A practical approach to the identification of low temperature heated bone using TEM

Abstract

Unlike burnt bone, cooked bone leaves no obvious trace. The ability to identify cooked bone has far reaching implications within the fields of palaeo-anthropology, archaeology and forensic science. There is, however, only limited literature on low temperature heated bone. This is not due to a lack of significance, but from an absence of any means of identifying such heating at temperatures insufficient to cause charring. The situation is further complicated by the fact that diagenetic alteration to bone may mimic heat induced changes.

A previous Transmission Electron Microscopy (TEM) study showed that heat induced morphological changes to the collagen fibrils occur after low temperature heating of fish bone. This paper investigates whether these findings could be replicated on mammal bone. A series of experiments were carried out using sheep humeri. These results were compared with cooked and uncooked bones recovered from experimental burials, representing a variety of different environments (moorland, woodland and garden soil). Morphological changes to the fibrils were seen following only very mild heating events, such as short-term roasting of fleshed bone. However, similar changes were observed in unheated bone which had been buried in a low pH (3.5–4.5) soil for 7 years. Within a given burial cooked and uncooked bone was easily distinguishable. The technique, therefore, has direct application in forensic studies and may be of value in distinguishing heated from unheated bone within a given archaeological assemblage.

Introduction

The social and practical implications of the technological breakthrough heralded by the first use of fire cannot be underestimated [17]. Evidence of heated bone in later pre-history can be used not only to imply human influence, but also to elucidate dietary habits, cooking techniques [24] and socio-cultural or funerary practices [16]. Unfortunately, it is only possible to identify the presence of heated bone if charred, from abrasive features (pot polish) or by inference from the presence of associated features, such as hearths [8]. This means that there is a wealth of evidence not being utilised from boiled and roasted bone, which though heated, does not reach a temperature that will induce charring and therefore remains undetected. If minimal heating of bone could be identified perceptions of the past might change. As an example, establishing cannibalism in the archaeological record is a topic which remains of great interest and debate especially with regard to sites in the American Southwest [15], [24] and Fiji [4]. Recognising cooked human bone could not only help to distinguish war related mutilation or secondary burial from cannibalism [7] but the reported incidence of cannibalism would arguably increase [6].

The importance of identifying heated bone from archaeological sites has led to a number of methods of identification. There are three main stages through which a bone passes as it is heated; first water is boiled away, followed by combustion of the organic matter and finally modification and decomposition of the mineral. Methods have identified changes in the appearance of the bone as well as detecting thermal alteration within the organic and mineral phases. At a macroscopic level, studies have focused on the observation of thermally induced colour changes and patterns of warping and cracking on the bone surface [11], [20]. Within the bone microstructure, the chemical modification and loss to the organic fraction of bone has been investigated through analytical techniques monitoring carbon, hydrogen, and nitrogen (CHN) concentrations [13], [22], as well as glycine/glutamic acid (gly/glu) ratios and ammonia (NH₃) levels [23]. Changes to the mineral have also been identified using x-ray diffraction [20] and infrared spectroscopy [22].

The situation however is complicated by the fact that weathering processes cause similar alteration to bone, such as cracking, cortical exfoliation and colour change to the bone surface [1], an increase in crystal size and organisation within the bone mineral [22], [25], a loss of collagen and modification to the remaining amino acid composition [9], [26]. Indeed a recent study failed to discriminate between boiled and buried bone [19].

Furthermore, attempting to detect evidence of cooking provides an additional problem in that when heating food the objective is to retain the moisture. This being the case it is unlikely that the bone would reach even the initial stage of thermal alteration where water is boiled out of the bone especially if the surrounding flesh acts as an insulator from the external temperature. Consequently, the aforementioned changes used to identify heated bone will only occur at temperatures greatly beyond those achieved during domestic roasting or boiling.

In this study we explore an alternative approach, namely monitoring heat-induced changes to the packing of the collagen fibril. Evidence from previous studies of mineralised sheep tendon [21] and fish bone [18] suggest that mild heating leads to disorganisation of mineralised collagen fibrils, observable using TEM. It was unclear if mammal bone collagen would react in a similar way, nor if the alterations could be unequivocally attributed to heating, or could also be caused by diagenesis (cf. [19]).

This paper comprises an investigation to monitor changes to collagen fibrils in sheep bones subjected to low temperature heating, burial for 7 years, or both. Primarily TEM analysis was used to identify alterations to the collagen component, supplemented with infrared investigation to monitor any corresponding changes to the mineral component.