The North Sea is considered a unique heritage site that yielded a huge amount of zoological and archaeological data. More than 200 palaeozoological and archaeological fossil bone samples from the North Sea bed are dated by 14C. About 2/3 of these dated bones are Pleistocene in age; the majority of the bones are from extinct species (in particular woolly mammoth); about 1/3 of the sample date to the Holocene. The presented dataset is important in its kind, but interpretation is limited because of a lack of context of the finds. The stable isotopes (13C, 15N) of the dated samples provide additional information on palaeoenvironmental conditions and dietary habits in the past. We present primarily a Groningen list of data; a few fossils dated in other laboratories are included for completeness.

Expensive and time-consuming preparation procedures for radiocarbon and stable isotope analyses can be conducted on archaeological bone samples even if no collagen is preserved. Such unsuccessful preparation can lead to the partial destruction of valuable archaeological material. Establishing a rapid prescreening method for evaluating the amount of bone collagen while minimizing the impact of sampling constitutes a challenge for the preservation of archaeological collections. This study proposes and discusses a new methodology to detect and quantify collagen content in archaeological bone samples by attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy. A total of 42 Pleistocene to modern bone samples were selected according to their nitrogen content measured using an elemental analyzer. Comparison of collagen content estimation using ATR-FTIR and mass spectrometry reveals that some of the studied samples are contaminated by a nitrogen source coming from the burial environment. Two different FTIR calibration approaches were tested on the uncontaminated samples: peak-to-peak ratio and multivariate regression (PLS). The two approaches yield similar results with a good correlation of ATR-FTIR analyses and N wt% from 0.7 to 4wt% (R²=0.97–0.99; standard error of estimation ±0.22 to 0.25wt%). While collagen content remains difficult to detect in poorly preserved bones (less than ~3wt%), ATR-FTIR analysis can be a fast alternative for sample screening to optimize the sampling strategy and avoid partial destruction of valuable samples that do not contain enough collagen for further analysis.