International Journal of Osteoarchaeology
Int. J. Osteoarchaeol. 18: 262 279 (2008)
Published online 31 August 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/oa.934
Dental Pathology and Diet at
Apollonia, a Greek Colony on
the Black Sea
*
A. KEENLEYSIDE
Department of Anthropology, Trent University, Peterborough, Ontario, K9J 7B8, Canada
ABSTRACT Dental pathology has the potential to provide insight into the composition of the diet and to
reveal dietary differences based on age, sex and social status. Human skeletal remains from
the Greek colonial site of Apollonia (5th to 2nd centuries BC) on the Black Sea coast of Bulgaria
were analysed for various forms of dental pathology in order to: assess the prevalence of
dental disease in the population; compare the dental pathology data from Apollonia with
dietary data derived from ancient literary texts and from previous stable isotopic analysis of the
colonists remains; explore variations in dental disease with respect to age and sex; and
compare the prevalence of dental pathology in the Apollonians with that of other Greek
populations. The composition of the diet, as indicated by the dental pathology data, is
consistent with the stable isotopic evidence from Apollonia and with the ancient literary texts,
both of which indicate the consumption of a relatively soft, high carbohydrate diet. The higher
frequency of dental caries, abscesses, calculus, and antemortem tooth loss in older adults
compared with younger ones reflects the age-progressive nature of these conditions. The lack
of significant sex differences in caries, abscesses, calculus and tooth loss corresponds with
the stable carbon and nitrogen isotopic data derived from bone collagen, which indicate no
significant sex differences in the consumption of dietary protein. In contrast, these findings
conflict with the ancient literary texts, which refer to distinct dietary differences between males
and females, and with the stable carbon isotopic values derived from bone carbonate, which
indicate sex differences with respect to the overall diet. Despite the lack of marked sex
differences in dental pathology, overall trends point to subtle dietary differences between
males and females. A greater degree of tooth wear in males also hints at possible sex
differences in the use of the teeth as tools. Copyright ß 2007 John Wiley & Sons, Ltd.
Key words: dental pathology; diet; Greek colonial populations; Black Sea
conditions and modes of subsistence (Littleton &
Introduction
Frohlich, 1993; Tayles et al., 2000), investigated
dietary changes over time (Walker & Erlandson,
Dental pathology is commonly recorded in archa-
1986; Lillie, 1996), and examined age-, sex-, and
eological samples, and its use in palaeodietary
status-based differences in diet (Frayer, 1984;
reconstructions is based on the recognised link
Lukacs, 1996; Cucina & Tiesler, 2003). As an
between dental pathology and diet (Nelson et al.,
indirect source of information on diet, dental
1999: 333). Studies of dental caries, antemortem
pathology has also been combined with direct
tooth loss, and other forms of dental disease
evidence of diet derived from stable isotopic
have explored the relationship between these
analysis of bone collagen and bone carbonate
(Sealy & van der Merwe, 1988; Lubell et al., 1994;
* Correspondence to: Department of Anthropology, Trent Univer-
White, 1994; Lillie & Richards, 2000; Prowse,
sity, Peterborough, Ontario, K9J 7B8, Canada.
e-mail: [e-mail] 2001).
Copyright # 2007 John Wiley & Sons, Ltd. Received 10 July 2006
Revised 30 April 2007
Accepted 11 May 2007
Dental Pathology in a Greek Colony 263
The earliest examination of dental pathology dental pathology in the Apollonians with that of
in Greek skeletal samples was conducted by other Greek populations.
Angel (1944), who compared dental pathology in
samples of early and modern Greek teeth with the
Archaeological and biocultural context of the
goal of assessing temporal changes in dental
sample
health. Subsequent studies have examined dental
morphology and disease in approximately 1500
According to literary sources, the Greek colony
Middle Minoan (1750 1550 BC) teeth from
of Apollonia was founded in 610 BC by the city of
Knossos (Carr, 1960), explored the relationship
Miletus in Asia Minor, on the site of the
between dental caries and selenium and fluoride
present-day town of Sozopol on the Black Sea
levels in archaeological and modern Greek teeth
coast of Bulgaria (Figure 1). Thought to have been
(Hadjimarkos & Bonhorst, 1962), examined the
occupied by Thracians at the time of the colony s
impact of changing subsistence patterns on dental
foundation, the site had a number of desirable
wear in Mesolithic and Neolithic remains from
characteristics, including an excellent harbour,
Franchthi Cave (Smith & Cook, 1991), and
good fishing, and an easily defensible location
compared the skeletal and dental evidence for
(Nedev & Panayotova, 2003). Its strategic
social status in a Late Bronze Age skeletal sample
location contributed to its establishment as a
from the Athenian Agora (Smith, 2000).
major trading centre, and Apollonia became a
Dental pathology data have been recorded in a
prosperous urban centre during the 5th and 4th
small number of Greek colonial samples, most of
centuries BC, minting its own coins, producing its
them from southern Italy (Becker & Donadio,
own ceramics, erecting monumental public
1992; Becker, 1995, 1996; Henneberg, 1998;
buildings, and establishing secondary settlements
Henneberg & Henneberg, 1998, 2003). In
in the surrounding territory (Nedev & Panayo-
contrast, little is known about the dental health
tova, 2003). Archaeological evidence indicates a
of Greek colonists on the Black Sea. Few studies
trading relationship between the colonists and
of ancient Greek populations have explored sex
the local Thracian population, and close con-
differences in dental health, and none have
nections between the two groups are further
compared data on dental pathology with that
indicated by the presence of Thracian names on
derived from literary sources and stable isotopic
analyses to test the utility of each as sources
of information on diet (for an excellent example
of this approach applied to Roman remains see
Prowse, 2001). The ancient literary texts refer to
distinct sex differences in diet among the ancient
Greeks. At Apollonia, stable carbon isotope
values derived from bone carbonate indicate
sex differences with respect to the overall diet,
while stable carbon and nitrogen isotope values
measured in bone collagen revealed no sex
differences in the consumption of dietary protein
(Keenleyside et al., 2006).
The aim of this study is to assess the prevalence
of dental pathology in a skeletal sample from the
Greek colonial site of Apollonia on the Black Sea
coast of Bulgaria, explore variations in dental
disease with respect to age and sex, compare the
dental pathology data from Apollonia with
dietary data derived from ancient literary texts
and from previous stable isotopic analysis of the
Figure 1. Map of Apollonia (Sozopol) showing the
colonists remains, and compare the prevalence of location of the necropolis.
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
264 A. Keenleyside
several Greek tombstones excavated from the the bacterial fermentation of dietary carbo-
necropolis (Nedev & Panayotova, 2003). Like hydrates, especially sugars (Larsen, 1997: 65).
other Greek colonies on the Black Sea coast, the Of the factors underlying the development of
inhabitants of Apollonia attempted to recreate caries, diet has been identified as the most
the same way of life they had enjoyed in their important (Powell, 1988). Diets containing
homeland (Tsetskhladze, 1997). While they refined sugars are more cariogenic than those
brought the religious and political institutions consisting of unrefined carbohydrates, while diets
of their mother city with them, and maintained high in protein and fat and low in carbohydrates
commercial links with Miletus and other cities in are known to impede caries formation (Powell,
Asia Minor, they remained independent and 1988). The texture of food also plays a role in the
self-sufficient (Nedev & Panayotova, 2003). development of caries, with soft, sticky foods
In 1938, the chance discovery of a grave in an being more cariogenic than hard, abrasive foods
area known as Kalfata, located on the shore of the (Larsen, 1997). Other factors influencing tooth
Black Sea approximately 2.5 km south of Sozopol decay include tooth crown morphology, attrition,
(Figure 1), prompted scholars to direct their enamel hypoplasia, oral hygiene, fluoride con-
attention to this area, and from 1946 to 1949, sumption, oral bacteria and the composition of
large-scale excavations uncovered 801 graves saliva (Dawes, 1970; Lillie, 1996; Hillson, 2000).
consisting of simple unlined pits, stone cists, tile Numerous studies of skeletal samples world-
graves, wooden coffins and urn burials (Venedi- wide have revealed higher rates of caries in
kov, 1963). Extending to a depth of five metres in agricultural populations compared with hunter-
some places, most of the burials were typical gatherers (Turner, 1979; Larsen, 1995; Hillson,
Greek inhumation burials containing single 1996), reflecting their greater consumption of
individuals laid out on their backs in an extended carbohydrates. Agricultural diets have, in fact,
position, with their heads oriented primarily to been inferred from high caries rates in the
the east. The remains dated from the second half absence of other types of dietary evidence
of the 5th century to the beginning of the 2nd (Cohen, 1989: 107). In the western Mediterra-
century BC, based on associated grave goods nean region, for example, the transition to
(Venedikov, 1963). A small number of cremation agriculture was accompanied by an increase in
and flexed burials, dating from the mid to late 4th the rate of caries (Meiklejohn & Zvelebil, 1991).
century BC, have been tentatively identified as Hunting and gathering populations, in contrast,
Thracian burials (Venedikov, 1963). Since 1992, typically show an absence of caries, reflecting
excavations led by Dr Kristina Panayotova of the their non-cariogenic, high protein diet; also,
Institute of Archaeology in Sofia, of the Kalfata populations heavily reliant on marine resources
necropolis and several adjacent sites have typically show low rates of caries (Macchiarelli,
uncovered an additional 400 burials (Panayotova, 1989; Nelson et al., 1999).
1998; Nedev & Panayotova, 2003). Palaeopatho- One of the possible consequences of dental
logical analysis of some of these remains has caries is antemortem tooth loss, indicated by the
revealed evidence of iron deficiency anaemia, resorption of alveolar bone in the tooth socket.
trauma and non-specific infections among the Other causes of tooth loss include attrition,
colonists (Keenleyside & Panayotova, 2005, trauma and periodontal disease, the latter being
2006). The sample utilised in this study is derived one of the main causes of antemortem tooth loss
from this most recent excavation. in modern populations (Hillson, 2005: 306).
Dental caries can also lead to periapical
abscessing, which occurs when bacteria invade
Dental pathology and diet the exposed pulp cavity and proceed down the
root canal into the alveolar bone, causing an
Dental caries, the most common form of dental infection of the surrounding tissue and the
pathology, is an age-progressive disease process formation of an abscess (Hillson, 2005). Pulp
 characterized by the focal demineralization of exposure resulting from trauma or attrition can
dental hard tissues by organic acids produced by also lead to abscesses.
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
Dental Pathology in a Greek Colony 265
Dental calculus, or mineralised plaque, is More than 100 different factors have been linked
common in archaeological remains. Occurring to the development of enamel hypoplasia, with
both above (i.e. supragingival) and below (i.e. particular attention being paid in the literature to
subgingival) the gums, it is most frequently found nutritional stress and childhood infections. Since
on teeth closest to the salivary glands (Roberts & it is rarely possible to link enamel defects with
Manchester, 1995). The relationship between their underlying aetiology, they are considered to
diet and the formation of dental calculus is not be non-specific indicators of physiological stress
straightforward (Hillson, 2001: 265), and high (Goodman & Rose, 1990). In some archaeological
calculus rates have been associated with both samples, a higher prevalence of hypoplastic
high protein and high carbohydrate diets defects in younger individuals compared with
(Meiklejohn & Zvelebil, 1991; Lieverse, 1999; older adults has been interpreted as reflecting an
Lillie & Richards, 2000). When combined with increased likelihood of individuals who experi-
dental caries, however, it can be used to enced stress during childhood dying at a younger
determine the relative contribution of proteins age (Duray, 1996).
versus carbohydrates to the diet. High calculus
rates and low caries rates have been found to
characterise populations subsisting on a high The Ancient Greek diet
protein, low carbohydrate diet (Lillie, 1996: 140),
while heavy calculus deposits and high caries Information on the ancient Greek diet derives
rates have been observed in populations that from a variety of sources, including literary texts,
consumed a high carbohydrate diet (White, most notably Athenaeus Deipnosophistae (Professors
1994: 283). The formation of calculus is also of the Dinner-Table) and Galen s On the Properties of
influenced by a number of non-dietary factors, Foodstuffs, archaeological evidence in the form of
including oral hygiene, the mineral content of plant remains, animal bones, food preparation
drinking water, the rate of salivary flow, cultural utensils, storage vessels, ritual food offerings
practices such as chewing betel nuts and coca associated with graves, and stable isotopic
leaves, and the use of the teeth as tools (Lieverse, analysis of human skeletal remains. These sources
1999). indicate that the ancient Greeks relied on a staple
Dental wear can provide valuable information diet of cereals, legumes, olive oil and wine for
on the types of foods consumed, their consist- their subsistence. Olive oil was the main source of
ency and texture, and their methods of prep- dietary fat (Martin, 1996), and wine mixed with
aration. Known to vary widely among popu- water was the preferred beverage. Cereal crops,
lations, high levels of tooth wear recorded in most notably wheat and barley, comprised the
agricultural populations have been attributed to primary source of protein and carbohydrates
the consumption of coarse cereal grains and the (Garnsey & Whittaker, 1983: 7; Rathbone, 1983:
use of grinding stones which can add a significant 46; Garnsey, 1999: 15). Dry legumes such as
amount of grit to the diet (Larsen, 1997). lentils, broad beans, peas and chickpeas were also
Differential wear between the anterior and an important source of protein. Several varieties
posterior dentition, seen in some archaeological of millet were cultivated but were considered  an
populations, has been interpreted as reflecting emergency resource for small farmers (Garnsey,
different causes of tooth wear (e.g. diet vs. tool 1988: 52). Similarly, oats were grown mainly as a
use) (Littleton & Frohlich, 1993). fodder crop (Brothwell & Brothwell, 1969: 101). A
Enamel hypoplasia refers to a deficiency in variety of fresh fruits, vegetables and nuts were
enamel thickness caused by a disruption in the also consumed, and honey, dates and figs were
secretory phase of enamel matrix formation the main sweeteners (Brothwell & Brothwell,
(Goodman et al., 1984). Formed during tooth 1969; Grmek, 1989; Dalby, 1996).
crown development, hypoplastic defects are not Of the domesticated animals, sheep, pigs and
remodelled in later life, and therefore provide goats were the primary sources of meat (Dalby,
permanent markers of stress episodes during 1996), although meat was relatively scarce and
infancy and childhood (Goodman et al., 1984). generally made up only a small component of the
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
266 A. Keenleyside
diet (Jameson, 1983: 10; Garnsey, 1999: 16, 123). with some of the graves contained food remains
Goat and sheep milk was consumed in diluted interpreted as evidence of funerary feasts
form (Brothwell & Brothwell, 1969: 50), and (Panayotova, 1998). These included animal
cheese made from the milk of both animals was bones, grape seeds, acorns, almonds and hazel-
also eaten (Brothwell & Brothwell, 1969: 52). Fish, nuts, and ceramic fish plates and grills (Panayo-
including shellfish, were a highly valued part of tova, 1998: 105; Nedev & Panayotova, 2003:
the diet of many Greeks (Dalby, 1996) and 138 9). Astragali from sheep/goats, found in the
garum, or fish sauce, was a popular ingredient in graves of both children and adults (Panayotova,
many dishes. 1998), suggest the consumption of one or both
According to literary sources, sex differences in species.
diet existed among the ancient Greeks. Males Stable isotopic analysis of bone collagen and
were given preferential access to foods, while carbonate samples taken from the remains of 54
women were denied meat and other nourishing of the adults examined for dental pathology in
food items (Garnsey, 1999). Consequently, this study revealed that the colonists relied on a
women were at higher risk of nutritional mixed diet of terrestrial and marine resources
deficiency and disease, and may have experienced (Keenleyside et al., 2006). Statistical comparisons
a higher incidence of malnutrition than men by age and sex revealed no significant differences
(Garnsey, 1999). Given this evidence for sex- in d13C or d15N values for bone collagen,
based variation in diet, one should therefore suggesting similarities in the consumption of
expect to find sex differences in dental pathology dietary protein. In contrast, males had signifi-
in ancient Greek skeletal samples. Such differ- cantly higher d13C values for bone carbonate,
ences might, for example, take the form of lower suggesting sex differences with respect to the
caries rates in males resulting from a greater overall diet (Keenleyside et al., 2006). As such, it
consumption of meat and fish. Fluoride and is hypothesised that sex differences should be
strontium, elements present in high quantities in seen in the prevalence of certain forms of dental
marine foods (Siebert & Trautner, 1985; Malde pathology at Apollonia.
et al., 1997), are known to impede the formation
of carious lesions.
Materials and methods
The dentitions of 162 adult skeletons were
Archaeological and stable isotopic evidence of examined in this study (Table 1). Sex determi-
diet at Apollonia nation of the skeletons was done using standard
morphological and metric criteria of the crania
The archaeological evidence recovered from and pelvic bones (Bass, 1987; Ubelaker, 1989).
Apollonia suggests that during the 5th to 2nd Age estimation was based on pubic symphysis
centuries BC, the population subsisted on a varied morphology (Brooks & Suchey, 1990), auricular
diet of grains, nuts, fish, meat and shellfish surface morphology (Lovejoy et al., 1985) and
(Panayotova, 1998). Ritual fireplaces associated ectocranial suture closure (Meindl & Lovejoy,
Table 1. Age and sex distribution of the sample
Age (years) Number of individuals Number of teeth
Males Females Indeterminate Males Females Indeterminate
18 35 22 40 3 418/511 851/1031 65/72
36 50 37 23 3 762/908 424/594 73/73
50þ 14 20 0 171/388 175/436 0/0
Total 73 83 6 1351/1807 1450/2061 138/145
Number of observable teeth/number of observable teeth plus teeth lost antemortem plus teeth lost post-mortem.
Statistically significant differences between males and females, P < 0.05.
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
Dental Pathology in a Greek Colony 267
1985). Individuals were assigned to one of three respect to the number of individuals (x2 ź 5.29,
age categories: 18 to 35 years, 36 to 50 years, and P ź 0.025) and the number of observable sockets
50þ years. (x2 ź 189.92, P ź 0.001) and preserved teeth
All permanent dentitions were examined macro- (x2 ź 217.33, P ź 0.001). In contrast, there were
scopically for evidence of pathology. Dental significantly more males than females in the
caries, antemortem tooth loss, abscesses and middle-aged category (36 50 years) when com-
calculus were recorded by individual and by parisons were made by individual (x2 ź 8.66,
tooth/tooth socket following the methodology P ź 0.01), by socket (x2 ź 186.12, P ź 0.001),
outlined in Buikstra & Ubelaker (1994). Carious and by tooth (x2 ź 211.34, P ź 0.001). Frequen-
lesions were recorded as present only if they had cies were calculated for each type of dental
penetrated the tooth enamel. The number and pathology recorded, and statistical comparisons
location of lesions per tooth were also noted. by age and sex were conducted using the
Teeth were scored as being lost antemortem if chi-square test and the Mann-Whitney U test.
their sockets exhibited evidence of resorption. All statistical analyses were performed using
Dental abscesses were identified by the presence SPSS, Version 14.
of a drainage channel through the alveolar bone
at the apex of the tooth root, and their location
(buccal vs. lingual) was also recorded. The Results
location and degree of dental calculus were
recorded for each affected tooth, following the When recorded by individual, dental caries was
criteria outlined in Buikstra & Ubelaker (1994). the second most common dental disease observed
The degree of dental wear was recorded for each in the sample, affecting 53.8% (85/158) of
tooth using Smith s (1984) 8-stage system for the individuals (Table 2). In contrast, only 7.7%
incisors, canines and premolars, and Scott s (227/2939) of teeth had carious lesions (Table 3).
(1979) 10-stage system for the molars; average Of the four tooth types, the molars were the most
scores were calculated for each individual and frequently affected (13.9%) (Table 4), reflecting
each tooth type (after Prowse, 2001). Finally, their more complex morphology. The majority of
enamel hypoplasia was classified according to the lesions (54.2%) were located on the interprox-
Developmental Defects of Enamel (DDE) Index imal surfaces (Table 5), while only 2.9% were
(Commission on Oral Health, 1982). Hypoplas- located on the buccal or lingual surfaces. While
tic defects were identified by visual examination both males and females had the greatest number
and were scored as present or absent for each of of carious lesions on the interproximal surfaces of
the six anterior maxillary and mandibular teeth. their teeth, males had considerably more inter-
All observed defects were classified as either proximal lesions than females (64.5% compared
linear arrangements of pits, horizontal grooves, or with 44.5%), and females, in contrast, had more
both, and the total number of defects was cervical lesions than males (19.5% compared with
recorded for each affected tooth. Those teeth 6.4%) (Table 5). When statistical comparisons
that exhibited severe attrition resulting in the loss were made by the number of individuals affected,
of more than a third of the tooth crown, heavy there was no significant difference in the pre-
calculus deposits, severe caries or trauma were valence of caries between males and females for
excluded from the analysis. any of the three age groups compared (Table 2).
Since the age and sex distribution of a sample When comparisons were made by the number of
can affect interpretations of dental pathology affected teeth, young adult and middle-aged
data, statistical comparisons were made by age females had a slightly higher frequency of carious
and sex to determine if there were any significant teeth than males (Table 3), while males over the
differences in the number of individuals and the age of 50 had a slightly higher frequency than
number of observable teeth/tooth sockets. These females. There were no statistically significant sex
comparisons revealed that the sample contained differences in caries rates for any of the age
significantly more females than males in the groups examined (Mann-Whitney U test), and
young adult age category (18 35 years) with for both sexes, the rates increased with age.
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
268 A. Keenleyside
Table 2. Frequency of affected individuals by age and sex
Age (yrs) Males % Females % Indeterminate % Total %
Dental caries
18 35 11/22 50.0 16/39 41.0 0/3 0 27/64 42.2
36 50 25/37 67.6 14/23 60.9 0/2 0 39/62 62.9
50þ 9/14 64.3 10/18 55.5 0/0 0 19/32 59.4
Total 45/73 61.6 40/80 50.0 0/5 0 85/158 53.8
AMTL
18 35 6/22 27.3 6/40 15.0 0/3 0 12/65 18.5
36 50 16/37 43.2 14/23 60.9 0/3 0 30/63 47.6
50þ 13/14 92.9 19/20 95.0 0/0 0 32/34 94.1
Total 35/73 47.9 39/83 47.0 0/6 0 74/162 45.7
Abscesses
18 35 2/18 11.1 4/40 10.0 0/2 0 6/60 10.0
36 50 11/36 30.5 4/22 18.2 1/2 50.0 16/60 26.7
50þ 6/14 42.9 4/19 21.0 0/0 0 10/33 30.0
Total 19/68 27.9 12/81 14.8 1/4 25.0 32/153 20.9
Calculus
18 35 19/22 86.4 28/39 71.8 2/3 66.7 49/64 76.6
36 50 31/37 83.8 18/23 78.3 3/3 100.0 52/63 82.5
50þ 11/13 84.6 13/18 72.2 0/0 0 24/31 77.4
Total 61/72 84.7 59/80 73.7 5/6 83.3 125/158 79.1
Enamel hypoplasia
18 35 6/18 33.3 9/35 25.7 1/3 33.3 16/56 28.6
36 50 9/32 28.1 4/16 25.0 0/3 0 13/51 25.5
50þ 1/7 14.3 1/8 12.5 0/0 0 2/15 13.3
Total 16/57 28.1 14/59 23.7 1/6 16.7 31/122 25.4
Statistically significant, P < 0.05.
Table 3. Frequency of affected teeth by age and sex
Age (yrs) Males % Females % Indeterminate % Total %
Dental caries
18 35 19/418 4.5 43/851 5.0 0/65 0 62/1334 4.6
36 50 59/762 7.7 50/424 11.8 2/73 2.7 111/1259 8.8
50þ 28/171 16.4 26/175 14.9 0/0 0 54/346 15.6
Total 106/1351 7.8 119/1450 8.2 2/138 1.4 227/2939 7.7
AMTL
18 35 24/511 4.7 7/1031 0.7 0/72 0 31/1614 1.9
36 50 40/908 4.4 44/594 7.4 0/73 0 84/1575 5.3
50þ 142/388 36.6 158/436 36.2 0/0 0 300/824 36.4
Total 206/1807 11.4 209/2061 10.1 0/145 0 415/4013 10.3
Abscesses
18 35 2/358 0.6 6/954 0.6 0/42 0 8/1354 0.6
36 50 15/721 2.1 7/521 1.3 2/49 4.1 24/1291 1.9
50þ 13/309 4.2 8/393 2.0 0/0 0 21/702 3.0
Total 30/1388 2.2 21/1868 1.1 2/91 2.2 53/3347 1.6
Calculus
18 35 176/414 42.5 285/840 33.9 13/65 20.0 474/1319 35.9
36 50 405/753 53.8 188/418 45.0 26/72 36.1 619/1243 49.8
50þ 87/161 54.0 104/163 63.8 0/0 0 191/324 58.9
Total 668/1328 50.3 577/1421 40.6 39/137 28.5 1284/2886 44.5
Enamel hypoplasia
18 35 25/140 17.9 43/307 14.0 2/22 9.1 70/469 14.9
36 50 33/254 13.0 3/124 2.4 0/26 0 36/404 8.9
50þ 2/43 4.6 3/67 4.5 0/0 0 5/110 4.5
Total 60/437 13.7 49/498 9.8 2/48 4.2 111/983 11.3
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
Dental Pathology in a Greek Colony 269
It is well-known that antemortem tooth loss
(AMTL) can have an effect on the calculation of
caries rates. To compensate for the antemortem
loss of teeth, some of which may have been
carious, Lukacs (1995) developed a  caries correc-
tion factor which involves estimating the number
of teeth lost antemortem due to caries by examin-
ing the proportion of teeth with pulp exposure
due to caries. The caries correction factor was not
employed in this study, however, as data on pulp
exposure due to caries was not recorded.
Another calculation that relates AMTL to
caries is the Diseased Missing Index (DMI: Kelley
et al., 1991: 205), which involves adding together
the total number of carious teeth and teeth lost
antemortem and expressing this number as a
percentage of the total number of teeth and soc-
kets observed. Application of this method to the
caries data from Apollonia yields a frequency of
17.3% (312/1807) for males and 15.9% (328/
2061) for females, values that are close to double
the standard caries rates (Table 6). However,
statistical comparisons again revealed no signifi-
cant sex differences in the DMI (Mann-Whitney
U test).
Concern has also been raised about the
possible effect of post-mortem tooth loss on
caries rates. To compensate for the post-mortem
loss of teeth that may have been carious, Erdal &
Duyar (1999) developed a  proportional correc-
tion factor designed for use on samples with a
high rate of post-mortem loss of the anterior
teeth. This method involves calculating the caries
rate of the anterior and posterior teeth and
multiplying it by a ratio representing the propor-
tion of these teeth in one quadrant of the mouth
(e.g. 3 anterior teeth to 5 posterior teeth, or 0.6).
This correction factor was not applied to the
Apollonia data, however, as the ratio of anterior
to posterior teeth in the sample (0.6) matches the
expected ratio.
AMTL was also common in the sample, affect-
ing 45.7% (74/162) of individuals (Table 2) and
10.3% (415/4013) of teeth (Table 3). Among
tooth types, the molars were most frequently
involved (16.1%). There were no significant sex
differences by individual for any of the age groups
compared (Table 2). Similarly, when compari-
sons were made by the number of affected teeth
(Table 3), males and females had a similar overall
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
Total
Mandible
Jaw
Maxilla
Incisors
Canines
65/368 (17.7)
46/615 (7.5)
44/456 (9.6)
67/527 (12.7)
111/983 (11.3)
Tooth type

Premolars

Molars
29/1146 (2.5)
13/873 (1.5)
4/450 (0.9)
7/878 (0.8)
33/1489 (2.2)
20/1858 (1.1)
53/3347 (1.6)
144/1038 (13.9)
55/821 (6.7)
13/404 (3.2)
15/676 (2.2)
116/1338 (8.7)
111/1601 (6.9)
227/2939 (7.7)
228/1419 (16.1)
90/1040 (8.6)
25/531 (4.7)
72/1023 (7.0)
189/1863 (10.1)
226/2150 (10.5)
415/4013 (10.3)
433/1025 (42.2)
311/795 (39.1)
193/399 (48.4)
347/667 (52.0)
485/1308 (37.1)
799/1578 (50.6)
1284/2886 (44.5)
Statistically significant, P < 0.05.
Table 4. Dental pathology by tooth type and dental arcade
Dental pathology
Caries
Abscesses
AMTL
Calculus
Enamel hypoplasia
270 A. Keenleyside
Table 5. Frequency of carious lesions by location
Location Males Females Indeterminate Total %
n % n % n %
Occlusal 8/110 7.3 9/128 7.0 0/2 0 17/240 7.1
Interproximal 71/110 64.5 57/128 44.5 2/2 100.0 130/240 54.2
Smooth 2/110 1.8 5/128 3.9 0/2 0 7/240 2.9
Cervical 7/110 6.4 25/128 19.5 0/2 0 32/240 13.3
Root 11/110 10.0 11/128 8.6 0/2 0 22/240 9.2
Large 11/110 10.0 21/128 16.4 0/2 0 32/240 13.3
Surface of origin cannot be determined.
prevalence of AMTL. The prevalence of tooth (125/158) of individuals (Table 2) and 44.5%
loss in both sexes increased with age. (1284/2886) of teeth (Table 3). The incisors were
Abscesses were the least common type of the most commonly affected teeth, followed by
dental pathology observed in the sample, affect- the canines, molars and premolars (Table 4). Also,
ing 20.9% (32/153) of individuals (Table 2) and significantly more mandibular teeth were affected
only 1.6% (53/3347) of teeth (Table 3). They than maxillary teeth (x2 ź 53.20, P 0.001)
were found more frequently on the buccal side of (Table 4). The majority of affected teeth in both
the alveolus than the lingual side, and affected the males and females had only slight calculus
molars more frequently than the other tooth deposits. There were no significant sex differ-
types (Table 4). On an individual basis, signifi- ences when comparisons were made by the
cantly more males had abscesses than females number of affected individuals (Table 2). When
overall (x2 ź 3.87, P 0.05) (Table 2). When comparisons were made by the number of
comparisons were made by the number of affected teeth (Table 3), males had a higher
affected teeth (Table 3), however, males had prevalence than females overall (50.3% compared
only a slightly higher prevalence of abscesses with 40.6%), and young adult and middle-aged
than females (2.2% compared with 1.1%). For males had a higher rate than females of the same
both sexes, the prevalence of abscesses increased age groups. None of these differences, however,
with age. was statistically significant (Mann-Whitney U
Dental calculus was the most common con- test). For both sexes, the frequency of calculus
dition observed in the sample, affecting 79.1% increased with age.
The majority of teeth in the sample showed
slight to moderate occlusal surface wear, as
Table 6. Comparison of caries index and DMI by age and
indicated by the average wear scores calculated
sex
for each tooth type (Table 7). The incisors,
canines and premolars had average scores ranging
Males Females Indeterminate Total
from 3.3 to 4.1 out of 8 (using the Smith, 1984
Caries index
system), and the molars had average scores
18 35 4.5 5.0 0.0 4.6
ranging from 12.7 to 20 out of 40 (using the Scott,
36 50 7.7 11.8 2.7 8.8
50þ 16.4 14.9 0.0 15.6
1979 system). While the use of two different
Total 7.8 8.2 1.4 7.7
scoring systems makes it difficult to draw a direct
DMI
comparison of the degree of wear between the
18 35 8.4 4.8 0.0 5.8
36 50 10.9 15.8 2.7 12.4
anterior vs. the posterior teeth, both sets of teeth
50þ 43.8 42.2 0.0 43.0
show low to moderate levels of wear. Males in all
Total 17.3 15.9 1.4 16.0
three age categories had higher average tooth
Total number of carious teeth 100/total number of wear scores than females for each tooth type
observed teeth.
examined (Table 8), but the differences were
Total number of carious teeth and teeth lost ante-
not statistically significant for any of the age
mortem 100/total number of teeth and sockets
observed. groups (Mann-Whitney U test). When statistical
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
Dental Pathology in a Greek Colony 271
Table 7. Average tooth wear scores for each tooth type individuals (Table 2) and 11.3% (111/983) of
teeth (Table 3). Statistical comparisons by the
Tooth type No. of teeth Average tooth
number of affected individuals revealed no signi-
observed wear score
ficant sex differences (Table 2). In contrast,
Maxilla
comparisons of the number of affected teeth
I1 147 4.1
revealed that males had a slightly higher
I2 165 3.3
frequency of defects than females (13.7% com-
C 187 4.0
P1 190 3.6
pared with 9.8%) (Table 3). For both sexes,
P2 189 3.6
young adults had a higher prevalence of defects
M1 178 19.7
than older adults.
M2 164 15.2
M3 105 12.7
Mandible
I1 169 3.9
Discussion
I2 198 3.7
C 215 3.7
P1 228 3.2
Dental caries
P2 197 3.3
M1 202 20.0
M2 213 16.0
In contrast to other studies of caries which have
M3 147 14.1
found significant differences in caries rates bet-
ween the upper and lower dentition (e.g.
Thylstrup & Fejerskov, 1994; Lingström &
Borrman, 1999), there was no significant
comparisons of the posterior teeth (molars) vs.
difference in caries rates between the maxillary
anterior teeth (premolars, canines and incisors)
and mandibular teeth in the Apollonia sample.
were conducted, however, males had a signifi-
The significantly higher frequency of carious
cantly greater degree of wear in the anterior teeth
lesions, by number of affected teeth, in older
than females overall (Mann-Whitney U test,
individuals compared with younger ones is,
P ź 0.004), and young adult males had a signi-
however, consistent with previous studies of
ficantly greater level of wear on the anterior teeth
caries in archaeological populations, and reflects
than females of the same age group (Man-
the age-progressive nature of this disease
n-Whitney U test, P ź 0.041). For both sexes, the
(Hillson, 2000, 2001).
degree of tooth wear increased with age for all
Sex differences in caries rates have been ob-
tooth types.
served in both archaeological and modern popu-
Enamel hypoplasia involving the canines and/
lations, with females typically showing higher
or incisors was recorded in 25.4% (31/122) of
rates than males (Walker & Hewlett, 1990;
Hillson, 2001: 253). This has been attributed
to a higher carbohydrate intake by females, earlier
Table 8. Average tooth wear scores by age and sex
eruption of teeth, and behavioural differences in
Category Molars Premolars Canines Incisors subsistence pursuits (Larsen, 1997; Walker &
Hewlett, 1990). Changes in salivary composition
Sexes combined
during pregnancy and lactation have also been
18 35 years 13.1 2.3 2.7 2.7
36 50 years 18.7 3.9 4.4 4.4 implicated in the development of caries in women
50þ years 26.4 5.6 5.5 5.5
(Laine, 2002). In the Greek colonial sample from
Males
rural Metaponto, a significantly higher number of
18 35 13.5 2.5 3.2 3.1
36 50 19.4 4.0 4.5 4.5 females (64%) were found to have carious lesions
50þ 28.7 5.8 5.9 6.3
than males (46%) (Henneberg & Henneberg,
Total 18.2 3.9 4.4 4.2
2003: 34). Sex differences in caries rates in the
Females
18 35 12.8 2.2 2.5 2.5 Apollonia sample are much less pronounced.
36 50 17.7 3.7 4.1 4.4
While young adult and middle-aged females had
50þ 24.3 5.4 5.1 4.9
a slightly higher frequency of carious teeth
Total 15.3 3.0 3.3 3.3
than males, statistical comparisons revealed no
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
272 A. Keenleyside
significant sex differences with respect to either It is interesting to compare the dental caries
the caries rate or the DMI, suggesting that males data from Apollonia with those derived from
and females were consuming foods with similar other coastal sites, particularly those for which
cariogenic properties. stable isotopic data have also been obtained. The
Comparison of the prevalence of caries in the Neolithic site of Alepotrypa Cave and the Greek
Apollonia sample with that of other archaeolo- colonial site of Metaponto in southern Italy are
gical samples is hindered by the fact that different particularly noteworthy in this regard. As indica-
methodologies may have been used to diagnose ted in Table 9, the rate of dental caries (10.5%)
and record carious lesions (Hillson, 2001: 249). for the rural cemetery from Metaponto is slightly
Further limiting such comparisons is the varying higher than that recorded at Apollonia (7.7%);
age and sex composition of archaeological however, the average d13C and d15N values
samples, the failure to distinguish between tooth measured in the Metaponto sample ( 19.6% and
types affected and the different forms of carious 10.5% respectively: Henneberg & Henneberg,
lesions observed, and differing patterns of ante- 2003) approximate those measured in the
mortem and post-mortem tooth loss. Despite Apollonians ( 18.5% and 10.1% respectively)
these limitations, it is useful to compare the (Keenleyside et al., 2006), indicating that both
prevalence of caries (uncorrected) at Apollonia populations relied heavily on terrestrial resources.
with that of other archaeological samples, includ- For Alepotrypa the caries rate was lower (3.2%),
ing those with varying subsistence strategies. As but this calculation included both permanent and
expected, there is considerable variability in the deciduous teeth. The average d13C and
prevalence of caries between populations with d15N values measured in the remains from
similar modes of subsistence (Turner, 1979; Alepotrypa ( 19.95% and 7.17% respectively)
Larsen, 1995: 188). This is not surprising given also point to a primary reliance on a terrestrial C3
methodological differences in the recording and diet (Papathanasiou, 2005: 385).
reporting of caries rates. At 7.7% of teeth, the
prevalence of caries at Apollonia falls at the lower
end of the range reported by Turner (1979) for Antemortem tooth loss
agricultural populations (2.1 26.9%, average ź
10.43%), and within the range reported for The rate of AMTL in the Apollonia sample is
populations with a mixed economy (1.0 10.3%, similar to that seen in other agricultural popu-
average ź 4.84%). It also falls within the range lations and probably reflects the higher rate of
reported for other Greek skeletal samples caries typically seen in such populations (Larsen,
(3.2 36.9%) (Table 9). 1995). A significantly higher number of older
Table 9. Prevalence of dental caries in Greek skeletal samples
Sample/time period Frequency by tooth Reference
Neolithic (3rd millennium BC) 32/267 (12.0%) Angel (1944)
Neolithic (Alepotrypa Cave) 14/436 (3.2%) Papathanasiou (2005)
Middle Bronze Age (2000 1400 BC) 34/601 (5.7%) Angel (1944)
Mycenean (1400 1150 BC) 22/356 (6.2%) Angel (1944)
Lefkandi (Iron Age) 11/113 (9.7%) Musgrave & Popham (1991: 289)
Early Iron Age (1150 650 BC) 40/724 (5.5%) Angel (1944)
Knossos (7th or 8th c. AD) 14/90 (15.5%) Musgrave (1976: 46)
Ailias (Middle Minoan) 135/1498 (9.0%) Carr (1960: 119 22)
Kato Zakro (Middle Minoan) 11/152 (7.2%) Becker (1975: 275)
Khania (Late Minoan) 76/206 (36.9%) McGeorge (1992: 39)
Armenoi (Late Minoan) 367/2069 (17.7%) McGeorge (1992: 39)
Classical (650 150 BC) 36/724 (5.0%) Angel (1944)
Metaponto (7th 2nd c. BC) 192/1828 (10.5%) Henneberg & Henneberg (1998: 535)
Coastal sites.
Permanent and deciduous teeth combined.
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
Dental Pathology in a Greek Colony 273
individuals had lost teeth prior to their death than had only slight deposits. Dental calculus, like
younger individuals, a finding consistent with the caries, abscesses and AMTL, was significantly
age-related increase in AMTL seen in modern more common in older individuals compared
populations. Causes of AMTL include caries, with younger ones. This phenomenon has been
attrition, periodontal disease, trauma and absces- observed in other populations and reflects the
ses (Hillson, 2000). Metabolic diseases such as progressive build-up of calculus with age (Beis-
scurvy can also lead to tooth loss (Wols & Baker, wanger et al., 1989). Few studies of Greek skeletal
2004: 68). Determining the cause(s) of AMTL, remains report the prevalence of dental calculus;
particularly in archaeological samples, may be however, the high rate of calculus in the
difficult. The fact that dental caries and AMTL Apollonia sample is consistent with that of other
most frequently affected the molar teeth in the agricultural populations, and points to a reliance
sample suggests that caries may have been a major on a high carbohydrate diet and/or poor oral
cause of tooth loss in this population. Similar hygiene.
rates of tooth loss among males and females As noted earlier, the relationship between
correspond with the caries data. Periodontal calculus and diet is not straightforward (Hillson,
disease was not scored in this study, but may also 2001), and high calculus rates have been
have contributed to tooth loss among the colo- associated with both high protein and high
nists. Low to moderate levels of tooth wear in the carbohydrate diets (Meiklejohn & Zvelebil, 1991;
sample and the small number of cases of pulp Lieverse, 1999; Lillie & Richards, 2000).
exposure resulting from wear suggest that tooth Significant sex differences, reported for some
wear was not a major contributor to AMTL populations, have been interpreted as reflecting
among the colonists. inequality in access to dietary protein (Lillie &
Richards, 2000). In the case of Apollonia, males
had a higher prevalence of calculus than females
Dental abscesses overall (50.3% compared with 40.6%), and
young adult and middle-aged males had a higher
The prevalence of abscesses in the sample is quite rate than females of the same age groups. None of
low. While it is consistent with the low rate of these differences, however, was statistically
caries and the low to moderate rate of tooth wear significant (Mann-Whitney U test), suggesting
recorded in the sample, it may also reflect the that males and females consumed a similar diet.
extraction of teeth prior to pulp exposure and
infection, the loss of abscessed teeth and the
subsequent remodelling of the alveolar bone long Tooth wear
before death, and/or the inability to observe
abscesses that have not yet penetrated the Occlusal surface wear is an indicator of the type
alveolar bone (Hillson, 2000; Wols & Baker, of foods consumed and methods of food
2004). The significantly higher frequency of preparation. Among ancient Greek populations,
abscesses by tooth socket in older individuals high levels of tooth wear have been attributed to a
compared with younger ones is mirrored in other diet of uncooked and unrefined food items such as
archaeological samples, and reflects the higher coarse cereal grains, the consumption of poorly
rate of caries and tooth wear typically seen in this washed foods, and the use of grinding stones
age group. Similar rates of abscesses in males and (Becker, 1975: 275; McGeorge, 1992: 39 40). In
females mirror the data on caries and AMTL. contrast, low levels of wear are associated with
the consumption of soft foods. While compari-
sons of tooth wear in the Apollonia sample with
Dental calculus those of other archaeological samples are
hindered by the use of different methodologies
The Apollonian diet does not appear to have to record tooth wear, as well as the failure of
promoted the formation of large amounts of publications to report the methods used, the low
calculus on the teeth, as most of the affected teeth to moderate levels of wear recorded in the
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
274 A. Keenleyside
Apollonia sample are consistent with those sample had enamel hypoplasia (Henneberg &
recorded in other agricultural populations, Henneberg, 2001).
including Greek colonial samples from Pithe- Hypoplastic defects in Greek skeletal remains
koussai (Becker, 1995: 279) and rural Metaponto have been interpreted as evidence of nutritional
(Henneberg & Henneberg, 1998: 536), and stress, childhood infections and/or weaning
suggest the consumption of a relatively soft diet. stress. Angel (1984: 64), for example, attributed
The association between tooth wear and age is the significant increase in the prevalence of
reflected in greater levels of tooth wear in older enamel hypoplasia in the Middle Bronze Age to
individuals compared with younger ones. The childhood epidemics resulting from increased
finding that males had a significantly greater population size and density. Henneberg and
degree of wear on the anterior teeth than females colleagues have suggested that the high fre-
raises the possibility that non-dietary factors, quency of defects observed in the rural sample
perhaps activities associated with fishing, may from Metaponto most likely reflects the presence
have contributed to the wearing down of their of some form of chronic, endemic infection,
front teeth. possibly treponemal disease, which appears to
have been common in the population (Henne-
berg & Henneberg, 1998). With respect to the
urban sample from Metaponto, Henneberg &
Enamel hypoplasia Henneberg (2001) suggested that the high
prevalence of hypoplastic defects may reflect
The prevalence of enamel hypoplasia in the infectious diseases spread by crowded living
sample is relatively low. The lack of significant conditions.
sex differences in the frequency of hypoplastic The factors underlying the development of
defects among the colonists suggests that males enamel hypoplasia in the Apollonian colonists
and females both suffered from physiological probably included both nutritional stress and
stresses during early childhood. The higher exposure to infectious diseases. As indicated by
frequency of the condition in younger adults the stable isotopic data, the colonists relied
compared with older individuals is consistent heavily on a terrestrial diet of C3 plants, suggest-
with studies of other archaeological populations ing that nutritional deficiencies may have been
(Larsen, 1997), and supports the hypothesis that common in the population. Palaeopathological
individuals who experience stress during child- analysis of 184 skeletons excavated from the site
hood are more likely to die at a younger age revealed the presence of cribra orbitalia in 28% of
(Duray, 1996). individuals, and iron deficiency anaemia has been
Comparison of the prevalence of enamel proposed as the most likely cause of these lesions
hypoplasia in the Apollonia sample with that (Keenleyside & Panayotova, 2006). Periodic food
of other samples is hindered by methodological shortages, common in antiquity (Garnsey, 1999),
differences in the scoring of this condition, most may also have contributed to the formation of
notably the lack of a minimum defect size these defects. Ancient literary sources indicate
considered to be a hypoplastic defect (Goodman that Greek colonies on the Black Sea coast of
et al., 1980: 527; Goodman & Rose, 1990: 92; Romania and Ukraine suffered periodically from
Hillson, 2000). Nevertheless, the frequency of food shortages during the Hellenistic period
individuals with defects in the Apollonia sample (Garnsey, 1988: 13, 163, 274), and it is likely that
(25.4%) is not significantly higher than that Apollonia experienced similar food shortages on
reported by Bisel & Angel (1985) for Hellenistic occasion.
period Greeks (18.8%) from other archaeological The possibility that the hypoplastic defects
sites. It is lower than that reported for Classical seen in the Apollonian sample resulted from
period Greeks (37.9%) (Bisel & Angel, 1985), and infectious diseases must also be considered.
significantly lower than that reported for Meta- While the skeletons analysed to date show no
ponto, where 78% (88/113) of individuals in the evidence of chronic infection such as that seen at
rural sample and 96% of individuals in the urban Metaponto (Henneberg & Henneberg, 1998),
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
Dental Pathology in a Greek Colony 275
this does not preclude the possibility of infectious The lack of marked sex differences in dental
diseases having been a health problem among the pathology is somewhat surprising given the
colonists. The population of Apollonia is esti- textual and isotopic evidence for dietary differ-
mated to have numbered approximately 3000 ences between males and females. It is useful,
citizens during the 5th to 4th centuries BC however, to look at overall trends in the dental
(Danov, 1948), and excavations have revealed pathology data. Higher rates of caries in females
that during this time period, the centre of the and higher rates of calculus in males do hint at
settlement was characterised by high-density subtle dietary differences in the form of greater
occupation (Nedev & Panayotova, 2003). It is consumption of carbohydrates by females and
therefore conceivable that the population may greater consumption of protein by males. This
have been exposed to pathogenic microorgan- corresponds with the stable carbon isotopic
isms associated with crowded living conditions. values derived from bone carbonate, in particular
Other factors that have been linked to the the isotopic spacing between bone collagen and
development of enamel hypoplasia include the apatite (carbonate), which can be used as an
stresses associated with weaning (Corruccini et al., indicator of trophic level (Krueger & Sullivan,
1985; White, 1994). Infant feeding vessels 1984). Large differences between d13Capatite and
recovered from child graves in Greece suggest d13Ccollagen reflect a more herbivorous diet, while
that weaning may have begun relatively early small differences reflect a more carnivorous diet.
(Fildes, 1986: 25), and literary sources indicate In the Apollonia sample, the apatite-collagen
that cereals were the basic weaning food among spacing is higher for females than males (8.8%
the ancient Greeks (Garnsey, 1999). A recent compared with 7.9%), suggesting greater carbo-
stable isotopic study of infant feeding practices at hydrate consumption by females and greater
Apollonia has revealed that infants were weaned protein consumption by males.
onto a diet of cereals and dairy products, foods When interpreting dental pathology data,
that would have predisposed them to nutritional consideration must also be given to the non-
deficiencies (Kwok, 2007). dietary use of the teeth. While stable isotopic
data and ancient literary texts may provide useful
information on diet, they tell us nothing about
non-dietary uses of teeth that might be reflected
Comparison of the dental pathology data in certain forms of dental pathology. As noted
with the stable isotopic evidence and ancient earlier, greater levels of tooth wear in males
literary sources compared with females, particularly in the
anterior teeth, raises the possibility that non-
The composition of the diet, as indicated by the dietary factors, perhaps activities associated with
dental pathology data, is consistent with the fishing, may have contributed to greater tooth
stable isotopic evidence from Apollonia and with wear among males.
ancient literary texts, both of which indicate a
heavy reliance on carbohydrates. The lack of
significant sex differences in dental pathology is Conclusions
also consistent with the stable carbon and
nitrogen isotopic data derived from bone The prevalence of dental pathology observed at
collagen, which indicate no significant sex Apollonia is typical of that seen in other
differences in the consumption of dietary protein agricultural populations, and indicates the con-
(Keenleyside et al., 2006). In contrast, the dental sumption of a relatively soft, high carbohydrate
pathology data conflict with the ancient literary diet. It is consistent with the stable isotopic
texts, which highlight distinct dietary differences evidence for diet at Apollonia and with the
between males and females, and with the stable ancient literary texts, both of which indicate a
carbon isotopic values derived from bone heavy reliance on carbohydrates. The prevalence
carbonate, which indicate sex differences with of caries among the colonists is similar to that
respect to the overall diet. observed in the Greek colony of Metaponto, for
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
276 A. Keenleyside
which isotopic data also indicates a primary References
reliance on a terrestrial C3 diet.
The higher frequency of dental caries,
Angel JL. 1944. Greek teeth: ancient and modern.
abscesses, calculus and AMTL in older adults
Human Biology 16: 283 297.
compared with younger ones reflects the age- Angel JL. 1984. Health as a crucial factor in the
changes from hunting to developed farming in
progressive nature of these conditions. Sex
the eastern Mediterranean. In Paleopathology at the
differences in dental pathology are not pro-
Origins of Agriculture, Cohen MN, Armelagos GJ
nounced, and, as such, are consistent with the
(eds). Academic Press: New York; 51 73.
stable carbon and nitrogen isotopic data derived
Bass WM. 1987. Human Osteology: A Laboratory and Field
from bone collagen, which indicate no significant
Manual of the Human Skeleton, (3rd edn). Special Pub-
sex differences in the consumption of dietary
lication No. 2. Missouri Archaeological Society:
protein (Keenleyside et al., 2006). In contrast,
Columbia.
these findings conflict with the ancient literary
Becker MJ. 1975. Human skeletal remains from Kato
texts, which attest to distinct dietary differences
Zakro. American Journal of Archaeology 79: 271 276.
between males and females, and with the stable
Becker MJ. 1995. Human skeletal remains from the
carbon isotopic values derived from bone precolonial Greek emporium of Pithekoussai on
Ischia (NA): culture contact in Italy from the early
carbonate, which indicate sex differences with
eighth to the second century B.C. In Settlement and
respect to the overall diet.
Economy in Italy 1500 B.C. to A.D. 1500, Christie N
Despite the lack of strong evidence for sex
(ed.). Papers of the Fifth Conference on Italian
differences in dental pathology, overall trends in
Archaeology: Oxford; 273 281.
the data, particularly when combined with the
Becker MJ. 1996. Appendix: The human skeletons
isotopic data, do hint at subtle dietary differences
from the Archaic cemeteries. In The Archaic Cemeteries.
between males and females, with males consum-
Morgantina Studies, Vol. 5, Lyons CL. Princeton
ing more protein and females consuming more
University Press: Princeton; 227 237.
carbohydrates. As such, the integration of dental
Becker MJ, Donadio A. 1992. A summary of the
pathology data with other types of evidence
analysis of cremated human skeletal remains from
provides greater insight into the diet of past the Greek colony of Pithekoussai at Lacco Ameno,
Ischia, Italy. Old World Archaeology Newsletter 16:
populations than reliance on one source of
15 23.
information alone. A greater degree of tooth
Beiswanger BB, Segreto VA, Mallatt ME, Pfeiffer HJ.
wear in males also hints at sex differences in the
1989. The prevalence and incidence of dental
non-dietary use of the teeth, an observation not
calculus in adults. Journal of Clinical Dentistry 1:
recorded in the ancient texts.
55 58.
Bisel SC, Angel JL. 1985. Health and nutrition in
Mycenaean Greece: a study in human skeletal
remains. In Contributions to Aegean Archaeology: Studies
Acknowledgements
in Honor of William A. McDonald, Wilkie NC, Coulson
WDE (eds). Center for Ancient Studies, University
I would like to thank Kristina Panayotova, Insti- of Minnesota Publications in Ancient Studies:
tute of Archaeology in Sofia, Bulgaria, and Dimi- 197 210.
Brooks S, Suchey JM. 1990. Skeletal age determi-
tar Nedev, Director of the Archaeological
nation based on the os pubis: a comparison of
Museum in Sozopol, Bulgaria, for providing
the Acsádi-Nemesk#ri and Suchey-Brooks
access to the skeletal sample examined in this
methods. Human Evolution 5: 227 238.
study. I am grateful to Tracy Prowse, whose
Brothwell DR, Brothwell P. 1969. Food in Antiquity. A
research inspired this paper, and to Peter Garnsey,
Survey of the Diet of Early Peoples. Thames and Hudson:
Richard Lazenby, and the reviewers of an earlier
London.
draft of this manuscript for their helpful com-
Buikstra JE, Ubelaker D. (eds). 1994. Standards for Data
ments. Fieldwork in Bulgaria was supported by a
Collection from Human Skeletal Remains, Arkansas
Leverhulme Research Fellowship and a grant from
Archaeological Survey Research Series No. 44.
the Social Sciences and Humanities Research Arkansas Archaeological Survey: Fayetteville,
Arkansas.
Council of Canada (#410-2005-0452).
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
Dental Pathology in a Greek Colony 277
Carr HG. 1960. Some dental characteristics of the toric populations from Illinois. Human Biology 52:
Middle Minoans. Man 60: 119 122. 515 528.
Cohen MN. 1989. Health and the Rise of Civilization. Yale Goodman AH, Martin DL, Armelagos GJ, Clark G.
University Press: New Haven. 1984. Indications of stress from bone and teeth. In
Commission on Oral Health. 1982. An epidemio- Paleopathology at the Origins of Agriculture, Cohen MN,
logical index of developmental defects of dental Armelagos GJ (eds). Academic Press: New York;
enamel (DDE Index). International Dental Journal 32: 13 49.
159 167. Grmek MD. 1989. Diseases in the Ancient Greek World
Corruccini RS, Handler JS, Jacobi KP. 1985. Chrono- (transl. M. Muellner and L. Muellner). Johns Hop-
logical distribution of enamel hypoplasias and kins University Press: Baltimore.
weaning in a Caribbean slave population. Human Hadjimarkos DM, Bonhorst CW. 1962. Fluoride- and
Biology 57: 699 711. selenium-levels in contemporary and ancient Greek
Cucina A, Tiesler V. 2003. Dental caries and ante- teeth in relation to dental caries. Nature 193:
mortem tooth loss in the Northern Peten area, 177 178.
Mexico: a biocultural perspective on social status Henneberg RJ. 1998. Dental Health and Affiliations of
differences among the Classic Maya. American Jour- Inhabitants of the Ancient Greek Colony in Metaponto, Italy
nal of Physical Anthropology 122: 1 10. (6th 3rd Century BC). PhD thesis, University of the
Dalby A. 1996. Siren Feasts: A History of Food and Witwatersrand, South Africa.
Gastronomy in Greece. Routledge: London. Henneberg M, Henneberg RJ. 1998. Biological
Danov KH. 1948. Kam starata istoriya na Sozopol, characteristics of the population based on analysis
Zapadnoto Chernomorie I Trakiya (On the ancient of skeletal remains. In The Chora of Metaponto: The
history of Sozopol, the Western Pontic coast and Necropoleis Vol. II, Carter JC (ed.). University of
Thrace). Izvestiya na Bulgarskoto Istorichesko Druzhestvo Texas Press: Austin, TX; 503 562.
22-24: 172 183. Henneberg M, Henneberg RJ. 2001. Analysis of
Dawes C. 1970. Effects of diet on salivary secretion human skeletal and dental remains from Metaponto
and composition. Journal of Dental Research 70: (7th 2nd c. BC). AttiTaranto 40: 461 474.
1263 1272. Henneberg RJ, Henneberg M. 2003. The diet of the
Duray S. 1996. Dental indicators of stress and reduced Metapontine population as reconstructed from the
age at death in prehistoric Native Americans. Amer- physical remains. In Living Off the Chora: Diet and
ican Journal of Physical Anthropology 99: 275 286. Nutrition at Metaponto, Carter JC (ed.). University of
Erdal YS, Duyar I. 1999. A new correction procedure Texas Press: Austin, TX; 29 36.
for calibrating dental caries frequency. American Hillson SW. 1996. Dental Anthropology. Cambridge
Journal of Physical Anthropology 108: 237 240. University Press: Cambridge.
Fildes V. 1986. Breasts, Bottles and Babies: A History Hillson SW. 2000. Dental pathology. In Biological
of Infant Feeding. Edinburgh University Press: Anthropology of the Human Skeleton, Katzenberg MA,
Edinburgh. Saunders SR (eds). Wiley-Liss: New York; 249 286.
Frayer DW. 1984. Tooth size, oral pathology and class Hillson SW. 2001. Recording dental caries in archaeo-
distinctions: evidence from the Hungarian Middle logical human remains. International Journal of
Ages. Anthropologiai Kzlemenyek 28: 47 54. Osteoarchaeology 11: 249 289.
Garnsey P. 1988. Famine and Food Supply in the Graeco- Hillson SW. 2005. Teeth (2nd edn). Cambridge Uni-
Roman World: Responses to Risk and Crisis. versity Press: Cambridge.
Cambridge University Press: Cambridge. Jameson M. 1983. Famine in the Greek world. In Trade
Garnsey P. 1999. Food and Society in Classical Antiquity. and Famine in Classical Antiquity, Garnsey P, Whittaker
Cambridge University Press: Cambridge. CR (eds). Cambridge Philological Society, Supple-
Garnsey P, Whittaker CR (eds). 1983. Trade and Famine mentary Volume 8: 6 16.
in Classical Antiquity. Cambridge Philological Keenleyside A, Panayotova K. 2005. A bioarchaeo-
Society, Supplementary Volume 8: Cambridge. logical study of the Greek colonial population of
Goodman AH, Rose JC. 1990. Assessment of systemic Apollonia Pontica. Archaeologia Bulgarica 9(2):
physiological perturbations from dental enamel 21 38.
hypoplasias and associated histological structures. Keenleyside A, Panayotova K. 2006. Cribra orbitalia
Yearbook of Physical Anthropology 33: 59 110. and porotic hyperostosis in a Greek colonial popu-
Goodman AH, Armelagos GJ, Rose JC. 1980. Enamel lation from the Black Sea. International Journal of
hypoplasias as indicators of stress in three prehis- Osteoarchaeology 16: 373 384.
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
278 A. Keenleyside
Keenleyside A, Schwarcz HP, Panayotova K. 2006. Lukacs JR. 1995. The  Caries Correction Factor : a new
Stable isotopic evidence of diet in a Greek colonial method of calibrating dental caries rates to com-
population from the Black Sea. Journal of Archaeolo- pensate for antemortem loss of teeth. International
gical Science 33: 1205 1215. Journal of Osteoarchaeology 5: 151 156.
Kelley MA, Levesque DR, Weidl E. 1991. Contrasting Lukacs JR. 1996. Sex differences in dental caries rates
patterns of dental disease in five early Northern with the origin of agriculture in South Asia. Current
Chilean Groups. In Advances in Dental Anthropology, Anthropology 37: 147 153.
Kelley MA, Larsen CS (eds). Wiley-Liss: New York; Macchiarelli R. 1989. Prehistoric   fish-eaters  along the
203 214. Eastern Arabian coasts: dental variation, morphology,
Krueger HW, Sullivan CH. 1984. Models for carbon and oral health in the Ra s al-Hamra community
isotope fractionation between diet and bone. In (Qurum, Sultanate of Oman, 5th-4th millennia BC).
Stable Isotopes in Nutrition, Turnlund JE, Johnson PE American Journal of Physical Anthropology 78: 575 594.
(eds). American Chemical Society Symposium Malde MK, Maage A, Macha E, Julshamn K, Bjorvatn
Series 258: 205 222. K. 1997. Fluoride content in selected food items
Kwok CS. 2007. Baby Bones, Food, and Health: Stable from five areas in East Africa. Journal of Food Con-
Isotopic Evidence for Infant Feeding Practices in the Greek sumption and Analysis 10: 233 245.
Colony of Apollonia (5th 2nd centuries BC). MA thesis, Martin TR. 1996. Ancient Greece. From Prehistoric to Hellenic
Department of Anthropology, Trent University. Times. Yale University Press: New Haven, CT.
Laine MA. 2002. Effect of pregnancy on periodontal McGeorge PJP. 1992. Part II. The burials. In Late
and dental health. Acta Odontologica Scandinavica 60: Minoan III Burials at Khania: the Tombs, Finds and
257 264. Deceased in Odos Palama, Hallager BP, McGeorge
Larsen CS. 1995. Biological changes in human popu- PJP (eds). Studies in Mediterranean Archaeology
lations with agriculture. Annual Review of Anthropology Vol. XCIII: Göteborg; 29 44.
24: 185 213. Meiklejohn C, Zvelebil M. 1991. Health status of
Larsen CS. 1997. Bioarchaeology. Interpreting Behaviour European populations of the agricultural transition
from the Human Skeleton. Cambridge University Press: and the implications for the adoption of farming. In
Cambridge. Health in Past Societies: Biocultural Interpretations of Human
Lieverse AR. 1999. Diet and the aetiology of dental Remains in Archaeological Contexts, Bush H, Zvelebil M
calculus. International Journal of Osteoarchaeology 9: (eds). British Archaeological Reports International
219 232. Series 567: Oxford; 129 145.
Lillie MC. 1996. Mesolithic and Neolithic populations Meindl RS, Lovejoy CO. 1985. Ectocranial suture clo-
of Ukraine: indications of diet from dental pathol- sure: a revised method for the determination of skeletal
ogy. Current Anthropology 37: 135 142. age at death based on the lateral-anterior sutures.
Lillie MC, Richards M. 2000. Stable isotope analysis American Journal of Physical Anthropology 68: 57 66.
and dental evidence of diet at the Mesolithic Musgrave JH. 1976. Appendix A: the human remains.
Neolithic transition in Ukraine. Journal of Archae- In An Early Christian Osteotheke at Knossos, Catling
ological Science 27: 965 972. HW, Smyth D eds. Annual of the British School
Lingström P, Borrman H. 1999. Distribution of dental at Athens 71: 40 46.
caries in an early 17th century Swedish population Musgrave JH, Popham MR. 1991. The Late Helladic
with special reference to diet. International Journal of III C Intramural Burials at Lefkandi, Euboea. Annual
Osteoarchaeology 9: 395 403. of the British School at Athens 86: 273 296.
Littleton J, Frohlich B. 1993. Fish-eaters and farmers: Nedev D, Panayotova K. 2003. Apollonia Pontica
dental pathology in the Arabian Gulf. American (end of the 7th 1st centuries B.C.). In Ancient Greek
Journal of Physical Anthropology 92: 427 447. Colonies in the Black Sea, Vol. 1, Grammenos D,
Lovejoy CO, Meindl RS, Pryzbeck TR, Mensforth RP. Petropoulos E (eds). Archaeological Institute of
1985. Chronological metamorphosis of the auricu- Northern Greece: Thessaloniki; 95 155.
lar surface of the ilium: a new method for the Nelson GC, Lukacs JR, Yule P. 1999. Dates, caries and
determination of age at death. American Journal of early tooth loss during the Iron Age of Oman.
Physical Anthropology 68: 15 28. American Journal of Physical Anthropology 108: 333 343.
Lubell D, Jackes M, Schwarcz HP, Knyf M, Meikle- Panayotova K. 1998. Apollonia Pontica: recent dis-
john C. 1994. The Mesolithic-Neolithic transition coveries in the necropolis. In The Greek Colonisation of
in Portugal: isotopic and dental evidence of diet. the Black Sea Area, Tsetskhladze GR (ed.). F. Steiner:
Journal of Archaeological Science 21: 201 216. Stuttgart; 97 114.
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa
Dental Pathology in a Greek Colony 279
Papathanasiou A. 2005. Health status of the Neolithic Smith SK, Cook DC. 1991. Mesolithic to modern dental
population of Alepotrypa Cave, Greece. American wear at Franchthi Cave, Greece. Paper presented at the
Journal of Physical Anthropology 126: 377 390. Annual Meeting of the Society for American
Powell ML. 1988. Status and Health in Prehistory. Smith- Archaeology, New Orleans, 25 April.
sonian Institution Press: Washington, DC. Tayles N, Dommett K, Nelsen K. 2000. Agriculture
Prowse TL. 2001. Isotopic and Dental Evidence for Diet from and dental caries: the case of rice in prehistoric
the Necropolis of Isola Sacra (1st 3rd centuries AD), Italy. Southeast Asia. World Archaeology 32: 68 83.
PhD thesis, Department of Anthropology, McMas- Thylstrup A, Fejerskov O. 1994. Textbook of Clinical
ter University, Hamilton, Ontario. Cariology. Munksgaard: Copenhagen.
Rathbone D. 1983. The grain trade and grain Tsetskhladze G. 1997. The artistic taste in Phanagoria.
shortages in the Hellenistic East. In Trade and Famine Greek colonies on the Black Sea. Apollo (July 1997):
in Classical Antiquity, Garnsey P, Whittaker CR (eds). 7 9.
Cambridge Philological Society, Supplementary Turner CG II. 1979. Dental anthropological indica-
Volume 8: 45 55. tions of agriculture among the Jomon People of
Roberts CA, Manchester K. 1995. The Archaeology of Central Japan. American Journal of Physical Anthropol-
Disease (2nd edn). Cornell University Press: ogy 51: 619 636.
Ithaca, NY. Ubelaker DH. 1989. Human Skeletal Remains (2nd edn).
Scott EC. 1979. Dental wear scoring technique. Taraxacum: Washington, DC.
American Journal of Physical Anthropology 51: 213 Venedikov I. 1963. Les fouilles de la necropole en
218. 1947-1949. In Apoloniya razkopkite v nekropola na Apol-
Sealy JC, van der Merwe NJ. 1988. Social, spatial and oniya 1947-1949. Sofia.
chronological patterning in marine food use as Walker PL, Erlandson JM. 1986. Dental evidence from
determined by d13C measurements of Holocene prehistoric dietary change on northern Channel
human skeletons from the southwestern Cape, Islands, California. American Antiquity 51: 375 383.
South Africa. World Archaeology 20: 87 102. Walker PL, Hewlett BS. 1990. Dental health, diet, and
Siebert G, Trautner K. 1985. Fluoride content of social status among central African foragers and
selected human food, pet food and related materials. farmers. American Anthropologist 92: 383 398.
Zeitschrift für Ernährungswissenschaft 24: 54 66. White CD. 1994. Dietary dental pathology and cul-
Smith BH. 1984. Patterns of molar wear in hunter- tural change in the Maya. In Strength in Diversity,
gatherers and agriculturalists. American Journal of Herring DA, Chan L (eds). Canadian Scholar s
Physical Anthropology 63: 39 56. Press: Toronto; 279 302.
Smith SK. 2000. Skeletal and dental evidence for Wols HD, Baker JE. 2004. Dental health of elderly
social status in Late Bronze Age Athens. In Palaeodiet Confederate veterans: evidence from the Texas
in the Aegean, Vaughan SJ, Coulson WDE (eds). State Cemetery. American Journal of Physical Anthro-
Oxbow Books: Oxford; 105 113. pology 124: 59 72.
Copyright # 2007 John Wiley & Sons, Ltd. Int. J. Osteoarchaeol. 18: 262 279 (2008)
DOI: 10.1002/oa