Graphical presentation 0f intercropping data for two levels of error variation

Abstract

This paper is concerned with problems in the analysis of intercropping experiments when the random sample comes from a bivariate population. The airn is to find how to represent and interpret information obtained from two levels of error variation, such as in experiments using a split-plot design. The literature is reviewed and a method of calculating a linear transformation is derived and evaluated for the case when two different residual correlations (main-plot and split-plot) occur