Abstract
On the basis of the most recent publications, the present theories of classification are criticized. Four principal taxonomic systems compete now with each other. Two of them are essentially typologic: the traditional taxonomy (intuitive approach); the numerical or phenetic taxonomy (blind approach); the two others are evolutionist: the evolutionary taxononomy (mixed approach); and the phylogenetic or cladistic taxonomy (sister-group approach). The present disagreement between the Simpson-Mayr school and the Henning school is discussed. Partly new leading principles are proposed for the attainment of the reconstruction of nature's own hierarchy, i.e. the reconstitution of natural groups. A strict monophyletic origin is basically required: the taxon must exclusively include all, and only, the species which have descended from the same common ancestor. Through the establishment of morphoclines, i.e. phylogenetic series of transformation of homologous characters, plesiomorphous (primitive) characters and apomorphous (advanced) characters can be distinguished. The criteria necessary to the distinction are explained. The anagenetic advance of a clade can be rightly estimated through the agreement of several morphoclines. A subsequent process is constituted by the classification by grade i.e. the use of a definite level reached forward in evolution, that is to say a distinct grade (level) of organization. Many difficulties confront the taxonomist: convergence, parallelism, mosaic evolution