Nitrogen content variation in archaeological bone and its implications for stable isotope analysis and radiocarbon dating

The collagen component of ancient bones is routinely isolated for radiocarbon dating and stable isotope studies. However, it is impossible to tell the state of collagen preservation from visual inspection of bones. At the Oxford Radiocarbon Accelerator Unit (ORAU), the percent nitrogen by weight (%N) of a ∼5 mg sample of bone powder is measured on a mass spectrometer and used as a proxy for protein content. A previous study showed that samples with %N > 0.76 are considered likely to produce sufficient collagen for radiocarbon dating (Brock et al., 2010b). However, the extent of variation between bone %N and collagen yield is unclear, as is the intra-bone variation in %N. Here, we report a series of tests performed on Palaeolithic bones known to have variable collagen preservation. This new study shows significant variation in %N within the same bone and that there is sometimes a lack of correlation between %N and collagen yield. These results suggest that for bone samples from difficult environments or from Pleistocene contexts, it may be worth sub-sampling for %N in different locations of the bone (if possible) and then attempting to extract collagen from marginally preserved bones (%N around 0.2–0.7%), as they may still yield sufficient collagen for isotope and dating studies. [Display omitted] •229 bones from an archaeological site were analysed to evaluate variability in nitrogen content (%N) and collagen yields.•28% of bones produced collagen yields incorrectly predicted by their %N values.•Sub-sampling of bones revealed significant variation in %N within individual bones.