Multivariate analysis in process morpholoy of plants

This study presents the quantification of process morphology by means of multivariate analysis, particularly discriminant analysis (DA) and principal components analysis (PCA). According to process morphology, form is process. This process is analyzed in terms of 13 parameters (variables) of growth. The values of these parameters, which are quantified, represent specific processes such as determinate or indeterminate growth, radial or dorsiventral growth, branching or lobing, etc. Any particular structures such as a thallus, a telome, a stem, or a leaf is seen as a specific combination of the specific processes. Process combinations representing a wide variety of typical and atypical or controversial structures of flowering plants as well as typical structures of other plant groups (a thallus, a telome, an enation, a microphyll and megaphylls) were selected. DA and PCA were then used to quantitatively determine the relations of these process combinations. The general picture that emerged is a patterned continuum. The pattern is due to the fact that some regions of the continuum are much denser than others, i.e. they are occupied by a greater number of process combinations. Obviously the dense regions are those process combinations representing the structural categories, root, stem/shoot, leaf, trichome as well as thallus, telome and others. Since the emphasis in this study was on flowering plants, the relations of process combinations in algae, bryophytes and more primitive vascular plants were not demonstrated in detail. However, the present approach could be used to establish evolutionary change in the pattern of process combinations. It also provides a basis for phylogenetic reconstruction if one agrees that phylogeny is a change in the dynamics of successive ontogenies.