Scholarly article on topic 'Stable dietary isotopes and mtDNA from Woodland period southern Ontario people: results from a tooth sampling protocol'

Stable dietary isotopes and mtDNA from Woodland period southern Ontario people: results from a tooth sampling protocol Academic research paper on "History and archaeology"

CC BY-NC-ND
0
0
Share paper
Academic journal
Journal of Archaeological Science
OECD Field of science
Keywords
{"Mitochondrial DNA" / "Ontario archaeology" / "Late Woodland" / Iroquois / "Molecular anthropology" / Bioarchaeology / Palaeodiet / "Bone and tooth chemistry" / Enamel / Dentin / "Carbon isotopes" / "Nitrogen isotopes" / Fisheries / "Genetic diversity"}

Abstract of research paper on History and archaeology, author of scientific article — Susan Pfeiffer, Ronald F. Williamson, Judith C. Sealy, David G. Smith, Meradeth H. Snow

Abstract Bioarchaeological research must balance scholarly commitment to the generation of new knowledge, descendants' interests in their collective past, and the now common practice of rapid re-interment of excavated human remains. This paper documents the first results of a negotiated protocol built on the retention of one tooth per archaeologically derived skeleton, teeth that can then be used for destructive testing associated with ancient DNA and stable isotope investigations. Seven archaeological sites dating from the 13th to 16th centuries provided 53 teeth, 10 of which were subdivided between DNA and isotope labs. All tooth roots yielded haplogroup results, and five provided more detailed sequence results. Stable isotopes of carbon and nitrogen document heavy reliance on maize among all individuals, as well as reliance on a diverse range of fish. This work establishes baseline mtDNA information for Northern Iroquoians, and confirms the value of using dental tissues for dietary reconstruction. Particularly when human remains are fragmentary or co-mingled, this approach holds promise for ongoing incorporation of bioarchaeology into reconstructions of past peoples' lives.

Academic research paper on topic "Stable dietary isotopes and mtDNA from Woodland period southern Ontario people: results from a tooth sampling protocol"

ELSEVIER

Contents lists available at ScienceDirect

Journal of Archaeological Science

journal homepage: http://www.elsevier.com/locate/jas

Stable dietary isotopes and mtDNA from Woodland period southern Ontario people: results from a tooth sampling protocol^

Susan Pfeiffer a'b'*, Ronald F. Williamson c, Judith C. Sealyb, David G. Smith d, Meradeth H. Snowe

a Department of Anthropology, University of Toronto, 19 Russell Street, Toronto M5S2S2, Canada b Department of Archaeology, University of Cape Town, Private BagX3, Rondebosch 7701, South Africa cArchaeological Services Inc., 528 Bathurst Street, Toronto M5S 2P9, Canada d Department of Anthropology, University of California, One Shields Avenue, Davis 95616, USA e University of Montana, Missoula, MT 59812, USA

ARTICLE INFO

ABSTRACT

Article history: Received 21 May 2013 Received in revised form 11 October 2013 Accepted 6 November 2013

Keywords: Mitochondrial DNA Ontario archaeology Late Woodland Iroquois

Molecular anthropology

Bioarchaeology

Palaeodiet

Bone and tooth chemistry

Enamel

Dentin

Carbon isotopes Nitrogen isotopes Fisheries Genetic diversity

Bioarchaeological research must balance scholarly commitment to the generation of new knowledge, descendants' interests in their collective past, and the now common practice of rapid re-interment of excavated human remains. This paper documents the first results of a negotiated protocol built on the retention of one tooth per archaeologically derived skeleton, teeth that can then be used for destructive testing associated with ancient DNA and stable isotope investigations. Seven archaeological sites dating from the 13th to 16th centuries provided 53 teeth, 10 of which were subdivided between DNA and isotope labs. All tooth roots yielded haplogroup results, and five provided more detailed sequence results. Stable isotopes of carbon and nitrogen document heavy reliance on maize among all individuals, as well as reliance on a diverse range of fish. This work establishes baseline mtDNA information for Northern Iroquoians, and confirms the value of using dental tissues for dietary reconstruction. Particularly when human remains are fragmentary or co-mingled, this approach holds promise for ongoing incorporation of bioarchaeology into reconstructions of past peoples' lives.

© 2013 The Authors. Published by Elsevier Ltd. All rights reserved.

1. Introduction

The lower Great Lakes of North America were the focus of dramatic cultural developments prior to European contact. The Transitional to the Late Woodland period, ca. 1500 to 350 BP, saw the development of distinctive Iroquoian and Algonquian speaking

q This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited.

* Corresponding author. Department of Anthropology, University of Toronto, 19 Russell Street, Toronto M5S 2S2, Canada. Tel.: +1 416 978 4004; fax: +1 416 978 3217.

E-mail addresses: susan.pfeiffer@utoronto.ca (S. Pfeiffer), rwilliamson@iasi.to (R

F. Williamson), Judith.sealy@uct.ac.za (J.C. Sealy), dgsmith@ucdavis.edu (D.

G. Smith), meradeth.snow@umontana.edu (M.H. Snow).

cultural groups. Southern Ontario was the home of the Wendat (known to the French as the Huron), the Tionontaté (Petun), and the Attiwandaronk (Neutrals), all of whom spoke Northern Iroquoian languages, as well as a number of Algonquian-speaking groups including the Odawa, Nipissing, and Algonquian (Fig. 1). There were interactions, as well, with the Iroquois Confederacy (Haudenosaunee) who lived in clustered tribal groupings across central New York State and included (from west to east) the Seneca, Cayuga, Onondaga, Oneida and Mohawk. All of these nations had unique cultural traits and histories owing to their geographic separation and development in distinct tribal territories, which they continued to occupy into the historic period.

Northern Iroquoian society often forms a distinct focus of Late Woodland archaeology. In Ontario, the Late Woodland period is subdivided into Early (A.D. 900-A.D. 1300), Middle (A.D. 1300-A.D. 1400) and Late Iroquoian Periods (A.D. 1400-A.D. 1650). Maize

0305-4403/$ - see front matter © 2013 The Authors. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/jjas.2013.11.008

Fig. 1. Distribution of First Nations populations of the Great Lakes, prior to disruptions associated with European contact. Labels in CAPS indicate broader language groups of the region.

agriculture was an important organizing principle as large palisaded base settlements around which maize was grown developed from about A.D. 1000 onward (Pihl et al., 2008). Around the turn of the fourteenth century, a fully developed agricultural system including large, year-round occupied villages, new socially inte-grative institutions and distinctive material culture all point for the first time to a recognizably "Iroquoian" pattern as described in early European reports of their life. Population growth followed (Warrick, 2008). By the mid-fifteenth century, communities were coalescing in the face of violent conflict. By the early sixteenth century, these communities had consolidated with populations between 1500 and 2000 people with new fully integrated identities (Birch and Williamson, 2013). One of the distinguishing socially integrative features of ancestral and historic Huron-Wendat populations was the Feast of the Dead, associated with secondary burial in ossuaries of as many as 500 deceased community members (Williamson and Steiss, 2003).

Early to mid-seventeenth century accounts from explorers and missionaries provide information that can form a basis for hypothesis building (Thwaites, 1896-1901; Trigger, 1976). This approach must be supplemented, or even challenged, when sites are several hundred years older and when topics of interest are beyond the scope of historic observations.

Among First Nations descendant communities, interest in ancestral populations is growing. Permission to study the remains of archaeologically discovered ancestors is often granted, contingent on conditions that only small fragments of human tissue be retained and that questions deemed pertinent to the descendants be addressed. In part because of the analytical limitations of comingled ossuary remains, a protocol of retaining one tooth per person is a pragmatic approach. The approximate age and the sex of the person can be estimated from the morphology of the maxilla or

mandible from which a permanent molar is taken, or the age in the case of a deciduous tooth. Insofar as modern teeth are removed as part of dental treatment and tooth loss is seen as a typical life experience, teeth are rarely seen as being imbued with spiritual power, so that this approach is acceptable to many descendants. The research described here arose as an exploratory project, designed to maximize the relevant information that could be obtained from a single human tooth. This project also marks the first opportunity to examine the mitochondrial DNA composition of ancient Northern Iroquoian speakers. Relatively little is known about the relationship of these groups to other aboriginal populations of Northeast North America. As the mtDNA of Algonquian groups of this region becomes better known (Dewar et al., 2010; Shook and Smith, 2008), information about northern Iroquoians is a missing piece of the puzzle.

The teeth come from excavations in which the human remains were reinterred after brief descriptive study and tooth sampling. In some cases, study was limited to in-ground examination. One tooth per person was retained with permission of the relevant First Nations groups, following a consultative process consistent with the Ontario Cemeteries Act. Each archaeological site in the study illustrates different research opportunities and challenges. Only one tooth from one site is of probable Attiwandaronk ancestry; all others are probable Wendat. One site, Staines Road, provided numerous teeth but they are from a disturbed context that compromised the generation of population information. Another site, Mantle, yielded two different contexts for human remains: a formal cemetery of primary interments and loose teeth found within the living site. Despite varying contexts, study of the teeth can generate information about the proportion of the diet that came from maize and the sources of animal protein. Despite differences in the excavation contexts, the study provides

methodological information about the isotopic offsets among human collagen, dentin and enamel tissues in a consistent and comparable format.

This study also provides information about the mitochondrial DNA of the ancestors of Northern Iroquoian speakers, contributing to our understanding of the peopling of Northeast North America. North and South America are home to some of the earliest studies of mtDNA and its distributions. The first four haplogroups identified (A, B, C, and D), based on differences from the Anderson or Cambridge Reference Sequence, are found in both modern and ancient native populations. Two more mtDNA lineages have been described in ancient remains from the Americas, haplogroups X (Malhi and Smith, 2002), and M (Malhi et al., 2007) which have so far been reported only in North America, but the latter haplogroup has not been reported in any modern population. Together, these six haplogroups are thought to comprise the vast majority, if not all, of New World mtDNA diversity.

Prior studies of dietary patterns using carbon and nitrogen isotopes from bone collagen and dental tissues have documented a rather rapid adoption and ultimately heavy reliance on maize (Katzenberg et al., 1995; Schwarcz et al., 1983,1985; van der Merwe et al., 2003b). Investigating the nature of the transition to maize horticulture is important not only to answering questions regarding nutrition and health, but also to understanding the evolution of Iroquoian cultural complexity. High proportions of maize, perhaps high enough to be detrimental to health, may have been transient (Katzenberg et al., 1995; Schwarcz et al., 1985; van der Merwe et al., 2003b). Recent studies also report that neighbouring, non-Iroquoian groups ingested substantially more maize than would be expected in that their settlement systems lack year-round occupied villages (Dewar et al., 2010; Watts et al., 2011). These possibilities introduce questions about the temporal and spatial patterning of dietary regimens. Early work on isotopic dietary signals was based on bone collagen. This is a protein and as such, the carbon it contains comes primarily from protein foods, with some contribution from dietary carbohydrates and fats. The carbon isotope signal in collagen is therefore skewed in favour of protein foods. In this region, riverine, lacustrine and anadromous fish species were important sources of dietary protein. Their diverse isotopic values make the interpretation of human protein-based tissues challenging. Wishing to clarify the dietary carbohydrate contribution of maize to Iroquoian communities, researchers have explored apatite values from bone (Harrison and Katzenberg, 2003) and tooth crowns (van der Merwe et al., 2003b). The latter study of the early fourteenth century Moatfield ossuary proposed a design that focuses on dental tissue, including tooth enamel apatite and collagen from dentin. The potential for distinguishing isotopic differences between individuals who relied on different food sources or who grew up in different regions introduces the possibility of tracking local group identity, especially during periods of coalescence. The dynamic nature of the cultural landscape also generates questions about population origins and relationships.

2. Materials and methods

2.1. The archaeological sites

The teeth used in this study were derived from seven archaeological sites located in southern Ontario (Table , Fig. 2).

The Damiani site (Borden designation AlGv-231) is a 1.5 ha late fifteenth century village situated in the upper reaches of the Humber River in the City of Vaughn comprising 22 long houses, a palisade, and hundreds of features. Its location, structure and artifactual assemblage indicate Wendat cultural affinity. Five burial features were excavated by Archaeological Services Inc. (ASI) in

Table 1

Archaeological sites providing human teeth for this study.

Site name Borden Approx. Date Sources

number

Damiani AlGv-231 AD 1450—1500 ASI 2012

Hidden Spring AlGu-368 AD 1450—1500 ASI 2010

Hutchinson AkGt-34 Mid-14th century Robertson 2004

Mantle AlGt-334 AD 1500—1530 Birch and Williamson 2013

Staines Road AkGt-65 Late 13 century Williamson and Steiss 2003

Teston Road AlGv-2 AD 1450—1500 ASI 2005

Wainfleet N/A Late Woodland ASI 2009

2009. Of the five, only one was of an adult (burial 2). The burial of infants and juveniles beneath house floors is a common feature of Iroquoian sites. Teeth available for study include a second maxillary molar from the adult woman and a deciduous maxillary molar from a child of about three years of age.

The late fifteenth century Hidden Spring site (Borden designation AlGu-368), excavated by ASI in 2008, is located about 15 km east of Damiani in the Don River watershed and is from a similar time period. Portions of the two long houses that were excavated included three burial features. Each feature included at least some poorly preserved adult remains, as well as juveniles in two of the three. Sex and specific ages at death remain unknown. The sampled teeth include three first permanent molars (two mandibular and one maxillary), one maxillary third molar and, from one of the juveniles, a deciduous second mandibular molar. In two of the adult teeth, some adhering alveolar bone was also available.

The fourteenth century Hutchinson site (Borden designation AkGt-34), excavated by ASI in 2001, is located along the Rouge River in the City of Toronto, Ontario. Two long houses were identified along with two small separate mortuary areas that contained intact burials as well as fragmentary skeletal remains of at least twelve individuals. It has been proposed that the site was used to prepare the dead for burial (Robertson, 2004). Given its immediate proximity to the Staines Road ossuary and their similarity in age, it is possible this was a preparation site for that ossuary. One tooth from each of five adults, plus alveolar bone from one of the five, were available for study. Sex and specific ages at death remain unknown.

The early sixteenth century Mantle site (Borden designation AlGt-334), located near Stouffville, Ontario, York Region, is the most structurally complex of the sites in this study. Excavations by ASI were undertaken from 2002 to 2005 (Birch and Williamson, 2013). The 2.94 ha occupation site included 98 overlapping long houses, 51 of which may have been occupied at one time (Birch and Williamson, 2013:78), providing housing for 1500—2000 people. From fragmentary human remains found within the village, nine permanent teeth were available for study. A sampling preference for right first molars combined with dispersal of teeth from across this very large site makes double sampling of any one person very unlikely. Mantle represents the coalescence of multiple small villages into one well-planned and well-integrated community (Birch and Williamson, 2013:53—64). Beyond the village palisade was a cemetery with at least thirty-four primary interments. Exceptionally cosmopolitan artifacts from the village, combined with single, primary interments — an interment style unusual for ancestral Wendat — contribute to a conclusion that people from quite far away may have lived and died at Mantle. Analysis of the skeletons was limited to only on-site observations, plus the removal of one tooth from a number of the adult burials. Samples include the maxillary first molar from one woman, and three first molars and one second premolar from each of four men.

The Staines Road ossuary (Borden designation AkGt-65) was discovered at a locale adjacent to the Hutchinson site in northwest Scarborough (eastern metropolitan Toronto). Investigated and

0 15 30

Fig. 2. Regional map illustrating site locations. Shading reflects surficial geology, lighter tones being clays and darker tones primarily till soils, highly relevant variables to communities dependent on maize crops.

excavated by ASI in 2001, this ossuary of at least 308 individuals had been removed from its original location, and fragmented, perhaps mechanically, then mixed into a pile of soil fill and modern garbage. Its aboriginal origin was confirmed through a radiocarbon date from bone collagen (Beta-156359) and the associated 513C value of —11.3. The date between cal B.P. 920 and 680, which must apply to all of the remains as most Wendat ossuaries are formed during one ceremonial event (Williamson and Steiss, 2003), makes the site a very early example of ossuary burial, so its large size is especially surprising. Soil adhering to the human remains as well as site taphonomy suggests the original ossuary pit was adjacent to the deposit. Of the numerous loose teeth that have been retained from this large collection, thirteen right first maxillary molars and two left first molars were chosen for this study. Sex and specific ages at death remain unknown.

The mid- to late fifteenth century Teston village site and ossuary (Borden designation AlGv-2) are located on the west bank of the West Don River in the City of Vaughan. While the village is situated on flat, high tableland, the ossuary is located on a small knoll on the tableland, one to two hundred metres to the southwest of the village (the precise limits of which are not known). The ossuary was discovered in 2005 during construction associated with the widening and relocation of Teston Road. It was found to have a diameter of 280 cm and only a portion of the feature was disturbed. From these disturbed materials a number of permanent maxillary molar teeth were retained for future study, consistent with the site disposition agreement. Displaced remains were returned to the ossuary and it has been permanently protected. Ossuaries of this size have been known to contain the remains of 300-400 individuals (Williamson and Steiss, 2003).

A maxillary first permanent molar from the Wainfleet site has also been included. It comes from an isolated burial of a man

investigated by police and documented by ASI in 2009, in Wain-fleet, Ontario, Niagara Region. The locale and associated artifacts suggest that this burial is ancestral to Attiwandaronk (Neutral) people.

2.2. Documentation

Prior to destructive testing, dental measurements were taken of maximum dimensions, as well as analogous dimensions at the cement-enamel junction. Each tooth was photographed in a standard format with scale from two perspectives, and a negative impression was made of each crown. A selection of ten teeth from four of the sites was sent to the Molecular Anthropology Laboratory at the University of California in Davis, California, where portions of the roots (approximately 0.5 g) were removed for analysis. The tooth crowns were returned to the University of Toronto, from which the entire sample of 53 teeth was transported to the University of Cape Town Archaeometry Laboratory for isotopic analysis.

2.3. Ancient DNA methods

The Molecular Anthropology Laboratory (MAL) at the University of California in Davis, California, maintains a positively pressurized clean room that is separated from the modern DNA extraction and PCR laboratories and to which access is strictly controlled. Precautions to minimize risk of contamination at every step of the extraction process include UV irradiating all supplies, performing duplicate independent extractions from, and amplification of, each sample, daily bleaching of all surfaces and tools within the room, use of protective clothing and running negative controls at each step. The mtDNA haplogroup and HVSI&II control region sequences

of anyone having any contact with the samples and aDNA facilities in which the analyses were conducted were noted, to ensure that any source of laboratory contamination of the aDNA samples could be identified.

The MAL conducted two independent DNA extractions on each sample, following the standard protocols described in Kemp et al. (2006). Each sample was cleaned with 6% sodium hypochlorite (full strength household bleach; Kemp and Smith, 2005), and soaked in 2 mL of molecular grade 0.5 M, pH 8.0, EDTA (Gibco) for at least 48 h (an average time of seven days) to remove calcium. Three milligrams of Proteinase K (Invitrogen, Fungal Proteinase K) was added to each sample and incubated at 65 °C overnight (an average time of 18 h). A three-step phenol-chloroform extraction was carried out including two extractions with equal volumes (2 mL) of phenol:chloroform:isoamyl alcohol (24:24:1; OmniPur), followed by one extraction with an equal volume of chloroform:isoamyl alcohol (24:1; Amresco). The DNA was precipitated in a solution of isoproponol and 5 M ammonium acetate overnight (for at least 12 h) to remove PCR inhibitors. The DNA was pelleted by centri-fugation at 3100 rpm for at least 45 min, washed in 1 mL of 80% purified ethanol, centrifuged for at least another 45 min, and allowed to air-dry for 15-20 min. Each sample was purified using a Wizard PCR Preps Purification System (Promega) following the manufacturer's instructions, save for the final elution of the DNA using 1200 ul of ddH2O.

Regions containing the mutations diagnostic of haplogroups A, B, C, and D were amplified at least twice from each extraction using the polymerase chain reaction (PCR) and primers, cycling conditions and methods cited in Kemp et al. (2006). After amplification of a PCR product of the expected size was confirmed by electrophoresis on 6% polyacrylamide gels, the DNA samples were digested with the Hae III (for haplogroup A) and AluI (for haplogroups C and D) restriction enzymes (see Kemp et al., 2006 for further detail). The amplicons were incubated for at least 4 h at 37 °C and then run on a gel to check for digestion. Visual inspection of digestions for each sample reveals the presence or absence of the restriction sites characteristic of haplogroup A, C, or D. Haplogroup A is identified by a Hae III site gain at np 663, haplogroup C by an Alu I site gain at np 13261, and haplogroup D by Alu I site loss at np 5176. Haplogroup B, identified by a 9-base pair deletion in region V of the mtDNA genome, is confirmed by its shorter fragment length after 6% polyacrylamide gel electrophoresis. Haplogroup X was not tested for using restriction fragment analysis, as it would have been indistinguishable from M. Snow who conducted the analysis and belongs to haplogroup X. As all of the samples were identified as belonging to other haplogroups, and none of the sequences demonstrate any haplogroup X diagnostic markers when sequenced, this was not problematic to this research.

Fisher's exact tests were conducted using Genepop (Raymond and Rousset, 1995) to compare haplogroup frequency distributions from relevant pairs of populations. Interpretation of the results were completed with the sample size in mind, particularly how it may influence the statistical significance of haplogroup similarities between populations. Samples that could be assigned a haplogroup were sequenced to independently confirm their hap-logroup assignments and provide more detailed genetic characterization of similarities among these samples and those from other populations.

Sequencing was carried out by amplifying four overlapping 150-bp segments of the D-loop of the mtDNA control region from position 16045-16394, according to the Anderson reference sequence (Anderson et al., 1981). Samples were sequenced using an ABI 3130 DNA sequencer. The primers and parameters used throughout this process are those given in Kemp et al. (2007). The completed sequences were analyzed and concatenated using the Sequencher

and Mesquite (Maddison and Maddison, 2007) computer programs and their mutations were compared with those expected based on their haplogroup assignments to confirm those assignments. Samples assigned to haplogroup A and B are expected to exhibit most of their haplogroup's characteristic HVSI mutations, described in Table 5.

The DNA sequences containing derived mutations were included in a haplotype median joining (MJ) network constructed using default parameters in Network 4.6.1.1 (Bandelt et al., 1999) together with other samples cited in Table 5. The MJ Network predicts the order in which specific mutations occurred that differentiate samples from the basal haplotype(s) in the network. Sequences that share derived mutations appear on the same branch of the network and reflect close ancestor/descendant relationships.

2.4. Dietary stable isotope methods

Three different types of tissue were prepared and analyzed for this research: bone, enamel and dentin. The laboratory work was done in large part by a group of six students during a University of Toronto "Science Abroad" course, hosted by the University of Cape Town. To analyze the 513C and 515N in the collagen of bone and dentin, the collagen was extracted using standard procedures employed for many years in the stable isotope laboratory at the University of Cape Town (Sealy, 1997, 2010). Small chunks (with diameters of a few millimetres) of dentin or bone are surface-cleaned with fine sandpaper, decalcified in ca. 2% HCl, rinsed in distilled water, then treated overnight with 0.1 M sodium hydroxide to remove humic acids. If rootlets are present, they are removed from the wet collagen by hand, using fine tweezers. The collagen samples are soaked in distilled water until the pH remains unchanged, then freeze-dried. The dry collagen is weighed and the % yield (% of the starting weight of whole bone or dentin) calculated. Approximately 0.6 mg of dry collagen is weighed into a tin cup that is tightly folded to exclude air. Each sample is analysed in duplicate. The tin capsules are loaded into an automated elemental analyzer and combusted at 1020 °C. The resultant gases are introduced into a Thermo-Finnigan Delta Plus XP or a Delta V Plus mass spectrometer via a continuous flow (Conflo) inlet, using helium as the carrier gas. Standard deviations of repeated measurements of a homogeneous laboratory standard (Merck gelatine, n = 12) are less than 0.2& for both 513C and 515N. Results are reported relative to international standards Vienna PDB and AIR.

The 513C in tooth enamel carbonate was prepared and analysed as follows: For each tooth, a diamond-tipped drill was used to burr a line on either the buccal or lingual surface. This line extended from the occlusal surface to the cement-enamel junction, in order to sample tooth enamel formed over the duration of development of the crown. Care was taken to remove only enamel and not the underlying dentin. About 5 mg of the resultant enamel powder is collected and transferred to a microcentrifuge tube. In the stable isotope laboratory, the pre-treatment protocol used is that of Lee-Thorp et al. (1997), with slight modifications, as specified below. The powder is reacted with 1.75% sodium hypochlorite for 45 min, then washed three times with distilled water. It is treated with 0.1 M acetic acid for 15 min to remove diagenetic carbonates, then again washed three times with distilled water and freeze-dried. The pre-treated powders are analysed on a Finnigan-MAT 252 stable isotope ratio mass spectrometer coupled to a Finnigan Gas Bench II. Carbon dioxide is produced from enamel apatite by reaction with 100% phosphoric acid at 72 °C, then swept in a stream of helium through hygroscopic Nafion™ tubing to remove water, then through a glass capillary gas chromatograph, a second water trap, and finally into the mass spectrometer. The standard deviation of repeated measurements of homogeneous material (NBS 18, n = 12)

is lower than 0.2& for 513C, and frequently lower than 0.1 %„ Results are reported relative to the international standard Vienna PDB.

3. Results

3.1. Ancient DNA

All ten of the teeth that were sent to the MAL, were assigned to three of the six Native American haplogroups (haplogroups A, B, C), and for seven (70%) of the samples, this haplogroup assignment was replicated to ensure the extracted DNA was endogenous to the sample (Table 2). In no instance was there any contradiction between the first and second extraction. Due to the nature of this study and clear in-house confirmation of results, replication in a separate lab was not deemed necessary.

Five (50%) of the ten samples belonged to haplogroup A, four (40%) belonged to haplogroup B, and a single sample (10%) belonged to haplogroup C. The sample size is too small to allow for statistical comparison of these results to other populations in the region, as sampling bias would play a significant role.

It was possible to obtain replicated sequence data from the first hypervariable region for five (50%) of the 10 samples (GenBank accession numbers KF536720—KF536724). Four of these samples belonged to haplogroup A and were sequenced between np 16037— 16403. A single haplogroup B sample was sequenced between np 16058—16399. The haplogroup B sample carried a shorter sequence range due to a failure to replicate sequence data at the primer pairs that covered the final portion of the HVR.

The sequences were compared to other modern and ancient samples from populations in the region (see Table 3) according to which haplogroup they were assigned. No exact matches were detected between the samples presented here and other populations in the region.

Of the four haplogroup A samples, all carried an additional mutation at np16325 (Table 4). This mutation is shared with thirty-seven other samples from the region, including samples from: Norris Farms, Wisconsin Chippewa, Sisseton/Wapheton Sioux, Quapaw, Micmac, Cheyenne, Wisconsin Chippewa, Turtle Mountain Chippewa, and Minnesota Chippewa. This mutation is found within many different populations from a broad region, and several different language families, and therefore provides limited information about any population relationships the Iroquoian hap-logroup A samples may bear.

Among the haplogroup A samples, two samples carried mutations with np 16111: Wainfleet and Damiani B2 (Table 4). The mutation at np 16111 is typical of the Native American haplotype A2, which is one of the founding Native American lineages in the New World and found throughout many populations. Three samples exhibited derived mutations with the potential to reveal ancestor/descendant relationships when compared with other samples and all three sequences were confirmed by replication. The

Table 2

Summary of results from the ten tooth roots sampled for mitochondrial DNA.

Sample ID Haplogroup Replication Sequence GenBank accession number

Mantle B2 C

Mantle B7a B X

Mantle B7b B

Mantle B8 A X

Mantle B24 B X

Teston Rd 1 A X

Teston Rd 2 B X

Damiani B4 A

Damiani B2 A X

Wainfleet A X

Table 3

North American populations used for comparison with haplogroup A and B sequences generated. Data was collected from Shook and Smith (Shook and Smith, 2008), and Bolnick and Smith (2007).

Population (abbrev.) Sample name N Ancient Language Haplogroup

HIND Hind 3 Y A

(Glacial

Morse Morse 1 Y A

Orend Orendorf 4 Y A

MNC Minnesota Chippewa 4 Algonquian A

TMC Turtle Mountain 55 Algonquian A

Chippewa

WCH Wisconsin Chippewa 24 Algonquian A

Cheyen Cheyenne 9 Algonquian A

CheyenArap Cheyenne/Arapaho 2 Algonquian A

ChippSW Chippewa SW 2 Algonquian A

Delaware Delaware 1 Algonquian A

Iowa Iowa 1 A

Kikapoo Kikapoo 2 Algonquian A

Micmac Micmac 2 Algonquian A

Orcche Oklahoma Red 4 Iroquoian A

Cross Cherokee

Pawnee Pawnee 1 Caddoan A

Potowato Potawatomi 1 Algonquian A

Quapw Quapaw 1 Siouan A

Sac&Fox Fox 1 Algonquian A

Shawnee Shawnee 2 Algonquian A

StilcCh Stillwell Cherokee 2 Iroquoian A

SWSioux Sisseton/Wapheton 9 Siouan A

Wchippew Wisconsin Chippewa 1 Algonquian A

Ojibwa Ojibwa 1 Algonquian A

NorFarm Norris Farms 11 Y A

PyrLnsm 2 A

HW Hopewell Mound 14 Y A

Total 160

HIND Hind 4 Y B

(Glacial

Zimmer Zimmer Site 2 Y B

(Glacial Kame)

TMC Turtle Mountain 8 Algonquian B

Chippewa

MCH Minnesota Chippewa 4 Algonquian B

Cheyen Cheyenne 3 Algonquian B

Kickapoo Kikapoo 1 Algonquian B

ORCChe Oklahoma Red 3 Iroquoian B

Cross Cherokee

Pawne Pawnee 1 Caddoan B

Stilwch Stillwell Cherokee 7 Iroquoian B

Swsioux Sisseton/Wapheton 4 Siouan B

Ojibwa Ojibwa 1 Algonquian B

NorFarm Norris Farms 5 Y B

PyrLnsm 4 B

HW Hopewell Mound 3 Y B

Total 50

haplogroup A sample Mantle B8 carried additional derived mutations at np 16377, 16401 and 16403. Teston Road sample 1 also carried a derived mutation at 16051. The single haplogroup B sample, Mantle B24, was defined by the typical haplogroup B mutations at np 16189 and 16217, and also carried the additional mutations at np 16182 and 16223.

The MJ networks for haplogroups A and B are given in Figs. 3 and 4, respectively. These networks include the three samples cited above with derived mutations (Mantle B8 and Teston Road1 for haplogroup A and Mantle B24 for haplogroup B) together with reference samples cited in Table 3. The three derived mutations in Mantle B8 (16377, 16401 and 16403) define a branch of the

X KF536720

X KF536721

X KF536722

X KF536723

X KF536724

Mutations observed in sequences listed in Table 3. Numbering is based on the Anderson reference sequence (Anderson et al., 1981). Dashes (—) indicate the sample was the same as the Anderson sequence at that location. Base pair locations marked with an asterisk (*) as mutations characteristic of haplogroup A, and those marked with a pound sign (#) are characteristic of haplogroup B. Derived mutations are shown in bold italic. The C > T mutation at 16223 is diagnostic of haplogroup A, but derived in members of haplogroup B.

Sample ID Haplogroup 16051 16111 16182 16183 16189# 16217# 16223* 16290* 16319* 16325 16362* 16377 16401 16403

Anderson A C A T T T C C G T T C C C

Mantle B2 A — — — — — — T T A C C T T T

Teston Rd#1 A G — — — — — T T A C C — — —

Damiani B2 A — T — — — — T T A C C — — —

Wainfleet A — T — — — — T T A C C — — —

Mantle B24 B — — C C C C T — — — — — — —

haplogroup A network that the sample shares with two sequences from prehistoric sites in the Illinois River Valley: one sample from the Morse site and two samples from the Orendorf site. The Morse site represents the Red Ocher tradition of the early Woodland period, 3000-2500 ybp (Morse and Morse, 1964), while the Orendorf site is a much younger Middle Mississippian site. The derived mutation 16182 in the Mantle B24 sample defines a branch of the haplogroup B MJ network shared with two aDNA samples from the Hind site in SW Ontario (Essex County) belonging to the Glacial Kame tradition (Donaldson and Wortner, 1995) that is approximately coeval with the Red Ocher tradition in the Illinois Valley. The co-occurrence of these derived haplogroup A and B mutations in the sites described here are consistent with the hypothesis of a relationship between the earlier Red Ocher and Glacial Kame traditions (Ritzenthaler and Quimby, 1962; Donaldson and

Wortner, 1995). Studies of additional samples from these sites might clarify, especially if whole mtDNA genome sequences could be generated, might provide a test of this hypothesis.

3.2. Dietary stable isotopes

Nearly all samples yielded well preserved collagen. Indicators of collagen quality, i.e. percentages of C and N by weight, and atomic C:N ratios for all samples are given in Table 5. Three bone collagen samples yielded atomic C:N ratios outside the range 2.9-3.6. The isotopic results for those samples are considered unreliable (Ambrose 1990; van Klinken 1999) and have been discarded. The remaining C:N ratios ranged from 3.1 to 3.4, with an average of3.13, not far from 3.2, which is the composition of fresh collagen. Three molar roots (noted on Table 5) were sampled at both the upper and

K.V1502Q

16ÜB МЛ1501О

™=»0

Fig. 3. Haplotype MJ Network for haplogroup A samples reported in this and other studies. Mutations that differentiate adjacent haplotypes are shown along branches. Haplotypes are represented by yellow circles whose size is proportional to the sample size and median vectors are identified by small red circles. Sample abbreviations are those provided in Table 4.

Fig. 4. Haplotype MJ Network for haplogroup B samples reported in this and other studies. Mutations that differentiate adjacent haplotypes are shown along branches. Haplotypes are represented by yellow circles whose size is proportional to the sample size and median vectors are identified by small red circles. Sample abbreviations are those provided in Table 4.

lower ends (cemento-enamel junction and apex), to explore possible dietary differences during the years while the root was forming. Averaged values for those pairs are reported here. The reported isotopic results can be regarded with confidence, without concern for diagenesis. Isotopic values from 53 teeth are summarized by each of the seven sites in Table 6, Figs. 5 and 6.

3.2.1. Nitrogen

Mean 515N for dentin from all 53 humans in this study is

11.5 ± 1.0&, with a range from 9.8 to 13.4&. For 13 individuals, adhering alveolar bone was also analysed. Mean 515N for dentin is

11.6 ± 0.5&, and for bone 11.5 ± 0.7& (Table 6). Values for the two tissues are the same, within measurement error. The same pattern was noted at the Moatfield ossuary (van der Merwe et al., 2003a; van der Merwe et al., 2003b). Among the three pairs of within-tooth values, the differences were 0.5, 1.0 and 1.5&, from a maxillary M1, mandibular M2 and maxillary M2 respectively. The first two values, both from Staines Road, are within measurement error, given that dentin is not an entirely homogenous material. The M2 from Teston Road has values of 11.6& at the crown and 10.1 & at the apex, reflecting a dietary shift during mid-childhood.

515N values for dentin for all 53 individuals fall in the same general range as those reported for southern Ontario peoples prior to European contact (van der Merwe et al., 2003a). Candidate dietary proteins include beans, maize (although lacking the essential amino acid lysine), hunted meat, meat of domesticated dogs and fish from nearby lakes and rivers. 515N values of plants are very variable, depending on processes of nitrogen cycling in soils. In

relatively humid environments such as this, plants typically have 515N values close to zero, with a range of a few per mille either side. In Mesoamerica, cultivated beans have recently been reported to have a 515N value of 4.0 ± 2.6& (n = 13) (Warinner et al., 2013), with lowland crops marginally more enriched than highland. The most important source of meat was white tailed deer. Analysis of 515N in dentin from 45 archaeological white-tailed deer teeth from sites in the lower Great Lakes yielded a mean value of 5.9 ± 0.9& (Pfeiffer et al., 2013). This is in good agreement with the value of 5.5 ± 0.6& (n = 8) reported by Katzenberg (1989), who also reported archaeological dog bones to have mean 515N of 9.6 ± 0.2& (n = 8). Consumers typically have 515N 3-4& higher than their food (Hedges and Reynard, 2007) so that humans relying on beans, maize, deer and similar terrestrial foods might be expected to have 515N no higher than 10&, unless they were consuming a great many dogs. In fact, most of the humans analysed in this study have more positive 515N, almost certainly as a result of the consumption of fish. Bones of many types of fish have been recovered from archaeological sites; a few have 515N values comparable to those of deer, but the most desirable food species have values well over 8& (Katzenberg, 1989; van der Merwe et al., 2003b). Values for humans in this study are consistent with dietary reliance on diverse types of fish including salmon, trout, burbot (a form of cod) and walleye, all of which have 515N values above 10&.

Dentin 515N values are different among the various sites (Fig. 5), with the most positive mean values at Hutchinson and the least positive at Mantle indicating fish was clearly an important item of

Isotope values and collagen quality indicators for all tooth enamel, dentin and bone samples in this study. Values for dentin from UCT 13680,13687 and 13694 are the means of two samples of tissue from the cement—enamel junction and the tip of the root.

Site Name Identifier Tooth type UCT no. Sl3C s15n wt %C wt % N C/N C/N mtDNA

Tooth/Bone Enamel Dentin Bone Dentin Bone Dentin Bone Dentin Bone Dentin Bone

Mantle-cemetery B8 Lmand Ml l3660 -4.3 -l3.l 10.6 43.5 15.9 3.2 X

Mantle-cemetery B2 L max Ml l366l -4.6 -l0.8 10.6 4l.8 l5.3 3.2 X

Mantle-cemetery B24 R max Ml l3662 -3.4 -l0.6 10.4 35.5 l3.0 3.2 X

Mantle-cemetery B7a R max Ml l3663 -2.7 -ll.6 9.9 42.6 l4.8 3.4 X

Mantle-cemetery B7b R mand PM2 l3664 -5.4 -l2.7 l2.l 44.2 16.3 3.2 X

Mantle-village 435-180 surface L max Ml l3665 -5.7 -9.8 10.2 42.9 16.0 3.1

Mantle-village 465-155 layer 1 R mand Ml l3666 -3.3 -9.9 10.4 4l.8 l5.6 3.1

Mantle-village 435-180 Q5, fill L max M2 l3667 -2.9 -9.l ll.l 4l.7 l5.6 3.1

Mantle-village 485-240 trench L max C l3668 -7.4 -l0 10.5 42.4 l5.6 3.2

Mantle-village 495-129 fill L mand PM2 l3669 -3.2 -ll.6 l0.7 42.9 l5.4 3.3

Mantle-village 522-141 charcoal L max Ml l3670 -3.9 -9.5 9.8 42.6 l5.5 3.2

Mantle-village 335-160 fill RmaxPMl l367l -2.7 -8.8 10.5 4l.9 l5.5 3.2

Mantle-village 465-155 Q4, layer 1 R mand M3 l3672 -3 -l0.7 10.6 28.7 l0.7 3.1

Mantle-village 500-165 fill R max M3 l3673 -2.4 -ll.2 11.5 4l.5 l5.l 3.2

Mantle-village 499-133 fill R mand Ml l3674 -3.2 -9.2 l0 4l.6 l5.3 3.2

Staines Sample 1 R max M2 l3675/l3724 -5.7 -9.7 -ll.9 l2 l2 42.6 43.l l5.5 l4.8 3.2 3.4

Staines Sample 2 R max Ml l3676/l3725 -6.2 -ll.3 l2.6 43.8 9.4 l5.7 2.7 3.3 4.1

Staines Sample 3 R mand Ml l3677/l3726 -5.9 -l0.7 -ll.9 11.8 l2 4l.6 4l.6 l5.0 l4.6 3.2 3.3

Staines Sample 4 R max Ml l3678/l3727 -3 -9.2 -12.5 10.9 l2.8 4l.l 36.3 l5.0 l2.9 3.2 3.3

Staines Sample 5 R mand Ml l3679 -l.9 -ll.3 l2 42.l l5.4 3.2

Staines Sample 6 L max Ml l3680 -2.l -ll.4 l2.3 43.2 l5.7 3.2

Staines Sample 7 L mand Ml l368l/l3728 -3.2 -l0.9 -l4.l 11.3 10.6 42.2 42.0 l5.6 l4.8 3.2 3.3

Staines Sample 8 R mand Ml l3682 -3.l -ll.9 l2.7 4l.8 l5.3 3.2

Staines Sample 9 R mand M2 l3683 -2.9 -l3.4 11.6 43.l l5.4 3.3

Staines Sample 10 R max Ml l3684 -4.5 -l0.2 11.5 43.0 l5.5 3.2

Staines Sample 11 L mand Ml l3685 -3.8 -l0.5 11.4 28.l 10.0 3.3

Staines Sample 12 R max Ml l3686 -5.9 -ll.4 l2.9 43.3 l5.8 3.2

Staines Sample 13 R mand M2 l3687 -4 -9.6 11.3 36.8 l3.4 3.2

Staines Sample 14 R max Ml l3688 -3.4 -l3.2 ll.7 43.5 l5.8 3.2

Staines Sample 15 R max Ml l3689 -3.9 -l2 l2.5 43.0 l5.7 3.2

Teston Sample 1 L max M2 l3690 -3.9 -9.8 11.2 4l.8 l5.5 3.2 X

Teston Sample 2 R max M2 l369l -3.7 -l0.6 10.9 45.0 16.5 3.2 X

Teston Sample 3 L max Ml l3692/l37l6 -4.l -ll.7 -ll.9 11.6 ll 42.l 23.2 l5.2 8.1 3.2 3.4

Teston Sample 4 L max Ml l3693/l37l7 -5 -9.9 10.4 23.0 4.6 8.6 l.2 3.1 4.3

Teston Sample 5 R max M2 l3694/l37l8 -2.l -ll.3 -12.5 10.9 11.8 42.6 24.4 l5.8 8.6 3.2 3.3

Teston Sample 6 L max Ml l3695/l37l9 -2.l -ll -l2 l2.4 11.6 39.2 4l.8 l3.9 15.1 3.3 3.2

Teston Sample 7 L max Ml l3696/l3720 -4.2 -ll.5 -12.5 11.6 l0.7 43.2 25.0 16.0 8.8 3.2 3.3

Teston Sample 8 L max Ml l3697/l372l -5.4 -ll.6 ll.9 ll 11.3 46.3 32.0 16.5 11.5 3.3 3.2

Teston Sample 9 L max Ml l3698/l3722 -2.2 -ll 9.8 42.5 5.2 l5.l l.3 3.3 4.6

Teston Sample 10 L max M2 l3699/l3723 -4.6 -ll.4 -ll.8 10.9 10.9 35.5 35.0 l3.2 l2.6 3.2 3.2

Hidden Spring B1 L mand Ml l3700/l37l4 -3.9 -ll.4 -l3.3 l2.5 11.9 42.l 31.0 l5.6 l0.5 3.2 3.4

Hidden Spring B2a L max M3 l370l -3.3 -l0.7 l2.3 42.6 l5.7 3.2

Hidden Spring B2b R mand dm2 l3702 -3 -l2.8 10.8 42.l l4.4 3.4

Hidden Spring B3a L mand Ml l3703 -4.8 -ll.9 l2.2 43.2 16.2 3.1

Hidden Spring B3b L max M2 l3704/l37l3 -2.2 -l0.7 -l3 11.6 11.8 43.0 34.4 l5.8 12.1 3.2 3.3

Damiani B2 L max M2 l3705 -2.9 -9.7 l2 42.8 l5.6 3.2 X

Damiani B4 R max dml l3706 -3.8 -ll.9 l2.2 4l.5 l3.9 3.5 X

Hutchinson B1 L max M3 l3707 -2.8 -ll.4 l2.4 4l.l l5.2 3.2

Hutchinson B2 R max M3 l3708 -5.2 -l0.5 l3.8 35.4 l3.2 3.1

Hutchinson B2a R max M2 l3709/l37l5 -3.3 -l0.5 -12.2 l2 10.4 42.8 19.2 l5.7 6.5 3.2 3.4

Hutchinson F88 L mand M2 l37l0 -3.l -ll.6 l3.4 42.l l5.7 3.1

Hutchinson F92 R max M3 l37ll -2.5 -l3.4 l3.2 43.2 l5.5 3.3

Wainfleet Sample 1 R max Ml l37l2 -3.5 -9.4 11.5 34.8 l2.6 3.2 X

diet at Hutchinson but may not have been so at Mantle. This is a pattern consistent with the quantities of fish bone recovered from regional pre-coalescent period sites (AD 1300—1450) compared to later coalescent or post-coalescent sites (Birch and Williamson, 2013:104). Pre-coalescent sites in the same drainage as Mantle have significant percentages of fish bone in their faunal assemblages ranging from 33% to 50%. Similarly aged sites in the Don and Rouge River drainage systems to the west have 50—60% representation of fish. The fourteenth-century Robb site in the Rouge River system, near to the Hutchison and Staines Road sites, and possibly one of the contributing populations to the Staines Road ossuary, had an even higher representation of 70% fish. Fish represented 8% and 14% of the animal bone from the large, coalescent Draper and Parsons sites respectively and only 6% at Mantle. While there are

many taphonomic factors that might affect the survival and recovery of fish bone, these percentages for the coalescent and post-coalescent Mantle sites are consistent with these isotopic data and suggest that fishing was not as important an activity as it had been in pre-coalescent times. This was perhaps because considerably more effort was needed to grow sufficient maize to feed expanding and aggregated populations as well as to hunt deer, not just for meat but also for the thousands of deerskins necessary to clothe coalescent and post-coalescent populations (Birch and Williamson, 2013:111—118). Moreover, Mantle and its successor villages were situated at greater distances from Lake Ontario than earlier sites. Substantially more effort would have been needed to acquire and transport large quantities of processed lake fish northward along non-navigable waterways.

Values of d13C from enamel apatite and d13C and d15N from dentin collagen, summarized by individual sites and the total sample; mean, standard deviation in parentheses, and sample size provided.

Archaeological site

513C enamel apatite (&)

513C dentin collagen (&)

S15N dentin collagen (&)

Damiani (n = 2) -3.4 (0.6) -10.8 (1.6) 12.1 (0.1)

Hidden Spring (n = 5) -3.4 (l.o) -11.5 (0.9) 11.9 (0.7)

Hutchinson (n = 5) -3.4(1.1) -11.5 (1.2) 13.0 (0.7)

Mantle Cemetery (n = 5) -4.1 (1.1) -11.8 (1.1) 10.7 (0.8)

Mantle Village (n = 10) -3.8 (1.0) -10.0 (0.9) 10.5 (0.5)

Staines Road (n = 15) -4.0 (1.4) -11.1 (1.2) 11.9 (0.6)

Teston Road (n = 10) -3.7 (1.2) -11.0 (0.7) 11.1 (0.7)

Wainfleet (n = 1) -3.5 -9.4 11.5

Total (N = 53) -3.8 (1.2) -11.0(1.1) 11.5(1.0)

Moatfield (vdM et al., 2003) -4.2 (1.6) -11.3 (1.4) 12.6 (0.9)

3.2.2. Carbon

Mean 513C for dentin collagen from all 53 humans in this study is —11.0 ± 1.1 %„ with a range from —8.8 to —13.4&. Among the three pairs of within-tooth values, the differences were 0.2, 0.1 and 0.3& for dentin from a maxillary M1, mandibular M2 and maxillary M2 respectively. These are not significantly different. Comparison of 513C in dentin and bone is informative, since dentin forms at the time when the tooth develops, and therefore reflects diet in early life. Bone is slowly resorbed and reformed throughout life, so its isotopic values are likely to reflect diet somewhat later in life. Examination of the tissues from individuals in this study shows that dentin is more enriched: mean 513C for dentin is —10.9 ± 0.8&, and for bone —12.4 ± 0.7& (t (12) = 5.16, p < 0.05) (Table 6; values are for 13 individuals for whom both tissues were available). If these data are combined with similar comparisons for 11 individuals from the nearby Moatfield ossuary (van der Merwe et al., 2003a; van der Merwe et al., 2003b), the picture remains the same: dentin is more enriched than bone (t (23) = 4.28, p < 0.05). The only study from further afield that reports 513C values in both dentin collagen and bone collagen from the same individuals does not show this pattern (France and Owsley, 2013). The difference may derive from shifts in dietary emphasis during peoples' lifetimes, with greater reliance on C3 foods in later life. A larger sample size will be required to investigate this possibility in more detail. The dentin values of the 13 individuals in Table 6 do not diverge from

Fig. 5. Values for S15N (&) from dentin collagen from each archaeological site. The median value is indicated by the line inside each box; box extends from the 25th (bottom) to 75th (top) percentiles, whiskers extend to 2.5th percentile or minimum value (bottom) and 97.5th percentile or maximum value (top), circles and star indicate outliers.

Fig. 6. Values for S C (&) from enamel apatite and dentin collagen from each archaeological site. The median value is indicated by the line inside each box; box extends from the 25th (bottom) to 75th (top) percentiles, whiskers extend to 2.5th percentile or minimum value (bottom) and 97.5th percentile or maximum value (top), circle and stars indicate outliers.

those of the other forty, so the bone values can be used to compare with previously published bone-based studies.

Mean 513C for enamel from all 53 humans is —3.8 ± 1.2& with a range from —1.9 to —7.4& (Fig. 6). The mean value for 513C in enamel from Moatfield is very similar, at —4.2 ± 1.6&. The Staines Road site, which is thought to predate Moatfield, has mean enamel 513C of —4.0 ± 1.4& (n = 15). Hence, these values reflect no clear temporal trajectory in the magnitude of maize consumption. These values are generally about 1& less negative than bone apatite values from Iroquoian sites of the same era (Harrison and Katzenberg, 2003).

Values for 513C and 515N from dentin as well as 513C from enamel are very similar to those reported from the 44 teeth studied from the early fourteenth century Moatfield ossuary (van der Merwe et al., 2003b). Standard deviations are also quite similar, despite the broader temporal and spatial scope of the current study.

With regard to reconstructing the proportion of the diet contributed by maize, this study brings attention to the imprecision of using 513C of bone collagen as a proxy for maize intake if people also ate aquatic foods. Wild foods in this region are C3, as are beans and squash. Forty-five archaeological deer teeth (this study) yielded mean 513Cdentm of —22.4 ± 1.0&; since flesh is approximately 2& more negative than collagen, deer meat would be expected to have a value of about —24&. Domesticated dogs had mean 513C of —11.1 ± 0.9& (for bone collagen) (Katzenberg, 1989). Elevated 513C values in human tissues likely derive primarily from maize consumption, since this was a staple food, with some contribution from dog meat and fish. 513C of archaeological fish bones from this area range from —11.5 to —24.9&, although most lie between —16 and — 21 & (Katzenberg, 1989; van der Merwe et al., 2003a,b). To date, there is no evidence for the dietary dominance of any particular species. Fish are, therefore, enriched in 13C compared with other foods except maize. As a result, if maize intake is held constant, more fish will lead to more negative 513C values.

Carbon in bone and dentin collagen derives primarily from the protein component of the diet, with some contribution from carbohydrates and fats. Carbonate in tooth enamel is a better index of average 513C of the whole diet (Ambrose and Norr, 1993; Howland et al., 2003; Tieszen and Fagre, 1993). We might, therefore, expect fish to reflect more strongly in collagen than in enamel carbonate. Plotting the difference between 513Cdentm and 513Cenamel against 515Ndentin is inconclusive (y = 0.3288x + 8.8202, r2 = 0.098; one outlier omitted); there appears to be too much complexity in the system to track inputs in detail in this way. Nevertheless, as we

explore these questions more closely, it becomes apparent that some aspects of earlier estimates of proportional maize contribution to the diet need to be reconsidered.

4. Conclusions

The purpose of this research was to demonstrate the efficacy of one tooth per person as a protocol for addressing pertinent archaeological questions. Results generate insights into fundamental questions about the population background of Northern Iroquoians and their subsistence approaches. While the sample size for DNA analysis was small, the preservation was quite good, allowing for both haplogroup and haplotype assessment. While the samples did not provide a shared haplotype sequence with any modern population, additional sampling may allow for this type of finding. In general, the DNA results point to a great deal of diversity in the sampled sites, and the individuals tested are nested within the modern and ancient populations in this region that have previously been sampled. Derived mutations in several samples assigned to haplogroups A and B represent potentially tribal-specific mutations that may be identifiable in ancestors and/or descendants of the prehistoric populations sampled. Dietary isotopes show complex patterns of exploitation of the Great Lakes fishery and reliance on maize as a staple foodstuff throughout the region, with an apparent decrease in importance of fish in the diet of aggregated populations. Exploration of enamel apatite here and in future studies should help to clarify the timing and magnitude of shifts in this subsistence package.

Upon reflection, these results speak to the need for a population-based methodology in future research. Except in special circumstances (e.g., see Kemp et al., 2007) testing a small number of recent humans for solely mtDNA (and not the entire genome) provides an inadequate characterization of a population. The diversity seen in both the mitochondrial DNA results and the isotopic results is substantial. The Middle and Late Woodland periods were times of population movement, mixing and diversification in the Lower Great Lakes. The dynamism of the period raises the bar for researchers, making the precision and accuracy of our work ever more important.

From the one tooth per person, there is sufficient additional tissue to support the quantification of environmental isotopes such as strontium and oxygen, work that is underway with the samples studied here. Stronger relationships between researchers and descendant communities are crucial if we are to collectively learn about the past. While the discovery of human remains may be destabilizing in many regional contexts, it is important that field archaeologists pursue permission for some level of study, including the retention of tissue samples. We suggest that a single human tooth can be used to generate important information that would have been otherwise unavailable.

Acknowledgements

The Huron-Wendat Nation of Wendake, Quebec and the Six Nations of the Grand River, Oshwekan, Ontario, provided written consent for the retention of tooth samples and subsequent analyses. The collection, management and documentation of the samples, both human and deer, was provided by Dr. Crystal Forrest. We thank all the staff and faculty associated with the short course in principles and practice of stable light isotopes, offered by the Archaeometry Laboratory at the University of Cape Town. We acknowledge the hard work and intellectual contributions of students Arden Azim, Michelle Cameron, L. Elizabeth Doyle, Anne Drury, Heather Kristjanson, Stephen McIsaac and Magdalena Sobol, all participants in the University of Toronto Science Abroad

program. In the UCT laboratory, Vincent Hare, John Lanham and Ian Newton supervised the isotope analyses. Grants from the National Research Foundation of South Africa and the University of Cape Town contributed to laboratory costs. The ancient DNA work was supported by the National Science Foundation (MHS; BCS-08-50311). We offer special thanks to Dr. Beth Shook for providing comparative sequence data for analysis.

References

Ambrose, S.H., 1990. Preparation and characterization of bone and tooth collagen for isotopic analysis. J. Archaeol. Sci. 17, 431-451.

Ambrose, S.H., Norr, L., 1993. Experimental evidence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone collagen and carbonate. In: Lambert, J.B., Grupe, G. (Eds.), Prehistoric Human Bone: Archaeology at the Molecular Level. Springer-Verlag, Berlin, pp. 1 -37.

Anderson, S., Bankier, A.T., Barrell, B.G., Debruijn, M.H.L., Coulson, A.R., Drouin, J., Eperon, I.C., Nierlich, D.P., Roe, B.A., Sanger, F., et al., 1981. Sequence and organization of the human mitochondrial genome. Nature 290 (5806), 457-465.

Bandelt, H.-J., Forster, P., Roehl, A., 1999. Median-joining networks for inferring intraspecific phylogenies. Mol. Biol. Evol. 16, 37-48.

Birch, J., Williamson, R.F., 2013. The Mantle Site: An Archaeological History of an Ancestral Wendat Community. AltaMira Press, Lanham, Maryland.

Bolnick, D.A., Smith, D.G., 2007. Migration and social structure among theHopewell: Evidence from ancient DNA. Am. Antiq. 72, 627-644.

Dewar, G., Ginter, J.K., Shook, B.A.S., Ferris, N., Henderson, H., 2010. A bioarchaeological study of a Western Basin tradition cemetery on the Detroit River. J. Archaeol. Sci. 37 (9), 2245-2254.

Donaldson, W.S., Wortner, S., 1995. The Hind site and the Glacial Kame burial complex in Ontario. Ontario Archaeol. 59, 5-95.

France, C.A.M., Owsley, D.W., 2013. Stable carbon and oxygen isotope spacing between bone and tooth collagen and hydroxyapatite in human archaeological remains. Int. J. Osteoarchaeol. Early View. http://dx.doi.org/10.1002/oa.2300.

Harrison, R.G., Katzenberg, M.A., 2003. Paleodiet studies using stable carbon isotopes from bone apatite and collagen: examples from Southern Ontario and San Nicolas Island, California. J. Anthropol. Archaeol. 22 (3), 227-244.

Hedges, R.E.M., Reynard, L.M., 2007. Nitrogen isotopes and the trophic level of humans in archaeology. J. Archaeol. Sci. 34, 1240-1251.

Howland, M.R., Corr, L.T., Young, S.M.M., Jones, V., Jim, S., Van der Merwe, N.J., Mitchell, A.D., Evershed, R.P., 2003. Expression of the dietary isotope signal in the compound-specific delta(13) values of pig bone lipids and amino acids. Int. J. Osteoarchaeol. 13 (1-2), 54-65.

Katzenberg, M.A., 1989. Stable isotope analysis of archaeological faunal remains from southern Ontario. J. Archaeol. Sci. 16, 319-329.

Katzenberg, M.A., Schwarcz, H.P., Knyf, M., Melbye, F.J., 1995. Stable isotope evidence for maize horticulture and paleodiet in Southern Ontario, Canada. Am. Antiq. 60 (2), 335-350.

Kemp, B.M., Smith, D.G., 2005. Use of bleach to eliminate contaminating DNA from the surface of bones and teeth. Forensic Sci. Int. 154, 53-61.

Kemp, B.M., Monroe, C., Smith, D.G., 2006. Extraction and analysis of DNA from archaeological specimens. In: Glascok, M.D., Speakman, R.J., Popelka Filcoff, R.S. (Eds.), Archaeological Chemistry: Analytical Techniques and Archaeological Interpretation, pp. 78-98.

Kemp, B.M., Malhi, R.S., McDonough, J., Bolnick, D.A., Eshleman, J.A., Rickards, O., Martinez- Labarga, C., Johnson, J.R., Lorenz, J.G., James Dixon, E., Fifield, T.E., Heaton, T.H., Worl, R., Smith, D.G., 2007. Genetic analysis of early Holocene skeletal remains from Alaska and its implications for the peopling of the Americas. Am. J. Phys. Anthropol. 132, 605-621.

Lee-Thorp, J.A., Manning, L., Sponheimer, M., 1997. Problems and prospects for carbon isotope analysis of very small samples of fossil tooth enamel. Bull. Soc. Geol. France 168 (6), 767-773.

Maddison, W.P., Maddison, D.R., 2007. Mesquite: a Modular System for Evolutionary Analysis, 2.01 ed.

Malhi, R.S., Kemp, B.M., Eshleman, J.A., Cybulski, J., Smith, D.G., Cousins, S., Harry, H., 2007. Mitochondrial haplogroup M discovered in prehistoric North Americans. J. Archaeol. Sci. 34 (4), 642-648.

Malhi, R.S., Smith, D.G., 2002. Brief communication: haplogroup X confirmed in prehistoric North America. Am. J. Phys. Anthropol. 119 (1), 84-86.

Morse, D., Morse, P., 1964.1962 excavations at the Morse site: a Red Ocher cemetery in the Illinois Valley. Wis. Archaeol. 45, 79-98.

Pfeiffer, S., Williamson, R.F., Sealy, J.C., Forrest, C., 2013. Environmental Isotopes Indicate Interactions Among Iroquoians. "Trading Spaces: the Archaeology of Interaction, Migration and Exchange" University of Calgary, Alberta, Canada, November 7-9, 2013. Chacmool: Trading Spaces: the Archaeology of Interaction, Migration and Exchange University of Calgary, Calgary, Alberta.

Pihl, R.H., Monckton, S.G., Robertson, D.A., Williamson, R.F., 2008. Settlement and subsistence change at the turn of the first millennium: the view from the Holmedale site, Brantford, Ontario. In: Hart, J.P. (Ed.), Current Northeast Pale-oethnobotany II. State University of New York Press, Albany, pp. 151-172.

Raymond, M., Rousset, F., 1995. An exact test for population differentiation. Evolut. Biol. 49 (6), 507-518.

Ritzenthaler, R.E., Quimby, G.I., 1962. The red ocher culture of the upper Great Lakes and adjacent areas. Fieldiana Anthropol. 36 (11), 243-275.

Robertson, D.A., 2004. The Hutchinson site: A place to prepare for the final journey. Ont. Archaeol. 77/78, 95-120.

Schwarcz, H.P., Katzenburg, M.A., Knyf, M., Melbye, J., 1983. Analysis of Stable Isotopes of Carbon and Nitrogen in Prehistoric Southern Ontario Populations. Vancouver, pp. 1 -12.

Schwarcz, H.P., Melbye, J., Katzenburg, M.A., Knyf, M., 1985. Stable isotopes in human skeletons of southern Ontario: reconstructing paleodiet. J. Archaeol. Sci. 12, 187-206.

Sealy, J., 1997. Stable carbon and nitrogen isotope ratios and Coastal diets in the Later Stone Age of South Africa: a comparison and critical analysis of two data sets. Anc. Biomol. 1 (2), 131-147.

Sealy, J.C., 2010. Isotopic evidence for the antiquity of cattle-based pastoralism in southernmost Africa. J. Afr. Archaeol. 8 (1), 65-81.

Shook, B.A.S., Smith, D.G., 2008. Using ancient mtDNA to reconstruct the population history of Northeastern North America. Am. J. Phys. Anthropol. Preview 137(1), 14-29.

Thwaites, R.G., 1896-1901. The Jesuit Relations and Allied Documents. Burrows Brothers Company, Cleveland, OH.

Tieszen, L.L., Fagre, T., 1993. Carbon isotopic variability in modern and archaeological maize. J. Archaeol. Sci. 20 (1), 25-40.

Trigger, B.G., 1976. The Children of Aataentsic: A History of the Huron People to

1660. McGill-Queen's University Press, Montreal. van der Merwe, N.J., Pfeiffer, S., Williamson, R.F., Thomas, S.C., 2003a. Isotopic analysis and the diet of the Moatfield community. In: Willamson, R.F., Pfeiffer, S. (Eds.), Bones of the Ancestors: The Archaeology and Osteobiography of the Moatfield Ossuary. Canadian Museum of Civilization, Ottawa, Ontario, pp. 205-222. van der Merwe, N.J., Williamson, R.F., Pfeiffer, S., Thomas, S.C., Allegretto, K.O., 2003b. The Moatfield Ossuary: isotopic dietary analysis of an Iroquoian community, using dental tissue. J. Anthropol. Archaeol., 245-261. Van Klinken, G.J., 1999. Bone collagen quality indicators for palaeodietary and

radiocarbon measurements. J. Archaeol. Sci. 26 (6), 687-695. Warinner, C., Robles Garcia, N., Tuross, N., 2013. Maize, beans and the floral isotopic

diversity of highland Oaxaca, Mexico. J. Archaeol. Sci. 40 (2), 868-873. Warrick, G.A., 2008. A Population History of the Huron-Petun, A.D. 500-1650.

Cambridge University Press, Cambridge, U.K. Watts, C.M., White, C.D., Longstaffe, F.J., 2011. Childhood diet and Western Basin Tradition foodways at the Krieger site, southwestern Ontario, Canada. Am. Antiq. 76 (3), 446-472. Williamson, R.F., Steiss, D.A., 2003. A history of Iroquoian multiple burial practice. In: Williamson, R.F., Pfeiffer, S. (Eds.), Bones of the Ancestors: The Archaeology and Osteobiography of the Moatfield Ossuary. Canadian Museum of Civilization, Ottawa, Ontario, pp. 89-132.