Content uploaded by James Watson
Author content
All content in this area was uploaded by James Watson
Content may be subject to copyright.
The Atacama Desert of northern Chile and
southern Peru is a 966-km (600-mi) strip of
land along the Pacific coast and bounded to
the east by the Andes Mountains. It is the driest
desert in the world (Rech et al. 2003) yet supported
some of the earliest permanently settled foraging
groups centuries prior to the introduction of agri-
culture in the region (Arriaza 1995). The extremes
of the Atacama are balanced by the immense biotic
richness of the Pacific Ocean along the coast, nar-
row but fertile river valleys that provide direct
access to diverse resources in higher elevations,
and the rich marshy deltas that connect these two
biological corridors. Ancient populations living in
the region had to adapt both culturally and biolog-
ically to an environment of extremes, and nowhere
is this more evident than in the transition to agri-
culture in the Lower Azapa Valley of northwest
Chile. Here we explore the cultural and biological
processes involved in, and the consequences of,
dietary change during the Formative transition in
skeletal remains from a series of archaeological
sites near Arica, Chile.
The foraging- to- farming transition represents
FORMATIVE ADAPTATIONS, DIET, AND ORAL HEALTH IN THE
AZAPA VALLEY OF NORTHWEST CHILE
James T. Watson, Iván Muñoz Ovalle, and Bernardo T. Arriaza
Indicators of oral health were recorded in a sample of 200 Formative period (1500 B.C.–A.D. 500) skeletons from archae-
ological sites located in the Lower Azapa Valley of northwest Chile. This period represents a major shift in subsistence
strategies in the Atacama Desert, as coastal groups adopted agriculture and moved deeper into the valley. Frequencies of
caries and antemortem tooth loss were compared between site locations (coast vs. valley) and by archaeological phase
(early vs. late) to interpret the degree to which these incipient agriculturalists were reliant on domesticated resources. Over-
all, frequencies of caries (11.9 percent) and tooth loss (11.6 percent) are somewhat higher than for other prehistoric groups
practicing a mixed subsistence strategy. However, residents of the interior valley exhibited significantly more dental decay
and tooth loss than those along the coast. Our results identify that although the Formative period residents of the Lower
Azapa Valley practiced a mixed subsistence strategy, the degree of reliance on agricultural production differed between the
coast and the valley. We propose that these differential patterns in oral health are tied to local investment, adaptive cycles,
and niche construction.
Indicadores de la salud oral fueron registrados en una muestra de 200 esqueletos del periodo Formativo (1500 a.C.–500 d.C.)
excavada en varios sitios arqueológicos del Valle de Azapa en el noroeste de Chile. Este periodo representa un cambio signi-
ficativo en la estrategia de subsistencia dentro del Desierto de Atacama, conforme los grupos costeros adoptaron la agricul-
tura y varios de ellos se mudaron al valle. Se compararon las frecuencias de caries y pérdida de dientes entre las localidades
(costa y valle) y entre las fases arqueológicas (temprana y tardía) para interpretar en qué grado estos grupos de agricultores
incipientes dependían de recursos domesticados. En general, las frecuencias de caries (11,9 percent) y de pérdida de dientes
(11,6 percent) son ligeramente más elevadas en comparación a otros grupos que practican una estrategia de subsistencia
mixta. Sin embargo, los residentes del valle exhibieron más patología oral que los que vivían en la costa. Estos resultados
identifican que durante el periodo Formativo los residentes del Valle de Azapa practicaban una estrategia de subsistencia mixta
pero el porcentaje de dieta agrícola era diferente entre la costa y el valle. Es muy probable que la diferencia entre los patrones
de salud oral esté ligada a una inversión local, ciclos de adaptación y la construcción de nichos.
James T. Watson 䡲Arizona State Museum and School of Anthropology, University of Arizona, P.O. Box 210026, Tucson,
AZ 85721 (watsonjt@email.arizona.edu)
Iván Muñoz Ovalle 䡲Departamento de Arqueología y Museología, Universidad de Tarapacá, Arica, Chile
Bernardo T. Arriaza 䡲Instituto de Alta Investigación, Departamento de Antropología y Centro de Investigaciones del
Hombre en el Desierto, Universidad de Tarapacá, Arica, Chile
Latin American Antiquity 21(4), 2010, pp. 423–439
Copyright ©2010 by the Society for American Archaeology
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 423
"(&(
one of the most profound changes in human behav-
ior during the last 10,000 years. Its impacts on
human health are important to understand because
they offer insights into the etiology and distribu-
tion of many modern diseases. The global effects
of subsistence changes to human health have been
well documented from numerous skeletal samples
(see Cohen and Armelagos 1984; Cohen and Crane-
Kramer 2007; Larsen 2000; and Steckel and Rose
2002 for synthetic treatments). Archaeological
research has shown that, in general, the shift to
agriculture caused a decline in human health
(Cohen and Armelagos 1984). The causes of this
decline relate to two broad phenomena: larger
sedentary populations and changes in subsistence.
Larger sedentary populations created poor sanitary
conditions, which promoted the spread of com-
municable diseases (Goodman et al. 1984; Larsen
1997, 2000, 2002). Changes in subsistence activi-
ties, dietary breadth, and diet composition caused
more dental disease and created nutritional inade-
quacies (Nelson et al. 1994; Steckel et al. 2002).
Regionally specific changes in dietary compo-
sition associated with the “Neolithic Revolution”
(Childe 1925) often have the first measurable
impact on the human body (detectable in skeletal
remains). Consumption of mechanically processed
and carbohydrate- rich domesticated cultigens,
which often stick to grooves and fissures on tooth
surfaces and between teeth and gums, increases
oral bacterial loads and consequently the incidence
of oral disease (Hillson 2001, 2002; Larsen 1981,
1995). As a result, agriculturalists tend to have ele-
vated rates of dental decay, malocclusion, and prob-
lems associated with nutrition during enamel
formation (amelogenesis) when compared with for-
aging societies (Bridges 1989; Larsen 1995; Turner
1979). An associated decrease in occlusal surface
wear additionally results in plaque buildup, stabi-
lizing bacterial colonization and increasing poten-
tials for cariogenesis and periodontitis (Hillson
2001).
Numerous bioarchaeological case studies have
documented a significant difference in the fre-
quency of caries between subsistence strategies
(foragers vs. agriculturalists) and an increase over
time— from the Paleolithic/Archaic to the
Neolithic/Formative period (see Turner 1979 for a
global perspective). A few exceptions to this pat-
tern have been identified, such as the consumption
of wild tubers in Southeast Asia (Tayles et al. 2000)
and succulents in the North American desert west
(Hartnady and Rose 1991; Watson 2008). These
examples highlight the importance of considering
human adaptations and their consequences under
“local” ecological circumstances, considering the
constraints and variability available to groups under
different environmental and technological condi-
tions in prehistory. The prehistoric residents of the
Lower Azapa Valley had access to three diverse
local environmental zones (coast, marsh, and river
valley) when cultigens and farming technology
arrived in the area prior to ca. 1500 B.C. Our pri-
mary goal in this article is to examine oral health
indicators in Formative period (1500 B.C.–A.D.
500) skeletal samples from the Lower Azapa Val-
ley to interpret the degree to which these incipient
agriculturalists utilized the various wild and domes-
ticated resources available.
Formative Period In The Lower Azapa Valley
The Formative period (1500 B.C.–A.D. 500) in the
Lower Azapa Valley presents the first evidence for
settlement away from the coast— in the valley
interior— and the beginning of local agricultural
investment. The skeletal samples analyzed in this
study were recovered from two groups of Forma-
tive period sites located in the Lower Azapa Valley
(Figure 1). Coastal sites (Playa Miller-7, Quiani-
7) are located directly adjacent to the shoreline and
in close proximity to the marshy mouth of the San
José River (Focacci 1974, 1980). Valley sites
(Azapa-14, Azapa-70, Azapa-71, Azapa-75, and
Azapa-115) are located close to each other on Pleis-
tocene benches overlooking the San José River
floodplain approximately 12.5 km from the coast
(Focacci 1980; Focacci and Erices 1972–1973;
Santoro 1980a, 1980b).
Material culture recovered from these sites is
similar in the broader sense that cultigens and items
representing agricultural technology are present in
both areas (Rivera 1984, 1991). Differences are
present, but these are minor compared to the clear
impact that the arrival of agriculture had on sub-
sistence strategy, as observed in the archaeobotan-
ical record of these sites (Muñoz 2004). Subtle
differences in artifact assemblages separate the For-
mative period in the Lower Azapa Valley into early
and late archaeological phases, but these are largely
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 424
analytical tools (Arriaza and Standen 2008). The
early phase (1500–500 B.C.) is termed Faldas de
Morro at coastal sites and Azapa at valley sites
(Focacci 1980; Santoro 1980b). The late phase (500
B.C.–A.D. 500) is termed El Lauchoat coastal sites
and Alto Ramírez at valley sites (Focacci 1974;
Santoro 1980b).
Early theories about the arrival of cultigens and
agricultural technology in the Lower Azapa Valley
centered on demic diffusion (Kelley et al. 1991;
Kolata 1993; Muñoz 1989; Rivera 1984, 1991;
Rothhammer et al. 1989; Rothhammer et al. 2002;
Santoro 1980a, 1980b; Varela and Cocilovo 2002).
Researchers suggested that, despite evidence for
earlier contact, the Formative period represents the
first migration of altiplano farmers down the val-
ley to settle with an existing technology, including
domesticated camelids and a reliance on agricul-
ture. Pastoralism and the cultivation of quinoa,
manioc, corn, beans, squash, and hot chilies (aji)
represent a drastic change in dietary base from the
previous marine and terrestrial foraging base of the
Chinchorro culture (Muñoz 2004). However, evi-
dence from both coastal and valley sites indicates
that a mixed subsistence strategy was practiced for
the duration of the Formative period in the Lower
Azapa Valley (Muñoz 2004; Santoro 1980a).
Recent work has demonstrated a cultural and
biological continuity that connects the earliest res-
idents in the area to subsequent populations that
spread up the valley (Cassman 1997; Munizaga
1964, 1974, 1980; Sutter 1997, 2006; Sutter and
Mertz 2004). Of particular relevance to this study
is the exhaustive work of Sutter (1997, 2000, 2006;
Sutter and Mertz 2004), who examined a suite of
inheritable discrete dental traits in the same skele-
tal samples considered here and documented close
genetic relationships between groups living at both
coastal and valley sites. He further demonstrated
that these relationships continued over time,
extending from Archaic period Chinchorro foragers
to later post- Tiwanaku settlements in the Lower
Azapa Valley. Sutter’s findings are additionally sig-
nificant to this study because they allow us to con-
sider differences in oral health across biologically
related groups, a factor that could otherwise facil-
$*)('%*$&# !
Figure 1. Location of sites in Lower Azapa Valley.
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 425
itate genetically based differences in susceptibility
to oral diseases (Hillson 2002).
Subsistence
Archaic period Chinchorro groups comprised
approximately 6,000 years of stable marine and
terrestrial foraging (Arriaza 1994, 1995). Chin-
chorro settlements flourished along the coast of
northern Chile and southern Peru, largely relying
on the rich biotic diversity of marine resources
available at settlement’s edge. A diet apparently low
in the consumption of carbohydrates is corrobo-
rated by Kelley et al. (1991), who found only five
carious teeth out of 823 (.6 percent caries rate) at
the site of Morro-1. The introduction of domesti-
cated cultigens and animals and agricultural and
pastoral technology prior to the beginning of the
Formative period therefore represents a major shift
in subsistence strategy and lifestyle from the pre-
vious inhabitants of the area.
Some continuity can be observed in the contin-
ued importance of marine resources to residents of
the Lower Azapa Valley (Muñoz 2004). Coastal res-
idents continued to rely heavily on the ocean as evi-
denced by large shell middens, abundant remains
of sea mammals, and artifact assemblages largely
based in marine foraging technology. These mate-
rial remains are complemented by ceramics and
plant processing technology as well (Muñoz 2004).
Marine resources continued to be exploited through
later occupations along the coast (Muñoz 1993;
Muñoz and Focacci 1985). Domesticated animals
were also exploited to some extent by coastal res-
idents. Focacci (1974) documents an immense
diversity of grave offerings at the site of Playa
Miller-7, including tools made from camel bones
as well as sea lion and birds.
Formative period sites from the valley also pro-
duced sizable quantities of marine faunal remains,
including various species of fish, shellfish, and sea
mammals, in addition to artifacts associated with
marine foraging (Muñoz 2004). Excavation at
Tumulus 6 at the site of Azapa-70 revealed a vari-
ety of funerary objects that reflected a direct rela-
tionship with marine foraging, such as bone
harpoons, loincloths, and fishing nets (Muñoz
1989). Similar discoveries during excavations at
Azapa-115 separated closely interred individuals
dating from the terminal Formative period and the
beginning of the Medio period— the subsequent
period defined by influence from Tiwanaku
(Muñoz 1995–1996). Grave accompaniments
among the Formative period burials reflected ties
to coastal foraging and perhaps a “coastal” iden-
tity, vs. Tiwanaku- influenced design styles in pot-
tery and textiles among the early Medio period
burials (Muñoz 1995–1996).
The importance of marine resources in the diet
of Formative period valley residents was similarly
demonstrated using stable isotopes in a sample of
seven mummies recovered from the site of Azapa-
75 (Aufderheide et al. 2002). The primary conclu-
sions of the study are that pulmonary diseases were
one of the major causes of death in these groups,
and the authors specifically attribute investment in
agropastoralism in the valley interior as a causative
agent compared to groups along the coast (Aufder-
heide et al. 2002). Aufderheide et al. (2002) addi-
tionally used nitrogen isotopes from bone collagen
to determine that 52 percent of dietary protein was
contributed by marine meat. The remaining protein
was contributed by terrestrial meat sources, maize,
and other plants (Aufderheide et al. 2002). These
findings provide a marked contrast to the bone iso-
tope ratios of earlier Chinchorro samples, which
estimated that 80 to 90 percent of the diet was com-
posed of marine resources (Aufderheide et al.
1993). They also note that coprolites of the valley
residents lacked evidence for fish tapeworms com-
mon among coastal groups and therefore suggest
that they were cooking the marine foods they con-
sumed, a significant difference in food processing
compared to coastal groups.
Diverse assemblages of plant remains recov-
ered from Formative period sites, both wild and
domesticated, are equally ubiquitous in coastal and
valley locations (Belmonte 1998; Belmonte et al.
1988; Erices 1975; Muñoz 1981; Rivera 1977).
Erices (1975) and Focacci (1974) have identified
a mix of wild and domesticated flora from Playa
Miller-7. Erices (1975) documents similar assem-
blages from later Medio (A.D. 390–1000) and
Desarrollo Local (ca. A.D. 1200) period contexts
at sites Azapa-6 and Playa Miller-4, respectively.
Muñoz (1981) and Rivera (1977) examined plant
remains from funerary contexts and coprolites from
four Formative period valley sites (Azapa-12, -70,
-83, and -122) to identify maize consumption in
addition to a variety of wild flora. Belmonte (1998)
found significant continuity in the plants exploited
"(&(
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 426
from the Chinchorro site of Playa Miller-3 to the
Formative period valley sites of Azapa-8, -11, -75,
and -141. He also includes a botanical inventory of
the local wild species in the coastal valleys of Arica.
Table 1 demonstrates the variety of plants recov-
ered from Formative period contexts in two con-
temporaneous sites in the Lower Azapa Valley and
identifies a similarity in their distribution between
coast and valley.
In their study of mummies from Azapa-75,
Aufderheide et al. (2002) used carbon isotope ratios
to estimate that they likely consumed a mixed diet
of approximately 64 percent nonmaize terrestrial
plants, including both wild and domesticated
resources (Aufderheide et al. 2002). Radiocarbon
dates run on soft tissue from three individuals in
the sample returned dates ranging from 351 B.C.
to A.D. 504, which spans the Alto Ramírez phase
(Aufderheide et al. 2002). The combination of
radiocarbon dates, isotopic analyses, and health
assessments indicates that although the investment
in agropastoralism was certainly affecting the
health of these individuals, the diet still consisted
of a mix of terrestrial and marine resources and both
wild and domesticated species. This pattern lasted
through the end of the Formative period.
Burial
Concomitant with changes in subsistence, a marked
shift in burial practices also occurred in the area
just prior to the Formative period. Various mum-
mification practices common to the Archaic period
residents of the coast cease (Arriaza 1995), and a
new form of constructing social identity through,
and for, the dead begins with the construction of
tumuli (tumulos)—layered earthen burial mounds
(Focacci 1980; Muñoz 1989, 1995–1996, 2004).
These represent major technological and ideolog-
ical changes and likely mark a significant social
transition in the Lower Azapa Valley.
Continuity along the coast can be observed in
the shared space used for disposal of the dead
through time (Arriaza and Standen 2008). Forma-
tive period sites on the coast are cemeteries located
on approximately 5 km of beach south of where
the San José River empties into the Pacific. Playa
Miller-7 is a densely packed cemetery that dates to
the late phase stretched along the colluvial bench
just above the beach (Focacci 1974). Quiani-7 is a
small cemetery that dates to the early phase located
at the base of a rocky precipice south of Playa
Miller. Although limited materials were recovered
from this cemetery, evidence suggests that these
burials represent a transition between Archaic for-
aging and Formative agropastoralism along the
coast. In addition, 11 burials from the contempo-
raneous site of Camarones-15 (located on the coast
at the adjacent river mouth to the south) were added
to bolster the limited number of remains from the
early phase on the coast. Burial patterns at all of
these sites include individual tombs in a variety of
positions and orientations with an assortment of
$*)('%*$&# !
Table 1. Macrobotanical Remains from Formative Period Sites in the Lower Azapa Valley.
Wild Plants (species) PLM-7aAZ-70bDomesticated Plants (species) PLM-7aAZ-70b
Algarrobo (Prosopis chilensis)X Achira (Canna edulis)XX
Amaranthaceae (Amaranthus sp.) X Ají (Capsicum annum)X
Cane (Arundo donax)X Bean (Phaseolus vulgaris)XX
Chilca (Pluchea sp./Bacharis sp.) X Coca (Erithroxylum coca)X
Cola de Caballo (Equisetum sp.) XX Maíze (Zea mays)XX
Cotton (Gossypium barbadense)X X Manioc (Manihot utilissima)X X
Grasses (Poaceae sp.) X Pallar (Phaseolus lunatus)XX
Junquillo (Scirpus sp.) X Potato (Solanum tuberosum)X
Malvaceae (Tarasa sp.) X Quinua (Chenopodium quinoa)X X
Mucumae (Mucuma elliptica)X X Squash (Cucurbita sp.) XX
Pacae (Inga feuillet)XXSweet Potato (Ipomea batatas)X X
Pimiento (Schinus molle)X
Sauce (Salix humboldtiana)X
Sorona (Tessaria absinthioides)X
Totora (Typha angustifolia)XX
a(Erices 1975)
b(Muñoz 1985; Romero et al. 2004)
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 427
grave accompaniments. Artifacts varied consider-
ably, from fishing hooks of bone and copper to pot-
tery and domesticated cultigens. The patterns
observed in these cemeteries, including body treat-
ment and associated artifacts, are similar to those
encountered in Formative period sites deeper in the
Azapa Valley.
The Formative period sites located in the Azapa
Valley are burial locations characterized by tumuli,
the burial mounds of varying size located on the
fluvial terraces adjacent to the San José River. They
all share similarities in construction, consisting of
burials covered by layers of vegetal material, river
stones, and sediment (Muñoz 2004). Most of the
plants used in the construction of these layers are
from local wild species commonly found at local-
ized marshy areas created by the river. The con-
struction layers do not appear to occur in any
particular order, however. Artifacts recovered from
within the layers include fragments of pottery and
textiles, lithic debitage, ground stone, and seashells.
Burials within the tumuli are generally either
complete individuals, discrete burials of several
crania, or incomplete individuals; some were dis-
turbed, and others were redeposited with grave
offerings. There are no apparent patterns to burial
position or orientation. The type and quantity of
grave goods and offerings also vary considerably,
from diverse textile objects and ceramic urns, to
fishing and foraging tools, to agricultural tools and
products (Muñoz 2004). Many individuals com-
pletely lack evidence for associated funerary
deposits. The majority of offerings found in the con-
structed layers of the tumuli consist of agricultural
products, including corncobs, squash, and chili,
and likely reflect the importance of production to
these groups. Radiocarbon dates recovered from
tumuli indicate centuries of contiguous use and
reuse in their construction (Muñoz 2004).
The skeletal samples considered here represent
approximately 2,000 years of occupation of the
Lower Azapa Valley during the Formative period.
These sites embody a period when domesticated
cultigens and animals are incorporated into exist-
ing local subsistence strategies. Domesticates and
agropastoral technology allow some groups to
move deeper into the valley to exploit additional
niches. Dependence on these resources becomes
more important among subsequent groups in the
area. These skeletal samples provide the opportu-
nity to examine the degree to which cultigens and
agricultural investment were relied upon, between
the coast and the valley and over the course of the
Formative period.
Materials
The skeletal samples analyzed in this study are part
of the osteological collection held in repository at
the Universidad de Tarapacá Museo Arqueológico
San Miguel de Azapa, Chile. We examined upper
and lower dental segments from crania of 200 adult
individuals (age >15) from seven Formative period
sites (Table 2) in the Lower Azapa Valley, and one
contemporaneous site from the Camarones Valley
coast, in northwest Chile. Frequencies of caries
and antemortem tooth loss (AMTL) are consid-
ered by location and archaeological phase to look
for differential patterns in resource dependence
among these incipient agriculturalists. Table 2 dis-
plays the composition of the study sample from the
Lower Azapa Valley. The preservation of dentition
in the sample is generally good, however, a num-
ber of individuals are missing teeth lost post-
mortem, a pattern that differentially affected
anterior teeth to a greater extent than the posterior
teeth.
Sex was estimated for individuals in the sample
by examining standard macroscopic aspects of the
pelvis and/or cranium (Buikstra and Ubelaker
1994). The sample includes relatively even num-
bers of males (n= 101) and females (n= 99), and
although they are not evenly distributed within sites
(see Table 2), chi- square tests identify that when
samples are pooled the mean number of males and
females does not differ significantly between coast
and valley locations (2= .71; df = 1; p= .39) or
between early and late archaeological phases (2=
.18; df = 1; p= .67).
Age was estimated for the sample using dental
eruption (Ubelaker 1978) and relative rates of den-
tal attrition following generalized standards estab-
lished by Brothwell (1989), epiphysis fusion, cranial
suture closure, and macroscopic degenerative
changes in the pubic symphysis following standards
established by Buikstra and Ubelaker (1994). The
sample was classified into four ordinal, nine- year age
cohorts (15–24, n= 49; 25–34, n= 58; 35–44, n=
71; 45–54, n= 22), resulting in a skewed distribu-
tion of individuals by age. The greatest portion of
"(&(
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 428
individuals in the sample (35.5 percent) died between
the third and fourth decade of life. The smallest por-
tion of individuals (11 percent) represents the final
decades of life. Chi- square tests demonstrate that
there are more older individuals in the samples from
the coastal sites (2= 17.13; df = 3; p= .01) but that
the mean number of individuals does not differ sig-
nificantly between early and late archaeological
phases (2= .48; df = 3; p= .92). Mean ages of the
samples are displayed by site in Table 2.
The age structure of skeletal samples is an
important consideration, especially when measur-
ing caries and tooth loss because they are both age-
progressive diseases and therefore cannot be
compared across samples or variables with differ-
ent age structures (Hillson 2001). Results of the chi-
square tests identify that although age does not
differ significantly between archaeological phases,
it could significantly affect comparisons of oral
pathology between coast and valley samples. The
frequencies of both caries and AMTL observed
within the coastal sample are generally low even
among the oldest individuals in the sample. Ulti-
mately, tooth loss was not found to be significantly
impacted when age was controlled for in our sub-
sequent analyses.
Methods
Although a suite of oral health indicators can be
observed in archaeological populations, here we
chose to focus on the frequency of dental caries and
antemortem tooth loss because these represent
comparable quantitative measures that have been
documented to commonly correlate with differing
subsistence strategies. Caries are the result of the
focalized demineralization of dental hard tissues
(i.e., enamel, dentin, and cementum) by acid pro-
duced from bacterial fermentation of carbohydrates
(Hillson 2002; Larsen 1995). AMTL has numer-
ous etiologies and can result from extreme attrition,
caries, trauma, or periodontal disease (Hillson
2002). Although the specific link between caries
prevalence and tooth loss in archaeological popu-
lations is not well understood, it is important to deal
with both phenomena together because they are
minimally indicative of a certain level of dental
stress in a population (Hillson 2001). It is also
important to adjust caries rates for teeth lost dur-
ing life— although there is no agreement on the
method (Brothwell 1989; Erdal and Duyar 1999;
Hillson 2001; Lukacs 1992, 1995).
Dental data were recorded on every individual
in the skeletal sample with at least one dental seg-
ment present. These data included the number and
type of observed dental segments per individual
(tooth and/or alveolar portion), condition of each
dental segment (present or lost, ante- or post-
mortem), number of caries per tooth, and nature and
location of the carious lesion following Moore and
Corbett (1971). Carious lesions were recorded
under strong light and only where decay of the
tooth surface(s) or root(s) was observed, and, when
necessary, a magnifying glass and dental probe
were utilized to aid in identification. A carious
lesion was recorded only when demineralization
had formed a distinct cavity in the tooth. The preser-
vation of the skeletal remains in these samples is
excellent and provides accurate identification of
carious lesions. Antemortem tooth loss was
recorded only when partial or complete alveolar
resorption was observed. Problems can arise with
this method in assessing AMTL because teeth lost
near or at the time of death will be recorded as post-
mortem loss, but Lukacs (1995) suggests that prob-
lems in the interpretation of AMTL due to this type
$*)('%*$&# !
Table 2. Composition of the Study Sample from the Lower Azapa Valley.
Site Location Phase (Culture) Individuals (n) Dental Segments (n) M/F Ratio Mean Age
Cam-15 Coast Early (Faldas de Morro) 11 250 1.8 34.50
Quiani-7 Coast Early (Faldas de Morro) 4 110 3.0 40.00
Plm-7 Coast Late (El Laucho) 80 1,551 .8 35.75
Az-14 Valley Late (Alto Ramírez) 30 782 2.0 31.00
Az-70 Valley Late (Alto Ramírez) 25 590 1.5 31.20
Az-71 Valley Early (Azapa) 5 105 4.0 26.00
Az-71 Valley Late (Alto Ramírez) 6 124 1.0 33.30
Az-75 Valley Late (Alto Ramírez) 27 712 .1 28.50
Az-115 Valley Late (Alto Ramírez) 12 358 1.4 37.50
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 429
of error are likely negligible. In addition, we
recorded teeth with noncarious pulp exposure when
the pulp cavity was open (no secondary dentin pre-
sent) but lacked evidence of surrounding carious
decay (Moore and Corbett 1971). Noncarious pulp
exposure most commonly results from extreme
wear of the occlusal surface or damage (trauma) to
the tooth crown, can also contribute to tooth loss
(Hillson 2002), and is therefore important to con-
sider when calculating caries frequencies (Lukacs
1996).
We employ standardized frequency calculations
for caries and AMTL to assess the sample between
variables (Buikstra and Ubelaker 1994; Lukacs
1989, 1995). Table 3 displays caries and AMTL fre-
quencies for the total sample and by site location
and archaeological phase. The prevalenceprovides
information about the frequency of individuals
within the sample with either pathology and is cal-
culated by dividing the number of individuals
affected with at least one pathological expression
by the total number of individuals for each group-
ing. The rate describes the observed percentage of
teeth or alveolar segments affected and provides
information on the degree to which dental segments
are pathological in each group. This calculation is
free of bias from missing elements and is calcu-
lated by dividing the number of pathological teeth
or alveolar segments by the total number of loca-
tions observed. The observed rate was corrected
using Lukacs’s (1995) “Corrected Caries Rate” to
account for tooth loss.
In order to facilitate the comparison of pathol-
ogy rates between variables and additionally con-
trol for age in the analysis, we followed the example
of Pechenkina et al. (2002) and used general lin-
ear modeling analysis (Gill 2001) with age as the
covariate and location and phase as grouping fac-
tors. General linear modeling analysis can be
employed to evaluate differences among samples
with dissimilar age structures (Pechenkina et al.
2002). In addition to comparisons of pathology fre-
quencies, we compared caries and tooth loss by
individual dental segments and anterior vs. poste-
rior dentition to identify if particular teeth or groups
of teeth are more affected than others. We also com-
pared sites of caries initiation on the tooth surface
to identify if differences exist in the types of cari-
ous lesions within the sample. Chi- square and
independent- samples t- tests were used to test caries
and tooth loss by segment, tooth groups, and initi-
ation site between site locations and archaeologi-
cal phases. All statistical tests were performed using
the SPSS 14.0 for Windows program (Statistical
Package for Social Sciences, Chicago).
Results
The frequencies of caries (11.9 percent) and ante-
mortem tooth loss (11.6 percent) calculated from
Formative period samples from the Lower Azapa
Valley are somewhat elevated compared to other
prehistoric groups practicing a mixed subsistence
strategy (Turner 1979). However, when each
marker is considered by location and archaeolog-
ical phase the frequencies vary considerably (Table
3, Figure 2a–b). Significant covariance was
observed with age for both caries and tooth loss
(Table 4). Significant differences were found
between sites located on the coast and those located
in the valley interior (p< .05). Caries were rare
among coastal residents compared to their valley
counterparts, who display an observed rate almost
five times greater (Tables 3–4, Figure 2a). The rate
of tooth loss was also greater among valley resi-
dents (Tables 3–4, Figure 2b).
Although there appears to be a significant
increase in caries from the early to late phase of the
Formative period in the Lower Azapa Valley (Table
3), the burden of oral disease observed among indi-
viduals is extremely variable, and therefore the dif-
ference is nonsignificant (Table 4). The rate of
antemortem tooth loss also remained constant
between the early and late phases. It is possible that
the limited sample size from valley sites dating to
the early phase (n= 5) may affect these results. Nei-
ther caries nor tooth loss demonstrated a signifi-
cant location–phase interaction, which indicates
that these patterns preserve when considering the
results by location between phases or vice versa
(Table 4).
Overall, the number of caries varied signifi-
cantly by individual tooth and between anterior and
posterior dentitions (Figure 3a). The high number
of anterior teeth lost postmortem, particularly
incisors, certainly plays a major role in the distri-
bution of caries between front and back teeth. The
frequencies, however, which are free of bias from
missing data, identify a significantly higher pro-
portion of carious teeth in the posterior (14.7 per-
"(&(
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 430
cent) vs. the anterior (6.9 percent) dentition (t=
–2.26; df = 30; p= .03). This pattern is similar
between site locations and archaeological phases.
Evidence for tooth loss during life appears to be
more evenly distributed across the dentition (Fig-
ure 3b). Overall, slightly more posterior teeth (18.5
percent) were lost on average compared to anterior
teeth (16 percent), but these difference are not sta-
tistically significant (t= –.99; df = 30; p= .33). Indi-
viduals from the coast lost more anterior teeth (16.8
percent) compared to posterior teeth (14.8 percent),
but the difference is also not significant (t = .44; df
= 30; p= .66). Individuals from valley sites, how-
ever, demonstrate a significantly higher proportion
of posterior tooth loss (20.9 percent) on average
compared to anterior teeth (15.8 percent [t= –2.55;
df = 30; p= .02]). No statistically significant dif-
ferences are observed in proportions of anterior vs.
posterior tooth loss between archaeological phases
in the sample.
Occlusal surface (38.3 percent) and interproxi-
mal (24.5 percent) caries constitute the majority of
the carious lesions on teeth in the Lower Azapa Val-
ley. Large caries (14.9 percent)—lesions for which
no site of initiation could be identified— are the
third most frequent type of lesion observed in the
sample, followed by noncarious pulp exposure
(12.4 percent). Smaller proportions of cervical (6.4
percent), smooth surface (2.8 percent), and root (.7
percent) caries are also present in the sample.
Since caries frequencies were found to differ
significantly between coastal and valley sites, we
additionally compared surfaces of caries initiation
by site location to gain additional information on
the dietary effects of caries etiology in the sample
(see Figure 4). The overwhelming majority of
lesions recorded on teeth from coastal sites com-
prise nearly equal proportions of occlusal surface
caries (44.1 percent) and noncarious pulp expo-
sure (35.6 percent). This pattern stands in stark
contrast to lesions recorded on teeth from valley
sites, which include large proportions of occlusal
surface (36.8 percent), interproximal (27.4 per-
cent), and large (17 percent) caries. Noncarious
pulp exposure was observed on a very small per-
centage of teeth (6.3 percent) from sites in the val-
ley. As such, there are significant differences not
only in the frequencies of caries between coast
and valley sites but additionally in the surface of
origin of these lesions.
$*)('%*$&# !
Figure 2. (a) Error bar graph of caries rates and 95 percent confidence intervals by site location; (b) error bar graph of
antemortem tooth loss (AMTL) rates and 95 percent confidence intervals by site location.
ab
Table 3. Frequencies of Oral Health Indicators from the Lower Azapa Valley.
Caries Prevalence AMTL Prevalence Caries Rate AMTL Rate
Total Sample 49.74 50.75 11.90 11.59
Coast 21.28 41.49 3.75 9.56
Valley 75.76 58.65 19.65 13.41
Early Phase 31.58 42.11 5.44 10.56
Late Phase 51.72 51.11 12.61 11.70
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 431
Discussion
The association between subsistence practices and
oral pathology has been studied worldwide under
vastly different environmental circumstances, and
assessments of oral health indicators provide use-
ful inferences about the dietary habits of prehistoric
populations (Larsen 1981; Molnar and Molnar
1985; Moore and Corbett 1971; Tayles et al. 2000;
Turner 1979; Walker and Erlandson 1986). The tran-
sition from food foraging to an agricultural lifeway
has traditionally been blamed for an incremental rise
in oral pathology among human groups as a result
of increased consumption of processed,
carbohydrate- rich agricultural products (Cohen and
Armelagos 1984; Larsen 2002). Rates of dental dis-
ease reflecting overall differences in diet are there-
fore generally lower among food foragers than
among agriculturalists. Turner (1979), e.g., exam-
ined the frequency of carious teeth in samples from
around the world and found that foraging groups
range between .0 and 5.3 percent, mixed subsistence
groups range between .4 and 10.3 percent, and agri-
culturalists range between 2.3 and 29 percent.
Overall, the frequency of caries for the Forma-
tive period skeletal sample from the Lower Azapa
Valley (Table 3) is within the middle range of
agriculture- based economies as defined by Turner’s
(1979) study. Frequencies are generally compara-
ble for antemortem tooth loss as well. Fully
agriculture- dependent prehistoric populations from
the region exhibit higher caries frequencies, rang-
ing between 14 and 48 percent, and more tooth loss,
ranging between 16 and 51 percent (Kelley et al.
1991). Yet the archaeological evidence from this
time period clearly identifies a mixed subsistence
economy. Both coastal and valley residents con-
tinued to exploit marine and terrestrial resources
heavily as well as invest in the variety of domesti-
cated plants and animals that had been introduced
into the region (Muñoz 2004).
There is also evidence to suggest that the inland
groups continued to exploit marine resources
directly. In addition to faunal remains and burial
associations observed at some valley sites, Standen
and colleagues (1997) identified a significant
increase in the presence and size of external audi-
tory exostoses (EAEs) from earlier Chinchorro
"(&(
Figure 3. (a) Distribution of carious and noncarious teeth by dental segment; (b) distribution of alveolar segments lack-
ing evidence of resorption (present) and alveolar segments exhibiting evidence of resorption (antemortem tooth loss
[AMTL]) by dental segment.
ab
Table 4. Results for Two-way Analysis of Variance in General Linear Model,
Age as a Covariate and Location and Phase as Grouping Factors.
Location Phase Location × Phase Age (covariate)
Dependent Variables F(df = 1) pF(df = 1) pF(df = 1) pF(df = 1) p
Caries 15.49 0.01* 0.02 0.89 0.19 0.67 6.34 0.01*
Tooth loss 4.03 0.05* 0.09 0.76 0.08 0.77 100.05 0.01*
*Significant at < .05 level.
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 432
fisher- foragers to the subsequent Formative period
incipient agriculturalists in the same samples exam-
ined here. This pathology can result from chronic
ear infections and is largely associated with marine
foraging and subsistence- related diving in cold
water (Okumura et al. 2007). Approximately 50
percent of the individuals observed from
Camarones-15, Quiani-7, and Playa Miller-7 and
20 percent of the individuals from Azapa-14 and
Azapa-70 exhibited EAE (Standen et al. 1997).
Standen and her colleagues (1997) suggest that the
Formative period groups in the Lower Azapa Val-
ley increased investment in marine resources as a
mechanism to supplement their agricultural sub-
sistence base.
Fewer caries and less tooth loss suggest that for-
aging likely continued to be more important than
agropastoral investment among coastal residents
(Table 4, Figure 2a–b). These data complement the
conclusions of Standen and her colleagues (1997)
for the coastal sites but appear to be contradictory
for the valley sites. The higher frequency of caries
and tooth loss suggests a heavy investment in agri-
cultural production in particular among valley res-
idents. The oral health data identify a clear
separation of the Formative period sample by local-
ity. Given the high incidence of EAE at inland For-
mative sites, it is plausible that either some coastal
residents were interred in valley cemeteries or some
valley residents were dedicated to harvesting
marine resources for the community. Given that
almost one- quarter of the valley residents exhibit
diving- related pathology, it appears more likely
that this is evidence for subsistence specialization
within valley communities.
Proximity to the coast not only was an advan-
tage but also likely constrained the Formative
period residents from moving farther into the val-
ley until they had fully mastered agropastoralism.
Llagostera (1989) envisions a similar progressive
process for the Archaic period Chinchorro. He sug-
gests that the ability to forage in this area was the
result of sequential successes at mastering the mul-
tiple dimensions of accessing the variety of
resources available. The same was likely true as
cultigens, domesticated animals, and new tech-
nologies spread down the valleys and along the
Andean coast. Coastal residents adopted and slowly
began to master these different dimensions within
their existing worldview, altering it as they mas-
tered each new element. Evidence for this process
comes from the presence of early cultigens at late
Archaic sites along the coast (Focacci 1974, 1980)
and the continued reliance on marine resources at
Formative period sites in the valley (Muñoz 2004).
At the very least, marine resources would have pro-
vided an ideal buffer against the uncertainties of
mastering agricultural production. Given the cli-
matic instability inherent in the eastern Pacific, it
is also plausible that the reverse was true as well,
whereby agriculture provided a buffer for fluctua-
tions in marine resources (Arriaza et al. 2001).
The fact that frequencies of oral health indica-
tors do not differ between archaeological phases in
the sample (Table 4) could imply that location-
specific investment adaptations began as early as
$*)('%*$&# !
Figure 4. Pie charts illustrating the distribution of caries by surface of initiation between coast and valley locations.
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 433
the arrival of cultigens in the area and remained sta-
ble for the duration of the Formative period in the
Lower Azapa Valley. The data indicate that although
a mixed subsistence economy was employed at
both locations, each group placed greater invest-
ment in those resources that were immediately
available. Coastal residents continued and perhaps
expanded their exploitation of marine resources,
while valley residents expanded their investment
in cultivation and pastoralism along the San José
River. These represent the beginnings of “imbed-
ded” strategies that continued until the arrival of
the Spanish in the region. Additionally, if our appli-
cation of Llagostera’s model to agropastoral adap-
tation in the Lower Azapa Valley is an accurate
description of the cultural processes involved, then
the earliest groups to move farther up the river had
already mastered some portion of agricultural pro-
duction.
Higher rates of caries and tooth loss observed
among inland inhabitants are not likely due to
increased access to cariogenic wild plants. There
are very few wild plant species in the Lower Azapa
Valley that are as equally cariogenic as the intro-
duced cultigens such as maize, potatoes, manioc,
and beans (see Table 1). Most are weedy or seedy
plants with low sugar content. These same plants
were recovered in similar contexts from both
coastal and valley sites, therefore reinforcing the
fact that they were equally available to constitute
part of the diet at both locations. With evidence for
use of domesticated cultigens, there is no reason to
think that coastal residents experienced more tooth
decay compared to their Archaic predecessors yet
less than contemporaneous valley residents due to
an increase in the exploitation of only wild
resources. Muñoz (1989, 2004) recognizes the use
of both wild cane and maize stalks in the con-
struction of layers in several Formative period bur-
ial mounds in the valley interior, suggesting their
relatively equal ubiquity.
The loss of more posterior teeth among valley
residents is additionally indicative of an increase
in the consumption of carbohydrates and a softer,
more processed diet associated with agriculture.
Some researchers would be inclined to attribute
this differential pattern to differences in tooth wear
between coastal and valley groups. Often foraging
groups exhibit greater anterior wear and lose more
teeth to extreme attrition and exposure of the pulp
cavity (Molnar 1971; Smith 1984). With decreases
in overall rates of occlusal surface wear, foodstuffs
are more likely to stick to grooves and fissures of
the posterior teeth and create ideal circumstances
for caries development (Larsen 2002; Lukacs
1989). These patterns of tooth loss mirror those
observed by Kelley et al. (1991) in earlier Chin-
chorro skeletal samples. However, tooth loss among
the samples here is generally similar, and valley res-
idents lost the same frequency of anterior teeth as
their coastal counterparts. In a related study, we also
failed to find any significant differences in overall
attrition rates between coastal and valley sites (Wat-
son et al. 2011). The loss of more posterior teeth at
valley sites likely results from greater consumption
of carbohydrates and caries development and tends
to mirror the patterns observed in later fully
agriculture- dependent groups in the area (Kelley et
al. 1991).
Differences observed in the locations of caries
initiation between coastal and valley samples also
reflect diversity in subsistence strategy and dietary
base. Although both groups display similar pro-
portions of occlusal surface caries, the coastal sam-
ple contains an almost equal proportion of teeth
exhibiting noncarious pulp exposure, which likely
results from wear and chipping of the crown (Wat-
son et al. 2011). The remaining majority of caries
in the valley sample reflect initiation sites that result
from foodstuffs attaching to or becoming lodged
along the remaining surfaces of the tooth, a pattern
more common among agricultural groups (Hillson
2002).
The oral health data presented here demonstrate
that although Formative period groups residing in
the Lower Azapa Valley practiced a mixed subsis-
tence strategy, they also maintained clear internal
differences in dietary base between coastal sites and
valley locations. Caries and tooth loss among For-
mative period coastal groups are higher than those
observed among the Chinchorro (Kelley et al.
1991), their dedicated Archaic foraging predeces-
sors. The presence of cultigens during the Forma-
tive period functioned to effectively increase the
amount of dietary carbohydrates that coastal resi-
dents consumed and resulted in a corresponding
decrease in oral health. Caries and tooth loss among
the Formative period valley groups are lower than
those observed among the subsequent dedicated
agriculturalists to inhabit the area (Kelley et al.
"(&(
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 434
1991). The continued exploitation and consump-
tion of both wild plants and marine resources dur-
ing this time functioned to limit the degree of
dependence on domesticated cultigens, which
resulted in effectively suppressing the level of oral
disease that is observed in later groups.
Kelley et al. (1991) focus on documenting a lon-
gitudinal trend in agricultural dependence based on
similar oral health indicators from the Archaic
onward. However, they fail to consider the role of
local environment and social dynamics that may
have led to greater investment in agriculture among
valley residents. Our study demonstrates that
related, closely located, contemporaneous groups
actively made very different subsistence choices
during the critical period when cultigens and agri-
culture were introduced into the area. These
processes were not the same across all of the Ata-
cama Desert along the coast.
Population Expansion and Adaptation
We attribute the patterns in oral health during the
Formative period in the Lower Azapa Valley to dif-
ferential investment in niche- specific resources.
These adaptations may also reflect a larger biocul-
tural process of human expansion and adaptation
in the area with the introduction of agriculture. It
is possible that the arrival of domesticated cultigens
along the Chilean coast upset the adaptive balance
that the Chinchorro had achieved and which
allowed them to maintain stable populations for
over 6,000 years. Llagostera’s (1989) vision of
Chinchorro foraging recognizes “process” as a cru-
cial element to their adaptation to the Atacama
coast. Most anthropologists identify change as a
constant in culture; therefore the introduction of
cultigens and perhaps some level of agricultural
technology would have provided an obvious cata-
lyst to facilitate further change by promoting the
exploration and exploitation of different niches
available in the Lower Azapa Valley. This process
resulted in the movement of groups or segments of
the population deeper into the valley and created a
closer relationship with the niche being exploited.
It is also important to consider that these choices
simultaneously resulted in niche construction.
The process of population expansion and niche
construction in the Lower Azapa Valley can be
explained using Resilience Theory (Redman 1992,
1999, 2005; Redman and Kinzig 2003; Redman et
al. 2004). Redman (2005) highlights a four- stage
cycle that groups go through to adapt to major
change: (1) reorganization, renewal, and innova-
tion; (2) growth and exploitation; (3) conservation
and resource “ locking- up”; and (4) release and
rapid change. The spread of cultigens and agricul-
tural technology in the area was part of a continent-
wide phenomenon that carried with it the elements
of major disruptive change, constituting the arrival
of innovation (stage 1). Populations utilizing this
new technology would have experienced a disrup-
tion in traditional subsistence patterns to include
new varieties of plant and animal resources. Seg-
ments of the coastal group moved farther up the
valley— possibly the result of population growth—
to begin better exploitation of the new resources
and niche construction (stage 2) and eventually
would have developed a particular “conservation”
for the application of the new technology (stage 3).
Subsequent changes associated with agriculture,
Formative period cultural developments, and diver-
gence from coastal residents can be attributed to
“release” (stage 4).
This model has the advantage of contextualiz-
ing the arrival of agriculture within the existing
marine- based adaptation along the coast and the
subsequent expansion up the valley. Resilience
Theory assumes that systems function as an adap-
tive cycle, fluctuating between stability and trans-
formation (Redman 2005). Stable foraging
adaptations along the coast were disrupted by the
arrival of cultigens. Subsequently, these groups
grew, diversified, and moved into the valley to
exploit new niches. After this transformation
process, the Lower Azapa Valley experienced
another period of stability lasting approximately
2,000 years until the arrival of influence from
Tiwanaku (Muñoz 2004).
One of the key elements to the application of
Resilience Theory is population growth (Redman
2005). Population growth is largely considered one
of the primary results of the introduction of agri-
culture in a wide variety of societies across the
planet (Bocquet- Appel 2008; Childe 1925). In con-
sidering population growth commonly associated
with the Neolithic transition, Cohen (2008) con-
tends that the moderate fertility/moderate mortal-
ity pattern observed among stable archaic foragers
(see discussion in Bocquet- Appel 2008) is remi-
$*)('%*$&# !
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 435
niscent of K- selected strategies as described in pop-
ulation biology literature. He additionally contends
that the high fertility/high mortality pattern
observed among subsequent expanding agricul-
tural groups is reminiscent of r- selected strategies.
This model fits the circumstances identified for the
Chinchorro, which sustained stable populations in
a crowded niche along a narrow coastline where
equilibrium was maintained for several millennia.
With the introduction of cultigens and agriculture,
the population expanded farther away from the
coast into open niches (primary colonizers) and
began to rely on the exploitation of a wider variety
of products.
Both archaeological data and skeletal remains
attest to continued affinity and interaction among
coastal and valley interior residents during the For-
mative period. Yet our results indicate that decisions
regarding resource investment and food choices
differed within this biocultural continuum. The dif-
ferences observed in oral health appear intuitive,
but the archaeological evidence indicates a similar
mixed subsistence economy at both locations. What
instead might be suggested is that the importance
of domesticated resources is not underscored by
their presence in burial contexts during the For-
mative period.
Conclusions
Evidence from archaeological and skeletal remains
identify that Formative period groups in the Lower
Azapa Valley practiced a shared mixed subsistence
economy and were likely related culturally and bio-
logically. Despite apparent connections, patterns in
oral health demonstrate that the dietary base var-
ied appreciably between coastal and valley resi-
dents. Coastal groups exhibit fewer caries and less
tooth loss compared to valley groups. Sites of caries
initiation are largely confined to occlusal surfaces
in coastal groups but vary across the whole tooth
in valley groups. Floral and faunal data recovered
from these sites identify that each group was con-
suming the same mix of domesticated and wild
resources, yet valley residents suffered from more
dental decay and tooth loss. Although severe wear
can contribute to tooth loss, more lost teeth among
valley residents mirrors the trend in caries and indi-
cates that both are likely the result of greater con-
sumption of dietary carbohydrates. These patterns
suggest that coastal groups were more invested in
marine foraging and valley groups were more
invested in agropastoral production.
Negative impacts on oral health that resulted
from the foraging- to- farming transition in the
Lower Azapa Valley were likely buffered, in par-
ticular, by the rich marine resources at their dis-
posal. In this case, after several millennia of stable
and socially complex fishing- foraging, people
began to invest in agriculture, possibly as a means
of increasing dietary variety, stability, and produc-
tivity. Some of these groups moved farther inland
and began to focus on the exploitation of ecologi-
cal niches better suited for agropastoral production.
These local investments or “imbedded” strategies
mark the beginning of the Formative period in the
area and continued relatively unchanged for almost
2,000 years until the arrival of influence from
Tiwanaku and a complete shift to dependence on
agropastoralism in the Azapa Valley.
Acknowledgments. Study of the collections from Azapa was
conducted under the auspices of the project “Túmulos fune -
rarios: Monumentos del paisaje desértico indicadores de la
organización territorial de las poblaciones formativas de los
valles occidentales del norte de Chile y sur del Perú”
(Fondecyt 1085106; Universidad de Tarapacá, Arica). The
work was supported by the William J. Fulbright Visiting
Scholar Grant, the Commission for the International
Exchange of Scholars, the Fulbright Commission Chile, the
Universidad Tarapacá de Arica, and the Museo Arqueológico
San Miguel de Azapa. Many thanks go to Raúl Rocha for
providing Figure 1 and to John McClelland, the journal edi-
tors, and the anonymous reviewers for providing helpful
comments on previous versions of this manuscript.
References Cited
Arriaza, Bernardo T.
1994 Tipología de las momias chinchorros: Evolución de
las prácticas de momificación. Chungará 26:11–24.
1995 Chinchorro Bioarchaeology: Chronology and
Mummy Seriation. Latin American Antiquity 6:35–55.
Arriaza, Bernardo T., and Vivian Standen
2008 Bioarqeuología. Historia biocultural de los antiguos
pobladores del extremo norte de Chile. Editorial Univer-
sitaria, Santiago.
Arriaza, Bernardo T., Vivian G. Standen, Eliana Belmonte, Fred
Neils, and Eugenia Rosello
2001 The Peopling of the Arica Coast During the Prece-
ramic: A Preliminary View. Chungará 33:31–36.
Aufderheide, Arthur C., Saras Aturaliya, and Guillermo Focacci
2002 Pulmonary Disease in a Sample of Mummies from
the Az-75 Cemetery in Northern Chile’s Azapa Valley.
Chungará 34(2):253–263.
Aufderheide, Arthur C., Iván Muñoz, and Bernardo T. Arriaza
1993 Seven Chinchorro Mummies and the Prehistory of
"(&(
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 436
Northern Chile. American Journal of Physical Anthropol-
ogy 91:189–201.
Belmonte, Eliana
1998 Estudio comparativo de especies vegetales de contexto
arqueológico (Museo Arqueológico San Miguel de Azapa).
Informe Final de Investigación Proyecto No. 3740-96.
Universidad de Tarapacá, Arica.
Belmonte, Eliana, Eugenia Rossello, and Nelly Rojas
1988 Análisis de restos vegetales contenidos en coprolitos
de camélidos (desembocadura del Río Camarones). Chun-
gará 20:47–61.
Bocquet- Appel, Jean- Pierre
2008 Explaining the Neolithic Demographic Transition. In
The Neolithic Demographic Transition and Its Conse-
quences, edited by Jean- Pierre Bocquet- Appel and Ofer
Bar- Yosef, pp. 35–56. Springer, New York.
Bridges, Patricia S.
1989 Changes in Activities with the Shift to Agriculture in
the Southeastern United States. Current Anthropology
30:385–394.
Brothwell, Donald R.
1989 The Relationship of Tooth Wear to Aging. In Age
Markers in the Human Skeleton, edited by M. Yasar Iscan,
pp. 303–318. Charles C. Thomas, Springfield, Illinois.
Buikstra, Jane E., and Douglas H. Ubelaker
1994 Standards for Data Collection from Human Skeletal
Remains. Arkansas Archaeological Survey Research Series
No. 44. Fayetteville.
Cassman, Vicki
1997 A Reconsideration of Prehistoric Ethnicity and Sta-
tus in Northern Chile: The Textile Evidence. Unpublished
Ph.D. dissertation, Department of Anthropology, Arizona
State University, Tempe.
Childe, V. Gordon
1925 The Dawn of European Civilization. Kegan Paul, Lon-
don.
Cohen, Mark N.
2008 Implications of the NDT for World Wide Health and
Mortality in Prehistory. In The Neolithic Demographic
Transition and Its Consequences, edited by Jean- Pierre
Bocquet- Appel and Ofer Bar- Yosef, pp. 481–500.
Springer, New York.
Cohen, Mark N., and George J. Armelagos (editors)
1984 Paleopathology at the Origins of Agriculture. Acad-
emic Press, Orlando.
Cohen, Mark Nathan, and Gillian M. Crane- Kramer
2007 Ancient Health: Skeletal Indicators of Agricultural
and Economic Intensification. University Press of Florida,
Gainesville.
Erdal, Yilmaz S., and Izzet Duyar
1999 Brief Communication: A New Correction Procedure
for Calibrating Dental Caries Frequency. American Jour-
nal of Physical Anthropology 108:237–240.
Erices, Sergio
1975 Evidencias de vegetales en tres cementerios prehis-
pánicos, Arica Chile. Chungará 5:65–71.
Focacci, Guillermo
1974 Excavaciones en Playa Miller 7. Chungará 3:23–74.
1980 Síntesis de la arqueología del extremo norte de Chile.
Chungará 6:3–23.
Focacci, Guillermo, and Sergio Erices
1972–1973 Excavaciones en túmulos de San Miguel de
Azapa (Arica- Chile). Actas del VI Congreso de Arque-
ología Chilena, pp. 47–62. Universidad de Chile, Santiago.
Gill, Jeff
2001 General Linear Models: A Unified Approach. Sage
Publications, Thousand Oaks, California.
Goodman, Alan H., Debra L. Martin, and George J. Armelagos
1984 Indications of Stress from Bone and Teeth. In Pale-
opathology at the Origins of Agriculture, edited by Mark
N. Cohen and George J. Armelagos, pp. 13–49. Academic
Press, Orlando.
Hartnady, Phillip, and Jerome C. Rose
1991 Abnormal Tooth- Loss Patterns Among Archaic-
Period Inhabitants of the Lower Pecos Region, Texas. In
Advances in Dental Anthropology, edited by Marc A. Kel-
ley and Clark Spenser Larsen, pp. 267–278. Wiley- Liss,
New York.
Hillson, Simon
2001 Recording Dental Caries in Archaeological Human
Remains. International Journal of Osteoarchaeology
11(4):249–289.
2002 Dental Anthropology. 3rd ed. Cambridge University
Press, Cambridge.
Kelley, Marc A., Dianne R. Levesque, and Eric Weidl
1991 Contrasting Patterns of Dental Disease in Five Early
Northern Chilean Groups. In Advances in Dental Anthro-
pology, edited by Marc A. Kelley and Clark Spenser
Larsen, pp. 203–213. Wiley- Liss, New York.
Kolata, Alan
1993 The Tiwanaku: Portrait of an Andean Civilization.
Blackwell, Cambridge.
Larsen, Clark Spenser
1981 Skeletal and Dental Adaptations to the Shift to Agri-
culture on the Georgia Coast Population. Current Anthro-
pology 22:422–423.
1995 Biological Changes in Human Populations with Agri-
culture. Annual Review of Anthropology 24:185–213.
1997 Bioarchaeology: Interpreting Behavior from the
Human Skeleton. Cambridge Studies in Biological Anthro-
pology. Cambridge University Press, Cambridge.
2000 Skeletons in Our Closet: Revealing Our Past Through
Bioarchaeology. Princeton University Press, Princeton.
2002 Post- Pleistocene Human Evolution: Bioarchaeology
of the Agricultural Transition. In Human Diet: Its Origin
and Evolution, edited by Peter S. Ungar and Mark F.
Teaford, pp. 19–36. Bergin and Garvey, Westport, Con-
necticut.
Llagostera, Agustín
1989 Caza y pesca marítima (9.000 a 1.000 a.C.). In Cul-
turas de Chile. Prehistoria. Desde sus orígenes hasta los
albores de la Conquista, edited by Jorge Hidalgo, Vigilio
Schiappacasse, Hans Niemeyer, Carlos Aldunate, and Iván
Solimano, pp. 57–79. Editorial Andrés Bello, Santiago.
Lukacs, John R.
1989 Dental Paleopathology: Methods for Reconstructing
Dietary Patterns. In Reconstruction of Life from the Skele-
ton, edited by M. Yasar Iscan and Kenneth A. R. Kennedy,
pp. 261–286. Wiley- Liss, New York.
1992 Dental Paleopathology and Agricultural Intensifica-
tion in South Asia: New Evidence from Bronze Age
Harappa. American Journal of Physical Anthropology
87:133–150.
1995 The “Caries Correction Factor”: A New Method of
Calibrating Dental Caries Rates to Compensate for Ante-
mortem Loss of Teeth. International Journal of Osteoar-
chaeology 5:151–156.
1996 Sex Differences in Dental Caries Rates with the Ori-
gin of Agriculture in South Asia. Current Anthropology
37:147–153.
Molnar, Stephen
1971 Human Tooth Wear, Tooth Function, and Cultural
Variability. American Journal of Physical Anthropology
34:175–190.
$*)('%*$&# !
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 437
Molnar, Stephen, and Iva Molnar
1985 Observations of Dental Diseases Among Prehistoric
Populations of Hungary. American Journal of Physical
Anthropology 67:51–63.
Moore, William J., and M. Elisabeth Corbett
1971 The Distribution of Dental Caries in Ancient British
Populations 1: Anglo- Saxon Period. Caries Research
5:151–168.
Munizaga, Juan R.
1964 Comparación de poblaciones del norte de Chile.
Revista del Centro de Estudios Antropológicos 2:87–95.
1974 Paleopatologıa Chilena. Nueva Época 1. Universidad
de Chile, Santiago.
1980 Esquema de la Antropología física chilena. Chungará
6:124–136.
Muñoz, Iván
1981 Investigaciones arqueológicas en los túmulos fune -
rarios del valle de Azapa (Arica). Chungará 6:57–95.
1985 Tempranos Cultivos de Plantas en Poblaciones Pre-
hispánicas del Valle de Azapa (Arica- Chile). Idesia 9.
1989 El Período Formativo en el Norte Grande (100 a.C. a
500 d. C.). In Culturas de Chile. Prehistoria. Desde sus
orígenes hasta los albores de la Conquista, edited by Jorge
Hidalgo, Vigilio Schiappacasse, Hans Niemeyer, Carlos
Aldunate, and Iván Solimano, pp. 107–128. Editorial
Andrés Bello, Santiago.
1993 Spatial Dimensions of Complementary Resource Uti-
lization at Acha-2 and San Lorenzo. In Domestic Archi-
tecture, Ethnicity, and Complementarity in the
South- Central Andes, edited by Mark S. Aldenderfer, pp.
94–102. University of Iowa Press, Iowa City.
1995–1996 Poblamiento humano y relaciones intercultu -
rales en el Valle de Azapa: Nuevos hallazgos en torno al
período Formativo y Tiwanaku. Diálogo Andino
14–15:241–278.
2004 El periodo Formativo en los valles del norte de Chile
y sur de Perú: nuevas evidencias y comentarios. Chungará,
Volumen Especial, 2004:213–225.
Muñoz, Iván, and Guillermo Focacci
1985 San Lorenzo: testimonio de una comunidad de agricul-
tores y pescadores postiwanaku en el Valle de Azapa
(Arica- Chile). Chungará 15:7–30.
Nelson, Ben A., Debra L. Martin, Anne C. Swedlund, Paul Fish,
and George J. Armelagos
1994 Studies in Disruption: Demography and Health in
the Prehistoric American Southwest. In Understanding
Complexity in the Prehistoric Southwest, edited by
George Gumerman and Murray Gell- Mann, pp. 59–112.
Santa Fe Institute Studies in the Sciences of Complex-
ity, Proceedings Vol. 16. Addison- Wesley, Reading,
Massachusetts.
Okumura, Maria Mercedes M., Célia H. C. Boyadjian, and
Sabine Eggers
2007 Auditory Exostoses as an Aquatic Activity Marker: A
Comparison of Coastal and Inland Skeletal Remains from
Tropical and Subtropical Regions of Brazil. American
Journal of Physical Anthropology 132:558–567.
Pechenkina, Ekaterina A., Robert A. Benfer Jr., and Wang Zhi-
jun
2002 Diet and Health Changes at the End of the Chinese
Neolithic: The Yangshao/Longshan Transition in Shaanxi
Province. American Journal of Physical Anthropology
117:15–36.
Rech, Jason A., Jay Quade, and William S. Hart
2003 Isotopic Evidence for the Source of Ca and S in Soil
Gypsum, Anhydrite and Calcite in the Atacama Desert,
Chile. Geochimica et Cosmochimica Acta 67(4):575–586.
Redman, Charles. L.
1992 The Impact of Food Production: Short- Term Strate-
gies and Long- Term Consequences. In Human Impact on
the Environment: Ancient Roots, Current Challenges,
edited by Judith E. Jacobsen and John Firor, pp. 35–49.
Westview Press, Boulder.
1999 Human Impact on Ancient Environments. University
of Arizona Press, Tucson.
2005 Resilience Theory in Archaeology. American Anthro-
pologist 107(1):70–77.
Redman, Charles L., Steven R. James, Paul R. Fish, and J.
Daniel Rogers (editors)
2004 The Archaeology of Global Change: The Impacts of
Humans on Their Environment. Smithsonian Books, Wash-
ington, D.C.
Redman, Charles L., and Ann P. Kinzig
2003 Resilience of Past Landscapes: Resilience Theory,
Society, and the Longue Durée. Conservation Ecology
7(1):14.
Rivera, Mario A.
1977 Análisis experimental de coprolitos, provenientes de
los sitios Az-83 y Az-84, pertenecientes a la fase Alto
Ramírez, período Intermedio Temprano; norte de Chile.
Podium presentation at VII Congreso Nacional de Arque-
ología Chilena Altos de Vilches, Talca.
1984 Altiplano and Tropical Lowland Contacts in North-
ern Chilean Prehistory: Chinchorro and Alto Ramírez
Revisited. In Social and Economic Organization in the Pre-
hispanic Andes, edited by David L. Browman, Richard L.
Burger, and Mario A. Rivera, pp. 143–160. Vol. 194. BAR
International Series, Oxford.
1991 The Prehistory of Northern Chile: A Synthesis. Jour-
nal of World Prehistory 5:1–47.
Romero, Álvaro G., Calogero M. Santoro, Daniela Valenzuela,
Juan Chacama, Eugenia Rosello, and Luigi Piacenza
2004 Túmulos, ideología y paisaje de la fase Alto Ramirez
del Valle de Azapa. Chungará, Volumen Especial,
2004:261–272.
Rothhammer, Francisco, José Cocilovo, Elena Llop, and Silvia
Quevedo
1989 Orígenes y microevolución de la población chilena.
In Culturas de Chile Prehistoria: desde sus orígenes hasta
los albores de la Conquista, edited by Jorge Hidalgo, Vig-
ilio Schiappacasse, Hans Niemeyer, Carlos Aldunate, and
Iván Solimano, pp. 403–413. Andrés Bello, Santiago.
Rothhammer, Francisco, Calogero Santoro, and Mauricio Mo -
raga
2002 Craniofacial Chronological Microdifferentiation of
Human Prehistoric Populations of the Azapa Valley, North-
ern Chile. Revista Chilena de Historia Natural
75:259–264.
Santoro, Calogero M.
1980a Estratigrafía y secuencia cultural funeraria. Fases:
Azapa, Alto Ramírez y Tiwanaku. Chungará 6:24–45.
1980b Fase Azapa. Transición del Arcaico, al desarrollo
agrario inicial en los Valles Bajos de Arica. Chungará
6:46–56.
Smith, B. Holly
1984 Patterns of Molar Wear in Hunter- Gatherers and Agri-
culturalists. American Journal of Physical Anthropology
63:39–56.
Standen, Vivian G., Bernardo T. Arriaza, and Calogero M. San-
toro
1997 External Auditory Exostosis in Prehistoric Chilean
Populations: A Test of the Cold Water Hypothesis. Amer-
ican Journal of Physical Anthropology 103:119–129.
Steckel, Richard H., and Jerome C. Rose (editors)
"(&(
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 438
2002 The Backbone of History: Health and Nutrition in the
Western Hemisphere. Cambridge University Press, Cam-
bridge.
Steckel, Richard H., Jerome C. Rose, Clark Spenser Larsen, and
Philip L. Walker
2002 Skeletal Health in the Western Hemisphere from 4000
B.C. to the Present. Evolutionary Anthropology
11:142–155.
Sutter, Richard C.
1997 Dental Variation and Biocultural Affinities Among
Prehistoric Populations from the Coastal Valleys of
Moquegua, Peru, and Azapa, Chile. Unpublished Ph.D.
dissertation, Department of Anthropology, University of
Missouri, Columbia.
2000 Prehistoric Genetic and Culture Change: A Bioar-
chaeological Search for Pre- Inka Altiplano Colonies in
the Coastal Valleys of Moquegua, Peru, and Azapa, Chile.
Latin American Antiquity 11:43–70.
2006 The Test of Competing Models for the Prehistoric
Peopling of the Azapa Valley, Northern Chile, Using Matrix
Correlations. Chungará 38(1):63–68.
Sutter, Richard C., and Lisa M. Mertz
2004 Non- Metric Cranial Trait Variation and Prehistoric
Biocultural Change in the Azapa Valley, Chile. American
Journal of Physical Anthropology 123:130–145.
Tayles, Nancy, Kate Domett, and K. Nelson
2000 Agriculture and Dental Caries? The Case of Rice in
Prehistoric Southeast Asia. World Archaeology
32(1):68–83.
Turner, Christy G.
1979 Dental Anthropological Indications of Agriculture
Among the Jamon People of Central Japan. American
Journal of Physical Anthropology 51:619–635.
Ubelaker, Douglas H.
1978 Human Skeletal Remains: Excavations, Analysis,
Interpretation. Aldine, Chicago.
Varela, Hector H., and José A. Cocilovo
2002 Genetic Drift and Gene Flow in a Prehistoric Popu-
lation of the Azapa Valley and Coast, Chile. American
Journal of Physical Anthropology 118:259–267.
Walker, Philip L., and Jon M. Erlandson
1986 Dental Evidence for Prehistoric Dietary Change on
the Northern Channel Islands. American Antiquity
51:375–383.
Watson, James T.
2008 Prehistoric Dental Disease and the Dietary Shift from
Cactus to Cultigens in Northwest Mexico. International
Journal of Osteoarchaeology 18:202–212.
Watson, James T., Bernardo T. Arriaza, Vivian Standen, and Iván
Muñoz
2011 Occlusal Dental Wear and the Formative Transition
Along the Northern Chilean Coast. International Journal
of Osteoarchaeology, in press.
(Need dates)
$*)('%*$&# !
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 439
Statement of Ownership, Management, and
Circulation. 1. Publication title: Latin A merican
Antiquity, 2. Publication number: 1045-6635. 3.
Filing date: October 1, 2010. 4. Issue Frequency:
Four times yearly in March, June, September,
December. 5. Number of issues published
annually: Four. 6. Annual Subscription Price:
$149.00. 7. Complete Mailing Address of Known
Office of Publication: Society for American
Archaeology, 900 Second St NE, Suite 12,
Washington, DC 20002-3560; contact person:
John Neikirk; telephone: 202/789-8200. 8.
Complete Mailing Address of Headquarters of
General Business Office of Publisher: Society for
American Archaeology, 900 Second St NE, Suite
12, Washington, DC 20002 -3560; Editor:
Christopher A. Pool, Department of
Anthropology, University of Kentucky,
Lexington, KY 40506; Managing Editor: John
Neikirk, Society for American Archaeology, 900
Second St NE, Suite 12, Washington, DC 20002 -
3560. 10. Owner: Society for American
Archaeology, 900 Second St NE, Suite 12,
Washington, DC 20002-3560. 11. Known
bondholders mortgagees, and other security
holders, etc.: None. 12. The purpose, function,
and nonprofit status of this organization
and the exempt status for federal income
purposes has not changed during the preceding
12 months. 13. Publication name: Latin
American Antiquity. 14. Issue date for circulation
data below: September 2010. 15. Extent and
nature of circulation: Average no. copies each
issue during the preceding 12 months: a. Total
number of copies (net press run): 2075; b. Paid
and/or requested circulation: (1) Paid/ requested
outside-county mail subscriptions stated on form
3541: 1862; (2) Paid in-county mail subscriptions
stated on form 3541: None; (3) Sales through
dealers and carriers, street vendors, counter sales,
and other non-USPS paid distribution: None; (4)
Other classes mailed through USPS: None; c.
Total paid and/or requested circulation: 1862; d.
Free or Nominal Rate Distribution: (1) 6, (2)
None, (3) None, (4) None; e. Total Free or
Nominal Rate Distribution: 6; f. Total
distribution: 1868; g. Copies not distributed: 207;
h. Total: 2075; i. Percent paid and/or requested
circulation: 99.7%. Single issue published
nearest to filing date: a. 2050; b. (1) 1937; (2)
None; (3) None; (4) None; c. 1905; d (1). 6; e. 6;
f. 1943; g. 107; h. 2050; i. 99.7%. 16. Publication
of Statement of Ownership: December 2010. 17.
Signed by John Neikirk, Managing Editor,
October 1, 2010.
LAQ21(4) Watson_Layout 1 12/2/10 11:09 AM Page 440
