Identification of ancient charcoal fragments is a valuable tool in reconstructing past environments and determining natural and anthropogenic disturbances, and for understanding past cultures and societies. Although in Europe such studies are fairly straightforward, utilising charcoal records from the tropics is more complicated due to the species-richness of the natural vegetation. Comprehensive databases have greatly aided identification but often identification of charcoalified woods from the tropics relies on minute anatomical features that can be difficult to observe due to preservation or lack of abundance.
This article illustrates the relative potential of four imaging techniques and discusses how they can provide optimal visualisation of charcoal anatomy, such that specific difficulties encountered during charcoal examination can be evaluated and fine anatomical characters can be observed enabling high-level identification of charcoal (and wood) taxa. Specifically reflected Light Microscopy is often used to quickly group large numbers of charcoal fragments into charcoal types. Scanning Electron Microscopy and High-Throughput X-ray Computed Tomography are employed to observe fine anatomical detail. More recently X-ray Computed Tomography at very high resolution has proved successful for imaging hidden or ‘veiled’ anatomical features that cannot be detected on exposed surfaces but need three-dimensional volumetric imaging.
This study aimed to define the variability in the microstructure of Norway spruce within an annual ring by examining differences between earlywood and latewood. In particular, we were interested in obtaining new information on bordered pit occurrence and locations relative to tracheid ends, plus the lumina dimensions and longitudinal overlap of tracheids that collectively define the longitudinal hydraulic pathways. A stacked series of X-ray micro-CT scans of an annual ring of Norway spruce were made and stitched together longitudinally to form a three-dimensional volume. The imaging resolution was carefully chosen to capture both longitudinal and transverse anatomical details. Measurements of tracheid length, overlap, radial lumen diameter, and bordered pit location were made semi-automatically using image analysis. The distribution of radial lumen diameter was used to define earlywood and latewood. Then bordered pit linear density and spatial distribution, tracheid length and overlap were analysed, presented and contrasted for earlywood and latewood. Further differences between earlywood and latewood were found only in bordered pit linear density. Clear trends in radial lumen diameter and pit linear density were observed with radial position within the growth ring. These results provide new information on the variability of the Norway spruce microstructure within an annual ring.