DEVELOPMENT OF AN INFECTION MODEL FOR BOTRYTIS BUNCH ROT OF GRAPES BASED ON WETNESS DURATION AND TEMPERATURE

Grape berries were dipped in conidial suspensions of Botrytis cinerea and incubated for 4, 8, 12, 16, or 20 h of wetness at temperatures ran ging from 12-30 C. Berries were infected after 4 h of wetness at all t emperatures tested. Incidence of berry infection increased with increa sing wetness duration at all temperatures. A multiple regression model described the legit of infection as a function of the interaction of wetness duration and temperature (R(2)=0.75). This model was incorpora ted into an infield environmental monitoring station and evaluated for two seasons on Thompson Seedless table grapes in the central valley o f Chile. Applications of captan (3.2 kg/ha a.i.) or vinclozolin (1 kg/ ha a.i.) were made according to the Botrytis model or a standard pheno logical spray program that consisted of four sprays (at bloom, cluster thinning, veraison, and preharvest) plus additional sprays after majo r rain events. Disease incidence and severity at harvest were similar whether applications were made according to a standard program (six to nine applications in 1991-1992, four to five in 1992-1993) or accordi ng to the Botrytis model (two to four applications in 1991-1992, zero to five in 1992-1993). In some vineyards, postharvest disease was sign ificantly less when sprays were made according to the Botrytis model r ecommendations compared with the standard spray program.