This article describes, after brief references to other methodological traditions, the dialectometric methods used in Salzburg for the global processing of mass data stored in linguistic atlases. The main goal is the discovery of deep structures hidden in these data and the quantitative patterns underlying them. It explains the main steps of the quantitative analysis and the subsequent visualization of the computed data. The following seven map types are properly introduced and discussed: working maps, similarity maps, parameter maps, isogloss syntheses, beam maps, dendrograms, and correlation maps. All the discussed examples are drawn from the French linguistic atlas Atlas linguistique de la France (ALF).
