Questions Tree diversity is key to the functioning of forest ecosystems. However, which components of tree diversity are responsible for tree diversity effects on associated organisms, and in which context, is poorly understood. Location ORPHEE Experimental site, Cestas-Pierroton in the southwest of France. Methods We used a large-scale tree diversity experiment in which we controlled water availability by an irrigation treatment to address tree diversity and water stress effects on the diversity and height of forest understorey vegetation. We assessed the species richness and height of understorey vegetation in irrigated and non-irrigated 20 m x 20 m plots, either being monocultures ofPinus pinaster,Betula pendulaorQuercus robur, or either mixtures of two to five species amongPinus pinaster,Betula pendula,Quercus robur,Quercus ilexandQuercus pyrenaica. Results Tree species composition, i.e. the proportion of birch or pine, had a significant effect on understorey plant assemblage, species richness and diversity. The proportion of the fast-growing deciduous angiospermBetula pendulawas negatively correlated to understorey plant richness and diversity, and positively correlated with understorey vegetation height. Understorey vegetation was higher in irrigated plots than in non-irrigated plots, but irrigation had no clear effect on the species richness, diversity or composition of understorey plant assemblages. Conclusions Forest tree species composition and in particular the relative proportions of different tree species had stronger effects on understorey plants than tree species richness per se. These effects were consistent across irrigation treatments. Even in young forest plantations, effects of tree mixture on understorey vegetation may be observed and seem mainly driven by the functional type of tree in the canopy and initial dynamics of plant regeneration in planted forests.