The paper presents lexicographic work carried out for Danish in the EU project ELEXIS. In ELEXIS the focus is on the development of automatic methods to link senses across two dictionaries based on initially manually created datasets. In the paper, I describe the task of creating the Danish dataset and discuss how the results of the manual alignment can also be used in Danish lexicographic projects. The alignment is carried out for more than 500 identical lemmas in the historic ODS (covering Danish from 1700-1950) and the modern dictionary DDO (the Danish Dictionary), and I report the different types and degrees of sense match between them. I also take a closer look into the cases where there is no sense match from ODS to DDO, and argue that this information can be useful in lexicographic projects where historic dictionaries for Danish before 1700 are to be linked to more modern ones. The opposite cases where the DDO senses have no match in ODS, give on the other hand insights into the development of new senses in Danish and constitute relevant data in modern Danish lexicography. Finally, I also describe how the created dataset is used as a gold standard in experiments with automatic alignment of the senses carried out by the ELEXIS partners.