«THE GENETIC ARCHITECTURE OF THE LATE PREHISTORIC AND PROTOHISTORIC RAPANUI A DISSERTATION SUBMITTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF ...»
THE GENETIC ARCHITECTURE OF THE LATE PREHISTORIC
AND PROTOHISTORIC RAPANUI
A DISSERTATION SUBMITTED TO THE GRADUATE DIVISION OF THE
UNIVERSITY OF HAWAI‘I IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHYIN
ANTHROPOLOGYAPRIL 2008 By John Vede Dudgeon
Terry Hunt, Chairperson Rebecca Cann Michael Pietrusewsky Nina Etkin Terrence Lyttle Carl Lipo © John V. Dudgeon, 2008 iii Acknowledgments This dissertation has been a long time in the making. I am grateful to the those people in my life who have seen fit to support me in this endeavor and can now appreciate whatever success I achieve as a result. In some ways, the dissertation document represents the epitome of collaborative effort, influenced by the suggestions of advisors, mentors, editors, proofreaders and academic and personal consultants. There are those who help you conceptualize, perform and complete the task at hand (the thesis), those who’s teaching and mentoring lay the foundation for successful scholarship and those who by their close personal relationship inspire you to do greater things. I would like to thank each of these people in turn.
A great debt of gratitude is due my dissertation committee. Their thoughtful support, theoretical and methodological guidance and excellent editorial commentary have improved this document substantially. Terry Hunt gave me my first opportunity to do fieldwork in the Pacific and brought me to Rapa Nui for several field seasons as an assistant and instructor. He worked tirelessly to open doors in the local community and helped me develop the professionalism and diplomacy required when working in another culture’s archaeology. His enthusiasm and theoretical rigor in the most vexing problems for Pacific archaeologists challenged me to ask the important why questions, when all I was originally after was what and where.
Rebecca Cann opened her lab and taught me protocols and good lab practice, as well as teaching me to strive for anthropologically-sufficient genetic explanations in my research. My many philosophical conversations with her stressed the importance of synthetic approaches to the archaeological record, couching evolutionary descriptions in iv terms of both proximate and ultimate causes. Mike Pietrusewsky proved to be an excellent editor and advisor on physical anthropological methods and data analysis. His advice was both pragmatic and useful, and his insistence on my understanding and documenting the theoretical and historical significance of osteological variation substantially improves the clarity and meaning of genetic variation in this context.
Terry Lyttle provided insight and important caveats to the description and interpretation of population genetic data and helped me to acknowledge the limitations and potential of this type of research. In large measure, his critique will hone and direct future research and publication. A special note of thanks goes to Nina Etkin, who well and truly exceeded her role as committee member and graduate chair. Her professionalism, persistence and advocacy on my behalf were crucially important.
Carl Lipo, as committee member, mentor, colleague and friend, provided endless hours of stimulating conversation on method and theory, helped me develop the primary and secondary research questions, and engaged in highly-
thought experiments to explore the universe of possible conclusions even before I had generated any empirical data to evaluate.
Others who deserve recognition for their support of this thesis include: Hector Neff, who tutored me in the methods and theory of provenance by trace element geochemistry of archaeological materials; Zed Mason, who supported my personal dissertation research interests in the Institute for Integrated Research in Materials, Environments and Society (IIRMES) at Cal State, Long Beach; the Human Identification (HID) group at Applied Biosystems, for several hours of evaluation and interpretation of genotype data; and Joan Chen and Felis Wolven at the DNA Microchemical Core
genotyping experiments. On Rapa Nui, special thanks go to Francisco Torres, Director of the P. Sebastian Englert Museum, former governor Sergio Rapu and my field school student assistants, Virginia Cannon, Amber Harrison, Elena Kouneski and Megan Pickus.
This research was partially supported by funding from the National Science Foundation, BCS 0604712 and 0321361, and by University College Extension, CSULB.
Acknowledgments go to: Robert Dunnell, my M.A. thesis advisor, who taught me the value of simple, concise writing, the importance of logical, stepwise reasoning to archaeological problems, and persistence in the face of damning critique; Mark Madsen, friend and colleague, for his ability to elucidate complicated subject matter; and Julie Field, Ethan Cochrane and Jacce Mikulanec, colleagues and conspirators in the anthropology department, for their friendship and support. Thanks to Joe Genz, for helping me with paperwork, delivering drafts to the committee and being a good friend;
and Chris King, for practical and dispassionate advice on the sociological aspects of dissertation committees and his close friendship over many years.
Lastly, I would like to thank my family: my father, step-mother and brother, Matthew for their interest in and appreciation of the importance of this dissertation for me. My mother, who didn’t get a chance to see this to its completion, always supported my decisions and was forever my staunchest defender in the face of criticism. Most of all I would like to thank my wife Amy, for enduring the seemingly endless late nights in the lab and wild mood swings between rewrites, with a grace and sense of optimism that kept my ship righted and gave me safe harbor throughout my research and writing.
This thesis represents an initial effort to evaluate the genetic and chemical evidence for residential, spatial and interaction-based explanations for social and community structure within the late prehistoric (AD 1680-1722) and protohistoric (AD 1722-1868) inhabitants of Rapa Nui (Easter Island). The primary data for this thesis consists of samples of archaeological human teeth housed in the Museo Anthropológico Padre Sebastian Englert (MAPSE), which were recovered from interment locations adjacent to coastal settlement and monumental architectural features from twenty five site-defined areas around the island between 1979 and 1981.
Out of a total of 487 unique-to-individual dental elements that were selected from the museum-curated skeletal collection, 167 specimens were subjected to minimallydestructive minor and trace elemental analysis, to determine if the geochemistry of the location of recovery was correlated with the chemical fingerprint archived within the permanent enamel of each tooth crown. Additionally, a total of 88 unique-to-individual dental specimens were subjected to DNA extraction from the tooth root, and a study of the overall variation within the island as a group, as well as the patterns of gene flow between sites and site clusters was performed using short tandem repeat (STR, or microsatellite) sequences derived from genomic DNA.
Laser ablation inductively coupled plasma mass spectrometry of the subsurface dental enamel was used to test for the presence of minor and trace elements incorporated into the crystalline structure of the enamel during formation. The minor and trace elements in the enamel were compared to the elemental signature from soils and sediments that were pedogenically derived from the different volcanic flow events on the
assess the likelihood of residential stability. If the signature of the trace elements incorporated in the dental enamel during permanent tooth formation in early life is similar to the trace element compositional variation in the sediments of the location of recovery of the skeleton, and the skeletons can be chemically grouped by their location of recovery, then the hypothesis of residential stability is supported.
Mahalanobis posterior classification is used to determine the goodness of fit of the chemical groups with respect to the spatial distribution of skeletons over varying geochemical regions. These results show that while it is not possible to assign individuals to particular geochemical regions because of biopurification of many minor and trace elements, posterior classification is greater than 80% successful in grouping individuals by region of recovery, indicating probable long-term residential stability of individuals within each geochemical region. These samples are the unit of subsequent genetic analysis of drift and gene flow between archaeological sites and geographic regions.
Genetic analysis of the recovered DNA (ancient DNA) was performed using a reagent kit optimized for old, degraded or damaged DNA. Strict contamination prevention procedures were employed during extraction and PCR to avoid sampling modern DNA contaminants, and short tandem repeat (STR) microsatellite loci were chosen to access the biparental contribution data from each specimen, to increase the source of discriminatory variation over maternally and paternally-inherited genetic data (mtDNA and Y-chromosome DNA).
The genetic data are evaluated using a nested set of hypotheses constructed around the tribal or lineage-based descent systems described in the protohistoric
osteological studies using craniometrics and discrete traits of the skeleton to identify intra-island population affinities. Inbreeding coefficients and analysis of molecular variance (AMOVA) are used to describe the direction and frequency of gene flow and isolation between the ethnographic and craniometric groups, and phylogenetic trees visually demonstrate possible patterns of relatedness. Additionally, the genetic data are used to test some ideas about the size and timing of island colonization, by estimating effective population size in the current sample and determining whether the genetic architecture of the sample is representative of recent genetic bottleneck, or whether recent demographic expansion better explains the observed distribution of genetic variation.
The data and conclusions developed from the chemical and genetic analysis of the enamel and bone from the Rapa Nui skeletal collection are intended to augment current archaeological, physical anthropological and ethnohistoric evidence and explanations.
While much evidence has been advanced to explain the position of the Rapa Nui people in greater Polynesian context, relatively less work has focused on the evidence for intraisland biological evolution. Despite manifest issues with temporal assignment of the Rapa Nui skeletal series, I accept the premise made by many that the skeletal collection is representative of late prehistoric or early protohistoric population. Based on this assumption, I offer some evidence and test explanations for the observed distribution of genetic variation in the skeletal series, and offer a research program for future study in the curated skeletal collection to significantly expand the current understanding of the nature of the archaeological landscape on Rapa Nui.
List of Figures
List of Tables
Chapter 1. Introduction
Islands as laboratories
Isolation and interaction
Rapa Nui prehistory
The Rapa Nui skeletal collection
Why Rapa Nui?
Chapter 2. Research Design
Introduction to the study questions
Brief history of space
Brief history of time
Study of dental trace elements
Study of ancient DNA from teeth
Study hypotheses developed for ancient DNA from teeth
Hypothesis set one: no spatial genetic variation
Hypothesis set two: low-level or idiosyncratic spatial genetic variation.................. 41 Hypothesis set three: extensive, regionally-patterned spatial genetic variation....... 42 Hypothesis set four: anthropologically explainable, differential and regionallypatterned spatial genetic variation
Chemical Analysis of Dental Enamel
Dental enamel in the archaeological record
Suitability of dental enamel for studying chemical variation
Instrumental sampling methods
Genetic Analysis of Rapanui Skeletal Material
x Genetic structure in the skeletal series
Rapa Nui population genetics
Challenges in using microsatellites for old (ancient) DNA
Effect of error rates on genotyping
Chapter 3. Chemical Analysis of Rapa Nui Dentition
Sample selection methodology and rationale
Comparative analysis of trace elements in Rapa Nui sediments and soils
Regional sediments and soils
Sediments and soils as glass beads
Volcanic flow morphology on Rapa Nui
Previous geochemical research
Results of the sediment trace element analysis
Results of the tooth enamel trace element analysis
Mahalanobis distance calculation and posterior classification
Interpretation of dental enamel results
Chapter 4. Genetic Analysis of Rapa Nui from Archaeological Teeth.
......117 Analysis parameters
Selection of dental element
Recording, sectioning and abrasive cleaning
Minimization of downstream contamination
Grinding root section to bone powder
Decalcification and collagen conversion
Selection of multiplex kit