The paper discusses several aspects of the use of ontological knowledge and some concomitant Semantic Web technologies in the development of software tools for operative access to digitized cultural heritage and archival collections. The emphasis falls on a number of general issues like semantic mark-up of content, information integration, interoperability of ontologies etc. Some domain-specific problems e.g. the scope of the ontologies that are needed for the purpose (and which ones should the heritage sector develop and which ones will be possible to borrow from other sectors) have been also analyzed. Two successful projects directed to the implementation of ontology-driven access to various types of cultural heritage repositories have been analyzed as examples of best practices in the area. The experience in building software tools for knowledge-based search in repositories of digitized manuscripts and archival materials gained at the Institute of Mathematics and Informatics of the Bulgarian Academy of Sciences and the Faculty of Mathematics and Informatics of Sofia University has been discussed as well.