Determining Phytotoxic Pesticide Interactions in Soil

Abstract

Abstract Methods of determining and describing total phytotoxic actions resulting from pesticides combined in soil have not always been appropriate. The best design appears to be a simple factorial. The criteria selected for analysis should be in a form suitable for statistical evaluation. Regression estimate analysis can identify, characterize, and express magnitude of a physiological interaction. An analysis of variance indicates only the presence of an interaction, Duncan's Multiple Range Test identifies an interaction, and Colby's method indicates magnitude of differences. The regression estimate analysis overcomes some of the statistical and interpretive limitations of the heretofore used Duncan's Multiple Range Test and Colby's method for phytotoxic interactions.