Soils are complex and heterogeneous environments, which result in a variety of microhabitats, in terms of pore size, water regime, availability of substrates, nature of pore wall. These habitats may be colonized by different microorganisms and the local conditions at the microscale may be different from the average conditions estimated on bulk soil. As a result it is still difficult to establish accurate predictive relationships between bulk soil characteristics and microbial functions. Describing the spatial distribution of microorganisms at the microhabitat scale and relating it to their activity may be a necessary step for understanding and quantifying the factors that control microbial functions. Although much work has been devoted to developing techniques to analyze the diversity of soil microorganisms, comparatively little effort has been spent to connect these measures with habitat and function. The objective of this presentation is to review current knowledge on the spatial distribution of microorganisms at different scales within the soil matrix. The spatial distribution of microorganism of microorganisms has been studied either after the fractionation of soil structure into different sub-units, i.e. aggregates of different sizes and stabilities or by quantifying microorganisms and their activities in spatially referenced samples at different scales. In all cases, the spatial distribution of microorganisms is extremely heterogeneous, in terms of total microbial numbers as well as for specific groups or their activities. Using geostatistics, the distribution of microorganisms was found to be spatially structured rather than random at the microscale (< 1 cm). Microbial hot spots have been related to organic matter patches, such as incorporated plant residues. The spatial distribution of microorganisms at the microscale has tremendous functional importance as it controls the accessibility to organic matter for heterotrophs and biochemical processes such as aerobic mineralization, denitrification and methanogenesis. It also regulates the exposure of microorganisms to pollutants. However, only a few studies have attempted to characterize microscale microbial hot-spots and evaluate their contributions to overall function at the scale of soil peds. Very little is known on the dynamics of the observed spatial distributions, e.g. on the duration of microbial hotspots. Developments in this area are crucial to integrating the biology and physics of the soil ecosystems. These will depend on the development of methods that can couple the location of microorganisms in undisturbed soil samples with their identification and or activity, NanoSIMS being one such method. Sophisticated modeling tools are required to integrate the knowledge gained on individual habitats. These may be stochastic which takies into account the observed spatial variability of microorganisms and activities. Or it can be deterministic that is based on a conceptual representation of soil a mosaic of microhabitats and on the knowledge of their characteristics in terms of microbial populations, activities and driving factors.