Comparison of multivariate analyses using taxonomic data of Oxalis

Publisher Website
Google Scholar
Abstract
Several multivariate techniques are used to analyze the resemblances between the 25 North American taxa of Oxalis section Ionoxalis. Results from each of three clustering strategies are subjected to stepwise discriminant analysis for refinement and reallocation. The resultant phenetic classifications are compared with each other and with a recent conventional treatment using the same data. Affinities of the taxa, along with new relationships indicated by these analyses, are discussed. A polythetic hierarchical clustering method using an information statistic produces the most efficient and reliable phenetic classification and is recommended for use in systematic studies.

This publication has 2 references indexed in Scilit: