The experiments were conducted in the laboratories of the "Departamento de Fitopatologia" of the "Universidade Federal de Viçosa" and in commercial greenhouses and packinghouses in Barbacena, State of Minas Gerais, Brazil. Two experiments, repeated once, were conducted. In experiment I, rose flowers were pulsed with citric acid, salicylic acid, sucrose, and calcium sulfate, and in experiment II, rose flowers were pulsed with STS. The STS was evaluated separately from the other substances, because it may be toxic to flowers when they are exposed to it for too long, and because the other substances need more than 10 h to have an effect. In each experiment two assays were performed. In the first assay, we evaluated the effect of treatments on latent infections derived from the field and on infections occurring naturally after harvest. In the second assay, we evaluated the effect of treatments on flowers artificially inoculated with B. cinerea. In both experiments, inoculation was accomplished by using a DeVilbis atomizer and applying a volume of about 0.1 ml of a spore suspension of 104 conidia/ml to each flower. In the repetition of the experiments, the inoculated flowers were treated with double the volume of inoculum used the first time to determine if the efficiency of the treatments was altered with an increase in the amount of inoculum.
A scale of ratings showing the descriptions of the successive stages of floral opening was developed specifically to be used in the present work (Table 1).
A completely randomized experimental design with three replications of each treatment was used in all experiments.
Pulsing with citric acid, salicylic acid, sucrose, and calcium sulfate
Rose flowers of cv Kiss were harvested at three different stages: at the commercial harvest stage (stage I), which corresponds to a rating of 0 in Table 1; 24 h before stage I (stage II); and 48 h before stage I (stage III). A total of 900 flowers, 20 per replication of a total of three replications of each stage and substance, were placed into vases containing, separately, the solutions of citric acid at 0.8 mM, salicylic acid at 7.2 mM, sucrose at 2.4 mM, calcium sulfate at 50 mM, or water as controls. After pulsing for 15 h, the flowers were divided in two groups of 450 flowers each. One group was inoculated, as previously described, and covered with plastic bags. The other group was not inoculated to evaluate the effect of the treatments on latent infections originated in the field or naturally after harvest. Both inoculated and uninoculated flowers were packed separately and stored at 10 oC and RH>90%. After 36 to 48 h, the flowers were placed in vases containing tap water, and both the vase life and the severity of disease were evaluated using the scale (Table 1), and a scale used by Araújo (1995), which varies from 1 to 10, where 1 = 0%, 2 = 0-2%, 3 = 2-5%, 4 = 5-10%, 5 = 10-15%, 6 = 15-25%, 7 = 25-50%, 8 = 50-75%, 9 = 75-100%, and 10 = 100% of the diseased area in each petal, respectively.
Severity was considered the area of the petal that was diseased, relative to the total petal area. Before statistical analysis each severity rating was converted to a disease proportion, in which the midpoint of the interval was used (Campbell & Madden, 1990). The values of maximum severity and area under the disease progress curve (AUDPC) were compared among the treatments. The equation proposed by Shanner & Finney (1977) was used to calculate AUDPC, where:
in which yi is disease severity at time ti, in days, and yi +1 is disease severity at ti + 1.
Statistical analysis was done using ANOVA and Tukey's test (P=0.05).
Pulsing with silver thiosulfate (STS)
A total of 240 flowers of the cv. Kiss were harvested in stage 0 of the scale of ratings (Table 1) and placed separately in 20 l buckets half filled with tap water (four buckets with 60 flowers each). One group of 60 flowers was then transferred to a bucket containing 10 l of a 1 mM STS solution and let pulse for 120 min. Sixty-min after the first group was transferred to the STS solution, another group of 60 flowers was transferred to a bucket with the same solution. Finally, 90 min after the first group was transferred, another group of 60 flowers was transferred to the STS solution. In this way the treatments were pulsing for 120 min, 60 min, and 30 min, respectively. The control flowers were transferred to a bucket containing water at the same time the first group of flowers was transferred to the STS solution.
At the end of the pulsing time (2 h total), the sixty flowers of each bucket were divided into three groups of 20 flowers each, which were considered as replications. Ten flowers of each replication group of each treatment were inoculated as described before. The other ten flowers were not inoculated to evaluate disease severity due to latent infections from the field or infections initiated after harvest. Once inoculated, the flowers were covered with plastic bags, packed separately from the non-inoculated ones, and stored at 10 oC and RH>90%. After storing for 36 to 48 h the flowers were placed in vases, which were stored at 20 oC and RH>80% under continuous light, and both the vase life and the severity of disease were evaluated as described. Analysis of variance and linear regression (P = 0.05) were used for statistical analysis.