Proceedings of the International School of Physics “Enrico Fermi”
Course CLIV, M. Martini, M. Milazzo and M. Piacentini (Eds.)
IOS Press, Amsterdam 2004
Stable isotopes and diet: You are what you eat
R. H. Tykot
Department of Anthropology, University of South Florida - Tampa, FL 33620, USA
Introduction
Archaeologists frequently use many different sources of information to reconstruct
ancient diets, including faunal and macrobotanical remains, pollen and phytoliths, as
well as pottery residues, coprolites, and indirect sources such as skeletal pathology, dental
wear patterns, artistic depictions, and ethnographic observations. In most cases, however,
the result is only the determination of the menu, and with the exception of animal foods,
without any quantitative estimate of their importance. Furthermore, these data almost
always pertain to groups of people and represent a time span of multiple generations if
not a century or more.
In the late 1970s, though, a new area of dietary research developed based on the
isotopic composition of human bone. This work began with the observation that radiocarbon
dates on the remains of certain plants such as maize were offset from dates
obtained on other remains from the same archaeological context. With the determination
that this was due to maize having a different photosynthetic pathway than most
plants, which results in a different relative quantity of carbon-14 (and carbon-13) in its
tissues, it was realized that carbon isotope ratios in consumer tissues (e.g. bone) would
also be affected, and therefore the measurement of carbon isotope ratios could be used
to indicate the importance of maize in human diets [1, 2]. Following the observation
that nitrogen isotope ratios also vary between different food sources, especially marine
vs. terrestrial [3], the stable isotope analysis of human bone quickly became a widely
applied technique, especially in areas where isotopically diverse food sources are known
to have existed.
Analyses of human bone offer the specific advantage relative to faunal and floral
Societ`a Italiana di Fisica 433
434 R. H. Tykot
studies that the dietary results obtained are for individuals and not necessarily a larger
group, meaning that comparisons can be made between individuals of different age, sex,
and/or socioeconomic status, as well as between different sites and over multiple time
periods. Reliable results have been obtained from samples millions of years old, and
using modern instruments which require only tiny samples, it is also possible to study
seasonal variation in incremental growth tissues such as hair and teeth, widening even
further the subsistence questions which can be addressed by stable isotope analysis.
Principles of stable isotope analysis
Carbon and nitrogen isotope ratios in human bone may be used to reconstruct prehistoric
diet because of differential fractionation, between certain plant groups, of atmospheric
carbon dioxide during photosynthesis and of nitrogen during fixation or absorption
(fig. 1). There are two stable isotopes each of carbon (12
C, 13
C) and nitrogen (14
N,
15
N), with 12
C and 14
N by far the most common in nature. Small differences in the
ratios of these isotopes (12
C/13
C, 15
N/14
N) can be measured by isotope ratio mass spectrometry
using samples smaller than 1 milligram. High-precision isotope measurements
are reported using the delta notation (δ13
C, δ15
N) relative to internationally recognized
standards and are expressed in parts per thousand or per mil ( ).
Experimental data have indicated that different bone tissues reflect different components
of the diet [4,5]. In general, bone collagen is disproportionately produced from the
protein portion of the diet, while bone carbonate and tooth enamel carbonate (both a
calcium hydroxyphosphate, called apatite) are produced from a mixture of dietary protein,
carbohydrates and fats. Stable isotope analysis of both bone collagen and apatite
thus permits quantitative estimates of several dietary components. Both bone collagen
and bone apatite are constantly being resorbed and replenished, so that their isotopic
composition reflects dietary averages over at least the last several years of an individual’s
life, while the composition of tooth enamel will reflect diet during the age of crown
formation.
Since teeth form at different ages (beginning in utero for deciduous teeth, and from
0 to 12 yrs of age for permanent teeth), the analysis of multiple teeth from the same
individual may reveal dietary shifts caused by weaning (first the introduction of solid
foods and later the cessation of breast feeding) [6], while significant differences between
tooth (juvenile diet) and bone (adult diet) values may be the result of subsistence changes
due to migration. Analysis of hair segments also may reveal short-term or seasonal
dietary changes [7]. Overall, stable isotope analysis of multiple tissues can provide a
quantifiable dietary life history of an individual [8]. While collagen is rarely preserved
in bones predating the Upper Paleolithic, and is often badly degraded in hot and moist
environments (e.g., lowland Mesoamerica), bone apatite has provided reliable results for
the Holocene, and tooth enamel for early hominids [9] and into the Miocene.
Typically, grasses originally native to hot, arid environments follow the C4 (HatchSlack)
photosynthetic pathway, and will have δ13
C values averaging about −12.5 ;
trees, shrubs, and grasses from temperate regions, which follow the C3 (Calvin-Benson)
Stable isotopes and diet 435
Fig. 1. – Carbon isotope fractionation in terrestrial foodwebs.
photosynthetic pathway, will have δ13
C values averaging about −26.5 . In some forested
areas, a canopy effect occurs due to incomplete atmospheric mixing, resulting in even
more negative carbon isotope ratios [10]. Empirical data for large mammals combined
with experimental data for rats and mice indicate that bone collagen is enriched about
+5 relative to diet, although this value is affected by the proportion of protein in the
total diet and differences in δ13
C values between protein and energy sources [11, 3, 4].
Experimental data on rats also convincingly demonstrates that bone apatite is enriched
about +9.5 relative to the whole diet, regardless of the mixture or isotopic composition
of the foods consumed [3], although empirical data for larger herbivores suggest the dietapatite
spacing is about +12 . Stable carbon isotope analysis is particularly useful in
New World dietary studies since maize is often the only C4 plant contributing significantly
to human diets; its contribution to bone collagen and to bone apatite may be estimated by
interpolation. Some caution is warranted, however, if succulent plants were present, since
they utilize the alternative CAM (crassulacean acid metabolism) photosynthetic pathway
which results in carbon isotope ratios similar to those of C4 plants. Nevertheless, CAM
plants are unlikely to have been major sources of dietary protein, whether consumed
436 R. H. Tykot
directly or indirectly through herbivorous faunal intermediaries.
The carbon isotope ratios of marine and freshwater organisms are more variable,
depending on local ecological circumstances, and often overlap with those of terrestrial
plants and their consumers. These foods typically have much higher nitrogen isotope
values, however, and their high protein content will contribute much more carbon to
bone collagen than will maize (about 10% protein) or other plant foods. The analysis
of bone apatite, which is derived from all food groups, should allow the identification of
just a few percent maize or other C4 resources in an otherwise C3-based diet.
The nitrogen isotope ratios for plants depend primarily on how they obtain their
nitrogen —by symbiotic bacterial fixation or directly from soil nitrates— and these values
are similarly passed along through the food chain accompanied by an approximately 2–
3 positive shift for each trophic level, including between mother and nursing infant.
Human consumers of terrestrial plants and animals typically have δ15
N values in bone
collagen of about 6–10 whereas consumers of freshwater or marine fish, seals and sea
lions may have δ15
N values of 15–20 [3]. While most plants follow either the C3 or C4
photosynthetic pathway and have similar carbon isotope ratios in most ecological settings,
nitrogen isotope ratios vary according to rainfall, altitude and other factors [12], and both
carbon and nitrogen isotope ratios vary considerably among marine organisms [13].
It is critical, therefore, to establish a site-relevant isotopic baseline for interpreting
human skeletal data (fig. 2). Analyses of faunal remains provide a good estimate both of
the animals themselves and the plants they consume. Establishing an isotopic baseline
is particularly important in coastal areas where marine and riverine resources, as well as
C4 and/or CAM plants may have been available for direct or indirect consumption by
humans. Atmospheric carbon isotope ratios have become depleted by about 1.5 since
the industrial revolution, so values obtained on modern terrestrial plants and animals
must be adjusted accordingly for most archaeological studies.
Other surveys and summaries of stable isotope analysis in archaeology may be found
in several recent publications [14-17].
Analytical methods
For the isotopic analysis of bone, it is first necessary to separate the specific tissue
to be analyzed. For collagen, this involves demineralization of the bone using acid (with
different laboratories using different acids, concentrations, and temperatures), and separation
from any residual lipids. In my own laboratory at the University of South Florida,
bone collagen is extracted using well-established laboratory procedures [18]. Whole bone
is demineralized in 2% hydrochloric acid (72 h), base-soluble contaminants are removed
using 0.1 M sodium hydroxide (24 h each before and after demineralization), and residual
lipids are dissolved in a 2:1:0.8 mixture of methanol, chloroform, and water (24 h).
One milligram samples of the resulting collagen pseudomorphs —which make for a
very useful visual indicator of sample integrity, which is often a problem for archaeological
samples— are placed in tin capsules and analyzed for δ13
C and δ15
N in continuous flow
mode using a CHN analyzer coupled with a Finnigan MAT stable isotope ratio mass
Stable isotopes and diet 437
Fig. 2. – Carbon vs. nitrogen isotope ratios in common New World food groups.
spectrometer, also located at USF. Previously, it was necessary to convert collagen into
CO2 and N2 offline, through high-temperature combustion in a vacuum-sealed quartz
tube, which was followed by cryogenic distillation to separate these gases. Each gas
sample was then introduced to the mass spectrometer using a manifold system. The
CHN analyzer integrates the combustion step with separation in time of the resulting
CO2 and N2 en route to the mass spectrometer, which is also able to switch quickly
between the different mass ranges being measured (28 and 29 for N2; 44, 45 and 46 for
CO2, with mass 46 mostly a result of a single oxygen-18 and its measurement allowing
for a formulaic determination of how much of the mass 45 signal is from oxygen-17 and
therefore how much from carbon-13). With either method of sample introduction, gas
yields and C:N ratios are used to confirm the integrity of the collagen samples.
Bone apatite and tooth enamel samples are prepared using procedures designed to
remove non-biogenic carbon without altering the biogenic carbon isotope values [19].
Powder samples are obtained by drilling from the center of carefully cleaned bone samples,
or from tooth enamel after the surface layer has been removed. Approximately 10 mg of
powder are immersed in 2% sodium hypochlorite to dissolve organic components (24 h
for enamel, 72 h for bone apatite). Non-biogenic carbonates are then removed in 1.0 M
buffered acetic acid for 24 hours. The integrity of apatite and enamel samples is assessed
through yields obtained in each stage of the pretreatment process. Samples are analyzed
on a second Finnigan MAT mass spectrometer equipped with a Kiel III individual acid
bath carbonate system, which eliminates the need both for off-line production of CO2
by reaction of the sample with acid in a vacuum-sealed glass tube, and the cryogenic
438 R. H. Tykot
purification of the resulting gas sample.
In addition to reference gases against which samples are measured, several solid standard
samples are analyzed at the beginning of each run and then after every six or seven
archaeological samples to ensure reliability of all results. The analytical precision for
stable isotope ratio mass spectrometry is typically 0.1 for δ13
C and 0.2 for δ15
N.
Recent applications
I have been involved in a number of collaborative research projects around the world
using stable isotope or elemental analysis to assess ancient diets. Results have been
presented at various conferences, some have been published, and others are in press or
some stage of preparation for publication. Examples of some of these projects and the
results obtained are given below.
Patagonia (Argentina and Chile). – In southernmost South America, ethnographic
observations of historic groups of hunter-gatherers have revealed several discrete subsistence
adaptations, including great dependence on marine foods for those living on the
coasts and inlets of Tierra del Fuego. Growing archaeological evidence, however, has suggested
at least that this historic pattern does not extend into antiquity. Isotopic analysis
of bone collagen, bone apatite, and tooth enamel from a large number of sites in central
and southern Patagonia, provided through collaborations with Julieta Gomez Otero,
Juan Belardi, Ricardo Guichon and Luis Borrero, indicates that the coastal populations
had a broad range of diets, in many cases depending more on terrestrial than marine
foods (fig. 3). Some coastal groups clearly exploited high-trophic level marine resources
such as sea lions and penguins, while others had more mixed diets which included fish,
mollusks, guanaco, and land birds. The result appears to be that a continuum of subsistence
combinations were employed, rather than the discrete selected diets proposed by
ethnographic observations [20].
Highland Ecuador and Peru. – In highland Ecuador and Peru, collaborations with
Richard Burger and J. Stephen Athens have provided the opportunity to examine the
introduction and increasing importance of maize in the highland Andes. New collagen
carbon and nitrogen isotope data from Pacopampa and La Chimba support earlier results
based on collagen carbon from Huaricoto and Chavin de Huantar [21] which suggest
modest importance of maize during the first millennium BC in the highland Andes.
The apatite data, however, suggest that, while maize was a similar proportion of the
whole diet in both regions during the 1st millennium BC, it was a greater contributor
to dietary protein in highland Ecuador. In later periods maize became a dietary staple,
with differential access (e.g. the consumption of maize beer) according to social status
at sites such as La Florida [22]. Although there appears to be an increase in δ15
N values
over time, which could be interpreted as due to greater consumption of animals (higher
trophic level than plants) or the availability of small quantities of fish by trade with
people living at lower elevations, this is most likely an effect of the differences in altitude
of the sites in our study.
Stable isotopes and diet 439
Fig. 3. – Carbon vs. nitrogen in human collagen from Patagonia, showing variation in importance
of seafood among coastal populations.
Coastal Ecuador and Peru. – The diets associated with complex societies on the coasts
of Ecuador and Peru have been the subject of great debate because of the conceived
importance of agriculture —especially maize— as the basis for sedentary societies and
the emergence of political hierarchies, and the reported presence of maize pollen and
phytoliths in the region by 5000 BC. The precocious Valdivia culture on the coast of
Ecuador has ceramics and ceremonial sites by 3000 BC, while monumental platform
structures appear on the coasts of Peru in the 3rd millennium BC and ceramics shortly
afterwards. Analyses of a time series of individuals from coastal Ecuador [23,24], however,
suggest that maize is not noticeable in diets until the end of the Valdivia period (phases
7–8, about 2000 BC), long after the appearance of settled villages with ceramics and
ceremonial architecture, and only became a dietary staple in the second half of the 1st
millennium BC. In Peru, also, analyses of remains from the Lurin (Mina Perdida, Cardal,
Tablada de Lurin) [25] and Viru [26] valleys suggest that maize was not a dietary staple
until well into the Initial Period. For interpreting coastal diets, where a combination
of riverine, marine, terrestrial C3 and C4 resources could have been consumed, it is
important to obtain stable isotope data from both human bone collagen and apatite, as
well as the available fauna and flora.
Belize and Guatemala. – Skeletal remains from the Preclassic site of Cuello, excavated
by Norman Hammond of Boston University, have been analyzed for bone collagen, bone
apatite, and tooth enamel [27], as have samples from Classic sites at La Milpa (Hammond
440 R. H. Tykot
Fig. 4. – Bone apatite carbon vs. collagen nitrogen for Belize, suggesting a big increase in the
importance of maize to the whole diet by the beginning of the Classic period.
excavations) and the Programme for Belize (Fred Valdez/Richard Adams excavations).
In conjunction with data published by others, a chronological trend towards increasing
maize reliance is evident both in bone collagen and bone apatite [28]. For whole diet as
represented in bone apatite, however, the biggest shift in carbon isotope ratios is observed
between the Preclassic and Early Classic, while only a statistically insignificant change
is then observed through the Late, Terminal, and Postclassic periods (fig. 4). This is in
contrast to the largest shift in carbon isotope ratios in bone collagen, which occurred
between the Terminal and Postclassic periods. I interpret this to indicate that the real
increase in the importance of maize to the whole diet came during the first millennium
BC, when complex Maya society was first established, while changes in its importance
as a protein source may be correlated with the collapse of Classic Maya society.
At Cuello, and probably other sites, some of the maize signature likely came from
the consumption of maize-fed dogs. At La Milpa, diets also seem to vary based on social
status and residence location. At Iximch´e in highland Guatemala, a number of individuals
dating to the Late Postclassic period have been tested for bone collagen (by David Reed
and Stephen Whittington) and tooth enamel. The results reveal an extremely high
dependence on maize, more so than most other Guatemalan populations at least of the
Late Classic period, and certainly more so than contemporary populations at Lamanai in
Belize. This high dependence may be related to a combination of local ecological as well
as socioeconomic factors. Within the group of individuals tested, most of whom were
decapitated and may have formed a skull rack (tzompantli), several individuals stand out
Stable isotopes and diet 441
as having significantly different diets. We suggest that they may have been warrior-elites
from other communities who consumed different diets either because of differences in
local ecology, and/or had access to different foods because of their high status [29].
Florida. – Stable carbon and nitrogen isotope analysis of human skeletal remains
from several coastal, inland, and estuarine archaeological sites in Florida was performed
to reconstruct indigenous subsistence adaptations prior to European contact, and to
compare them with those associated with Mississippian and other cultural groups in the
southeastern United States. The samples selected represent a time range from the Archaic
Period up to the time of European contact about 1500 AD, and our preliminary results
suggest that the local ecological setting was the prime factor in determining indigenous
subsistence adaptations.
Native Americans living at inland settlements such as Cross Creek and Melton Mound
have carbon isotope ratios in their collagen and apatite which are consistent with dependence
entirely on C3 plants, wild animals that ate C3 plants, and freshwater fish, with
the latter also suggested by their high nitrogen isotope ratios. In contrast, those tested
from coastal sites such as Bay Pines and Bayshore Homes have the same nitrogen isotope
ratios but significantly increased carbon isotope ratios in collagen and apatite, indicating
the significant consumption of marine resources. While maize could have been consumed
to a minor extent at these coastal sites, and not be apparent from the carbon isotope
ratios, its clear absence in isotopic values for inland sites, and the lack of any archaeological
evidence for maize consumption anywhere in central Florida, suggests that it had
not yet been introduced to this region prior to European contact. This is particularly
significant in that other sociopolitically complex Native American cultures were dependent
on agriculture, but it appears that such cultures were able to develop in Florida
based on hunting and gathering strategies that had been practiced for millennia [30].
Europe. – In Mediterranean Europe an important question is that of the importance
of marine resources before, during, and after the introduction of domesticated plants and
animals in the Neolithic. In this region, it is possible to quantify the contribution of
fish and other marine resources to prehistoric diets, especially where C4 plants were not
consumed directly by humans, or by animals as fodder. Surprisingly, analyses of many
samples from coastal sites such as Grotta dell’Uzzo in Sicily, Arene Candide and other
sites in peninsular Italy, and the islands of Malta and Crete, all have carbon isotope
ratios which are not consistent with marine food consumption as a staple during the
Mesolithic, Neolithic, or Bronze Age periods (fig. 5).
This is in contrast at least with work on Atlantic Europe by Michael Richards (Bradford
University), who has documented heavy dependence on seafood in the Mesolithic,
and a resurgence in its importance in the Roman period [31, 32]. Changes which do
occur in Mediterranean Europe include dietary variation between elite and non-elite
people during the Bronze Age, for example at sites like Mycenae, and the introduction of
African millet (a C4 plant) in the 1st millennium BC, now documented at sites in Greece
and Slovenia.
442 R. H. Tykot
Fig. 5. – Carbon vs. nitrogen in human collagen for several Mediterranean inland, coastal, and
island sites, with seafood only measurably important for some Mycenae elites.
China. – Current interpretations are that millet was domesticated in northern China
and rice in southern China by the beginning of the neolithic period, and then spread
to other regions of China and Asia. The chronology of this spread and the quantitative
importance of these domesticates, however, is not well known. In collaboration with
Anne Underhill at the Field Museum and colleagues in China, human bone and tooth
samples were selected from a number of burials at the Longshan period (2600–1900 BC)
site of Liangchengzhen in Shandong province site, and isotopically analyzed along with
samples of flora and both terrestrial and aquatic fauna. Our preliminary results suggest
that in northern China, millet was a staple crop used for domesticated animals, but to a
far lesser degree for direct human consumption, probably because of the widespread use
of rice by this period. Isotopic analyses of residues from two ceramic vessels in tombs
were shown to be from fish, and support the general archaeological interpretation that a
diversity of domesticated and wild foods were consumed.
Microsampling of teeth. – Like hair, teeth are incremental growth structures, which
presents both advantages and limitations for isotopic analysis. In most cases, archaeologists
attempt to be minimally destructive in their analyses, so the removal of a small
sample of tooth enamel is required (with the tooth mostly preserved). The diet revealed
by isotopic analysis of such a small sample, however, will be very short term, perhaps
as little as a few weeks. A series of analyses of sequentially formed enamel layers may
then be used to examine short-term or seasonal variations in diet, and even to correlate
dietary changes with pathologies such as enamel hypoplasia. Preliminary results from
our project in collaboration with Douglas Ubelaker show potentially large isotopic dif-
Stable isotopes and diet 443
ferences within a single human tooth (4 per mil) [33]. On the other hand, “routine”
studies in which comparisons are made of individuals (or of multiple teeth from a single
individual) from which a single small enamel sample has been tested must allow for the
possibility of seasonal dietary variation obscuring variability which otherwise would have
been based on eruption age and weaning practices, migration, etc.
Strontium isotope analysis
Strontium isotope ratios (87
Sr/86
Sr) have been shown to be useful in identifying the
geographic area in which people lived, since strontium from the local environment is
incorporated into body tissues from the water, plants and animals consumed. In particular,
tooth enamel samples are compared with bones from the same individuals to see
if that person has moved between birth/childhood and adulthood. In central Europe, it
has been shown that the Beaker People of the 3rd millennium BC were highly mobile,
and thus probably directly responsible for the distribution of their widespread cultural
attributes [34]. At the large site of Teotihuacan (Mexico), it has been demonstrated that
many of the people living in the part of the city identified as a Oaxacan section actually
were born in Monte Alban or another place in Oaxaca and later moved to Teotihuacan
[35]. Current studies in my lab and elsewhere include early neolithic peoples and the
spread of agriculture in Europe; seasonal movements and transhumance; and the origins
of sacrificial victims found at sites in Mesoamerica and the southwest US.
REFERENCES
[1] Vogel J. C. and van der Merwe N. J., Am. Antiq., 42 (1977) 238.
[2] van der Merwe N. J. and Vogel J. C., Nature, 276 (1978) 815.
[3] Schoeninger M. J. and DeNiro M. J., Science, 220 (1983) 1381.
[4] Ambrose S. H. and Norr L., in Prehistoric Human Bone: Archaeology at the Molecular
Level, edited by Lambert J. B. and Grupe G. (Springer-Verlag, New York) 1993, pp. 1-
37.
[5] Tieszen L. L. and Fagre T., in Prehistoric Human Bone: Archaeology at the Molecular
Level, edited by Lambert J. B. and Grupe G. (Springer-Verlag, New York) 1993, pp. 121-
155.
[6] Wright L. E. and Schwarcz H. P., Am. J. Phys. Anthropol., 106 (1998) 1.
[7] O’Connell T. C. and Hedges R. E. M., Am. J. Phys. Anthropol., 108 (1999) 409.
[8] Sealy J., Armstrong R. and Schrire C., Antiquity, 69 (1995) 290.
[9] Sponheimer M. and Lee-Thorp J. A., Science, 283 (1999) 368.
[10] van der Merwe N. J. and Medina E., Geochim. Cosmochim. Acta, 53 (1989) 1091.
[11] van der Merwe N. J., in The Chemistry of Prehistoric Human Bone, edited by Price
T. D. (Cambridge University Press, Cambridge) 1989, pp. 105-125.
[12] Ambrose S. H., J. Arch. Sci., 18 (1991) 293.
[13] Schoeninger M. J. and DeNiro M. J., Geochim. Cosmochim. Acta, 48 (1984) 625.
[14] Schwarcz H. P. and Schoeninger M. J., Yearbook Phys. Anthropol., 34 (1991) 283.
[15] Schoeninger M. J., Evolut. Anthropol., 4 (1995) 83.
[16] Katzenberg M. A. and Harrison R. G., J. Arch. Res., 5 (1997) 265.
444 R. H. Tykot
[17] Larsen C. S., Bioarchaeology: Interpreting Behavior from the Human Skeleton
(Cambridge University Press, Cambridge) 1997.
[18] Ambrose S. H., J. Arch. Sci., 17 (1990) 431.
[19] Koch P. L., Tuross N. and Fogel M. L., J. Arch. Sci., 24 (1997) 417.
[20] Tykot R. H. et al., unpublished data.
[21] Burger R. L. and van der Merwe N. J., Am. Anthrop., 92 (1990) 85.
[22] Ubelaker D. H., Katzenberg M. A. and Doyon L. G., Am. J. Phys. Anthropol., 97
(1995) 403.
[23] van der Merwe N. J., Lee-Thorp J. A. and Raymond J. S., in Prehistoric Human
Bone: Archaeology at the Molecular Level, edited by Lambert J. B. and Grupe G.
(Springer-Verlag, Berlin) 1993, pp. 63-97.
[24] Tykot R. H. and Staller J. E., Curr. Anthropol., 43 (2002) 666.
[25] Tykot R. H. and Burger R., in preparation.
[26] Ericson J. E., West M., Sullivan C. H. and Krueger H. W., in The Chemistry of
Prehistoric Human Bone, edited by Price T. D. (Cambridge University Press, Cambridge)
1989, pp. 68-104.
[27] Tykot R. H., van der Merwe N. J. and Hammond N., in Archaeological Chemistry.
Organic, Inorganic and Biochemical Analysis, edited by Orna M. V. (American Chemical
Society, Washington, DC) 1996, pp. 355-365.
[28] Tykot R. H., Archaeological Chemistry. Materials, Methods, and Meaning, edited by
Jakes K. (American Chemical Society, Washington, DC) 2002, pp. 214-230.
[29] Whittington S. L., Tykot R. H. and Reed D. M., submitted to Ancient Mesoamerica.
[30] Tykot R. H., Kelly J. A. and Milanich J., to be published in Geoarchaeol. Bioarchaeol.
Studies.
[31] Richards M. P. and Mellars P. A., Antiquity, 72 (1998) 178.
[32] Richards M. P., Hedges R. E. M., Molleson T. I. and Vogel J. C., J. Arch. Sci.,
25 (1998) 1247.
[33] Wilson J. W., Tykot R. H. and Ubelaker D. H., Am. J. Phys. Anthropol., 114 (2001)
166.
[34] Price T. D. et al., in Archaeometry 98: Proceedings of the 31st Symposium: Budapest,
April 26-May 3 1998, edited by Jerem E. and Biro K. T. (Archaeopress, Oxford) 2002,
pp. 33-40.
[35] Price T. D., Manzanilla L. and Middleton W. D., J. Arch. Sci., 27 (2000) 903.