Structure is an important characteristic of documents. In the document management community it has been realized that there is a need for querying and retrieval of documents based on the structure of the documents. In this paper we propose a knowledge-based system for representation and retrieval of documents. The model on which the system is based is an extension of the description logic model of information retrieval. We discuss the advantages of our model and we show how our model can cope with many of the desirable queries involving the structure of documents.