VIEW ARTICLE

Techniques

Use of Near-Infrared Reflectance Spectrometry and Multivariate Data Analysis to Detect Anther Smut Disease (Microbotryum violaceum) in Silene dioica. M. Nilsson, Department of Forest Ecology, Swedish University of Agricultural Sciences, S-901 83 Umeå; T. Elmqvist(2), and U. Carlsson(3). (2)(3)Department of Ecological Botany, University of Umeå, S-901 87 Umeå, Sweden. Phytopathology 84:764-770. Accepted for publication 9 March 1994. Copyright 1994 The American Phytopathological Society. DOI: 10.1094/Phyto-84-764.

Near-infrared reflectance (NIR) spectral data was used in principal component analysis (PCA) to detect infection of Silene dioica by Microbotryum violaceum. Rosette leaf samples were accurately identified as either healthy (97%) or infected (96%) when NIR data was analyzed by PCA. The two classes overlapped slightly when principal component models were used to classify unknown samples. A method to measure the degree of infection is also presented. The use of NIR and PCA for both detection and quantification of fungal biomass in plant material should be useful for studying plant-pathogen interactions and as a method for assessing disease incidence in crops.

Additional keywords: chitin, pattern recognition.