All definite NPs carry a presupposition of existential uniqueness. Thus we can access an entity which is presented as the only one of the type expressed by the N of the definite NP. In the case of associative definites which introduce a new entity, this entity must be easily accessible through NP1. Prototypical whole-part relations are a case in point where such accessibility exists between the entities denoted by NP1 and NP2. The fact that a definite associative NP2 must be supported by an accessibility relation explains why the use of a definite associative NP2 come across as strange and awkward. The last part of this paper is devoted to a discussion needed in proposing two new constraints of my own, which are bearing on associative definite NPs. It is argued that nouns referring to abstract properties and nouns denoting body parts of animate entities should be constrained as candidates for associative definite NPs. (Hoseo University)