at SciVerse ScienceDirect
Journal of Archaeological Science 40 (2013) 1435e1443Contents lists availableJournal of Archaeological Science
journal homepage: http : / /www.elsevier .com/locate/ jasCommentary
Silo science and portable XRF in archaeology: a response to Frahm
Robert J. Speakman a,*, M. Steven Shackley b
aCenter for Applied Isotope Studies, University of Georgia, Athens, USA
bGeoarchaeological XRF Laboratory, University of California, Berkeley, USAa r t i c l e i n f o
Article history:
Received 1 August 2012
Received in revised form
20 September 2012
Accepted 22 September 2012
Keywords:
PXRF
Obsidian
Reliability
Validity
Silo science* Corresponding author.
E-mail address: archsci@uga.edu (R.J. Speakman).
0305-4403/$ e see front matter 
 2012 Elsevier Ltd.
http://dx.doi.org/10.1016/j.jas.2012.09.033a b s t r a c t
A recent study by Frahm (2013) on the application of portable XRF (PXRF) for chemical characterization
of obsidian ignores fundamental issues of reliability and validity in the measurements, and justifies
“internally consistent” measurements as acceptable. We argue this form of science is unacceptable, point
out several flaws in Frahm’s paper, and provide some examples of PXRF measurements that are valid and
reliable and conform to international standards as published.

 2012 Elsevier Ltd. All rights reserved.The recent literature has been spate with experiments and
analyses of archaeological materials, often obsidian, with portable
X-ray fluorescence spectrometry (PXRF). Many have been focused
on the project at hand, and some have been concerned with the
quality of the resultswith regard to the study. A common thread that
concerns us is consistently ignoring the issues of reliability and
validity in the measurements, and justifying these “internally
consistent” results as necessary and sufficient for compositional
analysis. While the philosophical underpinnings of this perspective
are unfortunate, particularly in Americanist archaeology, the long-
term results might be more problematic. Ellery Frahm’s recent
study (Frahm, 2013) is especially troubling given that it is obvious
that he fully understands the importance of reliable and valid
studies in archaeology (Frahm, 2012). If, as promulgated by Frahm
and others, that it is perfectly fine to provide results that are only
internally consistent, and do not conform to established interna-
tional standards and data, then we are entering a time of “silo
science” where each researchers’ data is self-contained, indepen-
dent, and cannot be verified externally. We find this to not only be
unacceptable, but another aspect of a social science that is ejecting
the “science” aspect of our discipline to “play scientist” with
portable XRF technology. Frahm’s “internally consistent” results
using the Niton PXRF, while valid as far as it goes, is not necessary or
acceptable. There are other PXRF options that conform to interna-
tional standards as published, and are valid and reliable en toto.All rights reserved.As Grave et al. (2012) have noted (which Frahm acknowledges
in his paper), “archaeological applications [of PXRF] continue to be
cautiously treated, principally because of a perceived lack of
analytic rigor.” Grave et al. are absolutely correct on this point.
In publishing his paper, Frahm essentially argues that a lack of
analytical rigor is acceptabledtrust me my results are
internally consistent. Most archaeological PXRF users lack
experiencedwhether it be X-ray physics, basic analytical chem-
istry, statistics, or the fundamentals of provenance studies. Many
users approach PXRF from a “black box” perspective in which the
inner workings of the XRF instrument are not understood by the
user, nor does the user care to learn how and why the instrument
functionsdthe only importance is that the sample is analysed and
that numbers are generated (Speakman et al., 2010, 2011). This
often results in the user not recognizing the most basic problems
that can and do occur with XRF analyses. Because of the manner in
which the capabilities of PXRF are [mis]represented by many
manufacturers (e.g., promoting ideas of internal consistency,
ability to measure low Z elements accurately, etc.), users tend to
disregard conventional knowledge concerning XRF fundamentals.
Consequently, we believe that many PXRF-based studies of
pottery, obsidian, metals, and other materials that are presented
at meetings and/or published are founded on poor science. A
professional photographer recently told one of us that as
a consequence of widespread availability of digital cameras, now
everyone is a photographer. It would seem that the same holds
true for portable analytical instrumentationdnow everyone can
be a scientist.
1 In fact, the manufacturer has taken the internal components of this small
portable benchtop system and repackaged them into a handheld configuration
referred to as the ProSpector. The internal electronics (tube, detector, multichannel
analyser) and software for the two systems are virtually identicaldthe only
difference is packaging.
R.J. Speakman, M. Steven Shackley / Journal of Archaeological Science 40 (2013) 1435e14431436While we agree that archaeological interpretations can be
made from Frahm’s data, the ultimate results do not fulfil the rules
of validity and reliability, the foundation of good scientific
research. If the results of any experiment cannot be compared and
evaluated by a subsequent experiment outside the original
experiment, then it is unreliable even though it is internally
consistent. X-ray fluorescence spectrometry (what Frahm refers to
as labXRF) in archaeology has attempted to rely on and refine valid
and reliable measurements such that any XRF or NAA laboratory
can use those results and analyses of archaeological or source data
to arrive at the same result. Most scientists would agree that this
is not only a laudable goal, but a necessary one in order to avoid
“silo science” in archaeologydi.e., my results are internally
consistent and so I don’t care if it’s correct within the discipline or
the results can be tested externally. A recent PXRF obsidian
analysis exercise at the 2012 Society for American Archaeology
meeting (organized by the authors, Michael Glascock, and Arlen
Heginbotham) was structured to address this very issue (publi-
cation of this study is forthcoming).
One can easily interpret Frahm’s paper as approving this
emerging “silo science”, and that reliability and validity no longer is
the cornerstone of scientific enquirydat least in compositional
studies in archaeology. Although Frahm’s paper deals specifically
with obsidian, which analytically is fairly straightforward to
chemically characterize, there are unintended consequences.
Specifically, there are those who would interpret and apply his
approach to the study of ceramics, soils, bone chemistry, and other
material classes.
In his introduction, Frahm states that to date fewer than two
dozen studies based on PXRF of obsidian have been published. This
statement is incorrect. We are aware of about 70 publications of
PXRF of obsidian (Appendix A); we suspect that many more exist in
the Asian, European, and South American literature.
Frahm’s introduction states that to achieve extreme portability,
these analysers sacrifice performance. While this is true of certain
models of all manufactures, the reality is that many PXRF systems
currently on the market have superior detector resolution and
electronics than laboratory-based instruments manufactured 5e10
years ago. More specifically, Frahm states that PXRF instruments
use less-than-ideal geometric arrangements of miniaturised, low-
power components that run on batteries, and the processing of X-
rays is done by on-board electronics and software rather than
external systems. We are unclear as to what this “less-than-ideal”
geometry specifically refers to, but we have serious doubts that any
manufacturer would fail to optimize their tube-sample-detector
geometry. We know that Bruker PXRF systems are configured
with the tube and detector at a 62
 angle with sample placement
occurring at the convergence of this angle; Thermo laboratory-
based units are about 70
 with sample placement occurring at
the convergence of this angle. Other XRF systems (portable and
fixed laboratory) use angles that range from 45 to 90
, none of
which create problems with current XRF engineering. Frahm also
cites miniaturization of components as problematicdWhy? Mini-
aturization of electronic components has been ongoing for decades
in all areas of technology. For example, radios and televisions no
longer use transistors and modern cell phones are only a fraction of
the size they were 15 years ago. Did miniaturization of electronic
components for radios, television, and phones result in a sacrifice to
performance? We think not. Finally, we fail to understand why
Frahm would state that the processing of X-rays being done by
onboard electronics is a sacrificedthat’s where the X-ray detector
is located. There’s no reason that this function needs to occur
outside the “box” whether it be a portable or a fixed-laboratory
systemdwe have come a long way from the days when a single
instrument required a large space that was dedicated toa mainframe computer, multichannel analysers, amplifiers, and
other assorted electronic components.
Although Frahm designates acronyms for laboratory-based
energy dispersive XRF (Frahm’s labXRF) and portable-XRF
(Frahm’s HHpXRF), it is apparent that Frahm truly does not
understand the instrumentation used to generate data for his own
publications. Frahm states that his labXRF analyses of obsidianwere
conducted at the University of Missouri Research Reactor (MURR),
and according to Frahm (2010: 403) the instrument used to analyse
the artifacts at MURR (and confirmed via personal communication
from Michael Glascock to RJS) was an ElvaX EDXRF system. The
ElvaX instrument is designed and marketed not as a laboratory XRF
system, but rather a portable-XRF instrument. Given that one of us
(Speakman) has used this specific instrument to conduct research
in Alaska and Panamadoutside of the laboratorydand MURR
scientists have deployed the ElvaX instrument in Argentina
(Barberena et al., 2011; Giesso et al., 2011), the ElvaX system clearly
does not fit Frahm’s definition of labXRF.1 Consequently in Frahm’s
manuscript, Figs. 6ael and 7aec are not figures that plot traditional
laboratory-based EDXRF data (Frahm’s labXRF) versus PXRF data
(Frahm’s HHpXRF) or laboratory EDXRF data versus NAA data, but
rather these figures are simply comparing (1) data generated from
two different portable XRF instruments and (2) PXRF data to NAA
data. To reiterate, these figures do not project any data generated by
a laboratory-based EDXRF system as Frahm has indicated.
However, it is of interest here that Frahm’s plots of labXRF
(really PXRF) data versus NAA data show higher correlated values
than plots comparing the ElvaX PXRF data to Niton PXRF data. The
reason for this is quite simpledthe MURR ElvaX instrument was
calibrated by Mike Glascock using obsidian that had been analysed
by INAA and/or other analytical methods at MURR with the express
intent of generating data that would be as comparable to MURR’s
extant NAA obsidian database. In other words the instrument was
calibrated to generate valid and reliable results for obsidian (for an
example, see INAA and PXRF data published in Knight et al., 2011).
In contrast, the Niton XRF data are poorly correlated with MURR
PXRF data because the Niton was not calibrated for obsidian.
In Fig. 1, we provide a comparison of data that compare data
from a Thermo Fisher laboratory-based EDXRF system and a Bruker
PXRF. The calibration for the Thermo XRF was created by Shackley
using international powdered geologic reference materials
whereas the PXRF data were generated using a ‘canned’ factory
calibration based on the analysis of 40 or so obsidian samples
provided to Bruker by MURR (Glascock and Ferguson, 2012). In
contrast to Frahm’s XRF comparison figures, we observe highly
correlated values for the 12 plotted samples. In some cases there
are systematic offsets between the Thermo and Bruker data, but
because the two datasets are highly correlated, it is possible to
correct the data through the use of a secondary reference material
(i.e., RGM-1, RGM-2, SRM-278)dthe same method that any two
laboratories, XRF or otherwise, would use to intercalibrate data.
This underscores the importance of publishing data not only for
archaeological samples, but also data generated for international
reference standards as has been the practice of Shackley and others
for more than 20 years.
Frahm states that “Empirical correction curves using custom
calibrations are regarded as the best practice for labXRF but are
rarely used in HHpXRF obsidian sourcing”dthis statement simply
Fig. 1. Direct comparison of values (expressed as ppm) obtained for 12 samples using a portable XRF (x-axis) and laboratory-based EDXRF (y-axis).
R.J. Speakman, M. Steven Shackley / Journal of Archaeological Science 40 (2013) 1435e1443 1437is not accurate. Beginning in 2006, Speakman and Mike Glascock
began using empirical calibration schemes based on obsidian
reference materials to calibrate the ElvaX PXRF, initially at MURR
and later at the Smithsonian. In 2007, Speakman in collaboration
with Bruker developed the “green” filter that is commonly used for
obsidian analyses involving Bruker instruments and at the same
developed the first empirical obsidian calibration scheme for the
Bruker PXRF. Since then, we along with our MURR colleagues have
continued to refine calibration standards for empirical analyses.
These standards have been used on dozens of instruments
since2008.2 XRF, portable or lab-based, requires one to create
element references, standards libraries, and calibrations that can be2 In Appendix A, data for all papers that include Glascock or Speakman as
coauthors are based on custom empirical calibrations. Furthermore, all publications
in Appendix A that are generated using Bruker instruments also are based on
custom empirical calibrations. Likewise, most PXRF measurements reported in
Heginbotham et al. (2011)da round robin of copper-based metalsdare based on
custom empirical calibrations.constantly modified. XRF instruments rarely have “off the shelf”
calibrations, and if they do they are typically instructional and
ultimately require the user to make some adjustments. Having said
that, some vendors do not allow the user to make such
adjustments.
Over the past decade, we have collectively examined almost
every major PXRF instrument on the market. The one single issue
that we have continually raised with vendors has been centred
around this fundamental issue of software and the ability of the
user to create and modify empirical calibrations. Ultimately, this is
why somemanufacturers have been more successful than others in
selling instruments for archaeological-based research. We believe
that PXRF manufacturers can and should provide better
softwaredthere is no compelling reason not to. For example,
Thermo-Fisher manufactures several high-quality lab based XRF
instruments. The software for these instruments is fully developed
and there is absolutely no reason why they could not provide that
software for use with spectra generated on their Niton handheld
instruments. The same holds true for other manufacturers.
Historically there is a reason for thisdcommercial portable units
Fig. 2. Results in ppm (y-axis) versus time (x-axis) for 307 consecutive 200-s counts of
Bruker obsidian standard XRF08. Note that the overall trend for each element is “flat”
indicating that instrumental drift is not an issue. Likewise note that the relative
deviation between analyses is minimal and that no outliers are present. Refer to
Table 1 for summary statistics for this experiment.
R.J. Speakman, M. Steven Shackley / Journal of Archaeological Science 40 (2013) 1435e14431438were designed intentionally to be black boxes for use by novices
outside of a laboratory environment primarily in the mining and
metals recycling industries. They were never really intended, at
least initially, to be true research instruments (or really developed
with the intent for analysis of silicates), hence the software and
user interfaces were “dummied” down to limit analytical choices
and facilitate ease of use.
Although Frahm and others (e.g., Dybowski, 2012) have
convincingly demonstrated that some PXRF instruments cannot be
used to generate valid and reliable obsidian data (using factory
calibrations), we believe some viable options do exist. For example,
Bruker has recently released an obsidian calibration based on a set
of forty slab-cut obsidian source samples commissioned from the
MURR Archaeometry Laboratory (Glascock and Ferguson, 2012).
Analysed by multiple analytical techniques, the set supersedes (in
terms of accuracy) earlier calibrations developed by MURR and
Speakman. A cursory examination of data generated using this
calibration suggests high degrees of accuracy, precision, and
reproducibility (see Speakman, 2012 for the full unpublished
report).
In an experiment designed to assess reproducibility, a Bruker
PXRF was set up to analyse a single obsidian sample continuously
for 17 h with spectra being saved every 200 s. All samples were
measured at 40 kV, 25 mA, with a 12 mil Al, 1 mil Ti, 6 mil Cu filter
placed in the X-ray path for a 200-s live-time count. Peak intensities
for the Ka peaks of Mn, Fe, Zn, Ga, Rb, Sr, Y, Zr, Nb, and La peak of Th
were calculated as ratios to the Compton peak of rhodium, and
converted to parts-per-million (ppm) using Bruker’s factory-
installed calibration for obsidian. This resulted in data for 307
analyses that are summarized in Table 1. Individual data points for
these analyses are plotted in Fig. 2 (Zn, Ga, Th, Rb, Sr, Y, Zr, Nb) and
Fig. 3 (Mn and Fe). Data for these 307 analyses exhibit relatively low
variation (expressed as %RSD). For Fe, Rb, Y, Zr, and Nb, the %RSD is
2% or lower which is comparable to data generated on most
laboratory-based EDXRF instruments. Mn, Zn, and Ga exhibit %RSD
values of ca. 3e6% which is typical for these elements. Sr has the
highest error, as would be expected given the low concentration in
this particular sample (which also approaches the limits of EDXRF
detection for Sr in silicate matrices). In other words, the high %RSD
for Sr in this sample is entirely due to counting statistics, and in no
way a reflection of the instrument itself, beyond inherent instru-
mental error.
A Horwitz curve (Horwitz et al., 1980) (Fig. 4) for the 40 obsidian
calibration standards illustrates our point concerning Sr. Horwitz
curves are commonly used to evaluate upper and lower ends of
calibration curves. In this plot Sr concentrations (x-axis, based on
the average value of 5 measurements) are plotted against %RSD (y-
axis). A power trendline is fit against all data points. In this
particular plot the x-axis is constrained to samples with less than
100 ppm Sr to facilitate visual evaluation of the low end of the
curve. The plot shows (as expected) that as concentration decreases
(i.e., lower counting statistics), %RSD increases; thereby illustrating
the point that measurement error is correlated with concentration
(counting statistics). The closer one gets to instrument detection
limits, the greater the analytical error. At concentrations of aboutTable 1
17-h stability test (307 consecutive analyses at 200 s each).
Mn Fe Zn Ga Th
Average 438.7 7446.2 134.1 27.0 41
St dev 23.7 72.6 4.6 1.7 1
%RSD 5.4 1.0 3.4 6.2 3
a High %RSD (% relative standard deviation) a consequence of low Sr concentration in th
instrument used for the analysis (all EDXRF instruments would exhibit similar errors at th
is the norm.7 ppm Sr we observe less than 5%RSD. At concentrations above
15 ppm Sr the error decreases to about 2e3%RSD. To reiterate, the
high %RSD reported in Table 1 is not a reflection of XRF perfor-
mance; similar results would be obtained with any properly cali-
brated EDXRF spectrometer regardless of manufacturer.
Of particular import for Figs. 2 and 3 is that the overall trend for
each element is “flat”, indicating that instrumental drift (stability)
is not an issue. For researchers conducting analyses under field
conditions (or in museums), reanalysis of samples is oftentimes not
possible. An instrument that drifts throughout the day will impart
greater analytical error on the experiment. For obsidian studies this
could mean that distinctions among compositional groups could be
obscured. Likewise it is important to note that the relative deviation
between analyses is minimal and that no outliers are present. In
most laboratory systems, the energies can be calibrated at any point
to mitigate this problem; many PXRF systems do this as well simply
by restarting the instrument.
We also assessed the accuracy of Bruker’s obsidian calibration
against independent quality control standards. As stated above and
elsewhere (Shackley, 2010), such standards must periodically be
analysed and published to establish validity. Tables 2 and 3 present
data obtained from replicate analyses of pressed-powdered inter-
national obsidian standards (RGM-1 and NIST SRM-278) relative to
their recommended values and published literature values. In all
cases the Bruker data are acceptable, thereby establishing validity.
Ultimately though, the problem with converting the spectra to
ppm and associated problems related to accuracy and reproduc-
ibility lie with the user, not the manufacturer. In Frahm’s case an
argument could be made that the Niton PXRF will not allow the
user to do this. However, the fact of the matter is that Niton spectraRb Sra Y Zr Nb
.7 364.4 1.1 83.9 163.7 237.5
.5 3.4 0.5 1.6 1.7 2.4
.5 0.9 41.1 2.0 1.0 1.0
is sample which results in low counting statistics. In no way is this a reflection of the
is concentration andmatrix). For samples containingmore than 15 ppm Sr, 2e3%RSD
Fig. 3. Results in ppm (y-axis) versus time (x-axis) for 307 consecutive 200-s counts of
Bruker obsidian standard XRF08. Note that the overall trend for each element is “flat”
indicating that instrumental drift is not an issue. Likewise note that the relative
deviation between analyses is minimal and that no outliers are present. Refer to
Table 1 for summary statistics for this experiment.
R.J. Speakman, M. Steven Shackley / Journal of Archaeological Science 40 (2013) 1435e1443 1439(and for that matter, any other digital format XRF spectra) can be
exported and manipulated using 3rd party software. Several of the
participants of a recent inter-laboratory study based on analysis of
copper-based metals employed such an approach with Bruker XRF
instruments (Heginbotham et al., 2011). Likewise, Speakman et al.
(2011) used a 3rd party software program to calculate numbers
for lower atomic mass elements in a recent study of pottery. Hence
it is not impossible, it just requires a little more expertisedthe kind
of expertise that comes from years of XRF experience and training.
It is very convenient to blame the instrument or manufacturer for
data that lack accuracy and reproducibility, but in reality it is the
user who is ultimately responsible for accuracy and reproducibility.
To reiterate here, if a user cannot develop a good calibration withFig. 4. Horwitz curve for the 40 obsidian calibration standards used to calibrate Bruker
PXRF instrument. Sr concentration (x-axis) is based on the average value of five
measurements. Curve is fit to all data points, but the plot is constrained to samples
with less than 100 ppm Sr to facilitate the evaluation of the low end of the curve. This
plot shows that as concentration decreases (i.e., lower counting statistics), that %RSD
increases thereby illustrating the point that measurement error is correlated with
concentration. The lower one gets to instrument detection limits, the greater the
analytical error. Again, this is not a reflection of instrument performance; similar
results would be obtained with any properly calibrated EDXRF spectrometer.their instrument, either on their own or in conjunction with the
manufacturer, then they need to employ a third-party software
packagedsuch software is available and just about any spectra
(whether it be Thermo, Bruker, or InnovX) can be converted. Data
lacking accuracy and reproducibility cannot and should not be
blamed on the instrument manufacturer (see McAlister, 2011).
From Frahm’s own introduction, “.using deliberately sub-
optimal conditions is not recommended for analytical work.”
Frahm is correct, deliberately using sub-optimal conditions is never
recommendeddso why publish it in the first place? Most of us who
have been analysing obsidian for years recognize that one can use
any number of data correction schemes to extract numbers that are
only internally consistent. Most of us involved in chemical char-
acterization studies of obsidian choose not to and instead focus our
efforts on generating and publishing data that are accurate, precise,
and reproducible. What then was the motivation for Frahm’s
manuscript? Frahm was previously aware from his earlier work
(Frahm, 2007) that Niton PXRF instruments provided substandard
results for obsidian when using their factory calibrationdso why
would he intentionally design an experiment using the same (or
similar) instrument and same (or similar) samples with the intent
of using deliberately sub-optimal conditions? The fact is that data
in Frahm’s paper look suspiciously like data presented at a 2007
meeting of the Geological Society of America (Frahm, 2007) for
which Frahm has published the PowerPoint presentation on-line:
(https://gsa.confex.com/gsa/viewHandout.cgi?uploadid¼250). On
the basis of Slide Number 5 in this presentation, which questions
why so much spread exists in the data, it would appear that Frahm
was not attempting to deliberately use suboptimal conditions in his
2007 study. Instead it appears that Frahm was making a genuine
attempt to produce reliable data. A publication based on this study
would have been idealdFrahm apparently followed the manufac-
turer’s analytical protocol and determined that data were subop-
timal using their canned calibration. It is Frahm’s demonstration
that Niton PXRF systems are not replacements for other XRF
instruments and that data are likely incompatible with extant
databases that would have had the greater benefit to archaeological
scientists. In contrast to Frahm, a recent publication by Dybowski
(2012) that compared obsidian data generated by laboratory-
based WDXRF and Niton portable XRF concluded “there is only
one plausible explanation for the differences in the pXRF and the
WDXRFS measurements, which is that errors exist within the Niton
pXRF software (see also McAlister, 2011 for a similar conclusion).
Thus, I am suspicious of the Niton and its continued use in XRF
studies for archaeological applications.” Although we would agree
with Dybowski where obsidian is concerned, the factory “metals”
calibration for most PXRF instruments (Niton, Bruker, and
InnovX)dat least for major elementsdappears to be acceptable for
archaeological research. The problem with metals though is that
surface measurements made by XRF do not necessarily reflect the
bulk composition of the metal.
There is significant variability among the three major PXRF
instruments (e.g., Niton, InnovX, and Bruker) used for the vast
majority of archaeological and museum-based research. Based on
our knowledge of these instruments, it does not appear that all are
equal for analyses of obsidian. We are optimistic, however, that all
manufacturers will ultimately produce instruments and associated
software that will provide the kind of results that are necessary for
consistently valid and reliable analyses for obsidian and other
sample matrices. However, as we have demonstrated above, there
are options available now, and as we have argued, “internally
consistent” results, while valid as far as it goes, is not necessary or
acceptable. We call on all reviewers of peer-reviewed manuscripts
and proposals to seriously evaluate data being put forward from
PXRF, and for that matter, all analytical studies employing portable
Table 2
Replicate analyses of USGS RGM-1 and comparison to published values.
Mn Fe Zn Th Rb Sr Y Zr Nb
Bruker tracer (n ¼ 5) 321  28 13,075  69 40  2 16  1 157  3 104  1 26  1 223  3 10  1
USGS recommended 279  50 13,010  210 32 15  1.3 150  8 110  10 25 220  20 8.9  0.6
Shackley (2012) 302  14 13,116  308 n.r. 16  3 151  3 106  3 25  2 219  5 9  2
Skinner and Davis (1996) 291  47 13,480  745 37  7 n.r. 152  3 107  9 24  3 217  8 11  1
Hughes (2007) 278  10 13,079  140 n.r. n.r. 143  4 105  3 23  3 214  4 8  3
n.r.dnot reported.
Table 3
Replicate analyses of NIST SRM-278 and comparison to published values.
Mn Fe Zn Th Rb Sr Y Zr Nb
Bruker tracer (n ¼ 5) 432  27 14,521  100 55  4 13  2 133  2 62  1 41  1 281  2 18  1
NIST recommended 403  2 14,269  140 n.r. 12.4  0.3 127.5  0.3 63.5  0.1 n.r. n.r. n.r.
Glascock (2006) 397  23 14,500  900 53  5 12.6  0.6 133  6 64  5 39  5 290  30 n.r.
Shackley (2012) 383  7 14,329  37 n.r. 15  5 130  2 67  1 40  2 276  2 15  2
n.r.dnot reported.
R.J. Speakman, M. Steven Shackley / Journal of Archaeological Science 40 (2013) 1435e14431440instrumentation. Perhaps, as one of us has argued elsewhere, the
acceptance of unreliable and invalid data in archaeology is perva-
sive and allows for users of PXRF instrumentation to readily accept
this “internally consistent” silliness (Shackley, 2010). Why would
any archaeologist accept results that could not be verified, unless
they simply did not understand the need for verification. If that is
indeed the case, 21st century archaeology is moving down
a hazardous path. And lastly, we still cannot understandwhy Frahm
would want to analyse materials in this way and publish it in an
archaeological science journal.
Conflict of interest statement
Robert J. Speakman maintains a professional relationship with
Bruker AXS specifically with respect to instrument and application
development in archaeological science. Speakman’s laboratory, The
Center for Applied Isotope Studies at the University of Georgia,
occasionally conducts work for Bruker AXS on a fee for service
basis. Speakman also is a principal in Rusty Trowel, LLC (registered
in the State of Georgia), a company that provides consulting
services, in the form of XRF instrument training, to Bruker AXS
customers. M. Steven Shackley declares no competing interests.
Acknowledgements
We thank Mike Glascock, Eric Dyrdahl, and two anonymous
reviewers for commenting on earlier drafts of this paper. Mike
Glascock, Mark McCoy, Eric Dyrdahl, and others also are acknowl-
edged for their assistance in compiling the PXRF obsidian
bibliography.
Appendix A. Bibliography of PXRF Obsidian publications
1. Barberena, R., A. Hajduk, A. F. Gil, G. A. Neme, V. Duran, M. D.
Glascock, M. Giesso, K. Borrazzo, M. de La Paz Pompei, M. L.
Salgan, V. Cortegoso, G. Villarosa and A. A. Rughini, 2011.
Obsidian in the south-central Andes: Geological, geochemical,
and archaeological assessment of north Patagonian sources
(Argentina). Quaternary International 245(1):25e36.
2. Blomster, J. P. and M. D. Glascock, 2010. Procurement and
consumption of obsidian in the Early Formative Mixteca Alta:
a view from the Nochixtlan Valley, Oaxaca, Mexico. In Crossing
the Straits: Prehistoric Obsidian Source Exploitation in the
North Pacific Rim, edited by Y. V. Kuzmin and M. D. Glascock,pp. 183e200. British Archaeological Reports International
Series 2152. Archaeopress, Oxford.
3. Boulanger, M. T., R. S. Davis and M. D. Glascock, 2012. Prelimi-
nary characterization and regional comparison of the Dasht-i-
Nawur obsidian source near Ghazni, Afghanistan. Journal of
Archaeological Science 39(7):2320e2328.
4. Burger, R. L. and M. D. Glascock, 2009. Intercambio prehistorico
de obsidiana a larga distancia en el Norte Peruano. In Revista
del Museo del Arqueologia, Antropologia e Historia, edited by E.
Vergara Montero, pp. 17e50. vol. 11. Universidad Nacional de
Trujillo, Facultad de Ciencias Sociales, Trujillo, Peru.
5. Burley, D. V., P. J. Sheppard and M. Simonin, 2011. Tongan and
Samoan volcanic glass: pXRF analysis and implications for
constructs of ancestral Polynesian society. Journal of Archaeo-
logical Science 38(10):2625e2632.
6. Caria, M. A., P. S. Escola, J. P. Gomez Augier and M. D. Glascock,
2009. Obsidian circulation: new distribution zones for the
Argentinian Northwest. International Association of Obsidian
Studies Bulletin 40:5e11.
7. Carter, T., 2009. Elemental Characterization of Neolithic Arte-
facts Using Portable X-Ray Fluorescence. Çatalhöyük Research
Project: Çatalhöyük 2009 Archive Report, pp. 126e128. Avail-
able online at http://www.catalhoyuk.com/downloads/
Archive_Report_2009.pdf.
8. Cecil, L. G., M. D. Moriarity, R. J. Speakman and M. D. Glascock,
2007. Feasibility of field-portable XRF to identify obsidian
sources in central Peten, Guatemala. In Archaeological Chem-
istry: Analytical Techniques and Archaeological Interpretation,
edited by M. D. Glascock, R. J. Speakman and R. S. Popelka-
Filcoff, pp. 506e521. ACS Symposium Series No. 968. Amer-
ican Chemical Society, Washington, DC.
9. Craig, N., R. J. Speakman, R. S. Popelka-Filcoff, M. S. Aldenderfer,
L. Flores Blanco, M. Brown Vega, M. D. Glascock and C. Stanish,
2010. Macusani obsidian from southern Peru: A characteriza-
tion of its elemental composition with a demonstration of its
ancient use. Journal of Archaeological Science 37(3):569e576.
10. Craig, N., R. J. Speakman, R. S. Popelka-Filcoff, M. D. Glascock, J.
D. Robertson, M. S. Shackley and M. S. Aldenderfer, 2007.
Comparison of XRF and PXRF for analysis of archaeological
obsidian from southern Peru. Journal of Archaeological Science
34(12):2012e2024.
11. Cruickshank, A. (2011). A Qualitative and Quantitative Analysis
of the Obsidian Sources on Aotea (Great Barrier Island), and
their Archaeological Significance. Unpublished MA thesis,
University of Auckland.
R.J. Speakman, M. Steven Shackley / Journal of Archaeological Science 40 (2013) 1435e1443 144112. Davidson, J., 2011. Archaeological investigations at Maungarei:
A large Maori settlement on a volcanic cone in Auckland, New
Zealand. Tuhinga 22:19e100.
13. Drake, B. L., A. J. Nazaroff, K. M. Preufer, 2009. Error assessment
of portable X-ray fluorescence spectrometry in geochemical
sourcing. SAS Bulletin 32(3):14e17.
14. Dybowski, D. J., 2012. PXRF/WDXRF Inter-Unit Data Compar-
ison of Arizona Obsidian Samples. Journal of Arizona Archae-
ology 2(1):15e21.
15. Eerkens, J. W., K. J. Vaughn, M. Linares-Grados, C. A. Conlee, K.
Schreiber, M. D. Glascock and N. Tripcevich, 2010. Spatio-
temporal patterns in obsidian consumption in the Southern
Nasca Region, Peru. Journal of Archaeological Science
37(4):825e832.
16. Elias, A. M., D. E. Olivera, M. D. Glascock and P. Escola, 2009.
Procedencia de obsidians de sitos arqueologicos tardios y
tardios-inkas de Antofagasta de la Sierra (Prov. de Cata-
marca, Puna Meridional Argentina) a traves de XRF. In
Arqueometria Latinoamericana: Segundo Congreso Argen-
tino y Primero Latinoamericano, edited by T. Palacios, pp.
109e114. Comision Nacional de Energia Atomica (CNEA),
Buenos Aires, Argentina.
17. Forster, N. and P. Grave, 2012. Non-destructive PXRF analysis of
museum-curated obsidian from the Near East. Journal of
Archaeological Science 39(3):728e736.
18. Frahm, E., 2007. An evaluation of portable X-ray fluorescence
for artifact sourcing in the field: Can handheld devices differ-
entiate Anatolian Obsidian sources? Paper presented at the
2007 Annual Meeting of the Geological Society of America.
Presentation available on-line: https://gsa.confex.com/gsa/
viewHandout.cgi?uploadid¼250.
19. Frahm, E., 2012. Non-destructive sourcing of Bronze Age Near
Eastern obsidian artefacts: Redeveloping and reassessing
electron microprobe analysis for obsidian sourcing. Archaeo-
metry 54:623e642.
20. Frahm, E., 2013. Validity of “Off-the-Shelf” Handheld Portable
XRF for Sourcing Near Eastern Obsidian Chip Debris. Journal of
Archaeological Science 40(2):1080e1092.
21. Frahm, E., 2013. Empires and resources: Central Anatolian
obsidian at Urkesh (Tell Mozan, Syria) during the Akkadian
period. Journal of Archaeological Science 40(2):1122e1135
22. Frahm, E., 2012. Notes from the President. IAOS Bulletin 47:2e5.
23. Frahm, E., 2010. The Bronze-age obsidian industry at Tell
Mozan (ancient Urkesh), Syria. Ph.D. dissertation, Department
of Anthropology, University of Minnesota. Available online at
University of Minnesota’s Digital Conservancy: http://purl.
umn.edu/99753.
24. Freund, K. P., 2010. Lithic Technology and Obsidian Exchange
Networks in Bronze Age Sardinia, Italy (ca. 1600e850 B.C.).
Graduate School Theses and Dissertations. Paper 3429. http://
scholarcommons.usf.edu/etd/3429.
25. Freund, K. P. and R. H. Tykot, 2011. Lithic technology and
obsidian exchange networks in Bronze Age Nuragic Sardinia
(Italy) Archaeological and Anthropological Sciences
3(3):151e164.
26. Galvão, T., Lopes, F., Moraes, A., Appoloni, C., 2009. Study of
analytical sensitivity of two portable X-ray fluorescence
systems for archaeological obsidians analysis. 2009 Interna-
tional Nuclear Atlantic Conference, Rio de Janeiro, Brazil, 27
September to 2 October 2009. Conference paper available
online at: http://www.fisica.uel.br/gfna/E16_1081.pdf.
27. Ghorabi, S., F. Khademi Nadooshan, M. D. Glascock, A. Hejabari
Noubari and M. Ghorbani, 2010. Provenance of obsidian tools
from Northwestern Iran using X-ray fluorescence analysis andneutron activation analysis. International Association of
Obsidian Studies Bulletin 43:14e25.
28. Giesso, M., M. A. Beron and M. D. Glascock, 2008. Obsidian in
the Western Pampas, Argentina: source characterization and
provisioning strategies. International Association of Obsidian
Studies Bulletin 38:15e18.
29. Giesso, M., V. Duran, G. Neme, M. D. Glascock, V. Cortegoso, A.
Gil and L. Sanhueza, 2011. A study of obsidian source usage in
the Central Andes of Argentina and Chile. Archaeometry
53(1):1e21.
30. Glascock, M. D., 2010. Comparison and contrast between XRF
and NAA: used for characterization of obsidian sources in
Central Mexico. In X-Ray Fluorescence Spectrometry (XRF) in
Geoarchaeology, edited by M. S. Shackley, pp. 161e192.
Springer, New York.
31. Glascock, M. D., A. Beyin and M. E. Coleman, 2008. X-ray fluo-
rescence and neutron activation analysis of obsidian from the
Red Sea Coast of Eritrea. International Association of Obsidian
Studies Bulletin 38(Winter 2008):6e10.
32. Glascock, M. D. and M. Giesso, 2012. New perspectives on
obsidian procurement and exchange at Tiwanaku, Bolivia. In
Obsidian and Ancient Manufactured Glasses, edited by I. Liritzis
and C. M. Stevenson, pp. 86e96. University of New Mexico
Press, Albuquerque, NM.
33. Glascock, M. D., Y. V. Kuzmin, A. V. Grebennikov, V. K. Popov, V.
E. Medvedev, I. Y. Shewkomud and N. N. Zaitsev, 2011. Obsidian
provenance for prehistoric complexes in the Amur River basin
(Russian Far East). Journal of Archaeological Science
38(8):1832e1841.
34. Glascock, M. D., P. C. Weigand, R. Esparza Lopez, M. A. Ohner-
sorgen, M. Garduno Ambriz, J. B. Mountjoy and J. A. Darling,
2010. Geochemical characterisation of obsidian in Western
Mexico: the sources in Jalisco, Nayarit, and Zacatecas. In
Crossing the Straits: Prehistoric Obsidian Source Exploitation in
the North Pacific Rim, edited by Y. V. Kuzmin and M. D. Glas-
cock, pp. 201e218. British Archaeological Reports International
Series 2152. Archaeopress, Oxford.
35. Goebel, T., R. J. Speakman and J. D. Reuther, 2008 Results of
geochemical analysis of obsidian artifacts from the Walker
Road Site, Alaska. Research in the Pleistocene 25:88e90.
36. Goldstein, R. C., 2010. Negotiating Power in the Wari Empire: A
Comparative Study of Local-imperial Interactions in the
Moquegua and Majes Regions during the Middle Horizon
(550e1050 CE). Ph.D. dissertation, Department of Anthro-
pology, Northwestern University.
37. Golitko, M., 2011. Chapter 13: Provenience Investigations of
Ceramic and Obsidian Samples Using Laser Ablation Induc-
tively Coupled Plasma Mass Spectrometry and Portable X-Ray
Fluorescence. Fieldiana Anthropology 42:251e287.
38. Golitko, M., J. Mierhoff, G. M. Feinman, P. R. Williams, 2012.
Complexities of collapse: the evidence of Maya obsidian as
revealed by social network graphical analysis. Antiquity
86:507e523.
39. Golitko, M., J. Meierhoff and J. E. Terrell, 2010. Chemical char-
acterization of sources of obsidian from the Sepik coast (PNG).
Archaeology in Oceania 45:120e129.
40. Goodale, N., D. G. Bailey, G. T. Jones, C. Prescott, E. Scholz, N.
Stagliano and C. Lewis, 2012. pXRF: A study of inter-instrument
performance. Journal of Archaeological Science 39(4):875e
883.
41. Houlette, C., J. Rasic, and J. Speakman, 2012. Prehistoric
Obsidian Procurement and Transport in Gates of the Arctic
National Park and Preserve. Alaska Park Science 10(1):12e15.
42. Jia, P. W.-m., T. Doelman, C.-j. Chen, H. Zhao, S. Lin, R. Torrence
and M. D. Glascock, 2010. Moving sources: A preliminary study
R.J. Speakman, M. Steven Shackley / Journal of Archaeological Science 40 (2013) 1435e14431442of volcanic glass artefact distributions in Northeast China using
PXRF. Journal of Archaeological Science 37(7):1670e1677.
43. Khazaee, M., M. D. Glascock, P. Masjedi, A. Abedi and F. K.
Nadooshan, 2011. The origins of obsidian tools from Kul Tepe,
Iran. International Association of Obsidian Studies Bulletin
45:14e17.
44. Knight, C. L. F., A. M. Cuellar, M. D. Glascock, M. L. Hall and P. A.
Mothes, 2011. Obsidian source characterization in the Cordil-
lera Real and eastern piedmont of the north Ecuadorian Andes.
Journal of Archaeological Science 38(5):1069e1079.
45. Laguens, A. G., M. Giesso, M. I. Bonnin andM. D. Glascock, 2007.
Mas alla del horizonte: cazadores-recolectores e intercambio
a large distancia en Intihuasi (provincia de San Luis, Argentina).
Intersecciones en Antropologia 8:7e17.
46. Levine, M., A. Joyce and M. D. Glascock, 2011. Shifting patterns
of obsidian exchange in Postclassic Oaxaca, Mexico. Ancient
Mesoamerica 22(1):123e133.
47. Liritzis, I., Zacharias, N., 2011. Chapter 6: Portable XRF of
archaeological artifacts: current research, potentials and limi-
tations, in: Shackley, M.S. (ed.), X-ray Fluorescence Spectrom-
etry (XRF) in Geoarchaeology. Springer, pp. 109e142.
48. McAlister, A. J., 2011, Methodological issues in the geochemical
characterisation and morphological analysis of stone tools:
a case study fromNuku Hiva, Marquesas Islands, East Polynesia.
Ph.D. dissertation, Department of Anthropology, University of
Auckland.
49. McCoy, M.D. (2011). Geochemical characterization of volcanic
glass from Pu’u Wa’awa’a, Hawaii Island. Rapa Nui Journal,
25(2):41e49.
50. McCoy, M., Mills, P., Lundblad, S., Rieth, T., Kahn, J., Gard, R.,
2011. A cost surface model of volcanic glass quarrying and
exchange in Hawai’i. Journal of Archaeological Science 38,
2547e2560.
51. McCoy, M.D., Ladefoged, T.N., Blanshard, A. and Jorgensen, A.
(2010). Reconstructing lithic supply zones and procurement
areas: an example from the Bay of Islands, Northland, New
Zealand. Journal of Pacific Archaeology 1(2): 174e183.
52. Meierhoff, J., M. Golitko and J. Morris, 2010. Sourcing obsidian
from the ancient Maya farming community of Chan, Belize
using portable XRF. SAS Bulletin 33(2):5e8.
53. Millhauser, J. K., E. Rodriguez-Alegria and M. D. Glascock, 2011.
Testing the accuracy of portable X-ray fluorescence to study
Aztec and Colonial obsidian supply at Xaltocan, Mexico. Journal
of Archaeological Science 38(11):3141e3152.
54. Moore, P.R., 2011. Cultural distribution of obsidian along the
Waikato-King Country Coastline, North Island, New Zealand.
Journal of Pacific Archaeology 2(1):29e39.
55. Mosley, B. & McCoy, M.D. (2010) Sourcing obsidian and pitch-
stone from the Wakanui Site, Canterbury, New Zealand. Rapa
Nui Journal 24(2):6e15.
56. Nazaroff, A. J., K. M. Prufer and B. J. Drake, 2010. Assessing the
applicability of portable X-ray fluorescence spectrometry for
obsidian provenance research in the Maya lowlands? Journal of
Archaeological Science 37(4):885e895.
57. Parry, W. J. and M. D. Glascock, in press. Obsidian blades from
Cerro Portezuelo: sourcing artifacts from a long-duration site.
Ancient Mesoamerica.
58. Phillips, S. C. and R. J. Speakman, 2009. Initial source evaluation
of archaeological obsidian from the Kuril Islands of the Russian
Far East using portable XRF. Journal of Archaeological Science
36(6):1256e1263.
59. Pool, C. A., C. L. F. Knight andM. D. Glascock, in press. Formative
obsidian procurement at Tres Zapotes, Veracruz, Mexico:
implications for Olmec and epi-Olmec political economy.
Ancient Mesoamerica.60. Rademaker, K. D., M. D. Glascock, B. Kaiser, D. Gibson and D. R.
Lux, in press. Multi-technique geochemical characterization of
the Alca obsidian source, Peruvian Andes. Geology.
61. Reimer, R., 2007. The Mountains and the rocks are forever:
Lithics and landscapes of SKWXWÚ7MESH UXWUMIXW. Open
Access Dissertations and Theses. Paper 6735. http://
digitalcommons.mcmaster.ca/opendissertations/6735.
62. Reuther, J. D., N. S. Slobodina, J. T. Rasic, J. P. Cook, and R. J.
Speakman et al., 2011. Gaining MomentumdThe Status of
Obsidian Source Studies in Alaska: Implications for Under-
standing Regional Prehistory. In From the Yenisei to the
Yukon Interpreting Lithic Assemblage Variability in Late
Pleistocene/Early Holocene Beringia, edited by Ted Goebel
and Ian Buvit, pp. 270e286, Texas A&M University Press,
College Station.
63. Rosania, C. N., M. T. Boulanger, K. T. Biro, S. Ryzhov, G. Grinka
and M. D. Glascock, 2008. Revisiting Carpathian obsidian.
Antiquity 82(318).
64. Schreyer, S., 2012. The Inka in the Northern Ecuadorian Andes:
Economic aspects of empire consolidation. MA Thesis,
Department of Anthropology, California State University,
Fullerton.
65. Seelenfreund, A., M. Pino, M. D. Glascock, C. Sinclaire, P.
Miranda, D. Pasten, S. Cancino, M. I. Dinator and J. R. Morales,
2010. Morphological and geochemical analysis of the Laguna
Blanca/Zapaleri obsidian source in the Atacama Puna. Geo-
archaeology 25(2):245e263.
66. Sheppard, P. J., G. J. Irwin, S. C. Lin and C. P. McCaffrey, 2011.
Characterization of New Zealand obsidian using PXRF. Journal
of Archaeological Science 38(1):45e56.
67. Sheppard, P., Trichereau, B., Milicich, C., 2010. Pacific obsidian
sourcing by portable XRF. Archaeology in Oceania 45: 21e30.
68. Slobodina, Natalia S., Joshua D. Reuther, Jeffrey T. Rasic, and
John P. Cook, Robert J. Speakman (2009) Obsidian Procurement
and Use at the Dry Creek Site (HEA-005), Interior Alaska.
Current Research in the Pleistocene 26:115e117.
69. Smith, M. E., A. L. Burke, T. S. Hare and M. D. Glascock, 2007.
Sources of imported obsidian at Postclassic sites in the Yaute-
pec Valley, Morelos: a characterization study using XRF and
INAA. Latin American Antiquity 18(4):429e450.
70. Speakman, R.J., 2012. Evaluation of Bruker’s Tracer Family
Factory Obsidian Calibration for Handheld Portable XRF Studies
of Obsidian. Report Prepared for Bruker AXS, Kennewick, WA.
Available on-line: http://www.bruker-axs.com/fileadmin/user_
upload/PDFse/handhelds/Bruker_Obsidian_Report.pdf.
71. Speakman, R. J., C. E. Holmes, and M. D. Glascock, 2007. Source
Determination of Obsidian Artifacts from Swan Point (XBD-
156), Alaska, Current Research in the Pleistocene 24:143e145.
72. Torrence, R., P. White, and S. Kelloway, 2012. Expanding the
Range of PXRF to Ethnographic Collections. IAOS Bulletin
46:9e14.
73. Tykot, R.H., 2010. Sourcing of Sardinian Obsidian Collections in
the Museo Preistorico-Etnografico ‘Luigi Pigorini’ Using Non-
Destructive Portable XRF,” in C. Lugliè (ed.), L’ossidiana del
Monte Arci nel Mediterraneo. Nuovi apporti sulla diffusione,
sui sistemi di produzione e sulla loro cronologia. Atti del 5

Convegno internazionale (Pau, Italia, 27e29 Giugno, 2008),
85e97, NUR, Ales.
74. Tykot, R. H., 2007. Early Neolithic obsidian trade in Sardinia: the
Coastal Site of Santa Caterina di Pittinuri (Cuglieri e OR). In
Préhistoire et protohistoire de l’aire tyrrhénienne/Preistoria e
protostoria dell’area tirrenica, 217e220, edited by C. Tozzi &
M.C. Weiss. Unione Europea, Interreg III A Francia e Italia,
“Isole” Toscana, Corsica, Sardegna. ASSE III e Scambi trans-
frontalieri Misura 3.1. Felici Editori. 2007.
R.J. Speakman, M. Steven Shackley / Journal of Archaeological Science 40 (2013) 1435e1443 144375. Tykot, R.H., L. Lai, C. Tozzi, 2011. Intra-Site Obsidian Subsource
Patterns at Contraguda, Sardinia (Italy), in I. Turbanti-Memmi
(ed.), Proceedings of the 37th International Symposium on
Archaeometry, 13the16th May 2008, Siena, Italy, 321e328.
Springer.
76. Vazquez, C., O. Palacios, M. Parra Lue-Meru, G. Custo, M. Ortiz
and M. Murillo, 2012. Provenance study of obsidian samples
by using portable and conventional X-ray fluorescence spec-
trometers. Performance comparison of both instrumenta-
tions. Journal of Radioanalytical and Nuclear Chemistry
292:367e373.
77. Williams, P. R., L. Dussubieux and D. J. Nash, 2012. Provenance
of Peruvian Wari obsidian: comparing INAA, LA-ICP-MS, and
portable XRF. In The Obsidian and Ancient Manufactured
Glasses, edited by I. Liritzis and C. M. Stevenson, pp. 75e85.
University of New Mexico Press.
References
Barberena, R., Hajduk, A., Gil, A.F., Neme, G.A., Duran, V., Glascock, M.D., Giesso, M.,
Borrazzo, K., de La Paz Pompei, M., Salgan, M.L., Cortegoso, V., Villarosa, G.,
Rughini, A.A., 2011. Obsidian in the south-central Andes: geological,
geochemical, and archaeological assessment of north Patagonian sources
(Argentina). Quaternary International 245 (1), 25e36.
Dybowski, D.J., 2012. PXRF/WDXRF inter-unit data comparison of Arizona obsidian
samples. Journal of Arizona Archaeology 2 (1), 15e21.
Frahm, E., 2007. An evaluation of portable X-ray fluorescence for artifact sourcing in
the field: can handheld devices differentiate Anatolian obsidian sources? Paper
Presented at the 2007 Annual Meeting of the Geological Society of America.
Presentation. Available on-line: https://gsa.confex.com/gsa/viewHandout.cgi?
uploadid¼250.
Frahm, E., 2010. The Bronze-age Obsidian Industry at Tell Mozan (Ancient Urkesh),
Syria. Ph.D. dissertation, Department of Anthropology, University of Minnesota.
Available online at University of Minnesota’s Digital Conservancy: http://purl.
umn.edu/99753.
Frahm, E., 2012. Evaluation of archaeological sourcing techniques: reconsidering and
re-deriving Hughes’ four-fold assessment scheme. Geoarchaeology 27, 166e174.
Frahm, E., 2013. Validity of “Off-the-Shelf” handheld portable XRF for sourcing near
easternObsidian chip debris. Journal of Archaeological Science 40 (2),1080e1092.
Giesso, M., Duran, V., Neme, G., Glascock, M.D., Cortegoso, V., Gil, A., Sanhueza, L.,
2011. A study of obsidian source usage in the central Andes of Argentina and
Chile. Archaeometry 53, 1e21.Glascock, M.D., 2006. Tables for Neutron Activation Analysis, sixth ed. Research
Reactor Center, University of Missouri, Columbia.
Glascock, M.D., Ferguson, J.R., 2012. Report on the Analysis of Obsidian Source
Samples by Multiple Analytical Methods. University of Missouri Research
Reactor, Columbia.
Grave, P., Attenbrow, V., Sutherland, L., Pogson, R., Forster, N., 2012. Non-
destructive pXRF of mafic stone tools. Journal of Archaeological Science 39,
1674e1686.
Heginbotham, A., Bezur, A., Bouchard, M., Davis, J., Eremin, K., Frantz, J.H., Glins-
man, L., Hayek, L., Hook, D., Kantarelou, V., Karydas, A., Lee, L., Mass, J., Matsen,
J., McCarthy, B., McGath, M., Shugar, A., Sirois, J., Smith, D., Speakman, R.J.,
2011. A round-robin study to evaluate the inter-laboratory reproducibility of
quantitative X-ray fluorescence spectroscopy on Historic copper alloys. In:
Mardikian, P., Chemello, C., Watters, C., Hull, P. (Eds.), Metal 2010, Proceedings
of the Interim Meeting of the ICOM-CC Metal Working Group, Charleston,
South Carolina 11e15 October 2010. Clemson University, South Carolina, pp.
244e255.
Hughes, R., 2007. The geologic sources of obsidian artifacts from Minnesota
archaeological sites. The Minnesota Archaeologist 66, 53e68.
Horwitz, W., Kamps, L.R., Boyer, K.W., 1980. Quality assurance in the analysis of
foods and trace constituents. Journal of the Association of Official Analytical
Chemists 63, 1344e1354.
Knight, C.L.F., Cuellar, A.M., Glascock, M.D., Hall, M.L., Mothes, P.A., 2011. Obsidian
source characterization in the Cordillera Real and eastern piedmont of the north
Ecuadorian Andes. Journal of Archaeological Science 38 (5), 1069e1079.
McAlister, A.J., 2011, Methodological Issues in the Geochemical Characterisation and
Morphological Analysis of Stone Tools: a Case Study from Nuku Hiva, Marquesas
Islands, East Polynesia. Unpublished Ph.D. dissertation, Department of
Anthropology, University of Auckland, New Zealand.
Shackley, M.S., 2012. ThermoFisher Scientific Quant’X Analysis and Instrumenta-
tion. http://www.swxrflab.net/anlysis.htm (accessed 18.07.12.).
Shackley, M.S., Nov. 2010. Is There Reliability and Validity in Portable X-ray
Fluorescence Spectrometry (PXRF)? The SAA Archaeological Record.
pp. 17e18&20.
Skinner, C.E., Davis, M.K., 1996. X-ray Fluorescence Analysis of an Obsidian Biface
from the Fort Hill Site, Highland County, Ohio. Report 1996-48 prepared for the
Ohio Historical Society, Columbus, Ohio, by Northwest Research Obsidian
Studies Laboratory, Corvallis, Oregon.
Speakman, R.J., 2012. Evaluation of Bruker’s Tracer Family Factory Obsidian Cali-
bration for Handheld Portable XRF Studies of Obsidian. Report prepared for
Bruker AXS, Kennewick, WA. Available on-line: http://www.bruker-axs.com/
fileadmin/user_upload/PDFse/handhelds/Bruker_Obsidian_Report.pdf.
Speakman, R.J., Little, N.C., Creel, D., Miller, M.R., Iñañez, J.G., 2011. Sourcing
ceramics with portable XRF spectrometers? A comparison with INAA using
Mimbres pottery from the American Southwest. Journal of Archaeological
Science 38 (12), 3483e3496.
Speakman, R.J., Phillips, S.C., Florey, V., Little, N.C., Iñañez, J.G., 2010. Approaches to
Micro-XRF analysis of obsidian. Paper Presented at the 38th International
Symposium on Archaeometry, Tampa Bay, Florida.