In the extent to which it is used, the concept of the potential natural vegetation (PNV) is one of the most successful novelties in vegetation science over the last decades. However, previous applications of the concept have shown that the theoretical principles were used inconsistently or interpreted in an incorrect sense. The present problems in application (which become evident when visualizing historical aspects of the concept) mainly result from (a) inconsistent treatment of the construction criteria; (b) failure to distinguish between the "potential natural vegetation", the "reconstructed natural vegetation" and the vegetation developing during succession, (c) the lack of a precise definition for reference terms to construct potential natural vegetation (e.g. treating reversible vs. irreversible changes of vegetation). For a sensible application of the concept it is suggested (a) to construct the potential natural vegetation on the basis of natural site conditions as well as permanently effective site changes as a consequence of human impact, (b) to consider the PNV to be in balance with all site conditions taken as basis for its construction. In practice, however, the construction basis may also derive from a particular question underlying the making of a PNV-map. A suggestion for a re-definition of the term "potential natural vegetation" as well as a key for PNV-mapping (valid for landscapes of Northern Germany) are given.