Disease Control and Pest Management
Sensitivities (ED _{50} values) of 300 monoconidial isolates of Venturia inaequalis to the sterol demethylation inhibitor (DMI) flusilazole were determined, based on the inhibitory effect on mycelial growth. Isolates were collected from three different orchards: in orchards 1 and 2, DMIs had never been used, whereas various DMIs had been tested for 12 yr in orchard 3. ED_{50} values for individual isolates were lognormally distributed, ranging from 0.0006 to 0.17, 0.0016 to 0.14, and 0.0007 to 0.065 ?g ml^{–1} in orchards 1, 2, and 3, respectively. Population means of the log transformed ED_{50} values were 0.0068, 0.01, and 0.076 ?g ml^{–1} for orchards 1, 2, and 3, respectively. Although the mean sensitivities were similar for all three sites, the mean ED_{50} value of orchard 2, in which DMI fungicides had never been used, was significantly higher than the mean of the two other orchards. Furthermore, the population of orchard 3, which was exposed to DMI fungicides for 12 yr, had not become more resistant to flusilazole compared to unexposed populations. Thus, differences in mean sensitivities of V. inaequalis populations are not necessarily related to the history of use of DMIs. Regardless of small differences among orchards, the variance of sensitivities determined for the three populations was homogenous, and all ED_{50} values could be combined in one distribution. Sample sizes necessary to detect differences in mean population sensitivities were determined based on the variation among all 300 isolates. A sample size of 50 was sufficient to detect a difference of 1.6 times the mean ED_{50} value. Sample sizes of >50 did not greatly improve the precision of the test, whereas with sample sizes of <15, the detectability of sensitivity differences among populations was decreased. The magnitude of growth inhibition at a single fungicide concentration close to the mean ED_{50} value of the baseline population was found to include a precise measure of flusilazole sensitivities and, thus, represented an appropriate alternative for the monitoring of population sensitivities. |