|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
ISOTOPIC DETERMINATION OF REGION OF ORIGIN IN MODERN PEOPLES:
APPLICATIONS FOR IDENTIFICATION OF U.S. WAR-DEAD FROM THE
LAURA A. REGAN
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
Laura A. Regan
To Ronald D. Reed, Ph.D., Brigadier General, USAF
11 November 1948 20 April 2005
It is with bittersweet appreciation that I thank an incredible mentor for taking a risk and
agreeing to this absolutely insane adventure; for the opportunities he provided and his
unwavering confidence in me. It is with great regret that I cannot share my success in
this endeavor with him. I was fortunate to know him and I dedicate this work to his
I could not have met all the crazy deadlines of this breakneck-paced program
without the assistance of a great many people. First, I wish to extend my sincere
appreciation to the members of my doctoral committee, Drs. Anthony B. Falsetti (Chair),
David Daegling, Thomas Holland, Connie Mulligan, and David Steadman. Not one of
you ever indicated you had any reservations about my ability to complete this program in
a blistering 3 years. Your faith in me fueled me on when I was doubting myself.
Boss, I have never encountered a professor who is so nurturing of his students, yet
has no qualms about telling us when we're being knuckleheads. I cannot express my
gratitude for all you have taught me, your unwavering friendship, your constant
confidence in me, and your personal support throughout this program. You've opened
countless doors for me. I will never be able to repay you for all of your kindness,
generosity, and all of the laughs. You take good care of your "kids." Please take good
care of yourself as well. No setting yourself on fire any more.
Dr. Daegling opened my mind and challenged me to think critically on a whole
new plane. The academic rigor of his courses was both mildly overwhelming and
incredibly fulfilling. Receiving an "A" in his course was truly something to covet. A
great thank you goes out to Dr. Thomas Holland, Joint POW/MIA Accounting
Command-Central Identification Laboratory Scientific Director, for supporting me and
this project, allowing me to intern with him for three incredible months, and putting me
on that week-long C-17 ride to Vietnam. I have partaken in some once in a lifetime
experiences through your generosity, and those memories I will always cherish. Please
don't forget about me. I'll be looking for ajob in about 8 years.
Dr. Mulligan broke down my internal block when it came to understanding genetics
and taught me a great deal about attention to detail and organization. She held my feet to
the fire and made me not only address but fully understand the flaws in my work, vastly
improving the quality of my scientific work. The journey to enlightenment could sure be
frustrating though. Dr. Steadman was a constant source of enthusiasm and energy. His
positive attitude kept me going and allowed me to overcome a long seeded loathing of
avian fauna that arose during my days in undergraduate Vertebrate Zoology. Birds are
I owe Dr. Andy Tyrell a great deal of gratitude as well, for getting me se up at CIL
and through his continued guidance and assistance. To Col. (ret) Thomas and Col. Merle
Sprague, thank you for allowing me to mooch off of you for 3 months. You opened your
hearts and home to me. I am truly grateful and a better person for knowing you. I would
also like to pass along my appreciation to LTC Mark Gleisner, for showing me the basics
of drilling teeth and along with the rest of the CIL dental guys, answering my many,
I owe a great deal to Col. Nancy Perry and Maj. Albert Ouellette, 10th Dental
Squadron, U.S. Air Force Academy for agreeing to assist me with this project and
especially to Albert, who provided over 1000 freshly extracted third molars to me during
the course of this study (are you sure you guys do not have a quota?). Thanks also go out
to Drs. Jack Meyer and Ray Berringer from the North Florida/South Georgia Veterans
Health System, Veterans Affairs Dental Clinic..
I am indebted to Dr. Bruce MacFadden, Florida Museum of Natural History, who
really exposed me to the possibilities of isotope studies and in whose class this project all
took shape. He graciously allowed me the use of his laboratory to prepare samples, took
keen interest in my progress, and always greeted me with a smile, no matter what the
circumstances. I also learned a great deal about the basics of isotope work from Dr.
Joann Labs Hochstein and am grateful for her tutelage as I was starting out.
I would like to thank Dr. John Krigbaum for planting the seed of awareness of
stable isotope studies and bailing me out during a great time of need. Your genuine
concern for your students is well known. I would like to acknowledge the contributions
of George Kamenov, who showed me the ropes of heavy isotopes and gave me great
insight into their power and Dr. Jason Curtis, who worked around my crazy schedule,
even when his was just as bad, and always had time to answer my questions, even when
he was out of the country.
To the "Frogs," I cannot wait to rejoin your ranks. A special thank you goes out to
Col (ret) James Kent. Sir, you have been there for the course of my journey in academia.
I thank you for your patience, guidance, and gentle pushes in the right direction. Look-I
did not change my major once this degree program!
I would like to thank my family and friends for all of their love and support over
the years. You mean the world to me. Anna, you are the most selfless friend anyone
could ever be blessed with. I don't know what I would have done without you but I do
know I can never repay your kindness nor the countless times you bailed me out of a
difficult situation. Greg, you were a constant sounding board and helped me through
some very difficult times. Hang in there my friend. There is light at the end of the
tunnel, and no, it isn't a train. Shanna and Erin, I can't tell you how my stress melted
away when I was in your company-and thanks to some apple juice-laced wine. I'll miss
our girls' dinners more than you will ever know. Thanks to Laurel and her technical
wizardry and extraordinary and often utilized dog sitting skills. I also wouldn't have
been able to launch this project if it hadn't been for the assistance of Alicia during the 3
months I was away. I can't tell you how much your help eased my mind. I owe much
appreciation to Carlos for teaching me the basics of tooth identification and to Miss
Shiela for assisting me with numerous mind-numbing tasks. Have a great Air Force day!
I'd also like to thank the rest of the Pound Lab rats and lab rats by-proxy: Dr. Mike
Warren, Shuala, Joe, Trey, Paul, Ron, Nicolette, Kathy, Debbie, Pat, Melissa, Megan, and
Jennifer; and my friends Chad, Laurie, and Erin. You added immeasurable levity to my
life during a period of extreme stress and thoroughly deprogrammed me. I couldn't ask
for a better cohort to be associated with. It's going to be tough going back to the real
I also owe a huge debt of gratitude to two undergraduate assistants, Ursula Zipperer
and Ana del Alamo, who spent countless hours helping me with the most mundane of
tasks. Lastly, I must express my heartfelt appreciation to Calvin and Hobbes. I would
not have survived this program, especially the first year, without you guys, but it would
have been nice if you had not eaten the door ..twice.
To all the men and women who have gone before me in service to our nation and to
those who currently serve, I salute you. I am proud to be among your company. It has
been an incredible honor to complete this project with the hopes of reuniting families
with their long, lost, loved ones. Until they are home ...
My tuition was provided by the United States Air Force. This research was funded
in part by the Joint POW/MIA Accounting Command-Central Identification Laboratory,
the C.A. Pound Human Identification Laboratory, a William R. Maples Scholarship, and
my savings account.
TABLE OF CONTENTS
A C K N O W L E D G M E N T S ................................................................................................. iv
L IST O F T A B L E S ........ .......................................................... ..................... xii
LIST OF FIGURES ......... ....... .................... .. ....... ........... xiv
ABSTRACT ........ .............. ............. ........ ..................... xvi
1 ST A B L E ISO T O PE S ............ .............................................................. ......... .. ....... 1
Stu dy Isotopes ............................................... 6
C arb on ............................................................. . 6
O x y g en ....................................................... 7
Strontiu m ................................................................... 8
L e a d ........................................................................................................1 0
Fractionation ................... ...... ......................................... ...............11
Frequently Sampled Human Tissues .............................. ...............13
B o n e ............................................ ...... ............................................................ 1 4
T e e th ............................................ ....... ........................................1 5
H air ........... ........ ..... .............................. .......... 17
F ingernails and T oenails ........................................................ 18
S k in ............................................ ................................................1 9
Com plications .................................. .......................... .... .... ........ 20
Diagenesis ................................ ......... 20
Anthropogenic Contamination ................................ ............... 27
G global E conom y ..................................... ....................................................... ...... 28
2 APPLICATIONS OF STABLE ISOTOPE ANALYSES ........................................30
T ra c in g S tu d ie s ...................................................................................................... 3 0
F ractionation Studies .............................................................32
Z o ology an d E ecology ............................................................................................. 3 3
A rch aeology .................................3.............................5
D iet A ssessm ent .............................................................35
Introduction of m aize ......................................................... ..... .......... 35
W meaning practices ......................................................... 37
R region of O rigin .................. ........................................................39
M material Culture.............. .. ................ ................ .. 41
Forensic Investigations ............. .... ............. ................................... 42
3 HUMAN FORENSIC IDENTIFICATION......................................................50
M military Identification M measures ........................................ .......................... 54
P resent Study ................................................................... 58
4 M ATERIALS AND M ETHODS ........................................ ......................... 66
D en tal P ro to c o ls .................................................................................................... 6 7
S am pling ............................................................................................................... 73
Central Identification Laboratory ................................................. .........74
United States Air Force Academy and Veterans Affairs................ .............. ....77
Carbon and Oxygen Sample Preparation......................................... ............... 80
Central Identification Laboratory Samples ............................... ............... 80
United States Air Force Academy Samples ................................................83
Strontium and Lead Sample Preparation ............ .........................................84
Statistical A analyses .......................................................... ... ...... ..... 90
5 ANALYTICAL COMPARISON OF EAST ASIAN AND AMERICAN
S A M P L E S ................................................................9 3
L eight Isotop es ....................................................... 93
C a rb o n ........................................................................................................... 9 3
O x y g e n .............................................................................................................. 1 0 8
Acetic Acid Test ...... .................. .......... ........114
Heavy Isotopes .......... .. .. ................ .......... 118
Strontium ............... ......... .......................118
Lead .............. ............................................ ..... ..... ......... 125
M ulti-elem ent A approach ..................................................................................... 133
6 VARIATION WITHIN USAFA SAMPLES ................. ................. ...........139
Y e a r o f B irth ....................................................................................................... 1 3 9
S e x .........................................................................1 4 1
R a c e .................................................................................................................... 1 4 2
T tobacco U se ................................................................ 143
D ie t ........................................................................ 1 4 6
R e sid e n c y ............................................................................................................ 1 4 6
S tro n tiu m ....................................................... 14 6
Lead ....................................... 150
Regionality ......................................... ......................................... 151
Relationship Between 6180 Values and Latitude ...................................... 156
Duplicate Residences ................................. ........................... ........... 159
Comparison to the Literature ................................. ...............................161
7 SU M M A R Y /C ON CLU SION ........................................................ .....................165
A REPLICATED VETERANS AFFAIRS BINDER................... ........................ 172
B CENTRAL IDENTIFICATION LABORATORY SAMPLING.............................203
C UNITED STATES AIR FORCE ACADEMY SURVEY RESULTS ....................210
D EXAM PLE PRISM LOAD SHEET.......................... .................... ............... 248
E COLUMN CHEMISTRY VESSEL AND IMPLEMENT CLEANING
IN STRU CTION S ............................................ .. .. .... ......... ......... 249
F MISCELLANEOUS HEAVY ISOTOPE RESULTS WORKSHEETS ..................251
L IST O F R E FE R E N C E S ..................................................................... ..... .................258
B IO G R A PH IC A L SK E T C H ........................................ ............................................278
LIST OF TABLES
1-1 Stable isotope standard materials and calibrants............... ..... ................. 5
1-2 Mean age of completion of permanent crown mineralization..............................16
2-1 Mean and standard deviations for selected groups of immigrant teeth (enamel).....49
3-1 Form s of forensic identification. ............................... ............................ ................ 54
3-2 Numbers of unaccounted for U.S. prisoners of war and/or those missing in
a ctio n ............................................................................ 5 9
3-3 United States casualties in Southeast Asia by race. ............................................60
3-4 United States military listed as unaccounted for in Southeast Asia by race. ...........60
4-1 Crown formation/tooth eruption.............. ............................... 74
4-2 Isotope sam pling m atrix. ............................................... ................................ 78
5-1 Summary statistics and general linear model results of all isotopes examined for
CIL samples compared to USAFA samples (CIL outlier excluded). All values
are in % o ............................................................................. .9 4
5-2 Carbon and oxygen isotope results. All values are in %o. ....................................95
5-3 Central Identification Laboratory outlier run data. ..............................................102
5-4 613C value comparison. Twelve most enriched CIL samples and 12 most
depleted USAFA samples (CIL outlier excluded). All values measured in %0.....103
5-5 Summary statistics and general linear model results of all isotopes examined for
American and foreign USAFA comparison (CIL outlier excluded). All values
are in %o. ................... ..... ........... ................... ................. 105
5-6 Summary statistics and general linear model results of all isotopes examined for
CONUS and overseas USAFA comparison (CIL outlier excluded). All values
are in %o. .......................... ........... ......................... ................ 105
5-7 East Asian 6180 values, in ascending order. .................................. .................109
5-8 Partial list of USAFA 6180 values, in ascending order (30 most depleted and 30
m ost enriched) ................................................................. ... ......... 110
5-9 Results of acetic acid test, with intertooth comparison when available...............16
5-10 Strontium isotope values for CIL and USAFA samples, in ascending order.........119
5-11 Comparison of the means for multiple runs of the GLM procedure for 87Sr/86Sr.
(CIL outlier excluded.) .......................................... ............ ................. 124
5-12 Lead isotope results for East Asia. ............................................... ............... 126
5-13 Lead isotope results for USAFA. .........................................................................127
5-14 Comparison of spiked lead concentration data (actual) with semi-quantitative
data. (All values are in ppm .). .................................. ................................ 133
6-1 USAFA-provided sampling demographics, American natal region only..............140
6-2 Locations during amelogenesis represented by sampled USAFA teeth. .............147
6-3 Strontium isotope values for American USAFA samples, in ascending order......148
6-4 Mean 207Pb/206Pb values for major U.S. lead ore deposits................................ 152
6-5 207Pb/206Pb values Americans reared in the United States, in ascending order......152
6-6 Region membership based on 6180 values.............. ..................... ........ ....... 153
6-7 Region-pair comparison for difference in 6180 means. .......................................155
6-8 Summary statistics for American USAFA 6180 values based on latitude. All
values are in %o. ........... ..... .................................................. .................. 158
6-9 6180 values corresponding to cities in which multiple participants resided. (All
values in %o.) .......................... ........... .............. 160
6-10 Comparison of Alberta fur trader lead values to USAFA donor from Alberta......164
B-l Central Identification Laboratory sampling data. .............................................204
C-l United States Air Force Academy survey data. ................................................211
F-l Semi-quantitative Sr concentration calculation matrix .......................................252
F-2 Semi-quantitative Sr concentration calculation matrix.......................................255
F-3 Comparison of the means for multiple runs of the GLM procedure for Pb. (CIL
outlier excluded)........ ...... .......................................... ......... 257
LIST OF FIGURES
4-1 Joint POW/MIA Accounting Command survey. ............................................. 70
4-2 Pre-drilling photo of CIL-033 #19 with data card. Note: the accession number is
purposely, partially obscured. ............................................................................76
4-3 Pre-drilling photo of AFA-093 #32...................................... ..................... 80
4-4 Loaded tray for PRISM mass spectrometer analysis. ............. .............. 82
5-1 Carbon and oxygen isotope results with overlapping value overlay......................99
5-2 Carbon and oxygen isotope results for American and foreign USAFA
com prison ...................................................... ................. 104
5-3 Carbon and oxygen isotope results for CONUS and overseas USAFA
com prison ...................................................... ................. 106
5-4 Latitudinal dispersion of major natal regions featured in this study. East Asia is
on the right. Information drawn from Rand McNally Atlas (1998)......................111
5-5 Weighted Annual 6180 for Asia. Map reproduced from IAEA (2001) ..............112
5-6 Weighted Annual 6180 for North America. Map reproduced from IAEA (2001). 112
5-7 Plot of strontium values compared to 208Pb/204Pb. .......................................... 121
5-8 Plot of 87Sr/86Sr compared to 206Pb/204Pb ........... ............................. ............. 122
5-9 Box and whisker plot of 87Sr/86Sr values. ................................................... 123
5-10 Comparative histogram of CIL and USAFA sample Sr concentrations (semi-
quantitative).................................... ................................ .......... 125
5-11 Plot of 206Pb/204Pb compared to 208Pb/204Pb. ............................... ............... 128
5-12 Plot of 206Pb/204Pb compared to 207Pb/204Pb. .......... ........................................ 128
5-13 Plot of 208Pb/206Pb compared to 207Pb/206Pb. .......... ........................................ 129
5-14 Comparative histogram of CIL and USAFA sample Pb concentrations (semi-
quantitative).................................... ................................ ..........131
6-1 Strontium isotope composition of the U.S. showing inferred s87Sr values, as
calculated by age variations in basement rocks. Image reproduced from Beard
and Johnson (2000), with permission ............... ........................... .............. 149
6-2 Region m ap based on 6180 values..................................... ......................... 153
6-3 Plot of latitude compared to 6180 values. Error bars equal 1 std dev .................157
6-4 Range of 6180 values produced from this study for specific cities ......................161
D -1 Exam ple PRISM load sheet......................................................... ............... 248
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
ISOTOPIC DETERMINATION OF REGION OF ORIGIN IN MODERN PEOPLES:
APPLICATIONS FOR IDENTIFICATION OF U.S. WAR-DEAD FROM THE
Laura A. Regan
Chair: Anthony B. Falsetti
Major Department: Anthropology
This study is novel in that it is the first of its kind to compile a reference sample of
isotopic values associated with known natal regions to be utilized in forensic work.
Stable isotopes of carbon, oxygen, strontium, and lead were examined to determine if
natal origins could be assessed isotopically between Southeast Asian and American
dental remains as well as regionally within the United States. Teeth believed to be of
East Asian origin were compared to the extracted third molars of recent American dental
patients. Living subjects completed surveys detailing physiological, behavioral, and
residential information that affect isotope values. The least squares means for all isotope
values examined exhibited significant differences between the East Asian and American
cohorts. Based on this information, a discriminant function was created that correctly
classified individuals, through resubstitution and cross-validation, as belonging to one of
these two groups by 95% or better. The sexes differed significantly as to their carbon
ratios with females displaying more enriched values than males. Significant differences
were also noted for 613C means among those who have never used tobacco products and
those who partook of smokeless tobacco. American strontium values displayed a distinct
trend toward homogenization, with the mean value for 87Sr/86Sr varying only slightly
from that of seawater. In order to identify natal origin among Americans, nine regions
were created within the United States based on 6180 values. Good discrimination was
noted between the mountain states and the southern states. A discriminant function
analysis proved disappointing though, and additional sampling from most states is needed
to improve the statistical robusticity of the model. The results of this study will have
wide-reaching effects across the medico-legal spectrum. This body of research will serve
as the foundation for a database of modern, human, geolocational isotope values that will
assist not only in the identification of fallen servicemen and women, but in the
identification of victims of mass fatality incidents, undocumented aliens who perish
attempting entry into the U.S., and local skeletal "Jane and John Doe" cases.
Ascertaining the national origin of unidentified human remains is problematic,
especially with the passage of time. Often, the number of identifiable bony elements is
so few, fragmentary and/or degraded by the chemical properties of the soil, that
estimating biological profiles and DNA analyses cannot effectively be performed. This
challenge is particularly acute for the Joint POW/MIA Accounting Command's Central
Identification Laboratory (JPAC-CIL). The identification of unknown remains believed
to be missing U.S. service personnel is frequently hampered by high levels of degradation
and fragmentation as a result of circumstances of loss and subsequent taphonomic
regimes. If the geo-political region of origin for a set of remains could be established, it
would facilitate the construction of identification shortlists, especially from large, open-
ended decedent populations. This, in turn, would provide a highly effective means of
excluding possible candidates for identification, notably for human remains whose
provenience is either unknown or suspect. One potential tool in determining
geolocational origins of skeletal material is that of stable isotope analyses. Developed
primarily in the geochemical community (Fogel et al. 1997), stable isotope work has
revolutionized the anthropological realm, beginning with pioneering, archaeological,
dietary studies in the late 1970s (DeNiro & Epstein 1978a and 1978b, van der Merwe &
Isotopes of a particular element are atoms whose nuclei contain the same number of
protons but differ in their number of neutrons (Hoefs 2004). It is the number of protons
in the atom that determines what the element is as well as how many electrons the atom
has (Herz & Garrison 1998). An atom at rest has a neutral charge; therefore, the normal
state for an atom is to have the same number of protons within the nucleus as electrons
outside of the nucleus. As stated previously, isotopes vary because of the differing
number of neutrons within the nucleus. This neutron variation, will in turn, affect the
atomic masses of different isotopes of the same element because the atomic mass is a
measure of the sum of the number of protons and neutrons (Hoefs 2004).
For example, carbon has an atomic number of "6," meaning an atom of carbon
contains 6 protons within the nucleus. Even though the number of protons is constant
within a carbon atom, it can take on three isotopic forms: 12C, 13C, 14C. A carbon atom
with a mass of 12 (denoted 12C) has 6 protons and 6 neutrons, one less neutron than a
carbon atom with a mass of 13 (13C) and two fewer neutrons than 14C.
Since chemical reactions are largely determined by the ionic or atomic electron
configuration, the varying isotopes of an individual element will have the same chemical
properties (Schwarcz & Schoeninger 1991). Different isotopes of a single element will
have different kinetic and thermodynamic properties when they undergo chemical
reactions though, because of differences in reaction rates and heat capacity influenced by
their different atomic masses (Urey 1947). So, while isotopes of a like element will react
the same chemically, they will react at different rates, due to their different atomic masses
and sizes. Different metabolic and chemical processes therefore change the ratios
between the isotopes in a characteristic manner (van der Merwe 1982). It is also noted
that as atomic weight increases, the differences in thermodynamic properties between
isotopes generally decrease (Urey 1947). In other words, light isotopes such as those of
hydrogen, carbon, and oxygen will have a much greater variation in their thermodynamic
and kinetic characteristics than heavier isotopes such as strontium and lead.
Stable isotopes are not radioactive (Hoefs 2004), thus they do not spontaneously
change into another atom or another isotope of the same element (Herz & Garrison
1998). Revisiting the carbon example, when considering the three isotopic forms of
carbon (12C, 13C, 14C), the former two are stable isotopes, while the latter is radioactive
(van der Merwe 1982), and commonly utilized for archaeological dating purposes.
Stable isotopes may also be characterized as radiogenic or nonradiogenic. A
particular isotope is classified as radiogenic if it is the product of the decay of a "long-
lived" radioactive isotope (Schwarcz & Schoeninger 1991). Strontium ( Sr) and lead
(206Pb, 207Pb, 208Pb) are the primary radiogenic isotopes used in nutritional ecology
studies. 87Sr forms from the radioactive decay of rubidium (87Rb) while 206Pb and 207Pb
arise from the decay of uranium (238U in the case of 206Pb and 235U for 207Pb) and 208Pb
results from the decay of thorium (232Th) (Herz & Garrison 1998). These radiogenic
isotopes vary considerably in abundance with respect to their associated non radiogenic
isotopes (86Sr and 204Pb) (Schwarcz & Schoeninger 1991) and serve as useful analytical
Eighty-one elements have stable isotopes of varying numbers (Herz & Garrison
1998). All of the biochemically important elements, with the exception of fluorine, have
more than one stable isotope (Schwarcz & Schoeninger 1991). Four of these; carbon,
oxygen, strontium, and lead; were examined in this study will be discussed in detail being
on page 6.
Measurements of stable isotopic ratios are performed by a mass spectrometer, an
instrument that determines the relative abundances of different isotopic masses in a
variety of elements (Thirlwall 1997). For carbon, the mass spectrometer determines the
raw ratio of 13C/12C, which it then compares to the ratio of a marine carbonate standard,
known as Pee Dee belemnite (PDB, now referred to as V-PDB, based on the Vienna
Convention; Hoefs 2004). The difference between the sample ratio and the V-PDB
standard ratio is what is known as the relative 13C content and is the value reported and
used for inferential purposes (van der Merwe 1982). The equation is as follows:
element = (ratiOsample/ratiOstd -1) x 1000%o = value in %o
613C 13C/12same x 1000%o (1-1)
This measure is denoted by the symbol 6 (delta) and measured in parts per mil (%o)
(van der Merwe 1982). If the hypothetical 13C/12C ratio of a sample was calculated as 12
per mil less than the V-PDB standard, the 613C value would be -12%o and considered
depleted compared to the sample. It is important to note that the V-PDB standard does
not equal zero (it equals 2.0671 x 10-6; Hoefs 2004) and results should not be interpreted
as deviations from the zero point.
Oxygen values for 180/160 are calculated similarly. When 6180 is calculated in
concert with 613C, the V-PDB standard is used along with a conversion factor (Dr. Jason
Curtis, personal communication). When isotopic calculations are performed singly or in
combination with hydrogen, the internationally accepted standard of standard mean ocean
water (SMOW or V-SMOW) is used (Hoefs 2004). The heavy isotopes of strontium and
lead are not generally normalized to a conventional standard, but instead, results are
expressed directly as ratios (Herz & Garrison 1998) and the standards are used for mass
spectrometer calibration adjustments.
Stable isotope standards have been drawn from a variety of sources over the years.
Some of the most commonly utilized in zoological and anthropological studies are listed
in Table 1-1.
Table 1-1. Stable isotope standard materials and calibrants.
Element Ratio Standard (Std) Std Notation Std Value
Hydrogen1 D/H (2H1H) Standard Mean SMOW or 155.76 x 10-6
Ocean Water V-SMOW
Carbon1 13C/12C Belemnitella PDB or V- 2067.1 x 10-6
Americana from PDB
Nitrogen1 15N/14N Air nitrogen N2 (atm) 3676.5 x 10-6
Oxygen' 1O/O60 Standard Mean SMOW or V- 2067.1 x 106
Ocean Water SMOW
Belemnitella PDB or V- 2067.1 x 10-6
Americana from PDB
Strontium2 7Sr/86Sr Strontium NBS-987 or 0.7045
carbonate/bulk NIST 987
Lead3 208Pb/204Pb Lead metal wire NBS-981 or 36.696
207Pb/204Pb NIST 981 15.491
SFrom Hoefs (2004)
2 From Beard and Johnson (2000)
3 George Kamenov (2006)
In 1968, Margaret Bender first reported that the major photosynthetic pathways of
plants manifest themselves in distinct carbon isotope ratios. This discovery served as the
catalyst for the multitude of carbon isotope studies documented in the literature today.
When interpreting carbon isotope signatures, one must harken back to the days of basic
biology class and discussions of the differences in the two major photosynthetic systems.
C3 photosynthesis occurs in the majority of cultivated and wild plants in temperate
regions (Schwarcz & Schoeninger 1991), such as wheat, rice, and barley, and produces
an initial three-carbon metabolite (van der Merwe 1982, Schwarcz & Schoeninger 1991,
MacFadden et al. 1999b). C4 photosynthesis, found in more drought-resistant plants,
produces an initial four-carbon compound in cultigens such as sugar cane, maize and
millet (van der Merwe 1982, Schwarcz & Schoeninger 1991, MacFadden et al. 1999b).
These different metabolic processes produce different isotopic ratios, which are then
incorporated into plant tissues. C4 plants exhibit more rapid carbon dioxide intake
leading to values between -9%o and -16%o. C3 plants on the other hand, have slow rates
of carbon dioxide uptake leading to values from -20%o to -35%o (van der Merwe 1982).
What makes carbon isotope analyses so powerful is that these to ranges do not
overlap. Intermediate values are found in plants utilizing a third photosynthetic pathway,
CAM or crassulacean acid metabolism (van der Merwe 1982, MacFadden et al. 1999b).
These plants are primarily succulents such as cactus and pineapple, and as such, they
neither factor significantly into most human diets nor the present research.
Plants demonstrate preferential uptake of 12C to 13C, thus they are depleted in 13C
compared to12C (Bender 1968). These two carbon species are differentially incorporated
into body tissues (i.e., they are fractionated in a characteristic manner) during digestive
processes (Durrance 1986). As a result of the differences in photosynthetic pathways in
plants, it is also possible to determine approximate proportions of C3 versus C4 plants in
an individual's diet based on the 613C value (Schwarcz & Schoeninger 1991).
Carbon isotopes also convey information regarding the use of marine foods in an
organism's diet. Marine animals present isotopic signatures intermediary to C3 and C4
food chains (Schoeninger & DeNiro 1984, Larsen et al. 1992). Marine mammals and fish
display 613C values that are enriched by roughly 6%o over animals that feed on C3
foodstuffs, and depleted by about 7%o compared to animals that feed on C4-based foods
(Schoeninger & DeNiro 1984). The best indicator of a reliance on marine food sources is
the information provided through a joint 613C and 815N analyses (Schoeninger & DeNiro
1984, Ambrose & Norr 1993).
Oxygen is the most abundant elemental component of the earth's crust (Herz &
Garrison 1998) and its isotopic ratios provide an indication of the point of origin of
remains. Isotopes of oxygen take the form of 160, 170, and 180 (Mattey 1997). Oxygen
is primarily incorporated into body tissues via atmospheric oxygen, water, and oxygen
bound in food (Sponheimer & Lee-Thorp 1999b). Because the 6180 value of atmospheric
oxygen is relatively constant, it is believed that oxygen isotopic signatures are primarily
representative of imbibed water, and to a lesser extent, the macronutrients found in
foodstuffs (Sponheimer & Lee-Thorp 1999b). The oxygen isotopes in water are
preserved in bone, teeth, and other tissues and are reflective of a particular environment
and climate, decreasing with increasing latitude, increasing altitude, and as you move
inland (Dupras & Schwarcz 2001, Kendall & Coplen 2001, Rubenstein & Hobson 2004).
Analytically available oxygen is present in both the phosphate and carbonate ions
of hydroxyapatite in the mineral phase of skeletal tissues. Most studies have examined
phosphate oxygen because the P-O chemical bond is much stronger than the C-O bone,
suggesting that phosphate oxygen is less susceptible to diagenesis than carbonate oxygen
(lacumin et al. 1996, Sponheimer & Lee-Thorp 1999b). Lengthy and harsh chemical
procedures are required to extract the phosphate oxygen from apatite however, while the
carbonate portion is easily obtained from the CO2 produced during mass spectrometry for
carbon isotopes (Sponheimer & Lee-Thorp 1999b). Bone carbonate has shown a strong
positive correlation to local meteoric water with an r2 value = 0.98 (lacumin et al. 1996).
Additionally, both carbonate and phosphate are better preserved by highly-mineralized
tooth enamel versus more porous dentin and bone phosphate (lacumin et al. 1996).
Strontium has been used to characterize prehistoric mobility patterns since the mid
1980s (Budd et al. 2004, Millard et al. 2004). There are four stable isotopes of strontium:
"Sr, 17Sr, 86Sr, and 84Sr. Only 87S is the product of radioactive decay (radiogenic), being
a product of the beta decay of rubidium 87. This radioactive decay pair, 8Rb-87Sr, has
consequently produced distinctively different 87Sr abundances in different parts of the
earth over its history (Beard and Johnson 2000) that have proven quite valuable in tracing
the origin of matter to a particular locale.
Strontium signatures depend purely on local geology since they reflect the
underlying bedrock of a particular area. Strontium isotopic ratios vary with the age and
type of bedrock underlying the soil. So the quantity of strontium in a particular rock will
depend not only on the amount of rubidium parent material found in the rock, but the age
of the rock as well as the original amount of 87Sr present in the rock when it was formed.
Strontium varies in plant tissue with the age and type of geological substrate or bulk
composition. Older soils are more enriched compared to younger soils as are calcium-
rich soils compared to calcium-poor soils. Additionally, atmospheric deposition or dry
fall from natural sources can also affect strontium values (Beard and Johnson 2000).
Anthropogenic factors that can influence isotope ratios include nuclear fallout, airborne
pollution from fossil fuels, and land-use practices that expose bedrock (Rubenstein and
Strontium is incorporated into human tissue following the calcium pathway
because this non-nutrient, non-toxic element has chemical properties similar to calcium
(Aberg et al. 1998). During nutrient uptake strontium often replaces calcium in bones
and therefore can be used to trace the flow of minerals from the soil through the food web
(Rubenstein and Hobson 2004). "Strontium concentrations in plants and animals are
controlled by trophic position, but the isotopic composition is invariant; that is, Sr does
not fractionate. Thus, bones and teeth in an individual will have different Sr abundances
but identical 87Sr/86Sr ratios" (Herz & Garrison 1998), with human enamel demonstrating
lower strontium content than bone (Price et al. 1994, Grupe et al. 1997, Beard & Johnson
2000). If food sources are local then, all participants in the food chain, regardless of what
tissue is sampled, should reflect the same isotopic signature.
Additionally, strontium abundance has commonly been examined to discriminate
between the meat and vegetable components in an organism's diet. Toots and Voorhies
(1965) published the seminal study in this area, discovering significant differences
(p-value <0.001) not only between the mean strontium concentrations for fossil Pliocene
carnivores and herbivores, but among the herbivorous grazers and browsers themselves.
The basis for this is that for each trophic level above the soil, there is a metabolic
discrimination against strontium in mammalian epithelium, as opposed to calcium
(Radosevich 1993). As one increases in trophic levels among the consumers, the
contribution from food sources to skeletal strontium is decreased at each step (Toots &
Voorhies 1965). Plants will retain 50-100% of the strontium found in the soil, with each
progressive trophic level exhibiting a reduction of 33% strontium over the lower level
(Radosevich 1993). Keep in mind that this refers to strontium abundance (or
concentration) and not the 687Sr value. So theoretically, someone such as a vegan should
have a higher strontium concentration than an ardent follower of the Adkins' diet, by
approximately 33%. Radosevich (1993) cautions against blindly accepting these
measurements however, without first considering factors such as parent material and soil
chemistry variation influencing plant uptake and physiological differentiation, as well as
behavioral changes in feeding strategies, trophic placement, and cultural practices.
"Lead is one of the most heavily utilized metals in human history" (Sangster et al.
2000). Lead has four naturally occurring stable isotopes: 204Pb, 206Pb, 207Pb, and 208Pb.
As previously discussed, the latter three isotopes are radiogenic. Because 204Pb is not
radiogenic, it serves as stable reference isotope (Sangster et al. 2000). Similar to
strontium, the isotopic composition of lead in a particular locale (or ore deposit) is
dependent upon four factors: 1) the length of time before lead was separated by
geological processes in the source reservoir; 2) the decay rate of the parent isotopes; 3)
the initial ratio of the abundance of the parent material to the abundance of lead in the
source reservoir; and 4) the initial isotopic constitution of the reservoir lead (Sangster et
al. 2000). The variations in parent isotope decay rates result in systematic differentials in
the ratios of 206Pb, 207Pb, and 208Pb to each other, as well as to 204Pb (Sangster et al.
2000). Most archaeological studies are based on the ratios of the radiogenic isotopes to
204Pb, whereas environmental studies tend to also form ratios from only the radiogenic
isotopes themselves (Dr. George Kamenov, personal communication). Additionally, lead
is favored by many researchers because like strontium, it does not exhibit fractionation in
nature (Stille & Shields 1997).
Lead is assimilated into skeletal elements in a similar manner to strontium, in that it
accumulates from the blood through calcium pathways and substitutes for calcium in the
carbonate hydroxyapatite fraction of hard tissues (Vogel et al. 1990). Juveniles exhibit a
higher propensity to absorb ingested lead than adults (Reinhard & Ghazi 1992), likely
due to the rapid modeling of bone occurring during the growth phase and because small
children tend to frequently put objects in their mouths. Lead particles are thought to enter
the body through ingestion, either through food stuffs/fluids or lead objects, or inhalation
(Gulson 1996). Environmental contamination by lead is found through mining
operations, waste dumps, emissions from lead smelting, coal combustion, and leaded
gasoline (Aberg et al. 1998). Furthermore, acid rain can transmit contamination from
emissions/combustion over great distances.
Prior to drawing conclusions regarding the delta value of a material, additional
issues such as fractionation effects must be factored in. Fractionation is the disparate
partitioning of isotopes between two substances or tissues (Hoefs 2004). Without it,
biological processes would be homogenous and some of the most powerful inferences in
isotopic analyses would not be possible.
Differential fractionation manifests itself in a variety of forms. One example is the
different rates carbon is fractionated as one progresses through the food chain. Carbon
found in the atmosphere is present with a near constant 13C/12C ratio of about 1:99
(Chisholm 1989). As plants incorporate carbon into their tissues during photosynthesis,
isotopic fractionation occurs altering the 13C/12C ratio. Since C3 and C4 photosynthetic
pathways differ chemically, they produce different degrees of fractionation. This is
beneficial, and in fact, essential in the case of carbon isotope studies, because the 613C
values can be utilized to classify between C3 and C4 plants and diets based on a complete
separation of approximately 14%o between groups allowing for discrimination between
them (DeNiro & Epstein 1978a, 1978b, Chisholm 1989, Ambrose & Norr 1993).
The selective metabolism and recombination of plant chemicals within organisms
feeding upon them, results in fractionation of elemental isotopes, leading to differences in
613C values between diet and bone collagen of primary consumers of +3%o to +5.3%o. An
additional fractionation factor of about +1%o must be accounted for as you increase in
trophic level (i.e., from primary to secondary consumer) (Schoeninger 1985, Chisolm
1989, Schoeninger 1989, Ambrose 1993). The 613C values of mammal hydroxyapatite
trend even further from the whole diet, with rats on experimentally controlled diets
showing an enrichment of +9.6%o (DeNiro & Epstein 1978b) and other mammals
displaying enrichments of+12%o to +13%o (Lee-Thorp et al. 1989). Additionally,
preferential uptake among different tissues within the same organism has been noted and
can further complicate matters, with animal muscle generally showing 613C values 3%o to
4%o less positive (-3%o to -4%o) compared to bone collagen (Schoeninger 1989).
Oxygen undergoes fractionation due to environmental factors such as evaporation,
condensation, and freezing and is also strongly influenced by temperature and humidity
(Stille & Shields 1997, lacumin 1996, Hertz & Garrison 1998, Kendall & Coplen 2001).
This leads to differential isotope incorporation in plant tissues and is reflected in the
differing values of herbivores thought to be a result of foraging habits. For instance,
oxygen isotope ratios were found to vary by as much as 8%o to 9%o in herbivores based
on whether they were browsers or grazers (lacumin et al. 1996).
A difference of approximately 9%o has also been measured between the carbonate
and phosphate fractions of bone and teeth from a variety of mammals (lacumin et al.
1996) as well as marine invertebrate shells (Longinelli & Nuti 1973). This consistent
enrichment of carbonate 6180 values, regardless of the animal, seems fairly constant as
long as temperature remains within the range of 00 C to 370 C. Outside of this
temperature range, the fractionation is not as predictable (lacumin et al. 1996).
One reason strontium and lead analyses appear so attractive is the general
consensus that these elements do not undergo fractionation in nature. Strontium and lead
do not appear to exhibit this trend due to their significantly large atomic masses (Stille &
Shields 1997) versus the lighter isotopes such as carbon and oxygen. Such being the
case, comparisons can be drawn then utilizing organisms from different trophic levels as
well as between different tissues, without having to employ conversion factors.
Frequently Sampled Human Tissues
A variety of human tissues have proved useful in isotopic studies within the
anthropological disciplines in recent years. Tissues primarily available to forensic
anthropologists include bone, teeth, hair, desiccated skin, and finger/toenails; each
presenting its own benefits and drawbacks potential isotopic use and preserving records
of residency and diet at different points of the individual's life.
Bone is arguably the most utilized tissue in archaeological isotope studies
(Schwarcz & Schoeninger 1991). It is a composed of three primary constituents: 1)
water; 2) an inorganic mineral fraction (hydroxyapatite); and 3) an organic matrix
(Schwarcz & Schoeninger 1991). Bone isotope studies utilize both hydroxyapatite
(apatite) and collagen, which is found in the organic phase. Dry bone is composed of
approximately 70% inorganics and 30% organic (Katzenberg 2000). The overwhelming
majority of the inorganic phase is comprised of the protein collagen (85% to 90%)
Bone has a turnover rate of between 10-30 years (Ambrose 1993) owing to the fact
that different bone components remodel at different rates. On average, trabecular bone
remodels much more rapidly than its denser cortical counterpart (Teitelbaum 2000).
Regardless of the speed of turnover, it is clear that bone delta values slowly change
throughout an individual's life as stable isotopes are constantly incorporated into this
continually remodeled tissue.
Apatite is a calcium phosphate product of which the carbonate portion arises from
dissolved carbon dioxide (C02) in the blood plasma. Fractionation does occur between
these two reservoirs with the bone carbonate portion 613C value enriched by
approximately 12%o over plasma CO2 (DeNiro and Epstein 1978b). Bone carbonate
therefore reflects the total metabolic carbon pool found in an individual's diet,
incorporating carbon equally from all dietary energy sources and representing the
isotopic signature of the whole diet (DeNiro and Epstein 1978b, Ambrose & Norr 1993).
Collagen is the most abundant protein in the body (Champe & Harvey 1987),
constituting roughly one-quarter of all proteins occurring in mammals (Stryer 1975). The
collagen found in bone, dentin, skin, and tendon is molecularly similar and falls under the
category of Type I collagen (Schwarcz & Schoeninger 1991). Controlled experiments
using rats demonstrated that collagen underestimates the non-protein component of the
diet, but is an excellent representative of the protein portion because of its heavy nitrogen
constituent (Ambrose & Norr 1993, Tieszen & Fagre 1993). One difficulty with bone
collagen is that is does degrade over time, much more so than apatite.
Teeth are especially useful in isotopic studies because of their robustness and
ability to survive in environs where bone would normally degrade. Unlike bone, tooth
enamel tends to be highly inert in terms of mineral exchange with the environment (Price
et al. 2002, Lee-Thorp & Sponheimer 2003), consequently they represent small, closed
systems. Because enamel is non-cellular and heavily mineralized with 96% or greater of
the weight of the enamel comprised of the inorganic constituent (Hillson 1996), it
withstands the effects of diagenesis very well and long preserves an accurate biogenic
isotopic signal (Lee-Thorp & Sponheimer 2003). Dentin and cement, on the other hand,
are much heavier in organic (roughly 20% and 25%, respectively) (Hillson 1996) and
much more susceptible to contamination.
Additionally, the inorganic nature of enamel, and specifically the apatite, reflects
the whole diet of the individual while the collagen in dentin, because of its high nitrogen
content, primarily mirrors the protein content of the diet (van der Merwe 1982, Harrison
and Katzenberg 2003). This is one of the drawbacks of using enamel. You cannot
analyze nitrogen isotopes.
Moreover, since teeth are genetically conservative, there is little variation in the
development and specifically, the period of mineralization of the tooth, although females
are slightly precocious in terms of dental formation, completing most stages of dental
growth before males (Fanning & Brown 1971, Hillson 1996). This observation was
confirmed by the 1976 study by Anderson et al. of the mineralization in permanent
dentition, although the authors state the degree of variability between the sexes has been
reported to be similar. Their calculations of the mean age of attainment of mineralization
in the adult teeth are presented below in Table 1-2.
Table 1-2. Mean age of completion of permanent crown mineralization.
1st 2nd 1st 2nd 1st 2nd 3rd
Incisor Incisor Canine Premolar Premolar Molar Molar Molar
Maxillary 3.70.28 4.00.48 4.90.53 5.81.0 6.30.65 3.80.30 6.70.72 13.31.58
Mandibular 3.6+0.21 4.00.46 4.80.59 5.61.21 6.30.70 3.70.14 6.70.71 13.31.51
Maxillary 3.60.14 3.80.40 4.1+0.49 5.10.56 5.90.65 no data 6.30.66 12.71.49
Mandibular 3.60.20 3.70.28 4.1+0.49 5.00.54 5.90.74 no data 6.30.66 12.81.63
Source: Anderson et al. (1976)
With in- and outflow of materials ceasing once amelogenesis is complete,
examining the permanent enamel provides a snapshot of the nutritional ecology of that
individual during the period of crown mineralization for that specific tooth. Dentin
primarily is laid down during and after amelogenesis, thus the bulk of it is formed during
childhood. Secondary dentin lines the pulp chamber and has a slow, continued formation
during adulthood, with turnover rates similar to bone (Hillson 1996).
Sampling can be done in bulk, which will average the isotope value for the entire
tissue component, or serially. In serial analyses, very specific regions of the enamel or
dentin, corresponding to even finer time periods, are sampled and compared. This takes
much greater skill in drilling and one must be sure what point in the individual's life the
area represents, but this method can also allow even finer resolution of dietary studies
over a period of years.
Several studies have turned to hair as an alternative sampling tissue (van der
Merwe et al. 1993, Yoshingaga et al. 1996, O'Connell & Hedges 1999, White et al. 1999,
Bonnichsen et al. 2001, Ayliffe et al. 2004, Cryan et al. 2004, West et al. 2004, Roy et al.
2005). Hair holds a great untapped potential in forensic isotope work. Often, hair masses
are found in association with skeletal remains. It is extremely durable, proving insoluble
to a variety of fluids, and can remain intact for thousands of years (Bonnichsen et al.
2001). The shaft is sheathed in a cuticle, a hard protective covering that is resistant to
chemical and microbial insult (Lubec et al. 1987, Macko et al. 1999b).
Because of its hardiness and the fact that the average human sheds 50-100 hairs a
day (Macko et al. 1999b), sampling is easy and essentially non-invasive. Non-keratinous
material, such as the root (bulb) is not normally sampled (Ayliffe et al. 2004) because of
its signature is not reflective of the shaft. Hair is also readily renewable, growing roughly
1 cm/month in humans (Yoshinaga et al. 1996), with isotope shifts demonstrating about
an 8-day delay (in the case of beard hair) from change of diet to hair exposure from the
follicle (Sharp et al. 2003). The isotopic composition of hair offers information
concerning an individual's diet and recent geolocational background. Thus a section of
hair provides a snapshot of an individual's nutritional ecology at a particular point in time
and a chronological record of the same along its length (White et al. 1999, Roy et al.
Hair is easier to isotopically analyze than bone and only very small samples (much
less than bone) are required (Cryan et al. 2004, Roy et al. 2005). Hair is made of
approximately 95% keratin, the proteinacious component (Taylor et al. 1995). Because
of its high protein content, hair requires minimal chemical processing and no chemical
purification. Conversely, bone and dentin collagen must be chemically extracted and
purified before the protein fraction can be analyzed (Ambrose 1993).
Roy et al. (2005) produced consistent carbon and nitrogen isotope values with
human hair specimen weights as low as 100 rig, corresponding to a length of 2 cm of
hair. The authors expect that strand lengths as small as 5 mm could be analyzed without
significant loss of precision when determining 613C alone, as nitrogen was the limiting
factor in their study.
The 613C values of hair keratin correlate well with that of total dietary protein, with
keratin being enriched by +1%o to +4.8%o relative to protein in the diet and depleted by
-2%o to -3%o compared to bone collagen (DeNiro & Epstein 1978b, Ambrose & Norr
1993, Tieszen & Fagre 1993, Yoshinaga 1996) in lab animals and contemporary humans.
Carbon isotope signatures from hair keratin and bone collagen are related but cannot be
directly equated (O'Connell & Hedges 1999).
Fingernails and Toenails
Like hair, finger- and toenails also offer a non-invasive way to examine isotopic
values in both the living and dead. The keratin composition of human nail material
makes them an excellent source of collagen values, as well as a variety of elemental
isotopes. They also provide a recent geolocational reference for a specific individual
with a whole nail representing approximately 6 months of growth in adults (note: authors
did not state the length of the nail) (Fraser et al. 2006) and 2 to 3 months growth from
cuticle to fingertip in infants (Fuller et al. 2006a). While some state that hair and
fingernail values are "similar" to each other (O'Connel et al. 2001, Fuller et al. 2006a), it
has been noted that nails are depleted in 13C and 180 compared to hair by mean values of
and -0.55%o and -1.6%o, respectively (Fraser et al. 2006).
Fogel et al. (1989, 1997) published the details of a landmark study examining the
weaning of modern infants as reflected in the differences in 615N between mothers and
infants. Fingernails prove an excellent medium for studying diets in modem infants
because they are metabolically inert, resistant to degradation, and have such a fast
synthesis rate (Fuller et al. 2006a). What the authors found was that the isotopic values
of the babies' fingernails were enriched in 15N by approximately +3%o from that of their
mothers, indicating that the infants were feeding at a higher trophic level than their
mothers. This was confirmed by Fuller et al. (2006a), who found infant 613C values
enriched by +1%o over their mothers' and 15N enrichment of +1.7%o to +2.8%o compared
to maternal values (for more detail, see Chapter 2). Such conclusions have been
extrapolated to bone and tooth isotopic analyses of weaning practices of archeological
populations (Schurr 1997, Herring et al. 1998, Schurr 1998, Wright & Schwarcz 1998,
Wright & Schwarcz 1999, Dupras et al. 2001, Mays et al. 2002, Clayton et al. 2006,
Fuller et al. 2006b).
Often desiccated skin is adherent to bone on remains submitted for forensic
analysis. This skin serves as an additional potential reservoir for isotopic values and can
be relatively easily removed from associated bone. Carbon turnover rates for skin and
hair are much faster than for bone, giving these tissues the ability to confer information
regarding diet and provenance much closer to death than hard tissues. Skin has an
estimated carbon turnover rate of roughly 15 days (Tieszen et al. 1983). The integrity of
skin after the decomposition process takes hold however is suspect, as skin appears
highly susceptible to contamination (White et al. 1999). For those studies in which viable
skin samples were obtained, relative to hair sample 613C values, skin appears to be
consistently depleted from -0.2%o to -2.7%o (White & Schwarcz 1994, White et al. 1999)
Radosevich (1993) aptly states that a reason for uncritical acceptance of methods or
assumptions is often the simple desire for a new technique to work. Stable isotope
analyses have seemingly been hailed as near-omniscient and people may turn a blind eye
to the limitations of such methods. On the other hand, modeling biological systems is an
extremely complex undertaking. There are times when a reductionist approach can
overwhelm the model in minutiae; where accounting for all the potentials of error
eclipses the actual data. All factors with the potential for confounding the data need to be
explored and understood, but often a relative weight can be assigned to them so the
model is not overloaded. There are much potential for error in stable isotope analyses;
but, as long as they are recognized apriori and dealt with, isotope ratios can provide
valuable insight into past and present systems.
After the initial glow wore off following the popularization of stable isotope
techniques, researchers began to find chinks in the analytical armor. Often confusing or
contrary results were obtained leaving researchers to scratch their heads as to what it all
meant and if isotope studies were really worth all the hype. In 1981, A. Sillen was one of
the first to propose that perhaps post-depositional contamination, or diagenesis, was
responsible for at least a portion of this noise, but the effects of diagenesis were largely
ignored or dismissed in most studies (Price et al. 1992).
Diagenesis is a subset of the study of the postmortem processes which can affect
bone appearance and integrity, commonly known as taphonomy (literally meaning the
"laws of burial;" Sandford 1992). These processes take both physical and chemical
forms. When diagenesis in an anthropological context is discussed, it is in reference to
the postmortem alterations in the chemical constituents and physical properties of bone
following deposition in soil. Diagenesis takes the form of both contamination and
leaching and arises from several different mechanisms (Sandford 1992).
The dense mineralization of enamel affords teeth a great measure of protection
against effects, but it is important to keep in mind that no skeletal element is impervious
to postmortem modification. The porous structure of bone however, makes it susceptible
to infiltration by foreign elements, especially when it has been physically degraded. The
intrinsic skeletal chemistry and microstructure of osseous tissue therefore leads to a
dynamic relationship between it and the environment in which it is interred (Sandford
Mary Sandford (1992) lists several different means by which the environment
interacts with the structure of bone, leading to alteration:
Elements may be precipitated as discrete "void-filling" mineral phases in
the small cracks and pores of bone.
Soluble ions present in soils may be exchanged for those that normally
occupy lattice positions in bone hydroxyapatite.
Bone apatite can "seed" formation of recrystallization through a variety of
Microorganisms break down bone collagen releasing elements through its
dissolution and the action of acid metabolites on hydroxyapatite.
Additional extrinsic factors such as the chemical environment of the burial sight
and the properties of the enveloping sediment influence the incidence and rate of
processes as well. Soil pH is one of the most important variables that affect change in
bone. Gordon and Buikstra (1981) first quantified the relationship, determining that it is
strongly negatively correlated, thus as soil pH decreases, degradation of bone increases.
The authors also noted that skeletal age was significant as well, with juvenile bone being
more susceptible to decay.
Temperature, microorganismal activity, groundwater, and precipitation also play a
role, as does the local geochemical environment to include soil texture, mineralogy, and
organic content. Sandford (1992) also mentions further intrinsic factors bearing on
processes such as bone density, size, microstructure, and biochemistry.
Recent investigations have shed light on bone alteration leading to several
generalizations: 1) elements differ in their susceptibility to diagenesis; 2) certain
categories of bones are more susceptible to diagenesis--less bone density, greater
porosity, or large quantities of amorphous material may predispose certain classes of
bones, such as immature bone, to taphonomic processes; 3) denser cortical bone
withstands diagenesis much better than the lattice-like trabecullar bone; 4) Direction and
intensity of change is not necessarily temporally or spatially uniform (Sandford 1992); 5)
the color and condition of skeletal material can be used as a general indicator of the
degree of diagenesis (Carlson 1996). The more the color approximates the color of fresh
bone, the less likely it is to have undergone change.
The majority of changes seen in bone arise due to precipitation of authigenic
carbonate or other minerals, exchange reactions in original carbonate or phosphate, and
uptake or loss of various trace elements. Recrystallization can also occur, producing
various phosphate-containing compounds with trace levels of elements often replacing
calcium at higher concentrations than found in modern bone (Schoeninger et al. 2003).
The same processes that bring about diagenetic change are ones that will eventually
return bones the lithosphere. The overwhelming majority of all deposited skeletal
material disappears relatively quickly, especially if exposed to taphonomic factors such
as acidic soil, alternate wet/dry conditions, strong solar radiation, and/or injurious
invasion by microorganisms (Lee-Thorp 2002). If we as anthropologists are fortunate to
encounter remains in the first place, we should not be discouraged from utilizing isotopic
resources in attempting to uncover clues about the lifestyle of the individualss. We must
keep in mind that these processes are not uniform over space and time, and thus even old
remains can produce valuable results.
Questions still remain however, as to what measures can be taken to minimize the
impact of diagenesis on isotopic interpretation. So what is a researcher to do? The first
step is to attempt to determine if processes have occurred and to what extent. In reality,
these processes are always occurring, but whether they exact a measurable effect upon
bone is another question. To begin with, a scientist should ask themselves several
questions. The first is what is/are the elements) of interest? Studies show that isotopes
of such elements as strontium and lead are little changed in bone due to diagenetic means
(Beard and Johnson 2000, Carlson 2002), thus scientists should have greater latitude in
using bones that have been interred for any period of time. Do the bones belong to an
adult or child? Because smaller bones have greater surface area to volume ratios, they
are more susceptible to change since there is more surface area for processes to act upon.
The absolute volume of cortical bone is reduced in juveniles as well, as they are still
growing, so bones are less shielded from environmental assailants. What bones are
available for sampling? Remains higher in cortical bone preserve better, so if presented
with a few cranial vault fragments, a researcher may be wise to opt out of isotope
analysis versus if a femoral shaft is available. Also, intact bone is always preferable to
One should also assess the environment the bones are interred in. Sandford (1992)
believes chemical analysis of soil is a mandatory requirement for gaining insight as to the
condition of bone. Soil samples should be recovered from feature fill in direct
association with bone (Gordon and Buikstra 1981). Samples can be prepared and pH
determined in situ utilizing a portable pH meter. These values can then be applied to
something similar to a regional variant of Gordon & Buikstra's (1981) regression
formulae for pH and state of preservation. (It is interesting that while the authors provide
several regression formulae, for example, in adult assemblages, preservation = -1.3(pH)
+12.5, there is no scale provided in which to interpret the preservation value.)
Further testing can compare total elemental concentrations of bone and associated
soil. Following the assumptions of the concentration gradient theory, significant
contamination of bone by soil is considerably less likely if soil concentrations of a
specific element are disproportionately different than those same elements in bone. If a
more homogenous elemental state has been reached between bone and soil, it is a good
indication that significant change has transpired (Sandford 1992, Carlson 1996).
Other factors such as temperature and exposure to water should also be accounted
for. It is well accepted that higher temperature leads to degradation of collagen and that
warm, moist habitats encourage microbial proliferation. Exposure to water can also lead
to increased rates of both contamination and leaching of minerals into the surrounding
soil. The best environs for the preservation of DNA are those that are cool, dark, and dry
(Smith 2005). That is because these same conditions optimize the resilience of the whole
bone complex, so isotopic fractions will be best preserved as well. Heavy bone erosion,
trauma, burning, associated human alterations such as boiling and internment/funeral
practices, and carnivore and rodent activity compromise the structural integrity of the
bone itself leaving it more vulnerable to processes.
Instrumental analyses can be completed as well to include electron microprobes
and x-ray diffraction (Sandford 1992), and backscatter scanning electron microscopy
(Collins et al. 2002). These methods attempt to look at the structure of bone and analyze
it for changes in crystalline architecture, chemical constituency, and microbial activity.
Analyses of collagen content of bone may also provide insight, since some have observed
low yield in collagen is often associated with aberrant stable isotope readings
Osteological comparisons can also be completed in conjunction with soil analyses
examining constancy in values (Sandford 1992). Intrabone comparisons look for
statistically significant correlations between elements and known contaminants or
"indicator elements." Interbone comparisons look for agreement with the assumption
that different types of bone, such as ribs and femora, should reflect varying degrees of
diagenesis. Interspecies comparisons can indicate activity when measured elemental
values vary from those predicted on the basis of dietary patterns. Additionally, if
interpopulational data were available as we are attempting to collect, congruency to
published values could be ascertained (Sandford 1992).
Further precautions are essential during sample preparation in the laboratory.
Standard protocols attempt to minimize effects of diagenesis through mechanical
abrasion, to physically remove contaminants from outer bone surfaces, and acid washing.
None of the aforementioned methods are fail safe, but their use enhances overall
understanding of the processes active in a certain area and attempts to circumvent
diagenetic effects by careful sampling selection and preparation.
Many subscribe to the notion that the longer a set of remains has been interred, the
greater the alteration to the material. It is unwise to use temporal criterion in isolation in
making a decision about employing isotopic analyses though. As in any scientific
situation, you must take measure of as many variables as possible in order to make the
most informed decision. Diagenesis is a complex mechanism and time is but one factor
that comes into play. Cases in the literature abound detailing the successful extraction of
viable isotopic material from fossils that would have proved opportunities lost if the
authors had decided against isotopic analyses simply because they were working with
very old material. Studies have examined diets in ancient, human mummies (White &
Schwarcz 1994, White et al. 1999) and Neolithic Icemen (Macko et al 1999a, 1999b;
Miller et al. 2003), and the diet and paleoecology ofAustralopithecus africanus (van der
Merwe et al. 2003) and 5 million-year-old horses (MacFadden et al. 1999b), to cite but a
few. Differential preservation is a rule, rather than an exception and thus each interment
must be individually assessed for the appropriateness of isotopic analyses.
All organisms alter their environment. Human beings are unique though, in that we
are the only species on the planet that is actually altering the basic conditions of life on
Earth (Vitousek et al. 1997). We have altered landscapes, climate, and biogeochemical
cycles. Many of the wastes generated by our industrial metabolism play no useful role in
nature, cannot be recycled (i.e., nuclear waste) or overwhelm the current processing
capabilities of the biosphere (McMichael 2001). The ecological footprint of the human
species is enormous. Everything we do leaves traces of our kind behind.
This anthropogenic effect extends to isotopic signal variation. Industrial pollution
is implicated in the changing of isotopic values when contemporary populations are
compared to paleological assemblages and can outright alter or mask the isotopic
signatures a researcher is attempting to interpret. This can complicate analyses and lead
to false conclusions if not identified. To account for this, several correction factors have
been established to ease temporal analyses. Because nearly all of the anthropological
work done with stable isotopes has been in bioarchaeological contexts, these corrections
are essential in drawing conclusions.
The carbon isotope ecology of terrestrial systems is controlled by atmospheric
carbon dioxide (van der Merwe et al. 2000). This has changed dramatically in the years
since the Industrial Revolution, with fossil fuel emissions altering the 13C/12C ratio of the
atmosphere by -1.5%o in the last 150 years. (van Klinken et al. 2000). To correct for this
change in 613C values, "Industrial Effect" (van der Merwe et al. 2000) or "fossil fuel
effect" (van Klinken et al. 2000) calibrations must be factored into results, normally by
adding 1.5%o to convert modern samples to pre-industrial values (van Klinken et al.
We also have significantly altered the lead content of certain environs. Budd et al.
(2000) state, "It is widely believed that the contamination of the atmosphere by
anthropogenic lead has led to far greater human exposure today than that which prevailed
in the distant past, but this has proved difficult to quantify." A marked increase in the
mobilization of lead in Europe and North America occurred after industrialization.
Drilling of Greenland ice-cores has revealed a ten-fold rise in lead concentration, with
rates skyrocketing from roughly 10 parts per billion (ppb) to 100 ppb in the last 100 years
(McMichael 2001). This is due primarily to environmental contamination due to the use
of leaded gasoline (which is still utilized in many nations), lead-based pigments and
compounds, lead-acid batteries, and through mining operations, soldering, and coal
combustion (Sangster et al. 2000).
Today's global economy has the potential to homogenize biogeochemical
signatures in contemporary people. Because of world-wide trade, especially when it
comes to food importation, what people eat may not necessarily reflect where they came
from. Strontium values are especially vulnerable to being washed out by the effects of
the global food market. Archaeological research does not usually concern itself with
such matters because food tended to be locally grown and consumed. After the Industrial
Revolution and the establishment of global trade networks, food in the U.S. was very
rarely grown in the localities were people lived. So, on a trip to the refrigerator one may
find bananas from Guatemala, grapes from Chile, and free range, grass-fed beef from
Increasing consumption of bottled water from non-local sources further
complicates matters, affecting not only strontium values, but oxygen and hydrogen as
well. This situation may be further complicated by the importation of fertilizer produced
in foreign countries (Price et al. 2002). Such soil additives will affect not only plant
intake but run-off will affect, and may significantly change, the isotopic values of
groundwater (Bohlke & Horan 2000).
APPLICATIONS OF STABLE ISOTOPE ANALYSES
Examples of the varied usages of isotopes in the literature abound. Stable isotope
analyses are an extremely effective means of recreating paleoecology (e.g., Amundson et
al. 1997, Cerling et al. 1997), tracking animal movements (e.g., Burton and Koch, 1999
Rubenstein & Hobson 2004), assessing migratory patterns of humans (e.g., Beard and
Johnson 2000, Dupras and Schwarcz 2001), and determining diet (e.g., DeNiro & Epstein
1978, van der Merwe 1982, MacFadden et al. 1999b). Within anthropology, stable
isotope analyses have been primarily relegated to realm of archaeology, but by applying
the technologies currently used in geology, paleontological and modern zoology, and
archaeology to forensic science, an effective means for presumptive identification
One exciting application of stable isotopes that transcends disciplinary bounds is
that of tracing studies. In a tracing study, an element is introduced into a system with a
known delta value and tracked through the system or at the termination of certain
processes to see how that element normally moves through the system. This approach is
frequently used in clinical nutrition studies to understand the uptake of various nutrients
(see Abrams 1999 for a review). Stable isotopes offer many benefits over more
traditional radioactive approaches in that they present little of a safety concern for
pregnant women or children and are less difficult and less expensive to remove than
radioactive wastes (Abrams 1999).
For instance, isotopic tracer studies were used to measure the efficiency of zinc
utilization at different doses. Patients were given labeled zinc solutions, and then urine
samples were collected to determine absorption rates. Based on this approach the study
concluded aqueous zinc doses greater than 20mg resulted in quite small and diminishing
increases in absorptivity (Tran et al. 2004). Magnesium tracer studies demonstrated that
absorption of isotopically labeled magnesium could be accurately monitored through
urine sampling versus more invasive blood and fecal sampling methods (Sabatier 2003).
Additionally, tracer studies in Nigerian children with rickets determined that those with
the disease did not express impaired abilities to absorb calcium when compared to
healthy counterparts, although fractional calcium absorption did increase after resolution
of the active disease (Graff et al. 2004.) Stable isotopes were even utilized to measure
calcium metabolism of two cosmonauts and one astronaut aboard the Mir space station
prior to, during, and after a 3-month spaceflight (Smith et al. 1999). Further non-human
trials utilized three diets of different isotopic compositions to determine the turnover time
of carbon isotopes in horse tail hair West et al. 2004) and tail hair and breath CO2
(Ayliffe et al. 2004). These baseline studies could then be applied to other wildlife
studies in an attempt to understand the dietary history of mammals
Ecological studies have also utilized isotope tracers to examine nutrient flow in
various systems. One recent study added isotopically-labeled nitrogen to a creek for 6
weeks and monitored 1N in dissolved, aquatic, and terrestrial riparian food web
components. High levels of incorporation of the tracer into the tissues of resident
organisms led researchers to believe that streams within undisturbed primary forests may
be highly efficient at uptake and retention of nitrogen (Ashkenas 2004). Another project
examined root turnover in relation to forest net primary production by fumigating a stand
with labeled 13C in the form of 13CO2 over a 5-year period, then sampling fine roots.
Their results suggest that root production and turnover in forests have likely been
overestimated and that sequestration of anthropogenic atmospheric carbon in forest soils
may be lower than currently believed (Matamala et al. 2003).
Fractionation studies have proven quite illustrative in a variety of genres as well.
Examination of carbon and nitrogen stable isotopes has yielded greater understanding of
the decompositional processes found within soil organic matter (Kramer et al. 2003).
Fractionation studies have also proven useful in attempting to measure the contribution of
gluconeogenesis to glucose production in humans. Here, body water was enriched with
2H20 and the ratio of 2H bound to carbon-5 versus carbon-2 of blood glucose was
measured (Katanik et al. 2003). Oxygen isotope fractionation has also been employed in
niche separation studies of African rain forest primates occupying overlapping
microhabitats. Oxygen isotope ratios from bone carbonate were positively correlated
with relative dependence of leaves in the diet, a fact obscured by carbon isotope analyses
(Carter 2003). A final study led to the discovery of what is commonly known as the
"canopy effect" (van der Merwe &Medina 1991). Van der Merwe and Medina
discovered the re-use of plant-fractionated, respired CO2 in dense vegetation can cause
systematic bias between plant and animal species living on the forest floor versus those
living in the forest canopy and open environments (also in van Klinken et al. 2000). Due
to the "canopy effect," the 613C value of atmospheric CO2 is lowest near the forest floor.
"Leaves fixing this 13C-depleted CO2 have lower 613C values than those higher up in the
canopy. Combined with the effects of low light intensity, high humidity and high CO2
concentrations on water use efficiency, this creates a vertical dine in leaf 13C values"
Zoology and Ecology
Within zoology and ecology, the examples of stable isotope use seem limitless.
One of the first such studies examined carbon ratios of two sympatric fossil hyrax
species, determining one a was browser, based on the C3-like signature these animals
displayed, while the other was chiefly a grazer, feeding on tropical grasses, which utilized
a C4 photosynthetic pathway (DeNiro & Epstein 1978a). Similar studies have shed new
light on the diet and ecology of 5-million-year-old horses (MacFadden et al. 1999b) and
Cenozoic sirenians from Florida (MacFadden et al. 2004). Stable isotopes have proven
especially insightful for scientists attempting to determine feeding strategies of marine
organisms, and in fact, "Most work on mammal and reptile movements using stable
isotopes has been done in the marine environment" (Rubenstein & Hobson 2004).
Carbon isotopes have been used to determine food sources for Red Sea barnacles
(Achituv et al. 1997) and examine photosymbiosis in fossil mollusks (Jones et al. 1988).
Delta 13C and 615N were useful in assessing not only the foraging strategies of Pacific
pinnipeds (Burton & Koch 1999), but tracking their migratory movements as well and
have been used in dietary studies of North Atlantic bottlenose dolphins (Walker et al.
1999). Moreover, adult female loggerhead turtles were sampled from around Japan to
determine the relationship between body size and feeding habitats (Hatase et al. 2002).
Claws (Bearhop et al. 2003) and feathers (Rubenstein et al. 2002, Bowen et al.
2005) have also been utilized to determine diets and habitat use of migratory birds whose
summering and wintering grounds are separated by thousands of kilometers; so too has
hair been examined in bats for evidence of seasonal molt and long-distance migration
(Cryan et al. 2004). Wing membranes from monarch butterflies have been sampled for
hydrogen and carbon isotopes to identify natal regions within the United States and
revealed that 13 discrete wintering colonies in Mexico were fairly well mixed as to the
origins of the individuals (Wassenaar & Hobson 1998). Biogeochemical fingerprints of
African elephant bone have been assessed to determine change in diet and habitat use
(Koch et al. 1995). Isotopic analyses have even been extended to determining the
allocation of reproductive resources in butterflies (O'Brien et al. 2004) and assessing
prey quality in predatory spiders (Oelbermann & Scheu 2002).
Stable isotope ratios also allow us a glimpse into the past. Based on 613C enamel
values of worldwide fossil mammals and modern endemic and zoo-housed African
mammals, Cerling et al. (1997) has postulated that between 8 and 6 million years ago,
there was a global shift to increased C4 plant biomass and a corresponding decrease in
atmospheric carbon dioxide. This has implications today as increasing levels of
atmospheric carbon dioxide could bring about a major biotic alteration towards a world
dominated by C3 plants, which would have widespread ecological consequences. Carbon
values have also been analyzed from ancient pollen in attempts to reconstruct
paleovegetational and paleoclimatic conditions with the hope of someday tracing the
origin of the C4 photosynthetic pathway (Amundson et al. 1997).
One further study has taken an innovative approach to tying stable isotopes to
hominin evolution. Wynn (2004) examined paleosols of Turkana Basin, Kenya, and
ascertained that modern hominins evolved during a period of waxing and waning
diversity of savanna-adapted fauna in an environment that trended towards increasing
aridity. Those hominins best suited to generalization of resources were the most capable
of surviving through evolutionary "pruning events" as savanna ecosystems changed
Mention of these zoological and ecological studies does not even scratch the
surface as to the diversity of isotopic studies that have been and are continuing to be
conducted within these disciplines. Isotope use in these fields is gaining momentum and
results generally enjoy widespread acceptance, ensuring their continued use far into the
Within anthropology, stable isotope analyses have primarily been relegated to the
realm of archaeology. Here, they have been used extensively to answer a litany of
questions in a variety of contexts concerning the human experience.
Introduction of maize
In archaeological contexts, stable isotopes have been used extensively to infer diet,
mobility patterns, and origins of material culture. When considering diet, a great amount
of effort has been expended attempting to determine when exactly maize became
prominent dietary component in various human populations (Vogel & van der Merwe
1977, van der Merwe & Vogel 1978, DeNiro & Epstein 1978b, Farnsworth et al. 1985,
Norr 1995). In fact, the majority of early archaeological stable isotope studies were
aimed at resolving the temporal and geographic origins of maize introduction, especially
into North America (Schwarcz & Schoeninger 1991). Striking changes in 613C values of
collagen resulted from the introduction of maize into human dietary patterns. These
values markedly decreased from roughly -21.4%o to -12.0%o during the period of A.D.
1000-1200 indicating that proportion of carbon from C4 plants went from 0 to more than
70% in some individuals (van der Merwe & Vogel 1978, van der Merwe 1982). This
agrarian shift also had other implications, with the development of permanent settlements
and an abandonment of the hunter/gatherer life strategy and all associated changes
inherent in the transition to a sedentary lifestyle. Not all agree on the timing of the
introduction of maize to North America though, with individuals such as Farnsworth et
al. (1985) concluding that maize was incorporated into human diets much earlier than
indicated in the fossil record. Today, the consensus seems to be that there is a temporal
variation in the conversion to maize agriculture within North America (Schwarcz &
Schoeninger 1991). Age effects in a prehistoric maize horticultural population (Ontario
Iroquois) have also been examined, with significantly higher 613C values found in infants
and young children suggesting a weaning diet high in maize (Katzenberg et al. 1993).
Isotopic dietary studies have been applied to fossils as old as Australopithecus
africanus, where individuals demonstrated an unusually varied diet, a large portion of
which was C4-based (Sponheimer & Lee-Thorp 1999a, van der Merwe et al. 2003).
From this information, the authors speculate that by about 3 million years ago, hominins
had become savanna foragers for a significant part of their diet. Based on carbon and
nitrogen stable isotope values, the diet of a Neolithic Alpine "Ice Man" was determined
to likely be primarily vegetarian, at least in the period closest to his death, based upon
hair values (Macko et al. 1999a).
Stable isotope dietary studies have also been performed on individuals from other
Mesolithic/Neolithic sites (Krigbaum 2003, Richards et al. 2003, Milner et al. 2004), the
Bronze Age in Northern Jordan (Al-Shorman 2004), prehistoric Chile (Macko et al.
1999b), preclassic and historic Mayan Belize (White & Schwarcz 1989, Tykot et al.
1996), ancient Egypt (Macko et al. 1999b, While et al. 1999) and Sudan (White &
Schwarcz 1994), prehistoric South Africa (Sealy et al. 1992, Lee-Thorp et al. 1993), and
prehistoric Micronesia (Ambrose et al. 1997). In addition, indigenous Easter Islanders
(Fogel et al. 1997), additional Native North American groups throughout time (Price et
al. 1985, Larsen et al. 1992, Fogel et al. 1997, Hedman et al. 2002, Roy et al. 2005,
Yerkes 2005), and colonists from the Chesapeake area (Ubelaker & Owsley 2003) have
also been examined.
One significant aspect of diet that has received much recent attention is that of
infant feeding. Breastfeeding practices, to include weaning, have wide implications for
population dynamics in earlier human groups (Mays et al. 2002). Breastfeeding is a
major determinant of fecundity and interval between births in societies lacking reliable
artificial contraceptive measures, (Vitzthum 1994, as cited in Mays et al. 2002) and thus,
can be a major factor in determining life histories of certain population groups.
Ultimately, the success of infant feeding will have far-reaching impacts in terms of
population health and growth, for it is the essential first step for realizing adulthood.
Bone chemistry has been critical in this area for archaeological interpretation of
remains. In 1989, Fogel et al. published a groundbreaking study comparing the
fingernails of mothers and newborns from birth through weaning, to determine the utility
of using isotopes for such analyses. Fetuses and newborns have a 615N roughly
equivalent to that of their mothers (Herring et al. 1998, Mays 2000). This makes sense
because a fetus receives nutrition through materials exchanged across the maternal and
fetal circulatory flows in the placenta. Once born and breastfeeding begins, neonates
change their trophic stratigraphy. They effectively become carnivores relative to their
mother (Ambrose 1993). Nursing infants are feeding at one trophic level above their
lactating mothers and hence, should show an enriched 615N level of +2%o to +4%o over
their mothers (Fogel et al. 1989, Fogel et al. 1997). This is exactly what Fogel et al.
found. During the period of breastfeeding, they measured infant 15N ratios approximately
+3%o higher than their mothers (Fogel et al. 1989, Fogel et al. 1997). These results were
further confirmed by Fuller et al. (2006a), who found infant 615N values enriched by
+1.7%o to +2.8%o compared to maternal values. As a child is weaned from its mother's
breast, its 615N level will begin to fall back towards a standard adult average, because the
shift from milk proteins to proteins obtained from solid foods registers as a decrease in
15N bone collagen values (Wright and Schwarcz 1998, Fuller et al. 2006b).
Since Fogel et al. (1989), numerous researchers have applied their findings to
various archeological assemblages ranging from mid-Holocene South Africa (Clayton et
al. 2006), the Roman period of Egypt (27 BC to AD 395) (Dupras et al. 2001), Mediaeval
England (Mays et al. 2002), pre-contact North America (Schurr 1997) to 19th century
Ontario (Herring et al. 1998). Wright and Schwarcz (1998, 1999) have taken a slightly
different slant by including oxygen isotopes in their analyses of prehistoric Guatemalans.
Their studies are based on the fact that, "Human breast milk is formed from the body
water pool and, thus, is heavier in 6180 than the water imbibed by a lactating mother"
(Wright & Schwarcz 1998). Infants who only breastfeed are enriched in their oxygen
ratios compared to their mothers, because of the mothers' metabolic processing of the
water incorporated into breast milk (Wright & Schwarcz 1998, Wright & Schwarcz
1999). Additionally, many studies have incorporated carbon delta values with the
standard nitrogen values to determine the approximate ages of supplementary food
consumption by children in their respective populations, providing further validation of
the conclusions drawn from nitrogen data. (Wright & Schwarcz 1998, Clayton et al.
2005, Fuller et al. 2006b)
Following in the footsteps of Fogel et al. (1989), many more contemporary studies
have been carried out in the hopes of applying the results to archaeological work.
O'Connell et al. (2001) compared pairings of hair keratin and bone collagen taken from
patients undergoing orthopedic surgery in the United Kingdom as well as pairings of nail
and hair keratin from living subjects to examine the utility of applying similar 613C and
615N results to archaeological work. Lead isotopes in modem people have also been
examined to determine comparative lead loads and digenetic effects in prehistoric teeth
(Budd et al 1998, Budd et al. 2000), with Budd et al. (2000) concluding that Neolithic
human enamel lead values were only an order of magnitude lower than modern juveniles.
Region of Origin
Paleodiet analyses have also been applied to detect human mobility since the mid-
1980s (Sealy & van der Merwe 1985, 1986). Such studies are predicated upon the notion
that individuals practicing seasonal migration from coastal to inland areas should have
similar 613C values, while those permanently inhabiting such diverse areas should
demonstrate distinct carbon ratios (Sealy & van der Merwe 1985, 1986). In addition to
carbon, there are a wide variety of isotopes that can be drawn on to infer information
concerning human migration. Schwarcz et al. (1991) were the first to demonstrate the
use of bone phosphate 6180 to in attempting to identify the geographical origin of 28
soldiers from the War of 1812, interred in the Snake Hill cemetery, New York. Their
findings of uniformity among 6180 values indicated the group all spent a major portion of
their lives living in the same geographical area. These values differed however from
oxygen isotope analyses performed on interments in southwestern Ontario and Antietam,
Maryland. Dupras and Schwarcz (2001) used oxygen isotopes to distinguish immigrants
from native peoples from a third-century cemetery in the Dakhleh Oasis, Egypt, and both
oxygen and strontium isotopes have been used to determine the geographic origin of
remains found from Viking occupation-era graves in Great Britain (Budd et al. 2004).
Strontium isotope ratios have been used extensively in transhumance studies from
Neolithic Europe (Grupe 1997, Budd et al. 2000, Bentley et al. 2002, Bentley et al. 2003,
Muller et al. 2003, Bentley et al. 2004) as well as Bronze Age and Romano-British sites
(Budd et al. 2000a, Montgomery et al. 2005, Fuller et al. 2006b), and prehistoric and
historic South Africa (Sealy et al. 1995). Two studies used strontium isotopes to
discriminate between immigrants and life-long residents of 14th century Grasshopper
Pueblo, Arizona (Price et al. 1994, Beard & Johnson 2000). Beard and Johnson (2000)
determined local strontium values by analyzing local field mice. Individuals outside of
this range were deemed immigrants to the area, with those having the greatest 687Sr
differences being the most recent additions to that area. Aberg et al. (1998) demonstrated
that strontium and lead isotopes could definitively distinguish between west coastal and
rural inhabitants of Medieval Norway. The authors further concluded that Medieval
residents subsisted on local products while contemporary people relied on imported or
industrially processed food to a greater degree.
Carlson (1996) discovered that lead isotope values corresponding to different
sources of anthropogenic and natural lead can indicate cultural affinity among Native
Americans and fur traders buried in a 19th century fur trade cemetery. Montgomery et al.
(2005) found lead to be a bit ambiguous, with results suggesting that lead isotopes
provide dissimilar types of information depending on what era is being examined. In
some instances it seemed to serve as a geographical marker, while in others it served
better as a cultural indicator.
Not all archaeological applications of isotopic signatures are anatomically-based.
The origins of various forms of material culture have also been traced using these
techniques and well as additional dietary analyses. The earliest attempt to determine
provenance through isotope use was attempted in 1965 by Robert Brill and colleagues on
lead and glass artifacts (Brill & Wampler 1967, Herz & Garrison 1998). Not only were
lead objects associated with specific mining regions in antiquity, but samples separated
by nearly a millennium in time were found to have virtually identical lead isotopic
signatures and are believed to have come from the same mine (Brill & Wampler 1967).
The source quarries of ancient marbles have been interpreted through 613C and 6180
values (Craig & Craig 1972) and today, an extensive database exists for the isotope
values of principle classical quarries so marble items can now often be associated with
the areas in which they originated (Herz & Garrison 1998). Oxygen values have traced
emerald trade routes from the Gallo-Roman period through the 18th century (Giuliani et
al. 2000) and the mining locations of lead artifacts, such as musket balls and coils, found
among Omaha Native Americans have been identified (Reinhard & Ghazi 1992).
Building materials such as the timbers for the prehistoric great houses of Chaco
Canyon, New Mexico, have been traced to their individual mountain growing areas
(English et al. 2001). Major constituents of prehistoric and historic diet have also been
accomplished through the analysis of cooking residues found on potsherds or within
intact kitchenware (Hastorf& DeNiro 1985, DeNiro 1987, Hart et al. 2003). When
carbon and nitrogen analyses are combined for proven plant encrustations, they can
distinguish among three plant groupings: 1) legumes; 2) non-leguminous C3 plants; and
C4 or CAM vegetation (Hastorf & DeNiro 1985).
Stable isotope analyses have been applied to a wide variety of contexts within the
forensic sciences. Within Europe two major organizations have emerged to advance the
development and application of isotopic work in this field. The Forensic Isotope Ratio
Mass Spectrometry (FIRMS) network and the Natural Isotopes and Trace Elements in
Criminalistics and Environmental Forensics (NITECRIME) European Union Thematic
Network both aim to raise awareness of the benefits of isotopes to forensic investigations,
encourage collaboration, and develop and validate new methodologies (Benson et al.
Stable isotopes have shown great promise as an analytical asset in the war on drugs,
specifically in determining the origin of illicit narcotics. The 62H (also denoted as 6D),
613C, 615N of components extracted from 3,4-methylenedioxymethylamphetamine or
"ecstasy" have shown that individual tablets can be traced back to a common batch
(Carter et al. 2002). Carbon and nitrogen isotopes have been further used to link heroin
and cocaine samples to the four major geographic regions in which they are grown
(Mexico, Southwest Asia, Southeast Asia, and South America). Morphine, which is
derived from heroin, demonstrated the most pronounced regional difference (Ehleringer
et al. 1999). Further studies were able to determine the country of origin in 90% of 200
coca-leaf samples, the source material for cocaine, as deriving from Bolivia, Columbia,
or Peru (Ehleringer et al. 2000).
Isotopic techniques have been used by food and spirit regulatory agencies as well to
ensure quality control. There is an international concern with not only simple validation
of food label claims, but with food adulteration as well. One application is within the
beer industry (Brooks et al. 2002). The primary ingredients in beer are water, malted
barley, hops, and yeast. All other "non-essential" ingredients are called adjuncts. In
many nations, the use of unlabelled adjuncts is forbidden by law. Carbon delta values
have proven very effective at detecting adjuncts and testing brewers' claims as to the
purity of their ingredients (Brooks et al. 2002). Additionally, 613C values have proven
invaluable in determining whether forms of glycerol are animal or vegetal in origin
(Fronza et al. 1998) and 613C and 615N values of eggs have been used to establish
whether chickens were given animal or plant protein as feed (Rossmann 2001).
Furthermore, oxygen values have been utilized to verify the regional origin of dairy
products, especially certain cheeses, which must be produced from milk of a particular
region (Rossmann 2001).
Food adulteration is of concern to authorities because it is essentially the
misrepresentation of an altered foodstuff as an authentic product. Here, a premium food
product is extended or completely replaced with cheaper materials, yet fraudulently sold
as a higher-end item (Parker et al. 1998). Stable isotope analyses have established
themselves as a particularly usefully analytical methodology in fighting this trend. The
most advanced applications of stable isotope analyses are within wine quality control,
where the European Union has established an official wine stable isotope parameter
database (Rossmann 2001). Carbon isotopes can detect the addition of exogenous
glycerol deceptively added to wine to disguise poor quality (Calderone et al. 2004). They
have also been used to differentiate between whiskies and assist in authenticating specific
whisky products (Parker et al. 1998) and determine the botanical origin of Brazilian
brandies (Pissinatto et al. 1999). Isotopic fractionation of hydrogen and oxygen resulting
from juice concentration processes have also been documented and utilized to quantify
added sugars in orange and grape juice (Yunianta et al. 1995); while 613C values have
been used for over 20 years to control for the authenticity of honey (Rossmann 2001).
Stable isotope analysis has also been employed by criminal investigators in cases
involving the use of firearms (Stupian et al. 2001). Bullet individualization via lead
isotope analysis was first reported in 1975 (Stupian 1975). Lead isotopic information can
indicate whether a fatal bullet shared a common origin with a box of ammunition
collected from a suspect or provide a detective with a tool independent of standard
ballistic methods to potentially link bullets from multiple crime scenes (Stupian et al.
2001). In instances where there is a shoot-out with several types of firearms and/or
ammunition, it may even be possible to conclude which bullet and/or weapon caused a
particular gunshot entry (Zeichner et al. 2006).
Another forensic isotope breakthrough occurred in 1975, when Nissenbaum
reported 613C could distinguish between trinitrotoluene (TNT) samples originating from
different countries. Other areas of forensic isotope applications include connecting the
sources of automobile (Deconinck et al. 2006) and architectural paints (Reidy et al.
2005), packaging tapes (Carter et al. 2004), and glass fragments (Trejos et al. 2003) to
Similar measures have been drawn upon to detect environmental toxins in soils,
waters, and plants. Isotopes can assist in identifying a geographical relationship between
a source and a spilled product, whether the contamination might be from an oil spill,
illegal dumping, pipeline breaks, or leaking storage tanks (Philip et al. 2003). For
instance, in the case of a crime scene, such applications may be able to link engine oil on
the victim of a hit and run with a particular vehicle (Philip et al. 2003). Source
identification of environmental perchlorate contamination has been performed with
chlorine and oxygen isotopes (Bohlke et al. 2005). Perchlorate, in even small amounts,
can adversely affect thyroid function by interfering with iodine uptake (Bohlke et al.
2005), but hopefully, by identifying the source of such chemicals, this form of pollution
can be stemmed.
These techniques have further been extended to biowarfare defense efforts. Horita
and Vass (2003) determined that cultured bacteria (Bacillus globigii and Erwinia
aglomerans) faithfully inherit the isotopic signature of hydrogen, carbon, and nitrogen
from the media waters and substrates they were grown on, proving "stable-isotope
fingerprint" can be created for chemical and biological agents. Because of these
properties, Kreuzer-Martin et al. (2003) were able to undertake sophisticated tracing
studies involving oxygen and hydrogen isotopes. Culture media was prepared with water
spiked with known isotopic quantities of hydrogen and oxygen. The 6180 and 6D found
within strains ofBacilus subtilis spores grown on this media were then traced back to
specific water sources establishing that the origin of microbes can be pinpointed to
particular areas based on the water content of the media on which they are grown.
Stable isotopes are also prominent in wildlife forensic issues. Several studies have
used a trivariate approach, combining carbon, nitrogen, and strontium isotope ratios to
create geolocational fingerprints for elephant ivory and bone (van der Merwe et al. 1990,
Vogel et al. 1990). It is hoped this will aid in conservation efforts by assisting in efforts
to stem the illegal trade of ivory. Similar goals are also being applied to the bounty of
information concerning animal migrations (Bowen et al. 2005).
While great advances have been made in the applications of stable isotope analyses
to the forensic sciences, human stable isotope studies in the medico-legal realm are
relatively recent phenomena. To date, very few studies examining stable isotope ratios as
they pertain to region of origin in contemporary human populations have been presented
or published. When examining the literature, it appears that the bulk of isotopic research
in modern humans is in the form of isotopic tracers for nutritional studies (see also
Abrams and Wong 2003, Mellon and Sandstrom 1996). Many, as previously discussed,
are also used as proxies for archaeological comparison (Fogel et al. 1989, O'Connell et
al. 2001, Fuller et al. 2006a).
Several studies have been conducted to investigate lead exposure and identify the
sources of lead absorbed in contemporary, living children by examining their deciduous
teeth (Alexander and Heaven 1993, Gulson & Wilson 1994, Gulson 1996) and other
tissues and excretions (i.e., blood and urine, Angle et al. 1995). Alexander and Heaven
(1993) measured 206Pb/207Pb ratios and lead abundance in teeth finding significant
difference among the lead isotope ratios. When compared against various environmental
sources of lead, the authors were able to identify differences in sources in northwest
England. While these studies were not utilized for geolocational purposes, they
nonetheless could be applied as such, (although anthropological studies tend to utilize
isotopes compared to 204Pb), and provide a good example of the multiple uses for isotope
One weaning study went one step further than those previously discussed and has
exciting forensic potential. Fuller et al. (2006) analyzed bovine milk-based and soy-
based formulas to determine if unique isotopic signatures exist that could identify infants
being fed different forms of supplementation. The authors purchased seven different
formulas sold within California and found that while the 613C values overlapped between
formulas derived from cow's milk and soy, the soy products demonstrated significantly
lower 615N values. This again, is a reflection of trophic level effects in nitrogen values.
Fraser et al. (2006) have begun compiling a database of modern human hair and
nail values examining the stable isotopes of hydrogen, carbon, nitrogen and oxygen. The
authors sampled hair and fingernails from 20 individuals living in Belfast, Northern
Ireland for a minimum of 6 months as well as an additional 70 individuals from 9
countries representing 4 European nations, Syria, the United States, Australia, India, and
Sudan. They did not report having yet applied the database results to a forensic situation,
but preliminary data is at least at the ready should the need arise. Similarly, at the 3rd
European Academy of Forensic Science Meeting in Istanbul, Turkey, Cerling et al.
(2003) presented results of a multi-element study of modem human hair. The authors
discovered regional differences in the 6D, 613C, 615N, and 6180 values of long-time
residents of particular locations and appear to still be collecting samples.
Beard and Johnson (2000) were the first to demonstrate the utility of strontium
isotopes in a human forensic setting. In their paper, they determined the region of origin
of an illegally harvested deer using the 87Sr/86Sr ratio of antler, then also applied this
information in an attempt to differentiate between the teeth of three commingled
Americans associated with the Vietnam conflict. They were able to match the natal area
of one individual, but the two others presented overlapping values. If the study had also
utilized alternative isotope comparisons, perhaps the authors might have been able to
discern between the remaining two individuals.
Also, preliminary data for a study using strontium isotope values in an attempt to
determine the geolocational fingerprints for Mexican-borne individuals residing in the
U.S. was presented at the 2005 annual meeting of the American Academy of Forensic
Sciences (Juarez 2005). Several bay-area dental clinics provided the author with 25
permanent lst molars of individuals originating from four different Mexican states.
Samples were accompanied by information as to the subjects' regions of origin within
Mexico, their ages, and sex. Initial results indicate four specific ranges of strontium
isotope ratios, one for each of the four states involved in the study. Within-state variation
proved too great however, to discriminate location further.
Additionally, a presentation at the 2001 annual meeting of the American
Association of Physical Anthropologists addressed the use of strontium isotopes and its
applications in forensic science (Schutkowski et al. 2001). The abstract makes reference
to the presentation of a multi-regional sample demonstrating differences in regional and
local strontium isotope ratios. Bone and tooth signatures were examined to determine if
mismatches of individual values with local isotope ratios demonstrate changes in
domicile. The areas of study were likely western European, as the authors at the time of
publication practiced in the United Kingdom and Germany. Unfortunately, it appears this
data has yet to be published in a western source.
Gulson et al. (1997) detail a pilot study comparing the lead isotope values in teeth
of native Australians to those of Australian migrants from Eastern and Southern Europe
(Table 2-1). While the actual data presented by Gulson et al. are not particularly useful in
cases of American service members this paper does indicate lead isotope ratios have the
potential to discriminate region of origin.
As can be seen from this short review, isotopic analyses and applications serve a
wide variety of functions. The incredible inferential value of isotopic analyses in
anthropology is clear. Examples of the power of isotopic studies abound in the literature
and continued advances will only further solidify how essential their inclusion is within
an anthropologist's analytical toolbox.
Table 2-1. Mean and standard deviations for selected groups of immigrant teeth
Australia CIS* Yugoslavia Lebanon Poland
(n=29) (n=14) (n=13) (n=8) (n=6)
Mean 206Pb/204Pb 16.56 17.98 18.23 17.62 18.07
SD 0.17 0.06 0.15 0.29 0.20
Mean 207Pb/206Pb 0.9318 0.8664 0.8566 0.8825 0.8617
SD 0.0088 0.0033 0.0063 0.0136 0.0088
Source: Gulson et al. (1997)
*CIS denotes the former Soviet Union
HUMAN FORENSIC IDENTIFICATION
Assuming that isotopic analyses do prove fruitful for forensic practitioners, this
technique will be added to a bounty of available measures for use in the personal
identification process. Those specializing in the forensic arts acknowledge that there is
stratification when it comes to the probative value of identification data. In attempting to
tease a name from a body, certain characteristics of the person will be much more unique
and individualizing than others. The most powerful measure of identification is a
positive identification, the essential component of which is the possession by the
decedent of unique characteristics (Ubelaker 2000). Because these characteristics are not
replicated in anyone else, they exclude all other individuals from consideration.
Even with the high resolution of DNA, the method of choice today for positive
identifications tends to be dental comparisons (Col. Brion Smith, personal
communication). Dental records are still consulted when available. The Computer-
Assisted Postmortem Identification system (CAPMI) is based on the presence of dental
restorations and has increased the efficiency of matching and comparing
antemortem/postmortem records (Friedman et al. 1989), especially in the case of mass
fatalities. Dental radiographic matches are much quicker and less costly than DNA
evaluations, although the number of individuals with no dental anomalies (Friedman et al.
1989, Col. Brion Smith, personal communication) is rising due to advances in dental
hygiene and medicine and mass fluorination of community water sources. In those cases
where the skin of the fingers is still intact, fingerprints may establish a positive
identification as well.
When these measures prove inconclusive, genetic fingerprinting utilizing nuclear
and mitochondrial DNA is another option. With the development of the polymerase
chain reaction procedure, which enabled rapid amplification of genetic material, and
lowered costs, DNA analysis is much more practical (Herrero 2003) than in days past.
Nuclear DNA is known to be a unique identifier (unless the subjects are identical twins).
Many investigators, including the Department of Defense (DoD), test 16 bands from the
available microsatellite loci pool (Col. Brion Smith, personal communication). From
experimental observations, the average odds that one band will be shared by any two
unrelated individuals is approximately 0.25 (Sudbery 2002). So the resolution of a 16-
band testing procedure is 0.2516 = 2.33 x 10-10; that is, there is a 0.000000000233 chance
that two unrelated individuals will share all 16 bands tested. Put another way, if you take
the reciprocal of this figure you see that there is a 1 in nearly 4.3 trillion chance that
someone unrelated has the same DNA profile. Since this number is considerably larger
than the world's population, nuclear DNA testing is said to provide for unique
This calculation is made with the assumptions that all individuals are unrelated and
that the chance that bands will be shared is the same for all people. In truth, people are
related and ethnic affinities may lead to higher rates of band sharing than among the
general world populace. Even so, after accounting for such complications, nuclear DNA
analyses are still considered positive and unambiguous identification (Sudbery 2002).
The resolution of mtDNA, on the other hand, is not as fine. Because mtDNA is
passed through maternal lineages only, recombination does not occur. Mutations aside,
this accounts for the integrity of mtDNA as it is passed from mother to child. This
constancy of code allows for familial tracing by comparing sequences of certain base pair
lengths among those who are maternally related. This is a very powerful tool indeed, and
allows for a distinctive discriminating function from nuclear DNA. The downside to it
though is that is cannot distinguish among relatives and can be preserved for generations,
leading to populations of people with the same or similar mtDNA profile (Col. Brion
Smith, personal communication).
Many also consider various forms of radiographic comparison to equate to positive
identification. This is especially true with the frontal sinus. The sinus becomes
radiographically visible between 7 and 9 years of age, and barring trauma or disease,
remains relatively unchanged throughout life (Ubelaker 1999). In a comparison
completed by Ubelaker (1999), the author noted than in a radiographic comparison of 35
radiographs (595 comparisons), no two frontal sinuses were alike. The number of
differences between individuals average to approximately 8, with a range of 3 to 15. If
additional antemortem radiographs exist documenting unique skeletal anomalies (i.e.,
pathology or trauma), these characteristics may also serve as a basis for positive ID.
One further skeletal anomaly for consideration is that of prosthetic devices (Burns
1999). While it may not be unique that an individual has a total knee replacement, what
will be unique is the serial number that is imprinted upon the prosthetic device along with
the manufacturer's emblem. Hospitals must document these serial numbers. With a little
detective work, the serial or lot numbers can be traced back to the manufacturer who in
turn, can direct an investigator to the hospital to which the device was sold (Warren
2003). Some also consider the comparison of still photos and the skull at the same angle,
or what is known as video superimposition, to be conclusive as well (Ubelaker 1999).
Hope for a positive ID can often prove frustrating and futile though, when there are
no reference samples on file for that individual. An individual must have antemortem
information available to compare against if a positive identification is to be achieved. So,
for instance, while DNA may have successfully been extracted from a set of remains, an
identification cannot be accomplished when there is no nuclear DNA on file or source
material available and no relatives of maternal lineage for the decedent can be located.
One step below a positive ID is exclusionary evidence for identification. When
remains are presented for identification, they will arrive from one of two environs, either
an open environment, or one which is closed (Warren 2003). Open environments are
those in which the person laying before you could be anyone in the world who was up
until recently, alive. For example, a body found in the woods could be an indigent, a
local, or a tourist from another country. In the case of a light aircraft crash however, the
potential for identification is much higher. If a passenger manifest was filed listing two
adult males and child of 12, and assuming it was correct, then there is the potential for an
exclusionary identification. The child will be easy to distinguish from the adults due to
developmental differences in the skeleton. If one of the adults is identified via
antemortem radiographs and the other has no antemortem comparison data, then the latter
would usually be identified by exclusionary methods, since ideally, in a closed system
there is no one else it could be.
Burns (1999) also lists identification by means of a preponderance of evidence.
This is often linked with tentative identifications, or what are also known as presumptive
identifications. There is much greater uncertainty with presumptive identifications
because they are based on evidence found associated with the body, such as personal
effects, and/or verbal testimony of witnesses, last known whereabouts of the body, and
familial recollections of undocumented conditions the individual may have suffered from
(Burns 1999). See Table 3-1 for a recap of identification measures.
Table 3-1. Forms of forensic identification.
Type of I.D. Basis for I.D.
Tentative identification Clothing
Location of body
Identification by preponderance of Anomalies known by family or friends, but
evidence without the existence of written records
Identification by exclusion "Everyone else is identified and there is no
evidence that this is not the only person
Positive identification Dental identification
Unique skeletal anomalies
Reproduced from Burns (1999)
Military Identification Measures
"Over the past 200 years, the United States has set the standard for the
identification and return of its servicemembers [sic] to their families" (AFIP 2004).
Since as early as the American Revolution, efforts have been made to recover, identify,
and provide individual burial for American military personnel (AFIP 2004). As the years
have progressed, standards and expectations for identification have increased and the
technology with which to do it has made sweeping advances.
The United States DoD employs all of the standard personnel identification
measures previously mentioned. What is unique about the military as a population
however, is that their physical attributes and markers are much better documented than
the general populace (i.e., they have much better antemortem records). Members have
fingerprints on file and flight crews have footprints documented as well. Meticulous
medical and dental records are fairly centralized. With few exceptions, blood cards are
on file for all current total force members in the case their DNA needs to be sequenced.
Individuating marks are noted such as scars, large birthmarks and moles, and tattoos as
well as information such as hair and eye color, race, stature, weight, and age.
Even so, such measures are not without their complications. Dental radiographs
are commonly not available of military members unaccounted for from previous
conflicts, especially World War II and the Korean War (Adams 2003a). The Office of
the Armed Forces Medical Examiner notes that greater than 5% of all service members
have no dental restorations, the primary means for dental identifications, and the number
is rising (AFIP 2004). In a study of 7030 living U.S. soldiers, it was revealed that 9%
had a full complement ofunrestored teeth (Friedman et al. 1989). In a pooled data set of
over 29,000 individuals from the Third national Health and Nutrition Examination
Survey (NHANES III) and the Tri-Service Comprehensive Oral Health Survey
(TSCOSO), Bradley Adams found that 12.77% had "perfect teeth" (2003b). Not only has
the number of dental restorations declined in younger individuals, but the complexity of
them has decreased as well (Friedman et al. 1989).
Historically, only 70% of service personnel actually have their fingerprints on file
with the Federal Bureau of Investigation with a further 15-30% of the fingerprint cards
submitted by the services rejected as "unclassifiable" (AFIP 2004). These numbers will
likely be reduced significantly though, with the wide-spread implementation of digital
fingerprinting DoD-wide, which instantly scans recorded images for acceptability
immediately after each individual print is taken. Additionally, radiographic analyses may
not be possible on highly fragmented remains (AFIP 2004). Identifications made based
on material evidence associated with remains can be very problematic as well.
Traditional items such as dog tags are not necessarily accurate either. As an example,
during current operations in Iraq and Afghanistan dog tags have been known to have been
blown off one individual and burned into the chest of another (Dr William Rodriguez,
On the leading edge of identification efforts for the U.S. government is the Armed
Forces DNA Identification Laboratory (AFDIL). AFDIL is the focal point for the DoD
in all matters concerning DNA identification efforts for military personnel and special
federal government projects. In addition to performing laboratory testing, AFDIL
manages the DoD DNA Registry. This function is responsible for maintaining blood
cards for DNA testing as well as providing administrative oversight of the database of all
sequenced data. Besides the Registry, AFDIL is also responsible for the DNA
Repository, which administers the AFDIL Family Reference Specimen database for
mtDNA matching when nuclear DNA is unavailable (AFIP 2004).
It is a common misconception that the military maintains DNA profiles on all its
personnel. It does not. Instead, AFDIL houses nearly 4.5 million blood stain cards for
active duty, reserve, guard component, retired military members and additional
specialized government personnel (Col. Brion Smith, personal communication).
According to Colonel Brion Smith, Chief Deputy Medical Examiner for the Forensic
DNA Division, Office of the Armed Forces Medical Examiner, there are two basic
reasoning behind the logic of this. The first is that it is cost prohibitive to perform DNA
analyses for every member of the armed forces. It is much less expensive to house blood
stain cards and generate the same information on an "as needed" basis. The second is
that storing the profiles of all who serve presents an ethical dilemma, especially when it
comes to who should be permitted access to the information and for what purposes. This
is further complicated by the fact that medical and dental records, to included DNA
information and blood cards, stay on file for 50 years after the service member retires
(Col. Brion Smith, personal communication).
An additional benefit of this system is it gives examiners the option aposteriori to
decide which test is best suited based on the conditions of the remains. If the body is in
good condition, nuclear DNA would be the preferred method. If the remains are charred
and disassociated, mtDNA might be most appropriate. Furthermore, a fully utilized
blood card can provide 30-40 punches, allowing less common tests such as Y short
tandem repeats to be completed (Col. Brion Smith, personal communication) or
providing the opportunity for future testing utilizing methods that have yet to be
developed or hit the mainstream. The beauty of this practice then is that technicians are
not restricted to only performing a form of analysis that matches the information present
in a data base so there is greater flexibility in analyses and hopefully the best method for
the materials available.
The utilization of both nuclear and mitochondrial DNA is dependent upon the
situation and essential to military identification. All individuals who die in current
combat, training, or in otherwise duty-related capacities are sampled for DNA analyses
upon intake to the Dover Air Force Base Port Mortuary (Col. Brion Smith, personal
communication), the DoD central receiving and processing center for all military
deceased. Even when other conventional methods of positive identification are available
such as radiographic dental comparisons, a DNA fingerprint will be generated. This will
delay returning a casualty to their families unless identity is questionable, but instead, is
performed to prevent questions surfacing at a later date as to correct identification and to
reassure family members that the body being returned to them is kin. (Col. Brion Smith,
This project was established to test the utility of stable isotope analyses for
identification of region of origin for modern, unidentified, human skeletal material that
has poorly-documented or unknown provenience. Initial efforts have focused on the
approximately 1,800 service members who remain unaccounted for from the Vietnam
conflict. It is hoped however, that this information will eventually be refined to use in the
identification of all those who remain unaccounted for and for those potentially
recoverable from previous conflicts (Table 3-2).
Often, the true national origin of remains recovered by the Joint Prisoner of
War/Missing in Action Accounting Command (JPAC) is uncertain. In addition, it is not
uncommon for de-contextualized, poorly preserved and/or highly fragmented remains to
be unilaterally turned over to the Central Identification Laboratory (CIL) by a foreign
agency. CIL personnel attempt to determine whether remains are U.S. service
Table 3-2. Numbers of unaccounted for U.S. prisoners of war and/or those missing in
Conflict Number Unaccounted For
World War II 78,000
(35,000 considered recoverable)
Korean Conflict 8,100
Vietnam War 1,800
Cold War 120
First Gulf War 1
Source: JPAC (2006)
personnel through a variety of means. The identification of unknown remains believed to
be missing U.S. service personnel is frequently hampered by high levels of degradation
and fragmentation as a result of circumstances of loss and subsequent taphonomic
regimes. These effects often combine to prevent effective DNA sampling strategies.
Teeth often prove excellent at distinguishing among the populations in questions. U.S.
military personnel had access to regular dental care. In countries such as Vietnam, this
was not the case for the majority of the population. In addition to untreated dental
insults, the occlusal surfaces of the molars and other teeth are frequently worn down from
the grit present in native diets exposing the underlying dentin (Mark Gleisner, personal
communication). The teeth then of modern Vietnamese often present similarly to
historical/prehistorical Native Americans. Every effort is also made to extract DNA from
a set of remains, although such efforts are often unsuccessful because of the poor state of
Additionally, the number of U.S. casualties during the Vietnam conflict of Asian
ancestry was relatively small. In 1985, the DoD reported the number of "Mongolian"
fatalities in Southeast Asia occurring from the period of 1 January 1961 to 30 April 1975
or as a result of injuries sustained in operations during said period was 114 or 0.002%.
Those listed of "Malayan" ancestry who died under the same circumstances was 253 or
0.004% (Reports 1985). See Table 3-3 for a complete listing of casualties by race.
More importantly for this study, only 5 servicemen of Asian ancestry remain
unaccounted for out of 1,760 total (JPAC 2006). The complete racial breakdown for
service members still listed as missing in Southeast Asia can be found in Table 3-4.
Besides military members, 32 American civilians are also listed as missing in Southeast
Asia. The racial backgrounds of these individuals were unavailable, but it is interesting
to note that two missing civilians are female. All of the military members unaccounted
for are male.
Because of the very low likelihood of a U.S. service member being a female or of
Asian ancestry, biological profiles can be useful in excluding individuals from
consideration. This is assuming enough of the skeleton remains to create a biological
profile. When the biological information is combined with documented information
Table 3-3. United States casualties in Southeast Asia by race.
Race (reported by DoD) Total U.S. Casualties
"American Indian" 226
Source: Reports (1985)
Table 3-4. United States military listed as unaccounted for in Southeast Asia by race.
Race (reported by JPAC) Total U.S. Military Missing
"Asian/Pacific Islander" 5
"American Indian/Alaska Native" 2
Source: JPAC (2006)
concerning troop engagement and staging areas and locations of downed aircraft, remains
may be returned to the originating nation if the evidence points overwhelmingly to the
fact that the remains are not of an American. Unfortunately, such an assessment is an
extremely complicated venture and in a great many cases it is simply impossible to make
such a distinction.
This project was initiated in the hopes that the results will assist in resolving this
dilemma. A two-pronged approach for this study has been utilized based on the
operating hypotheses that: 1) discernable differences exist between the isotopic ratios
incorporated into American and Southeast Asian tooth enamel and that these differences
can be used to determine region of origin; and 2) regional differences in natal isotopic
signatures are also discernable within populations raised within the U.S.
Because of the paucity of data in contemporary studies, it is near impossible to
predict the likelihood of the ability of this study to distinguish natal Vietnamese from
American-born individuals. It is encouraging that Juarez (2005) found significant
variation among the strontium isotope values for Mexican-born peoples from four
different states, even with her limited sample. If historical, human, migratory studies
(Montgomery et al. 2005, Miller et al. 2003, Montgomery et al. 2000, Dupras &
Schwarcz 2001, Aberg et al. 1998) are any indicator though, there is a high probability
that the chosen stable isotopes will be able to discriminate between these two
populations. The reasoning behind this is that the geochemical properties of different
continental systems should vary significantly and this difference will be further
magnified by the fairly culturally distinct dietary practices of the two populations.
None of the studies mentioned in Chapter 2 examined stable isotope use in a
forensic context in any great depth. The largest sample size was Gulson et al. (1997)
with 68, but it was a combined pool of permanent and deciduous teeth. This study will
utilize approximately 300-600 total samples and thus will have greater power.
Furthermore, all studies make mention of overlapping isotopic values which makes
discrimination virtually impossible. It is hoped this tendency will be reduced by
introducing multi-element analyses to forensic work. Theoretically, a multivariate
approach should allow finer resolution, especially since the deposition of the elements
depends largely on very different factors: carbon isotope ratios are based on cultural food
preferences; oxygen on meteoric water, altitude, and distance from major bodies of water;
and strontium and lead reflect the underlying bedrock and soil.
Carbon isotope ratios reflect the photosynthetic pathways of ingested plants and
echo cultural food preferences. It is expected that individuals who have subsisted on a
traditional, rice-based (C3 plant) Southeast Asian diet will differ significantly in their
carbon isotope signature from individuals who have subsisted on a heavier corn-and
sugar-based (C4 plants) American diet. Wild rice in the U.S. has produced results
ranging from -26.3%o to -29.7%o (Hart et al. 2003) and purified rice starch has been
averaged to -26.6%o (Ambrose & Norr 1993). This contrasts markedly to maize (corn)
values varying between -14.0%o (van der Merwe 1982) and -11.84%o (Hart et al. 2003)
and purified cane sugar at -11.2%o (Ambrose et al. 1997). Americans also eat a large
variety of wheat products. Wheat is a C3 plant, but it is enriched compared to rice, with
bread wheat leaves measuring -23.7%o (van der Merwe 1982 ). It stands to reason then
that those relying on a rice-based diet, such as the Vietnamese, would exhibit more
negative carbon isotope values than their American counterparts, whose corn and sugar
constituents of the diet, will shift the carbon isotope values in a less negative direction.
Due to the fractionation effects highlighted in Chapter 1, one must keep in mind
that the reported values will not trend directly with plant values. Mammal hydroxyapatite
will demonstrate an enrichment of +9.6% to +13% (DeNiro & Epstein 1978b, Lee-Thorp
et al. 1989) over plant material. Mixtures of the dietary plant constituent will also affect
an organism's overall 613C value as well as dependency on marine food resources
(Schoeninger & DeNiro 1984).
Since the majority of state borders within the continental U.S. are not based on
geomorphologic formations, it is unlikely that regional identification will be as
straightforward. This should be partially ameliorated through a multi-signature approach.
Due to the novelty of this approach, it is difficult to say with any certainty how precise
regional identification of geopolitical origin will become. Based on the limited success
of Beard and Johnson (2000) however, it appears natal origin within the U.S. can be
narrowed down to a regional level based on major geological formations.
Because 613C values represent dietary intake, they will not indicate regional origins,
since modern diets are primarily culturally based. The stable isotope ratios for oxygen,
strontium, and lead on the other hand, are aptly suited for this task. It is assumed that
individuals from Alaska, Hawaii, and the American territories will be identifiable. The
geographical distances between these areas and the continental United States (CONUS)
are vast, with a variety of different, but interrelated, environmental factors influencing
oxygen isotope distribution such as latitude, temperature, altitude, coastal affinity,
precipitation patterns, and humidity (lacumin 1996, Hertz & Garrison 1998, Kendall &
The geologic history of the major land masses nearly represents the 4.5 billion-year
history of the earth (Beard & Johnson 2000). Because of this, there are large differences
in the isotope compositions of different parts of the planet. relative to the analytical error
of the 87Sr/86Sr measurements (+ 0.00001 to +0.00003). Within the U.S., the ages of
crust varies from under 1 million years old in Hawaii to nearly 4 billion years old in areas
of Michigan and Minnesota (Beard & Johnson 2000). This age effect produces
significant variations in the strontium isotope composition within different regions of the
U.S. and is the basis for analytical techniques attempting to discern region of origin in
different peoples. Another strength of strontium is that its isotopes are thought to be little
influenced by fractionation (Toots & Voorhies 1965, Ambrose 1993, Carlson 1996, Hertz
& Garrison 1998, Beard & Johnson 2000, Budd et al. 2000) thus the isotope ratio remains
constant from soil to top carnivore as you move through the ecosystem. Soil samples can
then be checked against values to determine the geolocational origins of a tissue sample.
It is difficult to speculate whether strontium isotope analyses can identify natal
geolocation to the regional level in contemporary peoples. The analyses may seem fairly
straight forward on the surface, but there are underlying factors for modern man that may
inhibit its deductive power. Of primary concern is homogenization of strontium values
due to the global food trade.
An array of geological process are also responsible for the formation of these areas,
hence the bedrock composition is quite varied. Discerning among individuals reared
within the CONUS will likely prove more difficult, and overlapping values are expected.
By using the three different geologically-based isotopes in concert however, it is hoped
that general patterns will emerge.
In isolation, isotope delta values have limited evidentiary value and will rarely lead
to any form of identification. The same could be said of other bases of identification.
Clothing alone will not lead to a presumptive identification. Someone has to recognize
the clothing as belonging to the decedent before it has any realized significance. If the
geo-political region of origin for a set of remains could be ascertained however, it would
provide a direction in which to concentrate identification efforts. In mass disasters and in
closed environments, isotopes could be combined with other methods, leading to
exclusionary identifications or directing where to focus further analyses for potential
positive IDs. Such techniques are relatively inexpensive and quick. Isotopic analyses
can be performed for under $100 at the University of Florida and in the case of enamel,
can be completed in roughly 1 week's time. If stable isotope analyses are performed at
the onset of the identification process, it could save countless man-hours and dollars for
the military, preventing unnecessary analytical efforts if the remains are not deemed
MATERIALS AND METHODS
This study is groundbreaking in that it is the first of its kind to compile a reference
sample of isotopic values associated with known natal regions to be utilized in forensic
work. More importantly, the information gleaned from this study will be applied in
support of the Joint POW/MIA Accounting Command's mission to achieve the fullest
possible accounting of all Americans missing as a result of our nation's past conflicts.
A two-pronged approach for this project was utilized based on the operating
hypotheses that: 1) discernable differences exist between the isotopic ratios incorporated
into American and Southeast Asian tooth enamel and that these differences can be used to
determine region of origin; and 2) regional differences in natal isotopic signatures are
also discernable within populations raised within the U.S.
Teeth were utilized for this project because they are much more robust than bone
and little affected by diagenetic processes. This reduces the sample preparation time by
several days to a week. By only examining the enamel, isotopic values can be studies for
a known period of the subject's life, because in- and outflow of materials in enamel cease
at the termination of amelogenesis (Hillson 1996). It is also much easier to obtain
modern teeth than modem bone for sampling. When teeth are extracted, the standard
protocol is to dispose of them as biomedical waste, so there is little objection to obtaining
them for study. It is much more difficult to acquire samples of contemporary bone for
legal and cultural reasons. Objection is further fueled by the fact that isotopic sampling is
a destructive process.
Teeth are genetically conservative displaying little variation in the period of
mineralization of the tooth, although females are slightly precocious (Fanning & Brown
1971, Anderson et al. 1976, Hillson 1996), with Garn et al. reporting that females were in
advance of males by an average of 3% (1958). Different ethnic groups have shown
slightly different timing patterns as well, but all differences whether sex-related or ethnic,
equate to not much more than a few months between groups (Hillson 1996). This fact
should not impact this study however, as all crown mineralization is completed prior to
individuals being eligible for military service. All teeth supplied for this study had fully
Materials utilized in this study were supplied by three different institutions. The
Joint POW/MIA Accounting Command's Central Identification Laboratory (CIL),
Hickam Air Force Base, HI, permitted access to their "Mongoloid hold" collection for the
creation of an East Asian reference sample. The "Mongoloid hold" collection contains
remains of individuals recovered from East Asia or unilaterally turned over to the CIL,
whose governments have refused repatriation, once the remains were determined not to
belong to U.S. service personnel. Donated contemporary teeth and surveys completed by
their donors were also provided by the 10th Dental Squadron, United States Air Force
Academy (USAFA), Colorado Springs, CO and the Malcolm Randall Veterans Affairs
Medical Center (North Florida/South Georgia Veterans Health System) Dental Clinic,
henceforth referred to as the "VA," Gainesville, FL.
Prior to utilizing live subjects in this research, all appropriate permissions were
obtained and training completed (Appendix A). The USAFA Institutional Review Board
(IRB) granted IRB exempt status to this project (HQ USAFA IRB FAC2005026H). Prior
to conducting this study with the VA, the protocol was approved by the University of
Florida's Health Center Institutional Review Board (IRB-01 approval #474-2005), and
both the VA's Sub-Committee for Clinical Investigation and Research and Development
Committee. Additionally, a research template had to be created for incorporation into
each study participant's electronic medical record, via the VA's Computerized Patient
Record System (CPRS).
To further assist all parties engaged in this research, information binders were
distributed to both dental facilities. These packets included copies of all IRB and
committee approval letters, dental staff instructions, a subject identifier log, copies of all
required forms, a blank and completed, example survey, background information related
to this specific research project, pre-paid FedEx shipping forms (for USAFA), and a CD
with all electronic media on it (see Appendix A for a reproduction of the VA binder).
This project was essentially a piggy-back study attached to the normal patient
dental care of those individuals who are selected by USAFA and the VA for tooth
extraction(s) for valid medical reasons. The study, in and of itself, had no bearing on
whether an individual was selected for dental extraction(s). All patients scheduled for
dental extraction(s) during the study period were queried as to their willingness to
participate in the study. Complete inclusion of all consenting subjects cut down on bias
that would be introduced with nonrandom, arbitrary sampling by the dental staff. Dental
administration personnel proctored all forms. Upon receipt from the patient, they
reviewed the forms for completeness and verified birth date and sex with the subject's
Patients, to include Air Force Academy cadets, active-duty military, and military
retirees and/or veterans, already identified for tooth extraction for oral health reasons,
were asked to participate in a brief survey (Figure 4-1) and donate their extracted teeth
for analysis. The survey and, in the case of the VA subjects, associated combined Health
Insurance Portability and Accountability Act/informed consent form (Appendix A) were
administered upon initial intake while the patient filled out requisite preoperative
paperwork. This paperwork was in addition to the normal documentation required for
The HIPAA form was compulsory to protect participant health information.
Researchers must obtain patient authorization before they are allowed to disclose
protected health information. It was required in this instance because we were requesting
information such as location of residence and birth date, which cannot be ascertained
from observation alone. The informed consent form was required to secure subject
participation in the study. The form detailed the background, procedures, benefits and
risks of the research project, and obtained witnessed, signed consent of the individual that
they knowingly and voluntarily participated in the study. Dental staff were available to
answer any questions and an example of a completed questionnaire and study background
information (Appendix A) was made available. Survey completion and tooth donation
were the only requirements of subjects.
The data acquired from each subject included:
Date of birth
Tobacco product use
* Location of residence, birth to age 18
* Date of prior dental extraction at each facility (if applicable)
Thank you for your participation in this study. Its purpose is to provide a powerful new tool to assist in identifying our fallen
servicemen and women. The information you provide will be used to determine if geographic regions of the U.S. have specific
isotopic signatures that become incorporated into dental tissues. We will be looking at the mineral elements in your teeth. No
DNA analysis will be performed. When compared with isotopic signatures developed for geographic areas of Southeast Asia, it
is hoped the information will identify the origin of unknown remains recovered by JPAC's Central Identification Laboratory.
Additionally, the information gleaned from this study may prove useful in identifying remains recovered from further conflicts
such as World War II and encountered in mass disasters such as airliner crashes and the events of I 11 September 2001.
Instructions: Please fully answer all 8 questions. Incomplete data may exclude your teeth from the study. If you
have any questions, please ask your attending dental staff.
I) Date of birth (day/month/year)
2) Sex (circle one) Male Female
3) What race do you consider yourself?
4) Have you ever regularly used tobacco products (i.e. cigarette, chew, snuff)? Yes No
If yes, what products did/do you use, what was the time period, and what was the frequency (i.e. I pack a day)?
Tobacco Product Used From Used Until Frequency
5) Which of the following categories would you consider your childhood diet to the age of 18. Please circle only
one category unless you had a major diet shift. If so, please indicate the ages at which you followed each diet.
Meat Eater Vegetarian Vegan
6) What locations have you lived in, starting with birth and extending to age 18? Please be as specific as possible.
If you require more room, please use the back of this sheet.
(year) (year) City State Country
7) Please indicate the approximate location of each of the above areas on the attached map. For area (1) write a (,
area (2) write a 8, etc. If you lived outside of the U.S. for any portion of your childhood, please disregard for the
extent of your domicile outside of the U.S. Do include the numbers for any corresponding time lived in the U.S.
8) Have you undergone any prior dental extractions at this facility within the past year? (circle one) Yes No
If yes, please indicate the date to the best ofyour recollection. (day/month/year)
Approved UF-IRB-01 474-2005
Figure 4-1. Joint POW/MIA Accounting Command survey.
0atonaatasII o S[AESA IC I-
.. --.. CANADA
C ---- iMONTAA 1 NOR'H Oa- OTA
o M NNE5orA
SOUTH DKOTA CON S '..IN :. O~ I
I ',' lCOW I ....'--- "- i i
-, ... NEBRASKA *OH.',
LLNOIS IIN DLLANAI
S. 5.UCer A .
DKL HOMA J-
IA|R KANSAS . U -
NEW ~ EXIC.. .
pAL ~' GORGIA
TEXAS I ,
Figure 4-1. Continued.
The aim of the survey was to control for as many sources of error or variation as
possible in the data as well as create an isotopic mapping capability for natal region. All
questions, with the exception of the prior dental surgery question, were pertinent to the
study in that they account for factors that may possibly lead to differential maturation in
teeth or absorption/deposition of the various isotopes being studied.
The first item, date of birth, allowed for temporal comparison of specimens
between JPAC and VA samples versus USAFA samples. While dental development is
relatively genetically conservative, there is some minor variation in dental development
rates between sexes and major ethnic groupings. This information; date of birth, sex, and
self-perceived race; served as potential blocking factors during data analysis.
The effects tobacco use upon isotope analyses for teeth, have thus far not been
addressed in the literature. While enamel isotopic fates are locked in after amelogenesis
terminates, it is unknown whether tobacco use may trigger diagenetic changes within
teeth that may affect isotope values. It is commonly known to stain teeth, and may need
to be accounted for in preparation protocols and in interpretation of results.
Dates and locations of childhood residence were critical for making sense of the
oxygen, strontium, and lead stable isotope results. The validity of the residence
information was confirmed by individuals visually approximating these areas on a map.
It was also useful if someone could not remember the exact name of a town/city in which
they lived but did know approximately where it was in relation to neighboring areas.
The other questions served to control for potentially confounding variables. Dental
extraction history was necessary to prevent counting one individual who underwent
multiple extractions over multiple days as more than one subject. Survey questions were
limited to one page with the map encompassing a second page.
Pertinent information corresponding to each patient was also recorded by the dental
staff on each survey. Here, each facility assigned a unique subject identifier number to
each patient (i.e., VA-001). Additionally, the position in the arcade that each tooth came
from (tooth number) was noted according to the Universal/National System dental
numbering scheme for permanent dentition as well as the date of extraction.
Teeth were extracted following standard dental protocols for each facility. Care
was taken to preserve as much of the crown as possible. Each tooth was placed into its
own vial, which was labeled with the subject identifier number and tooth number, All
vials from a particular individual were then placed in a resealable bag and the bag stapled
to the associated survey. The surveys and teeth from USAFA were shipped via FedEx to
the C.A. Pound Human Identification Laboratory (CAPHIL). Surveys and teeth from the
VA were picked up weekly and CPRS updated by the author. Teeth provided by both
facilities were not stored in any solution or fixative.
Teeth were selected for sampling using the following hierarchy. Those teeth whose
cessation of amelogenesis was most similarly timed with the third molars were preferred,
with other teeth chosen on a decreasing sliding scale (Table 4-1). The younger in an
individual's life crown completion occurred for a specific tooth, the less desirable the
tooth was for sampling. Additionally, molars were preferable as they have the largest
surface area available for enamel removal. As a matter of course, mandibular teeth were
chosen over maxillary teeth and right over left. Furthermore, for the East Asian reference
samples from the CIL, teeth still present in the alveoli were selected over loose teeth,
Table 4-1. Crown formation/tooth eruption.
Crown Initiation' Crown Completion2 Tooth Eruption3
Tooth (upper/lower) in yrs (upper/lower) in yrs (upper/lower) in yrs
3rd molar 7.0-10.0/7.0-10.0 13.3/13.3 17-21/17-21
2nd molar 2.5-3.0/2.5-3.0 6.7/6.7 12-13/11-13
2nd premolar 2.0-2.5/2.0-2.5 6.3/6.3 10-12/11-12
Ist premolar 1.5-2.0/1.5-2.0 5.8/5.6 10-11/10-12
Canine 0.3-0.4/0.3-0.4 4.9/4.8 11-12/9-10
2nd incisor 0.8-1.0/0.25-0.3 4.0/4.0 8-9/7-8
1st incisor 0.25-0.3/0.25-0.3 3.7/3.6 7-8/6-7
1st molar 0.0/0.0 3.8/3.7 6-7/6-7
range in Shour & Massler (1940)
2 mean values in Anderson et al. (1976)
range in ADA (1999)
since the actual tooth number could be verified more easily with the presence of the
associated bone. In nearly all CIL cases however, a full arcade was not present to choose
from, thus the best option according to the aforementioned sampling scheme was selected
based on the resources available.
Each individual was assigned a unique identifier, not each tooth. Therefore, if two
teeth were utilized from the same individual, the samples would be given the same
identifier, with an additional tooth number identifier. This numbering scheme prevented
inflation of actual individual numbers. Furthermore, due to potential intertooth variation
in stable isotope values, enamel was not combined from multiple teeth to achieve the
desired weight of enamel powder.
Central Identification Laboratory
The CIL is an American Society of Crime Laboratory Directors (ASCLD)-certified
crime lab. As a result, all sampling conducted at the CIL conformed to their standard
operating procedures, to ensure compliance with ASCLD requirements. Prior to
sampling, potential specimens were researched utilizing a list of "Mongoloid holds"
provided by Dr. Andrew Tyrrell, the Casualty Automated Recovery and Identification
System (CARIS) database, and a thorough personal investigation of the entire evidence
storage area. Once suitable specimens had been identified, individual accessions were
checked out from the evidence manager and transferred to the CIL autopsy suite for the
Study identifiers with a "CIL" prefix and a three-digit suffix were associated with
lab accession numbers from the CIL Mongoloid hold collection. Two teeth, if available,
were selected from each accession following the above procedures. In all cases, at least
one intact and undisturbed tooth was left with the case in the event that future
identification efforts, such as DNA sequencing, were required. Each tooth selected for
sampling was assigned an additional sample number (01A or 02A) mirroring the lab's
DNA sampling procedures. Information cards for each tooth were created for photo
cataloging and provided the following information (Figure 4-2):
CIL accession number
Individual designator (if applicable)
Subject identifier number
From 14 June 2005 to 06 July 2005, a total of 112 teeth were sampled from 61
individuals believed to have originated from or been recovered from the following areas:
Vietnam (48 individuals); Cambodia (4 individuals); Laos (3 individuals); the Korean
Figure 4-2. Pre-drilling photo of CIL-033 #19 with data card. Note: the accession
number is purposely, partially obscured.
peninsula (3 individuals); the Solomon Islands (2 individuals); and the Philippines (1
individual). Teeth were eased out of their respective alveoli by hand or drilled out, when
necessary, using an NSK UM50 TM slow-speed dental drill with either a #2 or #4 carbide
dental drill bur, taking care to minimize damage to each alveolus. A photo, to include an
information card, reference scale (ruler), and empty collection vial was taken of each
tooth to document what each element looked like prior to drilling (Figure 4-2). Separate
photos of the buccal or lingual and occlusal surfaces were taken. (In cases where the
teeth had to be drilled out of their respective alveoli, pictures of the unaltered arcades or
portions thereof were taken using the same format.)
Each tooth was then placed into a vial of 3%, household-use hydrogen peroxide
and cleaned via a Branson Bransonic 2510 tabletop ultrasonic cleaner for 30 minutes.
When finished, teeth were removed from the solution and manually cleaned with a
toothbrush. The enamel surface of the teeth was prepared for drilling by cleaning off
excess calculus, soil, and/or staining using the same apparatus and a #8 carbide dental
Samples of approximately 100 mg of pristine enamel were drilled off of each tooth
using the same set-up (see Appendix B for drilling data). Care was taken not to drill into
the dentine. Enamel powder was collected on creased weighing paper and transferred to
labeled 1.5 mL microcentrifuge tubes. The drilled tooth, collection vial, scale (ruler), and
information card were again photographed to document the end-stage condition of the
tooth and for chain of custody purposes. The teeth were then returned to their original
storage bag along with the information cards with the associated elements for that
particular accession number. The bag was resealed with evidence tape, and the tape
initialed and dated on both sides. The remains were then turned in to the evidence
manager. Drill burs and weighing paper were discarded after each use and the drill
cleaned of adherent enamel powder.
Chain of custody forms were completed for all specimens, transferring possession
of the enamel powder and any associated enamel chips to the author (Appendix B). The
microcentrifuge tubes containing the enamel specimens were then transported from CIL
to CAPHIL through the services of FedEx.
The author also attempted to gain access to human teeth from native populations
while performing duty-related activities in Vietnam from July and into August 2005.
Such efforts were abandoned however, when provincial officials stated that regional and
higher government officials would be required to approve any request to procure human
United States Air Force Academy and Veterans Affairs
The Air Force Academy collected surveys and a total of 948 teeth from 274
individuals between late August 2005 and late April 2006 (Table 4-2; see also Appendix
C for a list of survey results). Of these, one third molar was selected from each
Table 4-2. Isotope sampling matrix.
# # Total # Individuals Run Total # Teeth Run Total # Runs
Source Inds Teeth C O Sr Pb C O Sr Pb C O Sr Pb
CIL 61 112 61 61 36 36 64 64 36 36 65 65 36 42
USAFA 274 948 230 230 36 36 238 238 36 36 279 279 36 36
Total 335 1060 291 291 72 72 302 302 72 72 344 344 72 78
of 228 individuals for inclusion in the primary study. One tooth each from two
individuals of unknown natal region were utilized for additional testing examining the
necessity of using acetic acid to process the enamel. Samples originating from three
different individuals were not used because of experimenter error in labeling the samples
and erroneous or missing information provided by the subject. Furthermore, after sample
AFA-185 from USAFA, samples were selectively chosen to fill in the geographic gaps
until optimally, each state had a minimum of five individuals represented. This approach
was chosen to reduce costs. Additionally, individuals from duplicate cities or those
people born prior to 1980 were sampled as well. Collection of specimens from the VA
began in mid-February 2006 and is ongoing. Unfortunately, because of the low number
of teeth provided by the facility and the poor condition of these teeth (i.e., little to no
enamel present) no samples were run for the current study. Sample collection is still
ongoing though, with the hope that the teeth can be used at a later date.
Upon receipt at CAPHIL, teeth were soaked in 3% hydrogen peroxide in their
original vials for 2 days. Teeth were then rinsed of the hydrogen peroxide with tap water
and scrubbed with a toothbrush to remove surface contaminants, such as blood. Any
adherent periodontal tissue or accessible neurovascular bundles were also removed.
Clean teeth were allowed to air dry overnight in a ventilation hood and each tooth was
stored in a separate, clean, labeled, resealable, plastic bag.
All USAFA samples contained at least one third molar, with the majority of individuals
providing all four. Only third molars were run from this facility. Whole teeth, in the best
overall condition were preferentially selected for drilling. If multiple teeth from an
individual were of the same quality, sampling selection was based on the same criteria as
mentioned for the CIL samples: mandibular teeth were chosen over maxillary teeth and
right over left. Teeth exhibiting unusual crown anomalies or staining patterns and/or
teeth in which the author disagreed with the dental staff numbering were photographed
prior to drilling only. The remaining teeth were not photo-documented. Photo content
consisted of the tooth, subject identifier number, tooth number, and a scale (Figure 4-3).
Two photos, one of the buccal or lingual surface, and one of the occlusal surface were
taken. Teeth were cleaned in distilled water within individual capped vials with a
Branson Bransonic 1510 tabletop ultrasonic cleaner for 30 minutes. After air-drying,
teeth were cleaned of any surface contaminants to include alveolar bone remnants using a
NSK UM50 TM slow-speed dental drill with a #8 carbide dental drill bur. Samples of
approximately 100-200 mg of pristine enamel were drilled off of each tooth using the
same set-up. Care was taken not to drill into the dentine. Enamel powder was collected
on creased weighing paper and transferred to labeled 1.5 mL microcentrifuge tubes. Drill
bits, weighing paper, and latex gloves were discarded after drilling each tooth and the
drill cleaned of adherent enamel powder.
Figure 4-3. Pre-drilling photo of AFA-093 #32.
Carbon and Oxygen Sample Preparation
Central Identification Laboratory Samples
Chemical preparation of the enamel powder was performed in the stable isotope
laboratory at the Florida Museum of Natural History, Gainesville, FL, according to the
protocol developed by Dr. Pennilynn Higgins, museum postdoctoral fellow. The powder
of one tooth from each individual was selected based on the integrity of the sample (i.e.,
whether there was the possibility of dentin or other contaminants mixed in with the
enamel) and greatest mass of powder available for analysis. Organic residues were
removed from the sample powder by adding 1 mL 30% hydrogen peroxide (H202) to
each microcentrifuge tube. Tubes were shaken utilizing a Thermolyne Maxi-Mix 1,
16700 mixer and the lids lifted up to prevent gas pressure build-up inside the tubes. The
opened vials were stored in a closed reaction cabinet. Samples were periodically shaken
with the mixer, every 1 to 2 days, to re-suspend the enamel powder that had settled at the
bottom of the vial.
On a weekly basis, the H202 was removed by centrifuging samples for 20 minutes
at 10,000 RPM in an Eppendorf 5415D microcentrifuge and pipetting off the H202.
Pipette tips were discarded between each sample to prevent cross-contamination. Fresh
H202 was then added following the same protocol. Samples were reshaken, lids opened,
and placed back in the reaction cabinet. The absence of escaping air bubbles from the
solution usually indicates the sample powder is finished reacting and ready for the next
phase of treatment. After consulting with Drs. Bruce MacFadden, Florida Museum of
Natural History, and John Krigbaum, University of Florida Department of Anthropology,
the samples were decanted of all H202 after 51 days in solution, even though nearly half
still appeared to be reacting. This was likely due to the large quantity of powder being
processed, with most enamel samples measuring 100 mg or greater. Samples were then
twice rinsed with 1 mL deionized water utilizing the same procedure for removing H202
(i.e., water added, then tubes shaken, centrifuged down, and decanted).
After rinsing the samples with deionized water and decanting all water from them,
secondary carbonates were removed via an acetic acid bath. Rinsed samples, free from
water, were bathed in 1 mL 0.1 N acetic acid, shaken, and allowed to sit for 30 minutes.
The acetic acid was pipetted off after centrifuging the microcentrifuge tubes at 10,000
RPM for 5 minutes. Samples were then twice rinsed with deionized water and the water
removed in the same manner as previously discussed. Samples were allowed to air-dry in
their open microcentrifuge tubes inside of a desiccator for 2 weeks. This ensured all
liquid had evaporated from the enamel powder.
An additional side test examining the necessity of performing the acetic acid step
was performed. Theoretically, teeth extracted from living subjects should not have to
undergo the acetic acid bath, because teeth in the living are not subject to diagenetic
changes associated with the build-up of secondary carbonates due to taphonomic factors.
The acetic acid bath was performed on all samples because there is no know precedence
to do without acetic acid for forensic isotope purposes. Eight USAFA and two CIL
samples were split in half with one half undergoing the full protocol previously outlined
and the second sample of each pair undergoing the H202 bath and rinses only. Values
were then compared to determine if this step of the protocol is indeed required.
Portions of the dried enamel measuring between 1.2 mg and 1.5 mg were then
loaded into stainless steel boats at the University of Florida, Department of Geological
Sciences, Light Isotope Laboratory. Each boat was placed into 1 of 44 numbered slots on
a brass tray (Figure 4-4) and the tray placed into a desiccator until they were run on the
laboratory's VG/Micromass (now GV Instruments) PRISM Series II isotope ratio mass
spectrometer with an Isocarb common acid bath preparation device. Load sheets were
accomplished for each tray listing the sample name and weight for each position in the
tray (Appendix D). All samples were loaded by the author. The PRISM was operated by
Dr. Jason Curtis and Kathy Curtis.
The first run was organized as follows: slots 1-4, standards ofNBS-19 measuring
between 60 gg and 120 gg; slots 5-20, alternating enamel powder and blank positions;
Figure 4-4. Loaded tray for PRISM mass spectrometer analysis.
slots 21-22, NBS-19 standard; slots 23-42, alternating enamel powder and blank
positions; slots 43-44, NBS-19 standard. This arrangement allowed for the analysis of
18 samples. Empty or blank positions were included in the first run to check for
contaminants within the samples. Mass spectrometer readings for the blank positions,
indicate leaching of slow-reacting sample into these positions and hence likely
contamination. Contamination tends to be much more of an issue with fossilized samples
versus modern or historical (Koch et al 1997). Because the first run ran clean with no
indication of contamination, the blanks were replaced with sample for all subsequent
runs. The sample line-up for all subsequent runs therefore was as follows: slots 1-4,
NBS-19 standard; slots 5-20, sample; slots 21-22, NBS-19 standard; slots 23-42,
sample; slots 43-44, NBS-19 standard. This arrangement allowed for the analysis of 36
samples for each run of the mass spectrometer.
United States Air Force Academy Samples
The Academy samples were prepared in the same manner as the CIL samples with
three exceptions. The first change to the processing protocol entailed reducing H202
exposure time to 24 hours. This change was made upon the recommendation of Dr.
Bruce MacFadden, Florida Museum of Natural History, and Dr. Pennilynn Higgins,
Stable Isotope Ratios in the Environment Analytical Laboratory, Department of Earth and
Environmental Sciences, University of Rochester. It was implemented because whole
teeth were cleaned for 2-3 days using a 3% H202 solution to remove external organic,
but more importantly, because teeth were collected directly from living subjects.
Organics associated with diagenetic transfer due to burial were not encountered with
these samples as they were with the CIL samples.