Advances in Human Palaeopathology by Ron Pinhasi, Simon Mays - Cultura Clássica (2025)

Prévia do material em texto

<p>Advances in Human Palaeopathology</p><p>Advances in Human Palaeopathology Edited by Ron Pinhasi and Simon Mays</p><p>© 2008 John Wiley & Sons, Ltd. ISBN: 978-0-470-03602-0</p><p>Advances in Human</p><p>Palaeopathology</p><p>Edited by</p><p>Ron Pinhasi</p><p>Department of Archaeology, University College Cork, Cork, Ireland</p><p>Simon Mays</p><p>Centre for Archaeology, English Heritage, UK</p><p>Copyright © 2008 John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester,</p><p>West Sussex PO19 8SQ, England</p><p>Telephone �+44� 1243 779777</p><p>Email (for orders and customer service enquiries): cs-books@wiley.co.uk</p><p>Visit our Home Page on www.wiley.com</p><p>All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in</p><p>any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under</p><p>the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright</p><p>Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of</p><p>the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & Sons</p><p>Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to</p><p>permreq@wiley.co.uk, or faxed to (+44) 1243 770620.</p><p>Designations used by companies to distinguish their products are often claimed as trademarks. All brand names</p><p>and product names used in this book are trade names, service marks, trademarks or registered trademarks of their</p><p>respective owners. The Publisher is not associated with any product or vendor mentioned in this book.</p><p>This publication is designed to provide accurate and authoritative information in regard to the subject matter</p><p>covered. It is sold on the understanding that the Publisher is not engaged in rendering professional services.</p><p>If professional advice or other expert assistance is required, the services of a competent professional should be</p><p>sought.</p><p>Other Wiley Editorial Offices</p><p>John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA</p><p>Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA</p><p>Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany</p><p>John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064, Australia</p><p>John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809</p><p>John Wiley & Sons Canada Ltd, 6045 Freemont Blvd, Mississauga, ONT, L5R 4J3, Canada</p><p>Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be</p><p>available in electronic books.</p><p>Library of Congress Cataloging in Publication Data</p><p>Advances in human palaeopathology / edited by Ron Pinhasi, Simon Mays.</p><p>p. ; cm.</p><p>Includes bibliographical references and index.</p><p>ISBN 978-0-470-03602-0 (alk. paper)</p><p>1. Paleopathology. I. Pinhasi, Ron. II. Mays, Simon. III. Title: Advances in human paleopathology.</p><p>[DNLM: 1. Paleopathology. QZ 11.5 A244 2008]</p><p>R134.8.A38 2008</p><p>616.07—dc22</p><p>2007026251</p><p>British Library Cataloguing in Publication Data</p><p>A catalogue record for this book is available from the British Library</p><p>ISBN 978-0-470-03602-0</p><p>Typeset in 10/12pt Times by Integra Software Services Pvt. Ltd, Pondicherry, India</p><p>Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire</p><p>This book is printed on acid-free paper responsibly manufactured from sustainable forestry</p><p>in which at least two trees are planted for each one used for paper production.</p><p>To Hattie and Maralyn</p><p>Simon Mays</p><p>To Rita</p><p>Ron Pinhasi</p><p>Contents</p><p>Preface ix</p><p>Simon Mays and Ron Pinhasi</p><p>Contributors xiii</p><p>Part 1: Analytical Approaches in Palaeopathology</p><p>1. The Chemical and Microbial Degradation of Bones and Teeth 3</p><p>Gordon Turner-Walker</p><p>2. How Representative Are Human Skeletal Assemblages for Population</p><p>Analysis? 31</p><p>Ron Pinhasi and Chryssi Bourbou</p><p>3. Epidemiological Approaches in Palaeopathology 45</p><p>Ron Pinhasi and Katy Turner</p><p>4. Macroscopic Analysis and Data Collection in Palaeopathology 57</p><p>Anne L. Grauer</p><p>5. Radiography and Allied Techniques in the Palaeopathology of Skeletal</p><p>Remains 77</p><p>Simon Mays</p><p>6. Computed Tomography Scanning and Three-Dimensional Visualization</p><p>of Mummies and Bog Bodies 101</p><p>Niels Lynnerup</p><p>7. Histological Studies on Ancient Bone 121</p><p>Gordon Turner-Walker and Simon Mays</p><p>8. Molecular Palaeopathology of Human Infectious Disease 147</p><p>Helen D. Donoghue</p><p>9. Databases 177</p><p>William White</p><p>viii Contents</p><p>Part 2: Diagnosis and Interpretation of Disease in Human Remains</p><p>10. Differential Diagnosis of Skeletal Lesions in Infectious Disease 191</p><p>Donald J. Ortner</p><p>11. Metabolic Bone Disease 215</p><p>Simon Mays</p><p>12. Tumours and Tumour-like Processes 253</p><p>Don Brothwell</p><p>13. Advances in the Palaeopathology of Teeth and Jaws 283</p><p>Alan Ogden</p><p>14. Trauma 309</p><p>Pia Bennike</p><p>15. Congenital Anomalies 329</p><p>Ethne Barnes</p><p>16. Growth in Archaeological Populations 363</p><p>Ron Pinhasi</p><p>Index 381</p><p>Preface</p><p>Traditionally, in palaeopathology the principal emphasis was on descriptions of individual</p><p>cases, principally in order to demonstrate the diagnosis of specific conditions and to help</p><p>establish the antiquity of various diseases. In recent decades, however, although the case study</p><p>still has a place, there has been a greater emphasis on population studies. In part, this reflects</p><p>a move away from a medico-historical orientation to one where addressing archaeological</p><p>questions takes precedence. A dominant theme is now evaluating disease frequencies at a</p><p>population level and integrating this with cultural data pertaining to the populations under</p><p>study from archaeological (or historical) sources in order to address questions of broader</p><p>archaeological interest.</p><p>Several factors underpin this approach in palaeopathology, among which are the following.</p><p>1. An understanding of biases and limitations of the skeletal record caused by differential</p><p>skeletal survival and other factors.</p><p>2. Rigorous quantification of disease or lesion frequency in fragmentary and incomplete</p><p>remains.</p><p>3. The accurate ascription of causation to bony pathologies (be it diagnosis of specific</p><p>disorders or ascription of more general causes to non-specific lesions, such as dental</p><p>enamel hypoplasias).</p><p>As regards the first of these, our understanding of factors affecting skeletal decomposition</p><p>in the burial environment and the mechanisms of diagenesis has been greatly increased</p><p>in the last 15 years by the application of physical, chemical and microscopic analyses to</p><p>ancient bone. As regards the second point, we are increasingly coming to grips with the</p><p>problems of quantifying lesions and diseases in incomplete and fragmentary skeletons, and</p><p>the potential of applying epidemiological methodologies to ancient remains has begun to be</p><p>appreciated. There is also an increasing realization that we may need to go beyond recording</p><p>of simple prevalence rates in order to unlock more fully the information on earlier human</p><p>populations contained in skeletal pathology. More workers are now attempting to record</p><p>differences in severity of lesions objectively, and it is becoming increasingly common for</p><p>workers to record whether lesions were active or healed at time of death. As regards the</p><p>third point,</p><p>product of</p><p>diagenesis in archaeological bones (Hassan and Ortner, 1977; Newesely, 1989; Piepenbrink,</p><p>1989). Brushite has also been identified as a product formed in fresh bone after prolonged</p><p>immersion in an acidic solution (Lee-Thorpe, 1991; Nielsen-Marsh, 1997). However, since</p><p>brushite is more readily soluble than HAP, it seems unlikely that it forms a major component</p><p>of degraded bones in normal archaeological soils except in special circumstances, such as</p><p>skeletons interred in stone vaults, where acid decomposition products may be expected to</p><p>accumulate and soluble species will not be leached away from the bones.</p><p>The pH of the groundwater also determines which other ions are present in solution</p><p>and, therefore, available for ion exchange with the bone mineral. Thus, bones from acidic</p><p>burial environments tend to be brown in colour because transition-metal ions (such as iron</p><p>and manganese) and humic acids are soluble in the local groundwater. Conversely, bones</p><p>from alkaline soils tend to be white or cream coloured because many of the metal ions</p><p>are locked up as insoluble oxyhydroxides or carbonates. Several studies have been made</p><p>on the trace elements in human bone, usually with a view to addressing whether trace</p><p>elements can be used as dietary discriminants (Lambert et al., 1979, 1982, 1983) and to</p><p>what extent diagenesis has altered the original chemical composition (Badone and Farquhar,</p><p>1982; Lambert et al., 1985a,b; Price et al., 1985; Buikstra et al., 1989; El-Kammar et al.,</p><p>1989; Grupe and Piepenbrink, 1989; Pate et al., 1989). A comprehensive review of trace</p><p>element and isotopic analyses is given in Price (1989).</p><p>One of the first changes to be recorded in archaeological and fossil bones was an increase</p><p>in their ‘crystallinity’ compared with that of fresh bone. This increase can be detected using</p><p>XRD or through studies of the infrared spectra of bones and is often expressed as the ‘crys-</p><p>tallinity index’ (Bartsiokas and Middleton, 1992; Hedges et al., 1995) or infrared splitting</p><p>factor (Weiner and Bar Yosef, 1990; Nielsen-Marsh and Hedges, 2000a,b). Increases in crys-</p><p>tallinity have been found to be strongly correlated with other diagenetic parameters, including</p><p>a reduction in the carbonate content of archaeological bone (Hedges et al., 1995; Nielsen-</p><p>Marsh and Hedges, 2000a). There are several probable explanations for these observed</p><p>increases in crystallinity: dissolution and loss of the smallest crystals; dissolution and sub-</p><p>sequent recrystallization to larger and thermodynamically more stable crystals; a reordering</p><p>of the internal crystal structure; and slow growth of existing crystals by apposition. While</p><p>all these mechanisms are possible and may play some role in diagenesis of ancient bones</p><p>and teeth, it is the second, i.e. dissolution and recrystallization, that on present evidence</p><p>appears to dominate. Clearly, this process carries with it the likelihood that exogenous ions</p><p>(e.g. Sr2+, Zn2+, CO2−</p><p>3 , F−, etc.) or various isotopes (e.g. C, O) may be incorporated in the</p><p>reprecipitating crystals, with all that implies for the interpretation of subsequent chemical</p><p>analyses. Of course, the implications of increases in crystallinity and the incorporation of</p><p>14 Advances in Human Palaeopathology</p><p>exogenous mineral species are not always negative. These processes fall under the general</p><p>heading of fossilization, and it is undoubtedly true that the vast majority of bones simply</p><p>would not survive over geological time-scales without a certain level of permineralization</p><p>and the infilling of internal voids with calcite or silicates.</p><p>The mineral phase is not the only component of bone that is susceptible to chemical</p><p>degradation over time. The protein contents of archaeological bones are generally very low</p><p>compared with fresh bone, and in truly fossil bones the levels of organics are reduced</p><p>to chemical traces of amino acids and osteocalcin. The survival of proteinaceous mate-</p><p>rial in the archaeological record is generally restricted to very special circumstances and</p><p>environments. Mummified remains of humans are, of course, found, but most often in</p><p>environments with very restricted liquid water, such as arid deserts, frozen soils and ice,</p><p>mountain caves, etc. In deep cultural layers that lie beneath the water table, tanned leather</p><p>can survive for many centuries, and the natural tanning effects found in sphagnum bogs</p><p>are in part responsible for the spectacular preservation of ‘bog bodies’ such as Tollund</p><p>Man, Grauballe Man and Lindow Man. In most environments, however, liquid water not</p><p>only permits the growth of microbes, but it can also accelerate loss of protein via hydrol-</p><p>ysis. In normal soils, unmineralized collagen degrades rapidly via biological degradation</p><p>in which microorganisms use extracellular proteolytic enzymes to break the long collagen</p><p>molecule into smaller peptides that can be assimilated by bacteria and fungi. In miner-</p><p>alized collagen, the intimate association between the protein and mineral has a powerful</p><p>stabilizing effect that influences both microbial and chemical degradation of bones. The</p><p>resistance of bone to microbial attack arises from the absence of microscopic pores larger</p><p>than 8 nm. Microbial collagenases are large molecules with sizes ranging between 60 and</p><p>130 kDa (Bond and van Wart, 1984) and they are unable to penetrate the smaller pores</p><p>between the HAP and the collagen (Nielsen-Marsh et al., 2000; Gernaey et al., 2001).</p><p>Enzymatic hydrolysis of mineralized collagen would require that the mineral be removed</p><p>first, and this is indeed what happens in the microbial degradation of bones and teeth</p><p>(see below).</p><p>In the absence of enzymatic degradation, collagen can persist in bones for many hundreds</p><p>or even thousands of years. Bones recovered from deep gravel quarries at the Pleistocene</p><p>site of Shropham in Norfolk, UK, retain up to 85 % of their original collagen after more than</p><p>120 ky (Turner-Walker, unpublished data). In fact, mineralized collagen resists chemical</p><p>hydrolysis far longer than kinetic studies of unmineralized collagen suggest (Collins et al.,</p><p>1995). The chemical affinity between collagen and HAP is such that the presence of the</p><p>mineral not only excludes any molecule larger than water, it also physically constrains</p><p>(straitjackets) the collagen helix to a far greater degree than is achieved by tanning for</p><p>example. Unmineralized collagen will shrink or melt at a temperature of about 68�C. For</p><p>tanned leathers this temperature is typically in the range 75–85�C, whereas for mineralized</p><p>collagen this transformation does not occur until over 150�C (Nielsen-Marsh et al., 2000).</p><p>The straightjacketing effect of the HAP restricts the ability of the helix to expand. However,</p><p>since there is always some water held in the microporosity of bones (even in dry soils),</p><p>collagen will undergo slow chemical hydrolysis that can be accelerated by either an increase</p><p>or reduction in local pH and by increasing the temperature. Collagen is most stable against</p><p>hydrolysis when the pH lies in the range pH 3–7.5. At pH 1 the rate of hydrolysis is 10</p><p>times faster than at neutral pH, and at pH 12 the rate is 100 times faster (Collins et al., 1995:</p><p>Figure 1). Even though chemical hydrolysis may cause chain scissioning in the collagen</p><p>molecules, cutting the long fibrils into shorter peptide units, the strong affinity between</p><p>The Chemical and Microbial Degradation of Bones and Teeth 15</p><p>HAP and collagen combined with the small sizes of the micropores will severely restrict</p><p>diffusion of the fragments out of the bone structure. Therefore, unless there is an infiltration</p><p>of humics, which can affect cross-links between the damaged fibrils, a situation arises in</p><p>which archaeological bones can retain relatively high</p><p>collagen content and low macroporosity</p><p>but have considerably reduced mechanical strength (Collins et al., 1995; Turner-Walker and</p><p>Parry, 1995). Recent transmission electron microscope studies of degraded collagen from</p><p>bones excavated from experimental burials have shown that the structure of the fibrils</p><p>begins to break down after only a few years, exhibiting localized swellings and an apparent</p><p>unravelling of the tightly packed collagen molecules. This damage is limited to short sections</p><p>in the middle and ends of the fibrils and is also seen in cooked bones (Koon, 2006).</p><p>In the absence of microbial attack then, collagen and bone mineral are locked into a state</p><p>of mutual protection. The HAP protects the organic fraction from microbial enzymolysis</p><p>and retards the rate of chemical hydrolysis. The collagen in turn surrounds the tiny crys-</p><p>tallites of HAP and inhibits their dissolution by percolating groundwaters. Once one of the</p><p>components begins to break down, however, bone begins to degrade in the burial envi-</p><p>ronment. Loss of protein increases the microporosity and allows water to penetrate further</p><p>into the mineral phase. If the protein fraction is stripped from fresh bone using hydrazine</p><p>hydrate (NH2NH2·H2O) then the porosity changes dramatically: macroporosity shows an</p><p>almost fourfold increase from 0.075 cm3 g−1 to 0.300 cm3 g−1 and there is a corresponding</p><p>decrease in the microporosity from 0.059 cm3 g−1 to 0.031 cm3 g−1 – nearly half of its</p><p>initial value (Nielsen-Marsh and Hedges, 2000a). Because mineral makes up the bulk of the</p><p>volume of fresh bone, any loss through dissolution has a large influence on the degraded</p><p>bone’s porosity. In the case of demineralization, however, the bone loses rigidity and tensile</p><p>strength and is susceptible to shrinking, warping and cracking. More crucially, unless there</p><p>is some component in the groundwater either to inhibit microbial enzymolysis or to induce</p><p>cross-linking (tanning) in the exposed collagen matrix, the organic matter is quickly degraded</p><p>by soil microorganisms. In the case of many bog bodies, one of the components of sphagnum</p><p>peat, i.e. sphagnan, is responsible for both the loss of the HAP and, simultaneously, tanning</p><p>of the collagen and deactivation of microbial enzymes (Painter, 1983, 1991a,b; 1995, 1998;</p><p>Turner-Walker and Peacock, in press).</p><p>MICROBIAL DIAGENESIS OF BONES AND TEETH</p><p>During the decomposition of a corpse, the role of microorganisms is dominant and loss of</p><p>soft tissues, i.e. skeletonization, is largely mediated by bacteria and fungi, although autolysis</p><p>also plays an important role in the early stages of decay of the body. When a cell dies,</p><p>a cocktail of enzymes (proteases and DNases) are released which quickly break down the</p><p>surrounding cell components and tissues. The onset of this autolysis is very rapid, but it is</p><p>short-lived. Thereafter, bacterially mediated tissue destruction takes over with large numbers</p><p>of microorganisms being released from the gut into the abdominal cavity. The sequence of</p><p>autolytic decomposition follows that of tissues with the highest rates of synthesis of adenosine</p><p>triphosphate, the fuel that drives the body’s metabolism. Thus, the intestines, stomach, liver</p><p>and organs related to digestion are the first to deteriorate, together with the heart, blood</p><p>and circulatory systems. These are followed by the lungs, kidneys and bladder, brain and</p><p>nervous tissues, and later the skeletal muscles. Connective tissues, which are predominantly</p><p>collagen, are highly resistant to autolysis (Gill-King, 1997).</p><p>16 Advances in Human Palaeopathology</p><p>As the autolysis phase draws to an end, an almost entirely anaerobic environment is</p><p>created that is favourable to the proliferation of bacteria liberated by the decomposition of</p><p>the gut and, to a lesser extent, the local soil bacteria. In a healthy adult colon, 96–99 % of the</p><p>microbial florae are anaerobes and these work quickly on the body tissues, the fermentation</p><p>releasing the decomposition gases characteristic of putrefaction (Gill-King, 1997). There</p><p>has been some suggestion that the early release of microorganisms from the gut causes</p><p>more rapid degradation of the bones located around the abdomen (Child, 1995), but this is</p><p>not always borne out by examining skeletal element survival rate in large assemblages of</p><p>skeletons (Waldron, 1987). It has also been noted that diagenesis in bone from domesticated</p><p>animals that were slaughtered and butchered is often less pronounced than in equivalent</p><p>human bone from inhumations, raising the possibility that the early stages of putrefaction</p><p>have some bearing on later degradation by bacteria (Jans et al., 2004). However, it is</p><p>equally likely that these differences arise from the relative proportions of Haversian bone in</p><p>humans and animals. Domestic animals are typically slaughtered soon after reaching sexual</p><p>maturity and, therefore, have proportionally higher primary lamellar bone than humans,</p><p>who typically have more porous secondary or Haversian bone (Turner-Walker et al., 2002:</p><p>Figure 5).</p><p>Once reduced to a skeleton, the diagenesis of bones is mediated almost entirely by</p><p>microorganisms, the presence of which has a profound influence on their preservation</p><p>potential. Of course, local groundwater, oxygen availability, pH and temperature will not</p><p>only influence what kinds of microorganisms are present, but also how quickly they multiply.</p><p>From the earliest histological investigations of ancient bones, fungi were implicated in the</p><p>post-mortem destruction of bone tissues. Certainly, fungi can readily be found on excavated</p><p>bones, which are frequently washed in contaminated water and often are relegated to low-</p><p>priority storage facilities where damp and poor air circulation encourage mould growth.</p><p>Marchiafava et al. (1974), Hackett (1981) and Piepenbrink (1986) all conducted experiments</p><p>in fungal attack on buried bone in an attempt to replicate tunnelling and other features</p><p>associated with diagenesis. Marchiafava et al. (1974) compared experimentally buried fresh</p><p>human vertebrae with Neanderthal specimens using transmission electron microscopy and</p><p>optical microscopy. Mould specimens that developed around and within the vertebrae were</p><p>cultivated on agar for identification and subsequent inoculation into both sterilized soil</p><p>and bone autoclaved at 200�C for 20 min. Only one fungus, Mucor, was successfully</p><p>cultivated in isolation on inoculated sterile bone buried in sterilized earth. In retrospect,</p><p>this study would appear to have been fundamentally flawed. Autoclaving at 200�C for 20</p><p>min is approximately equivalent to boiling the bone for over 300 h. This would reduce</p><p>the collagen to a hydrolysed gelatine mass that would make an ideal food for a wide</p><p>range of microorganisms, but which formed a poor model for uncooked archaeological</p><p>bones.</p><p>Hackett (1981) experimented in the reproduction of what he termed microscopical focal</p><p>destruction in bone using samples of sterilized compact bone which he had buried in garden</p><p>soil at room temperature for 1 year. On excavation and microscopic examination, at least</p><p>two of these showed evidence of tunnelling and dissolution and reprecipitation of bone</p><p>mineral. The results of this experiment were ultimately inconclusive, however, since the</p><p>most promising specimens failed to show tunnelling in subsequent experiments. Towards</p><p>the end of his paper, Hackett suggested that the narrow Wedl tunnels found in exhumed and</p><p>fossil bone may result from the activity of certain bacteria, deriving their nourishment from</p><p>the debris left by fungi.</p><p>The Chemical and Microbial Degradation of Bones and Teeth 17</p><p>Piepenbrink (1986) also investigated the fungal degradation of buried bones using a wide</p><p>variety of analytical techniques, including histology, microradiography</p><p>and microbiological</p><p>incubation. He identified and isolated several species of fungi from stained areas in exhumed</p><p>bones. These fungi were subsequently found to colonize sterilized bone rapidly, but none</p><p>produced tunnelling or any of the other features associated with diagenetic alteration of</p><p>bone, such as loss of birefringence. As part of an investigation of the effects of microbial</p><p>degradation on trace element concentrations, Grupe and co-workers (Grupe and Piepenbrink,</p><p>1989; Grupe et al., 1993) inoculated fresh, irradiation-sterilized pig bone with several species</p><p>of fungi and bacteria. However, they also reported that no tunnelling could subsequently</p><p>be detected in any of the samples examined, although some superficial staining and loss of</p><p>birefringence was seen in the periosteal layers.</p><p>Research undertaken subsequent to the 1990s has shifted the focus away from fungi; now,</p><p>bacteria are recognized as playing a fundamental role in the destruction of bone tissues in</p><p>archaeological contexts. At the time of writing, there was no smoking gun that convinc-</p><p>ingly identified a particular species of soil organism as being responsible for destruction</p><p>of histology in bones or teeth. However, despite the numerous classifications for different</p><p>types of destruction seen in histological sections – Wedl or centrifugal tunnelling, linear</p><p>longitudinal tunnelling, budded tunnelling and lamellate tunnelling (Hackett, 1981; True-</p><p>man and Martill, 2002; Jans et al., 2004) – only two broad classes of bacteria seem to</p><p>be involved: aerobic bacteria in normal archaeological soils (Nielsen-Marsh and Hedges,</p><p>1999; Turner-Walker et al., 2002; Turner-Walker and Syversen, 2002) and cyanobacteria in</p><p>marine environments (Bell et al., 1991). Much of this shift from fungi to bacteria has arisen</p><p>as a result of applying more powerful techniques to the problem of diagenesis and using</p><p>microscopy techniques with higher resolutions than available earlier, particularly backscatter</p><p>scanning electron microscopy (BSEM), which has replaced microradiography and, to a large</p><p>extent, optical microscopy as the technique of choice (Bell, 1990; Bell et al., 1991, 1996;</p><p>Turner-Walker et al., 2002; Turner-Walker and Syversen, 2002).</p><p>The higher resolution of current SEM techniques over previous light microscopy of</p><p>thin sections has revealed a fine structure to the microscopical focal destruction described</p><p>by other researchers. The tunnels identified by Hackett and other researchers, and which</p><p>appeared to have diameters around 5–10 �m, are actually comprised of numerous smaller</p><p>pores with diameters that range from 0.1–1.0 �m (Figure 1.5). In BSEM images these</p><p>pores can be seen to be confined to localized zones, each 10–40 �m across. These zones</p><p>are frequently surrounded by an electron-dense region that either delineates the extent of</p><p>the tissue destruction or completely fills the intervening area between the small pores</p><p>(Figure 1.5d). In other places the pores lie within a general area of lower electron density</p><p>than the surrounding unaffected bone. This patchwork of demineralized and hypermineralized</p><p>zones is responsible for many of the features seen in light microscopy of archaeological</p><p>bone sections. The development of the sub-micrometre pore network is also responsible for</p><p>changes in the optical properties of ancient bones viewed in this section. The threadlike</p><p>tunnels created by the bacteria also disrupt the optical properties of bone tissues, reducing</p><p>its transparency in affected regions and causing them to appear opaque when viewed in</p><p>polarized light (Figure 1.5b). Staining from soil water may leave the affected bone black or</p><p>dark brown in thin sections (Turner-Walker and Syversen, 2002).</p><p>The soil bacteria responsible for the destruction of bone tissues infiltrate the interior of</p><p>the bones via the network of vascular channels but appear to be inhibited from attacking the</p><p>periosteal surfaces in many environments by the presence of humic substances in the soil</p><p>18 Advances in Human Palaeopathology</p><p>Figure 1.5 (a) Transmitted light image of medieval human bone from Wharram Percy, UK. Most</p><p>of the tissue is affected by diagenetic degradation, which obscures much of the histological features</p><p>(compare with Figure 1.2a and b) Section viewed in polarized light. Almost all the birefringence is</p><p>obscured or lost, demonstrating a disruption of the collagen – HAP bond. (c) High-magnification image</p><p>showing details of the affected bone. (d) An equivalent BSEM image of similar bone from Wharram</p><p>Percy. The affected bone is revealed as penetrated by numerous pores or tunnels. Demineralization</p><p>and reprecipitation of HAP is also evident</p><p>water (Figure 1.6a). These humics may act either on the collagen molecules, creating cross-</p><p>links that reduce the effectiveness of bacterial enzymes (Hedges, 2002), or by deactivating</p><p>the collagenases themselves (Jans et al., 2004). The penetration of bacteria through the</p><p>compact bone tissue is influenced by the microarchitecture of the tissues. For example,</p><p>bacteria seem unable to cross the cement lines that mark the boundaries between secondary</p><p>osteons (Haversian systems) and the surrounding primary lamellar bone or that mark the</p><p>reversal of resorption in remodelled bone (Figure 1.6b). In cross-sections of affected bone,</p><p>some bacterial colonies can be seen to tunnel normal to the plane of the section (i.e. along the</p><p>long axis of the bone), whereas in other places they create meandering tunnels that stream</p><p>parallel to the plane of the section (Figure 1.6c and d respectively). This suggests that the</p><p>bacteria follow the orientation of the collagen fibres in different parts of the bone and are</p><p>able to exploit planes of weakness in the tissues.</p><p>This preferred orientation in the spread of bacteria through calcified tissues is much more</p><p>marked in longitudinal sections of diagenetically altered teeth. Even in cases where bacterial</p><p>destruction obscures much of the histological details of the tissues it is possible to distinguish</p><p>the boundary between the dentine, where elongated destructive foci are aligned along the</p><p>dentinal tubules, and the cementum, where the destructive foci are larger and more globular,</p><p>The Chemical and Microbial Degradation of Bones and Teeth 19</p><p>Figure 1.6 BSEM images of bones from various sites. (a) Inhibition of bacterial attack at the periosteal</p><p>surface. (b) Bacterial tunnelling stopped by cement line surrounding osteon (arrowed). (c) Bacterial</p><p>tunnelling passing normal to the plane of the cross-section. Note the bright borders of hypermineralized</p><p>HAP. (d) Meandering pathways of bacterial tunnelling parallel to plane of cross-section</p><p>showing no preferred orientation (Figure 1.7a and b). Where recent and archaeological teeth</p><p>show evidence of dental caries there is no similarity with the kinds of diagenetic tissue</p><p>destruction seen in bone, cementum or dentine. Rather, there is a general demineralization</p><p>of the dentine that reveals the growth patterns of dentinal tubules (Figure 1.7c), but none of</p><p>the tunnelling or reprecipitated mineral seen in typical bacterial alteration of skeletal tissues.</p><p>Similarly, demineralization resulting from caries highlights the internal structure of enamel</p><p>(Figure 1.7d).</p><p>The observation that bone mineral is dissolved and reprecipitated in the zones affected</p><p>by bacterial attack supports theoretical considerations that microbial enzymes are unable to</p><p>degrade mineralized collagen. Further support for the necessity to demineralize bone tis-</p><p>sues prior to enzymatic degradation is provided by studies of remodelling in living bone.</p><p>Bone resorption is undertaken by mature osteoclasts using a combination of acid dissolution</p><p>of bone mineral and destruction of organic matrix by proteolytic enzymes. This essential</p><p>initial step, i.e. removal of bone mineral, determines the rate and extent of bone removed</p><p>from the resorption pit (Ortner and Turner-Walker, 2003). In bone that has been recov-</p><p>ered from acid or free-draining leaching soils, the reprecipitated HAP that is no longer</p><p>protected by its intimate association with collagen is susceptible to dissolution and loss.</p><p>The ragged holes left behind erase any trace of the smaller porosity and leave the bone</p><p>extremely fragile. In extreme cases the bone may disappear entirely from the archaeological</p><p>record.</p><p>20 Advances in Human Palaeopathology</p><p>Figure 1.7 Diagenetic degradation of teeth. (a) Longitudinal section of root of archaeological human</p><p>incisor from Wharram Percy, UK. The pattern of bacterial tunnelling shows clearly the border between</p><p>the dentine and the overlying cementum. (b) Detail of image in (a). (c) Demineralized dentine in a</p><p>modern tooth with dental caries. This is easily distinguished from degradation by soil bacterial in</p><p>archaeological specimens. (d) Demineralized enamel in modern dental caries specimen</p><p>Of the bone that does survive and is recovered from excavation sites, the bacterial attack</p><p>described in the previous paragraphs is almost ubiquitous and can be found in bones of</p><p>almost all ages, from decades to millions of years. Only in bones recovered from contexts</p><p>representing rapid burial in anoxic sediments or those from very cold climatic regions are</p><p>these features absent. In medieval skeletons from Trondheim in mid-Norway, for example,</p><p>bacterial destruction of bone tissues is not in evidence. A combination of low average soil</p><p>temperatures and graves cut into waterlogged, organic-rich or clay soils has led to spectacular</p><p>preservation of the bones, which consequently have a high residual collagen content and</p><p>excellent preservation of lipids and other biomolecules (Figure 1.8a). These observations</p><p>suggest two conclusions. First, the bacterial attack seen in so many bones derives from the</p><p>action of aerobic soil bacteria. Second, since the early stages of putrefaction of the corpses</p><p>in Trondheim presumably followed a similar path as those in more temperate regions, the</p><p>influence of gut bacteria on subsequent destruction of bone tissue may not be as important</p><p>as has been suggested by some researchers. Bones from waterlogged anoxic environments</p><p>often contain pyrite framboids in their internal porosity (Figure 1.8b). These clusters of finely</p><p>divided iron sulphides are a characteristic by-product of the metabolism of certain anaerobic</p><p>sulphate-reducing bacteria (SRB). These SRB are primitive organisms that are incapable of</p><p>metabolizing large organic molecules such as peptide fragments and, thus, cannot destroy</p><p>mineral tissues directly. Instead, they use the sulphate ion as an oxidizing agent for simple</p><p>The Chemical and Microbial Degradation of Bones and Teeth 21</p><p>Figure 1.8 (a) Medieval human femur from Trondheim, Norway. The histology is perfectly preserved</p><p>with no evidence of bacterial degradation. Note the actively resorbing osteons and numerous cement</p><p>lines showing several remodelling episodes. (b) Bone from a Neolithic cemetery in Ypenburg, the</p><p>Netherlands. A rise in the local water table in antiquity has caused loss of the bacterially degraded</p><p>tissues, and anoxic conditions have favoured the colonization of the pore spaces by SRB. Note the</p><p>well-developed pyrite framboids. (c) Animal bone from a Mesolithic site in the Vale of Pickering,</p><p>UK. This is similar to the bone from Ypenburg, but oxidation of the pyrite to sulphuric acid has given</p><p>rise to crystallization of lenticular gypsum crystals in the pore spaces. (d) Animal bone from the late</p><p>Neolithic site of Aartswoud, the Netherlands. The settlement was on salt marshes and tidal flats, and</p><p>this marine environment is reflected in the characteristic tunnelling by cyanobacteria. Note tunnelling</p><p>is limited to the outer millimetres</p><p>organic compounds, such as acetate, lactate and propionate, found in decaying organic matter.</p><p>There is a corresponding reduction of the sulphate ion to sulphide, which combines with</p><p>metal ions, such as iron to give iron sulphides. If there is a change in the burial environment</p><p>to more oxidizing conditions, then this finely divided pyrite can undergo oxidation with</p><p>the consequent release of sulphate and hydrogen ions. The resulting fall in pH can cause</p><p>local dissolution of HAP and give rise to deposition of gypsum (CaSO4·2H2O, Figure 1.8c;</p><p>Turner-Walker, 1998a,b) or vivianite (Fe3(PO4)2·8H2O) within and on the bones (Mann et al.,</p><p>1998; Maritan and Mazzoli, 2004).</p><p>The other group of organisms associated with tunnelling destruction of bone tissues are</p><p>the cyanobacteria. These are phototrophic organisms and, consequently, are restricted in</p><p>their habitats. Nevertheless, they are implicated in destruction of bones from tidal or estu-</p><p>arine deposits, where they may thrive down to depths of several metres of water. Bone</p><p>destruction in specimens from these sites exhibit a different pattern of attack (Figure 1.8d).</p><p>Destruction proceeds from the periosteal surface inwards, or sometimes from the larger</p><p>22 Advances in Human Palaeopathology</p><p>physiological pores. The size (5–10 �m), close spacing and tortuous branching habit of</p><p>these tunnels are very similar to those made by the endolithic filamentous cyanobacterium</p><p>Mastigocoleus testarum, responsible for bioerosion in marine shells and vertebrate skeletons</p><p>(Davis, 1997; Kaehler, 1999: figure 6E). In cross-section the pores display ragged borders,</p><p>rather than the smooth, globular cross-sections of the bacterially degraded bones from ter-</p><p>restrial burial sites, and branch more frequently. The tunnels attributed to cyanobacteria do</p><p>not appear to respect the natural micro-architecture of the bone in the same way that terres-</p><p>trial soil bacteria do, and there is no evidence for local demineralization or reprecipitation</p><p>of HAP.</p><p>Diagenetic Pathways</p><p>There are two predominant mechanisms of bone degradation in archaeological soils, which</p><p>may or may not proceed simultaneously. These are bacterial degradation of the tissues and</p><p>chemical hydrolysis of bone collagen. In most bones from aerated soils, both mechanisms</p><p>proceed simultaneously, albeit at different rates, and the net result is a gradual loss of collagen</p><p>content over time (Figure 1.9). Bacterial degradation is by far the most rapid pathway by</p><p>several orders of magnitude. Bone buried in tropical countries may be rapidly consumed</p><p>within a few centuries, whereas bone buried in colder or waterlogged sediments may survive</p><p>for several thousands of years, or even hundreds of thousands of years. The two mechanisms</p><p>may be distinguished at the limits of resolution of scanning electron microscopes. Figure 1.10</p><p>shows a backscatter image of bone from the medieval site of Wharram Percy in the UK,</p><p>viewed with a high-resolution field-emission scanning electron microscope. The large circular</p><p>voids on the left-hand side of the image were created by soil bacteria. The area affected</p><p>is bounded by a band of dense, reprecipitated HAP that appears bright in backscatter.</p><p>The countless tiny holes filling the right-hand side of the image are the voids left by the</p><p>hydrolysis and leaching of the collagen fibres, leaving a negative cast of undissolved HAP.</p><p>The occasional swirling bands reflect changes in the orientation of collagen fibres in the</p><p>lamellar structure of the tissue.</p><p>The state of preservation of any skeleton, or assemblage of bones, depends upon its early</p><p>taphonomic history and the particular diagenetic trajectory it follows. The former may be</p><p>controlled by cultural, economic</p><p>or social factors, whereas the latter may be controlled</p><p>solely by geographical location, climatic factors and the character of the burial environment.</p><p>Thus, preservation may differ markedly depending upon whether a corpse is interred in a</p><p>stone-lined crypt, a wooden coffin or directly in the soil. The degree of bacterial degradation</p><p>(or its total absence) may depend upon depth of burial (above or below the water table),</p><p>the local average soil temperature and the dissolved oxygen content of the soil waters.</p><p>Graves cut into alkaline chalky soils, particularly those in hot countries, show evidence of</p><p>collagen hydrolysis and shrinkage cracks, as well as some limited bacterial attack (Jans</p><p>et al., 2004). Waterlogged acidic or neutral soils produce bones with little or no bacterial</p><p>attack and a high residual collagen content, but with some potential for demineralization and</p><p>permineralization with exogenous mineral species. Bodies that find their way into sphagnum</p><p>bogs, either by accident or as some ritual sacrifice, exhibit another type of preservation in</p><p>which the collagen matrix survives as part of a ‘tanned’ bog body, but where the mineral</p><p>may have been completely lost.</p><p>It is worth noting here that the relationships between bone preservation at the microscopic</p><p>level (as revealed by histological studies) and gross preservation as perceived by visual</p><p>The Chemical and Microbial Degradation of Bones and Teeth 23</p><p>Figure 1.9 (a) Graph of calcium versus phosphorus (in weight percent) for archaeological bones</p><p>from different sites. The open circles represent bones excavated from aerated soils. The filled square is</p><p>modern sheep bone. The open triangles represent chemically deproteinated modern bone (i.e. pure bone</p><p>mineral). The crosses represent bones excavated from a waterlogged site lying beneath peat deposits.</p><p>It is clear that some bones suffered bacterial degradation (movement along collagen loss trajectory)</p><p>prior to demineralisation in the peat deposits. The single filled circle is bone from a very alkaline soil</p><p>that contained diagenetic calcite. (b) Graph of calcium versus residual protein for bones from different</p><p>sites. Symbols are as described for (a). Unpublished data from Turner-Walker (1993)</p><p>24 Advances in Human Palaeopathology</p><p>Figure 1.10 High-resolution field-emission scanning electron microscope backscatter image of</p><p>medieval human femur from Wharram Percy, UK. The circular holes (approximately 500 nm in</p><p>diameter) on the left side of the image were created by soil bacteria. Hydrolysis and leaching of the</p><p>collagen fibres is revealed as the numerous small holes (40–50 nm in diameter) filling the rest of the</p><p>image. The central band that appears bright in backscatter represents HAP that has been dissolved by</p><p>bacteria and reprecipitated at a higher density</p><p>inspection of archaeological skeletons are far from clear. Some skeletal assemblages (partic-</p><p>ularly those from alkaline soils) that are grossly very well preserved may exhibit very poor</p><p>preservation histologically, with considerable destruction of cortical bone by soil bacteria.</p><p>Often the outer millimetre of bone tissue where it has been in direct contact with the soil</p><p>is preferentially preserved (Figure 1.6a) compared with the interior. Bacterial degradation</p><p>may have to be sufficiently advanced to render archaeological bone prone to fragmentation</p><p>and dissolution before it substantially reduces the ability of osteoarchaeologists to extract</p><p>useful data from an assemblage. Conversely, skeletons that exhibit especially good his-</p><p>tological preservation, such as those from medieval Trondheim, can sometimes show very</p><p>variable gross preservation, including erosion of bone surfaces. This observation has obvious</p><p>implications for the palaeopathological diagnosis of many diseases that affect the surface</p><p>texture or morphology of bones and raises the question as to what extent palaeohistological</p><p>work can aid the diagnosis of disease in archaeological skeletons (Turner-Walker and Mays,</p><p>Chapter 7).</p><p>What it is important to appreciate is that, over archaeological time-scales, burial envi-</p><p>ronments can change, either as a direct result of human activities (drainage of wetlands,</p><p>build up of deep cultural layers in urban contexts, etc.) or by natural geological or climatic</p><p>processes. The soil conditions from which bones are excavated may not necessarily represent</p><p>those of antiquity; therefore, the preservation state of the bones may not reflect the preserv-</p><p>ing qualities of the surrounding soil. Bones, therefore, travel through time along different</p><p>diagenetic trajectories or pathways, which in some cases may be approximately linear but</p><p>in other cases may make abrupt changes in course according to the evolution of the burial</p><p>The Chemical and Microbial Degradation of Bones and Teeth 25</p><p>environment (Figure 1.9a). It is possible that a history of the burial environment is preserved</p><p>in the histology of the bones themselves, and in certain circumstances that is clearly the case</p><p>(Turner-Walker, 1998b; Turner-Walker and Jans, in press).</p><p>Despite the considerable advances made over the past decade in the understanding of</p><p>diagenetic alteration of archaeological bones, much still remains to be done. There is still the</p><p>unresolved question of which microorganism (or group of microorganisms) is responsible</p><p>for the tunnelling that is such a common feature in ancient bone. It would also be interesting</p><p>to trace the antiquity of the organism(s) responsible by tracking this tunnelling through the</p><p>fossil record. The exact relationships between microbial destruction of mineralized tissues</p><p>and the survival of biomolecular evidence are also unresolved, although all evidence to date</p><p>points to a regular relationship between histological preservation and the ability to extract</p><p>intact biomolecules. In particular, the precise location of preserved DNA in archaeological</p><p>bones and teeth, and how this influences the success or failure of decontamination and</p><p>extraction protocols, would seem a pressing question. In conclusion, it may be expected that</p><p>the coming decade will see the answers to these and many other questions relating to the</p><p>potential data locked away in archaeological and fossil bones.</p><p>REFERENCES</p><p>Agarwal SC, Grynpas MD. 1996. Bone quantity and quality in past populations. Anat Rec 246:</p><p>423–432.</p><p>Badone E, Farquhar RM. 1982. Application of neutron activation analysis to the study of element</p><p>concentration and exchange in fossil bones. J Radioanalytic Chem 69: 291–311.</p><p>Baron H, Hummel S, Herrmann, B. 1996. Mycobacterium tuberculosis complex DNA in ancient human</p><p>bones. J Archaeol Sci 23: 667–671.</p><p>Bartsiokas A, Middleton AP. 1992. Characterization and dating of recent and fossil bone by X-ray</p><p>diffraction. J Archaeol Sci 19: 63–72.</p><p>Behrensmeyer AK, Hill AP. 1980. Fossils in the Making: Vertebrate Taphonomy and Palaeoecology.</p><p>University of Chicago Press: Chicago.</p><p>Bell LS. 1990. Palaeopathology and diagenesis: an SEM evaluation of structural changes using</p><p>backscattered electron imaging. J Archaeol Sci 17: 85–102.</p><p>Bell LS, Boyd A, JonesSJ. 1991. Diagenetic alteration to teeth in situ illustrated by backscattered</p><p>electron imaging. Scanning 13(2): 173–183.</p><p>Bell LS, Skinner MF, Jones SJ. 1996. The speed of post mortem change to the human skeleton and its</p><p>taphonomic significance. Forensic Sci Int 82(2): 129–140.</p><p>Bethel PH, Carver MOH. 1987. Detection and enhancement of decayed inhumations at Sutton Hoo.</p><p>In Death Decay and Reconstruction, Boddington A, Garland, AN, Janaway RC (eds). Manchester</p><p>University Press: Manchester; 10–21.</p><p>Bond MD, van Wart HE 1984. Characterisation</p><p>of the individual collagenases from Clostridium</p><p>histoliticum. Biochemistry 23: 3085–3091.</p><p>Braun M, Cook DC, Pfeiffer S. 1998. DNA from Mycobacterium tuberculosis complex iden-</p><p>tified in North American, pre-Columbian human skeletal remains. J Archaeol Sci 25:</p><p>271–277.</p><p>Briggs DEG. 1992. Conodonts: a major extinct group added to the vertebrates. Science 256: 1285–1286.</p><p>Buikstra JE, Frankenberg S, Lambert JB, Xue L. 1989. Multiple elements: multiple expectations. In</p><p>The Chemistry of Prehistoric Human Bone, Price TD (ed.). Cambridge University Press: Cambridge;</p><p>155–210.</p><p>26 Advances in Human Palaeopathology</p><p>Carver M. 2005. Sutton Hoo: A Seventh-Century Princely Burial Ground and its Context. British</p><p>Museum Press: London.</p><p>Cattaneo C, Gelsthorpe K, Phillips P, Sokol RJ. 1995. Differential survival of albumin in ancient bone.</p><p>J Archaeol Sci 22: 271–276.</p><p>Child AM. 1995. Microbial taphonomy of archaeological bone. Stud Conserv 40: 19–30.</p><p>Collins MJ, Riley MS, Child A, Turner-Walker G. 1995. A basic mathematical simulation of the</p><p>chemical degradation of ancient collagen. J Archaeol Sci 22: 175–183.</p><p>Davis PG. 1997. The bioerosion of bird bones. Int J Osteoarchaeol 7: 388–401.</p><p>De Cupere VM, Van Wetter J, Rouxhet PG. 2003. Nanoscale organization of collagen and mixed</p><p>collagen – pluronic adsorbed layers. Langmuir 19: 6957–6967.</p><p>De Vries H, Oakley KP. 1959. Radiocarbon dating of the Piltdown skull and jaw. Nature 184: 224–226.</p><p>Dupras TL, Schwarcz HP. 2001. Strangers in a strange land: stable isotope evidence for human</p><p>migration in the Dakhleh Oasis, Egypt. J Archaeol Sci 28: 1199–1208.</p><p>Eastoe JE, Eastoe B. 1954. The organic constituents of mammalian compact bone. Biochem J 57:</p><p>453–459.</p><p>Efremov IA. 1940. Taphonomy: a new branch of palaeontology. Pan-Am Geol 74: 81–93.</p><p>El-Kammar A, Hancock RGV, Allen RO. 1989. Human bones as archaeological samples: changes</p><p>due to contamination and diagenesis. In Archaeolaeolgical Chemistry IV, Allen RO (ed.). American</p><p>Chemical Society: Washington, DC.</p><p>Evershed RP, Turner-Walker G, Hedges REM, Tuross N, Leyden A. 1995. Preliminary results for the</p><p>analysis of lipids in ancient bone. J Archaeol Sci 22: 277–290.</p><p>Geigl E-M. 2002. On the circumstances surrounding the preservation and analysis of very old DNA.</p><p>Archaeometry 44(3): 337–342.</p><p>Gernaey AM, Waite ER, Collins MJ, Craig OE, Sokol RJ. 2001. Survival and interpretation of</p><p>archaeological proteins. In Handbook of Archaeological Science, Brothwell DR, Pollard AM (eds).</p><p>Wiley: Chichester; 323–329.</p><p>Gilbert MTP, Rudbeck L, Willerslev E, Hansen AJ, Smith C, Penkman K, Prangenberg, K,</p><p>Nielsen-Marsh C, Jans M, Arthur P, Lynnerup N, Turner-Walker G, Biddle M, Kjølbye-</p><p>Biddle B, Cooper A, Collins MJ. 2005. Biochemical and physical correlates of DNA con-</p><p>tamination in archaeological bones and teeth excavated at Matera, Italy. J Archaeol Sci 32:</p><p>785–793.</p><p>Gilbert MTP, Hansen AJ, Willerslev E, Turner-Walker G, Collins MJ. 2006. Insights into the processes</p><p>behind the contamination of degraded human teeth and bone samples with exogenous sources of</p><p>DNA. Int J Osteoarchaeol 15: 1–9.</p><p>Gill-King H. 1997. Chemical and ultrastructural aspects of decomposition. In Forensic Taphonomy:</p><p>The Postmortem Fate of Human Remains, Haglund WD, Sorg MH (eds). CRC Press: Boca Raton;</p><p>93–108.</p><p>Giraud-Guille MM. 1988. Twisted plywood architecture of collagen fibrils in human compact bone</p><p>osteons. Calcified Tissue Int 42: 167–180.</p><p>Götherström A, Collins MJ, Angerbjörn A, Liden K. 2002. Bone preservation and DNA amplification.</p><p>Archaeometry 44: 395–404.</p><p>Grupe G, Piepenbrink H. 1989. Impact of microbial activity on trace element concentrations in</p><p>excavated bones. Appl Geochem 4: 293–298.</p><p>Grupe G, Dreses-Werringloer U, Parsche F. 1993. Initial stages of bone decomposition: causes and</p><p>consequences in prehistoric human bone. In Archaeology at the Molecular Level, Lamberts JB,</p><p>Grupe G (eds). Springer: Berlin; 257–274.</p><p>Hackett CJ. 1981. Microscopical focal destruction (tunnels) in exhumed human bones. Med Sci Law</p><p>21: 243–265.</p><p>Hassan AA, Ortner DJ. 1977. Inclusions in bone material as a source of error in radiocarbon dating.</p><p>Archaeometry 19: 131–135.</p><p>Hedges REM. 2002. Bone diagenesis: an overview of processes. Archaeometry 44: 319–328.</p><p>The Chemical and Microbial Degradation of Bones and Teeth 27</p><p>Hedges REM, Millard AR. 1995. Bones and groundwater towards the modelling of diagenetic processes.</p><p>J Archaeol Sci 22: 155–165.</p><p>Hedges REM, Millard AR, Pike AWG. 1995. Measurements and relationships of diagenetic alteration</p><p>of bone from three archaeological sites. J Archaeol Sci 22: 201–211.</p><p>Horowitz LK, Goldberg P. 1989. A study of Pleistocene and Holocene hyena coprolites. J Archaeol</p><p>Sci 16: 71–94.</p><p>Jans MME, Nielsen-Marsh CM, Smith CI, Collins MJ, Kars H. 2004. Characterisation of microbial</p><p>attack on archaeological bone. J Archaeol Sci 31: 87–95.</p><p>Kaehler S. 1999. Incidence and distribution of phototrophic shell-degrading endoliths of the brown</p><p>mussel Perna perna. Mar Biol 135: 505–514</p><p>Keeley HCM, Hudson GE, Evans J. 1977. Trace element contents of human bones in various states of</p><p>preservation: 1. The soil silhouette. J Archaeol Sci 4: 19–24.</p><p>Koon HEC. 2006. Detecting cooked bone in the archaeological record: a study of the ther-</p><p>mal stability and deterioration of bone collagen. Unpublished doctoral thesis, University of</p><p>York.</p><p>Lambert JB, Szpunar CB, Buikstra JE. 1979. Chemical analysis of excavated human bone from middle</p><p>and late Woodland sites. Archaeometry 21: 115–129.</p><p>Lambert JB, Vlask SM, Thometz AC, Buikstra JE. 1982. A comparative study of the chemical analysis</p><p>of ribs and femurs in Woodland populations. Am J Phys Anthropol 59: 289–294.</p><p>Lambert JB, Simpsom SV, Buikstra JE, Hanson D. 1983. Electron microprobe analysis of elemental</p><p>distribution in excavated human femurs. Am J Phys Anthropol 62: 409–423.</p><p>Lambert JB, Simpson SV, Szpunar CB, Buikstra JE. 1985a. Bone diagenesis and dietary analysis.</p><p>J Hum Evol 14: 477–482.</p><p>Lambert JB, Simpson SV, Weiner SG, Buikstra JE. 1985b. Induced metal ion exchange in excavated</p><p>human bone J Archaeol Sci 12: 85–92.</p><p>Lee-Thorpe JA, van der Merwe NJ. 1987. Carbon isotope analysis of fossil bone apatite. S Afr J Sci</p><p>83: 712–715.</p><p>Lee-Thorpe J, van der Merwe NJ. 1991 Aspects of the chemistry of modern and fossil biological</p><p>apatites. J Archaeol Sci 18: 343–354.</p><p>Lowenstam HA, Weiner S. 1989. On Biomineralization. Oxford University Press: Oxford.</p><p>Mann RW, Feather ME, Tumosa CS, Holland TD, Schneider KN. 1998. A blue encrustation</p><p>found on skeletal remains of Americans missing in action in Vietnam. Forensic Sci Int 97:</p><p>79–86.</p><p>Marchiafava V, Bonucci E, Ascenzi A. 1974. Fungal osteoclasia: a model of dead bone resorption.</p><p>Calcified Tissue Res 14: 195–210.</p><p>Maritan L, Mazzoli C. 2004. Phosphates in archaeological finds: implications for environmental</p><p>conditions of burial. Archaeometry 46: 673–683.</p><p>Mays S. 1998. The Archaeology of Human Bones. Routledge: London.</p><p>Mays S. 1999. Osteoporosis in earlier human populations. J Clin Densitom 2: 71–78.</p><p>Mays S. 2000. New directions in the analysis of stable isotopes in excavated bones and teeth. In Human</p><p>Osteology in Archaeology and Forensic Science, Cox M, Mays S (eds). Greenwich Medical Media:</p><p>London; 425–438.</p><p>Mays S, Turner-Walker G, Syversen U. 2006. Osteoporosis in a population from medieval Norway.</p><p>Am J Phys Anthropol 116: 34–44.</p><p>Muyzer G, Sandberg P, Knapen MHJ, Vermeer C, Collins M, Westbroek P. 1992. Preservation of the</p><p>bone protein osteocalcin in dinosaurs. Geology 20: 871–874.</p><p>Newesely H. 1989. Fossil bone apatite. Appl Geochem 4: 233–245.</p><p>Nielsen-Marsh CM. 1997. Studies in archaeological bone diagenesis. Unpublished doctoral thesis,</p><p>University of Oxford.</p><p>Nielsen-Marsh CM, Hedges REM. 1999.</p><p>Bone porosity and the use of mercury intrusion porosimetry</p><p>in bone diagenesis studies. Archaeometry 41: 165–174.</p><p>28 Advances in Human Palaeopathology</p><p>Nielsen-Marsh CM, Hedges REM. 2000a. Patterns of diagenesis in bone I: the effects of site</p><p>environments. J Archaeol Sci 27: 1139–1150.</p><p>Nielsen-Marsh CM, Hedges REM. 2000b. Patterns of diagenesis in bone II: effects of acetic acid</p><p>treatment and the removal of diagenetic CO2−</p><p>3 . J Archaeol Sci 27: 1151–1159.</p><p>Nielsen-Marsh CM, Gernaey AM, Turner-Walker G, Hedges REM, Pike AWG, Collins MJ. 2000. The</p><p>chemical degradation of bone. In Human Osteology in Archaeology and Forensic Science, Cox M,</p><p>Mays S (eds). Greenwich Medical Media: London; 439–454.</p><p>Olsen EC. 1980. Taphonomy: its history and role in community evolution. In Fossils in the Making:</p><p>Vertebrate Taphonomy and Palaeoecology, Behrensmeyer AK, Hill AP (eds). University of Chicago</p><p>Press: Chicago; 5–19.</p><p>Ortner DJ, Turner-Walker G. 2003. The biology of skeletal tissues. In Identification of Pathological</p><p>Conditions in Human Skeletal Remains (2nd edition), Ortner DJ (ed.). Academic Press: New York;</p><p>11–35.</p><p>Painter TJ. 1983. Residues of d-lyxo-5-hexosulopyranuronic acid in sphagnum holocellulose, and their</p><p>role in cross-linking. Carbohydr Res 124: C18–C21.</p><p>Painter TJ. 1991a. Lindow Man, Tollund Man and other peat-bog bodies: the preservation and antimi-</p><p>crobial action of sphagnan, a reactive glycuronoglycan with tanning and sequestering properties.</p><p>Carbohydr Polym 15: 123–142.</p><p>Painter TJ. 1991b. Preservation in peat. Chem Industr 17: 421–424.</p><p>Painter TJ. 1995. Chemical and microbiological aspects of the preservation process in sphagnum peat.</p><p>In Bog Bodies: New Discoveries and New Perspectives, Turner RC, Scaife RG (eds). British Museum</p><p>Press: London; 88–99.</p><p>Painter TJ. 1998. Carbohydrate polymers in food preservation: an integrated view of the Maillard</p><p>reaction with special reference to discoveries of preserved foods in sphagnum-dominated peat bogs.</p><p>Carbohydr Polym 36: 335–347.</p><p>Pate FD, Hutton JT, Norrish K. 1989. Ionic exchange between soil solution and bone: towards a</p><p>predictive model. Appl Geochem 4: 303–316.</p><p>Petchley F, Higham T. 2000. Bone diagenesis and radiocarbon dating of fish bones at the Shag River</p><p>mouth site, New Zealand. J Archaeol Sci 27: 135–150.</p><p>Piepenbrink H. 1986. Two examples of biogenous dead bone decomposition and their consequences</p><p>for taphonomic interpretation. J Archaeol Sci 13: 417–430.</p><p>Piepenbrink H. 1989. Examples of chemical changes during fossilization. Appl Geochem 4:</p><p>273–280.</p><p>Pike AWG. 1993. Bone porosity, water and diagenesis: towards a grand unified theory of bone</p><p>diagenesis. Unpublished undergraduate dissertation, University of Bradford.</p><p>Pike AWG, Nielsen-Marsh CM, Hedges REM. 2001. Modelling bone dissolution under different</p><p>hydrological regimes. In Archaeological Sciences ’97, Millard AR (ed.). BAR International Series</p><p>939. Archaeopress: Oxford; 127–132.</p><p>Posner AS. 1985. The structure of bone apatite surfaces. J Biomed Mater Res 19: 241–250.</p><p>Price TD (ed.). 1989. The Chemistry of Prehistoric Human Bone. Cambridge University Press:</p><p>Cambridge.</p><p>Price TD, Schoeninger MJ, Armelagos GJ. 1985. Bone chemistry and past behaviour: an overview.</p><p>J Hum Evol 14: 419–448.</p><p>Roux W. 1887. Über eine Knochen lebende Gruppe von Fadenpilzen (Mycelites ossifragus). Zeitschr</p><p>Wiss Zool 45: 227–254.</p><p>Sansom IJ, Smith MP, Armstrong HA, Smith MM. 1992. Presence of the earliest vertebrate hard</p><p>tissues in conodonts. Science 256: 1308–1311.</p><p>Sarathchandra P, Pope F, Ali S. 1999. Morphometric analysis of type I collagen fibrils in the osteoid</p><p>of osteogenesis imperfecta. Calcif Tissue Int 65: 390–395.</p><p>Schultze H-P. 1996. Conodont histology: an indicator of vertebrate relationship? Mod Geol 20:</p><p>275–286.</p><p>The Chemical and Microbial Degradation of Bones and Teeth 29</p><p>Smith CI, Craig OE, Prigodich RV, Nielsen-Marsh CM, Jans MME, Vermeer C, Collins MJ.</p><p>2005. Diagenesis and survival of osteocalcin in archaeological bone. J Archaeol Sci. 32:</p><p>105–113.</p><p>Stott AW, Evershed RP, Jim S, Jones V, Rogers JM, Tuross N, Ambrose S. 1999. Cholesterol as a</p><p>new source of palaeodietary information: experimental approaches and archaeological applications.</p><p>J Archaeol Sci 26: 705–716.</p><p>Taylor GM, Crossey M, Saldanha J, Waldron T. 1996. DNA from Mycobacterium tuberculosis iden-</p><p>tified in mediaeval human skeletal remains using polymerase chain reaction. J Archaeol Sci 23:</p><p>789–798.</p><p>Trueman CN, Martill DM. 2002. The long term preservation of bone: the role of bioerosion.</p><p>Archaeometry 44: 371–382.</p><p>Turner-Walker G. 1993. The characterisation of fossil bone. Unpublished PhD thesis, Department of</p><p>Archaeology, University of Durham.</p><p>Turner-Walker G. 1998a. The West Runton fossil elephant: a pre-conservation evaluation of its</p><p>condition, chemistry and burial environment. Conservator 22: 26–35.</p><p>Turner-Walker G. 1998b. Pyrite and bone diagenesis in terrestrial sediments: evidence from the West</p><p>Runton fresh water bed. Bull Geol Soc Norfolk 48: 3–26.</p><p>Turner-Walker G, Jans MME. in press. Reconstructing taphonomic histories using histological analysis.</p><p>Palaeogeog Palaeoclimatol Palaeoecol.</p><p>Turner-Walker G, Parry TV. 1995. The tensile strength of archaeological bone. J Archaeol Sci 22:</p><p>185–191.</p><p>Turner-Walker G, Peacock EE. in press. Preliminary results of bone diagenesis in Scandinavian bogs.</p><p>Palaeogeog Palaeoclimatol Palaeoecol.</p><p>Turner-Walker G, Syversen U. 2002. Quantifying histological changes in archaeological bones using</p><p>BSE-SEM image analysis. Archaeometry 44: 161–168.</p><p>Turner-Walker G, Nielsen-Marsh CM, Syversen U, Kars H, Collins MJ. 2002. Sub-micron spongiform</p><p>porosity is the major ultra-structural alteration occurring in archaeological bone. Int J Osteoarchaeol</p><p>12: 407–414.</p><p>Tuross N. 2003. Recent advances in bone, dentin and enamel biochemistry. In Identification of Patho-</p><p>logical Conditions in Human Skeletal Remains, 2nd edition, Ortner DJ (ed.). Academic Press: New</p><p>York; 65–72.</p><p>Van Klinken GJ. 1999. Bone collagen quality indicators for palaeodietary and radiocarbon</p><p>measurements. J Archaeol Sci 26: 687–695.</p><p>Waldron T. 1987. The relative survival of the human skeleton: implications for palaeopathology.</p><p>In Death Decay and Reconstruction, Boddington A, Garland AN, Janaway RC (eds). Manchester</p><p>University Press: Manchester; 55–64.</p><p>Wandeler P, Smith S, Morin PA, Pettifor RA, Funk SM. 2003. Patterns of nuclear DNA degeneration</p><p>over time a case study in historic teeth samples. Mol Ecol 12: 1087–1093</p><p>Wedl C. 1864. Über einen im Zahnbein und Knochen keimenden Pilz. Mineral Biol Erd 50: 171–193.</p><p>Weiner S, Bar-Yosef O. 1990. States of preservation of bones from prehistoric sites in the near east.</p><p>J Archaeol Sci 17: 187–196.</p><p>Weiner S, Traub W. 1992. Bone structure; from ångströms to microns. FASEB J 6: 879–885.</p><p>Weser U, Kaup Y, Etspüler H, Kenward N, Hedges REM. 1996. Biochemically and immunologi-</p><p>cally active alkaline phosphatase in archaeologically important bone samples. J Archaeol Sci 23:</p><p>723–730.</p><p>Yamamoto S, Nakamura F, Hitomi J, Shigeno M, Sawaguchi S, Abe H, Ushiki T. 2000. Atomic force</p><p>microscopy of intact and digested collagen molecules. J Electron Microsc 49: 423–427.</p><p>2</p><p>How Representative</p><p>Are Human Skeletal</p><p>Assemblages for Population</p><p>Analysis?</p><p>Ron Pinhasi1 and Chryssi Bourbou2</p><p>1 Department of Archaeology, University College Cork, Cork,</p><p>Ireland</p><p>2 28 th Ephorate of Byzantine Antiquities, Chania, Crete, Greece</p><p>INTRODUCTION</p><p>In recent decades the palaeopathological study of health and disease patterns has demon-</p><p>strated a shift from the descriptive realm of case studies to the</p><p>more theoretical realm of</p><p>palaeoepidemiology and cross-population analysis. Bioarchaeologists can address research</p><p>questions regarding diachronic changes in health status in past populations, changes in</p><p>morbidity and mortality profiles, and the prevalence of certain pathological conditions.</p><p>However, any population-level analysis is affected by a set of biases and limitations</p><p>that typify archaeological skeletal samples. These include methodological aspects, such</p><p>as age and sex estimation, excavation and recovery methods, and survival biases of skeletal</p><p>samples.</p><p>The aim of this chapter is to discuss some of the sources of bias that potentially exert</p><p>significant effects on population-based palaeoepidemiological assessments. The next section</p><p>of the chapter addresses biases in the representativeness of archaeological samples. A short</p><p>review of aspects of the interpretation of morbidity status in palaeopopulations and the effects</p><p>of some demographic parameters on the interpretation of palaeopathological data are then</p><p>provided. The chapter concludes with a demonstration of an age-correction epidemiological</p><p>method to archaeological material.</p><p>Advances in Human Palaeopathology Edited by Ron Pinhasi and Simon Mays</p><p>© 2008 John Wiley & Sons, Ltd. ISBN: 978-0-470-03602-0</p><p>32 Advances in Human Palaeopathology</p><p>REPRESENTATIVENESS OF ARCHAEOLOGICAL SAMPLES</p><p>Under ideal conditions, the skeletal sample studied represents the total deaths in the popula-</p><p>tion during the time period under investigation (Alesan et al., 1999). However, this is hardly</p><p>ever the case, as some of the skeletons, particularly those of foetuses and infants, do not</p><p>survive or were buried elsewhere. Moreover, the archaeologist usually does not excavate the</p><p>total cemetery due to various budgetary, time and logistical constraints. In most cases, lack</p><p>of written documentation makes it impossible to ascertain whether the excavated sample is</p><p>a true random sample of the cemetery skeletal population. Therefore, we do not know how</p><p>well the skeletal sample represents the living population and we have no means to resolve</p><p>this issue, as various non-random taphonomic, cultural, and demographic factors may have</p><p>played a major part in the sampling process (Waldron, 1994; Hoppa, 2002). The sections</p><p>below discuss some of the more important influences on the degree to which a cemetery</p><p>sample is representative of the original population.</p><p>Bone Survival and Recovery</p><p>A number of physical, chemical and biological factors play a significant role in determining</p><p>the preservation of human remains (Waldron 1987, 1994; Mays, 1998; Bourbou, 1999; Caffel</p><p>et al., 2001; Turner-Walker, Chapter 1). A buried body has a better chance of surviving</p><p>than an exposed one, the latter being more vulnerable to environmental degradation and</p><p>animal attacks (Henderson, 1987). Different components of the skeleton exhibit different</p><p>preservation patterns, as they vary in their physical strength and structure. Trabecular bone</p><p>seems to decay more rapidly in the soil than cortical bone, partially due to the larger surface</p><p>area of trabecular bone favouring chemical exchange between bone and soil (Lambert et al.,</p><p>1982; Grupe, 1988) and partially due to its mechanical fragility. Studies on the relative</p><p>survival and recovery of bones of adults from the Romano-British site at West Tenter Street,</p><p>London (Waldron, 1987), and on medieval skeletons from the Blackfriars site in Ipswich,</p><p>England (Mays, 1992), confirm that small bones and those with high trabecular and low</p><p>cortical bone contents are poorly represented. The low mineralization of subadult bones and</p><p>their fragile nature contribute to their poor preservation (Currey and Butler, 1975; Gordon</p><p>and Buikstra, 1981; Specker et al., 1987; Nicholson, 2001).</p><p>The underrepresentation of specific skeletal parts is of great relevance to palaeopathol-</p><p>ogists. For example, in leprosy, peripheral nerve damage causes loss of sensation and</p><p>eventually ulceration and secondary infections in both hand and foot (Steinbock, 1976;</p><p>Aufderheide and Rodríguez-Martín, 1998). Consequently, some of the hand and foot pha-</p><p>langes are expected to be missing from the archaeological assemblage owing to disease.</p><p>A systematic analysis of the prevalence of leprosy in a skeletal population must take into</p><p>consideration whether missing hand and foot phalanges should be possibly attributed to both</p><p>leprosy and bone survival and recovery aspects. Similarly, some of the main diagnostic</p><p>criteria for leprosy involve resorption of the anterior nasal spine, rounding of the margins of</p><p>the nasal aperture, perforation of the hard palate and ante-mortem loss of the central upper</p><p>incisors (Møller-Christensen, 1961). In archaeological assemblages, one or more of these</p><p>indicative features is frequently damaged by taphonomic processes.</p><p>The survival of bones with pathological lesions will vary depending on the nature and</p><p>severity of the specific condition. Bone is more susceptible to decomposition and post-</p><p>mortem damage in the case of a predominantly lytic process (e.g. osteomalacia, Paget’s</p><p>How Representative Are Human Skeletal Assemblages for Population Analysis? 33</p><p>Figure 2.1 Lumbar vertebrae of a 10th-century AD adult male from Turkey with numerous non-</p><p>sclerotic punched-out lytic lesions caused by multiple myeloma</p><p>disease) than in a predominantly blastic process (e.g. osteophytosis, DISH). An example of</p><p>the effect of metastatic lytic bone lesions on bone completeness and preservation is illustrated</p><p>in Figure 2.1.</p><p>In recent years, more researchers have considered taphonomic processes as an important</p><p>aspect of archaeological and forensic anthropological analysis (see contributions in Haglund</p><p>and Sorg (1997, 2002), Nicholson (2001) and Hunter and Cox (2005). Dry and wet sieving</p><p>of soil remaining in the grave after the skeleton has been lifted is now standard practice. It</p><p>facilitates recovery of small bones and other remains, such as calcified cysts, kidney stones,</p><p>calcified blood-vessel walls, lymph nodes, pleural plaques, loose teeth or infant bones (Wells</p><p>and Dallas, 1976; Morris and Rodgers, 1989; Baud and Kramer, 1991; Baker et al., 2005).</p><p>Subadult Underrepresentation</p><p>A major concern in cemetery excavations is the possible underrepresentation of young</p><p>individuals (Angel, 1969; Acsádi and Nemeskéri, 1970; Weiss, 1973a; Johnston and Zim-</p><p>mer, 1989). Guy et al. (1997) compared the number of infants aged 0–1 years from Hungarian</p><p>archaeological cemeteries (10th–12th centuries AD), as a proportion of every 1000 burials,</p><p>to figures from 18th–20th century burial registers from Europe. They noted an underrepre-</p><p>sentation of infants in the archaeological samples and argue that this cannot be explained by</p><p>taphonomic factors, but is more likely due to a mixture of factors, including type of burial</p><p>and associated burial practices, and archaeological recovery strategies.</p><p>The underrepresentation of infants or any subadult age cohort should be assessed on a</p><p>case-by-case basis. Acsádi and Nemeskéri (1970) noted that, in Hungarian medieval material,</p><p>infants were underrepresented. This skeletal material was hand recovered, so underrepre-</p><p>sentation of infants might be due to poor recovery of perinatal/infant bones rather than</p><p>34 Advances in Human Palaeopathology</p><p>taphonomic factors. Saunders et al. (2002) examined age distributions of subadults and</p><p>adults of the 19th century AD skeletal population from the St Thomas Anglican church,</p><p>Belleville, Ontario, and assessed the skeletal age-at-death pattern against available parish</p><p>records. They reported a general excess of infants and a</p><p>deficiency of elderly adults in the</p><p>skeletal sample when compared with the registers. The difference between the parish records</p><p>and the skeletal representation is particularly pronounced for infants aged 0–1 years and for</p><p>adults aged >60 years. The underrepresentation of older adults may be because these are</p><p>often classified osteologically as adults of indeterminate age and, hence, are excluded from</p><p>palaeodemographic analysis. The high proportion of infants in the skeletal sample might</p><p>be due to the use of sieving to recover small bones (Saunders et al., 2002); and that the</p><p>numbers were higher than expected from parish records may indicate that burials of infants</p><p>were sometimes not recorded in the burial registers. Mays (1993) examined infant age at</p><p>death in several Romano-British sites over a wide variety of soil conditions and found that</p><p>skeletons of perinatal age are often better preserved than those of older infants. Several</p><p>studies of prehistoric skeletal samples from Italy indicate that subadult remains are under-</p><p>represented in cemeteries associated with settlements, whereas they are overrepresented in</p><p>mortuary deposits in caves (Skeates, 1991, 1997; Robb, 1994).</p><p>Pseudopathology</p><p>Pseudopathology refers to post-mortem skeletal changes that may be mistakenly diagnosed</p><p>as ante-mortem pathological conditions. Soil may be responsible for warping of bones and</p><p>erosion of the bone cortex, which can be mistaken for specific pathological conditions, like</p><p>rickets or trauma, and cribra orbitalia or periostitis respectively (Wells, 1967, Ubelaker,</p><p>1991; Buikstra and Ubelaker, 1994). However, the most prevalent type of pseudopathology</p><p>is post-mortem fractures. These may be caused by damage from excavation tools, pressure</p><p>from overlying soil, and rough handling of excavated remains (Aufderheide and Rodríguez-</p><p>Martín, 1998). Perimortem fractures need to be distinguished from post-mortem damage.</p><p>The best way to differentiate these is to examine variations in bone colour at the frac-</p><p>ture sites. Lighter ‘fresh’ colour indicates recent post-mortem trauma, whereas regions of</p><p>perimortem trauma should be of the same colour as the rest of the bones (Ubelaker and</p><p>Adams, 1995; Bennike, Chapter 14). However, post-depositional breaks occurring during</p><p>the interval between deposition and recovery may also leave weathered edges; hence, differ-</p><p>entiation between perimortem and post-mortem fractures is not always straightforward. It is</p><p>also important to investigate the direction of the force that operated on the bone in relation</p><p>to the anatomical position of the skeleton in order to assess whether the given blows could</p><p>have been applied during life. For example, in a study of a skeleton from Georgia, USA</p><p>(Ubelaker and Adams, 1995), it was noted that whereas fractures on the left tibia and fibula</p><p>were in a similar location on the diaphysis, they displayed fracture patterns which suggest the</p><p>forces that produced them came from nearly opposite directions. Had the trauma occurred</p><p>perimortem, with the bones still in articulation, the pattern would imply the occurrence of</p><p>two separate traumatic events originating from opposite directions, but involving the same</p><p>area of the lower leg. This is very unlikely; hence, it seemed more likely that this trauma</p><p>pattern was sustained post-mortem, a suggestion supported by the clean edges of the breaks.</p><p>Rounded perforations of bones caused by various post-mortem processes may be mis-</p><p>taken for pathological or traumatic lesions. For example, two nearly symmetric perforations</p><p>observed on the fifth metatarsals of a female skeleton from a Byzantine tomb from Ramat</p><p>How Representative Are Human Skeletal Assemblages for Population Analysis? 35</p><p>Figure 2.2 Fifth metatarsals showing post-depositional perforations from the Byzantine site of Ramat</p><p>Handiv, Israel</p><p>Hanadiv, Israel, could be mistakenly attributed to ante-mortem trauma (Figure 2.2), and</p><p>lesions of this type have in the past been misdiagnosed as pathology (Elliot-Smith, 1908).</p><p>Rodents and carnivores often cause gnawing damage to bones (Figure 2.3) that may poten-</p><p>tially be mistaken for trauma or cannibalism (Haglund and Sorg, 2002). Bone gnawing by</p><p>rodents can be distinguished from that of carnivores by the characteristic parallel series of</p><p>furrows created by the incisors (Haglund, 1992).</p><p>Burial Practices</p><p>Individuals in cemeteries are often buried in specific zones according to age, social status,</p><p>sex, and other aspects. Sex- and age-specific zones are known from excavations of Swedish</p><p>medieval Christian cemeteries (Gejvall, 1960; Kjellström et al., 2005). At some Christian</p><p>cemeteries in England, spatial clusters of infant burials have been identified; for example, at</p><p>medieval Wharram Percy, England, a concentration of infant burials was found close by the</p><p>north wall of the church (Mays, 1997). Similarly, Boddington (1987) reports age-specific</p><p>spatial clusters in the early medieval English cemetery at Raunds.</p><p>Segments of the population, such as social outcasts or specific age groups, may be accorded</p><p>different burial rites from the rest of the population (Barley, 1995; Parker-Pearson, 1999)</p><p>and, therefore, may be missing or underrepresented in skeletal samples. An example is the</p><p>Irish cilliní. This was a special resting place (e.g. deserted churches and graveyards, ancient</p><p>megalithic tombs and secular earthworks, sea or lake shores) used from early Christian times</p><p>to the 20th century for stillborn babies, unbaptized infants and some other members of society</p><p>36 Advances in Human Palaeopathology</p><p>Figure 2.3 Carnivore tooth puncture marks on ilium. Reproduced with permission from Dr. C.</p><p>Rodríguez–Martín, Instituto Canario de Bioantropalogía</p><p>(e.g. mentally retarded, strangers, the shipwrecked, criminals, famine victims, suicides), who</p><p>were considered unsuitable for burial in consecrated ground (Murphy and McNeil, 1993;</p><p>Hurl and Murphy, 1996; Donnelly et al., 1999). An interesting case of social exclusion is an</p><p>adult skeleton recovered from an Iron Age well in Athens (Little and Papadopoulos, 1998).</p><p>The middle-aged male has multiple healed fractures to the skull and the vertebral column.</p><p>The authors suggest that the head injuries probably caused permanent neurological defects,</p><p>and that this may be the explanation for the unusual burial treatment of this individual.</p><p>Exclusion from the communal burial ground is also known in the case of individuals</p><p>who suffered from leprosy. Written sources indicate that, from the 13th century AD, lepers</p><p>in Europe were generally buried in cemeteries associated with leprosaria rather than being</p><p>taken back to their place of origin (Roberts et al., 2002). By contrast, leprosy cases in the</p><p>Avar period (8th–9th centuries AD) in Hungary and Lower Austria were buried in the same</p><p>cemetery as the rest of the local population (Pinhasi et al., unpublished data).</p><p>Analysis of the age and sex distribution of a skeletal sample may provide indirect evidence</p><p>for a catastrophic (hence unnatural) assemblage that may be the outcome of a single event.</p><p>The age distribution of skeletal assemblages from catastrophic episodes often resembles the</p><p>living age distribution, with greater numbers of older children, adolescents, and young adults</p><p>than in mortality profiles typical of most archaeological cemeteries (Paine, 2000). Hence,</p><p>a mortality profile of this type suggests deaths resulted from a catastrophe of some sort,</p><p>perhaps due to violence, natural disaster or epidemic.</p><p>MORBIDITY STATUS</p><p>Over the last 15 years there has been much debate regarding the interpretation of prevalence</p><p>data in palaeopathology. In particular, a concept termed the ‘osteological paradox’ by Wood</p><p>et al. (1992) has been influential. Wood et al. (1992) point out that a morbidity</p><p>profile</p><p>How Representative Are Human Skeletal Assemblages for Population Analysis? 37</p><p>reconstructed from a skeletal sample, since it represents non-survivors, may not properly</p><p>reflect the morbidity profile of the once living population from which it derived. The reason</p><p>is the possibility that selective mortality and heterogeneity in risk of death may confound</p><p>the interpretation of skeletal disease prevalence. A high prevalence of skeletal lesions may</p><p>indicate not an unhealthy population but one in which disease was regularly survived for</p><p>long enough to cause bone changes. By contrast, a population showing few skeletal lesions</p><p>may do so not because they were generally healthy, but because their resistance to disease</p><p>was too poor for them regularly to survive it long enough for bone lesions to develop.</p><p>Such interpretations were held by Wood et al. (1992) to be rather counter-intuitive and to</p><p>run counter to ‘conventional’ palaeopathological interpretations, which equate low skeletal</p><p>lesion prevalence with good health and high bone lesion prevalence with poor health – hence</p><p>the osteological ‘paradox’.</p><p>Since the Wood et al. (1992) article, the concept of the osteological paradox and its</p><p>implications have been hotly debated (references in Milner et al. (2000)), but the extent to</p><p>which it actually undermines conventional palaeopathological interpretation remains unclear.</p><p>For example, a recent examination of stress in relation to growth of subadults from two</p><p>medieval skeletal assemblages from Denmark – a disadvantaged group from a leprosarium</p><p>in Næstved and a relatively privileged medieval sample from Æbelholt – showed that the</p><p>subadults from Næstved displayed higher frequency and severity of nutritional and disease</p><p>stress (Bennike et al., 2005). This study, and others like it (e.g. Cohen, 1997), suggest</p><p>that in some cases the morbidity status of an archaeological sample is a valid indication</p><p>of the actual health profile of the original population. Currently, the debate regarding the</p><p>osteological paradox has yet to be resolved. Perhaps the best we can do is to be conscious</p><p>that different interpretations of prevalence data are possible, and to attempt to use supporting</p><p>data in order to advance some interpretations at the expense of others, rather than attempting</p><p>to interpret prevalence data in isolation from other evidence. For example, analysis of the</p><p>palaeodemographic profile of the population(s) under study and, in the case of infectious</p><p>lesions, recording whether lesions were active at death or healed (Mays, 1997) may help</p><p>to elucidate the extent to which the osteological paradox is likely to apply to the particular</p><p>material under study.</p><p>PALAEODEMOGRAPHY</p><p>Palaeodemographic research published during the 1970s compared the age-at-death profiles</p><p>and life tables of prehistoric skeletal samples in order to reconstruct the profiles in the original</p><p>populations from which they were the physical remains (Acsádi and Nemeskéri, 1970; Weiss,</p><p>1973a,b; Klepinger, 1979). However, it soon became evident that there are difficult problems</p><p>of ascertainment of age of specimens and there is bias in the skeletal deposition of certain</p><p>age groups (Weiss, 1976). During the next 10–15 years it became evident that these major</p><p>methodological concerns could not be ignored (Bocquet-Appel and Masset, 1982) and, in</p><p>fact, in most cases they have not yet been overcome. In the following sections, we selectively</p><p>focus only on the aspects that have a direct effect on palaeopathological analysis rather than</p><p>on current issues in the field of palaeodemography (for the latter, see various contributions</p><p>in Hoppa and Vaupel (2002)).</p><p>38 Advances in Human Palaeopathology</p><p>Age-at-Death Tables and the Health Status of Past Populations</p><p>Calculation of age-at-death mortality tables of past populations is of particular interest to any</p><p>population-based palaeopathological analysis. These require the calculation of crude mortality</p><p>rates for skeletal populations (Jackes, 1992). The mortality profile of a past population can be</p><p>interpreted in relation to the prevalence of various specific and non-specific stress indicators</p><p>in order to derive a ‘health index’ for the specific population (Steckel et al., 2002). A major</p><p>source of bias in the calculation of age-at-death categories is the poor accuracy and high error</p><p>ranges associated with many ageing methods, and these are particularly severe in the case</p><p>of older adults (Cox, 2000). In the last two decades, palaeodemographic research has moved</p><p>from the calculation of mortality tables and life tables to the use of hazards models and</p><p>maximum likelihood estimators (Wright and Yoder, 2003). Such studies assess the nature of</p><p>the demographic structure of the skeletal sample (Milner et al., 2000; Hoppa and Vaupel,</p><p>2002). However, none of the statistical manipulations can overcome the inherent problems</p><p>with adult skeletal age indicators.</p><p>The Application of Age Adjustment</p><p>Age at death is clearly important for bioarchaeologists who wish to assess the age-specific</p><p>health profile of the population from which the study sample is drawn. However, many</p><p>bioarchaeologists are only concerned with the analysis of the general prevalence of given</p><p>pathological condition in skeletal samples. Even in this latter case, it nevertheless remains</p><p>important to control for the effects of age: different diseases strike different age groups with</p><p>different frequencies; and bony pathologies generally accumulate with age, so that, in skele-</p><p>tal series, older age cohorts tend to show greater lesion prevalences. Direct age adjustment</p><p>involves the use of prevalence parameters obtained from a reference or hypothetical ‘stan-</p><p>dard’ population that should be divided into the same age categories as the archaeological</p><p>population under study. The researcher may use clinical epidemiological data on a population</p><p>that they believe to be the most appropriate standard for their archaeological material. Alter-</p><p>natively, they may combine age-specific prevalence taken from previous bioarchaeological</p><p>studies and use that as their standard. By using a single standard population one eliminates</p><p>the possibility that observed differences between the archaeological samples are the result of</p><p>age differences in the populations (Gordis, 2000). By applying each age-specific prevalence</p><p>rate to the population in each age group of the standard population it is then possible to</p><p>calculate the expected rate for each age group and to standardize the rates in each new</p><p>archaeological population according to these rates (Gordis, 2000: 52–54).</p><p>Application of Standardized Mortality Ratio to the Study of Gout: A Case Study</p><p>Another approach is indirect age adjustment by applying standardized mortality ratios (SMRs;</p><p>Gordis, 2000: 54). The SMR is the observed prevalence of a disease in the archaeological</p><p>sample divided by the expected prevalence of the disease in the base population. SMRs are</p><p>then calculated for each age group and can be compared for other archaeological samples.</p><p>An SMR = 1 will indicate that the prevalence in the archaeological population is identical to</p><p>the one of the standard population, whereas SMR > 1 indicates a higher prevalence. Again,</p><p>by addressing prevalence by age group it is possible to avoid at least some of the more crude</p><p>palaeodemographic biases and to derive comparable figures.</p><p>An unpublished study of joint disease in an early medieval skeletal sample from the site of</p><p>Zwölfaxing, Lower Austria, revealed a high prevalence of gout. Of a total of 130 skeletons</p><p>How Representative Are Human Skeletal Assemblages for Population Analysis?</p><p>the increasing use of medical imaging techniques, microscopic examination of</p><p>lesions and biomolecular analyses has been a major aid to the description and/or diagnosis</p><p>of disease in human remains. In addition, more workers are integrating the study of spe-</p><p>cific or non-specific disease with aspects such as growth and are examining associations</p><p>between the occurrence of different types of lesion or disease. These studies aid the inter-</p><p>pretation of skeletal disease. It was in the light of these developments that we conceived the</p><p>current volume.</p><p>x Preface</p><p>We made the decision to concentrate mainly (but not exclusively) on skeletal remains</p><p>rather than preserved soft tissue. This is simply because in most instances the skeleton is</p><p>all that survives. We have organized this volume into two parts. Part 1 deals with analytical</p><p>issues in palaeopathology. The first contribution, by Gordon Turner-Walker, deals with the</p><p>diagenesis of buried skeletal tissues. He describes the changes wrought by chemical and</p><p>microbial agents in the organic and inorganic components of skeletal tissues. He considers</p><p>some of the determinants of the rate of diagenesis; chief among these is the availability</p><p>of water in the burial environment, together with its pH and the presence or otherwise</p><p>of dissolved ionic species. As Turner-Walker points out, a sound understanding of post-</p><p>depositional changes to hard tissues is essential when attempting to interpret pathological</p><p>conditions in skeletons. Developing this theme in Chapter 2, Pinhasi and Bourbou discuss</p><p>how skeletal survival, as well as other factors such as excavation methods and ancient burial</p><p>practices, may bias a skeletal sample and, hence, complicate the interpretation of disease at</p><p>a population level. They also emphasize the importance of controlling for age at death in</p><p>population studies in palaeopathology and provide a case study to illustrate one approach</p><p>to this.</p><p>The third contribution, by Pinhasi and Turner, discusses some analytical approaches to</p><p>the quantitative study of disease frequency in palaeopopulations: palaeoepidemiology. They</p><p>discuss key paleoepidemiological concepts and provide hypothetical examples to illustrate</p><p>the application of some of these concepts to skeletal data.</p><p>Chapters 4–8 focus on techniques for examining pathological changes in ancient human</p><p>remains. Anne Grauer discusses macroscopic data collection in skeletal palaeopathology in</p><p>Chapter 4. She notes that, despite the advent of technologically advanced techniques, gross</p><p>visual examination of specimens remains the foundation of palaeopathological investigation.</p><p>She discusses the historical background of study and evaluates attempts toward standardizing</p><p>terminology and data collection. She identifies a number of problems and issues germane to</p><p>this area, and offers suggestions as to how these might be resolved.</p><p>The next four chapters concentrate on the application of medical imaging and histolog-</p><p>ical and biomolecular techniques in palaeopathology. Radiography is the oldest and still</p><p>the most frequently used augment to visual examination of specimens in palaeopathology.</p><p>In Chapter 5, Simon Mays discusses plain-film radiography, quantitation of cortical bone</p><p>thickness from radiographs (radiogrammetry) and various radiological methods of measur-</p><p>ing bone density. The principles of these techniques are explained and their contribution to</p><p>palaeopathological description and diagnosis discussed. In Chapter 6, Lynnerup discusses</p><p>the imaging by CT of hard and soft tissues in mummies and bog-bodies. An important focus</p><p>of both Mays’ and Lynnerup’s contributions is on the issues raised and problems encoun-</p><p>tered when applying imaging techniques developed for medical application to ancient human</p><p>remains.</p><p>Turner-Walker and Mays discuss the microscopic study of disease in skeletal remains</p><p>in Chapter 7. Focusing principally on light and electron microscopy, they discuss sample</p><p>preparation techniques, the effects of diagenesis on the histological appearance of buried</p><p>bone and the role of microscopic studies of skeletal lesions in palaeopathology. Histological</p><p>structures may be studied in a qualitative manner and any abnormalities noted may assist</p><p>in diagnosis of disease. They may also be studied quantitatively (histomorphometry) to</p><p>investigate the extent of progressive metabolic conditions such as osteoporosis or to estimate</p><p>age at death. The contribution concludes with a discussion of the potential role of newer</p><p>microscopic techniques.</p><p>Preface xi</p><p>Donoghue covers the fast-moving field of biomolecular study of ancient infectious disease</p><p>in Chapter 8. Focusing principally on the study of DNA, the degradation and authentication of</p><p>ancient DNA are discussed and the contribution of biomolecular study to the palaeopathology</p><p>of various specific infections is evaluated. The potentially important contribution to be</p><p>made by ancient DNA studies to our understanding of the evolution of disease-causing</p><p>microorganisms is also considered.</p><p>The systematic gathering of large amounts of osteological data raises questions of how</p><p>best to organize these data and make them available to the wider scholastic community. In</p><p>the last chapter in this section, Bill White reviews issues concerned with the establishment</p><p>and maintenance of computerized databases of human remains. He identifies several differ-</p><p>ent types of database, ranging from simple inventories to help researchers locate archived</p><p>collections, to those which include considerable osteological detail with the intent that schol-</p><p>ars use the data directly in their research. He presents an evaluation of the strengths and</p><p>weaknesses of some of the major extant databases of human remains, and discusses possible</p><p>future directions.</p><p>In Part 2 we concentrate on the diagnosis and interpretation of various classes of disease.</p><p>We have not attempted to be comprehensive in our coverage of the different categories</p><p>of disease, but rather have endeavoured to select those where recent advances have made</p><p>themselves most felt.</p><p>Don Ortner discusses diagnostic issues in the evaluation of skeletal infectious disease in</p><p>Chapter 10, with an emphasis on macroscopic study. He reviews the major infectious diseases</p><p>that can be identified on the skeleton and provides an extensive photographic illustration of</p><p>lesions, and emphasizes the need for careful description of lesions and rigorous differential</p><p>diagnosis.</p><p>In his chapter on metabolic bone disease, Simon Mays reviews the pathophysiology,</p><p>palaeopathological diagnosis and interpretation of vitamin D deficiency, vitamin C defi-</p><p>ciency, osteoporosis and Paget’s disease of bone. Most studies of the former two conditions</p><p>have been conducted in order to investigate biocultural questions concerning living condi-</p><p>tions and diet. Palaeopopulation studies of the latter two have been mainly orientated toward</p><p>increasing our understanding of the risk factors for these poorly understood conditions which</p><p>continue to be important contributors to morbidity and mortality today.</p><p>A review of tumours and tumour like processes is provided by Don Brothwell in Chap-</p><p>ter 12. He gives a wide-ranging review of the archaeological evidence for both benign and</p><p>malignant tumours in hard and soft tissues, and considers the potential for relating changes</p><p>in frequency through time to environmental or cultural factors. The need for collation of</p><p>widely scattered data and for rigorous statistical analysis is emphasized.</p><p>In his chapter on dental disease, Alan Ogden concentrates on some key recent develop-</p><p>ments in our understanding. He describes a new type of dental enamel hypoplasia, discusses</p><p>diagnostic criteria for periodontal disease and presents a simple scoring system. He then</p><p>proceeds to</p><p>39</p><p>over 16 years of age (77 males and 53 females), 14.2 % were diagnosed with gout (using</p><p>a combination of macroscopic, microscopic and radiological methods). The sex-specific</p><p>prevalence was 15.6 % for males and 5.7 % for females.</p><p>The prevalence of gout in modern populations is shown in Table 2.1. The highest modern</p><p>prevalence reported is 10.2 % among Maori males (Prior et al., 1966). The prevalence</p><p>in the archaeological material from Zwölfaxing exceeds even this, but it is possible that</p><p>the Zwölfaxing prevalence was confounded by the age structure of the sample. In order to</p><p>investigate this, the SMR was calculated for the aged and sexed samples. The total population</p><p>was divided to broad age categories: 16–30, 31–50, and > 50 years. Next, a standard reference</p><p>population was selected from the literature. A standard reference population should closely</p><p>resemble genetically the archaeological sample under investigation. In the current case, no</p><p>adequate data on gout prevalence are available for modern Austria. In addition, the choice of</p><p>standard was constrained by the fact that many epidemiological studies do not present their</p><p>data in ways that allow the necessary comparisons. The standard reference data chosen to</p><p>calculate the SMRs were from modern England (Harris et al., 1995). They comprise 300 376</p><p>individuals (146 973 males, 153 403 females) of which 2865 were diagnosed of having gout.</p><p>The prevalence in Zwölfaxing was standardized following the above-described method. The</p><p>results are presented in Table 2.2. None of the archaeological cases diagnosed with gout was</p><p>younger than 30 years of age; hence, the SMR value of both males and females is nil for this</p><p>age category. With the age category of 31–50 years, the Zwölfaxing SMR is 10-fold higher</p><p>than in the reference British population for males and 66-fold higher for females. For the age</p><p>category of >50 years the SMR is 6-fold higher for males and 20-fold higher for females.</p><p>The results confirm the high prevalence figures obtained before age adjustment. However,</p><p>owing to the small sample size, the SMR for Zwölfaxing females should be interpreted</p><p>Table 2.1 A summary table of the prevalence of gout among various modern populations</p><p>Sample/location</p><p>Period</p><p>of study Sex Sample size Age</p><p>Gout</p><p>prevalence</p><p>(%) Reference</p><p>Framingham,</p><p>USA</p><p>1950s–</p><p>1960s</p><p>Both 5 127 30–59 0.25 (at age 44) Hall et al.</p><p>(1967)</p><p>England 1991 Both 300 376 − 1.64 (ÿ)</p><p>0.95 (þ)</p><p>Harris et al.</p><p>(1995)</p><p>Paris, France 1980 Both 4 663 35–44 1–2 Zalokar</p><p>et al. (1981)</p><p>Greece 1996–</p><p>1999</p><p>Both 8 547 ≥19 0.47 Andrianakos</p><p>et al. (2003)</p><p>New Zealand,</p><p>Maori</p><p>1960s Both 370 (ÿ) 20–>70 10.2 (þ) Prior et al.</p><p>(1966)</p><p>385 (ÿ) 1.8 (þ)</p><p>Cook Islands,</p><p>Rarotonga</p><p>and Pukapuka</p><p>1964 ÿ 431 Adult 2.4 (Rarotogan)</p><p>5.3 (Pukapuka)</p><p>Prior et al.</p><p>(1966)</p><p>China, Shanghai 1997–</p><p>1998</p><p>Both 3 190 (ÿ) ≥15 0.63 (þ)</p><p>0.059 (þ)</p><p>Dai et al.</p><p>(2003)3 394 (þ)</p><p>40 Advances in Human Palaeopathology</p><p>Table 2.2 Computation of SMR for gout in a medieval Austrian population from Zwölfaxing using</p><p>indirect age adjustment based on 1991 epidemiological data on the age- and sex-specific prevalence in</p><p>Englanda</p><p>Modern England Zwölfaxing</p><p>Prevalence rate (per 1000) Total Observed cases SMR</p><p>Age (years) Male Female Male Female Male Female Male Female</p><p>16–30 0.96 0.09 15 13 0 0 0.00 0�00</p><p>31–50 18.14 1.12 44 27 8 2 10.02 66�14</p><p>>50 35.48 3.84 18 13 4 1 6.26 20�03</p><p>a Modern data from Harris et al. (1995).</p><p>with caution; the more reliable figures are those obtained for the males. Of course, the</p><p>methods used for the diagnosis of gout in skeletal samples differ to a fair extent from</p><p>those used to diagnose the disease in modern clinical studies. The application of rigorous</p><p>palaeopathological methodology for differential diagnosis, together with taphonomic aspects</p><p>of bone representation and degree of preservation, will tend to result in underestimation of</p><p>the true prevalence of the condition in the archaeological population, supporting the idea</p><p>that the greater frequency of gout in the Zwölfaxing population than the modern reference</p><p>population is a robust result. All archaeological specimens with gouty lesions were at the</p><p>chronic tophaceous disease stage, which in modern clinical cases develops, on average, 10</p><p>or more years after the onset of acute intercritical gout (Edwards, 2001). Therefore, it seems</p><p>that the onset of gout in some of the specimens studied from the necropolis population was</p><p>during (or prior to) the third decade of life. The age distribution suggests that the age of</p><p>onset of gout in this population is, on average, much earlier than in the modern reference</p><p>population, although some cases of early onset (early 20s) of chronic gout in males are</p><p>reported in the medical literature (Resnick et al., 1975).</p><p>CONCLUSIONS</p><p>Factors that may affect the nature of an archaeological sample can be broadly grouped in</p><p>two categories: the first, uncontrolled by the researcher, includes biases caused by burial</p><p>practices, differential bone survival, and age- and sex-specific biases in the representativeness</p><p>of the excavated sample; the second, controlled by the researcher, includes the excavation</p><p>methods and recovery strategies, the application of specific age-correction algorithms, and</p><p>the like. Palaeopathologists should not be discouraged by these biases. The future lies in a</p><p>paradigmatic shift towards a population-based approach in palaeopathology. As Wright and</p><p>Yoder (2003: 49) assert:</p><p>Accurate age-at-death estimates are critical for interpreting the impact of pathological</p><p>lesions on well-being at the population level. Analysis of pathological lesion abundance</p><p>by age-at-death cohorts may be a useful approach for evaluating the significance of lesions</p><p>in terms of morbidity and mortality.</p><p>How Representative Are Human Skeletal Assemblages for Population Analysis? 41</p><p>Future work on palaeopathological and palaeoepidemiological aspects should, therefore,</p><p>incorporate and synthesize palaeodemographic factors. Such an approach will potentially</p><p>lead to a better understanding of the complex relationship between mortality and morbidity</p><p>profiles in skeletal samples of past populations.</p><p>ACKNOWLEDGEMENTS</p><p>We thank Professor Yizhar Hirschfeld, Department of Archeology, Hebrew University,</p><p>Jerusalem, Israel, for the permission to publish Figure 2.2.</p><p>REFERENCES</p><p>Acsádi G, Nemeskéri J. 1970. History of Human Life Span and Mortality. Akadémiai Kiadó: Budapest.</p><p>Alesan A, Malgosa A, Simo C. 1999. Looking into the demography of an Iron Age population in the</p><p>western Mediterranean (1): mortality. Am J Phys Anthropol 110: 285–301.</p><p>Andrianakos A, Trontzas P, Christoyannis F, Dantis P, Voudouris C, Georgountzos A, Kaziolas G,</p><p>Vafiadou E, Pantelidou K, Karamitsos D, Kontelis L, Krachtis P, Nikolia Z, Kaskani E, Tavaniotou</p><p>E, Antoniades C, Karanikolas G, Kontoyanni A. 2003. Prevalence of rheumatic disease in Greece:</p><p>a cross-sectional population based epidemiological study. J Rheumatol 30: 1589–1601.</p><p>Angel JL. 1969. The bases of paleodemography. Am J Phys Anthropol 30: 427–438.</p><p>Aufderheide AC, Rodríguez-Martín RM. 1998. The Cambridge Encyclopaedia of Human Palaeopathol-</p><p>ogy. Cambridge University Press: Cambridge.</p><p>Baker BJ, Dupras TL, Tocheri MW. 2005. The Osteology of Infants and Children. Texas A&M</p><p>University Press: College Station, TX.</p><p>Barley N. 1995. Dancing on the Grave. Encounters with the Dead. John Murray: London.</p><p>Baud CA, Kramer C. 1991. Soft tissue calcifications in paleopathology. In Human Paleopathology.</p><p>Current Syntheses and Future Options, Ortner DJ, Aufderheide AC (eds). Smithsonian Institution</p><p>Press: Washington, DC; 87–89.</p><p>Bennike P, Lewis ME, Schutkowski H, Valentin F. 2005. Comparison of child morbidity in two</p><p>contrasting medieval cemeteries from Denmark. Am J Phys Anthropol</p><p>128: 734–746</p><p>Bocquet-Appel JP, Masset C. 1982. Farewell to paleodemography. J Hum Evol 11: 321–333.</p><p>Boddington A. 1987. Raunds, Northamptonshire: analysis of a country churchyard. World Archaeol</p><p>18: 411–425.</p><p>Bourbou C. 1999. Digging up burials: excavation techniques, handling and transportation of bones.</p><p>Archaeol Art 67: 55–61.</p><p>Buikstra JE, Ubelaker, DH. 1994. Standards for Data Collection from Human Skeletal Remains.</p><p>Arkansas Archaeological Survey, Research Seminar Series 44. Arkansas Archaeological Survey:</p><p>Fayetteville,AR.</p><p>Caffell A, Roberts CA, Janaway RC, Wilson A. 2001. Pressures on osteological collections-the impor-</p><p>tance of damage limitation. In Human Remains: Conservation, Retrieval and Analysis, Williams E</p><p>(ed.). British Archaeological Reports International Series 934. Archaeopress: Oxford; 187–197.</p><p>Cohen MN. 1997. Does paleopathology measure community health? A rebuttal of ‘the osteologi-</p><p>cal paradox’ and its implication for world history. In Integrating Archaeological Demography:</p><p>Multidisciplinary Approaches to Prehistoric Population, Paine RR (ed.). Centre for Archaeolog-</p><p>ical Investigations Occasional Paper No. 24. Southern Illinois University Press: Carbondale, IL;</p><p>242–260.</p><p>Cox M. 2000. Ageing adults from the skeleton. In Human Osteology in Archaeology and Forensic</p><p>Science, Cox M, Mays S (eds). Greenwich Medical Media: London; 61–82.</p><p>42 Advances in Human Palaeopathology</p><p>Currey JD, Butler G. 1975. The mechanical properties of bone tissue in children. J Bone Jt Surg 57:</p><p>810–814.</p><p>Dai SM, Han XH, Zhao DB, Shi YQ, Meng JM. 2003. Prevalence of rheumatic symptoms, rheumatoid</p><p>arthritis, ankylosing spondylitis, and gout in Shanghai, China: a COPCORD study. J Rheumatol 30:</p><p>2245–2251.</p><p>Donnelly S, Donnelly C, Murphy EM. 1999. The forgotten dead: the cilliní and disused burial grounds</p><p>of Ballintoy, County Antrim. Ulster J Archaeol 58: 109–113.</p><p>Edwards NL. 2001. Gout. A: Epidemiology, pathology and pathogenesis. In Primer on the Rheumatic</p><p>Diseases, 12th edition, Klipper JH (ed.). Arthritis Foundation: Altanta, GA; 307–312.</p><p>Elliot-Smith G. 1908. The alleged discovery of syphilis in prehistoric Europeans. Lancet 2: 521–524.</p><p>Gejvall N-G. 1960. Westerhus. Mediaeval Population and Church in the Light of the Skeletal Remains.</p><p>Vitterhets Historie och Antikvitets Akademien: Lund.</p><p>Gordis L. 2000. Epidemiology, 2nd edition. WB Saunders: Philadelphia.</p><p>Gordon CG, Buikstra JE. 1981. Soil pH, bone preservation and sample bias at mortuary sites. Am</p><p>Antiquity 46: 566–571.</p><p>Grupe G. 1988. Impact of the choice of bone samples on trace element data in excavated human</p><p>skeletons. J Archaeol Sci 15: 123–129.</p><p>Guy H, Masset C, Baud CA. 1997. Infant taphonomy. Int J Osteoarchaeol 7: 221–229.</p><p>Haglund WD. 1992. Contribution of rodents to postmortem artifacts of bone and soft tissue. J Forensic</p><p>Sci 37: 1459–1465.</p><p>Haglund WD, Sorg MH (eds). 1997. Forensic Taphonomy. The Postmortem Fate of Human Remains.</p><p>CRC Press: Boca Raton, FL.</p><p>Haglund WD, Sorg MH (eds). 2002. Advances in Forensic Taphonomy. Method, Theory and</p><p>Archaeological Perspectives. CRC Press: Boca Raton, FL.</p><p>Hall AP, Barry PE, Dawber TR, McNamara PM. 1967. Epidemiology of gout and hyperuricemia:</p><p>a long term population study. Am J Med 42: 27–37.</p><p>Harris CM, Lloyd DCEA, Lewis J. 1995. The prevalence and prophylaxis of gout in England. J Clin.</p><p>Epidemiol 48: 1153–1158.</p><p>Henderson J. 1987. Factors determing the state preservation of human remains. In Death, Decay and</p><p>Reconstruction. Approaches to Archaeology and Forensic Science, Boddington A, Garland AN,</p><p>Janaway RC (eds). Manchester University Press: Manchester; 43–53.</p><p>Hoppa RD. 2002. Paleodemography: looking back and thinking ahead. In Paleodemography: Age</p><p>Distributions from Skeletal Samples, Hoppa RD, Vaupel JW, (eds). Cambridge University Press:</p><p>Cambridge; 9–23.</p><p>Hoppa RD, Vaupel JW (eds). 2002. Paleodemography: Age Distributions from Skeletal Samples.</p><p>Cambridge University Press: Cambridge.</p><p>Hunter J, Cox M. 2005. Forensic Archaeology. Advances in Theory and Practice. Routledge: London.</p><p>Hurl DP, Murphy EM. 1996. Life and death in a County Antrim tower house. Archaeol Ireland 10:</p><p>20–23.</p><p>Jackes M. 1992. Paleodemography: problems and techniques. In The Skeletal Biology of Past Peoples,</p><p>Saunders S, Katzenberg MA (eds). Wiley-Liss: New York; 189–224.</p><p>Johnston FE, Zimmer LO. 1989. Assessment of growth and age in the immature skeleton. In Recon-</p><p>struction of Life from the Skeleton, İşcan MY, Kennedy KAR (eds). Wiley-Liss: New York;</p><p>11–22.</p><p>Kjellström A, Tesch S, Wikström A. 2005. Inhabitants of a sacred townscape: an archaeological and</p><p>osteological analysis of skeletal remains from late Viking Age and medieval Sigtuna, Sweden. Acta</p><p>Archaeol 76: 87–110.</p><p>Klepinger LL. 1979. Paleodemography of the Vadivia III phase at Real Alto, Ecuador. Am Antiquity</p><p>44: 305–309.</p><p>Lambert JB, Vlasak SM, Tometz AC, Buikstra JE. 1982. A comparative study of the chemical analysis</p><p>of ribs and femurs in Woodland populations. Am J Phys Anthropol 59: 284–294.</p><p>How Representative Are Human Skeletal Assemblages for Population Analysis? 43</p><p>Little L, Papadopoulos K. 1998. A social outcast in early Iron Age Athens. Hesperia 67: 375–404.</p><p>Mays SA. 1992. Taphonomic factors in a human skeletal assemblage. Circaea 9: 54–58.</p><p>Mays SA. 1993. Infanticide in Roman Britain. Antiquity 67: 883–888.</p><p>Mays SA. 1997. Life and death in a mediaeval village. In Death and Burial in Mediaeval Europe:</p><p>Papers of the ‘Mediaeval Europe Brugge 1997’ Conference, volume 2, De Boe G, Verhaege F</p><p>(eds). Instituut voor het Archeologisch Patrimonium Rapporten 2. Instituut voor het Archeologisch</p><p>Patrimonium: Zellik; 121–125.</p><p>Mays SA. 1998. The Archaeology of Human Bones. Routledge: London.</p><p>Milner GR, Wood JW, Boldsen JL. 2000. Paleodemography. In Biological Anthropology of the Human</p><p>Skeleton, Katzenberg MA, Saunders SR (eds). Wiley-Liss: New York; 467–497.</p><p>Møller-Christensen V. 1961. Bone Changes in Leprosy. Munksgaard: Copenhagen.</p><p>Morris AG, Rodgers AL. 1989. A probable case of prehistoric kidney stone disease from the Northern</p><p>Cape Province, South Africa. Am J Phys Anthropol 79: 521–527.</p><p>Murphy EM, McNeil TE. 1993. Human remains excavated at Doonbought Fort, County Antrim, 1969.</p><p>Ulster J Archaeol 56: 120–138.</p><p>Nicholson RA. 2001. Taphonomic investigations. In Handbook of Archaeological Science, Pollard</p><p>MA, Brothwell D (eds). Wiley: Chichester; 179–190.</p><p>Paine RP. 2000. If a population crashes in prehistory, and there is no paleoedemographer there to hear</p><p>it, does it make a sound? Am J Phys Anthropol 112: 181–190.</p><p>Parker-Pearson M. 1999. The Archaeology of Death and Burial. Sutton: Stroud.</p><p>Prior IAM, Rose BS, Harvey HPB, Davidson F. 1966. Hyperuricemia, gout, and diabetic abnormality</p><p>in Polynesian people. Lancet 1: 333–338.</p><p>Resnick D, Reinke RT, Taketa RM. 1975. Early-onset gouty arthritis. Radiology 114: 67–73.</p><p>Robb J. 1994. Burial and social reproduction in the peninsular Italian Neolithic. J Mediterr Archaeol</p><p>7: 27–71.</p><p>Roberts CA, Lewis ME, Manchester K (eds). 2002. The Past and Present of Leprosy. British</p><p>Archaeological Reports International Series S1054. Archaeopress: Oxford.</p><p>Saunders SR, Herring A, Sawchuk L, Boyce G, Hoppa R, Klepp S. 2002. The health of the middle</p><p>class: the St. Thomas Anglican church cemetery project. In The Backbone of History: Health and</p><p>Nutrition in the Western Hemisphere, Steckel RH, Rose JC (eds). Cambridge University Press:</p><p>Cambridge; 130–161.</p><p>Skeates R. 1991. Caves, cult and children in Neolithic Abruzzo, central Italy. In Sacred and Profane,</p><p>Garwood P, Jennings D, Skeates R, Toms J (eds). Oxford University Centre for Archaeology:</p><p>Oxford; 122–134.</p><p>Skeates R. 1997. The human use of caves in east-central Italy during the Mesolithic, Neolithic</p><p>and</p><p>Copper Age. In The Human Use of Caves, Bonsall C, Tolan-Smith C (eds). British Archaeological</p><p>Reports International Series 667. Archaeopress: Oxford; 79–86.</p><p>Specker BL, Brazerol W, Tsang RC, Levin R, Searcy J, Stechen J. 1987. Bone mineral content in</p><p>children 1 to 6 years of age. Am J Dis Child 141: 343–344.</p><p>Steckel RH, Sciulli PW, Rose JC. 2003. A health index from skeletal remains. In The Backbone of</p><p>History: Health and Nutrition in the Western Hemisphere, Steckel RH, Rose JC (eds). Cambridge</p><p>University Press: Cambridge; 61–93.</p><p>Steinbock RT. 1976. Paleopathological Diagnosis and Interpretation. Charles C Thomas: Springfield.</p><p>Ubelaker DH. 1991. Perimortem and postmortem modification of human bone. Lessons from forensic</p><p>anthropology. Anthropologie 29: 171–174.</p><p>Ubelaker DH, Adams BJ. 1995. Differentiation of perimortem and postmortem trauma using</p><p>taphonomic indicators. J Forensic Sci 40: 509–512.</p><p>Waldron T. 1987. The relative survival of the human skeleton: implications for paleopathology. In</p><p>Death, Decay and Reconstruction. Approaches to Archaeology and Forensic Science, Boddington</p><p>A, Garland AN, Janaway RC (eds). Manchester University Press: Manchester; 55–64.</p><p>Waldron T. 1994. Counting the Dead. The Epidemiology of Skeletal Populations. Wiley: Chichester.</p><p>44 Advances in Human Palaeopathology</p><p>Weiss KM. 1973a. Demographic Models for Anthropology. Memoirs of the Society for American</p><p>Archaeology No. 27. Society of American Archaeology: Washington, DC.</p><p>Weiss KM. 1973b. A method for approximating age-specific fertility in the construction of life tables</p><p>for anthropological populations. Hum Biol 45: 195–210.</p><p>Weiss KM. 1976. Demographic theory and anthropological inference. Ann Rev Anthropol 5: 351–381.</p><p>Wells C. 1967. Pseudopathology. In Diseases in Antiquity, Brothwell DR, Sandinson AT (eds). Charles</p><p>C. Thomas: Springfield; 5–19.</p><p>Wells C, Dallas C. 1976. Romano-British pathology. Antiquity 50: 53–55.</p><p>Wood JW, Milner GR, Harpending HC, Weiss KM. 1992. The osteological paradox. Problems of</p><p>inferring health from skeletal samples. Curr Anthropol 33: 343–370.</p><p>Wright LE, Yoder CJ. 2003. Recent progress in bioarchaeology: approaches to the osteological paradox.</p><p>J Archaeol Res 11: 43–70.</p><p>Zalokar J, Lellouch J, Claude JR. 1981. Serum urate and gout in 4663 young male workers. Sem Hop</p><p>57: 664–670 (in French).</p><p>3</p><p>Epidemiological Approaches</p><p>in Palaeopathology</p><p>Ron Pinhasi1 and Katy Turner2</p><p>1Department of Archaeology, University College Cork, Cork</p><p>Ireland</p><p>2Division of Epidemiology, Public Health and Primary Care,</p><p>Imperial College, Praed Street, St Mary’s Campus, London W2</p><p>1PG, UK</p><p>INTRODUCTION</p><p>Epidemiological investigation of disease involves the elucidation of the aetiology of a spe-</p><p>cific disease or group of diseases by combining epidemiological data with information from</p><p>other sources (genetics, biochemistry, microbiology), and the evaluation of consistency of</p><p>epidemiological data with aetiological hypotheses developed either clinically or experimen-</p><p>tally (Lilienfeld and Stolley, 1994). Palaeoepidemiology can be broadly defined as ‘� � � an</p><p>interdisciplinary area that aims to develop more suitable epidemiological methods, and to</p><p>apply those in current use, to the study of disease determinants in human populations in</p><p>the past’ (de Souza et al., 2003: 21). What is the relationship between palaeoepidemiology</p><p>and medical epidemiology? Or, in other words, what concepts and methods of medical</p><p>epidemiology are of use to palaeopathologists working on skeletal remains? In an attempt</p><p>toward addressing these issues, some of the basic concepts of epidemiology will be briefly</p><p>reviewed, as these are also central to palaeoepidemiological studies.</p><p>Traditionally, epidemiology was a discipline inferring causal relationships between risk</p><p>factors and disease. Pearce (2005: 9) points out that ‘� � � the key feature of epidemiologi-</p><p>cal studies is that they are quantitative (rather than qualitative) observational (rather than</p><p>experimental) studies of the determinants of disease in human populations (rather than</p><p>individuals)’. Epidemiology has a descriptive dimension that involves the identification and</p><p>documentation of disease trends, differential diagnosis of disease and injury and other related</p><p>phenomena (Rockett, 1999). Epidemiology is a multi-step process that involves what Gordis</p><p>(2000) defines as ‘epidemiologic reasoning’, a process that begins with descriptive analysis</p><p>Advances in Human Palaeopathology Edited by Ron Pinhasi and Simon Mays</p><p>© 2008 John Wiley & Sons, Ltd. ISBN: 978-0-470-03602-0</p><p>46 Advances in Human Palaeopathology</p><p>but which should also proceed to address casual relationships between disease, demographic,</p><p>social and cultural factors.</p><p>Both the descriptive and analytical sides of epidemiology are of relevance to palaeopathol-</p><p>ogists; as yet, however, most palaeopathological studies do not incorporate epidemiological</p><p>methods in their analyses. To some extent, the limited use of epidemiological methods by</p><p>palaeopathologists reflects a general scepticism as to the degree to which a given skeletal</p><p>assemblage is a reliable representative sample of the parent population. Additionally, epidemio-</p><p>logical theory and methods were developed for medical rather than palaeopathological research,</p><p>so they require some modification to make them suitable for research using skeletal material.</p><p>RESEARCH DIRECTIONS IN THE STUDY OF DISEASE</p><p>AT A POPULATION LEVEL IN PALAEOPATHOLOGY</p><p>A significant body of research exists that involves assessing overall health status in a skeletal</p><p>population using a series of disease markers. In order to produce workable quantities of data,</p><p>markers are usually chosen that generally have reasonably high rates of occurrence in skeletal</p><p>series, e.g. dental enamel hypoplasias, cribra orbitalia, dental caries, ante-mortem tooth loss,</p><p>lesions associated with skeletal infections, degenerative joint disease and trauma. For exam-</p><p>ple, in the 1980s, indicators such as these were assessed in agricultural and pre-agricultural</p><p>populations from various parts of the world in order to assess the effects of the transition</p><p>to agriculture on the health status of earlier human groups (Cohen and Armelagos, 1984;</p><p>Cohen, 1989). Building on this work, Steckel et al. (2002, 2003) combined data on different</p><p>pathological features in order to produce overall ‘health indices’ for palaeopopulations from</p><p>various geographical and temporal contexts in the Americas in order to assess their health</p><p>profiles.</p><p>Although the ‘health index’ approach is potentially a useful one, there are several issues</p><p>that are of concern. An assumption in calculating a health index is that the palaeopathological</p><p>conditions that comprise it can be combined and compared between skeletal populations.</p><p>This approach is complicated by the fact that the occurrences of the different palaeopatho-</p><p>logical conditions are, to varying extents, correlated with one another, as some factors may</p><p>be involved in the causes of more than one condition. In order to control for this from an epi-</p><p>demiological perspective, it is essential to use methods such as logistic regression in order to</p><p>detect the main uncorrelated variables, rather than simply combining pathological features to</p><p>form a single index. Many of the skeletal pathological features used to derive a health index</p><p>have multiple and incompletely understood aetiologies. They are not diseases, but simply</p><p>lesions that can be identified on bone. Therefore, the health index is a palaeopathological</p><p>construct that is not directly comparable with any medical epidemiological index. Although</p><p>the health index may provide a measure of morbidity in a past population, it cannot shed light</p><p>on patterns of particular diseases. By contrast, medical epidemiological reasoning focuses</p><p>on the study of associations between social and environmental factors and specific diseases,</p><p>rather than on health status, which is in any event a rather nebulous concept. It may be more</p><p>useful for palaeopathological study to focus, where possible, on specific diseases in order to</p><p>maintain the link between the study of disease in the past and medical research on disease</p><p>in present-day populations. By doing so, palaeoepidemiological research will not become</p><p>a detached subdiscipline with ill-defined palaeopathological ‘conditions’ and will, instead,</p><p>continue to benefit from an ongoing discourse with medical epidemiological research.</p><p>Epidemiological Approaches in Palaeopathology 47</p><p>The study of ancient DNA of pathogens is beginning to make an impact in palaeoepidemi-</p><p>ology. Whilst early palaeopathological studies of pathogen DNA focused on single cases or</p><p>small numbers of individuals, more recent work has begun to analyse for pathogen DNA in</p><p>large numbers of skeletons, opening the way toward biomolecular-based palaeoepidemiolog-</p><p>ical work. For example, Aufderheide et al. (2003) analysed ancient DNA of Trypanosoma</p><p>cruzi from soft tissue samples taken from 283 naturally desiccated mummies from Chile and</p><p>Peru. The specimens ranged in date from 7050 BC to approximately the time of the Spanish</p><p>conquest, AD 1500. Of the 283 mummies, 115 (40.6 %) were positive for T. cruzi. No</p><p>statistically significant differences in the prevalence of the pathogen (indicated by positive</p><p>test results) were noted between any of the cultural groups. There was no sex difference,</p><p>but analysis of prevalence rates by age indicated that the prevalence of the disease was</p><p>significantly higher among infants of 0–2 years of age. The transmission of Chagas’ disease</p><p>depends on the ability of the insect vector to infest the wild animals’ nests or lairs, provid-</p><p>ing opportunities for the insect’s blood meal and transmission of the infectious agent (T.</p><p>cruzi). Aufderheide et al. (2003) assert that the lack of a significant diachronic trend in the</p><p>prevalence rates of Chagas’ disease among the human populations studied suggests that the</p><p>earliest human groups that colonized the Andean coast offered the Chagas vector a physical</p><p>environment for access to a blood meal that was equivalent to the nests and lairs of various</p><p>indigenous feral (host) animals.</p><p>Palaeomicrobiology, the study of the antiquity and molecular evolution of pathogens,</p><p>most usually involves the study of DNA from modern pathogens rather than ancient DNA.</p><p>However, it is likely that, as work on large samples of skeletons becomes more common and</p><p>techniques for amplifying and studying ancient DNA improve, the study of pathogen DNA</p><p>from ancient skeletons will begin to make a significant contribution to the understanding of</p><p>the evolution and spread of microbial human pathogens (Chapter 8).</p><p>Another related research direction involves the study of major epidemiological transitions</p><p>associated with particular cultural–historical changes in human history. Barrett et al. (1998)</p><p>provide a detailed account of major epidemiological changes in human host–pathogen sys-</p><p>tems that are associated with cultural/evolutionary changes during the Palaeolithic, Neolithic</p><p>and the Industrial Revolution periods. They adopt an evolutionary historical perspective,</p><p>using an expanded framework of epidemiologic transition theory that views major changes</p><p>in host–pathogen systems as being directly related to corresponding changes in human modes</p><p>of subsistence and social organization. Barrett et al. (1998) suggest that the first epidemi-</p><p>ological transition occurred about 10 000 years ago, when the first agricultural settlements</p><p>emerged in the Near East. The transition involved a drastic increase in infectious disease</p><p>and mortality associated with changes in aggregation, social organization, domestication of</p><p>animals (and the emergence of zoonotic infections), diet and other socio-cultural aspects</p><p>of the Neolithic lifestyle. The second epidemiologic transition roughly coincided with the</p><p>Industrial Revolution in mid-19th century Europe and North America. This period involved</p><p>a marked decline in infectious disease mortality within developed countries, the disappear-</p><p>ance of infectious diseases pandemics and a rise in chronic and degenerative non-infectious</p><p>diseases. The third epidemiologic transition occurs during the last 25 years in the con-</p><p>text of globalization, global trade, changes in disease ecology and mass migrations and a</p><p>drastic increase world travel. It is associated with the appearance of numerous new dis-</p><p>eases, an increase in the incidence and prevalence of pre-existing infectious diseases, and</p><p>re-emerging drug-resistant strains of pathogens, such those responsible for tuberculosis and</p><p>syphilis.</p><p>48 Advances in Human Palaeopathology</p><p>The study of the evolution and spread of disease requires interdisciplinary collaboration</p><p>between epidemiologists, anthropologists and geneticists. The contribution of palaeopathol-</p><p>ogy to such research involves the following. First, the detection and analysis of certain</p><p>infectious diseases that leave diagnostic lesions on bone. The identification of early cases of</p><p>leprosy, brucellosis, syphilis and other diseases in antiquity will at least provide geneticists</p><p>with a preliminary age for the antiquity of diseases that leave traces on bone and also poten-</p><p>tially indicate the geographic regions in which a disease existed during past epochs. Second,</p><p>a systematic analysis of large archaeological skeletal collections from various time periods</p><p>can give information concerning changes in disease prevalence over time. A good example</p><p>is the evidence for a sharp increase in prevalence rates for leprosy during late medieval</p><p>times, which was calculated from an observed increase in the number of skeletons with</p><p>palaeopathological lesions pathognomonic of leprosy (Roberts, 2002). Third, palaeopathol-</p><p>ogists should oversee the taking of bone samples for ancient DNA to ensure that they are</p><p>from archaeological specimens with unambiguous archaeological context. Fourth, the study</p><p>of bone degradation (Chapter 1) provides information about chances of recovery of ancient</p><p>DNA sequences from a given archaeological bone sample. Fifth, sampling for the pathogen</p><p>DNA requires palaeopathological knowledge about the disease at hand. For example, prefer-</p><p>ential sites for sampling for Mycobacterium leprae are within in the nasal cavity, rather than</p><p>from lesions in the hand and foot bones, which are principally due to secondary infections.</p><p>EPIDEMIOLOGICAL STUDIES OF SKELETAL SAMPLES:</p><p>RELEVANT CONCEPTS AND LIMITATIONS</p><p>An approach of clear value in palaeoepidemiology simply involves the application of medical</p><p>epidemiological methods to the investigation of disease in archaeological skeletal samples</p><p>(Waldron, 1991a–c, 1994). Fundamental observations in epidemiology are measures of the</p><p>occurrence of disease; the main measurements are those of risk, incidence and prevalence</p><p>(Rothman, 2002). In a population with N individuals and in which A individuals developed</p><p>the disease of interest during a period of time, risk is calculated (Rothman, 2002: 24) as</p><p>Risk = A</p><p>N</p><p>= Number of subjects developing disease during a time period</p><p>Number of subjects followed for the same period</p><p>The average risk in the population is also known as the ‘incidence proportion’. In most</p><p>cases risk is used in reference to a single person’s risk of developing the disease, whereas</p><p>incidence proportion refers to groups of people (Rothman, 2002). Incidence rate is similar</p><p>to incidence</p><p>proportion, but instead of measuring the number of subjects with the disease</p><p>as a proportion of the number of subjects that were initially followed, cases A are divided</p><p>by a specific period of time T , which is the summation, across all individuals, of the time</p><p>experiences by the population being followed (Rothman, 2002). Palaeopathological studies</p><p>are cross-sectional in nature and, therefore, cannot measure incidence (Waldron, 1994).</p><p>It is best, instead, to focus on the assessment of prevalence (Baker and Pearson, 2006).</p><p>Prevalence is the number of people P in a population of N individuals who have specific</p><p>disease (Rothman, 2002). The prevalence is P/N and is often multiplied by 1000 and reported</p><p>in epidemiological studies as a rate per 1000 (Gordis, 2000). Prevalence may be measured</p><p>as point prevalence, i.e. over a short period of time, or as a period prevalence, in which the</p><p>Epidemiological Approaches in Palaeopathology 49</p><p>time period is longer (Gordis, 2000). It is important to note that prevalence does not take</p><p>into consideration aspects such as when the disease developed and its duration.</p><p>Researchers who intend to apply a palaeoepidemiological approach to the study of disease</p><p>in archaeological skeletal samples must take into consideration certain problems and limi-</p><p>tations. First, it is necessary to assess the degree to which a sample is truly representative</p><p>of the population (Waldron, 1994; Chapter 2). In the case of palaeoepidemiological studies,</p><p>representativeness does not necessarily apply to the true biological parent population, but</p><p>rather to the group of interest to the epidemiological analysis. In fact, the sample used</p><p>by palaeopathologists will almost never be random in the epidemiological sense (Waldron,</p><p>1994). Nonetheless, this does not prevent the palaeopathologist from deriving a random</p><p>subsample from the parent skeletal ‘population’.</p><p>A second concern is the state of preservation of specimens in a given skeletal sample.</p><p>Palaeopathologists must make decisions concerning what to do with skeletons in which some</p><p>skeletal parts are damaged or missing, particularly in cases in which pathognomonic fea-</p><p>tures of a given disease require the preservation of specific skeletal features. In studies of</p><p>archaeological populations, bone preservation may vary not only between archaeological sites,</p><p>but also between and within cultural layers in the excavated area. The researcher, therefore,</p><p>should assess the preservation of specimens from the various archaeologically defined areas</p><p>or strata in order to decide whether differential preservation may bias prevalence estimates.</p><p>A third aspect is the time-scale involved. Most modern epidemiological studies focus on the</p><p>time interval of years or decades (Waldron, 1994). Palaeoepidemiology cannot usually assess</p><p>prevalence and other epidemiological aspects in this temporal resolution. Mean prevalence</p><p>of a disease over a time span of several hundred years may obscure variations in disease</p><p>prevalence during the time interval (Waldron, 1994). This, however, is more of an issue</p><p>in the study infectious diseases with characteristic episodic peaks and troughs than it is</p><p>for non-infectious conditions such as osteoarthritis. Because the palaeopathologists cannot</p><p>usually address palaeoepidemiological aspects in refined temporal resolution, the focus is</p><p>more often on broad chronological phases or culturally defined subdivisions of the skeletal</p><p>sample, or on aspects such as gender or social status.</p><p>A fourth aspect is the assessment of error in the diagnosis of conditions. Only a small</p><p>number of palaeopathological studies (e.g. Waldron and Rogers, 1990) involve a systematic</p><p>assessment of observer errors and error in the diagnosis of disease from lesions on bone.</p><p>This would involve: evaluation of repeatability of diagnosis of specific conditions by the</p><p>same observer and by other observers; assessment of the accuracy of the method, in terms</p><p>of the degree to which it does not exclude cases with the condition or include those without;</p><p>the skill required to assess the condition on the basis of the specific set of criteria; and</p><p>the investment of time that is required in order to evaluate the condition. Clearly, a more</p><p>comprehensive approach to the diagnosis of a specific condition may entail the recording of</p><p>an extensive set of diagnostic features. However, multiple features may be highly correlated,</p><p>so it is sufficient to include the minimal number of criteria that are pathognomonic. Recording</p><p>of fewer skeletal features means that fewer specimens need be excluded from study due to</p><p>incompleteness or time constraints on the work.</p><p>Disease Prevalence in Past Populations: Case Studies</p><p>In the following sections, two hypothetical data sets are provided in order to demonstrate</p><p>new methods for the calculation of prevalence rates in skeletal populations which take into</p><p>account missing data, differential diagnostic criteria and undiagnosed specimens</p><p>50 Advances in Human Palaeopathology</p><p>Calculation of Prevalence Rates Based on Differentially Weighted Criteria</p><p>Table 3.1 is a hypothetical study of prevalence of leprosy based on a set of morphological</p><p>criteria and on the methodology described by Law (2005). The columns represent a series</p><p>of observed pathological features. All conditions are recorded on a scale of ‘0-3’, where</p><p>‘0’ indicates the trait is absent and ‘1–3’ indicate the escalating scale of degree of severity</p><p>for the presence of the feature. Features that cannot be recorded because parts are missing</p><p>or damaged are marked with ‘D’. An ‘if’ condition was then applied so that a case was</p><p>diagnosed with leprosy if at least one of the rhinomaxillary criteria was given a score of ‘3’,</p><p>or one of the changes on the fibula/tibia and one or more of the changes affecting the joints</p><p>of the hand and/or feet are present. This can be reduced to the following logical expression:</p><p>Ci = 1 if {(RM1 or RM2 or RM3 or RM4 = 3) OR [(VG or SPE ≥ 3) AND (NBD or CDR</p><p>≥ 3)]}; else Ci = 0</p><p>where Ci is case i in a skeletal population and a value of ‘1’ denotes a specimen diagnosed</p><p>with leprosy and ‘0’ a specimen not diagnosed with leprosy.</p><p>The drawback of this system is that in some specimens it may be impossible to record</p><p>one or more of the rhinomaxillary traits due to post-mortem damage. It is difficult to</p><p>decide whether a skeleton should be included in which, for example, the anterior nasal</p><p>spine and/or the alveolar process of the premaxilla are damaged. Clearly, no high score</p><p>for the rhinomaxillary syndrome may be obtained in such instances simply because of</p><p>post-depositional damage. Nevertheless, it is possible with some slight modifications to use</p><p>the above methodology to calculate prevalence rates of both infectious and non-infectious</p><p>conditions. Pathognomonic aspects may be included with a logical condition so that a score</p><p>of ‘1’ is only obtained once these are present. Moreover, data can then be used by other</p><p>researchers who can modify the conditional phrase in order to compare their study with</p><p>others. Alternatively, a probabilistic approach may be adopted by applying a dichotomous</p><p>(absent, present) assessment to a set of features rather than giving particular weight to</p><p>Table 3.1 Hypothetical study of the prevalence of leprosy by a set of morphological criteriaa</p><p>Rhinomaxillary Fibula and tibia Hands and feet</p><p>Case no. RM1 RM2 RM3 RM4 VG SPE NBD CDR N (complete) Ci</p><p>1 0 1 3 D 1 2 3 2 1</p><p>2 2 2 2 2 2 2 0 0 1 0</p><p>3 0 1 0 0 0 0 3 3 1 0</p><p>4 3 0 D D 0 0 1 1 1</p><p>5 D D D 0 0 1 3 2 0</p><p>5 D 0 0 D 1 2 D D 0</p><p>6 0 3 3 2 1 2 3 3 1 1</p><p>aN (complete) denotes the total number of specimens that minimally have one complete tibia, fibula, hand and foot</p><p>bones and preserved facial morphology allowing the diagnosis of rhinomaxillary features.</p><p>RM1–RM4 are the four</p><p>rhinomaxillary changes described by Møller-Christensen (1961) on the scale of 0–3, where 0 denotes the normal</p><p>non-pathological condition; D: parts are missing or damaged. VG: vascular grooves; SPE: subperiosteal exostoses;</p><p>NBD: new bone deposition; CDR concentric diaphyseal resorption.</p><p>Epidemiological Approaches in Palaeopathology 51</p><p>pathognomonic features (Boldsen, 2001). The drawback of the probabilistic approach is that</p><p>it gives equal weight to each feature and eliminates any scoring scale for the manifestation</p><p>of a given feature.</p><p>Age- and Sex-Specific Disease Prevalence in Skeletal Samples</p><p>The following is a hypothetical example of the calculation of prevalence of gout in a skeletal</p><p>population of 1100 individuals (550 males and 550 females). The sample is then stratified</p><p>by age and sex (Table 3.2).</p><p>Data in Table 3.2 are based on the assumption that it was possible to age and sex all of</p><p>the specimens in the skeletal population. The ‘unknown’ specimens are those in which no</p><p>pathognomonic gouty lesions were observed but in which a key diagnostic skeletal part or</p><p>parts were missing or damaged (e.g. the halluces in the case of gout).</p><p>The following values were calculated: total unknown Ut = 140; average unknown per sub-</p><p>group Ua = 140/6 = 23; and the proportion of unknown cases is Li = Ui/Ni. The researcher</p><p>may decide not to calculate prevalence when the proportion of skeletons whose diagnostic</p><p>status is unknown exceeds a specific value for a given subgroup or for the whole sample as</p><p>this may suggest that the overall preservation of the sample is too poor to allow a reliable</p><p>palaeoepidemiological investigation.</p><p>The prevalence of the disease P in the stratified cells excludes all unknown specimens. It</p><p>is impossible to assess what proportion of the unknown specimens had the disease. However,</p><p>it is possible to estimate minimum and maximum values by assuming that either all or none</p><p>of the unknown specimens had the disease.</p><p>Next, we set the null hypothesis that there is no significant difference in Li for the</p><p>subcategories. Simple chi-square analysis of the unknown skeletons by each subcategory</p><p>indicates that we should reject the null hypothesis that there is no significant age or sex bias</p><p>in the distribution of unknown cases (�2 = 12�73, p > 0�05, 4 d.f.). This may indicate that</p><p>there are problems with the analysis of prevalence rates in this population. There are, in fact,</p><p>no simple solutions to this scenario, as pooling the subcategories and calculating the crude</p><p>prevalence figure will not resolve this bias.</p><p>Next, we examine the null hypothesis of no significant difference in prevalence rates</p><p>for the subcategories. A chi-square analysis of the prevalence of gout in each subcategory</p><p>Table 3.2 Age and sex specific prevalence of gout in a hypothetical skeletal population</p><p>18–30 years 30–50 years >50 years</p><p>Males Females Males Females Males Females</p><p>Gout 10 15 30 35 35 45</p><p>No gout 95 95 150 150 150 150</p><p>Unknown U 20 20 30 30 30 10</p><p>N (gout + no gout)) 105 110 180 185 185 195</p><p>N (gout+no gout+unknown) 125 130 210 215 215 205</p><p>Proportion unknown L 0�16 0�15 0�14 0�14 0�14 0�05</p><p>Prevalence P 0�10 0�14 0�17 0�19 0�19 0�23</p><p>52 Advances in Human Palaeopathology</p><p>yielded a non-significant value (�2 = 8�59, p > 0�05, 4 d.f.), so there is no evidence for a</p><p>difference in the prevalence rates of the various subcategories.</p><p>Comparing Prevalence Rates Between Skeletal Populations and Confidence Intervals of</p><p>Prevalence</p><p>The comparison of the prevalence rates in several populations requires either direct or indirect</p><p>standardization of the age- and sex-specific rates (Waldron, 1994; Chapter 2). A standard</p><p>population can be either another skeletal population with comparable age and sex categories</p><p>(and for which prevalence rates of the same disease are available), epidemiological data on a</p><p>modern population, or an entirely artificial population (as in Table 3.2). The standardization</p><p>of prevalence rates requires that both populations are subdivided according to the same age</p><p>and sex categories and that the same methods are applied for the diagnosis of the disease</p><p>of interest. It calls, therefore, for the use of standardized ageing and sexing methods in</p><p>population-based studies in palaeopathology, and preferably for the use of well-defined age</p><p>intervals. At present, there are no standard prevalence data that take into consideration the</p><p>effect of specimens where diagnostic parts are missing or damaged. Nonetheless, researchers</p><p>that wish to apply such a method to the study of several populations can combine these</p><p>into a total sample, and use the prevalence rates obtained to derive standardized rates for</p><p>each of the populations. Alternatively, it is possible to standardize the rates using medical</p><p>epidemiological prevalence figures and then compare the observed rates with the expected</p><p>rates using indirect standardization. The outcome of such a standardization procedure is the</p><p>derivation of palaeoepidemiological data that are comparable and, hence, is a step towards</p><p>meta-analysis of palaeopathological studies and pooling and/or comparing data from different</p><p>skeletal samples.</p><p>Palaeopathologists usually calculate the prevalence of a disease in a skeletal sample</p><p>without calculating its associated confidence interval. Consequently, they do not take into</p><p>account the range of possible values for the calculated prevalence in the population. The</p><p>95 % confidence interval provides a range of values within which there is a 95 % chance</p><p>that the true figure lies. An estimate of the 95 % confidence interval for the true population</p><p>mean is provided by</p><p>CI95 = P ±1�96×SE�P�</p><p>where P is the prevalence rate of the disease, 1.96 is the z value for a 95 % confidence</p><p>interval of a normal distribution, and SE(P) is the standard error of the prevalence under the</p><p>binomial model:</p><p>SE�P� =</p><p>√</p><p>P�1−P�</p><p>N</p><p>where N is the total cases excluding those with diagnostic parts missing or damaged.</p><p>The bottom three rows in Table 3.3 provide standard errors of the prevalence from the</p><p>above data set (Table 3.2) and the minimum and maximum values of the 95 % confidence</p><p>intervals. It is evident that there are no great differences in standard error for the various</p><p>columns. The interval obtained (ranging between the minimum and maximum prevalence</p><p>figures in Table 3.2) and the standard error of the prevalence rate provide the researcher</p><p>with additional information about the prevalence of disease in each age category.</p><p>Epidemiological Approaches in Palaeopathology 53</p><p>Table 3.3 Standard errors of the prevalences in Table 3.2 and the associated minimum and maximum</p><p>values of the 95 % confidence interval</p><p>18–30 years 30–50 years >50 years</p><p>Males Females Males Females Males Females</p><p>Prevalence P 0.095 0.136 0.167 0.189 0.189 0.231</p><p>SE 0.029 0.033 0.028 0.029 0.029 0.030</p><p>CIMIN 0.039 0.072 0.112 0.133 0.133 0.172</p><p>CIMAX 0.151 0.200 0.221 0.246 0.246 0.290</p><p>Analytical Palaeoepidemiology: Case-Control Studies</p><p>A case-control study is based on a non-random sampling from the source population and,</p><p>hence, can begin with people with a disease (the cases) and compare them with people</p><p>without the disease (the controls) (Gordis, 2000). The case-control study is suitable for</p><p>palaeoepidemiological investigations that start from the identification of disease in bone or</p><p>other tissue (Waldron, 1994).</p><p>Matching cases to controls may be done in two ways, group matching and individual</p><p>matching. Group matching involves selecting the controls in such a manner that the proportion</p><p>of certain characteristics, such as sex and age, are identical to the proportions</p><p>among the</p><p>cases. Individual matching involves selecting, for each case, a control that matches its specific</p><p>variables of concern, such as age and sex (Gordis, 2000). The researcher may apply a 1:1</p><p>ratio of cases and controls or alternatively use multiple controls, applying a ratio of 1:2, 1:3</p><p>or even 1:4 cases to controls. Multiple controls are used in medical epidemiology when the</p><p>researcher wishes to increase the overall sample size without having to increase the number</p><p>of cases (which may be difficult with, for example, rare conditions).</p><p>An example of a palaeoepidemiological case-control study is the analysis of the association</p><p>between osteoarthritis of the hands and knee. This association exists in modern populations;</p><p>a case-control study was used to investigate whether it also extended back into the past.</p><p>One hundred and fifteen skeletons with hand osteoarthritis from 18th–19th century AD</p><p>cemeteries in London were examined (Waldron, 1997). These cases with hand osteoarthritis</p><p>were selected based on the condition that knee joints were also present for observation. The</p><p>115 cases were individually matched for sex and age with 115 controls that did not have</p><p>osteoarthritis of the hands and which had knee joints preserved to allow the assessment</p><p>of osteoarthritis. Eight cases had osteoarthritis of the knee in comparison with only two</p><p>controls. The results appear to confirm that an association between osteoarthritis of the hand</p><p>and osteoarthritis of the knee already prevailed in 18th–19th century AD British populations.</p><p>FUTURE DIRECTIONS</p><p>Both medical epidemiology and palaeopathology share a common interest in disease pat-</p><p>terns and change over time, and a focus on processes that concern populations rather than</p><p>individuals. However, palaeopathological research, which in most instances is based on the</p><p>54 Advances in Human Palaeopathology</p><p>analysis of archaeological skeletal samples, calls for the further development of new epi-</p><p>demiologically based methods and techniques (de Souza et al., 2003). It calls for a focus</p><p>on biocultural approaches that form a sound link between the biological and socio-cultural</p><p>aspects that affected the health status of past societies. These should be addressed using a</p><p>population-based approach which can, on the one hand, apply the causal analytical approach</p><p>of epidemiological reasoning (and, hence, focus on the assessment of specific disease factors)</p><p>and, on the other hand, anchor these to the specific archaeological contexts of the skeletal</p><p>samples analysed. The growing trend in palaeopathology to shift away from the descriptive</p><p>realm to the analytical realm requires the application of analytical quantitative methods and</p><p>a focus on hypothesis-driven research.</p><p>Prevalence rates of disease in skeletal samples are probably the most useful parameter</p><p>in palaeoepidemiological investigations based on the analysis of archaeological skeletal</p><p>samples. Prevalence figures for archaeological skeletal series must take into consideration</p><p>issues of preservation and representativeness of specimens in the sample, and particularly in</p><p>cases when poor preservation or missing skeletal features preclude the diagnosis of a given</p><p>disease.</p><p>Biomolecular methods potentially allow the derivation of prevalence rates of infectious</p><p>diseases from bone/tissue samples from past populations. Unlike palaeopathological studies,</p><p>in which prevalence rates are assessed from lesions, these rates are based on identification</p><p>of pathogen DNA fragments. Palaeoepidemiological investigations that apply ancient DNA</p><p>analysis of pathogens may, therefore, open a new window to our understanding of the</p><p>antiquity of various diseases and cultural – historical changes in prevalence rates. However,</p><p>such studies must take into consideration the factor of latency; that is, the detection of a</p><p>pathogen, such as Mycobacterium tuberculosis, in a skeleton does not necessarily indicate</p><p>active disease. The calculation of prevalence rates of infectious diseases using ancient</p><p>DNA must also take into consideration the bias of false negatives due to non-survival of</p><p>ancient DNA.</p><p>In future it may be possible to derive inferences about the evolution, spread and natural</p><p>history of a given disease from the use of epidemiologic mathematical models. Modern</p><p>infectious disease epidemiology is underpinned by a theoretical framework based on the</p><p>reproductive number R0 (Anderson and May, 1991). The reproductive number R0 is defined</p><p>as the average number of secondary cases caused by one infectious case in a fully susceptible</p><p>population. Therefore, if R0 > 1, then the infection is able to spread; and if R0 < 1, then the</p><p>epidemic dies out. An important influence on R0 is contact patterns between humans and</p><p>disease-causing pathogens. This contact may be a result of contact with infected humans or</p><p>with other reservoirs of disease. To our knowledge, the concept of R0 has yet to be used as</p><p>an analytical tool in palaeoepidemiology, but it could be potentially applicable to such work.</p><p>Contact patterns in past populations could be estimated from archaeological data, such as</p><p>the study of the internal architecture of houses, location of refuse pits, kitchens, distance</p><p>between animal pens and human living areas, and other aspects of settlement morphology.</p><p>Population size also plays a role in determining the persistence of disease, by modifying the</p><p>number of available susceptibles and, hence, the number of effective contacts. Estimates for</p><p>population size in different periods/regions can be obtained from archaeological studies of</p><p>average size and settlement pattern analysis.</p><p>The discipline of palaeopathology may need to evolve and develop a better cross-</p><p>disciplinary dialogue which is grounded in a firmer theoretical basis. So far, palaeopathol-</p><p>ogists have managed successfully to incorporate medical, demographic and archaeological</p><p>Epidemiological Approaches in Palaeopathology 55</p><p>concepts in their analysis and interpretation for disease in past populations, but much more</p><p>work is needed on disease aetiology, causation and spread in past populations. The time is ripe</p><p>to broaden the current scope and to work towards a better dialogue between palaeopathology</p><p>and other disciplines, such as disease ecology and molecular biology and the development</p><p>of a palaeoepidemiological perspective.</p><p>REFERENCES</p><p>Anderson R, May R. 1991. Infectious Diseases of Humans: Dynamics and Control. Oxford University</p><p>Press: Oxford.</p><p>Aufderheide AC, Salo W, Madden M, Streitz J, Buikstra J, Guhl F, Arriaza B, Renier C, Wittmers Jr</p><p>LE, Fornaciari G, Allison M. 2003. A 9,000 years old record of Chaga’s disease. Proc Natl Acad</p><p>Sci U S A 17: 2034–2039.</p><p>Baker J, Pearson OM. 2006. Statistical methods for bioarchaeology: applications of age-adjustment</p><p>and logistic regression to comparisons of skeletal populations with differing age-structures. J Arch</p><p>Sci 33: 218–226.</p><p>Barrett R, Kuzawa XW, McDade T, Armelagos GJ. 1998. Emerging and re-emerging infectious</p><p>diseases: the third epidemiologic transition. Ann Rev Anthropol 27: 247–271.</p><p>Boldsen JL. 2001. Epidemiological approach to the palaeopathological diagnosis of leprosy. Am J Phys</p><p>Anthropol 115: 380–387.</p><p>Cohen MN. 1989. Health and the Rise of Civilization. Yale University Press: New Haven, CT.</p><p>Cohen MN, Armelagos GJ. 1984. Paleopathology at the Origins of Agriculture. Academic Press:</p><p>Orlando, FL.</p><p>De Souza SM, de Carvalho DM, Lessa A. 2003. Paleoepidemiology: is there a case to answer? Mem</p><p>Inst Oswaldo Cruz 98(Suppl 1): 21–27.</p><p>Gordis L. 2000. Epidemiology (2nd edition). WB Saunders: Philadelphia.</p><p>Law A. 2005. A simple method for calculating the prevalence of disease in a past human population.</p><p>Int J Osteoarchaeol 15: 146–147.</p><p>Lilienfeld DE, Stolley PD. 1994. Foundations</p><p>of Epidemiology (3rd edition). Oxford University Press:</p><p>New York.</p><p>Møller-Christensen V. 1961. Bone Changes in Leprosy. Munksgaard: Copenhagen.</p><p>Pearce N. 2005. A Short Introduction to Epidemiology (2nd edition). Occasional Report Series No. 2,</p><p>Centre for Public Health Research. Massey University: Wellington.</p><p>Roberts CA. 2002. The antiquity of leprosy in Britain: the skeletal evidence. In The Past and Present of</p><p>Leprosy, Roberts CA, Lewis ME, Manchester K (eds). British Archaeological Reports, International</p><p>Series 1054. Archaeopress: Oxford: 213–222.</p><p>Rockett IR. 1999. Population and health: an introduction to epidemiology. Popul Bull 54: 1–44.</p><p>Rothman KJ. 2002 Epidemiology: An Introduction. Oxford University Press: New York.</p><p>Steckel RH, Rose JC, Larsen CS, Walker PL. 2002. Skeletal health in the western hemisphere from</p><p>4000 B.C. to the present. Evol Anthropol 11: 142–155.</p><p>Steckel RH, Sciulli PW, Rose JC. 2003. A health index from skeletal remains. In The Backbone of</p><p>History: Health and Nutrition in the Western Hemisphere, Steckel RH, Rose JC (eds). Cambridge</p><p>University Press: Cambridge; 61–93.</p><p>Waldron T. 1991a. Rates for the job. Measures of disease frequency in palaeopathology. Int J</p><p>Osteoarchaeol 1: 17–25.</p><p>Waldron T. 1991b. Variations in the rates of spondylolysis in early populations. Int J Osteoarchaeol</p><p>1: 63–65.</p><p>Waldron T. 1991c.The prevalence of, and the relationship between some spinal diseases in a human</p><p>skeletal population from London. Int J Osteoarchaeol 1: 103–110.</p><p>56 Advances in Human Palaeopathology</p><p>Waldron T. 1994. Counting the Dead: The Epidemiology of Skeletal Populations. Wiley: Chichester.</p><p>Waldron T. 1997. Association between osteoarthritis of the hand and knee in a population of skeletons</p><p>from London. Ann Rheum Dis 56: 116–118.</p><p>Waldron T, Rogers J. 1990. Inter-observer variation in coding osteoarthritis in human skeletal remains.</p><p>Int J Osteoarcheol 1: 49–56.</p><p>4</p><p>Macroscopic Analysis and</p><p>Data Collection in</p><p>Palaeopathology</p><p>Anne L. Grauer</p><p>Department of Anthropology, Loyola University of Chicago, 6525</p><p>N. Sheridan Road, Chicago, IL 60626, USA</p><p>INTRODUCTION</p><p>Palaeopathology has embraced technology. The incorporation of precision instrumentation,</p><p>along with the application of new biochemical and biomedical techniques such as DNA</p><p>analysis, radiographic imaging and trace element analysis, has provided researchers with</p><p>new insights into the past and a promise of new theoretical directions to follow. Ironically,</p><p>however, the foundation of palaeopathological investigation, i.e. macroscopic analysis and</p><p>data collection, is rarely the focus of discussion or debate regardless of the fact that it serves</p><p>as the starting point for virtually all methodological approaches.</p><p>The lack of discussion towards improving macroscopic techniques and data collection</p><p>is not surprising given the fact that, at its most cursory level, the researcher needs only</p><p>good (or slightly magnified) vision, a reasonable light source, a sound measuring device,</p><p>and pen and paper. This low-tech tool kit hardly sparks the imagination of the young, nor</p><p>does it garner the excitement that genetic research, biochemical analyses, and spectacular</p><p>visuals can achieve in the public eye. Even within our discipline, discussion and focus on</p><p>the recognition, assessment, and recording of lesions noticeable to the naked eye remain</p><p>negligible, as new techniques overshadow old arguments.</p><p>Macroscopic analysis, however, stands as the primary and most pervasive means by which</p><p>researchers worldwide begin skeletal assessment. As recognized by Lovell (2000: 219):</p><p>Visual observation is generally the first method employed when examining archaeological</p><p>remains for pathological lesions. In many cases it may be the only method required, while</p><p>in some circumstances it may be the only method available.</p><p>Advances in Human Palaeopathology Edited by Ron Pinhasi and Simon Mays</p><p>© 2008 John Wiley & Sons, Ltd. ISBN: 978-0-470-03602-0</p><p>58 Advances in Human Palaeopathology</p><p>Arguably, without macroscopic evaluation of human skeletal material even the most tech-</p><p>nologically sophisticated research would be hindered. For instance, is every recovered or</p><p>curated skeleton to be earmarked for DNA, histological, or biochemical analysis in our</p><p>effort to answer our questions about the past? No. We have neither the time nor resources</p><p>to tackle this endeavour. The unrealistic initiative would fall short in answering many of</p><p>our questions. We would still find ourselves relying on our visual assessments of skeletal</p><p>remains to select new datasets for evaluation, to highlight human variation, and to isolate</p><p>variables.</p><p>This being the case, macroscopic evaluation of skeletal material ought to warrant as much</p><p>of our time, concern and effort as any new technique. However, in spite of the significance</p><p>of macroscopic analysis, it seems that macroscopic work has become a poor cousin to the</p><p>technologically innovative, cutting-edge means by which new information is acquired. This</p><p>must change.</p><p>HISTORICAL BACKGROUND</p><p>In order to understand the problems and promise of macroscopic analysis and data collection</p><p>in palaeopathology it is essential to understand the road that palaeopathology has taken.</p><p>Palaeopathology is a relatively new field with a long history. While curiosity about the</p><p>antiquity of human life developed amidst Renaissance Europe’s fascination with the ‘ancient</p><p>world’, specific focus on the history of diseases through the examination of human remains</p><p>(both skeletal and mummified) only emerged during the 20th century. Numerous workers</p><p>have explored the origins of palaeopathology and have chronicled its growth and devel-</p><p>opment (e.g. Moodie, 1923; Williams, 1929; Wells, 1964; Janssens, 1970; Jarcho, 1966a;</p><p>Brothwell and Sandison, 1967; Buikstra and Cook, 1980; Buikstra and Beck, 2007). All find</p><p>a similar trajectory: that scientific understanding of the history of human diseases began as</p><p>a physician’s ‘hobby’ and developed as an anthropological endeavour much later in the 20th</p><p>century.</p><p>Indeed, a review of publications produced in the late 19th and early 20th centuries clearly</p><p>highlights this trend. Rudolf Virchow (1821–1902), a renowned physician, Frederic Wood</p><p>Jones (1879–1954), an anatomist by training, Grafton Elliot Smith (1871–1937), chair and</p><p>professor of anatomy at the Governmental School of Medicine in Cairo, and Sir Marc Armand</p><p>Ruffer (1859–1917), a professor of bacteriology and the first director of the British Institute</p><p>of Preventative Medicine (heralded by Sandison (1967) as the ‘pioneer of palaeopathology’),</p><p>to name a few, focused much of their skill and attention on the recognition of specific</p><p>pathological lesions in human bone and the presence of anatomical variation between human</p><p>groups. Relying on their diagnostic skills and familiarity with human anatomy developed</p><p>during their medical training, and utilizing the current and burgeoning body of medical</p><p>terminology, the presence of particular diseases witnessed in particular skeletal specimens</p><p>became the focal point of interest. To the credit of these early researchers, their works</p><p>contributed greatly to macroscopic analysis. The diagnosis and recognition of skeletal lesions</p><p>became more firmly based on known clinical manifestations of the diseases. Suggested</p><p>diagnoses were increasingly corroborated by histological and radiographic investigation,</p><p>and the analysis of mummified remains allowed these early researchers to link soft-tissue</p><p>conditions more firmly with skeletal alterations.</p><p>Macroscopic Analysis and Data Collection in Palaeopathology 59</p><p>By the close of the first quarter of the 20th century a new approach towards understanding</p><p>the antiquity of disease began to emerge. Exemplified by Hooton (1930), a substantial</p><p>redirection takes hold within palaeopathology. Researchers now focused on disease processes</p><p>in humans, which were seen as complexly intertwined with ecological variables, human</p><p>behaviour and human culture (Kerley and Bass, 1967). Calvin Wells, for instance, asserted</p><p>that ‘the pattern of disease of injury that affects any group of people is never a matter of</p><p>chance. It is invariably the expression of stresses and strains to which they were exposed,</p><p>a response to everything in their environment and behaviour’ (Wells, 1964: 17). Similarly,</p><p>Brothwell’s broader and bolder perspective on human culture, environmental conditions and</p><p>disease explored prehistoric populations up to and including the Iron Age in Europe and the</p><p>Middle East (Brothwell, 1967).</p><p>Palaeopathological research also began to adopt greater diagnostic rigour. Møller-</p><p>Christensen, for instance, in his studies of the presence of leprosy in human populations</p><p>offered clear criteria for diagnosis (Møller-Christensen, 1967), meticulously described patho-</p><p>logical lesions (Møller-Christensen, 1961, 1978), offered differential diagnoses, and used</p><p>known clinical manifestations and measurements of the disease to aid in the diagnoses of</p><p>archaeological specimens (Møller-Christensen, 1974). In spite of these efforts, and those of</p><p>others, concerns over the uneven adoption of diagnostic rigour within palaeopathology were</p><p>voiced (Jarcho, 1966b; Brothwell and Sandison, 1967: xii).</p><p>In response, a number of important efforts were launched to deal with methodological</p><p>needs. Brothwell and Sandison’s seminal work, for instance, which stemmed ‘from a feeling</p><p>among students of early disease, that the time has come for some form of palaeopathological</p><p>stock-taking and pooling of recently collected data’ (Brothwell and Sandison, 1967: xiv),</p><p>provided the field with a compendium of skeletally recognized conditions and diseases.</p><p>Here, insights into the antiquity and geographical scope of skeletal lesions, as well as radio-</p><p>graphic, histological, and photographic documentation, were provided. Later, Steinbock</p><p>(1976: ix), in an effort to ‘provide a basic framework for those interested in diagnos-</p><p>ing and interpreting bone lesions’, led the way by closely synthesizing clinical data with</p><p>archaeological specimens, providing information on the pathogenesis of conditions, gross</p><p>morphology, radiographic appearance, differential diagnosis and photographic examples of</p><p>the conditions in archaeological samples. More recent efforts to meet palaeopathologists’</p><p>need for diagnosis and interpretation of lesions by Ortner and Putschar (1981), Manch-</p><p>ester (1983), Aufderheide and Rodriguez-Martin (1998), and Ortner (2003), along with</p><p>the extensive database of images from Ortner’s collection, currently residing on the Ohio</p><p>State University server (http://global.sbs.ohio-state.edu/cd-contents/Ortner-slides), have pro-</p><p>vided researchers bases upon which careful visual examination of bone lesions can lead to</p><p>diagnoses. For macroscopic analysis, this has meant that a compendium of clinically and</p><p>archaeologically derived information, along with visual images created by palaeopatholo-</p><p>gists, is available for researchers seeking to compare and understand specimens under their</p><p>investigation.</p><p>Alongside the new emphases on cultural/environmental contexts of diseases and new</p><p>rigorous means of providing diagnoses in the 1960s came changes in the understanding of</p><p>the nature and scope of human disease. Derived in part from Audy’s (1967) expanded and</p><p>holistic definition of health and disease, along with Selye’s (1973) development of the general</p><p>adaptation syndrome, which explored the impact of unspecified physiological disruption</p><p>on the human body, palaeopathologists began to avert their attention from lesions solely</p><p>associated with singular pathogens and to direct it toward a broad spectrum of abnormal</p><p>60 Advances in Human Palaeopathology</p><p>bone. ‘Multiple stress indicators’ and ‘non-specific indicators of stress’ (Huss-Ashmore</p><p>et al., 1981; Armelagos, 1997; Goodman and Martin, 2002) included conditions such as</p><p>aetiologically non-specific periosteal reaction and indicators of growth disruption.</p><p>The new theoretical approaches launched many methodological changes within</p><p>palaeopathology. For one, they initiated the need to incorporate many historically ignored</p><p>variables into skeletal analyses. The age at death and sex of the individual, the nutritional</p><p>and social environment, and developmental and genetic factors became important considera-</p><p>tions in recognizing ‘disease clusters’ and interpreting the presence of ‘stressors’ (Armelagos</p><p>et al., 1982; Armelagos and van Gerven, 2003). Researchers placed an emphasis on popula-</p><p>tion analysis rather than on individuals and required contextual analyses, both archaeological</p><p>and cultural. They demanded continued improvement in the recording and diagnosis of dis-</p><p>ease through stringent comparison with clinical data and with the support of technological</p><p>tools.</p><p>The impact of these changes on data collection was immense. With earlier focus on</p><p>individual specimens and the diagnosis of specific conditions there was little need for sys-</p><p>tematic means of data collection. In fact, data collection often consisted of gathering other</p><p>researchers’ diagnosed specimens. Case studies, which dominate the early literature, dealt</p><p>with sample sizes of one. The new osteoarchaeological and biocultural approaches, however,</p><p>required the development of more standardized recording of skeletal variables to meet the</p><p>needs for population-based approaches and cross-cultural comparison. In 1992, respond-</p><p>ing to continued dissatisfaction with terminology and classification of skeletal lesions, the</p><p>newsletter of the Paleopathology Association (Ragsdale, 1992) published a series of nouns</p><p>and modifiers to be used or avoided when describing bone lesions. A similar effort to</p><p>standardize terminology followed with Thillaud’s (1992: 4) suggestions for macroscopic ter-</p><p>minology and Lovell’s (2000) synthesis of previous work. The success of these endeavours</p><p>was limited. While they drew attention to the need for careful, well-formulated descriptions,</p><p>palaeopathologists made independent decisions on whether to adopt the suggestions or ignore</p><p>them. Hence, suggested ‘terms to avoid’ can still be found in the literature.</p><p>Efforts to improve the standardization of data recording in order to facilitate data sharing</p><p>and population-based study were also initiated. The Paleopathology Association began to</p><p>tackle this challenge in 1988, with discussion of the need and direction that standardization</p><p>might take (Paleopathology Newsletter, September 1988). In 1989, Jonathan Haas, then the</p><p>Vice President for Collections and Research at the Field Museum of Natural History in</p><p>Chicago, convened a workshop to develop standards for the collection of osteological data.</p><p>These efforts were further spurred in 1989 by the enactment of United States Public Law 101-</p><p>185 (The National Museum of the American Indian Act) and in 1990 by the United States</p><p>Public Law 101-601 (The Native American Graves Protection and Repatriation Act), which</p><p>created the immediate need for massive and careful data collection, and for collaboration</p><p>between researchers (Rose et al., 1996).</p><p>A number of contributions to palaeopathology stemmed from these efforts. One was the</p><p>Paleopathology Association Skeletal Database Committee Recommendations (Rose et al.,</p><p>1991), which sought to provide guidance in what types of data to record and methods</p><p>to use (Rose et al., 1991: 1). Another contribution, Standards for Data Collection From</p><p>Human Skeletal Remains (Buikstra and Ubelaker, 1994),</p><p>provided guidelines for terminology</p><p>(which were supported and exemplified by photographs and description), data collection</p><p>(including, for instance, inventory, taphonomy, metric and non-metric analyses, dentition,</p><p>and pathology), and created specific codes for recording variables in order for data to be</p><p>Macroscopic Analysis and Data Collection in Palaeopathology 61</p><p>more fully comparable between researchers. Focus was turned sharply away from determining</p><p>aetiology of conditions and offering diagnoses, towards the codification of skeletal remains.</p><p>Arising from the recognition that vast amounts of data would be generated and that the</p><p>computerization of collected data would enhance researchers’ ability to maintain and collect</p><p>information, a free-standing relational database called the Standardized Osteological Database</p><p>(SOD) was created and made available without cost to interested parties (this database</p><p>remains available as a downloadable application at http://www.cast.uark.edu/cast/sod). More</p><p>recently, building on their familiarity with the strengths and weaknesses of entering alphanu-</p><p>meric codes into text-based screens, as designed and required by Buikstra and Ubelaker</p><p>(1994) and SOD, researchers at the Smithsonian Institution Repatriation Osteology Labo-</p><p>ratory have developed a relational database with an enhanced graphical user interface. The</p><p>strengths of this new system rest on the researcher’s ability to link different data tables</p><p>together through key fields, to prevent the duplication of data or records, to compare many</p><p>records simultaneously, and to manipulate large amounts of data (Ousley et al., 2006). For</p><p>further discussion of databases see Chapter 9.</p><p>Efforts to tackle issues inherent to data collection were also recognized and initiated in</p><p>Britain. Focusing on current research method and theory, the British Association of Biological</p><p>Anthropology and Osteoarchaeology (BABAO) developed a guideline for data collection that</p><p>emphasized utility and function over detail and strict replicability (Brickley and McKinley,</p><p>2004). These efforts, culminating with the production of the Guidelines to the Standards</p><p>for Recording Human Remains (Brickley and McKinley, 2004), provide researchers with a</p><p>well-built foundation upon which individual recording procedures should be developed and</p><p>shared.</p><p>METHODOLOGIES</p><p>As the historical background to macroscopic analysis and data collection illustrates, changes</p><p>in the theoretical foci within palaeopathology have influenced what alterations in the human</p><p>skeleton warrant analysis, and discussions of how alterations are recorded and interpreted.</p><p>There is no single method by which macroscopic diagnoses and data collection is accom-</p><p>plished. However, guidelines developed by Buikstra and Ubelaker (1994), Ortner (1991,</p><p>1992, 1994, 2003), and Brickley and McKinley (2004) indicate that a standard of rigour can</p><p>be created and adopted. Synthesizing the work of these contributors, six suggestions toward</p><p>rigorous macroscopic diagnosis are set out in the following sections. All or most of them</p><p>ought to be adopted.</p><p>Assessment and Recording of Processes and Variables</p><p>Ortner and Putschar (1981: 36) argue that:</p><p>in a descriptive system for abnormal bone conditions, there are several essential elements.</p><p>These include (1) an unambiguous terminology, (2) precise identification of the location</p><p>and distribution of abnormal bone, and (3) a descriptive summary of the morphology of</p><p>the abnormal bone.</p><p>62 Advances in Human Palaeopathology</p><p>Unambiguous Terminology</p><p>Although attempts to standardize terminology within palaeopathology have met with lim-</p><p>ited success (see Ragsdale (1992), Buikstra and Ubelaker (1994) and Lovell (2000) for</p><p>suggestions and guidance), unambiguous terminology is rooted in established physiological</p><p>processes, and clinical derivation. The term ‘periostitis’, for instance, while commonly used</p><p>in the palaeopathological literature to indicate the presence of fusiform bone hypertrophy</p><p>involving the periosteum, carries with it the clinical suggestion that an inflammatory process</p><p>(involving vascular changes and phagocytic activity) is involved. As proliferative bone reac-</p><p>tions of the periosteum can be triggered by a variety of conditions (other than inflammation),</p><p>and macroscopic analysis of dry bone may mask changes to the cortex and/or endosteum,</p><p>‘periostosis’ is a less ambiguous and more precise term to adopt, as it suggests the presence</p><p>of hypertrophy of the periosteum without inserting an unsubstantiated cause. Researchers</p><p>must choose their terminology carefully.</p><p>Identification and Recording</p><p>Precise recording of anatomical location and distribution of abnormal bone is also integral to</p><p>macroscopic analysis. This is because different pathogens, along with physiological changes</p><p>in body functions and abilities, differentially affect areas and/or groups of bones. To this</p><p>end, Ortner (2003: 49–50) offers guidance in recording the distribution of lesions within the</p><p>skeleton, as do Buikstra and Ubelaker (1994). The critical aspect of recording anatomical</p><p>location lies in the need for details that include not only the precise bone affected, but also</p><p>the component of the bone involved (e.g. the epiphysis or proximal third of the diaphysis),</p><p>the aspect of the bone (e.g. anterior/lateral or posterior/medial), and affected features (e.g.</p><p>being circumscribed by the presence of vascular channels, affecting foramina, or traversing</p><p>sutures).</p><p>Descriptive Summaries</p><p>Providing a detailed and descriptive summary of the morphology of the abnormal bone is</p><p>key to macroscopic evaluation. Since variables associated with bone change are essential</p><p>components leading toward possible diagnosis, recognition and recording of all known</p><p>variables is mandatory. As an example, abnormal bone formation is recognizably complex.</p><p>New bone tissue might be composed of woven or compact bone, requiring the type of</p><p>deposited bone to be differentiated and recorded. The pace of formation also varies, resulting</p><p>in the development of different organizational structures (e.g. plaques of bone tissue deposited</p><p>over well-organized compact bone, or rapidly formed spicules of bone). The point of origin of</p><p>bone formation serves as another variable, as endosteal surfaces, cortical structures, and/or</p><p>periosteal surfaces can be impacted in isolation or serve as the starting point of diffuse</p><p>change.</p><p>Creating a Detailed Bone Inventory</p><p>Recognizing the presence of pathological lesions in bone and understanding disease processes</p><p>requires documenting the presence and condition of all bones and bone fragments in the</p><p>skeletal sample under investigation. Reports of the absence or low frequency of a particular</p><p>pathological condition within an individual or group is only meaningful if the investigator</p><p>Macroscopic Analysis and Data Collection in Palaeopathology 63</p><p>has shown that requisite bones, or aspects of requisite bones, were available for examination</p><p>in all members of the population. Low frequency rates within a population may simply be due</p><p>to the repeated failure of the anatomical feature to have survived or to have been recovered.</p><p>As another example, recording that a left humerus was present for evaluation is meaningless</p><p>to an investigator seeking to explore the frequency of degenerative joint disease unless</p><p>the presence and condition of all articular facets was also recorded. Buikstra and Ubelaker</p><p>(1994) and Brickley and McKinley (2004) offer ways in which components of bone can</p><p>be evaluated and recorded, allowing researchers to evaluate the frequency of conditions</p><p>within a population and/or</p><p>discuss the categorization and significance of periapical voids in alveolar bone.</p><p>Pia Bennike discusses trauma in skeletal remains in Chapter 13. She outlines various</p><p>fracture types and their recognition and quantification in skeletal populations. She also</p><p>considers the significance of decapitation and mass graves. To illustrate her points, she</p><p>draws particularly on examples from Denmark and other parts of Scandinavia.</p><p>In her chapter on congenital anomalies, Ethne Barnes gives an account of the morphogen-</p><p>esis of different areas of the skeleton and the anomalies which arise from disturbances to that</p><p>process. Because genetic factors are important causes of most of the anomalies discussed,</p><p>xii Preface</p><p>their biocultural significance lies chiefly with what they can reveal of relationships between</p><p>populations and between individuals. She illustrates this last point with examples from the</p><p>palaeopathological literature.</p><p>In the final contribution, Pinhasi discusses the value of growth studies of past populations.</p><p>He considers some of the methodological issues pivotal to such studies. He emphasizes the</p><p>value of the study of multiple skeletal elements in order to provide a fuller picture of bone</p><p>growth, and of the potential of studies that attempt to ascertain the effect of disease on growth</p><p>in past populations. He illustrates these points with reference to key palaeopathological</p><p>publications.</p><p>Although each contribution reflects the author or authors’ own unique perspective, a</p><p>number of common themes do seem to emerge. The quantification of lesions and disease</p><p>frequency in archaeological skeletal remains continues to present challenges. The benefits of</p><p>collating data generated by different authors are clear; but, in reality, it is often difficult to</p><p>compare data between publications, not least because of the sometimes rapid developments</p><p>and advances in recording methodologies and diagnostic criteria. Gross study of skeletal</p><p>lesions is likely to remain the foundation of palaeopathology, and there is a continued</p><p>emphasis on the development and refinement of macroscopic diagnostic criteria. However,</p><p>although diagenesis complicates the interpretation of medical imaging and histological and</p><p>biomolecular analyses of ancient human remains, these techniques are likely to play an</p><p>increasing role in future. The increasing use of technologically advanced laboratory tech-</p><p>niques, together with the increased appreciation of the value of analytical models from other</p><p>disciplines such as epidemiology, and the need to integrate palaeopathological data with</p><p>historical and or archaeological data, means that collaboration with other disciplines is more</p><p>vital than ever for the continued development of palaeopathology as a field of study.</p><p>Simon Mays and Ron Pinhasi</p><p>Contributors</p><p>Ethne Barnes</p><p>Tucson, Arizona, USA</p><p>Ethne Barnes is a physical anthropologist and palaeopathologist consultant and independent</p><p>researcher, based in Tucson, Arizona. She is recognized for establishing the morphogenetic</p><p>approach to analysing developmental defects of the skeleton in palaeopathology. She holds a</p><p>PhD in physical anthropology from Arizona State University (1991), an MA in anthropology</p><p>(1977) and BSN (1974) from Wichita State University. She has former clinical and teaching</p><p>experiences prior to becoming consultant to the Corinth Excavations of the American School</p><p>of Classical Studies at Athens in 1994, and with INAH excavations in Sonora, NW Mexico,</p><p>in 1998. Research and consultations also include archaeological projects in other parts of</p><p>Greece, Turkey, China, South America and North America. Major publications include</p><p>Developmental Defects of the Axial Skeleton in Paleopathology (University of Colorado</p><p>Press, 1994) and Diseases and Human Evolution (University of New Mexico Press, 2005).</p><p>Pia Bennike</p><p>Laboratory of Biological Anthropology, Institute of Forensic Medicine, University of</p><p>Copenhagen, Blegdamsvej 3, DK-2200 Copenhagen, Denmark</p><p>Pia Bennike received her PhD from the University of Copenhagen, Denmark, in 1984.</p><p>She is currently associated Professor at the Laboratory of Biological Anthropology, Depart-</p><p>ment of Forensic Medicine, University of Copenhagen. She is teaching skeletal biology for</p><p>archaeologists and palaeopathology (PhD courses, and EAA Summer School), and is super-</p><p>visor for a number of medical and archaeological students. Her research encompasses most</p><p>areas of human osteoarchaeology and especially palaeopathology. Key publications include:</p><p>Palaeopathology of Danish Skeletons (Akademisk Forlag, 1985); ‘Ancient trepanations and</p><p>differential diagnosis’ in Trepanation: History, Discovery, Theory, Arnott R, Finger S, Smith</p><p>CUM (eds) (Routledge, 2003); ‘Rebellion, Combat, and massacre: a medieval mass grave at</p><p>Sandbjerg near Næstved’ in Warfare and Society: Archaeological and Social Anthropolog-</p><p>ical Perspectives, Otto T, Thrane H, Vandkilde H (eds) (Aarhus Universitetsforlag 2006);</p><p>and various publications in Danish.</p><p>She is currently vice-president of the European Anthropological Association (former</p><p>president 2000–2004) and president elect of the Paleopathology Association.</p><p>Chryssi Bourbou</p><p>28th Ephorate of Byzantine Antiquities, Chania, Crete, Greece</p><p>Dr Chryssi Bourbou is a Research Associate at the 28th Ephorate of Byzantine Antiquities</p><p>(Hellenic Ministry of Culture). Her main research interests focus on the bioarchaeology of</p><p>xiv Contributors</p><p>medieval Greek populations, with special emphasis on subadult mortality and reconstruction</p><p>of dietary, breastfeeding and weaning patterns. She has participated in many excavations and</p><p>research projects in Greece and elsewhere, and offered lectures and organized workshops on</p><p>bioarchaeology and palaeopathology at institutions and universities worldwide. She has par-</p><p>ticipated in various national and international conferences and has a number of publications,</p><p>including: ‘Infant mortality: the complexity of it all!’, Eulimene, 2001; ‘Health patterns of</p><p>proto-Byzantine populations (6th–7th centuries AD) in South Greece: the cases of Eleutherna</p><p>(Crete) and Messene (Peloponnese)’, International Journal of Osteoarchaeology, 2003; The</p><p>People of Early Byzantine Eleutherna and Messene (6th–7th Centuries A.D.): A Bioarchae-</p><p>ological Approach (Crete University Press, 2004) and is a co-editor of the volume New</p><p>Directions in the Skeletal Biology of Greece (OWLS, forthcoming). She is currently a post-</p><p>doctoral fellow of the State Fellowship Foundation in Greece (2005–2007) and has recently</p><p>received a Dumbarton Oaks Project Grant for Byzantine Studies (2006–2007).</p><p>Don Brothwell</p><p>Department of Archaeology, University of York, King’s Manor, York YO1 7EP, UK</p><p>Don Brothwell is now emeritus professor of human palaeoecology in the University of</p><p>York. He has a doctorate from the University of Stockholm, has taught in the universities of</p><p>Cambridge, London, and York, and for a period was Head of the Sub-Department of Anthro-</p><p>pology (now extinct) in the British Museum of Natural History. He still teaches, and currently</p><p>researches on fossil hominins and mammoths, recent Microtus, and the palaeopathology of</p><p>humans and other mammals. Recent publications include a chapter in The Myth of Syphilis:</p><p>The Natural History of Treponematosis in North America, Powell M, Cook D (eds) (Uni-</p><p>versity Press of Florida, 2005) and ‘Skeletal atrophy and the problem of the differential</p><p>diagnosis of conditions causing paralysis’, Anthropologia Portuguesa, 2000. He is currently</p><p>trying to find time to return to his first love, art (being originally an art school dropout).</p><p>Helen D. Donoghue</p><p>Centre for Infectious</p><p>allowing comparisons between archaeological populations to</p><p>be made.</p><p>Inclusion of Demographic Information</p><p>An increased understanding of the role of sex and gender in disease processes (Grauer</p><p>and Stuart-Macadam, 1998), along with the development of population approaches within</p><p>palaeopathology, have necessitated our inclusion of age at death and sex determination in</p><p>skeletal analysis. Conditions such as iron-deficiency anaemia have been recognized to affect</p><p>skeletal tissue differently depending on age of onset (Stuart-Macadam, 1985), and reproduc-</p><p>tive capacities can directly (as in cases of malignancy of reproductive organs) and indirectly</p><p>(as in the predilection of gout in males or osteoporosis in women) affect the presence of</p><p>diseases. It is essential, therefore, to collect and incorporate demographic information into the</p><p>macroscopic evaluation, as it serves as a vital component for differential diagnosis and cross-</p><p>population comparisons. The assessment of age at death and sex, however, must include the</p><p>use of as many techniques as possible and address, when appropriate, the potential effects</p><p>of disease process on growth and development.</p><p>Appreciation of Multiple Conditions</p><p>The tendency to link the presence of macroscopic lesions with singular causes or pathogens</p><p>is common within palaeopathology. However, both the limited pathognomonic response of</p><p>bone tissue and the synergistic interactions between diseases often render this goal simplistic</p><p>and misleading. As an example, many diseases/conditions provoke osteoblastic activity,</p><p>obscuring our ability to determine the specific pathogen or condition responsible for the</p><p>bone change. Equally confounding are the synergistic effects between disease processes.</p><p>Parasitic or chronic infection may trigger iron-deficiency anaemia, whereas the presence</p><p>of iron-deficiency anaemia (or low serum iron levels) might protect individuals from other</p><p>pathogenic infections. There is also the potential for a single individual to have several</p><p>concurrent diseases or conditions. An appreciation of possible complexities lessens the</p><p>tendency to be oversimplistic and increases the power of differential diagnosis.</p><p>Use of Differential Diagnosis</p><p>An important component of macroscopic analysis is the use of differential diagnosis. Creating</p><p>an exhaustive list of potential causes of a lesion is a reasonable starting point. Included on</p><p>the list must be taphonomic processes that potentially mimic in vivo bone change. Adding</p><p>demographic data, as well as the archaeological and environmental context of the specimen,</p><p>64 Advances in Human Palaeopathology</p><p>provide further evaluative tools. The first step toward differential diagnosis is effectively</p><p>to argue that the bone alteration under investigation is not due to post-mortem processes</p><p>(Chapter 2). This may require radiographic and/or microscopic investigation, as well as</p><p>biochemical tests for diagenesis. Conservatively, only after post-mortem changes have been</p><p>ruled out should further steps towards differential diagnosis be taken. The next step requires</p><p>adopting clinically created and palaeopathologically supported criteria for the presence of</p><p>disease. The former, based upon a wide range of test results performed on the genetic,</p><p>cellular, tissue and system levels, serves as a base line for diagnosis in clinical settings.</p><p>With only bone tissue available for evaluation in most archaeological samples, the adoption</p><p>of clinical criteria is often impossible. Palaeopathologically supported criteria of disease</p><p>seek to unite clinical and palaeopathological research by finding common ground through</p><p>histological evaluation, microscopy, and a range of imaging techniques (Aufderheide and</p><p>Rodriguez-Martin, 1998; Ortner, 2003; Schultz, 2001, 2003). While incorporating these</p><p>approaches can lead to a narrowing of possible aetiologies, they often leave the researcher</p><p>with more than a single possible cause. Waldron (1994) and Ortner (2003) aptly suggest that</p><p>palaeopathologists relinquish the tendency to assert the presence of specific diseases derived</p><p>from macroscopic analysis and to utilize broader disease categories (e.g. metabolic disorder</p><p>or arthropathy) when warranted.</p><p>Use of Multiple Lines of Inquiry</p><p>Similar to the diagnosis of disease in clinical settings where multiple tests are conducted</p><p>and a variety of data are accumulated prior to diagnosis, palaeopathological analysis of</p><p>skeletal lesions on dry bone must adopt multiple lines of inquiry whenever possible. To</p><p>begin this process, determine the feasibility or appropriateness of different approaches.</p><p>For instance, is the preservation of the skeletal material adequate for analysis? The use</p><p>of some techniques may require that minimal post-mortem damage has occurred to the</p><p>bone. Imaging techniques, along with the analysis of isotopes, trace elements, and/or DNA</p><p>from bone tissue, can be compromised by diagenesis and contamination (Chapters 1, 5</p><p>and 8). Next, weigh the promise of the techniques adopted against the costs (both finan-</p><p>cial and material). Running technologically complex tests can cost a considerable sum. Is</p><p>the information sought worthy of the expense? Perhaps more controversial is to pose the</p><p>question of whether the information sought is worth causing destruction to the sample.</p><p>While there is no absolute answer to these questions, it is the investigator’s responsi-</p><p>bility to stand accountable for decisions made. Last, use strong inference, since hasty</p><p>associations between macroscopic lesions and variation in bone microstructure, chem-</p><p>istry, or genetics can be misleading. Have all possible explanations for the association</p><p>or variation been explored? Does only one explanation stand as a possibility? With the</p><p>adoption of these steps, the incorporation of multiple lines of inquiry, alongside macro-</p><p>scopic analysis, can provide new and robust directions for palaeopathological diagnosis and</p><p>interpretation.</p><p>APPLYING TECHNIQUES</p><p>A number of examples of new methodological directions in macroscopic analysis can found</p><p>in the palaeopathology literature. Many of these stem from repeated efforts to improve the</p><p>Macroscopic Analysis and Data Collection in Palaeopathology 65</p><p>diagnosis of macroscopically recognizable lesions (Mays, 2001, 2005; Santos and Roberts,</p><p>2001; Brickley et al., 2005; Matos and Santos, 2006; Mays et al., 2006). They result in</p><p>nuanced understandings of disease processes in humans and aid in our attempts to understand</p><p>the presence and evolution of human disease.</p><p>New methodological directions are also evident in studies adopting decidedly biocul-</p><p>tural and/or evolutionary perspectives. Palfi et al. (1999), for instance, based on the</p><p>contributions of many researchers, offer a compendium of work focusing on the multi-</p><p>faceted approaches towards understanding tuberculosis. Here, the authors compile research</p><p>on radiological, biomolecular, histological, epidemiological, historical, and palaeopatholog-</p><p>ical accounts of the disease. More recently, and more synthetically, Roberts and Buikstra</p><p>(2003) offer an integration and interpretation of archaeological, historical, and clinical data on</p><p>tuberculosis. Successfully integrating detailed descriptions of skeletal lesions with discussion</p><p>of the systemic effects of the disease, and incorporating demographic data with differential</p><p>diagnosis, the authors exemplify how multiple lines of inquiry can begin to provide a ‘global</p><p>view’ of a disease.</p><p>Efforts to understand and interpret the presence of porotic lesions of the cra-</p><p>nium (frequently referred to as porotic hyperostosis) have also provided strength to</p><p>the macroscopic analysis of bone. Schultz (2001, 2003), utilizing histological exam-</p><p>ination and clinical data alongside macroscopic analysis, elucidated</p><p>a wide range of</p><p>medical conditions capable of creating porotic lesions of the skull and offered new</p><p>means for differential diagnoses. These insights, coupled with work specifically focus-</p><p>ing on macroscopic, radiographic and microscopic (scanning electron microscope) skeletal</p><p>manifestations of scurvy found in the Americas (Ortner and Ericksen, 1997; Ortner</p><p>et al., 1999, 2001), have led to the recognition of this condition and, equally impor-</p><p>tant, its ramifications in the Old World (Roberts and Cox, 2003; Brickley and Ives,</p><p>2006).</p><p>The growing emphasis on global and evolutionary perspectives within palaeopathology</p><p>has led to an increased need and more common adoption of standardized recording mea-</p><p>sures. Two efforts in particular illustrate the potential of these initiatives. First, the ‘Western</p><p>Hemisphere Project’, in its efforts to integrate environmental, socio-economic, and biolog-</p><p>ical variables towards an understanding of human health and nutrition over the past 7000</p><p>years, argued for the circumscribed recording of seven skeletal conditions, which were then</p><p>used to construct a health index (Steckel and Rose, 2002). Deliberate measures to train</p><p>participants in the recognition and recording of the skeletal conditions were undertaken to</p><p>ensure uniform data collection. Although the limitations and oversimplifications have been</p><p>a source of concern to many, including to those involved in the project, the potential for</p><p>collaboration, the development of a substantial database, along with opportunities and com-</p><p>mitments to refine and re-evaluate the methods used for interpretation renders this project</p><p>monumental in the development of bioarchaeological and palaeopathological method and</p><p>theory.</p><p>Another ambitious effort was undertaken by Roberts and Cox (2003). Here, the authors</p><p>chronicle health and disease in Britain over a 10 000-year span. Using the work of many</p><p>researchers over many years, they recoded and re-evaluated the presence of macroscopic</p><p>lesions. Although issues of data compatibility were undeniable obstacles, the authors’ deter-</p><p>mination to develop a new synthesis of previously published and unpublished data allowed</p><p>for an unprecedented examination of the relationships between human disease and biosocial</p><p>environments.</p><p>66 Advances in Human Palaeopathology</p><p>PROBLEMS AND ISSUES</p><p>Have we solved all our problems? Do we stand as a unified body of researchers capable</p><p>of effectively and efficiently recognizing, diagnosing, interpreting, and communicating our</p><p>ideas to others? Can we regularly and successfully share our data in an effort to answer new</p><p>questions and adopt new approaches? Unfortunately, the answer is ‘No’.</p><p>Differing ‘Environments’</p><p>The environment within which palaeopathological analysis takes place today may affect</p><p>macroscopic evaluation and data collection. No longer a physician’s hobby, skeletal analysis</p><p>in some countries is conducted by researchers working in two arenas: academia/medicine,</p><p>and ‘contract’ work or ‘applied palaeopathology’. In the first instance, researchers in</p><p>palaeopathology are driven to create research agendas which are fundable, publishable, and</p><p>problem oriented. That is, the researcher chooses a pertinent problem or question and seeks</p><p>to find the answer. In the second instance, skeletal analysis is conducted under contractual</p><p>obligation, with success measured by the collection of data and the creation of a database.</p><p>As seen in Table 4.1, a number of differences between palaeopathological work com-</p><p>pleted within academia and/or medicine and ‘under contract’ or ‘applied palaeopathology’</p><p>exist.</p><p>The repercussions of these differences have noticeable impacts on macroscopic analyses</p><p>and data collection. Within academe, the researcher must be careful to focus closely on</p><p>a stated problem/question and streamline the data presented (and collected). Reviewing</p><p>peers must agree that the data and analyses provided to the reader substantially support the</p><p>conclusion, and ‘extraneous’ information is uniformly excluded. As noted by Wright and</p><p>Yoder (2003: 56):</p><p>Table 4.1 Differences between palaeopathological work in academia and applied palaeopathology</p><p>Palaeopathology ‘under contract’</p><p>Palaeopathology within academia/medicine (applied palaeopathology)</p><p>Greater emphasis on problem-oriented</p><p>research and seeking answers to set</p><p>questions</p><p>Emphasis on thorough data</p><p>collection</p><p>Skeletal variables investigated are defined by</p><p>researcher and limited by the questions</p><p>being posed</p><p>As many skeletal variables as</p><p>possible are examined and recorded</p><p>Time-line of project usually created by</p><p>researcher</p><p>Time restraints often guided by</p><p>external contractual obligations</p><p>Successful in gaining permission and</p><p>acquiring funds to engage in invasive and/or</p><p>technologically complex techniques</p><p>Rarely successful in gaining</p><p>permission to alter skeletal material</p><p>and/or to use technologically</p><p>complex techniques</p><p>Macroscopic Analysis and Data Collection in Palaeopathology 67</p><p>Many palaeopathological studies focus on a single indicator of health, such as enamel</p><p>defects, stature, or porotic hyperostosis. In part, this narrow focus is facilitated by the</p><p>constraints of journal publishing formats, which encourage us to carve up large projects</p><p>into manageable units.</p><p>Hence, problem-oriented studies, which often limit the variables under investigation at the</p><p>onset of the project, might provide even less information to the reader due to publica-</p><p>tion restraints. Conversely, applied palaeopathological work may appear truncated and/or</p><p>appended to archaeological site reports; and when it is allowed to be synthesized with</p><p>archaeological and cultural information, it can remain unpublished or published without</p><p>peer review. Arguably, each difference impacts macroscopic analysis and data collection.</p><p>Importantly, the dichotomy is not necessarily created by work conducted by two separate</p><p>groups of individuals, but rather is created by the impetus for the study.</p><p>Changing Criteria</p><p>As if the tensions and limitations created by research conducted within two different profes-</p><p>sional arenas are not enough of a challenge to palaeopathology, macroscopic analysis and</p><p>data collection is also strapped by its greatest asset: progress. The analysis of degenerative</p><p>joint disease (DJD) serves as a good example of this conundrum.</p><p>The recognition of DJD in archaeological human populations is almost a century old</p><p>(e.g. Ruffer and Rietti, 1912). Years later, assisted by the use of clinical manifestations of</p><p>the condition, Jurmain (1975, 1977) and Steinbock (1976) offered means of recording and</p><p>identifying DJD by creating ordinal scales, or recognizable ‘degrees’ of severity. In 1981,</p><p>Ortner and Putschar (1981: 420), while being careful to assert that ‘marginal lipping can</p><p>represent a feature of joint remodeling without degenerative cartilage changes’, declared</p><p>that ‘for our purpose in the study of dry bone, it [marginal lipping] has to be included</p><p>in the manifestations of degenerative joint disease’. Hence, DJD in skeletal populations</p><p>became identified by marginal lipping (and/or other bony changes, such as eburnation) on</p><p>one or more articular surfaces of the skeleton (as undertaken by Owsley et al. (1987) and</p><p>Bridges (1991)).</p><p>More recently, Waldron and Rogers (1991) and Rogers and Waldron (1995) assert that:</p><p>the palaeopathological diagnosis of OA [osteoarthritis, which is frequently used as a</p><p>synonym for DJD] should be simple and straightforward; it depends first and foremost on</p><p>demonstrating the presence of eburnation. Where eburnation is absent, then we suggest</p><p>that it should be diagnosed only when at least two of the following are present: marginal</p><p>osteophyte and/or new bone on the joint surface; pitting on the joint surface; or alteration</p><p>in the bony</p><p>contour of the joint (Rogers and Waldron, 1995: 43–44).</p><p>Importantly, they warn that ‘OA must never be diagnosed if marginal osteophytosis is the</p><p>only abnormality � � �’ (Rogers and Waldron, 1995: 44). Thus, researchers who might, after</p><p>careful consideration and research, have recorded prevalences of marginal lipping on multiple</p><p>articular surfaces, and who have used an ordinal scale of severity based upon prior published</p><p>methods, might find now that, at best, they will need to re-evaluate their data in order to</p><p>determine the frequency and pattern of DJD in their populations or that, at worst, they may</p><p>find they failed to collect the requisite information to make any assertions.</p><p>68 Advances in Human Palaeopathology</p><p>Conflation and Confusion</p><p>Yet another issue rendering macroscopic evaluation and data collection problematic is the</p><p>tendency within palaeopathology to conflate and confuse description and diagnosis. Wells</p><p>(1964: 36) was quick to recognize these problems and suggested that ‘to reduce this chaotic</p><p>rabble of pathology into an orderly scheme of disease some form of classification must be</p><p>devised’. He provided two proposals: to group conditions based upon anatomy and phys-</p><p>iology, which creates rubrics ‘diseases of the skin, diseases of bone, and cardiovascular,</p><p>ophthalmic or neurological lesions’; or ‘to group diseases according to their cause: congen-</p><p>ital abnormalities, infection, allergy, and other basic principles’ (Wells, 1964). While both</p><p>systems, he found, have strengths and weaknesses, he chose to organize his book based on</p><p>the underlying cause of a disease (i.e. congenital abnormalities, injury, infection, degener-</p><p>ative conditions, etc.). Clearly, the strength of this approach is in allowing researchers to</p><p>evaluate and compare frequencies of conditions within and between populations, and allow-</p><p>ing new questions about human disease to be posed. The drawback to this system is that,</p><p>along with the assessment of the presence of abnormal bone tissue, an integrated judgment</p><p>involving the cause of the lesion is asserted. Data collected using this approach record the</p><p>presence/absence, severity, and location of the already-diagnosed lesion. If the criteria upon</p><p>which we make our diagnoses change (as discussed above), then the recorded data may</p><p>prove to be useless to future researchers.</p><p>Recognizing this dilemma, Buikstra and Cook (1980: 442) suggest that:</p><p>Data collection strategies should be based upon careful and objective description of</p><p>abnormal bony remodelling, avoiding diagnostic labels. Although the investigator may</p><p>emphasize certain conditions, such as periostitis and osteomyelitis, in most cases more</p><p>generalised description of pathologic change is preferable.</p><p>Staunchly agreeing with this approach, and offering considerable guidance and discussion on</p><p>the subject, Ortner and co-workers (Ortner and Putschar, 1981; Ortner, 1990, 2003; Ortner</p><p>and Aufderheide, 1991) provide researchers with substantial bases upon which macroscopic</p><p>evaluation and data collection can be built.</p><p>However, even guidelines created for objective description of bone change are limited</p><p>and may lead to conflation and confusion. As an example, skeletal lesions categorized as</p><p>porotic hyperostosis in Buikstra and Ubelaker (1994: 115) are included in their skeletal</p><p>pathology code key as a separate category from all other bone processes and conditions.</p><p>Limiting the lesion to the frontal, parietal, and occipital bones, they ask the researcher to</p><p>record the ‘degree’ of the lesion (barely discernable, porosity only, porosity with coalescing</p><p>foramina, and coalescing with diploic thickening), as well as whether the lesions are active</p><p>or healed. These specific variables were based upon strong evidence that the presence</p><p>of ‘porotic hyperostosis’, a term offered by Angel (1966), was associated with hereditary</p><p>haemolytic anaemia (Angel, 1966, 1967, 1978), as well as with acquired iron-deficiency</p><p>anaemia (Carlson et al., 1974; El-Najjar, 1976; Stuart-Macadam, 1991, 1992).</p><p>The development of recordable variables based upon limited diagnostic criteria has led</p><p>inadvertently to the conflation of diagnosis and description in the palaeopathological liter-</p><p>ature. The descriptive term ‘porotic hyperostosis’ can be found used de facto to indicate</p><p>the presence of iron-deficiency anaemia, as offered by Larsen (1997: 30): ‘The skele-</p><p>tal changes associated with iron deficiency anaemia are part of a generalised syndrome</p><p>Macroscopic Analysis and Data Collection in Palaeopathology 69</p><p>called porotic hyperostosis’. While Larsen is careful to insert the caveat ‘are part of a</p><p>generalised syndrome’, to many researchers who have presented their work at meetings,</p><p>and have published in site reports and papers, the term ‘porotic hyperostosis’ has become</p><p>unequivocally associated with iron-deficiency anaemia. However, more recent work, indi-</p><p>cating that porous lesions of the ectocrania can be caused by infection, metabolic disorders,</p><p>tumours, haemorrhagic processes, and cranial deformation (Ortner and Putschar, 1981;</p><p>Aufderheide and Rodriguez-Martin, 1998; Schultz, 2001, 2003; Ortner, 2003), requires a</p><p>re-examination of the variables and diagnostic criteria to be adopted by researchers prior to</p><p>interpretation.</p><p>Clearly, the conflation of descriptive terminology with absolute cause is dangerous. It</p><p>fails to follow the careful processes needed for differential diagnosis (in fact, it skips over</p><p>differential diagnosis completely). The end result might be an accumulating body of research</p><p>with limited comparative ability and questionable diagnoses.</p><p>Static Models and Evolutionary Processes</p><p>Germane to palaeopathology is the assumption that macroscopic skeletal manifestations</p><p>of today’s diseases are similar to those in antiquity. Diagnostic sophistication within</p><p>palaeopathology has been achieved through careful correlation of bone change in clini-</p><p>cal populations with similar recognizable change in archaeological populations. However, if</p><p>we model the pathogenesis of particular diseases on modern population studies, we ignore the</p><p>large body of work within medicine, immunology, parasitology, epidemiology, and genet-</p><p>ics which indicate that pathogens, along with host-pathogen interactions, have profoundly</p><p>changed over time. Although circumventing this issue might be impossible, acknowledging</p><p>that environmental, immunological, genetic, and nutritional differences between past and</p><p>present populations might influence pathogenesis can serve as new directions for research</p><p>into the presence and influence of past diseases.</p><p>Thoroughness Has Its Limitations</p><p>As aptly explained in this volume, there are a growing number of methodological consider-</p><p>ations in palaeopathological investigation. The impact on the field is great. For macroscopic</p><p>analysis, the inclusion of new imaging techniques along with biochemical and genetic anal-</p><p>yses has allowed researchers to diagnose more robustly the presence of particular diseases</p><p>and to become more sensitive to disease – environment interactions. However, there are</p><p>substantial limitations to the widespread adoption of these techniques within palaeopathol-</p><p>ogy (Roberts, 2002). First, multiple imaging techniques and chemical and/or genetic study</p><p>of bone tissue are expensive analytical methods. Palaeopathologists’ budgets rarely support</p><p>the implementation of these methods on more than a few individuals within a population.</p><p>Second, a number of the new methodological approaches require substantial time for com-</p><p>pletion to ensure that variables are adequately controlled and replicability is possible. Third,</p><p>many of the new methodologies require specialized technology and expertise in preparing</p><p>specimens, running protocols, and interpreting results. Rarely</p><p>will a palaeopathologist have</p><p>the budget, time or expertise to implement many different methods on all individuals within</p><p>the population under investigation.</p><p>70 Advances in Human Palaeopathology</p><p>LOOKING TOWARDS THE FUTURE</p><p>In spite of the limitations and frustration surrounding efforts to describe clearly, carefully</p><p>record, and accurately diagnose skeletal lesions, palaeopathology has taken great strides in</p><p>bringing to light the lives of our ancestors. As macroscopic analysis will likely continue to</p><p>serve as the premier means by which palaeopathological research is conducted, it is essential</p><p>that efforts to improve macroscopic evaluation are equal to our efforts to refine, improve,</p><p>and develop new technologically complex techniques. Towards this goal, a few suggestions</p><p>are offered here.</p><p>Plan to utilize multiple lines of evidence. Balancing the promise of including new tech-</p><p>niques and methods into macroscopic analysis with recognized budget and time limitations is</p><p>difficult. It is not impossible. As practitioners within our field more repeatedly and effectively</p><p>argue that incorporating multiple methodologies alongside macroscopic analysis is essential</p><p>(not tangential) in palaeopathological analysis, the more likely it is to become routine. An</p><p>a priori argument for the need of multiple imaging techniques, trace element analysis or</p><p>ancient DNA, for example, based on a preliminary understanding of the population to be</p><p>excavated, might be more influential when seeking funding for skeletal analysis than asking</p><p>for financial support after macroscopic analysis has been completed and questions arise</p><p>from it.</p><p>Focus on standardized data collection. The contributions of Buikstra and Ubelaker (1994),</p><p>Rose et al. (1991), Brickley and McKinley (2004), and Ousley et al. (2006) toward devel-</p><p>oping standards by which skeletal data can be collected have deeply impacted the discipline.</p><p>Not only have the studies offered means for structure and consistency in data collecting</p><p>within the field, they have initiated a dialogue between researchers that continues to shape</p><p>the field today. The overwhelming amount of data to be collected, as well as the concrete</p><p>suggestions guiding researchers in what and how data might be collected, are current topics</p><p>of discussion, sources of concern, and focal points of debate. The contributions, therefore,</p><p>can be measured not only by the comprehensive nature of the endeavours, but also by the</p><p>continuing dialogue that they have ignited. Clearly, efforts to facilitate discussion and revi-</p><p>sion of techniques used within the field must include the voices of colleagues worldwide, and</p><p>must strive to keep our field dynamic. Perhaps, most importantly, the development of stan-</p><p>dards for data collection is only useful if all skeletal analysts have access to the methods and</p><p>techniques.</p><p>We need to do our homework. Most researchers would agree that a first step toward</p><p>answering a question is to explore previously published material on the subject. All PhD</p><p>dissertations and most published journal articles place the investigator’s question into con-</p><p>text by reviewing the work that has been conducted in the past. A skeletal report ought,</p><p>likewise, to contain information on the origin of selected techniques adopted by the inves-</p><p>tigator. While researchers will often explore the frequency and patterns of the disease under</p><p>investigation cited in publications, or will adopt techniques used by other researchers, rarely</p><p>will an investigator seek to find out how data were recorded and evaluated. That is, they</p><p>fail to seek out how the previous researchers recorded lesions, what scores and scales were</p><p>created and implemented, what criteria were used, what variables were recognized, and what</p><p>level of reproducibility can be met.</p><p>For example, in their attempt to offer concrete guidelines for the collection of human</p><p>skeletal data, Mann and Murphy (1990: 13) make the following suggestion:</p><p>Macroscopic Analysis and Data Collection in Palaeopathology 71</p><p>If, after examining ten or fifteen skeletons, you find that you have been scoring a lesion</p><p>as moderate in severity and feel that it should only be scored as slight, then you should</p><p>modify the criteria for the trait or lesion in question.</p><p>I would argue that if an investigator has to change the criteria of data collection midway</p><p>through analysis, then the investigator did not do their homework. Developing criteria in</p><p>a vacuum, let alone changing criteria mid project, easily renders the data and subsequent</p><p>conclusions unusable to the investigator and to any other interested researchers. Ortner and</p><p>Aufderheide (1991), Grauer (2002), and Roberts and Cox (2003) recount these issues in their</p><p>efforts to compare populations across time and space, and as they attempt to correlate the</p><p>work of many researchers.</p><p>Efforts to ensure that comparisons between studies and between populations are possible</p><p>in the future ought to include detailed descriptions or photographing examples of each level</p><p>of lesion severity denoted, or each type of lesion classified, or each variable delineated in</p><p>the investigator’s macroscopic analysis. As offered by Robb (2000: 479):</p><p>In designing a skeletal research project, to avoid finishing empty handed or having to</p><p>repeat data collection, data analysis must be built into research designs from the start.</p><p>This is all the more so since skeletal research almost always requires striking a practical</p><p>balance between the number of specimens studied, the amount of detail recorded, and</p><p>research time and money.</p><p>Err on the side of caution. The tendency among researchers to label or diagnose skele-</p><p>tal lesions prematurely or inaccurately is unfortunately common (Ortner and Aufderheide,</p><p>1991; Waldron, 1994; Miller et al., 1996; Lovell, 2000). However, with palaeopathologists’</p><p>increasing emphasis on non-specific conditions which impact human health, interests have</p><p>shifted away from documenting the antiquity of specific diseases towards understanding dis-</p><p>ease processes, the evolution of disease in human groups, and the impact of human culture</p><p>on human biology. Within these research agendas all abnormal skeletal changes (even those</p><p>not labelled as pathological by the medical community) have the potential, eventually, to</p><p>provide insight into the past.</p><p>Hence, for the palaeopathologist, carefully recognizing and recording even slight bone</p><p>changes and abnormalities becomes an essential component of evaluation. This means that</p><p>palaeopathologists cannot simply record conditions with which they are familiar, or focus</p><p>on conditions that they can allegedly quickly diagnose. They cannot choose to minimize</p><p>the presence of variation within their population by creating categories of lesions that are</p><p>anatomically, physiologically or pathogenically illogical. Rather they must get in the habit of</p><p>carefully describing what they see. With amazing data collection and data storage technology</p><p>at our fingertips, we need not fear collecting too much information. Ortner and Aufderheide</p><p>(1991: 8) assert that a goal is to:</p><p>develop and apply a descriptive methodology to the analysis and publication of patho-</p><p>logical specimens that does not necessarily preclude classification or diagnosis but,</p><p>at the very least, permits the reader to reach his or her own conclusion regarding</p><p>the nature of the pathology, without having to accept the diagnostic opinion of the</p><p>author.</p><p>72 Advances in Human Palaeopathology</p><p>Consider collaboration. Collaboration in the sciences is not novel, it is de rigueur. As</p><p>robust macroscopic evaluation of skeletal remains benefits from employing multiple</p><p>lines of</p><p>evidence, so too will palaeopathological research and publication benefit from the inclusion</p><p>of many specialists contributing throughout the analysis and publication process. Individuals</p><p>with a wide range of specializations must be routinely asked to participate in palaeopatholog-</p><p>ical projects, and not be called upon as an occasional resource under particular circumstances.</p><p>Research in South America showcases the promise of collaboration, as scientists across the</p><p>globe, specializing in many different fields, work to synthesize data in an effort to create a</p><p>holistic, nuanced and evolutionary perspective of life in Peru, Chile, and Brazil.</p><p>Publish with an eye for comparison. The success of our discipline relies on our ability to</p><p>communicate and to provide a hefty foundation of data and ideas for future palaeopathological</p><p>research. This is only accomplished if the ideas contained within our missives are thoroughly</p><p>researched, robustly argued, and based upon sound, reproducible data. In spite of the tendency</p><p>for skeletal reports to be truncated and/or placed into appendices, and in spite of the tendency</p><p>for journal publishing formats to encourage us to carve up large projects into manageable units,</p><p>it remains our responsibility to get the message across. Seeking to synthesize the maximum</p><p>archaeological and cultural data with pathological analysis must become a greater priority for</p><p>palaeopathologists asked to analyse skeletal collections. Similarly, including more detail and</p><p>a broader scope must become a greater priority in our journal submissions, as this will allow</p><p>investigators worldwide to develop cross-cultural comparisons.</p><p>A recent analysis conducted by the Paleopathology Association (Stodder et al., 2006)</p><p>highlights the controversy over the utility and need for the publication of descriptive case</p><p>studies (Armelagos and van Gerven, 2003; Larsen 2005; Stojanowski and Buikstra, 2005)</p><p>and brings to light the scope and frequency of palaeopathological research published in</p><p>journal format. Palaeopathological research, they found, appeared in 286 different journals</p><p>from 1996 to 2005, with 15 journals publishing 10 or more articles on the subject (Stodder</p><p>et al., 2006: 9). Clearly, palaeopathological research can be interpreted from these findings</p><p>as contributing to a variety of disciplines and scientific fields.</p><p>CONCLUSION</p><p>There is no easy fix to rid palaeopathology of its difficulties and stumbling blocks. Advance-</p><p>ments in technology will not supersede the need to inspect human remains macroscopically,</p><p>record findings, and communicate conclusions. It is equally unlikely that the problems inherent</p><p>in the visual examination of skeletons, common in data collection, and impacting communi-</p><p>cation and cooperation, will disappear if ignored. This being the case, the onus is on us, the</p><p>palaeopathologists. We must focus on clear description, develop and use standardized record-</p><p>ing techniques, share our information, familiarize ourselves with recording techniques used in</p><p>comparative samples, employ many analytical techniques, and maintain a dialogue focused on</p><p>the importance of macroscopic evaluation and data collection within our discipline.</p><p>REFERENCES</p><p>Angel JL. 1966. Porotic hyperostosis, anemias, malarias, and marshes in prehistoric eastern</p><p>Mediterranean. Science 153: 760–763.</p><p>Macroscopic Analysis and Data Collection in Palaeopathology 73</p><p>Angel JL. 1967. Porotic hyperostosis or osteoporosis symetrica. In Diseases in Antiquity, Brothwell</p><p>D, Sandison AT (eds). Charles C. Thomas: Springfield, IL; 378–389.</p><p>Angel JL. 1978. Porotic hyperostosis in the eastern Mediterranean. Med Coll Virginia Q 14: 10–16.</p><p>Armelagos GJ. 1997. Palaeopathology. In History of Physical Anthropology, Vol. 2, Spencer F (ed.).</p><p>Garland: New York; 790–796.</p><p>Armelagos GJ, van Gerven DP. 2003. A century of skeletal biology and palaeopathology: contrasts,</p><p>contradictions, and conflicts. Am Anthropol 105: 53–64.</p><p>Armelagos GJ, Carlson DS, van Gerven DP. 1982. The theoretical foundation and development of</p><p>skeletal biology. In A History of American Physical Anthropology, 1930–1980, Spencer F (ed.).</p><p>Academic Press: New York; 305–328.</p><p>Audy R. 1967. Measurement and diagnosis of health. In Environ/Mental: Essays on the Planet as a</p><p>Home, Shepard P, McKinley D (eds). Houghton Mifflin: Boston, MA.</p><p>Aufderheide AC, Rodriguez-Martin C. 1998. The Cambridge Encyclopedia of Human Paleopathology.</p><p>Cambridge University Press: Cambridge.</p><p>Brickley M, Ives R. 2006. Skeletal manifestations of infantile scurvy. Am J Phys Anthropol 129:</p><p>163–172.</p><p>Brickley M, McKinley JI. 2004. Guidelines to the Standards for Recording Human Remains. IFA</p><p>Paper Number 7. BABAO: Southampton.</p><p>Brickley M, Mays S, Ives R. 2005. Skeletal manifestations of vitamin D deficiency osteomalacia in</p><p>documented historical collections. Int J Osteoarchaeol 15: 389–403.</p><p>Bridges PS. 1991. Degenerative joint disease in hunter-gatherers and agriculturalists from southeastern</p><p>United States. Am J Phys Anthropol 85: 379–391.</p><p>Brothwell D. 1967. The bio-cultural background to disease. In Diseases in Antiquity, Brothwell D,</p><p>Sandison AT (eds). Charles C. Thomas: Springfield, IL; 56–68.</p><p>Brothwell D, Sandison AT. 1967. Editorial prolegomenon: the present and future. In Diseases in</p><p>Antiquity, Brothwell D, Sandison AT (eds). Charles C. Thomas: Springfield, IL; xi–xiv.</p><p>Buikstra JE, Beck L. 2007. Bioarchaeology: The Contextual Analysis of Human Remains. Elsevier:</p><p>San Diego, CA.</p><p>Buikstra JE, Cook DC. 1980. Palaeopathology: an American account. Ann Rev Anthropol 9:</p><p>433–470.</p><p>Buikstra JE, Ubelaker DH. 1994. Standards for Data Collection from Human Skeletal Remains.</p><p>Arkansas Archaeological Survey Research Series No. 44. Arkansas Archaeological Survey:</p><p>Fayetteville, AR.</p><p>Carlson DS, Armelagos GJ, van Gerven DP. 1974. Factors influencing the etiology of cribra orbitalia</p><p>in prehistoric Nubia. J Hum Evol 3: 405–419.</p><p>El-Najjar MY. 1976. Maize, malaria and the anemias in the Pre-Columbian New World. Yrbk Phys</p><p>Anthropol 20: 329–337.</p><p>Goodman AH, Martin DL. 2002. Reconstructing health profiles from skeletal remains. In The Backbone</p><p>of History: Health and Nutrition in the Western Hemisphere, Steckel RH, Rose JC (eds). Cambridge</p><p>University Press: Cambridge; 11–60.</p><p>Grauer AL. 2002. Taking the trauma out of trauma: a callous approach. In Annual Meeting of the</p><p>Midwest Bioarchaeology and Forensic Anthropology Association, Indianapolis, IN.</p><p>Grauer AL, Stuart-Macadam P. 1998. Sex and Gender in Paleopathological Perspective. Cambridge</p><p>University Press: Cambridge.</p><p>Hooton EA. 1930. The Indians of Pecos Pueblo: A Study of their Skeletal Remains. Yale University</p><p>Press: New Haven, CT.</p><p>Huss-Ashmore R, Goodman A, Armelagos GJ. 1981. Nutritional inference from palaeopathology. Adv</p><p>Archaeol Method Theory 5: 395–474.</p><p>Janssens PA. 1970. Palaeopathology: Diseases and Injuries of Prehistoric Man. Humanities Press:</p><p>New York.</p><p>Jarcho S. 1966a. Human Paleopathology. Yale University Press: New Haven, CT.</p><p>74 Advances in Human Palaeopathology</p><p>Jarcho S. 1966b. The development and present condition of human palaeopathology in the United</p><p>States. In Human Palaeopathology, Jarcho S (ed.). Yale University Press: New Haven, CT; 3–30.</p><p>Jurmain RD. 1975. Distribution of degenerative joint disease in skeletal populations. PhD dissertation,</p><p>Harvard University, Cambridge, MA.</p><p>Jurmain RD. 1977. Stress and the etiology of osteoarthritis. Am J Phys Anthropol 46: 353–366.</p><p>Kerley ER, Bass WM. 1967. Paleopathology: meeting ground for many disciplines. Science 157:</p><p>638–644.</p><p>Larsen CS. 1997. Bioarchaeology: Interpreting Behavior from the Human Skeleton. Cambridge</p><p>University Press: Cambridge.</p><p>Larsen CS. 2005. Description, hypothesis testing, and conceptual</p><p>advances in physical anthropology:</p><p>have we moved on? In Annual Meeting of the American Association of Physical Anthropologists,</p><p>Milwaukee, WI.</p><p>Lovell NC. 2000. Paleopathological description and diagnosis. In Biological Anthropology of the</p><p>Human Skeleton, Katzenberg MA, Saunders SR (eds). Wiley-Liss: New York; 217–248.</p><p>Manchester K. 1983. The Archaeology of Disease. Bradford University Press: Bradford.</p><p>Mann RW, Murphy SP. 1990. Regional Atlas of Bone Disease. Charles C. Thomas: Springfield, IL.</p><p>Matos V, Santos AL. 2006. On the trail of pulmonary tuberculosis based on rib lesions: results from</p><p>the human identified skeletal collection from the Museu Bocage (Lisbon, Portugal). Am J Phys</p><p>Anthropol 130: 190–200.</p><p>Mays SA. 2001. Paleopathological and biomolecular study of tuberculosis in a medieval skeletal</p><p>collection from England. Am J Phys Anthropol 114: 298–311.</p><p>Mays SA. 2005. Paleopathological study of hallux valgus. Am J Phys Anthropol 126: 139–149.</p><p>Mays SA, Brickley M, Ives R. 2006. Skeletal manifestations of rickets in infants and young children</p><p>in a historical population from England. Am J Phys Anthropol 129: 362–374.</p><p>Miller E, Ragsdale B, Ortner D. 1996. Accuracy in dry bone diagnosis: a comment on paleopathological</p><p>methodology. Int J Osteoarchaeol 6: 221–229.</p><p>Møller-Christensen V. 1961. Bone Changes in Leprosy. Munksgaard: Copenhagen.</p><p>Møller-Christensen V. 1967. Evidence of leprosy in earlier peoples. In Diseases in Antiquity, Brothwell</p><p>D, Sandison AT (eds). Charles C. Thomas: Springfield, IL; 295–306.</p><p>Møller-Christensen V. 1974. Changes in the anterior nasal spine and the alveolar process of the maxillae</p><p>in leprosy. Int J Leprosy 42: 431–435.</p><p>Møller-Christensen V. 1978. Leprosy Changes of the Skull. Odense University Press: Odense.</p><p>Moodie RL. 1923. Paleopathology: An Introduction to the Study of Ancient Evidences of Disease.</p><p>University of Illinois Press: Urbana, IL.</p><p>Ortner DJ. 1990. Introduction. In Regional Atlas of Bone Disease, Mann RW, Murphy SP, (eds).</p><p>Charles C. Thomas: Springfield, IL; ix–xii.</p><p>Ortner DJ. 1991. Theoretical and methodological issues in paleopathology. In Human Paleopathology:</p><p>Current Syntheses and Future Options, Ortner DJ, Aufderheide AC (eds). Smithsonian Institution</p><p>Press: Washington, DC; 5–12.</p><p>Ortner DJ. 1992. Skeletal palaeopathology: probabilities, possibilities, and impossibilities. In Disease</p><p>and Demography in the Americas, Verano JW, Ubelaker DH (eds). Smithsonian Institution Press:</p><p>Washington, DC; 5–15.</p><p>Ortner DJ. 1994. Descriptive methodology in paleopathology. In Skeletal Biology in the</p><p>Great Plains, Owsley DW, Jantz RJ (eds). Smithsonian Institution Press: Washington, DC;</p><p>73–80.</p><p>Ortner DJ. 2003. Identification of Pathological Conditions in Human Skeletal Remains, second edition.</p><p>Academic Press: San Diego, CA.</p><p>Ortner DJ, Aufderheide AC (eds). 1991. Human Paleopathology: Current Syntheses and Future</p><p>Options. Smithsonian Institution Press: Washington, DC.</p><p>Ortner D, Ericksen M. 1997. Bone changes in the human skull probably resulting from scurvy in</p><p>infancy and childhood. Int J Osteoarchaeol 7: 212–220.</p><p>Macroscopic Analysis and Data Collection in Palaeopathology 75</p><p>Ortner DJ, Putschar WGJ. 1981. Identification of Pathological Conditions in Human Skeletal Remains.</p><p>Smithsonian Contributions to Anthropology, No. 28. Smithsonian Institution Press: Washington,</p><p>DC.</p><p>Ortner D, Kimmerle E, Diez M. 1999. Probable evidence of scurvy in subadults from archaeological</p><p>sites in Peru. Am J Phys Anthropol 108: 321–331.</p><p>Ortner D, Butler W, Cafarella J, Milligan L. 2001. Evidence of probable scurvy in subadults from</p><p>archaeological sites in North America. Am J Phys Anthropol 114: 343–351.</p><p>Ousely S, Dudar JC, Jones E, London M, Madden G, Mulhern D, Wilczak C. 2006. Standardizing</p><p>the standards: computerized documentation of skeletal pathology at the Smithsonian Institution. In</p><p>Workshop presented at the Annual Meeting of the Palaeopathology Association, Anchorage, Alaska.</p><p>Owsley DW, Orser CE, Mann RW, Moore-Jansen PH, Montgomery RL. 1987. Demography and</p><p>pathology of an urban slave population from New Orleans. Am J Phys Anthropol 74: 185–197.</p><p>Palfi G, Dutour O, Deak J, Hutas I. 1999. Tuberculosis Past and Present. Golden Books: Szeged.</p><p>Ragsdale B. 1992. Task force on terminology: provisional word list. Palaeopathol Newslett 78: 7–8.</p><p>Robb J. 2000. Analyzing human skeletal data. In Human Osteology in Archaeology and Forensic</p><p>Science, Cox M, Mays S (eds). Greenwich Medical Media: London; 475–490.</p><p>Roberts C. 2002. Palaeopathology and archaeology: the current state of play. In The Archaeology of</p><p>Medicine, Arnott R (ed.). British Archaeological Reports, International Series 1046. Archaeopress:</p><p>Oxford; 1–20.</p><p>Roberts C, Buikstra JE. 2003. The Bioarchaeology of Tuberculosis. University Press of Florida:</p><p>Gainesville, FL.</p><p>Roberts C, Cox M. 2003. Health and Disease in Britain. Sutton: Stroud.</p><p>Rogers J, Waldron T. 1995. A Field Guide to Joint Disease in Archaeology. Wiley: Chichester.</p><p>Rose J, Anton SC, Aufderheide AC, Eisenberg L., Gregg JB, Neiburger EJ, Rothschild B. 1991.</p><p>Skeletal Database Committee Recommendations. Paleopathology Association, Detroit, MI.</p><p>Rose JC, Green TJ, Green VD. 1996. NAGPRA is forever: osteology and the repatriation of skeletons.</p><p>Ann Rev Anthropol 25: 81–103.</p><p>Ruffer MA, Rietti A. 1912. On the osseous lesions in ancient Egyptians. J Pathol Bacteriol 16:</p><p>439–465.</p><p>Sandison AT. 1967. Sir Marc Armand Ruffer (1859–1917) pioneer of palaeopathology. Med Hist 11:</p><p>150–156.</p><p>Santos AL, Roberts CA. 2001. A picture of tuberculosis in young Portuguese people in the early 20th</p><p>century: a multidisciplinary study of the skeletal and historical evidence. Am J Phys Anthropol 115:</p><p>38–49.</p><p>Schultz M. 2001. Paleohistopathology of bone: a new approach to the study of ancient diseases. Yrbk</p><p>Phys Anthropol 44: 106–147.</p><p>Schultz M. 2003. Light microscopic analysis in skeletal palaeopathology. In Identification of Patho-</p><p>logical Conditions in Human Skeletal Remains, second edition, Ortner DJ (ed.). Academic Press:</p><p>San Diego, CA; 73–108.</p><p>Selye H. 1973. The evolution of the stress concept. Am Sci 61: 692–699.</p><p>Steckel RH, Rose JC. 2002. The Backbone of History: Health and Nutrition in the Western Hemisphere.</p><p>Cambridge University Press: Cambridge.</p><p>Steinbock RT. 1976. Paleopathological Diagnosis and Interpretation. Charles C. Thomas: Spring-</p><p>field, IL.</p><p>Stodder ALW, Johnson K, Chan A, Handwerk E, Rudolph KZ. 2006. Publishing patterns in pale-</p><p>opathology: findings of the Publications Explorations Committee. Paleopathol Newslett 134:</p><p>6–13.</p><p>Stojanowski CM, Buikstra JE. 2005. Research trends in human osteology: a content analysis of papers</p><p>published in the American Journal of Physical Anthropology. Am J Phys Anthropol 128: 98–109.</p><p>Stuart-Macadam P. 1985. Porotic hyperostosis: representative of a childhood condition. Am J Phys</p><p>Anthropol 66: 391–398.</p><p>76 Advances in Human Palaeopathology</p><p>Stuart-Macadam P. 1991. Anaemia in Roman Britain: Poundbury Camp. In Health in Past Societies.</p><p>Biocultural Interpretations of Human Skeletal Remains in Archaeological Contexts, Bush H, Zvelebil</p><p>M (eds). British Archaeological Reports, International Series 567. Tempus Reparatum: Oxford;</p><p>101–113.</p><p>Stuart-Macadam P. 1992. Anemia in past human populations. In Diet, Demography, and Disease:</p><p>Changing Perspectives on Anemia, Stuart-Macadam P, Kent S (eds). Aldine de Gruyter: New York;</p><p>151–170.</p><p>Thillaud PL. 1992. Retrospective diagnosis in paleopathology. Palaeopathol Newslett 80(suppl): 1–4.</p><p>Waldron T, Rogers J. 1991. Interobserver variation in coding osteoarthritis in human skeletal remains.</p><p>Int J Osteoarchaeol 1: 49–56.</p><p>Waldron T. 1994. Counting the Dead. Wiley: Chichester.</p><p>Wells C. 1964. Bones, Bodies and Diseases.</p><p>Thames and Hudson: London.</p><p>Williams HU. 1929. Human paleopathology, with some original observations on symmetrical</p><p>osteoporosis of the skull. Arch Pathol 7: 839–902.</p><p>Wright LE, Yoder CJ. 2003. Recent progress in bioarchaeology: approaches to the osteological paradox.</p><p>J Archaeol Res 11: 43–70.</p><p>5</p><p>Radiography and Allied</p><p>Techniques in the</p><p>Palaeopathology of Skeletal</p><p>Remains</p><p>Simon Mays</p><p>English Heritage Centre for Archaeology, Fort Cumberland,</p><p>Eastney, Portsmouth PO4 9LD, UK</p><p>INTRODUCTION</p><p>Imaging of skeletal lesions using plain-film radiography has been used on ancient human</p><p>remains for over 100 years, and it remains the most important augment to visual examination</p><p>of specimens in the description and diagnosis of disease in palaeopathology. Since the</p><p>discovery of radiographic imaging in the late 19th century, many allied techniques using</p><p>X-rays have been developed and used in clinical medicine. The most important development</p><p>in radiographic imaging is computed tomography (CT), which is covered elsewhere in this</p><p>book (Lynnerup, Chapter 6). However, as well as imaging techniques, methods involving</p><p>the use of X-rays yielding quantitative data have also been developed. Measurements of</p><p>cortical thickness, taken from radiographs, either manually using rulers or callipers or using</p><p>image analysis with computer software, provide quantitative data on cortical bone. Taking</p><p>measurements of cortical bone from radiographs is known as radiogrammetry. Bone density</p><p>can also be measured radiologically. Historically, a variety of techniques have been used to</p><p>do this, but currently one method, dual X-ray absorptiometry (DXA), dominates both clinical</p><p>and palaeopathological bone densitometry.</p><p>This chapter discusses radiographic imaging of palaeopathological lesions and the use of</p><p>radiogrammetry and bone densitometry (particularly DXA) in palaeopathology. In each case,</p><p>the technique and its development in clinical science is described. Its areas of application in</p><p>palaeopathology are discussed, and the issues raised by using methods and techniques devel-</p><p>oped for living patients on excavated skeletal remains are considered. Finally, some possible</p><p>future directions in radiography and allied techniques in palaeopathology are discussed.</p><p>Advances in Human Palaeopathology Edited by Ron Pinhasi and Simon Mays</p><p>© 2008 John Wiley & Sons, Ltd. ISBN: 978-0-470-03602-0</p><p>78 Advances in Human Palaeopathology</p><p>RADIOGRAPHIC IMAGING</p><p>Production of a radiograph requires the passing of X-rays through a specimen and the</p><p>capture of the negative image that results from the attenuation of the X-rays (for details,</p><p>see Lynnerup, Chapter 6). Traditionally, images have been captured on radiographic film,</p><p>which is then processed chemically in a dark room to develop and fix the image. Digital</p><p>image capture, which may include the facility to view the radiographic image in real time,</p><p>is becoming increasingly available (Lang et al., 2005). As with photography, digital image</p><p>capture will doubtless in time oust film, but the substantial investment that has been made</p><p>by institutions in film-based methods means that this will probably be a slower process than</p><p>has been the case for photography.</p><p>Plain-film radiography produces a two-dimensional image. CT is required to generate</p><p>a full three-dimensional image; but in order to at least partially reconstruct the three-</p><p>dimensional appearance of a lesion using radiography, two views of the specimen are</p><p>normally obtained, classically medio-lateral and antero-posterior. For further details of prac-</p><p>tical aspects of skeletal radiography of palaeopathological specimens the reader is referred</p><p>to Ortner (2003: 57–62).</p><p>History of Radiography in Palaeopathology</p><p>X-rays were discovered by Roentgen in November 1895. It was clear from his initial</p><p>experiments that the new rays were an effective means of visualizing bone (Fiori and</p><p>Nunzi, 1995), and barely 2 months later the first clinical radiographs were being made</p><p>(Spiegel, 1995). Archaeologists were quick to grasp the potential of the new rays for studying</p><p>ancient human remains. As early as March 1896, radiographic images were being produced</p><p>of Egyptian mummies (Böni et al., 2004). Most of the early applications of radiography</p><p>in archaeology were carried out on mummified remains, chiefly from Egypt, simply to</p><p>discern what lay within the wrappings (Böni et al., 2004). Perhaps the first publication</p><p>devoted solely to a radiographic skeletal abnormality in human remains came in 1898,</p><p>with a report of an abnormal number of sesamoids in the hand of an Egyptian mummy</p><p>(Clendinnen, 1898). Radiography soon began to be applied not only to mummified material,</p><p>but also to the study of dry bones, with the specific purpose of visualizing pathological</p><p>changes (Eaton, 1916; Means, 1925; Williams, 1929). However, radiography failed to become</p><p>routine in skeletal palaeopathology, and as late as 1963 Calvin Wells (1963: 401) bemoaned</p><p>that it was still ‘largely neglected’. In part, the neglect to which Wells refers may stem</p><p>from the opinion, promulgated particularly by palaeopathologists with medical backgrounds,</p><p>that radiography of specimens, and in particular diagnostic radiography, was beyond the</p><p>grasp of the palaeopathologist. Wells (1963: 410) himself stated that: ‘It cannot be too</p><p>strongly emphasised that the interpretation of the films must invariably be the work of a</p><p>professionally trained radiologist’. This view was also echoed by others during the 1960s</p><p>(e.g. Sandison, 1968). This attitude reflects the status of palaeopathology at that time as a</p><p>still nascent discipline ‘borrowing’ the skills of others, rather than a fully fledged discipline</p><p>with its own academic traditions, training and skills base. This began to change from</p><p>the 1970s. The Palaeopathology Association was set up as the professional organization</p><p>for practitioners, and this was a vital step in the development of palaeopathology as a</p><p>discipline in its own right. This period also saw the publication of the first textbooks</p><p>aimed specifically at collating diagnostic criteria for the purpose of the identification of</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 79</p><p>disease in ancient skeletal remains (Steinbock, 1976; Zimmerman and Kelley, 1982; Ortner</p><p>and Putschar, 1985). With this avowed aim, radiography could not be ignored, nor could</p><p>it be dismissed as an arcane set of skills which a palaeopathologist could not hope to</p><p>acquire. These handbooks included not only illustrations of dry bone specimens, but also</p><p>radiographic views of clinical cases of bone disease, of pathology museum specimens with</p><p>known conditions, and of diseased archaeological specimens. Palaeopathological diagnostic</p><p>criteria were formulated from a combination of both gross and radiographic features. These</p><p>and subsequent texts (Aufderheide and Rodríguez-Martin, 1998; Ortner, 2003) have helped</p><p>to ensure that radiography became a routine aid to description and diagnosis, and a fully</p><p>integrated core skill in palaeopathology.</p><p>Methodological Issues in Radiographic Imaging in Palaeopathology</p><p>In palaeopathology, lesions need first to be adequately described, both in terms of their</p><p>morphology and in terms of their distribution in the skeleton. Then, generally using recent</p><p>cases of known disease as a baseline, competing diagnoses can be considered and, in some</p><p>cases, a most likely cause determined. Radiography is of value in palaeopathology both in</p><p>description and in diagnosis.</p><p>For all but the most superficial lesions, radiography aids the elucidation of their mor-</p><p>phology by revealing those parts hidden</p><p>by overlying bone (Figure 5.1). Lesions completely</p><p>confined to the bone interior will, of course, be invisible on gross examination in intact</p><p>Figure 5.1 (a) The acetabular roof in this specimen from Wharram Percy shows united fissure</p><p>fractures and several holes in the joint surface which penetrate the subchondral bone. (b) Radiograph</p><p>of the specimen in (a), revealing that the holes in the joint surface communicate with a large lytic</p><p>area within the subchondral bone. This lesion represents a supra-acetabular cyst, most probably due to</p><p>passage of synovial fluid into the bone as a result of the hip injury</p><p>80 Advances in Human Palaeopathology</p><p>Figure 5.1 (Continued)</p><p>specimens, but may be identified on X-ray. Therefore, skeletons suspected from gross exam-</p><p>ination as having some systemic bone disease should be X-rayed in their entirety so that a full</p><p>assessment of the distribution of lesions in surviving elements can be obtained (Figure 5.2).</p><p>Using radiography, the appearance of lesions seen in archaeological skeletons can be</p><p>directly compared with lesion morphology in patients. Radiography provides a direct link</p><p>between the changes seen in palaeopathological specimens and those in living people with</p><p>known diseases. Radiographs from modern patients are useful in aiding palaeopathological</p><p>interpretation; but, in addition, because the technique was invented more than 100 years ago,</p><p>a substantial radiographic record of disease exists in the medical literature from the first</p><p>half of the 20th century. This means that, for plain-film radiography, unlike more recently</p><p>developed imaging modalities such as CT, there is a substantial published corpus depicting</p><p>cases of disease whose progress was unhindered by modern treatment, and such cases more</p><p>closely resemble those which we might expect to encounter in ancient skeletons. These cases</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 81</p><p>Figure 5.2 A case of metastatic carcinoma from Wharram Percy. (a) Endosteal new bone formation</p><p>is visible in some rib elements via post-depositional breaks. (b) Radiography reveals much more</p><p>widespread changes, with areas of patchy sclerosis in most ribs and both clavicles</p><p>are a particularly important augment to our baseline for the diagnosis of palaeopathological</p><p>specimens.</p><p>Post-depositional changes in skeletal remains may cause problems in the interpretation</p><p>of radiographs. Severe soil infiltration may prevent the production of diagnostically useful</p><p>radiographs (e.g. Ortner and Mays, 1998). In other cases it may produce radiographic</p><p>artefacts (consisting of rather ‘fluffy’, irregular areas of radiodensity) even if it does not</p><p>obscure the whole picture (Figure 5.3). Localized soil erosion of cortical or cancellous bone</p><p>may produce radiolucencies which mimic the effects of disease. Post-depositional uptake of</p><p>lead may alter radiographic images, simulating diseases such as osteopetrosis, but noticeable</p><p>effects on plain-film radiographs only occur when lead concentrations exceed about 1.5 %,</p><p>so that this effect is only likely when lead sources lie close to the body, as, for example, in</p><p>individuals interred in lead coffins (Molleson et al., 1998) or buried with lead objects as grave</p><p>goods. The possibility of post-depositional artefacts in palaeopathological radiographs means</p><p>that they should always be examined in conjunction with the specimen itself (Wells, 1967),</p><p>and the palaeopathologist should be aware of the archaeological context of the specimen.</p><p>82 Advances in Human Palaeopathology</p><p>Figure 5.3 Radiograph of the distal parts of a tibia from School Street, Ipswich. There is a shell of</p><p>new bone (involucrum) (1) surrounding the original cortex (sequestrum) (2), which has been partially</p><p>resorbed. The radiolucency (3) is a healed cloaca. These are typical features of chronic osteomyelitis.</p><p>The radiolucency (4) is a post-depositional artefact. The radio-opacities at (5) and elsewhere in the</p><p>metaphysial interior and in the medullary cavity are post-depositional ingress of soil</p><p>Wells (1963: 403) stated that ‘it is difficult to think of any pathological condition which is</p><p>not made more intelligible when the accompanying internal bony changes are revealed’. The</p><p>veracity of this statement is evident in palaeopathology textbooks, which present radiographic</p><p>images illustrating changes in most major classes of disease. However, it is fair to say that</p><p>radiography plays more of a role in the study of some conditions than in others. Radiography</p><p>and allied techniques are clearly vital to the study of conditions such as osteoporosis (Mays,</p><p>Chapter 11), which involve loss of bone mineral without change to the normal anatomical</p><p>shape of the bone. Radiography is also necessary for the study of lesions that are wholly</p><p>hidden within the bone, for example Harris lines (Mays, 1995; Ameen et al., 2005). It is</p><p>also important for the identification of diseases for which gross bony changes are obvious</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 83</p><p>but non-specific. An example is Paget’s disease of bone, where the gross appearance of</p><p>the thickened, pitted bones can resemble a variety of other conditions (e.g. treponematosis).</p><p>Although the gross specimen is difficult to interpret, the radiographic appearance is highly</p><p>characteristic (Mays, Chapter 11).</p><p>Even in diseases where dry bone changes are more helpful in diagnosis, radiography, in</p><p>combination with careful assessment of dry bone morphology, is usually of value. This may</p><p>particularly be the case in disease conditions where clinical diagnosis is heavily based on</p><p>radiographic features. Radiography has, since its introduction into clinical medicine, played a</p><p>major role in the identification of the different arthropathies (Rogers and Waldron, 1995: 3),</p><p>and it continues to be a major diagnostic tool in rheumatology today. It is also important</p><p>in description and differential diagnosis of many types of arthritis in palaeopathology. For</p><p>example, various arthropathies may cause joint ankylosis, but the type of bony union differs in</p><p>different conditions. For example, both DISH and sero-negative spondylolarthropathies may</p><p>cause ankylosis of the sacro-iliac joints. In DISH, bony union takes place via ligamentous</p><p>ossification so that union is confined to the margins and the joint space remains intact</p><p>(Rogers and Waldron, 1995: 50). By contrast, in sero-negative spondyloarthropathies, union</p><p>may occur across the whole joint surface with eventual trabecular continuity between sacrum</p><p>and ilium (Ortner, 2003: 572). Radiography may aid in distinguishing different types of joint</p><p>ankylosis.</p><p>In the radiographic study of lytic lesions, it is important to note not only lesion morphology,</p><p>but also the nature of lesion margins. Radiographically, a slowly developing lytic lesion</p><p>tends to have a margin that is well defined and often shows some sclerosis. In more</p><p>rapidly expanding lesions, the margin may be well defined but lack sclerosis, and very</p><p>aggressive lesions may be poorly circumscribed with a gradient of radiolucency rather</p><p>than a well-defined edge (Ortner, 2003: 52). Thus, for example, slow-growing urate crystal</p><p>deposits (tophi) in gout may lead to lytic lesions at the joints. These effectively represent</p><p>pressure defects of bone (Watt, 1989), and have well-defined, frequently sclerotic margins</p><p>(Figure 5.4). Similarly, benign tumours, which generally grow rather slowly, tend, when</p><p>they</p><p>lead to bony destruction, to produce well-circumscribed lesions with sclerotic margins.</p><p>By contrast, a malignant osteolytic lesion, such as may occur in metastatic carcinoma, may,</p><p>if the cancer is aggressive, produce defects which have poorly defined margins. Because</p><p>it reflects to a great extent the rate at which a lesion was expanding at time of death, the</p><p>radiographic appearance of lytic lesion margins may also vary with the phase of the disease</p><p>when the individual died. For example, in rheumatoid arthritis, margins of erosions are often</p><p>fuzzy in acute-phase disease but are more circumscribed during regressive phases (Jensen</p><p>and Steinbach, 1977).</p><p>Turning to blastic lesions, the distribution of endosteal sclerosis can only be adequately</p><p>evaluated radiographically. As well as aiding the documentation of changes in a specimen,</p><p>the distribution of any endosteal sclerosis also often informs diagnosis (Figure 5.5). The bone</p><p>in blastic lesions tends, if they are slow growing, to be dense and well corticated, whereas</p><p>more rapidly deposited bone tends, if deposited immediately prior to death, to consist</p><p>of poorly structured woven bone. Thus, for example, sub-periosteal bone deposits may</p><p>occur in metastatic carcinoma, particularly of the prostate. Since this is a lethal condition,</p><p>lesions are generally active at time of death and consist of diffuse woven bone of low</p><p>radiodensity. Conversely, sub-periosteal bone in conditions that are not rapidly fatal may</p><p>be well consolidated and of similar radiodensity to normal bone tissue. In instances where</p><p>blastic lesions consist of circumscribed bony overgrowth(s) rather than diffuse periostitis,</p><p>84 Advances in Human Palaeopathology</p><p>Figure 5.4 A case of gout from Barton-on-Humber. There are erosions, some of which show sclerotic</p><p>margins. Reprinted from Journal of Archaeological Science Vol. 14, 1987, Rogers et al., ‘Arthropathies</p><p>in palaeopathology: the basis of classification according to most probable cause’, p. 191. Copyright</p><p>(1987), with permission from Elsevier</p><p>visualization of the internal structure of the ‘lump’ or ‘bump’ by radiography aids diagnosis</p><p>(Figure 5.6).</p><p>For primary bone diseases, clinical radiographic diagnostic criteria can often be directly</p><p>applied to palaeopathological cases. This is the case, for example, with Paget’s disease of</p><p>bone (Mays, Chapter 11) and with skeletal neoplasms. However, for most conditions, the</p><p>application of clinical radiographic diagnostic criteria is less straightforward. For diseases that</p><p>affect soft tissue as well as bone, clinical radiographic diagnostic criteria frequently include</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 85</p><p>Figure 5.5 (a) Vertebral body osteosclerosis (stippled area) in renal osteodystrophy. The band-like</p><p>sclerosis of the inferior and superior parts is termed ‘rugger-jersey spine’. (b) In Paget’s disease of</p><p>bone, sclerosis may occur around the entire periphery of the vertebral body; an appearance termed</p><p>‘picture-frame’ sclerosis. After Resnick and Niwayama (1995: Fig. 64–22)</p><p>Figure 5.6 Internal structure of some circumscribed blastic lesions due to differing causes. (a) Bulging</p><p>of cortical contour due to well-consolidated ossified haematoma. (b) Osteoid osteoma: bulging of</p><p>cortical contour with radiolucent central focus. (c) Osteochondroma: the cancellous and cortical bone</p><p>of the exostosis is continuous with that of the underlying normal bone</p><p>86 Advances in Human Palaeopathology</p><p>changes that cannot be evaluated in the dry specimen. For example, joint-space narrowing is</p><p>a cardinal clinical radiographic diagnostic feature in osteoarthritis (Rogers et al., 1990). In</p><p>dry bones, joint-space narrowing clearly cannot be directly evaluated. In some diseases, the</p><p>clinical radiographic literature emphasizes alterations that can be recorded in palaeopathology</p><p>but for which radiography is redundant as they can be readily seen with the naked eye in dry</p><p>specimens. For example, in rickets, most radiographic clinical diagnostic criteria relate to</p><p>bending deformities, metaphysial broadening, concavity of metaphysial subchondral bone and</p><p>‘fraying’ of bone beneath the epiphysial plate (Thacher et al., 2000; Pettifor, 2003: 555–557).</p><p>These changes are obvious in the dry specimen, so palaeopathologists are selective in their</p><p>application of clinical radiographic diagnostic criteria to dry bones. Radiographic diagnostic</p><p>criteria for rickets in palaeopathology emphasize alterations to the internal bone structure</p><p>(e.g. coarsening/thinning of trabecular structure and loss of cortico-medullary distinction)</p><p>that are not visible grossly on the undamaged specimen (Mays et al., 2006a).</p><p>For diseases where bone changes are superficial and easily seen on gross examination,</p><p>radiography of palaeopathological specimens may be of rather limited value. For some such</p><p>conditions, dry bone diagnostic criteria have been specifically developed with reference</p><p>to radiographic criteria used by clinicians. For example, in osteoarthritis, in addition to</p><p>joint-space narrowing, osteophyte formation, and subchondral cyst formation and sclero-</p><p>sis are important radiographic diagnostic features for clinical cases (Rogers et al., 1990).</p><p>In palaeopathology, osteophyte development and joint surface porosis and eburnation are</p><p>important dry bone indicators of osteoarthritis (Rogers and Waldron, 1995: 44). Joint surface</p><p>porosis and eburnation correspond to subchondral cyst formation and subchondral sclerosis</p><p>identified in clinical radiographs: joint surface pores often communicate with subchondral</p><p>cysts, and eburnated bone is sclerotic on X-ray (Rogers and Waldron, 1995: 44). For other</p><p>conditions, palaeopathological diagnostic criteria have been developed based more on dry</p><p>bone cases than on clinical radiographic features. This is the case, for example, with leprosy</p><p>and scurvy (Ortner, 2003: 263–271, 383–393); radiography plays only a minor role in their</p><p>palaeopathological study.</p><p>A change in bone density of about 40 % is needed before any alteration is visible on</p><p>plain-film radiography (Ortner, 1991). This means that lesions may be invisible on X-ray</p><p>even when they are quite obvious on the dry specimen to the naked eye. Therefore, it is</p><p>often difficult to compare disease frequencies obtained from skeletal remains using dry bone</p><p>palaeopathological criteria with those generated from radiographic surveys of living patients.</p><p>For example, in a study of osteoarthritis at the knee, Rogers et al. (1990) found that only 2 of</p><p>the 24 skeletal specimens they studied were abnormal radiographically, whereas there were</p><p>obvious bony changes of osteoarthritis in 16 on visual assessment. The greater sensitivity</p><p>of visual inspection over radiography for detecting lesions in dry bone specimens (provided</p><p>changes are not entirely confined to the bone interior) also shows the value of, where</p><p>possible, developing dry bone diagnostic criteria for palaeopathology rather than relying on</p><p>clinical radiographic criteria.</p><p>In palaeopathology, most radiographic work is in the study of disease; applications to</p><p>the study of skeletal trauma are rather more limited. In cases of healed fractures, a fracture</p><p>line may be discernable radiographically, and this may aid diagnosis in such cases, but in</p><p>many instances the value of radiography is quite limited as far as diagnosis is concerned.</p><p>Whether a thickened cuff of bone represents a fracture callus is usually obvious to the naked</p><p>eye, particularly where, as is generally the case with fracture union in antiquity, there is</p><p>shortening or abnormal angulation of the bone. In cases</p><p>where the diagnosis is unclear, this</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 87</p><p>is generally because it may be a fracture that occurred long before death, so that remodelling</p><p>of the callus is very thorough (Figure 5.7). In instances where remodelling is advanced it</p><p>is not generally possible to visualize the fracture line on X-ray either; hence, the diagnosis</p><p>cannot be confirmed. Although it is often not very useful for fracture diagnosis, radiography</p><p>may, in cases where remodelling of the lesion is not too thorough, enable visualization of</p><p>the direction of the fracture and the disposition of the broken ends. In this way, the degree of</p><p>displacement on healing and, hence, the efficacy of any splinting or other treatment received</p><p>may be assessed (Grauer and Roberts, 1996). Radiography may also of value for visualization</p><p>of weapon points and other foreign objects embedded in bone. Corrosion often means that it</p><p>is difficult to discern the shape of embedded iron objects, but this may be readily revealed</p><p>on radiography (Figure 5.8).</p><p>RADIOGRAMMETRY</p><p>Measurement of cortical thickness using radiogrammetry was originated over 40 years ago as</p><p>a means of studying bone loss in clinical work on osteoporosis (Barnett and Nordin, 1960).</p><p>Measurements are generally taken, traditionally using callipers or rulers, on radiographs of</p><p>the second metacarpal. This bone is selected for a number of reasons (Garn, 1970: 5–8).</p><p>Adequate views can be obtained from standard postero-anterior hand radiographs. The bone</p><p>section approaches circularity, which means that the method is robust to minor discrepancies</p><p>in bone orientation, and, if desired, cortical area and other biomechanical parameters can</p><p>readily be estimated. Measurements are generally made at the midshaft (Figure 5.9). Total</p><p>bone width T and medullary width M are taken. Cortical thickness is simply given by</p><p>T −M , but for cross-sectional, population studies the cortical index CI = 100 × �T −M�/T</p><p>is preferable as it is a measure of cortical thickness standardized for bone size.</p><p>Major radiogrammetric studies of cortical bone in modern populations were undertaken</p><p>during the 1960s. Garn and co-workers conducted a monumental series of studies on the</p><p>second metacarpal. Looking at various world populations, they investigated both appositional</p><p>growth in cortical thickness in childhood and loss of cortical bone with advancing age in</p><p>adults. This work, summarized in Garn (1970), showed that bone is added to the outer</p><p>surface beneath the periosteum throughout the growth period. During infancy and childhood,</p><p>bone is simultaneously resorbed from the endosteal surface, enlarging the medullary cavity.</p><p>Endosteal resorption generally occurs at a slower rate than periosteal apposition. During</p><p>adolescence, bone is added at both the periosteal and endosteal surfaces. The upshot is that,</p><p>as well as an increase in overall bone width, there is an increase in thickness of cortical</p><p>bone throughout the growth period. Peak cortical thickness is normally attained during early</p><p>adult life. Slight periosteal bone apposition continues throughout adulthood, but, from about</p><p>middle age, bone begins once more to be resorbed from the endosteal surface at a rate</p><p>that outstrips periosteal apposition, leading to thinning of cortical bone. There is greater</p><p>endosteal loss in females, principally as a response to the hormonal changes that accompany</p><p>menopause (Garn, 1970).</p><p>In addition to Garn and colleagues’ work, a number of other studies have been done</p><p>on specific populations to obtain reference values for cortical bone. The largest of these</p><p>is by Virtama and Helelä (1969), who conducted a radiogrammetric study of most of the</p><p>tubular bones of the body in individuals from a Finnish population aged 1–90 years. Studies</p><p>88 Advances in Human Palaeopathology</p><p>Figure 5.7 (a)–(c) Inferior view of some clavicles, showing probable healed fractures. (a) Specimen</p><p>from a burial from Launceston Castle. The fracture line, at midshaft, is clearly evident (arrows). (b)</p><p>Specimen from burial V33 from Wharram Percy. Although less obvious than in the Launceston Castle</p><p>specimen, a fracture line can be discerned (arrows). (c) Specimens from burial V24 from Wharram</p><p>Percy. The right bone is shorter than its antimere and is somewhat thickened in its lateral third,</p><p>suggesting a united fracture in this region (arrow). (d)–(f) Infero-superior radiographs of the specimens</p><p>in (a)–(c). (d) Launceston Castle: as in the gross specimen, the fracture line is clearly evident. (e)</p><p>Wharram Percy, V33. Although the course of the fracture could be traced with careful examination of</p><p>the gross specimen, it is difficult to discern in the radiograph. (f) Wharram Percy, V24. No fracture line</p><p>can be seen. This sequence illustrates the limited value of radiography in the identification of united</p><p>fractures in skeletal remains. In no case did the radiograph add significantly to diagnostic interpretation</p><p>based on gross examination</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 89</p><p>Figure 5.8 (a) Frontal bone from a burial from Ipswich Blackfriars. There is an iron fragment</p><p>embedded in the left side. (b) Radiograph (superior view) showing the morphology of the embedded</p><p>fragment. It appears to be the tip of a projectile point, implement or weapon that has broken off in the</p><p>bone. It fails to penetrate the full thickness of the frontal bone</p><p>Figure 5.9 Schematic diagram of a second metacarpal, showing the methodology for measuring total</p><p>bone width T and medullary width M at the midshaft from postero-anterior radiographs for the purpose</p><p>of calculating cortical index</p><p>specifically of the second metacarpal have also been done on other populations, notably by</p><p>Bugyi (1965) on Hungarians and by Dequeker et al. (1971) on Belgians.</p><p>Measurements of metacarpal cortical bone are of value in investigating osteoporosis, as</p><p>they are a guide to the status of those elements (hip, spine and wrist) most at risk of osteo-</p><p>porotic fracture: metacarpal cortical bone is a useful indicator both of bone density (Adami</p><p>et al., 1996; Dey et al., 2000) and fracture risk (Crespo et al., 1998; Haara et al., 2006)</p><p>at these skeletal sites. Although successful in a research setting, the manual measurement</p><p>technique meant that radiogrammetric quantification of cortical bone was labour intensive</p><p>and time consuming, and these were significant disadvantages for its use for routine clinical</p><p>monitoring of osteoporosis. With the commercial availability of automated photon absorp-</p><p>tiometric methods for measuring bone density in the 1970s, and DXA in the 1980s, the use</p><p>of radiogrammetry declined, although occasional research studies based on its use continued</p><p>to appear in the clinical literature (e.g. Dequeker, 1976; Meema and Meema, 1987; van</p><p>Hemmert et al., 1993; Derisquebourg et al., 1994; Maggio et al., 1997).</p><p>Recently, there has been a revival of interest in radiogrammetry as a diagnostic tool in osteo-</p><p>porosis, due largely to the development of automated computerized versions of the traditional</p><p>manual measurement technique (e.g. Dey et al., 2000; Hyldstrup and Nielsen, 2001; Montalban</p><p>90 Advances in Human Palaeopathology</p><p>et al., 2001; Nielsen, 2001; Rosholm et al., 2001; Ward et al., 2003; Reed et al., 2004; Boonen</p><p>et al., 2005; Haara et al., 2006). Computerized digital image analysis versions of radiogramme-</p><p>try are referred to as digital X-ray radiogrammetry</p><p>Diseases and International Health, Department of Infection, University</p><p>College London, London, UK</p><p>Helen Donoghue received her PhD from the University of Bristol. She spent 6 years at the</p><p>MRC Dental Unit in Bristol, investigating oral microflora. For 4 years she was Lecturer in</p><p>Medical Microbiology at the University of Bradford and is now Senior Lecturer at University</p><p>College London. Her recent research has focused on DNA from pathogenic microorganisms</p><p>in archaeological material, using PCR. Most work has been done on ancient tuberculosis,</p><p>leprosy and, more recently, parasites such as schistosoma and leishmania. Key publications</p><p>include: ‘Widespread occurrence of Mycobacterium tuberculosis DNA from 18th–19th cen-</p><p>tury Hungarians’ (with Fletcher, Holton, Pap and Spigelman), American Journal of Physical</p><p>Anthropology, 2003; ‘Molecular analysis of Mycobacterium tuberculosis from a family of</p><p>18th century Hungarians’ (with Fletcher, Taylor, Van Der Zanden, and Spigelman), Microbi-</p><p>ology, 2003; and ‘Co-infection of Mycobacterium tuberculosis and Mycobacterium leprae in</p><p>human archaeological samples – a possible explanation for the historical decline of leprosy’</p><p>(with Marcsik, Matheson, Vernon, Nuorala, Molto, Greenblatt, and Spigelman), Proceedings</p><p>of the Royal Society of London, Series B, 2005. She is a Fellow of the Royal Society for</p><p>Tropical Medicine and Hygiene, a member of numerous microbiological societies and of the</p><p>Paleopathology Association.</p><p>Contributors xv</p><p>Anne L. Grauer</p><p>Department of Anthropology, Loyola University of Chicago, 6525 N. Sheridan Road,</p><p>Chicago, IL 60626, USA</p><p>Anne Grauer received her PhD from the University of Massachusetts–Amherst in 1989. She</p><p>is currently a Professor in the Department of Anthropology at Loyola University of Chicago.</p><p>Her research focuses on exploring issues of health and disease in medieval England and</p><p>within non-Native American groups in the USA. Of particular interest is the use of doc-</p><p>umentary evidence, combined with skeletal analyses, to understand the lives of women.</p><p>In 1993, she was awarded the National Science Foundation Presidential Faculty Fellow-</p><p>ship to support her incorporation of undergraduate students into scientific research. Key</p><p>publications include: Bodies of Evidence: Reconstructing History Through Skeletal Analy-</p><p>sis (editor, Wiley-Liss, 1995); Sex and Gender in Paleopathological Perspective (co-editor</p><p>Stuart-Macadam, Cambridge University Press, 1999); ‘Where were the women?’ in Human</p><p>Biologists in the Archives, Herring DA, Swedlund AC (eds) (Cambridge University Press,</p><p>2003). She has recently served on the editorial board of the American Journal of Physical</p><p>Anthropology, the Executive Board of the American Association of Physical Anthropologists,</p><p>and as Treasurer and Webmaster of the Paleopathology Association.</p><p>Niels Lynnerup</p><p>Laboratory of Biological Anthropology, The Panum Institute, Blegdamsvej 3, DK-2200</p><p>Copenhagen, Denmark</p><p>Niels Lynnerup received his PhD from the University of Copenhagen, Denmark, in 1995.</p><p>He is currently head of the Laboratory of Biological Anthropology at the Institute of Foren-</p><p>sic Medicine, University of Copenhagen. His biological anthropological research comprises</p><p>both the living (photogrammetry, gait analyses) and the dead (mass-grave analyses, palaeode-</p><p>mography, stable isotopes, aDNA, and the utilization of CT-scanning and 3D visualization</p><p>techniques to study mummies and bog bodies). Key publications include: ‘Life and death</p><p>in Norse Greenland’ in Vikings: The North Atlantic Saga (Smithsonian Institution Press,</p><p>2000); ‘3-D CAT-scan: anthropology, archaeology and virtual reality’ in Archaeological</p><p>Informatics, Pushing the Envelope, (British Archaeological Reports, 2002); ‘Age and fractal</p><p>dimensions of human sagittal and coronal sutures’ (with Brings Jakobsen), American Journal</p><p>of Physical Anthropology, 2003; and ‘Person identification by gait analysis and photogram-</p><p>metry’ (with Vedel), Journal of Forensic Sciences, 2005. He is a member of the board of</p><p>the Forensic Anthropology Society Europe (FASE).</p><p>Simon Mays</p><p>English Heritage Centre for Archaeology, Fort Cumberland, Eastney, Portsmouth PO4</p><p>9LD, UK</p><p>Simon Mays received his PhD from the University of Southampton, England, in 1987. He</p><p>is currently Human Skeletal Biologist for English Heritage and is a Visiting Lecturer at the</p><p>University of Southampton. His research encompasses most areas of human osteoarchaeol-</p><p>ogy. Key publications include: The Archaeology of Human Bones (Routledge, 1998); Human</p><p>Osteology in Archaeology and Forensic Science (Greenwich Medical Media, 2000, co-edited</p><p>with M. Cox); ‘Palaeopathological and biomolecular study of tuberculosis in a mediaeval</p><p>skeletal collection from England’ (with Taylor, Legge, Shaw & Turner-Walker), American</p><p>xvi Contributors</p><p>Journal of Physical Anthropology, 2001; ‘Skeletal manifestations of rickets in infants and</p><p>young children in an historic population from England’ (with Brickley and Ives), American</p><p>Journal of Physical Anthropology, 2006. He is a member of the managing committee of the</p><p>British Association for Biological Anthropology and Osteoarchaeology (BABAO), of the</p><p>Human Remains Advisory Panel of the UK Governmental Department of Culture, Media</p><p>and Sport, and is Secretary of the Advisory Panel on the Archaeology of Christian Burials</p><p>in England.</p><p>Alan Ogden</p><p>Biological Anthropology Research Centre, Department of Archaeological Sciences, Univer-</p><p>sity of Bradford, Bradford BD7 1DP, UK</p><p>Alan Ogden was trained as a dental surgeon, and was a Lecturer/Associate Specialist in</p><p>Restorative Dentistry at Leeds Dental Institute for 20 years, with an especial interest</p><p>in implants, and in 1988 was awarded his doctorate. He then trained in osteology and</p><p>palaeopathology at Bradford and has been a Research Fellow/Contract Osteologist in the</p><p>Bioanthropology Research Centre at the University of Bradford since 2001. He has run</p><p>the postgraduate courses on musculo-skeletal anatomy in 2003 and archaeology of human</p><p>remains in 2004 and regularly lectures on the anatomy and palaeopathology of teeth and</p><p>jaws. He has produced skeletal reports on more than 2000 medieval individuals from Norton</p><p>Priory, Chichester Leper Hospital and Hereford. He has reported on 60 middle Bronze Age</p><p>burials from a British Museum dig in the Lebanon, and he is currently re-examining 50</p><p>Neolithic individuals excavated in the 1860s from barrows in the Yorkshire Wolds. Recent</p><p>joint publications include ‘Tallow Hill Cemetery, Worcester: the importance of detailed study</p><p>of post-mediaeval graveyards’ and ‘A study of Paget’s disease at Norton Priory, Cheshire,</p><p>England’ British Archaeological Reports International Series 2005; ‘Morbidity, rickets, and</p><p>long-bone growth in post-medieval Britain’, Annals of Human Biology, 2006; ‘Gross enamel</p><p>hypoplasia in subadults from a 16th–18th Century London graveyard’, American Journal of</p><p>Physical Anthropology (in press). A former Curator and Honorary Member of the British</p><p>Society for the Study of Prosthetic Dentistry, he is a member of the American Association</p><p>of Physical Anthropologists, the Paleopathology Association and the British Association for</p><p>Biological Anthropology and Osteoarchaeology.</p><p>Donald J. Ortner</p><p>Department of Anthropology, National Museum of Natural History, Smithsonian Institution,</p><p>Washington, DC 20560, USA</p><p>Donald J Ortner holds a PhD in physical anthropology from the University of Kansas, USA,</p><p>and an honorary DSc degree from the University of Bradford, England. He is a biological</p><p>anthropologist in the Department of Anthropology, National Museum of Natural History,</p><p>Smithsonian Institution, USA, where he has worked during most of his professional career.</p><p>In 1988 he was appointed a Visiting Professor in the Department of Archaeological</p><p>(DXR) in the clinical literature. A frequently</p><p>used DXR method in clinical research is the Pronosco system, originated in Denmark. The soft-</p><p>ware uses an algorithm to locate ‘regions of interest’ (ROIs) around the narrowest parts of the</p><p>diaphyses of the second, third and fourth metacarpals from a standard postero-anterior hand</p><p>radiograph. In each metacarpal, more than 100 measurements are performed of T and M within</p><p>the ROI, and a weighted average of results from the three bones is used to calculate an over-</p><p>all cortical index (Hyldstrup and Nielsen, 2001). If desired, the system can combine cortical</p><p>thickness with analyses of porosity of cortex from the radiograph to produce an estimate of bone</p><p>mineral density (BMD; Böttcher et al., 2006). As well as providing a more rapid process than the</p><p>manual measurement technique, this helps to minimize method error (Hyldstrup and Nielsen,</p><p>2001). Owing to the methodological differences, DXR results are not directly comparable to</p><p>those generated from traditional, manual radiogrammetry. To my knowledge, DXR has yet to</p><p>be applied to archaeological human remains</p><p>Radiogrammetry in Palaeopathology</p><p>Population studies on cortical bone thickness began to be undertaken in palaeopathology soon</p><p>after the pioneering studies of Garn and others on modern populations (e.g. Dewey et al.,</p><p>1969; van Gerven et al., 1969; Armelagos et al., 1972). Most palaeopathological studies</p><p>used radiogrammetry, but some, particularly the early work, used direct measurement of cut</p><p>sections of bone (e.g. Dewey et al., 1969; Armelagos et al., 1972; Hummert, 1983; van</p><p>Gerven et al., 1985). Bone sectioning methods have the advantage that they allow direct</p><p>measurement of cortical areas, but the results using such methods are not directly comparable</p><p>to those generated by radiogrammetry on living subjects, and in any event cause damage to</p><p>collections that is unacceptable in most curatorial situations.</p><p>There have been two principal foci for studies of cortical thickness in palaeopathology.</p><p>First, echoing the work done on living subjects, studies in adults have been used to investigate</p><p>osteoporosis, and both peak cortical bone thickness and age-related patterns of its loss have</p><p>been investigated (Mays, Chapter 11). Second, the quantity of cortical bone accumulated</p><p>during growth has been used as a stress indicator in skeletal populations. Although, other than</p><p>in extremis, there is no relationship in adults between skeletal maintenance and diet either</p><p>in terms of protein or caloric intake (Garn, 1970), a consistent link has been found between</p><p>poor childhood nutrition and deficient appositional bone growth (Adams and Berridge, 1969;</p><p>Garn et al., 1969; Barr et al., 1972; Himes et al., 1975). Most workers studying cortical bone</p><p>as a stress indicator have focused on patterns of increase in cortical thickness with age in</p><p>juveniles (e.g. Cook, 1979; Huss-Ashmore et al., 1982; Hummert, 1983; van Gerven et al.,</p><p>1985; Mays, 1985, 1995, 1999), but some studies taking this perspective have compared</p><p>peak cortical bone levels in adults from different populations (Hatch et al., 1983; Pfeiffer</p><p>and King, 1983; Owsley, 1991; Rewekant, 2001).</p><p>Most palaeopathological osteoporosis work using radiogrammetry has used the second</p><p>metacarpal, reflecting this bone’s pre-eminence in monitoring cortical bone loss in osteo-</p><p>porosis in modern populations. A few studies have investigated loss of cortical bone in other</p><p>elements, principally the femur (e.g. Ekenman et al., 1995; Mays et al., 1998). By contrast,</p><p>most of the studies using cortical thickness as an indicator of non-specific stress during the</p><p>growth period have studied long-bones, most often the femur. That major long-bones rather</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 91</p><p>Table 5.1 Repeatability of cortical index taken at the midshaft of the second metacarpal and femur</p><p>from antero-posterior radiographs, and of DXA measurements of BMD in the proximal femur, in adult</p><p>skeletons from Wharram Percy, England. Data from Mays (1996) and Mays et al. (1998)a</p><p>Variable N Sm S R</p><p>MCCI 10 2.270 11.52 0.040</p><p>FEMCI 9 1.280 9.180 0.019</p><p>BMDN 144 0.017 0.192 0.008</p><p>BMDW 144 0.026 0.239 0.012</p><p>aMCCI: metacarpal cortical index; FEMCI: femur cortical index; BMDN; BMD at femur neck, BMDW: BMD</p><p>at Ward’s triangle; BMD values in g cm−2. N : number of repeat measurements, Sm: standard deviation of the</p><p>measurement (method error statistic); S: sample standard deviation; R: variance of measurement/sample variance,</p><p>i.e. S2</p><p>m/S.</p><p>than the metacarpal have been used at least in part probably reflects difficulties in taking</p><p>reliable measurements on the small metacarpals of young individuals, measurement error</p><p>being proportionately greater for smaller measurements.</p><p>In skeletal studies, provided that specimens showing soil erosion are excluded, accurate</p><p>measurements of cortical thickness are readily obtained using radiogrammetry. Metacarpal</p><p>radiogrammetry on archaeological specimens produces results which are closely comparable</p><p>to those obtained clinically. The metacarpal can be readily positioned on the film so that</p><p>its orientation mimics that of the bone in hand radiographs of living subjects, and given</p><p>the fairly small soft tissue thicknesses, differential radiographic enlargement of T and M</p><p>between clinical and dry bone studies is likely minor; the effect of radiographic enlargement</p><p>on CI is likely to be negligible.</p><p>The repeatability of radiogrammetric measurements on palaeopathological specimens</p><p>appears good. Replicability data for metacarpal and femoral cortical index in adult skeletons</p><p>from a large archaeological assemblage (from Wharram Percy, England), with re-radiography</p><p>of specimens, and analysed using the method error statistic, are shown in Table 5.1. The</p><p>results indicate that measurement error made up approximately 2 % and 4 % of sample vari-</p><p>ance for femoral and metacarpal cortical indices respectively. That intra-observer error is</p><p>low has been confirmed by other workers on archaeological material (e.g. Lazenby, 2002;</p><p>Ives and Brickley, 2004). However, inter-observer error may be a problem if workers are</p><p>using different criteria for defining the endosteal border in their measurement of medullary</p><p>width (Ives and Brickley, 2004).</p><p>BONE DENSITOMETRY</p><p>The oldest technique for measuring bone density from radiographs is photodensitometry</p><p>(Mack et al., 1939). A standard of known density, usually an aluminium step-wedge, is</p><p>exposed in the radiograph alongside the bone, and the standard used to estimate bone</p><p>density using an optical densitometer (Lees et al., 1998). This method has been little used</p><p>in palaeopathology (although see Ives and Brickley (2005)).</p><p>Single, and later dual, photon absorptiometry were introduced commercially to measure</p><p>bone density in osteoporosis in a clinical setting in the 1970s. These techniques used a</p><p>radionuclide source to generate photons, the attenuation of which by bone was used to</p><p>92 Advances in Human Palaeopathology</p><p>assess density (Lees et al., 1998). Some early palaeopathological work on osteoporosis used</p><p>photon absorptiometry (e.g. Perzigian, 1973; Laughlin et al., 1979), but the technology is</p><p>now obsolete.</p><p>DXA has replaced photon densitometry as a means of measuring bone density, and it is</p><p>currently the ‘gold standard’ by which bone loss is assessed clinically in osteoporosis (Kanis</p><p>and Glüer, 2000). DXA uses an X-ray source that emits beams at two different energy</p><p>levels. In the living patient this enables differing attenuation due to bone and soft tissue</p><p>to be calculated. The areas scanned are generally those sites which are most vulnerable to</p><p>osteoporotic fracture, namely the hip, spine and forearm. For the site scanned, the computer</p><p>software in the machine selects a standard ROI within which bone density is measured.</p><p>For example, for the hip the standard ROIs are the femoral neck, Ward’s triangle and the</p><p>greater trochanter. The bone mineral content in the ROI is measured, and the result divided</p><p>by the area scanned. This gives an ‘areal’ density (grams per square centimetre) rather</p><p>than a true volumetric density (Lees et al., 1998). Because of this, DXA BMD results are</p><p>not fully normalized for bone size. Larger bones will give greater apparent densities than</p><p>smaller ones simply because the X-rays have passed through a greater thickness of bone.</p><p>This will merely result in a minor degree of random noise in cross-sectional studies in adult</p><p>populations, but it is more of a difficulty when studying age-related change in BMD in</p><p>children (Nelson and Koo, 1999). To deal with this problem, formulae have been derived to</p><p>estimate volumetric densities taking into account bone dimensions (e.g. Kröger et al., 1992;</p><p>Boot et al., 1997).</p><p>Dual X-Ray Absorptiometry in Palaeopathology</p><p>DXA scanners became commercially available to clinicians in 1987 (Banks, 2001), and</p><p>within a few years the first DXA work on archaeological bone, to investigate osteoporosis,</p><p>was being undertaken (Hammerl et al., 1990). Since then, DXA has been quite widely</p><p>used to study osteoporosis in ancient populations (Mays, Chapter 11). The value of BMD</p><p>measured using DXA as a stress indicator in juvenile skeletons has also recently begun to</p><p>be investigated (McEwan et al., 2005).</p><p>In general, palaeopathological studies of BMD in osteoporosis using DXA are conducted</p><p>with the aim of comparing patterns seen in the study population with those in other groups,</p><p>either a modern reference population or some other ancient population (Mays, Chapter 11). A</p><p>general problem with making comparisons between different studies is that, due to hardware</p><p>and software differences between different DXA scanners, absolute BMD values from dif-</p><p>ferent machines cannot be directly compared (Genant et al., 1994; Hui et al., 1997; Boonen</p><p>et al., 2003). In order to do such comparisons, a cross-calibration between machines needs</p><p>to be carried out. In a clinical setting, this may be accomplished using a bone phantom, a</p><p>calibration standard containing inserts of different density (Lees et al., 1998). Alternatively,</p><p>cross-calibration data may be derived by scanning the same set of subjects in different</p><p>machines (e.g. Hui et al., 1997). Similarly, in a palaeopathological setting, a subset of the</p><p>specimens under study can be scanned on different machines and cross-calibration equations</p><p>obtained (Mays et al., 2006b).</p><p>In DXA scanners, attenuation of X-rays by soft tissue is taken into account by the computer</p><p>software when BMD is calculated. This means that, for scanning, archaeological specimens</p><p>generally need to be placed in a material whose density approximates to that of soft tissue,</p><p>such as water (Kneissel et al., 1994) or dry rice (Mays et al., 1998). Nevertheless, because</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 93</p><p>archaeological specimens lack marrow and soft tissue, absolute BMD values cannot be</p><p>compared with those on living subjects (Lees et al., 1993; Chappard et al., 2004). This</p><p>means that peak BMD cannot be compared between ancient and modern subjects. However,</p><p>provided that significant diagenetic changes in bone density in archaeological specimens</p><p>can be excluded, valid comparisons of age-related patterns in BMD between ancient and</p><p>modern data can be made and peak BMD can be compared between different skeletal</p><p>populations.</p><p>The possibility of diagenetic change in density of buried bone, either due to physical</p><p>contamination with extraneous matter from the burial environment or due to chemical</p><p>or microstructural alteration, is perhaps the most significant potential problem in bone</p><p>densitometry studies in palaeopathology using DXA or other methods. One way of evaluating</p><p>the possibility of diagenetic change in BMD is with reference to the results themselves.</p><p>For example, in cases where patterning in BMD is as expected on physiological grounds</p><p>(e.g. decline in BMD with age in both sexes but greater in females, greater and earlier age-</p><p>related loss of BMD at Ward’s triangle than at the femur neck) then this suggests that any</p><p>post-depositional change in BMD is minor on the basis that it is highly unlikely that some</p><p>complex pattern of differential diagenesis can have fortuitously reproduced such patterns</p><p>(e.g. Mays et al., 1998). However, even in cases such as this it is desirable that there be</p><p>some independent evidence that diagenesis in BMD has not occurred to any great extent. To</p><p>check for gross physical contamination, specimens should be screened prior to scanning for</p><p>the presence of internal soil infiltration using plain-film radiography (e.g. Mays et al., 1998).</p><p>The possibility of minor soil infiltration, too subtle to be evident on plain-film radiographs,</p><p>can be assessed using microscopic study. Chemical analysis or spectroscopic study of bone</p><p>samples is also of value to detect intrusive minerals (Mays et al., 2006b).</p><p>Microstructural diagenetic deterioration of bone, of the type described by Hackett (1981)</p><p>and Hedges et al. (1995), has the potential to alter bulk bone density. Relatively little work</p><p>has been done to determine whether it actually does, although a study has recently been</p><p>undertaken involving comparison of two medieval populations, from Norway and England</p><p>(Mays et al., 2006b). The bones from Norway (Trondheim) showed excellent microstructural</p><p>preservation of bone. Those from England (Wharram Percy) showed severe microstructural</p><p>diagenesis. Nevertheless, DXA revealed similar absolute BMD values, and similar age-related</p><p>patterns in BMD, in the two groups. Were microstructural diagenesis an important influence</p><p>on BMD, then some difference between the two groups, at opposite ends of the spectrum</p><p>of diagenetic alteration, would have been expected. Consistent with the notion that the</p><p>microstructural diagenesis in the Wharram Percy bones had not caused appreciable change</p><p>in bulk density, examination of histological sections (Turner-Walker and Syversen, 2002)</p><p>showed that alterations had resulted in discrete areas of hyper- and hypo-mineralization,</p><p>visible as areas of different tonal values under scanning electron microscopy. Elemental</p><p>analysis indicated that, although diagenesis caused dissolution and re-precipitation of bone</p><p>mineral, movement of bone mineral in this process was a highly localized phenomenon and</p><p>bulk calcium content was little changed (Turner-Walker and Syversen, 2002; Mays, 2003).</p><p>In summary, existing work clearly demonstrates that valid bone densitometry studies can</p><p>be carried out on ancient skeletal remains. However, the possibility of diagenetic change in</p><p>BMD should always be evaluated in the skeletal population under study.</p><p>The replicability of DXA results on ancient remains appears good. For example, repeat</p><p>scannings, with repositioning of specimens, on the Wharram Percy adults suggests that about</p><p>1–2 % of sample variance consisted of measurement error (Table 5.1). That the precision of</p><p>94 Advances in Human Palaeopathology</p><p>DXA on ancient remains is good is also supported by studies on other assemblages (Lees</p><p>et al., 1993; Poulsen et al., 2001).</p><p>Other Densitometric Techniques in Palaeopathology</p><p>Some more recently developed techniques can also produce BMD estimates from bone.</p><p>Energy-dispersive low-angle X-ray scattering (EDLAXS) is one such technique. This tech-</p><p>nique has some advantages over DXA. It produces an estimate of true (volumetric) BMD,</p><p>rather than the areal BMD of DXA, so that results are truly size-standardized. By adjust-</p><p>ing the measurement</p><p>area, BMD purely for trabecular bone can be obtained if desired. In</p><p>addition, the spectrum generated by EDLAXS of different minerals is unique. This means</p><p>that the presence of different minerals in a bone sample can be recognized and their quan-</p><p>tities calculated. It is, therefore, another way of investigating diagenesis. The technique</p><p>has been trialled using archaeological bone, and a good correlation with physical measures</p><p>of density has been reported (Farquharson et al., 1997; Farquharson and Brickley, 1997).</p><p>However, EDLAXS has no clinical application, precluding comparisons with living subjects,</p><p>and the equipment is not commercially available. For these reasons, future applications in</p><p>palaeopathology are likely to be limited.</p><p>Quantitative CT (qCT) is a method that has become widely used for measuring BMD in</p><p>clinical research. Like EDLAXS, it provides an estimate of volumetric BMD and permits the</p><p>isolation of a specific volume of bone so that exclusively trabecular bone density can be mea-</p><p>sured if desired. However, as with DXA, the fact that skeletal remains lack marrow and soft</p><p>tissue means that bones need to be placed in soft-tissue equivalents (e.g. water) for scanning,</p><p>and absolute BMD values are not directly comparable between skeletal remains and living</p><p>subjects. The high cost and higher radiation dose than for DXA have restricted the clinical</p><p>application of qCT (Lees et al., 1998; Banks, 2001; Kanis, 2002; Moyad, 2003). A small and</p><p>rather inconclusive pilot study (Gonzalez-Reimers et al., 2007) has been conducted on qCT</p><p>in ancient bones, but further work is needed to assess adequately its value in palaeopathology.</p><p>CONCLUSIONS</p><p>Plain-film radiography has traditionally been the most important augment to gross exami-</p><p>nation for identifying and interpreting skeletal lesions in palaeopathology. It seems likely</p><p>that this will continue to be the case for the foreseeable future. CT scanning may increase</p><p>in importance, but cost considerations mean that its use in palaeopathology is unlikely to</p><p>become routine. In time, digital image capture will doubtless oust film-based radiography,</p><p>as has been the case with photography. Provided that resolution is adequate, real-time radio-</p><p>graphy provides the potential for more rapid screening of remains for pathological lesions.</p><p>The increasing availability of portable digital radiographic equipment facilitates radiographic</p><p>work at locations that lack radiographic facilities, such as field situations.</p><p>Radiogrammetry offers a straightforward technique by which cortical bone can be quan-</p><p>tified. It provides data that are closely comparable to those obtained on living subjects,</p><p>facilitating comparison between ancient and modern populations. It uses facilities (plain-</p><p>film radiography, callipers) readily available to most palaeopathologists. In a clinical setting,</p><p>the rather time-consuming nature of traditional manual radiogrammetry was a significant</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 95</p><p>disadvantage, and the revival of clinical radiogrammetry was heralded by the advent of</p><p>automated DXR techniques. However, it remains to be seen whether currently available</p><p>DXR systems, such as Pronosco, can be applied to dry bones. In some ways DXR would</p><p>seem to offer few advantages to palaeopathologists over manual radiogrammetry. Rapidity,</p><p>which is the great advantage of DXR in a clinical context, is less critical in a research</p><p>setting. Clinically, the decreased method error of DXR is important, but this is less so</p><p>palaeopathologically, where the study is of patterning in cross-sectional population samples</p><p>rather than, for example, attempting to detect change in sequential radiographs of individual</p><p>patients. In addition, method errors introduced by imprecision in manual radiogrammetry</p><p>are often minor compared with those introduced into work on palaeopopulations by impre-</p><p>cision in other methods; for example, limitations in skeletal age determination, rather than</p><p>method error in radiogrammetry, are the major limitation on the resolution with which we</p><p>can study age-related patterns of decline in cortical bone in osteoporosis. The Pronosco</p><p>system of DXR relies on measurements from three metacarpals, a requirement that will</p><p>reduce sample size in fragmentary archaeological remains. Despite these disadvantages, it</p><p>would be useful to investigate the applicability of DXR to skeletal remains, if only so</p><p>that archaeological data can be compared with data currently being generated on modern</p><p>populations.</p><p>Turning to the assessment of bone density, advantages of using DXA in the study of</p><p>osteoporosis in archaeological populations are that it is the clinical ‘gold standard’ by which</p><p>the BMD in osteoporosis is assessed, and it can be used to estimate bone density specifically</p><p>at the skeletal sites vulnerable to osteoporotic fracture. A disadvantage in its use in skeletal</p><p>remains is that, because absolute BMD values are not directly comparable to those obtained</p><p>from living patients, only patterns of age-related bone loss rather than peak bone density can</p><p>be compared between living and skeletal populations. An additional problem (also shared</p><p>by other densitometric methods) is the potential for diagenesis to alter the density of buried</p><p>bone. Although review of the published literature reveals little evidence that diagenetic</p><p>change in BMD is a pervasive problem in palaeopathological studies, it is a possibility that</p><p>should be investigated in each archaeological population under study.</p><p>REFERENCES</p><p>Adami S, Zamberlan N, Gatti G, Zanfisi C, Braga V, Broggini M. 1996. Computed radiographic</p><p>absorptiometry and morphometry in the assessment of post-menopausal bone loss. Osteoporos Int</p><p>6: 8–13.</p><p>Adams P, Berridge SR. 1969. Effects of kwashiorkor on cortical and trabecular bone. Arch Dis Child</p><p>44: 705–709.</p><p>Ameen S, Staub L, Ulrich S, Vock P, Ballmer F, Anderson SE. 2005. Harris lines of the tibia across</p><p>two centuries: a comparison of two populations, medieval and contemporary in central Europe. Skel</p><p>Radiol 34: 279–284.</p><p>Armelagos GJ, Mielke JH, Owen KH, van Gerven DP, Dewey JR, Mahler PE. 1972. Bone growth and</p><p>development in prehistoric populations from Sudanese Nubia. J Hum Evol 1: 89–119.</p><p>Aufderheide AC, Rodríguez-Martin RM. 1998. The Cambridge Encyclopaedia of Paleopathology.</p><p>Cambridge University Press: Cambridge.</p><p>Banks LM. 2001. Dual energy X-ray absorptiometry (DXA). In Diagnostic Radiology: A Textbook of</p><p>Medical Imaging (4th edn), vol. 1, Grainger RG, Allison DJ, Adam A, Dixon AK (eds). Churchill</p><p>and Livingstone: London; 185–195.</p><p>96 Advances in Human Palaeopathology</p><p>Barnett E, Nordin BEC. 1960. The radiological diagnosis of osteoporosis: a new approach. Clin Radiol</p><p>11: 166–174.</p><p>Barr DGD, Shmerling DH, Prader A. 1972. Catch-up growth in malnutrition studied in celiac disease</p><p>after institution of a gluten-free diet. Pediatr Res 6: 521–527.</p><p>Böni T, Rühli FJ, Chhem RK. 2004. History of paleoradiology: early published literature, 1896–1921.</p><p>Can Assoc Radiol J 55: 203–210.</p><p>Boonen S, Kaufman J-M, Reginster J-Y, Devogelaer J-P. 2003. Patient assessment using stan-</p><p>dardized bone mineral density values and a national reference database: implementing uniform</p><p>thresholds for the reimbursement of osteoporosis treatments in Belgium. Osteoporos Int 14:</p><p>110–115.</p><p>Boonen S, Nijs J, Borghs H, Peeters H, Vanderschueren D, Luyten FP. 2005. Identifying post-</p><p>menopausal women with osteoporosis by calcaneal ultrasound, metacarpal digital radiogrammetry</p><p>and phalangeal radiographic absorbtiometry: a comparative study. Osteoporos Int 16: 93–100.</p><p>Boot AM, de Ridder MAJ, Pols HAP, Krenning EP, Keizer-Schrama SMPFdeM. 1997. Bone mineral</p><p>density in children and adolescents: relation to puberty,</p><p>calcium intake and physical activity. J Clin</p><p>Endocrinol Metab 82: 57–62.</p><p>Böttcher J, Pfeil A, Rosholm A, Schäfer M-L, Malich A, Petrovich A, Seidl B, Lehman G, Mentzel H-J,</p><p>Hein G, Wolf G, Kaiser WA. 2006. Computerized digital imaging techniques provided by digital</p><p>X-ray radiogrammetry as new diagnostic tool in rheumatoid arthritis. J Digital Imag 19: 279–288.</p><p>Bugyi B. 1965. Über Alternsabhängigkeit des Kortikalen Index des II Metakarpalknochens. Anat Anz</p><p>116: 378–383.</p><p>Chappard D, Moquereau M, Mercier P, Gallois Y, Legrand E, Baslé MF, Audran M. 2004. Ex vivo</p><p>bone mineral density of the wrist: influence of medullar fat. Bone 34: 1023–1028.</p><p>Clendinnen FJ. 1898. Skiagram of hand of Egyptian mummy, showing abnormal number of sesamoid</p><p>bones. Intercolon Med J Austr 3: 106.</p><p>Cook DC. 1979. Subsistence base and health in prehistoric Illinois Valley: evidence from the human</p><p>skeleton. Med Anthropol 4: 109–124.</p><p>Crespo R, Revilla M, Usabiago J, Crespo E, Garcia-Areno J, Villa LF, Rico H. 1998. Metacarpal</p><p>radiogrammetry by computed radiography in postmenopausal women with Colles’ fracture and</p><p>vertebral crush fracture syndrome. Calcif Tissue Int 62: 470–473.</p><p>Dequeker J. 1976. Quantitative radiology: radiogrammetry of cortical bone. Br J Radiol 49: 912–920.</p><p>Dequeker J, Remans J, Franssen R, Waes W. 1971. Ageing patterns of trabecular and cortical bone</p><p>and their relationship. Calcif Tissue Res 7: 23–30.</p><p>Derisquebourg T, Dubois B, Devogelaer JP, Meys E, Duquesnoy B, Nagant de Deuxchaisnes C,</p><p>Delacambre B, Marchandise X. 1994. Automated computerised radiogrammetry on the second</p><p>metacarpal and its correlation with absorptiometry of the forearm and spine. Calcif Tissue Int 54:</p><p>461–465.</p><p>Dewey JR, Bartley MH, Armelagos GJ. 1969. Rates of femoral cortical bone loss in two Nubian</p><p>populations. Clin Orthop 65: 61–66.</p><p>Dey A, McCloskey EV, Taube T, Cox R, Pande KC, Ashford RU, Forster M, de Takats D, Kanis JA.</p><p>2000. Metacarpal morphometry using a semi-automated technique in the assessment of osteoporosis</p><p>and vertebral fracture risk. Osteoporos Int 11: 953–958.</p><p>Eaton GF. 1916. The collection of osteological material from Machu Picchu (Peru). Mem Conn Acad</p><p>Arts Sci 5: 1–96.</p><p>Ekenman I, Eriksson SAV, Lindgren JU. 1995. Bone density in mediaeval skeletons. Calcif Tissue Int</p><p>56: 355–358.</p><p>Farquharson MJ, Brickley M. 1997. Determination of mineral composition of archaeological bone</p><p>using energy-dispersive low-angle X-ray scattering. Int J Osteoarchaeol 7: 95–99.</p><p>Farquharson MJ, Speller RD, Brickley M. 1997. Measuring bone mineral density in archaeolog-</p><p>ical bone using energy dispersive low angle X-ray scattering techniques. J Archaeol Sci 24:</p><p>765–772.</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 97</p><p>Fiori MG, Nunzi MG. 1995. The earliest documented application of X-rays to the examination of</p><p>mummified remains and archaeological materials. J R Soc Med 88: 76–69.</p><p>Garn SM. 1970. The Earlier Gain and Later Loss of Cortical Bone. Charles C Thomas: Springfield, IL.</p><p>Garn SM, Guzman MA, Wagner B. 1969. Subperiosteal gain and endosteal loss in protein – calorie</p><p>malnutrition. Am J Phys Anthropol 30: 153–156.</p><p>Genant HK, Grampp S, Glüer CC, Faulkner KG, Jergas M, Engelke K, Hagiwara S, van Kuijk C.</p><p>1994. Universal standardization for dual X-ray absorptiometry: patient and phantom cross-calibration</p><p>results. J Bone Miner Res 9: 1503–1514.</p><p>Gonzalez-Reimers E, Velasco-Vázquez J, Arnay de la Rosa M, Machado-Calvo M. 2007. Quantitative</p><p>computerized tomography for the diagnosis of osteopenia in prehistoric skeletal remains. J Archaeol</p><p>Sci 34: 554–561.</p><p>Grauer AL, Roberts CA. 1996. Palaeoepidemiology, healing and possible treatment of trauma in the</p><p>mediaeval cemetery population of St Helen-on-the-Walls, York, England. Am J Phys Anthropol 100:</p><p>531–544.</p><p>Haara M, Heliövaara M, Impivaara O, Arokoski JPA, Manninen P, Knekt P, Kärkkäinen A, Reunanen</p><p>A, Aromaa A, Kröger H. 2006. Low metacarpal index predicts hip fracture. Acta Orthop 77:</p><p>9–14.</p><p>Hackett CJ. 1981. Microscopic focal destruction (tunnels) in exhumed bone. Med Sci Law 21:</p><p>243–265.</p><p>Hammerl J, Protsch R, Happ J, Frohn J, Hor G. 1990. Osteodensitometrie des Femurhalses an his-</p><p>torischen Skeletten. In Osteologie-Interdisziplinar, Werner R, Mattiass HH (eds). Springer: Berlin;</p><p>139–142.</p><p>Hatch JW, Willey PS, Hunt EE. 1983. Indicators of status-related stress in Dallas society: transverse</p><p>lines and cortical thickness in long bones. Midcontinent J Archaeol 8: 49–71.</p><p>Hedges REM, Millard AR, Pike AWG. 1995. Measurements and relationships of diagenetic alteration</p><p>of bone from three archaeological sites. J Archaeol Sci 22: 201–209.</p><p>Himes JH, Matorell R, Habicht J-P, Yarbrough C, Malina RM, Klein RE. 1975. Patterns</p><p>of cortical bone growth in moderately malnourished preschool children. Hum Biol 47:</p><p>337–350.</p><p>Hui SL, Gao S, Zhou X-H, Johnson CC, Lu Y, Glüer CC, Genant HK. 1997. Universal standardisation</p><p>of bone density measurements: a method with optimal properties for calibration among several</p><p>instruments. J Bone Miner Res 12: 1463–1470.</p><p>Hummert JR. 1983. Cortical bone growth and dietary stress among subadults from Nubia’s Batn el</p><p>Hajar. Am J Phys Anthropol 62: 167–176.</p><p>Huss-Ashmore R, Goodman AH, Armelagos GJ. 1982. Nutritional inference from paleopathology. Adv</p><p>Archaeol Meth Theor 5: 395–474.</p><p>Hyldstrup L, Nielsen SP. 2001. Metacarpal index by digital X-ray radiogrammetry. J Clin Densitom</p><p>4: 299–306.</p><p>Ives R, Brickley MB. 2004. A procedural guide to metacarpal radiogrammetry in archaeology. Int J</p><p>Osteoarchaeol 14: 7–17.</p><p>Ives R, Brickley M. 2005. Metacarpal radiogrammetry: a useful indicator of bone loss throughout the</p><p>skeleton? J Archaeol Sci 32: 1552–1559.</p><p>Jensen PS, Steinbach HL. 1977. Roentgen features of the rheumatic diseases. Med Clin North Am 61:</p><p>389–404.</p><p>Kanis JA. 2002. Diagnosis of osteoporosis and assessment of fracture risk. Lancet 359: 1929–1936.</p><p>Kanis JA, Glüer CC. 2000. An update on the diagnosis and assessment of osteoporosis with</p><p>densitometry. Osteoporos Int 11: 192–202.</p><p>Kneissel M, Boyde A, Hahn M, Teschler-Nicola M, Kalchhauser G, Plenk H. 1994. Age- and sex-</p><p>dependent cancellous bone changes in a 4000y BP population. Bone 15: 539–545.</p><p>Kröger H, Kotaniemi A, Vainio P, Alhava E. 1992. Bone densitometry of the spine and femur in</p><p>children by dual-energy X-ray absorptiometry. Bone Miner 17: 75–82.</p><p>98 Advances in Human Palaeopathology</p><p>Lang J, Middleton A, Ambers J, Higgins T. 2005. Radiographic images. In Radiography of Cultural</p><p>Material (2nd edition), Lang J, Middleton A (eds). Elsevier: Oxford; 20–48.</p><p>Laughlin WS, Harper AB, Thompson DD. 1979. New approaches to the pre- and post-contact history</p><p>of Arctic peoples. Am J Phys Anthropol 51: 579–587.</p><p>Lazenby RA. 2002. Circumferential variation in human second metacarpal cortical thickness: sex, age</p><p>and mechanical factors. Anat Rec 267: 154–158.</p><p>Lees B, Molleson T, Arnett TR, Stevenson JC. 1993. Differences in proximal femur bone density over</p><p>two centuries. Lancet 341: 673–675.</p><p>Lees B, Banks LM, Stevenson JC. 1998. Bone mass measurements. In Osteoporosis, Stevenson JC,</p><p>Lindsay R (eds). Chapman and Hall: London; 137–160.</p><p>Mack PB, O’Brien AT, Smith JM, Bauman AW. 1939. A method for estimating the degree of</p><p>mineralisation of bones from tracings of roentgenograms. Science 89: 467.</p><p>Maggio D, Pacifici R, Cherubini A, Simonelli G, Luchetti M, Aisa MC, Cucinotta D, Adami</p><p>S, Senin U. 1997. Age-related cortical bone loss at the metacarpal. Calcif Tissue Int 60:</p><p>94–97.</p><p>Mays SA. 1985. The relationship between Harris line formation and bone growth and development. J</p><p>Archaeol Sci 12: 207–220.</p><p>Mays SA. 1995. The relationship between Harris lines and other aspects of skeletal development in</p><p>adults and juveniles.</p><p>J Archaeol Sci 22: 511–520.</p><p>Mays S. 1996. Age-dependent cortical bone loss in a medieval population. Int J Osteoarchaeol 6:</p><p>144–154.</p><p>Mays S. 1999. Linear and appositional long bone growth in earlier human populations: a case study</p><p>from mediaeval England. In Human Growth in the Past: Studies Using Bones and Teeth, Hoppa</p><p>RD, Fitzgerald CM (eds). Cambridge University Press: Cambridge; 290–312.</p><p>Mays SA 2003. Bone strontium:calcium ratios and duration of breastfeeding in a mediaeval skeletal</p><p>population. J Archaeol Sci 30: 731–741.</p><p>Mays S, Lees B, Stevenson JC. 1998. Age-dependent cortical bone loss in the femur in a medieval</p><p>population. Int J Osteoarchaeol 8: 97–106.</p><p>Mays S, Brickley M, Ives R. 2006a. Skeletal manifestations of rickets in infants and young children</p><p>in an historic population from England. Am J Phys Anthropol 129: 362–374.</p><p>Mays S, Turner-Walker G, Syversen U. 2006b. Osteoporosis in a population from mediaeval Norway.</p><p>Am J Phys Anthropol 131: 343–351.</p><p>McEwan J, Mays S, Blake G. 2005. The relationship of bone mineral density and other growth</p><p>parameters to stress indicators in a mediaeval juvenile population. Int J Osteoarchaeol 15:</p><p>155–163.</p><p>Means JH. 1925. A roentgenological study of skeletal remains of the prehistoric Mound Builder Indians</p><p>of Ohio. Am J Roentgenol 13: 359–367.</p><p>Meema HE, Meema S. 1987. Postmenopausal osteoporosis: simple screening method for diagnosis</p><p>before structural failure. Radiology 164: 405–410.</p><p>Molleson TI, Williams CT, Cressey G, Din VK. 1998. Radiographically opaque bones from lead-lined</p><p>coffins at Christ Church, Spitalfields, London – an extreme example of bone diagenesis. Bull Soc</p><p>Geol France 169: 425–432.</p><p>Montalban SJ, Rico LH, Cortes PJ, Pedrera ZJD. 2001. Cortical bone mass and risk fac-</p><p>tors for osteoporosis among postmenopausal women in our environment. Rev Clin Esp 201:</p><p>16–20.</p><p>Moyad MA. 2003. Osteoporosis: a rapid review of risk factors and screening methods. Urol Oncol 21:</p><p>375–379.</p><p>Nelson DA, Koo WWK. 1999. Interpretation of absorptiometric bone mass measurements in the</p><p>growing skeleton: issues and limitations. Calcif Tissue Int 65: 1–3.</p><p>Nielsen SP. 2001. The metacarpal index revisited: a brief overview. J Clin Densitom 4:</p><p>199–207.</p><p>Radiography and Allied Techniques in the Palaeopathology of Skeletal Remains 99</p><p>Ortner DJ. 1991. Theoretical and methodological issues in paleopathology. In Human Paleopathology:</p><p>Current Syntheses and Future Options, Ortner DJ, Aufderheide AC (eds). Smithsonian Institution</p><p>Press: Washington; 5–11.</p><p>Ortner DJ. 2003. Identification of Pathological Conditions in Human Skeletal Remains (2nd edition).</p><p>Academic Press: New York.</p><p>Ortner DJ, Mays S. 1998. Dry-bone manifestations of rickets in infancy and early childhood. Int J</p><p>Osteoarchaeol 8: 45–55.</p><p>Ortner DJ, Putschar WGJ. 1985. Identification of Pathological Conditions in Human Skeletal Remains.</p><p>Smithsonian Institution Press: Washington, DC.</p><p>Owsley DW. 1991. Temporal variation in femoral cortical thickness of North Ameri-</p><p>can Plains Indians. In Human Paleopathology: Current Syntheses and Future Options,</p><p>Ortner DJ, Aufderheide AC (eds). Smithsonian Institution Press: Washington, DC;</p><p>105–110.</p><p>Pettifor JM. 2003. Nutritional rickets. In Pediatric bone. Biology and Diseases, Glorieux F, Pettifor J,</p><p>Jüppner H (eds). Academic Press: New York; 541–565.</p><p>Pfeiffer S, King P. 1983. Cortical bone formation and diet among protohistoric Iroquoians. Am J Phys</p><p>Anthropol 60: 23–28.</p><p>Perzigian AJ. 1973. Osteoporotic bone loss in two prehistoric Indian populations. Am J Phys Anthropol</p><p>39: 87–96.</p><p>Poulsen LW, Qvesel D, Brixen K, Versterby A, Boldsen JL. 2001. Low bone mineral density in the</p><p>femoral neck of medieval women: a result of multiparity. Bone 28: 454–458.</p><p>Reed MR, Murray JRD, Abdy SE, Francis RM, McCaskie AW. 2004. The use of digital X-ray</p><p>radiogrammetry and peripheral dual X-ray absorptiometry in patients attending fracture clinic after</p><p>distal forearm fracture. Bone 34: 716–719.</p><p>Resnick D, Niwayama G. 1995. Diagnosis of Bone and Joint Disorders (3rd edition). WB Saunders:</p><p>London.</p><p>Rewekant A. 2001. Do environmental disturbances of an individual’s growth and development influence</p><p>the later bone involution process? A study of two medieval populations. Int J Osteoarchaeol 11:</p><p>433–443.</p><p>Rogers J, Waldron T. 1995. A Field Guide to Joint Disease in Archaeology. Wiley: Chichester.</p><p>Rogers J, Watt I, Dieppe P. 1990. Comparison of visual and radiographic detection of bony changes</p><p>at the knee joint. Br Med J 300: 367–368.</p><p>Rosholm A, Hyldstrup L, Baeksgaard L, Grunkin M, Thodberg HH. 2001. Estimation of bone mineral</p><p>density by digital X-ray radiogrammetry: theoretical background and clinical testing. Osteoporos Int</p><p>12: 961–969.</p><p>Sandison AT. 1968. Pathological changes in the skeletons of earlier populations due to acquired</p><p>disease, and difficulties in their interpretation. In The Skeletal Biology of Earlier Human Populations,</p><p>Brothwell DR (ed.). Pergamon: Oxford; 205–243.</p><p>Spiegel PK. 1995. The first clinical X-ray made in America. Am J Roent 164: 241–243.</p><p>Steinbock RT. 1976. Paleopathological Diagnosis and Interpretation. Charles C Thomas: Spring-</p><p>field, IL.</p><p>Thacher TD, Fischer PR, Pettifor JM, Lawson JO, Manaster BJ, Reading JC. 2000. Radiographic</p><p>scoring method for the assessment of the severity of nutritional rickets. J Trop Pediatr 46: 132–139.</p><p>Turner-Walker G, Syversen U. 2002. Quantifying histological changes in archaeological bones using</p><p>BSE-SEM image analysis. Archaeometry 44: 461–468.</p><p>Van Gerven DP, Armelagos GJ, Bartley MH. 1969. Roentgenographic and direct measurement of</p><p>femoral cortical involution in a prehistoric Mississippian population. Am J Phys Anthropol 31:</p><p>23–38.</p><p>Van Gerven DP, Hummert JR, Burr DP. 1985. Cortical bone maintenance and geometry of</p><p>the tibia in prehistoric children from Nubia’s Batn el Hajar. Am J Phys Anthropol 66:</p><p>275–280.</p><p>100 Advances in Human Palaeopathology</p><p>Van Hemert AM, Vanderbroucke JP, Hofman H, Valkenberg HA. 1993. Metacarpal bone loss in</p><p>middle-aged women: “horse racing” in a 9-year population based follow up study. J Clin Epidemiol</p><p>43: 579–588.</p><p>Virtama P, Helelä T. 1969. Radiographic measurements of cortical bone. Variation in a normal</p><p>population between 1 and 90 years of age. Acta Radiol 293(Suppl): 1–268.</p><p>Ward KA, Cotton J, Adams JE. 2003. A technical and clinical evaluation of digital X-ray</p><p>radiogrammetry. Osteoporos Int 14: 389–395.</p><p>Watt I. 1989. The radiology of erosions. In The Antiquity of the Erosive Arthropathies, Dieppe P,</p><p>Rogers J (eds). Arthritis and Rheumatism Council for Research, Conference Proceedings No. 5.</p><p>Arthritis and Rheumatism Council: London; 18–20.</p><p>Wells C. 1963. The radiological examination of human remains. In Science in Archaeology, Brothwell</p><p>D, Higgs E (eds). Thames and Hudson: London; 401–412.</p><p>Wells C. 1967. Pseudopathology. In Diseases in Antiquity, Brothwell DR, Sandison AT (eds). Charles</p><p>C Thomas: Springfield, IL; 5–19.</p><p>Williams HU. 1929. Human paleopathology with some original observations of symmetrical</p><p>osteoporosis of the skull. Arch Path 7: 839–902.</p><p>Zimmerman MR, Kelley MA. 1982. Atlas of Human Paleopathology. Praeger: New York.</p><p>6</p><p>Computed Tomography</p><p>Scanning and</p><p>Three-Dimensional</p><p>Visualization of Mummies</p><p>and Bog Bodies</p><p>Niels Lynnerup</p><p>Laboratory of Biological Anthropology, The Panum Institute,</p><p>Blegdamsvej 3, DK-2200 Copenhagen, Denmark</p><p>INTRODUCTION</p><p>Mummies and bog bodies present unique opportunities for the palaeopathologist. While</p><p>many pathological processes may be identified in the skeleton, bones are overall a ‘slow-</p><p>reacting’ tissue, so that pathological</p><p>changes mainly reflect chronic diseases (Ortner, 2003).</p><p>The presence of soft tissues may expand the scope of pathological studies, so that more</p><p>acute diseases and diseases that do not affect bone tissue may be identified (Cockburn and</p><p>Cockburn, 1980). The mummification and preservation of the various soft tissues, though,</p><p>is very variable. Interior organs, particularly of the digestive system, are often completely</p><p>decomposed, and organs may be very shrunken and difficult to identify morphologically due</p><p>to desiccation. Furthermore, various funerary rites that include embalming may entail the</p><p>complete removal of internal organs, including the brain (as seen for Egyptian mummies).</p><p>Generally, the most commonly preserved soft tissues are those with a high content of</p><p>collagen, such as the dermis, muscle fasciae and tendons. Still, even just the presence of</p><p>skin may give important clues to pathology and trauma; for example, penetrating wounds</p><p>and cuts, scars and even warts (Lowenstein, 2004).</p><p>However, the very presence of soft tissue, especially the skin, makes it difficult to examine</p><p>the body. One may conduct an autopsy, but while such mummy autopsies have been carried</p><p>out often (David, 1979; Aufderheide, 2003), most archaeologists, conservators, physical</p><p>anthropologists and pathologists are reluctant to permit this. An autopsy is an invasive</p><p>Advances in Human Palaeopathology Edited by Ron Pinhasi and Simon Mays</p><p>© 2008 John Wiley & Sons, Ltd. ISBN: 978-0-470-03602-0</p><p>102 Advances in Human Palaeopathology</p><p>and destructive procedure; the integrity of the mummy or bog body as an archaeological</p><p>specimen may be destroyed. Attitudes have perhaps shifted from an earlier, more clinical,</p><p>medical approach to mummies and bog bodies (in a certain sense seeing the body as any</p><p>other unidentified body that must be properly examined in a forensic way) to one that views</p><p>a mummy or a bog body as an invaluable archaeological artefact.</p><p>X-RAYS AND MUMMIES</p><p>The most well-known non-invasive technique for visualizing the internal structures of mum-</p><p>mies and bog bodies employs X-rays. Indeed, the first use of X-radiography in mummy</p><p>research occurred only 1 year after William Röntgen discovered X-rays in 1895 (Koenig,</p><p>1896). X-rays are electromagnetic waves. Because they are generated by the conversion of</p><p>the energy acquired by electrons accelerated through an electrical field gradient, they are</p><p>more often characterized by their photon energies. The electrical field is produced by an</p><p>anode and cathode in the X-ray tube. The cathode is made of a tungsten filament, and this</p><p>emits the electrons. When the electrons strike the anode, a small fraction of the energy is</p><p>emitted as X-rays. The X-rays then pass out of the tube, through adjustable diaphragms</p><p>in order to limit the beam to the minimally required area one wishes to examine (col-</p><p>limation). The intensity of the X-rays produced at a given voltage is determined by the</p><p>number of electrons impacting the anode, and is expressed in milliamperes. The X-ray dose</p><p>is proportional to the time during which the beam current flows. Intensity and dosage are</p><p>major parameters in clinical use, as X-rays are potentially hazardous due to their ioniz-</p><p>ing properties. As the X-rays pass through the object to be examined, they interact with</p><p>the object, being either scattered or absorbed. The combined effect of this is expressed as</p><p>attenuation. After passing through the object, the X-rays then strike a photographic plate.</p><p>Since the X-rays are thus attenuated differentially by different tissues and materials, they end</p><p>up striking the photographic plate at different intensities, resulting in differing grey-values.</p><p>After development, these are rendered from white (tissues or materials with high attenuation,</p><p>e.g. bone) to black (e.g. air in body cavities). Modern clinical radiographic equipment is</p><p>now fully digitized, i.e. the X-rays do not end on a photographic plate but rather in special</p><p>sensors which translate the attenuation to direct pixel-based images. The digital equipment</p><p>has the advantage of being able to control the beam modalities and the image building</p><p>more directly, enabling sharper pictures (Fleckenstein and Tranum-Jensen, 1993; Carlton</p><p>and Adler, 2001).</p><p>Radiography has the advantage of being almost universally available and easily performed.</p><p>As mentioned above, radiography has a long track record when it comes to mummies and</p><p>bog bodies, in effect having been the only method available for ‘looking inside’. Flinders</p><p>Petrie used X-rays in the studies of mummies in 1897 (Petrie and Griffith, 1898). Moodie</p><p>published the findings of his analyses of 17 Egyptian mummies at the Field Museum in</p><p>Chicago in 1931; this probably constitutes the first systematic radiographic analysis of a</p><p>major mummy collection (Moodie, 1931). All the royal mummies housed at the Cairo</p><p>Museum were radiographed in 1967 (Harris and Weeks, 1973; Harris and Wente, 1980).</p><p>Radiography of mummies has also been performed on site in some very remote locations</p><p>(Notman et al., 1987; Notman and Beattie, 1995).</p><p>Computed Tomography Scanning and 3D Visualization 103</p><p>The primary purpose of radiographing mummies has often been archaeological as well</p><p>as medical; for example, searching for amulets in Egyptian mummy wrappings (Chris-</p><p>tensen, 1969). However, the determination of sex and age, based on skeletal traits, has</p><p>usually been carried out whenever possible (e.g. Fawcitt et al., 1984). Grafton Elliot Smith</p><p>evaluated the epiphyseal union of the mummy of Tuthmosis IV in order to ascertain age</p><p>at death (Smith, 1912). Cranial traits, and cranial morphometry based on radiographs,</p><p>have been studied in order to try to establish kinship (Harris and Wente, 1980). This</p><p>pre-empted one of the uses of computed tomography (CT) scanning: the production of</p><p>three-dimensional solid models of skulls for facial reconstruction purposes (see below).</p><p>Neave (1979) managed to make a facial reconstruction of an Egyptian mummy based on</p><p>radiographs.</p><p>The pathological changes observed by X-raying mummies include arthritis, atheroma,</p><p>healed fractures, and parasite-induced changes (Brothwell and Sandison, 1967; Christensen,</p><p>1969; David 1979; Harris and Wente, 1980; Bloomfield, 1985). Most of the pathological</p><p>processes observed reflect diseases affecting bone or calcified structures, such as a radiodense</p><p>structure in an Egyptian mummy that was revealed to be a guinea worm parasite at a</p><p>later autopsy (David, 1979). This is a reflection of the limitations of radiography. Skeletal</p><p>structures are usually easily identifiable due to the high attenuation, but it may be very</p><p>difficult to discriminate between various soft-tissue remains, especially as these will be</p><p>superimposed upon another on the radiograph. In radiography, a three-dimensional structure</p><p>is presented in two dimensions, so the skin and fasciae of the ventral aspect of a mummy will</p><p>be superimposed over the remains of both the interior organs, membranes and fasciae, and</p><p>the skin and fasciae of the dorsal aspect. Since the soft tissues already may have near equal</p><p>attenuation coefficients, this inhibits greatly the possibilities for investigating soft-tissue</p><p>pathology.</p><p>COMPUTED TOMOGRAPHY SCANNING</p><p>A further development of radiographic imaging came with the advent of CT scanners</p><p>(computer-assisted tomography). Unlike conventional X-ray images, the regions of interest</p><p>in CT scannings are presented without superimpositions of juxtaposed structures. The X-ray</p><p>tube revolves around the item being analysed (Figure 6.1). Coupled with a specific array</p><p>for reading out the X-ray attenuation, this means that a CT scanner will generate an image</p><p>that represents a slice of an item (Figure 6.2). The item is moved through the CT scanner,</p><p>generating</p><p>hundreds or thousands of slices, which may be viewed as serial, two-dimensional</p><p>slice images (Hsieh, 2002).</p><p>The development of the first clinical CT scanners began in 1967 with Godfrey Hounsfield</p><p>(1976), and the first clinical CT scanner was installed in 1971. The first patient scanned was</p><p>diagnosed with a large cyst (Ambrose, 1975). Hounsfield was awarded the Nobel Prize for</p><p>his work, and he shared it with Allan M. Cormack, who had undertaken pioneering work on</p><p>the mathematical basis for image reconstruction. Since the introduction of the first scanners,</p><p>many technical advances have been made. These have produced much finer images and much</p><p>more rapid image acquisition times (from more than 1 min to less than 0.1 s per slice) (Hsieh,</p><p>2002). The advances in CT scanners are generally recorded as ‘generations’. First-generation</p><p>scanners had only one beam being recorded at a time, whereas second-generation scanners</p><p>had multiple beams, enabling a faster recording time. Third-generation scanners were able</p><p>104 Advances in Human Palaeopathology</p><p>Figure 6.1 CT scanning the Borremose Woman (photograph: N. Lynnerup)</p><p>to irradiate the entire object, and with the introduction of multislice scanners achieved better</p><p>volume rendering (Hsieh, 2002).</p><p>The CT scanner generates the slice images as an array of pixels (usually 512 × 512), with</p><p>each pixel having a value depending on the attenuation of the X-rays as they pass through the</p><p>object being scanned (Carlton and Adler, 2001). The attenuation is represented by Hounsfield</p><p>units (HU), which are scaled and calibrated arbitrarily according to the attenuation of water,</p><p>so that water has an HU of 1000 and air will have an HU of −1000 on the Hounsfield scale.</p><p>The Hounsfield units are then converted to a grey scale potentially covering 256 shades</p><p>of grey. This is more than the human eye can discern, so, consequently, the attenuation is</p><p>remapped to about 20 grey-scale shades (image reconstruction). This minimizing of data is</p><p>offset by adjusting the window width (i.e. the number of HUs differentiated on the grey-scale</p><p>Computed Tomography Scanning and 3D Visualization 105</p><p>Figure 6.2 A single slice through a Ptolemaic Period Egyptian mummy (Carlsberg Gyptotek Museum,</p><p>Copenhagen, Denmark). The many layers of wrapping are clearly seen. The scanning is of the head, and</p><p>the cranial vault is clearly visible. The brain has been removed, as was customary for the embalming</p><p>techniques (although there are some remains in the back of the vault). This has been done through the</p><p>nasal aperture and ethmoid, and other bony structures in this area are indeed fractured. Since the brain</p><p>has been removed, the CT scanner renders the cranial cavity as black. The skull fracture seen in the</p><p>back of the vault is post-mortem</p><p>shades in the reconstructed remap) and the window level (the midpoint value of the window).</p><p>Adjusting window width and window level enables viewing the structures of interest at the</p><p>best possible resolution (Fleckenstein and Tranum-Jensen, 1993).</p><p>CT scanning of mummies was first carried out in 1977 (Harwood-Nash, 1977; Lewin and</p><p>Harwood-Nash, 1977). Eleven mummies at the Museum of Fine Arts in Boston were CT</p><p>scanned by Marx and D’Auria (1986) in what was then the most systematic CT scanning</p><p>of a collection of mummies. Increasing computing power enabled imaging with an ever-</p><p>finer resolution and rapid three-dimensional visualizations (e.g. Marx and D’Auria, 1988;</p><p>Pickering et al., 1990). Basically, the CT-scanner computer ‘restacks’ the single slices and,</p><p>based on different algorithms, connects object boundaries between the slices (Zollikofer</p><p>and Ponce de León, 2005). CT scanners used in hospitals come equipped with various</p><p>computer programs that allow rapid image building and three-dimensional visualization, but</p><p>it must be noted that the software is developed for medical purposes, and tuned to tissues</p><p>and organ systems of the living. For example, the Hounsfield units associated with living</p><p>106 Advances in Human Palaeopathology</p><p>bone do not vary much (pathology aside), so it is easy to preprogram a three-dimensional</p><p>rendering of the skeletal tissues in a CT scanner, so that, for example, the skull of a scanned</p><p>patient can be visualized just by a click on the appropriate menu. Mummies, and especially</p><p>bog bodies may require much more work at this stage (see below). While the earlier CT</p><p>scanners operated with a slice thickness of several millimetres, new CT scanners can achieve</p><p>slice thicknesses of less than 1 mm, as well as the capacity for ‘overlapping’. This is an</p><p>important parameter for three-dimensional visualization, since the slice thickness will affect</p><p>the quality of the three-dimensional image. If the slice thickness is several millimetres or</p><p>more, then the resulting image will have a ‘terraced’ appearance. Obviously, this may also</p><p>lead to misinterpretations of a three-dimensional structure, so it is always important to note</p><p>the slice thickness when appraising CT scans (Hsieh, 2002; Zollikofer and Ponce de León,</p><p>2005). Large, detailed studies on embalming techniques and mummy wrappings of Egyptian</p><p>mummies, employing three-dimensional imagery, have lately been performed by Hoffman</p><p>et al. (2002) and Jansen et al. (2002). Recently, three Inca mummies were found on top of</p><p>Mount Llullaillaco in Argentina. Owing to the extreme altitude, the 500-year old mummies</p><p>were frozen. This meant that the internal organs were exceptionally well preserved and easily</p><p>visible upon CT scanning (Previgliano et al., 2003).</p><p>Three-dimensional visualization may be an important tool for the physical anthropologist.</p><p>Since CT scanning is based on X-rays, bone is generally easy to visualize. This means</p><p>that skeletal structures may be viewed so that they can be assessed morphologically, as</p><p>well as measured. Visualizing the skull and pelvis makes it possible to assess the sex-</p><p>specific traits virtually and use many of the methods described for ‘real’ bones (e.g. Buikstra</p><p>and Ubelaker, 1994; Mays, 1998; White and Folkens, 2005). Virtual ageing may also be</p><p>performed by focusing on dental development and epiphyseal morphology (Buikstra and</p><p>Ubelaker, 1994; Mays, 1998; White and Folkens, 2005).</p><p>The benefits of three-dimensional visualization also apply to the assessment of possi-</p><p>ble pathological changes. For example, a mummy that had already been scanned in 1983</p><p>(Vahey and Brown, 1984) was re-scanned in 1990 in order to make three-dimensional visu-</p><p>alizations of the skull and pelvis to confirm a possible fracture (Pickering et al., 1990).</p><p>Dental disease has also been visualized in this way (Melcher et al., 1997), as has an Egyp-</p><p>tian mummy where an intravital, foreign object (presumably a toe prosthesis) was found</p><p>(Nerlich et al., 2000). However, given all the mummies that have been CT scanned, patho-</p><p>logical or traumatic finds have been rather sparse. Probably one of the more well-known</p><p>cases involved the CT scanning of the Iceman, a glacier mummy found in Italy in 1992.</p><p>The mummy had been X-rayed and CT scanned after the find (zur Nedden and Wicke,</p><p>1992). However, not until a renewed scanning 10 years later was an arrowhead identified</p><p>in the left shoulder region (Gostner and Vigl, 2002). Rheumatoid arthritis has been diag-</p><p>nosed based on CT-visualized bone erosions and joint subluxation (Ciranni et al., 2002).</p><p>A CT scan of a natural mummy from the 19th century AD from a friary in Italy revealed</p><p>a distended bladder and a ring of dense tissue at the site of the prostate, indicative of</p><p>prostatic hyperplasia (Fornaciari et al., 2001). Bone tumours have been identified in two</p><p>Egyptian</p><p>mummies (Taconis and Maat, 2005). Bone pathology as observed by CT scanning,</p><p>indicative of tuberculosis, has been correlated with ancient-DNA analyses (Pap, personal</p><p>communication). It should be mentioned that CT scanning has also been used to locate patho-</p><p>logical processes or specific organs, in order to make precise incisions to perform biopsies</p><p>(Brothwell et al., 1990; Bennike, 2003), or to guide endoscopic examinations (Rühli et al.,</p><p>2002).</p><p>Computed Tomography Scanning and 3D Visualization 107</p><p>Problems in Mummy and Bog Body Computed Tomography Scanning</p><p>and Three-Dimensional Visualization</p><p>The presence of many layers of mummy wrappings, especially those closely adhering to the</p><p>skin, as well as organ removal, etc., may impede the interpretation of the body structures of</p><p>a mummy. Diagenetic changes, mainly desiccation, may also have an impact. However, the</p><p>skeletal structures are usually intact in both natural and artificial mummies. This means that</p><p>structures useful for ageing and sexing may still be visualized. Availability of the skeletal</p><p>images also results in having many reference points, which makes it easier to assess the</p><p>remains of internal organs and structures.</p><p>Special problems of radiography and CT scanning arise when the diagenetic changes</p><p>are so massive that the remaining tissues, including bone, are severely degraded. This is</p><p>perhaps best shown by bog bodies. Owing to the acidic bog environment, calcium is leached</p><p>from the bones, causing demineralization of the bony tissues, which consequently lose</p><p>their hardness and become pliable (van der Sanden, 1996). When a bog body is X-rayed,</p><p>bones are often very badly visualized. The bones have an appearance as if they were made</p><p>of glass. This may be demonstrated by the radiographs taken of the Grauballe Man, a</p><p>Danish bog body from the Iron Age, excavated in 1950 and X-rayed in 1955 (Figure 6.3).</p><p>Consequently, when CT scanning a bog body, applying the same range of Hounsfield</p><p>units for bones as in clinical work may result in the bones not being visualized at all.</p><p>Furthermore, the demineralization is not necessarily uniform, but may differ within the</p><p>skeletal system or a single bone due to the diagenetic microenvironment. This may generate</p><p>a patchy appearance of the bone, even though it is intact morphologically. The second</p><p>Figure 6.3 X-ray from 1955 of the left leg of the Grauballe Man (from Munck (1956))</p><p>108 Advances in Human Palaeopathology</p><p>difference is the fact that other tissues seem to acquire a more radiodense structure (i.e.</p><p>the attenuation of the X-ray beams is increased). This is especially seen in some of the</p><p>connective tissues, e.g. ligaments, fasciae and the subcutis. The reason for this is probably</p><p>a deposition of mineral salts (containing metals such as iron) from the soil in collagenous</p><p>tissues.</p><p>In the Iron Age, bogs in northern Europe had a special significance. Not only could they</p><p>be difficult (even dangerous) to pass, but they were also used for rituals and sacrifices.</p><p>The people of the Iron Age probably sacrificed weapons, boats, chariots, animals, and even</p><p>humans in the bogs. They knew that people thrown in would be ‘swallowed’ up by the</p><p>wet and cold bogs. They probably even knew that special preservation might take place</p><p>in the bogs, so that textiles, leather and even human bodies might be preserved for a long</p><p>time. Today, many bogs have disappeared, having been drained and then made into arable</p><p>land. One other use of the bogs was for peat. Most northern European finds of bog bodies</p><p>were discovered as a result of peat cutting (van der Sanden, 1996). In Denmark, peat was</p><p>excavated throughout the last few centuries, but it was especially in the 1930s–1950s that</p><p>peat was excavated on a large scale. It is from this period that most bog bodies were</p><p>found, as accidental finds. Peat is still excavated in Ireland, and the extent to which the bog</p><p>environment may have an effect on bog bodies can be illustrated by our work on three such</p><p>recently found bog bodies from Ireland (‘Clony Cavan Man’, ‘Old Man Croghan’ and ‘Derry</p><p>Cashel’). These bog bodies have been subjected to a series of scientific studies, including</p><p>CT scanning. The head of Clony Cavan Man shows extreme lateral flattening, with a facial</p><p>breadth of only approximately 2–3 cm. The humeri of Old Man Croghan are an example</p><p>of differential preservation: the right humerus is intact enough to be readily recognizable,</p><p>whereas the left humerus is very difficult to visualize (Figure 6.4).</p><p>Many bog bodies are also shrunken. Whereas some shrinkage may take place in the</p><p>bog, most is probably due to the drying out of bog bodies after excavation. Drying out</p><p>was previously the only preservation and conservation method for ‘wet’ bog bodies. In the</p><p>mid-20th century, more targeted methods were used, such as attempts to substitute the water</p><p>with alcohols and tannic oils, bark extracts and wax, while today freeze-drying is used.</p><p>Figure 6.4 Segmentation of the bones of the upper limbs of the Irish bog body ‘Old Man Croghan’.</p><p>Note the difference in preservation of the humeri</p><p>Computed Tomography Scanning and 3D Visualization 109</p><p>Whatever the method, some shrinkage is probably inevitable; and it may be pronounced,</p><p>especially for the older bog finds.</p><p>Thus, diagenetically modified tissues and organs may have to be delineated (or segmented)</p><p>manually on almost every single slice. At present, features allowing manual slice-by-slice</p><p>image editing are seldom available on the computer programs of the (clinical) CT scan-</p><p>ners. The images, therefore, need to be transferred to another program that allows editing</p><p>(post-image-capture processing) of the CT data.</p><p>Post-Capture Processing of Computed Tomography Scan Images</p><p>Several computer programs exist that allow post-image-capture processing of the CT data,</p><p>especially the very important ability to edit the single image-slices manually. At our</p><p>laboratory, we currently use a program package (MIMICS®) from Materialise®, Belgium</p><p>(www.materialise.be). After import of the CT-scanner data file, the single-slice images may</p><p>be shown. The program also allows for immediate multiplanar reformatting (Figure 6.5).</p><p>The single image elements may then be edited. In order to visualize, for example, the head</p><p>of the Grauballe Man, a colour coding is applied ‘over’ the single grey-scale pixels. Dif-</p><p>ferent tissues and organs may be given specific colours (Figure 6.5). This usually involves</p><p>a process of identifying the single structures on the single-slice images and then following</p><p>these structures through on the adjacent slices. For example, the bones of the skull may</p><p>Figure 6.5 Program interface of MIMICS® showing a CT slice image being segmented by application</p><p>of ‘masks’ (colour coding) based on thresholding of the grey-scale values</p><p>110 Advances in Human Palaeopathology</p><p>first be picked out, then the brain, etc. All the single slices have to be edited, so that that</p><p>ultimately the individual structures are completely delineated (segmented). The colour-coded</p><p>pixels associated with the various structures may then be extracted, and used as a basis for</p><p>three-dimensional rendering.</p><p>The manual process of delineating and extracting the relevant items (tissues, organs, etc.)</p><p>necessitates a certain anatomical knowledge, the more so as the bones often are bent and</p><p>deformed and organs are often shrunken. There is some room for subjectivity when perform-</p><p>ing the segmentation. Along with the previously stated</p><p>caveat on slice thickness (and indeed</p><p>on the inherent properties of X-ray-based image acquisition), this means that CT-scanning</p><p>images and three-dimensional renderings should not be viewed as a totally objective and</p><p>‘true’ representation of internal structures and tissues. Some pathological processes may be</p><p>falsely ascribed to diagenetic processes, and vice versa.</p><p>Once tissues and organs have been segmented and visualized, they can then be measured</p><p>and assessed morphologically. An initial benefit is the visualization of bone structures that</p><p>may be used for sexing and ageing. We segmented the Grauballe Man pelvis into the single</p><p>constituent bones and were thus able to display the auricular surface (Figure 6.6). Although</p><p>the resolution does not, for example, allow a direct application of the auricular ageing</p><p>technique (Lovejoy et al., 1985), some information may be salvaged.</p><p>Although some bog bodies have preservation of soft tissues that allow direct sexing, other</p><p>bog bodies, especially body parts, rely on identifying certain features associated with the</p><p>male and female skeleton. This also includes measurements. The Frer foot is a foot in a</p><p>leather shoe found in a Danish bog without any other associated body parts. We visualized</p><p>the calcaneus and were then able to measure the dimensions directly and set this in relation</p><p>to several tables. However, applying the raw measures in tables and formulae developed</p><p>on normal bones must be done with some caution. As noted above, not only do bones</p><p>become deformed in a bog, they may also shrink. When we analysed the Grauballe Man, we</p><p>noted an overall 10 % decrease in several bone measurements (both cranial and post-cranial</p><p>measurements), compared with the mean values of non-bog skeletons from the same period.</p><p>It is then uncertain whether this 10 % decrease reflects diagenetic change, or whether the</p><p>Grauballe Man was smaller than other people of his day and age.</p><p>Figure 6.6 The pelvis and an isolated innominate showing the auricular surface: Grauballe Man</p><p>Computed Tomography Scanning and 3D Visualization 111</p><p>Bog Body Lesions and Trauma</p><p>Grauballe Man was first examined in 1950. Forensic pathological and radiographic analyses</p><p>at that time showed that he had had his throat cut, and probably had sustained a cranial fracture</p><p>and a tibial fracture perimortally. This was interpreted by the archaeologists as evidence of</p><p>a sacrificial execution, whereby he had his leg broken in order to incapacitate him, then a</p><p>blow to his head and finally a sharp, mortal wound to his throat (Munck, 1956; Glob, 1965).</p><p>We CT scanned the body in 2001, and after image segmentation it was possible to</p><p>visualize the throat organs (Figure 6.7). The hyoid bone and the larynx were not fractured.</p><p>This indicates that when the Grauballe Man had his throat slit, that this was done with his</p><p>head held back. This supports the theory of execution: by holding his head back, his throat</p><p>was exposed, and his throat was then slit from ear to ear, probably by a person also standing</p><p>at his back. A direct blow frontally to his throat, for example by a sword or axe, would more</p><p>probably have fractured or lacerated the larynx or hyoid.</p><p>On the other hand, the previously noted cranial fracture did not seem as certain. When</p><p>visualizing the cranium, it was clearly seen how the appearance was more suggestive of</p><p>post-mortem influences (Figure 6.8). Unlike the Clony Cavan Man, the head had not been</p><p>crushed completely, but had bilateral impressions. On the right side, the bone was also</p><p>Figure 6.7 A three-dimensional rendering of hyoid and larynx: Grauballe Man</p><p>112 Advances in Human Palaeopathology</p><p>Figure 6.8 The skull of the Tollund Man (right) and the Grauballe Man (left). Note the impressions</p><p>in the parietal regions of both crania</p><p>fractured and bent inward. The fracture did not resemble a peri-mortem fracture as seen</p><p>in modern clinical cases. We felt this could be substantiated when we studied another bog</p><p>body. The Tollund Man is from the same time period as Grauballe Man, and found in a bog</p><p>near that of the Grauballe Man. Tollund Man’s head is extremely well preserved; yet, when</p><p>visualizing his skull, the same kind of concave impression may be seen (Figure 6.8).</p><p>The Grauballe Man’s leg fracture at first did seem to resemble a peri-mortem traumatic</p><p>event. Indeed, the initial appraisal was that this tibia fracture did look somewhat like present-</p><p>day traffic accidents, where a person is hit by the fender of a car. However, the fibula had</p><p>not fractured, and the leg had a rather pronounced rotation (Figure 6.9). This could be a case</p><p>of a peri-mortem fracture, followed by post-mortem degradation and rotation. When we first</p><p>analysed the Grauballe Man, we could not reach a decision as to whether the fracture was</p><p>peri- or post-mortem. However, a few years later, in 2004, we CT scanned the Borremose</p><p>Woman (an Iron Age bog body from northern Jutland) and we noted that she had a femoral</p><p>fracture. Due to the way her legs lay, it seemed clear to us that this fracture was post-mortem</p><p>due to soil pressure, the fractured femur lying across the other leg. Also, the fracture did not</p><p>present itself as identical to modern clinical cases. When perusing the single CT-images of the</p><p>Borremose Woman’s thigh, we noted that the fracture not only showed itself as lines between</p><p>bone segments, but also as a curious separation between trabecular bone and compact bone.</p><p>This, in fact, was the same kind of fracture also seen in the Grauballe Man tibia fracture.</p><p>We find that since the Borremose Woman most probably had a post-mortem fracture with</p><p>these characteristics, then probably the Grauballe Man tibia fracture is also post-mortem.</p><p>Pathology and Pseudopathology</p><p>Our examinations using CT scanning of bog bodies have meant that we were able to</p><p>visualize otherwise hidden structures, and post-image-acquisition editing further allowed us</p><p>Computed Tomography Scanning and 3D Visualization 113</p><p>Figure 6.9 Tibia fracture: Grauballe Man</p><p>to visualize demineralized bone. This, in turn, enabled us to reappraise some previously</p><p>described lesions. The acid bog diagenetics mean that bone will be demineralized, become</p><p>pliable and, upon subsequent excavation and drying out, also shrink and warp. This means</p><p>that, under these conditions, aetiological attribution of pathology and trauma lack certainty.</p><p>It is our contention that some previously described lesions associated with the bog bodies</p><p>are more an effect of post-mortem factors than trauma. This is, of course, complicated by</p><p>the fact that a perimortal lesion may also be diagenetically affected.</p><p>The aforementioned Irish bog body finds may also illustrate damage due to excavation,</p><p>aside from the post-mortem degradation due to the bog. Unlike earlier finds in the 19th</p><p>century and up until the 1950s, peat is now excavated from bogs by machine. Peat-cutting</p><p>machinery may sever body parts, and subsequent transport on conveyor belts may further</p><p>damage the body. Clony Cavan Man had lost his forearms and lower abdomen on this</p><p>account.</p><p>Based on our analyses, therefore, we suggest that bog body lesions and pathology should</p><p>be viewed in a two-axis continuum: one axis covering pre-, peri-, and post-mortem time</p><p>periods, and the other axis ranging from ‘true’ lesions to clearly pseudopathology, i.e. lesions</p><p>or pathology due to diagenetic change (Figure 6.10).</p><p>For example, the Grauballe Man throat wound is without doubt a peri-mortal lesion.</p><p>Also, it clearly presents itself as a lesion, with only little post-mortem change (the skin has</p><p>shrunk from the cut surfaces). This would yield a clear point</p><p>in the graph (number 1 on</p><p>Figure 6.10). The Tollund Man head impression is, on the other hand, clearly post-mortem,</p><p>due to diagenetic change (number 2 on Figure 6.10). The Grauballe Man cranial lesion is</p><p>most probably quite like the Tollund Man lesion; but, since there is some fracturing and</p><p>114 Advances in Human Palaeopathology</p><p>2</p><p>3</p><p>1</p><p>Antemortem Perimortem Postmortem</p><p>Pseudopathology</p><p>Pathology</p><p>Figure 6.10 Pathology and pseudopathology versus time-period when lesion was inflicted</p><p>bone loss, it cannot be definitively ruled out that a peri-mortem lesion was there origi-</p><p>nally, which has since been heavily degraded and warped post-mortem. This corresponds to</p><p>point 3 on Figure 6.10.</p><p>As a final case in point, we looked at the Borremose Woman’s head. Her skull was found</p><p>to be heavily fractured, and from the archaeological viewpoint this was seen as perhaps</p><p>how she had been executed: by a blow or multiple blows to the face (Ry Andersen and</p><p>Geertinger, 1984). From a forensic viewpoint this would be somewhat unusual: craniofacial</p><p>fracturing of that magnitude is mainly found in high-energy impacts, e.g. traffic accidents.</p><p>Inflicting these fractures would be difficult using a blunt weapon even if she was lying on</p><p>her back on the ground. The Borremose Woman was found lying face down in the bog.</p><p>Owing to the diagenetic effects, the fracturing of the facial skeleton might be more indicative</p><p>of sutures coming apart and soil pressure ‘flattening the face’ in an antero-posterior direction</p><p>(unlike the more lateral compression seen for Grauballe Man and Tollund Man, who both lay</p><p>with their head on the side). The three-dimensional visualization of her skull shows that the</p><p>cranial bones show major dislocation, with some bones separating at the sutures, while some</p><p>bones do present fracture lines. This means that a peri-mortem lesion cannot be ruled out,</p><p>perhaps only with minor fractures, but which has since been heavily degraded post-mortem,</p><p>resulting in the total separation of the cranial and facial bones. Indeed, we feel that she might</p><p>even have been dealt a single blow to the back of the head, resulting in an impact fracture to</p><p>the posterior cranium, with fracture lines extending forward to the facial bones and cranial</p><p>base (as, indeed, is seen in modern forensic pathology of head trauma). Further post-mortem</p><p>degradation, especially due to the soil pressure, could then explain the findings.</p><p>Computed Tomography Scanning and the Biocultural Perspective</p><p>As with other investigative techniques and methods employed for the study of mummies,</p><p>the data generated from CT scanning and three-dimensional visualization must ultimately</p><p>be put in a biocultural context. In this respect, the differentiation between natural and</p><p>artificial mummies is important. What can be seen in artificial mummies will require a</p><p>basic knowledge of the embalming techniques employed, which may alter or remove organs.</p><p>On the other hand, natural mummies reflect the diagenetic changes to which the body has</p><p>Computed Tomography Scanning and 3D Visualization 115</p><p>been subjected. At one end of the spectrum, the natural mummies from Mount Llullaillaco,</p><p>preserved in permafrost, are examples of an exceptional degree of preservation of internal</p><p>organs. Bog bodies stand at the other end, being very much degraded due to the acidic</p><p>bogs. This establishes limitations in terms of what can be visualized by CT scanning.</p><p>Overall, perusing the results of CT scanning of mummies and bog bodies, it seems to</p><p>this author that the major contribution of CT scanning to the biocultural interpretations of</p><p>mummy finds is in the identification of mummification methods for artificial mummies</p><p>and of trauma in natural mummies. Palaeopathological lesions are generally rare, especially</p><p>in non-mineralized tissues. For those pathological conditions identified, it may perhaps be</p><p>discussed to what extent minor pathologies have a major impact on biocultural understanding;</p><p>trauma, however, is clearly indicative of an active, maybe even sacrificial, cultural setting.</p><p>The Tyrolean Iceman had an arrow point lodged in his shoulder (Gostner and Vigl, 2002),</p><p>while nearly all the bog bodies have definite signs of having been put to death (van der</p><p>Sanden, 1996).</p><p>One should, perhaps, focus just as much on the physical anthropological data</p><p>derived from CT scannings. For example, three-dimensional visualizations make it</p><p>possible to ascertain age and sex, and these data may in themselves be important</p><p>when placing mummies in a biocultural perspective. The Mount Llullaillaco mummies</p><p>were all determined to be children or juveniles, which is highly informative of Inca</p><p>human sacrificial customs (Previgliano et al., 2003). Likewise, the Greenland mum-</p><p>mies from Qilakitsoq were all children (two) and women (six), again attesting to</p><p>Greenlandic Thule culture social structures. Very few male skeletons are found, presum-</p><p>ably because the males have a higher risk of death while out hunting. On the other</p><p>hand, if the male hunter dies, the rest of the family may die from starvation (Hart</p><p>Hansen, 1989).</p><p>CONCLUSION</p><p>The methods of CT scanning and subsequent three-dimensional visualization are powerful</p><p>analytical tools for palaeopathological and physical anthropological analyses of mummies</p><p>and bog bodies. The technique allows the visualization of internal structures, and especially</p><p>of bones and teeth. However, post-image-capture processing is often necessary to extract</p><p>the full information from the CT data. This is especially the case with bog bodies, where</p><p>there is much taphonomic alteration of the skeletal structures. We have CT scanned several</p><p>Danish bog bodies, and based on our findings we feel that some lesions previously claimed</p><p>as signs of peri-mortem trauma may have to be re-evaluated. Owing to the sometimes</p><p>extreme post-mortem, diagenetic influences on the bog bodies, it may be impossible ever to</p><p>be certain of the exact nature of the observed lesions. We propose to view the lesions as a</p><p>continuum, covering both pseudopathology and ‘true’ lesions, over pre- to post-mortem time</p><p>periods.</p><p>Finally, we wish to draw attention to the fact that CT scanning may not only be a</p><p>valuable analytical tool, but also an extraordinarily precise tool for documenting bog bodies</p><p>and mummies. The CT scanning process generates data, literally millimetre by millimetre</p><p>and inside out, in a digital format that may be accessed and shared with other scientists</p><p>or museums. Our cooperation with Argentine scientists in the investigation of the Mount</p><p>116 Advances in Human Palaeopathology</p><p>Llullaillaco mummies is an example of such a data sharing. The CT scanning data may also</p><p>be important in terms of future assessments of preservational status of the bog bodies. Using</p><p>so-called stereolithography, it is possible to produce 1:1 models of CT-scanned structures</p><p>directly from the computer visualizations (zur Nedden and Wicke, 1992; Hjalgrim et al.,</p><p>1996; Cesarani et al., 2004), detailing structures at a 1 mm resolution (Figure 6.11). This</p><p>has some very evident uses for exhibits, and has been used as basis for facial reconstruction</p><p>of two bog bodies (Figure 6.12). Aside from advances in resolution and faster CT scanners,</p><p>future prospects will probably include more CT-scanned bog bodies and mummies that will</p><p>lead to better comparative studies. These can be expected to result in a better understanding</p><p>of mummification, taphonomical changes, pseudopathologies, and the nature of observed</p><p>lesions and tissue preservation. In this respect, use of CT scanning in mummy studies as</p><p>a ‘screening tool’ for locating pathological</p><p>Sciences</p><p>at the University of Bradford, England. His major research interest is in human adapta-</p><p>tion, but he has a specific interest in calcified tissue biology and the effect of disease on</p><p>human evolution during the Holocene. The latter interest includes a focus on the impact of</p><p>major developments in human society, such as urbanism, on human health. Recent publi-</p><p>cations include: the second edition of Identification of Pathological Conditions in Human</p><p>Skeletal Remains (Academic Press, 2003); ‘Skeletal manifestations of hypothyroidism from</p><p>Contributors xvii</p><p>Switzerland’ (with Hotz), American Journal of Physical Anthropology, 2005; ‘Infectious</p><p>disease and human evolution’, 2006 McGraw-Hill Yearbook of Science & Technology (The</p><p>McGraw-Hill Companies, 2005). Administrative posts include 4 years as chairman of the</p><p>Department of Anthropology and 2 years as acting director of the National Museum of</p><p>Natural History, USA. He has done fieldwork in Jordan and has conducted research projects</p><p>in the USA, Europe, and Australia. He has served on several boards and review panels. From</p><p>1999 to 2001, he was president of the Paleopathology Association. He is a member of the</p><p>American Association of Physical Anthropologists, the Paleopathology Association and the</p><p>International Skeletal Society.</p><p>Ron Pinhasi</p><p>Lecturer in Prehistoric Archaeology, Department of Archaeology, University College Cork,</p><p>Cork, Ireland</p><p>Ron Pinhasi received his PhD from the University of Cambridge, England in 2003. He</p><p>spent two years in a Lise Meitner postdoctoral position at the Natural History museum,</p><p>Vienna, examining the health status of early medieval Austrian populations. He is currently</p><p>a lecturer in Archaeology, University College Cork, Ireland. His research focuses on growth</p><p>and development in past populations, the origin and spread of leprosy in Eurasia, and the</p><p>origins and spread of farming in the Near East. He carries out fieldwork in Israel and directs</p><p>prehistoric excavations in Armenia. Key publications include ‘Morbidity, rickets, and long</p><p>bone growth in post-medieval Britain – a cross-population analysis’ (with Shaw, White and</p><p>Ogden), Annals of Human Biology, 2006; ‘A cross-population analysis of the growth of long</p><p>bones and the os coxae of three early medieval Austrian populations’ (with Teschler-Nicola,</p><p>Knaus and Shaw), American Journal of Human Biology, 2005; ‘Tracing the origin and spread</p><p>of agriculture in Europe’ (with Fort and Ammerman), PLoS Biology, 2005; and ‘A regional</p><p>biological approach to the spread of farming in Europe: Anatolia, the Levant, south-eastern</p><p>Europe, and the Mediterranean’ (with Pluciennik), Current Anthropology, 2004. He is a</p><p>member of the European Archaeological Association, and the Paleopathology Association.</p><p>Katy Turner</p><p>Research Associate, Division of Epidemiology, Public Health and Primary Care, Imperial</p><p>College, Praed Street, St Mary’s Campus, London W2 1PG, UK</p><p>Katy Turner received her PhD from Imperial College London in 2004. She spent 18 months at</p><p>the Health Protection Agency, developing mathematical models to investigate the potential</p><p>impact of the National Chlamydia Screening Programme in England. She is currently a</p><p>research associate at Imperial College, investigating the evolutionary biology of human</p><p>pathogens. Her research focuses on the complex interactions between bacterial populations</p><p>(intra- and inter-species) and between host and pathogen, using a variety of mathematical</p><p>modelling and analytical approaches, together with the best available empirical data, to</p><p>address important public health questions. Key publications include: ‘The impact of the</p><p>phase of an epidemic of sexually transmitted infection on the evolution of the organism’ (with</p><p>Garnett), Sexually Transmitted Infections, 2002; ‘Modelling the effectiveness of chlamydia</p><p>screening in England’ (with Adams, Lamontagne, Emmett, Baster and Edmunds), Sexually</p><p>Transmitted Infections, 2006; and ‘Modelling bacterial speciation’ (with Hanage, Spratt and</p><p>Fraser), Philosophical Transactions of the Royal Society of London, Series B: Biological</p><p>Sciences, 2006.</p><p>xviii Contributors</p><p>Gordon Turner-Walker</p><p>School of Cultural Heritage Conservation, National Yunlin University of Science and</p><p>Technology, 123 University Road Sec. 3, Touliou, 640 Yunlin, Taiwan (ROC)</p><p>Gordon Turner-Walker was awarded a PhD by the University of Durham, England, in</p><p>1993. He worked as the archaeological conservator for Norfolk Museums Service before</p><p>taking up a 3-year postdoctoral fellowship at the Norwegian University of Science and</p><p>Technology investigating osteoporosis in the medieval population of Trondheim. He is cur-</p><p>rently Associate Professor of cultural heritage conservation at National Yunlin University</p><p>of Science and Technology, Taiwan. His main areas of research are post-mortem alterations</p><p>to bone chemistry and microstructure, the archaeology of osteoporosis and the degradation</p><p>of cultural materials in marine and terrestrial environments. Key publications include: ‘The</p><p>West Runton fossil elephant: a pre-conservation evaluation of its condition and burial envi-</p><p>ronment’ The Conservator, 1998; ‘Quantifying histological changes in archaeological bones</p><p>using BSE-SEM image analysis’ (with Syversen), Archaeometry, 2002; ‘Sub-micron spongi-</p><p>form porosity is the major ultra-structural alteration occurring in archaeological bone’ (with</p><p>Nielsen-Marsh, Syversen, Kars and Collins), International Journal of Osteoarchaeology,</p><p>2002; ‘Osteoporosis in a population from medieval Norway’ (with Mays and Syversen),</p><p>American Journal of Physical Anthropology, 2006. He is a Fellow of the International</p><p>Institute for Conservation of Historic and Artistic Works.</p><p>William White</p><p>Centre for Human Bioarchaeology, The Museum of London, 150 London Wall, London</p><p>EC2Y 5HN, UK</p><p>Bill White CChem, FRSC, FSA, is the Senior Curator of Human Remains at the Museum</p><p>of London’s Centre for Human Bioarchaeology. He began his career as an organic chemist</p><p>working in the pharmaceutical industry. Early on he developed an interest in archaeology and</p><p>obtained a Diploma in Archaeology at the University of London, England. After undertaking</p><p>the university post-diploma course in ‘Human Remains in Archaeology’ with Theya Molleson</p><p>of the Natural History Museum, South Kensington, he began to prepare the first of what</p><p>was to become a long series of bone reports, chiefly from archaeological sites excavated in</p><p>London. During the past 20 years or more he has analysed thousands of human skeletons,</p><p>albeit using continuously changing recording media. In 2003 Bill was appointed the inaugural</p><p>Curator of Human remains at the Museum of London, responsible inter alia for booking in</p><p>and invigilating postgraduate and postdoctoral researchers working on the huge collection</p><p>of archaeological skeletons at the museum. He also headed the team of osteoarchaeologists</p><p>who recorded 5000 of these skeletons onto an electronic relational database, the Wellcome</p><p>Osteological Research Database, under a grant from the Wellcome Trust and which went</p><p>online in 2007.</p><p>PART 1</p><p>Analytical Approaches in</p><p>Palaeopathology</p><p>Advances in Human Palaeopathology Edited by Ron Pinhasi and Simon Mays</p><p>© 2008 John Wiley & Sons, Ltd. ISBN: 978-0-470-03602-0</p><p>1</p><p>The Chemical and Microbial</p><p>Degradation of Bones</p><p>and Teeth</p><p>Gordon Turner-Walker</p><p>School of Cultural Heritage Conservation, National Yunlin</p><p>University of Science and Technology, 123 University Road</p><p>Sec. 3, Touliou, 640 Yunlin, Taiwan (ROC)</p><p>INTRODUCTION</p><p>changes for subsequent precise sampling or for</p><p>guiding endoscopic procedures will probably become more common.</p><p>Figure 6.11 Stereolithographic models of the Borremose Woman cranial bones</p><p>Computed Tomography Scanning and 3D Visualization 117</p><p>Figure 6.12 Facial reconstruction of the Borremose Woman (photo: N. Lynnerup)</p><p>REFERENCES</p><p>Ambrose J. 1975. A brief review of the EMI scanner. Proc Brit Inst Radiol 48: 605–606.</p><p>Aufderheide A. 2003. The Scientific Study of Mummies. Cambridge University Press: Cambridge.</p><p>Bennike P. 2003. Bodies and Skeletons from Danish Bogs. Do they tell the same story? In Mummies</p><p>in a New Millennium, Proceedings of the 4th World Congress on Mummy Studies, Lynnerup N,</p><p>Andreasen C, Berglund J (eds). Danish Polar Centre, Publication No. 11. Danish Polar Centre:</p><p>Copenhagen; 39–42.</p><p>Bloomfield JA. 1985. Radiology of Egyptian mummy. Austr Radiol 29: 64–66.</p><p>Brothwell D, Sandison AT (eds). 1967. Diseases in Antiquity. Charles. C. Thomas: Springfield.</p><p>Brothwell D, Liversage D, Gottlieb B. 1990. Radiographic and forensic aspects of the female</p><p>Huldremose body. J Danish Archaeol 9: 157–178.</p><p>Buikstra JE, Ubelaker DH. 1994. Standards for Data Collection from Human Skeletal Remains.</p><p>Arkansas Archaeological Survey Research Series, 44. Arkansas Archaeological Survey Fayetteville,</p><p>AR.</p><p>Carlton RR, Adler AM. 2001. Principles of Radiographic Imaging. An Art and a Science. Delmar:</p><p>Albany, NY.</p><p>Cesarani F, Martina MC, Grilletto R, Boano R, Roveri AMD, Capussotto V, Giuliano A, Celia M,</p><p>Gandini G. 2004. Facial reconstruction of a wrapped Egyptian mummy using MDCT. Am J Radiol</p><p>183: 755–758.</p><p>Christensen OE. 1969. Un examen radiologique des mommies Egyptienne des musées Danois. La</p><p>Semaine Hôpit 45: 1990–1998.</p><p>Ciranni R, Garbini F, Neri E, Melai L, Giusti L, Fornaciari G. 2002. The “Braids Lady” of Arezzo: a</p><p>case of rheumatoid arthritis in a 16th century mummy. Clin Exp Rheumatol 20: 745–752.</p><p>Cockburn A, Cockburn E. 1980. Mummies, Disease and Ancient Cultures. Cambridge University Press:</p><p>Cambridge.</p><p>118 Advances in Human Palaeopathology</p><p>David RA. 1979. The Manchester Museum Mummy Project. Manchester University Press: Manchester.</p><p>Fawcitt RA, Jarvis H, Isherwood I. 1984. X-raying the Manchester mummies. In Evidence Embalmed.</p><p>Modern Medicine and the Mummies of Ancient Egypt, David R, Tapp E, editors. Manchester</p><p>University Press: Manchester.</p><p>Fleckenstein P, Tranum-Jensen J. 1993. Anatomy in Diagnostic Imaging. Munksgaard: Copenhagen.</p><p>Fornaciari G, Ciranni R, Ventura L. 2001. Paleoandrology and prostatic hyperplasia in Italian mummies</p><p>(XV–XIX century). Med Secoli 13: 269–284.</p><p>Glob PV. 1965. Mosefolket. Gyldendal: Copenhagen.</p><p>Gostner P, Vigl E. 2002. Report of radiological-forensic findings on the Iceman. J Archaeol Sci 29:</p><p>323–326.</p><p>Harris JE, Weeks KR. 1973. X-Raying the Pharaohs. Schribners and Sons: Chicago, IL.</p><p>Harris JE, Wente EF. 1980. An X-Ray Atlas of the Royal Mummies. Chicago University Press: Chicago,</p><p>IL.</p><p>Hart Hansen JP. 1989. The mummies from Qilakitsoq – paleopathological aspects. In The Mummies</p><p>from Qilakitsoq – Eskimos in the 15th Century, Hart Hansen JP, Gulløv HC (eds). Meddelelser om</p><p>Grønland, Man and Society Series, 12. Danish Polar Centre: Copenhagen; 69–82.</p><p>Harwood-Nash DC. 1979. Computed tomography of ancient Egyptian mummies. J Comput Assist</p><p>Tomogr 3: 768–773.</p><p>Hjalgrim H, Lynnerup N, Liversage M, Rosenklint A. 1995. Stereolithography: potential applications</p><p>in anthropological studies. Am J Phys Anthropol 97: 329–333.</p><p>Hoffman H, Torres WE, Ernst RD. 2002. Paleoradiology: advanced CT in the evaluation of nine</p><p>Egyptian mummies. RadioGraphics 22: 377–385.</p><p>Hounsfield GN. 1976. Historical notes on computerized axial tomography. J Can Assoc Radiol 27:</p><p>135–142.</p><p>Hsieh J. 2002. Computed Tomography. Principles, Design, Artifacts, and Recent Advances. SPIE:</p><p>Bellingham, WA.</p><p>Jansen RJ, Poulus M, Taconis W, Stoker J. 2002. High-resolution spiral computed tomog-</p><p>raphy with multiplanar reformatting, 3D surface- and volume rendering: a non-destructive</p><p>method to visualize ancient Egyptian mummification techniques. Comput Med Imag Graph 26:</p><p>211–216.</p><p>Koenig W. 1896. 14 Photographien mit Röngten-Strahlen. Johann Ambrosius Barth: Leipzig.</p><p>Lewin PK, Harwood-Nash DC. 1977. Computerized axial tomography in medical archaeology.</p><p>Paleopathol Newslett 17: 8–9.</p><p>Lovejoy CO, Meindl RS, Pryzbeck TR, Mensforth RP. 1985. Chronological metamorphosis of the</p><p>auricular surface of the illium: a new method for the determination of adult skeletal age at death.</p><p>Am J Phys Anthropol 68: 15–28.</p><p>Lowenstein EJ. 2004. Paleodermatoses: lessons learned from mummies. J Am Acad Dermatol 50:</p><p>919–936.</p><p>Marx M, D’Auria SH. 1986. CT examination of eleven Egyptian mummies. RadioGraphics 6: 321–330.</p><p>Marx M, D’Auria SH. 1988. Three-dimensional CT reconstructions of an ancient human Egyptian</p><p>mummy. Am J Radiol 150: 147–149.</p><p>Mays S. 1998. The Archaeology of Human Bones. Routledge: London.</p><p>Melcher AH, Holowka S, Pharoah M, Lewin PK. 1997. Non-invasive computed tomography and</p><p>three-dimensional reconstruction of the dentition of a 2,800-year-old Egyptian mummy exhibiting</p><p>extensive disease. Am J Phys Anthropol 103: 329–340.</p><p>Moodie RL. 1931. Roentgenologic Studies of Egyptian and Peruvian Mummies. Field Museum of</p><p>Chicago: Chicago, IL.</p><p>Munck W. 1956. Patologisk-anatomisk og retsmedicinsk undersøgelse af moseliget fra Grauballe.</p><p>KUML (J Jutland Archaeol Soc) 1956: 131–137.</p><p>Neave RAH. 1979. The reconstruction of skulls for facial reconstruction using radiographic techniques.</p><p>In Science in Egyptology, David AR (ed.). Manchester University Press: Manchester; 329–336.</p><p>Computed Tomography Scanning and 3D Visualization 119</p><p>Nerlich AG, Zink A, Szeimies U, Hagedorn HG. 2000. Ancient Egyptian prosthesis of the big toe.</p><p>Lancet 356: 2176–2179.</p><p>Notman DNH, Beattie OB. 1995. The paleoimaging and forensic anthropology of frozen sailors from</p><p>the Franklin Arctic expedition mass disaster (1845–1848): a detailed presentation of two radiological</p><p>surveys. In Human Mummies: A Global Survey of their Status and the Technique of Conservation,</p><p>Spindler K, Wilfing H, Rastbichler-Zissering E, zur Nedden D, Nothdurfter H (eds). Springer: New</p><p>York; 3–8.</p><p>Notman DNH, Anderson L, Beattie OB, Amy R. 1987. Arctic paleoradiology: portable radiographic</p><p>examination of two frozen sailors from the Franklin Expedition (1845–1848). Am J Roent 149:</p><p>347–350.</p><p>Ortner DJ. 2003. Identification of Pathological Conditions in Human Skeletal Remains. Academic</p><p>Press: San Diego, CA.</p><p>Petrie WMF, Griifith FLl. 1898. Deshashesh 1897. Fifteenth Memoir of the Egypt Exploration Fund.</p><p>Egypt Exploration Fund: London.</p><p>Pickering RB, Conces JF, Braunstein EM, Yurco F. 1990. Three-dimensional computed tomography</p><p>of the mummy Wenuhotep. Am J Phys Anthropol 83: 49–55.</p><p>Previgliano CH, Ceruti C, Reinhard J, Araoz FA, Diez JG. 2003. Radiologic evaluation of the</p><p>Llullaillaco mummies. Am J Radiol 181: 1473–1479.</p><p>Rühli FJ, Hodler J, Böni T. 2002. CT-guided biopsy: a new diagnostic method for paleopathological</p><p>research. Am J Phys Anthropol 117: 272–275.</p><p>Ry Andersen S, Geertinger P. 1984. Bog bodies investigated in the light of forensic medicine. J Danish</p><p>Archaeol 3: 111–119.</p><p>Smith GE. 1912. The Royal Mummies. Catalogue General des Antiquitees Égyptiennes de Musée du</p><p>Cairo. Musée du Cairo: Cairo.</p><p>Taconis WK, Maat GJR. 2005. Radiological findings in the human mummies and human heads. In</p><p>Egyptian Mummies. Radiological Atlas of the Collections in the National Museum of Antiquities in</p><p>Leiden, Raven MJ, Taconis WK (eds). Brepols: Turnhout; 53–80.</p><p>Vahey T, Brown D. 1984.</p><p>The physical survival of bone is integral to any kind of palaeopathological study. Not</p><p>only must the skeleton survive in the burial environment or tomb, it must retain sufficient</p><p>strength to be excavated, lifted, archived and studied. When assessing skeletal remains</p><p>for pathological conditions, it is also important to distinguish successfully between bone</p><p>lesions that arose ante- or peri-mortem as a result of disease or trauma, and damage caused</p><p>by post-mortem processes taking place in the burial environment. A sound understanding</p><p>of post-mortem changes to mineralized tissues is, therefore, essential when attempting to</p><p>interpret pathological conditions in skeletons, particularly those (the majority) that have</p><p>been buried in soils for centuries or millennia. Unlike some gross post-mortem patterns</p><p>of destruction caused by root action, insects or rodents, which are frequently visible on</p><p>the outer surfaces of the specimens, microbial and chemical degradation is microscopic in</p><p>nature and can influence the interiors of the bones as well as their surfaces. This unseen</p><p>deterioration not only contributes to the fragility of archaeological bones, but by altering the</p><p>chemistry and microstructure of the tissues it can also have a serious impact on chemical</p><p>or radiological analyses and on the radiocarbon dating of skeletons (Lee-Thorpe and van</p><p>der Merwe, 1987; van Klinken, 1999; Mays, 2000; Petchley and Higham, 2000; Dupras</p><p>and Schwarcz, 2001). The potential for leaching and the movement of soluble salts into</p><p>and from the bone structure also has a bearing on the interpretation of radiodensitometry</p><p>(Mays, Chapter 5 this volume) and measurements of bone density using clinical techniques</p><p>such as dual energy X-ray absorptiometry (Agarwal and Grynpas, 1996; Mays, 1999; Mays</p><p>et al., 2006). Thus, changes to skeletal tissues arising from their interaction with the burial</p><p>Advances in Human Palaeopathology Edited by Ron Pinhasi and Simon Mays</p><p>© 2008 John Wiley & Sons, Ltd. ISBN: 978-0-470-03602-0</p><p>4 Advances in Human Palaeopathology</p><p>environment and from the actions of soil microorganisms have an impact on almost all aspects</p><p>of palaeopathological study and the value of human skeletons as a source of information about</p><p>the past.</p><p>In recent decades, rapid developments in the field of biomolecular archaeology have demon-</p><p>strated thatphysicalandmicroscopic integrity isno longerenoughwhenconsidering the research</p><p>potential of an individual skeleton or assemblage. The integrity of any isotopic and molecular</p><p>evidence contained within bone and tooth tissues is equally important (Muyzer et al., 1992;</p><p>Cattaneo et al., 1995; Evershed et al., 1995; Baron et al., 1996; Taylor et al., 1996; Weser et al.,</p><p>1996; Braun et al., 1998; Stott et al., 1999; Götherström et al., 2002; Geigl, 2002). Recognition</p><p>of this has driven much of the research into how and why skeletal tissues degrade in the soil,</p><p>and the progress made in the understanding of these diagenetic processes during the last decade</p><p>of the 20th century and early years of the 21st century has been almost as dramatic as the huge</p><p>strides made in the analyses of DNA, lipid and protein residues over the same period.</p><p>Compared with other scientific studies of archaeological and fossil bones, the study of</p><p>bone deterioration is relatively young. The term taphonomy, to describe post-mortem pro-</p><p>cesses influencing bone survival, was introduced nearly 70 years ago by Efremov (1940),</p><p>and these ‘laws of burial’ were invoked to help interpret fossil and archaeological bone</p><p>assemblages. In its broadest sense, taphonomy concerns all aspects of the passage of organ-</p><p>isms from the biosphere (the living world) to the lithosphere or Earth’s crust (Olson, 1980).</p><p>The primary goal of taphonomic studies is to work backwards from the surviving bone</p><p>assemblages to the composition, structure and dynamics of the parent populations (human</p><p>or animal) using evidence recovered from the bones themselves, the nature of their con-</p><p>texts and an understanding of post-mortem processes (Olsen, 1980). The geological term</p><p>diagenesis is defined as the processes by which sediment is transformed into sedimentary</p><p>rock under conditions of low temperature and pressure. In recent years, this term has been</p><p>adopted to describe the changes undergone by skeletal tissues in the burial environment.</p><p>These changes may involve dissolution of bone tissue or its cementation by exogenic min-</p><p>erals, recrystallization of bone mineral or its replacement by other mineral species. These</p><p>alterations to bone tissue are often crudely referred to as fossilization (Behrensmeyer and</p><p>Hill, 1980) and a combination of taphonomic and diagenetic processes determine whether a</p><p>bone decays and ultimately disappears or persists throughout the course of archaeological or</p><p>geological time.</p><p>As early as the middle years of the 19th century, microscopic examination of ancient bones</p><p>had identified the potential importance of microorganisms in the destruction and degradation</p><p>of bone tissues. In 1864, Wedl examined thin sections of ancient bones under the light</p><p>microscope and described small channels or tunnels penetrating the bone tissues (Wedl,</p><p>1864). Roux, working in the late 19th century, also identified these features in fossil bones</p><p>and termed them bored channels or Bohrkanäle (Roux, 1887). The presence of fine, brown</p><p>filaments visible in these tunnels suggested to him the action of fungi in their formation.</p><p>Thus, from the outset, the action of fungi was implicated as the principal causal factor in the</p><p>destruction of dead bone tissues – an assumption that persisted for more than 100 years and</p><p>remains contentious today.</p><p>By the middle of the 20th century, chemical analysis of ancient skeletal tissues was being</p><p>used as a means of absolute dating, initially with the introduction of fluorine-content dating</p><p>and later followed by the radiocarbon revolution in archaeology. One of the earlier successes</p><p>for carbon-14 dating was the confirmation of the Piltdown find of an ‘English ape-man’</p><p>as a modern hoax (de Vries and Oakley, 1959). Suspicions had already been voiced after</p><p>The Chemical and Microbial Degradation of Bones and Teeth 5</p><p>the failure to find the significant levels of fluoride in the bones that would be expected for</p><p>a find of geological age. As a result of these developments, together with the introduction</p><p>of uranium-series dating, calcium-41 dating and amino acid racemization dating, scientists</p><p>became increasingly aware of the importance of understanding changes in the structure and</p><p>composition of bones and teeth. These problems were later underlined during attempts to</p><p>isolate faint dietary signatures, in trace element concentrations or in stable isotope variations,</p><p>from larger diagenetic chemical alterations.</p><p>Before discussing post-mortem changes to skeletal tissues it is necessary to take a closer</p><p>look at the nature of bones and teeth.</p><p>THE CHEMISTRY, ULTRASTRUCTURE AND</p><p>MICROSTRUCTURE OF SKELETAL TISSUES</p><p>Skeletal tissues have a very ancient ancestry in the evolutionary record. Work on a group</p><p>of fossil elements called conodonts has confirmed that these tooth-like structures represent</p><p>the grasping mouthparts of primitive marine animals resembling eels (Briggs, 1992). These</p><p>tiny fossils, measuring between 0.2 and 2 mm in length, are composed of the calcium</p><p>phosphate mineral carbonate fluorapatite, and investigations of their microstructure have</p><p>shown that they bear many features in common with the hard tissues (such as calcified</p><p>cartilage, bones and teeth) of more advanced vertebrates (Sansom et al., 1992; Schultze,</p><p>1996). These discoveries push back the origin of bony tissues, and consequently our ultimate</p><p>ancestors, to the late Cambrian period, over 500 million years ago.</p><p>The basic chemistry of the calcified tissues bone, antler and tooth dentine (including ivory)</p><p>is fundamentally the same, although they differ in their mode of growth and microstructure.</p><p>Tooth enamel is rather specialized and differs from the other calcified tissues in that it is</p><p>more crystalline and has a negligible organic content. Since bone is by far the most common</p><p>calcified tissue, it is perhaps appropriate to consider it first before outlining the ways in</p><p>which other tissues differ from it.</p><p>Bone</p><p>Living bone consists of three major components: organic matter, principally proteins; mineral</p><p>in the form of calcium phosphates; and water. Here, the inclusion of water as a major</p><p>constituent may seem pedantic, but the water contents of buried bones and the sediments</p><p>that surround them play as important a role in their future integrity over archaeological</p><p>time-scales as the chemistry and availability of biological fluids do during life. The organic</p><p>matter in dry bone accounts for approximately 22–23 % by weight (Turner-Walker, 1993)</p><p>and 40 % by volume (Nielsen-Marsh and Hedges, 2000a). About 90 % of this component is</p><p>made up of long fibrils of Type I collagen that give living bones their tensile strength and</p><p>a small degree of flexibility. Type I collagen molecules are highly organized, comprising</p><p>three stretched helical amino acid chains which are themselves twisted into a triple helix.</p><p>Collagen is characterized by a high glycine content, which makes up every third amino acid</p><p>(33 %), with high levels of proline and hydroxyproline, which together account for a further</p><p>20 %. Each triplet is approximately 300 nm in length and 1.5 nm in diameter (Yamamoto</p><p>et al., 2000, De Cupere et al., 2003).</p><p>6 Advances in Human Palaeopathology</p><p>The individual collagen molecules self-assemble or aggregate extracellularly and assume</p><p>a hierarchical architecture with triplets organizing into bundles, called microfibrils, which</p><p>ultimately form into fibrils and fibres. These fibre bundles align themselves with a quasi-</p><p>hexagonal packing (Figure 1.1). Type I collagen is insoluble under normal physical and</p><p>physiological conditions because of this well-ordered three-dimensional arrangement of the</p><p>fibres, the ionic and hydrophobic interactions between adjacent amino acid chains, and a</p><p>degree of cross-linking between the molecules. Strong aldehyde cross-links form between</p><p>the lysine and hydroxylysine of adjacent collagen molecules and the microfibril is further</p><p>stabilized by numerous intramolecular hydrogen bonds. Newly formed microfibrils are about</p><p>20 nm in diameter but grow in size with maturity up to approximately 90 nm, with an</p><p>average microfibril diameter in young adults of 75 nm (Sarathchandra et al., 1999). The</p><p>unmineralized collagen network or organic matrix also contains non-collagenous proteins</p><p>(including osteocalcin) and mucopolysaccharides which make up the remaining 10 % by</p><p>weight (Tuross, 2003). Some of these non-collagenous proteins can be extremely stable over</p><p>geological time-scales, strongly suggesting an intimate association with the mineral phase</p><p>(Muyzer et al., 1992; Smith et al., 2005).</p><p>Figure 1.1 Diagrammatic representation of the close packing of collagen molecules (triplets) into</p><p>fibrils. In reality the molecules are stabilized by intermolecular bonds. Progressive mineralization with</p><p>small platelets of hydroxyapatite (HAP) proceeds in the gaps between the ends of the molecules and</p><p>between adjacent triplets</p><p>The Chemical and Microbial Degradation of Bones and Teeth 7</p><p>The compressive strength of bone tissues is provided by the mineral component, which is</p><p>generally accepted to be a stoichiometrically imperfect, carbonate-containing HAP analogue</p><p>with a composition approximating to Ca10(PO4)6(OH)2, also called bioapatite. This mineral</p><p>phase also includes traces of other anionic and cationic species that variously adsorb on</p><p>crystal surfaces or substitute for Ca2+, PO2−</p><p>4 and hydroxyl ions in the lattice. The exact</p><p>nature of these mineral – ion interactions is not relevant to this discussion, but it is important</p><p>to understand that they are closely related to the small sizes of the bioapatite crystals and their</p><p>total available surface area. HAP crystals are plate-like in morphology and have currently</p><p>accepted dimensions of approximately 35 nm by 5 nm and with a thickness of about 2–3</p><p>nm (Lowenstam and Weiner, 1989; Nielsen-Marsh et al., 2000). It is widely recognized that</p><p>the average sizes of the HAP crystals in bone increase with the maturity of the tissue. The</p><p>extreme small sizes of the individual bone crystals, or more properly crystallites, present an</p><p>enormous active surface area for bone mineral, estimated at between 100 and 200 m2 g−1</p><p>(Posner, 1985; Newesely, 1989). It is unlikely, however, that this large active area is ever</p><p>realized, because of the intimate association between the collagen matrix and the HAP.</p><p>It has long been known that bone sections exhibit birefringence in polarized light, and</p><p>this optical property arises from the orientation of both the collagen fibres and the HAP</p><p>crystallites (Figure 1.2). These crystallites are embedded in the collagen matrix with their</p><p>c-axes aligned parallel to the long axes of the fibres. These fibres are aligned in lamellae in</p><p>which the fibre orientation in successive layers is rotated to give a plywood-like structure</p><p>(Giraud-Guille, 1988; Weiner and Traub, 1992). Evidence points to initial deposition of</p><p>HAP crystallites (primary mineralization) within gaps in the closely grouped collagen fibrils</p><p>(Figure 1.1), with the bulk of the mineral load progressively filling the interfibrillar spaces</p><p>(secondary mineralization), a process that may take several weeks or months. This results in</p><p>greater variability in mineral density between mature and more recent bone tissues in older</p><p>individuals, especially in osteonal or Haversian bone (Ortner and Turner-Walker, 2003).</p><p>There is an intimate association between the collagen molecules and HAP, and this chemical</p><p>affinity is strengthened by the non-collagenous protein osteocalcin, which makes up 2 %</p><p>Figure 1.2 (a) Transmitted light image of medieval human bone from Trondheim, Norway. Histo-</p><p>logical preservation is excellent, but staining around the central osteon illustrates the fine canalicular</p><p>network that connects the tissues with the soil environment. (b) The section viewed in polarized light</p><p>with a quarter-lambda plate. The spectacular birefringence arises from the alignment of collagen fibrils</p><p>and HAP in the bone lamellae</p><p>8 Advances in Human Palaeopathology</p><p>by weight of dry bone (Smith et al., 2005). Osteocalcin is known to bind both to HAP</p><p>and to collagen, and this relatively small protein plays an important role in the primary</p><p>mineralization of skeletal tissues.</p><p>Dry, fresh bone contains about 8 % water that is loosely bound and can be driven off</p><p>by heating in air at 105�C (Eastoe and Eastoe, 1954). However, for materials like bone</p><p>with a high microporosity, the total amount of bound water held by a sample depends</p><p>strongly on both the temperature and local relative humidity. For very small pores, quite high</p><p>temperatures are required to drive off all the liquid water held in small capillaries, and even</p><p>higher temperatures are necessary for chemically bound water. Determination of total bound</p><p>water in fresh bone is further complicated because, in thermogravimetric measurements,</p><p>weight losses at elevated temperatures are compounded by thermal decomposition of organic</p><p>matter and loss of bound carbonates from the bone mineral.</p><p>Measurements undertaken by Nielsen-Marsh and Hedges (2000a) of pore</p><p>volumes for</p><p>fresh bone using calibrated relative humidities indicated that the macroporosity (those pores</p><p>with radii between 4 and 20 nm) and microporosity (pores less than 4 nm in radius) were</p><p>0.075 cm3 g−1 and 0.059 cm3 g−1 respectively, giving a total pore volume below 20 nm of</p><p>0.134 cm3 g−1. This figure compares well with measurements of total pore volume for fresh,</p><p>compact bovine bone, which lie in the range 21–26 % by volume or 0.110–0.158 cm3 g−1</p><p>(data from Turner-Walker and Parry (1995)). These latter measurements (made from liquid</p><p>water absorption) included larger pores attributable to vascular channels and voids left by</p><p>degraded bone cells (osteocyte lacunae). More recently, mercury intrusion porosimetry has</p><p>refined the interpretation of bone porosity in the range 2 nm to 100 �m, and this technique</p><p>has had a significant bearing on current understanding of bone diagenesis (Nielsen-Marsh</p><p>and Hedges, 1999; Turner-Walker et al., 2002; Jans et al., 2004).</p><p>Bone is a physiologically active tissue, repairing itself when damaged – either at a macro-</p><p>scopic scale, as during the healing of a fracture, or microscopically, as in the constant</p><p>remodelling and replacement of bone to remove the microfractures that accumulate through</p><p>normal activity. Bone is also involved in calcium homeostasis, releasing or absorbing Ca2+</p><p>ions to maintain serum calcium levels within physiological limits. This requirement for</p><p>skeletal bone mineral to be immediately accessible hinges on both the large available surface</p><p>area of bone HAP and the considerable vascularity of bones. Living bone is penetrated by</p><p>numerous channels (Haversian canals and canals of Volkmann) averaging about 50 �m in</p><p>diameter, through which pass blood vessels and nerves (Figure 1.3). The branching archi-</p><p>tecture of these vessels provides a pathway between the countless bone cells or osteocytes</p><p>within the bone tissues and the circulating blood. A large number of cytoplasmic processes</p><p>extend from each osteocyte, connecting to neighbouring cells via canaliculi with a diameter</p><p>of approximately 200 nm. This extended network of fine channels penetrating bone allows</p><p>chemical messages to be transmitted throughout the tissue, as well as permitting nutrients</p><p>and mineral ions to be supplied to the bone matrix and metabolic waste products to be</p><p>removed (Figure 1.2a).</p><p>The microarchitecture of bone tissue varies, depending upon where it forms and the speed</p><p>at which it develops. Bone tissue associated with very rapid growth is called woven or fibre</p><p>bone. Fibre bone is not as dense or as well organized as other types of bone associated</p><p>with slower growth rates. The collagen microfibrils are irregular in thickness and lack the</p><p>linear orientation typical of later stages of bone development. Fibre bone forms early in the</p><p>growing skeleton but may be found in later life in abnormal bone tissue, such as fracture</p><p>callus and neoplasms (cancers) or beneath the periosteum as a response to infection. Mature</p><p>The Chemical and Microbial Degradation of Bones and Teeth 9</p><p>Figure 1.3 Three-dimensional representation of the micro-architecture of compact bone</p><p>bone has a more lamellar structure, forming either by apposition on the periosteal surface</p><p>(circumferential lamellar bone) or by remodelling of the interiors (Haversian or osteonal</p><p>bone). The microarchitecture of bone tissues clearly influences its mechanical properties,</p><p>porosity and, ultimately, its resistance to post-mortem degradation. However, a detailed</p><p>description of bone development and physiology lies outside the purposes of this chapter.</p><p>For a fuller account of the biology of skeletal tissues the reader is referred to Ortner and</p><p>Turner-Walker (2003) and Tuross (2003).</p><p>Tooth Dentine and Enamel</p><p>Teeth are complex structures that have properties that represent a trade-off between the need for</p><p>a hard, resistant material that can efficiently withstand many years of biting or grinding tough</p><p>foods and one that has good resistance to fracture. Good teeth are fundamental to the survival of</p><p>any animal, and nature has perfected many different designs to suit different diets and feeding</p><p>strategies. Unlike bones, which grow in situ and remain surrounded by soft tissues, teeth form</p><p>within the jaw and are later erupted through the gum into the mouth, where they are in frequent</p><p>and intimate contact with the outside world. Once in place, any remodelling or repair of damage</p><p>is strictly limitedbecause the tooth is effectively removedfromthecellular apparatus that formed</p><p>it. By way of compensation, humans develop two sets of teeth, the milk or deciduous teeth of</p><p>infancy and the permanent teeth which gradually replace the deciduous teeth. The permanent</p><p>dentition is complete by about 18 years of age.</p><p>10 Advances in Human Palaeopathology</p><p>Figure 1.4 Simplified cross-section of a tooth (incisor) and jaw</p><p>The mature human tooth can be divided into three parts: the crown, which is the part</p><p>visible above the gum; one or more roots, which anchor the tooth into the jaw; and the</p><p>neck or cervix, where the crown meets the root and which lies between the gum-line and the</p><p>socket (Figure 1.4). The bulk of the tooth is composed of dentine, which forms the underlying</p><p>load-bearing structure. Unlike bone, dentine is an avascular tissue with no blood supply. It</p><p>is also largely acellular and the living part of the tooth is restricted to the pulp cavity, which</p><p>extends from a small hole or foramen in the base of the root into the body of the tooth. The</p><p>pulp cavity contains blood vessels and nerves and is lined with cells called odontoblasts.</p><p>Numerous, tightly packed dental tubules extend radially out from the pulp cavity towards the</p><p>outer surfaces of the tooth. These tubules reflect the developmental growth of the tooth (in</p><p>the growing tooth, dentine is laid down on the interior surface of the enamel and proceeds</p><p>inwards) and provide a sensory mechanism for detecting loads on the teeth. The crown</p><p>of the tooth is encased in hard enamel, which is made up of parallel prisms composed of</p><p>almost pure HAP. Enamel has negligible organic content and is more crystalline than bone</p><p>HAP as a result of a larger crystallite size and their parallel alignment within prisms. Once</p><p>enamel is damaged by tooth wear or dental disease (caries) there is no natural mechanism</p><p>for effective repair. The outer surface of the tooth root is covered in a type of woven bone</p><p>called cementum which, together with the periodontal ligament, anchors the tooth in the</p><p>socket (Mays, 1998; Ortner and Turner-Walker, 2003). Healthy enamel has zero porosity,</p><p>apart from occasional growth defects. Although there has been little or no investigation of</p><p>the porosity of tooth dentine, it is clear that its porosity is low compared with that of bone.</p><p>The Chemical and Microbial Degradation of Bones and Teeth 11</p><p>Because of the absence of a vascular network in tooth dentine, its relatively low porosity</p><p>and the hard shell of impervious enamel that covers the exposed crown, it is generally</p><p>accepted that teeth are less susceptible to diagenesis than bones and, therefore, that they</p><p>represent a more reliable source of ancient DNA and other biomolecular information. Recent</p><p>evidence (Götherström et al., 2002; Wandeler et al., 2003; Gilbert et al., 2005, 2006) supports</p><p>the view that the potential for post-mortem and post-excavation contamination of dentine is</p><p>much lower than for bones (which are frequently handled by archaeologists and researchers).</p><p>Nevertheless, teeth are by no means immune to the diagenetic forces that affect bone</p><p>tissues.</p><p>CHEMICAL DIAGENESIS OF BONES AND TEETH</p><p>It is common</p><p>knowledge that bone tissue degrades in the soil. Bone-meal (ground bone)</p><p>has been used by gardeners as a fertilizer for centuries. This makes good sense, since bone</p><p>is rich in both nitrogen and phosphorus. Anecdotal evidence has suggested that water and</p><p>temperature play important roles in the deterioration of human corpses. For example, in Act</p><p>V: Scene 1 of Shakespeare’s Hamlet the sexton refers to the destructive power of water</p><p>on interred corpses: ‘� � � your water is a sore decayer of your whoreson dead body’. Also,</p><p>in some countries of northern Europe it was common practice in the past to pack wood</p><p>shavings around corpses before sealing the coffins if it was anticipated that the grave may</p><p>have to be reopened within the year to add the body of a close relative. The additional</p><p>insulation presumably raised the temperature of the body and accelerated decomposition of</p><p>the soft tissues, thus reducing the smell of decay. More rigorous research has confirmed the</p><p>importance of both soil hydrology (Pike, 1993; Hedges and Millard, 1995; Nielsen-Marsh,</p><p>1997; Pike et al., 2001; Nielsen-Marsh and Hedges, 2000a) and soil temperature (Collins</p><p>et al., 1995; Gernaey et al., 2001) on the deterioration of archaeological bones.</p><p>The role of water</p><p>The availability and movement of water within the soil, and hence through and around</p><p>archaeological bones, has an immense influence on their potential for survival. Water is the</p><p>medium of almost all chemical reactions that take place in the soil, and the presence of</p><p>water also supports microbial metabolism. Whilst in the body, bone mineral lies within a</p><p>relatively closed system and is surrounded by fluids that have a strictly controlled pH and</p><p>are approximately saturated with respect to HAP. In vivo dissolution and recrystallization of</p><p>bone mineral is mediated by bone cells which are themselves stimulated by a complex web of</p><p>systemic and local chemical signals, including physical stimuli, growth factors, parathyroid</p><p>hormone, and calcitriol – the active form of vitamin D (Ortner and Turner-Walker, 2003). In</p><p>sharp contrast to this, the soil represents an open system that is far from saturated in calcium</p><p>and phosphate ions (except perhaps in the case of deeply cut charnel pits containing many</p><p>hundreds of tightly jumbled bones). Bone mineral, therefore, is vulnerable to dissolution in</p><p>soil water, which can also bring in exogenous ions that may bind to the surface of the HAP or</p><p>substitute for Ca2+, PO2−</p><p>4 or CO2−</p><p>3 ions within the crystal lattice (Hedges and Millard, 1995).</p><p>Analyses of archaeological bones from different environments have demonstrated that</p><p>those bones that come from soil horizons where there is considerable fluctuation in the</p><p>12 Advances in Human Palaeopathology</p><p>groundwater content, i.e. water repeatedly flows around and through the bones, exhibit very</p><p>poor preservation compared with those that lie in permanently saturated conditions, i.e. below</p><p>the water table (Nielsen-Marsh et al., 2000: Figure 2). Clearly, susceptibility to dissolution</p><p>depends on many variables, but one obvious factor is the porosity of the bone. Water does</p><p>not act solely on the outer surfaces of bones; it penetrates the interiors via the network of</p><p>interconnecting vascular channels. However, once these small pores are filled and the pore</p><p>water is saturated in Ca2+ and PO2−</p><p>4 ions there may be considerable resistance to the flow</p><p>of water out of the bone, and dissolution of HAP is limited by a local diffusion gradient.</p><p>Diffusion rates through fine pore networks are generally very slow, leading to very limited</p><p>dissolution of bone mineral. These diffusive environments can be found in waterlogged</p><p>deposits or in sediments that resist movement of soil water, such as clays; and bones from</p><p>these environments frequently exhibit exceptionally good preservation.</p><p>If bones lie in an environment where there are repeated cycles of wetting and drying,</p><p>then, as the surrounding soil dries out, a hydraulic potential is generated that will draw water</p><p>(saturated in Ca2+ and PO2−</p><p>4 ions) out from the interiors of the bones. After heavy rain, the</p><p>bones experience a recharge regime in which the pores are refilled with water that is no</p><p>longer saturated in Ca2+ and PO2−</p><p>4 ions (Hedges and Millard, 1995). After many wetting and</p><p>drying cycles, the successive losses of calcium and phosphorus from the bone matrix cause</p><p>further increases in the porosity of the bones, which then become locked into a positive</p><p>feedback loop. Bones that are excavated from shallow, free-draining soils are generally less</p><p>well preserved than those from deep, waterlogged sites.</p><p>Bones buried in well-drained soils overlying sands or gravels that never become saturated</p><p>in water are particularly susceptible to leaching of the mineral phase. The rate and volume of</p><p>water flow through a bone in such a soil depends upon the relative hydraulic conductivities</p><p>of the soil and the bone (i.e. their relative porosities), and the total volume of water available</p><p>to flow (i.e. the amount of rainfall). This flow regime is potentially the worst situation for</p><p>the survival of archaeological bones, and in extreme cases (such as inhumations cut through</p><p>sands and gravels) can lead to total leaching of the skeleton, leaving only a soil silhouette</p><p>or ‘sand body’ (Keeley et al., 1977; Bethel and Carver, 1987; Carver, 2005).</p><p>The solubility of apatites in groundwaters is heavily dependent on the water’s pH and</p><p>the presence of other dissolved ionic species. As a general rule, all the calcium phosphates</p><p>become increasingly soluble as pH falls. Solid HAP will reach an equilibrium with stationary</p><p>water in contact with its surface according to</p><p>Ca10�PO4�6�OH�2 +14H+</p><p>� 10Ca2+ +6H2PO−</p><p>4 +2H2O</p><p>Thus, an increase in the hydrogen ion concentration (decrease in pH or increase in acidity)</p><p>will drive the equilibrium to the right and HAP dissolves. In alkaline soils, therefore, bone</p><p>mineral will tend to be protected from dissolution, unless there is a high dissolved carbon</p><p>dioxide concentration, in which case Ca2+ can precipitate out as bicarbonate or carbonate.</p><p>Removal of calcium ions, therefore, will also drive the equation to the right and HAP once</p><p>again can dissolve. HAP, therefore, acts as a buffer helping to stabilize local pH variations.</p><p>From the equation above it is also clear that soils low in phosphate may also lead to</p><p>demineralization of bones. HAP is most stable at pH 7.8 (Nielsen-Marsh et al., 2000).</p><p>HAP is thermodynamically one of the most stable forms of solid calcium phosphate.</p><p>Both bone mineral and synthetic HAP can be dissolved in mineral acids and reprecipitated</p><p>once more to a poorly crystalline HAP on the addition of alkali. In fact, carnivores that</p><p>The Chemical and Microbial Degradation of Bones and Teeth 13</p><p>consume large quantities of bone routinely excrete finely divided HAP in their faeces. X-ray</p><p>diffraction (XRD) studies on modern and ancient hyena coprolites have demonstrated that</p><p>their spectra are essentially the same as those of bone apatite (Horowitz and Goldberg,</p><p>1989). Although calcium and phosphorus are excreted via the gut of all vertebrates, the bulk</p><p>of the apatite in these coprolites will derive directly from bone dissolved in the stomach.</p><p>Therefore, it is thermodynamically favourable that bone mineral solubilized in the burial</p><p>environment, either by the active intervention of microorganisms or by the movement of</p><p>groundwaters over bone, will be reprecipitated as a poorly crystalline HAP (or carbonate</p><p>apatite depending upon the local availability of dissolved carbon dioxide) when the pH rises</p><p>or when the solubility product of either calcium or phosphate is exceeded (Nielsen-Mash</p><p>and Hedges, 2000b). Several authors have reported brushite (CaHPO4·2H2O) as a</p>