Background: Wood density is considered to be the most important predictor of wood quality but despite its importance, diffuse-porous tree species have been the subject of only a limited number of studies. The importance of European beech forests for Central Europe calls for profound research to examine the potential impact of a warmer climate on the quality of beech timber.Methods: In this study we analysed the influence of tree-ring width and tree-ring age on the wood density of beech,and whether the wood density response to these two parameters is modified by aspect. A linear mixed-effects model for wood density was constructed for mean density data measured with high frequency densitometry on stem discs from 72 beech trees sampled from two different aspects(northeast-NE and southwest-SW) of a valley in southwestern Germany.Results: Part of the variability of mean annual wood density was explained by cambial age: an increase in cambial age resulted in an increase in mean wood density. Tree-ring width and aspect had only a small influence on wood density.Wood density on the SW aspect was lower than on the NE with a difference of approximately 0.006 g/cm3. The between-tree variability was very high.Conclusions: The significant interaction between cambial age and aspect reflects the importance of site conditions at older tree ages: with increasing cambial age the difference between aspects becomes stronger. Our results give a better understanding of the importance of site conditions on the wood quality of beech. Background: Wood density is considered to be the most important predictor of wood quality but despite its importance, diffuse-porous tree species have been the subject of only a limited number of studies. The importance of European beech forests for Central Europe calls for profound research to examine the potential impact of a warmer climate on the quality of beech timber. Methods: In this study we analyze the influence of tree-ring width and tree-ring age on the wood density of beech, and whether the wood density response to these Two parameters are modified by aspect. A linear mixed-effects model for wood density was constructed for mean density data measured with high frequency densitometry on stem discs from 72 beech trees sampled from two different aspects (northeast-NE and southwest-SW) of a valley in southwestern Germany. Results: Part of the variability of mean annual wood density was explained by cambial age: an increase in cambial age resulted in an increase in mean wood density . Tree-ring width and aspect had only a small influence on wood density. Wood density on the SW aspect was lower than on the NE with a difference of approximately 0.006 g / cm3. The between-tree variability was very high. Conclusions: The significant interaction between cambial age and aspect reflects the importance of site conditions at older tree ages: with increasing cambial age the difference between aspects is stronger. Our results give a better understanding of the importance of site conditions on the wood quality of beech.