Soil water content and actual water repellency were assessed for soil profiles at two sites in a bare and grasscovered plot of a sand pasture, to investigate the impact of the grass removal on both properties. The soil of the plots was sampled six times in vertical transects to a depth of 33 cm between 23 May and 7 October 2002. On each sampling date the soil water contents were measured and the persistence of actual water repellency was determined of field-moist samples. Considerably higher soil water contents were found in the bare versus the grass-covered plots. These alterations are caused by differences between evaporation and transpiration rates across the plots. Noteworthy are the often excessive differences in soil water content at depths of 10 to 30 cm between the bare and grass-covered plots. These differences are a consequence of water uptake by the roots in the grass-covered plots. The water storage in the upper 19 cm of the bare soil was at least two times greater than in the grass-covered soil during dry periods. A major part of the soil profile in the grass-covered plots exhibited extreme water repellency to a depth of 19 cm on all sampling dates, while the soil profile of the bare plots was completely wettable on eight of the twelve sampling dates. Significant differences in persistence of actual water repellency were found between the grass-covered and bare plots.