The most obvious characteristic of the terrestrial interface is its heterogeneity. Phenomenologically, this one, human or natural, show limits that define the forms deployed in geographic space. These limits are certainly due to temporal dynamic, all as much as a scale dynamic. This one is manifested in the possible relationships and variables that exist between scales inasmuch as every scale can no conceive than relative to another that is used as reference. This leads to the scale relativity (SR) which should allow to define intrinsically the geographical space. The first objective is to show the possibility of using SR in geography. At the heart of the SR, we found that fractal geometry is indispensable to try to understand the organization in the scales of the world. So far fractals no were used than as a tool of describing more or less relevant. In SR, the fractal forms become a consequence of a formal space intrinsically irregular. The fractality can be thus a way of understanding of the world using the space of scales, that is to say its resolutions. The central objective of this study is thus to build a general fractal methodology necessary under investigation of an unspecified morphology through various examples resulting from the physical geography, the urban geography, the geohistory and the geography of the settlement. The final objective is to lead to solutions formal accessible at a broad community from geographers, which is not the case of the theory of the R.E in its current formalism. From an epistemological point of view, the development in geography of the R.E raises the question of the Re-naturalization of this discipline of the human and social sciences and of the constitution in analytical science, therefore more largely to propose a new definition of the geography
