Plant nutrient acquisition largely depends on the availability and constant replenishment of nutrients at root surfaces. The main process providing nutrients for root uptake is diffusion which is critically affected by soil physicochemical and biological factors. The effective diffusion path length increases with decreasing soil water content thus limiting diffusive nutrient supply in dry soils. Microdialysis, an increasingly popular sampling tool for monitoring nutrient fluxes in soil in situ, could be used to measure such a limitation provided it operates reliably under dry soil conditions. However, due to its mode of operation, the performance of microdialysis could be impeded in dry soils since a significant amount of perfusate might be lost via leaching to the surrounding soil. Until now, however, no data is available on the applicability of microdialysis in dry soils. In this project, we estimated leaching losses of perfusate to eight soils with different texture maintained at 10, 20, 30, 50 and 80% (w/w) of maximum water-holding capacity in dependence of perfusion flow rate. We found that microdialysis operates reliably even in dry soils when using a high perfusion flow rate (10 μl min−1). At lower perfusion flow rates (1 or 5 μl min−1), however, significant amounts of perfusate were leaching out to soils. Diffusive fluxes of nitrate and ammonium were strongly affected by soil water content, but not fluxes of phosphate. Even at the highest flow rate (where no losses of perfusate where observed) concentrations of both inorganic N forms in dialysates decreased with decreasing soil water content. Overall, we conclude that microdialysis can be applied in dry soils without high risk of perfusate leaching when a high perfusion flow rate (10 μl min−1) is used. Further, our results highlight that microdialysis can be used in much dryer soils than previously suggested (70% MWHC) and allows monitoring the effect of soil water content on nutrient supply for root uptake via diffusion.