Childhood Stress of an Iron Age Population from Taiwan: Using Linear Enamel Hypoplasia and Porotic Hyperostosis as Stress Indicators

Material Information

Childhood Stress of an Iron Age Population from Taiwan: Using Linear Enamel Hypoplasia and Porotic Hyperostosis as Stress Indicators
LIU, CHIN-HSIN ( Author, Primary )
Copyright Date:


Subjects / Keywords:
Age groups ( jstor )
Childhood ( jstor )
Dental enamel hypoplasia ( jstor )
Dentition ( jstor )
Hyperostosis ( jstor )
Lesions ( jstor )
Physical anthropology ( jstor )
Teeth ( jstor )
Tooth enamel ( jstor )
Weaning ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Chin-Hsin Liu. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
Resource Identifier:
436098788 ( OCLC )


This item is only available as the following downloads:

Full Text




Copyright 2005 by Chin-Hsin Liu


For my parents, Mr. Ken-Chih Liu and Mrs. Hsiu-Chen Lin


iv ACKNOWLEDGMENTS As this thesis comes to completion, a numb er of people deserve special mention for their assistance and effort in helping me accomplish this task. First and foremost, I would like to express my most sincere gratitude to my mentor and committee chair, Dr. John Krigbaum, for his inspiration, support and guidance. His welcoming emails brought me into this program, and his tireless scholarship ha s led me to this stage. I would also like to thank Dr. Michael Warren for serving as my committee member and for his careful and constructive comments. E-mail corresponde nce with Drs. Michael Pietrusewsky (University of Hawaii at Ma noa), Don Reid (University of Newcastle), Kate Domett (James Cook University) and Marc Oxenham (Australian National University) has been fruitful and I thank their genero sity and invaluable advice. I thank Dr. Cheng-Hwa Tsang and Professor I-Chang Liu, director and co-director of the Shih-S an-Hang (SSH) salvage project, fo r allowing me access to the human remains and original field records in their care. I am also grateful to Professor Michael Pietrusewsky--his work on the SSH collec tion provides important insight towards bioarchaeological study in Taiw an and has contributed a great deal to this research. Bioanthropological research and curation of the SSH remains by Professor Pietrusewsky, Professor Tsang and former researchers/assi stants of the SSH project, Ms. Ching-Fang Chang, Hsuman Lin and Chin-Hwei Thu, are acknowledged. Professor I-Chang Liu, chair of the Ar chaeology Division, was most supportive during my stay as a junior vi siting scholar in the Ts’ai Y an-Pei Research Center for


v Humanities and Social Sciences (Academia Sinica). I could not have successfully conducted this research without his consideration and genero sity. I am indebted to Ms. Bing-Li Lee, Mr. Chih-M ing Chang, Wen-Sheng Li ao, and four part-time staffers in the Archaeology Division, for their assistance thr oughout my visit. My gratitude goes to Ms. Li-Chi Chiang, Ya-Ling Chien, and Shei-Hwa Chang for their assistance with administrative procedures. My appreciati on goes to Mr. Yong-Bao Young, photographer at the Institute of History and Philology (Academia Sinica), for his time and patience during photographic documentation. Faculty and friends from the Depart ment of Anthropology, National Taiwan University, provided most sincere blessing a nd friendship during my stay in Taipei, and made my data-collection trip enjoyable. Among them, Professors Chow-M ei Lien, Shih-Chung Hsieh, and Maa-Ling Ch en are particularly thanke d for their encouragement and advice. Special thanks go to Dr. Pochan Chen for patiently responding to my endless queries and demands for references. I am tha nkful to Mr. Te-Ren Lee and Mrs. Feng-Ping Young for kindly offering me their cozy home during my stay. In addition to people affiliated with anthropology, dentists and family members Yu-ren Liu, D.D.S., and Yu-Hsing Liu, D.D.S., are acknowledged for sharing their useful knowledge regarding dental anatomy and clinical observation. At the University of Florida, I am grat eful to all my friends in the Anthropology Department for making my life interesting. Among them, I thank Erin Ehmke, Laurel Freas, Joe Hefner, Laurie Kauffman, Bryan Tucker, Anna Vick, and Heather Walsh-Haney for their critical eye, statistic al assistance, and oste ological knowledge. Special thanks go to Mauric io Hernandez for his incred ible efforts proof-reading, on


vi short notice, a draft of this thesis. I would also like to thank my dear roommates Shu-Yu Lin, Wan-Ping Chao and Shun-Pei Miao for ch eering me up and sharing all the ups and downs during this difficult time. I thank my aunt, uncle, and their families in the U.S. for their encouragement, and for making their homes my home away from home. Finally, I could not have accomplished any of this without my parent s’ whole-hearted suppor t. Their frequent phone calls are simply the best. This thesis is dedicated to them, Mr. Ken-Chih Liu and Mrs. Hsiu-Chen Lin, the most important peopl e in my life, for their faith, love, and support. Also, I wish my beloved Grandmother could have seen me through. Her wisdom and love will always be in my heart.


vii TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES...........................................................................................................xi ABSTRACT.....................................................................................................................xi ii 1 INTRODUCTION........................................................................................................1 Research Goal...............................................................................................................2 Site Studied................................................................................................................... 2 Stress Indicators............................................................................................................3 2 THE SHIH-SAN-HANG SITE: BA CKGROUND AND BIOARCHAEOLOGY......5 Discovery and Excavation............................................................................................5 Archaeological Context................................................................................................7 Cultural Context............................................................................................................8 Subsistence...................................................................................................................9 The Skeletons..............................................................................................................10 Bioarchaeological Studies of the SSH Site.................................................................12 3 MATERIALS AND METHODS...............................................................................17 Materials.....................................................................................................................1 7 Methods......................................................................................................................17 Inventory..............................................................................................................17 Sex Determination...............................................................................................18 Age Determination..............................................................................................20 Linear Enamel Hypoplasia..................................................................................21 Methods for LEH Observation............................................................................25 Porotic Hyperostosis............................................................................................28 Methods for Porotic Hyperostosis Observation..................................................33 Dental Pathology.................................................................................................33 Statistical Analysis..............................................................................................34


viii 4 RESULTS...................................................................................................................35 Linear Enamel Hypoplasias........................................................................................35 Deciduous Dentition............................................................................................35 Permanent Dentition............................................................................................36 LEH by tooth count......................................................................................36 LEH by individual count..............................................................................46 Timing of LEH formation............................................................................55 Overall assessment.......................................................................................55 Sex comparisons of LEH formation time.....................................................62 Porotic Hyperostosis...................................................................................................68 Comobidity of LEH and Porotic Hyperostosis...........................................................78 Dental Pathology........................................................................................................79 Dental Caries.......................................................................................................79 Dental Calculus...................................................................................................80 Dental Abscessing...............................................................................................80 5 DISCUSSION.............................................................................................................86 Dental Pathology and SSH Diet.................................................................................86 Enamel Defects and Stress..........................................................................................89 LEH Distribution among Tooth Types................................................................89 Prenatal Stress.....................................................................................................90 Childhood Stress..................................................................................................91 Overall assessment.......................................................................................91 LEH formation and weaning........................................................................92 Peak timing of LEH formation.....................................................................95 Sex Differences and Cultural Practices...............................................................97 Prevalence and number of LEH episodes....................................................97 LEH formation timing..................................................................................98 LEH and Mortality............................................................................................100 Porotic Hyperostosis.................................................................................................102 Morbidity of the SSH People....................................................................................105 6 CONCLUSION.........................................................................................................107 LIST OF REFERENCES.................................................................................................109 BIOGRAPHICAL SKETCH...........................................................................................118


ix LIST OF TABLES Table page 2-1. Paleopathology of the SSH site reported in Chang..................................................13 2-2. Paleopathology of the SSH site re ported in Pietrusewsky and Tsang......................14 3-1. Sex and age distribution of SSH skeletal remains studied.......................................17 3-2. Adjusted molar wear and age relationships after Brothwell....................................21 3-3. Assigned premolar, canine and incisor wear stages and age relationship after Smith........................................................................................................................21 3-4. SSH mean crown heights and regres sion formulae modified from Goodman et al............................................................................................................................. ..26 3-5. Chronology of enamel development modified from Reid and Dean.......................30 4-1. LEH prevalence on deciduous dentition by tooth count..........................................37 4-2. LEH prevalence on permanent dentition by tooth count..........................................38 4-3. Overall prevalence of LEH by dentition. Ages and sexes pooled............................39 4-4. Mean number of LEH on observed and affected teeth by tooth types.....................40 4-5. Overall LEH distribution by sex..............................................................................42 4-6. LEH distribution on anterior dentition by sex..........................................................42 4-7. LEH distribution on poste rior dentition by sex........................................................42 4-8. Overall LEH distribution by age groups..................................................................43 4-9. Overall LEH distribution by younger (subadult and young adult) and older (middle adult and old adult) age groups...................................................................43 4-10. Overall LEH distribution by maturity (subadult and adult).....................................44 4-11. Sex comparison of mean LEH count on observed and affected teeth by tooth types.........................................................................................................................4 5


x 4-12. Mean LEH count by age groups...............................................................................47 4-13. Comparison of mean LEH c ount on subadults and adults.......................................52 4-14. LEH distribution by individual count.......................................................................52 4-15. LEH distribution by sex and age groups by number of individuals.........................54 4-16. Developmental zones of LEH format ion across tooth type s using modified method from Goodman et al., by number of individuals.........................................57 4-17. Distribution of porotic hyperost osis according to sex and location.........................76 4-18. Distribution of orbital lesi ons by sex and healing stages.........................................77 4-19. Distribution of porotic hyperostosis according to age groups and location by number of individuals...............................................................................................77 4-20. Distribution of porotic hyperostosis according to maturity and location by number of individuals...............................................................................................78 4-21. Distribution of orbital lesions by matu rity and healing stage by number of individuals................................................................................................................78 4-22. Correlation between the occurrence of LEH and porotic hyperostosis by number of individuals............................................................................................................79 4-23. Prevalence of dental caries on deciduous dentition by tooth count.........................82 4-24. Prevalence of dental caries on permanent dentition by tooth count.........................83 4-25. Prevalence of dental calculus on permanent dentition by tooth count.....................84 4-26. Prevalence of dental abscessing on permanent dentition by tooth count.................85


xi LIST OF FIGURES Figure page 2-1. Location of the Shih-san-hang site.............................................................................6 3-1. A general model for the study of stre ss in skeletal populat ions. From Goodman and Armelagos..........................................................................................................23 4-1. Percentage of observed permanent t eeth affected with LEH by tooth types...........39 4-2. Mean number of LEH on observed teeth by tooth types..........................................40 4-3. Mean number of LEH on affected teeth by tooth types...........................................41 4-4. Sex comparison of mean LEH count on observed teeth by tooth types...................45 4-5. Sex comparison of mean LEH count on affected teeth by tooth types....................46 4-6. Age comparison of mean LEH coun t on observed teeth by tooth types..................50 4-7. Age comparison of mean LEH coun t on affected teeth by tooth types....................51 4-8. LEH distribution by individual count.......................................................................53 4-9. Distribution of LEH episode s according to tooth type.............................................53 4-10. Sex comparison of LEH distribut ion among tooth types by number of individuals................................................................................................................54 4-11. Age comparison of LEH distribut ion among tooth types by number of individuals................................................................................................................55 4-12. Developmental zones of LEH format ion among tooth types using modified method of Goodman et al.........................................................................................58 4-13. Age patterns by number of individua ls of LEH formation among maxillary anterior tooth types using modified charts from Reid and Dean..............................63 4-14. Age patterns by number of individua ls of LEH formation among mandibular anterior tooth types using modified charts from Reid and Dean..............................63


xii 4-15. Developmental zones of LEH formati on of each tooth type using modified charts from Reid and Dean. Count ed by number of individuals..............................64 4-16. Age patterns of LEH on male anteri or dentition using modified method of Goodman et al..........................................................................................................66 4-17. Age patterns of LEH on female anteri or dentition using modified method of Goodman et al..........................................................................................................67 4-18. Sex comparison on age patterns of LEH by tooth types using modified method of Goodman et al......................................................................................................69 4-19. Age patterns of LEH on male anterior dentition using modified charts of Reid and Dean...................................................................................................................72 4-20. Age patterns of LEH on female anterior dentition using modified charts of Reid and Dean...................................................................................................................73 4-21. Sex comparison on age patterns of LEH by tooth types using modified charts of Reid and Dean..........................................................................................................74


xiii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts CHILDHOOD STRESS OF AN IRON AGE POPULATION FROM TAIWAN: USING LINEAR ENAMEL HYPOPLASIA AND POROTIC HYPEROSTOSIS AS STRESS INDICATORS By Chin-Hsin Liu August 2005 Chair: John S. Krigbaum Major Department: Anthropology The growth and development trajectory of an individual is a direct reflection of his/her stress load. At the population level, pa tterns of stress load have been used to assess the overall well-being of a group. Childhood stress, reflected by morbidity and mortality, is particularly usef ul in understanding how well a population is adapted to their environment. Interaction between environmen tal perturbations (various stressors) and host response may result in a number of skelet al defects which can se rve as indicators of stress events during the course of an i ndividual’s life. In this study, linear enamel hypoplasia (LEH) and porotic hypero stosis are selected as stre ss indicators used to assess the health status of a skeletal population fr om Iron Age Taiwan. The distribution of stress markers by age and sex is here interpreted in biocultural context. Potential insults that induce stress are addressed with respect to how the population interacts with its environment and its overall biocultural milieu.


xiv The Shih-San-Hang (SSH) site dates to 2,000 to 500 years B.P. The site is characteristic for its locally innovative iron-smelting tradition and iron craftsmanship. The skeletons of 306 recovered individuals fr om SSH are analyzed in this study. Basic demographic information and scored pathol ogies are used to assess prevalence and timing of LEH and porotic hyperostosis by sex and age. Results suggest that the SSH people had a fairly stressful childhood based on incidence of LEH. The peak age of LEH fo rmation occurs between 2 to 5 years. Weaning-related stressors, such as co ntaminated food and water, prolonged breast-feeding, and low quality weaning di et, are all possible causes for the LEH observed. Females are more stressed than males, as indicated by signi ficantly higher LEH prevalence and greater average LEH counts, a nd age of peak LEH formation occurs later in females than males. These two observations may reflect cultural conceptions and bias by sex suggesting male preference and/or di fferential weaning practice. LEH is more common among subadults and young adults whic h is indicative of higher morbidity due to an impaired immune system during early childhood. Prevalence of porotic hyperost osis, an indicator of irondeficiency anemia, suggests greater impact of the condition among i ndividuals in late subadulthood and early adulthood. Both sexes are similarly at risk in suffering from anemic stress. Parasitic infection due to marine-orien ted subsistence and/or poor hyg iene are possible etiologies. The occurrence of LEH and porotic hyperostos is does not overlap by age, which suggests that the two markers are independent of one another. Overall, the SSH inhabitants had stressful childhoods, however, once individuals reached adulthood their health status improved.


1 CHAPTER 1 INTRODUCTION The growth and development trajectory of an individual is a direct reflection of his/her stress load. At the populat ion level, stress load patter ns have been used to assess the overall well-being of a group. Stress in this study is defined as “the physiological disruption of an organism resulting from e nvironmental perturbation” (Huss-Ashmore et al., 1982: 396). Thus, severity of a skeletal insult or defect is a result of the interaction between relative rates of environmental stressors a nd host response (Goodman et al., 1988). Inferring health and disease in preh istoric populations can be determined by analysis of skeletal defects, which in turn helps to reconstruct quality of past living conditions. Human skeletal re mains recovered from archaeol ogical sites may preserve evidence of stress markers that are invaluable to assess the adaptive milieu in biological and biocultural context (Skinner and Goodman, 1992). Childhood morbidity and mortality are two good proxies of a population’ s adaptiveness to its physical and cultural environment (Alesan et al., 1999; G oodman and Armelagos, 1989). High childhood morbidity and mortality not only signify poor living conditions, incl uding nutritional and physical stress, but also sugge st the lack of an effectiv e cultural buffering system in coping with the harsh condition of ch ildhood (Goodman and Armelagos, 1989). Many skeletal markers are recognized as indicators of specific stress, such as porotic hyperostosis and treponematosis, and non-specific stress, including enamel hypoplasia and Harris lines. Prev alence of these indicators in a population can lead to better understanding their well-being. However, caution should be used when interpreting


2 skeletal lesions since demographic nons tationary, selective mortality, and hidden heterogeneity in risk (differential frailty ) may have effects on the distribution of pathologies (Wood et al., 1992). Research Goal The goal of this study is to assess stre ss patterns of early childhood in Iron Age Taiwan. A biocultural approach followi ng Goodman and colleagues (Goodman and Armelagos, 1989; Goodman et al., 1988) is adop ted to facilitate the interpretation of observed results in broader cultural context. The distribution of stre ss patterns by age and sex permits salient issues to be addresse d. For example, what are the effects of environmental stressors on skeletal remain s, how a population in teracts with their immediate surroundings, and how well people adapt. Furthermore, the relationship between different pathologies and its implications is also interpreted. Site Studied Human skeletal remains rec overed from Shih-san-hang, an Iron Age site in Taiwan, are utilized in this study. Sh ih-san-hang (hereafter, SSH) is a coastal site that was occupied between 2000-500 years B.P. The site is most noted for its massive amount of iron slag remains, iron smelting technology and ir on artifacts. It is one of the earliest Iron Age sites known and is recognized as the Ir on Age “type site” in Taiwan. The quantity and quality of human skeletal remains recovered from the site make it an appropriate assemblage for conducting population-based st ress-related research in prehistoric Taiwan. Further, the SSH sample permits assessment of how well a population with advanced metallurgy adapted to its environmen t, and what cultural behaviors may have contributed to environmental and physiological stressors.


3 Stress Indicators Two stress indicators, linear enamel hypoplasia (hereafter, LEH) and porotic hyperostosis are chosen for th e purpose of this study. Th ese two pathologies are well established in bioarchaeological studies of growth and development, and general health assessments of prehistoric populations (Blo m et al., in press; Domett, 2001; Douglas, 1996; Keenleyside, 1994; Larsen, 1997; 2001; Powell, 1988; Slaus, 2000; Stodder, 1997; Wright, 1994). LEH is a result of a disturba nce during the deposition of enamel matrix whereby normal ameloblast activity is affected and reduced thickness of enamel results (Goodman and Rose, 1990). There are several ad vantages in using LEH to assess stress patterns in an archaeological popu lation. First, teeth are the most resistant elements in the human skeleton; therefore they have high preservation potential after burial. Second, unlike bone, once enamel has been deposit ed, remodeling does not occur. The developmental defects are preserved unless er ased by dental attrit ion. Teeth therefore provide a hard record of all developmen tal insults occurring during infancy and childhood, during which time enamel depositi on and mineralization occurs (Goodman and Rose, 1990). This characteristic has made teeth invaluable when assessing health during childhood, since immature bones are less li kely to preserve due to differential preservation (Guy et al., 1997). Lastly, the sche dule of enamel formation follows a clear trajectory, with slight vari ation among populations. This schedule is essential in delineating the timing of LEH events and a llowing population patterns to be revealed (e.g., Goodman et al., 1980; Reid and Dean, 2000). Porotic hyperostosis is a more specific stress indicator suggestive of anemia. Among the various types of anemia, nutritiona l anemia, particularly iron-deficiency anemia induced in part by parasites and/or disease, is most common. Porotic hyperostosis


4 is manifest as a porous or “sieve-like” lesi on often appearing on the orbital roof and/or cranial vault. After the anemic stress is al leviated or eased, the lesion becomes remodeled and may or may not leave its scar on the affected area. While completely remodeled lesions could bias the prevalence of anemic stress, active and trace lesions are indicative of stress load at the time of death (Hill and Armelagos, 1990; Holland and O’Brien, 1997; Mensforth et al., 1978; Stua rt-Macadam, 1987; 1992). This study attempts large-scale bioarch aeological research using Iron Age human skeletal remains from Taiwan. Results a nd their interpretation build upon previous studies (Chang, 1993; Pietrusewsky and Ts ang, 2003), and offer new perspectives regarding the health and well-be ing of the individuals recovered from SSH. This thesis also provides new data to facilitate future research when other comparable data from large skeletal populations become available.


5 CHAPTER 2 THE SHIH-SAN-HANG SITE: BACKGROUND AND BIOARCHAEOLOGY Discovery and Excavation The SSH site is located on the northwestern coast of Taiwan near the southern bank of the Damsui River estuary (25 10 N; 121 24 E) (Figure 2-1). S ituated on a sand dune, the elevation is ca. 5 m above sea level. Its present admini strative location falls in the Ding-Ku Village, Ba-Li Township, Taipei Count y. It was first discovered accidentally in 1955 by a pilot who flew over the site and detect ed sporadic readings from his compass. With the speculation of large amounts of iron ore existing und erground, a geologist surveyed the site and determined it was preh istoric iron slag that created the magnetic effect. The first archaeological excavation was led by the late professor Chang-ju Shih in 1959. His team confirmed that the iron slag was the byproduct of preh istoric metallurgic activities by the SSH inhabitants. Due to poor communication between govern ment and academic sectors and the lack of cultural heritage mana gement at the time of the recovery, a large sewage plant was scheduled for construction at the SSH location. After negotia tion, five intensive salvage excavations took place between 199092 under the direction of Dr. Cheng-hwa Tsang and Professor I-chang Liu. The excavat ed artifacts, faunal remains and human skeletal remains are curated in the Acad emia Sinica and Shih-san-hang Museum of Archaeology. The latter is a public museum built at the original site location for the purpose of recognizing the importance of cultu ral heritage, and as a memorial to the conflict between economic development and preservation Tsang and Liu, 2001b).


6 Figure 2-1. Location of th e Shih-san-hang site. N


7 Archaeological Context The total area of the SSH site was es timated to be 60,000 m2, but only 7,000 m2 were excavated. Forty-three radiocarbon dates re veal that the site was occupied between 2,000-500 years B.P., with its heyday ca. 1,800-800 years B.P. (Liu, 1995; Tsang and Liu, 2001b). Five major periods of Taiwan prehistory are generally accepted. They include Late Paleolithic to Early Neol ithic (c.a. >15,000 B.P.-5,000 B.P.), Middle Neolithic (c.a. 4,500 B.P.-3,500 B.P.), Late Neolithic (c.a. 3,500 B.P.-2,000 B.P.), and Metal/Iron Age (c.a. 2,000 B.P.-400 B.P.) (Liu, 1992; see Chen, 2000 for a detailed review). The SSH site was occupied during this last period of Taiwan’s prehistory. During the Metal/Iron Age, five geographica lly distinct cultural traditions have been identified: the SSH Culture (northern and north-coastal), the Fan-zih-yuan Culture (midwestern), the Niao-song Culture (s outhwestern), the Guei-shan Culture (southernmost), and the Jing-pu Culture (easte rn and east-coastal). Radiocarbon dates and artifact typology of the latte r four cultures suggest thei r time range being 1,700-400 years B.P., 1,700-1,250 years B.P., 1,550-1,470 years B.P., and 1,000-500 years B.P. (Chen, 2000; Liu, 1992; Tsang, 2000). All dates are well-correlated w ith that of the SSH site. Due to the enormity of the SSH site, it has b een recognized as the “type site” of the SSH Culture, and marks the beginning of Taiwan’s Metal/Iron Age. Generally speaking, the terminal prehistori c period (occurring from the Metal/Iron Age to the contact of the Han people) of nor thern and coastal Taiwan is represented by the SSH Culture. The SSH Culture is divide d into Early and Late periods. The Early period (2,000-1,000 years B.P.) further contains three subtypes: SSH Type, Hou-long-di Type, and Fan-she-hou Type; the Late period (1,000 years B.P. to the acculturation of Han) includes Pi-dao-ciao Type, Sin-gang Type, Jiou-she Type, and Pu-luo-wan Type


8 (Liu, 1995). The seemingly confusing and deta iled classification of protohistoric and prehistoric sites in Taiwan signifies the complexity of prehistoric culture and extent of interaction between discrete populat ion groups inhabiting the Island. Cultural Context The Metal Age is largely characterized by the paucity of stone tools recovered, and this is especially the case with the SSH Cu lture. At SSH site, for example, some stone hoes, hammers and other non-edge d artifacts were found. Edged stone tools such as adzes and unifacial tools were very rare. Iron slag (by product of iron-smelting) and its associated minerals are abunda nt at SSH. Many areas of the site were covered by a layer of slag that varied in thickne ss across different zones, but av eraged ca. 5 cm (Tsang et al., 1990). The recovery of an open and circular structure lined with sandstone boulders is interpreted as an iron smelting furnace (Tsa ng and Liu, 2001b). While the function of this structure needs to be further elucidated, th e presence of iron slag indicates proficient knowledge and skill in iron metallurgy. Trace element analysis of iron products from the site demonstrates that SSH iron artifacts were manufactured with raw materials contai ning magnetite. Chen (2000) suggests that the locally available black b each sand was the raw material used in iron-smelting. The process of extracting iron from this source is particularly more complex than when using other other raw ma terials, and may require additional smelting structures. With the exception of Korea, no site s in East and Southeast Asia have yielded iron remains similar in chemical composition to that documented at the SSH site. This iron-smelting tradition was practiced as early as 1,500 years B.P. in the site. Chen (2000) proposed this tradition to be independently innovated by the SSH people and unique in the Asian region.


9 Reddish-brown decorated pottery, incl uding cooking and storage vessels, constitutes 90 % of the SSH pottery tradition. It was tempered with fine sand and fired by high temperature. Some animal figurines made with clay were also found. Moreover, beads, glass bracelets, earrings , and other colored beads were recovered, especially when associated with burials. Iron-made knives, ar rowheads, axes, hoes and nails, as well as bronze-made artifacts such as vessels, r ounded-shaped containers, and dagger/knife handles were collected. Extraordinary remain s, including bronze bells and silver and gold ornaments were also excavated from the site (Chen, 2000; Tsang and Liu, 2001a, 2001b). The Chinese porcelains, both intact and fr agmentary, and bronze coins were also recovered. The decorative motifs and inscripti ons on these important artifacts indicate association with the North-South Dynast y (420-577 AD) through to the mid-ChÂ’ing Dynasty (1,644-1,850 AD) in Chinese histor y (Tsang and Liu, 2001b). The coins were not used in monetary exchange, but rather as decoration and burial goods. Interestingly, evidence of frequent canoeing and/or paddling, such as the presence of costal-clavicular grooves on adult skeletons and the bones of non-coastal fish species , suggest that the SSH people likely engaged in sea-faring activ ities (Chang, 1993). Therefore, researchers have suggested that the SSH people were indeed sea-faring and may have had well-formed exchange networks not only with other aboriginal groups in Taiwan, but also with people from coastal mainland China a nd/or neighboring Paci fic Islanders (Liu, 1995; Tsang and Liu, 2001a). Subsistence The subsistence pattern of a population is important, due to the fact that the consequences of their diet may leave ma rks on their skeletons. Agriculture-related artifacts and ecofacts, such as cultivation tools and rice grains, have been discovered


10 from a nearby site (Chih-san-yan) dated to the Late Neolithic period (Wang, 1984). Rice grain remains are preserved at the SSH s ite, although particular species (wild or domesticated) and amounts have not been reported (Tsang and Liu, 2001a, 2001b). While rice has been suggested to be part of th e diet (Tsang, 2000), a direct linkage between cultivating tools and rice grains is still not clear. Rice agricult ure, at least as part of a large and intensive cultivation system, ca nnot be safely inferred for the SSH site. Shell mounds are commonly found at the si te with more than twenty shellfish species identified to date. Fish bones are also abundant; howev er, specific identifications have yet to be conducted. Butchered faunal remains have been found often burnt, with some showing evidence of cut and bite mark s present which suggest human consumption of these resources (Tsang and Liu, 2001a , 2001b). Analysis of pig mandibles by Lin (1997) suggests that the SSH pe ople were in the process of domesticating wild boar, and hunting would have been a possible way of acquiring them. Other terrestrial protein sources include deer, Formosan barking deer ( Muntiacus reevesii ), chickens, and various bird species. Based on archaeological findings and th e prevalence of dental pathology, Pietrusewsky and Tsang (2003) suggest that the SSH people had a mixed diet. Marine and plant food resources were most likely acquired by hunting and gathering, although a certain degree of plant cultivation is also possible. The Skeletons Two hundreds and ninety burial units were recovered from the site during salvage excavation. More than 100 buria ls are difficult to identify in terms of the interment postures. The majority of identifiable burials were laid on their left side in flexed or semi-flexed posture (110 burials). Right side flexed and semi-flexed is the second most


11 common posture, which represen ts approximately 50 burials. A ll interment patterns were encountered in all excavation un its and levels, and seem to be randomly distributed. Head orientation falls between SW 15 to 60 , with most individuals between SW 45 to 60 . With respect to face directi on, two clusters were found with more individuals facing NW 15 to 60 while others were facing SE 30 to 75 (Tsang and Liu, n.d.). These two directions correspond with the direction of the Taiwan Strait and the Kuang-yin Mountains (the highest mountains visible from the site), respectively. Comb ined with the orientation of the skeleton and face directi on of the face the SSH people tended to bury their dead either facing the sea or the mount ains, although further cu ltural interpretations require further research (Tsang and Liu, 2001b). Cranial morphometric studies by Chang and Pietrusewsky (Chang, 1993; Pietrusewsky and Chang, 2003, N=17 and 13, re spectively, males on ly) suggest that shared biological similarities may exist be tween the Babuza, Pazeh and SSH people. The former two populations are Plains Aborigines who once lived in the central area of the western coast of Taiwan. It has been dem onstrated that the cranial morphology of SSH people shows less affinity with the Atayal tribe, the most geographically-related mountain aborigines, than with other aboriginal p opulations on the Island. In fact, the Atayal sample is even more distinct from the SSH than from the Polynesian/Micronesian sample included in the comparative data. Th erefore, biologically, the SSH people seem to show close affiliation with central-northern Ta iwanese Plains Aborigines and are likely to have genetic links with the Polynesian s (Chang, 1993; Pietrusewsky and Chang, 2003). Some researchers propose that the SSH people show close ties to the extinct northern Taiwanese Plains Aborigines, the Ketagalans and Kavalans, based on cultural


12 remains and geographical loca tion of SSH (Yang, 1961 cited in Pietrusewsky and Chang, 2003). Although craniomophologica l analysis was not performed on the Ketagalan and Kavalan crania, data on population migration, artifact typology, and te mporal association all support a close relationship of SSH occupa nts with its successive populations in the local region (Liu, 1992). The location of the SSH site near the coas t of Taiwan allowed the occupants to be involved in more fre quent contact with other populations, both in northern Taiwan and perhaps eastern Pacifi c Islanders. Intermarriage is one possible explanation for the complexity of biologi cal traits of the SSH people (Chang, 1993). Bioarchaeological Studies of the SSH Site While the archaeological and cultural aspects of the SSH site have been intensively studied, only subsamples of the human skelet al remains have received thorough analysis to date. Two studies on well preserved adu lt skeletons were conducted by Pietrusewsky and Chang (Chang, 1993; Pietrusewsky and Tsang, 2003). Results of their respective findings are presented in Tabl e 2-1 and 2-2. Among these samples, 21 individuals were studied by both sets of researchers, and young adults comprised the bulk of the samples (N= 16/32 and 18/23, respectively). Average stature is estimated at 165 cm for males and 155-160 cm for females. Compared to other Taiwanese aboriginal groups , stature of SSH adult males falls in the range of Plain Aboriginal clusters and is greater than Mountain Dwelling Aboriginals (Chang, 1993). SSH adult males, when compared with Asian and Oceanic populations, are similar in height to the Mongolians, southern Chinese and prehistoric Mariana Islanders, and were shorter than northe rn Chinese and prehistoric Hawaiians (Pietrusewsky and Tsang, 2003).


13 Table 2-1. Paleopathology of the SSH site reported in Chang (1993) Male Female Total Affected Observed % Affected Observed % Affected Observed % N 17 15 32 Mean stature (cm) 164.70 +/3.59 155.16 ---Cribra orbitalia --------------------------EH (C & I)a 108 183 59.0 127 162 78.4 235 345 68.1 EH (All tooth types)a 170 490 34.7 246 417 59.0 416 907 45.9 Caries 20 490 4.1 13 425 3.1 33 917 3.6 Abscessing 8 484 1.7 1 414 0.2 9 898 1.0 Calculus (moderate & marked) 32 491 6.5 29 419 6.9 61 908 6.7 AMTL 9 540 1.7 20 476 4.2 29 1016 2.1 Vertebral osteoarthritis (slight & moderate) 460 1394 33.0 503 1310 38.4 963 2704 35.6 Appendicular osteoarthritis (slight) 202 879 23.0 232 816 28.4 434 1695 25.6 Osteophytosis 62 614 10.1 75 586 12.9 137 1194 11.5 Note: a. statistical significant (p<0.05) derived fr om data in Appendices 13 and 14 in Chang (1993).


14 Table 2-2. Paleopathology of the SSH site reported in Pietrusewsky and Tsang (2003) Male Female Total Affected Observed% Affected Observed % AffectedObserved% N 15 8 23 Mean stature (cm) 165.2 159.9 ---Cribra orbitalia 7 22 31.8 7 14 50.0 14 36 38.9 LEH (C & I)a 37 133 27.8 32 53 60.4 69 186 37.1 Caries 6 395 1.5 0 190 0.0 6 585 1.0 Abscessing 4 405 1.0 0 193 0.0 4 598 0.7 Calculus (moderate) 114 393 29.0 48 167 28.7 162 560 28.9 AMTL 2 419 0.5 0 214 0.0 2 633 0.3 Vertebral osteoarthritis (slight & moderate) 88 1175 7.5 34 575 5.9 122 1750 7.0 Appendicular osteoarthritis (slight) 52 639 8.1 22 354 6.2 74 993 7.5 Osteophytosis 38 517 7.4 26 261 10.0 64 778 8.2 Note: a. statistical significant (p<0.05).


15 Dental health of the SSH population is gene rally quite good. Dental caries rate is low, ranging from 0 to 4.1%, with males having slightly more dental caries than females. Prevalence of moderate to marked dental cal culus differs largely be tween the two studies (7% and 29%), possibly due to differences in scoring procedure. Dental abscessing and antemortem tooth loss (AMTL) are rarely observed. A slightly higher frequency of AMTL presented by Chang (1993) could be the result of more older individuals being incorporated into her subsample, as AMTL is an age-related phenomenon. While calculus is often associated with periodontal disease, no relati onship between calculus and abscessing is reported. One case in ChangÂ’s study demonstrates direct evidence of abscessing caused by severe dental caries. Co mpared with prehistoric Southeast and West Asian and Oceanic populations, prevalence of AMTL, caries and abscessing of SSH are significantly lower. Moderate to marked dental calculus seems to be a significantly more frequent condition in SSH than in prehisto ric Oceanic skeletal remains. High calculus deposition is attributed to a high carbohydrate diet and a more alkaline oral environment. This could be created by betel nut chewi ng or poor dental hygiene (Pietrusewsky and Tsang, 2003). For stress indicators, both studies note dental enamel hypoplasia frequency as moderate to high (Chang, 1993; Pietru sewsky and Tsang, 2003). Females show significantly higher incidence of hypoplastic lesions than males. With canines (60%) and incisors (78%) are most affected. While non-specific in nature, nutritional deficiencies, metabolic disorders and infectious diseases co uld all be the causativ e effect of enamel defects. During childhood, SSH females may experience more physiological stress than males (Chang, 1993; Pietrusewsky and Ts ang, 2003). Although timing of hypoplastic


16 formation is not reported, the overall high prevalence of enamel defects in the SSH population suggests stress during childhood. In term s of cribra orbitalia, higher frequency is observed in females (50%) than males (32 %); however degree of severity and healing process was not noted. Overall pr evalence of cribra orbitalia (39%) at the SSH is related to anemia caused by parasitic in fection, iron-deficient diets, an d/or other chronic stressors in the environment (Pietrusewsky and Tsang, 2003). Prevalence of degenerative changes in SSH people is low among the young adult subsample analyzed by Pietrusewsky and Tsa ng (2003); however, degenerative changes become significant when including indivi duals of more advanced age (Chang, 1993). Based on these findings, it is inferred that SSH people might have engaged in moderate physical activity throughout their lifetime. Presence of vertebral stress fractures and spondylolysis reinforces this interpretation (Pietrusew sky and Tsang, 2003). Finally, evidence of specific infectious disease and/or trauma is extremely rare. In sum, according to Chang (1993) and Pietrusewsky and Tsang (2003), SSH people seem to have enjoyed a generally healthy life although a somewhat stressful childhood and hard-working adulthood was experi enced by most of the SSH inhabitants.


17 CHAPTER 3 MATERIALS AND METHODS Materials The aim of this study is to quantify and assess stress patterns in the SSH population, especially those occurring during childhood. Both Chang (1993) and Pietrusewsky and Tsang (2003), note the need for comparative study of the entire SSH skeletal series. Hence, all 306 individuals exca vated from the site during III, IV, V, VI and VII field periods (1990-1992) were ex amined regardless of preservation or completeness. Sampled remains are curated at the Division of Archaeology, Institute of History and Philology, Academia Sinica, Taiwan. Table 3-1 is the demographic composition of the individuals under study. Table 3-1. Sex and age distribution of SSH skeletal remains studied Male % Female % ?Sex % Total % SA 3 4.8 1 1.8 74 40.7 78 25.5 YA 34 50.7 37 65.0 19 10.4 90 29.4 MA 19 28.4 14 24.6 3 1.6 36 11.8 OA 2 3.0 3 5.3 0 0 5 1.6 ?Age 9 13.4 2 3.5 20 11.0 31 10.1 Total A 64 95.5 56 98.2 42 23.1 162 52.9 ?Age 0 0 0 0 66 36.3 66 21.6 Grand total 67 100.0 57 100.0 182 100.0 306 100.0 Note: SA= subadult; YA= young adult; MA= middle adult; OA= old adult; A= adult. Methods Inventory The majority of excavated human remains were processed by previous workers. Some fragile elements, particularly skulls , however, remain wrapped in adhered soil


18 matrix. When encountered, attempts to expose the dental area, orbital roof and the vault were carefully conducted. Skeletal remains from each burial unit were identified and registered on an inventory sheet with detail ed records for cranial bones. While each burial number was designated for one individual in the field, often buri al units contained elements from more than one individual. Mo st of the additional elements consisted of teeth and jaw bone fragments. For the purpose of this study, skeletal materials that do not belong to the primary individua l under a burial number were itemized separately from the primary skeleton. Minimum numb er of individuals (MNI) fo r contributed feature were identified and recorded. A catalog number was then assigned to each additional individual, CM49-2 and CM49-3, for example, and the primary individual in CM 49 was recorded as CM49-1. When possible teeth were iden tified and features scored on dental charts modified from Brothwell (1981). Teeth were designate d as “present”, “missing” or “unknown”. Teeth in the “present” categor y were pooled results of teet h preserved with and without their matching sockets. A tooth was assigned as “missing”, or lost postmortem, when its tooth socket was present and unf used without the survival of its matching tooth. Teeth in the “unknown” category are those where neit her the teeth nor matching sockets were preserved. Although “unknown” teeth are likely lost due to differential preservation, this category reflects potential agenesis, antemort em tooth loss, and/or postmortem loss. Sex Determination Sex of each individual was assessed with the multifactoral approach following suggestions and illustrations in Buikstra and Ubelaker (1994). For cranial morphology, development of nuchal crest, size and shape of mastoid process, contour of supraorbital margins, development of supraorbital ridge and shape of mental eminence (chin) were


19 observed. Overall size and robusticity of the cranium were noted as supporting criteria. Five sexually dimorphic features of the pe lvis, when presented, were scored. These include the ventral arc, subpubi c concavity, ischiopubic ramus, greater sciatic notch and the preauricular sulcus. Over all contour of the sacrum a nd os coxae, and position of acetabulum were incorporated as further conf irmation of estimated sex, when possible. Bass (1995) and Thieme and Schull (1957, c ited in Stewart, 1979) suggest that maximum clavicle length and sternum/manubr ium ratio are sexually dimorphic traits, although their accuracy less secu re than cranial and pelvic morphology. To increase the rate of identifiable skelet ons, these two features and their indices were observed following Bass (1995) and used as supplemen tary indices. In addition, based on the assumption that males tend to be more physical than females and have more robust musculature, linea aspera development and overall robusticity were incorporated in assessment of sex determination. Left side feat ures of crania and paired elements were observed when present, otherwise ri ght side features were scored. For sex indicators, pelvic features are mo st reliable due to dimorphic differences related to reproduction. Pelvic features are less vulnerable to modification by muscular development which may be caused by sexual differences in occupation practices, and could eliminate cultural bias when determ ining sex of young adults. Therefore, when pelvic features show sex discrepancies agains t cranial and/or other postcranial features, pelvic sex was taken as the estimated sex. Determination of sex in subadults can be di fficult, if not impossible. Therefore, sex identification was not attempted on subadult skeletons (under 20 years of developmental


20 age) unless their developmental age was cl ose to 20 years and mo rphological criteria were clear. Age Determination Age estimation on a skeletal population is th e most critical task for constructing a paleodemographic profile of a populati on without known age data. Multiple age indicators were utilized to increase the identi fiable rate of the skeletons and to avoid bias created by using certain age indicators alone (White, 2000). Caution was followed to prevent imposing an age category on individual s that failed to preserve enough reliable age indicators. For the purpose of this study, individual s were not assigned a specific age. Following Buikstra and UbelakerÂ’s (1994) suggestions, age groups were used, which include subadult (0-20 years), young adult (21-35 years), mi ddle adult (36-50 years) and old adult (50 years and older). If surviv ing features only provide information for distinguishing subadult from adult, the i ndividual was left in either category. Age of subadults was determined in the order of dental erup tion (Ubelaker, 1989), epiphyseal fusion (Bass, 1995; Brothwell, 1981; Ubelaker, 1989) and diaphyseal length (Ubelaker, 1989). For adults, a multifactor al approach was used due to deceasing accuracy of age estimation when individuals of advanced age were encountered. Indicators used include pubic symphysis (B rook and Suchey, 1990), auricular surface (Lovejoy et al., 1985), clavicle fusion (M cKern and Stewart, 1957), suture closure (Buikstra and Ubelaker, 1994; Meindl and Lovejoy, 1985), and dental attrition. Dental attrition rate and pattern is a population specific phenomenon that can be affected largely by coarseness of foodstuffs consumed and task-related abrasions. To detect and establish the relationship between tooth wear and skeletal age (epiphyseal


21 closure, pubic symphysis phase and auri cular surface degeneration) estimates, a subsample of 23 well-preserved, previously st udied individuals (Pietrusewsky and Tsang, 2003) from the SSH site were analyzed. Rela tionship between wear stages and ages presented in Brothwell (1981: 72) and Sm ith (1984: 45-46) was evaluated for the necessity of modification. Estimated age from skeletal indicators is slightly younger than age estimated following Brothwell’s (1981) method and thus de ntal attrition of SSH population is more severe than the hypothetical population from which Brothwell’s chart is derived. Therefore, corresponding age range of each wear stage for Brothwell and Smith (1984) has been modified and assigned as presente d in Tables 3-2 and 3-3. The range of age groups is relatively wide but it serves well for the purpose of this study. Table 3-2. Adjusted molar wear and age re lationships after Br othwell (1981: 72) Brothwell, ‘81 17-25 25-35 33-45 45+ SSH adjusted 15-20 21-35 35-50 50+ Table 3-3. Assigned premolar, canine and inciso r wear stages and ag e relationship after Smith (1984: 45-46) Smith, ‘84 1~3 3~5 5~7 7~8 SSH 15-20 21-35 35-50 50+ Linear Enamel Hypoplasia LEH is a linear or pitted lesion of de creased enamel thickness resulting from disruption of enamel formation. The enamel forming cells, ameloblasts, actively deposit enamel matrix following a scheduled rhyt hm, in a process known as amelogenesis (Goodman and Rose, 1990). Systemic physiologi cal perturbation (stress) or localized dental trauma could affect the metabolism of ameloblasts, and prevent the cells from


22 maintaining normal function. After the stresso r(s) are lifted, amelogenesis resumes and leaves a groove or pitting on the crown surface. Based on animal experiments and clinical observation, more than 100 specific diseases/d isorders contribute to LEH, including but not exclusive of parasitic inf ection, rickets, diabetes, rube lla, syphilis, tetanus, renal failure and brain damage (Koch et al., 1999; Nikiforuk and Fras er, 1981; Pindborg, 1982; Seow et al., 1995; Suckling, 1989; Suckling et al., 1986). Research on nutritionally stressed children in Mexico has establis hed a link between malnutrition and enamel defects (Goodman et al., 1991). Thus, multiple etiologies of LEH make it difficult to determine a specific when encountered in skelet al remains, without the aid of historical documents. In general, LEH is accepted as an indicator of non-specific and systemic stress suffered by an individual during enam el formation. Systemic stress is broadly defined as “any disruption of the normal func tions and homeostasis of a living organism” by Selye (1956, cited in Powell, 1988: 34). The nonspecific nature of stress led Selye to propose the “General Adaptation Syndrome”, re ferring to an organism’s or a population’s reaction to disruptions (str essors) caused by both environm ental and cultural factors (Goodman et al., 1988). In this model, the ultimate goal of an organism’s response to insults or stressors is to maintain hom eostasis. Arrested amelogenesis induced by non-specific stressors is a “tra de-off” of the body to conserve energy in order to cope with the impact of an increased stress load. A biocultural perspective has been frequen tly adopted in bioarchaeological studies (Blakely, 1977; Larsen, 1997; Goodman a nd Armelagos, 1989; Goodman et al., 1984, 1988). The biocultural approach incorporates both physiologi cal and cultural aspects in interpreting the roles of stress indicators fo r skeletal remains, and helps to discern the


23 relationship between populations and the environment (Figure 3-2). In this model, both environment and culture influence on indivi duals survival (e.g. natural resource, and cultural practices for buffering stress) and its pot ential sources of stress (failure to access, or limited access to resources, environmental or culturally induced stressors, and the combined effects of all these factors). If the host (an indivi dual or a group of people) lacks the ability or strength to resist th e stressors, they could in turn experience physiological disruptions (stress). At this st age, the host is sus ceptible and less fit. Depending on the type and magnitude of th e stressor and the hostÂ’s resistance, the consequences could be acute illness, chr onic physiological disruption or death. To his advocates, SelyeÂ’s central notion of stress is th at the response and recovery from an insult should be an adaptive process, and view ed in a broad biocultural context. Figure 3-1. A general model for the study of stress in skeletal populations. From Goodman and Armelagos (1989: 226). Numerous studies have demonstrated th at LEHs are most frequently found on anterior dentition, especially mandibular canines and maxillary central incisors. Goodman and Rose (1990) present a threshol d model for the development of hypoplasias. For a given individual, the threshold level of each tooth type is dictated by the combined effects of unknown underlying e tiology, nutritional intake an d history of illness for a


24 given individual. When the duration and ma gnitude of a stress episode exceeds the threshold, an enamel defect is manifeste d. The threshold also fluctuates along with developmental age and crown he ight (length); the cervical an d incisal third of the crown have higher thresholds, while the threshold is lower at the mid-third of the crown. It is also proposed that the susceptibility and sens itivity to stress varies across tooth types: anterior teeth have lower thresholds than pos terior teeth and are th erefore more easily affected by LEH (for non-threshold explan ation, see Hodges a nd Wilkinson, 1990). It follows that the development of LEH on canines and incisors is representative of an individualÂ’s pattern of physiol ogical stress. However, stress of greater magnitude may be recorded on posterior teeth, a nd serve as an indicator of unusually heavy stress loads (Wright, 1997). LEH is a record of stress during enamel fo rmation, and it is also a gauge of relative health and well-being. It has been associat ed with consequences of weaning-related events during childhood (e.g., Lanphear, 1990; Lovell and Whyte, 1999; Moggicecchi et al., 1994; Saunders, 1999; but see Blakey et al., 1994). Timing of hypoplastic events (sites of LEH) can be identif ied by calculating the distance from the LEH to the CEJ, placing the defect into a chr onological sequence of enamel development for that tooth type. Patterns (distribution and timing) of LEH across tooth types and between individuals studied permit dental physiological reactions to stress insults and stress patterns at the population level to be assessed. LEH is often associated with early mort ality (decreased life expectancy), and frequently is found on individuals who di e at a young age (Cook and Buikstra, 1979; Duray, 1996; Goodman and Armelagos, 1988; Kat zenberg et al., 1996; Rose et al., 1978;


25 Slaus, 2002; Wood, 1996). Nonetheless, the ca usal factor influencing LEH events is usually not the direct cause of death si nce LEH represents a signal of recovery (Palubeckaite, 2001). LEH prevalence by sex is typically not significant. However, it cannot be directly inferred that the physio logical and cultural f actors operating on both sexes are similar; rather, the intertwined re sults of both factors need to be explored (Boldsen, 1997; Guatelli-Steinberg and Lukacs, 1999). Methods for LEH Observation To assess the presence and absence of LEH, each tooth was exposed to a 60 Watt yellowish oblique light from a desk lamp in addition to an over head light source. A hand-held lens with 10x magnification was used to help define border(s) of the defect and to facilitate measurement. Only lesions obser vable without the aid of magnification were recorded as LEH, in order to eliminate ma rks of normal enamel growth or minute but ambiguous stress episodes (Goodman and Ro se, 1990). Location of LEH was measured as the distance from the midpoi nt of the lesion to the midpoi nt of the CEJ. Teeth with broken CEJ or obscured by calculus were not m easured. A digital dent al caliper capable of measuring to one-hundredth of a millimeter was utilized. Although a hypoplastic lesion will not be rem odeled once formed, it is possible for the lesion to be erased by dental attrition. Goodman and Rose (1990) note that LEH are more likely to occur on the mid-third of the buccal surface of the tooth. To control the effect of dental attrition, onl y teeth with more than twothirds of remaining average crown height were incorporated. The aver age crown height of the SSH population was calculated from analysis of unworn teeth (Table 3-4).


26 In this study, LEH data are analyzed at two levels, by tooth and by individual. At the tooth level, prevalence and distribution of LEH can be observed with respect dental physiology and developmental resistance against stress. Table 3-4. SSH mean crown heights and regre ssion formulae modified from Goodman et al. (1980) SSH unworn crown height Tooth N Mean s.d. Regression equationsa (x = distance of LEH from CEJ) UM2 18 6.904 0.746 Age= -0.652 x + 7.5 UM1 22 6.814 0.735 Age= -0.514 x + 3.5 UP4 12 7.085 0.715 Age= -0.494 x + 6.0 UP3 15 7.651 1.088 Age= -0.523 x + 6.0 UC 14 10.679 0.477 Age= -0.562 x + 6.0 UI2 11 10.010 0.684 Age= -0.350 x + 4.5 UI1 19 12.018 0.682 Age= -0.374 x + 4.5 LM2 28 6.463 0.820 Age= -0.619 x + 7.0 LM1 22 7.420 0.642 Age= -0.472 x + 3.5 LP4 14 7.063 0.738 Age= -0.708 x + 7.0 LP3 13 7.901 0.905 Age= -0.633 x + 6.0 LC 11 10.861 0.721 Age= -0.552 x + 6.5 LI2 12 9.934 0.642 Age= -0.403 x + 4.0 LI1 13 9.538 0.701 Age= -0.419 x + 4.0 Note: a. Ages are calculated in years. At this level, sides were pooled by tooth type. At the individu al level, an overall pattern of LEH condition for a population can be reveal ed, which provides insight to the adaptive well-being of the total sample. Here, one side of teeth of each tooth type was included per individual. The selec tion criteria are: 1) individuals observed with no LEH: the si de with larger crow n height remaining was used; 2) individuals observed with LEH on one si de: the side with LEH was chosen. This assumes LEH is a systemic stress indicator, and potential risks of counting localized LEHs as reflection of systemic stress are recognized; and


27 3) individuals observed with LEH on both si des: a) the side with a greater number of LEH episodes was selected; and b) the si de with more crown height remaining was chosen, if same LEH episodes were present on both sides. Developmental age of LEH formation was calculated following two methods. First, regression formulae proposed by Goodman et al . (1980) were adopted. An assumption is held in this method that the rate of enamel deposition on each tooth is constant, i.e., that enamel forms in a linear fashion. Furthermor e, effects of hidden cuspal enamel are not considered when developing the formulae reported in Goodman et al. (1980). As Goodman and Song (1999) have noted, the rela tionship between crown height and timing of enamel formation is not clear, and en amel probably forms in a nonlinear fashion. Hidden cuspal enamel, for example, could complicate estimate of crown formation duration, and in turn, affect results of LEH formation assessment. While many studies have addressed issues related to these unde rlying assumptions, the Goodman method has been widely utilized in reporting the ti ming of LEH formation (Hillson and Bond, 1997; Palubeckaite, 2001; Stodder, 1997; Wright, 1997) . Therefore, GoodmanÂ’s method is used here to provide comparable results. As Wright (1997) has suggested, average crown heights could vary considerably among popul ations. As presented above in Table 3-4, unworn crown height of each tooth type from the SSH sample was substituted in GoodmanÂ’s regression formulae. The second method to determine developmental age of LEH follows Reid and Dean (2000). Their enamel formation seque nces are derived fro m measuring spacing between adjacent daily cross st riations of tooth enamel, and calculating the duration of cuspal enamel formation. Ten equal zones divi de the tooth from the CEJ to the occlusal


28 edge, and timing of enamel formation at each border provided. In this study, SSH crown heights were substituted and Table 3-5 shows their values as modified from Reid and Dean (2000). Their chart is based on empirical observations of tooth development. It accounts for issues related to cuspal enamel and crown geometry, and therefore provides greater accuracy in estimating timing of LEH defects than the Goodman method (Goodman et al., 1980). Porotic Hyperostosis Porotic hyperostosis, a term first used by Angel (1966), is a cranial lesion induced by various types of anemia (but see Wapler et al., 2004). During an anemic episode, compromised red blood cells increase in number to maintain normal oxygen transportation. The external table of bone e xpands due to increased number of red blood cells, and the space between marrow cells is also enlarged. The outer layer of the cranial bone becomes very thin and often exposes the trabecular portion of the diplöe, and a porous area appears (Huss-Ashmore et al., 1982). In some cases, marrow hyperplasia can result in hypertrophy of bony structure, and therefore the cranial bone becomes thickened. The gross lesion is observed as a spongy or sieve-like area on various locations of the cranium. Radiographically, it is characterized as a “hair-on-end” pattern reflecting changes in bony texture (Stuart-Macadam, 1987). If the anemic episode is resolved, remodeling begins to heal the bony lesion, although th e duration of this process varies by individual. Therefore, stages of the porotic lesi on (active, mixed, and healed) are indicative of anemic proce sses experienced by an individual at the time of death (but see Stuart-Macadam, 1985, discussed below). Porotic hyperostosis is most frequently mani fested on orbital roof s (cribra orbitalia), parietals, occipitals and sometimes the tem poral and sphenoid bones. For a period of time


29 it was thought that cribra orbitalia and porotic hyperostosis were independent pathologies, Stuart-Macadam (1989), among othe rs, explicitly confir med that they are actually the result of the same etiology, anemia. Orbital lesions are possibly formed earlier before the vault is affected during anemic events (Caffey, 1937). While congenital anemia such as thalassemia and sick le-cell anemia can cause bony reactions, iron-deficiency anemia is the most commonly encountered type of anemia across populations and geographic areas (Stu art-Macadam, 1992; Wintrobe, 1993). Iron metabolism is a complex physiologi cal process within the human body. A variety of factors and their in teractions could affect iron st atus, including iron absorption, diet, blood loss, iron withholding and geneti cs (Stuart-Macadam, 1998). A single etiology of iron-deficiency anemia, therefore, is not easily specified. Two factors, however, are most often discussed in preh istoric settings: diet and blood loss. Prevalence of porotic hyperostosis (orbital and/or vault) is positiv ely associated with the intensification of agriculture and increasing dependence on cult ivated products. Unbalanced diet, decreased dependence on iron-rich foodstuffs acquired fr om hunting and gathering, and low iron concentration in plant food resources are of ten cited as reasons (Cohen and Armelagos, 1984; Pietrusewsky and Douglas, 2001). However, from the physiological perspective, irondeficiency anemia may not be blamed solely on insufficient diet. Iron metabolism in the human body is reported as an almost cl osed system. Approximately 90% of the iron needed for the production of new red blood cell is recycled from old red blood cells which are destroyed by the metabolic system. It follows, therefore that very little iron is incorporated from the diet (Hoffbrand and Lewis, 1981; Stuart-Macadam, 1998).


30Table 3-5. Chronology of enamel developmen t modified from Reid and Dean (2000) Maxillary dentition UC age (year) from CEJ (mm) UI2 age (year) from CEJ (mm) UI1 age ( from CEJ (mm) 5.3---------------------------------------0.000 5.1-----------------------------------0.000 5.0----------------------------------0.000 4.8---------------------------------------1.068 4.6-----------------------------------1.001 4.4----------------------------------1.200 4.3---------------------------------------2.136 4.1-----------------------------------2.002 3.9----------------------------------2.402 3.8---------------------------------------3.204 3.7-----------------------------------3.003 3.4----------------------------------3.604 3.4---------------------------------------4.272 3.3-----------------------------------4.004 2.9----------------------------------4.806 3.0---------------------------------------5.340 2.9-----------------------------------5.005 2.4----------------------------------6.008 2.7---------------------------------------6.407 2.7-----------------------------------6.006 2.0----------------------------------7.210 2.4---------------------------------------7.475 2.4-----------------------------------7.007 1.8----------------------------------8.412 2.2---------------------------------------8.543 2.2-----------------------------------8.008 1.6----------------------------------9.614 1.9---------------------------------------9.611 2.0-----------------------------------9.009 1.3---------------------------------10.816 1.7-------------------------------------10.679 1.8----------------------------------10.010 1.1---------------------------------12.018


31Table 3-5. Chronology of enamel development modi fied from Reid and Dean (2000) (continued) Mandibular dentition LC age from CEJ (mm) LI2 age (yea from CEJ (mm) LI1 age ( from CEJ (mm) 1.5-------------------------------------10.861 1.0-----------------------------------9.934 1.0----------------------------------9.538 1.7---------------------------------------9.775 1.1-----------------------------------8.941 1.1----------------------------------8.584 2.0---------------------------------------8.689 1.3-----------------------------------7.947 1.3----------------------------------7.630 2.3---------------------------------------7.603 1.5-----------------------------------6.954 1.5----------------------------------6.677 2.7---------------------------------------6.517 1.8-----------------------------------5.960 1.7----------------------------------5.723 3.1---------------------------------------5.431 2.1-----------------------------------4.967 2.0----------------------------------4.769 3.6---------------------------------------4.344 2.4-----------------------------------3.974 2.3----------------------------------3.815 4.2---------------------------------------3.258 2.8-----------------------------------2.980 2.6----------------------------------2.861 4.9---------------------------------------2.172 3.3-----------------------------------1.987 3.0----------------------------------1.980 5.6---------------------------------------1.086 3.7-----------------------------------0.993 3.4----------------------------------0.954 6.2---------------------------------------0.000 4.2-----------------------------------0.000 3.8----------------------------------0.000


32 Instead, loss of blood, by menstruation, trau ma and chronic bleeding, is the source of decreased iron concentration. Infection, most commonly from parasites, has long been linked to internal chronic blood loss. Research on the prevalence of porotic hyperostosis in populations with a high pa rasitic load has revealed that the incidence of iron-deficiency anemia is correlated to pa rasitic infection (Hengen, 1971; Merbs, 1992; Walker, 1986). Whether anemia is an “ad aptive reaction” to parasitic load (Stuart-Macadam, 1992) or simply a sign of in sufficient iron due to blood loss is not yet clear. As for the major factors of iron st atus, Holland and O’Brien (1997) have suggested that while parasitic infection (chronic blood loss) and its potential adaptive function is plausible, acquired iron deficiency (dieta ry-related) should be emphasized as well. According to Stuart-Macadam (1985), porotic hyperostosis is a representation of a childhood condition. First of all, porotic hype rostosis is most commonly observed in young children than in adults. Secondly, clinic al research reports that children (and premenopausal women) are most susceptible to developing iron-deficiency anemia. Thirdly, development of marrow physiology has made children even more vulnerable to manifest bony lesions when anemia impacts. Lastly, she speculates that bony lesions may actually be the delayed reactions to earlier anemic episodes, not a timely reflection of current anemia. Therefore, porotic hypero stosis observed among older subadults and adults is either a persisting anemic process due to genetic factors (initiated in early childhood), or a reflection of childhood anemic episodes. The notion of porotic hyperostosis as a childhood pathology is supporte d implicitly or explicitly by Blom et al. (in press), Mensforth et al. (1978) , and Salvadei et al. (2001).


33 Methods for Porotic Hyperostosis Observation Porotic hyperostosis was determined following the criteria suggested by Stuart-Macadam (1985) and reference pict ures in Buikstra and Ubelaker (1994). Preserved orbital roofs, parietal bones, and occipital bones were examined for the lesion. Lesions were assigned into one of the four categories of sever ity, including barely discernible, porosity only, porosity with co alescence of foramina but no thickening, and coalescence foramina with increased thickness (Buikstra and Ubelaker , 1994). Activity of the lesion was determined based on signs of healing. Active, healed, and mixed reactions at the time of death were scored. Since porot ic hyperostosis is a result of systemic disruption (Stuart-Macadam, 1989), bilateral symmetry of the le sion was also recorded if both sides of the bone were available. Dental Pathology Diet and subsistence patterns of a population are influential factors of a populationÂ’s well-being. Several dental pathol ogies/conditions including dental caries, calculus and abscessing are particularly useful in identi fying dietary and subsistence-related conditions (e.g., Cohen and Armelagos, 1984). Prevalence of dental caries is frequently used as an index to discern hunting-gathering from agricultural subsistence (Turner, 1979). Dent al calculus is an indicator of food consistency and oral alkalinity. As for dental abs cessing, it is a condition caused by periodontal disease and/or advanced dental caries (Larsen, 1997) . In this study, these three dental pathologies/conditions are incorporated to facilitate the r econstruction of SSH subsistence, and to permit the LEH and porotic hyperostosis data to be interpreted in a complimentary perspective. Presence or absence, and location of each grossly visible


34 lesion was recorded. Incidence of moderate and severe dental calculus was noted, following Brothwell (1981). Statistical Analysis Standard descriptive statisti cs and frequency tables were computed to characterize lesion prevalence and peak LEH formation zones. Student t-test was used when evaluating sex and age differences on mean LEH counts among tooth types. Chi-square statistics were performed in testing distri bution differences of the lesions among sex and age groups. When degree of freedom equals one , probability of FisherÂ’s exact test (FET) was reported. Otherwise, PearsonÂ’s chi-square value and probability were used. All tests were run using SPSS version 12.0 and the possibility of committing a Type I error ( ) was set to 0.05.


35 CHAPTER 4 RESULTS In this chapter, results of LEH observa tions are presented in two categories: by tooth and by individual. In the tooth category, overall prev alence, LEH prevalence among sex and age, and mean number of LEH among sex and age are reported. In the individual category, data of overall prevalence and by sex and age are provided. Prevalence of LEH on deciduous teeth is briefly discussed, but fo cus is placed on the permanent dentition. Overall patterns and sex comparisons of LEH formation timing are then demonstrated according to method used. In terms of poro tic hyperostosis, overall prevalence and distribution by sex and age is presented. To reveal the morbidity pattern of the SSH people based on the two pathologies, numbers of individuals affect ed with LEH and/or porotic hyperostosis are tabulated and a co rrelation statistic is provided. Lastly, prevalence of dental pathologi es (caries, calculus, and abscessing) are reported by the tooth count method. Linear Enamel Hypoplasias Deciduous Dentition Table 4-1 is the summary of LEH oc currence on deciduous dentition by tooth count. Among the observed deciduous teet h, two cases of LEH were scored: one maxillary canine and one central incisor, and each tooth presents one LEH episode. These two incidences belong to two very young ch ildren who do not have other macroscopic dental pathologies. Overall, there are virt ually no LEH events that manifested on deciduous dentition of indivi duals from SSH population.


36 Permanent Dentition LEH by tooth count Overall assessment. Percentage of teeth with at least one visible LEH lesion by tooth types presented in Tabl e 4-2 and Figure 4-1. LEH is most frequently observed on mandibular canines (56%), followed by maxillary canines (41%), central incisors (32%) and lateral incisors (26%). Maxillary denti tion exhibits slightly higher LEH prevalence than mandibular dentition. Post erior dentition (premolars a nd molars) is significantly less affected with LEH than anterior dentition (d f= 1, FET p= .000). When all tooth types are combined, overall LEH prevalence of th e SSH site is 18.2% (Table 4-3). The mean number of LEH events was calculated by dividing total number LEH observed either by the total number of teet h available for observation, or by the total number of teeth that are affected wi th LEH. While suffering from differential preservation among tooth types, by using th e number of all available teeth as the denominator provides a general trend of mean LEH episodes across tooth types. For the affected teeth category, multiple stress ep isodes on each tooth type can be clearly detected. Since only teeth that exhibit at l east one LEH episode were incorporated into the sample, the minimum number of LEH count is one. One shortcoming of the affected teeth category is that the result can be biased by small sample size in certain tooth types, especially those not frequently affected with LEH, and this limits meaningful statistical analysis. Therefore, to ensure more precise results, the mean number of LEH counts is presented in both categories.


37Table 4-1. LEH prevalence on deci duous dentition by tooth count Maxillary dentition m2 m1 c i2 i1 LEH OBS % LEH OBS % LEH OBS % LEH OBS % LEH OBS % R 0 22 0 0 22 0 1 14 7.1 0 7 0 1 5 20.0 L 0 21 0 0 21 0 0 11 0 0 4 0 0 8 0 Pooled 0 43 0 0 43 0 1 25 4.0 0 11 0 1 13 7.7 Mandibular dentition m2 m1 c i2 i1 LEH OBS % LEH OBS % LEH OBS % LEH OBS % LEH OBS % R 0 27 0 0 24 0 0 14 0 0 5 0 0 4 0 L 0 20 0 0 20 0 0 11 0 0 3 0 0 3 0 Pooled 0 47 0 0 44 0 0 25 0 0 8 0 0 7 0 Note: LEH= teeth affected with LEH; OBS= observed teeth.


38Table 4-2. LEH prevalence on perm anent dentition by tooth count Maxillary dentition M3 M2 M1 P4 P3 C I2 I1 LEH OBS % LEH OBS % LEH OBS % LEH OBS % LEH OBS % LEH OBS % LEH OBS % LEH OBS % R 6 65 9.2 19 100 19.0 9 112 8.0 9 96 9.4 8 95 8.4 32 87 36.8 25 87 28.7 23 74 31.1 L 6 67 9.0 17 103 16.5 16 96 16.7 5 90 5.6 12 88 13.6 38 82 46.3 18 78 23.1 23 70 32.9 Pooled 12 133 9.0 36 203 17.7 25 208 12.0 14 186 7.5 20 183 10.9 70 169 41.4 43 165 26.1 46 144 31.9 Mandibular dentition M3 M2 M1 P4 P3 C I2 I1 LEH OBS % LEH OBS % LEH OBS % LEH OBS % LEH OBS % LEH OBS % LEH OBS % LEH OBS % R 0 70 0 4 106 3.8 9 109 8.3 9 93 9.7 16 92 17.4 46 79 58.2 16 59 27.1 6 41 14.6 L 2 58 3.5 8 96 8.3 12 98 12.2 10 91 11.0 17 82 20.7 42 79 53.2 13 59 22.0 7 49 14.3 Pooled 2 128 1.6 12 202 5.9 21 207 10.1 19 184 10.3 33 174 19.0 88 158 55.7 29 118 24.6 13 90 14.4


39 0 10 20 30 40 50 60 UM3UM2UM1UP4UP3UCUI2UI1LI1LI2LCLP3LP4LM1LM2LM3Tooth typePercent (%) Figure 4-1. Percentage of observed permanent teeth affected with LEH by tooth types. Table 4-3. Overall prevalence of LEH by dentition. Ages and sexes pooled Maxillary MandibularAnteriorPosteriorOverall Affected 266 217 289 194 483 Observed 1391 1261 844 1808 2652 % 19.1 17.2 34.2 10.7 18.2 Table 4-4, Figures 4-2 and 4-3 present the overall mean number of LEH episodes, with ages and sexes pooled. Data show that mandibular canines are most affected by LEH, followed by maxillary canines and inci sors. Anterior dentition shows greater numbers of LEH than the posterior denti tion. Among anterior dent ition, canines from upper and lower jaws and maxillary incisors are particularly high in mean LEH counts. Interestingly, premolars that are said to ha ve higher stress threshol ds, exhibit slightly more mean LEH counts than mandibular incisors. While this could be an artifact of more


40 severe attrition on mandibular incisors, stronger stress loads in SSH site could not be ruled out. Table 4-4. Mean number of LEH on obser ved and affected teeth by tooth types Maxillary dentition Mean LEH counts on observed teeth Mean LEH counts on affected teeth Mandibular dentition Mean LEH counts on observed teeth Mean LEH counts on affected teeth UM3 0.11 1.70 LM3 0.02 1.00 UM2 0.20 1.11 LM2 0.06 1.08 UM1 0.12 1.00 LM1 0.11 1.05 UP4 0.10 1.29 LP4 0.11 1.05 UP3 0.13 1.21 LP3 0.25 1.26 UC 0.70 1.66 LC 0.96 1.72 UI2 0.39 1.49 LI2 0.30 1.21 UI1 0.51 1.49 LI1 0.17 1.15 0.00 0.20 0.40 0.60 0.80 1.00 1.20 UM3UM2UM1UP4UP3UCUI2UI1LI1LI2LCLP3LP4LM1LM2LM3Tooth typeMean number of LEH Figure 4-2. Mean number of LEH on observed teeth by tooth types.


41 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 UM3UM2UM1UP4UP3UCUI2UI1LI1LI2LCLP3LP4LM1LM2LM3Tooth typeMean number of LEH Figure 4-3. Mean number of LEH on affected teeth by tooth types. LEH prevalence and sex. Teeth from all sexed indi viduals were evaluated for LEH prevalence. Contrary to other studies , LEH prevalence at SSH shows significant difference between sexes, with females havi ng higher LEH incidence than males; based on tooth type analysis (Table 4-5, df= 1, FET p=.034). Twenty percent of teeth from females exhibit at least one LEH, while 17% of teeth from males have LEH. The sexual difference is especially pronounced on anterior dentition (Table 4-6), where prevalence of LEH on canines and incisors is 13% higher in adult females than adult males (df= 1, FET p=.000). However, posterior dentition of male s shows slightly higher prevalence of LEH than females (Table 4-7). LEH prevalence and age. Table 4-8 demonstrates the di stribution of LEH affected teeth by age. Subadults and young adults show si milar frequencies of teeth having at least one LEH event.


42 Table 4-5. Overall LEH distribution by sex Sex Female Male Total LEH no LEH 784 860 1644 w/ LEH 202 173 375 Total 986 1033 2019 % w/ LEH 20.5 16.7 18.6 Table 4-6. LEH distribution on anterior dentition by sex Sex Female Male Total LEH no LEH 187 226 413 w/ LEH 131 86 217 Total 318 312 630 % w/ LEH 41.2 27.6 34.4 Table 4-7. LEH distribution on posterior dentition by sex Sex Female Male Total LEH no LEH 597 634 1231 w/ LEH 71 87 158 Total 668 721 1389 % w/ LEH 10.6 12.1 11.4 Note: df= 1, FET p= .447 About 20% of teeth from these two age groups were observed with LEH present. Only 10% of middle adult teeth have LEH, where i ndividuals older than 50 years of age show 17% of their teeth affected. Prevalence of LEH across age groups is significantly different ( X2= 24.116, df= 3, p= .000) using chi-square analysis. The prevalence of the middle adult group is remarkably different fr om other age groups represented at SSH. When the age groups are categorized into younger (subadults and young adults) and older individuals (middle and old adul ts), prevalence of teeth affected by LEH is significantly different between younger and ol der individuals (Table 4-9) . Teeth belonging to younger members of the SSH population are twice as like ly to be affected with LEH as teeth of older members (df= 1, FET p= .000).


43 To investigate the potential impact of st ress load on an indi vidualÂ’s longevity, LEH prevalence in subadults was compared with that of the adults by grouping all teeth associated with skeletally mature individuals into one category. Here , the total number of teeth analyzed is slightly di fferent from above (Table 4-10), because some individuals could only be distinguished as adults from s ubadults and could not be assigned to a finer age group due to the lack of reliable aging crite ria. Teeth from these individuals were not incorporated in age group comparisons. LEH prevalence of subadult and adult teeth is 21% and 18%, respectively, and no statistical significance was found. Table 4-8. Overall LEH di stribution by age groups Age groups SA (0-20) YA (21-35) MA (36-50) OA (50+) Total LEH no LEH 425 1298 389 29 2141 w/ LEH 112 314 43 6 475 Total 537 1612 432 35 2616 % w/ LEH 20.9 19.5 10.0 17.1 18.2 Table 4-9. Overall LEH distribution by younger (subadult and young adult) and older (middle adult and old adult) age groups Age groups SA & YA MA & OA Total LEH no LEH 1723 418 2141 w/ LEH 426 49 475 Total 2149 467 2616 % w/ LEH 19.8 10.5 18.2 Mean number of LEH and sex. Mean number of LEH on observed and affected teeth is presented in Table 4-11 and Figures 4-4 and 4-5. In both samples, males generally have more LEH episodes on posterior dentition than females. Maxillary third premolars are the most sexually dimorphic in LEH counts. Fe males exhibit more LEH counts on anterior dentition when all observed teeth are considered. Female mandibular canines, among others, show greater LEH per tooth than ma les which is of statistical significance.


44 Table 4-10. Overall LEH distribution by maturity (subadult and adult) Maturity SA (0-20) Adult (21+) Total LEH no LEH 425 1732 2157 w/ LEH 112 371 483 Total 537 2103 2640 % w/ LEH 20.9 17.6 18.3 Note: df= 1, FET p= .91 In the sample of affected teeth, teeth in almo st all male tooth type s, especially canines and incisors, show multiple LEH episodes. Mean number of LEH and age. Table 4-12, Figures 4-6 and 4-7 show the mean LEH counts of each tooth type by age. Subadu lts generally exhibit more or comparable average LEH episodes than young adults, both in observed and affected categories. The differences are more pronounced in the observed teeth group, although this is not statistical significant. Comparing data of mi ddle adults to young adu lts and subadults, the results vary by tooth type. Middle adults have less mean LEH counts in the maxillary dentition when analyzed against all observable teeth, but do not show consistent trends in other comparisons. The unusually high number of mean LEH count of old adults is likely due to very small sample size, both in obs ervable and affected t eeth, and should not be considered a realistic reflection of the SSH population. In general, younger individuals (subadults and young adults) ha ve higher numbers of LEH than older individuals. Due to the skew ed nature of the older adults in the SSH sample, the Student t-test was not performe d to assess the differences in LEH count between younger individuals and the older ones.


45 Table 4-11. Sex comparison of mean LEH count on observed and affected teeth by tooth types Mean LEH counts on observed teeth Mean LEH counts on affected teeth Male Female Male Female Tooth type N LEH N LEH N LEH N LEH UM3 63 0.13 51 0.12 7 1.14 5 1.20 UM2 82 0.23 75 0.17 15 1.27 13 1.00 UM1 74 0.12 79 0.08 9 1.00 6 1.00 UP4 73 0.15 71 0.07 8 1.38 4 1.25 UP3 73 0.19a 70 0.06a 10 1.40b 4 1.00b UC 69 0.65 59 0.63 25 1.80 24 1.54 UI2 66 0.32 59 0.51 14 1.50 21 1.43 UI1 50 0.34 53 0.58 11 1.55 22 1.41 LM3 62 0.03 48 0.00 2 1.00 0 --LM2 80 0.09 46 0.04 7 1.00 4 1.25 LM1 72 0.17 42 0.10 11 1.09 6 1.00 LP4 73 0.11 42 0.17 7 1.14 10 1.00 LP3 68 0.24 44 0.36 11 1.45 19 1.16 LC 59 0.68c 44 1.18c 25 1.60 45 1.62 LI2 39 0.26 38 0.39 8 1.25 12 1.25 LI1 29 0.14 28 0.25 2 2.00 7 1.00 Note: Bold indicates statis tical significance. a. p= .046; b. p= .037; c. p= .013 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 UM3UM2UM1UP4UP3UCUI2UI1LI1LI2LCLP3LP4LM1LM2LM3Tooth typeMean number of LEH Male Female Figure 4-4. Sex comparison of mean LEH count on observed teeth by tooth types.


46 0.00 0.50 1.00 1.50 2.00 2.50 UM3UM2UM1UP4UP3UCUI2UI1LI1LI2LCLP3LP4LM1LM2LM3Tooth typeMean number of LEH Male Female Figure 4-5. Sex comparison of mean LEH c ount on affected teeth by tooth types. To assess the relationship between number of stress episodes and longevity, mean LEH counts were compared between subadult and ad ult teeth. Results are presented in Table 4-13. Subadults show more LEH counts than adu lts in the majority of tooth types when calculated using all observable teeth. Differences on maxillary first molars and mandibular fourth premolars are of statistical significance. When assessed in the affected teeth category, the tendency becomes less clear. Adults seem to have more multiple LEH episodes in each tooth than subadults do, with the difference on maxillary third premolars reaching statistical significance. However, the two most stress sensitive teeth, mandibular canines and maxillary central incisors, exhibi t more LEH events on subadults than adults, where the difference between subadults and adults on maxillary central incisors is significant. LEH by individual count At the individual level, an overall percentage is not provided because not all individuals represented in the SSH series have their complete dentition preserved.


47Table 4-12. Mean LEH count by age groups Mean number of LEH on observed teeth M ean number of LEH on affected teeth SA YA MA OA SA YA MA OA Tooth N LEH# N LEH#N LEH#N LEH#N LEH# N LEH#N LEH#N LEH# UM3 11 0.09 88 0.11 28 0.11 4 0.00 1 1.00 9 1.11 2 1.50 0 --UM2 40 0.28 115 0.21 42 0.07 4 0.25 11 1.00 20 1.20 3 1.00 1 1.00 UM1 53 0.21 116 0.09 46 0.04 1 0.00 11 1.00 11 1.00 2 1.00 0 --UP4 28 0.25 109 0.11 36 0.00 4 0.00 5 1.40 9 1.33 0 --0 --UP3 30 0.20 114 0.15 34 0.00 2 0.00 6 1.00 13 1.31 0 --0 --UC 37 0.78 103 0.76 24 0.08 2 1.00 19 1.53 45 1.73 2 1.00 1 2.00 UI2 33 0.30 94 0.44 29 0.24 4 0.50 7 1.43 28 1.46 4 1.75 2 1.00 UI1 38 0.63 81 0.54 21 0.24 0 --12 2.00 28 1.57 5 1.00 0 --LM3 8 0.00 77 0.03 38 0.00 2 0.00 0 --2 1.00 0 --0 --LM2 39 0.05 139 0.07 30 0.00 2 0.00 2 1.00 9 1.11 0 --0 --LM1 62 0.10 123 0.10 22 0.18 0 --6 1.00 11 1.09 4 1.00 0 --LP4 31 0.03 127 0.13 23 0.09 2 0.00 1 1.00 16 1.06 2 1.00 0 --LP3 28 0.21 119 0.27 23 0.22 1 0.00 5 1.20 24 1.33 5 1.00 0 --LC 36 0.92 100 0.97 19 0.74 3 1.00 17 1.94 60 1.62 9 1.56 2 1.50 LI2 31 0.23 75 0.29 13 0.46 1 0.00 6 1.17 18 1.22 5 1.20 0 --LI1 30 0.10 53 0.23 5 0.00 2 0.00 3 1.00 10 1.20 0 --0 ---


48 Counts of individuals are utili zed when reporting the prevalence. As reported earlier, two individuals with deciduous te eth have LEH. Therefore, only individuals with permanent dentition are incorporated in this section. Sa mple size of each tooth type varies depending on preservation, particular dental anatomy of the tooth (e.g., number of roots, shape and thickness of enamel), resistance against attr ition and eruption sequence. Percentages here are for descriptive purposes, and should not be read as overall SSH LEH prevalence at the individual level. As shown in Table 414, posterior dentition, especially molars are best preserved. There are fewer individuals with available an terior dentition. Overall assessment. Table 4-14 and Figure 4-8 show the proportion of individuals who are present/absent with at least one LEH episode. Individuals we re incorporated in the sample if one or both sides of a particul ar tooth type were available for observation. In agreement with findings presented by tooth counts, canines and incisors are the most affected teeth by individual assessment. Among 110 individuals with at least one maxillary canine preserved, 48% of them are affected with LEH. Sixty-two percent of individuals exhibit LEH on at least one mandi bular canine. While pos terior dentition is less likely affected, it is not free from manifesting enamel de fects during stress events. Among individuals with LEH, the vast majority are affected with one episode (Table 4-14 and Figure 4-9). More than two LEH episodes are rarely observed on posterior dentition. One individual was found with thr ee LEH events on maxillary second molar, and another two show three LEH events on hi s/hers mandibular third premolar. Multiple LEH episodes are frequently present on anteri or dentition. For the canines, more than 50% of individuals that are affected with LEH have two or more LEH counts. Five


49 individuals exhibit four epis odes on their canines, indicating recurring stress events. Only one of these individuals faile d to survive to adulthood. LEH prevalence and sex. Distribution of LEH prevalence among sexes in each tooth type is presented in Table 4-15 and Figure 4-10. On maxillary posterior dentition and mandibular molars, males are more likely to be affected with LEH events than females, although the differences are not statis tically significant. Females, in contrast, more frequently exhibit LEH episodes on their an terior teeth. All six anterior tooth types affected universally show higher proportions in female than in male individuals, particularly with mandibular canines. Despite preservation bias and uneven tooth counts, there are 102 i ndividuals (1/3 of all SSH human remains recovered) affected w ith at least one LEH ep isode on at least one tooth. Among them, there are 37 adult females, 29 adult males and 26 subadults of unknown sex. LEH prevalence and age. Table 4-15 and Figure 4-11 demonstrate the age distribution of individuals affected with LEH events. Subadults and young adults contribute to the vast majority of the aff ected individuals. In pa rticular, the young adult age group is most likely to be observed with one or more linear enamel defects. Middle and old adults are much less affected by LEH. LEH prevalence and age. Table 4-15 and Figure 4-11 demonstrate the age distributions of individuals affected w ith LEH events. Subadults and young adults contribute to the vast majority of the aff ected individuals. In pa rticular, the young adult age group is most likely to be observed with one or more linear enamel defects. Middle and old adults are much less affected by LEH.


50 0.00 0.20 0.40 0.60 0.80 1.00 1.20 UM3UM2UM1UP4UP3UCUI2UI1Tooth typeMean number of LEH SA YA MA OA 0.00 0.20 0.40 0.60 0.80 1.00 1.20 LM3LM2LM1LP4LP3LCLI2LI1Tooth typeMean number of LEH SA YA MA OA Figure 4-6. Age comparison of mean LEH count on observed teeth by tooth types.


51 0.00 0.50 1.00 1.50 2.00 2.50 UM3UM2UM1UP4UP3UCUI2UI1Tooth typeMean number of LEH SA YA MA OA 0.00 0.50 1.00 1.50 2.00 2.50 LM3LM2LM1LP4LP3LCLI2LI1Tooth typeMean number of LEH SA YA MA OA Figure 4-7. Age comparison of mean LEH c ount on affected teeth by tooth types.


52 Table 4-13. Comparison of mean LE H count on subadults and adults Mean number of LEH on observed teeth Mean number of LEH on affected teeth Subadult Adult Subadult Adult Tooth N LEH# N LEH# N LEH# N LEH# UM3 11 0.09 121 0.11 1 1.00 11 1.18 UM2 40 0.28 162 0.18 11 1.00 25 1.16 UM1 53a 0.21a 164a 0.09a 11 1.00 14 1.00 UP4 28 0.25 153 0.08 5 1.40 9 1.33 UP3 30 0.20 151 0.11 6c 1.00c 13c 1.31 c UC 37 0.78 132 0.66 19 1.53 51 1.71 UI2 33 0.30 130 0.40 7 1.43 36 1.44 UI1 38 0.63 105 0.47 12d 2.00d 33d 1.48d LM3 8 0.02 119 0.00 0 --2 1.00 LM2 39 0.07 162 0.05 2 1.00 10 1.10 LM1 62 0.10 145 0.11 6 1.00 15 1.07 LP4 31b 0.12b 163b 0.03b 1 1.00 18 1.06 LP3 28 0.21 144 0.26 5 1.20 29 1.28 LC 36 0.92 122 0.93 17 1.94 71 1.61 LI2 31 0.32 87 0.23 6 1.17 23 1.22 LI1 30 0.20 60 0.10 3 1.00 10 1.20 Note: Bold indicates statistical significance. a. p= .047; b. p= .035; c. p= .04; d. p= .024. Table 4-14. LEH distributi on by individual count LEH absent/present Tooth Individual observed No LEH 1 LEH 2 LEHs 3 LEHs 4 LEHs UM3 88 79 7 2 0 0 UM2 128 101 24 2 1 0 UM1 135 116 19 0 0 0 UP4 123 112 7 4 0 0 UP3 114 102 10 2 0 0 UC 110 57 22 23 6 2 UI2 106 72 21 10 3 0 UI1 96 62 18 11 5 0 LM3 88 86 2 0 0 0 LM2 124 114 9 1 0 0 LM1 124 109 14 1 0 0 LP4 112 97 14 1 0 0 LP3 112 85 22 3 2 0 LC 97 37 27 19 10 4 LI2 75 53 17 5 0 0 LI1 56 48 7 1 0 0


53 0 20 40 60 80 100 120 140 160 UM3UM2UM1UP4UP3UCUI2UI1LI1LI2LCLP3LP4LM1LM2LM3Tooth typeNumber of individuals w/ LEH no LEH Figure 4-8. LEH distributi on by individual count. 0 10 20 30 40 50 60 UM3UM2UM1UP4UP3UCUI2UI1LI1LI2LCLP3LP4LM1LM2LM3Tooth typeNumber of individuals 4 episodes 3 episodes 2 episodes 1 episode Figure 4-9. Distribution of LEH ep isodes according to tooth type.


54 Table 4-15. LEH distribution by sex and age groups by number of individuals Sex Age groups Male Female SA YA MA OA UM3 5 4 1 7 1 0 UM2 10 11 7 16 2 1 UM1 9 4 7 10 1 0 UP4 6 3 4 7 0 0 UP3 5 3 4 8 0 0 UC 17 18 15 33 2 1 UI2 12 15 7 27 3 2 UI1 9 15 9 20 4 0 LM3 2 0 0 2 0 0 LM2 6 3 2 7 0 0 LM1 7 6 3 9 3 0 LP4 5 8 1 12 2 0 LP3 8 15 5 18 4 0 LC 18 28 13 39 7 1 LI2 6 9 5 14 3 0 LI1 1 4 2 6 0 0 0 5 10 15 20 25 30 35 40 45 50 UM3UM2UM1UP4UP3UCUI2UI1LI1LI2LCLP3LP4LM1LM2LM3Tooth typeNumber of individuals Female Male Figure 4-10. Sex comparison of LEH dist ribution among tooth types by number of individuals.


55 Timing of LEH formation Two methods for correlating LEH to de velopmental age are followed in this study. The regression formulae of Goodman et al. (1980) and charts produced by Reid and Dean (2000) have been modified by substituting in SSH unworn crown heights. Overall distribution of LEH formation time by tooth type is presented separately by method. To provide for an aggregated pattern of childhood stress at the population level, individual counts are used. Distribution of LEH by developmental zones is then compared between sexes. However, data are assessed by utilizing count of LEH episodes in each developmental zone in attempt to lim it potential bias created by small sample size when individual counts were divided into sex groups. 0 10 20 30 40 50 60 70 UM3UM2UM1UP4UP3UCUI2UI1LI1LI2LCLP3LP4LM1LM2LM3Tooth typeNumber of individuals OA MA YA SA Figure 4-11. Age comparison of LEH dist ribution among tooth types by number of individuals. Overall assessment Method one: Goodman et al. (1980). Using formulae modified from Goodman and colleagues (1980), half-year intervals of LEH timing is adopted. Numbers of individuals that exhibit LEH episodes in pa rticular developmental zones across tooth


56 types are presented in Table 4-16. Figure 4-12 graphically illustrate the data by tooth type. Third molars were not assessed for LEH formation timing due to the lack of reference formulae. LEH are generally less likely to be obse rved during initial or terminal crown formation periods. Most LEH episodes are formed on the mid-third of the crown dimension. This is especially the case on th e maxillary dentition. While crown formation times vary by tooth type, maxillary dentition s eems to record a trend that individuals tend to suffer from stress loads between 2.0 to 4.0 years of age, especially between 3.5 to 4.0 years. Maxillary second molars are affected w ith LEH events later than other tooth types and show peak LEH formation between 5.0 to 5.5 years. The mandibular dentition, in contrast, doe s not show a clear pattern of defect formation as peak duration of LEH formation is more varied. Indi viduals between 2.0 to 5.0 years are more prone to LEH formation, wh ile the period of 3.0 to 3.5 has relatively fewer of LEH events. Interestingly, mandibular fourth premolars develop more LEHs at the sixth year postpartum. Method two: Reid and Dean (2000 ). Reid and Dean (2000) divide a tooth crown into ten zones of equal length. Developmental age of each zone varies by tooth type. Figures 4-13 and 4-14 represen t age patterns of LEH formati on by number of individuals on the maxillary and mandibular dentition, respectively. Number of individuals (N) incorporated in this method of each tooth type is the same with that used in the previous method. Most LEH episodes are observed on the mid-third of a tooth crown and continue to appear on zones near the cemento-ename l junction. Individuals with LEH events on the occlusal end of a crown are observed less frequently.


57 Table 4-16. Developmental zones of LEH form ation across tooth t ypes using modified method from Goodman et al. (1980) , by number of individuals Tooth (N) Zone UM2 (26) UM1 (18) UP4 (9) UP3 (11) UC (46) UI2 (33) UI1 (31) 7.0-7.5 0 6.5-7.0 0 6.0-6.5 9 5.5-6.0 6 0 0 1 5.0-5.5 11 2 0 3 4.5-5.0 3 3 4 2 4.0-4.5 1 2 3 16 1 1 3.5-4.0 0 4 6 22 17 6 3.0-3.5 0 1 1 0 11 14 17 2.5-3.0 2 0 0 15 15 13 2.0-2.5 7 0 3 4 12 1.5-2.0 6 6 1 0 1.0-1.5 1 0 0 1 0.5-1.0 1 0 0 0.0-0.5 0 0 0 LM2 (10) LM1 (14) LP4 (14) LP3 (26) LC (56) LI2 (21) LI1 (7) 0.0-0.5 0 0 0 0.5-1.0 0 0 0 0 1.0-1.5 2 0 0 0 0 1.5-2.0 2 0 0 0 0 2.0-2.5 4 0 1 7 7 2 2.5-3.0 7 0 4 3 11 5 3.0-3.5 0 0 0 3 11 3 0 3.5-4.0 0 1 10 8 0 0 4.0-4.5 4 2 6 25 4.5-5.0 1 2 3 29 5.0-5.5 2 3 5 14 5.5-6.0 2 5 1 5 6.0-6.5 2 1 1 6.5-7.0 0 1 Note: Bold indicates the most a ffected zone in a tooth type.


58 051015202530 3.0-3.5 3.5-4.0 4.0-4.5 4.5-5.0 5.0-5.5 5.5-6.0 6.0-6.5 6.5-7.0 7.0-7.5Develpmental ageNumber of individuals UM2 05101520253 0 0.0-0.5 0.5-1.0 1.0-1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5Developmental ageNumber of individuals UM1 051015202530 2.5-3.0 3.0-3.5 3.5-4.0 4.0-4.5 4.5-5.0 5.0-5.5 5.5-6.0Developmental ageNumber of individuals UP4 051015202530 2.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0 4.0-4.5 4.5-5.0 5.0-5.5 5.5-6.0Developmental ageNumber of individuals UP3 Figure 4-12. Developmental zones of LEH formation among toot h types using modified met hod of Goodman et al. (1980).


59 05101520253 0 0.0-0.5 0.5-1.0 1.0-1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0 4.0-4.5 4.5-5.0 5.0-5.5 5.5-6.0Developmental ageNumber of individuals UC 05101520253 0 1.0-1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0 4.0-4.5Developmental ageNumber of individuals UI2 051015202530 0.0-0.5 0.5-1.0 1.0-1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0 4.0-4.5Developmental ageNumber of individuals UI1 Figure 4-12. Developmental zones of LEH formation among tooth t ypes using modified method of G oodman et al. (1980). Continued.


60 051015202530 3.0-3.5 3.5-4.0 4.0-4.5 4.5-5.0 5.0-5.5 5.5-6.0 6.0-6.5 6.5-7.0Developmental ageNumber of individuals LM2 05101520253 0 0.0-0.5 0.5-1.0 1.0-1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5Developmental ageNumber of individuals LM1 0510152025302.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0 4.0-4.5 4.5-5.0 5.0-5.5 5.5-6.0 6.0-6.5 6.5-7.0Developmental ageNumber of individuals LP4 0510152025301.0-1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0 4.0-4.5 4.5-5.0 5.0-5.5 5.5-6.0Developmental ageNumber of individuals LP3 Figure 4-12. Developmental zones of LEH formation among tooth t ypes using modified method of G oodman et al. (1980). Continued.


61 0510152025300.5-1.0 1.0-1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0 4.0-4.5 4.5-5.0 5.0-5.5 5.5-6.0 6.0-6.5Developmental ageNumber of individuals LC 05101520253 0 0.0-0.5 0.5-1.0 1.0-1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0Developmental ageNumber of individuals LI2 051015202530 0.0-0.5 0.5-1.0 1.0-1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0Developmental ageNumber of individuals LI1 Figure 4-12. Developmental zones of LEH formation among tooth t ypes using modified method of G oodman et al. (1980). Continued.


62 All anterior tooth types, with the exception of the maxillary lateral incisor, show peak LEH formation at the seventh zone, 2.9 to 3.8 years on maxillary dentition and 2.3 to 4.2 years on mandibular dentition. Most SSH i ndividuals exhibit LEH episodes on their anterior dentition between the age of 2.0 to 5.0 years. Figure 4-15 is a graphical demonstration of LEH formation time by toot h types following Reid and DeanÂ’s method. Sex comparisons of LEH formation time Method one: Goodman et al. (1980). Figures 4-16 and 4-17 show age patterns of LEH formation in males and females calcu lated by regression formulae. Peak LEH formation times vary by tooth type for male maxillary dentition, wh ile two clusters of peak timing are distinguished at 2.5-3.0 and 3.54.0 years. For male mandibular dentition, age ranges of 2.0-2.5 and 4.0-4.5 years for peak LEH formation are observed. In females, the maxillary dentition exhibits peak LEH formation between 3.0 to 4.0 years, and the mandibular dentition shows 2.5-3.0 and 4.5-5.0 year peaks. Figure 4-18 presents clear sex differences with respect to LEH formation time by tooth type. Peak LEH formation zones in five female anterior tooth types appear 0.5 to 1.0 year later than in males. Female indi viduals show LEH pres ent between 2.0 to 4.0 years, while on mandibular canines LEH ar e observed between 2.0 to 6.0 years. Male individuals show slightly earlier LEH forma tion peaks than females, ranging from 1.5 to 3.5 years. Maxillary and mandibular canin es record peak LEH formation time much later in both sexes than incisors do. Males exhibit a slightly later LEH peak on the maxillary canines at 3.5 to 4.0 years, while females form most of the enamel defects 0.5 years before males.


63 0 5 10 15 20 25 30 35 40 1st 1.7-1.9 1.8-2.0 1.1-1.3 2nd 1.9-2.2 2.0-2.2 1.3-1.6 3rd 2.2-2.4 2.2-2.4 1.6-1.8 4th 2.4-2.7 2.4-2.7 1.8-2.0 5th 2.7-3.0 2.7-2.9 2.0-2.4 6th 3.0-3.4 2.9-3.3 2.4-2.9 7th 3.4-3.8 3.3-3.7 2.9-3.4 8th 3.8-4.3 3.7-4.1 3.4-3.9 9th 4.3-4.8 4.1-4.6 3.9-4.4 10th 4.8-5.3 4.6-5.1 4.4-5.0 Developmental ageNumber of individuals UC UI2 UI1 Figure 4-13. Age patterns by number of individuals of LEH formation among maxillary anterior tooth types using modified charts from Reid and Dean (2000). 0 5 10 15 20 25 30 35 40 1st 1.5-1.7 1.0-1.1 1.0-1.1 2nd 1.7-2.0 1.1-1.3 1.1-1.3 3rd 2.0-2.3 1.3-1.5 1.3-1.5 4th 2.3-2.7 1.5-1.8 1.5-1.7 5th 2.7-3.1 1.8-2.1 1.7-2.0 6th 3.1-3.6 2.1-2.4 2.0-2.3 7th 3.6-4.2 2.4-2.8 2.3-2.6 8th 4.2-4.9 2.8-3.3 2.6-3.0 9th 4.9-5.6 3.3-3.7 3.0-3.4 10th 5.6-6.2 3.7-4.2 3.4-3.8 Developmental ageNumber of individuals LC LI2 LI1 Figure 4-14. Age patterns by number of indivi duals of LEH formation among mandibular anterior tooth types using modified charts from Reid and Dean (2000).


64 0 0 2 6 15 26 13 2 0 150510152025303540 1.7-1.9 1.9-2.2 2.2-2.4 2.4-2.7 2.7-3.0 3.0-3.4 3.4-3.8 3.8-4.3 4.3-4.8 4.8-5.3Developmental ageNumber of individuals UC 0 0 2 2 5 13 8 15 6 105101520253035401.8-2.0 2.0-2.2 2.2-2.4 2.4-2.7 2.7-2.9 2.9-3.3 3.3-3.7 3.7-4.1 4.1-4.6 4.6-5.1Developmental ageNumber of individuals UI2 0 0 0 1 5 10 17 11 6 005101520253035401.1-1.3 1.3-1.6 1.6-1.8 1.8-2.0 2.0-2.4 2.4-2.9 2.9-3.4 3.4-3.9 3.9-4.4 4.4-5.0Developmental ageNumber of individuals UI1 Figure 4-15. Developmental zones of LEH form ation of each tooth type using modified charts from Reid and Dean (2000). Counted by number of individuals.


65 0 0 2 8 11 12 36 24 8 2051015202530354 0 1.5-1.7 1.7-2.0 2.0-2.3 2.3-2.7 2.7-3.1 3.1-3.6 3.6-4.2 4.2-4.9 4.9-5.6 5.6-6.2Developmental ageNumber of individuals LC 0 0 0 0 0 7 11 5 2 005101520253035401.0-1.1 1.1-1.3 1.3-1.5 1.5-1.8 1.8-2.1 2.1-2.4 2.4-2.8 2.8-3.3 3.3-3.7 3.7-4.2Developmental ageNumber of individuals LI2 0 0 0 0 0 2 5 0 0 00510152025303540 1.0-1.1 1.1-1.3 1.3-1.5 1.5-1.7 1.7-2.0 2.0-2.3 2.3-2.6 2.6-3.0 3.0-3.4 3.4-3.8Developmental ageNumber of individuals LI1 Figure 4-15. Developmental zones of LEH form ation of each tooth type using modified charts from Reid and Dean (2000). Counted by number of individuals. Continued.


66 0 2 4 6 8 10 12 140-0.50.5-1.01.0-1.51.5-2.02.0-2.52.5-3.03.0-3.53.5-4.04.0-4.54.5-5.05.0-5.55.5-6.06.0-6.5Developmental ageNumber of LEHs UC UI2 UI1 0 2 4 6 8 10 12 140-0.50.5-1.01.0-1.51.5-2.02.0-2.52.5-3.03.0-3.53.5-4.04.0-4.54.5-5.05.0-5.55.5-6.06.0-6.5Developmental ageNumber of LEHs LC LI2 LI1 Figure 4-16. Age patterns of LEH on male anterior dentiti on using modified method of Goodman et al. (1980).


67 0 2 4 6 8 10 12 140-0.50.5-1.01.0-1.51.5-2.02.0-2.52.5-3.03.0-3.53.5-4.04.0-4.54.5-5.05.0-5.55.5-6.06.0-6.5Developmental ageNumber of LEHs UC UI2 UI1 0 5 10 15 20 250-0.50.5-1.01.0-1.51.5-2.02.0-2.52.5-3.03.0-3.53.5-4.04.0-4.54.5-5.05.0-5.55.5-6.06.0-6.5Developmental ageNumber of LEHs LC LI2 LI1 Figure 4-17. Age patterns of LEH on female an terior dentition using modified method of Goodman et al. (1980). In the case of mandibular canines, females ha ve an LEH formation peak between 4.5 to 5.0 years and male peaks appear 0.5 years earl ier than females. Overall, based on the regression formula modified fr om Goodman et al. (1980), a tr end can be detected that


68 peak LEH formation periods on anterior de ntition is 0.5 years earlier in males than females. Method two: Reid and Dean (2000). Figures 4-19 and 4-20 are age distributions of LEH formation zones of male and female dentition derived from modified charts of Reid and Dean (2000). Male maxillary dentit ion shows peak LEH formation at the 7th zone, between 2.9 to 3.8 years. Most LEH are formed between 2.0 to 4.3 years. The LEH formation peak is at the 7th (3.6-4.2 year s) and 6th (2.1-2.4 year s) zone of male mandibular canines and lateral incisors, resp ectively. Female anterior dentition shows a similar pattern (Figure 4-20). Peak LEH forma tion zones occur in the 7th and 8th zones, whose developmental ages range from 2.94.3 years on maxillary teeth, and 2.3-4.9 years on mandibular teeth. Most LEH are ob served between 1.7 to 5.6 years. Both males and females tend to have enam el defects on the mid-third of the tooth crown. Zones towards the cemento-enamel junction are more prone to LEH formation than the occlusal end of the crown. The 7th z one is the region where most enamel defects aggregate. Sex comparisons of LEH formation time on each anterior tooth type are shown in Figure 4-21. Peak LEH formation zones ove rlap for male and female tooth crowns. Age patterns of LEH formation on maxillary dentition are similar in both sexes, where peak LEH formation time falls on the same deve lopmental zone of a particular tooth type. In terms of mandibular dentition, females show later peak LEH formation than males for canines and lateral incisors, with the differe nce ranging from 0.5 to 1 year in duration. Porotic Hyperostosis Results of porotic hyperostosis are pres ented by individual counts. The observed sample reflects individuals with at least one side of orbital ro of preserved and at least one side of the parietal bones available for observation.


69 0 5 10 15 20 250-0.50.5-1.01.0-1.51.5-2.02.0-2.52.5-3.03.0-3.53.5-4.04.0-4.54.5-5.05.0-5.55.5-6.06.0-6.5Develo p mental a g eNumber of LEHs Male Female 0 5 10 15 20 250-0.50.5-1.01.0-1.51.5-2.02.0-2.52.5-3.03.0-3. 53.5-4.04.0-4.54.5-5.05.0-5.55.5-6.06.0-6.5Developmental ageNumber of LEHs Male Female 0 5 10 15 20 250-0.50.5-1.01.0-1.51.5-2.02.0-2.52.5-3.03.0-3.53.5-4.04.0-4.54.5-5.05.0-5.55.5-6.06.0-6.5Developmental ageNumber of LEHs Male Female Figure 4-18. Sex comparison on age patterns of LEH by tooth types using modified method of Goodman et al. (1980). UC UI2 UI1


70 0 5 10 15 20 250-0.50.5-1.01.0-1.51.5-2.02.0-2.52.5-3.03.0-3. 53.5-4.04.0-4.54.5-5.05.0-5.55.5-6.06.0-6.5Develo p mental a g eNumber of LEHs Male Female 0 5 10 15 20 250-0.50.5-1.01.0-1.51.5-2.02.0-2.52.5-3.03.0-3. 53.5-4.04.0-4.54.5-5.05.0-5.55.5-6.06.0-6.5Develo p mental a g eNumber of LEHs Male Female 0 5 10 15 20 250-0.50.5-1.01.0-1.51.5-2.02.0-2.52.5-3.03.0-3.53.5-4.04.0-4.54.5-5.05.0-5.55.5-6.06.0-6.5Developmental ageNumber of LEHs Male Female Figure 4-18. Sex comparison on age patterns of LEH by tooth types using modified method of Goodman et al. (1980). Continued. LC LI1 LI2


71 If an individual has only part of the occipital bon e present, the indivi dual is not counted in the vault sample (Wright, 1994). In the same vein, the prevalence of individuals affected with porotic hyperostosis on the or bital roof and/or vau lt is calculated using individuals whose orbital roof and parietal bones are presen t. Finally, when the overall prevalence of porotic hyperostos is is considered regardless of lesion location, individuals are eligible when either one side of the or bital roof or one side of the parietal is observable. Table 4-17 reveals that 126 out of 306 ( 42%) identified individuals from the SSH site have at least one orbital roof or one parietal bone pr eserved. Twenty-three percent of the 126 individuals are affected with porotic hyperostosis on either the vault bones, the orbital roof, or both. One-third of 81 i ndividuals show orb ital lesions, although appearance of cranial po rotic hyperostosis is rarely seen. Individuals affected with porotic hyperostosis on both the vault and the orbital roof are rare. When the prevalence of porotic hyperost osis is considered by sex, males and females are similarly affected (23% and 24%, respectively). Ther e is no statistical significance on the prevalence of cribra orbitalia among the se xes. Since cranial porotic hyperostosis in the SSH population is rare, it is not surprising to find that prevalence between the sexes is not substantial. Males in general have slightly more orbital lesions, and female cranial bones are affected more often. Porotic hyperostosis activity is an in dicator of the progression of anemic conditions.


72 0 5 10 15 20 25 1st 1.7-1.9 1.8-2.0 1.1-1.3 2nd 1.9-2.2 2.0-2.2 1.3-1.6 3rd 2.2-2.4 2.2-2.4 1.6-1.8 4th 2.4-2.7 2.4-2.7 1.8-2.0 5th 2.7-3.0 2.7-2.9 2.0-2.4 6th 3.0-3.4 2.9-3.3 2.4-2.9 7th 3.4-3.8 3.3-3.7 2.9-3.4 8th 3.8-4.3 3.7-4.1 3.4-3.9 9th 4.3-4.8 4.1-4.6 3.9-4.4 10th 4.8-5.3 4.6-5.1 4.4-5.0 Developmental ageNumber of LEHs UC UI2 UI1 0 5 10 15 20 25 1st 1.5-1.7 1.0-1.1 1.0-1.1 2nd 1.7-2.0 1.1-1.3 1.1-1.3 3rd 2.0-2.3 1.3-1.5 1.3-1.5 4th 2.3-2.7 1.5-1.8 1.5-1.7 5th 2.7-3.1 1.8-2.1 1.7-2.0 6th 3.1-3.6 2.1-2.4 2.0-2.3 7th 3.6-4.2 2.4-2.8 2.3-2.6 8th 4.2-4.9 2.8-3.3 2.6-3.0 9th 4.9-5.6 3.3-3.7 3.0-3.4 10th 5.6-6.2 3.7-4.2 3.4-3.8 Developmental ageNumber of LEHs LC LI2 LI1 Figure 4-19. Age patterns of LEH on male anterior dentition using modified charts of Reid and Dean (2000).


73 0 5 10 15 20 25 1st 1.7-1.9 1.8-2.0 1.1-1.3 2nd 1.9-2.2 2.0-2.2 1.3-1.6 3rd 2.2-2.4 2.2-2.4 1.6-1.8 4th 2.4-2.7 2.4-2.7 1.8-2.0 5th 2.7-3.0 2.7-2.9 2.0-2.4 6th 3.0-3.4 2.9-3.3 2.4-2.9 7th 3.4-3.8 3.3-3.7 2.9-3.4 8th 3.8-4.3 3.7-4.1 3.4-3.9 9th 4.3-4.8 4.1-4.6 3.9-4.4 10th 4.8-5.3 4.6-5.1 4.4-5.0 Developmental ageNumber of LEHs UC UI2 UI1 0 5 10 15 20 25 1st 1.5-1.7 1.0-1.1 1.0-1.1 2nd 1.7-2.0 1.1-1.3 1.1-1.3 3rd 2.0-2.3 1.3-1.5 1.3-1.5 4th 2.3-2.7 1.5-1.8 1.5-1.7 5th 2.7-3.1 1.8-2.1 1.7-2.0 6th 3.1-3.6 2.1-2.4 2.0-2.3 7th 3.6-4.2 2.4-2.8 2.3-2.6 8th 4.2-4.9 2.8-3.3 2.6-3.0 9th 4.9-5.6 3.3-3.7 3.0-3.4 10th 5.6-6.2 3.7-4.2 3.4-3.8 Developmental ageNumber of LEHs LC LI2 LI1 Figure 4-20. Age patterns of LEH on female an terior dentition using modified charts of Reid and Dean (2000).


74 0 5 10 15 20 251.7-1.9 1.9-2.22.2-2.4 2.4-2.7 2.7-3.03.0-3.4 3.4-3.83.8-4.34.3-4.84.8-5.3Developmental ageNumber of LEHs Male Female 0 5 10 15 20 251.8-2.02.0-2.22.2-2.42.4-2.72.7-2.92.9-3.33.3-3.73.7-4.14.1-4.64.6-5.1Developmental ageNumber of LEHs Male Female 0 5 10 15 20 251.1-1.31.3-1.61.6-1.81.8-2.02.0-2.42.4-2.92.9-3.43.4-3.93.9-4.44.4-5.0Developmental ageNumber of LEHs Male Female Figure 4-21. Sex comparison on age patterns of LEH by tooth types using modified charts of Reid and Dean (2000). UC UI2 UI1


75 0 5 10 15 20 251.5-1.71.7-2.02.0-2.3 2.3-2.7 2.7-3.13.1-3.6 3.6-4.24.2-4.94.9-5.65.6-6.2Developmental ageNumber of LEHs Male Female 0 5 10 15 20 251.0-1.11.1-1.31.3-1.51.5-1.81.8-2.12.1-2.42.4-2.82.8-3.33.3-3.73.7-4.2Developmental ageNumber of LEHs Male Female 0 5 10 15 20 251.0-1.11.1-1.31.3-1.51.5-1.71.7-2.02.0-2.32.3-2.62.6-3.03.0-3.43.4-3.8Developmental ageNumber of LEHs Male Female Figure 4-21. Sex comparison on age patterns of LEH by tooth types using modified charts of Reid and Dean (2000). Continued. LI1 LI2 LC


76 A healed bony lesion suggests that an individu al overcame the anemic events and lived a period of anemia-free life before their death. Th e affected areas are therefore able to be remodeled and healed by osteoblast activity and bone maintenance and repair. As opposed to healed lesions, unheal ed lesions are indicative of active or progressive anemic episodes. Individuals with act ive lesions observable are interpreted as having suffered from anemic conditions at the time of death perimortem. It should also be noted that anemia is not always the cause of death for an individual. Table 4-18 presents differences by sex of the activity of porotic hyperostosis on the orbital region. Among individuals affected (N=19) whose sex can be determined, 10 have active lesions. The number of female individu als (N=8) is four times higher than male individuals (N=2). Males are more likely to be observed with healed lesions. Approximately 90% of females exhibit activ e orbital porotic hype rostosis, while only 20% of males exhibit active orbital lesions. The differen ce of prevalence between the sexes is statistically significant. For cranial porotic hyperostosis, all individuals affected (N=7) show healed bony lesions. Table 4-17. Distribution of porotic hype rostosis according to sex and location Males Females All FET A/Oa % A/O % A/O % Probability PH on orbit 10/32 31.3 9/34 26.5 26/81 32.1 .787 PH on vault 2/45 4.4 5/44 11.4 7/119 5.9 .266 Orbit & vault 1/30 3.3 3/33 9.1 4/76 5.3 .614 b Location pooled 11/47 23.4 11/45 24.4 29/126 23.0 1.000 Note: a. A/O= Affected/Observed; b. Two ce lls have expected counts less than 5. All chi-square analyses were perfor med at 1 degree of freedom. Prevalence of porotic hype rostosis among age groups is presented in Table 4-19. Almost all lesions are observed on subadults and young adults. Middle and old adults are rarely affected. Only two i ndividuals in the older age gr oups have porotic hyperostosis.


77 Table 4-18. Distribution of orbital lesions by sex and healing stages Sex Female Male Total Active 8 2 10 Activity Healed 1 8 9 Total 9 10 19 % Active 88.9 20.0 47.4 Note: df= 1, FET p= .005. Expected count s in most cells are less than 5. Subadults have a slightly higher incidence (47%) of orbital lesions than young adults (44%). Porotic hyperostosis, in contrast, is found more fre quently on cranial bones of young adults than in subadults. In fact, no s ubadults are observed with cranial lesions. Overall, the prevalence of porotic hyperostos is in young adults (35%) is 13% higher than that of the subadults (22%). Table 4-19. Distribution of porotic hyperost osis according to age groups and location by number of individuals SA YA MA OA Chi-square A/Oa % A/O % A/O % A/O % X2 p PH on orbit 8/17 47.1 17/39 43.6 1/20 5.0 0/4 0.0 12.649 .005 b PH on vault 0/32 0.0 6/47 12.8 1/26 3.8 0/4 0.0 5.183 .159 b Orbit & vault 0/15 0.0 4/37 10.1 0/19 0.0 0/4 0.0 ----Location pooled 8/36 22.2 19/55 34.9 2/28 7.1 0/4 0.0 9.140 .027 b Note: a. A/O= Affected/Observed; b. One or mo re cells have expected counts less than 5. All chi-square analyses were pe rformed at 3 degree of freedom. To determine the significance of LEH preval ence between subadults and adults, all skeletally mature individuals that satisfy th e selection criteria fo r porotic hyperostosis observation are combined (Table 4-20). Adu lts have a higher prevalence (31%) of the lesion than subadults (22%), but this trend is not statistically signifi cant. Subadults have a higher (47%) incidence of crib ra orbitalia than do adu lts (40%), although is not significant. Among the affected subadult indivi duals, 63% (N=5) have at least one active orbital lesion. In the case of adults, 56% (N =8) of the affected i ndividuals have active


78 lesions. There is no significant statistical difference between activity and maturity (Tables 4-21). Table 4-20. Distribution of porotic hyperostos is according to maturity and location by number of individuals Subadult Adult All FET A/Oa % A/O % A/O % Probability PH on orbit 8/17 47.1 18/46 39.1 26/81 32.1 .154 PH on vault 0/32 0.0 7/79 8.9 7/111 6.3 .187b Orbits & vault 0/15 0.0 4/61 6.6 4/76 5.3 --Location pooled 8/36 22.2 21/68 30.9 29/125 23.2 1.000 Note: a. A/O= Affected/Observed; b. One cell has expected counts less than 5. All chi-square analyses were perf ormed at 1 degree of freedom. Table 4-21. Distribution of orb ital lesions by maturity and he aling stage by number of individuals Maturity Subadult Adult Total Activity Active 5 10 15 Healed 3 8 11 Total 8 18 26 % Active 62.5 55.6 57.7 Note: df= 1, FET p= 1.000. Comobidity of LEH and Porotic Hyperostosis Although the etiologies of LEH and porotic hyperostosis ar e different, they are well known as general stress indicators for a population. Prevalence of LEH measures non-specific stress during childhood, while th e presence of poroti c hyperostosis is a marker of anemic events. If a positive relati onship between prevalence of the two lesions is found, it seems reasonable to argue that individuals with enamel defects have been “weakened” by childhood stress episodes and ar e less effective in coping with anemic conditions. Skeletal lesions of porotic hyperostosis are therefore manifested. To examine this argument, individuals w ith at least one tooth and at least one portion of parietal or orbital areas preserved are included in the correlation test. Table


79 4-22 shows the individual counts of both le sions and the statistics. Among 113 eligible individuals, about one-fifth of them have both lesions, while more than half (52%) of them have either lesion present. Pearson co rrelation suggests no relationship between the appearance of LEH and porotic hyperostos is. While low prev alence of porotic hyperostosis in the SSH population could have an effect on the test, the tendency that individuals affected with LEH are more likely to have porotic hyperostosis does not hold in the SSH population. Table 4-22. Correlation between the occurr ence of LEH and porotic hyperostosis by number of individuals PH present PH absent Total N % N % N % LEH present 21 18.6 52 46.0 73 64.6 LEH absent 7 6.2 33 29.2 40 35.4 Total 28 24.8 85 75.2 113 100.0 Pearson R2= .125 p= .188 Note: PH= porotic hyperostosis. Dental Pathology Dental Caries Dental caries is a pathology that has been widely used in investigating problems related to subsistence, general health and oral behavior . Caries rate in this study is assessed as observed caries rate. The “correct ed caries rate” met hod proposed by Lukacs (1995) is not used due to extremely low inci dence of caries-induced pulp cavity exposure (1 out of 41 teeth), and low ante mortem tooth loss (47 teeth). Among the deciduous dentition, six teeth are carious, mostly located on the posterior teeth (Table 4-23). The affected teeth belong to 3 individuals. Ov erall caries rate of the deciduous teeth is 2.4% (6/249). Observed caries rates of each permanent tooth type are presented in Table 4-24. Caries rate of permanent maxillary dentition is 0.5% (7/1452) and 0.9% (11/1226) on


80 mandibular dentition. Overall caries rate of the SSH permanent teeth is 0.7% (18/2678). The carious teeth belong to 11 individuals. Caries lesions are mostly located on the occlusal surfaces and buccal side of the toot h crowns. All caries appear on the posterior dentition, where molars are the most affected tooth types. Dental Calculus Dental calculus is a good indicator of food composition and oral hygiene of a population. Only one deciduous lower first mola r has calculus present. Prevalence of moderate to severe calculus accumulation on permanent teeth is shown in Table 4-25. Calculus appears more frequently on the an terior teeth. About 22% to 34% of anterior dentition is affected with moderate to seve re dental calculus. Mo lars, especially third molars, are the least affected tooth types. A clear trend is observed on mandibular dentition that the prevalence decreases wh en one moves from incisors to molars. Maxillary dentition is more likely to have calculus pr esent (26.2%, 380/1452) than mandibular dentition (17.7%, 251/1419). Overall prev alence of dental calculus of the site is 22.2% (631/2871). Dental Abscessing An abscessed tooth can result from various causes, including severe periodontal disease, dental caries, infec tion and exposed pulp cavity. With the alveoli being affected and eroded below normal height of the alveolar process, absce ssed teeth are likely to be lost antemortem. Severe dental abscessing, if left untreated, could result in soft tissue infection and may ultimately lead to deat h. Prevalence of dental abscessing of a population is therefore indicative of its di stribution of severe and long-term oral pathologies.


81 No abscessing is found on deciduous teeth. Table 4-26 presents prevalence across tooth types of permanent dentition. Only five teeth are affected w ith abscessing. Overall prevalence is extremely low (0.2%, 5/2871). Ab scessing is located on the buccal side of the alveolar process.


82Table 4-23. Prevalence of dental caries on deciduous dentit ion by tooth count Maxillary dentition m2 m1 c i2 i1 CAR OBS % CAR OBS % CAR OBS % CAR OBS % CAR OBS % R 0 22 0 1 20 5.0 0 14 0 0 2 0 1 5 20.0 L 1 21 4.8 0 19 0 0 11 0 0 4 0 0 6 0 Pooled 1 43 2.3 1 39 0 0 25 0 0 6 0 1 11 9.1 Mandibular dentition m2 m1 c i2 i1 CAR OBS % CAR OBS % CAR OBS % CAR OBS % CAR OBS % R 1 27 3.7 1 23 4.3 0 12 0 0 4 0 0 4 0 L 1 21 4.8 0 19 0 0 9 0 0 3 0 0 3 0 Pooled 2 48 4.2 1 42 2.4 0 21 0 0 7 0 0 7 0 Note: CAR= teeth affected with ca rious lesion; OBS= observed teeth.


83Table 4-24. Prevalence of dental caries on permanent dentition by tooth count Maxillary dentition M3 M2 M1 P4 P3 C I2 I1 CAR OBS % CAR OBS % CAR OBS % CAR OBS % CAR OBS % CAR OBS % CAR OBS % CAR OBS % R 2 66 3.0 2 106 1.9 1 117 0.9 1 97 1.0 1 99 1.0 0 99 0 0 77 0 0 63 0 L 2 55 3.6 2 105 1.9 0 111 0 0 96 0 0 94 0 0 91 0 0 77 0 0 66 0 Pooled 4 121 3.3 4 211 1.9 1 228 0.4 1 193 0.5 1 193 0.5 0 190 0 0 154 0 0 129 0 Mandibular dentition M3 M2 M1 P4 P3 C I2 I1 CAR OBS % CAR OBS % CAR OBS % CAR OBS % CAR OBS % CAR OBS % CAR OBS % CAR OBS % R 2 66 3.0 2 106 1.9 1 117 0.9 1 97 1.0 1 99 1.0 0 99 0 0 77 0 0 63 0 L 2 55 3.6 2 105 1.9 0 111 0 0 96 0 0 94 0 0 91 0 0 77 0 0 66 0 Pooled 4 121 3.3 4 211 1.9 1 228 0.4 1 193 0.5 1 193 0.5 0 190 0 0 154 0 0 129 0


84Table 4-25. Prevalence of dental calcul us on permanent dentition by tooth count Maxillary dentition M3 M2 M1 P4 P3 C I2 I1 CAL OBS % CAL OBS % CAL OBS % CAL OBS % CAL OBS % CAL OBS % CAL OBS % CAL OBS % R 4 65 6.2 27 98 27.6 36 108 33.3 28 98 28.6 31 97 32.0 29 95 30.5 24 88 27.3 22 83 26.5 L 6 61 9.8 27 101 26.7 40 108 37.0 59 92 31.1 28 92 30.4 34 92 37.0 20 86 23.3 21 88 23.9 Pooled 10 126 7.9 54 199 27.1 76 216 35.2 31 190 32.0 59 189 31.2 63 187 33.7 44 174 25.3 43 171 25.2 Mandibular dentition M3 M2 M1 P4 P3 C I2 I1 CAL OBS % CAL OBS % CAL OBS % CAL OBS % CAL OBS % CAL OBS % CAL OBS % CAL OBS % R 8 66 12.1 10 106 9.4 13 117 11.1 13 97 13.4 12 99 12.1 21 99 21.2 21 77 27.3 21 63 33.3 L 6 55 10.9 13 105 12.4 17 111 15.3 18 96 18.8 18 94 19.2 21 91 23.1 19 77 24.7 20 66 30.3 Pooled 14 121 11.6 23 211 10.9 30 228 13.2 31 193 16.1 30 193 15.5 42 190 22.1 40 154 26.0 41 129 31.8 Note: CAL= teeth affected with moderate to severe dental calculus; OBS= observed teeth.


85 Table 4-26. Prevalence of de ntal abscessing on permanent dentition by tooth count Maxillary dentition M3 M2 M1 P4 P3 C I2 I1 ABS OBS % ABS OBS % ABS OBS % ABS OBS % ABS OBS % ABS OBS % ABS OBS % ABS OBS % R 0 65 0 0 98 0 0 108 0 0 98 0 0 97 0 0 95 0 0 88 0 0 83 0 L 0 61 0 0 101 0 2 108 1.9 0 92 0 0 92 0 0 92 0 0 86 0 0 88 0 Pooled 0 126 0 0 199 0 2 216 0.9 0 190 0 0 189 0 0 187 0 0 174 0 0 171 0 Mandibular dentition M3 M2 M1 P4 P3 C I2 I1 ABS OBS % ABS OBS % ABS OBS % ABS OBS % ABS OBS % ABS OBS % ABS OBS % ABS OBS % R 0 66 0 1 106 0.9 0 117 0 0 97 0 0 99 0 0 99 0 0 77 0 0 63 0 L 0 55 0 0 105 0 0 111 0 1 96 1.0 0 94 0 1 91 1.1 0 77 0 0 66 0 Pooled 0 121 0 1 211 0.5 0 228 0 1 193 0.5 0 193 0 1 190 0.5 0 154 0 0 129 0 Note: ABS= teeth affected with dent al abscessing; OBS= observed teeth.

PAGE 100

86 CHAPTER 5 DISCUSSION To reconstruct the SSH diet and facilit ate interpretation of LEH and porotic hyperostosis patterning in bioc ultural context, prevalence of dental pathology and its implication on SSH subsistence is first di scussed below. Prenat al and childhood stress patterns, based on LEH prevalence and chronolog ical patterns of LEH formation, are then examined. The relationship of LEH incide nce (childhood morbidity) and mortality is reviewed and the prevalence a nd distribution of porotic hypero stosis is considered with respect to anemia and its potential etio logy. Lastly, overall childhood morbidity is assessed and a general pattern of SSH childhood stress is presented. Dental Pathology and SSH Diet The structures of the mouth and the teeth are the primary recipients and processors of all food consumed. As such, dental pathology is one of the best tools for biological anthropologists to assess eating habits of an individual, a nd the dietary patterns of a population. Three dental patholog ies are assessed in this st udy; dental caries, dental calculus and abscessing. It widely accepted, es pecially in New Worl d studies, that the occurrence of dental caries in creases with the intensifica tion of agriculture (Cohen and Armelagos, 1984; Larsen, 1997). Agricultural pr oducts, particularly maize, are more cariogenic due to their high carbohydrate and su gar content. In a widely cited paper, Turner (1979) collected a se ries of population caries rate s among societies with various subsistence strategies. The data exhibit a cl ear trend that hunter-g atherers have lower caries rates by tooth count (0 to 4.6%), while people in agricultural societies have higher

PAGE 101

87 caries rates (2.3 to 26.9%). Cari es rates for people with a subsistence mixed economy fall in between the previously mentioned rate s. Dental calculus is the accumulation and calcification of food residues and is related to the consistenc y of food consumed. Soft and sticky food tends to get caught around the gum line, while coarse foods are more effective at preventing food remains fr om accumulating. Although not strictly a pathology, prevalence of dental calculus is a ffected by oral pH. Calcification of food residues is more likely to occur in a higher pH, alkaline, environment. Dental abscessing is a severe periodontal pathology caused by various etiologies and could lead to antemortem tooth loss and/or systemic in fection. These acquired dental conditions are commonly used in bioarchaeological studies when assessing overall oral health for a population (Littleton and Frohlic h, 1993; Manzi et al., 1999). The caries rate in the SSH population is extremely low, only 0.7% of teeth and 11 adult individuals were affected with caries. One case out of 41 teeth showed an exposed pulp cavity was caused by caries. The caries rate observed on the deciduous teeth is 2.4% and 3 children were affected. This observati on is comparable with results presented in Pietrusewsky and Tsang (2003), however is lowe r than results presen ted in Chang (1993). Twenty two percent of permanent teeth are observed with moderate to severe dental calculus, a value that is intermediate between the findings of previous studies on the SSH remains. As noted by both aforementioned studies, dental abscessing is a rare phenomenon. Only 0.4% of teeth are affected as determined from results presented in this study. The dietary patterns of SSH people can be inferred from these findings. An extremely low caries rate suggests they reli ed mostly on food that was hunted and/or

PAGE 102

88 gathered from natural sources. Site location and recovered faunal remains at the SSH site indicate a subsistence pattern dependent on marine resources . Marine foodstuffs were effective resources to exploit in the warm, tropical coastal and estuarine setting around SSH site. Recovered remains of a wide range of marine faunal spec ies and the presence of abundant shell mounds suppor t such an environment. In terms of cultigens, only rice grains we re preserved at SSH site although clear signs of domestication and abundance remain unclear. Although a nearby site has yielded clear evidence of rice agriculture about 2000 years before SSH site (Wang, 1984), there is no safe ground to assume that the SSH people were intentionally cultivating rice for consumption. Tsang (2001) suggests that rice grains found at the SSH site were probably used for making beer rather than for food consumption, which is a very common practice among Taiwanese Aborigines. Therefore, even if rice was consumed by the SSH people, it does not seem to have been a staple crop, particularly when one considers the evidence for marine-oriented site locati on and the lack of direct ev idence for rice agriculture. Tayles et al. (2000) find rela tively low caries rates in pr ehistoric agri cultural sites from Thailand and propose that rice itself ha s low intrinsic cariogenicity. If crudely processed, rice is coarse and thus decreases the chance for cariogenic conditions to form. Fine polishing of rice grains is a very recen t practice in Southeast Asia (FAO, 1954 cited in Tayles et al., 2000). If this holds true in Taiwan prehistory, consumption of coarse rice of the SSH people helps in removing potential carious locations on teeth. Combined with low caries and abscessing rates, coastal site location and the lack of rice cultivation, it appears that SSH people subsisted mostly on hunted and gathered terrestrial foods, and collection of marine food resources.

PAGE 103

89 Prevalence of moderate to severe dental ca lculus (22%) suggests a rather alkaline oral environment, as related by Pietruse wsky and Tsang (2003). Betel nut chewing, another habitual custom of th e aboriginal people in Taiwa n, may also account for this phenomenon. In addition, certain amounts of s ticky foodstuffs, such as taro and yam, may also have been consumed. Such root plants are common in Taiwan and do not require complicated techniques or intense labor to grow. It is likely that the SSH people consumed roots as a major carbohydrate source. Furthermore, high oral pH and accumulation of calculus are effective ways of preventing a hospitable environment for cariogenesis. In sum, Iron Age SSH was basically a hunt er-gatherer society with terrestrial, marine and estuarine resources all playing a significant role to thei r broad spectrum diet. Rice and other terrestrial plant resources we re also mixed in the diet. Limited rice consumption and dependence on marine foods allowed the inhabitants to enjoy a balanced diet and moderate to good dental health. Enamel Defects and Stress LEH Distribution among Tooth Types As Goodman and colleagues suggest (G oodman and Rose, 1990; Goodman et al., 1984), and other researchers acknowledge (e.g., Lukacs et al., 2001; Wright, 1997), LEH is most likely to appear on maxillary and mandibular canines, followed by maxillary incisors. LEH prevalence among tooth types from the SSH site clearly agrees with this trend. The majority of SSH canines observed were affected by LEH, and fifty-six percent of mandibular canines, the most affected to oth type, show LEH lesions. This proportion is at least 25% greater than incisors and th ree times greater than that observed in the posterior dentition. Goodman and Rose (1990) suggest that anteri or teeth are better

PAGE 104

90 indicators of individual stress history due to their lower thre shold in coping with stress load their heightened sensitivity to record less severe stre ss events. Along this line, 11% of posterior teeth were aff ected by LEH. Posterior teeth, it follows, require stress loads with higher magnitude to discontinue nor mal amelogenesis, which indicates that powerful stress load was present at SSH and affected on physiological well-being. Anterior teeth are not only more likely to be affected with LEH, they also record more LEH episodes than posteri or teeth. Mandibular canines show the highest count of LEH in both observed and affected tooth samples, followed by maxillary canines and incisors. LEH counts of the ante rior dentition as a whole outnu mber that of the posterior dentition as well. This suggests that multiple stress episodes are more frequently recorded in the anterior dentition, and can effectively be used as a subsample for studies utilizing LEH to explore stress-rela ted health and well-bei ng (Goodman et al., 1980). Prenatal Stress Low prevalence of LEH on deciduous teet h (0.8%) at SSH suggests people who died before the age of six did not undergo stressful prenatal an d neonatal periods. The health status of the mothers was supposed ly good and early death should not be blamed on impaired prenatal health. Women are mo re likely to suffer from nutritional stress during pregnancy due to increased nutritiona l demands from both mother and fetus. Good maternal health suggests a well organized dist ribution system of diet ary resources that is valuable to gestating females. SSH women may have benefited from such a system with dependence on marine-based and broad spect rum food resources acquired by hunting and gathering (Hutchinson and La rsen, 1988; Ulijaszek, 1991).

PAGE 105

91 Childhood Stress Overall assessment At SSH, LEH are observed on 34% of anterior dentition, with ove rall prevalence of 18% for all tooth types inspecte d. These findings are comparable with results presented in Pietrusewsky and Tsang (2003), but considerab ly lower than that reported in Chang (1993). When compared to Bronze Age populations in Thailand, overall childhood stress at SSH is slightly higher, but comparable with Ban Lum Khao (3,350-2,450 BP) and Ban Na Di (2,550-2,350 BP). These two populat ions depended heavily on intensive agriculture and marine resources (Domett, 2001). Linear enamel defects in the SSH population is a co mmon pathology as observed and recorded in permanent teeth, suggesti ng that SSH people had a stressful childhood. However, it should be noted that although LEH is a marker left by an episode of physiological disturbance, it is also an i ndicator of recovery. A hypoplastic line (or reduced enamel thickness) is only observed after the ameloblasts resume normal function, at which time the stress load has been lifted. Therefore, individuals with LEH are actually those who survived stress episodes (Pal ubeckaite, 2001). A relatively high LEH prevalence on permanent dentition suggests that morbidity during childhood was high in the SSH population. Although many children we re temporarily ill sometime during childhood, it does not necessarily suggest that children we re frail throughout their childhood. Nonetheless, at least 102 out of 306 individuals in the SSH population had some type of stressful childhood experien ce. Weaning stress (discussed below), infectious disease and/or nutritional imbalance are possible factors th at contributed to a generally harsh childhood.

PAGE 106

92 Among the SSH sample studied, markers of active infectious disease are rarely observed. Only one case of infection, possi bly trauma-related, is encountered among the adult skeletons analyzed. No signs of infec tious disease were found on subadult remains. As suggested by Wood and colleagues (1992), many acute and/or systemic infectious diseases could have claimed an individualÂ’s li fe before that stress marker is manifested skeletally, if indeed a marker is ever to be manifested. Although this osteological paradox may account for the low frequency of infectious disease observed in the SSH sample, it is equally implausible to attribute infecti ous disease as the major cause of LEH. LEH formation and weaning LEH formation has been frequently associ ated with weaning-related stress events (Corruccini et al., 1985; G oodman et al., 1984, 1987). Instead of the timing when complete cessation of breast-feeding takes place, weaning in this study is recognized as a process of introduction of non-milk foods a nd reduction of dependence on breast milk (Katzenberg et al., 1996: 179). Based on this de finition, weaning could start from several months after birth and extend into late childhood, where sp ecific ranges vary greatly among populations. When considered from a biological and immunol ogical perspective based on nonhuman primate patterns, timing of complete cessation of breast-feeding is predicted to range between 2.5 and 7.0 years of age (Dettwyler, 1995). Research suggests that six years after birth is probably the most biologically reasonable time to completely exclude breast-milk from an individualÂ’s di et. By this age, human individuals usually have achieved adult immune competence, co inciding with the eruption of the first permanent molar (Dettwyler, 1995). Sim ilar to modern weaning practices, complementary non-milk items are provided earl y on, ranging before the first month to 2 to 3 years of age (Dettwyler and Fishman, 1992) . It is the conseque nce of breast-feeding,

PAGE 107

93 duration of the weaning proce ss and introduction of non-milk food that most likely cause stressful periods observed in early childhood. Breast-milk contains species-specific c oncentrations of hormones and bioactive compounds such as peptides, amino acids, glycop roteins, prostaglandins and prolactin, all critical for the development of the gastrointe stinal tract, the pituit ary gland, the pancreas and the brain (Stuart-Macadam, 1995: 9). Ir on and other nutrients are also readily available and easy to digest in breast-milk. Infants benefit from consuming breast-milk because it helps them to receive a balanced nutrition and develop a stronger immune system. Since breast-milk is suckled directly fr om mothers, contamination is not likely to have negative side-effects on the health of infants. Not surprisi ngly, it is frequently reported that morbidity and mortality ar e lower in children who are breast-fed (Cunningham et al., 1991; Duffy et al., 1986; Glass and Stoll, 1989). Although breast-milk is generally sterile and nutrient-rich, its composition changes after a period of lactation. Th e concentrations of most nutri ents decrease and become less bioavailable, usually after a few months of lactation (Stinson, 2000). Complementary foods (given in addition to breast-milk) and supplementary food (given as replacement of breast-milk) is expected to be incorporated in the diet of children, along with breast-milk (Dettwyler and Fishman, 1992). To emphasize th e benefit of breast-feeding and to avoid prolonged period of it, in 1992 the World Heal th Organization recommended for children to be breast-fed at least for one year. Introduction of complementary food during th e weaning process is a necessary but risky task. First, the interaction of foodstuff and breast-milk can lower the bioavailability of certain nutrients in the digestive tract (Oski, 1980 cited in De ttwyler and Fishman,

PAGE 108

94 1992). For example, certain chemical elements in maize and rice products tend to prevent iron from being absorbed (Ryan, 1997). Second, early introduction of unbalanced or non-nutritious weaning food, if accompanied with reduction of breast-milk, could dramatically decrease nutri tion intake and lower the im munological capacity of the infants and children. Certain complementary fo od, such as rice water or porridge, does not contain enough sustenance for rapidly grow ing children. In addi tion, excessive supply of water to the infant can displace the demand for breast-milk and also dilute the nutrient concentration of breast-m ilk (Dettwyler and Fishman, 1992). The rich supply of antibodies found in breast-milk is therefore not sufficiently consumed by the infant (Stinson, 2000). Third and most important, contaminated weaning food can be a severe threat to infant and childhood health. Child ren are hence likely to be exposed to environmental pathogens by consuming pollu ted foodstuffs, most commonly when the water is not obtained from clean sources (Stinson, 2000; Walker, 1986). Diarrhea and infectious diseases frequently occur among infants undergoing weaning, especially in socially disfavored groups and/or ov erpopulated communities (Holland and OÂ’Brien, 1997). Severe consequences, such as delaye d growth or death, can be induced by ingestion of contaminated food/water. At SSH, while no direct evidence of weaning has been found, based on reconstructed subsistence, it is possible to deduce that people weaned infants with a mixture of rice and other foods such as root crops and/or marine foodstuffs. Food and water contamination at coastal sites can be mo re severe than at si tes along freshwater river sources. Because of their weak and compromised immune system, food-induced sickness and/or infectious disease is more likely to impact weanling children.

PAGE 109

95 Ethnographic studies on “traditiona l societies” (referring to small-scale societies with relatively simple modern technologies, lik e electricity and running water; Dettwyler, 1995) have suggested that wean ing duration in these groups ra nges between 2 and 4 years. For a less agriculturally depende nt prehistoric occupation as SSH, an analogy of weaning duration could be reasonably adopted. In a ddition, the peak stress time of the SSH site from LEH analysis falls into the estimated biological timeframe of weaning synthesized by Dettwyler (1995). Therefore, it seems pl ausible to attribute the prevalence and distribution of LEH to weaning-re lated stress for the SSH population. Peak timing of LEH formation In this study, developmental ages correl ating to LEH formation are generated by utilizing two methods whose underlying assump tions are different. Patterns of enamel formation and buried cuspal enamel are critic al issues concerning the estimated time of hypoplastic formation. A method that has be en widely used (Goodman et al., 1980) assumes linear enamel formation and does not acknowledge the effects of cuspal enamel. Empirical observation in Reid and Dean (2000) in contrast, takes cuspal enamel effects into consideration and therefore seems to be a closer estimation of enamel development. The regression method in Good man et al. (1980) returns a wide range of peak LEH formation zones among tooth types. When onl y the anterior teeth are considered, the duration from 2.0 to 5.0 years is most freque ntly affected. When all tooth types are reviewed, peak LEH formation zones appear between 2.0 to 6.0 years. The results may be partly due to different durati ons of crown formation and the availability of recording LEH across tooth types. When LEH locations are converted to deve lopmental ages using the Reid and Dean (2000) method, results show narrower age ra nges than conversion using the method of

PAGE 110

96 Goodman et al. (1980), with the exception of the maxillary lateral incisor, peak LEH zones of five other anterior tooth types fall within the 7th zone of tooth crowns, 2.9-3.8 years for maxillary teeth; 2.3-4.2 for mandibular teeth. Combined with all teeth analyzed, peak LEH formation time is between 2.3 and 4.2 years. The Reid and Dean method more clearly delineates the timing of stress episodes than the method presented by Goodman and colleagues. In addition, by incorporati ng empirical observations of unworn tooth crowns, the Reid and Dean method is bette r equipped to account for the duration of cuspal enamel development. Importantly, linear enamel formation is not assumed. As a result, peak LEH formation time based on Reid and Dean is adopted here as representative of the most stressful pe riod for the SSH population. Results following Goodman et al. (1980), however, are useful for comparative studies. As discussed above, the weaning process may well be linked to the formation of LEH (Corruccini et al., 1985). So me researchers, however, find that LEH formation time does not correlate with documented weaning period (Blakey et al., 1994). Katzenberg and colleagues (1996) argue that the overlappi ng time spans of peak LEH formation and weaning process are possibly co incidental and may not repr esent a causal relationship. In the case of SSH population, moderate prevalence of LEH and aggregated timing of LEH formation suggests that stressful childhood is an endemic phenomenon among individuals. Children between th e ages of 2.3 and 4.2 years are more likely to be affected by systemic physiological disturbances or insults. Besides weaning-induced stress, infectious disease, trauma and chronic blood lo ss (parasitic infection) are possible factors impairing physiological balance of an individu al. Interestingly, no ma rkers of infectious disease or trauma were observed in SSH subadults. Subadults present with porotic

PAGE 111

97 hyperostosis are rarely seen (see below). While subadults are underrepresented in the SSH sample, what skeletons exist are generally free from specific identifiable diseases. As discussed earlier, the weaning process put s an individual at risk of nutritional imbalance and increases oneÂ’s susceptibility to food/water-induced infections resulting from contaminated foods. Considering the geographical location of SSH and its archaeological assemblage, it is very likel y that the complementary food during the weaning period was either marine and/or ri ce based. Marine resources have high potential for contamination if not processed carefull y, and rice products are low in nutrients and can act as inhibitors against the normal ab sorption of essential nutrients, such as iron (Ryan, 1997). Therefore, it is suggested here that the timing of LEH formation correlates with the weaning process at SSH. It is not mere ly a coincidence, but likely a reflection of cultural practice at SSH that provided a st ressful weaning environment experienced by many of its inhabitants. Sex Differences and Cultural Practices Prevalence and number of LEH episodes Females of SSH show higher prevalence of LEH than males. This is in accordance with results reported in Chang (1993) and Pietrusewsky and Tsang (2003). Further, females exhibit higher incidence of LEH on their anterior dentition, and have more multiple LEH episodes than do males. LEH prevalence at the individual level corresponds well with these findi ngs. Since canines and incisors are sensitive recorders of stress events, it is reasonable to infer th at SSH females went through a more stressful childhood than their male counterparts. Females seem to be more vulnerable to repetitive physiological disturbance than do males. E ssentially, their body fails to maintain a normal rhythm of enamel formation and the result is compromised, hypoplastic enamel.

PAGE 112

98 LEH formation timing SSH males tend to have peak LEH forma tion times 0.5 to 1 year earlier than females. This is made especially clear us ing timing derived from Reid and Dean (2000). Peak LEH formation in males falls mostly within the 7th zone of the crown, while femalesÂ’ falls mostly within the 8th zone. No matter which method is used, the distribution of LEH formation times for each tooth type, by sex, shows that males display earlier timing of peak defective zones. Females are said to be well-buffered by a stronger immune system than males (Ortner, 1998). Laboratory studies on various diseases, nutritional, and stress-related disturbances suggest that females cope better during periods of disease and poor diet than are males (Riopelle, 1990; Hoyenga and Hoyenga, 1982 summarized in Guatelli-Steinberg and Lukacs, 1999). As Stin i (1985) has pointed out, sexual differences with respect to oneÂ’s resistance to physiol ogical stress may be linke d to the reproductive demands of females. The exact mechanisms th at lead to females being less sensitive to environmental and physiological stress are unknown (Stinson, 2000), however, the concept of a well-buffered female is not universally accepte d (e.g., Stini, 1994). Weaning practice and diet seems to diffe r according to sex. Differential treatment of children, by sex, may have affected their respective health fre quencies during early childhood and into adulthood. Cultural practices regarding sex/gender roles can largely determine the magnitude and types of stressors . In a society where males are favored, be it for cultural and/or biological reasons, essential or valued resources are less likely to be distributed equally among the sexes. Girls may be intentionally neglected or may be fed with inferior dietary food resources. In extr eme conditions, when resources are limited or a population control policy is proclaimed, it is not uncommon for female infanticide to be

PAGE 113

99 committed in many parts of the world (O omman and Ganatra, 2002; Secondi, 2002; Smith and Smith, 1994). Increased value toward s males is common in Asia and is deep rooted in China. In present day Taiwan, the concept of equality between males and females is fairly recent. The prevalence and peak timing of LEH formation at SSH suggests that sex differences and childhood stress may be indicative of cultura l perceptions and differential treatment by sex during the weaning process. Females are more likely to have LEH and their LEH counts are significantly higher than that of males. Re sults do not support the idea that females are well-buffered agains t environmental and physiological stress. However, two possibilities could be inferred from these data. One is that females are more vulnerable (and sensitive) to stress even ts and therefore it is easier for hypoplastic defects to become manifest on their teeth. The other possibi lity is that SSH females are more frequently exposed to environmental and/or physiologi cal stressors, and perhaps to stronger stressors as well. It may be these tw o combined possibilities are to contribute to the LEH prevalence among females at SSH . Females were likely neglected during infancy and given less nutritious compleme ntary/supplementary food during the weaning process. This would place female children at greater risk (not necessary sick but more vulnerable to stress events) and expose th em to greater physiol ogical stress during the difficult periods of weaning. The 0.5 to 1.0 year difference between ma les and females suggests that female children at SSH are subject to stress related insults later in time than males of equivalent age. It is difficult to identify specific timing for the weaning process based only on macroscopic observations. While timing a nd duration of weaning may vary among the

PAGE 114

100 sexes, effects of the stronge r nature of female physiol ogy (well-buffered female) cannot be excluded from the causal factor for this trend. In short, the difference of LEH prevalence and peak formation time between SSH males and females suggests that cultural pr eference for males during weaning process may be taking place. However, better-buffere d physiology of females is an alternative explanation for them to exhibit dela yed patterns of peak LEH formation. LEH and Mortality Many studies have reported that evidence of growth disturbance is more frequently associated with individuals who died at a young age, especially subadults (Cucina et al., 2000; Douglas et al., 1997; Stodde r, 1997). Overall mortality and longevity seem to have high correlation with incidence of stress markers. For example, research on prehistoric Guam populations demonstrates that individuals who died as subadults and young adults (<21 years) experienced more stressful events than those who survived into middle and old adulthood (Stodder, 1997). It is possibl e that childhood morbidity, early mortality, and reduced longevity are the results of impaired immune function (caused by early stress) and maintenance of a continuous stress load throughout life (Stodder, 1997; Stuart-Macadam, 1998). At SSH, subadults (0-20 years) show slight ly higher prevalence and higher average LEH counts than adults. This is statistica lly significant, however, only for a few tooth types. It appears that those individuals who failed to reach adulthood experienced slightly more stressful periods and/or were more sensitive to stressors. Their health may have been compromised by enduring a greater stress load early in life which made them more vulnerable to infection and other insults later in life. It should be emphasized that the stressors causing LEH were probably not the di rect causes of mortality, as the presence of

PAGE 115

101 LEH represents a sign of resumed function of ameloblasts. Despite lower prevalence and average LEH counts per tooth, individuals wi th LEH who survived into adulthood may have had a stressful childhood similar to th ese individuals who died as subadults. In terms of comparison between age groups , prevalence of LEH is significantly higher in younger individuals ( 35 years). Teeth with multiple LEH episodes are less common in middle and old adult individuals. Wh ile there is positive association between mortality and the appearance of LEH, the eff ects of potential bias must be addressed among the data collected. Fi rst, older individuals ( 35 years) are underrepresented (ca. 17%), and there are fewer preserved teeth as a result. Poor preservation of older individuals and/or shorter life expectancy among the SSH population could account for this trend. Second, LEHs that formed earlier in life on the tooth crown could have been erased by severe attrition. However, teeth wi th less than 2/3 of the enamel were not scored for LEH to control for dental wear . It is possible that the small number of observable teeth for old adults makes their representation skewed, but LEH prevalence data were scored independent to tooth wear. Hence, the finding that individuals living well into adulthood with fewer hypoplastic lesi ons can be explained as a reflection of higher survivorship (lower rate of early mortality) compared to those individuals who have suffered fewer stressful episodes during childhood. There are no significant differences in LEH formation times between subadults and adults. There is no pattern that those who died as subadults have LEH earlier or later than those who survived into adulthood. This result suggests that timing of stressful periods or insults has little effect on overall mortality of individuals at SSH. While the health of an individual could be impaired by childhood stre ss, such as weaning (which may lead to

PAGE 116

102 early mortality), the timing of a stressful even t seems to have less discernible effect on oneÂ’s well-being. Porotic Hyperostosis It is commonly recognized that the etiology of porotic hyperostosis is complicated due to factors such as impaired immune sy stem, unbalanced diet, parasitic infection and life cycle of individual (e.g., G oodman, 1994; Holland and OÂ’Brien, 1997; Stuart-Macadam, 1992). This complex phenomen on has made the pinpointing of specific causes difficult, if not impossibl e. Instead of searching for the cause, research has focused on what the occurrence of poro tic hyperostosis signifies an d how the lesion may correlate within a broader environmental context (Holland and OÂ’Brien, 1997: 191). About a quarter of individuals with observ able affected cranial bones show signs of porotic hyperostosis at SSH. Walker (1986) examined a prehisto ric coastal population from the Channel Islands, California and sugge sts that the occurrence of orbital porotic hyperostosis is the synergetic effect of dia rrheal infection (contami nated water), parasitic infection (eating raw seafood) , prolonged breast-feeding a nd seasonal protein-calorie malnutrition. Indeed, its coastal location a nd marine-based economy make the Channel Islands prehistoric population comparable to SSH. The transition to agriculture and intensif ication, and dependence on staple crops is frequently associated with an increased pr evalence of porotic hyperostosis (Cohen and Armelagos, 1984). At SSH, however, dependenc e on agricultural foodstuffs is not supported by observations of dental pa thology and recovered materials in the archaeological record. It is likely that while the SSH people subsisted on iron-rich marine resources, negative effects of consuming cont aminated water and/or foodstuffs may have increased frequencies of iron-deficiency an emia. Population density, which cannot be

PAGE 117

103 directly assessed at SSH, is al so a potential factor with re spect to site hygiene and the spread of infectious disease. Loss of blood volume and iron concentration due to menstruation and birth leaves females more prone to anemic conditions than males. It follows that porotic hyperostosis, a cranial pathology induced by anemic conditions, is frequently found on female individuals. Stuart-Macadam ( 1989) has advocated that orbita l and vault lesions represent the same pathological process, although orbital areas tend to be affected earlier than vault areas. In SSH, when orbital lesions are consid ered for adults, females are less affected than males, as opposed to the findings in Pietrusewsky and Tsang (2003). On the vault, males are less affected than females. Howe ver, no statistical si gnificance was found on the prevalence of porotic hypero stosis among sexes regardless of lesion location. It is particularly interesting to obser ve that most females affected with orbital lesions exhibit active stages of anemia, and the affected male s typically exhibit healed marks. All vault lesions observed were healed on both sexes. It is possible that females experienced a heavier impact of anemic conditions resulting from intertwined etiol ogies as lower blood iron concentration and/or inferior food choi ces dictated by cultural behavior. Anemic adult females often died when the anemic episodes were still in progress, although anemia itself can not necessarily be identified as the direct cause of death. In terms of adult males, they seem to have recovered from anemic events before they died. Healed or remodeled lesions are indicators of recovery. Stuart-Macadam (1985) in her well-cite d paper has addressed that porotic hyperostosis is a childhood pa thology. Among SSH, however, adults have a slightly higher prevalence of porotic hyperostosis (e ither orbital or vault) than subadults.

PAGE 118

104 Subadults do show a higher prevalence on the orbital area and most of those are active lesions. Ages of the affected subadults vary from early childhood (4-5 years, 2 individuals), to later childhood (7-10 years, 3 individuals), to la te adolescence (17-18 years, 2 individuals). No vault lesions ar e found on subadult bones. The majority of the affected adults are younger than 35 years. In short, age patterns of porotic hypero stosis from the SSH population do not support Stuart-MacadamÂ’s notion that the le sion is a childhood pathology. Adults and subadults have comparable chances to manifest cranial porosity induced by iron-deficiency anemia. Interestingly, i ndividuals living into later adulthood ( 35 years) have significantly fewer cases of porotic hyperostosis than those who died earlier. Although this pattern seems to support a relationship between morbidity and early mortality, small sample size precludes such in ferences to be made (especially in the subadult category). Age and sex distribution of porotic hyperost osis at SSH provides biological data to infer subsistence practice and living environm ent. It is likely that young people (late teens and young adults) participated more frequently in hunting and gatheri ng activities, which increased their chance of exposure to parasi tes. Marine-borne parasites are common in coastal environments and can be easily ac quired by humans collecting shellfish, for example. If parasitic infec tion was indeed the cause of iron-deficiency anemia among some of the SSH people, and was associated w ith subsistence activities, equal prevalence of porotic hyperostosis among males and fe males suggests a non-specific division of labor between sexes. At SSH, both sexes were equally exposed to contaminated food/water resources and unsanit ary working environments.

PAGE 119

105 Young children affected with porotic hyperostosis may exhi bit this pathology as a result of parasitic infection, since the in cidence is very low in this age category. Interaction between contaminated post-wean ing foodstuffs and an immature immune system may account for such lesions. Overall, young children were not the main target of iron-deficiency anemia at SSH, although underr epresentation of subadults is does limit conclusive evidence of this finding. Morbidity of the SSH People The incidence of porotic hyperostosis and linear enamel hypoplasia do not show significant correlation among the SSH population examined in this study. Goodman and Rose (1990) have demonstrated that am elogenesis is sensitive to physiological disturbance and a short period of time (one week) is sufficient to form a macroscopically visible LEH. Therefore, LEH is more freque ntly observed when multiple slight/acute stress events occur during enamel formati on. Porotic hyperostosis, on the other hand, is a bony lesion that results from chronic anemia . Compared to the duration of enamel defects, bony tissue responds to systemic stress manifested as a skeletal lesion in a slow fashion. Hence, porotic hyperostosis is a reco rd of chronic, more severe physiological stress episodes. As a more specific stress indi cator, it is not surprisi ng to encounter fewer individuals affected with porotic hyperostosis. While no significant correl ation has been found between the prevalence of LEH and porotic hyperostosis, it is noteworthy that 75% of indivi duals who show evidence of porotic hyperostosis also e xhibit LEH. Iron-deficiency an emia, however, cannot be the causal factor for LEH formation. Age patterns of individuals who died with active porotic hyperostosis do not overlap the peak LEH form ation time on affected individuals. In fact, individuals who died with act ive lesions and LEH were all older than 8 years of age,

PAGE 120

106 which was a significant depa rture from peak LEH formation time (2 to 5 years postpartum). Individuals who di ed with active lesions but no LEH lend suggest that the occurrences of LEH and porotic hyperostosis are independent events. The appearance of LEH and porotic hyperostosis on the same indi vidual may in turn support the idea that early childhood stress could have impaired the strength of the immune system, and made the individual less effective in coping with iron-deficiency anemia. In terms of those individuals affected with LEH with no sign of porotic hyperostosis, several possibilities could accoun t for this observed pattern. First, these individuals may have experi enced anemic conditions that were not severe enough to manifest as skeletal lesions . Second, individuals could have died from severe anemia, possibly induced by acute infectious disease, before a lesion was formed on the bones. This is a concept noted by Wood et al. (1992) as one of the arguments of the “osteological paradox”. Third, unlike enamel defects, porotic hyperostosis can be erased by bone remodeling process. It is therefore possible that porotic hyperostosis, formed early in life, is no longer observable when an individual survives into later adulthood. The people of SSH had a fairly stressful early childhood due to st ressors associated with the weaning process, including pr olonged breast-feeding, unbalanced and/or contaminated weaning food, and differentia l weaning practice between the sexes. Markers of iron-deficiency anemia are typically found on young adults and show no significant differences between the sexes. Other specific infectious diseases or trauma-related condition were rarely encount ered for both subadults and adult skeletons examined. From this, it is inferred that once an individual survived the stress of early childhood, health conditions improved for the people of SSH people.

PAGE 121

107 CHAPTER 6 CONCLUSION This study is designed to reve al childhood stress patterns of the Shih-san-hang site, one of the earliest and largest Iron Age site s in Taiwan. Two stress indicators, linear enamel hypoplasia (LEH) and porotic hyperost osis, are interpreted in a biocultural context. The incidence of LEH suggests th at the SSH people had a fairly stressful childhood. The peak age of LEH formation range s from 2 to 5 years. Weaning-related stressors, such as contaminated food and water, prolonged breast-feeding and unbalanced weaning diet, are all possible causes. Females were found to be more stressed than males, as indicated by significantly higher LEH prevalence and greater average LEH counts on females. Age of peak LEH formation for females is later than males. Cultural concept (male preference), and practice (differential weaning behavior) may contribute to this difference in weaning time. LEHs are mo re frequently found on subadults and young adults, which may be a result of an impaire d immune system in early childhood (early morbidity) and early mortality, rather than extreme tooth wear biasing the results. Iron-deficiency anemia tends to have great er impact on late teens and young adults as indicated by the age distribution of poro tic hyperostosis among individuals at SSH. Males and females are similarly at risk of anemic stress. Parasitic infection due to marine-oriented subsistence activities a nd poor hygiene throughout their occupational environment are both considered major etio logies. Therefore, it is likely that young individuals, regardless of sex, were more involved in subsiste nce-related tasks in the SSH population. No overlapping age patterns were f ound for the occurrence of LEH or porotic

PAGE 122

108 hyperostosis, which are indicators of independe nt stress events. Hence, no direct causal effect was established between the two patholog ies. In addition, few traces of infectious disease and/or trauma were observed on the SSH remains. Overall, it can be concluded that the SSH inhabitants gene rally had a stressful childhood, a nd once they survived into adulthood, a good health status is maintained. To further understand health and dise ase of the SSH population, an exhaust paleopathological study of the entire SSH collection is warranted. In the future, comparative studies of the SSH site and othe r temporally and geogr aphically-related sites in Taiwan and Southeast Asia are essential to reveal trends associated with changes of subsistence patterns and cultural traditions.

PAGE 123

109 LIST OF REFERENCES Alesan A, Malgosa A, and Simo C (1999) L ooking into the demography of an Iron Age population in the western Mediterranean. I. Mortality. American Journal of Physical Anthropology 110: 285-301. Angel JL (1966) Porotic hyperostosis, anemias, malarias, and marshe s in the prehistoric eastern Mediterranean. Science 153: 760-763. Bass WM (1995) Human Oste ology: A Laboratory and Field Manual (4th Edition). Columbia, Missouri: Missouri Ar chaeological Society, Inc. Blakely RL (1977) Introduction: changing strate gies for the biological anthropologist. In RL Blakely (ed.): Biocultural Adaptation in Prehistoric America. Athens: The University of Georgia Press, pp. 1-9. Blakey ML, Leslie TE, and Reidy JP (1994) Frequency and chronologi cal distribution of dental enamel hypoplasia in enslaved Af rican-Americans: a test of the weaning hypothesis. American Journa l of Physical Anthropology 95: 371-383. Blom DE, Buikstra JE, Keng L, Tomczak PD , Shoreman E, and Stevens-Tuttle D (in press) Anemia and childhood mortality: latitudinal patt erning along the coast of pre-Columbian Peru. American J ournal of Physical Anthropology. Boldsen JL (1997) Estimating pa tterns of disease and morta lity in a medieval Danish village. In RP Paine (ed.): Integrating Archaeological De mography: Multidisciplinary Approaches to Prehistoric Population. Carbondale, IL: Center for Archaeological Investigations, Southern Illinois University, pp. 229-241. Brooks ST, and Suchey JM (1990) Skeletal ag e determination based on the os pubis: a comparison of the Acsadi-Nemeskeri and Suchey-Brooks methods. Human Evolution 5: 227-238. Brothwell DR (1981) Digging Up Bones (3rd Ed ition). Ithaca: Cornell University Press. Buikstra JE, and Ubelaker DH (1994) Sta ndards: For Data Co llection from Human Skeletal Remains. Fayetteville: Arkansas Archaeological Survey. Caffey J (1937) The skeletal changes in the chronic hemolytic anemias. American Journal of Roentgenography and Radiation Therapy 65: 547-560.

PAGE 124

110 Chang C-F (1993) An Osteolog ical Study of Human Remains Recovered from the Shih San Hang Site. Unpublished M.A. Thesis, National Taiwan University, Taipei, Taiwan. Chen K-T (2000) Ancient Iron Technology of Taiwan. Ph.D. Dissertation, Harvard University, Cambridge, MA. Cohen M, and Armelagos GJ, eds. (1984) Pa leopathology at the Orig ins of Agriculture. New York: Academic Press. Cook DC, and Buikstra JE (1979) Health a nd differential survival in prehistoric populations: prenatal dental defects. American Journa l of Physical Anthropology 51: 649-664. Corruccini RS, Handler JS, and Jacobi KP ( 1985) Chronological distribution of enamel hypoplasias and weaning in a Caribb ean slave population. Human Biology 57: 699-712. Cucina A, Mancinelli D, a nd Coppa A (2000) Life span an d physiological perturbations: assessment of demographic parameters and linear enamel hypoplasia in past populations. Homo 51: 56-67. Cunningham AS, Jelliffe DB, and Jeliffe EEP (1991) Breastfeeding and health in the 1980s: a global epidemiologic re view. Journal of Pediatrics 118: 659-666. Domett KM (2001) Health in Late Preh istoric Thailand. Oxford: Archaeopress. Douglas MT (1996) Paleopathol ogy in Human Skeletal Rema ins from the Pre-Metal, Bronze and Iron Ages, Northeastern Tha iland. Ph.D. Dissertati on, University of Hawaii, Honolulu, Hawaii. Douglas MT, Pietrusewsky M, and IkeharaQuebral RM (1997) Skeletal biology of Apurguan: a precontact Chamorro site on Guam. American Journal of Physical Anthropology 104: 291-313. Duffy LC, Ripenhoff-Talty M, Byers TE, La Scolea LJ, and Zielezny MA (1986) Modulation of rotavirus enteritis during br eastfeeding. American J ournal of Diseased Children 140: 1164-1168. Duray SM (1996) Dental indicat ors of stress and reduced ag e at death in prehistoric Native Americans. American Journal of Physical Anthropology 99: 275-286. Glass RI, and Stoll BJ (1989) The protective effect of human milk against diarrhea: a review of studies from Bangladesh. Acta Paediatrica Scandinavian Supplement 351: 131-136.

PAGE 125

111 Goodman AH, Allen LH, Hernandez GP, Amador A, Arriola LV, Chavez A, and Pelto GH (1987) Prevalence and age at developm ent of enamel hypoplasias in Mexican children. American Journal of Physical Anthropology 72: 7-19. Goodman AH, and Armelagos GJ (1988) Childh ood stress and decreased longevity in a prehistoric population. American Anthropologist 90: 936-944. Goodman AH, and Armelagos GJ (1989) Infa nt and childhood morbidity and mortality risks in archaeological p opulations. World Archaeology 21: 225-243. Goodman AH, and Song RJ (1999) Sources of vari ation in estimated ages at formation of linear enamel hypoplasia. In RD Hoppa and C FitzGerald (eds.): Human Growth in the Past: Studies from Bones and Teeth. New York: Cambridge University Press, pp. 210-240. Goodman AH, Armelagos GJ, and Rose JC ( 1980) Enamel hypoplasias as indicators of stress in three prehistoric populations from Illinois. Human Biology 52: 515-528. Goodman AH, Armelagos GJ, and Rose JC (1984) The chronologi cal distribution of enamel hypoplasias from prehistoric Dick son Mounds populations. American Journal of Physical Anthropology 65: 259-266. Goodman AH, Martinez C, and Chavez A ( 1991) Nutritional supplementation and the development of linear enamel hypoplasias in children from Tezonteopan, Mexico. American Journal of Clinical Nutrition 53: 773-781. Goodman AH, and Rose JC (1990) Assessment of systemic physiological perturbations from dental enamel hypoplasias and associat ed histological structures. Yearbook of Physical Anthropology 33: 59-110. Goodman AH, Thomas RB, Swedlund AC, and Armelagos GJ (1988) Biocultural perspectives on stress in prehistoric, hist orical, and contemporary population research. Yearbook of Physical Anthropology 31: 169-202. Guatelli-Steinberg D, and Lukacs JR (1999) Interpreting sex differences in enamel hypoplasia in human and non-human primates : developmental, environmental, and cultural considerations. Year book of Physical Anthropology 42: 73-126. Guy H, Masset C, and Baud C-A (1997) Infa nt taphonomy. Interna tional Journal of Osteoarchaeology 7: 221-229. Hengen OP (1971) Cribra or bitalia: pathogenesis a nd probable etiology. Homo 22: 57-76. Hill MC, and Armelagos GJ (1990) Porotic hyperostosis in past and present perspective. In JE Buikstra (ed.): A Life in Science: Papers in Honor of J. Lawrence Angel. Kampsville, IL: Center of American Archaeology. pp. 52-63.

PAGE 126

112 Hillson S, and Bond S (1997) Relationship of enamel hypoplasia to the pattern of tooth crown growth: a discussion. American Journal of Physical Anthropology 104: 89-103. Hodges DC, and Wilkinson RG (1990) Eff ect of tooth size on the aging and chronological distribution of enamel hypoplas tic defects. American Journal of Human Biology 2: 553-560. Hoffbrand AV, and Lewis SM (1981) Postgraduate Haematology. London: William Heinemann Medical Books Limited. Holland TD, and O'Brien MJ (1997) Parasites, porotic hyperostosis, a nd the implications of changing perspectives. American Antiquity 62: 183-193. Hoyenga KB, and Hoyenga KT (1982) Gender and energy balance: sex differences in adaptations for feast and famine. Physiological Behavior 28: 545-563. Huss-Ashmore R, Goodman AH, and Armelagos GJ (1982) Nutriti onal inference from paleopathology. In M Schiffer (ed.): Advances in Archaeological Method and Theory. New York: Academic Press, pp. 395-474. Hutchinson DL, and Larsen CS (1988) Determ ination of stress episode duration from linear enamel hypoplasias: a case study from St. Catherines Island, Georgia. Human Biology 60: 93-110. Katzenberg MA, Herring DA, and Saunders SR (1996) Weaning and infant mortality: evaluating the skeletal evidence. Yearbook of Physical Anthropology 39: 177-199. Keenleyside MA (1994) Skeletal Evidence of Health and Disease in Preand Post-Contact Alaskan Eskimos and Aleuts. P h.D. Dissertation, Mc Master University, Hamilton, Canada. Koch MJ, Buhrer R, Pioch T, and Scharer K (1999) Enamel hypoplasia of primary teeth in chronic renal failu re. Pediatric Nephrology 13: 68-72. Lanphear KM (1990) Frequency and distribut ion of enamel hypoplasias in a historic skeletal sample. American Jour nal of Physical Anthropology 81: 35-43. Larsen CS (1997) Bioarchaeology. Camb ridge: Cambridge University Press. Larsen CS, ed. (2001) Bioarchaeology of Span ish Florida: The Impact of Colonialism. Gainesville: University Press of Florida. Lin H-M (1997) Zooarchaeological Study of Sus Mandible from Shih-San-Hang Site. Unpublished MA Thesis, National Taiw an University, Taipei, Taiwan.

PAGE 127

113 Littleton J, and Frohlich B ( 1993) Fish-eaters and farmers: dental pathology in the Arabian Gulf. American Journal of Physical Anthropology 92: 427-447. Liu I-C (1992) The Archaeological Sites of Taiwan. Taipei, Taiwan: Taipei County Cultural Bureau. Liu I-C (1995) Preliminary study on the rela tionship between prehistoric cultures and aborigines. The Taiwan Folkways 45: 75-98. Lovejoy CO, Meindl RS, Pryzbeck TR, and Mensforth RP (1985) Chronological metamorphosis of the auricular surface of the ilium: a new method for the determination of adult skeletal age at death. American Journal of Physical Anthropology 68: 15-28. Lovell NC, and Whyte I (1999) Patterns of de ntal enamel defects at ancient Mendes, Egypt. American Journal of Physical Anthropology 110: 69-80. Lukacs JR, Nelson GC, and Walimbe SR ( 2001) Enamel hypoplasia and childhood stress in prehistory: new data fro m India and Southwest Asia. Journal of Archaeological Science 28: 1159-1169. Manzi G, Salvadei L, Vienna A, and Pssarell o P (1999) Discontinuity of life conditions at the transition from the Roman Imperial Age to the early Middle Ages: example from central Italy evaluated by pa thological dento-al veolar lesions. Am erican Journal of Human Biology 11: 327-341. McKern T, and Stewart TD (1957) Skelet al Age Changes in Young American Males Analyzed from the Standpoint of Identific ation. Technical Repor t EP-45. Natick, MA: Headquarters, Quartermaster Research and Development Command. Meindl RS, and Lovejoy CO (1985) Ectocrania l suture closure: a revised method for the determination of skeletal age at death base d on the lateral-anterior sutures. American Journal of Physical Anthropology 68: 57-66. Mensforth RP, Lovejoy CO, Lallo JW, and Armelagos GJ (1978) The role of constitutional factors, diet, and infec tious disease in the etiology of porotic hyperostosis and periosteal reactions in prehistoric in fants and children. Medical Anthropology 2: 1-59. Merbs CF (1992) A New World of infec tious disease. Yearbook of Physical Anthropology 35: 3-42. Moggicecchi J, Pacciani E, and Pintociste rnas J (1994) Enamel hypoplasia and age at weaning in 19th century Flor ence, Italy. American Journa l of Physical Anthropology 93: 299-306.

PAGE 128

114 Nikiforuk G, and Fraser D ( 1981) The etiology of enamel hypoplasia: a unifying concept. Journal of Pediatrics 98: 888-893. Oomman N, and Ganatr a BR (2002) Sex selection: the syst ematic elimination of girls. Reproductive Health Matters 10: 184-188. Ortner DJ (1998) Male-female immune reactiv ity and its implications for interpreting evidence in human skeletal paleopathology. In AL Grauer and P Stuart-Macadam (eds.): Sex and Gender in Paleopathologi cal Perspective. Cambridge: Cambridge University Press, pp. 79-92. Palubeckaite Z (2001) Patterns of linear enamel hypoplasia in Lithuanian Iron Age population. Variability and Evolution 9: 75-87. Pietrusewsky M, and Chang C-F (2003) Ta iwan aborigines and peoples of the Pacific-Asia region: multivariate craniome tric comparisons. Anthropological Science 111: 293-332. Pietrusewsky M, and Douglas MT (2001) Intensif ication of agriculture at Ban Chiang: is there evidence from the skeletons? Asian Perspectives 40: 157-178. Pietrusewsky M, and Tsang C-H (2003) A prel iminary assessment of health and disease in human skeletal remains from Shi San Hang: a prehistoric aborig inal site on Taiwan. Anthropological Science 111: 203-223. Pindborg JJ (1982) Aetiology of developmental en amel defects not related to fluorosis. International Dental Journal 32: 123-134. Powell ML (1988) Status and Health in Prehistory. Washington D.C.: Smithsonian Institution Press. Reid DJ, and Dean MC (2000) Brief comm unication: the timing of linear hypoplasias on human anterior teeth. American J ournal of Physical Anthropology 113: 135-139. Rose JC, Armelagos GJ, and Lallo JW (1978) Histological enamel i ndicator of childhood stress in prehistoric skeletal samples. American Journal of Physical Anthropology 49: 511-516. Riopelle AJ (1990) Postnatal pr otein deprivation in rhesus monkeys. American Journal of Physical Anthropology 83: 239-252. Ryan AS (1997) Iron-deficiency anemia in in fant development: implications for growth, cognitive development, resistance to in fection, and iron supplementation. Yearbook of Physical Anthropology 40: 25-62.

PAGE 129

115 Salvadei L, Ricci F, and Manzi G (2001) Poro tic hyperostosis as a ma rker of health and nutritional conditions during childhood: studies at the transition between imperial Rome and the early Middle Ages. American Journal of Human Biology 13: 709-717. Saunders SR, and Keenleyside A (1999) Enam el hypoplasia in a Canadian historic sample. American Journal of Human Biology 11: 513-524. Secondi GS (2002) Biased childhood sex ratios and the economics status of the family in rural China. Journal of Comparative Family Studies 33: 215-234. Seow WK, Needleman HL, Smith LEH, Ho ltzman D, and Najjar S (1995) Enamel hypoplasia, bilateral cataracts, and aqueduc tal stenosis: a new syndrome. American Journal of Medical Genetics 58: 371-373. Skinner MF, and Goodman AH (1992) Anthropol ogical uses of developmental defects of enamel. In S Saunders and M Katzenberg (e ds.): Skeletal Biol ogy of Past Peoples: Research Methods. New York: Wiley-Liss, Inc., pp. 153-174. Slaus M (2000) Biocultural anal ysis of sex differences in mortality profiles and stress levels in the late medieval population from Nova Raca, Croatia. American Journal of Physical Anthropology 111: 193-209. Slaus M, Kollmann D, Novak SA, and Novak M (2002) Temporal tre nds in demographic profiles and stress levels in medieval (6th-13th century) population samples from continental Croatia. Croatian Medical Journal 43: 598-605. Smith BH (1984) Patterns of molar wear in hunter-gatherers a nd agriculturalists. American Journal of Physical Anthropology 63: 39-56. Smith EA, and Smith SA (1994) Inuit sex -ratio variation: population control, ethnographic error, or parental manipulation? Current Anthropology 35: 595-624. Stewart TD (1979) Essentials of Forensic Anthropology. Springfield, IL: Charles C. Thomas. Stini WA (1994) Differences in female and male aging patterns in modern populations. American Journal of Physical Anthropology Supplement 18: 188. Stinson S (2000) Growth varia tion: biological and cultural f actors. In S Stinson, B Bogin, R Huss-Ashmore and D O'Reilly (eds.): Human Biology: An Evolutionary and Biocultural Perspective. New York: Wiley-Liss, pp. 425-463. Stodder ALW (1997) Subadult stress, morbidity, and longevity in Latte Period populations on Guam, Mariana Islands. Amer ican Journal of Physical Anthropology 104: 363-380.

PAGE 130

116 Stuart-Macadam P (1985) Porotic hyperostosis : representative of a childhood condition. American Journal of Physical Anthropology 66: 391-398. Stuart-Macadam P (1987) Porotic hyperostos is: new evidence to support the anemia theory. American Journal of Physical Anthropology 74: 521-526. Stuart-Macadam P (1989) Porotic hyperostos is: relationship between orbital and vault lesions. American Journal of Physical Anthropology 80: 187-193. Stuart-Macadam P (1992) Porotic hyperostosis : a new perspective. American Journal of Physical Anthropology 87: 39-47. Stuart-Macadam P (1998) Iron deficiency anem ia: exploring the difference. In A Grauer and P Stuart-Macadam (eds.): Sex and Gender in Paleopathological Perspective. Cambridge: Cambridge University Press, pp. 45-63. Suckling GW (1989) Developmental defects of enamel: histological and present-day perspectives of their pathogenesis. Advances of Dental Research 3: 87-94. Suckling GW, Elliott DC, and Thurley DC (1986) The macroscopic appearance and associated histological cha nges in the enamel organ of hypoplastic lesions of sheep incisor teeth resulting from induced pa rasitism. Archives of Oral Biology 31: 427-439. Tayles N, Domett KM, and Nelson K (2000) Ag riculture and dental caries? The case of rice in prehistoric South east Asia. World Archaeology 32: 68-83. Tsang C-H (2000) Recent advances in the Ir on Age archaeology of Taiwan. Indo-Pacific Prehistory Association Bulletin 20: 153-158. Tsang C-H, Kao J, and Liu I-C (1990) Archaeo logical Studies of the Early Chinese and Plains Aborigines of Taiwan. Second Year Re port. Taipei, Taiwan: Institute of History and Philology, Academia Sinica. Tsang C-H, and Liu I-C (2001a) Projects Report of the Exhibitions in the Shih-san-hang Museum. Taipei, Taiwan: Organization Co mmittee of the Shih-san-hang Museum, Taipei County. Tsang C-H, and Liu I-C (2001b) The Shih-s an-hang Site: Salvag e and Preliminary Research. Taipei, Taiwan: Cultural Affair s Bureau of Taipei County Government. Tsang C-H, and Liu I-C (n.d.) Unpublishe d field notes. Shih-san-hang Salvage Excavation Project, 1990-1992. Turner II CG (1979) Dental anthropologica l indications of agriculture among the Jomon people of central Japan. X. Peopling of the Pacific. American Journal of Physical Anthropology 51: 619-636.

PAGE 131

117 Ubelaker DH (1989) Human Skeletal Remains (2nd Edition). Washington D.C.: Taraxacum. Ulijaszek SJ (1991) Human dietary change. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 334: 271-279. Walker PL (1986) Porotic hyperostosis in a marine dependent California Indian population. American Journal of Physical Anthropology 69: 345-354. Wang S-C (1984) The Neolithic Si te of Chih-shan-yen. Taipei , Taiwan: Taipei Municipal Cultural Heritage Commission. Wapler U, Crubezy E, and Schultz M (2004) Is cribra orbitalia synonymous with anemia? Analysis and interpretation of cranial pathology in Sudan. American Journal of Physical Anthropology 123: 333-339. White TD (2000) Human Osteology (2nd Edition). San Diego: Academic Press. Wintrobe M (1993) Clinical Hematol ogy. Philadelphia: Lea and Febiger. Wood JW, Milner GR, Harpending HC, and We iss KM (1992) The osteological paradox: problems of inferring prehistoric health fr om skeletal samples. Current Anthropology 33: 343-370. Wood L (1996) Frequency and chronological di stribution of linear enamel hypoplasia in a North American colonial skeletal sa mple. American Journal of Physical Anthropology 100: 247-259. Wright LE (1994) The Sacrifice of the Earth? Diet, Health, and Inequality in the Pasion Maya Lowlands. Ph D. Dissertation, Un iversity of Chicago, Chicago, Illinois. Wright LE (1997) Intertooth patterns of hypoplasia expression: implications for childhood health in the Classic Maya coll apse. American Journal of Physical Anthropology 102: 233-247. Yang C-S (1961) A report on archaeologi cal surveys at the Shi-san-hang and Ta-pen-keng sites. Bulletin of the Department of Archaeology and Anthropology, National Taiwan University 17/18: 45-70.

PAGE 132

118 BIOGRAPHICAL SKETCH Chin-hsin Liu was born and raised in Pi ngtung, the southernmost county in Taiwan. After her high school education in Pingtung Gi rls’ High, she went to National Taiwan University and officially ste pped into the field of anthropo logy, although her interests in anthropology and bioarchaeology have been with her since she wa s eleven. Chin-hsin received her B.A. degree from National Taiw an University in anthropology in June, 2002, with a minor in political science and certification in the earth system program. To pursue her lifelong career as a biologica l anthropologist, she jo ined the graduate program in the Department of Anthropology at the University of Florida in August, 2002. There, she studies and works under the mentorship of Dr. John Krigbaum. She finds this institution offers exactly what she has been seeking for, and will stay to work towards her doctoral degree.