Spatial information describes types, relations, and various different aspects of space. This PhD thesis investigates how modular ontologies can model spatial information. Particularly, different perspectives on space are analyzed. A perspectival framework for spatial ontology modules is presented that allows the integration and combination of different facets of spatial information. This work discusses perspectives on space by distinguishing and categorizing quantitative, qualitative, abstract, domain-specific, and modal types of spatial information. Application examples are presented for spatial natural language interpretation, image recognition, and architectural design. The results are achieved by theoretical analyses of spatial domains as well as empirical and experimental findings from different disciplines related to the spatial domain. Technically, methods from formal ontology and ontological engineering are applied.