Biological control of grey mould (Botrytis cinerea Pers.: Fr.) of greenhouse-grown tomatoes in Crete using Bacillus brevis

Bacillus brevis Nagano, gramicidin S producing strain and B. brevis E1 an antibiotic negative mutant were tested for their efficacy to control disease in two growing seasons. B. brevis Nagano had been extensively tested in Scotland, UK and it was found to be very effective again at B. cinera in Chinese cabbage and tomato crops. Also the strain E1 which produces an unidentified biosurfactant which is believed to reduce leaf wetness duration was tested for the first time in planta. Along with these well characterized strains three native Bacillus strains were tested in the greenhouse. One B. subtilis and two B. pumilus strains producing antifungal compounds were tested in vitro, in vivo and in planta. B. brevis E1 proved to be the most effective treatment. Its effectiveness was assigned to the performance of the biosurfactant under conditions of temperatures ranging between 16-24°C when percentage relative humidity was marginally below 90%. B. brevis Nagano WT which also produces biosurfactant was moderately effective in both years. The reduction of antibiotic activity due to binding of gramicidin S on ethanol soluble components of the leaf surface is considered as a possible cause for its reduced activity. Both strains managed to delay epidemics and reduce the rate of disease in an overall study of disease progress. The consistent level of disease reduction (despite the variability of the epidemics over the two years), indicated that B. brevis E1 is a potential BCA. Disease reduction was affected mostly by environmental factors as well as the survival of the antagonists but clear conclusions on the duration of this effect were not drawn. The number of viable colony forming units of B. brevis on the phylloplane decreased following an inverse exponential pattern in most of the cases. An increase was occasionally monitored which was difficult to be attributed to any abiotic or biotic factor. Biological control agents did not seem to have any detrimental effect on naturally occurring fungi and bacteria, but they were inadequate to control B. cinerea quiescent infection of flowers.