The present article1 discusses difficulties in the practical application of spatio-temporal databases to geographical-historical research. Apart from undeniable assets of introducing database tools to historical studies, research practice implies also quite significant difficulties related to the unreliability, incompleteness or imprecision of historical information. These features, along with the subjectivism of historical inferential methods (their susceptibility to interpretation), should be taken into consideration when creating Historical Geographical Information Systems (HGIS). Thus assembled, historical information becomes easily accessible for secondary interpretation (source scheme). The critical scheme of any database, created jointly by historians, engineering ontologists, and data modelling specialists, should account for both the ‘expediency’and ‘processuality’of historical phenomena, as well as the complex nature of spatio-temporal objects (the ongoing dispute between endurantists and perdurantists).