The critical timing of infections that cause Botrytis bunch rot (BBR) has long been a focus of international debate: Some believe in the primacy of infections that occur at bloom (which remain latent thereafter until the preharvest period, when they become active and rot the berries), whereas others believe that post-veraison infections cause the most damage. In this project, we have shown that latent infections can readily become established when clusters are inoculated at bloom or during the early stages of berry development, but that the vast majority of these infections fail to become active, i.e., they never cause disease. In contrast, berries become much more susceptible to BBR after veraison, and inoculating just one or two berries per cluster at that time resulted in up to 50 to 60 additional diseased berries per cluster at harvest. Thus, severe disease losses appear to result largely from secondary disease spread during the post-veraison period. Nevertheless, bloom-time infections can serve the critical function of primary disease foci (i.e., the sources of secondary spread) if they do become active before harvest. Viewing bloom versus post-veraison infections as the primary and secondary phases of disease development, respectively, largely resolves the dispute over their relative importance, since it is the combination of the two that causes serious losses. High berry N content, high relative humidity during the preharvest period, and high soil water content after veraison were all shown to promote the activation of latent infections established at bloom; thus, these factors can help initiate an epidemic preharvest. Tightly compressed bunches were shown to dramatically facilitate post-veraison disease spread, and this appears to be the primary factor responsible for the long-observed correlation between cluster compactness and BBR severity. High berry N content also was shown to promote secondary spread under conditions of otherwise light to moderate disease pressure. Fungicide programs contributed to disease management by reducing both the initial establishment of primary infections (sprays at bloom) and their spread upon activation (sprays at bunch closure, veraison, and/or preharvest); the relative importance of specific spray timings depended on the particular year of the trial and weather events during these periods. When applied at veraison, the fungicide, cyprodinil (Vangard), showed some ability to eradicate latent infections that had been established at bloom, whereas fenhexamid (Elevate), iprodione (Rovral), and boscalid (Endura) did not. Following applications at bloom and bunch closure, Vangard also appeared to provide significantly greater residual protection within the berries at veraison than did the other tested materials.
