Biological control of Botrytis, Aspergillus and Rhizopus rots on table and wine grapes in Israel

Transcription

1 Postharvest Biology and Technology 20 (2000) Biological control of Botrytis, Aspergillus and Rhizopus rots on table and wine grapes in Israel Tirtza Zahavi a, *, Lea Cohen a, Batia Weiss a, Leonardo Schena b, Avinoam Daus a, Tania Kaplunov a, Johanan Zutkhi a, Ruth Ben-Arie a, Samir Droby a a Department of Fruit and Vegetable Storage, Agricultural Research Organization, The Volcani Center, Bet Dagan 50250, Israel b Dipartimento di Protezione della Plante dalle Malattie, Uni ersita degli Studi di Bari, Via Amendola 165/A, Bari 70126, Italy Received 20 December 1999; accepted 8 May 2000 Abstract One hundred and twenty-nine strains of epiphytic micro-organisms, isolated from table and wine grapes in Israel, were screened for antagonistic activity against Botrytis cinerea on table grapes. Two isolates (Candida guilliermondii, strain A42 and Acremonium cephalosporium, strain B11) were further evaluated for the control of decay in grapes caused by Aspergillus niger and Rhizopus stolonifer. Decay incidence caused by Botrytis cinerea, Aspergillus niger and Rhizopus stolonifer on wounded detached berries was reduced to 8, 14 and 22% respectively, by A42 and to 16, 82 and 60%, respectively, by B11. On small clusters with intact berries, decay was reduced to 30, 22 and 22%, respectively, by A42 and to 48, 39 and 30% respectively, by isolate B11. Both strains survived well under local vineyard conditions and during storage at 0 C and maintained relatively high cell counts on the berries. Field experiments were conducted in 1996, 1997 and 1998, with both table and wine grapes. Vines were sprayed with yeast suspension 2 5 times at 7 10 day intervals and decay was evaluated before harvest (wine grapes) or after storage (table grapes). A42 reduced decay caused by Botrytis cinerea in two of the three seasons in both table and wine grapes, and rots caused by Aspergillus niger in wine grapes were reduced significantly in 1997 and B11 reduced Botrytis cinerea development in the two years it was tested in wine grapes but in table grapes only in Morever, it did not control decay caused by Aspergillus niger Elsevier Science B.V. All rights reserved. Keywords: Vitis inifera; Yeast; Biocontrol; Grey mould; Candida guilliermondii; Acremonium cephalosporium 1. Introduction * Corresponding author. Tel.: ; fax: address: tirtzaz@yahoo.com (T. Zahavi). Botrytis cinerea, the cause of gray mold of grapes, is the dominant rot-causing pathogen in most regions of the world (Bulit and Dubos, 1988). In the warmer vine-growing areas of Israel Aspergillus niger and Rhizopus stolonifer are also /00/$ - see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S (00)

2 116 T. Zaha i etal. / Posthar est Biology and Technology 20 (2000) reported to cause decay and are of equal economic importance to B. cinerea (Barkai-Golan, 1981). Rots caused by B. cinerea and R. stolonifer usually begin to appear at the stage of veraison (fruit softening) when sugar content in the berries reaches 10 12% (Hill et al., 1981; Marois et al., 1992). A. niger usually causes decay only at higher sugar levels (15%), although it is one of the fungi found on the surface of healthy grapes at all stages (Zahavi, unpublished data). The relative importance of the several rot-causing fungi, both in the vineyard and in stored grapes, changes from one season to another, probably because of differing climatic conditions. Since there is no way of knowing, at the onset of veraison (the recommended time to start chemical control programs), which pathogen will be predominant, it is difficult to recommend a specific chemical control program in the vineyard. After harvest, SO 2 is generally very effective in preventing the development of decay (Ben Arie et al., 1991), but effectiveness depends on the pathogen and the inoculum load. Higher amounts of SO 2, which might be more efficient, cause bleaching of the berries and an off-flavour. Recently, SO 2 has been removed from the federal government s generally regarded as safe (GRAS) list in the USA and its future use on grapes is therefore uncertain. Public demand to reduce pesticide use, stimulated by greater awareness of environmental and health issues, as well as development of resistance of some of the pathogens to the fungicides, limits the application of chemicals on agricultural products. In recent years much research has focused on developing alternative control methods against pre- and postharvest decay in grapes as well as other agricultural commodities (Ferreira, 1990; Peng and Sutton, 1991; Wilson et al., 1991, 1993; Filonow et al., 1996; Fokkema 1996; Harman et al., 1996; Leibinger et al., 1997). In previous work, we evaluated the activity of the yeast strains, Kloeckera apiculata and Candida guilliermondii (strain U.S.7) in reducing postharvest decay of table grapes (Ben Arie et al., 1991; McLaughlin et al., 1992). When applied as a postharvest dip, both antagonists were found to protect injured and artificially inoculated grape berries and also to reduce the incidence of decay of naturally infected, non-injured grapes. However, one of the important quality parameters of table grapes is the presence of a bloom on the surface of the berry. A postharvest dip removes some of the bloom, thereby adversely affecting the perceived quality of the treated fruit. Therefore, the possibility of treating the fruit with antagonists before harvest was evaluated (Ben Arie et al., 1991). The efficacy of the yeast C. guilliermondii in stopping disease development in the vineyard and in reducing postharvest decay was demonstrated. However, the treatment did not retain its effectiveness and the incidence of Botrytis decay increased considerably during 4 weeks storage at 0 C. This may be attributed to poor colonization and survival of this particular strain, which had been isolated from the surface of lemon. We therefore decided to look in the vineyard for effective antagonists of grape pathogens that are better adapted to the pre- and post-harvest environment. In the present paper we describe the isolation of natural epiphytic yeast antagonists from Israeli vineyards, and the evaluation of their biocontrol activity against bunch rots under laboratory and field conditions. 2. Materials and methods 2.1. Yeast isolation Epiphytic micro-organisms were isolated by shaking five grape berries in 10 ml of sterile distilled water for 1 h at 200 rpm on a rotary shaker (Peng and Sutton, 1991). The wash was serially diluted and 30 l of each dilution were spread on basal yeast agar (BYA) containing 20 g glucose, 1 g yeast extract, 10 g protease peptone and 15 g agar amended with 250 mg Penicillin G (to suppress growth of bacteria) in 1 l of distilled water. The Petri dishes were incubated at room temperature for 4 days and yeast-like colonies were selected randomly according to colour and morphological characteristics, removed with a sterile plastic loop and transferred to fresh BYA plates to obtain pure cultures. Colonies were held

3 T. Zaha i etal. / Posthar est Biology and Technology 20 (2000) at 4 C until used. Selected isolates were identified by the Centraalbureau voor Schimmeelcultures (Baarn, The Netherlands) Bio-control assay on detached berries Cultures were grown on a rotary shaker at 200 rpm in 50 ml nutrient-yeast dextrose broth (NYDB) for 48 h at room temperature (20 25 C). Cells were then pelleted by centrifugation and resuspended in an equal volume of sterile distilled water to give a final concentration of cells/ml. Conidia of B. cinerea were obtained from 2 4 week old PDA cultures incubated at room temperature. Spores were suspended in sterile distilled water, filtered through two layers of cheesecloth and the spore concentration was adjusted to conidia/ml. Individual berries of Thompson Seedless grapes were removed from clusters, surface disinfected by dipping for 1 min in 1% (v/v) sodium hypochlorite (ph 11.5) and mounted on masking tape strips glued to PVC pads on the floor of the incubation box. The berries were punctured with a pin (2 mm deep) and 10 l of an antagonist cell suspension were pippetted onto the wound site and left to dry for 1 2 h, after which the berries were inoculated with 10 l of the conidial suspension. Each treatment was applied to three replicates of seven or eight berries. Following treatment, wet filter paper was placed in the boxes which were covered with polyethylene to maintain high relative humidity. The percent of decayed berries in each replicate was evaluated after 4 5 days at 20 C. The effect of two isolates (A42 and B11, which were the best antagonists in the screening tests), on the control of decay caused by A. niger and R. stolonifer and the effect of cell concentration on biocontrol activity were tested on detached berries as described above, with minor changes. The isolates were grown in NYDB and cell concentrations were adjusted to 10 8,10 7 and 10 6 cells/ml. Conidia of B. cinerea and A. niger and sporangiospores of R. stolonifer were obtained from 1 2 week old PDA cultures incubated at room temperature (20 25 C) and suspended in sterile distilled water at concentration of CFU/ml. Analysis of variance, for all laboratory experiments, was carried out using Duncan s multiple range test Biocontrol acti ity on grape clusters Biocontrol activity of isolates A42 and B11 was evaluated against B. cinerea, A. niger and R. stolonifer on small bunches of grapes. Clusters of Thompson Seedless grapes were divided to make bunches of ten berries. Antagonists were cultured as described above and diluted 1:10 in sterile distilled water. The bunches were dipped in the antagonist suspension, allowed to dry for 2 3 h and then sprayed with fungal suspensions ( CFU/ml) of B. cinerea, A. niger or R. stolonifer. Bunches were incubated at 20 C for 4 5 days and percent infection was determined for each bunch. Each treatment consisted of four replicates of five small bunches Sur i al of antagonist yeasts in the ineyard Survival of the antagonists under Israeli vineyard conditions and in storage was determined in the table-grape experiments. Samples were collected on the first spraying date, after the clusters had dried and thereafter before each spray. Five berries per plot were sampled aseptically into 150 ml sterile cups containing 20 ml of water and shaken on a rotary shaker at 200 rpm for 1 h. After serial 1:10 dilutions, 20 l of each dilution were plated in Petri dishes containing BYA and the plates were held at room temperature for 3 4 days, after which the number of colonies was counted. Antagonist survival was determined as CFU/berry on clusters that received a spray every week and on clusters that were sprayed only once, at the beginning of the experiment. Their survival was also followed after harvest on fruit packed as described and held at either 0 or 20 C. Data analysis (mean and standard error) was carried out after logarithmic transformation of the data. On two sampling dates, samples of recovered isolates were compared to pure cultures of the antagonists, using rapid DNA isolation and RAPD PCR with two different sets of primers, as described by Schena et al. (1999).

4 118 T. Zaha i etal. / Posthar est Biology and Technology 20 (2000) Field experiments The efficacy of the antagonists A42 and B11 against bunch rots of wine and table grapes was evaluated during 1996, 1997 and 1998 on Thompson Seedless and Superior Seedless (table grapes) and Sauvignon blanc (wine grapes) in vineyards located in the southern coastal plains (Lachish) and Golan heights (Yonatan). Untreated vines and chemically treated vines served as controls. Details of the field experiments conducted on wine and table grapes in are shown in Table 1. Antagonists were grown for 48 h in 1 l bottles with NYDB, on a rotary shaker as described above. Cells were pelleted, resuspended in tap water at the initial concentration and diluted ten fold (except in the 1996 experiment with table grapes, where the cultures were used at the original concentration). Experimental plots consisted of one to seven vines per treatment in the different experiments, arranged as randomized blocks with four replicates. The antagonists and chemical controls were applied two to five times until run-off, with a motor-driven back-sprayer. The incidence of decay in the wine grape experiments was determined on the day of harvest. Forty clusters were sampled from each plot and scored according to the causal agent of the decay and the percentage of rot. In the table grape experiments, no decay developed in the vineyard and rot was evaluated only after storage. Approximately 3 kg of grapes were harvested from each plot and packed in plastic boxes which were wrapped in polyethylene bags to create high relative humidity. In 1996, the last spray was applied on the day of harvest and the grapes were picked after the clusters had dried. In the Thompson Seedless experiments in 1997 and 1998 there were two harvests: the first one week after the penultimate spray, and the second after the clusters had dried following the last spray. Rot development was evaluated after 3 4 weeks storage at 0 C followed by 3 4 days at 20 C. After arc-sin transformation of the data, analysis of variance was carried out by Duncan s multiple range test using the SAS GLM (SAS Ins. Cary, NC) procedure. 3. Results 3.1. Screening for antagonists among epiphytic microflora The antagonistic activity, against B. cinerea of Table 1 Details of field experiments conducted on wine and table grapes in Wine Table Wine Table Wine Table Isolates A42, B11 A42, B11 A42, B11 A42, B11 A42 A42 Dilution a 1:10 1:1 1:10 1:10 1:10 1:10 Replicate size b 7 vines 1 vine 7 vines 3 vines 4 vines 3 vines Cultivar S.B. c T.S. S.B. T.S. S.B. T.S. & S.S Spray frequency Weekly Weekly Weekly Weekly 10-days 14 days (Superior) days (Thompson) No. of sprays (Superior) 4 (Thompson) Augmentation spray + +/ Chemical control Iprodione (0.075%) d & Iprodione (table grapes) Fluazinam Pyrimethanil (0.08%) Benzamidazole (0.025%) (0.05%) (wine grapes) a Dilution factor from the concentration at the stationary phase. b Four replicates in each experiment. c Sauvignon blanc, S.B.; Thompson Seedless, T.S.; Superior Seedless, S.S. d Active ingredient.

5 T. Zaha i etal. / Posthar est Biology and Technology 20 (2000) Table 2 Efficiency of vineyard yeast isolates in suppressing rot caused by B. cinerea on detached Thompson Seedless berries a Percent inhibition No. of isolates No. of isolates of rot b (1993) (1995) Number of tested isolates a Detached Thompson Seedless berries were wounded with a pin, treated with 10 l antagonist cell suspension and inoculated with 10 l of conidia/ml suspension of B. cinerea. b Percent of control Biocontrol on small bunches The infection levels in the small bunches, inoculated by spraying the pathogen cell suspension and without wounding the berries, were lower than those in the detached wounded berries. Final infection levels in the control treatments were 70, 72 and 76%, respectively, for Aspergillus, Botrytis and Rhizopus (Fig. 2). The isolates were equally effective in reducing the decay caused by all three fungi. B11 significantly (P=0.05) reduced decay development by Rhizopus, Aspergillus and Botrytis to 50, 54 and 63%, respectively, in comparison isolates collected over 2, years was tested on detached berries. Biocontrol activity of the various isolates in reducing the number of decayed berries ranged between 0 and 93% compared to decay development on wound-inoculated berries to which no isolates were applied (Table 2). Two isolates, A42 and B11, were selected for further studies. These were identified by Centraalbureau voor Schimmeelcultures (Baarn, The Netherlands) as Candida guilliermondii (A42) and Acremonium cephalosporium (B11) Biocontrol acti ity on detached berries The isolates (A42, B11) showing the highest antagonistic activity in the screening tests were tested for their efficacy to inhibit the three grape pathogens, B. cinerea, A.niger and R. stolonifer on detached single berries. Results presented in Fig. 1 show that both isolates were significantly effective (P=0.05) in reducing the development of B. cinerea, A. niger and R. stolonifer when applied at 10 8 CFU/ml. At lower concentrations both demonstrated reduced efficacy against the three pathogens. It is worth noting that, of the three pathogens, A. niger was the least affected by both isolates when applied at 10 7 and 10 6 CFU/ml. Fig. 1. Effect of antagonists on rot development in wounded detached berries. Berries were punctured once with a pin (2 mm deep). Ten microliters of antagonist cell suspension was pipetted onto each wound and when dried, berries were inoculated with 10 l of fungal spore suspension ( CFU/ml). Decay development was assessed 4 days after inoculation. Bars indicate means and standard errors.

6 120 T. Zaha i etal. / Posthar est Biology and Technology 20 (2000) number of CFU/berry found 3 weeks after a single spray did not differ from that found on clusters sprayed weekly (Fig. 3A). Both isolates survived well on fruit stored for one month at 0 or 20 C (Fig. 3B) Field experiments Figs. 4 and 5 summarize the effects of the different treatments on decay development in wine and table grapes, respectively. The predominant rot pathogen in the experiments on wine grapes, conducted in the Golan heights, was Aspergillus niger affecting 32, 7 and 37% of the clusters in the untreated control in 1996, 1997 and Fig. 2. Effect of antagonists on rot development on small grape bunches. Small bunches (6 10 berries) were dipped in an antagonist cell suspension (ca CFU/ml). When dry, the clusters were sprayed to run-off with a CFU/ml fungal spore suspension. Rot (percent of decayed berries) was evaluated twice, 4 6 days after inoculation. Bars indicate means and standard errors. with the control bunches, while A42 gave significant (P=0.05) reductions of 36, 58 and 43%, respectively. The extent of rot increased as the holding time at 20 C increased Sur i al of antagonists in the field During the experimental periods in 1996 and 1997, both isolates survived well under the conditions existing in the field and storage (Fig. 3A and B are the results from the 1996 experiment). The Fig. 3. Antagonist survival. Grape clusters were sprayed with a suspension of yeast cells (10 8 CFU/ml). When the clusters had dried and every following week, five berries were sampled from each plot into sterile cups with 20 ml of sterile water. The berries were shaken for 1 h on a rotary shaker and serial dilutions of the wash were plated on BYA in petri dishes. Colonies were counted after 3 4 days at room temperature. In A, lines represent survival on clusters that were sprayed weekly and bars represent survival 3 weeks after a single application. Survival during storage after a weekly field application (B) was followed in grapes held at 0 (solid line) or at 20 C (dashed lines).

7 T. Zaha i etal. / Posthar est Biology and Technology 20 (2000) the 1996 experiment, had no effect on the percentage of decayed bunches. In 1997, Fluazinam (0.05% a.i.) prevented the development of rot caused by Botrytis cinerea but had no effect on the incidence of Aspergillus niger. Only in 1998, when Pyramithanil (0.08% a.i.) was used as the chemical control, was there an adequate reduction in total decay incidence compared to the untreated control. Neither antagonists reduced decay incidence in 1996 compared with the untreated control. B11 had no effect in the 2 years it was applied and therefore was not included in the experiment in A42 reduced decay incidence significantly in both 1997 and 1998, and was as effective as the application of the chemicals. On stored table grapes, Botrytis cinerea was the predominant rot pathogen. Decay severity (percentage of decayed berries) was reduced by both isolates examined in 1996, compared with the non-treated control (Fig. 5). Only A42 reduced decay in 1997 and an augmentation spray of the yeast, applied on the day of harvest (marked in the graph A42+), did not increase the effect. B11 reduced decay severity in the 1996 experiment, although less than A42; in 1997 it had no effect and was, therefore, not included in the 1998 experiments. Two experiments were conducted in 1998, one with Thompson Seedless and the other with Superior Seedless. Only A42 was included in these experiments, but no reduction of decay was achieved. 4. Discussion Fig. 4. Effect of antagonist sprays on decay development in wine grapes. Vines were sprayed from veraison to harvest at 7 10 day intervals (five sprays in and four in 1998). Incidence of decay, caused by Botrytis cinerea and Aspergillus niger, was monitored 1 2 days before harvest. Bars indicate means and standard errors (see Table 1 for details of chemical treatment in each year). 1998, respectively, while rot caused by Botrytis cinerea developed on 8, 9 and 1.5% of the clusters, respectively. Benzamidazole (0.025% a.i.) and Iprodione (0.075% a.i.), the chemical control in The majority of organisms we isolated from grape berry surfaces showed some ability to reduce decay development in the initial tests. However, the natural epiphytic population isolated was very diverse in its propensity to reduce decay by rot fungi and only a small percentage of the isolates tested reduced decay development to a level that could be considered significant under commercial conditions (80%). None of the yeasts increased the level of berry rot, although, yeasts are considered as part of the sour rot complex in grapes (Bisiach et al., 1986).

8 122 T. Zaha i etal. / Posthar est Biology and Technology 20 (2000) The two antagonists selected for further study were efficient in reducing decay caused by Botrytis, Aspergillus and Rhizopus, both in wounded detached berries and on small clusters with intact berries that were artificially inoculated after application of the antagonists. Wound-inoculation with pathogenic fungi is ideal for initial screening of antagonists but is very different from the real situation in the field. The second method, whereby the bunches were dipped in the antagonist suspension and then sprayed with a spore suspension of the pathogen, represents a situation closer to field biological control systems, where the fungi are sparse, most berries are not wounded and the antagonist, applied as a spray, covers the fruit. The improved control obtained in this system seems promising. Fokkema (1996) states that in order to prevent a pathogen from establishing itself on the plant, it is important to have the biological control agent on the fruit surface before the arrival of the propagules of the pathogen. This can be achieved either through very frequent applications, or by using strains that can survive in the field (Benbow and Sugar, 1999). Initial screening in the laboratory is, therefore, only the first step in looking for antagonists to diseases that start their development on the plant. To obtain a significant effect in the vineyard it is necessary to choose among the antagonists that perform best in the laboratory, those that can survive and colonize grape berries in the vineyard. Our results show that the use of indigenous microflora ensures that a high proportion of them will survive under field conditions. Molecular methods, like RAPD-PCR can be used to compare different yeast strains (Zimand et al., 1994; Wolfgang et al., 1997) and in the present study, Fig. 5. Effect of antagonist sprays on decay development in stored table grapes. Vines were sprayed from veraison to harvest at 7 10 day intervals. At harvest, 3 kg of grapes were harvested, packed and held at 0 C for 4 5 weeks. Decay severity, caused by Botrytis cinerea, Rhizopus stolonifer and Aspergillus niger was monitored after 4 7 days at 20 C following cold storage. In the 1997 experiment, the + signs indicate clusters that received an augmentation spray on the day of harvest. Bars indicate means and standard errors.

9 T. Zaha i etal. / Posthar est Biology and Technology 20 (2000) RAPD-PCR in conjunction with techniques we developed for rapid extraction of yeast DNA, proved to be a useful tool to test whether the yeasts retrieved from the grape surface after field application were, in fact, the ones that had been applied. In field tests reduction of bunch-rots was significant in most cases. A pre-harvest augmentation spray did not increase the antagonistic effect. This can be explained by the fact that the antagonists survived well on the fruit for 2 weeks and more. The results of the table grape experiment in 1998 in which no reduction in decay development was achieved, may be due to the lower survival of the antagonist in that year during which extremely high temperatures occurred. Many factors can affect the survival of the bio-antagonist as they affect the development of the rot causing organisms (Stapelton and Grant, 1992; Elad and Kirshner, 1993). Dik (1990) showed that development of a yeast population on wheat depends on air temperature, humidity and broad-spectrum fungicides (Captan). Although the biological control agents we used in this study were indigenous to the vineyard, their efficient use and the number of times they should be applied before harvest will depend on factors affecting the physical and chemical microclimate. Those affects must be determined in order to improve their biocontrol efficacy. The future use of biological agents for disease control will depend greatly on the price of the product, which, in turn, will depend on the ease of growing the organism (Fokkema, 1996) and on the dilution factor needed from culture to commercial spray. When we worked with small clusters, instead of checking the exact concentration of the antagonist in suspension, we used a dilution factor of 1:10 from the concentration at the stationary phase of each antagonist. Different organisms reach the stationary phase at different concentrations and strains that can be effective at a high dilution rate will be less expensive as control agents. Chemical control of grape decay is difficult because of the diverse fungi involved. The chemicals we used gave adequate control of B. cinerea but only the newly registered fungicide Pyrimethanil (applied in 1998) protected the grapes from A. niger rot, the predominant fungus on wine grapes. The antagonists, though less efficient against this fungus, in most cases, reduced the percentage of diseased clusters. One of the problems with chemical control is the hazard to the consumers and an IPM program combining early chemical sprays, to reduce the level of fungal inoculum, with antagonist application close to harvest, could be the logical solution to benefit consumers, growers and the environment. References Barkai-Golan, R., An annotated check-list of fungi causing postharvest diseases of fruit and vegetables in Israel. Special Publication No. 194, Division of Scientific Publications, The Volcani Center, Bet Dagan, Israel. Ben Arie, R., Droby, S., Zutkhi, J., Cohen, L., Weiss, B., Sarig, P., Zeidman, M., Daus, A., Chalutz, E., Preharvest and postharvest biological control of Rhizopus and Botrytis bunch rots of table grapes with antagonistic yeasts. In: Wilson, C., Chalutz, E. (Eds.), Proc. Workshop Biol. Control Postharvest Dis. Fruits and Veg. U.S. Dep. Agric. Agric. Res. Serv. Publ. 92, pp Benbow, J.M., Sugar, D., Fruit surface colonization and biological control of postharvest diseases of pear by preharvest yeast applications. Plant Dis. 83, Bisiach, M., Minervini, G., Zerbetto, F., Possible integrated control of grapevine sour rot. Vitis 25, Bulit, J., Dubos, B., Botrytis bunch rot and blight. In: Pearson, R.C., Goheen, A.C. (Eds.), Compendium of Grape Diseases. APS Press, The American Phytopathological Society, MN, pp Dik, A., Population dynamics of phyllosphere yeasts: influence of yeasts on aphid damage, diseases and fungicide activity in wheat. Ph.D. Thesis, University of Utrecht. Elad, Y., Kirshner, B., Survival in the phylloplane of an introduced biocontrol agent (Trichoderma harzianum) and populations of the plant pathogen Botrytis cinerea as modified by biotic conditions. Phytoparasitica 21, Ferreira, J.H.S., In vitro evaluation of epiphytic bacteria from table grapes for the suppression of Botrytis cinerea. S. Afr. J. Enol. Vitic. 11, Filonow, A.B., Vishniac, H.S., Anderson, J.A., Janisiewicz, W.J., Biological control of Botrytis cinerea in apples by yeasts from various habitats and their putative mechanisms of antagonism. Biol. Cont. 7, Fokkema, N.J., Biological control of fungal plant diseases. Entomophaga 41, Harman, G.E., Latorre, B., Agosini, E., San Martin, R., Riegel, D.G., Nielsen, P.A., Tronsmo, A., Pearson, R.C.,

10 124 T. Zaha i etal. / Posthar est Biology and Technology 20 (2000) Biological and integrated control of Botrytis bunch rot of grape using Trichoderma spp. Biol. Cont. 7, Hill, G., Stellwaag-Kittler, F., Huth, G., Schlosser, E., Resistance of grapes in different developmental stages to Botrytis cinerea. Phytopathol. Zeitsch. 102, Leibinger, W., Breuker, B., Hahn, M., Mendgen, K., Control of postharvest pathogens and colonization of the apple surface by antagonistic microorganisms in the field. Phytopathology 87, Marois, J.J., Bledsoe, A.M., Bettiga, L.J., Bunch rots. In: Flaherty, D.L. (Ed.), Grape Pest Management. University of California, Davis, pp McLaughlin, R.J., Wilson, C.L., Droby, S., Ben-Arie, R, Chalutz, E., Biological control of postharvest diseases of grape, peach and apple with the yeasts Kloeckera apiculata and Candida guilliermondii. Plant Dis. 76, Peng, G., Sutton, J.C., Evaluation of microorganisms for biocontrol of Botrytis cinerea in strawberry. Can. J. Plant Pathol. 13, Schena, L., Ippolito, A., Zahavi, T., Cohen, L., Nigro, F., Droby, S., Genetic diversity and biocontrol activity of Aurobasidium pullulans isolates against postharvest rots. Postharvest Biol. Technol. 17, Stapelton, J., Grant, S., Leaf removal for non-chemical control of the summer bunch rot complex of wine grapes in the San Joaquin valley. Plant Dis. 76, Wilson, L.C., Wisnieski, E.M., Biles, L.C., Mclaughlin, R., Chalultz, E., Droby, S., Biological control of postharvest diseases of fruit and vegetables: alternatives to synthetic fungicides. Crop Prot. 10, Wilson, L.C., Wisnieski, E.M., Droby, S., Chalultz, E., A selection strategy for microbial antagonists to control postharvest diseases of fruit and vegetables. Sci. Hortic. 53, Wolfgang, L., Breuker, B., Matthias, H., Mendgen, K., Control of postharvest pathogens and colonization of the apple surface by antagonistic microorganizms in the field. Phytopathology 87, Zimand, G., Valinsky, L., Elad, Y., Chet, I., Manulis, S., Use of RAPD procedure for the identification of Trichoderma strains. Mycol. Res. 98,