1Nottingham University, Sutton Bonington Campus, Loughborough, Leics LE12 SRD, UK; 2ADAS Arthur Rickwood, Mepal, Ely, Cambs CB6 2BA, UK

Background and objectives
Interest in biological control has been stimulated by public concern about a perceived excessive use of synthetic pesticides in crop protection. To date, however, there has been only limited success in the development of biological control agents (BCAS) with consistent activity when used on crops grown under commercial conditions. Extremes of temperature, humidity and leaf wetness encountered in field cropping may not be conducive to the growth and activity of antagonists. Application of BCAs to crops grown in a protected environment could enhance the likelihood of successful disease control. Grey mould of lettuce, caused by Botrytis cinerea, was therefore selected as a model system to evaluate the feasibility of biological control of a disease of a protected crop.

Materials and methods
Existing collections of potential antagonists, including isolates which had previously shown antifungal activity, were supplemented by bacteria, actinomycetes and fungi isolated from from lettuce leaves and roots and from adjacent soil. All isolates were screened for anti-Botrytis activity using an in vivo assay which utilised either leaf-discs or small lettuce plants. Lead strains which showed promising activity were then used as spray treatments in subsequent polythene tunnel trials on lettuce (cv. Kathy) grown under commercial conditions. The BCAs were evaluated both singly and in combination, and in comparison with a standard fungicide.

Results and discussion
Over 700 bacteria, 150 actinomycetes and 32 strains of Trichoderma harzianum were evaluated in vivo for Botrytis control. Ten bacteria, three actinomyctes and three Trichoderma strains gave very goood control in the detached leaf-disc assays, comparable to that given by the fungicide ipridione (Rovral WP). The three most active antagonists obtained were all identified as Bacillus subtilis. Interestingly none of the most active strains were isolated from lettuce, but were obtained from oilseed rape roots in previous research [1]. In a series of the polythene tunnel trials, moderate disease levels were obtained by placing infected spreader plants within the crop. Significant reductions in Botrytis were achieved with several strains following application of 3 or 4 sprays at 2-week intervals from soon after planting. BCAs which gave good control in one trial did not always perform well in subsequent trials. The degree of control achieved by the best BCAs was around 50%, almost equal to that given by ipridione.

1. Fiddaman PJ, Rossall S, 1995. Plant Pathology 44, 695-703.