Vegetated Buffer Zones (BZ) are obligatory for farms joining the Agri-Environmental subsidy programme in Finland. BZs along watercourses serve as countermeasures against eutrophication because they reduce erosion and retain nutrients. However, preliminary results show that in old BZs the infiltration of water into the soil is substantially decreased. Macropores are filled with dead and living roots and hydrophobic organic matter delays wetting of the soil surface, both phenomena promoting increased surface runoff. In order to get a deeper comprehension of the hydrological processes taking place in BZs upon ageing of the vegetation, we investigated effects of perennial grass of different ages (15 vs. 3 years) on structural properties and water infiltration (I) in a clayey and in a sandy soil. We compared these sites to adjacent soils in cereal cultivation. Infiltration rate was measured by using cylinders (diameter 30 cm, height 20 cm) to 10 cm soil depth. To study soil porosity and water repellence, soil cores were sampled from 0-10 cm soil depths. Water repellence index was determined by comparing sorptivity of water and ethanol with miniaturized infiltrometer. The results help understand the phenomena occurring in BZs, evaluate their efficiency in water protection and give practical advice in the management of BZs. The ploughed layer of the clay soil (Typic Cryaquepts/Vertic Cambisols at Jokioinen, South-Western Finland) had micaceous clay contents of 51-56% and organic carbon (C) content of 2.1 % (6-16 cm soil depth). The sand soil represented Oxyaquic Dystrocryepts/Dystric Regosols at Maaninka in East-Central Finland, with a pH of 6.6, 8% of clay and C content of 1.4%. Both sites had slopes of 10-20%. The old BZs were vegetated with wild hay, and young ones with grass species harvested annually. During 15 years under grassy vegetation, a granular structure dominated in the top 0-6 cm of the clay soil, while, as a response to periodic shrinking and swelling of the clay, a medium-sized prismatic structure with a very firm consistency had developed in the lower part (6-16 cm) of the previous ploughed layer. This contrasts with the subangular blocky structure of the annually tilled cropland. The changes in soil structure were less pronounced in the fine sand, where the ploughed layer of the BZ consisted predominantly of weak medium-sized platy or angular blocky aggregates. The structure was slightly weaker in the fine sandy cropland. Data from May-June 2005 indicated that spring-sown fields with conventional tillage had very low infiltration (5-7 cm h^-1 after 15 min ponding) as compared with vegetated sites: Under aged vegetation infiltration reached 60 and 30 cm h^-1 for clay and sand, respectively. Young grasslands of clay and sand had medium infiltration, 50 and 10 cm h^-1, respectively. In vegetated clay soils, the high infiltration was due to preferential flows through cracks after dry early season, as infiltration under saturated conditions in late November was below 2 cm h^-1. Moreover, the preferential flows were promoted by soil water repellence. In clay and sandy soils top soil water repellence index (at 40°C) was highest in BZ with aged vegetation, 10.3 (standard error of the mean, SE 1.5) and 3.8 (SE 0.5), respectively, and lowest in cereal cultivation, 3.1 (SE 0.3) and 2.0 (SE 0.2), respectively. In both soil types, water repellence index in young BZ was only slightly higher, 3.6 (SE 0.8) for clay and 2.5 (SE 0.3) for sand, than in cereal cultivation. However, degree of water saturation has a big importance on water movement and functioning of buffer zones in clay soils. Because of swelling under saturated conditions, buffer zones conduct water very slowly, regardless of the type of vegetation. The swelling/shrinkage properties of the clay soil will be studied during the year 2006.