A non-targeted, ultra-high-resolution mass spectrometric, direct analysis of oak-wood extracts from two species (Quercus robur L. and Quercus petraea Liebl.) from three French forests, and of a wine aged in barrels derived therefrom has been performed to identify families of metabolites that could discriminate both the species and the geographical origin of woods. From 12 T ultra-high-resolution Fourier transform ion cyclotron resonance mass spectra of wood extracts, hundreds of mass signals were identified as possible significant biomarkers of the two species, with phenolic and carbohydrate moieties leading the differentiation between Q. robur and Q. petraea, respectively, as corroborated by both FTMS and NMR data. For the first time, it is shown that oak woods can also be discriminated on the basis of hundreds of forest-related compounds, and particular emphasis is put on sessile oaks from the Tronçais forest, for which sugars are significantly discriminant. Despite the higher complexity and diversity of wine metabolites, forest-related compounds can also be detected in wines aged in related barrels. It is only by using these non-targeted analyses that such innovative results, which reveal specific chemodiversities of natural materials, can be obtained.