The Web is an enormous source of information. The content of Web pages is a visible kind of information, but the Web also contains information of a more hidden kind, like the connections that hyperlinks create between different Web sites. The research field of webometrics tries, among other things, to create new knowledge from this hidden information embedded in the hyperlinks structure on the Web and to understand what kinds of real world phenomena and relationships it may represent. From a broad perspective, the goal of this research was to increase understanding of linking behaviour on the Web and especially of govenment relater Web linking. This research studied how municipalities in the region of Finland Proper are represented on the Web and what kind of hyperlink networks they have created. The present research investigates municioal interlinking, co-inlinking and co-outlinking networks and studies whether hyperlink data can be used to map offline relationships between the municipalities. The research gave unusually clear-cut results for a webometric analysis, both in terms of the discovered geographic factors influencing linking and in terms of classified linking motivations. The results showed that interlinking can be used to map geopolitical connections and cooperation between the municipalities. Co-in-linking to the municipalities was more of a measure of geographic similarity between the municipalities, while co-outlinking showed some potential to be a useful tool to map shared interests. Some of the major contributions of this research are the development of new methods for webometric research and increased knowledge about some existing methods from social network analysis that could be used for webometric research.