Summer 2003 was one of the warmest and driest periods in Central Europe for almost 100 years. The air temperature between June and August was anomalous, deviating from the 30 year average (1961-1990) by +2 degrees C to +5 degrees C. In northeastern Central Europe (51 degrees N-55-N, 10 degrees E-17 degrees E: north-eastern Germany and western Poland) a deficit in precipitation of 180 mm resulted from the 2003 drought. In this year the precipitation rate amounted to 440 turn p. a. only. European beech (Fagus sylvatica L.) is one of the main tree species in the ongoing conversions of coniferous plantations to deciduous forests, since beech dominates the potential natural vegetation (PNV) in large parts of Central Europe. However, the reputed drought sensitiveness of beech has led to discussions about the growth and regeneration of beech on dryer sites in southern Germany in the context of climate change predictions. In eight pure and mixed European beech forests in northeastern Central Europe the growth response of natural beech regeneration during the 2003 summer drought and the additional effects of shading was investigated. The selected stands, comprising four to six year old natural beech regeneration situated on sites with similar sandy soils, spanned a 600 kin geographic gradient from Northeast Germany to Northeast Poland (Masuria). During the extended drought period from end of July to mid August 2003, the plant water status of beech regeneration was assessed by measuring the predawn potential (psi(PD)) of 17 to 22 randomly selected saplings with a Scholander pressure chamber. All stands were then classified into class (1) without water stress (psi(PD) > -0.4 MPa), class (2) with moderate water stress (psi(PD) -0.4 MPa to -0.8 MPa) or class (3) with high water stress (psi(PD) < -0.8 MPa). Inventories of natural regeneration took place on permanently marked plots 0.12 ha to 0.25 ha in area. On six to eight subplots (20 m(2) area), up to 20 beech saplings were labeled and total aerial shoot length, root collar diameter, and terminal shoot length were recorded. A second inventory in September 2004 enabled relative length and diameter increment in 2003 and 2004 to be calculated in relation to plant size at the beginning of the growth period. Hemispherical photography was used to consider additional effects of below-canopy irradiance on plant performance by deriving the diffuse site factor (DIFFSF) of each subplot. Results showed that plant water status during July and August 200 had a considerable effect on the relative increment of saplings during 2003 and 2004. Increased water stress, indicated by decreased predawn plant potential, correlated to lower relative length and diameter increment. A carry-over effect of the summer drought on beech sapling performance was evident as a decrease in relative growth was observed during the 2004 wet growing season compared to sapling growth during the 2003 drought year. An analysis of covariance (ANCOVA) identified both water stress and diffuse site factor (DIFFSF) as parameters affecting relative and absolute growth increment of the saplings. Water stress had a stronger effect on plant performance than shading. However influence of canopy was more evident when water stress was increased. For the relative length increment in 2004, climatic water shortage and diffuse site factor (DIFFSF) were found to have an interactive effect. The latter may reflect the impact of soil resource depletion in addition to the impact of shelterwood competition on regeneration performance by limiting light availability. In addition, the limited ability of shaded beech regeneration to adapt to water stress can also play a role. This may be due to the lower osmotic potential and lower root/shoot ratios of these plants. The lower adaptation potential of shaded beech exacerbates the impact of competition from the overstorey. The predicted increase in summer drought events in Central Europe suggest that an optimization of soil water resource management in future silvicultural planning will be crucial for successful beech stand regeneration. In this context, a marked reduction in canopy after the successful establishment of young beech plants will reduce the risk of water stress, provided competition from ground vegetation is controlled. Irregular shelterwood systems creating gap openings with an initial area of up to 20 meters in diameter will provide those conditions, particularly in gap centres. Additional opportunities to increase the water stress tolerance of beech stands, such as the introduction of beech provenances from their eastern or southern boundaries where dryer climates prevail, need to be investigated.