Xylaria, or wood collections, can be considered sleeping beauties in terms of wood technological and biological output. In this study we focus on determining the natural durability of xylarium specimens from the Federal Xylarium in the Royal Museum for Central Africa. The Federal Xylarium contains over 80,000 specimens, covering more than 13,000 species, mainly from Central Africa. These specimens have a random format (disks, book-shapes…) and should be treated with care when sampling for testing, as these specimens are not only limited in size, but also have a significant historical and cultural value. This makes it impossible in most cases to retrieve standard specimen dimensions as required according to CEN/TS 15083-1 (2005). To gain information on the natural durability of these xylarium specimens, two tests were performed: a mini-block test similar to the test described by Bravery (1978) and an indirect assessment using chemical fingerprints. In total 1112 mini-block samples were prepared from 577 xylarium specimens, covering 33 tropical species and complemented with 545 reference specimen, covering 11 species for benchmarking. The natural durability of the wood samples was assessed using the white-rot fungus Trametes versicolor. Beech wood was included as a reference to check whether the fungus was sufficiently virulent. The resulting natural durability classes per species of the mini-block test are similar to those found in common literature. Erythropheum suaveolens and Millettia laurentii appear highly durable, while Pycnanthus angolensis and Terminalia superba performed worst. Chemical fingerprints for the xylarium test specimen were obtained using Direct Analysis in Real Time – Time of Flight Mass Spectrometry (DART TOFMS). Several ions/molecules appear to be correlated with natural durability, showing the potential towards quick screening of the natural durability potential of a certain specimen/species

Download document (647 kb)
free for the members of IRG. Available if purchased.

Purchase this document