Cluster analysis of Wenlockian (Silurian) conodonts of the central United States

Abstract

The purpose of this investigation was to evaluate statistical methods for Silurian condont relative abundance analysis. Data collected by other researchers were used to test the statistical outcome against traditional methods of evaluation. It was determined that cluster analysis may be useful for dealing with the high number of conodont elements associated with conodont-bearing rocks. The FASTCLUS procedure of the statistical software program SAS/STAT was used to process data. The ensuing smalytical procedure produced results that were similar to those obtained from the literature, indicating that this procedure is useful for determining general trends in the data. It appears that this procedure is most useful with data containing various species that are not present at all locations or stratigraphic levels. It may follow that this type of statistical analysis could be used to include additional data with less effort than traditional methods of analysis.