The city comprises a complex arrangement of physical and activities space.  Diverse spatial and social interactions associated with urban socio-economic development processes are reflected in spatial and temporal patterns (of any specialized density) at various scales.  Density is an effective descriptor of such processes and a common language for communication between various disciplines.  In Geographical Information Systems (GIS), density is frequently used in the representation, measurement and modelling of urban activities and phenomena.  In this paper, we first present an overview of these different approaches to defining density in GIS. Recent advances in spatial data capture have contributed to significant improvements in quantifying urban geographical interests/issues (e.g. population density, development density, plot ratio, and building density) from regional to neighbourhood scales.  Density can be represented in either discrete or continuous forms which can be estimated using GIS methods.  Second, we propose two new types of density: perceived density and virtual density which are derived from GIS analysis.  Perceived density describes and visualises how people normally measure density in their own practice, e.g. urban land use and transportation planners.  Virtual density is a construct that is often less tangible, but despite this it can play an important role in (spatial) decision making.  One such example is job opportunity; here measured by integrating population, employment and transport network data in a GIS environment.  We argue that these two densities require further attention in urban planning practice.  Third, we discuss other issues associated with the measurement and analysis of density.  Key themes include the positive and negative effects of mapping density; multi-scale issues of representation and uncertainty; cultural differences which may influence the reasonable use and interpretation of density; and, participatory mapping techniques which may offer improved interpretation of density for communication between different users/disciplines.