Multivariate Analytical Treatment of Quantitative Species Associations: An Example from Palaeoecology

Abstract

The multivariate statistical methods of Principal Component Analysis and Factor Analysis (Joereskog''s modification) provide potentially useful means of studying frequency data of species in a community. The mathematical model on which factor analysis is based seems to be more applicable than that of principal component analysis. Seventeen species of ostracods from boreholes in the Paleocene deposits of Western Nigeria were analyzed by both methods. Both indicate the existence of 5 approximately equally important environmental factors. The procedures are unable to disclose whether a species is, for example, stenohaline or euryhaline, stenothermal or eurythermal, but they can show whether the species is stenooic or euryoic. The methods are mainly useful in paleoecology but they may also be applied to collections of Recent material on which there are little or no data other than counts on relative abundance of species. Details of the calculations are provided. The actual arithmetic work was done on electronic computers.