) with a highly significant effect of treatment (df = 1, F = 24, P < 0.0001).

Discussion

Predation activity data demonstrate that field margins can be important sources of generalist beneficials to the strawberry fields. This was evident early in the season before specialist natural enemy activity commences, and thus at a time where generalist predators are particularly important (Symondson et al. 2002). How much these predators contribute to tortricid mortality will require further studies. Other tortricids may be active on strawberry, but were not encountered in our material, indicating that A. comariana is the most common species. Field data suggest that C. aretas can be an important parasitoid of the strawberry tortrix in Denmark. Further studies of its biology are ongoing, and may help us to get a better idea of the potential of this parasitoid. The B. thuringiensis trial was promising. Whether a B. thuringiensis treatment in autumn has practical relevance will depend on a larger scale trial where it is possible to assess if the treatment can significantly reduce the tortricid infestation the following year.

Acknowledgements

We are grateful to the growers and to the advisory service for support and assistance. Thanks to student Nanna Karkov Ytting for assistance with field work and insect handling. The research was supported by the Directorate for Food, Fisheries and Agri Business.

References Guerrieri, E. & Noyes, J. 2005: Revision of the European species of Copidosoma Ratzeburg (Hymenoptera: Encyrtidae), parasitoids of caterpillars (Lepidoptera) – Systematic Entomology 30: 97-174. Gurr, G.M., Wratten, S.D. & Luna, J.M. 2003: Multi-function agricultural biodiversity: pest management and other benefits. – Basic and Applied Ecology 4: 107-116. Jervis, M.A., Lee, J.C. & Heimpel, G.E. 2004: Use of behavioural and life-history studies to understand the effect of habitat manipulation. – In: Ecological Engineering for Pest Management. Advances in Habitat Manipulation for Arthropods. Eds: Gurr, G.M.; Wratten, S.D.; Altieri, M.A., Wallingford, CABI & CSIRO: 65-100. SAS Institute 1999: SAS/STAT User's Guide Version 8. – SAS/STAT Inc. Cary N.C. 124

Symondson, W.O.C.; Sunderland, K.D. & Greenstone, M.H. 2002: Can generalist predators be effective biocontrol agents? – Annual Review of Entomology 47: 561-594. Turner, J.R.G. 1968: The ecological genetics of Acleris comariana (Zeller) (Lepidoptera: Tortricidae), a pest of strawberry. – The Journal of Animal Ecology 37: 489-520.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 125-129

Biological control of aphids with Chrysoperla carnea on strawberry

Marion Turquet1, Jean-Jacques Pommier1, Mireille Piron2, Emilie Lascaux2, Guillaume Lorin1 1 Hortis Aquitaine, Regional station of experimentation, Domaine de Lalande, 47110 Sainte Livrade sur Lot, France; 2 Koppert France, 147 avenue des Banquets, 84300 Cavaillon, France

Abstract: Aphids cause considerable damage in strawberry crops and intensive applications of insecticides have come to a limit with the appearance of pest resistance. Aphid species such as Rhodobium porosum and Aphis gossypii are resistant to all chemicals approved on strawberry crops in France. The large variety of aphid species encountered on strawberry crops (about 10 different species), requires the use of non specific beneficial organisms such as lacewings. The lacewing Chrysoperla carnea is widely known as a natural beneficial insect against aphids. However, its effective use as a predator introduced in strawberry crops has still to be confirmed. An experiment carried out in 2007 aimed to assess its efficiency as a predator on strawberries. In this experiment, various treatments were compared on soil- and soilless-grown everbearer strawberry crops: 1 lacewing per plant, 5 lacewings per plant, and, in the soil-grown crop only, a control treatment without lacewings. The introductions were repeated every two weeks starting from planting. The results show that aphid infestations were successfully limited by the use of lacewings. The absence of chemical treatment during the release of lacewings was quite encouraging. Control of pest populations differed according to the amount of lacewings introduced: with 5 lacewings per plant, the predatory action could be characterised as both preventive and curative. With 1 lacewing per plant, the increase in aphid populations was limited but it is suggested that several treatments might be necessary to stop this increase. Furthermore, the introduction of 5 green lacewings per plant limited thrips infestation in soilless culture. These experiments indicated that lacewings can be used as an effective biocontrol agent of aphid populations in strawberry crops. Further studies are now required to develop control strategies using lacewing introductions that may be adopted by strawberry growers.

Key words: Fragaria ananassa, green lacewing, aphids

Introduction

Greenhouses provide an optimal environment for plant growth, thanks to the control of climatic and agronomic factors (temperature, moisture, sun radiation). These optimal conditions for plant growth are also favourable to a number of crop pests, such as aphids. High population development rates require continuous pest control, which leads growers to make multiple insecticide treatments in order to remain below the economic threshold. Insecticides, when used intensively, have shown their limits : toxicity of chemicals and the pollution they cause, side effects on human health, induction of pest resistance (aphid species such as Rhodobium porosum and Aphis gossypii are resistant to all insecticides approved on strawberry crops in France). The large variety of aphid species encountered on strawberry crops (about 10 different species) requires the use of non specific beneficial organisms, such as lacewings. Lacewings are widely known as natural beneficial insect against aphids (Lolivier et al. 2002; Çaldumbide et al. 2001). However, the efficiency of Chrysoperla carnea as a predator

125 126

introduced in strawberry crops has to be confirmed. The experiment carried out in the year 2007 intended to assess the efficiency of this predator on strawberry plants. In this experiment, different treatments were compared (1 lacewing per plant, 5 lacewings per plant and a control treatment without lacewing) on soil- and soilless-grown strawberries.

Material and methods

Soil culture The experimental design included 3 tunnels of 200 m² each (Figure 1). The variety grown was Charlotte (everbearer), planted in the third week of February (week 8), at a density of 3.8 plants/m². Each tunnel included 3 raised beds of 2 rows each. Each tunnel was separated from the next one by a tunnel not included in the trial, acting as a buffer zone.

Control without 5 lacewings/plant Buffer lacewings Buffer 1 lacewing/plant tunnel tunnel

Figure 1. Experimental design in the soil culture.

Soilless culture The trial was carried out as in soilless culture crop in a tunnel of 200 m². The varieties used were everbearing varieties (Charlotte, Mara des Bois and Cirafine) planted in the first week of March (week 10), at a density of 7.5 plants/m² (6 rows). One half of the tunnel received 1 lacewing/plant, the other 5 lacewings/plant. There was no control without lacewings in this trial.

Treatments and observations Young larvae of lacewings were introduced every two weeks starting from planting. For each method, weekly observations were carried out on 20 strawberry plants selected randomly in the whole plot. The presence or absence of aphids on each half leaflet were recorded using a class ranking system, with 0 corresponding to the absence of aphids and 1 to their presence. A final score per leaf, between 0 and 6 was obtained. Averages of aphid infestation records and of the numbers of thrips per flower on the 20 plants observed randomly were calculated, giving an overview of the global evolution of pest populations. An average of the numbers of beneficial insects per leaf (all stages were considered: egg, larva, pupa, adult) was also calculated.

Results and discussion

Aphid species which occurred in the two trials were Rhodobium porosum and Aphis gossypii.

Soil culture The control method required an aphicide treatment 2 months after the beginning of the experiment. Incidence of natural beneficial insects limited the aphid infestation from week 26 to the end of the experiment (Figure 2). In the method “5 lacewings/plant”, very low numbers of aphids and beneficial insects were observed while high numbers of lacewings were recorded (Figure 3). 127

Aphicid 4 0,3

0,25 3 0,2

2 0,15

0,1 1 0,05

0 0 Average of the infestation aphids note aphids infestation the of Average Average number of beneficial per leaf per beneficial of number Average S14 S15 S16 S17 S18 S19 S20 S21 S22 S23 S24 S25 S26 S27 S28 S29 S30 S31 S32 S33

April May June July August Aphids Lacew ings Hov erf lies A. aphidimyza

Figure 2. Evolution of aphids and beneficial insect populations in the control without lacewings in soil culture.

4 0,3

0,25 3 0,2 Lacewings 2 0,15

0,1 1 0,05

0 0 Average of theinfestation aphids note S1 4 S1 5 S1 6 S1 7 S1 8 S1 9 S2 0 S2 1 S2 2 S2 3 S2 4 S2 5 S2 6 S2 7 S2 8 S2 9 S3 0 S3 1 S3 2 S3 3 Average number of beneficialper leaf

April May June July August Aphids Hoverf lies A.aphidymyza Lacew ings

Figure 3. Evolution of aphids and beneficial insect populations with 5 lacewings per plant in soil culture.

4 0,3

0,25 3 0,2 Lacewings 2 0,15

0,1 1 0,05

0 0 Average of the infestation aphids note aphids infestation the of Average S14S15S16S17S18S19S20S21S22S23S24S25S26S27S28S29S30S31S32S33 leaf per beneficial of number Average

April May June July August Aphids Hov erf lies Lacew ings A.aphidymyza

Figure 4. Evolution of aphids and beneficial insect populations with 1 lacewing per plant in soil culture. 128

The method “1 lacewing/plant” showed good healthy conditions but aphid infestation scores were higher than in the method “5 lacewings/plant” (Figure 4).

Soilless culture The results in the soilless culture were similar to those in the soil-grown crop. Aphids were quite absent in the treatment “5 lacewings/plant” (Figure 5).

4 0,3

0,25 3 0,2

2 Lacewings 0,15

0,1 1 0,05

Average of the infestation aphids note 0 0 Average number of beneficial per leaf S14 S15 S16 S17 S18 S19 S20 S21 S22 S23 S24 S25 S26 S27 S28 S29 S30 S31 S32 S33

April May June July August Aphids Lacewings Hoverflies A. aphidimyza

Figure 5. Evolution of aphids and beneficial insect populations with 5 lacewings per plant in soilless culture.

Aphid infestation scores were low in the treatment “1 lacewing/plant” (Figure 6), but higher than in the method “5 lacewings/plant”. This experiment also showed that the number of thrips per flower varied according to the number of lacewings introduced.

4 0,3

0,25 3 0,2

2 Lacewings 0,15

0,1 1 0,05

0 0 Average of the infestation aphids note S14S15S16S17S18S19 S20 S21 S22 S23 S24 S25 S26 S27 S28 S29 S30 S31 S32 S33 Average number of beneficial per leaf April May June July August Aphids Lacewings Hoverflies A. aphidimyza

Figure 6. Evolution of aphids and beneficial insect populations with 1 lacewing per plant in soilless culture.

129

Conclusion

Generally, aphid infestations were successfully controlled by lacewings. The absence of chemical pest control treatment during the release of lacewings was quite encouraging. Control of pest populations differed according to the amount of lacewings introduced: with 5 lacewings per plant, the predator action can be described as both preventive and curative. With 1 lacewing per plant, the increase in aphid populations was limited but several treatments seemed necessary to stop it. Moreover, the positive action of the lacewings on the thrips populations in soilless culture was promising. While these results are encouraging, we should keep in mind that the economical aspects of this type of strategy have not been considered (nine releases of 1 lacewing/plant in soil culture:~0.5€/m², in soilless culture: ~1€/m²). Furthermore, the difficult implementation of this kind of protection strategy is real (long time required to release 1 or 5 lacewings/plant), and thus makes this kind of plan hardly achievable on large growing areas. These experiments confirm that lacewings can be used as predators of aphid populations in strawberry crops. Further studies are now needed to develop crop protection strategies based on the use of lacewings that can be adopted by strawberry growers.

References

Çaldumbide, C., Faessel, L., Travers, M., Thierry, D., & Rat-Morris, E. 2001: Les chrysopes communes, auxiliaires polyvalents. – Phytoma-La Défense des Végétaux 540: 14-19. Lolivier, F., Marrec, C. & Maisonneuve, J-C. 2002: Mise au point d’une méthode de lutte contre les pucerons des cultures légumières et ornementales avec Chrysoperla lucasina et Chrysoperla kolthoffi issues d’un élevage expérimental. – 2ème conférence internationale sur les moyens alternatifs de lutte contre les organismes nuisibles aux végétaux: 398-405. 130 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 131-136

Severing damage by Anthonomus rubi populations in the UK

Chantelle Jay, Jerry Cross, Chris Burgess East Malling Research, East Malling, West Malling, Kent ME19 6BJ, UK

Abstract: The strawberry blossom weevil (Anthonomus rubi) is a widespread pest of strawberry. Damage is caused by females severing flower buds. This work looked at different sampling techniques for the pest: suction sampling, beat sampling and sticky traps. Beat sampling and suction sampling could only be used in dry weather. Beat sampling was slightly more effective, both in terms of catch number and ease of use. Yellow sticky traps were more effective than other colours. To investigate the relationship between severed buds and weevil numbers, natural field populations were managed by using different spray programmes and flower bud severing was assessed throughout the season. The intensity of flower bud severing was significantly lower in the sprayed plots. A regression of the total number of severings per plot against the total number of weevils beat sampled per plot showed that at least 4.5 flower buds were severed per blossom weevil caught. Flower bud severing was also assessed on caged plots following artificial inoculation of different weevil numbers. Caged plots with four introduced females had double the number of severings compared with plots with one or two introduced females. This work contains information that can be used when planning Integrated Pest Management programmes.

Key words: Strawberry blossom weevil, pest monitoring, crop damage assessment

Introduction

The strawberry blossom weevil Anthonomus rubi [Herbst.] (Coleoptera: Curculionidae) is a widespread pest of strawberry. Overwintered adults emerge from late April and reach peak numbers in May. After mating, the female lays eggs singly in unopened flower buds, and severs the flower bud by making a series of small punctures round the peduncle. The flowers either fall to the ground, or remain on the plant but do not develop. The effect of severing on yield has been studied. Cross & Burgess (1998) showed that for small plants (cultivar Elsanta) with less than seven flower buds artificially severed per plant, the yield was decreased in direct proportion to the number of flower buds severed. Therefore no severing damage could be tolerated. It was found that larger plants could tolerate at least 50% flower bud severing without a loss in yield. Terrettaz et al. (1995) found that first year Elsanta plants could tolerate 10% severing, whilst second year plants could tolerate 25% severing with no loss in yield. In fact the yields increased in the second year plants by up to 10% with less than 25% severing. The tolerance shown may be due to crop compensation (English-Loeb et al. 1999), the number of flower buds present in relation to the yield potential of the plant being crutial. Crop compensation for severing damage affects the threshold levels for treatment. However this is not taken into account in published treatment thresholds. Current thresholds for A. rubi are based on either direct counts of adults or counts of severed buds. Previous thresholds have been set at less than one adult per m2 in Yugoslavia and Russia, although recent work relating to sweep samples has given thresholds of 1 to 4 adults per sweep sample (10 sweeps at a moderate walking pace) (Kovanci et al. 2005). In order to establish an action threshold based on pest abundance a relationship between weevil numbers and severed buds is needed. These experiments aimed to establish practicable

131 132

sampling methods for the strawberry blossom weevil, and to determine the relationship between the weevil and damage in a field situation.

Material and methods

Sampling methods A comparison of sampling methods was done at East Malling in Kent in 1997 on a second year strawberry field of the everbearer variety Tango planted in raised beds in double rows. There were twenty plots; each contained four parallel beds, each bed contained 50 plants. Three sampling/trapping methods were compared to determine the most efficient for A. rubi: sticky traps, suction sampling and beat sampling. The sticky trap experiment had four treatments: two colours of sticky traps (blue and yellow) plus white and transparent. These were arranged in a randomised block design with five blocks and four plots per block. Three traps of one trap colour were placed in each plot in different orientations, north/south, east/west and horizontal, with the orientation randomised within the plot. The traps were sticky on both sides and were 20 cm x 20 cm for the yellow, blue and white traps or 16 cm x 16 cm for the transparent traps. The sticky area on each side of the traps was 300 cm2 for the yellow traps, 250 cm2 for the white and blue traps and 170 cm2 for the transparent traps. These were secured on 2 cm x 2 cm x 50 cm wooden stakes with the bottom of the vertical traps 20 cm above the soil; horizontal traps were placed 30 cm above the soil. The traps were placed in the crop on the 7 April 1997 and were monitored and replaced at approximately weekly intervals until 30 September 1997. Traps were removed before routine fungicide and herbicide sprays were applied. The total numbers of A. rubi caught on the traps were adjusted before statistical analysis, to take into account the area that was sticky. Trap catches were compared using Chi- squared tests. Each of twenty plots were suction and beat sampled on separate days. Different plants were sampled and any catches were recorded and released back to the plots on the day of sampling. Suction sampling was done using a petrol engine sweeper/vac modified as described by Macleod et al. (1994). In each case 40 plants were sampled per plot by holding the tube of the vacuum sampler (diameter of 11.5 cm) over each plant and passing it slowly back and forth through the foliage for c. 5 s. Catches from each plot were emptied into a large plastic bag and were taken back to the laboratory for identification. Suction samples were taken once every two weeks from 17 April 1997 – 18 September 1997. Each plot was also beat sampled once every two weeks from 18 April 1997 – 23 September 1997 by tapping each of 40 plants five times over a rectangular washing up bowl 34 cm x 26 cm x 11 cm. The total numbers caught per plot for both techniques during the main sampling period from mid-May to late July were analysed using a paired t-test following square root transformation.

Experiments with natural populations of weevils manipulated by insecticides Two experiments were done on second year strawberries planted in raised beds at East Malling in Kent (1998) and at Efford in Hampshire (1999). Plots were 10 m long containing 50 plants in double rows at 0.4 m spacing, with a 2 m spacing between plots lengthwise and at least 3 m spacing between plots laterally. In both experiments the June bearer variety Elsanta was assessed and an everbearer (variety Evita) was also assessed at East Malling. The everbearers were deblossomed on 10 June at East Malling. In order to generate a range of A. rubi population densities, different programmes of spray applications of the broad-spectrum insecticides cypermethrin and/or chlorpyrifos were applied as treatments at the recommended spray rates. Experiments used a randomised block design with four blocks and three plots per block. Each block had an insecticide treatment and two unsprayed control plots per block. At East Malling in 1998, treated plots were sprayed 133

weekly with chlorpyrifos or cypermethrin from 13 May to 12 September. At Efford chlorpyrifos was sprayed on 29 April 1999. To gain an indication of the blossom weevil population in each experiment, forty plants per plot were beat sampled. Any weevils that were caught were released back onto the plots. At East Malling the June-bearers were sampled weekly from 12 May until 28 May and the everbearers from 14 May until 7 August. At Efford beat samples were taken on 9 occasions from 29 April until 3 June. At East Malling the number of severings in each flower hierarchy order (i.e. primary, secondary, tertiary or quarternary flowers on the inflorescence) was recorded for separate whole rows on 19 and 22 May for the June-bearers and 10 June and 7 July on the everbearers. The numbers of open flowers per plant were counted on 19 May on the June-bearers. At Efford the number of severed flowers on the same 24 plants per plot was assessed on 11 occasions between 28 April and 4 June.

Artificial introduction experiments To determine the effect of weevil number on blossom severing a known number of paired weevils were introduced into field cages at East Malling in 1999. The experiment was done on fourth year strawberries of the variety Elsanta planted in double rows in raised beds with 0.5 m spacing between the plants in the row and 0.4 m spacing between rows in the bed. These plants had large numbers of flowers. Each cage enclosed 10 plants and was constructed of four cloche hoops, covered with horticultural fleece which was secured at the ends and sides to prevent insect escape. To improve the manipulation of the trusses and observation within the cage, a small area at the side of the cage was opened during assessment and then resealed. The experiment was a randomised block design with three blocks and ten plots per block. The treatments were 1, 2 and 4 mating pair(s) of weevils and an untreated control, with two replicates per block for each treatment and four replicates per block for the untreated control. Weevils were collected from commercial strawberry plots on 30 April 1999 and mating pairs were placed in sample tubes. The pairs of weevils were introduced immediately into the cages. Assessments of the number of buds severed were done on 3, 6, 10-11, 13 and 17 May. The total numbers of flowers, of each order, which were severed, were counted. On the last assessment the total numbers of flowers which were severed and the potential number of flowers of each order were counted. Flower counts were made on 1, 4, 7, 12, 14 and 17 May to give an indication of the stage of the crop.

Results

Sampling methods Blossom weevils were sampled from mid-April to early May 1997. Weevil numbers were low at the end of April and during wet weather in early May it was not possible to beat or suction sample. Sampling was restarted in mid-May continuing until late-September. The mean transformed data was 1.97 and 2.47 weevils per plot by suction sampling and beat sampling respectively (P<0.01, 19 d.f.). Sticky traps caught low numbers of blossom weevil; 29 weevils were caught between 6 May and 18 July 1997. No weevils were trapped outside of these dates. Of the weevils caught, 22 were found on the yellow coloured traps, 5 on the blue, 2 on the white and none on the colourless. The traps had a sticky area of 300, 254, 252 cm2 (mean of four traps) for the yellow, blue and white traps respectively. The numbers of weevils caught were adjusted to compensate for the trap area, giving 73%, 19% and 8% of the catches on the yellow, blue and 134

white traps respectively. A Chi-squared test of the adjusted values for the three colours plus the control was significant at P<0.001 at 3 d.f. assuming a null hypothesis that equal numbers should be found on each trap colour. Of all trap catches, on the vertical traps, 4, 6, 4 and 3 A. rubi were found on the north, south, east and west faces respectively, although this was not analysed due to the low numbers. On the horizontal traps, 7 were found on the upper surface and 5 on the under surface. As there were half the number of horizontal traps compared to vertical traps, the results indicate that A. rubi are caught preferentially on the horizontal traps.

Experiments with natural populations of weevils manipulated by insecticides At East Malling, in the June-bearing crop 2 to 3 weevils were caught per 40 unsprayed plants on 12, 18 and 28 May. Less than 2 weevils were found per 40 plants in the sprayed plots. Blossom weevils came into the everbearer crop later, in late June, with peak numbers of up to 9 blossom weevils per 40 plants being caught on 15 July in the unsprayed plots. Numbers dropped rapidly by August and no blossom weevils were found after 7 August. There were fewer weevils in the sprayed plots with a maximum of 3 weevils per 40 plants on the 24 July. There were significantly less total buds severed on the sprayed June bearer plots compared to the unsprayed plots on both 19 May and 22 May (P<0.05 & 0.001) with 1.7 buds severed in the control plot on 19 May increasing to 4.54 on the 22 May. The number in the sprayed plot was constant with <1 severing per plant. The majority of severed buds were secondary and tertiary buds. There was a mean of 7.4 open flowers per plant on 19 May. There were also significantly less severings on the everbearers that had been sprayed compared to those that had not been sprayed. On the 10 June, a mean of 2.27 buds was severed on the sprayed plots compared to 9.82 on the control plots (P<0.001, SED 0.889). There were significantly more secondary, tertiary and quaternary severings for the assessment of the 10 June. At this stage there was a mean of 81 strawberries and flowers per plant. On 7 July there were 0.37 severings on the sprayed plot compared to 1.25 severings on the control plot, this was mainly due to secondary and tertiary severings (P<0.05, SED 0.272). At Efford, blossom weevils were caught from 2 April in all plots. The full spray treatment maintained the levels at less than 3 weevils per 40 plants sampled at any time. The control plots had a maximum of 5 weevils on the 19 and 25 May 1999 with numbers falling to 1 weevil per 40 plants by the 3 June. The total number of weevils caught per plot, over the season, was analysed, with a mean of 25.4 weevils per unsprayed plot and 12.5 per sprayed plot (SED 5.67, 7 d.f.). This was not significant due to the high variability. There was a significant difference in the number of severings on the sprayed and the unsprayed plots, with a total of 121 severings on the sprayed plots and 54 on the unsprayed plots (P<0.05, SED 24.7, 7 d.f.). Although the difference in blossom weevil numbers between treatments was not significant at Efford, a range of catches was generated and a regression of the total number of severings for each plot was plotted against the total number of weevils per plot. This accounted for 79.2% of the variance, and gave an estimate of the parameters of 4.50 (s.e.=0.34), i.e. one blossom weevil caught over the season would lead to 4.5 flowers damaged. In effect this is not a true representation as the weevils were released back onto the plots and could have been recaptured. As there were 9 catches a more representative figure would be to multiply the value by 9 i.e. 40.5 flowers per weevil.

Artificial introduction experiment The number of severings increased at each assessment date for each treatment, with up to 39.5, 29.2 and 74 severings by 17 May for 1, 2 and 4 introduced pairs respectively (Table 1); control plots had no severings. The treatment with four introduced pairs was approximately double that of the other two treatments at each assessment. The treatments were found to be 135

significantly different from the 6 to the 17 May, this was mainly due to the high number of flowers severed in the treatment with four pairs.

Table 1. The total number of flowers severed on 10 plants under cages by strawberry blossom weevils introduced on 30 April 1999.

Number of females introduced 1 2 4 Fpr. SED 3 May 5.8 5.2 8.0 0.755 3.91 6 May 14.2 13.8 32.0 0.038 7.14 10-11 May 25.8 18.2 56.0 0.022 12.41 13 May 32.5 23.3 72.2 0.020 15.79 17 May 39.5 29.2 74.0 0.051 17.09

Discussion

This research has provided information that will help when designing Integrated Pest Management programmes, providing data on the timing and intensity of severing incidence and on the effectiveness of sampling methods. Both the suction sampling and beat sampling were comparable in the number of insects that they caught. Neither technique is however suitable for use when the plant material is wet. This would also apply to techniques such as sweep sampling, which was not tested here. Both suction sampling and beat sampling are most effective later in the day when the air temperature has dropped and the adults are less active. Sticky traps can be used in the field. In this work yellow traps were most effective. This was also shown for related species such as the boll weevil by Leggett & Cross (1978), who also showed additional benefit of using a pheromone component in the trapping system. In the field experiment horizontal traps were effective, which is in agreement with Innocenzi et al. (2001) when horizontal white traps were used as an initial prototype trap for field evaluation and optimisation of the pheromone of A. rubi. Cross et al. (2006) used the pheromone component (Innocenzi et al. 2001) with different traps. Whilst in one experiment they found that yellow traps were more effective, in another they were no more effective than white traps. A sticky stake design also caught a high number of weevils and was practical. A commercial pheromone is now available in the UK from International Pheromone Systems Ltd which can be used with a sticky stake trap. The natural population experiments in 1998 and 1999 gave an indication of the timing of severing. However sampling for blossom weevils should ideally be done before severing occurs, from flower extension in the spring. The relationship between the total number of severings and the blossom weevil number caught by beat sampling provided an indication of potential damage in the crop. However this can only be applied to this particular sampling technique. The regression indicated that one blossom weevil caught would lead to 4.5 flowers damaged; this assumes that weevils are not re-caught. In these experiments the weevils could be re-caught increasing the potential damage. More work would be needed to relate these figures to economic sampling thresholds in specific years. The artificial introduction experiment gives an indication of damage that can be caused per female. The mean total number of flowers severed in the single mated female treatment was 39.5 by the end of the season, this value was lower per female for the other treatments, perhaps due to searching 136

competition. These values could be applied to any sampling technique including the use of pheromone systems.

Acknowledgements

This work was funded by the UK Department for Environment Food and Rural Affairs.

References

Cross, J.V. & Burgess, C.M. 1998: Strawberry fruit yield and quality responses to flower bud removal: A simulation of damage by strawberry blossom weevil (Anthonomus rubi). – Journal of Horticultural Science and Biotechnology 73: 676-680. Cross, J.V., Hesketh, H., Jay, C.N., Hall, D.R., Innocenzi, P.J., Farman, D.I. & Burgess, C.M. 2006: Exploiting the aggregation pheromone of strawberry blossom weevil Anthonomus rubi Herbst (Coleoptera: Curculionidae): Part 1. Development of lure and trap. Crop Protection 25: 144-154. English-Loeb, G., Pritts, M., Kovach, K., Rieckenberg, R. & Kelly, M.J. 1999: Compensatory ability of strawberries to bud and flower removal: Implications for managing the strawberry bud weevil (Coleoptera: Curculionidae). – Journal of Economic Entomology 92: 915-921. Innocenzi, P.J., Hall, D.R. & Cross J.V. 2001: Components of male aggregation pheromone of strawberry blossom weevil, Anthonomus rubi Herbst. (Coleoptera: Curculionidae) – Journal of Chemical Ecology 27: 1203-1218 Kovanci, O.B., Kovanci, B. & Gencer, N.S. 2005: Sampling and development of economic injury levels for Anthonomus rubi Herbst adults. – Crop Protection 24: 1035-1041. Leggett, J.E. & Cross, W.H. 1978: Boll weevils: The relative importance of color and pheromone in orientation and attraction to traps. – Environmental Entomology 7: 4-6. Macleod, A., Wratten, S.D. & Harwood, R.W.J. 1994: The efficiency of a new lightweight suction sampler for sampling aphids and their predators in arable land. – Annals of Applied Biology 124: 11-17. Terrettaz, R., Antonin, P., Carron, R. & Mittaz, C. 1995: Incidence économique des dégâts simulés de l’anthonome sur les fleurs du fraisier. – Revue Suisse Viticulture, Arbori- culture, Horticulture 27: 361-363. Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 137-138

Studies on control of the vine weevil Otiorhynchus sulcatus using entomopathogenic nematodes

Solveig Haukeland Bioforsk, Norwegian Institute for Agricultural and Environmental Research, Plant Health & Plant Protection Division, Hogskoleveien 7, 1432 Aas, Norway

Abstract: Entomopathogenic nematodes (EPNs) are commercially available for control of soil dwelling larvae of the vine weevil (Otiorhynchus sulcatus). In Europe several products are available comprising three different species, Heterorhabditis bacteriophora, H. megidis, and Steinernema kraussei. The latter species is for use at low temperatures. Results from several trials using H. megidis and S. kraussei against vine weevil in strawberry fields indicate that low temperature, soil type and possibly the application method are limiting factors that appear to reduce their efficacy. The use of EPNs against vine weevil larvae in the field will be discussed including results from cold-activity studies and application methods.

Key words: entomopathogenic nematodes, Steinernema kraussei, Heterorhabditis megidis, vine weevil, Otiorhynchus sulcatus, biological pest control

Introduction

The black vine weevil Otiorhynchus sulcatus, (F.) (Coleoptera: Curculionidae), is a serious pest of many economically important plants including annual and perennial ornamentals and soft fruit in the temperate regions of the world, (Moorhouse et al. 1992). It is endemic to the temperate areas of Europe and the spread of O. sulcatus to places such as USA and Australia is associated with movement of plants. The adults feed on the leaves, deposit eggs on host plants such as strawberry and the developing larvae cause serious economic damage by feeding on plant roots. Currently the most common control method is the use of an insecticide targeting the adults during their feeding at night. In strawberry production adult control is not recommended in summer because of the fruit harvest, applications are therefore recommend after harvest by which time most of the eggs have been laid. Entomopathogenic nematodes (epn) in the families Heterorhabditidae and Steiner- nematidae are currently being used as biological control agents against many pests, including the black vine weevil. It is already well documented that the larval and pupal stages of the vine weevil are susceptible to epn, (Cowles et al. 2005). Heterorhabditis spp. appears to be the most effective and in Europe H. megidis and H. bacteriophora are sold for vine weevil control. These species however do not work well at temperatures below 10°C. A few years ago S. kraussei (Nemasys L) was commercially available for use at low temperatures. Studies reported by Long et al. (2000) and Willmott et al. (2002) show that this species is effective against vine weevil larvae and pupae at temperatures well below 10°C. In this paper we summarize some results from studies conducted to investigate the efficacy of S. kraussei at low temperatures.

137 138

Entomopathogenic nematodes (epn) for control of vine weevil in the field

In the field it is the soil borne larvae that are targeted for control by epn. In Northern Europe the larval stages are present from summer until the following spring (with few exceptions). It is common to target the over-wintering larvae by applying epn in the autumn months. Field trials conducted in naturally infested strawberries against larval stages of O. sulcatus using H. megidis and S. kraussei have indicated that the former species performs better than the latter (Haukeland unpublished). S. kraussei does give some control in the field at low temperatures (below 10°C) but usually not more than about 50%. H. megidis on the other hand gives good control (90%) at higher temperatures (above 10°C). A heavy soil type might influence both epn species efficacy negatively. Another limiting factor may be the application method of epn in the field. Epn treatments are recommended applied as a drench with about 100ml per plant. When applied on strawberries grown under plastic mulch, there may be considerable spillage and loss of nematodes. Care must be taken to ensure that the epn solution reaches the root zone of the plants where the weevil larvae feed. In laboratory trials with S. kraussei and H. megidis against larval stages of O. sulcatus, there is no doubt that S. kraussei is comparatively cold active. However the level of control (% mortality), like in the field, is not very high, at 9°C up to 50% and at 6°C between 10 and 20% (Haukeland unpublished data). It may be that O. sulcatus is not a suitable host for this epn species. An immune response (melanization) against developing S. kraussei in infected O. sulcatus larvae has been observed by Steiner (1990) and personal observations. In conclusion, based on studies conducted so far, epn for control of O. sulcatus larvae in the field is best conducted with H. megidis (or H. bacteriophora) when soil temperatures are above 10°C. By ensuring a good application method and targeting the larvae that are about to over-winter, by applying epn in the autumn, damage to the crop may be reduced the following year.

References

Cowles, R.S., Polavarapu,S., Williams, R.N., Thies, A. & Ehlers, R-U. 2005: Soft fruit applications. – In: Nematodes as biocontrol agents. Grewal, P.S., Ehlers, R-U. and Shapiro-Ilan, D.I. (Eds.): 231-254. Long, S.J., Richardson, P.N. & Fenlon, J.S. 2000: Influence of temperature on the infectivity of entomopathogenic nematodes (Steinernema and Heterorhabditis spp.) to larvae and pupae of the vine weevil Otiorhynchus sulcatus (Coleoptera: Curculionidae). – Nemato- logy 2: 309-317. Moorehouse, E.R., Charneley, A.K. & Gillespie, A.T. 1992: A review of the biology and control of the vine weevil, Otiorhynchus sulcatus (Coleoptera: Curculionidae). – Annals of Applied Biology 121: 431-454. Steiner, W.A. 1996: Melanization of Steinernema feltiae (Filipjev) and S. kraussei (Steiner) in larvae of Otiorhynchus sulcatus (F.). – Fundamental and Applied Nematology 19: 67-70. Willmott, D.M., Hart, A.J., Long, S.J., Edmundson, R.N. & Richardson, P.N. 2002: Use of a cold-active entomopathogenic nematode Steinernema kraussei to control overwintering larvae of the black vine weevil Otiorhynchus sulcatus (Coleoptera: Curculionidae) in outdoor strawberry plants. – Nematology 4: 925-932.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 p. 139

Use of entomopathogenic fungi for vine weevil and thrips control

Tariq M. Butt, Farroq A. Shah, Minshad A. Ansari Department of Biological Sciences, University of Wales, Swansea, SA2 8PP, UK

Abstract: The insect pathogenic fungus Metarhizium anisopliae V275 was highly efficacious against black vine weevil (BVW) larvae and western flower thrips (WFT) pupae in a range of different plant growingmedia (peat, bark, coir, peat blends with 10% and 20% composted green waste). When M. anisopliae was used alone it gave 75-100% control of BVW in a range of plant species (Euonymus, Sedum, Tellima, strawberry), which was equal or higher than that provided by imidacloprid or fipronil. However, when used with either low rates of chemicals (1% imidacloprid or 10%fipronil) or entomopathogenic nematodes, it gave even higher control of BVW irrespective of the application method, plant species or growing media. M. anisopilae was more efficacious than chemicals in controlling WFT pupae in a range of growing media. Efficacy against WFT was independent of low dose chemicals. Indeed several entompathogenic fungi screened gave better control than the chemical pesticides. Whereas M. anisopliae alone caused 78-91% mortality of WFT pupae, imidacloprid and fipronil alone gave 40% control. Overall our studies show that M. anisopliae V275 is robust and compatible with conventional insecticides and offers much promise for the control of BVW and WFT as part of an integrated pest management program.

Key words: Metarhizium anisopliae, vine weevil, thrips, entomopathogenic nematodes, entomopathogenic fungi, growing media, insecticides

139 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 p. 140

Breeding for durable resistance to the large raspberry aphid, Amphorophora ideai, in field and protected raspberry plantations: Co- evolution and IPDM

Nick Birch1, Stuart Gordon1, Rex Brennan1, Nikki Jennings2, Carolyn Mitchell1 1 SCRI, Invergowrie, Dundee, DD2 5DA, UK; 2 MRS Ltd, Invergowrie, Dundee, DD2 5DA UK

Abstract: Plant breeders at SCRI and East Malling Research have been breeding raspberry cultivars with genetic resistance to the large raspberry aphid, Amphorophora ideai for more than forty years. This aphid is a major pest of both field- and tunnel-grown raspberries in the UK, because it transmits four damaging viruses. It can also damage the plant directly during feeding, via honeydew/fungal interactions and can be a contaminant on harvested fruit. Due to the prolonged cultivation of aphid- resistant raspberries in the UK, biotypes of this aphid species have counter-adapted to the resistance genes being deployed. Progressively the A1 and A10 resistance genes in different raspberry cultivars have been overcome. Minor gene resistance (multi-genic, partial resistance to A. idaei) is more durable, but less effective. A key to developing more durable strategies for controlling this pest without increased reliance on pesticides lies in understanding the co-evolutionary interactions between the pest, the resistant plant and natural enemies, occurring over space (tunnels, fields, regions, countries) and time (growing seasons). SCRI’s ongoing research on aphid resistance and deployment strategies will be discussed in relation to the optimal use of aphid resistance genes within an IPDM framework, taking into account the potential effects of changing climate and of crop management practices using decreased insecticides sprays to comply with consumer demands.

Key words: Amphorophora ideai, large raspberry aphid, breeding for pest resistance, durability, Integrated Pest and Disease Management

140 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 141-147

Aphid biology and the development of a program to manage the spread of Blueberry scorch virus in southwestern British Columbia, Canada

David A. Raworth1, Sneh Mathur1, Mark Sweeney2, Victoria Brookes1 1Pacific Agri-Food Research Centre, P.O. Box 1000, 6947 #7 Highway, Agassiz, British Columbia, V0M IA0, Canada; 2British Columbia Ministry of Agriculture and Lands, 1767 Angus Campbell Road, Abbotsford, BC, V3G 2M3, Canada

Abstract: Blueberry scorch virus (BlScV) was first found in New Jersey, USA in the 1970’s associated with Sheep Pen Hill Disease. Various strains of the virus have since been found in the Pacific Northwest, and more recently, the north-eastern United States and Europe. Certain BlScV strains cause severe yield reductions in some cultivars, and the pathogen is considered an important threat to the highbush blueberry (Vaccinium corymbosum L.) industry. We undertook studies of the aphid vectors during 2001-2007 and used the information to develop techniques to reduce the spread of BlScV. Here, we consolidate the results of that work and discuss the implications.

Key words: highbush blueberry, Ericaphis fimbriata, life history, integrated control

Note: Use of trade names or trademarks does not imply endorsement of the companies or products named nor criticism of similar ones not named.

Introduction

Blueberry scorch virus (BlScV) is the causal agent of an important disease of highbush blueberry, Vaccinium corymbosum L. (Ericaceae). BlScV has been found in: New Jersey (Stretch 1983; Martin et al. 1992; Cavileer et al. 1994), Washington State (Martin & Bristow 1988), Oregon (MacDonald & Martin 1990), Connecticut and Massachusetts (DeMarsay et al. 2004), U.S.A.; southwestern British Columbia (Hudgins 2001) and Québec (Canadian Food Inspection Agency 2005 Survey), Canada; the Netherlands (Martin et al. 2006), and northern Italy (Ciuffo et al. 2005). In southwestern British Columbia, BlScV was not detected in an extensive survey of highbush blueberry in 1989 (MacDonald & Martin 1990), however, it was detected in 20 out of 25 fields with symptomatic plants, throughout the southwestern British Columbia production area in 2000 (Hudgins 2001), and in 140 fields by the end of 2004 (Wegener et al. 2006). The source of BlScV in British Columbia is not known. Hudgins (2001) showed that most of the infected cultivars sampled (18 of 26) were more than 15 years old, older than the first extensive survey by MacDonald and Martin which revealed no infected fields; and this trend in detecting BlScV in fields more than 15 or 20 years old has persisted. Clearly, whatever the source, considerable virus spread has occurred since 1989. Whether the increase from 20 to 140 infected fields during the recent 4 years was due to spread or simply detection of infections established during the 1990’s is not known. Symptoms depend on virus strain (Wegener et al. 2006) and blueberry cultivar, and can include severe blighting of flowers and young leaves, twig dieback, and yield reductions of more than 85% in the third year of symptom expression (Bristow et al. 2000). The virus is a

141 142

member of the genus Carlavirus and is transmitted by Ericaphis fimbriata (Richards) (= Fimbriaphis fimbriata) (Hemiptera: Aphididae) (Remaudière & Remaudière 1997) in a non-persistent fashion with a 1- to 3-year latency period prior to symptom expression (Bristow et al. 2000). Little is known about the relative role of resident and migrant aphids in the transmission of BlScV, however, Lowery et al. (2004) showed in controlled laboratory studies that BlScV can be transmitted by species other than E. fimbriata. BlScV has been transmitted to an alternate herbaceous host, Nicotiana occidentalis, in the laboratory (Lowery et al. 2005); and it has been found in the field in cranberry (V. macrocarpon Ait.), which is asymptomatic (Wegener et al. 2004), and black huckleberry (V. membranaceum Dougl. ex Torr.), which is also asymptomatic (Wegener et al. 2007), complicating the picture of potential virus sources. Given a non-persistent mode of transmission, we undertook studies of all aphid species in commercial highbush blueberry fields in southwestern British Columbia with the aim of developing aphid management techniques that would reduce virus spread. The work was split into two parts: resident aphids; and winged migrant aphids.

Material and methods

Resident aphids Effective aphid management requires an understanding of the life cycle, population trends, and life history traits of the aphids on blueberry. Surveys of aphid populations in five commercial highbush blueberry fields were conducted during 2001-2003 by collecting leaf terminals (all leaves arising from a single bud) weekly through the field season (Raworth 2004). Life history studies of E. fimbriata were conducted on leaf disks in Petri dishes to obtain developmental rate data and age-specific fecundity and survival curves at different temperatures, and seasonal estimates of developmental time and fecundity on leaves collected from the field at different times (Raworth and Schade 2006). Dormant spray oil (98.9% mineral oil; 1 part oil to 50 parts water by volume) was tested to determine its effect on eclosion of E. fimbriata eggs on blueberry foliage during February and March (Raworth 2004). Several chemicals, thiamethoxam (Actara® 25WG), clothianidin (Clutch 50 WDG), pymetrozine (Fulfill 50 WG), pymetrozine + LI 700 surfactant, and imidacloprid (Admire 240 g/l) were tested in a randomized complete block field experiment with four replications to determine their efficacy in controlling aphids on blueberry early in the spring.

Migrant aphids There are about 400 species of aphids in British Columbia (Chan & Frazer 1993). We surveyed migrant aphid populations in five commercial highbush blueberry fields to determine which species were present and their population patterns during the field season. Green tile, yellow tile, and Tremclad yellow water pan traps, mounted at the top of the plant canopy, were used to collect aphids during 2001-2002 (Raworth et al. 2006). Eight potential vectors of BlScV were selected for transmission trials based on the water trap data and known virus-vector relationships (Chan et al. 1991): Anoecia corni (Fabricius), Aphis fabae Scopoli, Brachycaudus helichrysi (Kaltenbach), Capitophorus hippophaes (Walker), Euceraphis betulae (Koch), Hyalopterus pruni (Geoffroy), Pemphigus spyrothecae Passerini, and Periphyllus testudinaceus (Fernie). These species were collected from wild host plants in 2004. Including E. fimbriata as a positive control, 2,180 of these aphids were placed individually on BlScV-infected leaves for 1-40 minutes and then transferred to healthy plants to test for BlScV transmission. Tests were also made with aphids and infected material in sleeve cages using E. betulae or P. testudinaceous, total 420 aphids; and aphid transfers were 143

made from blueberry to cranberry, and cranberry to blueberry using E. fimbriata, E. scammelli (Mason), or A. fabae, total 1,440 aphids. All plants including healthy controls were maintained in a screenhouse for 3 years and were observed for symptoms and tested for BlScV by double-antibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA) in the spring of each year. Because 87 migrant aphid species were found in blueberry fields and transmission efficiencies for individual species were difficult to determine, we decided to treat migrant aphids as a group. Winged aphids have been successfully manipulated by increasing plant canopy reflectance (Lowery et al. 1990). Therefore, the efficacy of kaolin clay in reducing winged migrants was tested. Near a field with BlScV-infected plants, 100 healthy trap plants in groups of five were set up on rebar posts coated with Stickem Special® so that they would trap only winged aphids. Ten groups treated with kaolin clay and 10 untreated groups were completely randomized. Winged aphids were collected from the plants for 8 weeks during the spring and summer. After field exposure, the plants were maintained aphid free in a screenhouse and tested for BlScV by DAS-ELISA each spring for 3 years, confirming positive plants with reverse transcription polymerase chain reaction (RT-PCR) (Raworth et al. 2007).

BlScV titres BlScV transmission in commercial highbush blueberry must depend upon at least two factors, aphid densities, and virus titres. If either factor becomes zero, transmission will not occur. We have good information about the density of resident and migrant aphids, but little is known about BlScV titres in infected plants. We therefore conducted a survey of the temporal changes in BlScV titres in infected plants using DAS-ELISA optical density readings as an index of relative titres.

Results and discussion

Resident aphids The dominant aphid species on blueberry in southwestern British Columbia is E. fimbriata (Raworth 2004). This aphid emerges from eggs on blueberry in late February and early March, increases exponentially in numbers until early July, and then declines, producing sexuals in September and eggs in October. Traditionally, aphids on blueberry have been controlled chemically at mid-June when they are obvious to the grower. However, aphid samples indicated that this timing is too late. It is after the initial production of winged aphids in early May – wingless fundatrix (stem mother) aphids produce a large proportion of winged progeny, particularly on flower clusters (Raworth 2004). In addition, chemical controls in June may not be effective (Raworth 2004), and they may adversely affect predators and parasitoids which are active and reproducing (Raworth & Schade 2006). We therefore moved the recommended chemical application date to late March or early April – after the aphids have hatched from eggs, and before winged aphids are produced. The idea is to reduce the number of founding aphids by x %, and thereby reduce the production of winged aphids and the population peak in early July by the same percentage. For example, given a rate of increase of 0.072 ln (aphids) per 5.9 day-degrees (dd) > 4.1°C, 770 dd from 1 April to 30 June, and pre- and post-chemical-treatment densities of 0.001 or 0.0001 aphids per leaf terminal on 1 April, the respective pre- and post populations on 30 June would be 12.1 and 1.21 aphids per terminal. Because the probability of virus transmission should increase as a function of aphid density, the strategy should reduce the spread of BlScV. In early April, E. fimbriata is below detection levels, so chemical control is recommended without sampling. This is a major shift away from a basic principle of IPM, however, aphids are found in virtually every field, so the approach is considered reasonable, and will work best if adopted 144

by all growers in the area of concern. A chemical with systemic activity is necessary because the aphids are concealed within leaf and flower clusters. Both pymetrozine and thiamethoxam demonstrated good efficacy in the field trials, and are in the process of registration for blueberry crops in Canada, however, pymetrozine required a surfactant for adequate aphid control. Although application in early April should minimize the impact on predators and parasitoids, pymetrozine is probably the chemical of choice because it appears to affect natural enemies and bumblebees the least (Torres et al. 2003; Sterk et al. 2003; van de Veire & Tirry 2003). This early spring chemical application could be avoided if dormant spray oil effectively prevented eclosion of E. fimbriata eggs, however, no effect of dormant spray oil was observed in replicated studies with leaves and whole plants (Raworth 2004). Aphid life history studies indicated that plant quality – for the aphids – decreased during the field season so that population growth rate approached zero during August (Raworth & Schade 2006). The consistent declines in aphid populations during July and August among fields and years (Raworth 2004) are considered to be a result of these changes in plant quality combined with natural losses from predators and parasitoids. Our approach, therefore, is to let the system develop on its own during the summer. One aphid spray is recommended in early spring, and we are attempting to use biological control and crop management techniques to reduce the exponential rate of increase of the remaining resident aphid population during the spring. Ongoing studies include aphid parasitoid releases at specified times, based on day- degree summations, to augment existing parasitoid populations, and manipulation of water and plant nutrients to reduce aphid birth and developmental rates.

Migrant aphids Water trap samples indicated that at least 87 aphid species frequent blueberry fields as winged migrants in southwestern British Columbia (Raworth et al. 2006). There were many statistical interactions involving years, fields, and seasons, which varied with aphid species. This suggests that once specific vector species are known, controls would probably need to be tailored for each field and year. None of the eight aphid species selected for transmission tests transmitted BlScV. Overall, transmission studies support the view that BlScV is not readily transmitted, but transmission probability increases with high aphid numbers (Bristow et al. 2000). Two applications of kaolin clay reduced the number of alatae collected from treated highbush blueberry plants by as much as a factor of eight relative to untreated controls (Raworth et al. 2007). However, the efficacy of kaolin clay in the spring, when rainfall washes the material from the leaves, is questionable; three of five BlScV-infected plants became infected in May, a month with greater rainfall than later months, and two of the three plants had been treated with kaolin clay. Furthermore, application of kaolin clay after fruit set results in white residues amongst the bracts of the calyx, limiting its use on fruiting crops to early spring. Mixing kaolin clay with an effective surfactant-sticker, or utilizing reflective netting (Loebenstein & Raccah 1980), could solve the problem associated with rainfall, but both approaches require testing in blueberry crops.

BlScV titres Preliminary results of the BlScV titre survey indicated that titres decline during the season. The result is currently being confirmed, but if true, it has important implications for virus transmission. Transmission probabilities will depend, among other factors, on the temporal position of the BlScV titre curve relative to the position of the aphid density curve; the former, declining, and the latter increasing until early July and then declining.

145

Conclusions

Eradication of BlScV from blueberry production areas appears unlikely given the 1- to 3-year latency period, lack of symptoms with some strains in some varieties, and its symptomless presence in alternate hosts. The severity of symptoms in many cultivars and lack of symptoms in others (Bristow et al. 2000) suggests that growers should select a cultivar for production based on knowledge of the distribution and strains of BlScV in the area. They should also be cautious when purchasing planting stock, reviewing the location and procedures used by the propagator, and using only plants grown under an accepted certification or testing program; DeMarsay et al. (2004) and Martin et al. (1992) conjectured that initial spread of the disease into Connecticut and Massachusetts, and Washington and Oregon, respectively, was probably due to shipping latently infected plants to those areas. Knowledge of the distribution and strains of BlScV in the area will determine the action of the grower with respect to crop surveillance and control of resident aphids. If the cultivars grown are asymptomatic then the grower may ignore the problem, however, this practice has important implications: it will result in a source of BlScV that could infect symptomatic varieties planted at a later date; and the long-term status of asymptomatic cultivars is not known. If BlScV is present in an area with symptomatic cultivars, then area-wide aphid control, thorough crop surveillance, and roguing of infected plants – including roots – after chemical aphid control to minimize aphid spread while rouging is probably warranted. Wagener et al. (2006) presented some data to show that this is a useful approach. However, it is important to note that symptoms similar to those caused by BlScV can also be caused by other factors including frost damage, Blueberry shock virus (BlShV), botrytis flower blight, flower infection with mummy berry, or winter moth damage, (Martin et al. 2006), and Pseudomonas infection, all of which would be treated in different ways. Therefore, before action is taken, plants with symptoms should be tested by DAS-ELISA to confirm BlScV infection. Winged aphids have been implicated in the transmission of BlScV (Raworth et al. 2007), therefore effective measures for reducing the numbers of winged aphids that land and probe blueberry plants are urgently needed.

Acknowledgements

We thank Jerry Cross, Christian Linder, and their collaborators, for the opportunity to present this paper at the 2007 IOBC/WPRS Workshop on Integrated Soft Fruit Production at East Malling Research, U.K. Pacific Agri-Food Research Centre contribution # 767.

References

Bristow, P.R., Martin, R.R. & Windom, G.E. 2000: Transmission, field spread, cultivar response, and impact on yield in highbush blueberry infected with Blueberry scorch virus. – Phytopathol. 90: 474-479. Cavileer, T.D., Halpern, B.T., Lawrence, D.M., Podleckis, E.V., Martin, R.R. & Hillman, B.I. 1994: Nucleotide sequence of the carlavirus associated with blueberry scorch and similar diseases. – J. Gen. Virol. 75: 711-720. Chan, C.K. & Frazer, B.D. 1993: The aphids (Homoptera: Aphididae) of British Columbia 21. further additions. – J. Entomol. Soc. Brit. Columbia 90: 77-82.

146

Chan, C.K., Forbes, A.R. & Raworth, D.A. 1991: Aphid-transmitted viruses and their vectors of the world. – Technical Bulletin 1991-3E. Vancouver, British Columbia: Research Branch, Agriculture and Agri-Food Canada. Ciuffo, M., Pettiti, D., Gallo, S., Masenga, V. & Turina, M. 2005: First report of Blueberry scorch virus in Europe. – Plant Pathol. 54: 565. DeMarsay, A., Hillman, B.I., Petersen, F.P., Oudemans, P.V. & Schloemann, S. 2004: First report of Blueberry scorch virus on highbush blueberry in Connecticut and Massachusetts. – Plant Dis. 88: 572. Hudgins, H. 2001: Survey of Blueberry scorch virus in highbush blueberries in British Columbia, 2000. – Can. Plant Dis. Surv. 81: 144-147. Loebenstein, G. & Raccah, B. 1980: Control of non-persistently transmitted aphid-borne viruses. – Phytoparasitica 8: 221-235. Lowery, D.T., Sears, M.K. & Harmer, C.S. 1990: Control of turnip mosaic virus of rutabaga with applications of oil, whitewash, and insecticides. – J. Econ. Entomol. 83: 2352-2356. Lowery, T., French, C. & Raworth, D. 2004: Final report on blueberry scorch aphid transmission and host range. – Report to the British Columbia Blueberry Council Research Committee, Abbotsford, British Columbia. Lowery, D.T., French, C.J. & Bernardy, M. 2005: Nicotiana occidentalis: A new herbaceous host for Blueberry scorch virus. – Plant Dis. 89: 205. MacDonald, S.G. & Martin, R.R. 1990: Survey of highbush blueberries for scorch viruses. – Can. Plant Dis. Surv. 70: 96. Martin, R.R. & Bristow, P.R. 1988: A carlavirus associated with blueberry scorch disease. – Phytopathol. 78: 1636-1640. Martin, R.R., MacDonald, S.G. & Podleckis, E.V. 1992: Relationships between blueberry scorch and Sheep Pen Hill viruses of highbush blueberry. – Acta Hort. 308: 131-139. Martin, R.R., Bristow, P.R. & Wegener, L.A. 2006: Scorch and Shock: Emerging virus diseases of highbush blueberry and other Vaccinium species. – Acta Hort. 715: 463-467. Raworth, D.A. 2004: Ecology and management of Ericaphis fimbriata (Hemiptera: Aphididae) in relation to the potential for spread of Blueberry scorch virus. – Can. Entomol. 136: 711-718. Raworth, D.A. & Schade, D. 2006: Life history parameters and population dynamics of Ericaphis fimbriata (Hemiptera: Aphididae) on blueberry, Vaccinium corymbosum. – Can. Entomol. 138: 205-217. Raworth, D.A., Chan, C.-K., Foottit, R.G. & Maw, E. 2006: Spatial and temporal distribution of winged aphids (Hemiptera: Aphididae) frequenting blueberry fields in southwestern British Columbia and implications for the spread of Blueberry scorch virus. – Can. Entomol. 138: 104-113. Raworth, D.A., Mathur, S., Bernardy, M.G., French, C.J., Chatterton, M., Chan, C.-K., Foottit, R.G. & Maw, E. 2007: Effect of kaolin clay on migrant alate aphids (Hemiptera: Aphididae) in blueberry fields in the context of Blueberry scorch virus. – J. Entomol. Soc. Brit. Columbia 104. (in press) Remaudière, G. & Remaudière, M. 1997: Catalogue of the World’s Aphididae. – INRA, Paris. Sterk, G., Jans, K., Put, K., Wulandari, O.V. & Uyttebroek, M. 2003: Toxicity of chemical and biological plant protection products to beneficial arthropods. In: Colloque International Tomate sous abri - Protection Intégréé - Agriculture biologique, Avignon, France 17-19 September, eds. Roche, Edin, Mathieu and Laurens: 113-118. Stretch, A.W. 1983: A previously undescribed blight disease of highbush blueberry in New Jersey. – Phytopathol. 73: 375. [Abstract] 147

Torres, J.B., Silva-Torres, C.S.A. & de Oliveira, J.V. 2003: Toxicity of pymetrozine and thiamethoxam to Aphelinus gossypii and Delphastus pusillus. – Pesq. agropec. bras., Brasilia 38:459-466. van de Veire, M. & Tirry, L. 2003: Side effects of pesticides on four species of beneficials used in IPM in glasshouse vegetable crops : “worst case” laboratory tests. – IOBC/WPRS Bulletin 26(5): 41-50. Wegener, L.A., Punja, Z.K. & Martin, R.R. 2004: First report of Blueberry scorch virus in cranberry in Canada and the United States. – Plant Dis. 88: 427. Wegener, L.A., Martin, R.R., Bernardy, M.G., MacDonald, L. & Punja, Z.K. 2006: Epidemiology and identification of strains of Blueberry scorch virus on highbush blueberry in British Columbia, Canada. – Can. J. Plant Pathol. 28: 250-262. Wegener, L.A., Punja, Z.K. & Martin, R.R. 2007: First report of Blueberry scorch virus in black huckleberry in British Columbia. – Plant Dis. 91: 328. 148 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 149-154

Advances in IPM for black currant

Adrian Harris, Jerry Cross East Malling Research, New Road, East Malling Kent ME19 6BJ UK

Abstract: Black currant gall mite and leaf midge are the most important pests of black currant in the UK. Until recently, they were controlled by three foliar sprays of the synthetic pyrethroid acaricide fenpropathrin per season. However, fenpropathrin is being withdrawn from use throughout the EU in 2008. The gall mite is the vector of reversion virus, which causes sterility in black currant bushes, and is the principal factor limiting the life of plantations. Fenpropathrin is harmful to a very wide range of arthropods, including the natural enemies of gall mite and leaf midge. We aimed to synthesize and validate IPM methods for black currant 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 black currant production. A six year replicated experiment at East Malling Research examined the efficacy of combining early season sulphur sprays with a gall mite resistant variety (Ben Hope) or a gall mite resistant, reversion susceptible variety (Ben Gairn) in an integrated gall mite management programme compared to a reversion virus and gall mite susceptible variety (Ben Alder). The IPM programmes, especially the latter, were highly effective in controlling gall mite and reversion virus. Damaging leaf midge populations did not develop but aphids were a severe problem, especially on Ben Gairn, requiring supplementary sprays of pirimicarb.

Key words: black currant, black currant gall mite, Cecidophyopsis ribis, leaf midge, Dasineura tetensi, fenpropathrin, IPM, Integrated Pest managment

Introduction

Black currant gall mite and leaf midge are the most important pests of black currant in the UK. Until recently, they were controlled by foliar sprays of the broad-spectrum synthetic pyrethroid 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 black currant bushes, and is the principal factor limiting the life of black currant plantations in the U.K. The mite colonises developing black currant 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. A surface deposit is needed. 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 the work reported here was to synthesize and validate IPM methods for black currant 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 black currant production. Previous reports of our work on IPM in black currant crops are given in Cross & Ridout (2001), Cross & Harris (2004, 2005) and Cross (2006).

149 150

Materials and methods

A purpose planted black currant plantation at East Malling Research was used. It comprised a 5 x 5 Latin square of 25 plots, each plot being split 3 ways by variety. Each plot consisted of six rows of six black currant bushes; two rows of each of three varieties in a 3-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 at the whole plot level. 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 IPM experimental plantation at East Malling Research in 2000-2002 and modified treatments applied in 2003-2005.

Insecticide/acaricide sprays Insecticide/acaricide sprays Treatment 2000-2002 2003-2005 2 sprays sulphur (bud-burst, A 1 spray sulphur grape visible) 2 sprays of sulphur (bud-burst, B grape visible), 1 Dursban just 2 sprays sulphur pre-flower 2 sprays sulphur (bud-burst, C grape visible), 1 Aphox pre- 1 spray sulphur, 1 spray Masai flower 3 sprays Meothrin, pre flower, 3 sprays Meothrin, pre flower, D post flower and + 10-14 days post flower and + 10-14 days E Untreated Untreated

Table 2. Products and their dose rates of application

Product Active ingredient and Product dose formulation rate/ha Aphox pirimicarb 50% w/w WG 560 g Dursban chlorpyrifos 480 g/l EC 1 l United Phosphorus Flowable Sulphur sulphur 800 g/l SC 12.5 l Masai tebufenpyrad 20% w/v WB 0.5 kg Meothrin 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 black currant gall mite galls on each bush were counted and recorded. 151

375 Ben Alder (gall mite and reversion susceptible) 300

225

150

0 2000 2001 2002 2003 2004 2005 75 Ben Hope

er bush (gall mite resistant, reversion virus susceptible) p 60 alls g 45 all mite all

g 30

Mean number of 2000 2001 2002 2003 2004 2005 75

Ben Gairn 60 (partially resistant to gall mite, reversion virus resistant)

0 2000 2001 2002 2003 2004 2005 Figure 1. Mean numbers of gall mite galls per bush at the end of each year. A 1 x Sulphur, B 2 x Sulphur, C 1 x Sulphur + 1 x Masai, D 3 x Meothrin, E Untreated.

152

50 Ben Alder (gall mite and reversion susceptible)

0 2000 2001 2002 2003 2004 2005 2006

50 Ben Hope (gall mite resistant, reversion virus susceptible) 40

0 2000 2001 2002 2003 2004 2005 2006 Number of reversion virus infected bushes 60

Ben Gairn 50 (gall mite susceptible, reversion virus resistant)

0 2000 2001 2002 2003 2004 2005 2006 Figure 2. Total number of bushes (n = 60) showing symptoms of reversion virus at the early flowering growth stage each year. A 1 x Sulphur, B 2 x Sulphur, C 1 x Sulphur + 1 x Masai, D 3 x Meothrin, E Untreated.

153

Results and discussion

At the end of year 1, small numbers of gall mite galls were present in all of the treatments of Ben Alder (Figure 1). These increased year-on-year until, after the six years, almost all bushes of Ben Alder were infested. Treatments A, B and C that received sprays of sulphur did have significantly lower levels of gall mite than treatment D (3 x Meothrin) and the untreated control which had the highest numbers of galls. Ben Gairn, performed much better with very few galls present in year 1, numbers being too small for useful statistical analysis. Numbers of galls started to increase on Ben Gairn in year 2 with similar spray treatment effects being observed as in Ben Alder. No galls were present on Ben Hope at the end of years 1 and 2, infestation only starting in year 3. Again the same treatment trends occurred with significantly lower levels of infestation on treatments A, B and C. No reversion virus was found in years 1 or 2 (Figure 2). 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 had received treatment C and 1 bush which had received treatment D were found to be infected. No infection was observed on the other varieties at that time. 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 appeared to remain 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. However, the presence of infection in these bushes was not confirmed in DNA tests at SCRI. 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. This 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 black currant 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. The results clearly show that the variety Ben Hope was highly, though not completely, resistant to gall mite and that Ben Gairn was completely resistant in this trial. 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 results indicate Ben Gairn is the most gall mite resistant variety followed by Ben Hope which has a significantly lower susceptibility to gall mite than the Ben Alder standard. 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 apparently 100 % control of reversion virus. This work has shown it is important that varietal 154

resistance is protected by sprays of acaricides to lessen the likelihood of the development of resistance breaking strains of gall mite and/or reversion virus disease. The gall mite IPM methods were validated in commercial practice in a large scale experiment in a commercial plantation. The 3 x Meothrin and the IPM programmes performed similarly for gall mite control, but both were inadequate because the plantation was too heavily infested with gall mite galls at the outset. 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.

Acknowledgements

This work was funded by the UK Department for environment Food and Rural Affairs and the UK GlaxoSmithKline black currant growers research fund (Project HH3115TSF). We are most grateful to Rob Saunders and James Wickham for their enthusiastic support and encouragement and to the Scottish Crops Research Institute who did the tests for reversion virus on Ben Gairn.

References

Cross, J.V. 2006: Further investigation of IPM methods for black currant gall mite and leaf midge. – Proceedings of 5th IOBC workshop on Integrated Soft Fruit Production, Stavanger, Norway, October 2005. IOBC/WPRS bulletin 29 (9): 131-138. Cross, J.V. & Harris, A.L. 2004: Exploiting natural enemies in Integrated Pest Management in black currant crops. –IOBC/WPRS Bulletin 27 (4): 9-16. Cross, J.V. & Harris, A.L. 2005: IPM in black currants. – Plant it. Issue 8 August 2005: 4-5. Cross, J.V. & Ridout, M.S. 2001: Emergence of the black currant gall mite (Cecidophyopsis ribis) from galls in spring. – Journal of Horticultural Science and Biotechnology 76: 311- 319.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 155-159

Sub-lethal exposure of honey bees to crop-protection: Feeding behaviour and flower visits

Bruno Gobin1, Kevin Heylen2, Johan Billen2, Roger Huybrechts2, Lut Arckens2 1 Zoology Department, pcfruit; Fruittuinweg 1, B-3800 Sint-Truiden, Belgium; 2 Zoological Institute, K.U.Leuven, Naamsestraat 59, B-3000 Leuven, Belgium

Abstract: Production quantity and quality in soft fruit and top fruit depend on optimal bee pollination, but also on the use of crop protection agents. The implementation of strict regulations and pre-flower intervals have greatly reduced acute bee-poisoning incidents. In recent years, concern grew among fruit growers and beekeepers about possible behavioural modifications of residual doses. As part of a research project on sub-lethal effects, we investigated whether some common plant protection products repel bees. We compared consumption of contaminated sugar solution in the lab and counted bee visits to flowers in a field trial in which test products were sprayed either before or during flowering.

Key words: pollination, sublethal effects, feeding experiment, Apis mellifera

Introduction

Bee pollination is essential for many crops including the soft fruits to guarantee maximal fruit quality (shape and size) and quantity (yield weight) (Klein et al. 2007). Fruit growers in Belgium depend mostly on non-commercial beekeepers to provide sufficient colonies for adequate pollination. Under optimal circumstances, beekeepers and fruit growers have mutual benefits from this cooperation as both honey and fruit yields increase. Fruits can however not be grown commercially without the use of crop protection agents, either from organic or chemical origin. All registered products are thoroughly screened for their lethal effect on bees, and bee safety or toxicity must be indicated on the label. The use of products with toxic effects is banned during flowering, and precise pre-flowering intervals have to be respected in Good Agricultural Practice. Positive actions at all levels from registration authorities, agrochemical industries, research institutes, extension services and of course growers and beekeepers alike have contributed to a considerable decrease in acute bee poisoning incidents. At present, few incidents occur, a trend similar to that in other European countries (Fletcher & Barnett 2003; Seefeld 2005). However, all involved parties must remain vigilant to ensure continued safe use of bees as pollinators. In recent years, concerns arose over sub-lethal effects (i.e. behavioural changes) of chemical crop protection on bees. Recent winter losses were attributed to indirect exposure to low doses of insecticides, such as nectar contamination with systemic compounds. Specific behavioural effects can indeed be induced when individual or small groups of bees are exposed to sub-lethal doses of insecticides. Bees can show decreased activity or mobility (Medrzycki et al. 2003), reduced orientation capacity (Colin et al. 1998; Kirchner 1999), or lowered ability to judge food quality or availability (Bortolotti et al. 2003; Decourtye et al. 2003), repulsion (Waller et al. 1984) or shorter life expectancy (Smirle et al. 1984). However, interpretation of these effects in the context of long-term colony level effects proved to be both difficult and controversial. Experiments with exposure at a colony level (i.e. (Faucon et al. 2005) did not show significant differences of various colony parameters between treated

155 156

and untreated colonies throughout the season. At present, a growing consensus exists that the cause of bee losses are likely multifactorial, including bee diseases, parasites, weather conditions, pesticides exposure, biodiversity, … (Lewis et al. 2007). Still, even when no clear cause and effect can be found, it is just common sense to make certain that we keep potential pesticide effects minimal at all times. From the fruit growers’ viewpoint, potential short-term effects on foraging behaviour are relevant as they can bear on pollination efficacy. Fruit quality and production quantity can thus be directly affected if correct pre-flower intervals are not respected. However, growers sometimes interpret reduced fruit set following incorrect use as a sign of repellence instead of a harmful effect on bees, and thus considered less dangerous. This is a dangerous line of thought, as it requires bees to actively avoid contaminated flowers, an unlikely situation in large homogeneous fields or orchards. As part of an ongoing research project on sub-lethal effects on bees, we conducted a few lab and field trials to investigate whether repellence is a true phenomenon, or whether bees will feed equally on treated versus untreated food sources. We observed this either with maximal bee exposure (feeding) or in-crop exposure (treatments at flowering) of a selection of chemicals with various degrees of bee toxicity. In the field trials, we wanted to determine if sub-lethal contamination of crop protection agents used before or after start of flowering might affect foraging behaviour and thus pollination. In the field experiments, we specifically focused on direct effects on visits to flowers, and determined whether potential side effects in the field might be lower due to repellence. As a more practical goal, we use our data to convince growers to respect GAP recommendations to minimize side effects on bees and pollination. The main aim of the laboratory experiments was to verify consistency of exposure throughout a series of behavioural experiments that will be reported elsewhere. The secondary goal of both trial types was to provide data to demonstrate to growers that bees are not likely to decrease feeding on contaminated carbohydrate sources, and exposure risk is a practical reality. In this paper we present the data concerning these feeding trials.

Material and methods

In all trials, 4 plant protection products were used at either the standard registered dose rate in Belgium for field applications, or at a sub-lethal dose rate in lab feeding trials (Table 1).

Table 1. Dose rates of four plant protection products in lab and field trials.

Test product control imidacloprid indoxacarb fenoxycarb captan Formulated product Confidor Steward Insegar Fytocap water 200SL 30WG 25WG 80WG Lab: concentration in - 3.5 ppb 5.1 ppm 1000 ppm 120 ppm sugar solution Field: Dose per ha - 46 g a.i. 51 g a. i. 100 g a.i. 1200 g a.i. leafwall area in water* * leaf wall area is accurately calculated taking the height and length of the sprayed plot.

Field repellence trial Two field trials were performed in spring 2007, to determine the effects on flower visits of crop protection applications at reasonable interval before or during flowering. We chose to perform these trials in two Jonagold apple orchards at different localities, as apple flowering 157

period is sufficiently short for detailed daily observations throughout the flowering period. We do not expect crop differences, as repellence is a behavioural response of the bees to a contaminated flower. Each test product was applied at two time points in the trial, being either at green bud stage when flower petals are not yet exposed (6 April 2007), or at blossom start, when 10% flowers were open (16 April 2007). Thus each combination of “test product” x “application timing” was a different test object. Test objects were applied at registered dose rates in a randomised block design over two test rows with 4 replicates per test object and 5 trees per replicate. We counted bees visiting flowers in each replicate during 1 minute 4-5 times per day. On either side of a test row, one row was left untreated with any crop protection prior to and during the trial. Pollination hives were present within a range of 150 m from the trial site at both locations. We tested the hypothesis that spraying of flowers at the start of flowering would deter bee visits compared to untreated trees or those treated before flower.

Laboratory contamination trial In lab trials known numbers (range 10-15) of bee workers were confined to metal bee test cages lined with filter paper for oral contamination. A standard concentration of sugar solution was prepared from either water (control) or watery solutions of sub-lethal concentrations of the test products (see table 1 for dose of the test products). Test products were transferred to pierced Eppendorf feeding tubes, and weighed to the nearest tenth of a mg before and after exposure. Each bee cage had three Eppendorf-type feeding tubes to ensure access to the carbohydrate source ad libitum. All test products and controls were run simultaneously for 24h in a climate chamber at 25°C. Bees were either 1-week old, identifiable from regular paint marking of freshly hatched bees in our hives, or random bees of unknown age taken from the hive’s combs.

Results and discussion

Bee lab feeding experiments We observed no difference in mean feeding quantity per bee when bees were offered sugar secretion with sub-lethal doses of the tested compounds, compared to control when substances were provided during 24h. We did find large differences in total feeding quantity between one-week old bees, and bees of unknown age (Figure 1). As bees for the latter group were taken randomly from the combs, they likely include older, more experienced bees. However, as with younger bees, treatment effects were absent. Our trial set-up was not a choice-experiment in which bees could choose between contaminated and non-contaminated carbohydrate sources, but rather aimed at measuring whether bees would adapt their sugar consumption to minimal uptake when confronted with a contaminated food source. If crop protection agents are applied during flowering on large surface crops such as strawberries or top fruit, bees are indeed not likely to encounter alternative, uncontaminated food sources.

Field trials When applied in the field at two time points (10 days before flowering and at the start of flowering), we found no significant difference in flower visits between any treatment at either time-point and the water-treated control. Especially the lack of differences between two treatment time points for a given product is striking (Figure 2). At the first date, corresponding to green bud stage, flower organs are still fully protected from spray residue, while petals and part of the reproductive organs are fully exposed at the second time point. Surrounding leaves are however fully treated at both time-points, but even then one would expect a certain degradation of the products under weather exposure for the first treatment. In this randomized 158

block design trial setup, foraging bees could potentially choose between treated and untreated flowers, or even move out of the trial rows into untreated rows. So also in the field, repellent effects seem minimal, even when no pre-flower interval is respected. In sum, bees are not repelled by contaminated food sources. They do not reduce feeding when forced to feed on contaminated sugar secretions and they do not avoid contaminated flowers in a field trial. Reduced exposure risk because of repellence is thus not likely to occur in practice. These findings underline the importance to respect good agricultural practice guidelines and label recommendations on pre-flower intervals to limit bee exposure risk and continue optimal cooperation between bee-keepers and fruit growers. 0,5

0,4

0,3

0,2

0,1

0 captan fenoxycarb imidacloprid indoxacarb control

1 week old (N=15) unknown age (N=4)

Figure 1. Mean consumption of contaminated or control sugar secretion per bee in caged bee trials.

60 orchard 1 orchard 2 50

0 check Imidacloprid Captan Indoxacarb Fenoxycarb Imidacloprid Captan 10% Indoxacarb Fenoxycarb green bud green bud green bud green bud 10% flow ers flow ers 10% flow ers 10% flow ers open open open open

Figure 2. Mean number of bees visiting flowers of plants exposed to different crop protection products before and at the start of flowering. Neither orchard showed statistical differences between treatments (ANOVA p= 0.3 and 0.4).

Acknowledgements 159

This research was funded through IWT Agronomical Research Project nr. 04/669. We thank Eddy Willems for assistance in data collection.

References

Bortolotti, L., Montanari, R., Marcelino, J., Medrzycki, P., Maini, S. & Porrini, C. 2003: Effect of sub-lethal imidacloprid doses on the homing rate and foraging activity of honey bees. – Bull. Insectol. 56: 63-67. Colin, M.E., Le Conte, Y., Di Pasquale, S., Bécard, J.M. & Vermandère, P. 1998: Effet des tournesols issus de semences enrobées d'imidacloprid (Gaucho) sur les capacités de butinage de colonies d'abeilles domestiques. – Unpublished study report INRA. Decourtye, A., Lacassie, E. & Pham-Delègue, M.H. 2003: Learning performances of honeybees (Apis mellifera L.) are differentially affected by imidacloprid according to the season. – Pest Manag. Sci. 59: 269-278. Faucon, J.P., Aurières, C., Drajnudel, P., Mathieu, L., Ribière, M., Martel, A.C., Zeggane, S., Chauzat, M.P. & Aubert, M.F.A. 2005: Experimental study on the toxicity of imidacloprid given in syrup to honey bee (Apis mellifera) colonies. – Pest Manag. Sci. 61: 111-125. Fletcher, M. & Barnett, L. 2003: Bee pesticide poisoning incidents in the United Kingdom. – Bull. Insectol. 56: 141-145. Kirchner, W.H. 1999: Mad-bee-disease? Sublethal effects of imidacloprid (Gaucho) on the behaviour of honey bees. – Apidol. 30: 422-423. Klein, A.M., Vaissière, B., Cane, J.H., Cunningham, S., Kremen, C., Steffan Dewenter, I. & Tscharntke, T. 2007: Importance of crop pollination in changing landscapes. – Proc. R. Soci. Lond. B 274: 303-313. Lewis, G., Thompson, H.M. & Smagghe, G. 2007: In focus: Pesticides and honeybees – the work of the ICP-BR Bee Protection Group. – Pest Manag. Sci. 63: 1047-1050. Medrzycki, P., Montanari, R., Bortolotti, L., Sabatini, A.G., Maini, S. & Porrini, C. 2003: Effects of imidacloprid administred in sub-letal doses on honey bee behaviour. Laboratory tests. – Bull. Insectol. 56: 59-62. Seefeld, F. (ed.) 2005. Chemical study of bee damages caused by plant protection products. – In: Das „Bienensterben“ im Winter 2002/2003 in Deutschland – Zum Stand der wissen- schaftlichen Erkenntnisse. R. Forster, E. Bode & D. Brasse (eds.), Bundesamt für Verbraucherschutz und Lebensmittelsicherheit, Braunschweig: 99-113. Smirle, M.J., Winston, M.L. & Woodward, K.L. 1984: Development of a sensitive bioassay for evaluating sublethal pesticide effects on the honey bee (Hymenoptera:Apidae). – J. Econ. Ent. 77: 63-67. Waller, G.D., Erickson, B., Harvey, J.J. & Archer, B.T. 1984: Comparison of honey bees (Apis mellifera) loses from 2 formulation of methyl parathion applied to sun flowers (Helianthus annus). – J. Econ. Ent. 77: 230-233. 160 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 161-163

Raspberry certification: How it benefits the raspberry sector?

Cathy Eckert, Christophe Calvin Ctifl Lanxade, 24130 Prigonrieux, France; Ctifl Balandran, BP 32, 30127 Bellegarde, France

Abstract: In 2000, a raspberry plant certification scheme was established by Ctifl, under the aegis of the French Ministry of agriculture, in order to provide growers with healthy, quality plants. This project was a success thanks to close cooperation between research and experiment partners, technicians, growers and nurseries. Over the last seven years, the share of certified plants has been on the rise, as has the number of varieties: Meeker and Heritage were the first varieties to be certified. Tulameen followed in 2005, and Polka is scheduled to enter the certification scheme in 2007. Currently, research is focusing on the means to include Phytophtora fragariae and crumbly berry in the certifiable production schemes.

Key words: certification, plant, raspberry, varieties

Introduction

At the beginning of the 80ies, raspberry growers had lot of problems with raspberry plants because of crumbly berry, Phytophthora and most of all, mixture of different varieties. Since 1991, research programs have showed that some clones had got more crumbly berry than others. A more rigorous production scheme had to be established. So, growers asked their partners, INRA, Ctifl and plant protection services, to organize a raspberry certified plant production in France (Navatel & Maillard, 2000).

Production scheme of certified raspberry plants and results

The production scheme can be summarized as follow. In a first step, Ctifl ensures the supply of prepropagation plants issued from initial plants kept in pots in screen houses (Ctifl Lanxade station). At fruiting, this material is visually inspected (crumbly fruit test). Then the plants enter a micropropagation phase in an approved laboratory. After an acclimatization phase in nursery where the potted plants are visually inspected during vegetation phase, four different itineraries can be followed:

Itinerary 1: Direct production of soilless in vitro plants* Itinerary 2: Planting of propagation plants in open-field, then suckering** – suckering** – canes** Itinerary 3: Planting of propagation plants in open-field, then suckering*** – root cutting then head cuttings in pots – soilless plants* Itinerary 4: Planting of propagation plants in open-field, then suckering*** – root cutting – canes**

* inspection of potted plants during vegetation stage; **inspection of plants during vegetation phase; ***inspection of propagation plants in the plots during the vegetation phase

161 162

The time required to integrate a variety into this scheme varies from five to six years: three years of sanitary testing at the Ctifl Lanxade station then two or three years of propagation as described in the scheme. In 2006, three varieties and more than 150’000 plants have been put at the disposal of the growers (Table 1). The volume of certified plants reaches 150’000 to 200’000 plants per year on a total of ca. 1.5 millions plants produced in France for raspberry growers (Figure 1). The production of certified plants is stable. The different controls, made by the Ctifl under the aegis of the French Ministry, bring to growers guaranties on the variety authenticity, on the plant homogeneity and on the healthy quality plant.

Table 1. Numbers of plants certified in 2006 per variety. Soilless-raised Soilless Varieties Suckers Total in vitro plants plants Meeker 50 000 28 200 0 78 200 F1010 Heritage 51 500 12 120 0 63 620 F1007 Tulameen 16 750 16 750 F 722 101 500 57 070 0 158 570 Total 64% 36 % 100%

300'000 Tulameen Heritage Meeker 200'000

100'000 Number of certified plan

0 2000 2001 2002 2003 2004 2005 2006

Figure 1. Evolution of certified production per variety from 2000 to 2006.

163

Perspectives for the certified material

Management of Phytophtora fragariae pv. rubi Some criteria for plot selection, in order to avoid risks of contamination by P. fragariae pv. rubi have been established:

- light, free-draining soil that does not receive the run-off from other plots; - use of distributed water for irrigation and treatments, no water from hill reservoirs; - no Rubus crop since at least 10 years; - propagation plots must be at a distance of at least 20 m from any Rubus to avoid contact with suckers of wild Rubus or uncontrolled raspberry.

A project (2006 to 2008) is under way between Ctifl, INRA, regional experimental stations and nurseries. The ultimate aim is acceptance of the four plant production itineraries presented in Table 1, in order to guarantee the absence of Phytophthora. Nurseries may then choose the production itinerary that suits them best. A PCR test for detection of the disease is carried out at the various plant production stages, during two production seasons, spread over three years. The project aims to show that the various tested itineraries offer a sufficient guarantee against Phytophthora.

Management of crumbly fruit Several possible causes have been listed: sanitary (RBDV), climatic (poor pollen quality), genetic. The latter hypothesis requires more extensive research (chromosome counting) in order to achieve a better understanding of floral morphogenesis in raspberry. Every year, initial plants are tested for crumbly fruits, after growing in a production plot. A research program is beginning this year, at Montpellier INRA, about the genetic origin of crumbly berry.

Conclusions

Raspberry certification is a success thanks to close cooperation between researchers, technicians, growers and nurseries. The implementation of the plant certification scheme for raspberry in France has considerably improved crop homogeneity and durability. Growers would wish all new variety could be certified. Further efforts are required on Phytophtora and on crumbly fruits.

References

Navatel, J.C. & Maillard, F. 2000: La certification du framboisier: situation actuelle et perspectives. – Infos Ctifl 160: 41-45. 164 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 165-167

Raspberry root rot control in the Scottish raspberry certification scheme

Alexandra Schlenzig Scottish Agricultural Science Agency, 1 Roddinglaw Road, Edinburgh EH12 9FJ, UK

Abstract: The Scottish raspberry certification scheme demands freedom from Phytophthora fragariae var. rubi in all grades of raspberry propagation stocks. This paper compares bait test and nested PCR as test methods for this pathogen and recommends their use for the different grades of the certification scheme.

Key words: Phytophthora fragariae var. rubi, certification scheme, raspberries

The Scottish certification scheme

Certification schemes ensure that growers are provided with planting material of known health standard and purity. They are an important tool to prevent the spread of harmful pests and diseases. In Scotland raspberry certification has a long history, starting in 1930, when inspection was limited to trueness to type and apparent vigour and health at the time of inspection (Smith 2003). In the Eighties Phytophthora fragariae var. rubi was identified in Scottish raspberry plantations as being the major pathogen causing raspberry root rot. In response to the high incidence and the devastating effects of this disease, the Scottish Government made raspberry certification mandatory in the Soft Fruit Plants (Scotland) Order 1991 to improve the health of raspberry planting material. However, in 1995 the Order was replaced with the EC Marketing of Fruit Plant Material Regulation and the Scottish certification scheme became voluntary again.

Table 1. Standards of certification for raspberry spawn beds.

Isolation and Separation Disease tolerances (distance in meter) 1) 1)

2) 2) ubus From Fruit Planta-tions Soil borne virus Grade Max. age of spawn bed Purity From SE From E From S From wild R Rasp. Root Rot RBDV Other virus SE 4 yrs 100% 4 100 250 250 1000 0% 0% 0% 0% E 4 yrs 100% 100 4 100 100 100 0% 0% 0% 0.5% S 4 yrs 99.5% 250 100 4 4 4 0% 0% 0% 2% 1) Super Elite (SE), Elite (E), Standard (S) 2) nematode transmitted viruses

The origin of all certified planting material is the Nuclear Stock which is held at the Scottish Crop Research Institute (SCRI) under strict containment. The following stages of

165 166

propagation are called Foundation Grade, Super Elite, Elite, and Standard. The latter three grades are field grown with Elite or Standard usually the grades used by growers to set up fruit plantations. Tolerances for certain diseases apply for the lower grades of the certification scheme. However, for Phytophthora fragariae var. rubi there is a nil-tolerance applied throughout all stages (Table 1).

Testing methods for Phytophthora fragariae var. rubi used in the Scottish certification scheme

To ensure freedom from pathogens, the certified material is inspected and tested regularly. Testing is done either by SCRI (for the Nuclear Stock) or by the Scottish Agricultural Science Agency (SASA) for the other stages. Nuclear Stock material and Foundation grade material is tested annually for Phytophthora fragariae var. rubi. The plants are tested for the latent presence of the pathogen, since symptoms have never been observed in these grades. Field grown stages are inspected annually in July/August but tested only if symptoms are present which indicate the possible presence of the pathogen. Additionally, there is a voluntary test offered to the growers in winter. Since inspection for symptoms is not possible for this test due to the dormant state of the plants, a statistical sampling method is used for taking roots. The European and Mediterranean Plant Protection Organization (EPPO 1994) published guidelines for a Rubus certification scheme. Here the recommended method for testing for Phytophthora spp. in raspberry roots is a bait test as described in Duncan et al. (1993). Although this test is sensitive and reliable it has some disadvantages: it takes about six weeks to complete and a microscopic diagnosis is necessary at the end, at least for the plants developing symptoms. This requires considerable expertise. Moreover, the test is quite destructive for single test plants when the necessary amount of roots is sampled. This is impractical for testing Nuclear Stock plants where the roots are needed for further propagation. Therefore, the SCRI developed a nested PCR method for the testing of the Nuclear Stock material. SASA tested this PCR method in comparison to the bait test in a survey in winter 2002/3. The correlation between bait test and PCR was generally good. Both methods indicated the same infected four fields but within these fields the bait test found 12 infected samples compared to 10 found by nested PCR (Schlenzig et al. 2005). Sensitivity of both tests was determined by mixing various proportions of infected roots into healthy roots to make up a 300g sample. The result for the nested PCR was not consistent. In one test it was less sensitive than the bait test, in another it was more sensitive.

Conclusions

There are numerous factors influencing the choice of method in plant pathogen diagnosis. PCR has many advantages compared to traditional methods: it is quick, does not require expert knowledge of plant pathogens, and in most cases it is more sensitive than bioassays. On the other hand is it expensive and does not distinguish whether the detected pathogen is dead or alive. The problem with the PCR for field samples is the restricted sample size that is proccessable for the DNA extraction. This makes it necessary to take a subsample which represents only a very small proportion of the original sample. A sample from an infected field will contain a mixture of healthy roots and infected roots. The lower the level of infection the bigger becomes the likelihood that a small subsample taken for DNA extraction 167

does not contain any infected roots, leading to a false negative result. Increasing the number of subsamples does lower this risk but field samples tend to be large and there is a limit of how many subsamples are practical. Within the Scottish certification scheme the nested PCR is the method of choice for Nuclear Stock material and Foundation material because it is less destructive and the samples are small enough to give reliable results. During the growing season field inspections the samples are from plants showing symptoms. Therefore, also the nested PCR is used because it is used for confirmation rather than detection. For the winter field test though, the bait test is the preferred method because bigger samples are taken and can be processed and the likelihood of finding the pathogen is increased.

References

Duncan, J.M., Kennedy, D.M., Chard, J., Ali, A. & Rankin, P.A. 1993: Control of Phytophthora fragariae on strawberry and raspberry in Scotland by bait tests. – In: Plant Health and the European Single Market. Ebbels, D. (ed.): 301-305. EPPO 1994: Certification scheme. Pathogen-tested material of Rubus. – EPPO Bulletin 24: 865- 873. Schlenzig, A., Cooke, D.E.L. & Chard, J.M. 2005: Comparison of a baiting method and PCR for the detection of Phytophthora fragariae var. rubi in certified raspberry stocks. – EPPO Bulletin 35: 87-91. Smith, V.V. 2003: The role of certification schemes in integrated crop management of soft fruit in Scotland. – IOBC/wprs Bulletin 26 (2): 7-10. 168 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 169-176

Biofumigation to control Verticillium wilt of strawberry: Potency and pitfalls

Vincent V. Michel1, Sharmila Dahal-Tscherrig1, Habib Ahmed2, Arnaud Dutheil3 1 ACW Agroscope Changins-Wädenswil, Centre des Fougères, CH-1964 Conthey, Switzerland; 2 HES-SO, University of Applied Science Western Switzerland, CH-1950 Sion, Switzerland; 3 ENITA Clermont-Ferrand, BP 35, F-63370 Lempdes, France

Abstract: The efficacy of biofumigation to reduce the soil inoculum of Verticillium dahliae, causal agent of Verticillium wilt, was tested in several field and pot experiments. The reduction of the number of microsclerotia ranged from 19 to 85%. This variation is due to the biological nature of this method, several factors can influence its efficacy. Among the most important ones are plant species, incorporation of plants, and soil type.

Key words: biofumigation, biological method, Brassica juncea, soil type, Verticillium dahliae

Introduction

Verticillium wilt, caused by Verticillium dahliae and V. albo-atrum (Pegg & Brady 2002) is a major soil-borne disease of strawberry (Ellis et al. 1997). The broad host range of Verticillium wilt covers at least 74 families, including several important agricultural crops and weeds (Pegg & Brady 2002), makes it difficult to reduce soil inoclulum of Verticillium spp. by the means of crop rotation. The use of Verticillium wilt resistant cultivars can be restricted by other traits such as quality, yield or resistances to pests or other diseases. In Switzerland, no fungicide is registered to control Verticillium wilt and soil fumigants are prohibited since 1989. The phasing out of methyl bromide, the probably most effective means of control of Verticillium wilt, initiated research for alternatives to this fumigant (Gulino et al. 2003) already before its prohibition in 2005 (Fedorowicz 2005). Biofumigation using cruciferous plant is one of these alternatives and was mainly developped in countries with an intesive use of methyle bromide such as Australia, Italy and the USA (Kirkegaard & Matthiessen 2004; Morra 2004; Lazzeri et al. 2004). This method is based on the transformation of glucosinulates in the tissues of cruciferous plants into isothio- and thiocyanates after the incorporation of the plants in the soil. Isothio- and thiocyanates are volatile molecules that are toxic to a number of soil-borne pathogens (Matile 1980; Rollin & Palmieri, 2004). During the last four years, a series of field and pot trials were conducted at Agroscope ACW and on two farmer's fields to evaluate the potential of biofumigation to reduce the inoculum of Verticillium wilt in the soil. During these trials, some limits of the method for its pratical application were encountered and will also be reported in this publication.

Material and methods

Field experiment In 2003, a field experiment was conducted at the Agroscope ACW Domaine de Bruson, an experimental site situated at 1080 m asl near the Grand St-Bernard (Valais, Switzerland) in a

169 170

field highly contaminated with Verticillium spp. The experimental layout was a RCBD with 4 replications, each experimental plot measured 5 x 1.5 m. The experiment included brown mustard, red clover (Trifolium pratense) and rye (Secale cereale) as green manures, two composts as soil amendments, a nitrogen fertilizer and a control treatment. In August 2003, the experiment started with the sowing of brown mustard (culitvar ISCI-20, www.bluformula.com). In October 2003, this biofumigation crop was incorporated in the soil at the flowering stage by two passages with a rototiller. During winter season, plots were left fallow. Red clover and rye were resown in spring 2004 after the sowing in automne 2003 failed (poor seed quality). In April 2004, both green manures were mulched and incorporated in the soil. At the same time, the two composts amendments were applied. Both composts were windrow compost, produced at the same site but different in age i.e., six weeks old (young compost) and more than four months old (old compost). Two weeks later, 20 strawberry plants (Elsanta) were planted per experimental plot with a within row distance of 25 cm. Plants were irrigated with a drip irrigation and no fertigation was applied. P2O5, K2O and Mg fertilizer was added before planting to all treatments except the two compost treatments. Only the nitrogen fertilizer treatment received nitrogen in form of ammonium nitrate, split in three dosis of 28 kg N/ha in summer 2005. Leaf diseases and pests were controlled by standard fungicide and insecticide applications. The experiment ended after fruits were harvested in summer 2005. Soil samples for the determination of the number of V. dahliae microsclerotia were taken in Mai 2005.

On-farm trials In 2005, trials were conducted on two vegetable grower farms located in the central valley of Valais, close to Agroscope ACW Centre des Fougères. At both sites, plots mainly cropped with sweep pepper (Capsicum annum) were contaminated with V. dahliae. Trials with the brown mustard cultivar ISCI-20 were set up in such plots. On an organic farm at Riddes (Favre trial), two adjacent plots were used for the trial. On one plot, brown mustard was sown end of April, the second plot was left fallow. At mid-june, the brown mustard plants were finely mulched with a flail mower at flowering stage and immediately incorporated with a rototiller. Half of the plot with the brown mustard was covered directly after incorporation with a transparent plastic film for one day. Soil samples for V. dahliae microsclerotia counts were taken one week after incorporation. Sweet pepper plants (cultivar "Red beefhorn", an old cultivar highly susceptible to Verticillium wilt) were transplanted the day after soil sampling. Both plots were covered with plastic tunnels. Assessment of Verticillium wilt was done in mid-October 2005. On a conventionnal farm at Saxon (Egg trial), one plot was used for the trial. Brown mustard was sown on the whole plot beginning of Mai and was finely mulched with a flail mower at flowering stage and immediately incorporated with a rototiller at beginning of July. Sweet pepper plants (cultivar Somborka, susceptible to Verticillium wilt) were transplanted one week after incorporating. Soil samples for determination of V. dahliae microsclerotia were taken one day before incorporating brown mustard and one day after transplanting sweet pepper. During the whole trial the plot was covered with a plastic tunnel. Assessment of Verticillium wilt was done in mid-October 2005.

Pot experiments In 2006, two pot trials were conducted in the greenhouse of Agroscope ACW Centre des Fougères. In the first experiment, the influence of the plant species and soil coverage was investigated. In the second experiment, the effect of the soil type on biofumigation was tested. In spring 2006, brown mustard cultivars ISCI-99 (very rich in glucosinolates, Table 1) and ISCI-20 (rich in glucosinolates, Table 1) and canola cultivar Talent (very low in 171

glucosinolates) were grown in pots containing a commercial peat substrate. At beginning of flowering, the above ground part of the plants were harvested. Seventy grams of fresh matter containing equal parts of stems, leaves and flowers where then finely chopped with an onion mincer (Zyliss brand) and immediately mixed with 650 ml of soil from the control plot of the Favre trial. The soil-plant mixture was placed in a 1 l plastic pot and was watered until runoff of water. Soil without incorporating plant material served as control treatment. Soil of this treatment was also mixed thourougly before being placed in pots. After incorporating and watering, half of the pots of each treatment were packed in tightly closed airproof polyethylene bags. Four pots were prepared for each plant x plastic combination. Pots were incubated at room temperature in a dark room, plastic bags were removed after two days, and after five days of incubation the soil samples were taken and air-dried. In autumn 2006, two soils naturally infested with V. dahliae were used for a pot experiment (Table 2). The brown mustard cultivar was ISCI-99, experimental parameters and techniques were the same as in the previous pot experiment.

Assessment of V. dahliae microsclerotia in the soil Soil samples from the field and on-farm trials consisted of several bulked subsamples which were thoroughly mixed before air-drying for at least six weeks immediately after sampling. For pot trials, the half the soil of each pot was air-dried for at least six weeks. Dry samples were conserved at room temperature in a dark and dry place until analysis. The number of living Verticillium dahliae microsclerotia per g dry soil was determined with a dry-plating method (Butterfield & DeVay 1977) using Sorensen’s NP-10 selective medium (Kabir et al. 2004). Each soil sample was sieved at 0,5 mm mesh-size and five 100 mg aliquotes were dry- plated on the NP-10 medium using an Anderson sampler similar device. After two weeks of incubation in darkness at 24°C, soil was removed from the medium surface by adding tap water and scraping with a glass slide. The number of V. dahliae microsclerotia forming colonies was counted under a dissecting microscope.

Table 1. Characteristics of the two brown mustard (Brassica juncea) cultivars used for the pot experiment (Source: SOVESCI BLUFORMULA information sheet).

Biomass Glucosinolates Main Cultivar (kg fresh matter ha-1) (micromoles g-1) glucosinolate ISCI-20 110 x 103 13.3 Sinigrin ISCI-99 113 x 103 16.8 Sinigrin

Table 2. Characteristics of the two soils used for the pot experiment.

Soil Origine Sand Silt Clay Organic pH (H O) type (site) (%) (%) (%) matter (%) 2 Silty Riddes 47.5 44.1 8.4 1.7 7.7 Sandy Epines 80.5 14.3 5.2 1.5 7.8

Results and discussion

The biofumigation treatement using brown mustard reduced significanly the number of living microsclerotia by 19% compared to the control in the field experiment at Bruson (Figure 1). 172

In the on-farm trials, the reduction of living microsclerotia was 74% in the Favre-trial (Figure 2) and 48% in the Egg-trial (Figure 3). When soil was covered for one day immediately after incorporation of brown mustard with a transparent plastic film, microsclerotia were reduced in the Favre-trial by 88% compared to the control plot (Figure 2). In the pot experiment comparing different plants, the number of living V. dahliae microsclerotia was significantly reduced by the two brown mustard cultivars (Figure 4). ISCI- 99 with a very high glucosinolate-content reduced microsclerotia by 66%, ISCI-20 with a high-glucosinolate content by 55%. The low glucosinolate canola cultivar Talent had an intermediate effect on V. dahliae microsclerotia which were reduced by 32%. Placing the pots for two days in a polyethylene bag had no effect on the reduction of the number of microsclerotia. In the pot experiment comparing two different soils, the number of living microsclerotia was reduced by 85% in the silty soil (Figure 5), which was the same used for the first pot experiment. In contrast, no significant reduction of V. dahliae microsclerotia occurred in the sandy soil.

800

B B 600 BB B AB A microsclerotia microsclerotia 400

per g dry soil g dry per 200

0 Verticillium dahliae old young red rye brown nitrogen- control compost compost clover mustard fertilizer (no plant)

Figure 1. Number of living microsclerotia per g dry soil on 5 May 2005 in plots of the field experiment at Bruson. Treatments with the same letters are not significant different (Tukey-test at 5%).

In general, incorporating brown mustard cultivars specifically selected for biofumigation reduced the number of living microsclerotia in the soil significantly whereas red clover or rye had no effect. Canola with a low glucosinolate content resulted in an intermediate reduction of V. dahliae microsclerotia, indicating that not only the formation of toxic volatiles was responsible for the decline of the V. dahliae population in the soil. Eventually the incorporation of fresh organic matter also increases the soil microbial activity which in its turn has an impact on the V. dahliae microsclerotia survival in the soil. Down et al. (2004) mentioned that the reduction of V. dahliae after the incorporation of lettuce might be caused by microbial action. The relatively week reduction in the field experiment at Bruson might be explained by two factors. Before incorporation, the brown mustard was not mulched but just cut and then incorporated with a rototiller. However, just rotavating rape or mustard plants instead of mulching them prior to incorporation resulted in a five-fold lower total isothiocyanate 173

concentration in the soil (Matthiessen et al. 2004) which in their turn decreases the efficacy of the biofumigation. Additionally, the brown mustard was incorporated only at beginning of October at a moment when soil temperature dropped below 10°C. Such a reduction might have slowed down the microbial breakdown of the mustard tissues and thereby negatively

250 = standard error

200

150 microsclerotia

100 per g dry sol g dry per

50 Verticillium dahliae Verticillium 0 brown mustard, brown mustard, no plant, not not covered covered with covered with with plastic plastic plastic

Figure 2. Number of living microsclerotia per g dry soil on 23 June 2005, one week after incorporation of brown mustard in the on-farm trial at Riddes (Favre-trial). Half of the plot with brown mustard was covered immediately after incorporation for one day with a transparent polyethylene plastic film. Standard error relates to the five measurements per treatment.

200 = standard error

150 microsclerotia 100 per g dry sol g dry per 50

Verticillium dahliae Verticillium 0 before incorporation after planting sweet brown mustard pepper

Figure 3. Number of living microsclerotia per g dry soil on 30 June 2005 (before incorporation) and on 8 July 2005 (after planting sweet pepper) in the on-farm trial at Saxon (Egg-trial). Standard error relates to the five measurements per treatment. 174

120 average 100 with plastic without plastic 80 microsclerotia microsclerotia 60 per g soil per 40

20 AAAB B

Verticillium dahliae 0 ISCI-20 ISCI-99 canola no plant

Figure 4. Number of living microsclerotia per g dry soil in a pot trial one week after incorporation of plant material: ISCI-20 = brown mustard cultivar rich in glucosinolates; ISCI-99 = brown mustard cultivar very rich in glucosinolates; canola = low glucosinolate-content cultivar (Talent). Treatments with the same letters are not significantly different (Tukey-test at 5%).

60 A brown mustard no plant 40 microsclerotia microsclerotia

a 20 per g dry soil g dry per a B

0 Verticillium dahliae silty soil sandy soil

Figure 5. Number of living microsclerotia per g dry soil in a pot trial one week after incorporation of brown mustard (cultivar ISCI-99) in two different soils. Treatments with the same letters are not significantly different (Tukey-test at 5%), soil types were tested separately.

influence the formation of isothio- and thiocyanantes. Lower temperature also reduces the activity of some of these molecules (Matthiessen & Kirkegaard 2006). Most probably, all these factors together reduced the biofumigation effect at Bruson. 175

In contrast, the reduction of V. dahliae microsclerotia ranging from 48 to 88% in the on- farm trials showed that an important decrease of the pathogen inoculum in field soil can be achieved with biofumigation. Mulching brown mustard prior to incorporation and warm soil temperatures (20°C at 5 cm depth at the moment of incorporation, measured at the meteorological station at Riddes) certainly can explain this higher efficacy. The use of an adapted plant species is another important factor for a successful biofumigation. The two glucosinolate-rich brown mustard cultivars caused a more pronounced decline of the V. dahliae population than the low-glucosinolate canola cultivar. Differences between plant species for their ability to generate isothio- and thiocyanates are known (Kirkegaard & Matthiessen 2004). Differences in the glucosinolate content and composition vary not only between plant species but also between cultivars within one species. Therefore, the use of cultivars with a defined glucosinonlate profile generating a high amount of biocidal active compounds is a cornerstone for the implementation of biofumigation in farmer's field. Implementing of cultural control method of soil-borne diseases need knowledge on the efficacy of the new method in different environments. Testing in only one environment i.e., with only one soil type can be fatal. The efficacy of a mineral soil amendment to control bacterial wilt of tomato caused by Ralstonia solanacearum varied strongly in different environments (Michel et al. 1997). Soil type strongly influenced the effect of this soil amendment (Michel & Mew 1998). Such an effect of the soil type on the biofumigation efficacy also occurred in our pot trial comparing two different soils. Soil texture influences the losses of volatiles (Brown & Morra 2005) and therefore also the efficacy of biofumigation. Studies including a large number of soils with different soil types (texture, organic matter content, pH) are needed to determine in what kind of soil biofumigation is a potential method to control Verticillium wilt and most probably also other soil-borne diseases and pests.

Acknowledgements

We would like to thank André Ançay, Monique Benz, Roger Carron, and Michel Fellay for their support in the field experiment and laboratory analysis. A great thank goes to Lionel Favre and Bernard Egg who allowed us to conduct trials on their farm and for their interest in the biofumigation method.

References

Brown, J. & Morra, M. J. 2005: Glucosinolate-Containing Seed Meal as a Soil Amendment to Control Plant Pests. – Subcontract Report NREL/SR-510-35254, National Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, Colorado 80401-3393, USA. Butterfield, E.J. & DeVa, J.E. 1977: Reassessment of soil assays for Verticillium dahliae. – Phytopathology 67: 1073-1078. Down, G.J., Harris, D.C., Murray R.A. 2004: Destruction of Verticillium dahliae in soil follwoing the addition of sulphur-containing volatile compounds potentially produced from Brassica tissues. – Agroindustria 3: 293-294. Ellis, M.A., Converse, R.H., Williams, R.N. & Williamson, B. 1997: Compendium of raspberry and blackberry diseases and insects. – APS Press, St. Paul, USA Fedorowicz, J. 2005: The Montreal protocol: Partnerships changing the world. – UNDP, New York, USA (http://exchange.unido.org/cmsupload/1509_2791686912_ozone.pdf). 176

Gullino, L.M., Camponogara, A., Gasparrini, G., Rizzo, V., Clini, C. & Garibaldi, A. 2003: Replacing methyl bromide for soil disinfestations: The Italian experience and implications for other countries. – Plant Dis. 87: 1012-1021. Kabir, Z., Bhat, R.G. & Subbarao, K.V. 2004: Comparison of media for recovery of Verticillium dahliae from soil. – Plant Dis. 88: 49-55. Kirkegaard, J. & Matthiessen, J. 2004: Developing and refining the biofumigation concept. – Agroindustria 3: 233-239. Lazzeri, L., Leoni, O., Bernardi, R., Malaguti, L. & Cinti, S. 2004b: Plants, techniques and products for optimising biofumigation in full field. – Agroindustria 3: 281-288. Matile, P. 1980: "Die Senfölbombe": Zur Kompartimentierung des Myrosinasesystems. – Bio- chem. Physiol. Pflanzen 175: 722-731. Matthiessen, J.N. & Kirkegaard, J.A. 2006: Biofumigation and enhanced biodegradation: Opportunity and challenge in soilborne pest and disease management. – Crit. Rev. Plant Sci. 25: 235-265. Matthiessen, J.N., Warton, B., Shackleton, M.A. 2004: The importance of plant maceration and water addition in achieving high Brassica-derived isothiocyanante levels in soil. – Agro- industria 3: 277-280. Michel, V.V., Wang, J.-F., Midmore, D.J. & Hartman, G.L. 1997: The effects of intercropping and soil amendment with urea and calcium oxide on the incidence of bacterial wilt of tomato and survival of soilborne Pseudomonas solanacearum in Taiwan. – Plant Pathol. 46: 600-610. Michel, V.V. & Mew, T.W. 1998: Effect of a soil amendment on the survival of Ralstonia solanacearum in different soils. – Phytopathology 88: 300-305. Morra, M.J. 2004: Controlling soil-borne plant pests using glucosinolate-containing tissues. – Agroindustria 3: 251-256. Pegg, G.F. & Brady, B.L. 2002: Verticillium wilts. – CABI Publishing, Wallingford (UK): 552 pp. Rollin, P. & Calmieri, S. 2004: Sulfur-containing metabolites in Brassicales. – Agroindustria 3: 241-244.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 177-179

The influence of weed covering on short day strawberries in the autumn

Rudolf Faby VBOG Langförden, OVB Jork, Spredaer Str. 2, 49377 Vechta Langförden, Germany

Abstract: In our strawberry industry we have nearly 1000 ha with so called strong cold stored plants for delaying the harvest in the planting year. The plants stay for a second crop in the following year. Because of the reduced use of soil herbicides weed control becomes more difficult, especially for Stellaria media. It covers the whole plant from September until the end of the vegetation period, which leads to reduced light intensity during the period of flower induction and differentiation. To simulate this effect, strawberries were covered with a shading tissue as a single layer and also as a double layer from the 20th of September to the end of November. The trial was carried out for four years (2002 until 2005) with the cultivar Elsanta. The light reduction reduced the yield by 17 to 32% and with the double coverage the reduction ranged between 52 and 65%. There was no effect on the fruit size except for 2006, when the fruits under covering were 2.5 g heavier per fruit.

Key words: weed control, shading, Stellaria media, strawberries

Introduction

In northern Germany in the region between the rivers Weser and Ems there are nearly 1000 ha of strawberries with so called strong cold stored plants for delaying the harvest in the planting year. The plants stay for a second crop in the following year. Because of the reduced use of soil herbicides weed control becomes more difficult, especially for Stellaria media. It covers the whole plant from September until the end of the vegetation period, which leads to a reduced light intensity during the period of flower induction and differentiation.

Materials and methods

To simulate this effect, strawberries were covered with a green shading tissue as a single layer and also as a double layer from 20th of September to the end of November. The trial was carried out for four years (2002 until 2005) with the cultivar Elsanta. The plot length was 6.6 m with 25 plants per plot. There were four plants per m in a single row, with four replications. The distance between the rows was 1 m. We recorded the yield per plant and the average fruit size by counting and weighting of a sample of 500 g every harvest date. The average fruit size was calculated by multiplication with the yield at every picking date. The statistical analysis was done with the t-test. The light intensity was measured with a luxmeter. The unit is lux. One measurement took one minute. The dates in table 1 and 2 are the average of ten different measurements.

Results

Measurements showed that the covering of strawberries with Stellaria media on a day with sunshine can reduce light intensity by 53 up to 83%, depending on the density of the covering

177 178

(Table 1). On a cloudy day the reduction was stronger. Below the shading tissue the reduction underneath the single layer averaged 25% and underneath the double layer 47% on a day with sunshine (Table 2). Under cloudy conditions the reduction was 41 and 73%, respectively. The covering of the strawberry plants with the shading tissue in autumn reduced the yield every year. Under the single layer it varied between 17 and 32% (Table 3, 4, 5) and under the double layer between 52 and 65%. There was no influence of the reduced light intensity on the fruit size except in 2006, where the fruits under the shadow tissue were 2.5 g or 19% larger (Table 5). That means the number of fruits per plant was more strongly reduced than the yield.

Table 1. Light intensity (lux) under Stellaria media.

Light coverage with Stellaria media Date and weather conditions index none medium dense 30.09.2002 lux 40368 18788 6853 clear sky with sunshine relative 100 47 17 29.10.2002 lux 19169 3484 322 cloudy, bad weather relative 100 18 2

Table 2. Light intensity (lux) under shading tissue.

Light coverage with shading tissue Date and weather conditions index none medium dense 30.09.2002 lux 41495 30970 21889 clear sky with sunshine relative 100 75 53 29.10.2002 lux 19657 11677 7316 cloudy, bad weather relative 100 59 37

Table 3. The influence of reduced light intensity in the autumn of 2003 on the yield 2004, start of covering 22.09.03, end of covering 28.11.03

yield fruit size Coverage type g/plant relative g control 743 100 18.3 shading tissue single layer 618 83 18.8 shading tissue double layer 359 48 18.5 LSD (0.05) 76 - ns

Table 4. The influence of reduced light intensity in the autumn of 2004 on the yield 2005, start of covering 21.09.04, end of covering 22.11.04

yield fruit size Coverage type g/plant relative g control 714 100 12.2 shading tissue single layer 489 68 13.0 shading tissue double layer 253 35 12.2 LSD (0.05) 116 - ns 179

Table 5. The influence of reduced light intensity in the autumn of 2005 on the yield 2006, start of covering 22.09.05, end of covering 14.11.05

yield fruit size number of fruits Coverage type g/plant relative g per plant relative control 1069 100 13.3 80 100 shading tissue single layer 796 75 15.8 50 63 shading tissue double layer 402 38 15.8 25 31 LSD (0.05) 130 - 0.3 9 -

Discussion

The reduction of the light intensity during the period of flower induction and differentiation on the short day plant Elsanta reduced the yield in the following year. There were variations between the years but altogether we found a close correlation between the reduced light and the yield. Because the reduction of light intensity under a medium covering with Stellaria media was stronger than under the double layer, the losses in yield per plant can be much higher than 60%. The losses in yield per hectare are normally lower, because not every plant is covered by the weed and very often it started and increased during this period. On the other hand we found that covered plants were less healthy in the winter and very often additionally damaged by mice, which live under the weed. So it is essential to protect the strawberry plants from becoming covered by Stellaria media.

180

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 181-187

Evaluation of alternative chemicals for control of botrytis in raspberry

Angela Berrie1, Tim O’Neill2, Erika Wedgwood2, Barbara Ellerker1 1 East Malling Research, New Road, East Malling, Kent, ME19 6BJ, UK; 2 ADAS Horticulture Ltd, ADAS Boxworth, Battlegate Road, Boxworth, Cambridge, CB23 4NN, UK

Abstract: In 2006 and 2007 three replicated experiments were conducted on two sites (Kent 2006 and 2007; Cambridgeshire 2006) in which fungicides and alternative chemicals were evaluated for control of botrytis fruit rot in the summer-fruiting raspberry Glen Ample. The experiments were on open field (Kent) and protected (Cambridgeshire) crops. In all three trials the incidence of botrytis at harvest was very low, but rapidly developed in fruit samples post-harvest. In 2006 at both sites fungicides, in particular UKA379, UKA374, Talat (tolylfluanid + fenhexamid), Signum (pyraclostrobin + boscalid), Switch (cyprodonil + fludioxonil), Scala (pyrimethanil), Amistar (azoxystrobin) and Folicur (tebuconazole,) gave the most consistent control of botrytis in post-harvest tests. None of the alternative chemicals evaluated had any effect on botrytis incidence apart from Hortiphyte Plus which showed some reduction in rotting at one pick date at the Kent site. In 2007 in Kent, programmes of Hortiphyte Plus applied alone or in combination with Teldor (fenhexamid) were compared for control of botrytis fruit rot. The least botrytis was recorded in fruit from plots treated with Teldor alone or combined with Hortiphyte Plus. Hortiphyte Plus applied alone at the same timings did not reduce botrytis incidence.

Key words: raspberry, fungicide, botrytis, alternative chemicals

Introduction

Botrytis fruit rot (Botrytis cinerea) is the most common and one of the most important diseases of raspberry. The fungus rots fruit in the field but mainly causes rapid and devastating losses of picked fruit after harvest. Fruit infection usually occurs via the flowers (McNicol et al. 1985) where the fungus can remain latent until the fruit matures, when, under conditions of high humidity, rapid colonisation of the fruit occurs. The fungus also causes lesions on primocanes. Currently fungicides are relied on for control and are applied close to harvest. Intensive use of fungicides in this way is undesireable and unsustainable. Retail surveillance has shown that >50% of UK produced fruit have fungicide residues most of which are from fungicides targeted at control of botrytis. Much of the UK raspberry crop produced for supermarkets is now grown under protection which may provide the opportunity to reduce the reliance on pesticides by developing alternative, non-pesticidal approaches for control of the key diseases of raspberries. In 2006 a Horticulture LINK project (HL0175) was initiated with the main objective of developing sustainable methods of integrated management of botrytis, powdery mildew and raspberry pests on protected raspberry crops that do not rely on sprays of pesticides during flowering or fruit development so that quality fruit can be produced with a minimum risk of detectable pesticide residues at harvest. For botrytis, the project includes studies on inoculum sources and inoculum suppression as well as exploring crop and tunnel management as means to reduce the botrytis risk. The objective of the work described here was to evaluate alternative chemicals for control of botrytis fruit rot on raspberry.

181 182

Materials and methods

Three field experiments were conducted on two sites (Kent and Cambridgeshire) in 2006 and 2007.

Experiment 1, Kent, 2006 The experiment was established in an open-field plantation of raspberry Glen Ample planted as long canes in 2005 and left unsprayed and cropped in 2005 to establish B. cinerea inoculum in the crop. In 2006, the treatments (Table 1) were applied to plots using a CP15 knapsack sprayer at 1000 l/ha on three occasions (9 June, 16 June and 28 June). All treatments were replicated four times in a randomised block design. Each plot was 8 m long. Blocks were separated from adjacent blocks by an unsprayed guard row of raspberries. Crop development was very variable: at the start of spraying the growth varied from early flower to early green fruit on different plants within the same row. Plants at early flower at the time of the first spray were labelled and picking started when the labelled fruit were red. Prior to this the plots were cleared of all ripe fruit. Plots were regularly inspected for botrytis. At harvest a random sample of two punnets (approximately 200 fruit) of red fruit were picked from the central section of each plot and assessed for botrytis, powdery mildew and any other diseases. The fruit was similarly picked and assessed on three further occasions coinciding with the spray timings. At each harvest a sample of 100 healthy red fruit were taken for post-harvest tests. The fruit were placed in individual modules in trays, covered in polythene and damp incubated. Rot incidence was assessed after seven days incubation at ambient temperature (20-25°C). A sample of green fruit was taken from each plot in July, surface sterilised in 5% by volume ‘Domestos’ bleach and incubated on agar containing paraquat and chloramphenicol (PCA) under lights to check for latent B. cinerea infection in the fruit. The incidence of cane diseases in the plots was assessed in March 2007.

Table 1. Treatments applied to open-field raspberries (Experiment 1) in 2006, East Malling Research, Kent.

Product Active ingredient Product rate 1. Untreated - - 2. Wetcit alcohol ethoxylate 200 ml/100 l spray 3. Farmfos potassium phosphite 6 ml/l potassium phosphite + other 4. Hortiphyte Plus 6 ml/l nutrients potassium phosphate + 5. Farmfos + Wetcit 6 ml/l + 200 ml/100 l spray alcohol ethoxylate 6. Calcium chloride flake calcium chloride 8 g/l 7. UKA 379 experimental 1.44 kg/ha 8. UKA 374 experimental 0.4 kg/ha 9. Talat tolylfluanid + fenhexamid 3 kg/ha 10. Signum pyraclostrobin + boscalid 1.8 kg/ha 11. Switch cyprodonil + fludioxonil 1.0 kg/ha tolylfluanid + fenhexamid + 12. Talat (50%) + Wetcit 1.5 kg/ha + 200 ml/100 l spray alcohol ethoxylate

183

Experiment 2, Cambridgeshire, 2006 The experiment was set up in a polytunnel containing three rows of established Glen Ample raspberries. The tunnel was 72 m long and covered with Luminance TSB polythene before the first spray at green bud. All three rows were used in the experiment, but only the inner faces of the outer rows and one face of the central row were treated and sampled. Plots were 5 m long and the central 3 m was assessed for disease and phytotoxicity. The treatments in Table 2 were applied to plots as fine droplets at 1000 l/ha, with a pressure-assisted knapsack sprayer using a vertically held 2 m boom on three occasions on 30 May (5-15% flowering), 9 June (40-50% flowering) and 19 June (70-75% flowering). An additional spray was applied at green bud (16 May) for one product (Talat). All treatments were replicated four times in a randomised block design. Two untreated control treatments, each replicated four times were also included. Other treatments (other than botrytis fungicides) for pest and diseases control were applied to all plots as normal commercial practice. Fruit not required for samples was picked by the grower (and destroyed) to prevent fruit rotting on the crop. Plots were regularly assessed for botrytis on the leaves, flowers and fruit of raspberry floricanes. Open flowers were tagged at the time of each fungicide application so that treatment efficacy could be related back to application timing. One hundred red fruit per plot were sampled on the 29 June, 7 July and 24 July and assessed for visible disease. At each harvest, 50 healthy fruit per plot were incubated at ambient temperature (20-25oC) in a multicell tray (one berry per cell) sealed in a polythene bag for 10 days. Rot incidence was assessed after seven days. A sample of yellow fruit was taken from each plot on 5 July and surface sterilised with 5% by volume ‘Domestos’ bleach, rinsed and then incubated under lights in sealed transparent containers for 4 to 7 days on agar containing paraquat and chloramphenicol (PCA) to check for latent botrytis infection in the fruit.

Table 2. Treatments applied to protected raspberries (Experiment 2) in 2006, Cambridgeshire.

Product Active ingredient Product rate 1. Untreated - - 2. Teldor fenhexamid 1.5 kg/ha 3. Talat (3 sprays) fenhexamid + tolylfluanid 3 kg/ha 4. Rovral WP iprodione 1.5 kg/ha 5. Scala pyrimethanil 2 l/ha 6. Folicur tebuconazole 0.8 l/ha 7. Amistar azoxystrobin 1 l/ha 8. Talat (4 sprays) fenhexamid + tolylfluanid 3 kg/ha 9. Hortiphyte Plus potassium phosphite + other nutrients 6 ml/l 10. Calcium nitrate calcium nitrate 3.5 g/l 11. Orosorb orange oil + borax + surfactants 2 ml/l

Experiment 3, Kent, 2007 The trial was established in the open-field plantation of raspberry (Glen Ample) used in 2006. Each plot consisted of a double row 8 m long separated from adjacent plots by an unsprayed guard row. In 2007, the treatments (Table 3) applied were based on the results obtained in 2006 and consisted of comparisons of programmes of Teldor or Hortiphyte Plus alone or in combination. The treatments were applied to plots using a Solo self propelled small plot mini 184

sprayer at 1000 l/ha on five occasions (25 May, 4 June, 15 June, 26 June and 6 July). All treatments were replicated four times in a randomised block design. Crop development was again very variable. Plants at early flower at the time of the first spray were labelled and picking started when the labelled fruit were red. Prior to this the plots were cleared of all ripe fruit. Assessments for Botrytis and powdery mildew were conducted as in 2006.

Table 3. Treatments applied to open-field raspberries (Experiment 3) in 2007, East Malling Research, Kent.

Active Product No of sprays Treatment Spray timing ingredient rate applied 1.Untreated - - - 0 3 sprays from 2. Teldor fenhexamid 1.5kg/ha flowering at 10 day 3 intervals potassium 3 sprays at 10 day 3. Hortiphyte Plus phosphite + 6 ml/l intervals from 3 other nutrients flowering fenhexamid + 3 sprays at 10 day 1.5kg/ha 4. Teldor + potassium intervals from + 3 Hortiphyte Plus phosphite + flowering 6ml/l other nutrients 2 sprays at 10 day fenhexamid + 1.5kg/ha intervals from 5. Teldor + potassium + flowering then 2 + 3 Hortiphyte Plus phosphite + 6ml/l Hortiphyte plus only other nutrients at 10 day intervals sprays at 10 day potassium intervals from 6. Hortiphyte Plus phosphite + 6 ml/l flowering, but same 5 other nutrients number of sprays as treatment 5

Results

Experimentl 1, Kent, 2006 The incidence of B. cinerea in fruit at harvest was negligible (<1%). No powdery mildew was observed on any of the plots or guard rows. In post-harvest tests, more than 50% of the fruit from untreated plots at the first pick (Table 4) developed botrytis. All fungicide treatments (UKA379, UKA374, Talat, Signum and Switch) reduced the incidence of B. cinerea by at least 75%. None of the other chemicals evaluated, including Talat at half-label recommended dose with Wetcit, reduced the incidence of B. cinerea. The incidence of B. cinerea in fruit from picks 2 and 3 was too low for meaningful comparisons to be made, most likely because weather conditions at flowering were dry and not favourable for B. cinerea infection. In fruit from pick 4, the lowest incidence of B. cinerea was recorded in fruit from plots treated with Hortiphyte Plus or Talat. However, 185

the incidence of B. cinerea varied considerably between plots within treatments and these differences were not significant. The incidence of B. cinerea in green fruit samples (Table 4) varied from 15 to 37% infected fruit. The incidence between plots varied considerably and there were no consistent differences between treatments. No botrytis lesions were observed on the canes in any of the treatments when assessed in March 2007.

Table 4. Percentage botrytis-rotted fruit in post-harvest tests on raspberries harvested from plots treated in 2006 with various chemicals at East Malling Research, Kent (Experiment 1).

% botrytis rotted fruit Green fruit Treatment Pick 1 Pick 2 Pick 3 Pick 4 harvested 26 July, 13 July 19 July 26 July 31 July incubated on PCA 1. Untreated 56.5 3.0 3.6 15.1 30.4 2. Wetcit 66.0 4.8 1.4 22.2 26.9 3. Farmfos 58.8 3.5 6.8 18.1 29.3 4. Hortiphyte 51.8 1.8 2.0 5.4 14.9 5. Farmfos + Wetcit 64.0 5.8 10.2 21.1 37.1 6. Calcium chloride 47.5 1.8 5.5 30.0 31.6 7. UKA379 12.0 1.8 1.2 11.1 21.9 8. UKA374 9.0 2.5 1.8 10.9 14.7 9. Talat 11.3 3.3 3.1 7.7 18.8 10. Signum 7.5 3.3 6.0 21.4 28.4 11. Switch 9.3 3.5 9.6 24.2 34.9 12. Talat (50%) + Wetcit 44.5 3.5 8.5 30.1 35.8

Experiment 2, Cambridgeshire, 2006 No botrytis leaf or flower infection was seen during the experiment, and no visible botrytis occurred on the fruit hanging in the crop. No other diseases developed in the crop up to the time of the final harvest. The marketability of the fruit did not differ between treatments. In post-harvest tests, more than 40% of the fruit from untreated plots at the first pick (Table 5) developed botrytis. Plots treated with Amistar, Talat at both timings and Scala all had significantly fewer fruit with botrytis than the untreated. The incidence of B. cinerea in fruit harvested on 5 July was high with more than 80% infected fruit in untreated plots. Plots treated with Amistar or Talat at both timings had significantly fewer fruit with botrytis than the untreated but the incidence of botrytis was still high. In the final red fruit sample (24 July) the lowest incidence of botrytis was in fruit sampled from plots treated with Amistar or Folicur. None of the alternative chemicals evaluated (calcium nitrate, Orosorb, Hortiphyte Plus) reduced the incidence of B. cinerea at any of the sample dates. The incidence of latent B. cinerea in the yellow fruit samples (Table 5) varied from 15 to more than 60% infected fruit. The incidence of botrytis was significantly lower in fruit from plots treated with Amistar, Folicur, Scala or Talat (3 sprays). Some phytotoxicity was noted, mainly leaf chlorosis, on plants treated with Scala and Folicur.

186

Table 5. Percentage botrytis-rotted fruit in post-harvest tests on raspberries harvested from plots treated in 2006 with various chemicals (Experiment 2), Milton, Cambridgeshire.

% botrytis rotted fruit Treatment Red fruit Red fruit Red fruit Yellow fruit harvested 5 29 June 5 July 24 July* July, incubated on PCA 1. Untreated 40.6 83.5 36.4 55.3 2. Teldor 31.4 80.0 45.7 37.1 3. Talat (3 sprays) 16.5 38.5 33.5 18.1 4. Rovral WP 33.3 74.7 46.8 34.4 5. Scala 25.4 77.5 32.9 29.0 6. Folicur 28.2 84.5 21.2 24.0 7. Amistar 22.4 67.5 29.0 15.0 8. Talat (4 sprays) 18.8 50.5 34.9 37.0 9. Hortiphyte Plus 35.7 84.0 49.9 46.0 10. Calcium nitrate 45.0 81.0 35.7 56.5 11. Orosorb 45.3 89.0 43.4 64.2 *24 July, predicted means given from GLIM analysis to allow for a plot row effect.

Table 6. Percentage botrytis-rotted fruit in post-harvest tests on raspberries harvested from plots treated in 2007 with various chemicals at East Malling Research, Kent (Experiment 3).

% botrytis rotted fruit Green fruit Treatment Pick 1 Pick 2 Pick 3 Pick 4 harvested 2 July, 3 July 10 July 17 July 23 July incubated on PCA 1. Untreated 55.0 79.5 75.6 81.9 86.0 2. Teldor (3 sprays) 39.5 71.7 36.1 53.6 66.0 3. Hortiphyte Plus (3 sprays) 57.5 75.3 58.9 61.1 90.6 4. Teldor + Hortiphyte Plus 22.0 42.5 35.6 45.6 62.5 (3 sprays) 5. Teldor (2 sprays) + 26.7 62.8 26.7 47.5 87.0 Hortiphyte Plus (5 sprays) 6. Hortiphyte Plus (5 sprays) 76.9 57.2 69.7 69.2 80.8

Experiment 3, Kent, 2007 The weather conditions in 2007 during most of the flowering period were very wet and favourable for botrytis infection of flowers. Despite this the incidence of botrytis on fruit at harvest was very low. In post-harvest tests the incidence of botrytis fruit rot varied from 30- 80% (Table 6). In most of the fruit picks the highest incidence of botrytis was recorded in the fruit from untreated plots. The programmes based on Hortiphyte Plus alone did not reduce the incidence of botrytis compared to the untreated control. At most of the harvest dates Teldor alone or mixed with Hortiphyte Plus (treatments 2, 4 and 5) was most effective in reducing botrytis rot. The incidence of B. cinerea in green fruit samples (Table 6) varied from around 60 to more than 90% infected fruit. The lowest incidence of botrytis was recorded in fruit treated with Teldor alone or in mixture with Hortiphyte Plus (treatments 2 and 4). 187

Discussion and conclusions

In all three experiments, even in the open-field crop trial in Kent in 2007 when the weather conditions during flowering were wet and very favourable for botrytis infection, the incidence of botrytis on fruit at harvest was negligible. Samples of green or yellow fruit taken from each of the crops and incubated on PCA to encourage sporulation of latent infections showed that fruit were infected with botrytis at quite a high incidence which was reflected in the incidence of botrytis in the fruit samples after post-harvest incubation. In all three experiments the most consistent control was achieved with the fungicides. UKA379, UKA374, Amistar, Scala, Talat, Signum and Switch were the most effective fungicides. Of the alternative chemicals only Hortiphyte Plus showed any hint of an effect on botrytis in the 2006 Kent experiment. In the 2007 Kent experiment, programmes based solely on Hortiphyte Plus were ineffective, but where a fungicide (Teldor) was included in the program either with each treatment or only early in the programme botrytis control was significantly better than that in the untreated. Hortiphyte Plus (Hortifeeds) is a plant feed containing phosphite fertiliser and bio-stimulants (citrus and herbal oils). Phosphites are known to have effects on plant disease by inducing resistance in the plant (Ribeiro Junior et al. 2006) and it is possible that this explains the reduction in disease observed here. However, the reduction in botrytis incidence was not present where Hortiphyte Plus was used alone so it is possible that the reduction in botrytis rot was due entirely to the fungicide. The programme of sprays applied here was started at early flower. It is possible that there may be improved benefits of Hortiphyte Plus if application is started prior to flowering. This possibility will be explored in experiments in 2008.

Acknowledgements

The authors would like to thank Defra for sponsoring this work through the Horticulture LINK scheme.

References

McNicol, R.J., Williamson, B. & Dolan, A. 1985: Infections of red raspberry styles and carpels by Botrytis cinerea and its possible role in post-harvest grey mould. – Ann. Appl. Biol. 106: 49-53. Ribeiro Junior, P.M., Resende, M.L.V. De, Pereira, R.B., Cavalcanti, F.R., Amaral, D.R. & Padua, M.A. De 2006: Effect of potassium phosphite on the induction of resistance in cocoa seedlings (Theobroma cacao) against Verticillium dahliae. – Ciencia e Agrotecnologia 30 (4) 629-636. 188 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 189-192

Efficacy of Metschnikowia fructicola (Shemer®) against post-harvest soft fruit (berries) rots in northern Italy (Trentino)

Daniele Prodorutti1, Alessandro Ferrari2, Alberto Pellegrini2, Ilaria Pertot2 1Plant Protection Department, 2Safecrop Centre, Istituto Agrario di San Michele all’Adige, via E. Mach 1, 38010 S. Michele all’Adige (TN), Italy

Abstract: Shemer® is a commercial formulation of the antagonistic yeast Metschnikowia fructicola. It is usually applied to prevent the development of post-harvest rots caused by a wide range of phytopathogenic fungi. Experimental trials were carried out in 2005 and 2006 on berries pre-harvest treated with Shemer® and artificially inoculated with Botrytis cinerea. Fruits were stored at 1°C and thereafter at 25°C or kept constantly at 25°C. Efficacy of Shemer® against post-harvest rot of red currants, highbush blueberries, raspberries and strawberries was evaluated. Pre-harvest applications of Shemer® allowed a good control of grey mould on red currant, raspberry and strawberry while partially controlled the pathogen on blueberry. This experiment suggests that Shemer® could be successfully applied on soft fruits (berries), but further tests under commercial growing and storage conditions are necessary to confirm these data and to identify the most effective application method/time for each crop.

Key words: antagonistic yeast, post-harvest rot, efficacy, soft fruits

Introduction

Berries, as raspberries, blueberries, strawberries and currants, are economically important crops in marginal agricultural areas in some alpine valleys. For example, in Trentino Province (northern Italy) they provide growers with an important additional source of income in small farms where main crops (apple and grapevine) cannot be cultivated because of unsuitable climate. On these crops, fruit rots caused by fungi (Botrytis cinerea, Penicillium sp., Rhizopus sp., Cladosporium sp., Colletotrichum sp.) during and after cold storage result in high economical losses or can shorten their shelf life. Post-harvest fungicides are not currently permitted on the mentioned berries and safe control tools to prevent fruit pathogens are desired. Shemer® (Agrogreen, Israel) is a commercial formulation of the antagonistic yeast Metschnikowia fructicola (Kurtzman & Droby 2001), currently registered in Israel for use on grapes, strawberries, citrus and sweet potatoes. It is usually applied to prevent the development of post-harvest rots caused by a wide range of phytopathogenic fungi, including Aspergillus, Botrytis, Penicillium and Rhizopus species. M. fructicola was already tested on strawberry (Karabulut et al. 2004) and table grapes (Karabulut et al. 2003) providing interesting results against pre-harvest and post-harvest fruit rots. The aim of this study is to evaluate the efficacy of Shemer® against post-harvest rots caused by B. cinerea on red currant, raspberry, blueberry and strawberry.

Material and methods

Experimental trials were carried out in 2005 (raspberry, blueberry and red currant) and 2006 (red currant, highbush blueberry, raspberry and strawberry) on plants pre-harvest treated with

189 190

M. fructicola (Shemer®, 0.2%) and artificially inoculated with suspension of 1*106 conidia/ml of B. cinerea. Shemer® efficacy against post-harvest rots of fruits was evaluated. Preliminary trials were performed in 2005 considering one field of red currant, one of highbush blueberry and one of raspberry. Treatments with Shemer® (0.2%) were made one week and one hour before harvest, using water as untreated control and inoculating B. cinerea just before harvest. Ten plastic box replicates, containing about 60-80 g of berries, were harvested for each fruit/treatment/storage strategy combination. Berries were incubated directly at 25°C (storing strategy I) or stored at 1°C for ten days and after transferred at 25°C (storing strategy II). Fruit observations started 24 hours after their transfer at 25°C and continued until the disease incidence reached 80-90%. Disease evaluation was based on rot incidence (percentage of B. cinerea infected berries), using T-test in statistical analysis. In 2006 the experiment was repeated using the same method of 2005, but including two fields for each soft fruit species (red currant, highbush blueberry, raspberry and strawberry). The storing strategies were: directly at 25°C (strategy I); at 1°C for ten days and after at 25°C (storing strategy II) and (storing strategy III, only for currants and blueberries) at 1°C for 3 month (currants), at 1°C for 2 month (blueberries) and after at 25°C. The colonization of treated fruits by M. fructicola was monitored in 2006 by counting the yeast colonies developed on berry surface one day after harvest: fruits were washed, the suspension was serially diluted and cultured on chloramphenicol potato dextrose agar (PDA, Oxoid) and the typical pink colonies developed on Petri dishes were counted. The number of colonies forming unit (CFUs) for cm² of fruit surface was calculated.

Results and discussion

In 2005, pre-harvest applications of Shemer® allowed a good control of grey mould on red currant and raspberry while did not reduced significantly the pathogen incidence on blueberry (Table 1). Generally, rot development at 25°C was faster on fruit previously kept in cold storage (1°C) than on fruit constantly kept at room temperature. Moreover M. fructicola activity was not negatively affected by cold storage, even if B. cinerea ability to grow at low temperatures probably gave a competitive advantage to the pathogen compared to the yeast.

Table 1. Botrytis cinerea incidence in 2005 trials on fruits incubated at 25°C (strategy I) and at 1°C for 10 days and after at 25°C (strategy II). Data in the same row followed by different letters are significantly different at P≤0.05, T-test. Assessments after 15, 20 and 10 days at 25°C were used, for red currants, blueberries and raspberries respectively.

Storing Rot incidence (%) Crop strategy Control Shemer I 39 a 18 b Red currant II 53 a 32 b I 37 a 33 a Blueberry II 45 a 21 a

Raspberry I 56 a 9 b II 99 a 35 b

191

In 2006, in the different storing strategies red currants treated with Shemer® showed a significantly lower incidence of gray mould compared with the control in field 2. In field 1 rot incidence was very similar between treated and untreated berries (Table 2). This result in the first orchard can be explained by the low number of CFUs on the fruit surface (7.4*103/cm2), probably caused by some copper residues on fruits. A high level of M. fructicola fruit colonization was obtained in berries of field 2 (4*105 CFUs/cm2). Similarly to 2005, Shemer® treatments on blueberries carried out in 2006 did not control significantly the development of B. cinerea even if a high yeast colonization was measured on fruit surface (4-6*104 CFUs/cm2). Only in the field 1 with the storing method 3 it was possible to obtain significant results, but with a low percentage of infection both in treated and untreated berries (Table 2).

Table 2. Botrytis cinerea incidence in 2006 trials on fruits stored with the three strategies. Data in the same row, for each field and storing method, followed by different letters are significantly different at P≤0.05, T-test. Assessments after 15, 20, 10 and 7 days at 25°C were used for red currants, blueberries, raspberries and strawberries respectively.

Rot incidence (%) Storing Crop strategy Field 1 Field 2 Control Shemer Control Shemer I 19 a 22 a 17 a 7 b Red currant II 49 a 67 a 28 a 14 b III 59 a 62 a 83 a 68 b I 16 a 8 a 53 a 38 a Blueberry II 6 a 8 a 20 a 17 a III 18 a 7 b 47 a 37 a I 57 a 50 a 3 a 2 a Raspberry II 71 a 63 b 55 a 46 a I 47 a 13 b 45 a 26 b Strawberry II 62 a 53 a 82 a 52 b

On raspberries, gray mould incidence in 2006 on Shemer® treated fruits was similar to the untreated control, except for field 1 with the storing strategy II where statistically differences were observed (Table 2). In raspberry field 2, the low percentage of infection and the low efficacy of treatments were caused by the low number of CFUs grown on fruit surface (2.7*103 CFUs/cm2) and by bacteria infections that did not allow the development of B. cinerea. Shemer® applications on strawberries allowed a significant reduction of Botrytis infections in both fields (Table 2). Fruits were well colonized by the yeast: the number of colonies on berry surface varied from 8*104 CFUs/cm2 in field 1 to 1.2*105 CFUs/cm2 in field 2. Generally, to obtain a good level of Shemer® efficacy against fruit rots, the method of preparation and application is important: the suspension should be prepared with lukewarm water, applications in the field should be done starting at least one week before harvest to allow the yeast growth on fruit surface and avoiding applications at high temperatures (>35°C) (Blachinsky D., personal communication 2005). Moreover, pre-harvest applications usually allow a better control of fruit rots in comparison with post-harvest applications because of a higher yeast growth on fruit surface. 192

In our experiments the level of yeast growing and viability on fruit surface affected its efficacy to control post-harvest rots: when the number of CFUs/cm2 was lower than 1*104, efficacy reduced and no statistically differences were found between untreated and treated berries. In conclusion, pre-harvest applications of M. fructicola allowed a good control of grey mould on red currant, strawberry and raspberry while did not control the pathogen on blueberry. These results suggest that post-harvest rot control efficacy of Shemer® is strongly dependent on the combination fruit type/pathogen. Shemer® could be successfully applied to control rot pathogens especially for short storing periods, improving shelf-life and marketability of soft fruits; but further tests under commercial growing and storage conditions are necessary to confirm these data and to identify the most effective application method and time for each crop.

Acknowledgements

The research was supported by Safecrop Centre, funded by Fondo unico per la ricerca, Autonomous Province of Trento. Thanks are due to Carmela Sicher (SafeCrop-IASMA), Sandro Conci (IASMA), Andrea Pergher (Cooperative Sant’Orsola S.C.A.) and Daphna Blachinsky (Agrogreen, Israel) for assistance and collaboration.

References

Karabulut, O.A., Smilanick, J.L., Mlikota Gabler, F., Mansour, M. & Droby, S. 2003: Near harvest applications of Metschnikowia fructicola, Ethanol and Sodium bicarbonate to control postharvest disease of grape in central California. – Plant disease 87: 1384-1389. Karabulut, O.A., Texcan, H., Daus, A., Cohen, L., Wiess, B. & Droby, S. 2004: Control of preharvest and postharvest fruit rot in strawberry by Metschnikowia fructicola. – Biocontrol Science and Technology 14: 513-521. Kurtzman, C.P. & Droby, S. 2001: Metschnikowia fructicola, a new ascosporic yeast with potential for biocontrol of postharvest fruit rots. – Systematic and Applied Microbiology 24: 395-399. Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 193-196

Low doses of copper control leaf spot diseases caused by Mycosphaerella ribis and Drepanopeziza ribis in black currants

Arne Stensvand1, Andrew Dobson1, Sigrid Mogan2 1 Bioforsk, Norwegian Institute for Agricultural and Environmental Research, Høyskoleveien 7, 1432 Ås, Norway; 2 The Norwegian Agricultural Extension Service, Foss gård, 3400 Lier, Norway

Abstract: Low rates of copper fungicides were tested against Mycosphaerella ribis and Drepanopeziza ribis in black currants in four experiments. Copper oxide (product: NORDOX 75 WG) at rates of 20, 40 and 200 g of the product per 100 litres and copperoxychloride (product: Kopperkalk Bayer) at a rate of 30 g of the product per 100 litres were applied five times in the period from just after bud break until early to mid June (flowering in mid to late May). Rates recommended by the manufacturers are 200 or 250 g per 100 litres of copper oxide or copperoxychloride, respectively. All treatments reduced the attack of both diseases, and there was no difference between 40 or 200 g copper oxide. The lowest rate of copper oxide and the copperoxychloride had a slightly weaker effect. It may be concluded that 1/5 dose (40 g per 100 litres) of copper oxide was as effective as the recommended rate. The low copper doses gave insignificant phytotoxic effects.

Key words: copperoxychloride, copper oxide, fungicides, Ribes nigrum

Introduction

There are two ascomycetes causing leaf spot and early defoliation in black currants; Mycosphaerella ribis that causes Mycosphaerella leaf spot (also named grey leaf speck) and Drepanopeziza ribis that causes Drepanopeziza leaf spot (often named black currant leaf spot or anthracnose). Both fungi can cause early defoliation and subsequent yield reductions in susceptible cultivars. The premature defoliation of black currants weakens the plants and results in lower yields, shorter life span of the planting and increased possibility of winter injury (Darrow 1919; Crosby et al. 1934). From New York it was reported that Mycosphaerella leaf spot was the more serious of the two diseases in red and white currants and gooseberry (Crosby et al. 1934), but in most instances Drepanopeziza leaf spot is considered the most important (Suit 1945; Ríordáin 1967; Taapio 1971). Cultivars grown in Norway vary greatly in susceptibility to the two diseases, and cultivars selected as especially suited for organic production (Ben Tron, Ben Nare and Narve Viking) are all very susceptible to M. ribis, but not to D. ribis (Stensvand & Mogan 2006). In organic production, sulphur and copper products are often the most effective compounds against fungal diseases. In trials in New York against leaf spot diseases, Bordeaux mixture (copper sulphate and hydrated lime) was effective, but lime sulphur had no effect (Suit 1945). Copperoxychloride was effective against both leaf spot diseases in currants, but caused phytotoxic symptoms on certain currant varieties (Taapio 1971). From Norway, copper oxychloride was also reported to be effective against D. ribis, but it was phytotoxic and caused serious damage to the foliage (Gjærum 1970). There is a constant focus on reducing pesticide input, and even though copper is an essential element for plants, it is a heavy metal, and excessive use may have negative effects on the environment. Copper has been used for more than a century against many fungal and

193 194

bacterial diseases, and there are clear indications that it is effective also in relatively low doses, e.g. against apple scab. In the present work, we wanted to investigate if it was possible to lower the doses of copper against M. ribis and D. ribis in black currants without loosing control of the diseases caused by the two fungi.

Materials and methods

Experiments took place in two black currant fields in south eastern Norway; one was located at Lier in Buskerud county (Ben Nare, susceptible to M. ribis, planted in 2001, 0.5 x 3.5 m plant distance, experiments took place in 2005 and 2006) and the other at Ås in Akershus county (Ben Nevis, susceptible to D. ribis, planted in 1991, 1.5 x 4.0 m planting distance, experiments took place in 2006 and 2007). Fungicides were applied five times in May and June (flowering took place in mid to late May). The following copper fungicides (product, % active ingredient, and manufacturer in parenthesis) were used: Copper oxide (NORDOX 75 WG, 75% a.i., NORDOX AS, Oslo, Norway) at rates of 20, 40 and 200 g per 100 litres, copperoxychloride (Kopperkalk Bayer, 84% a.i., Bayer Crop Science, Leverkusen, Germany) at a rate of 30 g per 100 litres. The standard rates recommended by the manufacturers are approximately 200 and 250 g per 100 litres for NORDOX 75 WG and Kopperkalk Bayer, respectively. The spray plots were 5 m (10 plants) at Lier and 3.5 m (3 plants) at Ås), and assessments took place on 5 to 6 plants in the middle of the plots at Lier and on the one plant in the middle of the plots at Ås. At both locations a Hardi wheelbarrow high pressure sprayer (handgun sprayer) was used, and it was sprayed to runoff. Only assessments of foliar damage are given below, and they were made once either in late August or in September, after harvest. Normal harvest time at the two locations was early August. Each leaf on 5 vegetative shoots (normally 15 to 25 leaves on each shoot) in each plot was graded on a scale 1 to 6 where 1 = 0%, 2 = 0-1 %, 3 = 2-5%, 4 = 6-20%, 5 = 21-40% and 6 = > 40% of the leaf area with visible symptoms of the diseases. Heavily attacked leaves would often be defoliated at time of assessments, and missing leaves on the shoots were graded 6. Percentage of leaves with grade 4 or higher, i.e. above 5% of the leaf area attacked by the disease, was included in the statistical analysis. Analysis of variance and least significant difference (LSD 5%) between means were carried out.

Results and discussion

All treatments were significantly better than the untreated control in all four experiments (Table 1). There was no difference in attack between 40 or 200 g copper oxide per 100 litres, the latter being the recommended rate. The lowest rate of copper oxide and the copperoxychloride contained comparable levels of copper, and although there was a slightly weaker effect of copperoxychloride, there were no statistical differences. The lowest copper oxide and copperoxychloride rates had a slightly weaker effect (not significant for copper oxide against D. ribis) compared to 40 or 200 g copper oxide. From the results it may be concluded that by spraying five times with 1/5 dose (40 g per 100 litres) of copper oxide against leaf spot diseases in black currants, it was possible to control them effectively. Lowering the dose to 1/10 of the recommended rate was not as effective, but it was still much better than the unsprayed control. We observed some phytotoxic effect of the high dose of copper oxide, but it did not result in early defoliation, and it was not visible on the berries. We did not observe phytotoxic effects when using the lower doses. It is known that copper may damage black currants (Gjærum 1970; Taapio 1971). Taapio (1971) pointed out that the phytotoxic effect was important only on some 195

cultivars. However, by using low doses of copper, we avoided the damage on Ben Nare and Ben Nevis used in these experiments. Fungicide applications were carried out from just after bud break to two to three weeks after flowering, and this coincided well with observations on time of spore release. From Finland it was reported that ascospores of both M. ribis and D. ribis were produced from May to July (Taapio 1971). In Norway, ascospores of M. ribis were released from leaf litter on the ground from around bud break in mid to late April until late June, with the highest release in May and early June (Stensvand & Mogan 2006; Stensvand et al. 2007). The fungus was also present in old fruit cluster stalks, and conidia may be released from bud break to flowering, but with a peak in May (Stensvand & Mogan 2006; Stensvand et al. 2007). Taapio (1971) found that the most important time of fungicide applications against leaf spot diseases in currants was just before and immediately after flowering (two sprays altogether). From New York it was found that the dose of Bordeaux mixture could be more than halved compared to the standard recommendation if it was mixed with a spreader sticker (oil) (Suit 1945). The present experiments show that the dose rates may be reduced even more without loosing control.

Table 1. Copper oxide and copperoxychloride against Mycosphaerela ribis and Drepanopeziza ribis in black currant; five applications in May-June.

Percent leaves with more than 5% leaf spots Mycosphaerella ribis Drepanopeziza ribis Treatments (g product per 100 l) 2005 2006 2006 2007 1. Untreated 12.25 aa 54.67 a 30.33 a 22,17 a 2. Copperoxychloride (30 g) -b 4.67 b 1.67 b 10,60 b 3. Copper oxide (20 g) 0.75 b 3.67 b 0.67 bc 7.33 bc 4. Copper oxide (40 g) 0.00 c 0.00 c 0.00 c 2.40 bc 5. Copper oxide(200 g) 0.00 c 0.00 c 0.00 c 0.00 c P-value < 0.001 < 0.001 < 0.001 < 0.003 aMean values denoted with different letters were significantly different at P = 0.05 according to the Least Significant Difference (LSD 5%) method bNot included

References

Crosby, C.R., Rankin, W.H. & Mills, W.D. 1934: Diseases and insects of small fruits. – Cornell Extension Bulletin 306: 28 pp. Gjærum, H.B. 1970: Forsøk med fungicider mot bladfallssopp (Drepanopeziza ribis) på solbær. – Forskning og Forsøk i Landbruket 21: 393-402. Ríordáin, F.Ó. 1967: Leaf spot of black currants caused by Mycosphaerella ribis (Fuckel) Kleb. – Irish Journal of Agricultural Research 6: 131-132. Stensvand, A., Eikemo, H. & Mogan, S. 2007: Epidemiology and control of Mycosphaerella leaf spot in organic production of black currants in Norway. – Proceedings NJF 23rd Congress. Trends and Perspectives in Agriculture, Copenhagen, June 26-29, 2007: 190. Stensvand, A. & Mogan, S. 2006: Bærbuskbladflekk i økologisk solbærdyrking. – Norsk Frukt og Bær 9 (1): 4-5. 196

Suit, R.F. 1945: Currant leaf spot control. – New York State Agricultural Experiment Station Bulletin 709: 13 pp. Taapio, E. 1971: Ribes leaf spot diseases and their control. – Annales Agriculturae Fenniae 11: 94-99. Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 197-201

Effectiveness of a trifloxystrobin and tolylfluanid mixture for control of blackcurrant diseases

Agata Broniarek-Niemiec, Anna Bielenin Research Institute of Pomology and Floriculture, Pomologiczna18, 96-100 Skierniewice, Poland

Abstract: A formulated mixture of trifloxystrobin and tolylfluanid (TFS 63 g/kg + tolylfluanid 625 g/kg) was tested for control of blackcurrant diseases. The experiments were conducted in three localities in central Poland. In all experiments, the mixture was highly effective > 87% control) against American powdery mildew, and it was similar or more effective than the standard fungicides Punch Bis 400 EC (flusilazole, 400 g/l), Nimrod 250 EC (bupirimate, 250 g/l) or Topsin M 500 SC (thiophanate-methyl, 500g/l). The mixture also gave good control of leaf spot and white pine blister rust. Its efficacy against leaf spot ranged from 64 to 83% and it was similar to the fungicides Punch Bis 400 EC, Topsin M 500 SC and Dithane M-45 80 WG (mancozeb, 800 g/kg). The effectiveness of the mixture for control of white pine blister rust varied from 81 to 90% and it was similar to Punch Bis 400 EC and more effective than Topsin M 500 SC or Dithane M-45 80 WG.

Key words: American powdery mildew, leaf spot, white pine blister rust, blackcurrant, chemical control

Introduction

Poland has been a leading producer of blackcurrant in the world for many years. Annual fruit production of this crop ranges between 100-130 thousand tons harvested from about 27,000 hectares. The range of cultivars has changed in recent years. Two new Polish cultivars Tiben and Tisel, and the Scottish cultivar Ben Hope occupy 30% of the cropping area, but still the old, susceptible cultivars Ben Lomond and Ojebyn, which require intensive chemical treatment against fungal diseases, cover over 40% of the area. American powdery mildew, leaf spot and white pine blister rust are the most important diseases in commercial blackcurrant plantations in Poland. Chemical control is necessary for blackcurrant protection. Now, in Integrated Pest Management (IPM), mainly triazols, pirimidin and in a restricted amount mancozeb fungicides are used in control of blackcurrant diseases (Bielenin 2001; Broniarek- Niemiec & Bielenin 2002). The effectiveness of other fungicides, trifloxystrobin and tolylfluanid in mixture, for control of blackcurrant diseases was studied.

Material and methods

The experiments were conducted on blackcurrant Ben Lomond on three plantations localized in central part of Poland in 2006. Each combination consisted of 4 replications with 10 shrubs growing in one row. The effectiveness of trifloxystrobin and tolylfluanid (TFS 63 g/kg + tolylfluanid 625 g/kg) mixture was tested for control of American powdery mildew, leaf spot and white pine blister rust. The first treatment against American powdery mildew was done two weeks after blooming and three others were made every two weeks. Against leaf spot and white pine blister rust six treatments were applied, starting before blooming and repeated every two weeks. The last application was done after harvest. In these experiments the

197 198

following standard fungicides were used: Punch Bis 400 EC (flusilazole, 400 g/l), Topsin M 500 SC (thiophanate-methyl, 500g/l) and, additionally, Nimrod 250 EC (bupirimate, 250 g/l) against American powdery mildew and Dithane M-45 80 WG (mancozeb, 800 g/kg) against leaf spot and white pine blister rust. The evaluations of diseases severity were done before and after the harvest on a sample of 100 leaves per replication (in four replications), according to six degree scale (0- no symptoms, 5 – over 50% of leaf or shoot covered by fungus). The results were elaborated using analysis of variance with data transformation according to Bliss function. The differences between mean values were evaluated by Newman-Keuls test at the level 5%.

Results and discussion

In all experiments, the effectiveness of the tested mixture of trifloxystrobin and tolylfluanid against American powdery mildew was high, above 87% and it was higher or similar to the efficacy of the standard fungicides Punch Bis 400 EC, Nimrod 250 EC and Topsin M 500 SC (Table 1).

Table 1. Effectiveness of trifloxystrobin and tolylfluanid mixture in control of American powdery mildew (Sphaerotheca mors-uvae).

1st evaluation 2nd evaluation Active Dose Fungicide % infected Efficacy % infected Efficacy ingredient per ha shoots [%] shoots [%] Locality: A (Sójki) Untreated 74.34 c* - 86.79 c - trifloxystrobin TFS+tolylfluanid 1.2 kg 9.57 a 87.1 9.91 a 88.6 + tolylfluanid Punch Bis 400 EC flusilazole 112 ml 21.93 b 70.5 20.43 ab 76.5 thiophanate- Topsin M 500 SC 1.5 l 22.33 b 70.0 23.86 b 72.5 methyl Nimrod 250 EC bupirimate 1.5 l 11.45 a 84.6 14.33 ab 83.5 Locality: B (Dąbrowice) Untreated 31.48 c - 86.79 c - trifloxystrobin TFS+tolylfluanid 1.2 kg 2.12 a 93.3 9.91 a 88.6 + tolylfluanid Punch Bis 400 EC flusilazole 112 ml 8.87 b 71.8 20.43 ab 76.5 thiophanate- Topsin M 500 SC 1.5 l 8.63 b 72.6 23.86 b 72.5 methyl Nimrod 250 EC bupirimate 1.5 l 3.84 a 87.8 14.33 ab 83,.5 Locality: C (Grójec) Untreated 49.45 c - 51.75 d - trifloxystrobin TFS+tolylfluanid 1.2 kg 1.25 a 97.5 3.41 a 93.4 + tolylfluanid Punch Bis 400 EC flusilazole 112 ml 5.08 a 89.7 7.17 b 86.2 thiophanate- Topsin M 500 SC 1.5 l 12.69 b 74.3 12.74 c 75.4 methyl Nimrod 250 EC bupirimate 1.5 l 2.96 a 94.0 4.24 a 91.8 *Statistical analysis was performed separately for each localization, means followed by the same letter are not significantly different at p=0.05 199

These results correspond with our earlier experiments, in which strobilurin fungicides (trifloxystrobin, kresoxim-methyl and azoxystrobin) showed good effectiveness for control of American powdery mildew of blackcurrant and gooseberry (Broniarek-Niemiec & Bielenin 2003; Broniarek-Niemiec 2004). Also the results of other authors showed very good effectiveness (more than 90%) of kresoxim-methyl in control of this disease (Jörg et al. 2000). The tested mixture (TFS+tolylfluanid) gave good results for control of leaf spot as well. Its efficacy ranged from 64 to 83% and it was similar to fungicides Punch Bis 400 EC, Topsin M 500 SC and Dithane M-45 80 WG. In one locality (C), the efficacy of TFS+tolylfluanid was even higher than Topsin M 500 SC and Dithane M-45 80 WG (Table 2).

Table 2. Effectiveness of trifloxystrobin and tolylfluanid mixture in control of leaf spot (Drepanopeziza ribis).

1st evaluation 2nd evaluation Active Dose Fungicide % infected Efficacy % infected ingredient per ha Efficacy leaves [%] leaves [%] Locality: A (Sójki) Untreated 44.48 b* - 94.79 c - trifloxystrobin TFS+tolylfluanid 1.2 kg 11.65 a 73.8 20.19 a 78.7 + tolylfluanid Punch Bis 400 EC flusilazole 112 ml 15.15 a 65.9 17.56 a 81.5 thiophanate- Topsin M 500 SC 1.5 l 16.85 a 62.1 39.60 b 58.2 methyl Dithane M-45 80 WG mancozeb 3.0 kg 15.18 a 65.9 23.02 a 75.7 Locality: B (Dąbrowice) Untreated 23.75 b* - 43.59 b - trifloxystrobin TFS+tolylfluanid 1.2 kg 8.46 a 64.4 12.96 a 70.3 + tolylfluanid Punch Bis 400 EC flusilazole 112 ml 9.70 a 59.2 12.56 a 71.2 thiophanate- Topsin M 500 SC 1.5 l 12.71 a 46.5 15.90 a 63.5 methyl Dithane M-45 80 WG mancozeb 3.0 kg 11.90 a 49.9 18.23 a 58.2 Locality: C (Grójec) Untreated 22.70 c* - 33.12 c - trifloxystrobin TFS+tolylfluanid 1.2 kg 8.18 a 64.0 5,.69 a 82.8 + tolylfluanid Punch Bis 400 EC flusilazole 112 ml 9.99 ab 56.0 8.00 ab 75.9 thiophanate- Topsin M 500 SC 1.5 l 12.60 b 44.5 11.23 b 66.1 methyl Dithane M-45 80 WG mancozeb 3.0 kg 12.34 b 45.6 9.99 ab 69.8 *Statistical analysis was performed separately for each localization, means followed by the same letter are not significantly different at p=0.05

These results confirmed earlier ones obtained in experiments on control of leaf spot on blackcurrant by trifloxystrobin, where efficacy ranged from 67 to 81% (Broniarek-Niemiec 2004). Also Locke et al. (2002) in their experiments noted good control of blackcurrant leaf spot by trifloxystrobin. The second component of tested mixture, tolylfluanid, shows some 200

activity against leaf spot. When used in control of fruit grey mould of blackcurrant it also reduced leaf diseases by 40-50% (Bielenin 2002). The effectiveness of the tested mixture in control of white pine blister rust varied from 81 to 90% and it was similar to provided by fungicide Punch Bis 400 EC and higher than effectiveness of Topsin M 500 SC and Dithane M-45 80 WG (Table 3). Results of field experiments carried out by the authors indicated that the mixture of two components, trifloxystrobin and tolylfluanid, is very effective in controlling blackcurrant diseases. Trifloxystrobin, which belongs to the strobilurin group, may be valuable for broadening of the fungicide list recommended for protection of blackcurrant against its main fungal diseases.

Table 3. Effectiveness of trifloxystrobin and tolylfluanid mixture in control of white pine blister rust (Cronartium ribicola).

Active Dose % infected Efficacy Fungicide ingredient per ha leaves [%] Locality: A (Sójki) Untreated 17.90 c* - trifloxystrobin TFS+tolylfluanid 1.2 kg 2.69 a 85.0 + tolylfluanid Punch Bis 400 EC flusilazole 112 ml 2.73 a 84.8 thiophanate- Topsin M 500 SC 1.5 l 8.69 b 51.5 methyl Dithane M-45 80 WG mancozeb 3.0 kg 7.09 b 60.4 Locality: B (Dąbrowice) Untreated 13.06 c*- trifloxystrobin TFS+tolylfluanid 1.2 kg 2.47 a 81.1 + tolylfluanid Punch Bis 400 EC flusilazole 112 ml 4.43 ab66.1 thiophanate- Topsin M 500 SC 1.5 l 5.44 b 58.4 methyl Dithane M-45 80 WG mancozeb 3.0 kg 5.98 b 54.2 Locality: C (Grójec) Untreated 12.98 c*- trifloxystrobin TFS+tolylfluanid 1.2 kg 1.3 a 90.0 + tolylfluanid Punch Bis 400 EC flusilazole 112 ml 2.63 ab79.7 thiophanate- Topsin M 500 SC 1.5 l 5.17 b 60.2 methyl Dithane M-45 80 WG mancozeb 3.0 kg 5.39 b 58.5 *Statistical analysis was performed separately for each localization, means followed by the same letter are not significantly different at p=0.05

References

Bielenin, A. 2001: Control of blackcurrant and gooseberry diseases in Integrated Pest Management (IPM). – Materiały Konf. „Intensyfikacja produkcji porzeczek i agrestu”. Skierniewice, 6 czerwca: 117-122 (in Polish). 201

Bielenin, A. 2002: Control of Raspberry and Blackcurrant diseases by tolylfluanid. – Proc. 8th JS on Rubus and Ribes. Acta Hort. (ISHS) 585: 331-333. Broniarek-Niemiec, A. & Bielenin, A. 2002: Control of American powdery mildew and leaf spot on goosebrry plantations. – Zesz. Nauk ISK 10: 211-215 (in Polish with English summary). Broniarek-Niemiec, A. & Bielenin, A. 2003: Efficacy of some fungicides in control of American powdery mildew and leaf spot on gooseberry plantations. – Progress in Plant Protection / Postępy w Ochronie Roślin 43 (2): 545-547 (in Polish with English summary). Broniarek-Niemiec, A. 2004: Effectiveness of fungicide Zato 50 WG in control of blackcurrant diseases. – Progress in Plant Protection / Postępy w Ochronie Roślin 44 (2): 623- 625 (in Polish with English summary). Jörg, E., Harzer, U. & Ollig, W. 2000: Integrated approach for the control of American gooseberry mildew. – IOBC/wprs Bulletin 23 (11): 25-34. Locke, T., Bobbin, P., Atwood, J. & Owen, J. 2002: Effect of strobilurin fungicides on disease control and yield in blackcurrants. – Proc. 8th JS on Rubus and Ribes. Acta Hort. (ISHS) 585: 375-380. 202 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 203-209

Interactions between isolates of powdery mildew (Podosphaera aphanis) and cultivars of strawberry, Fragaria x ananassa

Xiangming Xu, Joyce Robinson, David Simpson East Malling Research, New Road, East Malling, Kent, ME19 6BJ, UK

Abstract: Experiments were conducted to elucidate the nature of host-pathogen interactions between strawberry cultivars and powdery mildew. Over a period of two years, each of twelve mildew isolates was inoculated against potted plants of a number of commercial cultivars in controlled environmental cabinets. There were considerable differences in the mildew development between repeated inoculations as well as between cultivars. The cultivars Symphony, Rosie and Elsanta appeared to be most susceptible. There were no indications of significant isolate × cultivar (race-specific) interactions, indicating that breeding for partial resistance, instead of complete resistance, may be necessary.

Keywords: race-specific, race non-specific, partial resistance, resistance breeding

Introduction

Powdery mildew, caused by Podosphaera aphanis Braun & Takamatus (formerly Sphaero- theca macularis (Wallr.: Fr.) Lind syn. S. humuli (DC.) Burrill), is a serious disease on strawberries (Fragaria x ananassa) (Miller et al. 2003; Blanco et al. 2004; Amsalem et al. 2006). Because of limited epidemiological knowledge and the lack of cultivars with durable resistance, strawberry powdery mildew is currently controlled by routine sprays of fungicides. However, it is difficult to control the disease with the fungicides available, especially in everbearers because of their long cropping period and the difficulty of scheduling spray applications with regular harvesting. For this reason, resistance to powdery mildew is now regarded as an essential attribute of any new everbearing cultivar. Furthermore, increasingly both June-bearing and everbearing strawberries are being grown under protection, which has led to more severe epidemics than in open field conditions (Xiao et al. 2001). Sources of resistance exist within Fragaria x ananassa but the inheritance of the resistance is evidently complex, as indicated by investigations in several breeding populations (Hsu et al. 1969; Mcnicol & Gooding 1979; Simpson 1987; Nelson et al. 1995; Davik and Honne 2005). In general, these studies suggested that both additive and non-additive genetic variance components are important, although the non-additive component is the smaller of the two. Breeding for durable resistance to mildew is also hampered by observations that the field resistance in new selections may be short-lived (i.e. expression of resistance in breeding lines is not consistent over time) or vary greatly between different environmental conditions (Mcnicol & Gooding 1979; Nelson et al. 1995). Previous research on the genetics of strawberry resistance to powdery mildew has focused on the relative contribution of additive and non-additive components in selected strawberry germplasm, relying on natural infection by field inocula. It has so far ignored one particular aspect of host-pathogen interactions, i.e. whether there is a race-specific or race-non-specific relationship. For research on this topic it is necessary to maintain different mildew strains separately. Information on race-specific or non-specific resistance can have a significant impact

203 204

on breeding strategies as well as on exploitation and deployment of cultivars. Race-specific resistance to powdery mildew has been widely demonstrated in plant species, including wheat (Wolfe & Mcdermott 1994), cucurbits (Del Pino et al. 2002), melon (Mccreight 2006) and groundsels (Clark 1997). The existence of race-specific resistance may allow the exploitation of resistance genes by growing mixtures of cultivars with different resistance genes to prolong the life of these genes (Newton 1997). This paper reports a study carried out at East Malling Research (EMR), England to determine the extent and the type (race-specific or race non-specific) of resistance to powdery mildew in selected popular strawberry cultivars in the UK. Strains of powdery mildew were obtained from several cultivars in several locations and inoculated against potted plants of a number of selected strawberry cultivars.

Materials and methods

Individual leaves with sporulating mildew lesions were sampled from seven cultivars at several sites in the UK; the petiole of each sampled leaf was inserted into a thin layer of water agar in a polystyrene tissue culture tub. ‘Pure’ mildew isolates were obtained by progressively diluting inoculum. Conidia from a single lesion of a sampled mildewed leaf were used to inoculate three detached leaves of several cultivars Elsanta, Rosie, Tango and Sophie in vitro in a polystyrene tub with mannitol-sucrose agar (MSA – 20 g mannitol, 10 g sucrose per litre) into which leaf petioles were inserted. Then three more leaves were similarly inoculated with a single lesion resulting from the first inoculation. Finally, a single lesion from the second inoculation was considered as a ‘pure’ isolate. Individual isolates were maintained and multiplied in vivo on plants of Elsanta within an isolation plant propagator (Figure 1). Positive air pressure was maintained inside each individual propagator unit to prevent contamination of plants from external inoculum. Initially, each isolate was inoculated onto healthy, micropropagated plants of several cultivars in the propagator; but the most consistent sporulation were obtained on Elsanta and hence this cultivar was used to maintain and multiply all isolates. These plants infected with the mildew isolates were regularly replaced with healthy plants with young susceptible leaves, which were also maintained in the propagator to avoid natural infection. Because of the limited space in the propagator, only a maximum of six isolates could be maintained at a given time. A total of 12 ‘pure’ isolates were obtained from field plants for studies on host-pathogen interactions over the two-year period (2004-2005) (Table 1). The 12 isolates was inoculated onto potted strawberry plants of 12 commercial cultivars of both June-bearers and everbearers (Table 1) in controlled environmental (CE) cabinets at 15°C and 75% rh. These plants potted up periodically from bare-root runners, purchased from a commercial propagator in the UK, as necessary and kept in a compartment of the GroDome (containment greenhouse facilities). Each isolate was inoculated onto two plants of each cultivar; typically on each plant two young expanding leaves (with three leaflets) were inoculated and individually tagged for disease assessment. Tagged young leaves on each plant were inoculated by gently brushing conidia with a paint brush. Inoculations were repeated at least once over time. To prevent contamination, only a single isolate was inoculated onto plants on a given day. A total of 12 cultivars were used in this study but it was not possible to inoculate each isolate onto all cultivars because there were no susceptible leaves available at the time of inoculation for some isolates or a cultivar was not available from commercial propagators in the UK. The number of mildew lesions and the state of sporulation was visually assessed daily once the first lesion was seen. If lesions started to merge, the percentage of leaf area with 205

mildew lesions was visually estimated instead. Recording stopped a week after the first lesion was observed, since any subsequent new lesions may have resulted from infection by conidia of the new colonies instead of the original inoculum.

Figure 1. An isolation plant propagator used to maintain individual isolates of strawberry powdery mildew, where plants are watered using a wick and maintained in a negative air pressure to prevent contamination from external conidia.

Results and discussions

A successful ‘infection’ following an inoculation is taken to indicate that a particular inoculation resulted in sporulating lesions, i.e. with a non-zero score on the new scale. This is a necessary condition for an epidemic to continue in field conditions and is used in this study as the criterion for determining compatible or non-compatible host-pathogen interactions. For a given isolate, the length of time to symptom appearance (incubation period) was within a week of inoculation and varied little among cultivars. All visual lesions produced conidia. Disease severity varied considerably among inoculated leaves, isolates and cultivars as well as between replicate inoculations over time. The observed data on the number of lesions were rescaled for a clearer presentation: (a) 0 – no lesions developed on any inoculated leaflets, (b) 1 – average number of lesions per leaflet is ≤ 1, (c) 2 – average number of lesions per leaflet is > l but ≤ 3, (d) 3 – average number of lesions per leaflet is > 3 but ≤ 5, and (e) 4 – average number of lesions per leaflet is > 5 or at least one of the 12 inoculated leaflets had at least 50% area colonised. Table 1 presents the summary of all inoculations. Inoculation outcome varied greatly between replicate experiments. For example, the YDR1 isolate failed to produce sporulating lesions on any of the five cultivars inoculated in one replicate but produced sporulating lesions on the same five cultivars in the other replicate. There was also considerable lesions on the same five cultivars in the other replicate. There was also considerable variability in mildew lesion number between replicate inoculations. For example, on average less than one

206

207

lesion resulted from the inoculation of EME1 on Rosie and Symphony in one replicate but more than five lesions in the other (Table 1). For a given isolate, variability in the response between cultivars appeared to be greater between replicate inoculations than within a replicate inoculation. This observed variability between replicate inoculations is most likely due to variation in the quantity and quality of inoculum as well as the quality of strawberry leaves used for inoculation. To control the amount of spores inoculated, a spore suspension of a desired concentration is normally used. However, it is well known that viability of powdery mildew conidia in general is significantly reduced when immersed in water for a period of time as short as 15 min (Butt 1978; Sivapalan 1993; Sivapalan 1993; Sivapalan 1994). To overcome this, tower inoculation methods are sometimes used to inoculate plants with powdery mildew conidia. Tower inoculation is more accurate in controlling the amount of spores deposited onto flat leaf surfaces, often detached, such as wheat, barley and rose (Linde & Debener 2003), but young strawberry leaves of potted plants differ greatly in their positions/angles and most of the young leaves are not fully unrolled when inoculated. In addition to the variation in the inoculum quantity, the quality of both mildew conidia and plant leaves also varied greatly with season. Hence, inoculation during April-October usually resulted in a greater number of lesions than at other times. In total, there were 79 isolate-cultivar combinations tested; 12 of these combinations were not repeated (Table 1). Only three of the 79 combinations failed to produce sporulating lesions, all with Cambridge Favourite. A single inoculation of the LEE2 isolate on Cambridge Favourite failed to produce sporulating lesions and repeated inoculation of HOE1 and IGE1 also failed to infect Cambridge Favourite (Table 1). Overall, Symphony, Elsanta and Rosie were most susceptible. Of the 12 isolates, HBE2 and YDA1 resulted in the greatest disease, whereas YDR1 and HOE1 resulted in the least disease. These results suggested lack of strong race-specific interactions between strawberry cultivars and powdery mildew. This is different from other mildews where race-specific interactions are well documented, including wheat (Wolfe & Mcdermott 1994), cucurbits (Del Pino et al. 2002), melon (Mccreight 2006) and groundsels (Clark 1997). The current finding of the lack of strong race-specific interactions is consistent with the previous consensus that general combining ability is much more important than specific combining ability for resistance against strawberry powdery mildew (Hsu et al. 1969; Mcnicol & Gooding 1979; Simpson 1987; Nelson et al. 1995; Davik & Honne 2005). This makes selection of parents easier in breeding programmes, as it can be based on the parental phenotype rather than requiring a progeny test. Lack of strong race-specific interactions could be due to two main reasons. First, only current commercial cultivars were used in this investigation, since this study was intended to generate results not only for breeding programmes but also for managing mildew on these cultivars in practice. Cultivated strawberry in general may also lack resistance to powdery mildew since there is a high degree of relatedness between commercial cultivars in North America (Sjulin & Dale 1987; Dale & Sjulin 1990) and the same is almost certainly true in Western Europe. Similarly, there was also lack of genetic variation among strawberry mildew strains. DNA analyses showed that samples of Israeli and Italian mildew isolates were nearly identical (99-100% similar) based on sequence analyses of a 359 bp fragment of the ITS1-2 region and 6 random fragments totalling 2800 bp (Pertot et al. 2007). There were some differences between cultivars in their susceptibility but these differences often vary between inoculation experiments. Of all the cultivars tested, Symphony, Rosie and Elsanta are most susceptible where Evita, Tango and Cambridge Favourite are more tolerant. Cambridge Favourite was previously shown to be a good source of resistance to powdery mildew (Mcnicol & Gooding 1979). Present results suggest that the 208

inheritance of mildew resistance is unlikely to be controlled by a few major genes. Thus breeding for partial resistance, instead of complete resistance, may be necessary, at least in the near future.

Acknowledgements

This work was funded by the UK Department of Environment, Food and Rural Affairs (Defra) – Project Number: HH3228SSF.

References

Amsalem, L., Freeman, S., Rav-David, D., Nitzani, Y., Sztejnberg, A., Pertot, I. & Elad, Y. 2006: Effect of climatic factors on powdery mildew caused by Sphaerotheca macularis f. sp. fragariae on strawberry. – European Journal of Plant Pathology 114: 283-292. Blanco, C., de los Santos, B., Barrau, C., Arroyo, F.T., Porras, M. & Romero, F. 2004: Relationship among concentrations of Sphaerotheca macularis conidia in the air, environmental conditions, and the incidence of powdery mildew in strawberry. – Plant Disease 88: 878-881. Butt, D.J. 1978: Epidemiology of powdery mildews. – In: Spencer, D.M. (ed.): The Powdery Mildews. Academic Press, London: 51-81. Clark, D. 1997: The genetic structure of natural pathosystems. – In: Crute, I.R., Holub, E.B. and Burdon, J.J. (eds.): The gene-for-gene relationship in plant-parasite interactions. CAB International, Wallingford, UK: 231-243. Dale, A. & Sjulin, T. 1990: Few cytoplasms contribute to North American strawberry cultivars. – HortScience 25: 1341-1342. Davik, J. & Honne, B.I. 2005: Genetic variance and breeding values for resistance to a wind- borne disease [Sphaerotheca macularis (Wallr. ex Fr.)] in strawberry (Fragaria x ananassa Duch.) estimated by exploring mixed and spatial models and pedigree information. – Theoretical and Applied Genetics 111: 256-264. del Pino, D., Olalla, L., Perez-Garcia, A., Rivera, M.E., Garcia, S., Moreno, R., de Vicente, A. & Tores, J.A. 2002: Occurrence of races and pathotypes of cucurbit powdery mildew in Southeastern Spain. – Phytoparasitica 30: 459-466. Hsu, C.S., Watkins, R., Bolton, A.T. & Spangelo, L.P. 1969: Inheritance of resistance to powdery mildew in cultivated strawberry. – Canadian Journal of Genetics and Cytology 11: 426-438. Linde, M. & Debener, T. 2003: Isolation and identification of eight races of powdery mildew of roses (Podosphaera pannosa) (Wallr.: Fr.) de Bary and the genetic analysis of the resistance gene Rpp1. – Theoretical and Applied Genetics 107: 256-262. McCreight, J.D. 2006: Melon-powdery mildew interactions reveal variation in melon cultigens and Podosphaera xanthii races 1 and 2. – Journal of the American Society for Horticultural Science 131: 59-65. McNicol, R. & Gooding, H. 1979: Assessment of strawberry clones and seedlings for resistance to Sphaerotheca macularis (Wall ex Frier). – Horticultural Research 1979: 35- 41. Miller, T.C., Gubler, W.D., Geng, S. & Rizzo, D.M. 2003: Effects of temperature and water vapor pressure on conidial germination and lesion expansion of Sphaerotheca macularis f. sp fragariae. – Plant Disease 87: 484-492. 209

Nelson, M.D., Gubler, W.D. & Shaw, D.V. 1995: Inheritance of powdery mildew resistance in greenhouse-grown versus field-grown California strawberry progenies. – Phyto- pathology 85: 421-424. Newton, A.C. 1997: Cultivar mixtures in intensive agriculture. – In: Crute, I.R., Holub, E.B. and Burdon, J.J. (eds.): The Gene-for-Gene Relationship in Plant-Parasite Interactions. CAB International, Wallingford, Oxon, UK: 65-80. Pertot, I., Fiamingo, F., Amsalem, L., Maymon, M., Freeman, S., Gobbin, D. & Elad, Y. 2007: Sensitivity of two Podosphaera aphanis populations to disease control agents. – Journal of Plant Pathology 89: 81-92. Simpson, D.W. 1987: The inheritance of mildew resistance in everbearing and day-neutral strawberry seedlings. – Journal of Horticultural Science 62: 327-332. Sivapalan, A. 1993: Effects of impacting rain drops on the growth and development of powdery mildew fungi. – Plant Pathology 42: 256-263. Sivapalan, A. 1993: Effects of water on germination of powdery mildew conidia. – Mycological Research 97: 71-76. Sivapalan, A. 1994: Development of powdery mildew fungi on leaves submerged under water. – Journal of Phytopathology 140: 82-90. Sjulin, T.M. & Dale, A. 1987: Genetic diversity of North American strawberry cultivars. – Journal of the American Society for Horticultural Science 112: 375-385. Wolfe, M.S. & McDermott, J.M. 1994: Population genetics of plant pathogen interactions: the example of the Erysiphe graminis – Hordeum vulgare pathosystem. – Annual Review of Phytopathology 32: 89-113. Xiao, C.L., Chandler, C.K., Price, J.F., Duval, J.R., Mertely, J.C. & Legard, D.E. 2001: Comparison of epidemics of Botrytis fruit rot and powdery mildew of strawberry in large plastic tunnel and field production systems. – Plant Disease 85: 901-909. 210

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 211-215

Potential role of cleistothecia in strawberry powdery mildew

Xiangming Xu, Joyce Robinson, Angela Berrie East Malling Research, New Road, East Malling, West Malling, Kent, ME19 6BJ UK

Abstract: Experiments were conducted to determine whether ascospores can act as a source of primary inoculum of strawberry powdery mildew. Leaves with cleistothecia were collected from field plants in autumn and then placed outside on bare soil or grass. At a monthly interval, cleistothecia development was monitored from leaf samples. Samples of cleistothecia were checked whether ascospores were full of contents and discharged from asci individually. Mature (discharged as individual ascospores, full of contents) were observed in March and some of these ascospores had germinated on the agar. Experiments were also carried out to determine whether ascospores could lead to infections on strawberry leaves. Over-wintered leaves with cleistothecia were thoroughly wetted first and then immediately hung above healthy strawberry plants inside a controlled environmental cabinet (15°C). Development of mildew on these plants was regularly monitored. In both 2006 and 2007, only one single lesion was observed on these plants. These results so far suggest that ascospores are likely to be an important source of primary inoculum.

Key words: ascospores, Podosphaera aphanis, overwintering

Introduction

Powdery mildew, caused by Podosphaera aphanis Braun & Takamatus (formerly Sphaerotheca macularis (Wallr.:Fr.) Lind syn. S. humuli (DC.) Burrill), is a serious disease on strawberries (Fragaria x ananassa) (Amsalem et al. 2006; Blanco et al. 2004; Miller et al. 2003), particularly late-season and everbearing types, and strawberries grown under protection in glasshouses or polytunnels. It can infect leaves, leaf petioles, flower trusses, flowers and fruit. Because of limited epidemiological knowledge and the lack of cultivars with durable resistance, strawberry powdery mildew is currently controlled by routine sprays of fungicides. However, it is difficult to control the disease with the fungicides available, especially in everbearers because of their long cropping period and the difficulty of scheduling spray applications with regular harvesting. In addition, increasingly both June-bearing and everbearing strawberries are being grown under protection, which has led to more severe epidemics than in open field conditions (Xiao et al. 2001). The under-protection production allows growers to schedule their harvests to coincide with periods of market demand and may also help to reduce grey mould, fruit rots and root rots (Freeman et al. 1997; Legard et al. 2001; Xiao et al. 2001). However, without the inhibitory effect of rain on conidia germination, sheltered crops tend to have more powdery mildew infection (Legard et al. 2001; Xiao et al. 2001). Long periods at around 20°C and the high relative humidity in tunnels provide favourable conditions for P. aphanis (Amsalem et al. 2006). Powdery mildew is believed to overwinter as mycelium on living infected leaves (Peries 1962). Strawberry powdery mildew is heterothallic and cleistothecia are often produced (Berrie et al. 2002; Maas 1998; Nakazawa & Uchida 1998; Peries 1962) but their role in the life cycle of this disease as perennation bodies has not yet been determined. Thus, current opinions view the strawberry mildew pathogen as a clonal population with limited ability to generate new genetic variation. This view is supported by new findings that cleithstothecia

211 212

were not present in Israel over a period of four years (Amsalem et al. 2006) and that there were lack of genetic variation among sampled mildew isolates from Italy and Israel (Pertot et al. 2007). However, recent research at East Malling Research (Berrie & Xu, unpublished) showed that strawberry plants may continue to grow over the winter in some regions and that leaves infected during the previous autumn senesce and die by spring and thus appear not to play any role in overwintering. Furthermore, epidemics in the following spring may start a few weeks earlier on sites with cleistothecia present the previous autumn than on those sites with cleistothecia absent, indicating that epidemics in the following spring may have been initiated by ascospores rather than conidia. The relative importance of ascospores and mycelia on leaves as an overwintering inoculum has critical implications on disease management and breeding/deployment of resistant cultivars. Experiments were conducted to monitor the occurrence and development of cleistothecia and ascospores over the winter and to determine the viability of ascospores.

Materials and methods

Sampling leaves with cleistothecia Mildewed leaves with cleistothecia were collected from two sites in England (Herefordshire and Kent) on several commercial plantings of different cultivars in autumn 2004-2006. These leaves were then mixed together and placed into bags made from nylon mesh with the hole size of 10 µm (about 20 leaves per bag) during the late-October and early November period. Ten bags of leaves with cleistothecia were then placed outside on grass or bare soil and secured on the ground with metal nails. In addition, a few bags of leaves were also put into an incubator set to 5°C. One bag from each condition (on grass, on bare soil or from the incubator) was randomly brought back to a laboratory for assessing cleistothecia development during February-May at a monthly interval.

Assessing cleistothecia development Cleistothecia from several randomly selected leaves were crushed on a glass slide within a drop of water and observed for the state of ascospore development, particularly concerning whether ascospores were full of cytoplasm contents, clumped together or individually discharged. IN 2006, the contents of crushed cleistothecia were also placed into the MTT solution (0.1% tetrazolium bromide (MTT) in mM potassium phosphate buffer at pH = 6.3) and incubated for 36 hours to check spore viability. Spores stained rose are considered to be viable, black and unstained non-viable (El-Hamalawi & Erwin 1986) When individual apparently ‘viable’ ascospores were found in the crushed samples on slides or in the MTT solution, samples of leaf discs with cleistothecia from the same bag were first soaked in water for 10 minutes, and adhered to the lid of a Petri dish, containing glass slides or agar. A few days later, these slides/agar plates were checked for whether individual ascospores had discharged onto the slides and the agar, and germinated. In 2007, these discs were surface sterilised first in order to reduce contamination by other surface micro- organisms. Surface sterilisation was done by immersing fruit in a 0.5% Domestos® solution (the a.i. is sodium hypochloride so that 0.025% wt/vol chlorine is available) for 15 min and then immersed in sterile distilled water for 15 min.

Infection of leaves by ascospores When individual ‘viable’ ascospores were found in the crushed samples on slides or in the MTT solution, further experiments were carried out to determine whether ascospores from these cleistothecia could lead to infections on strawberry leaves. Leaves with cleistothecia were thoroughly first wetted for 10 minutes and then immediately hung above healthy 213

strawberry plants inside a controlled environmental cabinet (15°C) (Figure 1). Development of mildew on these plants was regularly monitored. Similarly, leaves were surface sterilised in 2007.

Figure 1. Experimental set up of investigating whether ascospores may infect strawberry leaves in a controlled environment cabinet.

Results and discussion

In February, no mature (discharged as individual ascospores, full of contents) were observed though MTT staining suggested that most ascospores remain viable. Nevertheless, individual ascospores were clearly distinguishable inside asci (Figure 2). By March, individual ascospores were discharged onto glass slides or agar plates. No germination was observed on glass slides. However, at least 10% of discharged ascospores germinated on PDA in 2007 (Figure 3).

Figure 2. Two microscopic (X 100) photos of crushed cleistothecia of strawberry powdery mildew in February 2007 with visible individual ascospores inside an ascus.

214

Figure 3. Three germinated ascospores of strawberry powdery mildew on PDA.

On those healthy plants placed under the leaves with cleistothecia, only a single fresh lesion was observed in either 2006 or 2007. Although it cannot categorically be ruled out that this single lesion in each year was not due to contamination of the natural inoculum. Nevertheless, such contamination is not likely because the plants were placed inside the cabinets at least 10 days before the experiment commenced. At 15°C, lesions due to infection prior to being moved inside the cabinet would have become visible within 10 days. Similar results on the potential of mildew ascospores infecting strawberry were also obtained in another study (Arne Stensvand, personal communication). Thus, it is likely that ascospore is a source of primary inoculum in early spring that could initiate early mildew epidemics on strawberry. Hence, removing dead leaves infected the previous season may be important in order to reduce the level of primary inoculum in spring.

Acknowledgements

This work was funded by the UK Department of Environment, Food and Rural Affairs (Defra) – Project Number: HH3228SSF.

References

Amsalem, L., Freeman, S., Rav-David, D., Nitzani, Y., Sztejnberg, A., Pertot, I., & Elad, Y. 2006: Effect of climatic factors on powdery mildew caused by Sphaerotheca macularis f. sp fragariae on strawberry. – European Journal of Plant Pathology 114: 283-292. Berrie, A., Harris, D., Xu, X.-M., & Burgess, C. 2002: A potential system for managing Botrytis and powdery mildew in main season strawberries. – Acta Horticulturae 567: 647-650. Blanco, C., de los Santos, B., Barrau, C., Arroyo, F.T., Porras, M., & Romero, F. 2004: Relationship among concentrations of Sphaerotheca macularis conidia in the air, environmental conditions, and the incidence of powdery mildew in strawberry. – Plant Disease 88: 878-881. El-Hamalawi, Z. & Erwin, D. 1986: Physical, enzymic, and chemical factors affecting viability and germination of oospores of Phytophthora megasperma f.sp. medicinagis. – Phytopathology 76: 503-507. Freeman, S., Nizani, Y., Dotan, S., Even, S., & Sando, T. 1997: Control of Colletotrichum acutatum in strawberry under laboratory, greenhouse and field conditions. – Plant Disease 81: 749-752. Legard, D.E., Mertely, J.C., Chandler, C.K., Price, J.F., & Duval, J.D. 2001: Management of strawberry diseases in the 21st century: chemical and cultural control of botrytis fruit rot. 215

– In: The 5th North American Strawberry Conference (eds S.C. Hokanson & A.R. Jamieson), Niagara Falls: 58-63. Maas, J. 1998: Compendium of Strawberry Diseases. – The American Phytopathological Society Press, St. Paul, Minnesota, USA. Miller, T.C., Gubler, W.D., Geng, S. & Rizzo, D.M. 2003: Effects of temperature and water vapor pressure on conidial germination and lesion expansion of Sphaerotheca macularis f. sp. fragariae. – Plant Disease 87: 484-492. Nakazawa, Y. & Uchida, K. 1998: First record of cleistothecial stage of powdery mildew fungus on strawberry in Japan. – Annals of the Phytopathological Society of Japan 64: 121-124. Peries, O.S. 1962: Studies on strawberry mildew, caused by Sphaerotheca macularis (Wallr. ex Fries) Jaczewski. I. Biology of the fungus. – Annals of Applied Biology 50: 211-224. Pertot, I., Fiamingo, F., Amsalem, L., Maymon, M., Freeman, S., Gobbin, D. & Elad, Y. 2007: Sensitivity of two Podosphaera aphanis populations to disease control agents. – Journal of Plant Pathology 89: 81-92. Xiao, C.L., Chandler, C.K., Price, J.F., Duval, J.R., Mertely, J.C. & Legard, D.E. 2001: Comparison of epidemics of Botrytis fruit rot and powdery mildew of strawberry in large plastic tunnel and field production systems. – Plant Disease 85: 901-909. 216 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 p. 217

Ontogenic resistance against powdery mildew (Podosphaera macularis) in leaf tissue of strawberry

David M. Gadoury1, Arne Stensvand2, Robert C. Seem1, M. Cathy Heidenreich1 1Dept. Plant Pathology, Cornell University, N.Y. State Agric. Exp. Stn., Geneva, NY 14456 USA; 2Bioforsk, Norwegian Institute for Agricultural and Environmental Research, Høyskoleveien 7, 1432 Ås, Norway

Abstract: Strawberry powdery mildew, caused by Podosphaera macularis, has been reported to occur most abundantly on the lower surface of strawberry leaves, and this has been suggested to be due to differences in susceptibility between the upper and lower epidermis, differential conditions of temperatures and humidity, and direct damage to powdery mildew colonies by UV light. However, these factors have never been conclusively demonstrated to be involved. In controlled inoculation of both upper and lower surfaces of emergent (E1), expanding (E2) and expanded (E3) leaves of the mildew-susceptible cultivar Earliglow, we found no difference in susceptibility of the upper vs. lower leaf surfaces. Only the lower surface is exposed in tightly-folded E1 leaves, and exposure of E1 leaves to airborne inoculum resulted in severe infections of the lower surface and no infection of the upper surface. E2 leaves were equally infected on both surfaces, but the incidence was much higher on E1 leaves. Differential distribution of mildew colonies on upper vs. lower leaf surfaces is at least partly due to the combined effect of rapid development of ontogenic resistance and folding of E1 leaves.

Key words: strawberry powdery mildew, Podosphaera macularis, leaf susceptibility, leaf surface

217 218 Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 219-222

Colletotrichum acutatum: Survival in plant debris and infection of some weeds and cultivated plants

Päivi Parikka, Anne Lemmetty MTT Agrifood Research Finland, Plant Production Research, FI-31600 Jokioinen, Finland

Abstract: Colletotrichum acutatum can survive in strawberry residues in nordic conditions over one winter, maybe even for as long as almost two years. Plant debris stored either on soil surface or covered with soil was able to infect young strawberry plants in greenhouse tests still after one winter. Fungal DNA was amplified by PCR of test plants after two winters. Annual and perennial weed species common on strawberry fields were inoculated in greenhouse with C. acutatum spore suspension. The fungus sporulated on old leaves of all the artificially infected weed species but not on the young leaves and tops of the respective species. Old leaves of Tripleurospermum, Chenopodium, Lapsana, Plantago, Capsella, Viola, Epilobium and Rumex were heavily infected while the fungus was not recovered on young parts of Epilobium, Viola, Senecio and Lapsana. Infection in young leaves could be superficial, but in old leaves the fungus easily colonized senescent tissues. C. acutatum caused necrotic lesions on some plant species like Plantago major and Phacelia tanacetifolia. The fungus was able to survive in infected weed debris on soil surface over one summer.

Keywords: strawberry, black spot, Colletotrichum acutatum, infection, survival

Introduction

Colletotrichum acutatum causing black spot on strawberry is a quarantine pathogen in the EU (Anon. 1997). It was detected for the first time in Finland in 2000 in imported plant material (Parikka & Kokkola 2001). The fungus has not been detected in strawberry plants produced by Finnish nurseries or other plant species. Survival of C. acutatum in artificially infected strawberry leaves, crowns and fruit was studied in 2002-2003. The infected strawberry debris was stored over one winter on soil surface and covered with soil. The fungus was able to infect young strawberry plants used as bait plants after the winter period (Parikka et al. 2006). A longer-term survival test with infected strawberry plant material was carried out during 2002-2005. C. acutatum did not cause symptoms on bait plants after a two-year test period, but a positive PCR result was obtained of bait plants (Lilja et al. 2006). C. acutatum has many host plants, including ornamentals, fruit and berry crops (Anon. 1997). In Norway, the pathogen has been isolated from sweet cherry plants growing adjacent to strawberry and it was detected also on some berry crops and weed species (Stensvand et al. 2006). Berrie & Burgess (2003) reported of 35 weed species that had been investigated and found to be infected by C. acutatum in trials and that were also able to transmit infection to strawberry. However, only seven species had visible symptoms. C. acutatum can live on plants as epiphytic and endophytic infections without any visible symptoms (Leandro et al. 2001). This makes the pathogen difficult to observe and control. In Finland, the authorities give orders to destroy plants infected by C. acutatum to avoid further spread of the pathogen. When infected strawberry plants are planted and develop black spot symptoms in the field, the infection spreads also to the adjacent plants such as weeds. When the vegetation is killed, either with herbicides or during winter, the fungus may survive

219 220

in the remaining debris. The aim of the study is to investigate the role of weeds and fallow plants and their debris as alternative hosts for C. acutatum in Finnish conditions.

Materials and methods

Annual and perennial weed species common on strawberry fields were inoculated in greenhouse with C. acutatum spore suspension. Altogether 19 weed species were included in trials in 2006-2007. Annual species Chenopodium album, Lapsana communis, Senecio vulgaris, Stellaria media, Spergula arvensis, Viola arvensis, Capsella bursa-pastoris and Persicaria lapathifolia and perennial or biennial species Tripleurospermum maritimum, Stachys sp., Matricaria discoidea, Taraxacum vulgare, Plantago major, Rumex longilolius, R. acetosa, Mentha sp., Valeriana sp, Epilobium angustifolium and Trifolium repens were tested. Some plant species used in fallowing were also included in trials: Carum carvi, Phacelia tanacetifolia, Brassica juncea, Lupinus angustifolius and Helianthus annuus. The weeds and fallow plants were germinated from seeds and grown in a greenhouse on commercial peat substrate (Kekkilä Oy). The plants were inoculated with C. acutatum spore suspension (26 conidia /ml) with a hand sprayer. The relative humidity was kept at 100% for 24 hours after inoculation and then for one week at night. The temperature was set at 22-24°C for the 4-week test period. Potential symptoms were observed during the test. Old leaves and young top leaves were sampled after the test. The samples were incubated on potato dextrose medium (20g/l, Difco Laboratories, France) at room temperature (22ºC) to detect C. acutatum infection. Samples of some species were also surface-sterilized with ethanol before plating to detect infection inside the tissues. Leaves were also placed on moist filter paper in Petri dishes to observe sporulation in tissues. To investigate survival in plant debris, infected dry plant material of the weeds was placed into nylon-mesh bags according to Parikka et al. (2006) outdoors on soil surface at the beginning of June 2007. The test was terminated at the beginning of October 2007. C. acutatum was isolated on PDA from the dry plant debris before the bait test which was conducted according to Parikka et al. (2006).

Results and discussion

Colletotrichum acutatum was isolated from all the weed and fallow plant species four weeks after inoculation. The old senescent leaves were found to be infected in all the weed species inoculated. The fungus sporulated abundantly on the tissues placed on moist filter paper and on PDA medium. In young leaves, however, C. acutatum was not detected on Senecio, Epilobium, Viola, Chenopodium, Trifolium and Mentha. On the other species, there was less infection in young leaves than in the old ones except for the rosette-forming species such as Taraxacum, Capsella and Plantago where the infection level was similar both on old and young leaves. Young leaves of Carum, Helianthus and Lupinus had no infection in the top leaves and infection detected on Phacelia and Brassica was low. Surface sterilization removed C. acutatum from the young leaves but did not have the same effect on old leaves. This indicates that infection is epiphytic on the surfaces of young leaves, but had already invaded the tissues of old leaves. C. acutatum can survive on plants as epiphytic infection for some time (Leandro et al. 2001). Of the weeds investigated, Rumex longifolius has been reported to be infected by C. acutatum in Norway (Stensvand et al. 2006), and of other Rumex species, R. obtusifolius also showed symptoms in England (Berrie & Burgess 2003). In our trials, Rumex species did not 221

have visible symptoms. However, Plantago major had severe necrotic lesions on the leaves after inoculation and, according to Berrie & Burgess (2003), P. lanceolata also can have symptoms.We also observed stem lesions in inoculated Phacelia tanacetifolia plants. In earlier trials, strawberry plant material such as leaves and petioles artificially infected in greenhouse rapidly showed symptoms of black spot, and C. acutatum sporulated abundantly in dying tissues. When this material was stored outdoors over one winter, C. acutatum survived the winter on the soil surface or covered by soil (Parikka et al. 2006). However, when this material was left outdoors for a longer time in summer, survival was reduced at warmer temperatures and higher soil moistures. According to Eastburn & Gubler (1992), temperature and moisture have a strong influence on the survival of C. acutatum. High temperatures and excess moisture cause a rapid decline of the fungus while cold temperatures of the winter period can enhance survival, because the competing micro-organisms are not active. However, C. acutatum was detected by PCR in young strawberry plants inoculated with infected plant debris stored outdoors from September 2003 until September 2005 (Lilja et al. 2006). This indicates that C. acutatum can survive also summer temperatures and high moistures in soil. However, part of the infected strawberry tissues consisted of crowns which degrade slowly in soil and can also keep the infection better than tissues such as leaves which are more rapidly used by soil microflora and fauna. The young and thin tissues of C. acutatum-infected weed debris stored outdoors on soil surface over the summer period were partially degraded under warm and moist conditions. C. acutatum sporulated in the debris of Carum and Phacelia after the trial period of 4 months. Inoculation with the infected debris of Capsella caused black spot symptoms in petioles of bait plants and C. acutatum was also isolated from the lesions. The result shows that weed debris is a potential source of inoculum for strawberry plants still after some months on soil surface in warm conditions. Especially weeds with slowly degrading stems may be a source of inoculum for a longer time.

Acknowledgements

This study was funded by the Maiju and Yrjö Rikala Horticultural Foundation.

References

Anonymous 1997: Colletotrichum acutatum. – In: Quarantine Pests for Europe. Second Edition. Data sheets on quarantine pests for the European Union and for the European and Mediterranean Plant Protection Organization. (eds. Smith, I.M., McNamara, D.G., Scott, P.R., Holderness, M. and Burger, B.). University Press, Cambridge: 692-697. Berrie, A.M. & Burgess, C.M. 2003: A review of research on epidemiology and control of blackspot of strawberry (Colletotrichum acutatum) with special reference to weeds as alternative hosts. – IOBC/wprs Bulletin 26 (2): 163-168. Eastburn, D.M. & Gubler W.D.1992: Effects of soil moisture and temperature on the survival of Colletotrichum acutatum. – Plant Disease 76: 841-842. Leandro, L., Gleason, M.L, Nutter, F.W. Jr., Wegulo, S. & Dixon, P. 2001: Germination and sporulation of Colletotrichum acutatum on symptomless strawberry leaves. – Phyto- pathology 91: 659-664. Lilja, A., Parikka, P., Pääskynkivi, E., Hantula, J., Vainio, E., Vartiamäki, H., Lemmetty, A. & Vestberg, M. 2006: Phytophthora cactorum and Colletotrichum acutatum: survival and detection. – Agriculturae conspectus scientificus 71 (4): 121-128. 222

Parikka, P. & Kokkola, M. 2001: First report of Colletotrichum acutatum on strawberry in Finland. – Plant Disease 85: 923. Parikka, P., Pääskynkivi, E. & Lemmetty, A. 2006: Survival of Colletotrichum acutatum in dead plant material and soil in Finland. – Acta Horticulturae 708: 131-134. Stensvand, A., Talgø, V., Strømeng, G. M., Aamot, H.U., Børve, J., Sletten, A. & Klemsdal, S. 2006: Colletotrichum acutatum in Norwegian strawberry production and potential sources of inoculum in and around strawberry fields. – IOBC/wprs Bulletin 29 (9): 87- 91.

Integrated Plant Protection in Soft Fruits IOBC/wprs Bulletin 39, 2008 pp. 223-224

Wild and cultivated Potentilla spp. may serve as alternate hosts and possible reservoirs of strawberry viruses

David Yohalem, Karen Lower East Malling Research, New Road, East Malling, Kent ME19 6BJ, UK

Abstract: Strawberry [Fragaria x ananassa Duchesne = Potentilla x ananassa (Rozier) Mabb] has accumulated a variety of virus diseases in its relatively short history. Strawberry crinkle virus, Strawberry mild yellow edge virus and Strawberry mottle virus are vectored by the aphid Chaetosiphon fragaefolii (Cockerell), which is known to occasionally infest other members of the genus Potentilla. Several wild strawberry species have been shown to be hosts of the virus. We grafted material infected with strawberry viruses into plants of UC-5, an indicator strawberry cultivar and then grafted UC-5 leaves into other members of the genus sensu lato. Leaves from these plants were subsequently grafted back to UC-5 plants to demonstrate their potential as reservoirs for the virus. UC- 5 plants were indexed for strawberry viruses by end-point reverse transcriptase PCR subsequent to back grafting.

Key words: Fragaria, strawberry crinkle virus, strawberry mild yellow edge virus, strawberry mottle virus.

Introduction

There are several viral diseases of strawberry (Fragaria x ananassa Duch.) that – alone or in combination – can result in severe losses of yield, particularly in perennial cropping systems. Three of these are vectored by the strawberry aphid Chaetosiphon fragaefolii (Cockerell) (Blackman & Eastop 2000). The aphid is known to occasionally infest other members of the genus Potentilla (Blackman & Eastop 2000). Indeed, it has been pointed out that the genus is either polyphyletic or that Fragaria should be included within it (Mabberly 2002). F. vesca and F. chiloensis have been shown to be hosts of the virus, however we wished to find out if local Potentilla spp. could harbour the virus, perhaps asymptomatically. To this end we grafted strawberry virus-infected material to members of Potentilla sensu lato with subsequent use of indicator plants and reverse transcriptase PCR (RT-PCR) to detect latent infections. In these preliminary experiments, positive detection of Strawberry mild yellow edge virus was obtained from Potentilla.

Materials and methods

Plants and viruses F. x ananassa cultivars Elsanta and UC-5 were used as reservoir and indicator, respectively, for Strawberry crinkle virus (SCV), Strawberry mild yellow edge virus (SMYEV) and Strawberry mottle virus (SMV). Potentilla reptans L., collected by D. Yohalem and Fragaria vesca L. from the collection of the EMR strawberry breeding program were grown in glasshouses at EMR and shown to be free of the three viruses by RT-PCR using virus specific primers (Thompson et al. 2003) with a plant chloroplast RNA as positive control for extraction.

223 224

Grafting and virus detection Virus-infected leaves of Elsanta strawberry were grafted into UC-5 indicator plants. When symptoms appeared in the UC-5 plants, leaves from these were grafted to Potentilla and F. vesca plants. One month after grafting virus-infected leaves into the P. reptans and F. vesca, leaves from these plants were grafted onto new, uninfected UC-5 plants. One month later, the UC-5 plants were examined for symptoms and were extracted for RNA using Qiagen RNA Easy Kits (Qiagen, Crawley, UK) following the manufacturer’s protocol, except that SE buffer was used. Total extracted RNA was reverse transcribed using Illustra Ready-To-Go RT-PCR beads (GE Healthcare, Little Chalfont, UK) and cDNA was amplified, as above.

Results and discussion

Symptoms developed and virus infection was confirmed by RT-PCR, as expected, in all accessions of F. vesca. This confirmed the utility of the grafting protocols. No virus or virus-like symptoms of infection were observed in P. reptans for two months subsequent to having scions of infected UC-5 plants grafted; the plants appeared normal. However, when young leaves from these plants were grafted onto UC-5 indicator plants mild symptoms were observed in two of these grafted plants. SMYEV was detected by RT-PCR from RNA extracted from the UC-5s, indicating that P. reptans can act as a latent host of, at least, one strawberry virus. Further research needs to be done to exclude the P. reptans as a possible latent host of the other viruses. The possibility of both P. reptans and F. vesca serving as alternate hosts should be further investigated since both can be found in close proximity to many commercial strawberry fields. Clean planting material is still the best way to limit virus ingress into commercial plantings, but it may be that management of strawberry relatives either through isolation or eradication may be an important adjunct to aphid control in maintaining virus-free strawberry propagation beds.

Acknowledgements

We thank DEFRA for funding this project and T. Passey for his assistance in the laboratory.

References

Blackman, R.L. & Eastop, V.F. 2000: Aphids on the World’s Crops: An Indentification and Information Guide. 2nd Edition. – Wiley, NY: 324 pp. Mabberly, D.J. 2002: Potentilla and Fragaria (Rosaceae) reunited. – Telopea 9: 793-801. Thompson, J.R., Wetzel, S., Klerks, M.M., Vašková, D., Schoen, C.D., Špak, J. & Jelkmann, W. 2003. Multiplex RT-PCR detection of four aphid-borne strawberry viruses in Fragaria spp. in combination with a plant mRNA specific internal control. – J. Viro- logical Methods 111: 85-93.