This paper aims at showing the usefulness of the fractal dimension for characterizing the spatial structure of the built-up surfaces within the periurban fringe. We first discuss our methodology and expectations in terms of operationality of the fractal dimension theoretically and geometrically. An empirical analysis is then performed on the southern periphery of Brussels (Brabant Wallon). The empirical analysis is divided into two parts: first, the effect of the size and shape of the windows on the fractal measures is empirically evaluated; this leads to a methodological discussion about the importance of the scale of analysis as well as the real sense of fractality. Second, we show empirically how far fractal dimension and density can look alike, but are also totally different. The relationship between density and fractality of built-up areas is discussed empirically and theoretically. Results are interpreted in an urban sprawl context as well as in a polycentric development of the peripheries. These analyses confirm the usefulness but also the limits of the fractal approach in order to describe the built-up morphology. Fractal analysis is a promising tool for describing the morphology of the city and forsimulating its genesis and planning.