BioControl 47: 193–206, 2002.© 2002 Kluwer Academic Publishers. Printed in the Netherlands.

Biocontrol of grey mould by Ulocladium atrum appliedat different flower and fruit stages of strawberry

Pedro BOFF1,3, Jürgen KÖHL1∗, Matthijs GERLAGH1 andJoop DE KRAKER2

1Plant Research International, Wageningen, the Netherlands;2Department of Plant Sciences, Wageningen University, Wageningen, the Netherlands;3Permanent address: Agricultural Research and Rural Extension Agency of Santa CatarinaState (EPAGRI), Florianópolis, SC, Brasil∗author for correspondence: e-mail: j.kohl@plant.wag-ur.nl

Received 19 December 2000; accepted in revised form 4 May 2001

Abstract. Grey mould is an important disease of strawberries resulting from flower and fruitinfection by Botrytis cinerea Pers. The saprophytic fungus Ulocladium atrum Preuss is apromising biological control agent for control of B. cinerea in strawberry and other crops.The objective of this research was to determine the efficacy of U. atrum to control grey mouldby a single application of a spore suspension (2 × 106 conidia/ml) at different flower and fruitdevelopment stages. Four experiments were carried out in 1999, two under natural and twounder enhanced inoculum levels of B. cinerea. In each experiment, flowers and young fruitsin six distinct stages of development were sprayed with either water or U. atrum suspension.U. atrum suppressed B. cinerea sporulation on petals by 15 to 54%. One to four days afterspraying, U. atrum was present on less than 30% of stamens and did not affect the incidence ofB. cinerea on these flower parts. The efficacy of the U. atrum sprays in controlling grey mouldwas low to moderate, and resulted on average in a reduction of 21% in disease incidence onripe fruits. Low control efficacy was probably due to poor coverage with or colonisation ofstamens by U. atrum spores, and a relatively low level of suppression of the colonisation offlower parts by B. cinerea. Significant reductions of grey mould in comparison to the control(p ≤ 0.10; on average 41% reduction) were found most frequently when the antagonist wasintroduced at late flowering or early fruit stages. Therefore, these are the most suitable stagesto apply U. atrum. Further studies are needed to improve the spray coverage and persistenceof U. atrum inoculum.

Key words: antagonism, inoculum pressure, Botrytis cinerea

Introduction

Grey mould, caused by Botrytis cinerea Pers., is a major cause of fruit lossesin strawberry crops (Hancock, 1999). Flower parts play an important rolein fruit infection. Infection of stamens can result in the establishment ofa latent infection in the receptacle (Powelson, 1960; Bristow et al., 1986).

194 PEDRO BOFF ET AL.

The pathogen remains quiescent until ripening and may then cause fruit rotbefore or after harvest (Powelson, 1960; Jersch et al., 1989). Petals, adheringto the fruit surface and colonised by B. cinerea, may also constitute animportant source of fruit infection (Wilkinson, 1954; Jarvis, 1962). Directinfections of ripening fruit by conidia are of minor importance, and chemicalcontrol of grey mould was equally effective when fungicides were appliedonly during flowering as throughout flowering and fruit development (Jarvis,1962; Wilcox and Seem, 1994). Therefore, prevention of the establishmentof B. cinerea in flowers with regular fungicide applications is at presentthe conventional method of grey mould control (Maas, 1984). However,increasing fungicide resistance of B. cinerea and public concern about the useof pesticides on food crops prompt the need for alternative control strategies(De Waard et al., 1993; Sutton, 1995; Ghini, 1996). Biological control byvarious filamentous fungi and yeasts is one of the methods that has showngood perspectives for non-chemical management of grey mould in strawberry(Sutton, 1995; Lima et al., 1997). Antagonists of B. cinerea are able to reducethe inoculum production of the pathogen during the flowering period, or mayinterfere with the infection process (Tronsmo and Dennis, 1977; Peng andSutton, 1991). In perennial strawberry crops, crop debris is the major sourceof B. cinerea inoculum during flowering (Braun and Sutton, 1987), but inthe annual cropping system as commonly encountered in the Netherlands,this source of inoculum is negligible (Boff, 2001). Consequently, a biolo-gical control strategy aimed at suppression of sporulation of B. cinerea oncrop debris will be less effective in annual strawberry crops. Given the roleof flower parts in fruit infection by B. cinerea, antagonists may effectivelycontrol grey mould if they succeed to prevent or reduce pathogen colonisationof these flower parts.

The fungus Ulocladium atrum Preuss, indigenous to the Netherlands,was found to exclude B. cinerea from necrotic plant tissue more effectivelyunder field conditions than other antagonists such as Gliocladium catenu-latum Gilman & E. Abbott, Aureobasidium pullulans (de Bary) Arnaud orChaetomium globosum Kunze:Fr. (Köhl et al., 1995). Laboratory studiesshowed that U. atrum is able to suppress tissue colonisation and sporulationby B. cinerea on necrotic leaves, stamens, and petals of strawberry (Köhlet al., unpublished). Under field conditions, multiple applications of U. atrumresulted in significant reduction of pre- and post-harvest grey mould in annualstrawberry crops (Boff, 2001). The applied dosage could be lowered if thefrequency of application was increased. This may be due to an improvedcoverage of the flowers in a particular stage of development, because duringthe flowering period in strawberry of approximately three weeks, severaloverlapping generations of flowers are produced. When the effectiveness of

BIOCONTROL OF GREY MOULD BY ULOCLADIUM ATRUM 195

U. atrum strongly depends on the development stage of the flower or youngfruit at the time of application, application frequency should be great enoughto reach all flowers in that particular stage. If the effectiveness of U. atrum isnot strongly dependent on the development stage, timing and dosage shouldbe such that as many flowers as possible are covered with an effective dose ofU. atrum.

The objective of the present research was to determine the efficacy ofU. atrum to control grey mould by a single application to flowers or youngfruits of strawberry at different development stages. The results can be usedto further optimise timing and dosage of antagonist applications.

Materials and methods

Experimental design, treatments and crop management

Four field experiments were carried out with strawberry cv. Elsanta, on asandy soil near Wageningen, the Netherlands. Cold-stored transplants keptat –2 ◦C until the day before planting were transplanted at 22 April (exper-iments 1 and 2) and 18 May 1999 (experiments 3 and 4). Each experimentconsisted of 180 plants in a plot of 4.5 × 10 m, comprising three double-rowswith three plants per meter row length. Spacing between double rows was 1 mand within double rows 0.5 m. Experiments1 and 3 were conducted undernatural inoculum pressure and experiments 2 and 4 under enhanced inoculumpressure of B. cinerea. Enhanced inoculum pressure was achieved by placinginfected strawberry fruits inside the experimental plot, from first green budappearance till ripening. Infected fruits were deposited twice per week at arate of circa two fruits per plant between the double rows and at the plotmargin at 30 cm distance from the strawberry plants. The experimental plotswith natural and enhanced inoculum pressure of B. cinerea were separated bya 100 m grass buffer.

In all four experiments, treatments were repeated on different sets ofplants. Therefore, plots were divided into 18 sub-plots consisting of ten plantsin double rows. On each of three consecutive dates, three sub-plots weresprayed with a U. atrum suspension (2 × 106 conidia/ml water with 0.01%Tween-80), and three other sub-plots were sprayed with water with Tween-80. The dates of spraying were 5, 8, and 11 June 1999 for experiments 1 and2, and 30 June, 1 July and 2 July 1999 for experiments 3 and 4. Sub-plots ofboth treatments were paired in blocks, and in each plot U. atrum or Tween-water was applied once to all flower branches. During the application anduntil 5 min afterwards the sub-plots were surrounded by a plastic sheet toavoid contamination of neighbouring sub-plots. At the date of spraying, 20 to

196 PEDRO BOFF ET AL.

30 flowers or fruits of the same development stage were labelled per sub-plotprior to application of the spray. At each application date, six developmentstages were distinguished: green bud, white bud, young open flower (withyellow anthers), old flower (with dark anthers at petal fall stage), green fruitand early white fruit.

Production and field application of U. atrum inoculum

Fresh conidial suspensions of U. atrum were prepared on each spraying dateand kept at 4 ◦C until one hour before spraying. Conidia from a four weekold culture of U. atrum grown on oat grains (Köhl et al., 1995) were washedfrom the substrate using a camping washing machine (Nova MW 100; Nova,Maastricht, the Netherlands) filled with tap water containing 0.01% Tween-80. The resulting suspension was filtered through nylon gauze (200-µ mesh).The concentration was determined using a haemocytometer and adjusted to2 × 106 conidia/ml before spraying. At each spraying date, the viability ofU. atrum conidia in the suspension was checked by spraying conidia on wateragar and assessing conidial germination after 24 hours incubation at 18 ◦C inthe dark. Suspensions were sprayed with an airbrush (Vega, Thayere ChandlerInc, USA) at 100 kPa, held at 20 cm distance from the flower branches, at anapplication rate of 200 l/ha.

Spore load in the air

Spore load in the air was monitored using Rotorod samplers Mod. 20(Sampling Technologies, Los Altos Hills, CA, USA) with a non-retractingcollector. Rotorods were positioned at 0.3 m height in the centre of strawberryplots with natural and enhanced inoculum pressure, and in the grass buffer at50 m distance from the strawberry plot. Runs were carried out on six days(30 May and 2, 8, 10, 15, and 16 June 1999) in experiments 1 and 2, andon seven days (18, 23, and 25 June, and 04, 07, 15, and 22 July 1999), inexperiments 3 and 4. Two 15 min. runs per day, except on 18 June 1999 (onerun), were carried out between 9:00 and 12:00 hours, which is the most likelyperiod for spore release of B. cinerea (Jarvis, 1980). The spores of one rod perrun were stained with cotton blue (2 ml lactic acid + 4 ml glycerol + 1.5 mlcotton blue at 1% + 2 ml demineralised water) and conidia of B. cinereaon the 22 mm upper part of a rod were counted microscopically at 200×magnification. The number of conidia of B. cinerea per cubic meter of airwas calculated per run and averaged for two runs per sampling day.

BIOCONTROL OF GREY MOULD BY ULOCLADIUM ATRUM 197

Fungal colonisation of flowers

Density of U. atrum conidia on petals

Density and percentage germination of U. atrum conidia on petals wereassessed in experiments 3 and 4. In each plot, samples of ten young petalswere collected from arbitrarily selected unlabelled flowers four hours afterspraying. The samples were incubated in plastic petri dishes with moist filterpaper on the bottom for 15 hours at 18 ◦C in the dark, to allow germinationof U. atrum conidia. Subsequently, each petal was put on a glass slide witha drop of fluorochrome calcofluor white (Fluorescent brightener 28, Sigma;0.02% w/v in 1M tris HCl buffer, pH 8.0). Germinated and non-germinatedconidia within an arbitrarily chosen square of 1 mm2 were counted using afluorescence microscope (Zeiss Axioskop, filter 5, Germany). A conidiumwas considered germinated if at least one stained germ tube longer than halfof a conidium diameter was observed.

Colonisation of petals by B. cinerea

Petals from unlabelled young flowers were collected four hours after sprayingand, when available, three days after spraying. Thirty to 45 petals per treat-ment per spraying date, with a maximum of two petals from each flower,were placed on water agar (1.5% of agar in tap water) in sterile plasticpetri dishes (14 cm diameter) and incubated for 14 days, at 18 ◦C in thedark. The percentage of petal area with B. cinerea sporulation was estimatedusing a stereomicroscope at 20× magnification. The data were averaged pertreatment, per spraying date.

Incidence of B. cinerea and U. atrum on stamens and sepals

Incidence of B. cinerea and U. atrum on stamens and sepals was assessedon old unlabelled flowers (with dark anthers) collected one and four daysafter spraying of U. atrum. Ten to 15 flowers were sampled per treatment perspraying date. Flowers were placed into sterile petri dishes (∅14 cm) on wateragar containing paraquat, at 10 mg per l1 (Gramoxone 200, Zeneca, UK), insuch a way that the base of sepals touched the medium, in order to kill thesepals and allow both fungi to colonise the dead tissue. Petri dishes wereincubated for 12 days at 18 ◦C in the dark. All stamens and sepals of eachsampled flower were examined for the presence of sporulation of B. cinereaand U. atrum with a stereomicroscope, at 10–100× magnification. The incid-ence of B. cinerea and U. atrum on stamens and sepals was calculated perflower and averaged per treatment, per spraying date.

198 PEDRO BOFF ET AL.

Grey mould

All ripe symptomless fruits and fruits with symptoms of grey mould werepicked twice per week. Healthy fruits were always picked separately fromdiseased fruits to avoid contamination during harvesting in view of post-harvest evaluations. Healthy and non-sporulating diseased fruits were placedin individual square open plastic pots (5 cm high). Pots from the same blockwere placed side by side into plastic trays (50 cm length × 30 cm width× 7 cm height) with moist filter paper on the bottom. Each tray was sealedwith a plastic bag and incubated for 72 hours at 18 ◦C in the dark to allowdevelopment of specific symptoms. After incubation, fruits were checked forthe presence of sporulation of B. cinerea. The incidence of grey mould onfruits at harvest or after post-harvest incubation was calculated per plot, perstage of development at the date of spraying.

Data analysis

The effect of sampling time (one vs. four days after spraying) on the incidenceof U. atrum on stamens and sepals was analysed with ANOVA, taking thethree spraying dates as replications over time. The effects of spray treatment(U. atrum vs. Tween-water) on the incidence of B. cinerea on stamens andsepals, and on the sporulation area of B. cinerea on petals were analysed in thesame way. The effect of spray treatment (U. atrum vs. Tween-water) on theincidence of grey mould on fruits was analysed separately for the six distinctstages of flower and fruit development at the date of spraying. The incidencedata were analysed with ANOVA, taking both spray treatment and date ofapplication as factors, with three replicate blocks per treatment combina-tion. ANOVAs were followed by F-tests. Only results regarding the effect ofspray treatment are reported here, as interaction between spraying treatmentand spraying date was not significant in any case. Significant associationsbetween several selected variables were determined by computation of simplelinear correlation coefficients (r). All statistical analyses were performed withthe computer package Genstat 5 version 4.1 (Genstat Committee, AlgorithmGroup Inc.).

BIOCONTROL OF GREY MOULD BY ULOCLADIUM ATRUM 199

Results

Density and survival of U. atrum

The percentage of germinated conidia of U. atrum from the applied suspen-sions on water agar was high and ranged between 89–95% for the differentexperiments. Samples of treated petals from young open flowers incubated inmoist chambers gave similar results with a percentage of germinated conidiaof 86% ± 5.3 (mean ± standard deviation). The initial density of U. atrumconidia on petals sprayed with a suspension at a concentration of 2 × 106

conidia/ml, and at a rate of 200 l/ha, was 5655 ± 1621 conidia per cm2 (mean± standard deviation) when petals were collected four hours after spraying.No later microscopical assessments of density and spore germination of U.atrum on petals could be done because the petals shrivelled. Conidia ofU. atrum were not detected on petals collected from untreated young openflowers.

Incidence of U. atrum on stamens ranged from 20 to 37% when flowerswere collected one day after spraying (Figure 1A). Four days after spraying,the incidence was 12 to 26%. Incidence of U. atrum on sepals ranged from60 to 78% per flower collected one day after spraying and from 45 to 80% forsepals collected four days after spraying (Figure 1B).

Spore load and incidence of B. cinerea on flower parts in control plots

The spore load in the air inside the strawberry plots with enhanced inoculumaveraged 63 (experiment 2) and 167 (experiment 4) conidia per m3 over thewhole experiment, whereas the natural inoculum pressure of B. cinerea insidethe strawberry field was 13 and 14 conidia per m3, respectively, in experi-ments 1 and 3. The average spore load of B. cinerea outside the strawberryplots was 10 (experiments 1 and 2) and 14 (experiments 3 and 4) conidiaper m3 of air. The average spore load per run per day was always higherin the strawberry fields with enhanced inoculum compared either to the plotwith natural inoculum or the grass buffer outside the experimental plots. Thespore inoculum load of B. cinerea inside the plots with natural inoculumpressure was not different from the spore load outside the plots. The incid-ence of B. cinerea on flower parts sampled from the control sub-plots variedhighly among the experiments, but was always higher in the experiments withenhanced inoculum pressure than in the experiments with natural inoculumpressure. The incidence of B. cinerea in samples of young petals rangedfrom 34 to 95%, and in samples of old petals from 42 to 97% (details notpresented). Sporulation of B. cinerea was found on average on 10 to 55% ofthe stamens and on 9 to 75% of the sepals of old flowers (Figure 2).

200 PEDRO BOFF ET AL.

Figure 1. Incidence of Ulocladium atrum on stamens (A) and sepals (B) of old flowerscollected one day and four days after spraying the antagonist (2 × 106 conidia/ml) and incub-ated on paraquat agar for 12 days, at 18 ◦C in the dark. Figure above each pair of columns isthe p-value (F-test). Pairs without a figure have p > 0.10. Bars indicate S.E.M.

Suppression of B. cinerea on flower parts

U. atrum suppressed B. cinerea colonisation on young petals collected fourhours after spraying significantly in two of four experiments (Figure 3A). Onpetals collected three days after spraying, a treatment effect on B. cinereasporulation was found in one of two experiments (Figure 3B). The suppres-sion of sporulation of B. cinerea by U. atrum on petals, calculated asa percentage of the control, was negatively correlated with the level ofB. cinerea sporulation on the untreated petals (r = –0.89, df = 4, p < 0.05).In plots with enhanced inoculum pressure of B. cinerea, the suppression ofsporulation of the pathogen by U. atrum on petals was lower (15–37%) ascompared to plots with natural inoculum (38–54%).

BIOCONTROL OF GREY MOULD BY ULOCLADIUM ATRUM 201

Figure 2. Incidence of Botrytis cinerea on stamens (A, B) and sepals (C, D) from old flowersof control and Ulocladium atrum treatments, collected one day (A, C) and four days (B,D) after spraying. Experiments 1 and 3 were conducted under natural B. cinerea inoculum;experiments 2 and 4 under enhanced levels of B. cinerea inoculum. Figure above each pair ofcolumns is the p-value (F-test). Pairs without a figure have p > 0.10. Bars indicate S.E.M.

Stamens sampled one day after application of U. atrum tended to havea lower incidence of B. cinerea in comparison to the control in one of fourexperiments (p = 0.08) (Figure 2A). The incidence of B. cinerea on stamenscollected four days after spraying of the antagonist was the same in bothtreatments (Figure 2B). The incidence of B. cinerea on sepals sampled oneday after spraying U. atrum tended to be lower than the control in two of fourexperiments (Figure 2C; P = 0.06 and P = 0.07 respectively). No differencein B. cinerea incidence between treatments was detected when flowers werecollected four days after spraying the antagonist (Figure 2D).

Grey mould reduction

The incidence of grey mould on fruits in the control sub-plot differed betweenexperiments. High levels were found under enhanced inoculum pressure inexperiment 2 (29–57%; Figure 4B), whereas in the other three experiments(1, 3 and 4) the incidence ranged from 5 to 17% (Figure 4A,C,D).

The reduction of grey mould incidence by a single application of U. atrum(2 × 106 conidia/ml) was more effective when sprayed on old flowers or

202 PEDRO BOFF ET AL.

Figure 3. Area with sporulation of Botrytis cinerea on petals from strawberry flowers ofcontrol and Ulocladium atrum treatments collected four hours (A) or three days (B) afterspraying. Figure above each pair of columns is the p-value (F-test). Pairs without a figurehave p > 0.10. Bars indicate S.E.M. No petals were collected at three days after spraying inexperiments 3 and 4.

young fruits than on buds or young flowers. Spraying of U. atrum reducedgrey mould when applied at the white bud stage in one of four experiments(p = 0.06, experiment 4), at the old flower stage in two of four experiments(p = 0.06, experiment 1; p = 0.01, experiment 2), at the green fruit stage intwo of four experiments (p = 0.03, experiment 3; p = 0.02, experiment 4), andat the white fruit stage in three of four experiments (p = 0.01, experiment 3and 4; p = 0.1, experiment 2). No reduction of grey mould incidence wasobserved after a single application of the antagonist at green bud and youngflower stages. For the eight cases for which a reduction of grey mould wasfound with a statistical probability of p < 0.10, a negative correlation wasfound between the level of reduction in the U. atrum treatment, calculated asa percentage of the control, and the incidence of grey mould in the control(r = –0.80, df = 6, p < 0.05).

BIOCONTROL OF GREY MOULD BY ULOCLADIUM ATRUM 203

Figure 4. Incidence of grey mould on strawberry fruits developed from flowers treated withUlocladium atrum at 2 × 106 conidia/ml (U. atrum) or with tween-water (Control), at differentdevelopment stages. (A) experiment 1, transplanting on 22 April 1999, natural level of Botrytiscinerea inoculum; (B) experiment 3, transplanting on 18 May 1999, natural level of B. cinereainoculum; (C) experiment 2, transplanting on 22 April 1999, enhanced level of B. cinereainoculum; (D) experiment 4, transplanting on 18 May 1999, enhanced level of B. cinereainoculum. Figure above each pair of columns is the p-value (F-test). Pairs without a figurehave p > 0.1. Bars indicate S.E.M.

Discussion

Single applications of U. atrum to strawberry flowers and young fruits tendedto reduce the incidence of grey mould on ripe fruits, on average by 21%.Statistically significant reductions (p < 0.10) ranged between 22 and 65%and were most frequently found when U. atrum was applied at the old flower,green fruit, or white fruit stages. The low to moderate control of grey mouldcan be explained by the poor coverage of flower parts with U. atrum conidiaor the inability of the antagonist to colonise specific flower parts, in particularthe stamens (Figure 1), and by the moderate suppression of the colonisation offlower parts by B. cinerea (Figures 2 and 3). The observed levels of suppres-sion of B. cinerea sporulation on petals (15–52%) were lower than expected,given the high densities of U. atrum conidia on sprayed petals (> 5,000conidia/cm2) and the high competitive ability of U. atrum on necrotic sterilepetals in the laboratory (Köhl et al., unpublished). Experiments and model

204 PEDRO BOFF ET AL.

simulations on competitive colonisation of substrates by U. atrum and B.cinerea have shown that a lower initial spore density can be compensatedby an earlier start (Kessel, 1999). Probably, B. cinerea has a time advantageon flower parts of strawberry, because the pathogen is able to infect petals,sepals, pistils and stamens before they senesce (Kamoen et al., 1985; Shirazeand Watanabe, 1985; Bristow et al., 1986; Bulger et al., 1987; Pappas andJordan, 1997). U. atrum as a saprophyte can probably only colonise necroticfloral tissues. At higher levels of B. cinerea inoculum pressure, U. atrumsprays resulted in lower percentages of reduction of petal colonisation byB. cinerea and of grey mould on fruits. The likely explanation is again thetime advantage of B. cinerea in competitive substrate colonisation, with ahigher probability of successful early infection of flower parts at a higherspore deposition rate. This time advantage in fields with higher inoculumpressure is probably more important than the higher numbers of B. cinereaspores per se. The low efficacy of U. atrum in reducing grey mould aftersingle applications may also be due to microclimatic conditions in the fieldlimiting the development of the fungus. However, recent results indicated thatU. atrum is not sensitive to low humidity (Köhl and Molhoek, 2001) and cancompete with B. cinerea in a broad temperature range (Köhl et al., 1999).

The best results with single sprays of U. atrum were obtained when thespray was applied at the old flower, green fruit, and white fruit stages, withsignificant (p < 0.10) grey mould reduction in seven of 12 cases (Figure 4). Inthese stages of flower and fruit development, petals and stamens have startedto senesce. The relative success of U. atrum, when applied at these stages,could be due to the high spore density present when the floral tissues becomesuitable for colonisation by the antagonist. U. atrum spores can remain viableon strawberry leaves in the field for several weeks (Boff, 2001). However,the number of spores decreased in time under field conditions due to lossesduring rainfall. On flower parts, such losses may occur during the periodfrom spraying U. atrum at young development stages until the moment whennecrotic tissue is available for colonisation by the antagonist. This may resultin spore densities too low to compete successfully with B. cinerea.

Optimisation of the application programme of U. atrum to control greymould differs from optimising a fungicide spray schedule (e.g. Jarvis, 1969),because not only the relationship between flower or fruit development stageand susceptibility to B. cinerea infection has to be considered, but also theeffect of development stage on colonisation by the antagonist. Our resultssuggest that, with the application technique and spore suspensions used inthis study, sprays can best be targeted at the old flower and young fruit stages.This is convenient, as these stages are of relatively long duration as compared

BIOCONTROL OF GREY MOULD BY ULOCLADIUM ATRUM 205

to, for example, the open flower stage. Multiple sprays at different stages offlower and fruit development may have an additive effect.

The efficacy of single U. atrum sprays needs to be improved. This maybe realised by enhancing the coverage of the flower parts, in particular thestamens, and by increasing the persistence of spores on flower parts. Thiscould be achieved by modifications in the application technique or by addingstickers to the spore suspension. Assuming that B. cinerea does have a timeadvantage in flower colonisation, especially when the inoculum pressure ishigh, U. atrum spores should be present in high numbers on flower partsas soon as these become available to the antagonist to minimise this timeadvantage. This could be achieved by targeting the earlier stages of flowerdevelopment, but only when both the coverage and persistence of U. atrumsprays is high.

Acknowledgements

Funding for this research was partly provided by the Brazilian Government– Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES;Proc. 2959/95-0), which included a full scholarship to the first author. Wealso acknowledge the European Commission (BIOSPORSUPPRESS; FAIR3CT96-1898) for partial financial support to the research. We wish to thankSaskia Burgers for statistical assistance.

References

Boff, P., 2001. Epidemiology and biological control of grey mould in annual strawberry crops.PhD thesis, Wageningen University, Wageningen, The Netherlands.

Braun, P.G. and J.C. Sutton, 1987. Inoculum sources of Botrytis cinerea in fruit rot ofstrawberries in Ontario. Can. J. Plant Pathol. 9: 1–5.

Bristow, P.R., R.J. McNicol and B. Williamson, 1986. Infection of strawberry flowers byBotrytis cinerea and its relevance to grey mould development. Ann. Appl. Biol. 109:545–554.

Bulger, M.A., M.A. Ellis and L.V. Madden, 1987. Influence of temperature and wetness dura-tion on infection of strawberry flowers by Botrytis cinerea and disease incidence on fruitoriginating from infected flowers. Phytopathology 77: 1225–1230.

De Waard, M.A., S.G. Georgopoulos, D.W. Hollomon, H. Ishii, P. Leroux, N.N. Ragsdale andF.J. Schwinn, 1993. Chemical control for plant diseases: problems and prospects. Annu.Rev. Phytopathol. 31: 403–421.

Elmer, P.A.G. and J. Köhl, 1998. The survival and saprophytic competitive ability of theBotrytis spp. antagonist Ulocladium atrum in lily canopies. Eur. J. Plant Pathol. 104:435–447.

Ghini, R., 1996. Ocorrência de resitência a fungicidas em linhagens de Botrytis cinerea, noestado de São Paulo. Fitopatol. Bras. 21: 285–288.

206 PEDRO BOFF ET AL.

Hancock, J.F., 1999. Strawberries. CABI, Wallingford, UK.Jarvis, W.R., 1962. The infection of strawberry and raspberry fruits by Botrytis cinerea Fr.

Ann. Appl. Biol. 50: 569–575.Jarvis, W.R., 1969. The phenology of flowering in strawberry and raspberry in relation to grey

mould control. Hort. Res. 9: 8–17.Jarvis, W.R., 1980. Epidemiology. In: J.R. Coley-Smith, K. Verhoeff and Jarvis WR (eds), The

Biology of Botrytis. Academic Press, London, UK. Pp. 219-250.Jersch, S., C. Scherer, G. Huth and E. Schlösser, 1989. Proanthocyanidins as basis for

quiescence of Botrytis cinerea in immature strawberry fruits. J. Plant Dis. Prot. 96:365–378.

Kamoen, O., G. Jamart, J. Van Vaerenbergh and P. Gouwy, 1985. Botrytis cinerea infectionson petals and green leaves. Med. Fac. Landbouw Rijksuniv. Gent 50/3b: 1053–1058.

Kessel, G.J.T., 1999. Biological control of Botrytis spp. by Ulocladium atrum, an ecologicalanalysis. PhD Thesis, Wageningen University, Wageningen, The Netherlands, 155 pp.

Köhl, J., C.H. Lombaers-van der Plas, W.M.L. Molhoek, G.J. Kessel and H.M. Goossen-vande Geijn (1999). Competitive ability of the antagonists Ulocladium atrum and Gliocladiumroseum at temperatures favourable for Botrytis spp. development. BioControl 44: 329–346.

Köhl, J. and W.M.L. Molhoek, 2001. Effect of water potential on conidial germination andantagonism of Ulocladium atrum against Botrytis cinerea. Phytopathology 91: 485–491.

Köhl, J., W.M.L. Molhoek, C.H. van der Plas and N.J. Fokkema, 1995. Effect of Ulocladiumatrum and other antagonists on sporulation of Botrytis cinerea on dead lily leaves exposedto field conditions. Phytopathology 85: 393–401.

Lima, G., A. Ippolito, F. Nigro and M. Salerno, 1997. Effectiveness of Aureobasidiumpullulans and Candida oleophila against postharvest strawberry rots. Postharv. Biol. andTechnol. 10: 169–178.

Maas, J.L., 1984. Compendium of Strawberry Diseases. American Phytopathological SocietyPress, St. Paul, USA.

Pappas, A.C. and V.W.L. Jordan, 1997. Phenology of fruit growth and susceptibility to greymould (Botrytis cinerea) of strawberry, raspberry and blackcurrant. Ann. Inst. Phytopathol.Benaki 18: 1–11.

Peng, G. and J.C. Sutton, 1991. Evaluation of microorganisms for biocontrol of Botrytiscinerea in strawberry. Can. J. Plant Pathol. 13: 247–257.

Powelson, R.L., 1960. Initiation of strawberry fruit rot caused by Botrytis cinerea. Phytopath-ology 50: 491–494.

Shirane, N. and Y. Watanabe, 1985. Comparison of infection process of Botrytis cinerea oncucumber cotyledon and strawberry petal. Ann. Phytopath. Soc. Japan 51: 501–505.

Sutton, J.C., 1995. Evaluation of micro-organisms for biocontrol: Botrytis cinerea and straw-berry, a case study. In: J.H. Andrews and I.C. Tommerup (eds), Advances in PlantPathology. Academic Press, San Diego, USA. pp. 173–190.

Tronsmo, A. and C. Dennis, 1977. The use of Trichoderma species to control strawberry fruitrot. Neth. J. Plant Pathol. 83: 449–455.

Wilcox, W.R. and R.C. Seem, 1994. Relationship between strawberry gray mold incidence,environmental variables, and fungicide applications during different periods of the fruitingseason. Phytopathology 84: 264–270.

Wilkinson, E.H., 1954. Observations on grey mould in strawberries. Plant Pathol. 3: 12.