Estimating the level of susceptibility to Phytophthora infestans in potato genotypes

Resistance and susceptibility are closely related terms but differ in their underlying assumptions and measurement. Standardized methods for determining the level of resistance and susceptibility in potato to Phytophthora infestans, the causal agent of late blight, have traditionally been semiquantitative and are not based on a true interval scale, thus making their use in most mathematical and statistical operations inappropriate. Recently, researchers have attempted to develop interval scales using regression analysis of the direct or transformed area under the disease progress curve (AUDPC). In this article, a similar approach is described based on the relative AUDPC (RAUDPC) of one or two reference cultivars and tested using a data set of field trials involving cultivars with varying levels of susceptibility evaluated in different environments in several countries. The coefficient of variation (CV) among trials of the AUDPC was reduced when the RAUDPC was used and even more so when the RAUDPC was made relative to the RAUDPC of cv. Bintje (RaRAUDPC), which was present in all trials. The RaRAUDPC was used in regression models to estimate scale values for eight potato cultivars in 13 to 15 locations (depending on cultivar). The CVs of scale values measuring variation among sites were similar to those of the RaRAUDPC. Using two cultivars gave a slight improvement in CV, which was statistically significant. The scale developed here has ascending numbers for increasing susceptibility, is simple, and can be constructed as a ratio measure, which permitted the calculation of mean, variance, and CV.