Résumé : The present study aimed at exploring the long-term impact of pure and mixed beech Fagus sylvatica and oak Quercus robur stands on the forest floor by documenting changes in the herb species’ behaviour and in humus index across a 200-yr chronosequence of forest stands. The research was conducted in central Belgium, in a 4383 ha beech-dominated forest. Analyses were carried out in stands which are replicated, of the same age, managed in the same way, and growing on the same soil type with the same land-use history. The results of this study indicate that stand aging is an important determinant of herb species occurrence in the studied area. Most of the species studied show a different response to stand age in pure compared to mixed stands. Our results clearly show a decrease of the humus quality with age in pure stands (beech as well as oak). On the other hand, we found that mixing beech and oak maintained or improved the humus status along the chronosequence according to the proportion of each tree. So the addition of some oak to the beech made it possible to keep a constant quality of the humus. We found that, even if the understory tree species is very scarce, it may be sufficient to maintain the humus status on the long term. In the present study, a cover of 1% oak in a beech stand was sufficient to show an effect of the minor species on these soils. This pattern contrasts with the widespread idea that substantial effects of the minor tree species on soils might not develop if the ratio of major/minor species is low.