International Journal of Osteoarchaeology

Int. J. Osteoarchaeol. 21: 669–678 (2011)

Published online 13 May 2010 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/oa.1173

Dirty Teeth and Ancient Trade: Evidence

of Cotton Fibres in Human Dental
Calculus from Late Woodland, Ohio
S. H. BLATT,a* B. G. REDMOND,b V. CASSMAN c AND P. W. SCIULLI a
a
Department of Anthropology, The Ohio State University, OH, USA
b
Department of Archaeology, The Cleveland Museum of Natural History, OH, USA
c
Department of Art Conservation, University of Delaware, DE, USA

ABSTRACT Analysis of ancient human dental calculus for the presence of inclusions related to diet and dental health has
been overlooked in anthropological literature. Small particles of archaeobotanical debris, which would
otherwise not be preserved in the archaeological record, can become incorporated into unmineralised
plaque on teeth during mastication and oral manipulation. When plaque mineralises into calculus, debris
is preserved in situ. Samples of dental calculus (n ¼ 18) were collected from the Danbury site (33OT16) in
Ottawa County, Ohio and viewed under a scanning electron microscope for inclusions. Analysis yielded a
variety of noticeable inclusions, including mineralised bacteria, calcium-phosphate crystalline structures and
numerous phytoliths. Here we report the first evidence of fibres consistent with cotton (Gossypium spp.)
embedded in the dental calculus from the Late Woodland component (900–1100 AD) of the Danbury site.
Prehistoric cotton has not been previously documented in Ohio. The distinct morphology of the Danbury cotton
and its presence in the Late Woodland component at Danbury suggests long-distance interaction at a time in
Ohio when movement of exotic goods appeared to have diminished. These microscopic remains provide
insight into paleoethnobotanical history of ancient Ohioans and attest to how analysis of dental calculus could
be used to supplement other paleodietary and archaeological analyses. Copyright ß 2010 John Wiley & Sons,
Ltd.

Key words: dental calculus; cotton; SEM; Late Woodland

Introduction are preserved in situ. The physical inclusion of bacteria

Dental calculus is an overlooked source of data in
anthropological studies. Calculus provides a unique
calcified reservoir of mineralised bacteria (Vander-
meersch et al., 1994; Arensburg, 1996; Hershkovitz
et al., 1997) and dietary debris (Armitage, 1975;
Dobney & Brothwell, 1986; Middleton, 1990; Mid-
dleton, 1992a; Middleton, 1992b; Cortello & Pochet-
tino, 1994; Middleton & Rovner, 1994; Fox et al., 1996;
Gobetz & Bozarth, 2001; Boyadjian et al., 2007;
Blondiaux & Charlier, 2008) seldom preserved else-
where in the archaeological record. Small particles of
archaeobotanical debris can adhere to and become
incorporated into the unmineralised plaque on teeth
during mastication and oral manipulation. When
plaque mineralises into calculus, debris and bacteria
* Correspondence to: Department of Anthropology, The Ohio State
University, OH, USA.
e-mail: blatt.22@osu.edu
and archaeobotanical debris in the matrix of dental
calculus is direct evidence for the contemporaneity of
both structures, because the mineralisation of dental
plaque occurs only in the presence of saliva during
the life of an individual (MacPhee & Cowley, 1975).
Such microscopic remains provide insight into the
dental health and paleoethnobotanical history of
ancient populations. Here we report the first findings
of preserved cellulose fibres consistent with the
characteristics of cotton (Gossypium spp.) embedded
within the matrix of the dental calculus of four
Late Woodland individuals from the Danbury site
(33OT16), Ohio.
The genus Gossypium is composed of a diverse group
of species (more than 50) with a global distribution and
marked morphological and cytogenetic diversity.
Gossypium constitutes a monophylytic taxon belonging
to the family Malvaceae. Wendel et al. (1992) and
Wendel (1995) provide ample evidence to establish
that New World tetraploid cottons are allopolyploids

Copyright # 2010 John Wiley & Sons, Ltd. Received 31 July 2009
Revised 25 February 2010
Accepted 17 March 2010
670
containing one genome similar to those found in the
Old World. Of the four domesticated cotton species
found worldwide, two have been domesticated in the
New World. Gossypium hirsutum is widely distributed in
Central and South America, the Caribbean and even
reaches distant islands in the Pacific (Solomon Islands
and Marquesas). Gossypium barbadense has a more
southerly indigenous range, centred in northwestern
South America but overlaps with the range of
G. hirsutum in the Caribbean. It is widely accepted
that, of the two varieties of pre-Columbian domesti-
cated cotton in the Americas, G. barbadense (commonly
known as Pima cotton or Egyptian cotton) was first
cultivated in Peru as early as 2500–1750 BC and then
introduced into the Caribbean, whereas G. hirsutum (or
Upland cotton) was first cultivated in Mesoamerica
between 300 BC and 500 AD and is first seen in the Gila
and Salt River regions of Arizona around 500 AD (Kent,
1983; Lee, 1984; Teague, 1998). Successful cultivation
of cotton requires a long frost-free period (160 days)
and moderate rainfall, usually from 600 to 1200 mm. In
general, these conditions are met within the seasonally
dry tropics and subtropics in the Northern and
Southern hemispheres (Wendel, 1995).
While there is some archaeological evidence of
contact between the Caribbean islands and southern
Florida during the Late Prehistoric period (1000–1650
AD), there is no evidence of pre-Columbian cotton in
eastern North America (Bullen, 1974; Seidemann,
2006). Knowledge of the role and extent of cotton
exchange between South-western Pueblo and Plains
groups and others has been limited due to a lack of
comprehensive archaeobotanical evidence. In fact,
there is no evidence of prehistoric cotton cultivation or
exchange east of the Spiro Mound site in LaFlore
County, Oklahoma (King & Gardener, 1981). There is
a similar lack of evidence for the movement of cotton
northward from known Preclassic through Postclassic
source areas in the Gulf lowlands of northeastern
Mexico (northern Veracruz and Tamaulpas) (Stark
et al., 1998; Evans, 2004).
Though societies in Ohio were producing yarns
made from twined native plant fibres at least as early as
the Middle Woodland (100 BC–500 AD), it was
unexpected to find multiple fibres with the physical
attributes of mature and immature Gossypium spp.
embedded in the dental calculus of four Late Woodland
(900–1100 AD) individuals from the Danbury site. The
cultural implications of this discovery are far-reaching
and could suggest long-range interaction between
either South-western Pueblo groups, or northern
Mesoamerican societies, and groups living in northern
Ohio during the Late Woodland period, a time in Ohio
Copyright # 2010 John Wiley & Sons, Ltd.
S. H. Blatt et al.
when trade of exotic goods is generally regarded as
having been attenuated.
Materials and methods
Dental calculus samples were obtained from a total of
18 teeth from 8 Danbury site individuals (05-02A, 05-
02B, 05-02C, 05-06, 05-07, 06-01B, 06-02A, 06-03A)
(see Table 1). The Danbury site is a multicomponent,
pre-Columbian habitation site located on the south
shore of Lake Erie in Ottawa County, Ohio (Red-
mond 2005, 2006a, 2006b, 2007) (Figure 1). Inves-
tigation of the site was first supervised by the Ohio
Valley Archaeological Consultants (OVAC) in 2003
during the construction of a housing development
and then from 2004 to 2007 by the Department of
Archaeology of The Cleveland Museum of Natural
History (Redmond, 2005, 2006a, 2007). Forty-two
burial features (BF), containing 152 individuals, have
been excavated from the Danbury site since 2003.
Accelerator mass spectrometry (AMS) dating of
carbon and bone collagen samples indicate habi-
tation at the site during the Late Archaic (2900–2500
BC), Early Woodland (800–1000 AD), Late Woodland
(900–1100 AD), and Late Prehistoric (1450–1550 AD)
periods. Burial treatments at the site included
primary flexed, semi-flexed, extended and bundled;
however, many of the burials appear to be secondary
interments combining multiple individuals (Red-
mond, 2006b).
A wide range of non-perishable burial goods were
found at the Danbury site, the majority being marine
shell beads, primarily disk beads and marginella shell
beads (Redmond, 2006a, 2006b). Prominent among
the marine shell artefacts and associated with the three
individuals from BF 05-02, are portions of two drilled
pendants made from a lightning whelk shell (Busycon
sinistrum) (Figure 2), a species whose ranges extend from
the Gulf of Mexico to the southern Atlantic (Trubitt,
2003; Redmond, 2006b). One half of a whelk pendant
was found in the mouth of individual A from BF 05-02,
and the matching half of this specimen was found
adjacent to the upper lumbar vertebrae of individual B.
The two halves of this first pendant tightly fit together,
suggesting they were broken only shortly before
interment. There are also three drill holes in the
specimen which display slight wear (Redmond,
2006b). The second pendant (Figure 2), found between
the left humerus and sternum of individual C, was made
from the outer whorl of large whelk had ground edges,
a single drill hole, and two lateral notches (Redmond,
2006a).
Int. J. Osteoarchaeol. 21: 669–678 (2011)
Dirty Teeth and Ancient Trade 671

Table 1. Specimen data including BF (burial feature) number, sex (P ¼ pelvis; Pb ¼ pubis; O ¼ other) and age (P ¼ pelvis; Pb ¼ pubis;
A ¼ auricular; O ¼ other; D ¼ dental stage; F ¼ epiphyseal fusion), stature in cm, body mass in kg, calculus score (0 ¼ absent; 1 ¼ mild,
covering less than one third of the crown; 2 ¼ moderate, covering from one third to two thirds of the crown; 3 ¼ severe, covering more
than two thirds of the crown), location of the calculus on each tooth (L ¼ lingual surface; B ¼ buccal surface) and the presence (indicated
by ‘x’) of inclusions such as bacteria, calcium phosphate (CaP) crystals, phytoliths and/or starch grains, unidentified fibres and
Gossypium sp. fibres within the calculus specimens

BF No. Sex Age Stature Mass Tooth Score Loc. Inclusions

(cm) (kg)
Bacteria CaP Phytoliths/Starch Unid. Gossypium
Crystals Fibre Fibre

05-02A F (P) 42–52 (A) 156.8 63.4 LRP4 1 L x x

05-02B M (P) 29–33 (A) 172.9 63.1 URM1 1 B
URM2 2 B x x
URP3 1 B x x
URI2 1 B
URC 1 B x
05-02C M (P, Pb) 35–40 170.7 64.1 ULI2 1 L x x
(Pb, A)
URM2 1 B
05-06 F (P, Pb) 25–30 159.3 59.0 LLM2 1 B x
(Pb, A)
ULM1 1 B
LRC 1 L x
LRI2 1 L x
LLC 1 L x
05-07 M (O) Adult (O) — — LRP3 1 L
1 L x
LM1 2 L x x x
06-01B F (P) 12–14 — — ULP4 1 B
(D, F)
06-02A M (P) 13–14 — — URI1 1 B
(D, F)
06-03A M (P, Pb) 50 þ (Pb, A) 163.2 64.1 LI2 1 L x x

Figure 1. Location of the Danbury site (33OT16) in northern Ohio.

Copyright # 2010 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 21: 669–678 (2011)
672
Figure 2. Lighting whelk shell pendants from BF 05-02. (A)
Reconstructed shell fragments from pendants associated with
individuals A and B; (B) pendant associated with individual C
(note the notching to the right of the drill hole).
A total of 186 disk beads and 51 whole marginella
beads also distinguished the individuals from BF 05-02.
Many of these were arranged around the head of
individual B and head and pelvic region of individual A.
Nearly all the disk beads appear to have been
manufactured from fragments of either the outer body
whorl or columella of whelk (Busycon spp.). The
marginella beads are perforated, complete shells and
were found distributed across the body of individual C
(Redmond, 2006a).
Accelerator mass spectrometry dating of bone
collagen samples from individuals B and C from BF
05-02 produced two sigma date ranges of Cal. 880–
1010 AD (Beta-207989) and Cal. 880–1020 AD
(Beta-207990), respectively. These dates place the
individuals and the associated shell artefacts in the Late
Woodland period, a time when such artefacts appear to
be rare in the Ohio region (Seeman & Dancey, 2000: p.
601; Stothers & Abel, 2002: p. 76). Consequently, the
whelk shell burial goods from the Danbury site provide
some of the best documented evidence of exchange
between the Late Woodland societies of northern
Ohio and the Gulf and/or southern Atlantic Coast.
Whelk burial artefacts have also been reported from
the nearby Libben site, a large Late Woodland period
cemetery (Phillips & Brown, 1978; Prufer & Shane,
1976), but have yet to be described in detail.
The bone collagen samples from individuals B (Beta-
207989) and C (Beta-207990) from BF 05-02 were also
subjected to stable carbon and stable nitrogen isotope
analysis. Stable carbon isotope results greater than
Copyright # 2010 John Wiley & Sons, Ltd.
S. H. Blatt et al.

15.0% indicate significant maize consumption and
in the Great Lakes region in particular, stable nitrogen
levels above 10.0% signal the consumption of
significant amounts of some freshwater fish and
waterfowl (Schurr & Redmond, 1991). Individuals A
and B exhibited stable nitrogen isotope results ranging
from 11.3 to 13.6% and stable carbon isotope results
ranging from
14.5 to
12.4%. The results indicate
that the Late Woodland inhabitants of the Danbury
site consumed significant amounts of maize and aquatic
fauna (Redmond, 2006a).
Recording of additional dietary information from the
Danbury Site, aside from the isotope data, is still
underway. A few maize kernels and wild fruit seeds
have been noted, but few botanical remains other than
wood charcoal have as yet been identified (Redmond,
2006b). Based on the small number of projectile points
recovered from the site (which were primarily derived
from locally available Pipe Creek chert, which
outcrops 15 km southeast of Danbury), it appears that
hunting or the butchering of game animals was not a
primary focus of subsistence activities conducted at
Danbury. Fishing and the processing of other raw
materials were, most likely, the main tasks conducted
there (Redmond, 2006b).
All human skeletal remains were analysed according
to Standards (Buikstra & Ubelaker, 1994). Sex was
assigned to each individual from pelvic morphology
(Phenice, 1969). Age was assigned to each individual
from auricular surface morphology (Lovejoy et al.,
1985) and when possible combined with the
morphology of the pubic symphyses (Meindl et al.,
1985). Subadults were aged according to their stage of
dental development and epiphyseal union (Sciulli,
2007). Visible pathologies or trauma were recorded
and stature (cm) (Sciulli & Hetland, 2007) and weight
(kg) of each individual was estimated (Auerbach & Ruff,
2004) (see Table 1). All of the human skeletal remains
and burial goods from Danbury have been or will be
repatriated.
Within the Danbury skeletal sample, dental calculus
was identified as a light buff to chocolate-brown
stained encrustation above or below the gingival line of
each tooth (Figure 3) and recorded by quantity and
location according to Brothwell’s (1981) index
(Table 1). A modified protocol of Middleton & Rovner
(1994) and Ciochon et al. (1990) was followed to
collect the calculus, adapted to the characteristics of
human dental calculus. The teeth were first cleaned
with a soft toothbrush in order to eliminate possible
adhering debris on the surface of the calculus. Pieces of
calculus were removed with the aid of a dental scalar
directly into individually sterilised vials. Each specimen
Int. J. Osteoarchaeol. 21: 669–678 (2011)
Dirty Teeth and Ancient Trade
Figure 3. Molars from BF 05-02B with varying degrees of cal-
culus encrustations on the lateral aspect of the posterior man-
dibular dentition.
was diluted with 5% hydrochloric acid (HCl) to
remove surface carbonates and postmortem contami-
nation. Specimens were allowed to digest for 2 h in
gamma-sterilised microcentrifuge vials stored within a
desicator cabinet.
After drying, specimens were affixed to aluminium
stubs with sterile tweezers and double-sided carbon
tape. Specimens were coated with two layers of carbon
and then stored in a humidity-controlled desiccator
cabinet until examination. Each specimen was
examined using a JEOL JSM 820 scanning electron
microscope (SEM) under 15Kv according to the
criteria suggested for biological specimens (Goldstein
et al., 2003). Only elements embedded within or clearly
coated with calculus were considered in situ, originating
from calculus calcification during the life of the
individual. Specimens were scanned at both low and
high magnification, with most inclusions being visible
beginning at 300. Micrographs were obtained using
Oxford Instrument’s INCA platform for SEM imaging.
SHB collected and prepared the calculus samples and
produced SEM images of all specimens.
Fibres were identified via SEM images according to
their distinct characteristics. Cotton seed fibres begin
as elongated cells that grow on seed coats. As the fibres
mature, the cell walls collapse inward to form
convolutions, giving cotton fibres their unique,
ribbon-like appearance with raised edges and minimal
convolutions (Jakes et al., 1994; Teague, 1998;
D’Orazio et al., 2000; Müller et al., 2004). Mature
cotton fibres display well-developed convolutions,
however, as the result of disease or pests, unfavourable
climatic conditions, or early harvest, the fibre in the
cotton boll (a segmented pod containing seeds from
Copyright # 2010 John Wiley & Sons, Ltd.
673
which cotton fibres grow) may be terminated before it
is fully developed, resulting in immature or ‘dead’ fibres.
Immature fibres are characterised as being devoid of or
having minimal convolutions, having little rigidity,
easily creasing and causing uneven dyeing problems
(David & Pailthrope, 1999).
Results
The SEM analysis of the calculus specimens from
the entire sample yielded a variety of noticeable
inclusions. These included the presence of both
mineralised bacteria (streptococci) and the outline of
degenerated bacteria (bacillus-type), calcium-phos-
phate crystalline structures, numerous identifiable and
unidentifiable phytoliths (amorphous, grass and non-
grass types), as well as several starch grains, a few
unidentified fibres that appear to be baste fibres, and
the presence of fibres consistent with cotton (Blatt,
2007 unpublished master’s thesis). The presence of
crystal structures and their lack of involvement with
any adjacent bacteria suggests that many of the
calculus samples represent immature calculus, and
thus their in situ inclusions, were mineralised less than
6 months prior to the deaths of these individuals
(Mandel, 1990). The presence of these inclusions as
well as the age and sex of the associated individuals,
degree of calculus and calculus location are provided
in Table 1. Due to the disaggregated nature of the
calculus matrix, not all specimens yielded identifiable
inclusions.
Evidence of cotton was recorded in the dental
calculus samples of four individuals from two burial
features (BF) 05-02 and BF 05-07 (Figure 4). The first,
BF 05-02 was a primary burial containing three adults
extended side by side. The eastern-most individual A
(BF 05-02A) was a female, approximately 45–52 years
old, stature of 156.8 cm and body mass of 63.4 kg, with
extensive degenerative joint disease of the scapula
glenoid and vertebrae (Auerbach & Ruff, 2004; Sciulli &
Hetland, 2007). Individual B (BF 05-02B), a male,
approximately 29–33 years old with a stature of
172.9 cm and body mass of 76.3 kg, lay to her left
(Auerbach & Ruff, 2004; Sciulli & Hetland, 2007).
Individual C (BF 05-02C) was also a male, approxi-
mately 35–45 years old at death with a stature of
170.7 cm and body mass of 64.1 kg (Auerbach & Ruff,
2004; Sciulli & Hetland, 2007). The human remains
from BF 05-07 were poorly preserved, but represented
those of an adult male; age, stature and body mass were
indeterminate due to the high degree of fragmentation
(Redmond, 2006a).
Int. J. Osteoarchaeol. 21: 669–678 (2011)
674
Figure 4. SEM micrographs of cotton fibres embedded in dental calculus from: (A) BF 05-02B; (B) BF 05-02A; (C) BF 05-02C; (D) BF 05-07.
The fibres from the above individuals were identified
via SEM images as being consistent with mature
and immature cotton fibres via a blind test by Vicki
Cassman and Kathleen Keifer, textile conservators at the
University of Delaware. As shown in Figure 4, the fibres
display the characteristic flattened, ribbon-like and
elongated distinct appearance of cotton seed fibre with
raised edges, rolled in a helicoidal manner and mostly
underdeveloped convolutions. Further, the aggregate of
fibres lack the typical Z or S torsion consistent with
fibres spin into yarns. Unfortunately, due to the calcified
nature of the fibres (and the fact they were in situ
embedded within the calcium-rich matrix of the calculus
rather than isolated), chemical microanalysis of their
precise cellulose content as well as assessment of the
relative thickness of the lumen and cell walls of the fibres
were precluded. A comparative collection of 34 fibre
genera common to prehistoric Ohio including varieties
of paw, dogbane, Indian hemp, willow, milkweed, nettle
and cattail was consulted (Jakes et al., 1994). Fibres from
certain genera of the plants in this collection display
distinct characteristics that were deemed useful for their
Copyright # 2010 John Wiley & Sons, Ltd.
S. H. Blatt et al.
specific identification, including specialised treatments
that can be microscopically observed via unique
absorbance bands or remaining extraneous plant cells.
None of the genera or treatment types in this collection
thoroughly resembled the calcified fibres from Danbury.
Discussion
Archaeological evidence of premature harvesting,
resulting in immature cotton fibres, increases in
frequency at sites with increased distance from the
equator, especially above of the Mogollon Rim and
Colorado Plateau of the southwest. Based on ethno-
historical evidence, Hill (1998) concluded that cotton
was cultivated primarily in the Rio Grande valley prior
to Spanish contact. According to Grossman (1873: p.
419) the Pima ‘raise. . .a small supply of very inferior
short cotton’. Forde (1963: p. 233) suggested that a
similar cotton crop produced by the Hopi ‘. . .reflects
the severity of its environment’. Although severe
environmental conditions or early harvesting due to a
Int. J. Osteoarchaeol. 21: 669–678 (2011)
Dirty Teeth and Ancient Trade
short growing season could retard development,
leading to immature convolution, it is difficult to
assess if this trait is an effect of environment or the
particular variety of cotton (Bohrer, 1962; Kent, 1983;
Teague, 1998).
The Hohokam of the Sonoran Desert had a well-
established cotton industry before 1000 AD, which could
be attributed to their construction of intricate irrigation
canals about 1500 years ago, preventing some of the
difficulties others had when growing cotton in the
region. Salinity is a concern for irrigated fields, but
Fryxell (1979) disregards this, describing native cottons
as having higher salt tolerance than modern cottons.
This suggests that prehistoric cotton from the South-
western United States could be a unique strain with
immature qualities (or that there is cotton experimen-
tation within the culture that is growing this cotton).
The immature length and underdeveloped convolutions
of the cotton found at the Danbury site is similar to that
of the Southwest and regions north of the equator.
Historically, cotton in the Southwest was traded as raw
material, yarn and finished fabric, however, archae-
ologically it appears that trade focused on the raw
material (Teague, 1998). There is evidence, based on
regional fabric patterns, that long distance trade existed
between groups in the Southwest (Teague, 1998);
however, no previous evidence exists of the pre-
Columbian cotton trade extending into the Midwest.
As noted earlier, the nearest presence of prehistoric
cotton to the Danbury site appears to be one textile
fragment tentatively identified as cotton from the Spiro
Mound sites in Leflore County, Oklahoma, occupied
from 850 to 1450 AD (King & Gardener, 1981). The
Spiro site is considered one of the most important
locations through which exotic goods were funnelled,
and there have been many artefacts from Spiro
indicating long-distance trade. Most of the shell cups
found at Spiro were made from lightning whelks
(Busycon sinistrum) from the Gulf of Mexico. Also, the
recent identification of Olivella beads (O. dama native
to the California coast and found at sites throughout
the Great Basin and southwest) from the Spiro Mounds
site similarly suggests the intensification of trade
between Puebloan and Plains peoples during the Late
Woodland and Late Prehistoric periods (Kozuch,
2002). The lithics from the Spiro Mounds site were
made of chert from the Cahokia, Illinois region, and
obsidian from Hidalgo, Mexico; and copper from the
Great Lakes region was also among the exotic artefacts
(Barker et al., 2002; Kozuch, 2002). It is possible then
that the cotton and lightning whelk shells found at the
Danbury site were obtained from a large exchange
complex like that of Spiro.
Copyright # 2010 John Wiley & Sons, Ltd.
675
Although the exact origins of the cotton fibres from
Danbury are uncertain, their presence suggests long
distance trade into northern Ohio during the Late
Woodland period. Both the whelk pendants (Busycon
sinistrum) and cotton fibres seem to share western
Gulf origins. The fact that these exogenous goods were
found together amongst a few select burials at Danbury
suggests the importance of the site as a semi-permanent
settlement of rotating and interacting populations.
How the cotton fibres became embedded into the
dental calculus of the Danbury individuals is unclear,
but several explanations are possible. Although
evidence of interproximal dental wear from western
Great Basin populations suggests the use of teeth as
tools for weaving (Larsen, 1985), there is no such
evidence observed from the Danbury teeth, nor are
there any southwestern ethnohistoric accounts of using
teeth while spinning cotton fibres (Teague, 1998). In
the southwest there are various spindle whorls,
hooked-sticks and looms for spinning yarns and
weaving fabrics, but no such evidence has been found
at the Danbury site for producing threads (Teague,
1998). However, the absence of these artefacts does
not wholly eliminate the possibility of inclusion of
fibres in dental calculus during spinning or fibre
preparation, which is not visible in the archaeological
record (Teague, 1998). Fluffing (separating the fibres)
is an important part of spinning preparation and
without modern carding tools, pulling fibres while
anchoring with ones teeth is an easy solution. It is
equally possible that while splitting plied cotton fibres
by hand the fibres were anchored in the mouth and
became incorporated into calculus. Most likely their
presence represents the reprocessing of cotton fibres
for the production of textiles or fishing line or nets.
The association of such exotic marine pendants as
the lightning whelk shells coupled with evidence of
cotton use also generates questions as to the relative
prestige of individuals A, B and C of BF 05-02 and BF
05-07, since such goods were absent in other burials.
Certainly the link between cotton as an exotic fibre, its
use as a tool for improved fishing technologies and an
important source of protein might be a link to social
status, though further investigations are needed to
document the extent to which these objects reflect
status within this population.
Conclusions
The analysis of human dental calculus provides an
innovative and direct assessment of ancient diet and
gives additional evidence when combined with stable
Int. J. Osteoarchaeol. 21: 669–678 (2011)
676
isotope studies. Calculus analysis has advantages over
stable isotopes of being non-destructive, less expensive
and generally more precise with regard to identifying
specific plant materials.
Though in practice it would be ideal to assess both
the physical and chemical nature of all archaeological
fibres, we hope that the data presented here will alert the
anthropological community to the value of microscopic
analysis of dental calculus in supplementing our
understanding of paleoethnobotanical histories. We
anticipate that additional studies of this kind will result
in the identification of cotton at sites outside the
Southwest, and yield more detailed results beyond those
of this report. In light of these new data, further research
is definitely needed to gain a more comprehensive view
of the spheres of interaction and trade of exotic goods in
prehistoric North America. While the identification of
plant fibres from archaeological contexts is essential for
understanding ethnobotanical history, identification is
limited by the wide range of plant material potentially
in use, the absence of plant parts used for standard
identification and the alteration of fibres from proces-
sing, use and degradation within the burial environment.
The methods used in this paper, however, provide
the potential for analysis of plant materials that would
otherwise be digested or lost from employing the
multiple washes or higher hydrochloric acid concen-
trations suggested by other authors. Analysis of dental
calculus is an ideal complementary method to stable
isotopic reconstruction, because it provides infor-
mation about foods eaten a short time before death
(from days to months), as opposed to the long-term
perspective on diet yielded by isotope analysis. Dental
calculus analysis offers an inexpensive and relatively
accurate means of assessing the herbaceous portion of
ancient diet, but like most paleodietary techniques, it
should be combined with other methods.
What is clear is that the cotton fibres discovered
within dental calculus from the Danbury site are not
indigenous to Ohio and were introduced to the Late
Woodland component of this site from an unknown
south-western source. The interaction between north-
ern Ohio and southern coastal regions is supported by
the evidence of both cotton and marine shell exchange
between these groups. The whelk shells do not point to
the Southwest as a source per se, rather they point to
another avenue of long distance exchange (from lower
Great Lakes to the Gulf of Mexico) which was most
likely channelled, along with cotton, through an
intermediary like Spiro. Although the precise variety of
cotton found at the Danbury site and the circumstances
behind such an undertaking are unclear, the ramifica-
tions of such broad-based exchange in prehistoric
Copyright # 2010 John Wiley & Sons, Ltd.
S. H. Blatt et al.
North America could be far-reaching and is deserving
of further examination.
Acknowledgements
Partial funding for this project was provided by the
Cleveland Chapter of the Archaeological Institute of
America. Scanning electron microscopy was performed
at the Microscopic and Chemical Analysis Research
Center (MARC), Department of Earth Sciences at The
Ohio State University. The authors express gratitude
to those who offered their comments, suggestions and
support of this project: Kristen Gremillion, Debbie
Guatelli-Steinberg, Kathleen Keifer, Greg Pellam,
Gabriela Jakubowska, Ana M. Casado, Gary Blatt
and Larry T. Boston. Appreciation is also offered to
the editor and two anonymous reviewers whose sug-
gestions and criticisms have improved this work. The
authors also thank the numerous dedicated volunteers
at the Cleveland Museum of Natural History, both in
the field and in the lab, without whom this work would
not have been possible. For the remaining imperfec-
tions, the first author is alone responsible.
References
Arensburg B. 1996. Ancient dental calculus and diet. Human
Evolution 11(2): 139–145.
Armitage PL. 1975. The extraction and identification of opal
phytoliths from the teeth of ungulates. Journal of Arche-
aological Science 2(3): 187–197.
Auerbach BM, Ruff CB. 2004. Human body mass estimation:
a comparison ‘‘morphometric’’ and ‘‘mechanical’’ methods.
American Jornal of Physical Anthropology 125(4): 331–342.
Barker FC, Griffen JB, Morgan RG, Neumann GK, Taylor
LJB. 2002. Mesoamerican origin for an obsidian scraper
from the Pre-Columbian southeastern United States.
American Antiquity 67: 103–108.
Blondiaux J, Charlier P. 2008. Palaeocytology in skeletal
remains: microscopic examination of putrefication fluids
and dental calculus of skeletal remains from French
archaeological sites. International Journal of Osteoarcheology
18: 1–10.
Bohrer VL. 1962. Ethnobotanical materials from Tonto
National Monument. In Archeological Studies at Tonto
National Monument, Arizona, Caywood LR (ed.). Southwest
Monuments Association Technical Series 2. Southwest
Monuments Association: Globe.
Boyadjian HCC, Eggers S, Reinhard K. 2007. Dental wash: a
problematic method for extracting microfossils from
teeth. Journal of Archaeological Science 34: 1622–1628.
Brothwell DR. 1981. Digging Up Bones. Cornell University
Press: New York; 159–160.
Int. J. Osteoarchaeol. 21: 669–678 (2011)
Dirty Teeth and Ancient Trade
Buikstra JE, Ubelaker DH. 1994. Standards for the Data Collec-
tion from Human Skeletal Remains: Proceedings of a Seminar at the
Field Museum of Natural History, organized by Jonathan
Haas. Field Museum of Natural History, Fayetteville,
AR: Arkansas Archeological Survey.
Bullen RP. 1974. Were there Pre-Columbian cultural con-
tacts between Florida and the West Indies: the archae-
ological evidence. Florida Anthropology 27: 149–160.
Ciochon RL, Piperno DR, Thompson RG. 1990.
Opal phytoliths found on the teeth of the extinct ape
Gigantopithecus blacki: implications for paleodietary
studies. Proceedings of National Academy of Science 87:
8120–8124.
Cortello AR, Pochettino MI. 1994. Starch grain analysis as a
microscopic identification feature in the identification of
plant material. Economic Botany 48: 171–181.
David SK, Pailthrope MT. 1999. Classification of textile
fibers: production, structure, and properties. In Forensic
Examination of Fibres (2nd edn), Robertson J, Grieve M
(eds). Taylor & Francis, Ltd.: London; 1–31.
Dobney K, Brothwell D. 1986. Dental calculus: its relevance
to ancient diet and oral ecology. In Teeth and Anthropology,
Cruwys E, Foley RA (eds). BAR International Series 291:
Oxford; 55–81.
D’Orazio L, Martuscelli E, Oresello G, Riva F, Scala G,
Taglialetela A. 2000. Nature, origin, and technology of
natural fibers of textile artifacts recovered in the ancient
cities around Vesuvius. J Arch Sci 27: 745–754.
Evans ST. 2004. Ancient Mexico and Central America, Archaeology
and Culture History. Thames and Hudson Press: London.
Forde CD. 1963. Habitat, Economy, and Society. E.P. Dutton:
New York.
Fox CL, Juan J, Albert RM. 1996. Phytolith analysis on
dental calculus, enamel surface, and burial soil: infor-
mation about diet and paleoenvironment. American Journal
of Physical Anthropology 101: 101–113.
Fryxell PA. 1979. The Natural History of the Cotton Tribe:
Malvaceae, Tribe Gossypiae. University Press: College
Station: Texas A&M.
Gobetz KE, Bozarth SR. 2001. Implications for Late Pleis-
tocene mastodon diet from opal phytoliths in tooth
calculus. Quaternary Research 55: 115–122.
Goldstein J, Newbury D, Joy D, Lyman C, Echlin P, Lifshin
E, Sawyer L, Michael J. 2003. Scanning Electron Microscopy
and X-ray Microanalysis, (3rd edn). Springer: New York.
Grossman FE. 1873. The Pima Indians of Arizona. In Annual
Report of the Board of Regents of the Smithsonian Institution Showing
the Operations, Expenditures, and Condition of the Institution for the
Year 1871. The Smithsonian Institute: Washington, D.C.
Hershkovitz I, Kelly J, Latimer B, Rothschild BM, Simpson
S, Polak J, Rosenberg M. 1997. Oral bacteria in Miocene
Sivapithecus. Journal of Human Evolution 33: 507–512.
Hill JB. 1998. Ecological variability and agricultural special-
ization among the Protohistoric Pueblos of Central New
Mexico. Journal of Field Archaeology 25(3): 275–294.
Jakes KA, Sibley LR, Yerkes R. 1994. A comparative collec-
tion for the study of fibres used in prehistoric textiles from
Copyright # 2010 John Wiley & Sons, Ltd.
677
Eastern North America. Journal of Archaeological Science 21:
641–650.
Kent KP. 1983. Prehistoric Textiles of the Southwest. School of
American Research: Sante Fe.
King, M.E., Gardener, J.S., 1981. The analysis of textiles
from Spiro Mound, Oklahoma. Annals of the New York
Academy of Sciences 21: 641–650.
Kozuch L 2002 Olivella beads from Spiro and the Plains
American Antiquity 67(4): 697–709.
Larsen CS. 1985. Dental modifications and tool use in the
western Great Basin. American Journal of Physical Anthropol-
ogy 67(4): 393–402.
Lee, J.A., 1984. Cotton as a world crop. In Cotton, Kohl , RJ
Lewis , CF (eds). Agronomu 24. American Society of
Agronomy: Madison.
Lovejoy CO, Meindl RS, Pryzbeck TR, 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 Anthro-
pology 68(1): 15–28.
MacPhee T, Cowley G. 1975. Essentials of Periodontology and
Periodontics, (2nd edn). Blackwell Scientific Publications:
Oxford.
Mandel ID. 1990. Calculus formation and prevention: an
overview. Compendium of Continuing Education in Dentistry,
Supplement 8: 235–241.
Meindl RS, Lovejoy CO, Mensforth RP, Walker RA. 1985. A
revised method of age determination using the os pubis,
with a review and tests of accuracy of other current
methods of pubic symphyseal aging. American Journal of
Physical Anthropology 68: 29–45.
Middleton WD. 1990. An improved method for extraction
of opal phytoliths from tartar residues on herbivore teeth.
Phytolitherian Newsletter 6: 2–5.
Middleton WD. 1992a. Preliminary report on the extraction
of opal phytoliths from tartar residues on herbivore teeth.
Phytolitherian Newsletter 7: 5.
Middleton WD. 1992b. Preliminary report on the extraction
of phytoliths from cervid dental tartar. Phytolitherian News-
letter 7(1).
Middleton WD, Rovner I. 1994. Extraction of opal phyto-
liths from herbivore dental calculus. Journal of Archaeological
Science 21: 469–473.
Müller M, Papiz MZ, Clarke DT, Roberts MA, Murphy BM,
Burghammer M, Riekel C, Pantos E, Gunneweg J. 2004.
Identification of textiles using microscopy and synchro-
tron radiation X-ray fibre defraction. Spectrohima Acta 59:
175–186.
Phenice TW. 1969. A newly developed visual method of
sexing the os pubis. American Journal of Physical Anthropology
30(2): 297–301.
Phillips P, Brown JA. 1978. Pre-Columbian Shell Engravings from
the Craig Mound at Spiro, Oklahoma, Part 1. Peabody Museum
Press: Cambridge, Massachusetts.
Prufer OH, Shane OC. 1976. The Portage-Sandusky-
Vermilion River Region in Ohio. In The Late Prehistory of
the Lake Erie Drainage Basin, A 1972 Symposium Revised,
Int. J. Osteoarchaeol. 21: 669–678 (2011)
678
Brose DS (Ed.), Scientific Publications of the Cleveland
Museum of Natural History: Cleveland, Ohio; 283–304.
Redmond, B.G., 2005. A report of the archaeological inves-
tigations at the Danbury site (33OT16): 2004 season.
Cleveland Museum of Natural History, Archaeological Research
Reports, No. 146.
Redmond, B.G., 2006a A report of the archaeological inves-
tigations at the Danbury site (33OT16): 2005 season.
Cleveland Museum of Natural History, Archaeological Research
Reports, No. 147.
Redmond, B.G., 2006b Saving the Danbury site (33OT16):
investigation of Woodland to Late Prehistoric settlement
and mortuary behavior along the Lake Erie shore. Ohio
Archaeological Council. [HTLM title] Available online:
March 15, 2006. Ohioarchaeology.org.
Redmond, B.G., 2007. A report of the archaeological inves-
tigations at the Danbury site (33OT16): 2006 season.
Cleveland Museum of Natural History, Archaeological Research
Reports, No. 147.
Schurr MR, Redmond BG. 1991. Stable isotope analysis of
incipient maize horticulturalists from the Gard Island 2
Site. Midcontinent Journal of Archaeology 16(1): 69–84.
Sciulli PW. 2007. Relative dental maturity and associated
skeletal maturity in prehistoric Native Americans of the
Ohio valley area. American Journal of Physical Anthropology
132: 545–557.
Sciulli PW, Hetland BM. 2007. Stature estimation for pre-
historic Ohio Valley native American populations based
on revisions of the fully technique. Archaeology of Eastern
North America 35: 105–113.
Seeman MF, Dancey WS. 2000. The Late Woodland Period
in Southern Ohio: basic issues and prospects. In Late
Copyright # 2010 John Wiley & Sons, Ltd.
S. H. Blatt et al.
Woodland Societies, Tradition and Transformation Across the
Midcontinent, TE, Emerson DL, McElrath AC Fortier
(eds). University of Nebraska Press: Lincoln, Nebraska;
583–611.
Seidemann R. 2006. The Bahamian problem in Florida
archaeology: oceanographic perspectives on the issue
of Pre-Columbian contact. KACIKE. Journal of Caribbean
Amerindian History and Anthropology 7(1): 33–61.
Stothers DM, Abel TJ. 2002. The Early Late Woodland
in the Southwestern Lake Erie Littoral Region. In
Northeast Subsistence-Settlement Change A.D. 700–1300, Hart
JP, Rieth CB (eds). State Museum Bulletin 496: New York;
73–96.
Stark BL, Heller L, Ohnersorgen MA. 1998. People with
cloth: Mesoamerican economic change from the
perspective of cotton in South-Central Veracruz. Latin
American Antiquity 9(1): 7–36.
Teague LS. 1998. Textiles in Southwestern Prehistory. University
of New Mexico Press: Albuquerque.
Trubitt, MBD. 2003. The Production and Exchange of
Marine Shell Prestige Goods. Journal of Archaeological
Research 11(3): 243–277.
Vandermeersch B, Arensburg B, Tillier A, Rak Y, Weiner S,
Spiers M, Aspillaga E. 1994. Middle Paleolithic dental
bacteria from Kebara, Israel. C.R. Academy of Science Paris
319(2): 727–731.
Wendel JF, Brubaker CL, Percival AE. 1992. Genetic diver-
sity in Gossypium hirsutum and the origin of Upland cotton.
American Journal of Botany 79: 1291–1310.
Wendel JF. 1995. Cotton. In Evolution of Crop Plants,
Simmonds N, Smartt J (eds). Longman: London; 358–
366.
Int. J. Osteoarchaeol. 21: 669–678 (2011)