Lithic Technology - Boston University

EXPERIMENTAL THREE-DIMENSIONAL PRINTING OF A
LOWER PALAEOLITHIC HANDAXE: AN ASSESSMENT OF THE
TECHNOLOGY AND ANALYTICAL VALUE
BRANDON R. OLSON, JODY M. GORDON, CURTIS RUNNELS AND
STEVE CHOMYSZAK


Boston University, USA
Wentworth Institute of Technology, USA
Chipped stone tools are difficult to illustrate in publications with line drawings or analog photographs, and previous
attempts to improve on stone tool illustration with stereoviews, coatings, or casts have not been widely adopted by
lithic analysts. New software makes it possible to create interactive photorealistic D digital images of stone tools in
the field or laboratory without specialized or expensive equipment. These images can be replicated by D printers,
and based on our experiments using three different printing media we demonstrate that models printed in ABS
plastic are acceptably inexpensive forms that reproduce the artifact features required by specialists. The combination of image-based modeling and D printing will revolutionize the illustration of artifacts and greatly mitigate
the need for extensive travel and help alleviate accessibility issues.
KEYWORDS: image-based modeling, D printing, digital surrogacy, Palaeolithic handaxe, D photogrammetry
New technologies allow archaeological artifacts to be
recorded and illustrated with precision in an ethical,
non-destructive, and inexpensive manner. Two innovative technologies in particular can be combined to
record and duplicate artifacts: image-based D
digital modeling and D printing. In the last two
years, D modeling software, techniques, and printers have rapidly become cheaper and more user
friendly, and the adoption of D technologies in
archaeological research has been growing exponentially. This trend is discernible in recent publications.
The Journal of Field Archaeology (. []) and
the Journal of Eastern Mediterranean Archaeology
and Heritage Studies (. and . []) have dedicated entire issues to digital approaches to archaeology, and the Journal of Archaeological Science
has experienced a > percent increase in published works focusing on D approaches in archaeology in the last decade (Figure ). Although
archaeologists began printing objects created with
D laser scanners five years ago (Kuzminsky and
Gardiner ; Niven et al. ), the rapid progress of technological developments in D modeling and printing continue to add useful tools to
the archaeologist’s expanding digital toolkit. Here
we offer the results of our own experiments with
image-based photorealistic digital modeling — as
© W. S. Maney & Son Ltd 
DOI: ./Y.
opposed to laser scanning — for D printing in
archaeology for use in off-site artifact study using
three presently-available and industry standard
printing media, viz. powder/binder printing,
fused deposition modeling, and polyjet printing.
THE PROBLEM
Efforts to record and illustrate artifacts to make
them available for analysis, comparison, and publication most often take the form of two-dimensional
analog photographs and illustrations. For years
archaeologists have endeavored to add a level of
three dimensionality to artifact recordings through
the use of shading (Heath ) and stippling in
illustrations and the creation of coatings, squeezes,
and casts (Rick and White ). With respect to
lithic artifacts, efforts to illustrate accurately the
form and appearance of chipped stone tools of
flint, obsidian, and other siliceous rocks have been
ongoing since the mid-nineteenth century when
the flaked stone artifacts were first recognized as
the artificial products of early human handicraft
(Addington : –). Photography has rarely
been employed for illustrating stone tools because
the reflective surfaces of the rocks that were used
to manufacture the tools make it difficult to light
Lithic Technology , Vol.  No. , –
EXPERIMENTAL THREE-DIMENSIONAL PRINTING OF A LOWER PALAEOLITHIC HANDAXE

FIGURE . A chart showing the number of articles published in the Journal of Archaeological Science focusing on D
approaches in archaeology. The query was made on April , .
the specimens adequately to bring out the texture.
There is also the problem of the depth of field,
which makes it difficult to focus both on the center
of sometimes thick, massive stone tools and on the
retouched edges in the same photograph. Thus,
even in the best two dimensional reflective analog
photograph the three dimensional form of the artifact is difficult to perceive or evaluate.
These difficulties are obviated, at least partially, by
the use of measured technical line drawings that
include the outline of the stone tool, the pattern of
the scars on the surface left by the removal of
flakes in the reduction and retouch stages of artifact
manufacture, and the use of radial shading lines in
the flake scar outlines to suggest the volume of the
artifact and the texture of its surface as they might
appear if the artifact was lit obliquely from the
upper left (Figure ). The methods and conventions
for making two dimensional line drawings were
developed by French and English archaeologists in
the s, to enable readers to “see” artifacts like
the Palaeolithic handaxes that convinced early
archaeologists of the antiquity of humans in
Europe, e.g., the natural-sized drawings of handaxes
found at Hoxne by John Frere in  (Daniel :
–). The specific conventions used today dispense with most efforts at realistic depictions of the
appearance of the artifacts, reducing the images to
outlines with flake scar patterns, often accompanied
by small arrows and other marks to indicate the
technical aspects of flaking such as the presence of
negative bulbs of percussion that mark the points
Lithic Technology , Vol.  No. , –

OLSON ET AL.
FIGURE . A traditional line drawing of an Acheulean handaxe from the Bosphorus Region of Turkey (Runnels and Özdog˘ an
). It has been slightly edited and is reproduced here courtesy of the President and Trustees of Boston University.
where flake detachments were initiated (Shea :
–).
To this day technical line drawings have been the
industry standard for the illustration of lithic artifacts. Although photographs have generally been
avoided, cameras and photographers are more
readily available than trained lithic illustrators, and
there has long been a desire to employ photography
to make effective illustrations. Many attempts have
been made over the last  years to improve the
“readability” of photographs, such as coating stone
tools with aluminum powder or chalk to reduce
the reflectivity of their surfaces, making nonreflective casts of the artifacts to be photographed
(Rick and White ), or using stereo photography
or holography to capture the three dimensional
appearance of the objects, e.g., the stereoviews of
handaxes and cleavers from the Lower Palaeolithic
site of Olorgesailie in Kenya (Isaac ). These
techniques have had limited success and have not
been widely adopted. There have been significant
improvements in stone tool illustration recently
with the introduction of polymer casting and D
laser scanning (Grosman et al. ). The polymer
casting technique is widely used today for the
production of artifact copies for use in teaching,
the laboratory, and public display. It is a difficult,
costly process that does not lend itself to the publication of such artifacts in print media or online
because the casts must again be rendered in line
drawings or photographed in order to be reproduced. The Computerized Archaeology Laboratory
at The Hebrew University of Jerusalem is currently
developing a system to use laser scanning to,
among other things, better facilitate quantitative
analyses of lithic features, but the technology is still
under development (Goren-Inbar ; Grosman
et al. , ; Malinsky-Buller et al. ).
Any serious study of stone tools requires the lithic
specialist to handle the artifacts and examine them
in three dimensions under variable lighting conditions. D casts are useful for this purpose, and
when made correctly, they allow anyone with the
proper technical training to use the cast as a proxy
for the actual artifact for most purposes. Unfortunately, trained technicians have to obtain permission and travel to distant locations to gain
access to the original artifacts in order to make the
necessary molds for the casting, which is often a prohibitive obstacle to the use of this technique.
Lithic Technology , Vol.  No. , –
EXPERIMENTAL THREE-DIMENSIONAL PRINTING OF A LOWER PALAEOLITHIC HANDAXE

The inadequacy of most methods of illustration
and reproduction of stone tools has led to the
demand for a more flexible method to reproduce
the physical appearance of stone tools at their
natural scale. This need is now being met by the
combination of D image-based digital models
and D printing. Our successful experiments
demonstrate that it is feasible to photograph
stone tools in order to make D images that can
be viewed on computers and other platforms
such as tablets and smartphones. Files with these
images can be sent by email and used to print
copies of the artifacts on D printers. We believe
that the combination of the two techniques,
which use commercially available software and
hardware and do not require special technical
training, can be readily and inexpensively deployed
to the field, whether the museum, the laboratory,
or the excavation site, allowing the rapid and accurate D imaging and reproduction in a printed
format of stone tools in a matter of hours. In our
view this is the most significant advance in the illustration of stone tools in over  years.
THE PROCESS
IMAGE-BASED MODELING
The use of image-based digital modeling in archaeology is relatively new, though a handful of studies
have attested to the technology’s accuracy and
utility in archaeological contexts (De Reu et al.
, ; Forte ; McCarthy ; Olson
et al. ). The recent popularity of the technology, however, has not been matched by a standardized terminology, and a variety of terms have been
adopted and published, e.g., photomodeling
(Opitz ), structure-from-motion modeling
(Green et al. in press); photogrammetry (Quartermaine et al. ); computational photography
(Rabinowitz ); and image-based modeling
(Olson and Placchetti ; Remondino ).
For the sake of clarity and consistency we adopt
the term “image-based modeling” to include the
methods, software programs, and specific algorithms that extract spatial information from
images to generate D data (point clouds, tin
grids, monochromatic models, photorealistic textured models, etc.). There are a number of software
programs that utilize an image-based approach,
including newly developed open source options
(Green et al. in press), but because of the affordability, ease of use, and quality outputs, we used
Agisoft PhotoScan to model a Lower Palaeolithic
handaxe for our experiments (Figure ).
FIGURE . A cutaway of a D model created using an imagebased modeling approach showing (A) the point cloud; (B) the
untextured model; (C) the textured model. The figure was
designed and created by Ryan A. Placchetti and Brandon
R. Olson.
To model the handaxe three dimensionally, we
took  photographs at five different angles of
capture using a tripod and a rotatable surface,
an effective photographic method that has been
tested previously (Olson et al. ). Accurate
D models are possible so long as the surface of
the object is completely photographed and there
Lithic Technology , Vol.  No. , –

OLSON ET AL.
is at least  percent overlap in successive photographs. The Lower Palaeolithic handaxe was
photographed for this study with an  MP
Canon Rebel Ti digital SLR camera with an
– mm lens. The photos were loaded into
PhotoScan to generate a tessellated D point
cloud, then a dense point cloud, a monochromatic
D model, and finally a fully photorealistic textured model (Figure ; attached D model). The
textured model was exported as an .obj file and
downloaded into Meshlab for scaling and conversion to an .stl file for printing.
D
PRINTING
With the assistance of the Manufacturing Center at
Wentworth Institute of Technology (WIT) in
Boston, we printed the handaxe using three types
of D printers (powder [Figure B], ABS plastic
[Figure C], and resin [Figure D]). WIT currently
has three D printers located in the D Print Lab
within WIT’s Manufacturing Center: Powder/
Binder Printing utilizing a ZCorp  Plus printer
and ZPrint software (henceforth powder), Fused
Deposition Modeling utilizing a uPrint D Printer
and Catalyst EX software (henceforth plastic), and
PolyJet Printing utilizing an OBJET  printer and
OBJET Studio software (henceforth resin). Each
printer utilizes a different printing technology to
build three-dimensional parts using a layered additive process. Each printer requires a computational
model in .stl (standard tessellation language)
format that embodies the D geometry of the
object to be printed. The printing process requires
preparation of the print files, monitoring/replacing
print materials, routine printer maintenance,
initiation of the print, removal of the print from
the printer, and typically some form of postprocessing. The entire printing process can take anywhere from minutes to days depending on the object
size, printing resolution, printing speed, and postprocessing. With the assistance of Mechanical
Engineering and the D Printing Lab at Wentworth
Institute of Technology in Boston, we printed copies
of the handaxe with all three D printers (Figure ).
POWDER PRINTING
Powder printing follows a process whereby a thin
layer of powder is adhered to previous layers using a
binder system that is sprayed through a liquid jet
printing head. The binder reacts with the powder
to transform it from a “dry” state into a “green”
state, meaning that the powder is stuck together,
but is still somewhat fragile. To fortify the green
powder print, a Magnesium Sulfate solution is
often applied to create a thin protective crust. Alternatively the print can also be infiltrated with Cyanoacrylate or a low viscosity two-part epoxy,
FIGURE . The printed Acheulean handaxe in the following forms (A) the original artifact; (B) the powder print; (C) the plastic
print; (D) the resin print.
Lithic Technology , Vol.  No. , –
EXPERIMENTAL THREE-DIMENSIONAL PRINTING OF A LOWER PALAEOLITHIC HANDAXE
either of which adds significant strength to the
object and, once infiltrated, the object can be
painted if desired.
The printer — the ZCorp  Plus — is capable of
creating layers as thin as . mm and can produce
relatively fine detailed prints within a  ×  ×
 mm build volume. An advantage of powder
printing is that all object surfaces are well supported
by the dry powder, which reduces the need for
additional object support and raft considerations
inherent in plastic and resin printing. Such an attribute allows many objects to be printed within the
D build volume of the printer during one continuous
print session, which greatly increases productivity.
While the level of detail possible with powder printing is relatively fine, increased detail can result in
increased object fragility, especially during infiltration because of manual handling during the excavation of the object from the printer, the removal of
excess powder from the print via a de-powdering
process involving a small stream of compressed air,
and the final post-processing.
PLASTIC PRINTING
Plastic printing uses a heated nozzle to melt a
filament of material that is deposited one layer at
a time to construct the object. The heated material
fuses to a previously deposited layer. The temperature of the nozzle and the speed at which it moves
are functions of the type of material being printed.
There are numerous polymer options including
Nylon, Polycarbonate, Polylactic Acid (PLA),
Ultem, PET, and ABS plastic.
The uPrint D Printer tested here is a Fused
Deposition Modeling (FDM) option that uses a
proprietary form of ABS plastic. It is capable of
printing in layer heights of . mm within a
build volume of  ×  ×  mm. Due to the
nature of FDM printing, parts with overhanging
features require support to prevent drooping
during the deposition and cooling process. The
Catalyst EX software used with the printer automatically calculates where the support material
needs to be added and adds the support features
to the printed model. In the case of the uPrint,
the support material is not ABS plastic, but a dissolvable plastic that is printed using a second
heated nozzle. It is easily dissolved in a heated
sodium hydroxide bath during post-processing.
Parts made with ABS plastic are less fragile than
powder prints and can be used immediately after
the support material is dissolved. It is even possible
to print pre-assembled parts that are operable so
long as proper clearances between moving parts

are designed into the assembly prior to printing.
Objects printed in plastic are not as strong as
those created using injection molding, which is a
common mass manufacturing method of creating
plastic parts, because the FDM process essentially
laminates (fuses) one layer of melted plastic onto
another. The fusing process does not provide the
same level of strength that would be found in a
molded part where the printing medium fills a
mold cavity while in a molten state and the entire
part subsequently cools to a uniform temperature.
This means that — at times — there are certain
utility considerations one must contemplate when
determining how to best orient the digital object
within the build volume prior to printing.
RESIN PRINTING
Resin printing uses a print head to spray a thin
layer of photocurable resin. Each layer of liquid
resin is converted into a solid by exposure to ultraviolet light. The benefit of this process is that it is
fast and can produce durable parts with great
detail and there are numerous resins with different
properties ranging from rigid plastic to flexible
rubber-like materials. As with plastic printing,
resin printing also requires a support system for
overhanging features. The support system is built
using a second photocurable resin that is easily
removed with a fine, high-pressure water jet. The
resin printer used for this test, an OBJET , is
capable of creating layers as thin as . mm
within a  ×  ×  mm build volume.
COSTS
In the end, it took approximately  hours to fully
model the Acheulean handaxe with an imagebased modeling technique, and, depending on the
desired printing method, up to seven hours for
each print. From the perspective of cost, the modeling of the objects required a digital SLR camera,
a tripod, and a license of PhotoScan ($ USD
with the educational discount). The cost of D
printing at WIT’s Manufacturing Center is based
solely on the cost of the materials used in each of
the printers, a nominal administrative markup,
and state sales tax. The cost of each print is estimated by the volume of printing material consumed, multiplied by the material’s price per unit
volume, which translated into the following
prices: powder ($. USD), plastic ($.
USD), and resin ($. USD). It must be noted
that one would encounter different pricing points
in a commercial setting. Figure  depicts an
Lithic Technology , Vol.  No. , –

OLSON ET AL.
FIGURE . A chart depicting an approximate normalized
material price range per cubic millimeter for each printing
media.
approximate normalized material price range per
cu mm for each of the three printers.
AN ASSESSMENT OF PRINTING TECHNOLOGIES
The principal use of printed D models of stone
tools will be for exhibition, teaching, and the comparison of specimens from different collections. D
models would permit colleagues, students, and
members of the public to evaluate the size,
volume, and appearance of artifacts with which
they have had little or no prior experience, or
where original specimens are unavailable. For
lithic technologists, on the other hand, the classification, measurement, publication, and comparison
of artifacts from different assemblages may still be
accomplished most economically by using conventional imagery supplemented with the use of D
digital images that can be viewed, manipulated,
and transmitted easily on electronic platforms
using commercially available software. In some
cases printed D models may be useful also for
analytical purposes. The form or the patterning
of the flake scars on stone tools may be difficult
to see in photographs or to depict in conventional
line drawings because the raw materials are
highly reflective like milky quartz, or have coarse,
grainy, and irregular surfaces like quartzites or
hydrothermally altered silicates. In such circumstances, even D digital models viewed on
computer screens, typically at a resolution of 
dpi, will suffer a significant loss of visual detail
and be difficult for the analyst to “read.” Printed
models show features that might otherwise be
hard to see in drawings or photographs. They
would also allow tactile examination, such as
putting one’s fingers in the negative bulbs of
percussion to find the places on the edge of the
tool where the flake scars originated, or to trace
shallow and indistinct edges of flake scars on the
face of the artifact. This sort of examination
could be particularly helpful in cases where the
surface of the stone tool is difficult to see, or
where the artifactual nature of the specimen
might be at issue. Our experience has shown that
in these cases it is best to handle the artifact
directly, and where this is not possible, the
printed D model may be the next best thing.
How well do the D printed models serve for the
study of stone tools by lithic analysts? And how do
the different media of reproduction (powder,
plastic, resin) affect the readability of the artifacts
being printed? To answer these questions we
selected a Lower Palaeolithic handaxe from the
French site of St. Acheul for modeling and printing
in all three materials for comparative purposes
(Figures  and ). The handaxe is from the Gabel
Museum of Archaeology in the Department of
Archaeology at Boston University and is a typical
specimen of the Acheulean Industrial Tradition. It
may be as much as , years in age and is
made of greenish brown flint (chert) of local origin
and has a heavily patinated or weathered surface.
QUALITATIVE COMPARISON
As part of their examination of such an artifact,
lithic analysts study the form of a stone tool in
order to describe and classify it correctly in terms
of the size, shape, and the placement of retouch,
among other attributes, that tend to vary from
culture to culture and period to period (for a full discussion of this process, see Shea : –,
especially –). Part of this examination involves
the study of the flaking process that was employed
to shape the artifact. The different methods of production and shaping produce different patterns of
flake scars on the surface of the artifact, and the
flake scars that result from the removal of flakes
by direct percussion with a hammerstone or a soft
hammer such as deer antler are shallow conchoidal
impressions. As flaking proceeds, the flake scars
may overlap in a parallel or sub-parallel series on
the surface of the finished artifact. The lithic
Lithic Technology , Vol.  No. , –
EXPERIMENTAL THREE-DIMENSIONAL PRINTING OF A LOWER PALAEOLITHIC HANDAXE
analyst must be able to see these patterns of flake
removals, and the shape, depth, and size of the
flake scars must be visible to permit the analyst to
correctly describe and classify the object.
For this purpose, the model printed in powder was
the least acceptable. Because we did not want to lose
minute surface details, we decided to forgo infiltration. On the one hand, subtle surface changes were
not obscured by the infiltration process, but on the
other hand the printed object had a dull chalky
surface with a grainy rough texture. Not only does
the material come off on one’s hands, it is also
difficult to see the finer flake scars on its surface.
The only advantage of using this material was its
low cost. The printed models would be useful only
for display or rough handling in the classroom
where expendable but cheap copies are desirable.
The resin model was of superior quality. It weighs
more, giving the model the look and feel of an actual
stone tool (albeit white in color, something that
could be corrected by using a color printer or
hand tinting). The flake scars are clearly visible on
the surface of the model, and the patterning, size,
and individual physical characteristics of the flake
removals can be readily discerned as the model is
examined in raking light. The surface reflects more
light than the powder model and has the same properties visually as found in the stone original. The
resin surface was more reflective, however, than
both of the other models, which tended to make it
difficult to define the boundaries, or edges, of the
flake scars. We found that it was more difficult to
see the surface detail under fluorescent lighting,
which washed out the surface by neutralizing the
reflectivity of the material. Natural light or incandescent lighting was superior for examining the
surface. Models in this material are more expensive
to print than those in the other media, but they
might be the best choice for use in displays.
The plastic models appear to be the most useful
for analytical purposes. The surface resolution of
detail is as good as that in the resin models, and
the cost of printing is significantly lower. There are
drawbacks to using plastic as a medium: the
models are lighter in weight than the resin print
(though, as with the resin, the print weight can be
increased to better approximate the actual artifact),
and thus do not feel as much like stone as the others.
The printing process also tends to build the plastic
up in layers that leave a surface texture with a distinctly visible striped or moiré pattern. This texture
can be distracting to the eye. Fortunately, this
layered texture does not interfere with the general
readability of the flake removals and surface

details of the artifact because the surface reflects
light in much the same way as noted on the resin
models. The plastic model had a crisp resolution of
the boundaries and edges of the flake scars.
QUANTITATIVE COMPARISON
Powder printing is not an ideal medium for artifact
analysis because of its fragility. As noted, to
strengthen a powder print, the object is often infiltrated with a Magnesium Sulfate solution, Cyanoacrylate, or a low viscosity two-part epoxy,
which create a protective crust that can obscure
subtle surface details. If not infiltrated, the print
deteriorates during handling. Owing to these
factors and those points raised in the qualitative
comparison, we offer here a quantitative comparison of only the plastic and resin prints.
In theory one would expect the resin handaxe to
maintain and present more surface detail than the
plastic form because it was printed at a higher resolution (an optimal layer thickness of . mm
versus . mm for the plastic, a  percent
difference). The combination of the thinner layers
and the viscosity of the liquid resin, however,
produce a smoothing effect during printing. The
smoothing effect is a result of minute surface displacement of each liquid resin layer prior to its
exposure to ultraviolet light. This smoothing
takes place during printing and produces a semireflective sheen on the printed object, which
slightly hinders the readability of the artifact in
certain lighting conditions. After exporting D
models of the plastic and resin prints as two dimensional raster grids and comparing , contiguous pixel values along the same linear path, the
difference between both printing media became
apparent (Figure ). Figure B depicts the pixel
values in centimeters for each object along the
paths shown in Figure A. The fact that the pixel
values for the plastic print are consistently higher
than those of the resin print is unimportant
because the difference could just as easily be attributed to photographic capturing conditions as to
differences in the media used to print the two
models. What is noteworthy, however, is the variability among pixel values within the object, which
is visible when one compares the rigidity of the
line representing values of the plastic model
(upper line in Figure B) to the smoother line
depicting resin pixel values (lower line in
Figure B). Furthermore, the average difference
in values of the same group of contiguous pixels
for the plastic print is . µm, while the average
Lithic Technology , Vol.  No. , –

OLSON ET AL.
FIGURE . A comparison of , contiguous pixel values for two D prints: (A) the line of pixels queried for the plastic (left)
and resin (right) prints; (B) a line graph showing the intensity and variability of pixel values in centimeters for both prints.
Lithic Technology , Vol.  No. , –
EXPERIMENTAL THREE-DIMENSIONAL PRINTING OF A LOWER PALAEOLITHIC HANDAXE
for the resin print is . µm. Therefore, the lessrigid contours of the line representing the resin
print (Figure B) and the smaller average difference
between contiguous pixel values (. µm versus
. µm) demonstrates that smoothing occurs by
way of liquid resin displacement during the printing process and prior to hardening. The higher viscosity of the ABS plastic prevents such a smoothing
effect. Although the smoothing only subtly affects
the preservation of topographic variation and
nuance of the printed model, the semi-reflective
surface hinders the readability of the object.
In sum, we found that both plastic and resin
work well for printing stone tools models that
are adequate for teaching, display, and laboratory
analysis. The choice of resin over plastic may be a
function of the unit cost of printing models and the
special needs of the specialist, but we are confident
that the plastic models are the best for most purposes, while the resin models are best reserved
for special purposes such as public display and
handling for non-analytic purposes.
CONCLUSIONS
In an attempt to define the new interest and
implementation of D workflows in all facets of
archaeology (planning, practice, documentation,
and dissemination), Rabinowitz cogently characterized the movement as the “age of digital surrogacy”
(Rabinowitz ). A digital surrogate is simply a
faithful digital representation of an original or
more specifically for our purpose here, the digital
outputs (point cloud, dense point cloud, tin grid,
georeferenced orthorectified photos, monochromatic
D models, textured D models, etc.) of an imagebased modeling workflow. What remains unclear,
however, is the relationship between the artifact
and the digital surrogate, and in our case the D
print of the digital surrogate. We are convinced
that the photorealistic models and D printings are
informative renderings that are superior to line illustrations in detail, and superior to casts in their ability
to be digitally transmitted and transferred.
It is clear that certain D printing formats retain
minute lithic characteristics. The powder model,
unless immersed in a durable coating, is too
fragile for continued handling. Although the
plastic model retains faint horizontal lines, which
are byproducts of printing, it retained the physical
characteristics of the stone artifact that a lithicist
needs to see in a reproduction, while the resin
model best approximates the original in physical
weight and feel, but at a modest loss in readability.

This entire process suggests that the combination
of inexpensive image-based modeling and printing
could revolutionize existing modes of archaeological analysis, dissemination, and education. This
combination of techniques makes it possible for
an archaeologist with little training to obtain a
digital camera, photograph an object, process the
images in a readily-available and easy-to-use commercial modeling software, print realistic D
models, and email the D models to others. And
it can be done in a day’s work. The process also
mitigates the need for extensive travel and artifact
accessibility issues after the object has been
modeled and printed. Three dimensional imagebased modeling and printing are revolutionary,
bringing the technology of the Stone Age to life
and for the first time, the technology helps solve a
-year old problem of lithic illustration.
ACKNOWLEDGEMENTS
We would like to thank the Department of Archaeology at
Boston University and the College of Arts and Sciences, as
well as the Manufacturing Center, at Wentworth Institute of
Technology for the use of their resources. The Chad DiGregorio Grant-in-Aid Fund at Boston University supported
this study, for which we are grateful.
REFERENCES
Addington, L. R.  Lithic Illustration: Drawing Flaked
Stone Artifacts for Publication. University of Chicago
Press, Chicago.
Daniel, G.  A Short History of Archaeology. Thames and
Hudson, London.
De Reu, J., P. De Smedt, D. Herremans, M. Van Meirvenne,
P. Laloo, and W. De Clercq  On Introducing an
Image-Based
D
Reconstruction
Method
in
Archaeological Excavation Practice. Journal of
Archaeological Science : –.
De Reu, J., G. Plets, G. Verhoeven, P. De Smedt, M. Bats,
B. Cherretté, W. De Maeyer, J. Deconynck,
D. Herremans, P. Laloo, M. Van Meirvenne, and
W. De Clercq  Towards a Three-Dimensional
Cost-Effective Registration of the Archaeological Heritage.
Journal of Archaeological Science : –.
Forte, M.  D Archaeology: New Perspectives and
Challenges — The Example of Çatalhöyuk. Journal of
Eastern Mediterranean Archaeology and Heritage
Studies : –.
Goren-Inbar, N.  Culture and Cognition in the
Acheulian Industry: A Case Study from Gesher Benot
Ya’aqov. Philosophical Transactions of the Royal
Society B: Biological Sciences : –.
Green, S., A. Bevan, and M. Shapland  A Comparative
Assessment of Structure from Motion Methods for
Archaeological Research. Journal of Archaeological
Science : –.
Lithic Technology , Vol.  No. , –

OLSON ET AL.
Grosman, L., A. Karasik, O. Harush, and U. Smilansky 
Archaeology in Three Dimensions: Computer-Based
Methods in Archaeological Research. Journal of Eastern
Mediterranean Archaeology and Heritage Studies : –.
Grosman, L., O. Smikt, and U. Smilansky  On the
Application of -D Scanning Technology for the
Documentation and Typology of Lithic Artifacts.
Journal of Archaeological Science : –.
Heath, S.  Closing Gaps with Low-Cost D. The
Archaeology of the Mediterranean World. <http://mediter
raneanworld.wordpress.com////closing-gapswith-low-cost-d/>.
Isaac, G. L.  Olorgesailie: Archeological Studies of a
Middle Pleistocene Lake Basin in Kenya. University of
Chicago Press, Chicago.
Kuzminsky, S. C., and M. S. Gardiner 
Three-Dimensional Laser Scanning: Potential Uses for
Museum Conservation and Scientific Research. Journal
of Archaeological Science : –.
Malinsky-Buller, A., L. Grosman, and O. Marder  A
Case of Techno-Typological Lithic Variability &
Continuity in the Late Lower Palaeolithic. Before
Farming /: –.
McCarthy, J.  Multi-Image Photogrammetry as a
Practical Tool for Cultural Heritage Survey and
Community Engagement. Journal of Archaeological
Science : –.
Niven, L., T. E. Steele, H. Finke, T. Gernat, and J.-J. Hublin
 Virtual Skeletons: Using a Structured Light
Scanner to Create a D Faunal Comparative
Collection. Journal of Archaeological Science :
–.
Olson, B. R., and R. A. Placchetti  A Discussion of the
Analytical Benefits of Image-Based D Modeling in
Archaeology. The Archaeology of the Mediterranean
World.
<http://mediterraneanworld.wordpress.com/
///a-discussion-of-the-analytical-benefits-of-imagebased-d-modeling-in-archaeology/>.
Olson, B. R., R. A. Placchetti, J. Quartermaine, and A.
E. Killebrew  The Tel Akko Total Archaeology
Project (Akko, Israel): Assessing the Suitability of
Multi-Scale D Field Recording in Archaeology. Journal
of Field Archaeology : –.
Opitz, R.  Three Dimensional Field Recording in
Archaeology: An Example from Gabii. The Archaeology
of the Mediterranean World. <http://mediterranean
world.wordpress.com////three-dimensional-fieldrecording-in-archaeology-an-example-from-gabii/>.
Quartermaine, J., B. R. Olson, and M. Howland  Using
Photogrammetry and Geographic Information Systems
(GIS) to Draft Accurate Plans of Qazion. Journal of
Eastern Mediterranean Archaeology and Heritage
Studies : –.
Rabinowitz, A.  The Work of Archaeology in the Age of
Digital Surrogacy. The Archaeology of the Mediterranean
World.
<http://mediterraneanworld.wordpress.com/
///the-work-of-archaeology-in-the-age-of-digitalsurrogacy/>.
Remondino, F.  Worth a Thousand WordsPhotogrammetry for Archaeological D Surveying. In
Interpreting Archaeological Topography: D Data,
Visualisation, and Observation, edited by R. S. Opitz, and
D. C. Cowley, pp. –, Occasional Publications of
the Aerial Archaeology Research Group .Oxbow, Oxford.
Rick, J. W., and T. D. White  Photography of Obsidian
Artifacts: A Useful Alternative. Newsletter of Lithic
Technology (): .
Runnels, C., and M. Özdog˘ an  The Palaeolithic of the
Bosphorus Region, NW Turkey. Journal of Field
Archaeology : –.
Shea, J.  Stone Tools in the Paleolithic and Neolithic
Near East. Cambridge University Press, Cambridge.
NOTES ON CONTRIBUTORS
Brandon R. Olson is a Ph.D. Candidate in the Department of Archaeology at Boston University. A classical archaeologist, he has
worked extensively in Cyprus, Turkey, and Israel. He specializes in the Hellenistic and Roman worlds, as well as Geographic
Information Systems and D modeling in archaeology. Correspondence to: Brandon R. Olson, Department of Archaeology,
Boston University,  Commonwealth Avenue, Boston, MA  U.S.A. Email: [email protected]
Jody Michael Gordon is an Assistant Professor of Humanities and Social Sciences at Wentworth Institute of Technology in
Boston where he teaches classes on ancient history, art, and architecture. He received his Ph.D. in classical archaeology from
the Department of Classics at the University of Cincinnati and he is an Assistant Director of the Athienou Archaeological
Project in the Republic of Cyprus. Correspondence to: Jody Michael Gordon, Department of Humanities and Social Sciences,
Wentworth Institute of Technology,  Huntington Ave. Boston, MA , U.S.A. Email: gordonj@wit.edu
Curtis Runnels is Professor of Archaeology in the Department of Archaeology at Boston University. He specializes in Aegean prehistory from the Palaeolithic to the Neolithic, and his recent research focuses on site location models for Mesolithic and Palaeolithic sites. He is currently engaged in the study of the Palaeolithic from Crete. Correspondence to: Curtis Runnels, Department of
Archaeology, Boston University,  Commonwealth Avenue, Suite EBoston, MA , U.S.A. Email: [email protected]
Steve Chomyszak is an Assistant Professor of Mechanical Engineering and Technology at Wentworth Institute of Technology in
Boston. He is a specialist in D printing and has taught a variety of classes in the Engineering Center at Wentworth. Correspondence to: Steve Chomyszak, Mechanical Engineering Wentworth, Institute of Technology,  Huntington Ave. Boston, MA
, U.S.A. Email: [email protected]
Lithic Technology , Vol.  No. , –