The application of cluster and discriminatory analysis in mapping granite intrusions

Abstract

This paper outlines an approach to the mapping of granitic rocks in which the numerical classificatory procedures of cluster and discriminatory analysis are used to supplement the largely subjective approach of conventional field mapping. Two examples are given. In one there is good agreement between the numerical and field methods. In the other example the numerical approach indicates the presence of two chemically distinct granites in what was believed, on the basis of field and petrographic evidence, to be a single homogeneous granite.