J Archaeol Sci 22 (2)

This paper develops a theory for describing those diagenetic changes in bone which involve its interaction with groundwater. Three main processes are considered; namely the uptake of uranium, the dissolution of bone, and the increase of crystallinity of the bone mineral (carbonate hydroxyapatite or dahllite). Here simple models of the chemistry involved are postulated (although how bone interacts with water on a molecular scale is not clearly known) in order to demonstrate the theory with explicit mechanisms and values. Greater emphasis is given to uranium uptake, since the model used is comparatively detailed, being based on the authors' previous work.

Reports a technique to simulate collagen diagenesis using a conceptual model by which the degradation is controlled by two key processes: depolymerisation of collagen by chemical hydrolysis of peptide bonds; dissolution of polypeptide fragments retained by fewer than a critical number of hydrogen bonds, biodeterioration being ignored. The model predicts that the curve describing weight loss is sigmoidal in accord with the sparse experimental data for non-mineralised collagen. The conclusion for studies of mineralised collagen is that the equivalent curve is exponential. It is thought that the experimental design of these studies was flawed. Other predictions of the model are detailed and the validity of applying it to archaeological bone discussed.

Four diagenetic parameters have been chosen to represent the state of diagenesis of bone buried on archaeological sites. They are: histological preservation, protein content, crystallinity, and porosity. How these parameters are measured is described and results from populations of bones from three different sites are presented. The results show the extent and variation in the degree of change, both within a site and between sites. In particular the correlations between diagenetic parameters are examined, which give clues about the processes which cause alteration. the value of porosity determinations (both at the intercrystalline level, and at coarser levels) in revealing the degree of diagenetic change in bone, and in underlying the dynamics of the interaction between buried bone and the surrounding water is stressed. The data also indicate, but do not prove, the following: microbiological attack is generally complete within less than 500 years; substantial levels of protein may remain in bone after maximal micromorphological alteration; loss of protein appears to be independent of other diagenetic change; the correlated loss of microporosity with increase in crystallinity suggests these changes may arise from the dissolution, perhaps with subsequent recrystallisation, of the smallest hydroxyapatite crystallites.

This paper presents research on the conditions under which progressive levels of burning may occur to archaeological bone, and how burning damage changes bones' crystal structure and increases friability. Experiments were conducted to simulate common patterns of high-temperature bone diagenesis and fragmentation previously documented in Palaeolithic shelter sites. Bones buried up to 6cm below the coal beds of the experimental fires were carbonised, but calcination occurred only with direct exposure to live coals. Analysis by infra-red spectroscopy reveals that marked changes in crystallinity accompany the macroscopic transformations in colour and friability of modern, fire-altered bone. These findings help qualify the behavioural and taphonomic implications of fragmented, burned bones in archaeological sites, especially with regard to potential stratigraphic associations between artefacts and hearth features in sites and the intensity of space use by human occupants. The identification of burning damage on archaeological bone is a separate issue. It was found that the molecular signatures of recrystallisation in modern burned bones partly overlap with recrystallisation caused by weathering after only one to two years' exposure in an arid setting and by partial fossilisation of archaeological bones over the long term. While infra-red and x-ray diffraction techniques effectively describe heat-induced changes in modern bone mineral, the same was not true for archaeological material. Cross-referencing readily visible colour phases with HCI-insoluble fraction data proves much more effective and economically feasible for the latter purpose.

Reports tests to see whether tooth enamel might be altered during burial by focusing on the carbonate content and electron spin resonance (ESR) signals used for dating. Manometry and Fourier transform infra-red spectroscopy were used to study carbonate in eleven fossil and five modern tooth enamels from horse, cow, hippopotamus, rhinoceros, and mammoth. It is shown that ungulate tooth enamel exhibits a narrow range of both carbonate content and ESR signal intensity. These characteristics are true both for modern and most fossil samples ranging in age up to several hundred years old. Evidence for possible diagenesis was found only for two samples, consisting of fragments of enamel which exhibited anomalous high and low carbonate contents (and ESR signal sizes).

Undegraded, recent Type I collagen was reacted with two different batches of humic acid, monitoring the speed of uptake, and measuring the final maximum uptake. The speed turned out to be quite high (hours rather than days), as was the average maximum uptake. The resulting complex was analysed using *13C and 14C measurements, and Fourier-transform IR spectroscopy, and then subjected to a number of different chemical treatments including enzymatic digestion. It was found that most chemical procedures only partially remove the humics, which is a significant finding in the light of effectiveness of radiocarbon/stable isotope chemistries. In the case of enzymatic digestion, collagenase action was not hindered by the 0.8kDa humic linked to the collagen, but was by the larger humic.

Lipid extracts were prepared for a number of samples of modern and ancient bones of humans and other animals (pig, cow, horse, and whale). Aliquots of the total lipid extracts were trimethylsilylated and analysed directly by gas chromatography (GC) and gas chromatography/mass spectrometry (GC/MS) without prior fractionation or purification. The analysis of archaeological bones and samples of adhering soil showed that contamination of the bone by migration of lipids from the burial environment was not a serious problem. Analyses of modern reference materials showed the lipid extracts of bone possessing adhering marrow fat to be dominated by acyl lipids. In contrast, the lipid extract of a sample of the modern cow bone with the marrow fat removed was dominated by cholesterol; acyl lipids were present at low abundance. The GC and GC/MS analyses of the ancient bones of varying age and contexts of recovery also showed only a very low abundance of acyl lipids. Most significant was the finding of an appreciable concentration of cholesterol together with its diagenetic degradation products, in all the samples that were examined. The possibility of using the cholesterol (and its degradation products) as a palaeodietary indicator is discussed. Evidence is also presented for the occurrence of bacterial marker compounds in the bone total lipid extracts.

Infra-red spectrometry, XRD spectrometry, and elemental analyses were used to compare the effects of the Nelson et al protocol with the solubility profile protocol on the crystal structure and crystallinity of the bone and fossil specimens. The results show that the procedures used by Nelson et al which include a preliminary washing step, result in severe recrystallisation of apatite both before and during the leaching step, which deleteriously affects any attempt to recover biological strontium. By contrast, no observable recrystallisation occurs in the solubility profile procedure. The results imply that the conclusions reached by Nelson et al are primarily artefacts of the specific methods used in their study, rather than due to any inherent mechanism of diagenesis.

The preservation of trophic structure and climatic information in Kent's Cavern Upper Pleistocene mammal bones and teeth was assessed by comparing the isotopic composition of modern and fossil equivalents. Yields of collagen from both bone and tooth were extremely variable, with values relative to modern bone ranging from 0% to 100%. No evidence of preferential preservation of tooth collagen was detected. Although the specimens have been subjected to extensive diagenetic alteration, the biological isotopic signals seem to have utility for palaeoecological reconstructions.