McDONALD INSTITUTE MONOGRAPHS Temple people Bioarchaeology, resilience and culture in prehistoric Malta By Simon Stoddart, Ronika K. Power, Jess E. Thompson, Bernardette Mercieca-Spiteri, Rowan McLaughlin, Eóin W. Parkinson, Anthony Pace & Caroline Malone Volume 3 of Fragility and Sustainability – Studies on Early Malta, the ERC-funded FRAGSUS Project Temple people McDONALD INSTITUTE MONOGRAPHS Temple people Bioarchaeology, resilience and culture in prehistoric Malta By Simon Stoddart, Ronika K. Power, Jess E. Thompson, Bernardette Mercieca-Spiteri, Rowan McLaughlin, Eóin W. Parkinson, Anthony Pace & Caroline Malone With contributions by Bruno Ariano, Robert Barratt, Sara Boyle, Dan Bradley, Laura T. Buck, Jacinta Carruthers, Joel D. Irish, Catherine Kneale, Emma Lightfoot, John S. Magnussen, Caroline Malone, Sarah Massingham, Valeria Mattiangeli, Rowan McLaughlin, John Meneely, Bernardette Mercieca-Spiteri, Argyro Nafplioti, Tamsin C. O’Connell, Anthony Pace, Margery Pardey, Eóin W. Parkinson, Ronika K. Power, Paula Reimer, John Robb, Jaap Saers, Jay T. Stock, Simon Stoddart, Jess E. Thompson, Alessandra Varalli, Nicholas Vella & Hannah Vogel Volume 3 of Fragility and Sustainability – Studies on Early Malta, the ERC-funded FRAGSUS Project This project has received funding from the European Research Council (ERC) under the European Union’s Seventh Framework Programme (FP7-2007-2013) (Grant agreement No. 323727). Published by: McDonald Institute for Archaeological Research University of Cambridge Downing Street Cambridge, UK CB2 3ER (0)(1223) 339327 eaj31@cam.ac.uk www.mcdonald.cam.ac.uk McDonald Institute for Archaeological Research, 2022 © 2022 McDonald Institute for Archaeological Research. Temple people is made available under a Creative Commons Attribution-NonCommercial- NoDerivatives 4.0 (International) Licence: https://creativecommons.org/licenses/by-nc-nd/4.0/ ISBN: 978-1-913344-08-5 Cover design by Dora Kemp, Caroline Malone and Ben Plumridge. Typesetting and layout by Ben Plumridge. On the cover: Stick figures from the Circle. Edited for the Institute by Matthew Davies (Series Editor). vContents Contributors xi Figures xiii Tables xix In memoriam George Mann xxi In memoriam Ann Monsarrat xxiii Acknowledgements xxv Preface xxvii Chapter 1 Introduction: people of early Malta and the Circle 1 Caroline Malone, Simon Stoddart, Jess E. Thompson & Nicholas Vella     1.1. Introduction 1     1.2. The origins of work at the Circle: funerary archaeology in Malta 1     1.3. Dating early Malta and changing approaches to the past: scientific questions       and approaches 5     1.4. Research goals of the Cambridge Gozo Project 5     1.5. Excavation of the Circle 7     1.6. Development of methodologies and progress of excavation (1987–94) 11       1.6.1. The rock-cut tomb methodology 11       1.6.2. Methodology for the main caves and deposits 11     1.7. The research methodologies of the Cambridge Gozo Project (1987–94) 15       1.7.1. Methodologies for bioarchaeological analysis 20     1.8. Recent work 21     1.9. Reflection 22     1.10. Impact of the project work, past and present 22 Chapter 2 New approaches to the bioarchaeology of complex multiple interments 27 Bernardette Mercieca-Spiteri, Ronika K. Power, Jess E. Thompson, Eóin W. Parkinson, Jay T. Stock, Tamsin C. O’Connell & Simon Stoddart     2.1. Introduction 27     2.2. The research questions of the FRAGSUS Population History Workgroup 27       2.2.1. The complex multiple interments at the Circle 27     2.3. Methods employed by the FRAGSUS Population History Workgroup 31     2.4. The first phase: isolation and archiving 32     2.5. The second phase: analyses of the isolated elements 34     2.6. The third phase: data analysis and interpretation 37     2.7. Other studies 37       2.7.1. Funerary taphonomic analysis 37       2.7.2. Metric analysis of the long bones: long bone cross-sectional geometry and body size 37       2.7.3. GIS digitizing 38     2.8. Challenges and concluding thoughts 38 Chapter 3 The chronology, structure and stratigraphy of the Circle 39 Eóin W. Parkinson, Rowan McLaughlin, Robert Barratt, Sara Boyle, John Meneely, Ronika K. Power, Bernardette Mercieca-Spiteri, Jess E. Thompson, Simon Stoddart & Caroline Malone     3.1. Introduction 39       3.1.1. Archaeological and environmental context 39       3.1.2. Digital organization of the excavation data (RMcL, SKFS) 42       3.1.3. The implementation of appropriate field methods (SKFS & CATM) 43       3.1.4. Intra-site GIS 43       3.1.5. 3D modelling and reconstruction 45 vi     3.2. Time: dating and chronological modelling 48     3.3. Rock-cut tomb: structure and stratigraphy 50     3.4. Temple Period burial complex 52       3.4.1. Structure 53       3.4.2. Stratigraphy 54     3.5. Discussion and conclusions 60 Chapter 4 Dental pathology in the Circle: oral health, activity and intervention in Neolithic Malta 63 Ronika K. Power, Bernardette Mercieca-Spiteri, Rowan McLaughlin, Jess E. Thompson, Jacinta Carruthers, Sarah Massingham, Laura T. Buck, Jaap Saers, Margery Pardey, Jay Stock & John Magnussen     4.1. Introduction 63     4.2. Dental pathology 64     4.3. Materials 65     4.4. Methods 65     4.5. Results: overview 66       4.5.1. Enamel hypoplasia 66       4.5.2. Fluorosis 70       4.5.3. Carious lesions 70       4.5.4. Hypercementosis 71       4.5.5. Fractures: crown and root 74       4.5.6. Extreme wear 84       4.5.7. Additional observations: case studies 97     4.6. Discussion 120       4.6.1. Limitations of study 120       4.6.2. The Circle in context 120     4.7. Conclusion 124 Chapter 5 Dental modification in the Circle: shaping bodies, shaping culture in Neolithic Malta 127 Ronika K. Power, Bernardette Mercieca-Spiteri, Jess E. Thompson, Rowan McLaughlin, Jacinta Carruthers, Laura T. Buck, Jaap Saers, Margery Pardey, Jay Stock & John Magnussen     5.1. Introduction 127     5.2. Background 128       5.2.1. ‘Active’ versus ‘passive’ modifications 129       5.2.2. History of study in biological anthropology 130       5.2.3. Dental modification in prehistoric Africa and Europe 131     5.3. Materials 133     5.4. Methods 133     5.5. Results: overview 134       5.5.1. Modification types 138     5.6. Case studies 155       5.6.1. Congenital variation, cultural intervention, or both? 155       5.6.2. Active modification, passive alteration, or both? 159     5.7. Discussion 161       5.7.1. Limitations of study 161       5.7.2. The Circle in context 162       5.7.3. Demographic insights 163       5.7.4. Aetiologies 164       5.7.5. Cross-cultural insights 165       5.7.6. Chronological insights 167     5.8. Conclusion 170 vii Chapter 6 Dental anthropology from the Circle: non-metric traits of the posterior dentition and population relationships in the Neolithic Mediterranean 173 Ronika K. Power, Bernardette Mercieca-Spiteri, Jay T. Stock & Joel D. Irish     6.1. Introduction 173     6.2. Materials 174     6.3. Methods 175       6.3.1. Data collection 175       6.3.2. Quantitative analyses 176     6.4. Results 178     6.5. Discussion and conclusions 181 Chapter 7 Physical activity and body size in Temple Period Malta: biomechanical analysis of commingled and fragmentary long bones 183 Eóin W. Parkinson & Jay T. Stock     7.1. Introduction 183     7.2. Materials 184       7.2.1. Sampling strategy 184       7.2.2. Comparative sample 184     7.3. Methods 184       7.3.1. Long bone cross-sectional geometry 184       7.3.2. Approaches to fragmented material 185       7.3.3. Statistical approach 187     7.4. Results 187       7.4.1. Upper and lower limb CSG properties 187       7.4.2. Body size 190     7.5. Discussion 190     7.6. Conclusion 192 Chapter 8 General pathology in the Circle: biocultural insights into population health, trauma and care in Neolithic Malta 195 Ronika K. Power, Bernardette Mercieca-Spiteri, Jess E. Thompson, Rowan McLaughlin, Jacinta Carruthers, Hannah Vogel, Laura T. Buck, Jaap Saers, Margery Pardey, Jay T. Stock & John S. Magnussen     8.1. Introduction 195     8.2. Materials 195     8.3. Methods 197     8.4. Results: overview 198       8.4.1. Spinal pathologies 199       8.4.2. Extreme extremities 199       8.4.3. Notable absences 199       8.4.4. Persons of interest 202       8.4.5. Taphonomic considerations 202     8.5. Results: case studies 204       8.5.1. Trauma: cranium and mandible 204       8.5.2. Other cranial pathology 207       8.5.3. Post-cranial pathology 221     8.6. Methodology case study: periosteal lesions in Context (960) 241       8.6.1. Context (960) adult remains 241       8.6.2. Context (960) non-adult remains 254     8.7. Discussion 272       8.7.1. Limitations of study 272       8.7.2. The Circle in context 273       8.7.3. The bioarchaeology of care 274 viii       8.7.4. Hyperostosis frontalis interna 276       8.7.5. Endocranial lesions 278       8.7.6. Nutrition 280       8.7.7. Interpersonal violence 282     8.8. Conclusion 284 Chapter 9 An isotopic study of provenance and residential mobility at the Circle and the Xemxija tombs 287 Argyro Nafplioti, Rowan McLaughlin, Emma Lightfoot, Ronika K. Power, Bernardette Mercieca-Spiteri, Catherine J. Kneale, Tamsin C. O’Connell, Simon Stoddart & Caroline Malone     9.1. Introduction 287     9.2. Background 287       9.2.1. Neolithic transition and population/settlement history in the Maltese Islands 287     9.3. Materials and methods 288       9.3.1. Principles of oxygen isotope ratio analysis 288       9.3.2. Principles of strontium isotope ratio analysis 288       9.3.3. The geological context 289       9.3.4. Samples 289       9.3.5. Sample preparation and analytical procedures 290     9.4. Results 291     9.5. Discussion 292       9.5.1. Variation in oxygen isotope ratio values 292       9.5.2. Variation in 87Sr/86Sr values 293     9.6. Conclusion 293 Chapter 10 An isotopic study of palaeodiet at the Circle and the Xemxija tombs 295 Rowan McLaughlin, Ronika K. Power, Bernardette Mercieca-Spiteri, Eóin W. Parkinson, Emma Lightfoot, Alessandra Varalli, Jess E. Thompson, Catherine J. Kneale, Tamsin C. O’Connell, Paula J. Reimer, Simon Stoddart & Caroline Malone     10.1. Introduction 295     10.2. Background: previous work 295     10.3. Food sources in prehistoric Malta 295       10.3.1. Tooth enamel samples 296       10.3.2. Bone collagen 296     10.4. Data analysis 296     10.5. Results 296     10.6. Discussion and conclusion 302 Chapter 11 aDNA: an investigation of uniparental genetic heritage in Neolithic Malta 303 Bruno Ariano, Valeria Mattiangeli, Rowan McLaughlin, Ronika K. Power, Jay T. Stock, Bernardette Mercieca-Spiteri, Simon Stoddart, Caroline Malone & Dan Bradley     11.1. Introduction 303       11.1.1. The genome and ancient DNA 303       11.1.2. Ancient DNA 303       11.1.3. Background: the genetic context of the Mesolithic in Europe 304       11.1.4. The genetic impact of the agricultural revolution 304       11.1.5. Arrival in Malta 305     11.2. Research questions 306     11.3. Methods 306       11.3.1. aDNA data collection and mitochondrial analysis 306       11.3.2. Contamination 306 ix       11.3.3. Y-chromosome haplogroup determination 306       11.3.4. Collection of publicly available data 306     11.4. Results 307       11.4.1. Mitochondrial contamination and history 307       11.4.2. Y-chromosome contamination and lineages 307     11.5. Discussion 309     11.6. Conclusion 310 Chapter 12 Reconstructing deathways at the Circle and the Xemxija tombs through funerary taphonomy 311 Jess. E. Thompson, Ronika K. Power, Bernardette Mercieca-Spiteri, Rowan McLaughlin, John Robb, Simon Stoddart & Caroline Malone     12.1. Introduction 311     12.2. Funerary taphonomy 312     12.3. Materials and sampling strategy 313     12.4. Methods 314     12.5. Results 316       12.5.1. Taphonomic analysis 316       12.5.2. Skeletal element representation 320       12.5.3. In situ articulation 326     12.6. Deathways at the Circle and the Xemxija Tombs 328     12.7. Conclusion 331 Chapter 13 The development of Late Neolithic burial places in Malta and Gozo: an overview 333 Anthony Pace     13.1. Introduction 333     13.2. Background 333     13.3. Central Mediterranean beginnings 340     13.4. The development of burial sites in Malta and Gozo 343       13.4.1. Għar Dalam and Skorba 343       13.4.2. Żebbuġ 343       13.4.3. Ġgantija 346       13.4.4. Tarxien 350     13.5. Neolithic architecture 352     13.6. Summary 354 Chapter 14 Conclusion: current inferences from the study of death in prehistoric Malta 355 Simon Stoddart, Caroline Malone, Ronika K. Power, Rowan McLaughlin & Jess E. Thompson     14.1. The state of death in prehistoric Malta 355     14.2. The sample 357     14.3. Who was buried? 358     14.4. When were they buried? 359     14.5. Can we identify changes in the health of the prehistoric populations? 360     14.6. Contemporary body concepts and belief in Neolithic Malta 360     14.7. Conclusion 363     14.8. Whither death studies in Malta? 363 References 365 Appendix 1 Archive images from the 1987–94 campaign 409 Appendix 2 Tables 451 Glossary 469 Index 471 xi Contributors Bruno Ariano Smurfit Institute of Genetics, School of Genetics and Microbiology, Trinity College Dublin, Dublin 2, D02 VF25, Ireland Email: arianob@tcd.ie Robert Peter Barratt School of Natural and Built Environment, Queen’s University Belfast, UK Email: rbarratt01@qub.ac.uk Sara Boyle (now Stewart) Ordnance Survey Northern Ireland, Land and Property Services, Lanyon Plaza, 7 Lanyon Place, Town Parks, Belfast, Northern Ireland. Laura T. Buck Research Centre for Evolutionary Anthropology and Palaeoecology, Liverpool John Moores University, United Kingdom / Cambridge Biotomography Centre, University of Cambridge, Cambridge, United Kingdom Email: l.buck@ljmu.ac.uk Dan Bradley Smurfit Institute of Genetics, School of Genetics and Microbiology, Trinity College Dublin Dublin 2, D02 VF25, Ireland Email: DBRADLEY@tcd.ie Jacinta Carruthers Department of History and Archaeology, Macquarie University, Sydney, Australia Email: jacinta.carruthers@mq.edu.au Joel D. Irish School of Biological and Environmental Sciences, Research Centre for Evolutionary Anthropology and Palaeoecology, Liverpool John Moores University, Liverpool, United Kingdom Email: J.D.Irish@ljmu.ac.uk Catherine J. Kneale McDonald Institute for Archaeological Research, Department of Archaeology, University of Cambridge, Cambridge, United Kingdom Email: cjk37@cam.ac.uk Emma Lightfoot c/o McDonald Institute for Archaeological Research, Department of Archaeology, University of Cambridge, Cambridge, United Kingdom. John S. Magnussen Department of Clinical Medicine, Faculty of Medicine Health and Human Sciences, Macquarie University, Sydney,Australia Email: john.magnussen@mq.edu.au Caroline Malone School of Natural and Built Environment, Queen’s University Belfast Email: c.malone@qub.ac.uk & caroline.stoddart@cantab.net Sarah Massingham Department of History and Archaeology, Macquarie University, Sydney, Australia Email: sarah.massingham@hdr.mq.edu.au Valeria Mattiangeli Smurfit Institute of Genetics, School of Genetics and Microbiology, Trinity College Dublin, Dublin 2, D02 VF25, Ireland Email: mattianv@tcd.ie Rowan McLaughlin School of Natural and Built Environment, Queen’s University Belfast, UK. Email: r.mclaughlin@qub.ac.uk John Meneely School of Natural and Built Environment, Queen’s University Belfast, UK Email: j.meneely@qub.ac.uk Bernardette Mercieca-Spiteri National Inventory, Research and Archaeology Unit, Superintendence of Cultural Heritage, Valletta Malta Email: bernardette.mercieca@gov.mt Argyro Nafplioti Ancient DNA Lab, Institute of Molecular Biology and Biotechnology (IMBB), FORTH, Heraklion, Greece / formerly McDonald Institute for Archaeological research, Department of Archaeology, University of Cambridge, Cambridge, United Kingdom. Email: argyro.nafplioti@googlemail.com Tamsin C. O’Connell McDonald Institute for Archaeological Research, Department of Archaeology, University of Cambridge, Cambridge, United Kingdom Email: tco21@cam.ac.uk Eóin W. Parkinson Department of Classics & Archaeology, University of Malta, Msida, Malta Email: ewparkinson24@gmail.com Anthony Pace Former Superintendent of Cultural Heritage Malta, Malta Email: anthonypace@cantab.net Margery Pardey Department of Clinical Medicine, Faculty of Medicine Health and Human Sciences, Macquarie University, Sydney, Australia Email: margery.pardey@mq.edu.au Ronika K. Power Department of History and Archaeology, Macquarie University, Sydney, Australia/ formerly McDonald Institute for Archaeological Research, Department of Archaeology, University of Cambridge, Cambridge, United Kingdom Email: ronika.power@mq.edu.au Paula J. Reimer School of Natural and Built Environment, Queen’s University Belfast, UK Email: p.j.reimer@qub.ac.uk John Robb McDonald Institute for Archaeological Research, Department of Archaeology, University of Cambridge, Cambridge, United Kingdom Email: jer39@cam.ac.uk Jaap Saers McDonald Institute for Archaeological Research, Department of Archaeology, University of Cambridge, Cambridge, United Kingdom Email: jpps2@cam.ac.uk Jay T. Stock Department of Anthropology, University of Western Ontario, London, Canada / Department of Archaeology, Max Planck Institute for the Science of Human History, Jena, Germany Email: jay.stock@uwo.ca Simon Stoddart McDonald Institute for Archaeological Research, Department of Archaeology, University of Cambridge, Cambridge, United Kingdom Email: ss16@cam.ac.uk Jess E. Thompson McDonald Institute for Archaeological Research, Department of Archaeology, University of Cambridge, Cambridge, United Kingdom Email: jet71@cam.ac.uk Alessandra Varalli Department of Humanities, Universitat Pompeu Fabra, Barcelona Email: alessandravaralli@gmail.com Nicholas Vella Department of Archaeology and Classics, University of Malta, Msida, Malta Email: nicholas.vella@um.edu.mt Hannah Vogel Department of History and Archaeology, Macquarie University, Sydney, Australia Email: hannah.vogel@hdr.mq.edu.au xii xiii Figures 0.1. George and Sheila Mann. xxi 0.2. George Mann at work. xxii 0.3. Sheila Mann cleaning bones. xxii 0.4. Anne Monsarrat. xxiii 1.1. Ġgantija’s World Heritage status inscribed at the Visitor Centre. 4 1.2. Preliminary sketch of the Circle, Bayer’s excavation and aerial view. 6 1.3. Unknown stone structures on Gozo. 7 1.4. The ‘Circle’ field before excavation in 1987. 7 1.5. Mikiel Bartolo, the elderly tenant of the field, in 1987. 7 1.6. First trenches in 1987–88. 8 1.7. Excavation of rock-cut tomb. 9 1.8. Site in 1989–90. 10 1.9. Site in 1990–92. 12 1.10. Excavations in 1993 of Contexts (783), (783), (951) and the ‘Shrine’. 13 1.11. Excavation recordings. 14 1.12. Clearance of the Circle in 1989. 16 1.13. Bone recording in the field. 17 1.14. Amalgamated bone plan for (783). 18 1.15. Digital scan of the Circle cavity in 2015. 19 1.16. Virtual reality study of the Circle caves. 19 1.17. GIS study of the excavated bones in (783). 20 1.18. Collecting the bones transported in the Royal Yacht Britannia in 1995. 21 1.19. Graph showing the output arising from the Circle research. 23 2.1. The site under excavation. 28 2.2. Skeletons in situ showing partial articulation. 30 2.3. Disarticulated bone in situ. 30 2.4. Commingled bones with original label. 31 2.5. Population History Workgroup examining bone and cleaning. 33 2.6. Population History Workgroup isolating teeth. 33 2.7. Population History Workgroup CT-scanning. 35 2.8. Population History Workgroup CT-scanning. 35 2.9. Population History Workgroup sampling. 36 3.1. Location map of the Circle on the Xagħra plateau, Gozo. 40 3.2. 3D reconstruction of the Circle marking the main areas. 40 3.3. Plan of the principal areas within the Circle. 41 3.4. 3D reconstructions of the Circle developed by the INSITE project of the University of Bristol. 43 3.5. Output from the Visibility Mapping tool for the Circle. 45 3.6. Use of computer model for celestial alignments. 46 3.7. Elements of the wire frame for a 3D reconstruction. 46 3.8. Reconstruction of external view of the Circle. 47 3.9. Reconstruction of the steps into the Circle. 47 3.10. Reconstruction showing low roof of inner parts of the Circle. 48 3.11. Kernel Density Estimation (KDE) models of radiocarbon-dated burials at the Circle. 49 3.12. Bayesian model of the chronology of selected contexts and events. 50 3.13. Section of the rock-cut tomb. 51 3.14. Plan of the rock-cut tomb. 51 3.15. GIS digitized human bone and overall density of human bone in the Shrine. 55 3.16. Density of burial activity in context (1328) and (1268). 56 3.17. Density of burial activity in context (1206) and (960). 57 3.18. Limits of the principal contexts of the Shrine. 57 3.19. Burial deposit 783 density. 58 3.20. Location of the path through the Display Zone and distribution of the fragmented standing figure. 59 xiv 4.1. A typical assortment of postmortem exfoliated loose human teeth from the Circle excavations. 66 4.2. Frequency distribution of dental pathology observations. 67 4.3. Frequency distribution of enamel hypoplasia observations by chronological phases. 68 4.4. Frequency distribution of enamel hypoplasia observations by proportion. 68 4.5. Frequency distribution of enamel hypoplasia observations in deciduous dentition. 69 4.6. Frequency distribution of enamel hypoplasia observations in deciduous dentition by proportion. 69 4.7. Frequency distribution of carious lesion observations by chronological phases. 71 4.8. Frequency distribution of carious lesion observations by type and proportion. 71 4.9. Examples of permanent mandibular incisors presenting mamelons from Context (783). 72 4.10. Frequency distribution of hypercementosis observations by contexts. 74 4.11. Frequency distribution of hypercementosis observations by chronology. 74 4.12. Frequency distribution of crown fracture observations by chronology. 75 4.13. Frequency distribution of crown fracture observations by tooth type. 76 4.14. Dynamic schema of crown fracture observations by tooth type and arcade side. 77 4.15. Frequency distribution of crown fracture observations across specific cusp locations. 77 4.16. Dynamic schema of crown fracture observations across specific cusp locations. 78 4.17. Frequency distribution of crown fracture observations across general cusp locations. 79 4.18. Dynamic schema of crown fracture observations across general cusp locations. 79 4.19. Complicated vertical crown and root fracture. 82 4.20. Complicated vertical crown and root fractures. 82 4.21. Complicated vertical crown and root fractures. 83 4.22. Complicated vertical crown and root fracture. 83 4.23. Complicated vertical crown and root fracture. 83 4.24. Frequency distribution of extreme wear observations by chronological phases. 85 4.25. Frequency distribution of extreme wear observations by proportion. 85 4.26. Frequency distribution of extreme wear observations by tooth type. 85 4.27. Dynamic schema of extreme wear observations by tooth type and arcade side. 86 4.28. Frequency distribution of extreme wear observations by location and orientation. 86 4.29. Frequency distribution of labial extreme wear observations. 87 4.30. Frequency distribution of lingual extreme wear observations. 88 4.31. Frequency distribution of extreme diagonal wear observations. 89 4.32. Frequency distribution of extreme approximately vertical wear observations. 89 4.33. Frequency distribution of extreme wear beyond the cementum enamel junction observations. 90 4.34. Frequency distribution of maxillary labial crescentic wear. 90 4.35. Frequency distribution of mandibular labial crescentic wear. 91 4.36. Intact adult left maxillary arcade. 91 4.37. Frequency distribution of extreme wear observations. 93 4.38. Dynamic schema of extreme wear observations across specific cusp locations. 93 4.39. Frequency distribution of extreme wear observations across general cusp locations. 94 4.40. Dynamic schema of extreme wear observations across general cusp locations. 95 4.41. Fragment of adult left mandibular corpus. 97 4.42. Mandibular corpus fragment. 98 4.43. Mandibular corpus fragment. 100 4.44. Examples of extreme calculus deposition. 102 4.45. Examples of ‘peg’ maxillary lateral incisors. 104 4.46. Examples of accessory root(let)s. 104 4.47. Examples of dilaceration, cervical carious lesion and hypercementosis. 105 4.48. Examples of fusion. 105 4.49. Examples of dwarfism. 106 4.50. Fragmentary left adult maxilla and mandible. 107 4.51. Fragmentary right adult maxillae. 108 4.52. Fragmentary refitting right and left adult maxillae. 109 4.53. Fragmentary right adult maxilla. 110 4.54. Fragmentary refitting right and left adult maxillae. 111 xv 4.55. Fragmentary articulating left and right adult maxillae. 112 4.56. Fragmentary articulating left and right adult maxillae. 113 4.57. Fragmentary maxillae from Context (1328) and Context (1206). 114 4.58. Fragmentary splanchnocranium from Context (1206). 115 4.59. Fragmentary articulating left and right adult maxillae from Context (845). 116 4.60. Fragmentary left adult mandibular corpus from Context (876). 117 4.61. Schematic chronological representation of dental pathology incidence. 122 5.1. Frequency distribution of dental modification observations by chronological phases. 135 5.2. Frequency distribution of tooth types featuring dental modification. 136 5.3. Dynamic schema of dental modification observations by permanent anterior tooth type and arcade side. 137 5.4. Dynamic schema of dental modification observations by deciduous anterior tooth type and arcade side. 137 5.5. Frequency distribution of dental modification types. 138 5.6. Examples of chipping modifications. 139 5.7. Frequency distribution of chipping modifications by chronology. 139 5.8. Frequency distribution of chipping by tooth types. 140 5.9. Frequency distribution of chipping by quadrants. 140 5.10. Examples of incisal notch modifications. 141 5.11. Frequency distribution of incisal notch modifications by chronology. 142 5.12. Frequency distribution of incisal notches by tooth types. 142 5.13. Frequency distribution of incisal notches by quadrants. 143 5.14. Examples of labial extreme wear modifications. 144 5.15. Frequency distribution of labial extreme wear by chronology. 145 5.16. Frequency distribution of labial extreme wear by tooth types. 145 5.17. Frequency distribution of labial extreme wear by quadrants. 146 5.18. Examples of lingual extreme wear modifications. 147 5.19. Frequency distribution of lingual extreme wear by chronology. 147 5.20. Frequency distribution of lingual extreme wear by tooth types. 148 5.21. Frequency distribution of labial extreme wear by quadrants. 148 5.22. Examples of diagonal profile modifications. 149 5.23. Frequency distribution of diagonal profile modification by chronology. 150 5.24. Frequency distribution of diagonal profile modification by tooth types. 151 5.25. Frequency distribution of diagonal profile modification by quadrants. 151 5.26. Frequency distribution of undulating profile modification by chronology. 152 5.27. Frequency distribution of undulating profile modification by tooth type. 153 5.28. Frequency distribution of undulating profile modification by quadrants. 153 5.29. Almost complete skull of probable female adult individual with variation of eruption and alignment. 156 5.30. 3D render and cross sections of cranium in anterior, right lateral and posterior views. 157 5.31. Adult splanchnocranium featuring modification of the left and right maxillary central incisors. 160 5.32. Schematic chronological representation of dental modification incidence. 169 6.1. Three-dimensional MDS of 12-trait MMD distances. 181 6.2. Two-dimensional scatterplot relative to the 13 comparative samples. 181 7.1. Examples of digitally reconstructed humeri. 186 7.2. Example of shape fitting method applied to fragmented femoral head. 187 7.3. Comparison of solid CSG properties of the humerus between Xagħra and central Italy. 188 7.4. Comparison of solid CSG properties of the femur between Xagħra and central Italy. 189 7.5. Comparison of solid CSG properties of the tibia between Xagħra and central Italy. 189 7.6. Side differences in TA and J of the humerus at Xagħra. 189 7.7. Side differences in cross-sectional shape of the humerus at Xagħra. 189 7.8. Comparisons of estimated stature and body mass between Xagħra and central Italy. 190 8.1. Typical disarticulated, fragmented and commingled human remains. 196 8.2. Frequency distribution of human remains of special interest from the Circle. 198 8.3. A representative selection of vertebrae presenting extreme pathology. 200 8.4. A representative selection of metacarpals, metatarsals and phalanges presenting extreme pathology. 201 8.5. A selection of indicators showing site used across the life course. 203 xvi 8.6. Images of mandible from Context (951) detailing healed fracture. 205 8.7. Images of adult mandible from Context (951) detailing healed fracture. 206 8.8. Fragmentary remains of healed nasal bone in cranium from Context (468). 207 8.9. Adult cranium from Context (783) with focal depression on ectocranial surface. 208 8.10. Adult cranium from Context (1238) with two focal depressions on the ectocranial surface. 209 8.11. Assorted adult cranial fragments from Context (951) with abnormal endocranial focal impressions. 210 8.12. Assorted adult cranial fragments from Context (951) with abnormal endocranial focal impressions. 211 8.13. Adult right parietal fragment from Context (951) with abnormal endocranial focal impressions. 212 8.14. Adult left parietal fragments from Context (951) with abnormal endocranial focal impressions. 213 8.15. Assorted adult cranial fragments from Context (518) with abnormal endocranial focal impressions. 214 8.16. Assorted adult cranial fragments from Context (838) with abnormal endocranial focal impressions. 215 8.17. Photographic and radiological images of adult frontal bone from Context (1268). 216 8.18. Photographic and radiological images of fragmentary adult cranium. 217 8.19. Photographic and radiological images detailing endocranial lesion in the frontal fossa. 218 8.20. Radiological images detailing a proliferative endocranial lesion within the frontal fossa. 220 8.21. Images of second cervical vertebra exhibiting severe degenerative joint disease and dysplasia. 221 8.22. Photographic and radiological images of twelfth thoracic vertebra from Context (1024). 222 8.23. Fragment of adult lumbar vertebral body exhibiting a central compression fracture. 224 8.24. Images of first lumbar vertebra exhibiting a central compression fracture. 225 8.25. Images of a fragment of adult lumbar vertebral body exhibiting a central compression fracture. 226 8.26. Fragment of adult lumbar vertebral body exhibiting an anterior compression fracture. 227 8.27. Images of right humerus detailing a complex non-reduced healed fracture. 229 8.28. Fragment of adult right humerus exhibiting dysplasia of the of the mid-diaphysis. 230 8.29. Images of right ulna detailing a healed fracture of the distal diaphysis. 231 8.30. Images of left ulna detailing a healed fracture of the distal diaphysis. 233 8.31. Images of left ulna a healed fracture of the distal diaphysis. 234 8.32. Images of adult right ulna detailing a healed fracture of the distal diaphysis. 235 8.33. Complete adult left ulna exhibiting healed fracture of the styloid process. 236 8.34. Fragmentary adult right ulna exhibiting healed fracture of the styloid process. 237 8.35. Images of left femur detailing a complete non-reduced healed transverse fracture. 238 8.36. Images of fragmentary tibia and fibula detailing fusion of the distal diaphysis and metaphyses. 240 8.37. Images of left radius depicting refitted fragments and periosteal lesions. 243 8.38. Images of left ulna depicting refitted fragments and periosteal lesions. 244 8.39. Images of left femur depicting refitted fragments and periosteal lesions. 246 8.40. Images of right tibia depicting periosteal lesions on the refitted fragments of the proximal diaphysis. 248 8.41. Images of left fibula depicting periosteal lesions traversing the refitted fragments. 250 8.42. Images of right fibula depicting periosteal lesions traversing the refitted fragments. 252 8.43. Fragmented adult left radius and ulna from Context (960). 253 8.44. Images of frontal from Context (960) displaying extensive periosteal lesions. 255 8.45. Images of frontal from Context (960) displaying extensive periosteal lesions. 257 8.46. Images of frontal from Context (960) displaying extensive periosteal lesions. 259 8.47. Images of left zygoma from Context (960) displaying extensive periosteal lesions. 260 8.48. Images of non-adult mandible from Context (960) displaying focal periosteal lesions. 263 8.49. Images of non-adult left rib from Context (960) displaying focal periosteal lesions. 266 8.50. Images of non-adult right rib from Context (960) displaying focal periosteal lesions. 269 8.51. Images of fragmented non-adult ulna displaying circumferential periosteal new bone. 269 9.1. Biplot of oxygen and carbon isotope ratios. 289 9.2. Histogram of enamel δ18Op values indicating a bimodal distribution. 290 9.3. Biplot of enamel δ18Op and strontium isotope ratio values. 292 9.4. Histogram of enamel strontium isotope ratios. 292 9.5. Boxplot of average enamel δ18O values for four categories of context stratified in time. 293 10.1. Stripcharts indicating the range and distribution of carbon and nitrogen isotope values. 298 10.2. Biplot of bone collagen carbon and nitrogen isotopic values. 298 10.3. Modelled contribution of foods to the palaeodiet. 299 xvii 10.4. Boxplots of the main isotopic results by context. 300 10.5. Nitrogen isotopic values of animal bone collagen. 301 10.6. Multiple regression modelling the dependence of nitrogen-15 values on time. 301 11.1. Reconstruction of the Circle. 305 11.2. Distribution of ancient mitochondrial haplogroups in Eurasia. 308 11.3. Distribution of ancient Y haplogroups in ancient Eurasia. 309 12.1. Heatmap illustrating the location of remains analysed from the Circle. 314 12.2. Cortical thinning, splintering and delamination indicating weathering on bone. 318 12.3. Insect modifications to a range of elements from the Xemxija and Circle rock cut tombs. 318 12.4. Rodent gnawing to elements from the Xemxija Tombs. 319 12.5. Burning on endocranial surface of fragments from Context (326) of the Circle rock-cut tomb. 320 12.6. Skeletal element representation in the Xemxija Tombs indicating a residual pattern. 322 12.7. Skeletal element representation profiles reflecting cranial curation. 323 12.8. Skeletal element representation profiles with an over-representation of long bones. 323 12.9. Skeletal element representation profiles indicating primary successive deposition. 324 12.10. Skeletal element representation profiles indicating primary successive deposition. 325 12.11. Articulated non-adult skeleton in spit 3 at the base of (783). 326 12.12. Remains of primary inhumations evident as axial skeletons in spit 3. 327 12.13. Perinate in (1206), spit 4, units 23 and 25. 327 12.14. Infant (1206), spit 3, skeleton 19. 327 12.15. Proposed model of deathways. 329 13.1. Distribution of rock-cut burial sites referred to in this chapter. 334 13.2. Old view of the Ħal Saflieni Hypogeum. 334 13.3. Buqana tomb and the burial cave at Bur Megħeż. 336 13.4. Bur Megħeż aerial photo and plan. 337 13.5. Regional presence of rock-cut tombs in the Central Mediterranean. 341 13.6. Schematic flow chart summarising changes in mortuary custom, landscape and settlement. 342 13.7. Żebbuġ Phase tombs. 344 13.8. Ġgantija Phase rock-cut tombs. 346 13.9. Spatial patterning of rock-cut tombs from the Żebbuġ through to the Tarxien phases. 348 13.10. Map of the Xagħra Plateau, Gozo. 349 13.11. Map of the Tarxien, Kordin and Santa Luċija plateau. 350 13.12. Map of the Xemxija promontory. 351 13.13. Plan of levels 2 and 3 of the Ħal Saflieni Hypogeum. 352 13.14. Site plan of the Circle. 353 A1.1. The lower field below the Circle in 1987. 410 A1.2. View to the southeast towards the excavation of the rock-cut tomb in 1989. 410 A1.3. Rock-cut tomb under excavation, with Bridget Trump in the entrance to the West chamber in 1988. 411 A1.4. West chamber of the rock-cut tomb in 1988, exposing human remains deposited in an ochred matrix. 411 A1.5. Assemblage of human remains and ceramic sherds under excavation in the East chamber. 412 A1.6. Large ceramic Żebbuġ sherd surrounded by red ochre and fragmented human remains. 412 A1.7. Deposit of commingled human and faunal remains in the rock-cut tomb. 413 A1.8. Group of disarticulated human remains towards the edge of the West chamber. 413 A1.9. Group of fragmented human remains in the West chamber. 414 A1.10. Large triton shell against ceramic sherd below the entrance to the East chamber. 414 A1.11. Miniature Saflieni phase ochre pots from the Circle. 415 A1.12. Mikiel Bartolo, the elderly tenant farmer, visiting the Circle during excavation in 1988. 415 A1.13. View to the northeast in 1989, overlooking early excavation of the East Cave and rock-cut tomb. 416 A1.14. Context (354) in the upper level of the North bone pit adjacent to the threshold. 416 A1.15. Context (622) in the mid-levels of the North bone pit. 417 A1.16. Context (622) with disarticulate bone elements and an articulate hand. 417 A1.17. Context (622) or (623) in the mid-levels of the North bone pit. 418 A1.18. A fragment of right ilium is articulated with a femur and patella in Context (799). 418 A1.19. The undisturbed remains of an adult male in the base levels of the North bone pit. 419 xviii A1.20. Excavation in 1993 showing upper levels in the ‘Shrine’ area. 419 A1.21. The stone bowl (841) prior to excavation. 420 A1.22. Articulated and nearly complete hand preserved at the base of the stone bowl in Context (842). 420 A1.23. Twin-seated figure (SF743) and ceramic strainer in situ in Context (831). 421 A1.24. Figurine cache (SF784) under excavation in Context (831) by conservator Carol Brown, 1991. 421 A1.25. Figurine cache (SF784) fully excavated and in situ in Context (831), 1991. 422 A1.26. Context (960) under excavation in the upper levels of the ‘Shrine’. 422 A1.27. Articulated and almost complete infant skeleton in Context (960). 423 A1.28. Row of crania and skulls deposited in varying positions in Context (960). 423 A1.29. Pit deposit (Contexts 980/979) in the upper level of the ‘Shrine’. 424 A1.30. Context (1206) in the mid-level of the ‘Shrine’. 424 A1.31. Most of an articulated adolescent skeleton within Context (1206). 425 A1.32. Semi-articulated thoracic region, pelvic girdle and upper legs of an adult skeleton. 425 A1.33. Articulated and nearly complete infant skeleton in Context (1206). 426 A1.34. Discrete bone group in Context (1206). 426 A1.35. Articulated adult skeleton in the lower levels of the ‘Shrine’, in Context (1268). 427 A1.36. Bone bundles in Context (1328). 427 A1.37. Miniature figurine (SF516) carved into a medial phalanx of Ovis/Capra. 428 A1.38. North Niche viewed from the North. 428 A1.39. Almost complete articulated leg in Context (845) within the North Niche. 429 A1.40. Deposit (863) resting against megalith (878) in the North Niche. 429 A1.41. 19th-century dog burial in Context (920) at the base of the pit excavated by Bayer. 430 A1.42. Dense deposit of mostly disarticulated bones in Context (933) in the upper levels of the Deep Zone. 430 A1.43. Contexts (951) and (1144), the major bone deposits in the Deep Zone. 431 A1.44. Fragmented deposits of disarticulate bone in Context (951). 431 A1.45. Bone group in Context (951). 432 A1.46. Group of fragmented crania in Context (951), deposited in varying positions. 432 A1.47. Semi-articulated lower leg and foot in Context (1257) in the lower levels of the Deep Zone. 433 A1.48. Small limestone figurine (SF1184) from Context (474) in the upper level of the Display Zone. 433 A1.49. Figurine SF775 in situ during excavation of Context (783) in the Display Zone. 434 A1.50. Context (518) in the West niche, adjacent to the Display Zone. 434 A1.51. Upper levels of Context (783), in the Display Zone, under excavation in 1991. 435 A1.52. Cranial cluster and various other elements abutting a megalith in Context (518). 435 A1.53. Articulate ribs and column in Context (783). 436 A1.54. Context (997), spit 4, in the West niche, exposing the skull and articulated cervical vertebrae. 436 A1.55. Semi-articulated skull and torso of an adult female in Context (518). 437 A1.56. Adult skull adjacent to an articulated infant lower arm and hand in Context (783). 437 A1.57. Articulated skull with cervical, thoracic and lumbar vertebrae in Context (783) in the Display Zone. 438 A1.58. Articulated thoracic region in Context (783). 438 A1.59. Articulated right lower leg and foot in Context (783). 439 A1.60. Deposit of disarticulated and semi-articulated remains in Context (783) in the Display Zone. 439 A1.61. Semi-articulated adult individual in Context (783). 440 A1.62. Semi-articulate torso with ribs and femurs associated in Context (783). 440 A1.63. Deposit under excavation in Context (783), revealing a mixture of semi-articulated skeletal regions. 441 A1.64. Figurine SF788 in situ during excavation of Context (783) in the Display Zone. 441 A1.65. Drawing of restored SF788. 441 A1.66. Head of SF784/5 with the emerging cache stone figures revealed in 1991. 442 A1.67. Group of terracotta figurines. 442 A1.68. Fragmented skull and cervical vertebrae exposed in Context (704). 443 A1.69. Cranium of an adult male deposited alongside a boar skull in the northern burial niche. 443 A1.70. Context (1067) in the East Cave contained a partial semi-articulated child skeleton. 444 A1.71. Non-adult remains in the Central pit above the East Cave roof collapse, in Context (436). 444 A1.72. Bone group including long bones, ribs and mandible in the Central pit above the East Cave roof. 445 A1.73. Articulated ribs and thoracic vertebrae, as well as hand bones, in the Central pit. 445 xix A1.74. Betyl Zone bordering the Southeast corner of the East Cave. 446 A1.75. View into the Central Zone of the East Cave. 446 A1.76. View of the Central Zone of the East Cave from the South showing remnant cave roof, 1991. 447 A1.77. View of the West Cave and East Cave systems from the southwest. 447 A1.78. Mikiel Bartolo, the tenant of the Circle field, visiting in 1988. 448 A1.79. Caroline Malone, Simon Stoddart and David Trump in 1994. 448 A1.80. Simon Stoddart and David Trump touring the site in 1991. 449 A1.81. David Trump sorting through ceramic sherds. 449 A1.82. Corinne Duhig refitting a fragmented cranium. 450 Tables 2.1. Studies carried out on the Circle assemblage. 32 2.2. Isolated bone elements. 34 4.1. Materials included in pathology study, including provenance and representation. 65 4.2. Pathology summary by context and chronology. 67 4.3. Summary of observations of frequency and reported locations of carious lesions. 71 5.1. Materials included in dental studies, including provenance and representation. 133 5.2. Dental modification of anterior teeth summary by context and chronology. 135 5.3. Extreme wear of anterior teeth summary by context, chronology and tooth surface observations. 136 6.1. Materials included in pathology study, including provenance and representation. 175 6.2. The 13 comparative dental samples. 176 6.3. Frequencies of the 39 dental traits by ASUDAS grade. 177 6.4. Dental trait percentages and number of individuals scored. 178 6.5. Component loadings, eigenvalues, and variance explained for 22 traits. 180 6.6. Pairwise distances between Malta and the 13 comparative samples. 180 7.1. Cross-sectional geometric properties used in the study. 185 7.2. Summary statistics and results of independent t-test of upper limb (humerus). 187 7.3. Summary statistics and results of independent t-test of lower limb (femur and tibia). 188 7.4. Descriptive statistics for CSG properties of the humerus. 188 7.5. Descriptive statistics and results of independent t-tests comparing body mass and stature. 190 10.1. Comparative terrestrial faunal isotope data from the Maltese Islands. 296 10.2. Comparative stable isotope measurements from ancient fish bone collagen. 296 10.3. Summary results for each site. 297 10.4. Offsets between the δ15Ncoll and δ13Ccoll human and animal values. 299 10.5. Number of samples of human bone from well-sampled contexts subjected to isotopic analysis. 300 11.1. Results from the contamination analysis. 307 11.2. Haplogroup assignment from Haplofind. 307 11.3. Sex assignment for each sample. 307 11.4. Values associated with contamination level using the X chromosome in male individuals. 307 12.1. Total number of fragments analysed from the Xemxija Tombs and contexts at the Circle. 313 12.2. Description of taphonomic variables recorded in this study. 315 12.3. Overall taphonomic results for the Xemxija Tombs and the Circle. 317 12.4. Examples of typical funerary practices and their expected skeletal part representation. 321 12.5. Chronological phasing of contexts and dominant funerary practices 328 13.1. Catalogue of known and likely late Neolithic burial sites in Malta and Gozo. 335 13.2. Prehistoric burial sequence based on ceramic remains identified from rock-cut tombs. 338 14.1. Known prehistoric funerary remains in the Maltese Islands. 356 A2.1. Radiocarbon dates. 451 A2.2. Strontium and oxygen isotope results. 453 A2.3. Palaeodietary results from human remains from the Circle. 456 A2.4. Palaeodietary results from human remains from Xemxija. 459 A2.5. MNE and BRI for all adult skeletal elements from the Xemxija Tombs. 460 xx A2.6. MNE and BRI for all non-adult skeletal elements from the Xemxija Tombs. 460 A2.7. MNE, MNI, BRI and FI for (276) in the West chamber of the rock-cut tomb. 461 A2.8. MNE, MNI, BRI and FI for (326) in the East chamber of the rock-cut tomb. 461 A2.9. MNE, MNI, BRI and FI for (354) in the North bone pit. 462 A2.10. MNE, MNI, BRI and FI for (799) in the North bone pit. 462 A2.11. MNE, MNI, BRI and FI for (783) in the Display zone. 463 A2.12. MNE, MNI, BRI and FI for (951) in the Deep zone. 463 A2.13. MNE, MNI, BRI and FI for (1144) in the Deep zone. 464 A2.14. MNE, MNI, BRI and FI for (1307) in the Deep zone. 464 A2.15. MNE, MNI, BRI and FI for (960) in the ‘Shrine’. 465 A2.16. MNE, MNI, BRI and FI for (1024) in the ‘Shrine’. 465 A2.17. MNE, MNI, BRI and FI for (1206) in the ‘Shrine’. 466 A2.18. MNE, MNI, BRI and FI for (436) in the Central pit. 466 A2.19. MNE, MNI, BRI and FI for (743) in the Central bone pit. 467 A2.20. MNE, MNI, BRI and FI for (595) in the Southwest niche. 467 A2.21. MNE, MNI, BRI and FI for (656) in the Southwest niche. 468 A2.22. MNE, MNI, BRI and FI for (734) in the Southwest niche. 468 xxi The re-discovery of the Xagħra Brochtorff Circle (1987–94) and the retrieval of prehistoric burials from the site represents the material that is the subject of this volume. Here we reassess and delve deeper into the detail of the excavated remains of a large prehis- toric population and other prehistoric burials known from Malta and Gozo. The original Xagħra fieldwork was intense, hot and hard, and it took place mostly at the height of summer, during university vacations. Such work was not for the faint-hearted; early morn- ing routines and 6-day weeks, crowded communal conditions – these were the standard experience for the young team of students and professional archaeologists who participated. It was an exciting learning experience for the ‘young ones’. For two much older men, retired from their careers, to choose to participate in this frenetic and noisy environment was unexpected, but enormously significant and supportive to what was then a major and pioneering undertaking. These gentlemen, Dr George Mann (a retired ENT consultant from Addenbrookes Hospital in Cambridge with a Masters in biological anthropol- ogy), and Kenneth Stoddart (just retired from a life of city commuting and business), brought maturity, wisdom, humour, compassion and humanity, as well as a vital breath of civilization to each annual season of work. We dedicated the 2009 volume to the mem- ory of Kenneth Stoddart. This volume appropriately is dedicated to the memory of George Mann. Dr George Edgar Mann (1923–2019) participated in the Gozo Project between 1990 and the completion of osteological study in 1996. Initially George, fresh from a post-retirement study of bioanthropology at Cambridge, came to assist Corinne Duhig who pre- pared the initial rock-cut tomb report. Professionally he had been a specialist consultant in otolaryngology at Addenbrookes Hospital in Cambridge, and had done his retirement MPhil dissertation on bony exos- toses in the outer meatus of the ear, caused by swim- In memoriam George Mann Caroline Malone Figure 0.1. George and Sheila Mann at work in the kitchen of the dig house, systematically recording a skeleton 1994. xxii In memoriam George Mann ming in cold water. The Gozo assemblage demanded a rapid revision of his knowledge of the post cranial skeleton, but soon up to speed, George then came every year to participate in each field season and post-excavation study season. He worked tirelessly with his wife Sheila, processing the excavated bones, separating out the animal bones for study by Ger- aldine Barber, and identifying the human remains himself with his team. He cheerfully accepted the spartan and crowded living conditions where he spent much time at the kitchen table or on the roof of rented holiday flats, sorting endless sacks of bone fragments into coherent identified catalogues. He measured, studied and quantified as he went and ensured every fragment was recorded. Towards the end of the fieldwork, some osteological material was transported to Britain, and George continued to log, measure, examine and interpret the human material in preparation for the 2009 report. His sys- tematic and painstaking recording work of the entire assemblage was of great importance, as the following pages reveal. Even with the ERC FRAGSUS Project resources, which provided funding at a level unim- agined in the earlier excavation years, it has been possible only to re-examine a sample of the vast oste- ological archive. George managed to ensure that we have the fundamental knowledge of the scope of the assemblage, and this is listed in the first report (see Malone et al. 2009d) and it forms the base for ongoing research of these remarkable ancient people and the Xagħra site. The record was written by hand, and the hundreds of sheets of record remain in the archives of the National Museum of Archaeology, ready for future studies, and whilst the original digital data- base of those handwritten records becomes ever more antiquated, George’s immense work remains a vital archive even as technology advances. All the teams, past and present, are delighted to dedicate this volume to George’s memory and his tremendous contribution to Maltese and osteological scholarship. Another key contributor to the work of the original Gozo Project was Ann Monsarrat, who lived on Gozo, and supported the project and its team with generosity and warmth over the many years of work and study. Figure 0.2. George Mann at work on the roof-top of the dig house in Gozo in 1994. Figure 0.3. Sheila Mann cleaning bones for George in the dig house 1994. xxiii Ann Monsarrat (1937–2020) made her home on Gozo, where she moved in 1968 with her husband Nicholas, the author of many novels about Malta and the sea. Gozo was a special place for Ann, a home with peo- ple that she truly loved, respected and admired. Ann was a remarkable person. She was welcomed and felt at home in the small village of San Lawrenz, where she lived for more than four decades. Her house was forever busy with people dropping in and sharing news, experiences, aspirations, the changing fortunes of Malta and Gozo and, of course, the difficulties of writing and the literary world. But beyond these and many other conversations, Ann was particularly interested in landscape – Gozo’s in particular – where archaeology, history and legends carved meaning out of a small island full of hills, valleys, majestic cliffs and skylines marked by parish church cupolas rising above quiet village houses. FRAGSUS owes a great deal to Ann. For, unbe- known to her, several good friends – all archaeolo- gists – whom she supported and entertained annu- ally during the excavation of the Xagħra Brochtorff Circle between 1987 and 1994, came together again to deliver another important project. Ann would have certainly been happy and excited with the results of FRAGSUS. A career journalist and a distinguished author in her own right, with works such as And the Bride wore; Thackeray: An Uneasy Victorian; Gozo: island of oblivion, a graphic literary itinerary, Ann was particularly interested in the archaeology of Malta and Gozo. She was always keen to follow research developments and new discoveries, and was eager to see young scholars, budding archaeologists, pho- tographers, historians, artists, writers, journalists, and so many others making headway in areas that she understood to be important in promoting Maltese cultural identity. Ann was in fact a formidable advo- cate of Maltese arts, culture and cultural heritage. Her work on the governing board of Saint James Cavalier Centre for Creativity in Valletta, and her continuous presence in Gozitan cultural circles, as well as her various contributions to numerous publication pro- jects reflected an enthusiasm and positiveness which was contagious and encouraging. Ann’s enthusiasm shone every time she visited the Xagħra Brochtorff Circle excavations, during our long walks along the ta’ Ċenċ promontory, during visits to the Cittadella, or when listening to the sounds rumbling from the depths of blocked shafts at the legendary clock-mak- er’s salt-works on the north coast of Gozo. These were real places with real stories, some illustrated in prints, others silently waiting to be teased out from In memoriam Ann Monsarrat Anthony Pace Figure 0.4. Anne Monsarrat (with kind permission of her family). xxiv In memoriam Ann Monsarrat stone monuments, field terraces and beautiful natural spots. Perhaps these were places whose biographies could best be understood by visiting and experienc- ing them in person. One of the last places Ann and I visited together was the archaeological site at Ras il-Wardija on Gozo’s western coast. The site is not an easy one to interpret, but from a spot rising several metres above the sur- rounding area, we shared an almost bird’s-eye view of Dwejra with the distant Azur Window below us, and we chatted about the meaning of the site and its links to the sea: seascapes, ancient mariners, people lost at sea, shipwrecks; and also of builders who constructed beautiful places and made beautiful art, making the Maltese Islands their home for at least seven thousand years. In these pages, the FRAGSUS team pays tribute to Ann Monsarrat. xxv The FRAGSUS Project 2013–18 is indebted to numer- ous individuals and institutions and acknowledges support throughout the course of a long and energetic project. In particular, we thank the European Research Council which awarded an Advanced Research Grant (Advanced Grant no. 323727) to Caroline Malone, as the Principal Investigator of the Project and the extended team in late 2012, enabling the project to take place between May 2013 and April 2018. Without this funding and the trust placed in us by the ERC and the grant assessors, this project could never have taken place. We express our gratitude for this opportunity to expand knowledge of Malta and indeed much else con- nected with prehistoric societies and their world in the early Mediterranaean. The research team also wants to record our indebtedness to the administrators of the grant within our own institutions, since this work required detailed and dedicated attention. In particular we thank Rory Jordan in the Research Support Office (Queen’s University Belfast – QUB), Laura Cousens (Cambridge University – UoC), Glen Farrugia and Cora Magri (University of Malta – UM), the Curatorial, Finance and Designs & Exhibitions Departments in Heritage Malta (HM) and Stephen Borg at the Super- intendence of Cultural Heritage (SCH). All participants thank our Maltese colleagues for their hospitable and enthusiastic support of the research, from the CEO of Heritage Malta, the Minis- ter for Culture Dr Owen Bonnici, and the Malta High Commissioner, his excellency Norman Hamilton and the British High Commissioner, his Excellency Stuart Gill, Dr Anthony Pace, Former Superintendent of Cultural Heritage and his deputy, Nathaniel Cutajar and their staff (Christian Mifsud, Mevrick Spiteri, Bernardette Mercieca-Spiteri) for enabling permits for the many sites explored during the research programme. We thank too the University of Malta and the staff of the Department of Archaeology and Classics, who enabled facilities, storage, meeting and research venues, student assistants, and logistics. In particular, we thank Chris Gemmell, Dr Maxine Ana- stasi, Prof. Nicholas Vella, Prof. Anthony Bonanno, and Dr Reuben Grima. The staff of Heritage Malta were instrumental in enabling access to sites, stores, and expertise. We especially thank Sharon Sultana in the National Museum of Archaeology, Ruby Cutajar when she was in the National Museum of Archaeol- ogy, Katya Stroud, Joanne Mallia, Dr Josef Caruana in Malta and Dr George Azzopardi, Daphne Sant Caruana, Nicolene Sagona and John Cremona in Gozo. In Britain, the team thank the staff of the McDon- ald Institute, particularly Emma Jarman, and the Department of Archaeology for facilities and support and the Cambridge Biotomography Centre for access and use of facilities. At Queen’s University Belfast, we thank the team in the 14CHRONO Lab under the direc- tion of Prof. Paula Reimer for the swift and effective processing of AMS dates and isotopic analysis and, in particular, Stephen Hoper, Michelle Thompson, Jim McDonald, and Dr Ron Reimer who made this possi- ble. We also thank John Meneeley for his exceptional work on scanning sites and processing data to provide images for ongoing research. At Trinity College Dublin, we thank Prof. Dan Bradley and his staff for supporting the aDNA study, which was funded separately from the ERC work, through the Wellcome Trust, and for the opportunity to add this important aspect of research to the study of Malta. We must also extend our thanks to all the participants, past and present who under- took the fieldwork at the Xagħra Brochtorff Circle in the 1980s–90s and to those who worked tirelessly to record the vast assemblage over the following years. For recent assistance we thank Chiara Marabelli who assisted in processing and cleaning human remains for study as part of the new bone team. For the preparation of this volume, we thank Dr Rob Barratt, Dr Rowan McLaughlin, Dr Eóin Parkinson, Dr Jess E. Thompson and Olivia Shelton for assisting in Acknowledgements xxvi Acknowledgements Deputy Director and Editor of the Monograph Series, for guiding the volume into its final stages. In particu- lar, we thank Ben Plumridge who undertook the task of production. Ben ‘cut his teeth’ in publication in 2007–9 when he was the principal assistant in the preparation of the 2009 Mortuary Customs in Prehistoric Malta (the original site report and the forerunner of the three companion volumes from FRAGSUS), making the work of the late Dora Kemp, his McDonald predecessor, as production editor so much easier. The archaeology of the island of Malta has thus continued to benefit from his support after his appointment as Production Editor of the McDonald series by Simon Stoddart when he was acting Deputy Director. To all those involved, we thank you. the editing and assembly of this very substantial work. Prof. Simon Stoddart was the overall editor who under- took the major editing of the volume (including the index and glossary), the design of the opening and clos- ing chapters and was in the grant application defined as responsible for the Human Population work group and the Landscape group in Cambridge. Prof. Caroline Malone undertook the preliminary assembly of the text, the checking and setting of tables and images, as well as much proofing and editing. We thank Emma Jarman in the McDonald Institute for her important role in coordinating the submission of material, and Dr James Barrett and Prof. Cyprian Broodbank (when James left Cambridge) for their oversight of the McDonald Monograph series. We thank Dr Matt Davies, the new xxvii This volume is the third in the FRAGSUS Project series. Volume 1: Temple Landscapes (edited by Charles French, Chris O. Hunt, Reuben Grima, Rowan McLaughlin, Simon Stoddart & Caroline Malone, 2020) focuses on the changing landscapes of early Malta, and pro- vides the background for the following two volumes. Volume 2: Temple Places (edited by Caroline Malone, Reuben Grima, Rowan McLaughlin, Eóin W. Parkin- son, Simon Stoddart & Nicholas Vella, 2020), reports on the archaeological studies of six sites through an examination of their chronological sequence, material culture and economic role in the Neolithic world of Malta. These discoveries set the scene against which Volume 3: Temple People (edited by Simon Stoddart, Ronika K. Power, Jess E. Thompson, Bernardette Mer- cieca-Spiteri, Rowan McLaughlin, Eóin W. Parkinson, Anthony Pace and Caroline Malone, 2022) are reas- sessed. This volume also has an additional role since it follows on more directly from the 2009 publication: Mortuary Customs in Prehistoric Malta (edited by Car- oline Malone, Simon Stoddart, Anthony Bonanno & David Trump, 2009). That volume revealed one of the largest prehistoric burial assemblages yet discovered in the Mediterranean, amounting to some 220,000 bones, with a rich assemblage of animal bone, figurative sculpture, symbolic artefacts and architectural remains. The new volume concentrates on the human remains, taking their evidence to a new level. In the light of better understanding of the changing environment and resources of a small island world, the early people of Malta emerge as a remarkable community telling an important tale of prehistoric resilience and survival. Preface Caroline Malone and Simon Stoddart 303 11.1. Introduction 11.1.1. The genome and ancient DNA Modern genetic studies and the use of the biological component offer considerable potential in the study of past individuals and populations. The genome refers to the full genetic component of an organism which is passed through generation from parents to offspring. The key components of this structure are molecules called nucleotide base which are codified in four different letters: A(adenine), T(thymine), G(guanine) and C(cytosine). These molecules are bonded together within the DNA (Deoxyribonucleic acid) structure. The way these bases are ordered in a genome is, with the exception of monozygotic twins, unique to each individual; for example, two unrelated individuals differ for 3 million nucleotides over approximately 3 billion that compose a human genome. The study of DNA can help to estimate how different people are related, past and present. Using modern biological techniques, DNA can be extracted intact from living persons without losing information. However, when dealing with samples that date far back into the past, dead cells cannot preserve the integrity of the genetic information, and therefore old/ancient DNA is difficult to reassemble. Nevertheless, the ancient DNA (aDNA) field has an important role in dealing and using this type of information to study the history and evolution of ancient organisms. In recent years, thanks to the advancement of a new generation of DNA sequencing techniques, aDNA studies have revolutionized most of the previous concepts about genetics and history and shed light on the origin of different species. 11.1.2. Ancient DNA The field of aDNA emerged in 1984 when Russ Higu- chi and colleagues (Higuchi et al. 1984) extracted a fragment of DNA from a dry tissue of a quagga, an historical relative of the horse family. Soon after, Pääbo (1985) reported the first aDNA extraction from an ancient mummy. aDNA became even more pro- lific, giving the opportunity to analyse a multitude of material, such as bones (Hagelberg et al. 1989), hair (Gilbert et al. 2004) and even parchment (Teasdale et al. 2015). Many hundreds of ancient genomes from differ- ent periods and parts of the world have been screened with high resolution, making it possible in particular, to shed light on human migrations (Mathieson et al. 2015) and on animal domestication (Daly et al. 2018; Zeder et al. 2006). Despite this recent progress, there are still some challenges that arise when dealing with ancient DNA samples. Due to spontaneous damage that occurs after death, the DNA in ancient samples is usually pres- ent in short fragments with a size range between 50 to 70 nucleotides. With smaller and more numerous fragments, it is more difficult to assemble the DNA molecule in its original form. Moreover, due to the lack of a repair system in dead cells, spontaneous mutations in nucleotide base pairs accumulate. A study published by Skoglund et al. (2014) showed that the amount of a particular type of DNA mutations, deamination, in a sample is proportional to its age. If not taken in consideration, these damages can lead to erroneous interpretation of DNA results during pop- ulation and evolutionary genetic analyses. In recent years, the deamination problem has been partially solved thanks to particular software that can target and quantify these specific patterns of postmortem damages (Jónsson et al. 2013). A third problem that emerges when working with ancient samples is the low quantity of endoge- nous DNA present. These values can be as low as 0.1% (Stoneking & Krause 2011), posing a problem from bacterial and human genome contamination. For this reason, it is important that the extraction of DNA from ancient samples is carried out in special cleanroom Chapter 11 aDNA: an investigation of uniparental genetic heritage in Neolithic Malta Bruno Ariano, Valeria Mattiangeli, Rowan McLaughlin, Ronika K. Power, Jay T. Stock, Bernardette Mercieca-Spiteri, Simon Stoddart, Caroline Malone & Dan Bradley 304 Chapter 11 some marked genetic distinctions from the WHG group (Haak et al. 2015). These individuals, who lived approximately 8000 years ago, are now considered part of a genetically distinct cluster identified as Eastern Hunter-Gatherer (EHG). This group can be considered a mix between WHG populations and Upper Palaeolithic individuals from Siberia (Mal’ta and Afontova Gora) (Raghavan et al. 2014; Fu et al. 2016). The influence of this group on other populations has been detected in hunter-gatherer individuals from Sweden and the Balkans (Gonzales Fortes et al. 2017; Lazaridis et al. 2014; Lazaridis 2018) and in populations from the steppe during the Bronze Age period (Haak et al. 2015). A third genetic cluster is formed by two indi- viduals found in western Georgia that are now identified as members of a Caucasus Hunter Gatherer (CHG) group. This population diverged from the WHG group long before the Last Glacial Maximum, approximately between 40 and 50 thousand years ago. It is a population that had a strong influence in both Mesolithic and Neolithic populations from Iran, and its influence is still present in the genomes of modern populations from Southern Caucasus (Jones et al. 2015). 11.1.4 The genetic impact of the agricultural revolution The adoption of agriculture was a turning point in human history which occurred in different parts of Eurasia and the Middle East between 12,000 and 7000 bc. In the Levant and Southern Anatolia between 11,000 and 9600 bc, local hunter-gatherer populations began to adopt a farming and sedentary lifestyle, accompanied by animal and plant domestication. With the help of aDNA studies it was discovered in 2016 that the origin of Near Eastern farming had two genetically distinct roots, one residing in Anatolia and the other in Iran (Broushaki et al. 2016). Between c. 6,600 and 6,500 bc Iranian farmers spread genetically towards eastern Eurasia whilst the Anatolian farming communities became well-established in north-west- ern Anatolia and had begun to move into Europe via Greece and the Balkans (Lazaridis 2018; Lazaridis et al. 2014). The arrival of farmers in Europe represented a genetic replacement with limited admixture from the local hunter-gatherer populations. This admix- ture became evident in 2009, when aDNA showed a genetic discontinuity between these two populations in Europe during the Neolithic period (Malstrom et al. 2009). More recent studies have emphasized this observation, giving a better view of the phenomenon. From the lower part of the Danube, the Anatolian farming culture reached the Hungarian plain by 5500 bc and gave birth to different farming groups facilities, where particular procedures are adopted to keep the bacterial and human contamination levels as low as possible (MacHugh et al. 2000). Once the DNA has been extracted, two common approaches are used for obtaining the sequence data, shotgun genome sequencing and targeted capture. The first method consists of fragmenting and sequencing the available genome of a sample. This technique has been extensively used for modern DNA analysis and can also be applied to ancient genomes as long as the samples are of good quality. The main advantage of this method is the opportunity to cover every position in a genome and study mutations that are still unknown or present in low frequency in a comparator population. The targeted capture method, on the other hand, usually focuses on a predefined set of high frequency variants (referred as SNPs) that are enriched using a custom-built probe. This technique has the advantage of obtaining more data compared with the WGS approach, especially when dealing with samples with low DNA quantity. With more than 1.2 million SNPs covered (Mathieson et al. 2015), this technique has become frequently used for ancient DNA anal- ysis. However, the main drawback of this approach resides in the limited number of analyses that can be performed using these variant positions. For example, the majority of rare mutations that are important for Mendelian diseases are not covered by the capture method and therefore cannot be directly studied. 11.1.3. Background: the genetic context of the Mesolithic in Europe The Mesolithic period dates from the end of the Epipalaeolithic period, around 12,000 years ago, and it was heralded by rapidly rising temperatures accompa- nied by the establishment of a Holocene forest biome across Europe. These conditions contrasted with the preceding tundra and glacial conditions (Clark et al. 2009). During the Mesolithic, human populations were scattered in groups around Europe, living in small groups, and following a typical hunter-gath- erer (HG) existence. Different published studies have investigated the genetic background of these pop- ulations, dividing them into three main groups. On the western side of Europe, individuals from Spain, Hungary and Luxembourg have been reported as genetically similar, and for this reason they have been identified as the Western Hunter-Gatherer (WHG) group. Also included in this group are individuals from eastern Europe that displayed a similar pattern of genetic affinity (González-Fortes et al. 2017; Jones et al. 2017). On the eastern side of Europe, two Mes- olithic individuals from Russia were found to have 305 aDNA: an investigation of uniparental genetic heritage in Neolithic Malta evidence (see Volume 1, Chapters 3 & 4), with clear archaeological traces present in the archipelago by 5500 bc (see Volume 2, Chapter 2). 11.1.5. Arrival in Malta The evidence supplied by archaeology, particularly the affinities between Għar Dalam and early Neolithic Impressed Wares of Southern Italy, strongly suggest that the source population of the Neolithic expansion into the Maltese Islands were located in Southern Italy and Sicily (see Volume 2, Chapter 10). Theories of an earlier colonization of Malta have been debated, but since hunter-gatherer populations require a large space for foraging, it seems unlikely that Malta would have been a viable long-term home before the advent of agriculture (Malone 1997–8). From the first evidence of human settlement, the early Maltese society evolved through different cultural phases: Għar Dalam, Grey Skorba, Red Skorba and finally Żebbuġ, signalling the start of the Temple Period and an increasingly distinc- tive island culture. In this last phase, the use of rock-cut tombs, containing collective burials and distinctive pottery defined the island culture (Malone et al. 1995). Subsequent cultural phases (the Temple Period) witnessed an unprecedented development in Maltese society, culminating in the Tarxien phase between 2800 and 2400 bc (Volume 2, Chapter 2). During the Tarxien phase, collective burial in the elaborate Circle cave (Starčevo, Körös and Criş). Some centuries later, from the same region, another cultural movement started to spread into north-west Europe with a new form of decorated pottery called the Linearbandkeramik (LBK) (Cunliffe 2015). A second culturally different wave of Neolithicization moved from the Adriatic Balkans through to the Mediterranean coast where it is associated with the pottery of the Impressed and Cardial traditions pottery style. The Impressed Ware culture was more closely associated with regions across Italy towards the Ligurian coast, whilst a var- iant of this pottery group, the Cardial Ware culture, arrived in Provence and extended towards the Atlantic and Portugal (Price 2000). It is important however, to point that these cultures were different, even though they were all close genetically to the same Anatolian Neolithic source (Olalde et al. 2015, Mathieson et al. 2018). The earliest Neolithic settlements in Italy, which date from about 6200 bc, are located along the lowland coastal areas of south-east Italy (the Apulian Salento peninsula and Tavoliere) (Malone 2003; Natali & Forgia 2018). Very high densities (c. one site per 3 km²) of ditched settlements across the area signal a major population increase (Whitehouse 2013). Adop- tion of the Neolithic economy then rapidly spread westward into Calabria (Morter & Robb 2010) and Sicily (Leighton 1999; Natali & Forgia 2018), reaching Malta by at least 5800 bc, based on environmental Figure 11.1. Reconstruction of the Circle (Malone et al. 2009d). 306 Chapter 11 fasta sequence. The contamination rate was calculated as the ratio of the number of mismatches over the total count of positions in the consensus sequence. When the mismatches included deaminated bases, these were counted as an upper limit value of contamina- tion. Fastq files were aligned to the human Revised Cambridge Reference Sequence, (rCRS, NC_012920.1) using the tool mpileup from the software samtools (Li et al. 2009). Only SNP calls with a base quality above 30 (parameter -Q30) were then retained for further analyses. The genome coverage of each sample was calculated using the tool qualimap (Okonechnikov et al. 2016). A consensus mitochondrial Fasta sequence was first obtained for each sample using bcftools software (Li et al. 2011) (parameter -c) and then given to the software Haplofind (Vianello et al. 2013) for the haplogroup assignment (Table 11.2). From this analysis, we considered as valid only the haplogroups that were at the most terminal part of a branch and had an assignment score of at least 0.9 and where the assignment did not derive from a transition SNP. 11.3.2. Contamination There are two common ways of checking for sample contamination in ancient DNA samples; the first method consists of checking for the presence of molec- ular damage at the 5’ and 3’ end of aligned reads. The second method is used also in modern DNA analyses and involves checking for the haploid state of the mitochondrial and X-chromosome DNA in male individuals. Given that all our samples were already treated for postmortem damages, we concentrated upon this last method for our contamination analyses. 11.3.3. Y-chromosome haplogroup determination Samples that were identified as male were evaluated for Y-chromosome haplogroup lineage. This task was executed using the software Yleaf v2 (Ralf et al. 2018) and the ISOGG (International Society of Genetic Gene- alogy) 2019 database as reference (https://isogg.org/ tree/ISOGG_YDNA_SNP_Index.html). SNPs annotated with the ‘~’ label were excluded from this analysis (Table 11.3). 11.3.4. Collection of publicly available data To contextualize our haplogroup results with other published ancient samples, we downloaded a well curated dataset of ancient DNA metadata from AmtDB (Ehler et al. 2018). We then used this resource to compare the geographical distribution of all sample haplogroups (both mitochondrial and Y-chromosome), focusing in particular on Neolithic, and Bronze Age periods. The samples were finally filtered for latitude and longitude thus restricting our analysis to Eurasia. complex on Gozo (Fig. 11.1) and at the Ħal Saflieni Hypogeum in Malta represent exceptional mortuary sites. The Circle excavations unearthed the individuals analysed for this study in the early 1990s (Malone et al. 2009d) and are the subject of additional study in this volume. The ancient DNA work we report here was undertaken in collaboration with the FRAGSUS Project (2013–2018) as part of a programme of envi- ronmental and archaeological research, including an extensive re-assessment of the Circle, applying additional radiocarbon dating and stable isotope studies. The overall aim of this research has been to understand better the cultural, economic and envi- ronmental dynamics of prehistoric Malta (Malone et al. 2019; Ariano et al. 2022). 11.2. Research questions Since ancient times the Mediterranean Sea has repre- sented one of the most important routes for migration in southern Europe. For example, during the late Neolithic period there is proof of both a cultural and a direct genetic connection between Portuguese and Greek Neolithic populations (Hofmanova et al. 2016). Despite this evidence, the prehistoric population his- tory of South Europe remains under-explored in terms of genetic studies. In contrast, most aDNA publications have focused on the history of Central and Northern European populations, with little attention paid to southern Europe. The reason for this absence is because of the particularly warm climate conditions that tend to accelerate the degradation process of aDNA samples. Importantly, the Maltese work we are reporting here is the genetic analysis of one of the most southerly archipelagos of the Mediterranean. Specifically, we obtained uniparental genetic data (mitochrondrial DNA and Y-chromosome haplotypes) from 3 ancient individuals that lived in Malta during the transition between the Neolithic and Bronze Age periods. Thanks to this data we addressed the question of whether the Maltese were genetically more similar to Neolithic or to Bronze Age populations in Eurasia. 11.3. Methods 11.3.1. aDNA data collection and mitochondrial analysis For this project we used data submitted by Ariano et al. (2022) from 3 petrous bones from the Circle. Reads obtained for each sample were aligned to the human reference genome (hg19/GRCh37). Both private and Haplogroup defining mutations were taken from the software Haplofind (Vianello et al. 2013) output. For each individual, these mutations were then used to measure the number of mismatches with the consensus 307 aDNA: an investigation of uniparental genetic heritage in Neolithic Malta 11.4. Results 11.4.1. Mitochondrial contamination and history A common method for estimating DNA contamination of a sample is to check the rate of heterozygous sites present in the mitochondrial DNA. The contamination percentages of our high coverage samples, not consid- ering sites that can derive from transition, range from values of 0.3% to 0.78% (Table 11.1). These values can be considered as acceptable for a no-contamination hypothesis. Once assured about the quality of our samples, we used the software Haplofind to investi- gate mitochondrial haplogroups, with the following results (Table 11.2): • MLT5 belongs to the haplogroup K1a which is a subgroup of the major branch K. This branch has already been described in individuals that come from Anatolia during the Pottery and pre-Pottery Neolithic period (Mathieson et al. 2015). • The individual MLT6 belongs to the hap- logroup V which, although low in frequency, has been found in populations from central Europe associated with LBK, Únětice and Pitted ware culture, and from Neolithic pop- ulations in Portugal (Haak et al. 2015). • MLT9 belongs to the haplogroup H4a1, which is a derived branch of haplogroup H. This major group evolved first in the Near East during the Neolithic period and afterward spread into western Europe (Torroni et al. 1998). It appears in fact to be frequent in France during Middle Neolithic period and Iberia during the Epi-Cardial Neolithic period. By inspecting the distribution of ancient haplogroups, it appears that the Maltese belonged to mitochondrial branches that were particularly widespread during the Neolithic period. Interestingly, samples that matched the Maltese haplogroups during the Bronze Age period (details in Fig. 11.2) tended to come from central Europe and the Iberian Peninsula and belonged to the Bell Beaker culture. 11.4.2. Y-chromosome contamination and lineages The results from Y chromosome screening indicate that two of our samples (MLT5 and MLT9) were male. We then used SNP information from the ISOGG database to define haplogroups and we found that the two individuals each belonged to one of two common European Neolithic haplogroup branches. MLT5 belongs to haplogroup H2. This haplogroup is rarely found in modern European populations and its earliest evidence dates back to a pre–pottery sample Table 11.1. Results from the contamination analysis. No sample shows significant traces of contamination, both excluding and including Transition sites (MD). Sample ID Mean coverage Site contamination % Site contamination no-MD % MLT5 128.26 1.422 0.533 MLT6 106.8 1.548 0.787 MLT9 184.87 0.563 0.340 Table 11.2. Haplogroup assignment from Haplofind. The assignment score gives a probability of a sequence to be part of an haplogroup. The Haploscore gives an assignment score taking into account the previous major haplogroup from the same branch. Sample ID Mitochondrial coverage Haplo- group Haplo- score Assignment score MLT5 128.26 K1a 0.8 0.96 MLT6 106.8 V 1 0.98 MLT9 184.87 H4a1 1 0.99 Table 11.3. Sex assignment for each sample. When a sample did not reach a sufficient confidence interval it is indicated as ‘Not Assigned’. For male individuals also the Haplogroup is assigned using the ISOGG database as reference. Sample ID Only ChrY Ratio ChrY/ChrY+ChrX SE 95% CI Sex assignment Haplogroup MLT5 208312 0.1162 0.0002 0.115-0.116 Male H2 MLT6 43469 0.0178 0.0001 0.017-0.018 Not assigned - MLT9 177879 0.1224 0.0003 0.121-0.122 Male G2a2a1a3 Table 11.4. Values associated with contamination level using the X chromosome in male individuals. Sample ID Contamination % SE P-value MLT5 0.6 0.0014 6.789e-11 MLT9 1.1 0.0017 1.128e-08 308 Chapter 11 in the Levant between 7300–6750 bc (Lazaridis et al. 2016). In more recent times this haplogroup was found in an Anatolian farmer and a European Neolithic sample belonging to the Starcevo culture. MLT9 has the haplogroup G2a2a1a3, one of the subclades of the major branch G commonly present in Europe during the Neolithic period (Broushaki et al. 2016). From examination of the incidence these haplogroups in ancient Eurasia, their prevalence during the Neolithic period compared with later times is clear (details in Fig. 11.3). There is a trend for matches to follow a more southern distribution. In the post-Neolithic comparison, only two H2 matches were found, in an Early Bronze Age sample from Bulgaria. Haplogroup G2a2a1a3 was interestingly found in 3 samples from Neolithic-Copper Age in Spain and Portugal. Other close subclades are common among Early European farmers and rarely feature in the Bronze period sample where they are mostly replaced by haplogroups R1a and R1b (Haak et al. 2015). Figure 11.2. Distribution of ancient mitochondrial haplogroups in Eurasia. Each point is a sample with the shape representing the haplogroup to which it belongs. A red colour indicates a match with one of the Maltese haplogroups encountered in this work, dark grey points show the geographical distribution of unmatched samples. Panel A: distribution of haplogroups during the Neolithic. Panel B: distribution of haplogroups in Bronze and Iron Age samples. 309 aDNA: an investigation of uniparental genetic heritage in Neolithic Malta 11.5. Discussion Mitochondrial DNA and Y-chromosome sequences from Neolithic Maltese individuals from the Temple Period (3rd millennium cal. bc) were analysed. Y chro- mosome haplogroup information showed that MLT5 and MLT9 are both part of Neolithic haplogroups common during the Neolithic period. Interestingly the MLT9 haplogroup was also found in samples from Copper Age Iberia pointing to a possible connection with the Cardial culture. These haplogroups almost disappeared during the Bronze Age and Iron Age periods with the only three matches found in Bronze Age individuals from Eastern Europe and Central Asia. Mitochondrial haplogroups results mirrored these fi ndings with samples that matched the Maltese mostly as Neolithic farmers and Bell Beaker samples from Western Europe. Both these results point to a Western European Neolithic or Bell Beaker ancestry of our ancient Maltese and we believe further analysis Figure 11.3. Distribution of ancient Y haplogroups in Eurasia. Each point is a sample with the shape representing haplogroup. A red symbol indicates a match with one of the Maltese haplogroups encountered in this work. Panel A: distribution of haplogroups during Neolithic. Panel B: distribution of haplogroups in Bronze Age samples. 310 Chapter 11 and spread northwest towards northern Europe. The other route was associated with Impressa-Cardial pot- tery culture and followed a westward Mediterranean route reaching the Atlantic in France and Iberia. Malta’s early settlers were likely part of this latter route with their uniparental markers resembling other southern European Neolithic samples most strongly. 11.6.2. The eastern influence By the 2nd millennium bc, the Bronze Age period populations from the steppe migrated from eastern to western Europe, displacing preceding local cultures (Olalde et al. 2018). Exotic pottery coming from eastern Europe, even before the Bronze Age period, could suggest a connection between the Maltese and other populations (for example, Thermi, Bell Beakers and the potential Balkan Cetina style). No genetic evidence in our samples implies contact with eastern populations. 11.7. Future perspectives The field of ancient DNA study is in continuous devel- opment, especially as the financial cost of sequencing analysis reduces. Although haploid lineage markers can give hints about ancestry, using autosomal mark- ers will help us to answer more important questions about migration and admixture. Therefore, our first next step will be to deepen our investigations by using methods to detect admixture, kinship and population structure from autosomal markers. of autosomal markers will clarify and refine estimates of their ancestry. 11.6. Conclusion The populations of the Maltese islands, located in the south of the Mediterranean Sea, were shaped by a succession of different cultures during the Neolithic period. The first group settled on the islands just after 6000 bc, probably as an Early Neolithic population. After an initial oscillation between growth and decline (see Volume 1, Chapter 2) an apogee of culture and population density was reached during Temple Period, especially in the Tarxien phase between c. 2800 and 2400 bc, which saw the construction of unparalleled sophisticated megalithic structures. Then this culture seemingly collapsed, and a number of questions have vexed scholars of early Malta ever sense: who were these ancient inhabits of Malta, and which ancient population did they resemble the most? To answer these questions, we offer here a first assessment of Maltese ancient DNA data using three individuals that lived during the Tarxien phase of the Temple Period. 11.6.1. The Neolithic routes The culture of Neolithic farming spread from north- west Anatolia into western Europe following two main routes. One route was associated with the Linearband- keramik culture (LBK) and followed the Danube valley 477 Introduction – Interim Knowledges Temple people The ERC-funded FRAGSUS Project (Fragility and sustainability in small island environments: adaptation, culture change and collapse in prehistory, 2013–18) led by Caroline Malone has focused on the unique Temple Culture of Neolithic Malta and its antecedents. This third volume builds on the achievements of Mortuary customs in prehistoric Malta, published by the McDonald Institute in 2009. It seeks to answer many questions posed, but left unanswered, of the more than 200,000 fragments of mainly commingled human remains from the Xagħra Brochtorff Circle on Gozo. The focus is on the interpretation of a substantial, representative subsample of the assemblage, exploring dentition, disease, diet and lifestyle, together with detailed understanding of chronology and the affinity of the ancient population associated with the ‘Temple Culture’ of prehistoric Malta. The first studies of genetic profiling of this population, as well as the results of intra-site GIS and visualization, taphonomy, health and mobility, offer important insights into this complex mortuary site and its ritual. Remarkable evidence on the bioanthropology of care practised by these populations, together with a relatively low level of interpersonal violence, and examples of longevity, reveal new aspects about the Neolithic Maltese. Detailed case studies employing computerized tomography describe disease such as =scurvy and explore dietary issues, whilst physical activity and body size have been assessed through biomechanical analysis, supported by taphonomic study, isotopic analyses, a review of mortuary practices during prehistory and a robust new chronology. The results form a rich contextualized body of material that advances understanding of cultural change within the context of small island insularity, and provides biological comparisons for the graphic figurative art of early Malta. These data and the original assemblage are conserved in the National Museum of Archaeology in Valletta as a resource for future study. ISBN 978-1-913344-08-5 Editors: Simon Stoddart is Professor of Prehistory in the Department of Archaeology, University of Cambridge. Ronika K. Power is Professor of Bioarchaeology in the Department of History and Archaeology at Macquarie University, and the Director of the Centre for Ancient Cultural Heritage and Environment (CACHE). Jess E. Thompson is a research associate on the ‘ANCESTORS’ project at the McDonald Institute, Cambridge. Bernardette Mercieca-Spiteri is the osteological officer of the Superintendence of Cultural Heritage of Malta. Rowan McLaughlin is a Pathway Fellow at the Hamilton Institute, National University of Ireland, Maynooth, where he is principal investigator of the IRC-funded project ‘A deep history of Ireland for the Information Age’. Eóin W. Parkinson is a Leverhulme Research Fellow at the Department of Classics and Archaeology of the University of Malta. Anthony Pace was Superintendent of Cultural Heritage on Malta during the course of the FRAGSUS Project. Caroline Malone is Professor in the School of Natural and Built Environment, Queen’s University Belfast. Published by the McDonald Institute for Archaeological Research, University of Cambridge, Downing Street, Cambridge, CB2 3ER, UK. Cover design by Dora Kemp and Ben Plumridge. ISBN: 978-1-913344-08-5 9 781913 344085