An Evaluation of Some Procedures Used in Simple Ordinations

Abstract

Simple ordinations of the Bray and Curtis (1957) and perpendicular axes (Orloci 1966) types of construction were produced from 4 sets of data using measures of inter stands distance derived from the Bray and Curtis (1957) similarity coefficient (C) and an Euclidian measure of distance (D) (Orloci 1966). The use of D results in a more efficient interpretation of tne original interstand distances within the ordinations while the Bray and Curtis method of construction is marginally more efficient than the perpendicular axes method. However the use of D appears to obscure gradients of composition and fails to separate dissimilar communities, particularly where the stands are rich in species, and in this respect the resulting ordinations are less efficient than those using derivatives of C. The differences would seem to be in part inherent in the actual measures of distance used but are also influenced by the mode of standardization of the original species'' scores.